Signaling Pathways and Molecular Mediators in Metastasis

Signaling Pathways and Molecular Mediators in Metastasis

Alessandro Fatatis Editor

Signaling Pathways and Molecular Mediators in Metastasis

Editor Alessandro Fatatis Drexel University College of Medicine 245 N. 15th Street Philadelphia, PA 19102 USA [email protected]

ISBN 978-94-007-2557-7 e-ISBN 978-94-007-2558-4 DOI 10.1007/978-94-007-2558-4 Springer Dordrecht Heidelberg London New York Library of Congress Control Number: 2011944018 © Springer Science+Business Media B.V. 2012 No part of this work may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, microfilming, recording or otherwise, without written permission from the Publisher, with the exception of any material supplied specifically for the purpose of being entered and executed on a computer system, for exclusive use by the purchaser of the work. Printed on acid-free paper Springer is part of Springer Science+Business Media (www.springer.com)

To Mara, Enrico and Fabrizio, for introducing me to life. To Olimpia and our son Andrea, for making my life worthwhile.

Preface

Since the dawn of modern medicine, the spreading of a tumor has always been regarded as an unfavorable event. However, until recently patients would refer to this condition as a tumor that had ‘returned’ and simply prepare for additional rounds of treatment, as if the recurrence would be a replica of the original disease, just in a different site. Today, patients and physicians alike are fully aware that the detection of metastases is a negative prognostic factor, in most cases leading to serious repercussions on the quality of life and overall survival. Indeed, the treatment of metastatic patients with curative intent is frequently a daunting task, albeit not always bond to fail. The first chapter of this book provides an excellent overview of the problem. The gravity of metastatic disease is often due to the multiplicity of lesions to be treated, combined to locations not easily amenable of surgical excision or irradiation. Although cytotoxic, targeted and in selected cases hormone-deprivation systemic therapies can initially bypass some of these limitations, the onset of resistance mechanisms will eventually obligate physicians to change chemotherapy protocol, switch to different targeted drugs and eventually resolve to palliative measures. Furthermore, we now recognize that metastatic lesions might share very little with their tumors of origin as a myriad of events occurring either at primary or secondary sites can dramatically alter genotypic and phenotypic features of cancer cells during the progression of the disease. Thus, aiming to the successful treatment of metastatic lesions based on information gained from primary tumors should be considered a dangerous overlook. Several chapters of this book provide a compelling review of the current knowledge on the changes occurring in different organ microenvironments that permit malignant colonization and subsequent progression of clinically overt metastases. These changes affect not only epithelial cancer cells, but also the resident cells of the surrounding stroma, immune cells and bone marrow-derived cells that can be locally recruited to create a pre-metastatic niche. A significant percentage of solid tumors are currently diagnosed at their initial stages. However, the progress in screening and diagnostic procedures only marginally translates into major survival benefits for patients, as too many still succumb to vii

viii

Preface

metastatic disease often years after the initial diagnosis. A better understanding of the mechanisms and molecular mediators promoting local tumor progression and invasion into the circulatory and lymphatic systems will lead to appropriate therapeutic strategies aiming to limit the extent and duration of cancer spreading. A number of chapters in this book very effectively address these particular issues. This volume presents the work of scientists at the forefront of the metastasis research field and it is a testimony of the power of their intellect, dedication and efforts to improve the range of treatment options for cancer patients and effectively counteract the most lethal complication of their disease. I knew some of them personally prior to undertaking this project and learned more about the others because of it. It has been a true privilege for me to work on this text with such a group of brilliant co-authors. They all have been enthusiastic about this book from the very beginning and demonstrated a remarkable willingness to participate, despite the variety of academic and clinical commitments they had to attend and the demands of their research groups and medical teams. For this, and for their contribution of competent writing and effective illustrations, I am extremely grateful. I am also indebted to all the staff at Springer, especially Melania Ruiz (Publishing Editor), Ilse Hensen (Publishing Assistant) and Sunil Padman (Project Manager). Their enthusiasm and knowledgeable assistance during the different phases of this project have been very much appreciated. The ensuing of metastatic disease is too often responsible for the demise of patients that would be otherwise successfully treated for their primary tumors. The scientists involved in this project, along with many others in the field, intend to change this grim scenario forever. Philadelphia, PA

Alessandro Fatatis

Contents

1

Introduction ............................................................................................. Robert B. Den and Adam P. Dicker

Part I 2

1

Local Invasion

Reactivation of Epithelial-Mesenchymal Transition in Invasive and Metastatic Cancer ........................................................ Michael D. Amatangelo and Mark E. Stearns

13

3

Biomechanical ECM Switches and Tumor Metastasis ........................ Jacquelyn J. Ames, Calvin P.H. Vary, and Peter C. Brooks

71

4

Moving Aggressively: S100A4 and Tumor Invasion ............................ Reniqua P. House, Sarah C. Garrett, and Anne R. Bresnick

91

5

Regulation of TGF-b Signaling and Metastatic Progression by Tumor Microenvironments ............................................................... 115 Michael K. Wendt and William P. Schiemann

6

Rho GTPases and Their Activators, Guanine Nucleotide Exchange Factors (GEFs): Their Roles in Glioma Cell Invasion ............................................................................................ 143 Bo Hu, Marc Symons, Bodour Salhia, Shannon P. Fortin, Nhan L. Tran, James Rutka, and Shi-Yuan Cheng

7

Impact of Tumor Hypoxia, Src, and Met Signaling in the Dissemination of Tumor Cells ..................................................... 171 Yao Dai, Wenyin Shi, Nikolett Molnar, and Dietmar Siemann

Part II 8

Homing and Colonization at Distant Sites

Cellular and Molecular Biology of Cancer Cell Extravasation .......... 197 J. Matthew Barnes and Michael D. Henry

ix

x

Contents

9

Role of CXCL12 and CXCR4 in Tumor Biology and Metastasis ......................................................................................... 221 Ramesh K. Ganju, Yadwinder S. Deol, and Mohd W. Nasser

10

Transcription Factors Stat5a/b and Stat3 in Prostate Cancer Growth and Metastases ............................................................. 245 Tuomas K. Mirtti, Pooja Talati, and Marja T. Nevalainen

11

Survival and Growth of Prostate Cancer Cells in the Bone: Role of the Alpha-Receptor for Platelet-Derived Growth Factor in Supporting Early Metastatic Foci ......................................... 261 Qingxin Liu, Yun Zhang, Danielle Jernigan, and Alessandro Fatatis

12

Androgen Receptor Regulation of Prostate Cancer Progression and Metastasis .................................................................... 277 R.S. Schrecengost, M.A. Augello, and Karen E. Knudsen

13

Integrin Signaling in Angiogenesis and Metastatic Cancer Progression in the Brain............................................................ 311 Mihaela Lorger and Brunhilde Felding-Habermann

14

Tumor and Host Determinants of Pulmonary Metastasis in Bladder Cancer ................................................................ 331 Neveen Said and Dan Theodorescu

15

The Influence of Metastatic Breast Cancer on the Bone Microenvironment ............................................................. 347 Andrea M. Mastro, Donna M. Sosnoski, Venkatesh Krishnan, and Karen M. Bussard

Index ................................................................................................................. 369

Chapter 1

Introduction Robert B. Den, MD and Adam P. Dicker, MD, PhD

Abstract Local therapies have continued to evolve with advances in surgical and radiation therapy techniques. This has contributed to improvements in survivorships in early stage disease. However, survival rates for metastatic disease remain poor. There has been increasing evidence that local therapy directly influences metastatic growth and overall survival. Both clinical and laboratory evidence support this hypothesis and an increasing understanding of the role of circulating tumor cells has further galvanized the idea of crosstalk between distant sites and the local disease. The current clinical arena is ripe for new innovations and the scientific advancements described in the remainder of this text set the stage upon which such progress will occur.

Cancer is the second leading cause of death in the United States for both men and women [1]. Currently, there are 1.5 million new cases diagnosed per year with a declining cancer-specific mortality rate of 600,000 per year due in large part to early detection and improved therapeutics. Five-year survivorship is markedly decreased between localized and metastatic disease. Regardless of the initial tumor type, 5-year overall survival with stage IV malignancy ranges from 2% to 30% in comparison to

R.B. Den, MD Assistant Professor, Department of Radiation Oncology and Cancer Biology, Kimmel Cancer Center, Jefferson Medical College of Thomas Jefferson University, Philadelphia, PA 19107, USA A.P. Dicker, MD, PhD () Professor and Chair of Radiation Oncology, Professor of Pharmacology and Experimental Therapeutics, Department of Radiation Oncology, Kimmel Cancer Center, Jefferson Medical College of Thomas Jefferson University, Philadelphia, PA 19107, USA e-mail: [email protected]

A. Fatatis (ed.), Signaling Pathways and Molecular Mediators in Metastasis, DOI 10.1007/978-94-007-2558-4_1, © Springer Science+Business Media B.V. 2012

1

2

R.B. Den and A.P. Dicker 100

25

90 20

70 60

15

50 40

10

DM Rate

Local Control

80

30

20

5

10 0

0 <70

70

75

80

Radiation Dose (Gy) Local Control

Distant Metastatic Free Survival

Fig. 1.1 Increasing radiation dose results in both an increase in local control as well as a decrease in distant metastatic free survival (Data based on Zelefsky et al. [22])

localized disease where 5-year survival is between 20% and 100% [1]. Thus, it is clear that the development of distant disease portends to patient death. For the vast majority of malignancies, patients are diagnosed when the disease is localized. Local control rates continue to improve and have shifted the management of cancer from pure palliation into a chronic disease. Local control has greatly improved due to advances in surgical technique [2], improved understanding of pathologic features that are prognostic for further adjuvant [3–5], and more sophisticated planning and delivery of radiation therapy [6, 7]. In the field of prostate cancer, multiple phase III trials have demonstrated the benefit to radiation dose escalation [8–15]. These trials have shown statistically significant improvements in biochemical free survival without compromise of higher toxicities in either genitourinary, gastrointestinal, or sexual metrics as reported by both patients and physicians [16–21]. Retrospective series have demonstrated that increased radiation doses reduced the rate of persistent positive disease on post treatment prostate biopsy [22] (Fig. 1.1). A meta-analysis of these trials demonstrated a linear correlation between total dose and biochemical progression free survival in all risk groups [23]. Biochemical progression has been shown to be a surrogate endpoint for prostate cancer specific mortality [24]. At the same time, the combination of androgen suppression therapy with radiation therapy has consistently demonstrated improvements in not only local control [22] and biochemical control, but in overall survival benefit as well. Multiple studies have shown that even short term (4–6 months) of androgen ablation with radiation therapy translates into improved cause-specific survival [25–32]. Further, in patients with high risk or locally advanced disease long-term hormonal suppression (2–3 years) in combination with radiotherapy results in improved local control and overall survival [33–38].

1

Introduction

3

Another primarily hormonally driven tumor, breast cancer, has seen decreased rate of in field breast recurrences with increased doses of adjuvant radiation therapy [39–41] and the addition of systemic agents such as tamoxifen [42]. While hormonal manipulation improved local control, it has been demonstrated in multiple large randomized studies to be insufficient to prevent local regional recurrences as monotherapy [43–45]. Further, local control has been demonstrated to lead to improvement in overall survival as demonstrated in the Early Breast Cancer Trialists’ Collaborative Group (EBCTCG) most recent meta-analysis [46]. A 20% absolute improvement in local control with the addition of radiation therapy at 5 years translated into a 5% improvement in overall survival at 15 years. This was seen both for women undergoing breast conserving surgery as well as those receiving post mastectomy radiation therapy. A different approach to improve local therapy was adopted in the field of head and neck cancer with regard to radiation therapy. Instead of increasing the dose of radiation, alternative fractionation schemes were used to provide greater biologically equivalent doses. There are numerous phase III trials supporting non-standard fractionation for head and neck cancer resulting in a 10% absolute improvement in local control [47–49]. In addition, the addition of chemotherapy, specifically a platinumbased regimen, has consistently increased both local control and overall survival [50–53]. Further, given the high levels of EGFR seen in head and neck cancers, the targeting of this receptor using the monoclonal antibody Cetuximab combined with radiation therapy results in improved local control and survival as well [54]. While the addition of systemic agents to local therapy (surgery or radiation) has clearly shown benefit in improvement in overall survival, more recently there has been increasing evidence that the corollary is true; the addition of local therapy to systemic therapy translates into increased overall survival. Two phase III trials within prostate cancer have demonstrated the survival benefit with the addition of radiation therapy to androgen suppression therapy [55, 56]. The SPCG-7/SFUO-3 trial [55] revealed a 12% absolute improvement in cause specific survival and a 9% improvement in overall survival with the addition of radiation therapy. The Intergroup trial [56] showed a similar 8% improvement in cause specific survival and a 5% absolute improvement in overall survival. These patients with locally advanced and high-risk features had previously been thought to have subclinical microscopic metastatic disease and it was presumed that there was no role for localized therapy. Further, the addition of radiation did not adversely impact the quality of life of these patients [57]. A post hoc analysis of SWOG 8894 revealed that amongst patients with metastatic cancer those who underwent prior radical prostatectomy had a better response to androgen ablation and better survival than those with an untreated prostate [58]. In a study of hormone refractory metastatic prostate cancer the patients who underwent prior prostatectomy/radiation had better survival than those who had no prior local treatment [59]. These clinical studies suggest that removing the prostate in metastatic prostate cancer might result in a more durable response to systemic treatment. The benefit of local control to improve overall survival even once the disease has metastasized is clearly demonstrated in kidney cancer. Cytoreductive nephrectomy is well established as a critical component of the management of metastatic renal cell carcinoma. This practice is based on results of two large randomized trials [60, 61] that

4

R.B. Den and A.P. Dicker

Primary Tumor Radiation + Surgery

Primary Tumor Radiation + Surgery

Distant metastasis and primary tumor reduced or eliminated

Distant metastasis and primary tumor reduced or eliminated

Fig. 1.2 Treatment of the primary tumor with a localized modality directly impacts sites of metastatic disease resulting in improvements in cancer specific and overall survival

compared surgery with interferon to interferon monotherapy. Both trials [62] showed significant survival advantages with the addition of surgery (median survival 13.6 vs. 7.8 months) and recently, retrospective data [63] suggested the benefit of nephrectomy in the setting of vascular endothelial growth factor targeted therapy, which has become the new standard of care for patients with metastatic renal cell carcinoma. Recently, there has been increasing interest in understanding the interplay between local disease and metastatic deposits. Historically, cancer cells have been thought to possess the ability to leave the primary tumor and seed metastatic deposits and a clear temporal relation between local failure and metastasis has been documented. William Halstead, who posited that breast cancer spread contiguously from the primary site through local and regional nodes before reaching metastatic sites, pioneered this relationship in the breast cancer literature. Currently, it is unclear whether locally persistent disease results in metastatic formation or is a prognostic factor for more virulent disease. If the former is true, then improvement in local therapy should translated into improved distant control (Fig. 1.2). In a trial randomizing men with advanced prostate cancer following prostatectomy to adjuvant radiation therapy or observation not only was there an increase in overall

1

Introduction

5

survival and but in distant metastatic free survival as well [64]. As well, Kuban et al. [14] demonstrated that the addition of 4 days of localized treatment corresponded to a reduction in the distant metastatic rate of 96% vs. 83% at 8 years. Zelefsky et al. [22] also demonstrated that post treatment biopsy positivity rate was significantly associated with worse distant metastases free survival (DMFS) and increased prostate cancer specific death. Coen et al. [65] showed a statistically significant difference in DMFS between those patients with local control vs. locally persistent disease (87% v 80%, 77% v 61%, 72% v 37%) at 5, 10, and 15 years respectively. Further, it was demonstrated that there was an increasing risk of distant metastasis over time in patients who ultimately develop local failure. Thus, through improved treatment of the primary site, there is decreased widespread metastatic burden. The identification of circulating tumor cells and the observation that these cells can colonize tumors of origin [66] has led to the “tumor self seeding” hypothesis [67]. This theory highlights the increasing importance of not only localized therapies, but also improvements in chemotherapy and targeted biologics. There has also been increasing excitement in the use of “local” approaches to treat oligometastatic disease. Stereotactic body radiation therapy or surgery has been used in the setting of patients with limited distant disease burden with success [68]. Currently, there are phase I trials examining the incorporation of targeted agents with radiation therapy in this patient population [69]. Historical response to single agent chemotherapy and combination therapy has poor results [70]. With increasing understanding of the underlying molecular pathways and drivers of cancer, improvements in survival have been demonstrated [71]. In fact, the BATTLE (Biomarker-Integrated Approaches of Targeted Therapy for Lung Cancer Elimination) trial presented at the 2009 ASCO meeting [72] and updated at the 2010 AACR Meeting [73] demonstrated the ability to select patients based on mutational status to different biologic agents. Further, these findings suggested that this approach improves the response rate in patients when compared to historical controls treated with traditional chemotherapy. The true benefit of such an approach is that it allows the optimal therapy to be delivered to a patient population which generally has overall poor functional status and impaired reserve to tolerate aggressive therapy. Clinically there is great interest in improvements to current treatment options and a need for more robust understandings of the molecular drivers of both formation and spread of metastases. While improvements in local control have increased greatly over the past decade, large breakthroughs in the management of metastatic disease have yet to be fully realized. Thus, the scientific advancements described in the remainder of this work set the stage upon which such progress will occur.

References 1. Jemal A, Siegel R, Xu J, Ward E (2010) Cancer statistics, 2010. CA Cancer J Clin 60(5):277–300 2. Quirke P, Steele R, Monson J, Grieve R, Khanna S, Couture J, O’Callaghan C, Myint AS, Bessell E, Thompson LC, Parmar M, Stephens RJ, Sebag-Montefiore D (2009) Effect of the plane of

6

3.

4.

5.

6. 7. 8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

R.B. Den and A.P. Dicker surgery achieved on local recurrence in patients with operable rectal cancer: a prospective study using data from the MRC CR07 and NCIC-CTG CO16 randomised clinical trial. Lancet 373(9666):821–828 Kao J, Lavaf A, Teng MS, Huang D, Genden EM (2008) Adjuvant radiotherapy and survival for patients with node-positive head and neck cancer: an analysis by primary site and nodal stage. Int J Radiat Oncol Biol Phys 71(2):362–370 Bernier J, Cooper JS, Pajak TF, van Glabbeke M, Bourhis J, Forastiere A, Ozsahin EM, Jacobs JR, Jassem J, Ang KK, Lefebvre JL (2005) Defining risk levels in locally advanced head and neck cancers: a comparative analysis of concurrent postoperative radiation plus chemotherapy trials of the EORTC (#22931) and RTOG (# 9501). Head NeckOct 27(10):843–850 Pignatti F, van den Bent M, Curran D, Debruyne C, Sylvester R, Therasse P, Afra D, Cornu P, Bolla M, Vecht C, Karim AB (2002) Prognostic factors for survival in adult patients with cerebral low-grade glioma. J Clin Oncol 20(8):2076–2084 Palta JR, Deye JA, Ibbott GS, Purdy JA, Urie MM (2004) Credentialing of institutions for IMRT in clinical trials. Int J Radiat Oncol Biol Phys 59(4):1257–1259; author reply 9–61 Apisarnthanarax S, Chao KS (2005) Current imaging paradigms in radiation oncology. Radiat Res 163(1):1–25 Dearnaley DP, Sydes MR, Graham JD, Aird EG, Bottomley D, Cowan RA, Huddart RA, Jose CC, Matthews JH, Millar J, Moore AR, Morgan RC, Russell JM, Scrase CD, Stephens RJ, Syndikus I, Parmar MK (2007) Escalated-dose versus standard-dose conformal radiotherapy in prostate cancer: first results from the MRC RT01 randomised controlled trial. Lancet Oncol 8(6):475–487 Hoskin PJ, Motohashi K, Bownes P, Bryant L, Ostler P (2007) High dose rate brachytherapy in combination with external beam radiotherapy in the radical treatment of prostate cancer: initial results of a randomised phase three trial. Radiother Oncol 84(2):114–120 Peeters ST, Heemsbergen WD, Koper PC, van Putten WL, Slot A, Dielwart MF, Bonfrer JM, Incrocci L, Lebesque JV (2006) Dose-response in radiotherapy for localized prostate cancer: results of the Dutch multicenter randomized phase III trial comparing 68Gy of radiotherapy with 78Gy. J Clin Oncol 24(13):1990–1996 Zietman AL, DeSilvio ML, Slater JD, Rossi CJ Jr, Miller DW, Adams JA, Shipley WU (2005) Comparison of conventional-dose vs. high-dose conformal radiation therapy in clinically localized adenocarcinoma of the prostate: a randomized controlled trial. JAMA 294(10):1233–1239 Zietman AL, Bae K, Slater JD, Shipley WU, Efstathiou JA, Coen JJ, Bush DA, Lunt M, Spiegel DY, Skowronski R, Jabola BR, Rossi CJ (2010) Randomized trial comparing conventionaldose with high-dose conformal radiation therapy in early-stage adenocarcinoma of the prostate: long-term results from proton radiation oncology group/american college of radiology 95–09. J Clin Oncol 28(7):1106–1111 Pollack A, Zagars GK, Starkschall G, Antolak JA, Lee JJ, Huang E, von Eschenbach AC, Kuban DA, Rosen I (2002) Prostate cancer radiation dose response: results of the M. D. Anderson phase III randomized trial. Int J Radiat Oncol Biol Phys 53(5):1097–1105 Kuban DA, Tucker SL, Dong L, Starkschall G, Huang EH, Cheung MR, Lee AK, Pollack A (2008) Long-term results of the M. D. Anderson randomized dose-escalation trial for prostate cancer. Int J Radiat Oncol Biol Phys 70(1):67–74 Sathya JR, Davis IR, Julian JA, Guo Q, Daya D, Dayes IS, Lukka HR, Levine M (2005) Randomized trial comparing iridium implant plus external-beam radiation therapy with external-beam radiation therapy alone in node-negative locally advanced cancer of the prostate. J Clin Oncol 23(6):1192–1199 Dearnaley DP, Sydes MR, Langley RE, Graham JD, Huddart RA, Syndikus I, Matthews JH, Scrase CD, Jose CC, Logue J, Stephens RJ (2007) The early toxicity of escalated versus standard dose conformal radiotherapy with neo-adjuvant androgen suppression for patients with localised prostate cancer: results from the MRC RT01 trial (ISRCTN47772397). Radiother Oncol 83(1):31–41 Beckendorf V, Guerif S, Le Prise E, Cosset JM, Lefloch O, Chauvet B, Salem N, Chapet O, Bourdin S, Bachaud JM, Maingon P, Lagrange JL, Malissard L, Simon JM, Pommier P,

1

18.

19.

20.

21.

22.

23.

24.

25.

26.

27. 28.

29.

30.

31.

32.

Introduction

7

Hay MH, Dubray B, Luporsi E, Bey P (2004) The GETUG 70Gy vs. 80Gy randomized trial for localized prostate cancer: feasibility and acute toxicity. Int J Radiat Oncol Biol Phys 60(4):1056–1065 Peeters ST, Heemsbergen WD, van Putten WL, Slot A, Tabak H, Mens JW, Lebesque JV, Koper PC (2005) Acute and late complications after radiotherapy for prostate cancer: results of a multicenter randomized trial comparing 68Gy to 78Gy. Int J Radiat Oncol Biol Phys 61(4):1019–1034 van der Wielen GJ, van Putten WL, Incrocci L (2007) Sexual function after three-dimensional conformal radiotherapy for prostate cancer: results from a dose-escalation trial. Int J Radiat Oncol Biol Phys 68(2):479–484 Talcott JA, Rossi C, Shipley WU, Clark JA, Slater JD, Niemierko A, Zietman AL (2010) Patient-reported long-term outcomes after conventional and high-dose combined proton and photon radiation for early prostate cancer. JAMA 303(11):1046–1053 Storey MR, Pollack A, Zagars G, Smith L, Antolak J, Rosen I (2000) Complications from radiotherapy dose escalation in prostate cancer: preliminary results of a randomized trial. Int J Radiat Oncol Biol Phys 48(3):635–642 Zelefsky MJ, Reuter VE, Fuks Z, Scardino P, Shippy A (2008) Influence of local tumor control on distant metastases and cancer related mortality after external beam radiotherapy for prostate cancer. J Urol 179(4):1368–1373; discussion 73 Viani GA, Stefano EJ, Afonso SL (2009) Higher-than-conventional radiation doses in localized prostate cancer treatment: a meta-analysis of randomized, controlled trials. Int J Radiat Oncol Biol Phys 74(5):1405–1418 Denham JW, Steigler A, Wilcox C, Lamb DS, Joseph D, Atkinson C, Matthews J, Tai KH, Spry NA, Christie D, Gleeson PS, Greer PB, D’Este C (2008) Time to biochemical failure and prostate-specific antigen doubling time as surrogates for prostate cancer-specific mortality: evidence from the TROG 96.01 randomised controlled trial. Lancet Oncol 9(11):1058–1068 Denham JW, Steigler A, Lamb DS, Joseph D, Mameghan H, Turner S, Matthews J, Franklin I, Atkinson C, North J, Poulsen M, Christie D, Spry NA, Tai KH, Wynne C, Duchesne G, Kovacev O, D’Este C (2005) Short-term androgen deprivation and radiotherapy for locally advanced prostate cancer: results from the Trans-Tasman Radiation Oncology Group 96.01 randomised controlled trial. Lancet Oncol 6(11):841–850 D’Amico AV, Manola J, Loffredo M, Renshaw AA, DellaCroce A, Kantoff PW (2004) 6-month androgen suppression plus radiation therapy vs. radiation therapy alone for patients with clinically localized prostate cancer: a randomized controlled trial. JAMA 292(7):821–827 D’Amico AV, Chen MH, Renshaw AA, Loffredo M, Kantoff PW (2008) Androgen suppression and radiation vs. radiation alone for prostate cancer: a randomized trial. JAMA 299(3):289–295 Laverdiere J, Gomez JL, Cusan L, Suburu ER, Diamond P, Lemay M, Candas B, Fortin A, Labrie F (1997) Beneficial effect of combination hormonal therapy administered prior and following external beam radiation therapy in localized prostate cancer. Int J Radiat Oncol Biol Phys 37(2):247–252 Laverdiere J, Nabid A, De Bedoya LD, Ebacher A, Fortin A, Wang CS, Harel F (2004) The efficacy and sequencing of a short course of androgen suppression on freedom from biochemical failure when administered with radiation therapy for T2–T3 prostate cancer. J Urol 171(3):1137–1140 Pilepich MV, Krall JM, al-Sarraf M, John MJ, Doggett RL, Sause WT, Lawton CA, Abrams RA, Rotman M, Rubin P et al (1995) Androgen deprivation with radiation therapy compared with radiation therapy alone for locally advanced prostatic carcinoma: a randomized comparative trial of the Radiation Therapy Oncology Group. Urology 45(4):616–623 Pilepich MV, Winter K, John MJ, Mesic JB, Sause W, Rubin P, Lawton C, Machtay M, Grignon D (2001) Phase III radiation therapy oncology group (RTOG) trial 86–10 of androgen deprivation adjuvant to definitive radiotherapy in locally advanced carcinoma of the prostate. Int J Radiat Oncol Biol Phys 50(5):1243–1252 Roach M 3rd, Bae K, Speight J, Wolkov HB, Rubin P, Lee RJ, Lawton C, Valicenti R, Grignon D, Pilepich MV (2008) Short-term neoadjuvant androgen deprivation therapy and external-beam

8

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

R.B. Den and A.P. Dicker radiotherapy for locally advanced prostate cancer: long-term results of RTOG 8610. J Clin Oncol 26(4):585–591 Pilepich MV, Winter K, Lawton CA, Krisch RE, Wolkov HB, Movsas B, Hug EB, Asbell SO, Grignon D (2005) Androgen suppression adjuvant to definitive radiotherapy in prostate carcinoma–long-term results of phase III RTOG 85–31. Int J Radiat Oncol Biol Phys 61(5):1285–1290 Bolla M, de Reijke TM, Van Tienhoven G, Van den Bergh AC, Oddens J, Poortmans PM, Gez E, Kil P, Akdas A, Soete G, Kariakine O, van der Steen-Banasik EM, Musat E, Pierart M, Mauer ME, Collette L (2009) Duration of androgen suppression in the treatment of prostate cancer. N Engl J Med 360(24):2516–2527 Bolla M, Gonzalez D, Warde P, Dubois JB, Mirimanoff RO, Storme G, Bernier J, Kuten A, Sternberg C, Gil T, Collette L, Pierart M (1997) Improved survival in patients with locally advanced prostate cancer treated with radiotherapy and goserelin. N Engl J Med 337(5): 295–300 Bolla M, Collette L, Blank L, Warde P, Dubois JB, Mirimanoff RO, Storme G, Bernier J, Kuten A, Sternberg C, Mattelaer J, Lopez Torecilla J, Pfeffer JR, Lino Cutajar C, Zurlo A, Pierart M (2002) Long-term results with immediate androgen suppression and external irradiation in patients with locally advanced prostate cancer (an EORTC study): a phase III randomised trial. Lancet 360(9327):103–106 Hanks GE, Pajak TF, Porter A, Grignon D, Brereton H, Venkatesan V, Horwitz EM, Lawton C, Rosenthal SA, Sandler HM, Shipley WU (2003) Phase III trial of long-term adjuvant androgen deprivation after neoadjuvant hormonal cytoreduction and radiotherapy in locally advanced carcinoma of the prostate: the Radiation Therapy Oncology Group Protocol 92–02. J Clin Oncol 21(21):3972–3978 Horwitz EM, Bae K, Hanks GE, Porter A, Grignon DJ, Brereton HD, Venkatesan V, Lawton CA, Rosenthal SA, Sandler HM, Shipley WU (2008) Ten-year follow-up of radiation therapy oncology group protocol 92–02: a phase III trial of the duration of elective androgen deprivation in locally advanced prostate cancer. J Clin Oncol 26(15):2497–2504 Romestaing P, Lehingue Y, Carrie C, Coquard R, Montbarbon X, Ardiet JM, Mamelle N, Gerard JP (1997) Role of a 10-Gy boost in the conservative treatment of early breast cancer: results of a randomized clinical trial in Lyon, France. J Clin Oncol 15(3):963–968 Bartelink H, Horiot JC, Poortmans PM, Struikmans H, Van den Bogaert W, Fourquet A, Jager JJ, Hoogenraad WJ, Oei SB, Warlam-Rodenhuis CC, Pierart M, Collette L (2007) Impact of a higher radiation dose on local control and survival in breast-conserving therapy of early breast cancer: 10-year results of the randomized boost versus no boost EORTC 22881–10882 trial. J Clin Oncol 25(22):3259–3265 Polgar C, Fodor J, Orosz Z, Major T, Takacsi-Nagy Z, Mangel LC, Sulyok Z, Somogyi A, Kasler M, Nemeth G (2002) Electron and high-dose-rate brachytherapy boost in the conservative treatment of stage I-II breast cancer first results of the randomized Budapest boost trial. Strahlenther Onkol 178(11):615–623 Fisher B, Bryant J, Dignam JJ, Wickerham DL, Mamounas EP, Fisher ER, Margolese RG, Nesbitt L, Paik S, Pisansky TM, Wolmark N (2002) Tamoxifen, radiation therapy, or both for prevention of ipsilateral breast tumor recurrence after lumpectomy in women with invasive breast cancers of one centimeter or less. J Clin Oncol 20(20):4141–4149 Potter R, Gnant M, Kwasny W, Tausch C, Handl-Zeller L, Pakisch B, Taucher S, Hammer J, Luschin-Ebengreuth G, Schmid M, Sedlmayer F, Stierer M, Reiner G, Kapp K, Hofbauer F, Rottenfusser A, Postlberger S, Haider K, Draxler W, Jakesz R (2007) Lumpectomy plus tamoxifen or anastrozole with or without whole breast irradiation in women with favorable early breast cancer. Int J Radiat Oncol Biol Phys 68(2):334–340 Hughes KS, Schnaper LA, Berry D, Cirrincione C, McCormick B, Shank B, Wheeler J, Champion LA, Smith TJ, Smith BL, Shapiro C, Muss HB, Winer E, Hudis C, Wood W, Sugarbaker D, Henderson IC, Norton L (2004) Lumpectomy plus tamoxifen with or without irradiation in women 70 years of age or older with early breast cancer. N Engl J Med 351(10):971–977

1

Introduction

9

45. Fyles AW, McCready DR, Manchul LA, Trudeau ME, Merante P, Pintilie M, Weir LM, Olivotto IA (2004) Tamoxifen with or without breast irradiation in women 50 years of age or older with early breast cancer. N Engl J Med 351(10):963–970 46. Clarke M, Collins R, Darby S, Davies C, Elphinstone P, Evans E, Godwin J, Gray R, Hicks C, James S, MacKinnon E, McGale P, McHugh T, Peto R, Taylor C, Wang Y (2005) Effects of radiotherapy and of differences in the extent of surgery for early breast cancer on local recurrence and 15-year survival: an overview of the randomised trials. Lancet 366(9503):2087–2106 47. Overgaard J, Hansen HS, Specht L, Overgaard M, Grau C, Andersen E, Bentzen J, Bastholt L, Hansen O, Johansen J, Andersen L, Evensen JF (2003) Five compared with six fractions per week of conventional radiotherapy of squamous-cell carcinoma of head and neck: DAHANCA 6 and 7 randomised controlled trial. Lancet 362(9388):933–940 48. Fu KK, Pajak TF, Trotti A, Jones CU, Spencer SA, Phillips TL, Garden AS, Ridge JA, Cooper JS, Ang KK (2000) A Radiation Therapy Oncology Group (RTOG) phase III randomized study to compare hyperfractionation and two variants of accelerated fractionation to standard fractionation radiotherapy for head and neck squamous cell carcinomas: first report of RTOG 9003. Int J Radiat Oncol Biol Phys 48(1):7–16 49. Horiot JC, Le Fur R, N’Guyen T, Chenal C, Schraub S, Alfonsi S, Gardani G, Van Den Bogaert W, Danczak S, Bolla M et al (1992) Hyperfractionation versus conventional fractionation in oropharyngeal carcinoma: final analysis of a randomized trial of the EORTC cooperative group of radiotherapy. Radiother Oncol 25(4):231–241 50. Calais G, Alfonsi M, Bardet E, Sire C, Germain T, Bergerot P, Rhein B, Tortochaux J, Oudinot P, Bertrand P (1999) Randomized trial of radiation therapy versus concomitant chemotherapy and radiation therapy for advanced-stage oropharynx carcinoma. J Natl Cancer Inst 91(24): 2081–2086 51. Forastiere AA, Goepfert H, Maor M, Pajak TF, Weber R, Morrison W, Glisson B, Trotti A, Ridge JA, Chao C, Peters G, Lee DJ, Leaf A, Ensley J, Cooper J (2003) Concurrent chemotherapy and radiotherapy for organ preservation in advanced laryngeal cancer. N Engl J Med 349(22):2091–2098 52. Chua DT, Ma J, Sham JS, Mai HQ, Choy DT, Hong MH, Lu TX, Min HQ (2005) Long-term survival after cisplatin-based induction chemotherapy and radiotherapy for nasopharyngeal carcinoma: a pooled data analysis of two phase III trials. J Clin Oncol 23(6):1118–1124 53. Al-Sarraf M, LeBlanc M, Giri PG, Fu KK, Cooper J, Vuong T, Forastiere AA, Adams G, Sakr WA, Schuller DE, Ensley JF (1998) Chemoradiotherapy versus radiotherapy in patients with advanced nasopharyngeal cancer: phase III randomized Intergroup study 0099. J Clin Oncol 16(4):1310–1317 54. Bonner JA, Harari PM, Giralt J, Azarnia N, Shin DM, Cohen RB, Jones CU, Sur R, Raben D, Jassem J, Ove R, Kies MS, Baselga J, Youssoufian H, Amellal N, Rowinsky EK, Ang KK (2006) Radiotherapy plus cetuximab for squamous-cell carcinoma of the head and neck. N Engl J Med 354(6):567–578 55. Widmark A, Klepp O, Solberg A, Damber JE, Angelsen A, Fransson P, Lund JA, Tasdemir I, Hoyer M, Wiklund F, Fossa SD (2009) Endocrine treatment, with or without radiotherapy, in locally advanced prostate cancer (SPCG-7/SFUO-3): an open randomised phase III trial. Lancet 373(9660):301–308 56. Warde PR, Mason MD, Sydes MR, Gospodarowicz MK, Swanson GP, Kirkbride P, Kostashuk E, Hetherington J, Ding K, Parulekar W (2010) Intergroup randomized phase III study of androgen deprivation therapy (ADT) plus radiation therapy (RT) in locally advanced prostate cancer (CaP) (NCIC-CTG, SWOG, MRC-UK, INT: T94–0110; NCT00002633). ASCO annual meeting, 2010: Journal of Clinical Oncology, 2010, p 18s 57. Fransson P, Lund JA, Damber JE, Klepp O, Wiklund F, Fossa S, Widmark A (2009) Quality of life in patients with locally advanced prostate cancer given endocrine treatment with or without radiotherapy: 4-year follow-up of SPCG-7/SFUO-3, an open-label, randomised, phase III trial. Lancet Oncol 10(4):370–380 58. Thompson IM, Tangen C, Basler J, Crawford ED (2002) Impact of previous local treatment for prostate cancer on subsequent metastatic disease. J Urol 168(3):1008–1012

10

R.B. Den and A.P. Dicker

59. Carducci MA, Saad F, Abrahamsson PA, Dearnaley DP, Schulman CC, North SA, Sleep DJ, Isaacson JD, Nelson JB (2007) A phase 3 randomized controlled trial of the efficacy and safety of atrasentan in men with metastatic hormone-refractory prostate cancer. Cancer 110(9): 1959–1966 60. Flanigan RC, Salmon SE, Blumenstein BA, Bearman SI, Roy V, McGrath PC, Caton JR Jr, Munshi N, Crawford ED (2001) Nephrectomy followed by interferon alfa-2b compared with interferon alfa-2b alone for metastatic renal-cell cancer. N Engl J Med 345(23):1655–1659 61. Mickisch GH, Garin A, van Poppel H, de Prijck L, Sylvester R (2001) Radical nephrectomy plus interferon-alfa-based immunotherapy compared with interferon alfa alone in metastatic renal-cell carcinoma: a randomised trial. Lancet 358(9286):966–970 62. Flanigan RC, Mickisch G, Sylvester R, Tangen C, Van Poppel H, Crawford ED (2004) Cytoreductive nephrectomy in patients with metastatic renal cancer: a combined analysis. J Urol 171(3):1071–1076 63. Choueiri TK, Xie W, Kollmannsberger C, North S, Knox JJ, Lampard JG, McDermott DF, Rini BI, Heng DY (2011) The impact of cytoreductive nephrectomy on survival of patients with metastatic renal cell carcinoma receiving vascular endothelial growth factor targeted therapy. J Urol 185(1):60–66 64. Thompson IM, Tangen CM, Paradelo J, Lucia MS, Miller G, Troyer D, Messing E, Forman J, Chin J, Swanson G, Canby-Hagino E, Crawford ED (2009) Adjuvant radiotherapy for pathological T3N0M0 prostate cancer significantly reduces risk of metastases and improves survival: long-term follow up of a randomized clinical trial. J Urol 181(3):956–962 65. Coen JJ, Zietman AL, Thakral H, Shipley WU (2002) Radical radiation for localized prostate cancer: local persistence of disease results in a late wave of metastases. J Clin Oncol 20(15):3199–3205 66. Kim MY, Oskarsson T, Acharyya S, Nguyen DX, Zhang XH, Norton L, Massague J (2009) Tumor self-seeding by circulating cancer cells. Cell 139(7):1315–1326 67. Norton L, Massague J (2006) Is cancer a disease of self-seeding? Nat Med 8:875–878 68. Macdermed DM, Weichselbaum RR, Salama JK (2008) A rationale for the targeted treatment of oligometastases with radiotherapy. J Surg Oncol 98(3):202–206 69. Kao J, Packer S, Vu HL, Schwartz ME, Sung MW, Stock RG, Lo YC, Huang D, Chen SH, Cesaretti JA (2009) Phase 1 study of concurrent sunitinib and image-guided radiotherapy followed by maintenance sunitinib for patients with oligometastases: acute toxicity and preliminary response. Cancer 115(15):3571–3580 70. Rajeswaran A, Trojan A, Burnand B, Giannelli M (2008) Efficacy and side effects of cisplatinand carboplatin-based doublet chemotherapeutic regimens versus non-platinum-based doublet chemotherapeutic regimens as first line treatment of metastatic non-small cell lung carcinoma: a systematic review of randomized controlled trials. Lung Cancer 59(1):1–11 71. Maemondo M, Inoue A, Kobayashi K, Sugawara S, Oizumi S, Isobe H, Gemma A, Harada M, Yoshizawa H, Kinoshita I, Fujita Y, Okinaga S, Hirano H, Yoshimori K, Harada T, Ogura T, Ando M, Miyazawa H, Tanaka T, Saijo Y, Hagiwara K, Morita S, Nukiwa T (2010) Gefitinib or chemotherapy for non-small-cell lung cancer with mutated EGFR. N Engl J Med 362(25): 2380–2388 72. Kim ES, Herbst RS, Lee JJ, Blumenschein G, Tsao A, Wistuba I, Alden C, Gupta S, Stewart D, Hong WK (2009) Phase II randomized study of biomarker directed treatment for non-small cell lung cancer (NSCLC): The BATTLE (Biomarker-Integrated Approaches of Targeted Therapy for Lung Cancer Elimination) clinical trial program. ASCO annual meeting: Journal of Clinical Oncology, 2009 73. Kim ES, Herbst RS, Blumenschein GR, Tsao A, Alden CM, Tang X, Liu S, Stewart DJ, Heymach JV, Tran HT, Hicks ME, Erasmus J, Gupta S, Powis G, Lippman SM, Wistuba II, Hong WK (2010) The BATTLE trial (Biomarker-integrated Approaches of Targeted Therapy for Lung Cancer Elimination): personalizing therapy for lung cancer. AACR 101st annual meeting, 2010

Part I

Local Invasion

Chapter 2

Reactivation of Epithelial-Mesenchymal Transition in Invasive and Metastatic Cancer Michael D. Amatangelo and Mark E. Stearns

Abstract During development cells of epithelial and mesenchymal origin convert between the two phenotypes in what has been described as Epithelial-Mesenchymal Transition (EMT) and Mesenchymal-Epithelial Transition (MET). The common characteristics exhibited during EMT are a loss of epithelial cell contacts, a reorganization of cytoskeletal components to promote a motile phenotype, and a remodeling of the surrounding extracellular matrix to allow for invasion. These events are tightly regulated and required for proper cellular organization and organogenesis during development. Studies in cancer models have identified an analogous plasticity of some epithelial cancer cells which acquire mesenchymal features as a means to escape the primary tumor mass. During the initial stages of tumor metastasis a complex series of events occur in which cancer cells leave the original tumor site and migrate to other parts of the body via the bloodstream and/or the lymphatic system. All metastatic cells must first acquire the abilities to disseminate, migrate and invade the surrounding tissue to allow for metastasis to occur. Thus, a reactivation of developmental pathways resulting in an EMT-like program is one possible mechanism by which cells acquire these capabilities and are able to form distal metastasis. Intriguingly, many similarities between developmental and oncogenic EMT have been identified and has led to our understanding of common signaling pathways (including TGF-beta, Ras and Wnt), transcriptional regulators (including the Snail, Zeb and Twist families) and microRNAs (including let-7 and miR-200 families) which regulate EMT. Aberrant regulation of these pathways and factors is associated with increased metastatic potential in vitro and in animal models and correlate with poor clinical outcomes. This chapter focuses on the EMT program in cancer, its regulation, its parallels to developmental EMT and its significance to the progression to metastatic disease.

M.D. Amatangelo • M.E. Stearns (*) College of Medicine, Drexel University, Philadelphia, PA, USA e-mail: [email protected]

A. Fatatis (ed.), Signaling Pathways and Molecular Mediators in Metastasis, DOI 10.1007/978-94-007-2558-4_2, © Springer Science+Business Media B.V. 2012

13

14

2.1

M.D. Amatangelo and M.E. Stearns

Defining EMT and Its Relevance in Cancer

Epithelial-mesenchymal transition (EMT) describes the progression of cellular phenotype from an epithelial to mesenchymal state. Within differentiated tissues, epithelial cells are rigid and perform surface-barrier and secretory functions, whereas mesenchymal cells are highly migratory and perform scaffolding, anchoring and maintenance functions. Normal epithelial cells are constrained in a two dimensional cobblestone sheet and are connected by specialized structures including tight junctions, adherence junctions, desmosomes and gap junctions. Within this sheet, epithelial cells exhibit an organized polarity that is maintained by these junctions in which the cells have a free apical (luminal) surface and an adhesive basal-lateral surface which anchors to the basement membrane (see Fig. 2.1). The basement membrane serves as a key extracellular partition made of extracellular matrix proteins that separates and impedes the movement of epithelial cells into the surrounding stroma. In contrast, mesenchymal cells exhibit a spindled shaped, bi-polar

Fig. 2.1 Reactivation of the developmental Epithelial-Mesenchymal Transition (EMT) program promotes progression towards metastatic disease in cancer. During EMT cells lose the rigid ‘cell-cell’ junctions characteristic of epithelial tissues, including E-cadherin based adherence junctions, which allows for cells to disseminate from the primary tumor mass. Cells also change the composition of their cytoskeleton, losing epithelial cytokeratins and gaining vimentin and FSP-1 expression, changing their morphology and promoting cell motility. Furthermore, cells begin to express mesenchymal proteases, such as MMP-2 and MMP-9, and lay down extracellular matrix proteins, like fibronectin (FN), which help degrade the basement membrane (BM) surrounding epithelial tissues and creates tracks in the surrounding stroma to promote local invasion. Such changes allow cells access to the lymphatic and circulatory systems to advance the formation of distant metastases. EMT can be induced by genetic changes (Ras hyperactivity) and factors (such as TGF-beta) present in the microenvironment from tumor cells, stromal fibroblasts and tumor associated inflammatory cells and is regulated by a variety of oncogenic and developmental transcription factors and miRNAs

2 Reactivation of Epithelial-Mesenchymal Transition in Invasive and Metastatic Cancer

15

morphology and lack the apical-basal polarity of epithelial cells. They also migrate individually and do not create rigid contacts with neighboring cells. This results in mesenchymal cells forming loosely organized, fibrotic tissues. The prototypical mesenchymal cell is the fibroblast, which is a highly motile cell responsible for maintaining connective tissue and the stroma surrounding epithelial tissues. Because each of these cell types represents a distinct lineage, each with a unique gene expression profile, this transition represents a considerable change in cellular physiology and biochemistry. EMT events for the most part have been described as part of germ layer reorganization and tissue remodeling during embryonic development. Specifically, these EMT events are critical for mesoderm formation during gastrulation, neural crest maturation, organ morphogenesis as well as wound healing and tissue repair in the adult [1–4]. Cellular and embryonic studies of EMT have resulted in three observational changes in phenotype which have since defined this phenomenon [5, 6]. First, EMT is associated with morphological changes in which epithelial cells no longer exhibit the cobblestone network and apical-basal polarity of normal epithelia. Instead, cells become more dispersed and fibroblast-like in their morphology. Second, EMT is associated with changes in protein expression wherein cells down regulate the expression of cell-cell junction proteins like E-cadherin and epithelial cytokeratin filaments and begin to express mesenchymal associated proteins such as fibronectin, vimentin and N-cadherin. Last, cells change their physiology from a rigid stationary cell to a more motile phenotype and begin to express matrix proteins and proteases that aid in their migration and invasion through tissues. Most solid human tumors (>90%) are carcinomas which arise from epithelial glands. Surgery to remove the primary tumors is an effective treatment for many cancers and patients treated prior to metastases are often cured. Unfortunately, treatment of advanced cancer is often complicated by metastatic disease where cancer cells have migrated to distant sites. Thus, the majority of cancer deaths are caused by the ability of cancer cells to become detached from the neoplastic epithelia and form metastases where surgery becomes impossible. It has become increasingly apparent that in order for cancer cells to accomplish this, an EMT-like program must be activated to prime nodular epithelial cells for the dissemination, movement and invasion required for metastatic spread [7]. The concept that EMT events are involved in the formation of metastatic cancer are primarily based on mechanistic studies done in vitro and in mouse models and observations that loss of epithelial characteristics and acquisition of mesenchymal markers within tumors is often associated with advanced disease. It must be noted, however, that EMT changes do not appear to occur in consistent patterns in cancers nor is the commitment to a mesenchymal phenotype always permanent and the reverse process, mesenchymalepithelial transition (MET), is also observed [8]. Although many similarities in the major concepts of developmental and pathological EMTs exists, Kalluri and Weinberg have classified EMT events into three separate categories to help clarify key differences in the functional consequences and regulation of these events [9]. Accordingly, EMT events associated with early

16

M.D. Amatangelo and M.E. Stearns

embryo and organ development are considered type-1 EMT. This type of EMT is tightly regulated, does not induce fibrosis or systemic, uncontrolled invasion and results from cells which have not fully matured. Type-2 EMT events are associated with wound healing, tissue regeneration and fibrosis. This type of EMT is less controlled and occurs in adult tissues in response to inflammation. Type-3 EMT occurs in carcinoma cells in which changes in oncogenes and tumor suppressor genes in conjunction with tumor associated inflammation utilize the EMT machinery to induce a migratory phenotype and invasion. While there are functional differences between tumor progression and normal embryonic development it also has become quite clear that many similarities also exist [10]. Thus, it is limiting to assert that type-1 and type-3 EMT events are distinctly different and it should be considered that a type-3 EMT event is a reactivation of developmental pathways observed in type-1 EMT resulting from the cellular and environmental changes in cancer. This reactivation in the context of cancer consequently enhances epithelial cell plasticity and promotes aberrant invasive and migratory activities of the cancer cell. Unfortunately, and unlike type-1 and type-2 EMT events, it has been difficult for pathologists to conclusively document type-3 EMT events associated with tumor progression and metastasis in humans [11]. EMT has, however, been directly observed at the leading edge of a spontaneously driven breast cancer model in vivo in mice utilizing stromal and epithelial cell specific cre-transgene markers [12]. Although this study definitively showed cells of epithelial origin becoming mesenchymal during tumor progression and being associated with increased invasion and metastasis, not all tumors that gave rise to metastases exhibited robust EMT. Therefore, it could be concluded that EMT is not the only mechanism by which tumors become invasive and metastatic. Invasive carcinomas may invade surrounding tissues as multicellular epithelial sheets which maintain cell-cell junctions and polarity in a process known as collective migration [13]. However, it has recently been demonstrated that cells restricted to collective invasion were only capable of lymphatic invasion and metastasis to adjacent organs, but not entry into blood vessels or dissemination to distant organ sites [14]. Furthermore, while in many tumors the presence or absence of local lymph node metastasis is a strong predictor of distant metastatic disease, in others, as many as 30% of patients free of lymph-node metastasis still develop disease at distant sites [15]. In agreement with these observations, cancers which display clear forms of collective migration, such as squamous cell carcinomas, rarely form distant metastases [16]. Collectively migrating cells also appear to always follow the tracks of a leading stromal cell but may also follow tracks or signals of migrating epithelial cells that have undergone EMT, suggesting that a cooperative interaction by cells with each other and the surrounding stroma is required for collective invasion [17]. The overall picture emerging from such observations is that collective migration may not be a sufficient for the formation of distant metastases. Regardless, it is clear that changes in the local extracellular matrix and adaptation of a motile phenotype are required for metastasis and such changes can occur as a result of EMT.

2 Reactivation of Epithelial-Mesenchymal Transition in Invasive and Metastatic Cancer

17

We suggest that type-3 EMT events play a significant role in the formation of distant metastasis. These events may represent transient processes of EMT-MET conversions whereby invasive cells which undergo EMT regularly revert back to a normal morphology following specific interactions with the surrounding stroma or to a new environment. Cells which exhibit the plasticity to undergo EMT and MET conversions are likely to be highly adaptable, metastatic and capable of forming distant metastases. While proof that such events occur during metastasis is not conclusive, given the discrete nature in which EMT might occur, standard pathological examination is not conducive to observing EMT or associating EMT with advanced and metastatic cancer. This is primarily because type-3 EMTs occur to different extents and exhibit different phenotypes, with some cells retaining more epithelial traits then others. Interestingly, many cells undergoing EMT adopt a cancer stem cell-like phenotype and are able to produce metastases in mice at very low titer [18, 19]. These results suggest that cells undergoing EMT might not have to en mass in order to promote the formation of metastasis. Accordingly, such events might go unobserved during pathological observation of tumors and the frequency of cells displaying EMT markers might not correlate with metastatic spread. So, while it is apparent in animal and in vitro models that different type-3 EMT events are related to metastatic potential, it is unclear whether the extent which a cell loses epithelial characteristics or gains mesenchymal traits or whether the quantity of cells undergoing EMT has any correlation to metastasis in humans. Thus, evidence in support of EMT will require real-time Imaging since the dynamic nature of such EMTs are difficult to capture in fixed tissues where only a small number of tumor cells might exhibit EMT markers at any one time. Furthermore, standard histological examination of human tumors is observational and not mechanistic in nature and therefore limited in its ability to understand the contribution EMT might have to metastatic progression. Until experimental techniques are available which resolve these issues in humans, the importance of EMT to human cancer metastasis will remain controversial.

2.2

EMT Biomarkers and Their Functional Significance

In order to generate cells with specific functions cells during development must exhibit a level of plasticity that allows them to give rise to or morph into other phenotypes. As this process proceeds, cells must change the repertoire of proteins they express in order to function in their new role. Identification of such changes in protein expression can be used as biomarkers to identify cells associated with a specific purpose. A variety of biomarkers specific to cells of epithelial and mesenchymal origin have been identified to demonstrate the three subtypes of EMT. Indeed, the changes in these biomarkers, including changes in ‘cell-cell’ adhesion, cytoskeleton dynamics and matrix remodeling, have functional consequences and clinical significance as they all play important roles in metastatic progression.

18

2.2.1

M.D. Amatangelo and M.E. Stearns

Disruption of Epithelial Cell-Cell Junctions

E-cadherin is an adherence junction protein that is expressed at ‘cell-cell’ junctions in nearly all epithelial cells but is absent in mesenchymal cells [20]. Furthermore, E-cadherin expression is critical for proper organization and differentiation of epithelial cells during development, as well as for maintaining the apical-basal polarity that is a hallmark of epithelial tissues [21, 22]. Decreases in E-cadherin expression are nearly always observed during EMT associated with development, tissue fibrosis, wound healing and cancer progression and thus its loss is the major biomarker for an EMT event [23]. The most convincing evidence for EMT being associated with invasive and metastatic disease is that loss of E-cadherin-based adherence junctions is consistently associated with progression to invasive carcinoma and poor prognosis in most human epithelial cancers, including carcinomas of the breast, colon, prostate, stomach, liver, esophagus, skin, bladder, kidney and lung [24]. The loss of E-cadherin at ‘cell-cell’ junctions promotes metastasis by enabling cells to detach in response to the shear forces found in lymphatic vessels, venules and arterioles, facilitating their dispersion from the tumor mass [25]. Impairment of E-cadherin-mediated cell adhesion during tumor progression has been shown to occur by deletion, mutation, chromatin rearrangement and hypermethylation. In addition, loss of E-cadherin promoter activity has been found to occur in many metastatic malignancies [24, 26]. In fact, knockdown of E-cadherin alone can induce wide-ranging transcriptional and functional changes which manifest in EMT and contributes to metastatic dissemination [27]. Conversely, several groups have demonstrated that forced re-expression of E-cadherin in malignant cells results in a reversion of the EMT phenotype and inhibits invasion and metastasis [28–31]. Additionally, in an in vivo model for spontaneous pancreatic cancer, maintenance of E-cadherin expression during beta-cell tumorigenesis inhibited invasion and arrested tumor development at the adenoma stage [30]. The implication being that loss of the E-cadherin adherence junction complex is one of the rate limiting step in EMT and progression to invasive cancer in vivo and thus serves as the gatekeeper of the epithelial phenotype [32]. Critical for maintaining E-cadherin based adherence junctions is its interaction with p120 CAS and alpha-, beta-, and gamma-catenin, which link E-cadherin to the actin cytoskeleton. Disruption of the intracellular E-cadherin–catenin complex alone is sufficient for a loss of ‘cell-cell’ adhesions and the tissue rigidity characteristics of epithelial tissues which occurs in tandem with tumor invasion [33]. In particular, beta-catenin is released from E-cadherin complexes into the cytoplasm when these ‘cell-cell’ junctions are disrupted and can act as a signaling molecule. The signaling activity of beta-catenin regulates cellular plasticity and type-1 EMT during heart cushion development [34]. Accordingly, beta-catenin localization has been used as a biomarker of EMT in both fibrosis and cancer [35, 36]. In normal epithelium, cytoplasmic beta-catenin is degraded by the ubiquitin-proteasome pathway through a multiprotein destruction complex containing APC and GSK-3beta. Upon activation of Wnt signaling pathways within the cell (which regulate

2 Reactivation of Epithelial-Mesenchymal Transition in Invasive and Metastatic Cancer

19

proliferation, morphology, motility, and fate during embryonic development) the activity of GSK-3beta is inhibited. This results in the accumulation of beta-catenin in the cytoplasm which can translocate into the nucleus where it can regulate gene expression. Thus, cytoplasmic or nuclear localization of beta-catenin is indicative of and a biomarker for EMT and an invasive phenotype [36]. For example, in advanced colorectal cancers the central region of the tumor exhibits membrane localized betacatenin associated with E-cadherin at cell junctions. In contrast, at the invasive front membrane staining is lost and beta-catenin is localized in the nucleus [36, 37]. Active Wnt/beta-catenin signaling and nuclear beta-catenin accumulation also correlates with EMT, invasion and a poor prognosis in breast cancers [38, 39]. During EMT loss of the E-cadherin cell junction complex is often accompanied by a concomitant up-regulation of mesenchymal cadherins, such as N-cadherin and cadherin-11 [40]. This process is described as “cadherin switching” and readily occurs during type-1 EMT when cells separate from the epiblast layer to ingress the primitive streak and when epithelial cardiomyocytes migrate toward the endocardium during heart morphogenesis [41, 42]. N-Cadherin is typically expressed by mesenchymal cells, fibroblasts and neuronal cells, however, aberrant expression of N-cadherin is also observed in invasive breast, prostate and melanoma cancer cells [43–45]. During type-3 EMT events, N-cadherin expression promotes loss of epithelial cell polarity and increased cell motility, invasion and metastasis, having the opposite effect as E-cadherin expression [46, 47]. Like N-cadherin, cadherin-11 is not expressed by normal epithelial cells but is induced during EMT events associated with development [48]. In addition, cadherin-11, is also expressed in aggressive melanoma, breast and prostate cancer cell lines and appears to coincide with greater cellular invasiveness and a poor clinical prognosis in patients [43, 49, 50]. Expression of these mesenchymal cadherins promotes tumor cell dissemination and invasion independent of E-cadherin down-regulation, highlighting their potent tumor promoting activities [51]. It appears that mesenchymal cadherins promote local invasion by allowing dynamic interactions with the endothelial and stromal components surrounding epithelial tumors [44]. In sum, loss of E-cadherin, nuclear beta-catenin localization and gain of mesenchymal cadherin expression are all currently recognized as key biomarkers for EMT and metastatic potential.

2.2.2

Changes in Cytoskeleton Dynamics

Studies of EMT have also revealed that the expression of vimentin and Fibroblast Specific Protein-1 (FSP1), also known as S100A4, are commonly associated with EMT during embryogenesis and highly invasive tumor cells. Vimentin is a mesenchymal intermediate filament which is known to play a role in maintaining cell integrity in response to mechanical stress [52]. Vimentin also plays a key functional role in the migration and contractibility of cells, as vimentin null mice have fibroblasts with a decreased ability to migrate and exhibit impaired wound healing [53]. During embryogenesis vimentin is expressed in mesodermal cells which exhibit a

20

M.D. Amatangelo and M.E. Stearns

highly motile and invasive phenotype [54, 55]. Specifically, vimentin is turned on at the onset of when cells first detach and migrate from the epithelium to form the mesoderm, and continues to be expressed in all mesenchymal cells making it an excellent biomarker for the mesenchymal phenotype [56]. Carcinoma progression to an invasive phenotype is also often accompanied by increased expression of vimentin in a wide range of cancers [57–61]. Conversely, knocking down the expression of vimentin in highly malignant colon, prostate and breast cancer cells inhibits their ability to migrate and become invasive [62, 63]. Vimentin still remains a controversial type-3 EMT marker, however, as pathologists do not always observe significant increases in vimentin expression in cancer and it is sometimes associated with benign tissue [64]. However, the limitations in histological examination of human tumor sections appear to be part of the problem. Advances in analysis of circulating tumor cells have recently associated vimentin expression with metastatic disease, thus vimentin positive circulating tumor cells might be indicative of tumor associated EMT [65, 66]. In addition to up-regulation of vimentin, rigid epithelial cytokeratins are also down-regulated during EMT and actin filaments are organized into stress fibers [67]. The consequence of this dramatic remodeling of the cytoskeleton is facilitating pseudopod formation at the leading edge of the cell to promote invasion and migration. Fibroblast Specific Protien-1 (FSP-1) appears to be another biomarker for EMT that is important for the transformation of epithelial cells to a metastatic phenotype. FSP-1 is a member of the S100 family of cytoplasmic proteins and is homologous to S100A4. Members of this family are calcium binding, low molecular weight proteins that function as both homo and heterodimers and are implicated in cytoskeletal-membrane interactions, calcium signal transduction, and cellular growth and differentiation [68]. FSP-1 expression was first observed in cells of mesenchymal origin during mesoderm formation and during inflammation induced fibrogenesis [69]. FSP-1 is absent in normal epithelial cells but is prevalent in mesenchymal cells including fibroblasts, monocytes, macrophages and lymphocytes, all of which exhibit a migratory phenotype [69–71]. Transfection studies have shown that FSP-1 is involved in the stimulation of cellular motility and it has been shown to colocalize with myosin IIA and actin filaments at the leading edge of migrating cells [72]. FSP-1 expression is commonly observed in cultured epithelial cells undergoing growth factor induced EMT as well as in EMT during renal fibrosis in transgenic mice [73, 74]. The FSP1 gene is a direct target of nuclear beta-catenin and is associated with promoting the progression of invasive and metastatic cancer [75]. Overexpression of FSP-1 in non-metastatic breast cancer and bladder cancer cells has been shown to promote local tumor invasion and metastasis to lungs and lymph nodes [76, 77]. Conversely, knock down of FSP-1 reduces cell motility, invasiveness and metastatic potential in vivo [78]. In addition, FSP-1 is expressed by invasive tumor cells undergoing EMT in PyV-mT induced mammary tumors in mice and mice over-expressing FSP1 crossed into tumorigenic backgrounds have offspring which exhibit increased frequency of metastasis even tough primary tumor incidence and size do not change [7, 79]. Finally, human patient studies have further revealed that in a panel of 349 breast cancer patients, FPS-1 expression was the

2 Reactivation of Epithelial-Mesenchymal Transition in Invasive and Metastatic Cancer

21

most significant predictor of patient survival [80]. Likewise, FSP-1 expression by tumor cells is a significant negative prognostic indicator in several other cancers [81]. In sum, cells undergoing EMT appear to undergo a reorganization of their cytoskeleton to support the mesenchymal phenotype. The studies highlighted above clearly demonstrate the functional significance of FPS-1 and vimentin expression and their role as an important positive marker for EMT and metastatic potential.

2.2.3

Matrix Remodeling

The extracellular matrix (ECM) is a ‘mesh- like’ scaffolding of proteins localized on the cell surface which provides support and structure for cells and tissues [82]. Cells adhere to the extracellular matrix through integrin receptors which bind to specific matrix proteins, link the ECM to the cytoskeleton and induce intracellular signaling cascades. Epithelial tissues are normally enclosed by an ECM rich in laminin and type-IV collagen proteins called the basement membrane, which is able to impede invasion and the advancement of benign growths [83]. Loss of the basement membrane is one of the key steps in the transition to invasive cancer and the most reliable indicator of poor prognosis in most epithelial carcinomas [84]. Therefore, it appears that during EMT cells significantly alter the way in which they interact with the ECM and there is a sustained remodeling of the adjacent ECM to allow for increased cell migration and invasion. ECM remodeling during EMT appears to involve increased secretion and organization of fibronectin. Fibronectin is a large molecular-weight matrix glycoprotein that is secreted and organized by fibroblasts. Normally, fibronectin is not secreted by epithelial cells, however, it is the most prominent ECM protein produced by cells undergoing type-1 EMT during gastrulation [85]. As a result of EMT, these mesenchymal cells secrete and organize fibronectin to create and facilitate a migratory path for the neural crest [86]. Abundant increases in fibronectin expression and secretion are also readily observed and serve as a type-3 EMT biomarker in vitro [87]. Functional studies have further revealed that fibronectin expression by melanoma cells enhances their invasive ability and synthetic peptides which inhibit fibronectin-integrin interactions inhibit melanoma cell invasion [88, 89]. Likewise, fibronectin promotes the invasion of prostate, breast and gastric cancer cells [90–92]. Furthermore, fibronectin is consistently up-regulated in a variety of solid tumors and is the major constituent of the fibrillar extracellular matrix characteristic of tissue fibrosis and the desmoplastic tumor associated stroma [93]. Tumor modeling studies further show that dense, fibrotic, fibronectin rich matrices create a rigid tumor microenvironment that enhances metastatic tumor progression and is a predictor of poor clinical outcome with breast cancer [94–96]. Fibronectin specifically binds to the fibronetcin receptor on cells, the alpha5beta1 integrin. Thus, many cellular responses to, as well as proper secretion and organization of fibronectin are dependent on expression of alpha5beta1 integrin. Interestingly,

22

M.D. Amatangelo and M.E. Stearns

alpha5beta1 integrin is also associated with type-1 EMT events and in conjunction with fibronectin is critical for generating productive cellular protrusions and inducing motility [97]. In regards to tumor progression, alpha5beta1 expression positively correlates with and is required for melanoma metastasis [98, 99]. Thus, co-expression of fibronectin and alpha5beta1 integrin may be needed to promote invasion and both proteins may serve as biomarkers for cells undergoing EMT. Matrix degradation at the leading edge of cells is also apparent during EMT and can be accomplished by the expression of a family of proteases called matrix metalloproteinases (MMPs). MMPs play an important role in matrix remolding in which they activate growth factors and other MMPs, disrupt ‘cell-cell’ and ‘cell-ECM’ contacts, and cleave several ECM components facilitating invasion. Approximately 23 different MMPs have been reported and each contains a pro-peptide, a catalytic domain with a Zn2+ binding site and a c-terminal tail that determines substrate specificity. Mechanistically, most MMPs are secreted in a latent pro-form and the pro-peptide must be cleaved off outside the cell to activate the enzyme. For example, MMP-2 is most notably cleaved and activated by MT1-MMP, a membrane associated MMP, and MMP-9 is efficiently activated by both MMP-2 and MMP-3 [100–102]. Unique to MMP-9 is its ability to also form a disulfide linked homodimer which is a strong stimulator of cellular migration and appears to be induced by similar pathways that activate EMT [103, 104]. Increases in MMP expression are well documented during type-1 and type-2 EMT events. During EMT associated with neural crest migration cells express MMP-8 and induction of MMP-2 has been shown to be required for EMT and mesenchyme formation in embryos [105–107]. Furthermore, MT1-MMP and MMP-2 have been shown to be induced and play an important part in regulating EMT during heart organogenesis [108, 109]. Cells undergoing type-2 EMT and migration in wounded bronchial epithelial tissues consistently express a variety of MMPs including MMP-9, MMP-3 and MMP-11 [110–112]. Several MMPs which efficiently degrade basement membrane components, including MMP-2, MMP-9, and MT1MMP, have also been linked to progression of benign tumors to invasive cancers of the ovaries, stomach, colon, breast and prostate [113–117]. In particular, expression of MT1-MMP plays an important role in cancer cell invasion and is specifically implicated in prostate cancer metastasis to the bone [118, 119]. Likewise, expression of MMPs has been associated with cells undergoing EMT. In vitro, it has been shown that breast, prostate and bladder cancer cells undergoing EMT, as characterized by decreases in E-cadherin and increases in vimentin, also frequently up-regulate MMP-2, MMP-9 and MT-MMP1 which contributes to their invasive potential [120–122]. MMP-2, MMP-9 and MT-MMP1 are also up-regulated in human breast cancer cell lines that are more mesenchymal in phenotype, further suggesting that increases in MMP production are a relevant marker for EMT [123]. Interestingly, the matrix remodeling that occurs during EMT, including both ECM protein organization and MMP production, also seems to regulate the EMT phenotype, perhaps acting as a positive feedback loop to maintain the mesenchymal state. Functionally blocking the association between alpha5beta1 integrin and fibronectin with a peptide mimetic can inhibit EMT induction and invasion of cells

2 Reactivation of Epithelial-Mesenchymal Transition in Invasive and Metastatic Cancer

23

[98]. Neutralizing antibodies to the beta1 integrin subunit and pharmacological inhibition of beta3 integrin subunit also inhibits induction of EMT [124, 125]. Furthermore, spontaneous breast tumors that develop in mice over- expressing MMP-3 are more metastatic and show mesenchymal characteristics and exposure of cultured mouse mammary epithelial cells to MMP-3 cleaves E-cadherin, dissociates cells and directly activates EMT [126, 127]. In addition, MT-MMP1 has also been shown to induce EMT in prostate cancer cells and MMP-9 has been shown to activate EMT in ovarian cancer cells [128, 129]. Therefore, it appears that the local extracellular changes which occur as a result of EMT also play important roles in regulating EMT. While it is generally thought that fibroblasts in the reactive stroma are responsible for tumor associated matrix remodeling, it is increasingly clear that such activities are also part of the EMT program. For example, MMP-2 expression by stromal cells is not associated with metastatic progression of prostate cancer, whereas MMP-2 expression by malignant epithelial cells is [130]. Thus, it can be concluded that tumor cell associated MMP production during EMT serves an important function in metastatic potential. Studies comparing the functional role of fibronectin expression by tumor associated fibroblasts versus cells undergoing EMT would also be helpful to understand how EMT related matrix remodeling effects metastatic potential. As the tumor microenvironment clearly influences many cell processes, it is most likely stromal cells and cells undergoing EMT work in concert to effect and regulate changes promoting dissemination, motility and invasion during tumor progression. Although the heterogeneity, genetic diversity and uniqueness of individual cancers make it difficult to draw general conclusions regarding EMT, the fact that many characteristics of EMT are associated with more advanced cancers and metastasis offers assurance that this process plays an important role in metastatic progression.

2.3

Cell Signaling Networks Regulating EMT

During development, type-1 EMT is tightly controlled by receptor kinases that are activated by different extracellular factors to induce signaling pathways in cells. Activation of these pathways directs gene transcription and the changes in protein expression that accompany EMT. Studies in cell lines and transgenic mouse models have confirmed that similar signaling pathways regulate type-2 EMT associated with fibrosis and type-3 EMT associated with carcinogenesis. It appears that factors involved in these pathways are often inappropriately expressed during tumor progression from tumor associated inflammation, tumor associated stroma or by the tumor cells. The spectrum of EMT inducers implicated in all three types of EMT is vast and most notably includes signaling pathways such as TGF-beta, EGF, Ras, HGF and Wnt/beta-catenin. Recently it has become clear that significant cross-talk between these pathways must occur to coordinate the elaborate transformation of

24

M.D. Amatangelo and M.E. Stearns

phenotype characteristic of EMT. Importantly, aberrations in these pathways are associated with tumor progression and metastatic potential in many different cancers and are considered attractive drugable targets.

2.3.1

TGF-beta

Transforming Growth Factor-beta (TGF-beta) is a multifunctional cytokine that acts as the master regulator of EMT in normal, developmental and malignant tissues. The TGF-beta superfamily of genes encodes for over 30 signaling peptides including the TGF-beta family, the activin family, the Growth and Differentiation Factors (GDFs) family and the Bone Morphogenetic Proteins (BMPs) family. As part of the TGFbeta family, mammals express three different TGF-beta isoforms: TGF-beta1, TGFbeta2, and TGF-beta3. Each of these appear to be functionally interchangeable in vitro, and although they exhibit a differential expression pattern in the developing embryo, each is associated with regions of active morphogenesis involving epithelial-mesenchymal interactions [131]. TGF-beta ligands act as homo- and heterodimmers and bind to transmembrane TGF-beta type-II receptors. Binding results in autophosphorylation which triggers transphosphorylation of intercellular serines and threonines and activation of the transmembranre TGF-beta type-I receptor. In the canonical TGF-beta signaling pathway, the activated type-I receptor phosphorylates and activates latent Smad proteins. Smads 1, 5 and 8 are activated by the BMP family, while the activin and TGF-beta families activate Smads 2 and 3. When activated, each of these proteins heterodimerizes with Smad4 in the cytoplasm forming a transcription factor complex that enters the nucleus and regulates the expression of TGFbeta responsive genes. One of the TGF-beta responsive genes is Smad7, which functions as a negative feedback loop for the canonical pathway. Smad7 expression competitively inhibits further Smad interactions with the TGF-beta type-I receptor as well as recruits the Smurf family of ubiquitin ligases to the TGF-beta receptor complex to promote its degradation [132, 133]. TGF-beta has also been shown to activate a variety of other noncanonical pathways, including, the AKT, mitogen activated protein kinase (MAPK), c-Jun N-terminal kinase (JNK) and NF-kappaB pathways [134–137]. However, direct activation of these pathways and the role this plays in TGF-beta induced EMT appears to be cell line and context dependent. Members of the TGF-beta superfamily play a critical role in all subtypes of EMT. The TGF-beta superfamily members, Nodal and Vg1, are required for and stimulate EMT in primitive streak cells during gastrulation [138, 139]. BMPs are essential to the induction of EMT during neural crest formation and expression of TGF-beta2 and 3 controls EMT during heart morphogenesis [140, 141]. Signaling through TGF-beta1, which is secreted by immune and epithelial cells, has also been implicated as the main culprit of EMT induced renal and lung fibrosis after injury [142, 143]. In cancer, TGF-beta and activin family members seem to play a more significant role in inducing type-3 EMT events and invasion whereas pathways of the BMP branch do not [144, 145]. TGF-beta1, -2 and -3 have each been shown to

2 Reactivation of Epithelial-Mesenchymal Transition in Invasive and Metastatic Cancer

25

induce EMT in ovarian cancer cells [146]. TGF-beta1 also appears to alters cells at the invasive edge of cancer from cohesive migration to single cell motility reminiscent of EMT, which results in a higher propensity to create distant metastases [14]. Furthermore, increases in TGF-beta1 mRNA and protein are commonly detected in advanced invasive and metastatic tumors of the breast, prostate, colon, skin and correlate with poor prognosis [147–150]. Mechanistic studies have revealed that TGF-beta stimulation and TGF-beta receptor activity result in down-regulation of E-cadherin and tight junction proteins such as ZO-1 and up-regulation of MMPs, fibronectin, FSP-1 and vimentin [151]. For example, during adherence junction disassembly, TGF-beta1 appears to function independent of Smad signaling but in cooperation with the activated oncogene Raf via clathrin-mediated endocytosis and lysosomal degradation of E-cadherin [152]. Additionally, TGF-beta1 treatment in conjunction with phosphatidylinositol 3 kinase (PI3K) can increase tyrosine phosphorylation of alpha- and beta-catenin resulting in disruption of the E-cadherin/catenin complex from the actin cytoskeleton thus reducing ‘cell-cell’ adhesion and promoting EMT [153]. The TGF-beta type-II receptor also plays a direct role in the dismantling of tight junctions. Recently it has been shown that TGF beta type-I receptors are located in the tight junctions where they interact with occludins and the polarity protein Par6 [154]. TGF-beta type-II receptors in turn phosphorylate both TGF-beta type-I receptors and Par6 upon TGF-beta1 stimulation. This controls the interaction of Par6 with the E3 ubiquitin ligase Smurf1 which leads to degradation of the small GTPase RhoA and tight junction dissolution. Additionally, expression of dominant negative TGFbeta type-II receptors in highly malignant cells has been shown to inhibit EMT, invasion and metastasis in mice, further demonstrating the importance of TGF-beta receptor activity in cancer progression and EMT [155]. TGF-beta signaling is also a strong inducer of matrix remodeling. TGF-beta1 promotes fibronectin induction through activation of JNK in a Smad4 independent manner while simultaneously promoting expression of the fibronectin receptor, alpha5beta1, as cells undergo EMT [98, 156]. TGF-beta has also been shown to induce expression of fibrillar collagens and the basement membrane protein laminin-5 in a Smad4 dependent manner [157, 158]. While laminin-5 has been shown to induce migration and EMT in hepatocellular carcinoma cell lines, its role in tumor progression remains controversial as in other systems its expression appears to be lost during EMT and progression to invasive cancer in vivo [98, 159]. Several immunohistochemical studies have however shown a positive correlation between expression of both protein and RNA of the gamma-2 chain, but not the alpha-3 or beta-3 chain, of laminin-5 and the invasive front of colorectal and gastric cancers (in contrast cancer cells deeper in the tumor were negative) [160–163]. TGF-beta through p38 MAPK signaling has also been shown to induce expression of both MMP-2 and MT-MMP1, both of which are known to cleave laminin-5 gamma-2 chain and expose a cryptic promigratory site that triggers cell motility in cancer cells [164– 166]. Thus, while the exact role of laminin-5 during EMT and cancer cell invasion remains obscure, it is clear that TGF-beta mediated ECM remodeling promotes migration, invasion and an EMT phenotype.

26

M.D. Amatangelo and M.E. Stearns

Tumor modeling studies have established that TGF-beta receptor singling through Smads 2 and 3 helps mediate the EMT response to TGF-beta family members and primes cells for metastasis [167]. Transgenic mouse models have revealed that Smad-specific knockouts or expression of mutant type-I TGF-beta receptors that specifically fail to activate Smad2 or Smad3 all block an EMT response in tumors [151]. Canonical TGF-beta signaling by Smad2 and Smad3 also depends on Smad4; likewise, knockout of Smad4 significantly reduces TGFbeta’s ability to down regulate E-cadherin expression and prevents EMT and cell migration in pancreatic tumors [168, 169]. Furthermore, Smad4 knockdown in a highly malignant breast cancer cell line was sufficient to inhibit EMT and bone metastasis in mice [170]. It should, however, be noted that Smads 2 and 3 seem to mediate distinct biological activities in response to TGF-beta and are not completely functionally redundant [171]. It appears that Smad2 activity inhibits EMT while Smad3 activity stimulates EMT and invasion. For example, Smad2 deletion has been shown to enhance the progression of squamous cell carcinomas by inducing EMT and in hepatocytes EMT induction depends on Smad3 but not Smad2 [172, 173]. Moreover, knockdown of Smad3 prevents TGF-beta induced EMT and invasion in ovarian cancer cells and nuclear Smad3 is as a marker for poor prognosis [146]. Smad3 deletion also prevents type-2 EMT during injury and type-1 EMT during gastrulation [174, 175]. Smad signaling is further complicated by its interactions with other signaling networks. Thus, despite the importance of Smad signaling as an inducer of EMT, it is becoming increasing clear that other signaling pathways must be activated in conjunction with Smad signaling to stimulate a robust EMT program.

2.3.2

EGF and Ras

While TGF-beta is the most widely studied cytokine which stimulates EMT, many cancer cell lines which exhibit proficient TGF-beta signal transduction, do not undergo TGF-beta mediated EMT [176]. It is therefore clear that other growth factors and their respective signaling pathways must play an important role in regulating EMT, such as the ErbB family of ligands and receptors. Notably, and unlike TGF-beta receptors, the ErbB family of receptors are frequently over-expressed in many solid human tumors, including lung, colon, breast, prostate, ovarian and bladder carcinomas [177, 178]. Over-expression of ErbB receptors increases their sensitivity to low concentrations of growth factors and thus amplifies the intracellular signaling networks activated by these receptors. Importantly, over-expression of ErbB receptors has been shown to correlate with poor clinical prognosis and promotes tumor cell motility, invasion and an EMT phenotype [179–181]. The ErbB family of receptors include 4 Receptor Tyrosine Kinases (RTKs) including the epidermal growth factor receptor (EGFR; ErbB1, HER1), ErbB2 (Neu, HER2), ErbB3 (HER3) and ErbB4 (HER4). ErbB ligands that activate these receptors are divided into three groups. The first group includes Epidermal Growth Factor

2 Reactivation of Epithelial-Mesenchymal Transition in Invasive and Metastatic Cancer

27

(EGF) and Transforming Growth Factor-alpha (TGF-alpha) which bind specifically to EGFR. The second group includes Betacellulin, Heparin-Binding Growth Factor (HB-EGF) and Epiregulin (EPR), which bind both EGFR and ErbB4. The final group is composed of the Neuregulins (NRGs) which can bind ErbB3 and ErbB4. Ligand binding and receptor activation results in either homo- or heterodimerization and leads to phosphorylation of specific intracellular tyrosine residues which serve as docking sites for proteins containing Src Homology 2 (SH2) and Phosphotyrosine Binding (PTB) domains. In particular, this leads to initiation of the Ras-Raf-MAPK and PI3K-AKT cascades as well as activation of Phospholipase C-gamma (PLCgamma) and Src. Notably, none of the EGF family of ligands binds ErbB2. However, ErbB2 is the preferred heterodimerization partner of the other ErbB family members, is constitutively present in the active state and ErbB2-containing heterodimers have the strongest signaling output [182, 183]. The signaling pathways activated by the ErbB receptors have been consistently implicated in stimulating EMT events in embryogenesis and cancer. Specifically, ErbB receptor signaling induces loss of ‘cell-cell’ junctions, loss of epithelial cell polarity, actin stress fiber polymerization, and induction of migration and invasion [182]. Thus, EGF ligands and receptors have a significant role in inducing EMT and the metastatic phenotype. EGF family member signaling has key functions across vertebrate species in inducing type-1 EMT events and is essential for heart morphogenesis [184, 185]. Direct evidence of EGF family member involvement in type-2 EMT events is lacking. Studies have, however, demonstrated a positive effect of EGF, TGF-alpha and HB-EGF on wound repair and EGF immunoreactivity is associated with the advancing epithelial margin [186]. In type-3 EMT there is stronger evidence which links EGF and its family of receptors to EMT events during tumor progression. Signaling through EGFR has been shown to induce EMT, invasion, and metastasis in several different human prostate and breast cancer cells [187–189]. One important role of EGF signaling in regards to EMT is the dismantling of the cobblestone network and cell polarity characteristic of normal epithelial tissues. For example, persistent exposure to EGF leads to the disruption of cell adherence junctions by caveolin-induced endocytosis of E-cadherin [187]. Over- expression of ErbB2 also disrupts apical–basal polarity organization by interacting with and inhibiting the function of the Par6/aPKC polarity complex [190]. Furthermore, inhibiting EGFR signaling has been shown to restore E-cadherin expression and prevent the mesenchymal phenotype [189]. EGF signaling through the MAPK pathway also induces expression of MT-MMP1 (which activates MMP2) and MMP-9 through activation of the Ets-1 and Ets-2 transcription factors [191–193]. ErbB2 over-expression also induces the expression of MMP-2 through p38 MAPK and AKT signaling, thereby promoting EMT and invasion [194]. Taken together, this evidence suggests that pathways induced by EGF signaling play a significant role in regulating EMT and tumor invasion. Dissection of the EGF signaling pathways have also provided significant insights into how EGF induces tumor cell motility once epithelial cell adhesions are dismantled. Several elegant studies have revealed that EGF signaling activates Coflin at the plasma membrane by PLC-gamma mediated hydrolysis of phosphatidylinositol (4,5)-bisphosphate (PIP2). The Arp2/3 complex and Cofilin then work together

28

M.D. Amatangelo and M.E. Stearns

to reorganize actin filaments at the leading edge by nucleating F-actin branches to form sheet like protrusions in the cell to facilitate lamellipodia mediated migration [195]. The significance of this is demonstrated by the fact that EGFR expression enhances prostate cancer cell motility and invasion and inhibition of EGFR signaling to block phosphorylation of PLC-gamma prevents invasion and metastasis to lungs in mice without an effect on proliferation [196]. Likewise, activation of PI3K by EGF is also an important regulator of actin polymerization and motility in carcinoma cells [197]. PI3K, composed of regulatory (p85) and catalytic (p110) subunits, binds to phosphorylated ErbB receptor family members and phosphorylates PIP2 to from phosphatidylinositol (3,4,5) trisphosphate (PIP3). The presence of PIP3 at the plasma membrane then activates Rho GTPases (Rho, Rac and Cdc42) which initiate and regulate the formation of lamelipodia, filapodia and invadapodia, the cell protrusions that drive cell motility [198]. Specifically, Rho GTPases promote motility through the activation of WAVE and WASP nucleating factors which work in concert with the Arp2/3 complex to form actin projections. Studies in a variety of cell lines have confirmed that both of these pathways stimulated by EGF are required for full induction of migration and invasion by metastatic cells [199, 200]. Albeit, the role of these pathways (and the different types of protrusions they induce) in the enhanced migration seen during EMT is poorly understood. The most significant downstream effector of EGF signaling promoting EMT is the Ras oncogene. The Ras family of GTP-binding proteins are downstream effectors of many growth factor receptors and mutations leading to hyper-activation of Ras signaling pathways are frequent in cancers [201]. The most well characterized action of Ras activation is recruitment of Raf kinases to the plasma membrane where Raf in turn activates the MAP kinase signaling pathway through MEK (MAPKK). GTP-Ras can also activate PI3K leading to phosphorylation and activation of AKT. Activation of both of these pathways makes Ras a powerful oncogene and promoter of cell proliferation. Ras also cooperates with Smads 2 and 3 to drive EMT and metastatic phenotypes where it promotes nuclear accumulation of Smad proteins [202]. In this regard, pharmacological inhibition of MEK1/2 activation also prevents TGF-beta from stimulating EMT [203]. While non-canonical MAPK activation by TGF-beta appears to be required for TGF-beta induced EMT, it is also apparent that constitutive Ras activation along with TGF-beta stimulation can act cooperatively to induce EMT when TGF-beta alone cannot [204, 205]. Ras-Raf-MAPK signaling and TGF-beta also work synergistically to induce endocytosis and lysosomal degradation of E-cadherin as well as increases in MMP-2, MMP-12 and MMP-13 [152, 206]. Furthermore, co-stimulation of cells with TGF-beta and EGF induces robust EMT and ErbB2 over-expression induces migration and invasive behaviors in cooperation with TGF-beta dependent on sustained elevated levels of activate MAPK [207, 208]. Interestingly, it appears that Ras-MEK1/2 activation of Erk2, but not Erk1, is required for EMT [209]. Considering that MEK1 and MEK2 have differential effects on Erk2 localization and function, it is plausible that EGF and Ras’s ability to differentially effect Erk2 activation might serve as a mechanism by which Ras and TGF-beta signaling work synergistically to induce EMT [210, 211]. In agreement with this hypothesis, other studies have shown that while the ability of

2 Reactivation of Epithelial-Mesenchymal Transition in Invasive and Metastatic Cancer

29

Ras to activate PI3K can induce cell scattering in TGF-beta treated cells, it is the ability of Ras to activate the Raf-MAPK pathway which promotes robust and sustainable EMT in response to TGF-beta treatment [212]. The studies highlighted in the above discussion suggest that TGF-beta must cooperate with other signaling pathways either from hyper-activation of oncogenes, such as Ras or ErB2, or exposure to other growth factors, such as EGF, to elicit EMT. The sources of multiple EMT inducing ligands might result from chronic inflammation or the reactive stroma associated with tumors, where macrophages and other immune and stromal cells secrete TGF-beta and EGF into the tumor microenvironment. In animal models of breast cancer, it is apparent that breast cancer cells often co-migrate with macrophages which secrete EGF and EGF is required for stimulation of carcinoma cell invasion and induces an EMT phenotype [213]. Consequently, this might be a mechanism by which early stage carcinoma cells that lack gross oncogenic abnormalities might be induced to undergo EMT and participate in the early dissemination model of tumor metastasis [214].

2.3.3

HGF and FGF

Besides TGF-beta and EGF, other growth factor receptor pathways also play a role in the induction of EMT and invasion; in particular fibroblast growth factor (FGF) and hepatocyte growth factor/scatter factor (HGF) seem to be an important. HGF is a multifunctional factor produced by many cell types, including normal fibroblasts, epithelial and endothelial cells. HGF binds to the RTK c-Met, which causes dimerization and autophosphorylation of tyrosines which create multisubstrate docking sites for the adaptor proteins Grb2 and Shc, and the p85 subunit of PI3K, resulting in activation of the Ras–MAPK, PI3K, PLC-gamma and Src kinase pathways. In addition, Gab1 is recruited to the receptor as well as the ubiquitin ligases c-Cbl and Cbl-b, which negatively regulate the c-Met signal. Gab1 is the major substrate for c-Met, and HGF binding promotes a prolonged Gab1 phosphorylation that is not produced upon activation of other RTKs, such as EGFR [215]. HGF was first identified as a fibroblast secreted factor that induced epithelial cell scattering and actin remodeling [216]. Since then, HGF has been found to induce EMT as well as increase cell motility and invasiveness [217]. c-Met over-expression is evident in many cancers including hepatocarcinomas and carcinomas of the colon, prostate, ovary and breast and correlates more invasive tumors and poor prognosis in each [218–222]. Likewise, HGF over-expression in the tumor microenvironment is also observed and is shown to directly induce localized invasion [223]. Mechanistically, HGF induces EMT and invasion by activating the PI3K pathway rather than Gab1, as specific inhibition of c-Met’s ability to activate PI3K prevents dismantling of ‘cell-cell’ contacts, cytoskeletal rearrangements and c-Met driven metastasis [224, 225]. This seems to rely on AKT-mTOR phosphorylation of p70 S6 kinase (p70S6K). Neither MAPK nor PLC-gamma activity are required for HGF to activate p70S6K. Additionally, HGF induction of p70S6K activity has been

30

M.D. Amatangelo and M.E. Stearns

shown to inhibit E-cadherin expression and promote EMT and invasion in ovarian cancer [226, 227]. Thus, in contrast from EGF studies indicating Ras-MAPK pathway as a key signaling network required for EMT, in other contexts, such as in HGF signaling, AKT activation appears to be sufficient to induce EMT as characterized by down-regulation of E-cadherin and up-regulation of vimentin [228]. The FGF family of ligands includes at least 23 member which signal through four high-affinity RTKs (and their alternatively spliced isoforms), designated FGFR-1 to −4, which can promote increased cellular motility and invasiveness in specific contexts [229]. Much like EGF and HGF, ligand binding to FGFRs results in receptor dimerization and activation of the intrinsic tyrosine kinase activity which activates several downstream signaling pathways including Ras-MAPK, PI3K, PLC-gamma and Src kinase. However, functional consequence of FGF activation appear to be very cell type specific and studies from one cell line or tissue may not be generally applicable to all tumors. In most cellular contexts FGFs appear to induce proliferation and migration, however it can also induce differentiation and/ or cell cycle arrest [230]. Thus, the mechanisms by which deregulated FGF signaling may promote tumorigenesis are still poorly understood despite the fact that there is solid evidence that FGF signaling supports EMT. FGF is a critical ligand in development, promoting mesodermal formation, vimentin expression and mesenchymal cell migration through the primitive streak during gastrulation [231]. FGFs also play a role in EMT during wound healing where they stimulate tissue remodeling and migration and epithelial cells as well as inflammation induced fibrosis [232]. In adult tissues FGFs are secreted by fibroblasts often after TGF-beta stimulation, and are localized along epithelial-stromal boundaries [233]. In studies regarding type-3 EMT, FGF-1 stimulation of bladder cancer cells has been shown to enhance cell migration and induce EMT [234]. In contrast to TGF-beta, EGF and HGF, FGF stimulation does not cause a decrease in expression of adherence junction proteins. Rather, E-cadherin based adherence junctions become internalized, cells fail to make ‘cell-cell’ junctions and desmosomal components disappear [235]. In in vivo prostate cancer modeling studies, active FGFR-1 has been shown to promote EMT and metastasis to liver and lymph nodes further substantiating that FGFs can play a role in tumor associated EMT and metastasis [236]. In sum, FGF clearly plays an important role in type-1 and type-2 EMTs, however, less has been explored about its role in type-3 EMTs and little is known about the intercellular signaling cascades that govern FGF stimulated EMT during tumorigeneis. Thus, the specific mechanisms by which FGF signaling may promote metastasis and invasion are poorly understood.

2.3.4

Wnt, Notch and Hedgehog

The Wnt/beta-catenin, Notch and Hedgehog pathways all play a critical role during embryogenesis and regulate type-1 EMTs during development [237]. Wnt growth factors constitute a family of 19 closely related glycoproteins that are secreted and

2 Reactivation of Epithelial-Mesenchymal Transition in Invasive and Metastatic Cancer

31

activate members of Frizzled family and LDL receptor-related protein 5/6 (LRP-5/6) receptors. Signaling through the Frizzled receptor causes release of the protein disheveled from the receptor complex which inhibits a multicomponent destruction complex composed of Axin, APC, Glycogen Synthase Kinase 3-beta(GSK3-beta) and Casein Kinase I (CKI). This complex regulates the stability of cytoplasmic beta-catenin by phosphorylation which targets it for degradation by the proteasome. As a consequence of activated Wnt signaling, beta-catenin is able to accumulate in the nucleus where it interacts with the TCF/LEF family of transcription factors which contain DNA binding sites and regulate gene expression. Wnt pathway activation is critical for induction of EMT in gastrulation and cardiac valve formation as stabilization of beta-catenin causes premature mesoderm formation and beta-catenin knockout mice fail to develop heart cushions during development [34, 238]. Activation of the Wnt pathway and beta-catenin nuclear accumulation also induces vimentin expression and EMT in several breast cancer cell lines as well as progressive loss of E-cadherin in colon cancer [37, 239, 240]. Furthermore, beta-catenin/TCF/LEF transcriptional activity induces invasion and metastasis through the expression of MT1-MMP, MMP-7 and MMP-26 [241–243]. Well characterized mutations in the Wnt/beta-catenin pathway have also been identified in the tumor suppressor gene APC which leads to inherited familial adenomatous polyposis, a spontaneous form of colon cancer that is associated with increased incidence of invasive colorectal cancers [244]. Aberrant Wnt signaling which causes nuclear accumulation of beta-catenin also contributes to EMT, invasion and increased metastatic potential in melanoma [245]. The Wnt/beta-catenin/LEF signaling pathway also exhibits extensive crosstalk with other receptor tyrosine kinases associated with EMT. Activation of c-Met by HGF results in tyrosine phosphorylation and cellular accumulation of beta-catenin, which amplifies TCF/LEF-mediated gene transcription [246]. beta-Catenin and TGF-beta/Smad signaling also cooperate via Smad and TCF/LEF interactions which can act as transcription factor complexes to synergistically activate target genes and EMT [247]. While most known receptor-ligand interactions governing intracellular signaling pathways involve diffusible proteins, Notch signaling involves protein ligands which remain bound to the cell surface thus inducing signaling via ‘cell-cell’ contacts. Activation of the Notch family of receptors, Notch-1, -2, -3 and -4, is mediated by the interaction with one of five different identified Notch ligands, Jagged-1/-2, and Delta-like-1/3/4 on a neighboring cell. Activation of Notch results in the cleavage of the Notch intracellular domain (NICD) inside the cell by gamma-secretase, which then translocates to the nucleus where it partners with CSL (C protein binding factor/Suppressor of hairless/Lag-1) to form an active transcriptional complex. Like Wnt-signaling, Notch is also critical for type-1 EMT associated with heart morphogenesis as well as neural crest migration during development [248, 249]. Like TGF-beta, Notch signaling also seems to act in both a tumor suppressive and tumor promoting role depending on cell context and cooperation with oncogenes. Activation of Notch has, however, been associated in vivo with promoting tumorigenesis and metastasis in the prostate cancer, promoting invasion in pancreatic cancers and over-expression of Notch1 and Jagged1 correlates with poor prognosis in

32

M.D. Amatangelo and M.E. Stearns

breast cancer patients [250–252]. Hypoxia within rapidly growing tumors has also been known to induce EMT and is linked to increased metastatic potential [253]. Studies have shown that activation of the PI3K-AKT pathway is critical for this process [254]. In an array of cell lines from different cancers, Notch signaling also seems to be required to convert this hypoxic stimulus and Hypoxia induced factor 1-alpha (HIF1-alpha) signaling into EMT, increased motility, and invasiveness [255]. The NICD/CSL transcription complex resulting from notch activation also directs the expression of multiple genes including the transcription factors, hairy enhancer of split (HES) and hairy enhancer of split-related YPRW motif (HEY) [256]. Interestingly, expression of HEY1 and Jagged-1 is required in a variety of cells for TGF-beta induced EMT as well, where the proteins themselves are induced by TGF-beta treatment in a Smad3 dependent manner [257]. These studies suggest that while Notch signaling itself maybe not direct EMT, its integration into other oncogenic pathways can facilitate EMT and invasive phenotypes in tumor cells. Another major pathway regulated during development is the Hedgehog pathway which is essential for stem cell maintenance and directing embryonic tissue patterning of the various organ systems [258, 259]. In mammals, three secreted Hedgehog ligands have been identified: Sonic Hedgehog (SHH), Indian Hedgehog (IHH), and Desert Hedgehog (DHH). Hedgehog signaling is mediated by Patched receptors (PTCH1 or PTCH2), which can bind each Hedgehog ligand, and by its co-receptor, Smoothened (SMO). When PTCH is unbound to ligand, signaling by SMO is inhibited. Upon binding of Hedgehog, SMO is released and initiates a cascade of signaling events that activates a microtubule-associated complex containing the proteins Fused (Fu), Suppressor of Fused (SuFu), and Costal-2 (Cos-2). Activation of the Fu-SuFu-Cos-2 complex results in the activation and nuclear translocation of the Gli family of transcription factors. The Gli family consists of Gli-1 and −2 which function as transcriptional activators and Gli-3 which acts primarily as a repressor [260, 261]. In regards to EMT, Hedgehog signaling has been shown to be required for EMT during somite development and heart morphogenesis [262]. Gli proteins also participate in conjunction with FGF and Wnt signaling in EMT during neural and mesodermal development [263]. Hedgehog signaling also promotes type-2 EMT events and is a significant contributor to the evolution of biliary fibrosis during chronic cholestasis [264]. Elevated Hedgehog signaling is also associated with type-3 EMTs which causes increased metastatic potential in prostate cancer and basal cell carcinoma cells [265, 266]. In fact, exogenous expression of Gli-1 in non-metastatic cell lines results in E-cadherin repression and enhancement of migration, invasion and establishment of metastasis in vivo [267]. In addition, cyclopamine, a potent inhibitor of Hedgehog signaling, inhibits Gli-1 induction of EMT and invasion. Hedgehog signaling is also associated with induction of the EMT in conjunction with the Wnt pathway in skin cancer further demonstrating the intricate crosstalk between EMT signaling pathways [265]. Collectively, the role of highly conserved developmental pathways such as Wnt, Notch and Hedgehog signaling in type-3 EMT events implies that metastatic cancer cells reactivate latent developmental programs to enable the required cellular plasticity needed for such drastic changes in phenotype.

2 Reactivation of Epithelial-Mesenchymal Transition in Invasive and Metastatic Cancer

33

Despite the established role of receptor signaling in EMT, the studies highlighted above clearly demonstrate that activation of a single pathway may not be sufficient to elicit robust EMT. While many oncogenic signals can result in abrogate epithelial polarity, tight and adherence junctions; acting alone they are insufficient to induce a mesenchymal, migratory and invasive phenotype [190]. These observations agree with the overwhelming evidence that EMT in various carcinomas involves an intricate interplay of multiple signaling pathways which crosstalk in a precise manner to elicit the drastic change in cellular phenotype seen during EMT. In addition, many of these pathways activate intracellular reactive oxygen species (ROS), which inhibit the activity of phosphatases thus amplifying signals [268]. To further complicate the dynamic signaling networks regulating EMT, various pathways can also regulate the secretion of other ligands and thus the activation of other pathways in an autocrine manner. For example, TGF-beta stimulation has been shown to induce PDGF secretion and PDGF receptor activation in hepatocellular carcinoma, and was required to induce nuclear beta-catenin and a migratory phenotype [269]. As the signaling networks regulating EMT become explored in greater depth the future challenge will be to identify the specific circumstances governing when and why certain pathways must be activated and when and why others are not necessary.

2.4

Transcriptional Regulators of EMT

Over the last 10 years much effort has been given to identifying and understanding several families of transcription factors which are up-regulated during EMT induced signaling. Members of the Snail (SNAI1 and SNAI2/Slug), Zeb (ZEB1 and ZEB2/SIP1), basic helix-loop-helix (Twist and Twist2), and homeodomain transcription factors are now considered to be master transcriptional regulators of EMT in both development and cancer related pathologies. In addition, chromatin associated proteins found to regulate EMT and plasticity during development have also been associated with the EMT phenotype and promote the invasive and metastatic potential of a variety of cancers. Given the obscurity of the vast array of signaling cascades which promote EMT, understanding the mechanisms of action of these transcription factors and their role in cancer metastases will likely be critical to development of therapeutic agents which prevent EMT, invasion and metastases.

2.4.1

Snail and Slug

SNAI1 (Snail) and SNAI2 (Slug) belong to the Snail superfamily and were the first such transcription factors shown to directly regulate EMT and invasive activities of cancer cells. Both Snail and Slug act as transcriptional repressors and bind to the promoter of E-cadherin and block its expression [270–272]. Snail family

34

M.D. Amatangelo and M.E. Stearns

protein expression regulates developmental EMT and their expression is regulated by cross-talk between various signaling molecules, including the TGF-beta, EGF, FGF, HGF, Notch, hedgehog and Wnt pathways [248, 257, 273, 274]. In particular, Ras activation of the MAPK and PI3K signaling pathways cooperate with TGF-beta to induce strong activation of the Snail promoter [275]. Snail also appears to function as the main transcriptional regulator of Hedgehog and betacatenin induced EMT during tumorigenesis [276]. Studies have also shown that both Snail and Slug are induced by HIF-1alpha during hypoxia to repress E-cadherin expression and promote EMT and migration in ovarian cancer cells [277]. Analysis of Snail family promoters has identified several conserved elements including AP1 and AP4 sites, Smad-binding elements, LEF1 binding sites and two E-boxes which regulate their transcriptional activity [278, 279]. Recent studies have also shown that Snail proteins can bind and represses their own promoters in a negative feedback loop [280]. Snail family members are also regulated through post-transcriptional modifications. For example, p21-activated kinase1 (PAK1) appears to promote the nuclear localization and the transcriptional repressor activity of Snail on E-cadherin and occludins by phosphorylation of Ser 246 [281]. Conversely, GSK3-beta binds to and phosphorylates Snail at its nuclear export signal (NES) and destruction box, promoting its cytoplasmic export and proteasomal degradation [282]. The post translational regulation of Slug is less understood. However, Slug, like Snail, appears to be a very unstable protein and is efficiently targeted for proteasomal degradation by the F-box protein, Partner of Paired (Ppa) independent of GSK3beta activity [283]. Structurally, members of the Snail family are zinc-finger transcription factors that share a highly conserved C-terminal region, containing four to six zinc fingers which bind consensus E-box elements in the promoters of genes (CAGGTG) [284]. Their repressor activity is dependent on the SNAG domain at their N-terminus, which is involved in the direct recruitment of a chromatin modifying repressor complex formed by the co-repressors SIN3A, and HDAC1 and HDAC2 [285]. Studies beyond E-cadherin have implicated the Snail family as repressors of a variety of other common epithelial genes including claudins and occludins, which are required for tight junctions and epithelial cell polarity, as well as epithelial cytokeratins [286]. Snail has also recently been shown to interact with Smad3 and Smad4 to form a transcriptional repressor complex which targets the promoters of CAR, occludin, claudin-3 and E-cadherin during TGF-beta driven EMT [287]. Thus, as a transcription factor, Snail appears to play a key role in controlling genes that regulate tight junction stability, epithelial cell polarity and morphology. Developmental biologists have shown that the expression of Snail family proteins triggers EMT during embryonic development; triggering mesoderm formation, gastrulation, neural crest migration and heart morphogenesis [288]. Studies in carcinoma cell lines and transgenic mouse tumor models have also confirmed that Snail and Slug expression induce type-3 EMT associated with carcinogenesis [279]. In pre-malignant breast epithelial cells, Slug in cooperation with activated Ras induces vimentin expression and migration [289]. Furthermore, whereas little

2 Reactivation of Epithelial-Mesenchymal Transition in Invasive and Metastatic Cancer

35

or no Snail expression is detected in well differentiated, non-invasive carcinomas, expression is high in a range of metastatic cell lines and invasive human cancers [290, 291]. In breast carcinomas, Snail and Slug expression are elevated and associated with repression of E-cadherin and increases in lymph node metastasis, tumor recurrence and poor prognosis [292, 293]. Studies in colorectal cancer have also indicated that both Snail and Slug expression are associated with the down regulation of E-cadherin, poor prognosis, and development of distal metastases [294, 295]. Likewise, Snail activates the MMP-2 promoter, up-regulates vimentin and induces invasion of squamous cell carcinomas, and has also been shown to induce MMP-7, MMP-9 and MT-MMP1 in a variety of cell lines to confer invasion and basement membrane degradation [296–299]. Snail over-expression is also observed in oesophageal squamous cell carcinomas and invasive edges of hepatocarcinomas, where it is associated with the down-regulation of E-cadherin and the development of distant metastasis [300, 301]. Similarly, highly metastatic prostate cancer cells require Slug to activate invasion and EMT [302]. Importantly, functional expression of Slug has also conclusively been shown to be essential for metastasis of aggressive melanoma cells in vivo, as its knock-out inhibited lung metastasis without effects on primary tumor growth [303]. Collectively, these studies implicate the Snail family of transcription factors in cancer associated EMT, invasion, poor clinical outcome, and development of metastases in a variety of cancer types.

2.4.2

Zeb1 and Zeb2

Zeb1 (delta-EF1) and Zeb2 (Sip1) are zinc finger transcription factors which have also emerged as direct transcriptional repressors of the E-cadherin gene and EMT, and are associated with increased invasive capacity in cells [304, 305]. Both Zeb1 and Zeb2 were identified as proteins capable of interacting with and modulating the activity of receptor activated Smads (Smads 1, 2 and 3) [306, 307]. Although the role of Zeb1 and Zeb2 in influencing Smad signaling is poorly understood, a conserved region downstream of the N-terminal zinc finger acts as a Smad interacting domain in both proteins, suggesting that Zeb proteins might play critical roles in integrating signals from TGF-beta and other signaling molecules [307]. Zeb proteins, like Snail family proteins, have also been shown to be post-transcriptionally modified by phosphorylation, however, the functional consequences of these events is unknown [308]. It is tempting to believe that such phosphorylation events might alter interactions with co-repressor and activators, regulating the transcriptional activity of Zeb factors and influence how they integrate signals from multiple pathways. Sumoylation also appears to be an important posttranslational modification on both Zeb1 and Zeb2, and is mediated by polycomb protein, Pc2, which has the functional consequence of decreasing Zeb mediated E-cadherin repression [309]. Zeb protein expression has been shown to be regulated by TGF-beta, NF-KappaB and hypoxia induced signaling [310, 311]. TGF-beta induction of Ets1 and

36

M.D. Amatangelo and M.E. Stearns

activation of Smads has been shown to be an upstream transcriptional activator of both Zeb1 and Zeb2 [312]. Treatment with TNF-alpha or induction of constitutive NF-kappB signaling in breast epithelial cells results in a decrease in E-cadherin and increase in vimentin in a Zeb1/2 dependent manner [311]. NF-kappaB signaling also appears to be essential for maintenance of the EMT phenotype and metastasis in Ras transformed breast cancer cells, suggesting that regulation of Zeb proteins might play a key role in EMT induction and metastasis promoted by NF-kappaB [313]. Zeb proteins also contain hypoxia response elements in their promoters and appear to mediate HIF-1alpha-dependent E-cadherin down-regulation in renal clear cell carcinoma cells, indicative of a role for Zeb proteins in inducing hypoxic mediated cell migration [310, 314]. Structurally, Zeb proteins contain two separated clusters of zinc fingers, one N-terminal and one C-terminal, which bind independently to bipartite e-box elements composed of one CACCT and one CACCTG sequence as found in the E-cadherin promoter [315]. Studies have indicated that the transcriptional repression activity of Zeb family proteins may be attributed to its association with the co-repressor CtBP (C-terminal binding protein) which aids in the recruitment of HDAC1, HDAC2 and the histone methyltransferases G9a and EuHMT to the promoters of genes [316]. Indeed, CtBP expression and its association with Zeb factors is associated with the repression of many epithelial genes, including E-cadherin and promotes cancer cell migration in response to hypoxia [317, 318]. Zeb1 has also recently been shown to interact with the SWI/SNF chromatin-remodeling protein Brg1 to down-regulate E-cadherin and induce vimentin independently of CtBP [319]. Interestingly, colocalization of Zeb1 and Brg1 is found in conjunction with nuclear beta-catenin staining at the invasive edge of colorectal tumors. Beyond E-cadherin, Zeb2 also down-regulates several other tight junction, adherence junction, desmosome and gap junction proteins by repression of promoter activity contributing to the dissociation of epithelial cells from their tissues [320]. Furthermore, conditional expression of Zeb2 in cells has been shown to be sufficient to disrupt ‘cell-cell’ adhesions and induces invasion in vitro [304]. Zeb proteins can also act as transcriptional activators, which is dependent on their ability to associate with p300 and displace CtBP [321]. While developmental biologists have observed Zeb family protein expressed in distinct regions during development of the central nervous system, heart, and in other cells of mesenchymal origin, genetic studies have shown their activities might not be functionally redundant or additive [322]. For example, in embryos lacking Zeb2 neural crest cells fail to undergo EMT and migrate, and they exhibit a host of abnormalities which Zeb1 expression cannot rescue [323]. Congruent with a role in type-3 EMT, expression of Zeb proteins are also associated with more aggressive, invasive and metastatic tumors. Zeb2 expression is associated with more advanced ovarian cancers, increases in circulating tumor cells and is an indicator of poor survival [291]. Zeb2 is also linked with advanced pancreatic tumors which are highly aggressive, have an extremely poor prognosis and are associated with a robust EMT and fibrotic response [324]. Head and neck squamous cell carcinoma cells which express high levels of Zeb1 are also more mesenchymal in phenotype and are highly invasive and

2 Reactivation of Epithelial-Mesenchymal Transition in Invasive and Metastatic Cancer

37

migratory [325]. Zeb1 is also associated with advanced colorectal cancers and seems to be specifically expressed in those that lack Snail expression suggesting that some redundancy amongst EMT inducing transcription factors might exist [321]. However, the relationship between Zeb and Snail family transcription factors is complicated. The expression of Snail indirectly induces expression of Zeb1 and Zeb2 in some cell lines to promote EMT [326, 327]. Moreover, Zeb family members have been shown to be necessary, but not sufficient, for TGF-beta induced EMT in other cells [307]. This suggests that there is a coordinated transcriptional response needed to induce and maintain EMT that likely involves activation of more than one transcription factor. The complexities of these interactions and their relationship to EMT and metastatic potential remain to be unraveled.

2.4.3

Twist1 and Twist2

The Twist proteins are members of the basic-helix-loop-helix (bHLH) family of transcription factors and were originally shown to be central to embryonic development and to control cell fate at the early stages of embryogenesis. More recent studies have shown Twist1 and Twist2 to play many roles in the development and progression of tumors, including transcriptional control of genes associated with EMT. bHLH transcription factors are classified into three major classes: Class A factors are ubiquitously expressed and include E12/E47; Class B factors are tissue restricted and include Twist1 and Twist2; and the inhibitory HLH proteins, which lack DNA binding sites, include the Id proteins. We have largely restricted this review to discussion of class B factors as expression of these has been most commonly linked to metastatic cancer progression. Like the Snail and Zeb family of proteins, the mechanisms leading to the aberrant reactivation of Twist factors seem to result from signaling pathways that normally direct the expression of genes during embryonic development. Wnt has been shown to induce Twist1 expression and Twist1 promoters are responsive to betacatenin/TCF complexes as well as c-Jun, and Ets transcription factors [328]. Twist1 is also induced in the mesoderm during embryogenesis by a combination of FGF and SHH signaling, but not by individual application of FGF or SHH alone [329]. The relevance of this has not been studied in cancer associated EMT however. In addition, stress conditions seem to play an important role in inducing Twist activity. Both Twist1 and Twist2, are induced during inflammation by a cytokine signaling pathway that requires the p65 NF-kappaB family of transcription factors [330]. Furthermore, like the Snail and Zeb transcription factors, HIF-1alpha also regulates the expression of Twist during hypoxia by binding directly to the hypoxia-response element (HRE) in the Twist promoter [331]. Significantly, knockdown of either Twist or HIF-1alpha prevents hypoxia induced EMT. Post-translational regulation of bHLH factors is more complex than that of Snail and Zeb factors because the specific expression pattern of members and their affinities for each other regulate their capacity to form functional dimers. Twist

38

M.D. Amatangelo and M.E. Stearns

transcription factors are unique in their ability to form functional homodimers and heterodimers with Class A bHLH proteins. The Id proteins, on the other hand, exhibit greater affinity for class A factors and actively alter dimerization pools by binding and sequestering class A factors [332]. It is therefore the composition of these partner pools that determines cellular response and thus the balance between different Twist homo and heterodimers directs the functional consequences of Twist expression [333, 334]. Both Twist1 and Twist2 genes have been found to be overexpressed and induce EMT and metastasis in a large set of human and murine tumors suggesting that the ability of Twist proteins to form homodimers and out compete Id proteins is sufficient for functionality [335, 336]. However, phosphorylation also influences bHLH activity by regulating their homo and heterodimerization affinity as well as DNA-binding capacity. For example, phosphorylation of the class A bHLH E47 on serine 140 increases its affinity to heterodimerize with other class A bHLH members [337]. Furthermore, Notch activation of Erk1 and Erk2 can phosphorylate class A bHLH factors and induce their proteasomal degradation [338]. Twist1 and Twist2 can also be phosphorylated on threonine 125 and serine 127, which is regulated by Protein Kinase A and Protein Kinase C and increases their DNA binding affinity in a dimerization partner dependent manner [339, 340]. Recently, AKT has also been linked to the phosphorylation of Twist1 at serine 42 which promotes cell survival and resistance to apoptosis [341]. AKT activation also appears to be required for Twist mediated EMT in response to hypoxia [254]. The basic structure of all bHLH family members involves two parallel amphipatic alpha-helices joined by a loop which is required for dimerization. bHLH proteins bind to DNA using a consensus E-box (CANNTG) site as homo- or heterodimers. bHLH protein complexes can also act as both transcriptional repressors and transcriptional activators. Transactivation activity seems to be dependent on the recruitment of histone acetyl transferase (HAT) proteins, such as p300 or the SPT–ADA–GCN5– acetyltransferase (SAGA) complex [342]. Twist proteins can also directly bind histone acetyltransferases, p300 and p300/CBP–associated factor (PCAF) and inhibit the ability of other transcription factors to promote transcription [343]. Both Twist1 and Twist2 also interact with and recruit HDAC complexes to exert transcriptional repression [344, 345]. Taken together, it might be implied that repression and activation functions of Twist proteins depend on dimer choice, however, such evidence in mammalian systems has not been consistent. Studies indicated that like Snail and Zeb family of transcription factors, Twist1 and Twist2 promote all three types of EMT events. In particular, Twist1 is a critical factor regulating TGF-beta’s ability to induce type-1 and type-3 EMT events as well as the TGF-beta induced fibrotic response during type-2 EMT [346–350]. Increased Twist expression is associated with invasive ductal carcinomas of the breast and positively correlates with EMT and the formation of distant solid metastasis in prostate cancer, oesophageal squamous cell carcinomas and hepatocarcinomas [351–354]. Others have shown that Twist expression is associated with circulating tumor cells in breast cancer and is essential for metastasis to the lung [335, 355]. Twist1 over-expression also correlates with nuclear beta-catenin and invasive cells in a mouse model of colorectal cancer [356]. Moreover, Twist

2 Reactivation of Epithelial-Mesenchymal Transition in Invasive and Metastatic Cancer

39

proteins have been shown to directly up-regulate N-cadherin and down-regulate E-cadherin, placing them at the center of the cadherin switch seen during some EMT events [357]. Furthermore, ectopic Twist expression results in the activation of several other mesenchymal genes including vimentin, fibronectin MMP-2 and alpha-smooth muscle actin, correlating with the adoption of a motile and invasive phenotype [335, 355, 358]. Overall, the significance of the overlap in Twist functions with Snail and Zeb factors remains unresolved. While it is clear that each of these factors is important in promoting EMT and metastatic potential in different systems, considerable work is required to understand how each of these transcriptions factors work in a coordinate manner to control gene expression generally associated with EMT, invasion and metastasis.

2.4.4

Homeodomain Proteins

Homeobox proteins are a group of highly conserved transcription factors that bind homeobox DNA sequences found within genes. The homeobox family of transcription factors acts as critical regulators of growth, migration and differentiation during embryogenesis. In addition, they have recently gained attention as promoters of invasion and type-3 EMT. Homeobox proteins generally do not function alone but in tandem with various members of the homeobox family which determines functional consequences of expression thus making it difficult to discern the roles of distinct members. Specifically, however, expression of members of the sine oculis homeobox (Six) protein family, Six1, Six2 and Six4, have been implicated in EMT events during development and cancer [10]. Six family protein expression is regulated at the level of transcription, by proteasome-mediated degradation and in some cancers by gene amplification [359]. Notably, Six1 expression is regulated during the cell cycle, with its maximum expression during S and G2 phase, and its transcription is induced by E2F1 [360, 361]. Six1 is then degraded via the proteasomal pathway at the G2/M phase transition via the anaphase-promoting complex [362]. Structurally, the Six family of transcription factors are characterized by a helixturn-helix motif and the presence of the homeobox nucleic acid recognition domain. While Six1, Six2 and Six4 show similarity in the binding to MEF3 sites on DNA (TCAGGTTC) to induce transcription, Six1 also binds structural DNA features, like nonsequence-specific HMG protein recognition sites [363–365]. The Six family of proteins appear to regulate a wide variety of genes that are critical for the migration of myogneic cells as well as organogenesis of the eye and kidney during development [366, 367]. Specifically, Six1 and Six2 are associated with cells of mesenchymal origin and Six2 appears to be critical for maintaining the mesenchymal phenotype of renal progenitor cells [368]. Interestingly, Six1 overexpression activates TGF-beta signaling and is associated with nuclear Smad3 accumulation and increased expression of TGF-beta responsive genes [369]. Furthermore, Six1 induces features of EMT in vivo in tumors derived from mammary specific Six1 over-expression, including loss of E-cadherin, cytokeratin-18,

40

M.D. Amatangelo and M.E. Stearns

nuclear accumulation of beta-catenin and expression of alpha-smooth muscle actin [370]. Over-expression of Six1 in mammary carcinoma cells in vitro also induces an EMT phenotype that is characterized by dissociation of ‘cell-cell’ contacts, decreases in cytokeratin-18 expression and increases in metastatic potential [371]. Consistent with other regulators of EMT, Six1 over-expression is seen in 90% of human metastatic breast cancer lesions and correlates with adverse outcomes in numerous other cancers including brain, cervical, prostate, colon, kidney, and liver [359, 369]. While the mechanism by which Six1 activates TGF-beta signaling and EMT remains unclear, it must be noted that inhibition of TGF-beta signaling is unable to completely reverse Six1-induced EMT, suggesting that Six1 activates additional pathways responsible for maintaining EMT [369]. Other homeobox transcription factors associated with development also appear to be inducers of EMT. Ladybird homeobox 1 (Lbx1) protein is a regulator of c-Met induced muscle precursor migration in the developing embryo [371]. Recently expression of Lbx1 was found to be associated with a robust EMT characterized by the down regulation of E-cadherin and occludins as well as the up-regulation of N-cadherin, vimentin and fibronectin [372]. Interestingly Lbx1 also induced Snail, Zeb1, Zeb2 and TGF-beta2 expression, although the contribution of TGF-beta2 induction toward this EMT response was not investigated. Significantly, inhibition of Lbx1 in metastatic cells also inhibits migration and increased Lbx1 expression is associated with more invasive subtypes of human breast cancers [372]. Msx2 is a member of muscle-segment homeobox genes and has also been implicated in developmental EMT events during organogenesis and neural tube migration [373, 374]. Msx2 is expressed at the sites of epithelial-mesenchymal interactions mediating the signaling between the two different tissues. Msx2 has also been shown to induce EMT in vitro in breast and pancreatic cancer cell lines as characterized by the loss of E-cadherin, nuclear accumulation of beta-catenin and up-regulation of N-cadherin, vimentin and Twist1 [375, 376]. Furthermore, Msx2 over-expression enhances the metastatic potential of cells and its expression is specifically observed in tumor cells that display an invasive phenotype in clinical samples. Interestingly, part of the mechanism by which Msx2 initiates EMT is by up-regulation of the signaling molecule Cripto [375]. Cripto-knockout mice lack of mesoderm formation and fail to initiate early body patterning, indicating Cripto plays a critical role in type-1 EMT events and inducing the mesenchymal phenotype. Likewise, Cripto expression increases the invasive and migratory properties of cells and induces down-regulation of E-cadherin and up-regulation of vimentin and N-cadherin [377, 378]. Cripto expression is, however, also associated with non-invasive tumor types in vivo, suggesting that induction of Cripto by Msx2 is necessary, but not sufficient to induce an EMT event that stimulates robust invasion and metastasis [10]. The Goosecoid (Gsc) homeobox transcription factors are also a well known activator of cell migration. During development, Gsc is expressed during primitive streak formation and its activity is required for embryonic cell migration and dorsoventral patterning of mesodermal tissues [379–381]. Gsc expression appears to be induced by TGF-beta signaling and down-regulates E-cadherin expression in cells possibly by inducing expression of Snail, Slug, Zeb2 and Twist1 [382, 383]. Studies

2 Reactivation of Epithelial-Mesenchymal Transition in Invasive and Metastatic Cancer

41

have shown that Gsc is also over-expressed in a majority of human breast tumors and ectopic expression of Gsc in human breast cells induces EMT and enhances the ability of breast cancer cells to form pulmonary metastases in mice [382]. While the specific roles of the various homeobox transcription factors during tumorigeneis and EMT are just starting to become unraveled, it is clear that they do play a significant role in regulating type-3 EMT events and metastatic potential in cancer cells.

2.4.5

HMGA2 and Polycomb-Group Proteins

The role of chromatin associated proteins which alter DNA and chromatin structure to regulate the transcriptional activity of several genes associated with EMT is just beginning to be investigated. Of particular interest in regards to EMT and tumor progression is the high mobility group A2 protein (HMGA2, also known as HMGI-C). While HMGA2 is widely expressed during early embryogenesis, it is almost completely absent in differentiated cells and normal adult tissues [384]. During development HMGA2 is expressed in all mesenchymal cell populations and its expression is required for heart organogenesis, all suggesting a possible role for HMGA2 in regulating the mesenchymal cell phenotype [385, 386]. Pathological studies have shown that HMGA2 is often re-expressed in tumors and correlates with occurrence of metastasis and poor prognoses in several human cancers including breast, pancreatic and lung carcinomas and in squamous carcinomas of the oral cavity [387–390]. Ectopic expression of HMGA2 in adult cells induces a mesenchymal phenotype characterized by robust down-regulation of E-cadherin and induction of Snail, Slug, Zeb1 and Twist1 [391]. Mechanistically, HMGA2 induces Snail by binding to the proximal region of the Snail promoter and synergizes with Smad3/Smad4 to activate Snail transcription [391]. Furthermore, HMGA2 induction appears to be required for TGF-beta stimulated EMT [347]. The HMGA family of proteins binds fairly ubiquitously to the minor groove of AT-rich DNA sequences. The DNA-binding domain is located in their N-terminus which contains three separate AT-hooks characterized by eight to nine basic amino acid repeats [392]. HMGA proteins do not display direct transcriptional activity, rather they function by altering chromatin architecture and interacting with transcriptional machinery to form multiprotein complexes on promoter/enhancer regions of genes [393]. HMGA proteins also participate in protein–protein interactions with other DNA bound proteins to affect transcriptional activity. For example, expression of HMGA2 enhances the transcriptional activity of E2F1 target genes by displacing HDAC1 from the pRB–E2F1 complex and inducing acetylation of both E2F1 and DNA-associated histones [394]. HMGA2 also binds the p50/p65 NF-kappaB heterodimer and enhances its transcriptional activity at various promoters [395, 396]. HMGA2 expression is induced by binding of the transcription factors c-Jun to AP1 sites, c-Myc to E-boxes and by TGF-beta stimulated Smad 4 binding to sites in the HMGA2 promoter [347, 397, 398]. Interestingly, HMGA2 appears to be one of the more post-transcriptionally modified proteins in the cell, via phosphorylation,

42

M.D. Amatangelo and M.E. Stearns

acetylation and sumolyation, however, the contribution of these events to HMGA2 function remain to be elucidated. While more work is clearly needed to fully understand HMGA2¢s role in EMT and its regulation of EMT associated transcription factors, the ability for it to interact with and enhance various signaling pathways, as well its promiscuous binding to DNA, might very well put it at the forefront as a key master regulator of EMT. Another family of chromatin associated proteins that have recently become of interest in regards to EMT are the Polycomb-group of Proteins (PcG) which are evolutionarily conserved chromatin modifying factors that repress transcription. These proteins regulate covalent modification of specific residues within aminoterminal tails of histones. There are two core PcG complexes which are made up of multiple protein subunits, the polycomb repressive complex 1 and 2 (PRC1 and PRC2 respectively). The PRC1 genes include Bmi-1, Mel-18, RING, Hpc and Hph, while PRC2 includes Eed, Ezh1/2, Suz12 and YH1. In general, PRC2 binds polycomb repressive elements (PREs) on DNA and through the action of the subunit enhancer of zeste homologue 2 (Ezh2) which trimethylates histone H3 at lysine 27 to induce transcriptionally inactive chromatin [399]. These changes in DNA conformation established by PRC2 are recognized by PRC1, which stabilizes the inactive state by ubiquitylating histone H2A to maintain transcriptional repression memory and impede RNA polymerase II initiation of transcription and elongation [400]. In addition to histone methylation, PcG complexes can also recruit proteins that facilitate other covalent modifications including the recruitment of sirtuin 1 (SIRT1) and HDAC2 to acetylate histones as well as DNA methyltransferases (DNMTs) which mediated direct methylation of DNA [401]. Notably, PRC2 proteins and their triMeK27-H3 marks reside at and transcriptionally repress many regulatory genes that control specific developmental lineages. Thus PcG proteins and their antagonist, the trithorax group of proteins, serve to fine tune gross changes in gene expression required for differentiation during embryonic organogenesis [402, 403]. Bmi-1, was the first PcG protein to be implicated in cancer and was initially identified as an oncogene that cooperates with Myc to induce proliferation and resistance to senescence [404]. Subsequently, Bmi-1 has been found to be over-expressed in aggressive and invasive breast cancer, prostate cancer, bladder cancer, melanoma, hepatocellular carcinomas and esophageal squamous cell carcinomas [405–410]. The combined over-expression of Bmi-1 and Ras in epithelial cells results in a mesenchymal-type change in cell morphology, increased invasive properties, and increased metastatic potential [411]. Conversely, knockdown of Bmi-1 in metastatic breast cancer cell lines reduces their invasive capacity. Together, these observations suggest that Bmi-1 may have a role beyond regulation of proliferation and senescence and may regulate EMT and invasion. Indeed, Ras and Bmi-1 expressed in combination can induce the expression of vimentin and fibronectin and decrease E-cadherin expression in normal breast epithelial cells [412]. Furthermore, Twist1 appears to be a direct regulator of Bmi-1 and Twist1 and Bmi-1 are mutually essential to promote EMT and act cooperatively to repress expression of E-cadherin [413]. Bmi-1 expression can also induce EMT and invasion in nasopharyngeal epithelial cells by the stabilization of Snail via modulation of GSK-3beta signaling [414].

2 Reactivation of Epithelial-Mesenchymal Transition in Invasive and Metastatic Cancer

43

Ezh2, a member of the PRC2 complex, has been shown to play a role in controlling actin polymerization in fibroblast cells, suggesting Ezh2 expression could also contribute to invasion by regulation of actin-dependent cell adhesion and migration [415]. Ezh2 might directly impact EMT by mediating transcriptional silencing of E-cadherin by trimethylation of H3 lysine 27 [416]. Additionally, Ezh2 has been shown to silence DAB2IP, a Ras GAP, and induce Ras hyperactivity and prostate cancer metastasis [417]. Taken together, these studies suggest a significant role for PcG in the regulation of EMT related pathways during tumorigenesis and highlight how epigenetic changes in cells might impact their ability to undergo EMT. In sum, it is clear that aberrant reactivation of developmental transcription factors is a critical event for EMT to occur, and that this enhances tumor cell migration, invasion and metastatic potential. One important concern at this juncture surrounds understanding the relative contribution of these different transcription factors and chromatin associated proteins during EMT. Future studies should focus on the relative interdependence of these transcription factors to induce EMT, whether a particular hierarchy of expression between factors exists and the relative robustness of EMT induced by each. For example, differences in potency of gene regulation exist as Slug inhibition of E-cadherin appears to be less effective than that of Snail, correlates with weaker binding affinity to the E-cadherin promoter and is associated with a different phenotypes in vivo [418]. In addition, a hierarchical pattern might exist where a dedifferentiated cell state induced in part by expression of PcG proteins, homoeobox proteins and HMGA2 is required for Snail and Slug expression in response to TGF-beta, leading to the activation of Zeb and Twist family members [419]. These transcription factors then orchestrate a down-regulation of adherence and tight junctions and a remodeling of the cytoskeleton and ECM to induce tumor cell invasion and metastasis as part of an EMT program. Some reports also suggest that EMT and tumor cell dissemination may be an early event in tumor progression before gross genetic abnormalities arise [214, 420]. Therefore, understanding the regulation of EMT in primary cell lines from early stage cancers might be key to understanding critical interactions amongst these transcription factors to control for active background pathways which may be contributing to the EMT phenotype.

2.5

MicroRNAs and EMT

Since the discovery of microRNAs (miRNAs), over 900 miRNA genes been identified in the human genome [421]. In many cases, it also appears that each mRNA species can influence global patterns of gene expression. Thus, miRNAs constitute one of the most significant classes of gene-regulatory molecules in humans. As such, miRNAs have been shown to be key players in a broad range of developmental and physiological processes and their deregulation has been closely linked to human diseases including cancer. MicroRNA profiling studies in a variety cancers have revealed that their expression varies considerably between malignant and normal

44

M.D. Amatangelo and M.E. Stearns

tissues [422–425]. While it appears that many miRNAs are down-regulated as tumors become less differentiated, consistent with a role for miRNAs in determining cellular differentiation states, miRNAs appear to act as critical components of both oncogenic and tumor suppressor pathways [426, 427]. Notably, as the genes and processes which miRNAs regulate expand, it has become increasing clear that miRNA expression indeed serves as an important regulator of cellular differentiation, cell fate, stemness and EMT. miRNAs are a diverse family of 18–24 nucleotide non-coding RNAs that posttranscriptionally regulate the stability and translational efficiency of target mRNAs. Generally, miRNAs function through imperfect complimentary base pairing with specific sequences in the 3¢ untranslated regions (UTRs) of target mRNAs. The key region of the miRNA that largely governs its target specificity is known as the seed sequence, and encompasses bases 2–7 from its 5¢ end [428]. In the nucleus, miRNAs are transcribed as a primary transcripts (pri-miRNAs), by RNA polymerase II or III, which can be several 1,000 nucleotides in length. These transcripts are 5¢-capped, spliced, and polyadenylated, and often contain more than one functional miRNA [429]. In the nucleus, pri-miRNAs are then cleaved into 70–80 nucleotide hairpin-loop structures called precursor miRNAs (pre-miRNA) by a multiprotein complex called the microprocessor which contains Drosha, an RNase III enzyme, and DGCR8/Pasha, a double-stranded RNA-binding protein [430]. Drosha cleavage generates a 2 nucleotide 3¢ overhang that is recognized by exportin 5 which transports the pre-miRNA into the cytoplasm in a RanGTP-dependent manner [431]. Once in the cytoplasm, pre-miRNAs are further processed by the RNase-III enzyme Dicer, which removes the stem loop and produces a 22 nucleotide RNA duplex. Of this duplex, one strand (the guide strand) stably associates with an Argonaute (Ago) protein within the RNA-induced silencing complex (RISC), while the other strand (passenger strand) is discarded [432]. This miRNA-RISC complex is then recruited to target mRNAs. There are multiple mechanisms by which miRNA expression can be changed during tumor development. To begin with, changes in miRNA expression between normal and cancer associated tissues can often be attributed to the location of miRNA genes in regions of chromosomal instability characteristic of certain cancers [433]. The expression and activation of specific transcription factors (often fueled by activation of signaling networks described above) also play an important role in regulating miRNA expression. Many miRNAs have been shown to be regulated by well-established tumor suppressor and oncogene pathways, such as p53, Myc and Ras in which miRNAs and their transcriptional regulators participate in complex feedback regulation loops which appear to drive tumorigenesis [434]. Interestingly, DICER is a haploinsufficient tumor suppressor and 27% of various tumors are found to have a hemizygous deletion [435]. Knockdown of Dicer1 also appears to increase the malignant and invasive potential of already transformed cell lines and to increase the rate of tumor formation in cancer mouse models [436, 437]. The phosphorylation of the Dicer cofactor TARBP2 by the MAPK pathway has also been shown to positively affect the processing of oncogenic miRNAs and negatively regulate the tumor suppressor let-7a [438]. Also, TGF-beta, through Smad signaling, promotes a rapid increase in expression of oncogenic miR-21 by promoting the

2 Reactivation of Epithelial-Mesenchymal Transition in Invasive and Metastatic Cancer

45

recruitment of pri-miR-21 to the Drosha microprocessor complex [439]. Thus regulation of miRNA processing also appears to play an important role in miRNA expression and function during tumorigenesis. While multiple mechanisms contribute to aberrant miRNA transcription and function in cancer, it has become apparent that various RNA binding proteins also play an important role in regulating the processing and stability of miRNAs in normal and malignant tissues. Lin28A (Lin28), Lin28B and RPS2 (Ribosomal Protein S2) all have been shown to be over-expressed in advanced tumors and bind to the pre-let7 hairpin and inhibit expression of the mature form of this potent tumor suppressor [440–442]. Furthermore, while it appears that pre-miRNA complexing with these proteins interferes with functional incorporation into the RISC silencing complex, Lin28 has also been shown promote miRNA degradation by binding to the GGAG sequence in the terminal loop and mediating terminal uridylation of preLet-7, 107, 143, and 200c miRNAs and recruiting 3¢ terminal uridylyl transferases [443, 444]. Many of these very miRNAs seem to be critical for all three types of EMT events. Recently it has been shown that over-expression of Lin28B is sufficient to induce EMT and invasion in hepatocellular carcinoma cell lines [445]. Given that miRNAs play critical roles in regulating multiple pathways and global gene expression patterns, it is not surprising that they also play an important role in orchestrating the extensive changes that occur during EMT. Thus, regulation of miRNA expression and function appears to be one of the most important regulators EMT and cancer cell invasion and metastasis. The extent of which is just now beginning to be realized. miRNAs represent a critical downstream component of key developmental, oncogenic and tumor suppressor pathways, and as has been discussed in previous sections, EMT appears to be driven by developmental and oncogenic pathways [427]. Thus, aberrant activation of these pathways in tumor cells directly influences miRNA expression, which may consequently affect EMT and metastasis [437]. This is elegantly illustrated by the process of miRNA expression controlled by the oncogenic transcription factor c-Myc, one of the most commonly activated oncoproteins in human cancer [446]. Myc directly induces expression of the miR17–92 cluster, which subsequently promotes proliferation, inhibits apoptosis, and promotes tumor angiogenesis [447]. Moreover, Myc activation leads to widespread repression of many additional miRNA genes including several with known tumor suppressor activity such as members of the Let-7 family, miR-15a/16–1, the miR-29 family, and miR-34a [446]. It appears that the predominant consequence of activation of Myc is widespread repression of miRNA expression as a direct result of Myc binding to miRNA promoters. This supports a critical role for miRNA repression in the Myc oncogenic mechanism of action. Furthermore, Myc is negatively regulated by the developmental miRNA, Let-7, whose functionality must be reduced for Myc expression [448]. Thus, there is a complex interplay and delicate balance of developmental and oncogenic transcription factor expression and miRNAs whose disruption appears to lie at the heart of tumorigenesis and the metastatic cascade.

46

2.5.1

M.D. Amatangelo and M.E. Stearns

The miRNA-200 Family

The best characterized miRNAs involved in regulating EMT are the miR-200 family. A number of reports have identified changes in their expression as a fundamental regulator of EMT [449–451]. Microarray analysis comparing mesenchymal cell lines to epithelial lines has revealed that a reduction in the levels of the miRNA-200 family and miRNA-205 are the most consistent markers associated with cells which lack E-cadherin and express vimentin [449]. miR-200 family members are also routinely found to be down-regulated during TGF-beta induced EMT; likewise, inhibition or ectopic expression of the miRNA-200 family members can induce EMT or epithelial cell characteristics respectively [451]. Accordingly, miR-200 family can reverse/inhibit EMT and are powerful inducers of epithelial differentiation. The miR-200 family consists of 5 highly homologous RNAs: miR-141, miR-429, miR-200a, miR-200b, and miR-200c. These miRNAs differ by only a single nucleotide in their seed sequence. miR-200b, miR-429, and miR-200c share the seed sequence 5¢-AAUACU-3¢ while miR-200a and miR-141 share the sequence 5¢-AACACU-3¢. Functionally, miR-200 family members act through the targeted silencing of the E-cadherin transcriptional repressors, Zeb1 and Zeb2, thus increasing the levels of E-cadherin and enforcing the epithelial phenotype [451]. miR-205 has also been shown to target both Zeb1 and Zeb2 [451]. Interestingly, Zeb1 and Zeb2 both suppress miR-200 family member expression, creating a double negative feedback loop that enables miR-200 family and Zeb proteins to respectively enforce the epithelial and mesenchymal cellular states [452]. The miR-200 family also inhibits the expression of other targets known to contribute to EMT, including TGFbeta2 by miR-141, beta-catenin by miR-200a, fibronectin by miR-200c and Bmi-1 by miR-200c [453–456]. Other EMT related targets of the miR-200 family may still be identified. In several cancer cells and models, expression of miR-200 family members inversely correlate with invasive and metastatic potential. For example, downregulation of the miR-200 family is part of a distinct miRNA signature of metastases from primary tumors in mice predisposed to develop pancreatic neuroendocrine tumors [457]. Metastasis and EMT, but not primary tumor growth rate, has also been shown to be entirely dependent on TGF-beta down-regulation of miR-200 family expression, where forced expression of the miR-200b cluster completely abrogates EMT and metastatic potential [458]. Additionally miR-200a expression inhibits the in vitro migration and invasion of highly metastatic and invasive nasopharyngeal carcinoma cells, miR-200b expression inhibits the invasive capacity of LNCaP prostate adenocarcinoma cells and MDA-MB-231 breast cancer cells and miR-200b inhibition supports invasion of non-invasive HT29 colon cancer cells [453, 459]. Interestingly, in cell lines isolated from a spontaneously arising mammary carcinoma which exhibit different metastatic potential, 4T1 cells, the most aggressively metastatic, express comparatively high levels of members of the miR-200 family [460]. These results seemingly contradict our current understanding of the tumor suppressor functions of the miR-200 family. However, this study highlights

2 Reactivation of Epithelial-Mesenchymal Transition in Invasive and Metastatic Cancer

47

three issues regarding EMT and metastasis. First, cells might need to re-establish epithelial characteristics by inducing a mesenchymal to epithelial transition (MET) in order to successfully form macroscopic secondary tumors. Therefore, expression of miR-200 family members might support the ability for cells to proliferate at secondary sites. Second, type-3 EMT events which lead to metastasis might not need to be full EMTs in that they do not require all the characteristics of an EMT to effectively disseminate and metastasize. It is quite likely that cells only need to surpass a threshold of sufficient EMT characteristics to successfully orchestrate their metastatic dissemination. In this regard, it should be noted that 4T1 cells undergoing EMT express E-cadherin while also expressing vimentin and Twist1 [460]. Lastly, given the heterogeneity of tumors, it is possible that a subset of highly metastatic 4T1 cells transiently loses miR-200 expression and induces an EMT which is undetectable but sufficient for invasion and metastasis. Regardless, the miR-200 family remains one of the most significant regulators of EMT. Future work is however needed to understand the results from 4T1 cells to clarify the roles of EMT and miR-200 expression in metastasis.

2.5.2

The Let-7 Family

The Let-7/miR-98 family was one of the first mammalian miRNAs to be identified and is ubiquitously expressed in normal adult tissues [461, 462]. The Let-7 family has 12 members (Let-7-a1, a2, a3, b, c, d, e, f1, f2, g, i and miR-98) with nine distinct mature miRNA sequences all with identical seed sequences (5¢-GAGGUA-3¢). Let-7 is expressed late in mammalian embryonic development and plays an evolutionarily conserved role of promoting cell and tissue differentiation [463, 464]. With regards to cancer, a plethora of human malignancies, including lung, ovarian, gastric, prostate cancer and melanoma, appear to down regulate Let-7 family members during tumor development [422, 465–468]. Multiple loci that encode Let-7 miRNAs are also frequently deleted in cancer cells [433]. The human Let-7 family of miRNAs possesses potent tumor suppressor activity by targeting multiple oncogenes, including those involved in EMT. Specifically Let-7 was shown to directly target Ras, negatively correlate with Ras expression and inhibit growth of human lung cancer cells [469, 470]. At least nine members of the Let-7 family efficiently bind multiple complementary sites in the 3¢-UTRs of all three human Ras genes [470]. Given the key role of Ras signaling in inducing EMT, Let-7 down-regulation might help to induce EMT by amplifying Ras signals. Indeed over-expression of Let-7 in lung cancer cells represses genes regulated by Ras [466]. Let-7 miRNAs are also a strong negative regulator of HMGA2, binding to multiple target sites in the 3¢ UTR [471]. HMGA2 is known to be expressed in a variety of tumors which also express low levels of Let-7 and correlates with poor prognosis [389, 472, 473]. Most importantly, HMGA2 has been described as a critical factor necessary for TGF-beta induced EMT in mouse primary mammary epithelial cells and its expression in conjunction with oncogenic Ras signaling seems to be required

48

M.D. Amatangelo and M.E. Stearns

for Snail expression and EMT in human pancreatic cancer cells [347, 474]. Taken together these studies suggest that Let-7 family members might be potent negative regulators of EMT. While detailed analysis of the direct role for Let-7 in EMT is lacking, interestingly, miRNA expression analysis of cancer cell lines suggests that Let-7 is a marker for less advanced cancers and negatively correlates with reactivation of embryonic mesenchymal genes [473]. Let-7a has also been shown to down-regulate expression of the proliferative transcription factor Myc and its target genes [448]. As eluded to above, Myc participates in a feedback loop, inhibiting Let-7 expression by transcriptional repression [446]. Myc also transcriptionally induces the Let-7 negative regulator, Lin28 [475]. Taken together with the observation that MAPK signaling through Erk2 can increase Myc activity, it appears that a signaling cascade involving Ras-MAPK-Myc-Lin28 down regulation of Let-7 might play a critical role in regulating metastasis and EMT [210, 475, 476]. Thus, disrupting the delicate balance of this loop is likely a potent facilitator of tumor progression, invasion and metastasis. Interestingly, recent findings have also connected Lin28 and Myc expression with stem cell maintenance and the pluripotent state, where expression of Myc and Lin28 appear to be interchangeable [477, 478]. Given that expression of Let-7 leads to differentiation, this data suggests that regulation of Let-7 expression by Myc and Lin28 might pose as a connection between EMT, pluripotency and cancer stem cells. For example, TGF-beta induction of EMT in human mammary epithelial cells not only involves up-regulation of EMT transcriptional regulators (Snail, Zeb2 and Twist), but also includes induction of classical stem cell markers [18, 479]. In mouse tumor modeling studies, these cells are able to form metastasis at very low titers and are thought to function as tumor initiating cells. Down-regulation of Let-7 expression appears to play a significant role in the activation of these tumor initiating cells. Tumor initiating cells isolated from breast cancers express very low levels of Let-7, and induction of Let-7 in these cells inhibits metastasis [480]. Ultimately it appears that the main function of Let-7 miRNAs are to suppress early embryonic genes which may have activities in the maintaining stem cell cell-like phenotype which is essential for EMT and induction of the metastatic phenotype. In future studies, development of therapeutics which prevents Lin28 or RPS2 binding to pre-Let-7 RNA may be highly beneficial in blocking EMT and metastasis.

2.5.3

Other miRNAs and EMT

In addition to the miR-200 family and let-7 family, several other miRNAs have been shown to alter the expression of EMT related genes, impacting cellular morphology, invasion and metastasis. Some miRNAs have been identified in TGF-beta induced EMT, and are directly regulated by TGF-beta stimulation. Levels of miR-155 have been shown to be increased during TGF-beta induced EMT in mammary epithelial cells via transcriptional activation by Smad4 [481]. While ectopic expression of

2 Reactivation of Epithelial-Mesenchymal Transition in Invasive and Metastatic Cancer

49

miR-155 does not induce EMT, it does induce disruption of cell polarity and tight junctions and loss of miR-155 suppresses TGF-beta induced EMT. miR-29a and miR-21 levels also appear to increase during TGF-beta induced EMT and are expressed at higher levels in mesenchymal cells when compared to epithelial cells [449, 482]. At this juncture the direct role of miR-29a in metastatic progression is unclear, however, miR-21 is found to be over-expressed in lung, breast, stomach, prostate, colon, brain, and pancreas carcinomas and stimulates invasion and metastasis in colon and breast cancer cells [483–485]. miR-21 transcription appears to be regulated by TGF-beta activation of the AP-1 and Zeb1 transcription factors in a Smad4 independent manner [486]. Twist1, as part of its ability to regulate EMT, activates miR-10b, which contributes to the migration, invasion and metastasis of breast cancer cells and its expression in primary tumors correlates with clinical progression [487]. miR-9 is also up-regulated by Myc and directly targets E-cadherin mRNA resulting in increased cell motility, invasiveness and metastatic potential [488]. Furthermore, miR-9 is expressed at high levels specifically in primary breast tumors from patients who have developed metastases [488]. Conflicting with these results, miR-10a and miR-9 do not appear to be expressed at high levels in actual distant metastasis [489, 490]. This is understandable if we assume EMT at the invasive front of tumors only leads to transient activation of miR-9 and miR-10b to promote dissemination and invasion and upon metastatic seeding, expression of these miRNAs is lost as cells undergo an MET. In conclusion, the network of oncogenes and miRNAs that play a role in controlling genes associated with EMT is in its infancy. Undoubtedly, many more pathways which regulate EMT will emerge that will help our understanding of the mechanisms involved in metastatic dissemination. As global regulators of gene expression and differentiation, the field of miRNAs remains an extraordinarily exciting opportunity to identify key regulators of EMT that are not context dependent. Most importantly, the prospects of developing novel diagnostics for cancer or discovering therapeutics which target miRNA regulatory pathways present enormous opportunities to improve cancer patient outcomes. As cancers are dynamic and heterogeneous in nature, future drug discovery efforts need to consider which cancer cells pose the greatest threat to patient survival. It is clear from the literature that the cellular characteristics adopted during an EMT, such as dissolution of epithelial cell contacts, cytoskeletal changes and matrix remodeling, all promote progression to metastatic and deadly disease. Furthermore, the key signaling pathways which induce EMT, namely TGF-beta, Wnt, MAPK and AKT, also promote tumor progression and metastasis. Targeting these EMT pathways is a promising therapeutic strategy as knockdown of several transcription factors regulating EMT, such as those of the Snail, Twist and Zeb families, effectively inhibit metastasis in mouse tumor modeling studies. Moreover, EMT is increasingly associated with dedifferentiation and tumor initiating cells, thus targeting cells with stem cell-like characteristics or pharmacological induction of differentiation may also have therapeutic value. Finally, therapeutic delivery of endogenous miRNAs or miRNAs which specifically target EMT regulators may represent a potentially potent therapy to prevent metastasis and extend patient survival.

50

M.D. Amatangelo and M.E. Stearns

References 1. Viebahn C (1995) Epithelio-mesenchymal transformation during formation of the mesoderm in the mammalian embryo. Acta Anat (Basel) 154(1):79–97 2. Duband JL et al (1995) Epithelium-mesenchyme transition during neural crest development. Acta Anat (Basel) 154(1):63–78 3. Krug EL, Runyan RB, Markwald RR (1985) Protein extracts from early embryonic hearts initiate cardiac endothelial cytodifferentiation. Dev Biol 112(2):414–426 4. Yan C et al (2010) Epithelial to mesenchymal transition in human skin wound healing is induced by tumor necrosis factor-alpha through bone morphogenic protein-2. Am J Pathol 176(5):2247–2258 5. Greenburg G, Hay ED (1982) Epithelia suspended in collagen gels can lose polarity and express characteristics of migrating mesenchymal cells. J Cell Biol 95(1):333–339 6. Hay ED (1995) An overview of epithelio-mesenchymal transformation. Acta Anat (Basel) 154(1):8–20 7. Xue C et al (2003) The gatekeeper effect of epithelial-mesenchymal transition regulates the frequency of breast cancer metastasis. Cancer Res 63(12):3386–3394 8. Davies JA (1996) Mesenchyme to epithelium transition during development of the mammalian kidney tubule. Acta Anat (Basel) 156(3):187–201 9. Kalluri R, Weinberg RA (2009) The basics of epithelial-mesenchymal transition. J Clin Invest 119(6):1420–1428 10. Micalizzi DS, Farabaugh SM, Ford HL (2010) Epithelial-mesenchymal transition in cancer: parallels between normal development and tumor progression. J Mammary Gland Biol Neoplasia 15(2):117–134 11. Tarin D, Thompson EW, Newgreen DF (2005) The fallacy of epithelial mesenchymal transition in neoplasia. Cancer Res 65(14):5996–6000; discussion 6000–1 12. Trimboli AJ et al (2008) Direct evidence for epithelial-mesenchymal transitions in breast cancer. Cancer Res 68(3):937–945 13. Friedl P, Gilmour D (2009) Collective cell migration in morphogenesis, regeneration and cancer. Nat Rev Mol Cell Biol 10(7):445–457 14. Giampieri S et al (2009) Localized and reversible TGFbeta signalling switches breast cancer cells from cohesive to single cell motility. Nat Cell Biol 11(11):1287–1296 15. Braun S et al (2000) Cytokeratin-positive cells in the bone marrow and survival of patients with stage I, II, or III breast cancer. N Engl J Med 342(8):525–533 16. Friedl P, Wolf K (2008) Tube travel: the role of proteases in individual and collective cancer cell invasion. Cancer Res 68(18):7247–7249 17. Gaggioli C et al (2007) Fibroblast-led collective invasion of carcinoma cells with differing roles for RhoGTPases in leading and following cells. Nat Cell Biol 9(12):1392–1400 18. Mani SA et al (2008) The epithelial-mesenchymal transition generates cells with properties of stem cells. Cell 133(4):704–715 19. Kong D et al (2010) Epithelial to mesenchymal transition is mechanistically linked with stem cell signatures in prostate cancer cells. PLoS One 5(8):e12445 20. Shimoyama Y et al (1989) Cadherin cell-adhesion molecules in human epithelial tissues and carcinomas. Cancer Res 49(8):2128–2133 21. McNeill H et al (1990) Novel function of the cell adhesion molecule uvomorulin as an inducer of cell surface polarity. Cell 62(2):309–316 22. Marrs JA et al (1995) Plasticity in epithelial cell phenotype: modulation by expression of different cadherin cell adhesion molecules. J Cell Biol 129(2):507–519 23. Hay ED, Zuk A (1995) Transformations between epithelium and mesenchyme: normal, pathological, and experimentally induced. Am J Kidney Dis 26(4):678–690 24. Birchmeier W, Behrens J (1994) Cadherin expression in carcinomas: role in the formation of cell junctions and the prevention of invasiveness. Biochim Biophys Acta 1198(1): 11–26

2 Reactivation of Epithelial-Mesenchymal Transition in Invasive and Metastatic Cancer

51

25. Byers SW et al (1995) Role of E-cadherin in the response of tumor cell aggregates to lymphatic, venous and arterial flow: measurement of cell-cell adhesion strength. J Cell Sci 108(Pt 5):2053–2064 26. Thiery JP et al (2009) Epithelial-mesenchymal transitions in development and disease. Cell 139(5):871–890 27. Onder TT et al (2008) Loss of E-cadherin promotes metastasis via multiple downstream transcriptional pathways. Cancer Res 68(10):3645–3654 28. Vleminckx K et al (1991) Genetic manipulation of E-cadherin expression by epithelial tumor cells reveals an invasion suppressor role. Cell 66(1):107–119 29. Frixen UH et al (1991) E-cadherin-mediated cell-cell adhesion prevents invasiveness of human carcinoma cells. J Cell Biol 113(1):173–185 30. Perl AK et al (1998) A causal role for E-cadherin in the transition from adenoma to carcinoma. Nature 392(6672):190–193 31. Gottardi CJ, Wong E, Gumbiner BM (2001) E-cadherin suppresses cellular transformation by inhibiting beta-catenin signaling in an adhesion-independent manner. J Cell Biol 153(5): 1049–1060 32. Berx G, Van Roy F (2001) The E-cadherin/catenin complex: an important gatekeeper in breast cancer tumorigenesis and malignant progression. Breast Cancer Res 3(5):289–293 33. Christofori G, Semb H (1999) The role of the cell-adhesion molecule E-cadherin as a tumoursuppressor gene. Trends Biochem Sci 24(2):73–76 34. Liebner S et al (2004) Beta-catenin is required for endothelial-mesenchymal transformation during heart cushion development in the mouse. J Cell Biol 166(3):359–367 35. Kalluri R, Neilson EG (2003) Epithelial-mesenchymal transition and its implications for fibrosis. J Clin Invest 112(12):1776–1784 36. Brabletz T et al (1998) Nuclear overexpression of the oncoprotein beta-catenin in colorectal cancer is localized predominantly at the invasion front. Pathol Res Pract 194(10):701–704 37. Brabletz T et al (2001) Variable beta-catenin expression in colorectal cancers indicates tumor progression driven by the tumor environment. Proc Natl Acad Sci USA 98(18):10356–10361 38. Prasad CP et al (2009) Expression analysis of E-cadherin, Slug and GSK3beta in invasive ductal carcinoma of breast. BMC Cancer 9:325 39. Logullo AF et al (2010) Concomitant expression of epithelial-mesenchymal transition biomarkers in breast ductal carcinoma: association with progression. Oncol Rep 23(2):313–320 40. Christofori G (2003) Changing neighbours, changing behaviour: cell adhesion moleculemediated signalling during tumour progression. EMBO J 22(10):2318–2323 41. Ong LL et al (1998) Trabecular myocytes of the embryonic heart require N-cadherin for migratory unit identity. Dev Biol 193(1):1–9 42. Hatta K, Takeichi M (1986) Expression of N-cadherin adhesion molecules associated with early morphogenetic events in chick development. Nature 320(6061):447–449 43. Tomita K et al (2000) Cadherin switching in human prostate cancer progression. Cancer Res 60(13):3650–3654 44. Hazan RB et al (1997) N-cadherin promotes adhesion between invasive breast cancer cells and the stroma. Cell Adhes Commun 4(6):399–411 45. Li G, Herlyn M (2000) Dynamics of intercellular communication during melanoma development. Mol Med Today 6(4):163–169 46. Hazan RB et al (2000) Exogenous expression of N-cadherin in breast cancer cells induces cell migration, invasion, and metastasis. J Cell Biol 148(4):779–790 47. Gravdal K et al (2007) A switch from E-cadherin to N-cadherin expression indicates epithelial to mesenchymal transition and is of strong and independent importance for the progress of prostate cancer. Clin Cancer Res 13(23):7003–7011 48. Simonneau L et al (1995) Cadherin 11 expression marks the mesenchymal phenotype: towards new functions for cadherins? Cell Adhes Commun 3(2):115–130

52

M.D. Amatangelo and M.E. Stearns

49. Matsuyoshi N et al (1997) Identification of novel cadherins expressed in human melanoma cells. J Invest Dermatol 108(6):908–913 50. Pishvaian MJ et al (1999) Cadherin-11 is expressed in invasive breast cancer cell lines. Cancer Res 59(4):947–952 51. Nieman MT et al (1999) N-cadherin promotes motility in human breast cancer cells regardless of their E-cadherin expression. J Cell Biol 147(3):631–644 52. Pekny M, Lane EB (2007) Intermediate filaments and stress. Exp Cell Res 313(10): 2244–2254 53. Eckes B et al (2000) Impaired wound healing in embryonic and adult mice lacking vimentin. J Cell Sci 113(Pt 13):2455–2462 54. Franke WW et al (1982) Formation of cytoskeletal elements during mouse embryogenesis. III. Primary mesenchymal cells and the first appearance of vimentin filaments. Differentiation 23(1):43–59 55. Duprey P, Paulin D (1995) What can be learned from intermediate filament gene regulation in the mouse embryo. Int J Dev Biol 39(3):443–457 56. Page M (1989) Changing patterns of cytokeratins and vimentin in the early chick embryo. Development 105(1):97–107 57. Thomas PA et al (1999) Association between keratin and vimentin expression, malignant phenotype, and survival in postmenopausal breast cancer patients. Clin Cancer Res 5(10):2698–2703 58. Gilles C et al (1996) Vimentin expression in cervical carcinomas: association with invasive and migratory potential. J Pathol 180(2):175–180 59. Hu L et al (2004) Association of Vimentin overexpression and hepatocellular carcinoma metastasis. Oncogene 23(1):298–302 60. Lang SH et al (2002) Enhanced expression of vimentin in motile prostate cell lines and in poorly differentiated and metastatic prostate carcinoma. Prostate 52(4):253–263 61. Otsuki S et al (2011) Vimentin expression is associated with decreased survival in gastric cancer. Oncol Rep 25(5):1235–1242 62. McInroy L, Maatta A (2007) Down-regulation of vimentin expression inhibits carcinoma cell migration and adhesion. Biochem Biophys Res Commun 360(1):109–114 63. Zhao Y et al (2008) Vimentin affects the mobility and invasiveness of prostate cancer cells. Cell Biochem Funct 26(5):571–577 64. Heatley M et al (1995) Vimentin and cytokeratin expression in nodular hyperplasia and carcinoma of the prostate. J Clin Pathol 48(11):1031–1034 65. Raimondi C et al (2011) Epithelial-mesenchymal transition and stemness features in circulating tumor cells from breast cancer patients. Breast Cancer Res Treat 130:449–455 66. Hofman V et al (2010) Detection of circulating tumor cells as a prognostic factor in patients undergoing radical surgery for non-small cell lung carcinoma: comparison of the efficacy of the CellSearch Assay and the isolation by size of epithelial tumor cell method. Int J Cancer 129(7):1651–1660 67. Thompson EW, Newgreen DF, Tarin D (2005) Carcinoma invasion and metastasis: a role for epithelial-mesenchymal transition? Cancer Res 65(14):5991–5995; discussion 5995 68. Barraclough R (1998) Calcium-binding protein S100A4 in health and disease. Biochim Biophys Acta 1448(2):190–199 69. Strutz F et al (1995) Identification and characterization of a fibroblast marker: FSP1. J Cell Biol 130(2):393–405 70. Takenaga K, Nakamura Y, Sakiyama S (1994) Cellular localization of pEL98 protein, an S100-related calcium binding protein, in fibroblasts and its tissue distribution analyzed by monoclonal antibodies. Cell Struct Funct 19(3):133–141 71. Cabezon T et al (2007) Expression of S100A4 by a variety of cell types present in the tumor microenvironment of human breast cancer. Int J Cancer 121(7):1433–1444 72. Kim EJ, Helfman DM (2003) Characterization of the metastasis-associated protein, S100A4. Roles of calcium binding and dimerization in cellular localization and interaction with myosin. J Biol Chem 278(32):30063–30073

2 Reactivation of Epithelial-Mesenchymal Transition in Invasive and Metastatic Cancer

53

73. Okada H et al (1997) Early role of Fsp1 in epithelial-mesenchymal transformation. Am J Physiol 273(4 Pt 2):F563–F574 74. Iwano M et al (2002) Evidence that fibroblasts derive from epithelium during tissue fibrosis. J Clin Invest 110(3):341–350 75. Stein U et al (2006) The metastasis-associated gene S100A4 is a novel target of beta-catenin/ T-cell factor signaling in colon cancer. Gastroenterology 131(5):1486–1500 76. Grigorian M et al (1996) Effect of mts1 (S100A4) expression on the progression of human breast cancer cells. Int J Cancer 67(6):831–841 77. Levett D et al (2002) Transfection of S100A4 produces metastatic variants of an orthotopic model of bladder cancer. Am J Pathol 160(2):693–700 78. Takenaga K, Nakamura Y, Sakiyama S (1997) Expression of antisense RNA to S100A4 gene encoding an S100-related calcium-binding protein suppresses metastatic potential of highmetastatic Lewis lung carcinoma cells. Oncogene 14(3):331–337 79. Ambartsumian NS et al (1996) Metastasis of mammary carcinomas in GRS/A hybrid mice transgenic for the mts1 gene. Oncogene 13(8):1621–1630 80. Rudland PS et al (2000) Prognostic significance of the metastasis-inducing protein S100A4 (p9Ka) in human breast cancer. Cancer Res 60(6):1595–1603 81. Helfman DM et al (2005) The metastasis associated protein S100A4: role in tumour progression and metastasis. Br J Cancer 92(11):1955–1958 82. Hay ED (1999) Biogenesis and organization of extracellular matrix. FASEB J 13(Suppl 2): S281–S283 83. Lee GY et al (2007) Three-dimensional culture models of normal and malignant breast epithelial cells. Nat Methods 4(4):359–365 84. Barsky SH et al (1983) Loss of basement membrane components by invasive tumors but not by their benign counterparts. Lab Invest 49(2):140–147 85. Duband JL, Thiery JP (1982) Appearance and distribution of fibronectin during chick embryo gastrulation and neurulation. Dev Biol 94(2):337–350 86. Poelmann RE et al (1990) The extracellular matrix during neural crest formation and migration in rat embryos. Anat Embryol (Berl) 182(1):29–39 87. Zeisberg M, Neilson EG (2009) Biomarkers for epithelial-mesenchymal transitions. J Clin Invest 119(6):1429–1437 88. Gaggioli C et al (2007) Tumor-derived fibronectin is involved in melanoma cell invasion and regulated by V600E B-Raf signaling pathway. J Invest Dermatol 127:400–410 89. Gehlsen KR et al (1988) Inhibition of in vitro tumor cell invasion by Arg-Gly-Asp-containing synthetic peptides. J Cell Biol 106:925–930 90. Zeng ZZ et al (2006) Role of focal adhesion kinase and phosphatidylinositol 3¢-kinase in integrin fibronectin receptor-mediated, matrix metalloproteinase-1-dependent invasion by metastatic prostate cancer cells. Cancer Res 66(16):8091–8099 91. Jia Y et al (2004) Integrin fibronectin receptors in matrix metalloproteinase-1-dependent invasion by breast cancer and mammary epithelial cells. Cancer Res 64(23):8674–8681 92. Yang Z et al (2007) Up-regulation of gastric cancer cell invasion by Twist is accompanied by N-cadherin and fibronectin expression. Biochem Biophys Res Commun 358(3):925–930 93. Allen M, Louise Jones J (2011) Jekyll and Hyde: the role of the microenvironment on the progression of cancer. J Pathol 223(2):162–176 94. Gorczyca W, Holm R, Nesland JM (1993) Laminin production and fibronectin immunoreactivity in breast carcinomas. Anticancer Res 13(4):851–858 95. Paszek MJ et al (2005) Tensional homeostasis and the malignant phenotype. Cancer Cell 8(3):241–254 96. Boyd NF et al (2005) Mammographic breast density as an intermediate phenotype for breast cancer. Lancet Oncol 6(10):798–808 97. Davidson LA et al (2006) Integrin alpha5beta1 and fibronectin regulate polarized cell protrusions required for Xenopus convergence and extension. Curr Biol 16(9):833–844 98. Maschler S et al (2005) Tumor cell invasiveness correlates with changes in integrin expression and localization. Oncogene 24(12):2032–2041

54

M.D. Amatangelo and M.E. Stearns

99. Qian F et al (2005) Interaction between integrin alpha(5) and fibronectin is required for metastasis of B16F10 melanoma cells. Biochem Biophys Res Commun 333(4):1269–1275 100. Fridman R et al (1995) Activation of progelatinase B (MMP-9) by gelatinase A (MMP-2). Cancer Res 55(12):2548–2555 101. Ramos-DeSimone N et al (1999) Activation of matrix metalloproteinase-9 (MMP-9) via a converging plasmin/stromelysin-1 cascade enhances tumor cell invasion. J Biol Chem 274(19): 13066–13076 102. Sato H et al (1994) A matrix metalloproteinase expressed on the surface of invasive tumour cells. Nature 370(6484):61–65 103. Dufour A et al (2008) Role of the hemopexin domain of matrix metalloproteinases in cell migration. J Cell Physiol 217(3):643–651 104. Olson MW et al (2000) Characterization of the monomeric and dimeric forms of latent and active matrix metalloproteinase-9. Differential rates for activation by stromelysin 1. J Biol Chem 275(4):2661–2668 105. Duong TD, Erickson CA (2004) MMP-2 plays an essential role in producing epithelialmesenchymal transformations in the avian embryo. Dev Dyn 229(1):42–53 106. Cai DH, Brauer PR (2002) Synthetic matrix metalloproteinase inhibitor decreases early cardiac neural crest migration in chicken embryos. Dev Dyn 224(4):441–449 107. Giambernardi TA et al (2001) Neutrophil collagenase (MMP-8) is expressed during early development in neural crest cells as well as in adult melanoma cells. Matrix Biol 20(8):577–587 108. Alexander SM et al (1997) Spatial and temporal expression of the 72-kDa type IV collagenase (MMP-2) correlates with development and differentiation of valves in the embryonic avian heart. Dev Dyn 209(3):261–268 109. Song W, Jackson K, McGuire PG (2000) Degradation of type IV collagen by matrix metalloproteinases is an important step in the epithelial-mesenchymal transformation of the endocardial cushions. Dev Biol 227(2):606–617 110. Buisson AC et al (1996) Gelatinase B is involved in the in vitro wound repair of human respiratory epithelium. J Cell Physiol 166(2):413–426 111. Legrand C et al (1999) Airway epithelial cell migration dynamics. MMP-9 role in cellextracellular matrix remodeling. J Cell Biol 146(2):517–529 112. Buisson AC et al (1996) Wound repair-induced expression of a stromelysins is associated with the acquisition of a mesenchymal phenotype in human respiratory epithelial cells. Lab Invest 74(3):658–669 113. Schmalfeldt B et al (2001) Increased expression of matrix metalloproteinases (MMP)-2, MMP-9, and the urokinase-type plasminogen activator is associated with progression from benign to advanced ovarian cancer. Clin Cancer Res 7(8):2396–2404 114. Sier CF et al (1996) Tissue levels of matrix metalloproteinases MMP-2 and MMP-9 are related to the overall survival of patients with gastric carcinoma. Br J Cancer 74(3): 413–417 115. Liabakk NB et al (1996) Matrix metalloprotease 2 (MMP-2) and matrix metalloprotease 9 (MMP-9) type IV collagenases in colorectal cancer. Cancer Res 56(1):190–196 116. Morgia G et al (2005) Matrix metalloproteinases as diagnostic (MMP-13) and prognostic (MMP-2, MMP-9) markers of prostate cancer. Urol Res 33(1):44–50 117. Jones JL, Glynn P, Walker RA (1999) Expression of MMP-2 and MMP-9, their inhibitors, and the activator MT1-MMP in primary breast carcinomas. J Pathol 189(2):161–168 118. Hotary K et al (2000) Regulation of cell invasion and morphogenesis in a three-dimensional type I collagen matrix by membrane-type matrix metalloproteinases 1, 2, and 3. J Cell Biol 149(6):1309–1323 119. Bonfil RD et al (2007) Prostate cancer-associated membrane type 1-matrix metalloproteinase: a pivotal role in bone response and intraosseous tumor growth. Am J Pathol 170(6):2100–2111 120. Gavrilovic J et al (1990) Expression of transfected transforming growth factor alpha induces a motile fibroblast-like phenotype with extracellular matrix-degrading potential in a rat bladder carcinoma cell line. Cell Regul 1(13):1003–1014

2 Reactivation of Epithelial-Mesenchymal Transition in Invasive and Metastatic Cancer

55

121. Pulyaeva H et al (1997) MT1-MMP correlates with MMP-2 activation potential seen after epithelial to mesenchymal transition in human breast carcinoma cells. Clin Exp Metastasis 15(2):111–120 122. Daja MM et al (2003) Characterization of expression of matrix metalloproteinases and tissue inhibitors of metalloproteinases in prostate cancer cell lines. Prostate Cancer Prostatic Dis 6(1):15–26 123. Zajchowski DA et al (2001) Identification of gene expression profiles that predict the aggressive behavior of breast cancer cells. Cancer Res 61(13):5168–5178 124. Bhowmick NA et al (2001) Integrin beta 1 signaling is necessary for transforming growth factor-beta activation of p38MAPK and epithelial plasticity. J Biol Chem 276(50): 46707–46713 125. Galliher AJ, Schiemann WP (2006) Beta3 integrin and Src facilitate transforming growth factor-beta mediated induction of epithelial-mesenchymal transition in mammary epithelial cells. Breast Cancer Res 8(4):R42 126. Sternlicht MD et al (1999) The stromal proteinase MMP3/stromelysin-1 promotes mammary carcinogenesis. Cell 98(2):137–146 127. Lochter A et al (1997) Matrix metalloproteinase stromelysin-1 triggers a cascade of molecular alterations that leads to stable epithelial-to-mesenchymal conversion and a premalignant phenotype in mammary epithelial cells. J Cell Biol 139(7):1861–1872 128. Cowden Dahl KD et al (2008) Matrix metalloproteinase 9 is a mediator of epidermal growth factor-dependent e-cadherin loss in ovarian carcinoma cells. Cancer Res 68(12): 4606–4613 129. Cao J et al (2008) Membrane type 1 matrix metalloproteinase induces epithelial-tomesenchymal transition in prostate cancer. J Biol Chem 283(10):6232–6240 130. Trudel D et al (2003) Significance of MMP-2 expression in prostate cancer: an immunohistochemical study. Cancer Res 63(23):8511–8515 131. Millan FA et al (1991) Embryonic gene expression patterns of TGF beta 1, beta 2 and beta 3 suggest different developmental functions in vivo. Development 111(1):131–143 132. Hayashi H et al (1997) The MAD-related protein Smad7 associates with the TGFbeta receptor and functions as an antagonist of TGFbeta signaling. Cell 89(7):1165–1173 133. Kavsak P et al (2000) Smad7 binds to Smurf2 to form an E3 ubiquitin ligase that targets the TGF beta receptor for degradation. Mol Cell 6(6):1365–1375 134. Gingery A et al (2008) TGF-beta coordinately activates TAK1/MEK/AKT/NFkB and SMAD pathways to promote osteoclast survival. Exp Cell Res 314(15):2725–2738 135. Kato M et al (2009) TGF-beta activates Akt kinase through a microRNA-dependent amplifying circuit targeting PTEN. Nat Cell Biol 11(7):881–889 136. Lee MK et al (2007) TGF-beta activates Erk MAP kinase signalling through direct phosphorylation of ShcA. EMBO J 26(17):3957–3967 137. Perlman R et al (2001) TGF-beta-induced apoptosis is mediated by the adapter protein Daxx that facilitates JNK activation. Nat Cell Biol 3(8):708–714 138. Conlon FL et al (1994) A primary requirement for nodal in the formation and maintenance of the primitive streak in the mouse. Development 120(7):1919–1928 139. Shah SB et al (1997) Misexpression of chick Vg1 in the marginal zone induces primitive streak formation. Development 124(24):5127–5138 140. Goldstein AM et al (2005) BMP signaling is necessary for neural crest cell migration and ganglion formation in the enteric nervous system. Mech Dev 122(6):821–833 141. Boyer AS et al (1999) TGFbeta2 and TGFbeta3 have separate and sequential activities during epithelial-mesenchymal cell transformation in the embryonic heart. Dev Biol 208(2):530–545 142. Sato M et al (2003) Targeted disruption of TGF-beta1/Smad3 signaling protects against renal tubulointerstitial fibrosis induced by unilateral ureteral obstruction. J Clin Invest 112(10): 1486–1494 143. Khalil N et al (1996) TGF-beta 1, but not TGF-beta 2 or TGF-beta 3, is differentially present in epithelial cells of advanced pulmonary fibrosis: an immunohistochemical study. Am J Respir Cell Mol Biol 14(2):131–138

56

M.D. Amatangelo and M.E. Stearns

144. Piek E et al (1999) TGF-(beta) type I receptor/ALK-5 and Smad proteins mediate epithelial to mesenchymal transdifferentiation in NMuMG breast epithelial cells. J Cell Sci 112 (Pt 24):4557–4568 145. Kowanetz M et al (2004) Id2 and Id3 define the potency of cell proliferation and differentiation responses to transforming growth factor beta and bone morphogenetic protein. Mol Cell Biol 24(10):4241–4254 146. Do TV et al (2008) Transforming growth factor-beta1, transforming growth factor-beta2, and transforming growth factor-beta3 enhance ovarian cancer metastatic potential by inducing a Smad3-dependent epithelial-to-mesenchymal transition. Mol Cancer Res 6(5):695–705 147. Eastham JA et al (1995) Transforming growth factor-beta 1: comparative immunohistochemical localization in human primary and metastatic prostate cancer. Lab Invest 73(5): 628–635 148. Krasagakis K et al (1998) Elevated plasma levels of transforming growth factor (TGF)-beta1 and TGF-beta2 in patients with disseminated malignant melanoma. Br J Cancer 77(9):1492–1494 149. Cardillo MR, Yap E, Castagna G (1997) Molecular genetic analysis of TGF beta1 in breast cancer. J Exp Clin Cancer Res 16(1):57–63 150. Langenskiold M et al (2008) Increased TGF-beta 1 protein expression in patients with advanced colorectal cancer. J Surg Oncol 97(5):409–415 151. Pardali K, Moustakas A (2007) Actions of TGF-beta as tumor suppressor and pro-metastatic factor in human cancer. Biochim Biophys Acta 1775(1):21–62 152. Janda E et al (2006) Raf plus TGFbeta-dependent EMT is initiated by endocytosis and lysosomal degradation of E-cadherin. Oncogene 25(54):7117–7130 153. Vogelmann R et al (2005) TGFbeta-induced downregulation of E-cadherin-based cell-cell adhesion depends on PI3-kinase and PTEN. J Cell Sci 118(Pt 20):4901–4912 154. Ozdamar B et al (2005) Regulation of the polarity protein Par6 by TGFbeta receptors controls epithelial cell plasticity. Science 307(5715):1603–1609 155. Oft M, Heider KH, Beug H (1998) TGFbeta signaling is necessary for carcinoma cell invasiveness and metastasis. Curr Biol 8(23):1243–1252 156. Hocevar BA, Brown TL, Howe PH (1999) TGF-beta induces fibronectin synthesis through a c-Jun N-terminal kinase-dependent, Smad4-independent pathway. EMBO J 18(5):1345–1356 157. Zapatka M et al (2007) Basement membrane component laminin-5 is a target of the tumor suppressor Smad4. Oncogene 26(10):1417–1427 158. Ignotz RA, Massague J (1986) Transforming growth factor-beta stimulates the expression of fibronectin and collagen and their incorporation into the extracellular matrix. J Biol Chem 261(9):4337–4345 159. Giannelli G et al (2005) Laminin-5 with transforming growth factor-beta1 induces epithelial to mesenchymal transition in hepatocellular carcinoma. Gastroenterology 129(5):1375–1383 160. Pyke C et al (1995) Laminin-5 is a marker of invading cancer cells in some human carcinomas and is coexpressed with the receptor for urokinase plasminogen activator in budding cancer cells in colon adenocarcinomas. Cancer Res 55(18):4132–4139 161. Pyke C et al (1994) The gamma 2 chain of kalinin/laminin 5 is preferentially expressed in invading malignant cells in human cancers. Am J Pathol 145(4):782–791 162. Koshikawa N et al (1999) Overexpression of laminin gamma2 chain monomer in invading gastric carcinoma cells. Cancer Res 59(21):5596–5601 163. Katayama M et al (2003) Laminin gamma2-chain fragment in the circulation: a prognostic indicator of epithelial tumor invasion. Cancer Res 63(1):222–229 164. Koshikawa N et al (2000) Role of cell surface metalloprotease MT1-MMP in epithelial cell migration over laminin-5. J Cell Biol 148(3):615–624 165. Giannelli G et al (1997) Induction of cell migration by matrix metalloprotease-2 cleavage of laminin-5. Science 277(5323):225–228 166. Kim ES, Kim MS, Moon A (2004) TGF-beta-induced upregulation of MMP-2 and MMP-9 depends on p38 MAPK, but not ERK signaling in MCF10A human breast epithelial cells. Int J Oncol 25(5):1375–1382

2 Reactivation of Epithelial-Mesenchymal Transition in Invasive and Metastatic Cancer

57

167. Padua D et al (2008) TGFbeta primes breast tumors for lung metastasis seeding through angiopoietin-like 4. Cell 133(1):66–77 168. Bardeesy N et al (2006) Smad4 is dispensable for normal pancreas development yet critical in progression and tumor biology of pancreas cancer. Genes Dev 20(22):3130–3146 169. Takano S et al (2007) Smad4 is essential for down-regulation of E-cadherin induced by TGFbeta in pancreatic cancer cell line PANC-1. J Biochem 141(3):345–351 170. Deckers M et al (2006) The tumor suppressor Smad4 is required for transforming growth factor beta-induced epithelial to mesenchymal transition and bone metastasis of breast cancer cells. Cancer Res 66(4):2202–2209 171. Brown KA, Pietenpol JA, Moses HL (2007) A tale of two proteins: differential roles and regulation of Smad2 and Smad3 in TGF-beta signaling. J Cell Biochem 101(1):9–33 172. Hoot KE et al (2008) Keratinocyte-specific Smad2 ablation results in increased epithelial-mesenchymal transition during skin cancer formation and progression. J Clin Invest 118(8):2722–2732 173. Ju W et al (2006) Deletion of Smad2 in mouse liver reveals novel functions in hepatocyte growth and differentiation. Mol Cell Biol 26(2):654–667 174. Saika S et al (2004) Smad3 signaling is required for epithelial-mesenchymal transition of lens epithelium after injury. Am J Pathol 164(2):651–663 175. Dunn NR et al (2004) Combinatorial activities of Smad2 and Smad3 regulate mesoderm formation and patterning in the mouse embryo. Development 131(8):1717–1728 176. Brown KA et al (2004) Induction by transforming growth factor-beta1 of epithelial to mesenchymal transition is a rare event in vitro. Breast Cancer Res 6(3):R215–R231 177. Salomon DS et al (1995) Epidermal growth factor-related peptides and their receptors in human malignancies. Crit Rev Oncol Hematol 19(3):183–232 178. Normanno N et al (2005) The ErbB receptors and their ligands in cancer: an overview. Curr Drug Targets 6(3):243–257 179. Price JT, Wilson HM, Haites NE (1996) Epidermal growth factor (EGF) increases the in vitro invasion, motility and adhesion interactions of the primary renal carcinoma cell line, A704. Eur J Cancer 32A(11):1977–1982 180. Ignatoski KM et al (2000) ERBB-2 overexpression confers PI 3¢ kinase-dependent invasion capacity on human mammary epithelial cells. Br J Cancer 82(3):666–674 181. Spano JP et al (2005) Impact of EGFR expression on colorectal cancer patient prognosis and survival. Ann Oncol 16(1):102–108 182. Normanno N et al (2006) Epidermal growth factor receptor (EGFR) signaling in cancer. Gene 366(1):2–16 183. Graus-Porta D et al (1997) ErbB-2, the preferred heterodimerization partner of all ErbB receptors, is a mediator of lateral signaling. EMBO J 16(7):1647–1655 184. Goishi K et al (2003) Inhibition of zebrafish epidermal growth factor receptor activity results in cardiovascular defects. Mech Dev 120(7):811–822 185. Iwamoto R, Mekada E (2006) ErbB and HB-EGF signaling in heart development and function. Cell Struct Funct 31(1):1–14 186. Santoro MM, Gaudino G (2005) Cellular and molecular facets of keratinocyte reepithelization during wound healing. Exp Cell Res 304(1):274–286 187. Lu Z et al (2003) Downregulation of caveolin-1 function by EGF leads to the loss of E-cadherin, increased transcriptional activity of beta-catenin, and enhanced tumor cell invasion. Cancer Cell 4(6):499–515 188. Ackland ML et al (2003) Epidermal growth factor-induced epithelio-mesenchymal transition in human breast carcinoma cells. Lab Invest 83(3):435–448 189. Yates CC et al (2007) Co-culturing human prostate carcinoma cells with hepatocytes leads to increased expression of E-cadherin. Br J Cancer 96(8):1246–1252 190. Aranda V et al (2006) Par6-aPKC uncouples ErbB2 induced disruption of polarized epithelial organization from proliferation control. Nat Cell Biol 8(11):1235–1245

58

M.D. Amatangelo and M.E. Stearns

191. Kheradmand F, Rishi K, Werb Z (2002) Signaling through the EGF receptor controls lung morphogenesis in part by regulating MT1-MMP-mediated activation of gelatinase A/MMP2. J Cell Sci 115(Pt 4):839–848 192. Reddy KB et al (1999) Mitogen-activated protein kinase (MAPK) regulates the expression of progelatinase B (MMP-9) in breast epithelial cells. Int J Cancer 82(2):268–273 193. Watabe T et al (1998) The Ets-1 and Ets-2 transcription factors activate the promoters for invasion-associated urokinase and collagenase genes in response to epidermal growth factor. Int J Cancer 77(1):128–137 194. Kim IY et al (2009) Overexpression of ErbB2 induces invasion of MCF10A human breast epithelial cells via MMP-9. Cancer Lett 275(2):227–233 195. DesMarais V et al (2004) Synergistic interaction between the Arp2/3 complex and cofilin drives stimulated lamellipod extension. J Cell Sci 117(Pt 16):3499–3510 196. Turner T et al (1996) EGF receptor signaling enhances in vivo invasiveness of DU-145 human prostate carcinoma cells. Clin Exp Metastasis 14(4):409–418 197. Hill K et al (2000) Specific requirement for the p85-p110alpha phosphatidylinositol 3-kinase during epidermal growth factor-stimulated actin nucleation in breast cancer cells. J Biol Chem 275(6):3741–3744 198. Kedrin D et al (2007) Cell motility and cytoskeletal regulation in invasion and metastasis. J Mammary Gland Biol Neoplasia 12(2–3):143–152 199. Mouneimne G et al (2004) Phospholipase C and cofilin are required for carcinoma cell directionality in response to EGF stimulation. J Cell Biol 166(5):697–708 200. Price JT et al (1999) Epidermal growth factor promotes MDA-MB-231 breast cancer cell migration through a phosphatidylinositol 3¢-kinase and phospholipase C-dependent mechanism. Cancer Res 59(21):5475–5478 201. Marshall CJ (1996) Ras effectors. Curr Opin Cell Biol 8(2):197–204 202. Oft M, Akhurst RJ, Balmain A (2002) Metastasis is driven by sequential elevation of H-ras and Smad2 levels. Nat Cell Biol 4(7):487–494 203. Xie L et al (2004) Activation of the Erk pathway is required for TGF-beta1-induced EMT in vitro. Neoplasia 6(5):603–610 204. Kim ES, Kim MS, Moon A (2005) Transforming growth factor (TGF)-beta in conjunction with H-ras activation promotes malignant progression of MCF10A breast epithelial cells. Cytokine 29(2):84–91 205. Horiguchi K et al (2009) Role of Ras signaling in the induction of snail by transforming growth factor-beta. J Biol Chem 284(1):245–253 206. Jechlinger M et al (2003) Expression profiling of epithelial plasticity in tumor progression. Oncogene 22(46):7155–7169 207. Seton-Rogers SE et al (2004) Cooperation of the ErbB2 receptor and transforming growth factor beta in induction of migration and invasion in mammary epithelial cells. Proc Natl Acad Sci USA 101(5):1257–1262 208. Zhau HE et al (2008) Epithelial to mesenchymal transition (EMT) in human prostate cancer: lessons learned from ARCaP model. Clin Exp Metastasis 25(6):601–610 209. Shin S et al (2010) ERK2 but not ERK1 induces epithelial-to-mesenchymal transformation via DEF motif-dependent signaling events. Mol Cell 38(1):114–127 210. Chuang CF, Ng SY (1994) Functional divergence of the MAP kinase pathway. ERK1 and ERK2 activate specific transcription factors. FEBS Lett 346(2–3):229–234 211. Skarpen E et al (2008) MEK1 and MEK2 regulate distinct functions by sorting ERK2 to different intracellular compartments. FASEB J 22(2):466–476 212. Janda E et al (2002) Ras and TGF[beta] cooperatively regulate epithelial cell plasticity and metastasis: dissection of Ras signaling pathways. J Cell Biol 156(2):299–313 213. Wyckoff J et al (2004) A paracrine loop between tumor cells and macrophages is required for tumor cell migration in mammary tumors. Cancer Res 64(19):7022–7029 214. Klein CA (2008) Cancer. The metastasis cascade. Science 321(5897):1785–1787 215. Maroun CR et al (1999) The Gab1 PH domain is required for localization of Gab1 at sites of cell-cell contact and epithelial morphogenesis downstream from the met receptor tyrosine kinase. Mol Cell Biol 19(3):1784–1799

2 Reactivation of Epithelial-Mesenchymal Transition in Invasive and Metastatic Cancer

59

216. Stoker M et al (1987) Scatter factor is a fibroblast-derived modulator of epithelial cell mobility. Nature 327(6119):239–242 217. Uehara Y, Kitamura N (1992) Expression of a human hepatocyte growth factor/scatter factor cDNA in MDCK epithelial cells influences cell morphology, motility, and anchorageindependent growth. J Cell Biol 117(4):889–894 218. Humphrey PA et al (1995) Hepatocyte growth factor and its receptor (c-MET) in prostatic carcinoma. Am J Pathol 147(2):386–396 219. Takeuchi H et al (2003) c-MET expression level in primary colon cancer: a predictor of tumor invasion and lymph node metastases. Clin Cancer Res 9(4):1480–1488 220. Ueki T et al (1997) Expression of hepatocyte growth factor and its receptor c-met protooncogene in hepatocellular carcinoma. Hepatology 25(4):862–866 221. Di Renzo MF et al (1994) Overexpression of the Met/HGF receptor in ovarian cancer. Int J Cancer 58(5):658–662 222. Beviglia L et al (1997) Expression of the c-Met/HGF receptor in human breast carcinoma: correlation with tumor progression. Int J Cancer 74(3):301–309 223. Kankuri E et al (2005) Induction of hepatocyte growth factor/scatter factor by fibroblast clustering directly promotes tumor cell invasiveness. Cancer Res 65(21):9914–9922 224. Royal I, Fournier TM, Park M (1997) Differential requirement of Grb2 and PI3-kinase in HGF/SF-induced cell motility and tubulogenesis. J Cell Physiol 173(2):196–201 225. Giordano S et al (1997) A point mutation in the MET oncogene abrogates metastasis without affecting transformation. Proc Natl Acad Sci USA 94(25):13868–13872 226. Pon YL et al (2008) p70 S6 kinase promotes epithelial to mesenchymal transition through snail induction in ovarian cancer cells. Cancer Res 68(16):6524–6532 227. Zhou HY, Wong AS (2006) Activation of p70S6K induces expression of matrix metalloproteinase 9 associated with hepatocyte growth factor-mediated invasion in human ovarian cancer cells. Endocrinology 147(5):2557–2566 228. Grille SJ et al (2003) The protein kinase Akt induces epithelial mesenchymal transition and promotes enhanced motility and invasiveness of squamous cell carcinoma lines. Cancer Res 63(9):2172–2178 229. Turner N, Grose R (2010) Fibroblast growth factor signalling: from development to cancer. Nat Rev Cancer 10(2):116–129 230. Colvin JS et al (1996) Skeletal overgrowth and deafness in mice lacking fibroblast growth factor receptor 3. Nat Genet 12(4):390–397 231. Ciruna B, Rossant J (2001) FGF signaling regulates mesoderm cell fate specification and morphogenetic movement at the primitive streak. Dev Cell 1(1):37–49 232. Darby IA, Hewitson TD (2007) Fibroblast differentiation in wound healing and fibrosis. Int Rev Cytol 257:143–179 233. Ornitz DM, Itoh N (2001) Fibroblast growth factors. Genome Biol 2(3):REVIEWS3005 234. Billottet C et al (2008) Modulation of several waves of gene expression during FGF-1 induced epithelial-mesenchymal transition of carcinoma cells. J Cell Biochem 104(3):826–839 235. Boyer B et al (1989) Rearrangements of desmosomal and cytoskeletal proteins during the transition from epithelial to fibroblastoid organization in cultured rat bladder carcinoma cells. J Cell Biol 109(4 Pt 1):1495–1509 236. Acevedo VD et al (2007) Inducible FGFR-1 activation leads to irreversible prostate adenocarcinoma and an epithelial-to-mesenchymal transition. Cancer Cell 12(6):559–571 237. Thiery JP (2003) Epithelial-mesenchymal transitions in development and pathologies. Curr Opin Cell Biol 15(6):740–746 238. Kemler R et al (2004) Stabilization of beta-catenin in the mouse zygote leads to premature epithelial-mesenchymal transition in the epiblast. Development 131(23):5817–5824 239. Gilles C et al (2003) Transactivation of vimentin by beta-catenin in human breast cancer cells. Cancer Res 63(10):2658–2664 240. Kim K, Lu Z, Hay ED (2002) Direct evidence for a role of beta-catenin/LEF-1 signaling pathway in induction of EMT. Cell Biol Int 26(5):463–476 241. Crawford HC et al (1999) The metalloproteinase matrilysin is a target of beta-catenin transactivation in intestinal tumors. Oncogene 18(18):2883–2891

60

M.D. Amatangelo and M.E. Stearns

242. Takahashi M et al (2002) Identification of membrane-type matrix metalloproteinase-1 as a target of the beta-catenin/Tcf4 complex in human colorectal cancers. Oncogene 21(38):5861–5867 243. Marchenko GN et al (2002) Promoter characterization of the novel human matrix metalloproteinase-26 gene: regulation by the T-cell factor-4 implies specific expression of the gene in cancer cells of epithelial origin. Biochem J 363(Pt 2):253–262 244. Jass JR et al (2003) APC mutation and tumour budding in colorectal cancer. J Clin Pathol 56(1):69–73 245. Weeraratna AT et al (2002) Wnt5a signaling directly affects cell motility and invasion of metastatic melanoma. Cancer Cell 1(3):279–288 246. Danilkovitch-Miagkova A et al (2001) Oncogenic mutants of RON and MET receptor tyrosine kinases cause activation of the beta-catenin pathway. Mol Cell Biol 21(17):5857–5868 247. Labbe E, Letamendia A, Attisano L (2000) Association of Smads with lymphoid enhancer binding factor 1/T cell-specific factor mediates cooperative signaling by the transforming growth factor-beta and wnt pathways. Proc Natl Acad Sci USA 97(15):8358–8363 248. Timmerman LA et al (2004) Notch promotes epithelial-mesenchymal transition during cardiac development and oncogenic transformation. Genes Dev 18(1):99–115 249. Endo Y, Osumi N, Wakamatsu Y (2002) Bimodal functions of Notch-mediated signaling are involved in neural crest formation during avian ectoderm development. Development 129(4):863–873 250. Shou J et al (2001) Dynamics of notch expression during murine prostate development and tumorigenesis. Cancer Res 61(19):7291–7297 251. Reedijk M et al (2005) High-level coexpression of JAG1 and NOTCH1 is observed in human breast cancer and is associated with poor overall survival. Cancer Res 65(18):8530–8537 252. Miyamoto Y et al (2003) Notch mediates TGF alpha-induced changes in epithelial differentiation during pancreatic tumorigenesis. Cancer Cell 3(6):565–576 253. Peinado H, Cano A (2008) A hypoxic twist in metastasis. Nat Cell Biol 10(3):253–254 254. Yan W et al (2009) PI3 kinase/Akt signaling mediates epithelial-mesenchymal transition in hypoxic hepatocellular carcinoma cells. Biochem Biophys Res Commun 382(3):631–636 255. Sahlgren C et al (2008) Notch signaling mediates hypoxia-induced tumor cell migration and invasion. Proc Natl Acad Sci USA 105(17):6392–6397 256. Kopan R, Ilagan MX (2009) The canonical Notch signaling pathway: unfolding the activation mechanism. Cell 137(2):216–233 257. Zavadil J et al (2004) Integration of TGF-beta/Smad and Jagged1/Notch signalling in epithelial-to-mesenchymal transition. EMBO J 23(5):1155–1165 258. Wicking C, Smyth I, Bale A (1999) The hedgehog signalling pathway in tumorigenesis and development. Oncogene 18(55):7844–7851 259. Lai K et al (2003) Sonic hedgehog regulates adult neural progenitor proliferation in vitro and in vivo. Nat Neurosci 6(1):21–27 260. Burglin T (2008) The Hedgehog protein family. Genome Biol 9(11):241 261. Jenkins D (2009) Hedgehog signalling: emerging evidence for non-canonical pathways. Cell Signal 21(7):1023–1034 262. Zhang XM, Ramalho-Santos M, McMahon AP (2001) Smoothened mutants reveal redundant roles for Shh and Ihh signaling including regulation of L/R asymmetry by the mouse node. Cell 105(6):781–792 263. Mullor JL et al (2001) Wnt signals are targets and mediators of Gli function. Curr Biol 11(10):769–773 264. Omenetti A et al (2008) Hedgehog signaling regulates epithelial-mesenchymal transition during biliary fibrosis in rodents and humans. J Clin Invest 118(10):3331–3342 265. Louro ID et al (2002) Comparative gene expression profile analysis of GLI and c-MYC in an epithelial model of malignant transformation. Cancer Res 62(20):5867–5873 266. Karhadkar SS et al (2004) Hedgehog signalling in prostate regeneration, neoplasia and metastasis. Nature 431(7009):707–712 267. Feldmann G et al (2007) Blockade of hedgehog signaling inhibits pancreatic cancer invasion and metastases: a new paradigm for combination therapy in solid cancers. Cancer Res 67(5):2187–2196

2 Reactivation of Epithelial-Mesenchymal Transition in Invasive and Metastatic Cancer

61

268. Wu WS (2006) The signaling mechanism of ROS in tumor progression. Cancer Metastasis Rev 25(4):695–705 269. Fischer AN et al (2007) PDGF essentially links TGF-beta signaling to nuclear beta-catenin accumulation in hepatocellular carcinoma progression. Oncogene 26(23):3395–3405 270. Barrallo-Gimeno A, Nieto MA (2005) The Snail genes as inducers of cell movement and survival: implications in development and cancer. Development 132(14):3151–3161 271. Bolos V et al (2003) The transcription factor Slug represses E-cadherin expression and induces epithelial to mesenchymal transitions: a comparison with Snail and E47 repressors. J Cell Sci 116(Pt 3):499–511 272. Cano A et al (2000) The transcription factor snail controls epithelial-mesenchymal transitions by repressing E-cadherin expression. Nat Cell Biol 2(2):76–83 273. Conacci-Sorrell M et al (2003) Autoregulation of E-cadherin expression by cadherin-cadherin interactions: the roles of beta-catenin signaling, Slug, and MAPK. J Cell Biol 163(4):847–857 274. Grotegut S et al (2006) Hepatocyte growth factor induces cell scattering through MAPK/ Egr-1-mediated upregulation of Snail. EMBO J 25(15):3534–3545 275. Peinado H, Quintanilla M, Cano A (2003) Transforming growth factor beta-1 induces snail transcription factor in epithelial cell lines: mechanisms for epithelial mesenchymal transitions. J Biol Chem 278(23):21113–21123 276. Li X et al (2007) Gli1 acts through Snail and E-cadherin to promote nuclear signaling by beta-catenin. Oncogene 26(31):4489–4498 277. Kurrey NK, Amit K, Bapat SA (2005) Snail and Slug are major determinants of ovarian cancer invasiveness at the transcription level. Gynecol Oncol 97(1):155–165 278. Sakai D et al (2005) Regulation of Slug transcription in embryonic ectoderm by beta-cateninLef/Tcf and BMP-Smad signaling. Dev Growth Differ 47(7):471–482 279. Barbera MJ et al (2004) Regulation of Snail transcription during epithelial to mesenchymal transition of tumor cells. Oncogene 23(44):7345–7354 280. Peiro S et al (2006) Snail1 transcriptional repressor binds to its own promoter and controls its expression. Nucleic Acids Res 34(7):2077–2084 281. Yang Z et al (2005) Pak1 phosphorylation of snail, a master regulator of epithelial-tomesenchyme transition, modulates snail’s subcellular localization and functions. Cancer Res 65(8):3179–3184 282. Zhou BP et al (2004) Dual regulation of Snail by GSK-3beta-mediated phosphorylation in control of epithelial-mesenchymal transition. Nat Cell Biol 6(10):931–940 283. Vernon AE, LaBonne C (2006) Slug stability is dynamically regulated during neural crest development by the F-box protein Ppa. Development 133(17):3359–3370 284. Nieto MA (2002) The snail superfamily of zinc-finger transcription factors. Nat Rev Mol Cell Biol 3(3):155–166 285. Peinado H et al (2004) Snail mediates E-cadherin repression by the recruitment of the Sin3A/ histone deacetylase 1 (HDAC1)/HDAC2 complex. Mol Cell Biol 24(1):306–319 286. Peinado H, Olmeda D, Cano A (2007) Snail, Zeb and bHLH factors in tumour progression: an alliance against the epithelial phenotype? Nat Rev Cancer 7(6):415–428 287. Vincent T et al (2009) A SNAIL1-SMAD3/4 transcriptional repressor complex promotes TGF-beta mediated epithelial-mesenchymal transition. Nat Cell Biol 11(8):943–950 288. Shook D, Keller R (2003) Mechanisms, mechanics and function of epithelial-mesenchymal transitions in early development. Mech Dev 120(11):1351–1383 289. Vuoriluoto K et al (2011) Vimentin regulates EMT induction by Slug and oncogenic H-Ras and migration by governing Axl expression in breast cancer. Oncogene 30(12):1436–1448 290. Blanco MJ et al (2002) Correlation of Snail expression with histological grade and lymph node status in breast carcinomas. Oncogene 21(20):3241–3246 291. Batlle E et al (2000) The transcription factor snail is a repressor of E-cadherin gene expression in epithelial tumour cells. Nat Cell Biol 2(2):84–89 292. Come C et al (2006) Snail and slug play distinct roles during breast carcinoma progression. Clin Cancer Res 12(18):5395–5402

62

M.D. Amatangelo and M.E. Stearns

293. Elloul S et al (2005) Snail, Slug, and Smad-interacting protein 1 as novel parameters of disease aggressiveness in metastatic ovarian and breast carcinoma. Cancer 103(8):1631–1643 294. Roy HK et al (2005) The transcriptional repressor SNAIL is overexpressed in human colon cancer. Dig Dis Sci 50(1):42–46 295. Shioiri M et al (2006) Slug expression is an independent prognostic parameter for poor survival in colorectal carcinoma patients. Br J Cancer 94(12):1816–1822 296. Yokoyama K et al (2003) Increased invasion and matrix metalloproteinase-2 expression by Snail-induced mesenchymal transition in squamous cell carcinomas. Int J Oncol 22(4): 891–898 297. Jorda M et al (2005) Upregulation of MMP-9 in MDCK epithelial cell line in response to expression of the Snail transcription factor. J Cell Sci 118(Pt 15):3371–3385 298. Miyoshi A et al (2005) Snail accelerates cancer invasion by upregulating MMP expression and is associated with poor prognosis of hepatocellular carcinoma. Br J Cancer 92(2): 252–258 299. Ota I et al (2009) Induction of a MT1-MMP and MT2-MMP-dependent basement membrane transmigration program in cancer cells by Snail1. Proc Natl Acad Sci USA 106(48): 20318–20323 300. Sugimachi K et al (2003) Transcriptional repressor snail and progression of human hepatocellular carcinoma. Clin Cancer Res 9:2657–2664 301. Takeno S et al (2004) E-cadherin expression in patients with esophageal squamous cell carcinoma: promoter hypermethylation, Snail overexpression, and clinicopathologic implications. Am J Clin Pathol 122:78–84 302. Emadi Baygi M et al (2010) Slug/SNAI2 regulates cell proliferation and invasiveness of metastatic prostate cancer cell lines. Tumour Biol 31(4):297–307 303. Gupta PB et al (2005) The melanocyte differentiation program predisposes to metastasis after neoplastic transformation. Nat Genet 37(10):1047–1054 304. Comijn J et al (2001) The two-handed E box binding zinc finger protein SIP1 downregulates E-cadherin and induces invasion. Mol Cell 7(6):1267–1278 305. Eger A et al (2005) DeltaEF1 is a transcriptional repressor of E-cadherin and regulates epithelial plasticity in breast cancer cells. Oncogene 24(14):2375–2385 306. Verschueren K et al (1999) SIP1, a novel zinc finger/homeodomain repressor, interacts with Smad proteins and binds to 5¢-CACCT sequences in candidate target genes. J Biol Chem 274(29):20489–20498 307. Postigo AA (2003) Opposing functions of ZEB proteins in the regulation of the TGFbeta/ BMP signaling pathway. EMBO J 22(10):2443–2452 308. Costantino ME et al (2002) Cell-specific phosphorylation of Zfhep transcription factor. Biochem Biophys Res Commun 296(2):368–373 309. Long J, Zuo D, Park M (2005) Pc2-mediated sumoylation of Smad-interacting protein 1 attenuates transcriptional repression of E-cadherin. J Biol Chem 280(42):35477–35489 310. Krishnamachary B et al (2006) Hypoxia-inducible factor-1-dependent repression of E-cadherin in von Hippel-Lindau tumor suppressor-null renal cell carcinoma mediated by TCF3, ZFHX1A, and ZFHX1B. Cancer Res 66(5):2725–2731 311. Chua HL et al (2007) NF-kappaB represses E-cadherin expression and enhances epithelial to mesenchymal transition of mammary epithelial cells: potential involvement of ZEB-1 and ZEB-2. Oncogene 26(5):711–724 312. Shirakihara T, Saitoh M, Miyazono K (2007) Differential regulation of epithelial and mesenchymal markers by deltaEF1 proteins in epithelial mesenchymal transition induced by TGFbeta. Mol Biol Cell 18(9):3533–3544 313. Huber MA et al (2004) NF-kappaB is essential for epithelial-mesenchymal transition and metastasis in a model of breast cancer progression. J Clin Invest 114(4):569–581 314. Katoh M (2009) Integrative genomic analyses of ZEB2: transcriptional regulation of ZEB2 based on SMADs, ETS1, HIF1alpha, POU/OCT, and NF-kappaB. Int J Oncol 34(6): 1737–1742

2 Reactivation of Epithelial-Mesenchymal Transition in Invasive and Metastatic Cancer

63

315. Remacle JE et al (1999) New mode of DNA binding of multi-zinc finger transcription factors: deltaEF1 family members bind with two hands to two target sites. EMBO J 18(18):5073–5084 316. Shi Y et al (2003) Coordinated histone modifications mediated by a CtBP co-repressor complex. Nature 422(6933):735–738 317. Zhang Q et al (2006) Redox sensor CtBP mediates hypoxia-induced tumor cell migration. Proc Natl Acad Sci USA 103(24):9029–9033 318. Grooteclaes ML, Frisch SM (2000) Evidence for a function of CtBP in epithelial gene regulation and anoikis. Oncogene 19(33):3823–3828 319. Sanchez-Tillo E et al (2010) ZEB1 represses E-cadherin and induces an EMT by recruiting the SWI/SNF chromatin-remodeling protein BRG1. Oncogene 29(24):3490–3500 320. Vandewalle C et al (2005) SIP1/ZEB2 induces EMT by repressing genes of different epithelial cell-cell junctions. Nucleic Acids Res 33(20):6566–6578 321. Pena C et al (2006) The expression levels of the transcriptional regulators p300 and CtBP modulate the correlations between SNAIL, ZEB1, E-cadherin and vitamin D receptor in human colon carcinomas. Int J Cancer 119(9):2098–2104 322. Vandewalle C, Van Roy F, Berx G (2009) The role of the ZEB family of transcription factors in development and disease. Cell Mol Life Sci 66(5):773–787 323. Van de Putte T et al (2003) Mice lacking ZFHX1B, the gene that codes for Smad-interacting protein-1, reveal a role for multiple neural crest cell defects in the etiology of Hirschsprung disease-mental retardation syndrome. Am J Hum Genet 72(2):465–470 324. Imamichi Y et al (2007) Collagen type I-induced Smad-interacting protein 1 expression downregulates E-cadherin in pancreatic cancer. Oncogene 26(16):2381–2385 325. Haddad Y, Choi W, McConkey DJ (2009) Delta-crystallin enhancer binding factor 1 controls the epithelial to mesenchymal transition phenotype and resistance to the epidermal growth factor receptor inhibitor erlotinib in human head and neck squamous cell carcinoma lines. Clin Cancer Res 15(2):532–542 326. Taki M et al (2006) Involvement of Ets-1 transcription factor in inducing matrix metalloproteinase-2 expression by epithelial-mesenchymal transition in human squamous carcinoma cells. Int J Oncol 28(2):487–496 327. Guaita S et al (2002) Snail induction of epithelial to mesenchymal transition in tumor cells is accompanied by MUC1 repression and ZEB1 expression. J Biol Chem 277(42):39209–39216 328. Howe LR et al (2003) Twist is up-regulated in response to Wnt1 and inhibits mouse mammary cell differentiation. Cancer Res 63(8):1906–1913 329. Hornik C et al (2004) Twist is an integrator of SHH, FGF, and BMP signaling. Anat Embryol (Berl) 209(1):31–39 330. Sosic D et al (2003) Twist regulates cytokine gene expression through a negative feedback loop that represses NF-kappaB activity. Cell 112(2):169–180 331. Yang MH et al (2008) Direct regulation of TWIST by HIF-1alpha promotes metastasis. Nat Cell Biol 10(3):295–305 332. Benezra R et al (1990) The protein Id: a negative regulator of helix-loop-helix DNA binding proteins. Cell 61(1):49–59 333. Firulli BA et al (2005) Altered Twist1 and Hand2 dimerization is associated with SaethreChotzen syndrome and limb abnormalities. Nat Genet 37(4):373–381 334. Connerney J et al (2006) Twist1 dimer selection regulates cranial suture patterning and fusion. Dev Dyn 235(5):1345–1357 335. Ansieau S et al (2008) Induction of EMT by twist proteins as a collateral effect of tumorpromoting inactivation of premature senescence. Cancer Cell 14(1):79–89 336. Puisieux A, Valsesia-Wittmann S, Ansieau S (2006) A twist for survival and cancer progression. Br J Cancer 94(1):13–17 337. Lluis F et al (2005) E47 phosphorylation by p38 MAPK promotes MyoD/E47 association and muscle-specific gene transcription. EMBO J 24(5):974–984 338. Nie L et al (2003) Notch-induced E2A ubiquitination and degradation are controlled by MAP kinase activities. EMBO J 22(21):5780–5792

64

M.D. Amatangelo and M.E. Stearns

339. Firulli BA et al (2007) Mutations within helix I of Twist1 result in distinct limb defects and variation of DNA binding affinities. J Biol Chem 282(37):27536–27546 340. Firulli BA et al (2003) PKA, PKC, and the protein phosphatase 2A influence HAND factor function: a mechanism for tissue-specific transcriptional regulation. Mol Cell 12(5): 1225–1237 341. Vichalkovski A et al (2010) PKB/AKT phosphorylation of the transcription factor Twist-1 at Ser42 inhibits p53 activity in response to DNA damage. Oncogene 29(24):3554–3565 342. Massari ME et al (1999) A conserved motif present in a class of helix-loop-helix proteins activates transcription by direct recruitment of the SAGA complex. Mol Cell 4(1):63–73 343. Hamamori Y et al (1999) Regulation of histone acetyltransferases p300 and PCAF by the bHLH protein twist and adenoviral oncoprotein E1A. Cell 96(3):405–413 344. Koh HS et al (2009) Twist2 regulates CD7 expression and galectin-1-induced apoptosis in mature T-cells. Mol Cells 28(6):553–558 345. Hayashi M et al (2007) Comparative roles of Twist-1 and Id1 in transcriptional regulation by BMP signaling. J Cell Sci 120(Pt 8):1350–1357 346. Bechtel W, Zeisberg M (2009) Twist: a new link from hypoxia to fibrosis. Kidney Int 75(12):1255–1256 347. Thuault S et al (2006) Transforming growth factor-beta employs HMGA2 to elicit epithelialmesenchymal transition. J Cell Biol 174(2):175–183 348. Yu W, Kamara H, Svoboda KK (2008) The role of twist during palate development. Dev Dyn 237(10):2716–2725 349. Lee YH et al (2008) Fibulin-5 initiates epithelial-mesenchymal transition (EMT) and enhances EMT induced by TGF-beta in mammary epithelial cells via a MMP-dependent mechanism. Carcinogenesis 29(12):2243–2251 350. Pozharskaya V et al (2009) Twist: a regulator of epithelial-mesenchymal transition in lung fibrosis. PLoS One 4(10):e7559 351. Mironchik Y et al (2005) Twist overexpression induces in vivo angiogenesis and correlates with chromosomal instability in breast cancer. Cancer Res 65(23):10801–10809 352. Kwok WK et al (2005) Up-regulation of TWIST in prostate cancer and its implication as a therapeutic target. Cancer Res 65(12):5153–5162 353. Yuen HF et al (2007) Upregulation of Twist in oesophageal squamous cell carcinoma is associated with neoplastic transformation and distant metastasis. J Clin Pathol 60(5):510–514 354. Lee TK et al (2006) Twist overexpression correlates with hepatocellular carcinoma metastasis through induction of epithelial-mesenchymal transition. Clin Cancer Res 12(18):5369–5376 355. Yang J et al (2004) Twist, a master regulator of morphogenesis, plays an essential role in tumor metastasis. Cell 117(7):927–939 356. Whitehead J et al (2008) Mechanical factors activate beta-catenin-dependent oncogene expression in APC mouse colon. HFSP J 2(5):286–294 357. Alexander NR et al (2006) N-cadherin gene expression in prostate carcinoma is modulated by integrin-dependent nuclear translocation of Twist1. Cancer Res 66(7):3365–3369 358. Mikheeva SA et al (2010) TWIST1 promotes invasion through mesenchymal change in human glioblastoma. Mol Cancer 9:194 359. Reichenberger KJ et al (2005) Gene amplification is a mechanism of Six1 overexpression in breast cancer. Cancer Res 65(7):2668–2675 360. Young AP, Nagarajan R, Longmore GD (2003) Mechanisms of transcriptional regulation by Rb-E2F segregate by biological pathway. Oncogene 22(46):7209–7217 361. Ford HL et al (1998) Abrogation of the G2 cell cycle checkpoint associated with overexpression of HSIX1: a possible mechanism of breast carcinogenesis. Proc Natl Acad Sci USA 95(21):12608–12613 362. Christensen KL et al (2007) Cell cycle regulation of the human Six1 homeoprotein is mediated by APC(Cdh1). Oncogene 26(23):3406–3414 363. Brodbeck S, Besenbeck B, Englert C (2004) The transcription factor Six2 activates expression of the Gdnf gene as well as its own promoter. Mech Dev 121(10):1211–1222

2 Reactivation of Epithelial-Mesenchymal Transition in Invasive and Metastatic Cancer

65

364. Ando Z et al (2005) Slc12a2 is a direct target of two closely related homeobox proteins, Six1 and Six4. FEBS J 272(12):3026–3041 365. Spitz F et al (1998) Expression of myogenin during embryogenesis is controlled by Six/sine oculis homeoproteins through a conserved MEF3 binding site. Proc Natl Acad Sci USA 95(24):14220–14225 366. Xu PX et al (2003) Six1 is required for the early organogenesis of mammalian kidney. Development 130(14):3085–3094 367. Grifone R et al (2005) Six1 and Six4 homeoproteins are required for Pax3 and Mrf expression during myogenesis in the mouse embryo. Development 132(9):2235–2249 368. Self M et al (2006) Six2 is required for suppression of nephrogenesis and progenitor renewal in the developing kidney. EMBO J 25(21):5214–5228 369. Micalizzi DS et al (2009) The Six1 homeoprotein induces human mammary carcinoma cells to undergo epithelial-mesenchymal transition and metastasis in mice through increasing TGF-beta signaling. J Clin Invest 119(9):2678–2690 370. McCoy EL et al (2009) Six1 expands the mouse mammary epithelial stem/progenitor cell pool and induces mammary tumors that undergo epithelial-mesenchymal transition. J Clin Invest 119(9):2663–2677 371. Gross MK et al (2000) Lbx1 is required for muscle precursor migration along a lateral pathway into the limb. Development 127(2):413–424 372. Yu M et al (2009) A developmentally regulated inducer of EMT, LBX1, contributes to breast cancer progression. Genes Dev 23(15):1737–1742 373. Satoh K, Ginsburg E, Vonderhaar BK (2004) Msx-1 and Msx-2 in mammary gland development. J Mammary Gland Biol Neoplasia 9(2):195–205 374. Ramos C, Robert B (2005) msh/Msx gene family in neural development. Trends Genet 21(11):624–632 375. di Bari MG et al (2009) Msx2 induces epithelial-mesenchymal transition in mouse mammary epithelial cells through upregulation of Cripto-1. J Cell Physiol 219(3):659–666 376. Satoh K et al (2008) Up-regulation of MSX2 enhances the malignant phenotype and is associated with twist 1 expression in human pancreatic cancer cells. Am J Pathol 172(4):926–939 377. Wechselberger C et al (2001) Cripto-1 enhances migration and branching morphogenesis of mouse mammary epithelial cells. Exp Cell Res 266(1):95–105 378. Strizzi L et al (2004) Epithelial mesenchymal transition is a characteristic of hyperplasias and tumors in mammary gland from MMTV-Cripto-1 transgenic mice. J Cell Physiol 201(2):266–276 379. Blum M et al (1992) Gastrulation in the mouse: the role of the homeobox gene goosecoid. Cell 69(7):1097–1106 380. Niehrs C et al (1993) The homeobox gene goosecoid controls cell migration in Xenopus embryos. Cell 72(4):491–503 381. Niehrs C, Steinbeisser H, De Robertis EM (1994) Mesodermal patterning by a gradient of the vertebrate homeobox gene goosecoid. Science 263(5148):817–820 382. Hartwell KA et al (2006) The Spemann organizer gene, Goosecoid, promotes tumor metastasis. Proc Natl Acad Sci USA 103(50):18969–18974 383. Taube JH et al (2010) Core epithelial-to-mesenchymal transition interactome gene-expression signature is associated with claudin-low and metaplastic breast cancer subtypes. Proc Natl Acad Sci USA 107(35):15449–15454 384. Rogalla P et al (1996) HMGI-C expression patterns in human tissues. Implications for the genesis of frequent mesenchymal tumors. Am J Pathol 149(3):775–779 385. Hirning-Folz U et al (1998) The expression pattern of the Hmgic gene during development. Genes Chromosomes Cancer 23(4):350–357 386. Monzen K et al (2008) A crucial role of a high mobility group protein HMGA2 in cardiogenesis. Nat Cell Biol 10(5):567–574 387. Rogalla P et al (1997) Expression of HMGI-C, a member of the high mobility group protein family, in a subset of breast cancers: relationship to histologic grade. Mol Carcinog 19(3):153–156

66

M.D. Amatangelo and M.E. Stearns

388. Abe N et al (2003) An increased high-mobility group A2 expression level is associated with malignant phenotype in pancreatic exocrine tissue. Br J Cancer 89(11):2104–2109 389. Meyer B et al (2007) HMGA2 overexpression in non-small cell lung cancer. Mol Carcinog 46(7):503–511 390. Miyazawa J et al (2004) Expression of mesenchyme-specific gene HMGA2 in squamous cell carcinomas of the oral cavity. Cancer Res 64(6):2024–2029 391. Thuault S et al (2008) HMGA2 and Smads co-regulate SNAIL1 expression during induction of epithelial-to-mesenchymal transition. J Biol Chem 283(48):33437–33446 392. Reeves R, Nissen MS (1990) The A.T-DNA-binding domain of mammalian high mobility group I chromosomal proteins. A novel peptide motif for recognizing DNA structure. J Biol Chem 265(15):8573–8582 393. Thanos D, Maniatis T (1992) The high mobility group protein HMG I(Y) is required for NF-kappa B-dependent virus induction of the human IFN-beta gene. Cell 71(5):777–789 394. Fedele M et al (2006) HMGA2 induces pituitary tumorigenesis by enhancing E2F1 activity. Cancer Cell 9(6):459–471 395. Mantovani F et al (1998) NF-kappaB mediated transcriptional activation is enhanced by the architectural factor HMGI-C. Nucleic Acids Res 26(6):1433–1439 396. Cleynen I et al (2007) HMGA2 regulates transcription of the Imp2 gene via an intronic regulatory element in cooperation with nuclear factor-kappaB. Mol Cancer Res 5(4):363–372 397. Hommura F et al (2004) HMG-I/Y is a c-Jun/activator protein-1 target gene and is necessary for c-Jun-induced anchorage-independent growth in Rat1a cells. Mol Cancer Res 2(5):305–314 398. Wood LJ et al (2000) HMG-I/Y, a new c-Myc target gene and potential oncogene. Mol Cell Biol 20(15):5490–5502 399. Cao R et al (2002) Role of histone H3 lysine 27 methylation in Polycomb-group silencing. Science 298(5595):1039–1043 400. Wang H et al (2004) Role of histone H2A ubiquitination in Polycomb silencing. Nature 431(7010):873–878 401. Sparmann A, van Lohuizen M (2006) Polycomb silencers control cell fate, development and cancer. Nat Rev Cancer 6(11):846–856 402. Boyer LA et al (2006) Polycomb complexes repress developmental regulators in murine embryonic stem cells. Nature 441(7091):349–353 403. Ezhkova E et al (2009) Ezh2 orchestrates gene expression for the stepwise differentiation of tissue-specific stem cells. Cell 136(6):1122–1135 404. van Lohuizen M et al (1991) Identification of cooperating oncogenes in E mu-myc transgenic mice by provirus tagging. Cell 65(5):737–752 405. Kim JH et al (2004) Overexpression of Bmi-1 oncoprotein correlates with axillary lymph node metastases in invasive ductal breast cancer. Breast 13(5):383–388 406. Yonemitsu Y et al (2009) Distinct expression of polycomb group proteins EZH2 and BMI1 in hepatocellular carcinoma. Hum Pathol 40(9):1304–1311 407. van Leenders GJ et al (2007) Polycomb-group oncogenes EZH2, BMI1, and RING1 are overexpressed in prostate cancer with adverse pathologic and clinical features. Eur Urol 52(2):455–463 408. He XT et al (2009) Association between Bmi1 and clinicopathological status of esophageal squamous cell carcinoma. World J Gastroenterol 15(19):2389–2394 409. Shafaroudi AM et al (2008) Overexpression of BMI1, a polycomb group repressor protein, in bladder tumors: a preliminary report. Urol J 5(2):99–105 410. Mihic-Probst D et al (2007) Consistent expression of the stem cell renewal factor BMI-1 in primary and metastatic melanoma. Int J Cancer 121(8):1764–1770 411. Hoenerhoff MJ et al (2009) BMI1 cooperates with H-RAS to induce an aggressive breast cancer phenotype with brain metastases. Oncogene 28(34):3022–3032 412. Datta S et al (2007) Bmi-1 cooperates with H-Ras to transform human mammary epithelial cells via dysregulation of multiple growth-regulatory pathways. Cancer Res 67(21):10286–10295 413. Yang MH et al (2010) Bmi1 is essential in Twist1-induced epithelial-mesenchymal transition. Nat Cell Biol 12(10):982–992

2 Reactivation of Epithelial-Mesenchymal Transition in Invasive and Metastatic Cancer

67

414. Song LB et al (2009) The polycomb group protein Bmi-1 represses the tumor suppressor PTEN and induces epithelial-mesenchymal transition in human nasopharyngeal epithelial cells. J Clin Invest 119(12):3626–3636 415. Su IH et al (2005) Polycomb group protein ezh2 controls actin polymerization and cell signaling. Cell 121(3):425–436 416. Cao Q et al (2008) Repression of E-cadherin by the polycomb group protein EZH2 in cancer. Oncogene 27(58):7274–7284 417. Min J et al (2010) An oncogene-tumor suppressor cascade drives metastatic prostate cancer by coordinately activating Ras and nuclear factor-kappaB. Nat Med 16(3):286–294 418. Peinado H et al (2004) Snail and E47 repressors of E-cadherin induce distinct invasive and angiogenic properties in vivo. J Cell Sci 117(Pt 13):2827–2839 419. Hugo HJ et al (2011) Defining the E-cadherin repressor interactome in epithelial-mesenchymal transition: the PMC42 model as a case study. Cells Tissues Organs 193(1–2):23–40 420. Weinberg RA (2008) Leaving home early: reexamination of the canonical models of tumor progression. Cancer Cell 14(4):283–284 421. Siomi H, Siomi MC (2010) Posttranscriptional regulation of microRNA biogenesis in animals. Mol Cell 38(3):323–332 422. Yanaihara N et al (2006) Unique microRNA molecular profiles in lung cancer diagnosis and prognosis. Cancer Cell 9(3):189–198 423. Iorio MV et al (2005) MicroRNA gene expression deregulation in human breast cancer. Cancer Res 65(16):7065–7070 424. Lee EJ et al (2007) Expression profiling identifies microRNA signature in pancreatic cancer. Int J Cancer 120(5):1046–1054 425. Ambs S et al (2008) Genomic profiling of microRNA and messenger RNA reveals deregulated microRNA expression in prostate cancer. Cancer Res 68(15):6162–6170 426. Lotterman CD, Kent OA, Mendell JT (2008) Functional integration of microRNAs into oncogenic and tumor suppressor pathways. Cell Cycle 7(16):2493–2499 427. Lu J et al (2005) MicroRNA expression profiles classify human cancers. Nature 435(7043): 834–838 428. Bartel DP (2009) MicroRNAs: target recognition and regulatory functions. Cell 136(2): 215–233 429. Kim VN, Han J, Siomi MC (2009) Biogenesis of small RNAs in animals. Nat Rev Mol Cell Biol 10(2):126–139 430. Gregory RI et al (2004) The Microprocessor complex mediates the genesis of microRNAs. Nature 432(7014):235–240 431. Yi R et al (2003) Exportin-5 mediates the nuclear export of pre-microRNAs and short hairpin RNAs. Genes Dev 17(24):3011–3016 432. Hammond SM et al (2001) Argonaute2, a link between genetic and biochemical analyses of RNAi. Science 293(5532):1146–1150 433. Calin GA et al (2004) Human microRNA genes are frequently located at fragile sites and genomic regions involved in cancers. Proc Natl Acad Sci USA 101(9):2999–3004 434. Krol J, Loedige I, Filipowicz W (2010) The widespread regulation of microRNA biogenesis, function and decay. Nat Rev Genet 11(9):597–610 435. Kumar MS et al (2009) Dicer1 functions as a haploinsufficient tumor suppressor. Genes Dev 23(23):2700–2704 436. Lambertz I et al (2010) Monoallelic but not biallelic loss of Dicer1 promotes tumorigenesis in vivo. Cell Death Differ 17(4):633–641 437. Kumar MS et al (2007) Impaired microRNA processing enhances cellular transformation and tumorigenesis. Nat Genet 39(5):673–677 438. Paroo Z et al (2009) Phosphorylation of the human microRNA-generating complex mediates MAPK/Erk signaling. Cell 139(1):112–122 439. Davis BN et al (2008) SMAD proteins control DROSHA-mediated microRNA maturation. Nature 454(7200):56–61 440. Viswanathan SR, Daley GQ, Gregory RI (2008) Selective blockade of microRNA processing by Lin28. Science 320(5872):97–100

68

M.D. Amatangelo and M.E. Stearns

441. Newman MA, Thomson JM, Hammond SM (2008) Lin-28 interaction with the Let-7 precursor loop mediates regulated microRNA processing. RNA 14(8):1539–1549 442. Wang M et al (2011) Role of ribosomal protein RPS2 in controlling let-7a expression in human prostate cancer. Mol Cancer Res 9(1):36–50 443. Heo I et al (2009) TUT4 in concert with Lin28 suppresses microRNA biogenesis through pre-microRNA uridylation. Cell 138(4):696–708 444. Heo I et al (2008) Lin28 mediates the terminal uridylation of let-7 precursor MicroRNA. Mol Cell 32(2):276–284 445. Wang YC et al (2010) Lin-28B expression promotes transformation and invasion in human hepatocellular carcinoma. Carcinogenesis 31(9):1516–1522 446. Chang TC et al (2008) Widespread microRNA repression by Myc contributes to tumorigenesis. Nat Genet 40(1):43–50 447. Mu P et al (2009) Genetic dissection of the miR-17 92 cluster of microRNAs in Myc-induced B-cell lymphomas. Genes Dev 23(24):2806–2811 448. Sampson VB et al (2007) MicroRNA let-7a down-regulates MYC and reverts MYC-induced growth in Burkitt lymphoma cells. Cancer Res 67(20):9762–9770 449. Park SM et al (2008) The miR-200 family determines the epithelial phenotype of cancer cells by targeting the E-cadherin repressors ZEB1 and ZEB2. Genes Dev 22(7):894–907 450. Korpal M et al (2008) The miR-200 family inhibits epithelial-mesenchymal transition and cancer cell migration by direct targeting of E-cadherin transcriptional repressors ZEB1 and ZEB2. J Biol Chem 283(22):14910–14914 451. Gregory PA et al (2008) The miR-200 family and miR-205 regulate epithelial to mesenchymal transition by targeting ZEB1 and SIP1. Nat Cell Biol 10(5):593–601 452. Bracken CP et al (2008) A double-negative feedback loop between ZEB1-SIP1 and the microRNA-200 family regulates epithelial-mesenchymal transition. Cancer Res 68(19): 7846–7854 453. Xia H et al (2010) miR-200a-mediated downregulation of ZEB2 and CTNNB1 differentially inhibits nasopharyngeal carcinoma cell growth, migration and invasion. Biochem Biophys Res Commun 391(1):535–541 454. Burk U et al (2008) A reciprocal repression between ZEB1 and members of the miR-200 family promotes EMT and invasion in cancer cells. EMBO Rep 9(6):582–589 455. Shimono Y et al (2009) Downregulation of miRNA-200c links breast cancer stem cells with normal stem cells. Cell 138(3):592–603 456. Cochrane DR et al (2009) MicroRNA-200c mitigates invasiveness and restores sensitivity to microtubule-targeting chemotherapeutic agents. Mol Cancer Ther 8(5):1055–1066 457. Olson P et al (2009) MicroRNA dynamics in the stages of tumorigenesis correlate with hallmark capabilities of cancer. Genes Dev 23(18):2152–2165 458. Gibbons DL et al (2009) Contextual extracellular cues promote tumor cell EMT and metastasis by regulating miR-200 family expression. Genes Dev 23(18):2140–2151 459. Sossey-Alaoui K, Bialkowska K, Plow EF (2009) The miR200 family of microRNAs regulates WAVE3-dependent cancer cell invasion. J Biol Chem 284(48):33019–33029 460. Dykxhoorn DM et al (2009) miR-200 enhances mouse breast cancer cell colonization to form distant metastases. PLoS One 4(9):e7181 461. Pasquinelli AE et al (2000) Conservation of the sequence and temporal expression of let-7 heterochronic regulatory RNA. Nature 408(6808):86–89 462. Sempere LF et al (2004) Expression profiling of mammalian microRNAs uncovers a subset of brain-expressed microRNAs with possible roles in murine and human neuronal differentiation. Genome Biol 5(3):R13 463. Lagos-Quintana M et al (2001) Identification of novel genes coding for small expressed RNAs. Science 294(5543):853–858 464. Grosshans H et al (2005) The temporal patterning microRNA let-7 regulates several transcription factors at the larval to adult transition in C. elegans. Dev Cell 8(3):321–330 465. Nam EJ et al (2008) MicroRNA expression profiles in serous ovarian carcinoma. Clin Cancer Res 14(9):2690–2695

2 Reactivation of Epithelial-Mesenchymal Transition in Invasive and Metastatic Cancer

69

466. Motoyama K et al (2008) Clinical significance of high mobility group A2 in human gastric cancer and its relationship to let-7 microRNA family. Clin Cancer Res 14(8):2334–2340 467. Schultz J et al (2008) MicroRNA let-7b targets important cell cycle molecules in malignant melanoma cells and interferes with anchorage-independent growth. Cell Res 18(5):549–557 468. Ozen M et al (2008) Widespread deregulation of microRNA expression in human prostate cancer. Oncogene 27(12):1788–1793 469. Kumar MS et al (2008) Suppression of non-small cell lung tumor development by the let-7 microRNA family. Proc Natl Acad Sci USA 105(10):3903–3908 470. Johnson SM et al (2005) RAS is regulated by the let-7 microRNA family. Cell 120(5):635–647 471. Lee YS, Dutta A (2007) The tumor suppressor microRNA let-7 represses the HMGA2 oncogene. Genes Dev 21(9):1025–1030 472. Mayr C, Hemann MT, Bartel DP (2007) Disrupting the pairing between let-7 and Hmga2 enhances oncogenic transformation. Science 315(5818):1576–1579 473. Shell S et al (2007) Let-7 expression defines two differentiation stages of cancer. Proc Natl Acad Sci USA 104(27):11400–11405 474. Watanabe S et al (2009) HMGA2 maintains oncogenic RAS-induced epithelial-mesenchymal transition in human pancreatic cancer cells. Am J Pathol 174(3):854–868 475. Dangi-Garimella S et al (2009) Raf kinase inhibitory protein suppresses a metastasis signalling cascade involving LIN28 and let-7. EMBO J 28(4):347–358 476. Zhu J, Blenis J, Yuan J (2008) Activation of PI3K/Akt and MAPK pathways regulates Myc-mediated transcription by phosphorylating and promoting the degradation of Mad1. Proc Natl Acad Sci USA 105(18):6584–6589 477. Nakagawa M et al (2008) Generation of induced pluripotent stem cells without Myc from mouse and human fibroblasts. Nat Biotechnol 26(1):101–106 478. Takahashi K et al (2007) Induction of pluripotent stem cells from adult human fibroblasts by defined factors. Cell 131(5):861–872 479. Morel AP et al (2008) Generation of breast cancer stem cells through epithelial-mesenchymal transition. PLoS One 3(8):e2888 480. Yu F et al (2007) let-7 regulates self renewal and tumorigenicity of breast cancer cells. Cell 131(6):1109–1123 481. Kong W et al (2008) MicroRNA-155 is regulated by the transforming growth factor beta/ Smad pathway and contributes to epithelial cell plasticity by targeting RhoA. Mol Cell Biol 28(22):6773–6784 482. Gebeshuber CA, Zatloukal K, Martinez J (2009) miR-29a suppresses tristetraprolin, which is a regulator of epithelial polarity and metastasis. EMBO Rep 10(4):400–405 483. Asangani IA et al (2008) MicroRNA-21 (miR-21) post-transcriptionally downregulates tumor suppressor Pdcd4 and stimulates invasion, intravasation and metastasis in colorectal cancer. Oncogene 27(15):2128–2136 484. Lu Z et al (2008) MicroRNA-21 promotes cell transformation by targeting the programmed cell death 4 gene. Oncogene 27(31):4373–4379 485. Zhu S et al (2008) MicroRNA-21 targets tumor suppressor genes in invasion and metastasis. Cell Res 18(3):350–359 486. Du J et al (2009) BMP-6 inhibits microRNA-21 expression in breast cancer through repressing deltaEF1 and AP-1. Cell Res 19(4):487–496 487. Ma L, Teruya-Feldstein J, Weinberg RA (2007) Tumour invasion and metastasis initiated by microRNA-10b in breast cancer. Nature 449(7163):682–688 488. Ma L et al (2010) miR-9, a MYC/MYCN-activated microRNA, regulates E-cadherin and cancer metastasis. Nat Cell Biol 12(3):247–256 489. Baffa R et al (2009) MicroRNA expression profiling of human metastatic cancers identifies cancer gene targets. J Pathol 219(2):214–221 490. Gee HE et al (2008) MicroRNA-10b and breast cancer metastasis. Nature 455(7216):E8–E9; author reply E9

Chapter 3

Biomechanical ECM Switches and Tumor Metastasis Jacquelyn J. Ames, Calvin P.H. Vary, and Peter C. Brooks

Abstract Data is rapidly emerging that physical or biomechanical changes in extracellular matrix molecules (ECM) may allow differential control of cell signaling pathways and therefore, these structural alterations may actively regulate diverse cell types that comprise the tumor stroma. These mechanical changes in the structure of ECM molecules may trigger the exposure of cryptic ECM elements that help create a local microenvironment that facilitates initiation and progression of inflammation, tumor growth and metastasis. To this end, we will highlight the accumulating experimental evidence for the existence of biomechanical ECM switches and discuss how the selective triggering of these switches may be used by diverse cell types to promote malignant tumor progression. Finally, we will discuss scientific and clinical advancements that have lead to the recent translation of these exciting biological concepts into human clinical trials for cancer imaging and therapy.

3.1

Introduction

Communication links between distinct populations of cells and between cells and the surrounding solid-state extracellular matrix (ECM) molecules are among the most fundamental requirements for essentially all biological processes. Implicit within this general concept is the importance of a metered flow of mechanical, biochemical and molecular information that can be precisely controlled to provide a selective means to impart specificity to such divergent biological processes as embryonic development and tumor progression. Bi-directional communication

J.J. Ames • C.P.H. Vary • P.C. Brooks (*) Maine Medical Center Research Institute, Center for Molecular Medicine, 81 Research Drive, Scarborough, ME 04074, USA e-mail: [email protected]

A. Fatatis (ed.), Signaling Pathways and Molecular Mediators in Metastasis, DOI 10.1007/978-94-007-2558-4_3, © Springer Science+Business Media B.V. 2012

71

72

J.J. Ames et al.

Fig. 3.1 Triggering biomechanical ECM switches. Accumulating evidence suggest that the local stromal microenvironment of a tumor plays critical roles in regulating tumor growth and metastasis. Localized structural changes within individual solid-state ECM molecules may represent “Biomechanical Switches”. The selective triggering of these biomechanical switches may provide a mechanism to fine tune cellular responses to mechanical changes within the tumor microenvironment

circuits are known to exist between cells and the ECM and these circuits can be connected by cell surface receptors such as integrins [1–3]. Interestingly, integrin-ECM interactions have been shown to regulate a wide array of cellular and molecular processes ranging from mechanical adhesion, motility, proliferation, survival and differentiation, to the regulation of gene expression and biochemical signaling [1–3]. Given the broad functional importance of these communication links, it is not be surprising that dysregulation of these information highways might result in profound changes in homeostasis, which in turn might govern pathological events such as tumor growth and metastasis. In some respects, the flow of biological information between cells and the ECM might be viewed in a similar manner to that of the flow of electricity within a network of circuits regulated by switches. It is well-accepted that the interconnected network of ECM molecules that compose the non-cellular microenvironment of a particular tissue is dynamic, with transient alterations occurring in the molecular composition and structural integrity [4, 5]. The dynamic nature of the ECM may provide a tissue specific mode for cells to respond in appropriate ways to transient changes resulting from external stimuli. Viewed from this perspective, structural or mechanical alterations in the integrity of individual ECM molecules might function as biomechanical switches, with the capacity to initiate, stop or divert the flow of biological information in different pathways when a specific need arises (Fig. 3.1). However, if problems occur in specific switches within this network of circuits, the resulting aberrant information flow may result in unintended or detrimental consequences. Within this context, we will discuss the diverse roles by which structural alterations in ECM molecules may function as biomechanical switches and examine the potential of these switches to contribute to tumor growth and metastasis.

3

Biomechanical ECM Switches and Tumor Metastasis

3.2

73

Tumor Cell Metastasis

It has been appreciated for some time that metastasis or the spread of tumor cells from the primary tumor mass to distant sites is a major obstacle to the successful management of malignant tumors. In fact, metastasis has been implicated as a leading cause for the majority of deaths associated with diverse forms of malignant tumors. Metastasis has often been described as a highly inefficient process due in large part to the requirement of tumor cells to complete all the interconnected steps within the metastatic cascade to successfully establish secondary lesions [6, 7]. Due to the physical distances and numerous cellular and non-cellular components that tumor cells may encounter during their journey, a wide array of factors influence the capacity of tumor cells to successfully establish secondary metastatic foci. Given these well-established caveats, it is not surprising that effective therapeutic strategies to manage metastatic disease have been slow to emerge. In general, the metastatic cascade (Fig. 3.2) can be thought of as an interdependent set of steps that facilitates the capacity of tumor cells to spread from the primary tumor to multiple sites throughout the body [6, 7]. These complex cellular events include alterations in cell-cell interactions via changes in expression and activity of cell-cell adhesion molecules such as E-Cadherin [8]. This overall reduction in cellcell adhesion can be accompanied by a multitude of additional changes, including increased expression of matrix altering enzymes as well as alterations in the expression and activation of numerous cell surface receptors that mediate interactions with soluble factors, other cells, and with the ECM [9, 10]. These changes and many more can lead to localized structural remodeling within the interconnected network of ECM molecules. Evidence suggests that multiple cell types within the tumor microenvironment including stromal fibroblasts and inflammatory infiltrates such as mast cells, neutophils and macrophages secrete an array of ECM altering enzymes including members of the serine, cystiene, heparinase and matrix metalloproteinase (MMP) families [11]. Several of these proteases have been shown to localize to the cell surface of tumor and stromal cells via diverse mechanisms, some of which include interactions with cell surface receptors such as integrins [12, 13]. Localized ECM remodeling as a consequence of either secreted or cell associated enzymes, results in more than simply degradation of individual molecules to provide a less restrictive pathway for cell movement. For example, many serine proteases as well as MMPs have been shown to release bioactive ECM peptides, while enzymes such as lysyl oxidase (LOX) can modify ECM molecules resulting in altered cross-linking leading to changes in ECM biomechanical properties [14, 15]. Finally, a number of proteolytic enzymes have been shown to contribute to the exposure of functionally active cryptic or hidden structural ECM domains that exhibit unique functions [16–18]. Collectively, these changes may govern divergent processes ranging from growth factor localization and changes in mechanical stability and stiffness to modifying numerous cellular interactions which ultimately impact the ability of cells to move, proliferate, differentiate and survive within a particular tissue microenvironment.

74

J.J. Ames et al.

Fig. 3.2 Tumor cell metastasis. Tumor cell metastasis involves a complex interconnected series of cellular and molecular events. This complex process is governed by numerous distinct cell types that functionally interact with each other as well as with an array of diverse extracellular matrix (ECM) molecules. Localized changes in cell-cell and cell-ECM interactions along with structural alterations within ECM molecules facilitate cellular invasion and ultimately the formation of secondary metastatic foci at remote site from the primary tumor lesion

A growing body of evidence now shows that changes in tensional and contractile forces can generate increased stiffness and rigidity of the ECM and may actually promote malignant cellular behavior. For example, increased interstitial pressure from a rapidly growing tumor mass may enhance the physical forces on individual ECM molecules such a collagen and fibronectin [19, 20]. In this regard, it is

3

Biomechanical ECM Switches and Tumor Metastasis

75

known that many of the multi-domain ECM molecules that make up the tumor microenvironment can respond to these alterations in mechanical forces by changing their native conformation and altering other biomechanical properties including their contractility and rigidity [21, 22]. Changes in ECM rigidity have been shown to alter integrin function and activation resulting in changes in focal adhesion formation, re-organization of the cytoskeleton and enhanced activation of critical intracellular signaling pathways such as MAPK/Erk and Rho signaling, which act to control cellular migration and invasion [23, 24]. Therefore, it is not surprising that structural remodeling resulting from either mechanical forces or proteases, facilitates invasion and migration away from the primary tumor mass into the surrounding interstitial tissue. Recent evidence suggests that under some circumstances, initial tumor cell invasion during metastasis may be associated with epithelial to mesenchymal transition (EMT) like processes [25]. Studies have shown that molecular regulators of EMT including Snail and Twist may play roles in carcinoma cell invasion and metastasis. EMT within this context may be transient, occurring predominantly at the invasive front of tumors and governed, at least in part, by extrinsic factors from the surrounding tissue microenvironment [26]. The potential transient nature of carcinoma-associated EMT may explain in part, the conflicting data for EMT during metastases of tumors from distinct histological origins. As cellular invasion and migration continue, tumor cells may sense specific gradients of soluble factors that help direct them towards lymphatic and/or blood vessels. These soluble factors range from growth factors; cytokines and chemokines to small ECM derived peptides [27]. Experiments have suggested that protease generated fragments of collagen and elastin can function as cellular chemoattractants and some studies have demonstrated that certain PPG amino acid containing peptides bind chemokine receptors and stimulate directed cellular motility [28]. Interestingly, new work suggests that macrophages, which are well–known to respond to chemokines, may contribute to the re-orientation of tumor cells toward blood vessels to facilitate intravasation [29]. Supporting this notion, evidence suggests that macrophages are often present within the immediate vicinity of tumor cells invading the host circulation and moreover, that reduction in macrophages may correlate with reduced tumor intravasation [30, 31]. In order to gain entry into the circulation, tumor cells are thought to remodel the structural integrity of basement membranes. Malignant cells may use a combination of proteolytic enzymes such as MMPs, serine proteases and heparinases to cleave the major components of the basement membranes including collagen type-IV, laminin and proteolgycans such as perlecan [32]. These altered basement membranes, along with changes in the biomechanical properties of the tumor cells themselves, help facilitate cellular invasion. Once in the circulation, tumor cells must survive physical shear forces as well as evade host immune surveillance. It has been suggested that platelet-tumor cell interactions result in cellular aggregates that help protect and hide the circulating tumor cells from mechanical shear forces and immunological attack [33, 34]. Multiple mechanisms have been described by which tumor cells arrest within the vasculature

76

J.J. Ames et al.

and these may involve a combination of mechanical trapping and weak interactions between tumor cells and specific cell-cell adhesion molecules such as selectins that are expressed on the luminal side of the endothelium. In addition, it has been appreciated for some time that the endothelia of distinct organs may display a unique repertoire of surface molecules termed vascular addressins, which may contribute to the ability of tumor cells to preferentially arrest at tissue-specific sites [35, 36]. This concept is thought to contribute to organ-specific metastasis. Following initiation of these weak cellular interactions, more stable physical connections between integrins and the underlying ECM proteins have been suggested to occur. While the mechanisms by which tumor cells exit the vasculature, a process termed extravasation, are complex, new insight is rapidly adding to the overall understanding of this highly dynamic process. For example, while endothelial cell contraction resulting in exposure of basement membrane components is thought to occur at the site of extravasation, evidence also suggests that cells often exit the vasculature at unique sites that express low levels of matrix proteins called LERs (Low Expression Regions). These LER’s may allow more efficient proteolytic and biomechanical remodeling of the basement membrane to promote cellular extravasation [37, 38]. Moreover, leukocytes such as neutrophils may contribute to the structural modification at LERs and the formation of these vascular exit gates may allow easier egress routes for tumor cells during extravasation. Once the tumor cells have escaped the confines of the circulation and entered the microenvironment of a distant organ a new set of challenges emerge. Tumor cells must either migrate into a biocompatible area or establish their own local microenvironment that allows continued survival. In some circumstances, tumor cells have been shown to co-opt an existing blood vessel to provide a sufficient source of localized nutrients needed for growth and expansion, while in other cases tumor cells initiate structural and mechanical remodeling of ECM, releasing matrix-bound growth factors, generating pro-migratory ECM fragments and secreting soluble molecules that function to promote recruitment of endothelial cells and bone marrow progenitor. Several studies have suggested that prior to tumor cell arrival at sites distant from the primary lesion, subsets of bone marrow progenitor cells may be recruited to create a localized pre-metastatic niche [39, 40]. These pre-metastatic niches may be created in part by altering the composition and possibly the structural integrity of the ECM by mechanisms involving alterations in the levels of fibronectin and MMPs [39]. In addition, infiltration of immune cells such as mast cells and neutrophils may also help modify or condition the microenvironment to potentiate cellular proliferation and support angiogenesis or vasculogenesis [41, 42]. In this regard, heparin, a major component of mast cell granules, has been shown to bind fibronectin, facilitating a biomechanical change that allows exposure of normally cryptic vascular endothelial growth factor (VEGF) binding sites [43]. These intriguing observations provide an important example by which structural alterations of ECM molecules may act as biomechanical switches to help create a local tissue microenvironment that is conducive to angiogenesis and tumor cell survival. In addition, experimental evidence suggests that biomechanical contraction of tissues may also be employed to physically translocate existing blood vessels to areas of

3

Biomechanical ECM Switches and Tumor Metastasis

77

hypoxia [44]. Whether this mechanism occurs in the context of tumor development remains to be confirmed. Taken together, these examples help illustrate some of the ECM remodeling mechanisms that might be utilized to ensure cell survival at sites distant from the primary lesion.

3.3

Importance of the Dynamic Nature of the ECM Microenvironment in Tumor Metastasis

It is well established that many secreted factors such as matrix-modifying enzymes; growth factors, cytokines and chemokines all play distinct roles in regulating the metastatic cascade. However, in addition to these soluble components, it has also been known for decades that cellular interactions with solid-state ECM molecules such as collagen (Coll), laminin (LN) and fibronectin (FN) play crucial roles to governing numerous aspects of both normal and pathological cell behavior. Just as farmers often need to plow or mechanically disrupt the hard packed soil of their fields in order to prepare nutrient rich soil for their seeds to grow and survive, it is also likely that tumor cells need to mechanically remodel the ECM to create a nutrient rich microenvironment to ensure their growth and survival. In many respects, this general concept was proposed in the late 1800s by Dr. Paget and was termed the seed and soil hypothesis [45]. Once the ECM has been remodeled and the biological information within the ECM rendered readily accessible, tumor cells need a means to detect and assimilate this information for use inside the cell. To this end, integrin receptors provide a key component that allows cells the capacity to tap into and utilize the vast amount of biological information contained within remodeled ECM molecules. These heterodimeric transmembrane receptors may act as functional hubs by connecting the extracellular compartment with the cell’s interior, thereby helping to establish bidirectional circuits that distribute the complex biological information among numerous intracellular signaling pathways [1, 46]. These important information circuits physically connect the extracellular environment to the cell’s signaling machinery via integrin-associated cytosolic adaptor proteins (Fig. 3.3). Following integrin engagement of extracellular ligands, conformational changes are induced within integrins. Studies suggest that these conformational changes are associated with a physical separation of the a and b integrin chains that is propagated across the plasma membrane to their cytoplasmic tails [47]. The physical separation of the a and b integrin cytoplasmic tails is thought to facilitate the formation of scaffolds containing signaling molecules such as Fak, and Src [48]. Recent estimates have suggested that these protein-protein signaling complexes could be composed of different combinations of many molecules and can lead to distinct patterns of phosphorylation that help initiate as well as modify downstream signaling cascades such as PI3K/Akt and MAP/Erk pathways [49]. Importantly, sustained activation of both PI3K/Akt and MAP/Erk signaling pathways have been implicated in regulating tumor cell proliferation, invasion, migration and survival; all critical processes required to complete the metastatic cascade.

78

J.J. Ames et al.

Fig. 3.3 Integrins as sensors of the extracellular microenvironment. Integrins are a family of heterodimeric cell surface receptors with diverse activities. A major functional role for integrins during tumor growth and metastasis involves acting as bidirectional sensors of changes within the intra and extracellular environments. Integrin receptors can gather and interpret unique changes within these compartments and differentially distribute this molecular information into distinct biochemical signaling pathways that regulate cellular behavior

As can be appreciated from this brief discussion, metastasis involves a number of potential biomechanical processes that might contribute to the ability of tumor cells to metastasize. These events range from changes in mechanical properties of individual tumor cells to allow alterations in cell shape and deformability to modulation of the physiochemical properties of individual ECM molecules that may impart distinct functions. To illustrate this important concept, we will discuss a few examples of how changes in physiochemical and mechanical properties of widely expressed ECM proteins such as FN, may act as a biomechanical switches to facilitate unique functions. It is known that changes in ionic strength and pH can vary substantially within a tumor microenvironment and these physiochemical parameters can alter the folding properties of ECM molecules such as FN [50, 51]. In this regard, structural alterations of FN can uncover cryptic self-association sites that are crucial for higher-order fibrillar network formation [52]. These mechanical changes in structural integrity can have a pronounced impact on numerous cellular signaling pathways resulting in alterations in adhesion, migration and gene expression that might impact tumor cell metastasis. Changes in the elastic properties of FN for example, may alter cellular accessibility to a variety of cryptic sites within this

3

Biomechanical ECM Switches and Tumor Metastasis

79

molecule [53]. In fact, it has been suggested that the precise physical distance and orientation between the integrin binding PHSRN synergy site within the 9th type-III FN module and that of the well-known cell binding RGD site within the 10th type-III FN module can alter the ability of specific integrin receptors to bind FN [54]. Experimental modeling suggests that mechanical stretching and unfolding of FN can increase the distance between these receptor binding sites, thereby converting FN from an a5b1- and aIIbb3-binding ligand to one that can be recognized by avb3 [54]. In this regard, it is well known that signaling through integrins such as a5b1 and avb3 can regulate distinct signaling cascades such as MAP/Erk and PI3K/ Akt pathways leading to modulation of protease expression, cellular proliferation, invasion and survival. For example, a5b1-mediated interactions with FN are associated with high levels of activated RhoA following epithelial cell spreading, which lead to phosphorylation and inactivation of the actin-modifying protein cofilin [55]. In addition, focal adhesion structures were shown to be translocated to spike-like protrusions at the cell periphery, which was associated with random migration patterns [55]. In contrast, avb3-mediated adhesion to FN results in little phosphorylation of coflin, extensive reorganization of the actin cytoskeleton resulting in actin polarization and a more directed cellular migration [55]. Given these clearly distinct cellular responses it might not be surprising that mechanical changes in FN structure may impact a cell’s migratory and invasive behavior within the tumor microenvironment. Thus, mechanically induced switching of integrin binding can be associated with initiation of different molecular signaling cascades. Another interesting example shows that proteolytic cleavage may facilitate exposure of enzymatically active FN sub-domains that exhibit independent disulfide isomerase activity as well as unique collagen degrading activity [56, 57]. These intriguing studies demonstrate the possibility that an ECM protein primarily known for its ability to regulate adhesive functions may be converted into a protein with enzymatic activities. Importantly, this biomechanically-induced switching of integrin utilization or ECM function is not restricted solely to FN, as studies have shown that alterations in the triple-helical structure of collagen can expose cryptic RGD sites that can lead to a shift from predominantly b1 integrin-mediated signaling to that of both b1- and avb3-mediated signaling [58]. Biomechanical switching of integrin binding to collagen was shown to protect melanoma and endothelial cells from apoptosis by providing a survival advantage associated with changes in the ratio of Bcl-2 to Bax expression [58, 59]. These examples are just a few of the novel cryptic functions that may be activated following the triggering of biomechanical switches within solid-state ECM molecules.

3.4

Biomechanical ECM Switches in the Interstitial Matrix

The ECM can be thought of as being organized into at least two major compartments, which include a thin, sheet-like arrangement of proteins that comprise the basement membrane and the loosely organized interconnected network of molecules

80

J.J. Ames et al.

that make up the interstitial matrix. Given the tissue-specific variations in the molecular composition of these distinct matrix compartments, it would be well beyond the scope of this review to discuss in detail all of the molecules that comprise the ECM. Therefore, we will limit our discussion to a few well-characterized examples to illustrate the significance of biomechanical switches in regulating tumor growth and metastasis. Rapidly proliferating primary tumor cells can elicit a strong angiogenic response, resulting in tumor-associated blood vessels that are often leaky, which in turn can facilitate the deposition of provisional matrix proteins from the circulation such as vitronectin (VN), fibrinogen (FB) and fibronectin (FN). As with many ECM proteins, the functional impact of these provisional matrix molecules is governed in part by their structural conformation. For example, it has been suggested that the cell-binding RGD site within monomeric VN in its native configuration within the circulation is cryptic and therefore not readily accessible [60]. However, when VN is deposited into the ECM it may undergo multimerization, resulting in activation of a biomechanical switch that allows exposure of its cell binding RGD site [60]. Integrin-mediated interactions with cryptic RGD sites have been suggested to regulate tumor cell adhesion and invasion especially in association with glioblastoma [61]. In addition to mechanical force induced triggering of cryptic ECM switches, proteolytic activation of cryptic ECM switches can also occur. In this regard, a similar paradigm can be observed with other RGD-containing ECM proteins such as Collagen and Tenacin-C, both of which harbor cryptic RGD sites that can be proteolytically exposed [59, 62]. It is important to point out that additional peptide sequences outside of the classical cell binding RGD site may also represent cryptic regulatory elements. To this end, proteolytic exposure of EGF-like repeats within proteins such as Tenacin-C and LN have been shown to bind EGF receptors and thereby stimulating downstream signaling, which has been implicated in regulating the behavior of a number of distinct tumor cell types [63, 64]. Interestingly, subtractive screening of an antibody display library allowed the isolation of a novel antibody that selectively binds to a cryptic epitope within activated VN [65]. This cryptic VN site was preferentially localized within human renal cell, colon and breast carcinomas as compared to their corresponding normal tissue counterparts [65]. While it remains unclear whether this specific VN cryptic site contains an RGD epitope or whether it plays a functional role in regulating tumor progression, the selective distribution of this cryptic site is consistent with the notion that a subset of biomechanical ECM switches may be triggered within the tumor microenvironment. In addition to VN, another important provisional ECM protein often seen in association with malignant tumors is fibrinogen (FB). A number of studies have implicated cellular interactions with FB in the regulation of angiogenesis and tumor metastasis [66]. Fibrinogen has been shown to contain cryptic integrin binding sites that may be exposed by proteolytic mechanisms. An interesting example of a cryptic biomechanical ECM switch within FB involves the proteolytic exposure of a cryptic avb6 binding site. Studies suggest that MMP-9 may cleave FB, triggering

3

Biomechanical ECM Switches and Tumor Metastasis

81

the exposure of an avb6-binding element that facilitates cellular interactions that promote activation of MAPk/Erk signaling [67]. As we discussed above, one of the most thoroughly studied ECM proteins within the interstitial matrix in terms of its cryptic functions is FN. Fibronectin is known to harbor cryptic self-association sites that, when exposed following mechanical unfolding, promote fibrillogenesis and the formation of a more elaborate interconnected network that can provide mechanical support for tumor cells as well as providing a number of distinct binding sites for additional ECM molecules such as fibrin, tenascin and collagen [68]. Moreover, as detailed above, changes in FN conformation following heparin binding may expose unique cryptic sites that facilitate VEGF binding. The elevated abundance of angiogenic growth factors may provide a biological reservoir that can be utilized to stimulate tumor-associated angiogenesis, which in turn may facilitate tumor cell survival as well as provide readily accessible sites for tumor cells to gain access to the circulation. Beyond providing binding sites for multiple ECM molecules and a variety of growth factors, FN provides critical integrin binding sites that have been shown to regulate diverse molecular and cellular processes. For example, cellular interactions with FN mediated by a5b1 have been shown to promote cell survival by mechanisms involving differential regulation of Bcl-2 as well as differentially regulating the expression of proteolytic enzymes such as MMPs [69]. In addition to regulating the expression of matrix-modifying enzymes, proteolytic cleavage of FN may expose sub-domains that actually exhibit intrinsic protease activity by themselves. Thus, it would be interesting to speculate that localized exposure of these unique matrix-modulating fragments within the tumor microenvironment contributes to tumor cell metastasis. In addition, the a4b1 FN receptor has been shown to play a role in regulating myeloid cell recruitment to metastatic sites [39]. Finally, while studies have demonstrated that cryptic sites within FN promote cell adhesion, other studies have also provided evidence for a cryptic site within the 14th type-III FN module reduces cell adhesion [70]. It is interesting to point out that a reduction in cell adhesion is necessary to facilitate cellular invasion and migration. Taken together, the dynamic mechanical and elastic properties of these multi-domain ECM molecules within the interstitial tumor microenvironment appear well-suited to serve as cryptic biomechanical switches.

3.5

Biomechanical ECM Switches in the Basement Membrane

A second major ECM compartment within the tumor microenvironment that is known to regulate tumor metastasis is the basement membrane. The molecular composition of basement membranes can vary substantially between different organ systems but collectively serves to regulate a variety of tissue-specific cellular processes ranging from tissue compartmentalization to gene expression. To illustrate the functional importance of cryptic biomechanical switches within basement membrane, we will focus primarily on laminin and collagen type-IV, the two major

82

J.J. Ames et al.

structural molecules within this compartment. The functional trimmer of LN is formed from distinct combinations of three chains termed a, b and g. Various combinations of these chains can result in the generation of LN isoforms with tissuespecific distribution patterns [71]. Cells can interact with LN by several cell surface molecules including integrins and non-integrin molecules such as a-dystroglycan, the 37/67Kda laminin receptor and some galactoside-binding lectins [71, 72]. As with many other ECM molecules, LN contains a number of functional domains that may regulate tumor metastasis including growth factor binding sites and integrin interaction elements. An extensive analysis of LN-1 has identified multiple peptides that exhibit differential biological activities in angiogenesis and tumor metastasis. For example, a YIGRS-containing peptide from the LN b1 chain was shown to reduce angiogenesis and tumor metastasis in experimental models [73]. In contrast, a SIKVAV-containing peptide from the a1 chain of LN was shown to increase metastasis [74]. These studies as well as others are consistent with the notion that unique cryptic sites within LN may play functional roles in controlling specific parts of the metastatic cascade. A well-characterized example of this concept can be illustrated by work demonstrating that MMP-mediated cleavage of the g2 chain of LN-5 can expose a cryptic site that promotes tumor cell motility [17]. To identify and characterize potentially functional cryptic ECM biomechanical switches in LN, our laboratory used a subtractive screening approach to examine a peptide phage display library in order to generate synthetic peptides that selectively bound denatured and proteolyzed LN-1 [75]. These studies resulted in the generation of a synthetic peptide termed STQ that bound denatured and proteolyzed LN-1, while exhibiting little if any interactions with intact LN. This cryptic LN site was preferentially exposed within basement membranes associated with malignant melanomas and tumor-blood vessels but failed to show significant exposure within corresponding normal tissues [75]. The STQ-peptide selectively inhibited tumor and endothelial cell adhesion, migration and proliferation on denatured LN, but failed to significantly alter these cellular events on intact LN, suggesting that this novel cryptic element can selectively modulate cellular behavior. The ability of this cryptic LN epitope to regulate these cellular processes may involve modulation of the cyclin-dependent kinase inhibitor P27KIP1 as the STQ-peptide induced elevated levels of P27KIP1 in melanoma cells seeded on denatured LN. Importantly, targeting of this selectively triggered biomechanical switch inhibited both angiogenesis and melanoma tumor metastasis in vivo [75]. Taken together, these findings suggest that the cryptic LN site defined by STQ-peptide may represent a functionally relevant biomechanical ECM switch that is triggered within the tumor microenvironment to promote tumor metastasis. A second major glycoprotein component of the basement membrane is collagen type-IV. Collagen type-IV is a triple-helical molecule composed of different combinations of a-chains with a characteristic tri-peptide repeat, where glycine is often present at every third amino acid position. This tri-peptide motif along with hydroxylation of proline residues have been shown to be important for the triple-helical structure of collagen and the packaging of collagen into more complex fibrillar networks [76, 77]. Similar to many other ECM glycoproteins, collagen type-IV also has discrete functional domains that provide unique binding sites for cell surface

3

Biomechanical ECM Switches and Tumor Metastasis

83

receptors, matrix altering enzymes and possibly secreted growth factors. Similar to what was accomplished with LN, studies on collagen type-IV have uncovered a wide array of peptide fragments with potent biological activities. The majority of these functional peptides were shown to exist within the non-collagenous (NC) domains and were shown to inhibit angiogenesis and tumor growth by mechanisms involving alterations in integrin signaling [78, 79]. Using a subtractive immunization approach, we developed a panel of antibodies that preferentially bound to denatured or proteolyzed collagen [80]. A monoclonal antibody (Mab HUIV26) generated from this procedure was shown to bind a cryptic epitope restricted to basement membrane collagen type-IV. This novel cryptic epitope may function as a biomechanical ECM switch as exposure of this site can be triggered by thermal denaturation as well as MMP-mediated cleavage [16]. While exposure of the HUIV26 cryptic site could not be readily detected in basement membranes of most normal tissues examined, it was exposed within basement membranes of malignant tumors and angiogenic blood vessels [81, 82]. The functional relevance of the HUIV26 biomechanical switch in tumor progression and angiogenesis was shown as a Mab directed to this site selectively inhibited angiogenesis, tumor growth and metastasis in several animal models [16, 81–84]. Importantly, studies have indicated that cellular interactions with this cryptic site were mediated by integrin avb3 in a non-RGD dependent manner. These studies provide an additional example of a mechanism by which cells may sense and utilize biological information following triggering of a unique basement membrane biomechanical switch. Using a similar subtractive immunization approach, we generated a second antibody termed HU177 that binds denatured and proteolyzed forms of multiple types of collagen including collagen type-IV [80, 85]. Peptide-based mapping analysis revealed that Mab HU177 recognized, in part, a consensus sequence containing the amino acids GPPG and also demonstrated a requirement for proline hydroxylation [85, 86]. As was observed with the HUIV26 cryptic site, little exposure of the HU177 cryptic epitope was detected within the ECM of normal tissues, while the HU177 cryptic site was readily detectable within the ECM of malignant tumors and angiogenic blood vessels [85, 87]. Importantly, antibodies specifically targeting the HU177 cryptic element were shown to inhibit both angiogenesis and tumor growth in vivo [85, 88]. Taken together, these studies suggest that the selective exposure of these functional biomechanical switches within the tumor microenvironment may represent clinically useful therapeutic and imaging targets.

3.6

Biomechanical ECM Switches as Therapeutic and Imaging Targets

Given the multiple cellular functions governed by solid-state ECM molecules, the non-cellular tissue microenvironment might represent a relatively untapped source of therapeutic and imaging targets. However, due to the ubiquitous expression of these molecules a clear strategy is needed to allow selective targeting of functional

84

J.J. Ames et al.

ECM sites associated with pathological conditions to avoid interactions in normal tissues. In this regard, the concept that selective triggering of biomechanical ECM switches may occur within the non-cellular tumor compartment may facilitate the development of novel approaches to selectively exploit the unique mechanical changes within tissues during tumor growth and metastasis. Evidence is beginning to accumulate that conformational alterations of ECM molecules may be useful for selective imaging of pathological processes such as tumor growth. One example involves the alternatively spliced form of FN termed ED-B Fibronectin [89]. Insertion of the ED-B domain into fibronectin was shown to create a conformational change that resulted in exposure of a cryptic epitope [90]. Interestingly, ED-B FN has been shown to be selectively expressed within tumorassociated vasculature, but not within quiescent adult blood vessels [89]. Preclinical studies have demonstrated that targeting this unique form of FN can lead to a preferential accumulation in experimental tumors growing in vivo. Clinical testing using L19, a radiolabeled antibody that specifically interacts with ED-B FN, has been initiated to examine its use in detection of colorectal and lung tumors [91]. With the initiation of these early proof of concept clinical investigations, along with the generation of antibodies and peptides that selectively bind novel biomechanical ECM switches in matrix molecules such as collagen, laminin and vitronectin, the clinical use of these novel reagents in combination with imaging modalities such as MRI, and Ultrasound may significantly enhance the ability to detect tumor growth and metastasis at an early stage. Similarly, the potential of targeting selectively exposed biomechanical ECM switches for the treatment of malignant tumors are beginning to be investigated in the clinical setting. For example, a murine Mab directed to the HU177 cryptic collagen epitope has recently been humanized and optimized (TRC093) for clinical use [92]. In preclinical studies, humanized Mab TRC093 was shown to selectively localize to tumor-associated blood vessels in animal models [88]. Moreover, TRC093 has been shown to inhibit tumor-associated angiogenesis and effectively enhanced the anti-tumor activity of Paclitaxel in vivo [88]. These and other studies have lead to a Phase-I human clinical trial of TRC093. Results of this clinical trial indicated that TRCO93 was well-tolerated with no dose-limiting toxicities observed [92]. Importantly, evidence of clinical activity was detected as stable disease was observed in several subjects and a reduction in liver lesions was noted in a patient with ovarian carcinoma. While all patients eventually progressed, these studies provide some of the first evidence consistent with the notion that selective targeting of cryptic biomechanical ECM switches may represent a useful therapeutic strategy.

3.7

Conclusion

The growing appreciation for the functional importance of the stromal microenvironment in governing malignant tumor development and metastasis has stimulated many investigators to search for innovative new strategies to exploit the increasing

3

Biomechanical ECM Switches and Tumor Metastasis

85

molecular understanding of the stroma to help control tumor progression. In addition to the progress that has been made in our knowledge of the diverse cell populations that contribute to tumor growth and metastasis, new molecular insight is rapidly emerging into the importance of unique biomechanical changes within individual ECM molecules as they pertain to regulation of the malignant phenotype. While it has been known for some time that the mechanical properties of a given ECM molecule can regulate its function, the concept that selective triggering of specific conformational changes within solid-state ECM molecules during pathological tissue remodeling may serve as biomechanical switches to regulate information flow is rapidly gaining acceptance. In fact, exciting basic research over the last decade has now led to examining the potential of targeting these unique biomechanical switches for both imaging and therapy in human clinical trials. It is likely that the translational applications of this new biological insight will expand into other clinical areas in the coming years. Acknowledgements This work was supported in part from the National Institute of Health Grant CA91645 to PCB, The Maine Cancer Foundation (CPHV and PCB) and the National Institute of Health Center for Research Resources P20-RR-15555 (CPHV and PCB) and NIH Grant HL083151 (CPHV). Jacquelyn Ames is supported by the Integrative Graduate Education and Research Traineeship (NSF-IGERT DGE0221625) from UMaine. We would like to apologize to those investigators whose important work was not cited do to space limitations.

References 1. Contois L, Akalu A, Brooks PC (2009) Integrins as “functional hubs” in the regulation of pathological angiogenesis. Sem Cancer Biol 19:318–328 2. Cretu A, Brooks PC (2007) Impact of the non-cellular tumor microenvironment on metastasis: potential therapeutic and imaging opportunities. J Cell Physiol 213:391–402 3. Bershadsky A, Kozlov M, Geiger B (2006) Adhesion-mediated mechanosensitivity: a time to experiment, and a time to theorize. Curr Opin Cell Biol 18(5):472–481 4. Järveläinen H, Sainio A, Koulu M et al (2009) Extracellular matrix molecules: potential targets in pharmacotherapy. Pharmacol Rev 61(2):198–223 5. Daley WP, Peters SB, Larsen M (2008) Extracellular matrix dynamics in development and regenerative medicine. J Cell Sci 121:255–264 6. Talmadge JE, Fidler IJ (2010) AACR centennial series: the biology of cancer metastasis: historical perspective. Cancer Res 70(14):5649–5669 7. Brooks SA, Lomax-Browne HJ, Carter TM et al (2008) Molecular interactions in cancer cell metastasis. Acta Histochem 112:3–25 8. Schmalhofer O, Brabletz S, Brabletz T (2009) E-cadherin, beta-catenin, and ZEB1 in malignant progression of cancer. Cancer Metastasis Rev 28(1–2):151–166 9. Langley RR, Fidler IJ (2007) Tumor cell-organ microenvironment interactions in the pathogenesis of cancer metastasis. Endocr Rev 28(3):297–321 10. Chiodoni C, Colombo MP, Sangaletti S (2010) Matricellular proteins: from homeostasis to inflammation, cancer, and metastasis. Cancer Metastasis Rev 29(2):295–307 11. Affara NI, Andreu P, Coussens LM (2009) Delineating protease functions during cancer development. In: Antalis TM, Bugge TH (eds) Methods in molecular biology, proteases and cancer, vol 539. Humana Press, Totowa

86

J.J. Ames et al.

12. Brooks PC, Strömblad S, Sanders LC et al (1996) Localization of matrix metalloproteinase MMP-2 to the surface of invasive cells by interaction with integrin alpha v beta 3. Cell 85(5):683–693 13. Stefanidakis M, Koivunen E (2006) Cell-surface association between matrix metalloproteinases and integrins: role of the complexes in leukocyte migration and cancer progression. Blood 108(5):1441–1450 14. Pasco S, Ramont L, Maquart FX et al (2004) Control of melanoma progression by various matrikines from basement membrane macromolecules. Crit Rev Oncol Hematol 49:221–233 15. Levental KR, Yu H, Kass L et al (2009) Matrix Crosslinking Forces Tumor Progression by Enhancing Integrin Signaling. Cell 139(5):891–906 16. Xu J, Rodriguez D, Petitclerc E et al (2001) Proteolytic exposure of a cryptic site within collagen type IV is required for angiogenesis and tumor growth in vivo. J Cell Biol 154(5):1069–1079 17. Giannelli G, Falk-Marzillier J, Schiraldi O et al (1997) Induction of cell migration by matrix metalloprotease-2 cleavage of laminin-5. Science 277(5323):225–228 18. Sekiguchi K, Hakomori S, Funahashi M et al (1983) Binding of fibronectin and its proteolytic fragments to glycosaminoglycans. Exposure of cryptic glycosaminoglycan-binding domains upon limited proteolysis. J Biol Chem 258(23):14359–14365 19. Kumar S, Weaver VM (2009) Mechanics, malignancy, and metastasis: The force journey of a tumor cell. Cancer Metastasis Rev 28(1–2):113–127 20. Yu H, Mouw JK, Weaver VM (2010) Forcing form and function: biomechanical regulation of tumor evolution. Trends Cell Biol 21(1):47–56 21. Vogel V (2006) Mechanotransduction involving multimodular proteins: converting force into biochemical signals. Annu Rev Biophys Biomol Struct 35:459–488 22. Geiger B, Spatz JP, Bershadsky AD (2009) Environmental sensing through focal adhesions. Nat Rev Mol Cell Biol 10(1):21–33 23. Choquet D, Felsenfeld DP, Sheetz MP (1997) Extracellular matrix rigidity causes strengthening of integrin cytoskeleton linkages. Cell 88(1):39–48 24. Giannone G, Sheetz MP (2006) Substrate rigidity and force define form through tyrosine phosphatase and kinase pathways. Trends Cell Biol 16(4):213–223 25. Yilmaz M, Christofori G (2009) EMT, the cytoskeleton, and cancer cell invasion. Cancer Metastasis Rev 28(1–2):15–33 26. Spaderna S, Schmalhofer O, Hlubek F et al (2006) A Transient, EMT-linked loss of basement membranes indicates metastasis and poor survival in colorectal cancer. Gastroenterology 131:830–840 27. Tran KT, Lamb P, Deng JS (2005) Matrikines and matricryptins: implications for cutaneous cancers and skin repair. J Dermatol Sci 40(1):11–20 28. Lin M, Jackson P, Tester AM et al (2008) Matrix metalloproteinase-8 facilitates neutrophil migration through the corneal stromal matrix by collagen degradation and production of the chemotactic peptide Pro-Gly-Pro. Am J Pathol 173(1):144–153 29. Condeelis J, Pollard JW (2006) Macrophages: obligate partners for tumor cell migration, invasion, and metastasis. Cell 124:263–266 30. Wyckoff JB, Wang Y, Lin EY et al (2007) Direct visualization of macrophage-assisted tumor cell intravasation in mammary tumors. Cancer Res 67(6):2649–2656 31. Escorcio-Correia M, Hagemann T (2010) Macrophages in the tumor microenvironment. In: Bagley RG (ed) The tumor microenvironment, cancer drug discovery and development. Humana Press, New York 32. Wolf K, Friedl P (2009) Mapping proteolytic cancer cell-extracellular matrix interfaces. Clin Exp Metastasis 26(4):289–298 33. Yahalom J, Eldor A, Brian S et al (1985) Platelet-tumor cell interaction with the subendothelial extracellular matrix: relationship to cancer metastasis. Radiother Oncol 3(3):211–225 34. Jain S, Harris J, Ware J (2010) Platelets. Linking Hemostasis and Cancer. Arterioscler Thromb Vasc Biol 30(12):2362–2367 35. Berg EL, Goldstein LA, Jutila MA et al (1989) Homing receptors and vascular addressins: cell adhesion molecules that direct lymphocyte traffic. Immunol Rev 108:5–18

3

Biomechanical ECM Switches and Tumor Metastasis

87

36. Bargatze RF, Streeter PR, Butcher EC (1990) Expression of low levels of peripheral lymph node-associated vascular addressin in mucosal lymphoid tissues: possible relevance to the dissemination of passaged AKR lymphomas. J Cell Biol 42(4):219–227 37. Wang S, Voisin MB, Larbi KY et al (2006) Venular basement membranes contain specific matrix protein low expression regions that act as exit points for emigrating neutrophils. J Exp Med 203(6):1519–1532 38. Voisin MB, Woodfin A, Nourshargh S (2009) Monocytes and neutrophils exhibit both distinct and common mechanisms in penetrating the vascular basement membrane in vivo. Arterioscler Thromb Vasc Biol 29(8):1193–1199 39. Kaplan RN, Riba RD, Zacharoulis S et al (2005) VEGFR1-positive haematopoietic bone marrow progenitors initiate the pre-metastatic niche. Nature 438(7069):820–827 40. Kaplan RN, Psaila B, Lyden D (2006) Bone marrow cells in the “pre-metastatic niche”: within bone and beyond. Cancer Metastasis Rev 25(4):521–529 41. Qian BZ, Pollard JW (2010) Macrophage diversity enhances tumor progression and metastasis. Cell 141(1):39–51 42. Grivennikov S, Greten FR, Karin M (2010) Immunity, inflammation, and cancer. Cell 140(6):883–899 43. Mitsi M, Forsten-Williams K, Gopalakrishnan M et al (2008) A catalytic role of heparin within the extracellular matrix. J Biol Chem 283(50):34796–34807 44. Kilarski WW, Samolov B, Petersson L et al (2009) Biomechanical regulation of blood vessel growth during tissue vascularization. Nat Med 15(6):657–664 45. Paget S (1889) The distribution of secondary growths in cancer of the breast. Lancet 133:571–573 46. Cabodi S, Di Stefano P, Leal Mdel P et al (2010) Integrins and signal transduction. Adv Exp Med Biol 674:43–54 47. Arnaout MA, Goodman SL, Xiong JP (2007) Structure and mechanics of integrin-based cell adhesion. Curr Opin Cell Biol 19(5):495–507 48. Zhao J, Guan JL (2009) Signal transduction by focal adhesion kinase in cancer. Cancer Metastasis Rev 28(1–2):35–49 49. Zaidel-Bar R, Itzkovitz S, Ma’ayan A et al (2007) Functional atlas of the integrin adhesome. Nat Cell Biol 9(8):858–867 50. Parks SK, Chiche J, Pouyssegur J (2011) pH control mechanisms of tumor survival and growth. J Cell Physiol 226(2):299–308 51. Goerges AL, Nugent MA (2004) pH regulates vascular endothelial growth factor binding to fibronectin: a mechanism for control of extracellular matrix storage and release. J Biol Chem 279(3):2307–2315 52. Vakonakis I, Staunton D, Rooney LM et al (2007) Interdomain association in fibronectin: insight into cryptic sites and fibrillogenesis. EMBO J 26(10):2575–2583 53. Vogel V, Thomas WE, Craig DW et al (2001) Structural insights into the mechanical regulation of molecular recognition sites. Trends Biotechnol 19(10):416–423 54. Krammer A, Craig D, Thomas WE et al (2002) A structural model for force regulated integrin binding to fibronectin’s RGD-synergy site. Matrix Biol 21(2):139–147 55. Danen EHJ, van Rheenen J, Franken W et al (2005) Integrins control motile strategy through a Rho-cofilin pathway. J Cell Biol 169(3):515–526 56. Langenbach KJ, Sottile J (1999) Identification of protein-disulfide isomerase activity in fibronectin. J Biol Chem 274(11):7032–7038 57. Boudjennah L, Dalet-Fumeron V, Pagano M (1998) Expression of collagenase/gelatinase activity from basement-membrane fibronectin–isolation after limited proteolysis of a bovine lens capsule and molecular definition of this thiol-dependent zinc metalloproteinase. Eur J Biochem 255(1):246–254 58. Montgomery AM, Reisfeld RA, Cheresh DA (1994) Integrin alpha v beta 3 rescues melanoma cells from apoptosis in three-dimensional dermal collagen. Proc Natl Acad Sci USA 91(19):8856–8860

88

J.J. Ames et al.

59. Petitclerc E, Strömblad S, von Schalscha TL et al (1999) Integrin a V b 3 Promotes M21 Melanoma Growth in Human Skin by Regulating Tumor Cell Survival Integrin alpha(v)beta3 Promotes M21 Melanoma Growth in Human Skin by Regulating. Cancer Res 59(11): 2724–2730 60. Seiffert D, Smith JW (1997) The cell adhesion domain in plasma vitronectin is cryptic. J Biol Chem 272(21):13705–13710 61. Gladson CL, Cheresh DA (1991) Glioblastoma expression of vitronectin and the alpha v beta 3 integrin. Adhesion mechanism for transformed glial cells. J Clin Invest 88(6):1924–1932 62. Denda S, Ulrich M, Crossin KL et al (1998) Utilization of soluble integrin-alkaline phosphatase chimera to characterize integrin alpha8beta1 receptor interactions with tenascin: murine alpha 8 beta 1 binds to the RGD site in Tenascin-C fragments, but not to native Tenascin-C. Biochemistry 37(16):5464–5474 63. Iyer AKV, Tran KT, Borysenko CW et al (2006) Tenascin cytotactin epidermal growth factorlike repeat binds epidermal growth factor receptor with low affinity. J Cell Physiol 211(3): 748–758 64. Panayotou G, End P, Aumailley M et al (1989) Domains of laminin with growth-factor activity. Cell 56(1):93–101 65. Bloemendal HJ, de Boer HC, Koop EA et al (2004) Activated vitronectin as a target for anticancer therapy with human antibodies. Cancer Immunol Immunother 53(9):799–808 66. Staton CA, Brown NJ, Lewis CE (2003) The role of fibrinogen and related fragments in tumour angiogenesis and metastasis. Expert Opin Biol Ther 3(7):1105–1120 67. Fouchier F, Penel C, Pierre Montero M et al (2007) Integrin alphavbeta6 mediates HT29-D4 cell adhesion to MMP-processed fibrinogen in the presence of Mn2+. Eur J Cell Biol 86(3):143–160 68. Singh P, Carraher C, Schwarzbauer JE (2010) Assembly of fibronectin extracellular matrix. Annu Rev Cell Dev Biol 26:397–419 69. Lee BH, Ruoslahti E (2005) alpha5beta1 integrin stimulates Bcl-2 expression and cell survival through Akt, focal adhesion kinase, and Ca2+/calmodulin-dependent protein kinase IV. J Cell Biochem 95(6):1214–1223 70. Miura S, Kamiya S, Saito Y et al (2007) Antiadhesive sites present in the fibronectin type IIIlike repeats of human plasma fibronectin. Biol Pharm Bull 30(5):891–897 71. Hallmann R, Horn N, Selg M et al (2005) Expression and function of laminins in the embryonic and mature vasculature. Physiol Rev 85:979–1000 72. Stipp CS (2010) Laminin-binding integrins and their tetraspanin partners as potential antimetastatic targets. Expert Rev Mol Med 12:e3 73. Iwamoto Y, Robey FA, Graf J et al (1987) YIGSR, a synthetic laminin pentapeptide, inhibits experimental metastasis formation. Science 238(4830):1132–1134 74. Yamamura K, Kibbey MC, Jun SH et al (1993) Effect of matrigel and laminin peptide YIGSR on tumor growth and metastasis. Semin Cancer Biol 4(4):259–265 75. Akalu A, Roth JM, Caunt M et al (2007) Inhibition of angiogenesis and tumor metastasis by targeting a matrix immobilized cryptic extracellular matrix epitope in laminin. Cancer Res 67(9):4353–4363 76. Kadler KE, Hill A, Canty-Laird EG (2008) Collagen fibrillogenesis: fibronectin, integrins, and minor collagens as organizers and nucleators. Curr Opin Cell Biol 20:495–501 77. Bhattacharjee A, Bansal M (2005) Collagen structure: the madras triple helix and the current scenario. IUBMB Life 57(3):161–172 78. Petitclerc E, Boutaud A, Prestayko A et al (2000) New functions for non-collagenous domains of human collagen type IV. Novel integrin ligands inhibiting angiogenesis and tumor growth in vivo. J Biol Chem 275(11):8051–8061 79. Roth JM, Akalu A, Zelmanovich A et al (2005) Recombinant alpha2(IV)NC1 domain inhibits tumor cell-extracellular matrix interactions, induces cellular senescence, and inhibits tumor growth in vivo. Am J Pathol 166(3):901–911 80. Xu J, Rodriguez D, Kim JJ et al (2000) Generation of monoclonal antibodies to cryptic collagen sites by using subtractive immunization. Hybridoma 19(5):375–385

3

Biomechanical ECM Switches and Tumor Metastasis

89

81. Hangai M, Kitaya N, Xu J et al (2002) Matrix metalloproteinase-9-dependent exposure of a cryptic migratory control site in collagen is required before retinal angiogenesis. Am J Pathol 161(4):1429–1437 82. Odaka C, Tanioka M, Itoh T (2005) Matrix metalloproteinase-9 in macrophages induces thymic neovascularization following thymocyte apoptosis. J Immunol 174(2):846–853 83. Roth JM, Caunt M, Cretu A et al (2006) Inhibition of experimental metastasis by targeting the HUIV26 cryptic epitope in collagen. Am J Pathol 168(5):1576–1586 84. Jo N, Ju M, Nishijima K et al (2006) Inhibitory effect of an antibody to cryptic collagen type IV epitopes on choroidal neovascularization. Mol Vis 12:1243–1249 85. Cretu A, Roth JM, Caunt M et al (2007) Disruption of endothelial cell interactions with the novel HU177 cryptic collagen epitope inhibits angiogenesis. Clin Cancer Res 13(10): 3068–3078 86. Freimark B, Clark D, Pernasetti F et al (2007) Targeting of humanized antibody D93 to sites of angiogenesis and tumor growth by binding to multiple epitopes on denatured collagens. Mol Immunol 44:3741–3750 87. Brooks PC, Roth JM, Lymberis SC et al (2002) Ionizing radiation modulates the exposure of the HUIV26 cryptic epitope within collagen type IV during angiogenesis. Int J Radiat Oncol Biol Phys 54(4):1194–1201 88. Pernasetti F, Nickel J, Clark D et al (2006) Novel anti-denatured collagen humanized antibody D93 inhibits angiogenesis and tumor growth: An extracellular matrix-based therapeutic approach. Int J Oncol 29(6):1371–1379 89. Ebbinghaus C, Sheuermann J, Neri D et al (2004) Diagnostic and therapeutic applications of recombinant antibodies: targeting the extra-domain B of fibronectin, a marker of tumor angiogenesis. Curr Pharm Des 10(13):1537–1549 90. Ventura E, Sassi F, Parodi A et al (2010) Alternative splicing of the angiogenesis associated extra-domain B of fibronectin regulates the accessibility of the B-C loop of the type III repeat 8. PLoS One 5(2):e9145 91. Santimaria M, Moscatelli G, Viale GL et al (2003) Immunoscintigraphic detection of the ED-B domain of fibronectin, a marker of angiogenesis, in patients with cancer. Clin Cancer Res 9(2):571–579 92. Robert F, Gordon MS, Rosen LS et al (2010) Final results from a phase 1 study of TRC093 (humanized anti-cleaved collagen antibody) in patients with solid cancer. ASCO annual meeting abst # 3038

Chapter 4

Moving Aggressively: S100A4 and Tumor Invasion Reniqua P. House, Sarah C. Garrett, and Anne R. Bresnick

Abstract Tumor cell invasion and metastasis are complex multi-step cellular processes that involve the ability of malignant cells to escape the primary tumor, migrate to distant sites within the body and establish secondary tumors. S100A4, a member of the S100 family of Ca2+-binding proteins, is a prognostic marker in a number of human cancers and has a direct role in promoting metastatic disease. S100A4’s metastasis promoting properties are attributed to diverse biological functions including the regulation of apoptosis, cell motility, invasion and angiogenesis. In this chapter we discuss the contribution of tumor and stromal-derived S100A4 in supporting metastatic dissemination and examine the biochemical properties of S100A4 and the S100A4/target protein interactions that may contribute to its function as a metastasis factor.

4.1

S100 Protein Family

S100A4 is a member of the S100 family of Ca2+-binding proteins and is implicated in the modulation of tumor cell motility, invasion, and angiogenesis (Fig. 4.1). The term “S100” arose from the observation that these proteins are soluble in 100% saturated ammonium sulfate at neutral pH [82]. In humans there are 21 S100 family members, the majority of which cluster on the long arm of human chromosome 1, with the remaining members spread throughout the genome on chromosomes 4, 5, 7, 21 and the X chromosome [74, 98]. The S100 proteins are expressed in a cell and tissue specific manner in vertebrates [22]. Due to the large number of family members and the diversity of their biological targets, S100 proteins regulate multiple

R.P. House • S.C. Garrett • A.R. Bresnick (*) Department of Biochemistry, Albert Einstein College of Medicine, 1300 Morris Park Avenue, Bronx, NY 10461, USA e-mail: [email protected]

A. Fatatis (ed.), Signaling Pathways and Molecular Mediators in Metastasis, DOI 10.1007/978-94-007-2558-4_4, © Springer Science+Business Media B.V. 2012

91

92

R.P. House et al.

Fig. 4.1 Proposed functions of S100A4-mediated metastatic dissemination. S100A4 (red circles) is expressed by diverse cells within the tumor microenvironment, including tumor cells, fibroblasts, macrophages and T cells. Intracellular S100A4 is associated with increased cell migration and invasion through interactions with cytoskeletal components such as myosin-IIA and via unidentified mechanisms. S100A4 is also released by tumor and stromal cells, and elicits a variety of cellular responses within the tumor microenvironment. Through interactions with annexin A2, S100A4 stimulates the conversion of plasminogen to plasmin on the endothelial cell surface to promote angiogenesis. Extracellular S100A4 also activates NF-kB in tumor cells; however, the receptors responsible for S100A4 signaling have not been identified

cellular processes and are involved in a number of human pathologies such as neurodegenerative and inflammatory disorders, fibrosis, cardiomyopathies and cancer [74, 98, 104]. The S100 family members share 25–65% sequence identity, and as expected exhibit overall similar three-dimensional structures. The majority of the S100 proteins exist as symmetric, anti-parallel dimers except for S100G, which is monomeric [2, 74, 98, 104]. For dimeric S100 proteins, each monomer consists of four helices and a small b-strand. Two helices from each subunit (1, 4, 1¢, and 4¢) form a tight X-type four-helix bundle that comprises the dimer interface [23, 95, 119] (Fig. 4.2). In addition, each S100 monomer contains two helix-loop-helix motifs, referred to as EF-hands, which bind calcium. The 12 residue canonical C-terminal EF-hand coordinates calcium using mostly side-chain carbonyl oxygens and exhibits a high affinity for calcium [23, 33, 73, 90, 132]. The N-terminal EF-hand, also known as the

4 Moving Aggressively: S100A4 and Tumor Invasion Fig. 4.2 Ribbon and surface diagrams of apo-S100A4 and Ca2+-S100A4 with one monomer shown in blue and one shown in green. Dark gray spheres denote calcium ions. Residues colored yellow exhibit significant chemical shift perturbations upon binding a myosin-IIA peptide comprised of residues 1908–1923

93

94

R.P. House et al.

“S100-specific” or pseudo EF-hand, differs from the canonical EF-hand found in Ca2+-binding proteins such as calmodulin and troponin C, and in the C-terminus of the S100 proteins [37, 73, 113]. The N-terminal pseudo EF-hand is comprised of 14 instead of 12 residues and coordinates calcium primarily via main-chain backbone carbonyl oxygens [23, 37, 73]. The reduced flexibility of the main-chain atoms may account for the 10 to 250-fold weaker affinity exhibited by the N-terminal EF-hand for calcium as compared to the C-terminal EF-hand [33, 73, 76]. Calcium has a pivotal role in S100 protein activation. Most S100 proteins, including S100A4, undergo a significant conformational rearrangement following Ca2+-binding to the C-terminal EF-hand [74, 98, 104]. In the apo state, helices 3 and 4, which flank the C-terminal EF-hand, have a roughly parallel orientation with respect to one another. Following Ca2+-binding, helix 3 in each S100A4 subunit rotates by 55° with respect to helix 4, exposing a hydrophobic cleft that serves as the binding pocket for target proteins (Fig. 4.2) [73, 119]. Of particular relevance to S100A4, NMR studies examining the binding of a 16 residue peptide from the target myosin-IIA demonstrated that residues in the hinge and helices 3 and 4 exhibited significant chemical shift perturbations following peptide addition [73]. These observations suggest that myosin-IIA binds in the Ca2+-induced hydrophobic cleft, which is consistent with other S100-target complexes [8, 63, 100, 103]. These structural studies highlight the dependency of S100 protein-target interactions on Ca2+activation and demonstrate that S100 proteins function as Ca2+-activated switches.

4.2

S100A4 as a Cancer Prognostic Marker

The first indication that S100A4 could serve as a prognostic marker came from the observations of Lukanidin and colleagues, who isolated S100A4 mRNA from metastatic CSML100 mouse mammary adenocarcinoma cells and noted that S100A4 mRNA was absent in nonmetastatic CSML0 cells [26]. Subsequently, S100A4 expression has been shown to be a predictive indicator of poor patient survival in several studies of breast cancer. In a study of 349 patients with stage I or II invasive breast cancer, the 19-year median survival rate was increased significantly in S100A4 negative patients (>228 months) compared to S100A4 positive patients (47 months) with 80% of the S100A4 negative patients and 11% of the S100A4 positive patients alive at the conclusion of the study [102]. The association between S100A4 expression and patient mortality was confirmed in two additional studies. In a cohort of 92 patients with stage I breast cancer, patients with S100A4 positive tumors had a significantly worse 10 year survival (29%) than those with S100A4 negative tumors (68.9%) [62]. Moreover, a 312 patient study of women with minimally invasive stage I or II breast cancer showed that the 18-year median for survival was >204 months for S100A4 negative patients as compared to 186 months for S100A4 positive patients [21]. Together these studies suggest that in early stage breast cancer S100A4 may be a prognostic indicator for tumor progression and patient mortality.

4 Moving Aggressively: S100A4 and Tumor Invasion

95

In contrast to these findings, two additional studies found no correlation between S100A4 expression and patient outcome. In a study of 66 stage I-IV breast carcinomas, S100A4 expression did not correlate with patient survival; however, there was an inverse correlation between S100A4 expression and the loss of estrogen receptor expression [92]. Similarly in a study examining S100A4 mRNA levels from 295 stage I-III breast cancer patients, S100A4 mRNA expression correlated with the loss of estrogen receptor expression and was associated with basal-type cancers, but was not predictive with respect to patient outcome [79]. The discrepancy between these two studies and studies demonstrating that S100A4 expression can predict patient outcome may result from several factors, including (1) small sample size and a relatively short follow-up period [92]; (2) the evaluation of S100A4 mRNA instead of S100A4 protein levels [79]; and (3) the inclusion of patients with more advanced disease [79, 92]. While there is substantive data supporting S100A4 as a predictive marker for patient outcome, additional studies using larger patient populations are needed to confirm that S100A4 is indeed a prognostic factor for early stage breast cancers. In addition to breast cancer, positive S100A4 expression correlates with (1) blood vessel invasion and poor patient survival in lung adenocarcinoma [77]; (2) lymph node and peritoneal metastases in gastric cancers [128, 140]; (3) recurrence and/or metastatic dissemination of colorectal cancers [39, 60]; (4) poor differentiation of pancreatic ductal adenocarcinomas [101]; (5) metastatic relapse in bladder cancer [1, 19]; (6) poor prognosis in renal cell carcinoma [6]; and (7) lymph node metastases in papillary thyroid carcinomas [80]. The association of S100A4 expression with metastatic disease and poor patient survival in a number of cancers suggests that S100A4 may have widespread use as a marker for disease progression. Current methods rely on tumor size and histological grade to predict potential metastatic dissemination. The identification and development of molecular markers, such as S100A4, which are better predictors of metastatic disease, will be necessary to improve treatment strategies and patient outcomes.

4.3

S100A4 Promotes Tumor Metastasis

Tumor invasion and metastasis are complex multi-step cellular processes that account for the majority of cancer related deaths [109, 111]. Although S100A4 has been considered a marker of metastatic disease, there is significant evidence demonstrating that S100A4 has a causal role in the metastatic process. In an orthotopic model of breast cancer, overexpression of S100A4 in nonmetastatic rat mammary adenocarcinoma cells resulted in an approximately 60% increase in the number of tumor-bearing animals with metastatic lesions in the lungs as compared to untransfected cells or control transfectants, which did not produce metastases [18]. In addition, S100A4 expression in hormone dependent nonmetastatic MCF-7 breast cancer cells promoted hormone-independent tumor invasion of the surrounding muscle and adipose tissue, as well as metastasis to the lungs and lymph nodes [42 ] . Conversely the loss

96

R.P. House et al.

of S100A4 expression in highly metastatic cells is associated with a reduction in tumor metastasis. Transfection of metastatic osteosarcoma cells with either a S100A4 specific ribozyme or S100A4 shRNA resulted in a 50% reduction in the number of animals with metastatic lesions, and a 75–85% decrease in the number of metastatic lung nodules, respectively [31, 70]. Similarly, transfection or retroviral infection of Lewis lung carcinoma cells with a S100A4 antisense oligonucleotide caused a significant reduction in cell motility and invasion in vitro and a corresponding 50–90% decrease in the number of metastatic lung lesions in vivo [115]. S100A4 transgenic mice do not develop tumors, indicating that S100A4 itself is not tumorigenic [4, 20], but S100A4 overexpression will facilitate metastasis in an already tumorigenic background. The mammary epithelium of wild-type mice and mice overexpressing S100A4 under the control of the mouse mammary tumor promoter are phenotypically indistinguishable. However, S100A4 overexpression in the mammary epithelium of MMTV-neu mice, which express the HER2/neu oncogene, resulted in a fourfold increase in the number of lung metastases that was associated with a 66-fold increase in the affected areas of the lung as compared to MMTV-neu mice [20]. S100A4 overexpression also enhanced tumor metastasis in GRS/A mice, which normally have a high incidence of tumor formation, but a low rate of metastasis [120]. Crossing GRS/A and MMTV-S100A4 mice produced progeny that exhibited a tenfold increase in the number of animals that developed metastatic disease as compared to non-transgenic littermates [4]. Notably, S100A4 expression in tumor cells had only a modest affect on tumor latency [20]. Conversely, when transgenic mice expressing the PyMT antigen, which normally exhibit a high incidence of metastasis, are crossed with mice carrying null alleles for S100A4, they display a 60% reduction in the number of metastatic lesions [46, 136]. Altogether these animal studies provide strong evidence that in tumor cells, S100A4 functions as a factor that facilitates the progression of metastatic disease.

4.4

S100A4 and the Tumor Microenvironment

The tumor microenvironment consists of extracellular matrix and diverse cell types, including fibroblasts, myofibroblasts, endothelial cells, inflammatory and immune cells, and various bone marrow derived progenitor cells that support the growth, local invasion and metastatic dissemination of tumors [51, 52]. Although S100A4 expression in tumor cells clearly induces a metastatic phenotype, there is accumulating data that S100A4 expression in cells within the tumor microenvironment has an equally important role in promoting tumor progression. S100A4, which is also known as fibroblast-specific protein (FSP1), has been described as being expressed exclusively in fibroblasts [112], and is used frequently as a marker to identify cancer-associated fibroblasts [52, 141] and fibroblasts in tissues undergoing remodeling and EMT [50, 61, 142]. However, there is substantial data demonstrating S100A4 expression in a variety of stromal cell types, including macrophages

4 Moving Aggressively: S100A4 and Tumor Invasion

97

[66, 89], myeloid dendritic cells [12], and lymphocytes [24]. Moreover, immunohistochemistry of human breast cancer biopsies showed that S100A4 is expressed in multiple cell types within the tumor stroma, with high S100A4 expression levels detected in fibroblasts, macrophages, and activated lymphocytes [15]. Studies examining the contribution of stromal S100A4 to tumor development showed that orthotopic injection of highly metastatic, S100A4-expressing mouse mammary adenocarcinoma cells (CSML100) into S100A4−/− mice resulted in reduced tumor uptake and decreased tumor incidences compared to S100A4+/+ mice [45]. Moreover, mice that did develop tumors in the S100A4 null background did not develop metastatic disease, demonstrating that S100A4 expression in host stroma contributes to tumor progression. Interestingly, co-injection of CSML100 cells and immortalized S100A4+/+ mouse embryonic fibroblasts (MEFs) into S100A4−/− mice reduced the number of tumor free animals and restored the ability of the CSML100 cells to metastasize in a low percentage of animals (~12%), whereas co-injection of S100A4−/− MEFs had no effect on tumor development or metastasis [45]. While these studies demonstrate a role for host-derived S100A4 in driving tumor progression, the inability of S100A4+/+ MEFs to rescue the full metastatic potential of CSML100 cells suggests that S100A4 expression in additional stromal cells within the tumor microenvironment contributes to tumor progression. Cancer-associated fibroblasts (CAFs) promote tumor progression through paracrine signaling with both tumor cells and other cells present in the tumor microenvironment. CAFs secrete vascular endothelial growth factor (VEGF), a chemoattractant for endothelial cells that promotes angiogenesis [9, 52], as well as chemotactic factors and matrix metalloproteinases that aid in tumor growth, invasion and metastasis [9, 52]. Although S100A4+/+ MEFs partially restore the metastatic capabilities of tumor cells in S100A4-deficient mice, the mechanisms by which fibroblast-derived S100A4 promotes tumor metastasis are unknown. S100A4 null MEFs exhibit impaired 2D and 3D motility, thus the metastasis-promoting effects of S100A4+/+ MEFS may be a consequence of the motile properties of these cells and their ability to more efficiently invade the extracellular matrix [45]. In addition, breast tumor cells have been reported to stimulate the release of S100A4 from fibroblasts, thus S100A4 may exert its prometastatic effects as an extracellular factor [30, 105]. The loss of S100A4 expression also modulates the recruitment of immune cells to the neoplastic stroma. Using the PyMT breast cancer model, recent studies demonstrated a significant reduction in the recruitment of CD45+ leukocytes and CD3+ T lymphocytes to adenoma/MIN tumors of S100A4−/− mice [46]. Moreover, increased S100A4 expression was observed in the tumor stroma during the transition from adenoma/MIN to early carcinoma, which was coincident with the infiltration of T-lymphocytes. In an orthotopic breast cancer model, co-injection of CSML100 cells and S100A4+/+ MEFs into S100A4−/− mice increased the recruitment of T lymphocytes to the tumor [46]. These observations suggest that stromal S100A4 may facilitate T lymphocyte recruitment to the site of tumor growth, possibly through the release of S100A4 by cancer-associated fibroblasts. T cells also express S100A4 [15] and it is unknown if the loss of S100A4 has a direct and adverse impact on T cell functions that could affect recruitment to mammary tumors.

98

R.P. House et al.

In human tumors, the population of infiltrating T-lymphocytes can fluctuate greatly with respect to CD4/CD8 expression and Th1/Th2 regulatory status and significantly impact disease progression and patient survival [11, 56, 71]. Further classification of the infiltrating T lymphocytes in S100A4−/− animals will be important for establishing the mechanism by which stromal S100A4 establishes a tumor microenvironment that favors metastatic progression. CD45 is a marker for multiple cell types in the tumor microenvironment, including tumor-associated macrophages, TIE2-expressing monocytes, B cells and T cells [51]. Therefore it will be of interest to determine if the reduced recruitment of CD45+ leukocytes to mammary tumors in S100A4−/− mice reflects the loss of other bone-marrow derived cells from the tumor microenvironment. In breast cancer, tumor-associated macrophages promote tumor angiogenesis, invasion and metastasis [40, 68, 133]. Based on immunohistochemistry with the macrophage marker F4/80, there was a trend towards reduced numbers of tumor-associated macrophages in S100A4−/− adenoma/MIN tumors [46]. This observation is intriguing as recent studies in a model of liver fibrosis demonstrated that S100A4 is expressed in a specific subpopulation of macrophages that are characterized by increased expression of proinflammatory cytokines, chemokines and osteopontin, which is highly expressed in tumor-associated macrophages [85, 89]. In addition, studies with S100A4−/− bone marrow-derived macrophages showed that S100A4 is required for the regulation of colony stimulating factor-1 signaling, and macrophage recruitment and chemotaxis in vivo [66]. These observations support S100A4 as a regulator of macrophage function and warrant a more thorough evaluation of the role of S100A4 expression in modulating the tumor-promoting functions of macrophages. As there is accumulating evidence that distinct subpopulations of macrophages support tumor-associated angiogenesis, modulate tumor cell migration, invasion and intravasation, and enhance the seeding and establishment of metastatic cells [96, 133, 135], it will be important to examine how S100A4 regulates the biological activities of these specialized subsets of macrophages.

4.5

S100A4-Target Interactions

At present 16 S100A4/target protein interactions have been reported (Table 4.1). The current list of S100A4 binding partners are involved in diverse biological processes and include cytoskeletal proteins, signaling molecules, transcription factors and extracellular proteins and growth factors. Although in vitro data support the interaction of S100A4 with each of these targets, for the majority of these proteins the biological relevance of the interaction with S100A4 has not been examined. A common feature of S100A4 expression in both cancer and non-transformed cells is the observation that S100A4 levels correlate strongly with cell motility. For instance, fibroblasts and epithelial tumor cells that overexpress S100A4 display increased migratory properties [110, 114]. Conversely, genetic deletion of S100A4 reduces the motility of fibroblasts and macrophages [45, 66], and

4 Moving Aggressively: S100A4 and Tumor Invasion

99

Table 4.1 S100A4 targets Target Cellular function Reference Intracellular Myosin-IIA Cytoskeleton [29, 59, 67, 81] Tropomyosin Cytoskeleton [116] F-actin Cytoskeleton [67, 129] liprin b1 Scaffolding/cell adhesion [57] MetAP2 Post-translational processing [27] p53 Transcription [28, 43, 122] p63 Transcription [121] p73 Transcription [121] MDM2 Ubiquitination [123] Smads 2 and 3 Transcription [78] Extracellular CCN3 Matricellular signaling [64] Annexin II Signaling [108] Amphiregulin Growth factor [54] RAGE Cell surface receptor [138] Tag7 Innate immunity [24] MFAP 4 Cell adhesion [130] MetAP2 methionine aminopeptidase 2, MDM2 murine double minute oncogene, CCN3 cysteinerich 61-connective tissue growth factor-nephroblastoma overexpressed gene, RAGE receptor for advanced glycation end products, Tag7 peptidoglycan recognition protein, MFAP4 microfibrilassociated glycoprotein 4

reduction of S100A4 expression in tumor cells correlates with decreased cellular motility [10, 115]. Consistent with a role in regulating motility, immunofluorescence studies showed that S100A4 localizes to the leading edge of polarized breast cancer cells [53, 65], and proteomic studies demonstrated that S100A4 is enriched in the pseudopodia of migrating cells [127]. Moreover, using a fluorescent S100A4 biosensor that reports on the Ca2+-bound, activated form of S100A4 [33], we observed activated S100A4 in the leading edge and dorsal ruffles of breast cancer cells migrating into a wound (Fig. 4.3). Given the connection between S100A4 expression, the enhancement of cell motility and the ability of S100A4 to promote tumor invasion, we will focus our discussion on S100A4 targets involved in modulating cell migration.

4.6 4.6.1

Intracellular Targets Myosin-II

Three myosin-II heavy chain isoforms are expressed in nonmuscle cells, myosinIIA (MYH9), myosin-IIB (MYH10) and myosin-IIC (MYH14). Although the three myosin-II isoforms share 64–80% sequence identity [38], they exhibit distinct

100

R.P. House et al.

Fig. 4.3 S100A4 activation in human breast cancer cells crawling into a wound. (a and b) Ratio images of two representative MDA-MB-231 cells injected with a 1:2 ratio of Flu-S100A4:MeroS100A4. Purple to red signifies low to high S100A4 activation and the white dashed line denotes the location of the wound edge. (c–f) Higher magnification image of boxed regions in left panel showing phase contrast (c and d) and ratio images (e and f). Mero-S100A4 activation is observed in cellular regions extending towards the wound and in dorsal ruffles (arrowheads). Scale bar = 10 mm. More information on the S100A4 biosensor can be found in reference [33]

patterns of tissue and cell expression, have different enzymatic activities, interact with different proteins, and have unique functional roles in vivo (reviewed in [124]). More importantly, myosin-II regulates and integrates multiple steps during cell motility, including cell polarization and protrusion, and the assembly and turnover of adhesions [35, 125]. In animal models, myosin-II regulatory pathways are activated in invasive breast cancer cells [126], and myosin-II activity is required for matrix deformation and breast cancer cell motility through a three-dimensional matrix [93, 134]. These observations suggest that myosin-II-based contractility is a critical component of tumor cell invasion.

4 Moving Aggressively: S100A4 and Tumor Invasion

101

In vitro characterization of the S100A4/myosin-II interaction showed that S100A4 binds to both myosin-IIA and myosin-IIB in a Ca2+-dependent manner; however, the affinity of S100A4 for myosin-IIB is at least tenfold weaker than that observed for myosin-IIA (myosin-IIB: Kd = 23.1 ± 2.7 mM; myosin-IIA: Kd = 2.7 ± 0.6 mM) [29, 67]. The S100A4 binding site maps to the C-terminal end of the coiled-coil of the myosin-IIA heavy chain (residues 1908–1923) and S100A4 binding promotes the monomeric, unassembled state of myosin-IIA [58, 67]. Although the S100A4 binding site overlaps a PKC phosphorylation site at Ser1916, S100A4 binding is unaffected by PKC phosphorylation; however, phosphorylation on Ser1943, which is located downstream of the S100A4 binding site, inhibits S100A4 binding [25]. Thus myosin-IIA phosphorylation on Ser1943 provides an additional regulatory mechanism for modulating the S100A4/myosin-IIA interaction. In addition, recent fluorescence cross-correlation spectroscopy studies showed that S100A4 specifically depolymerized myosin-IIA filaments in heteroassemblies comprised of myosin-IIA and myosin-IIB filaments, demonstrating that S100A4 selectively targets myosin-IIA filaments [81]. With respect to cell migration, S100A4 modulates cellular motility by effecting the localization and stability of cellular protrusions. Phenotypic analysis of mammary adenocarcinoma cells undergoing directional migration showed that S100A4 expression resulted in the formation of extensive forward protrusions and a suppression of side protrusions, resulting in enhanced cell polarization during chemotaxis [65]. Moreover, these effects were due to the specific interaction of S100A4 with myosin-IIA. Microinjection of an antibody that recognizes the S100A4 binding site on myosin-IIA and promotes myosin-IIA disassembly, into cells that do not express S100A4 produced a similar increase in forward protrusions [65]. Consistent with these observations, CSF-1 stimulated S100A4−/− macrophages form numerous, highly unstable forward and side protrusions and, as a consequence, turn more frequently and migrate with less persistence than wild-type macrophages [66]. In addition, the loss of S100A4 increased myosin-IIA filament assembly in vivo, which is in agreement with in vitro studies demonstrating that S100A4 promotes myosin-IIA filament depolymerization. Together these data support a role for S100A4 in maintaining cell polarization during chemotactic migration and suggest that S100A4 may modulate cell invasion via the regulation of myosin-IIA assembly.

4.6.2

Cytoskeletal Elements

The actin-based cytoskeleton is involved in the maintenance of cell morphology and is essential for cell migration and invasion [94, 137]. S100A4 is reported to exhibit Ca2+-dependent binding to F-actin [67, 129] and to the nonmuscle tropomyosin isoform 2 (TM2) [116]. Multiple tropomyosin splice variants are expressed in nonmuscle cells, which exhibit different affinities for F-actin, modulate myosin motor activity to varying extents and regulate the activity of actin filament severing proteins such as gelsolin and ADF/cofilin [47, 88]. S100A4-mediated regulation of

102

R.P. House et al.

tropomyosin function would be expected to significantly affect actin dynamics as well as cell polarization and migration; however, future studies will be required to evaluate how S100A4 binding impacts TM2 function and the subsequent consequences on cell motility and invasion.

4.6.3

p53

Altered p53 expression either through mutation or down regulation is thought to occur in approximately one-half of human cancers [48, 49, 62]. Wild-type p53 is activated in response to cellular stresses such as hypoxia, temperature shock and DNA damage, and regulates the expression of genes involved in cell cycle arrest and apoptosis to maintain the integrity of the genome. The initial report of the S100A4/ p53 interaction mapped the S100A4 binding site to the p53 C-terminal negative regulatory domain (residues 364–393) [43]; however, recent reports indicate that S100A4 binds weakly to the negative regulatory domain (Kd = 700 ± 50 mM), and instead binds a peptide derived from the p53 tetramerization domain with moderate affinity (Kd = 17 ± 1 mM) [28]. Notably, subsequent analytical centrifugation and size exclusion chromatography studies showed that S100A4 did not form a stable complex with either the C-terminal 100 residues of p53, which encompasses the tetramerization domain, or with full-length p53 [122]. These observations bring into question the relevance of the putative S100A4/p53 interaction. The cellular effects of S100A4 on p53 function have been equally complex. An evaluation of p53-regulated pro-apoptotic and pro-survival genes in S100A4 transfected cells demonstrated that expression was up-regulated for some genes (e.g., Bax); down-regulated for other genes (e.g. p21/WAF, thrombospondin-1) and for some genes such as MDM-2 expression was down-regulated initially and then upregulated [43]. These studies are difficult to interpret as there is no clear correlation in the expression patterns of the pro-survival or pro-apoptotic genes. Moreover, the density of the cell cultures significantly affected the gene expression profiles, suggesting that S100A4-mediated regulation of p53 transcriptional activity may depend on environmental signals. Lastly, S100A4 was reported to induce apoptosis in tumor cells expressing wild-type p53, whereas apoptosis was not triggered in cells expressing mutated p53 [43]. These observations led to the proposal that the induction of apoptosis in tumor cells expressing wild-type p53 might be a mechanism in which S100A4 expression selects for tumor cells expressing mutated p53, thus allowing for the survival of tumor cells with greater invasive potential. However, studies in colon carcinoma and lung cancer lines demonstrated high S100A4 expression in cells expressing wild-type p53 and the loss of S100A4 expression in cells deficient for p53 or expressing mutated p53 [17, 87], thus there is no clear consensus on p53 mutational status and S100A4 expression in cancer cells. Although p53 is considered primarily a regulator of cell proliferation and apoptosis, there is growing evidence supporting the involvement of p53 in the suppression of cell motility and invasion. The loss of wild-type p53 or expression

4 Moving Aggressively: S100A4 and Tumor Invasion

103

of mutant p53 enhances cell migration and invasion [83]. Normally p53 levels are maintained by the ubiquitin E3 ligase MDM2 [14]. Recent studies demonstrated that S100A4 binds with relatively high affinity to the N-terminal domain of MDM2 (Kd = 2.5 ± 0.6 mM) and inhibits MDM2-mediated ubquitination of p53 [123], which could provide a mechanism for stabilizing wild-type or mutant p53 proteins. Interestingly, the expression of mutant p53 protein in mouse embryonic fibroblasts, melanoma and pancreatic cancer cells enhanced RhoA-ROCK signaling and increased invasion into a 3D matrix [32, 118]. Thus S100A4 could indirectly modulate cell migration by inhibiting MDM2 function and preventing down-regulation of p53.

4.7

Extracellular S100A4

There is accumulating evidence that S100A4 is released by multiple cell types, including smooth muscle cells, fibroblasts, macrophages, tumor cells and chondrocytes [15, 30, 91, 105, 130, 139]. In addition, extracellular S100A4 is detected in the tumor interstitial fluid collected from pieces of freshly dissected invasive breast carcinomas [15] and in the synovial fluid collected from individuals with rheumatoid arthritis and osteoarthritis [55]. Notably, S100A4 does not contain any of the known signal peptide sequences required for export via the endoplasmic reticulum/ Golgi-dependent secretory pathway and therefore is thought to be released via an unconventional secretory process. Indeed, recent studies suggested that S100A4 release from fibroblasts occurs through a mechanism that involves the shedding of plasma membrane microvesicles [30]. The application of exogenous recombinant S100A4 has been reported to elicit a variety of cellular responses in multiple cell types. In neuronal cells, exogenous application of S100A4 promoted neurite outgrowth [84]. Extracellular S100A4 also stimulated angiogenesis in a corneal vascularization model and enhanced endothelial cell invasion into a 3D matrix [3, 106]. Further characterization showed that extracellular S100A4 stimulated the production of MMP-13 in endothelial cells, suggesting that S100A4 contributes to the angiogenic process by facilitating the degradation of the extracellular matrix [106]. Recent studies have indicated that exogenous S100A4 mediates mammary branching morphogenesis via the induction of MMP-3 expression [5], suggesting a role for extracellular S100A4 in normal mammary development. Despite these varied cellular responses, the receptors responsible for S100A4 binding and their associated signal transduction pathways remain largely unknown; although, in tumor cells extracellular S100A4 has been shown to activate the transcription factor NF-kB [13, 44, 106], which is a major contributor to the development and progression of cancer. A potential issue with all of these studies on the effects of extracellular S100A4 is the use of recombinant N-terminal hexahistidine-tagged S100A4, which forms dimers (2 S100A4 monomers), tetramers and large oligomers that range in size from 143 to 200 kDa (approximately 10–15 S100A4 monomers) [84]. Notably,

104

R.P. House et al.

these large S100A4 multimers are not observed with recombinant untagged S100A4 [36, 73]; however, many of the cellular responses described above are associated specifically with the oligomeric forms of S100A4 rather than S100A4 dimers or tetramers [3, 84, 106]. Size exclusion chromatography of human plasma and synovial fluid demonstrated the presence of dimeric S100A4 and also identified S100A4 in fractions corresponding to high molecular weight protein species [3, 55]. Although these high molecular weight species were interpreted to be S100A4 oligomers, S100A4 elution profiles were evaluated with S100A4-specific antibodies without consideration or assessment of additional proteins that may co-elute with S100A4. Thus these immunological detection approaches cannot distinguish between S100A4 oligomers or complexes comprised of S100A4 bound to other unidentified proteins. In addition, recent studies with S100B showed that imidazole, which is used in the purification of hexahistidine-tagged proteins, acts as a trigger to induce the formation of S100B tetramers and higher order oligomers that are stable following imidazole removal [117]. Despite these concerns, it is highly noteworthy that a few studies have shown that the application of untagged recombinant S100A4 can elicit cellular responses such as the stimulation of glioma cell motility [7], endothelial cell capillary tube formation [108] and activation of NF-kB [138]. As it is unknown if imidazole specifically induces S100A4 oligomerization or if the hexahistidine tag promotes S100A4 aggregation, the biological activities observed following the application of extracellular hexahistidine-tagged S100A4 need to be confirmed with untagged S100A4 or with S100A4 purified from conditioned medium derived from appropriate stromal or tumor cells.

4.7.1

Annexin A2

Annexins are cytoplasmic proteins that bind negatively charged phospholipids in a Ca2+-dependent manner; however, they are also found on the surface of activated endothelial cells and some tumor cells. In conjunction with tissue plasminogen activator (tPA), extracellular annexin A2 facilitates the conversion of plasminogen to plasmin, a serine protease that mediates tissue fibrinolysis [34]. S100A4 binds to the N-terminus of annexin A2 in a Ca2+-independent manner and increases the catalytic efficiency of tPA-mediated conversion of plasminogen to plasmin both in vitro and in cell-based assays with human cerebromicrovascular cells [108]. Plasmin mediates the degradation of the extracellular matrix and also activates latent matrix metalloproteinases and proangiogenic factors to promote angiogenesis. The release of S100A4 by tumor or stromal cells and subsequent binding to annexin A2 on the endothelial cell surface could allow for a local increase in tPA-mediated production of plasmin, thus providing a mechanism for facilitating angiogenesis. Consistent with a role for S100A4 in tumor angiogenesis, the mammary tumors from GRS/A mice that overexpress S100A4 in the mammary epithelium exhibited increased blood vessel densities compared to non-transgenic mice [3], whereas early and late carcinomas from S100A4−/− PyMT mice displayed reduced vascular densities [46].

4 Moving Aggressively: S100A4 and Tumor Invasion

4.8

105

S100A4-Targeted Therapeutics

At present there is substantial data supporting the involvement of S100A4 in several aspects of tumor progression, including cell migration, invasion and angiogenesis. Moreover, the universality of S100A4 expression suggests that S100A4 contributes to metastasis and disease progression in a variety of cancers and highlights S100A4 as a potential target for therapeutic intervention. Recent efforts have identified inhibitors for several S100 family members, demonstrating that these small Ca2+binding proteins can be targeted with small molecules. A combination of computeraided drug design and high throughput fluorescence polarization assays identified small molecules that disrupt the interaction of S100B and p53 [75, 131], and structurebased virtual screening allowed for the discovery of compounds that inhibit the binding of S100A10 to annexin A2 [99]. For S100A4, a fluorescent biosensor (Mero-S100A4) that reports on the conformational rearrangements associated with Ca2+-binding and protein activation, was used to identify several phenothiazines that block S100A4-mediated depolymerization of myosin-IIA filaments [33]. Moreover, phenothiazine-mediation inhibition of S100A4 occurs by a novel mechanism involving the sequestration of S100A4 into small molecule-induced oligomers [72] (Fig. 4.4). The use of S100A4 as a therapeutic target has also attracted the attention of the pharmaceutical industry. SP-MET-X1, developed by Supratek Pharma, is a modified version of the anti-allergy drug Amlexanox, which has been shown to bind several S100 proteins [69, 86, 97]. Since Amlexanox has an excellent safety record and is used widely for the treatment of aphthous ulcers and bronchial asthma, it is anticipated that long-term treatment may be used to prevent metastatic disease. Although the S100 proteins exhibit significant sequence identity, there is some indication that selectivity can be achieved with respect to S100 inhibitors. For example, the S100B inhibitor pentamidine, which disrupts the binding of S100B to p53 [16, 75], has no effect on the S100A4/myosin-IIA interaction (A.R. Bresnick, unpublished observations). Given the contribution of S100 proteins to a multitude of biological processes, the issues and challenges of specificity and selectivity will continue to be an active area of investigation as new S100 inhibitors are discovered.

4.9

Conclusions

S100A4 is a proven mediator of metastatic dissemination; however, the molecular mechanisms by which S100A4 promotes tumor invasion are not understood. Deciphering the contribution of S100A4 to the metastatic process is complicated by the observations that tumor and stromal-derived, as well as intracellular and extracellular S100A4 all contribute to metastatic progression. Although a number of intracellular, cell surface and extracellular S100A4 protein targets have been described, the details of how S100A4 regulates the functions of these protein targets

106

R.P. House et al.

Fig. 4.4 Structures of S100 protein/inhibitor complexes. (a) Ribbon diagram of the Ca2+–S100B/ pentamidine complex. The two S100B monomers are shown in blue and green. Two pentamidine molecules (pink) are bound per S100B monomer. Calcium ions are shown as gray spheres. (b) Ribbon diagram of the Ca2+-S100A4/trifluoperazine oligomer. Trifluoperazine (pink) binding induces the assembly of five Ca2+-S100A4/trifluoperazine dimers into a pentameric ring. Calcium ions are shown as gray spheres

and how these interactions synergize and influence tumor and stromal cell activities has not been examined. Biochemical and biological analysis of the mechanisms through which S100A4 modulates its protein targets will be essential for understanding how S100A4 promotes disease progression not only in cancer, but in other

4 Moving Aggressively: S100A4 and Tumor Invasion

107

pathological conditions as well [41, 107]. Moreover, this mechanistic information can be applied to the development of S100A4-targeted therapies. Most cancer therapeutics block tumor cell proliferation; however, the successful treatment of cancer by surgery and chemotherapy is limited by metastatic disease. As a consequence, there is a recognized need for cancer therapeutics that directly target metastatic disease by inhibiting the invasive capabilities of tumor cells. In particular, the signaling pathways and proteins that regulate cell motility and invasion, such as S100A4, are attractive candidates for drug development.

References 1. Agerbaek M, Alsner J, Marcussen N et al (2006) Focal S100A4 protein expression is an independent predictor of development of metastatic disease in cystectomized bladder cancer patients. Eur Urol 50:777–785 2. Akke M, Drakenberg T, Chazin WJ (1992) Three-dimensional solution structure of Ca(2+)loaded porcine calbindin D9k determined by nuclear magnetic resonance spectroscopy. Biochemistry 31:1011–1020 3. Ambartsumian N, Klingelhofer J, Grigorian M et al (2001) The metastasis-associated Mts1(S100A4) protein could act as an angiogenic factor. Oncogene 20:4685–4695 4. Ambartsumian NS, Grigorian MS, Larsen IF et al (1996) Metastasis of mammary carcinomas in GRS/A hybrid mice transgenic for the mts1 gene. Oncogene 13:1621–1630 5. Andersen K, Mori H, Fata J et al (2011) The metastasis-promoting protein S100A4 regulates mammary branching morphogenesis. Dev Biol 352:181–190 6. Bandiera A, Melloni G, Freschi M et al (2009) Prognostic factors and analysis of S100a4 protein in resected pulmonary metastases from renal cell carcinoma. World J Surg 33: 1414–1420 7. Belot N, Pochet R, Heizmann CW et al (2002) Extracellular S100A4 stimulates the migration rate of astrocytic tumor cells by modifying the organization of their actin cytoskeleton. Biochim Biophys Acta 1600:74–83 8. Bhattacharya S, Large E, Heizmann CW et al (2003) Structure of the Ca2+/S100B/NDR kinase peptide complex: insights into S100 target specificity and activation of the kinase. Biochemistry 42:14416–14426 9. Bhowmick NA, Neilson EG, Moses HL (2004) Stromal fibroblasts in cancer initiation and progression. Nature 432:332–337 10. Bjornland K, Winberg JO, Odegaard OT et al (1999) S100A4 involvement in metastasis: deregulation of matrix metalloproteinases and tissue inhibitors of matrix metalloproteinases in osteosarcoma cells transfected with an anti-S100A4 ribozyme. Cancer Res 59:4702–4708 11. Bohling SD, Allison KH (2008) Immunosuppressive regulatory T cells are associated with aggressive breast cancer phenotypes: a potential therapeutic target. Mod Pathol 21:1527–1532 12. Boomershine CS, Chamberlain A, Kendall P et al (2009) Autoimmune pancreatitis results from loss of TGFbeta signalling in S100A4-positive dendritic cells. Gut 58:1267–1274 13. Boye K, Grotterod I, Aasheim HC et al (2008) Activation of NF-kappaB by extracellular S100A4: analysis of signal transduction mechanisms and identification of target genes. Int J Cancer 123:1301–1310 14. Brooks CL, Gu W (2006) p53 ubiquitination: Mdm2 and beyond. Mol Cell 21:307–315 15. Cabezon T, Celis JE, Skibshoj I et al (2007) Expression of S100A4 by a variety of cell types present in the tumor microenvironment of human breast cancer. Int J Cancer 121:1433–1444 16. Charpentier TH, Wilder PT, Liriano MA et al (2008) Divalent metal ion complexes of S100B in the absence and presence of pentamidine. J Mol Biol 382:56–73

108

R.P. House et al.

17. Daoud SS, Munson PJ, Reinhold W et al (2003) Impact of p53 knockout and topotecan treatment on gene expression profiles in human colon carcinoma cells: a pharmacogenomic study. Cancer Res 63:2782–2793 18. Davies BR, Davies MP, Gibbs FE et al (1993) Induction of the metastatic phenotype by transfection of a benign rat mammary epithelial cell line with the gene for p9Ka, a rat calciumbinding protein, but not with the oncogene EJ-ras-1. Oncogene 8:999–1008 19. Davies BR, O’donnell M, Durkan GC et al (2002) Expression of S100A4 protein is associated with metastasis and reduced survival in human bladder cancer. J Pathol 196:292–299 20. Davies MP, Rudland PS, Robertson L et al (1996) Expression of the calcium-binding protein S100A4 (p9Ka) in MMTV-neu transgenic mice induces metastasis of mammary tumours. Oncogene 13:1631–1637 21. De Silva RS, Martin L, Roshanlall C et al (2006) Association of S100A4 and osteopontin with specific prognostic factors and survival of patients with minimally invasive breast cancer. Clin Cancer Res 12:1192–1200 22. Donato R (2003) Intracellular and extracellular roles of S100 proteins. Microsc Res Tech 60:540–551 23. Drohat AC, Baldisseri DM, Rustandi RR et al (1998) Solution structure of calcium-bound rat S100B(betabeta) as determined by nuclear magnetic resonance spectroscopy. Biochemistry 37:2729–2740 24. Dukhanina EA, Kabanova OD, Lukyanova TI et al (2009) Opposite roles of metastasin (S100A4) in two potentially tumoricidal mechanisms involving human lymphocyte protein Tag7 and Hsp70. Proc Natl Acad Sci USA 106:13963–13967 25. Dulyaninova NG, Malashkevich VN, Almo SC et al (2005) Regulation of myosin-IIA assembly and Mts1 binding by heavy chain phosphorylation. Biochemistry 44:6867–6876 26. Ebralidze A, Tulchinsky E, Grigorian M et al (1989) Isolation and characterization of a gene specifically expressed in different metastatic cells and whose deduced gene product has a high degree of homology to a Ca2+-binding protein family. Genes Dev 3:1086–1093 27. Endo H, Takenaga K, Kanno T et al (2002) Methionine aminopeptidase 2 is a new target for the metastasis-associated protein, S100A4. J Biol Chem 277:26396–26402 28. Fernandez-Fernandez MR, Veprintsev DB, Fersht AR (2005) Proteins of the S100 family regulate the oligomerization of p53 tumor suppressor. Proc Natl Acad Sci USA 102:4735–4740 29. Ford HL, Silver DL, Kachar B et al (1997) Effect of Mts1 on the structure and activity of nonmuscle myosin II. Biochemistry 36:16321–16327 30. Forst B, Hansen MT, Klingelhofer J et al (2010) Metastasis-inducing S100A4 and RANTES cooperate in promoting tumor progression in mice. PLoS One 5:e10374 31. Fujiwara M, Kashima TG, Kunita A et al (2011) Stable knockdown of S100A4 suppresses cell migration and metastasis of osteosarcoma. Tumour Biol 32(3):611–622 32. Gadea G, De Toledo M, Anguille C et al (2007) Loss of p53 promotes RhoA-ROCKdependent cell migration and invasion in 3D matrices. J Cell Biol 178:23–30 33. Garrett SC, Hodgson L, Rybin A et al (2008) A biosensor of S100A4 metastasis factor activation: inhibitor screening and cellular activation dynamics. Biochemistry 47:986–996 34. Gerke V, Moss SE (2002) Annexins: from structure to function. Physiol Rev 82:331–371 35. Giannone G, Dubin-Thaler BJ, Rossier O et al (2007) Lamellipodial actin mechanically links myosin activity with adhesion-site formation. Cell 128:561–575 36. Gibbs FE, Wilkinson MC, Rudland PS et al (1994) Interactions in vitro of p9Ka, the rat S-100related, metastasis-inducing, calcium-binding protein. J Biol Chem 269:18992–18999 37. Gingras AR, Basran J, Prescott A et al (2008) Crystal structure of the Ca(2+)-form and Ca(2+)binding kinetics of metastasis-associated protein, S100A4. FEBS Lett 582:1651–1656 38. Golomb E, Ma X, Jana SS et al (2004) Identification and characterization of nonmuscle myosin II-C, a new member of the myosin II family. J Biol Chem 279:2800–2808 39. Gongoll S, Peters G, Mengel M et al (2002) Prognostic significance of calcium-binding protein S100A4 in colorectal cancer. Gastroenterology 123:1478–1484 40. Goswami S, Sahai E, Wyckoff JB et al (2005) Macrophages promote the invasion of breast carcinoma cells via a colony-stimulating factor-1/epidermal growth factor paracrine loop. Cancer Res 65:5278–5283

4 Moving Aggressively: S100A4 and Tumor Invasion

109

41. Grigorian M, Ambartsumian N, Lukanidin E (2008) Metastasis-inducing S100A4 protein: implication in non-malignant human pathologies. Curr Mol Med 8:492–496 42. Grigorian M, Ambartsumian N, Lykkesfeldt AE et al (1996) Effect of mts1 (S100A4) expression on the progression of human breast cancer cells. Int J Cancer 67:831–841 43. Grigorian M, Andresen S, Tulchinsky E et al (2001) Tumor suppressor p53 protein is a new target for the metastasis-associated Mts1/S100A4 protein: functional consequences of their interaction. J Biol Chem 276:22699–22708 44. Grotterod I, Maelandsmo GM, Boye K (2010) Signal transduction mechanisms involved in S100A4-induced activation of the transcription factor NF-kappaB. BMC Cancer 10:241 45. Grum-Schwensen B, Klingelhofer J, Berg CH et al (2005) Suppression of tumor development and metastasis formation in mice lacking the S100A4(mts1) gene. Cancer Res 65:3772–3780 46. Grum-Schwensen B, Klingelhofer J, Grigorian M et al (2010) Lung metastasis fails in MMTV-PyMT oncomice lacking S100A4 due to a T-cell deficiency in primary tumors. Cancer Res 70:936–947 47. Gunning PW, Schevzov G, Kee AJ et al (2005) Tropomyosin isoforms: divining rods for actin cytoskeleton function. Trends Cell Biol 15:333–341 48. Hollstein M, Rice K, Greenblatt MS et al (1994) Database of p53 gene somatic mutations in human tumors and cell lines. Nucleic Acids Res 22:3551–3555 49. Hollstein M, Sidransky D, Vogelstein B et al (1991) p53 mutations in human cancers. Science 253:49–53 50. Iwano M, Plieth D, Danoff TM et al (2002) Evidence that fibroblasts derive from epithelium during tissue fibrosis. J Clin Invest 110:341–350 51. Joyce JA, Pollard JW (2009) Microenvironmental regulation of metastasis. Nat Rev Cancer 9:239–252 52. Kalluri R, Zeisberg M (2006) Fibroblasts in cancer. Nat Rev Cancer 6:392–401 53. Kim EJ, Helfman DM (2003) Characterization of the metastasis-associated protein, S100A4. Roles of calcium binding and dimerization in cellular localization and interaction with myosin. J Biol Chem 278:30063–30073 54. Klingelhofer J, Moller HD, Sumer EU et al (2009) Epidermal growth factor receptor ligands as new extracellular targets for the metastasis-promoting S100A4 protein. FEBS J 276:5936–5948 55. Klingelhofer J, Senolt L, Baslund B et al (2007) Up-regulation of metastasis-promoting S100A4 (Mts-1) in rheumatoid arthritis: putative involvement in the pathogenesis of rheumatoid arthritis. Arthritis Rheum 56:779–789 56. Kohrt HE, Nouri N, Nowels K et al (2005) Profile of immune cells in axillary lymph nodes predicts disease-free survival in breast cancer. PLoS Med 2:e284 57. Kriajevska M, Fischer-Larsen M, Moertz E et al (2002) Liprin beta 1, a member of the family of LAR transmembrane tyrosine phosphatase-interacting proteins, is a new target for the metastasis-associated protein S100A4 (Mts1). J Biol Chem 277:5229–5235 58. Kriajevska M, Tarabykina S, Bronstein I et al (1998) Metastasis-associated Mts1 (S100A4) protein modulates protein kinase C phosphorylation of the heavy chain of nonmuscle myosin. J Biol Chem 273:9852–9856 59. Kriajevska MV, Cardenas MN, Grigorian MS et al (1994) Non-muscle myosin heavy chain as a possible target for protein encoded by metastasis-related mts-1 gene. J Biol Chem 269:19679–19682 60. Kwak JM, Lee HJ, Kim SH et al (2010) Expression of protein S100A4 is a predictor of recurrence in colorectal cancer. World J Gastroenterol 16:3897–3904 61. Lawson WE, Polosukhin VV, Zoia O et al (2005) Characterization of fibroblast-specific protein 1 in pulmonary fibrosis. Am J Respir Crit Care Med 171:899–907 62. Lee WY, Su WC, Lin PW et al (2004) Expression of S100A4 and Met: potential predictors for metastasis and survival in early-stage breast cancer. Oncology 66:429–438 63. Lee YT, Dimitrova YN, Schneider G et al (2008) Structure of the S100A6 complex with a fragment from the C-terminal domain of Siah-1 interacting protein: a novel mode for S100 protein target recognition. Biochemistry 47:10921–10932 64. Li CL, Martinez V, He B et al (2002) A role for CCN3 (NOV) in calcium signalling. Mol Pathol 55:250–261

110

R.P. House et al.

65. Li ZH, Bresnick AR (2006) The S100A4 metastasis factor regulates cellular motility via a direct interaction with myosin-IIA. Cancer Res 66:5173–5180 66. Li ZH, Dulyaninova NG, House RP et al (2010) S100A4 regulates macrophage chemotaxis. Mol Biol Cell 21:2598–2610 67. Li ZH, Spektor A, Varlamova O et al (2003) Mts1 regulates the assembly of nonmuscle myosin-IIA. Biochemistry 42:14258–14266 68. Lin EY, Li JF, Gnatovskiy L et al (2006) Macrophages regulate the angiogenic switch in a mouse model of breast cancer. Cancer Res 66:11238–11246 69. Mack GS, Marshall A (2010) Lost in migration. Nat Biotechnol 28:214–229 70. Maelandsmo GM, Hovig E, Skrede M et al (1996) Reversal of the in vivo metastatic phenotype of human tumor cells by an anti-CAPL (mts1) ribozyme. Cancer Res 56:5490–5498 71. Mahmoud SM, Paish EC, Powe DG et al (2011) Tumor-infiltrating CD8+ lymphocytes predict clinical outcome in breast cancer. J Clin Oncol 29:1949–1955 72. Malashkevich VN, Dulyaninova NG, Ramagopal UA et al (2010) Phenothiazines inhibit S100A4 function by inducing protein oligomerization. Proc Natl Acad Sci USA 107: 8605–8610 73. Malashkevich VN, Varney KM, Garrett SC et al (2008) Structure of Ca2+-bound S100A4 and its interaction with peptides derived from nonmuscle myosin-IIA. Biochemistry 47:5111–5126 74. Marenholz I, Heizmann CW, Fritz G (2004) S100 proteins in mouse and man: from evolution to function and pathology (including an update of the nomenclature). Biochem Biophys Res Commun 322:1111–1122 75. Markowitz J, Chen I, Gitti R et al (2004) Identification and characterization of small molecule inhibitors of the calcium-dependent S100B-p53 tumor suppressor interaction. J Med Chem 47:5085–5093 76. Markowitz J, Rustandi RR, Varney KM et al (2005) Calcium-binding properties of wild-type and EF-hand mutants of S100B in the presence and absence of a peptide derived from the C-terminal negative regulatory domain of p53. Biochemistry 44:7305–7314 77. Matsubara D, Niki T, Ishikawa S et al (2005) Differential expression of S100A2 and S100A4 in lung adenocarcinomas: clinicopathological significance, relationship to p53 and identification of their target genes. Cancer Sci 96:844–857 78. Matsuura I, Lai CY, Chiang KN (2010) Functional interaction between Smad3 and S100A4 (metastatin-1) for TGF-beta-mediated cancer cell invasiveness. Biochem J 426:327–335 79. Mckiernan E, Mcdermott EW, Evoy D et al (2011) The role of S100 genes in breast cancer progression. Tumour Biol 32:441–450 80. Min HS, Choe G, Kim SW et al (2008) S100A4 expression is associated with lymph node metastasis in papillary microcarcinoma of the thyroid. Mod Pathol 21:748–755 81. Mitsuhashi M, Sakata H, Kinjo M et al (2011) Dynamic assembly properties of nonmuscle myosin II isoforms revealed by combination of fluorescence correlation spectroscopy and fluorescence cross-correlation spectroscopy. J Biochem 149:253–263 82. Moore BW (1965) A soluble protein characteristic of the nervous system. Biochem Biophys Res Commun 19:739–744 83. Muller PA, Vousden KH, Norman JC (2011) p53 and its mutants in tumor cell migration and invasion. J Cell Biol 192:209–218 84. Novitskaya V, Grigorian M, Kriajevska M et al (2000) Oligomeric forms of the metastasisrelated Mts1 (S100A4) protein stimulate neuronal differentiation in cultures of rat hippocampal neurons. J Biol Chem 275:41278–41286 85. Ojalvo LS, King W, Cox D et al (2009) High-density gene expression analysis of tumorassociated macrophages from mouse mammary tumors. Am J Pathol 174:1048–1064 86. Okada M, Tokumitsu H, Kubota Y et al (2002) Interaction of S100 proteins with the antiallergic drugs, olopatadine, amlexanox, and cromolyn: identification of putative drug binding sites on S100A1 protein. Biochem Biophys Res Commun 292:1023–1030 87. Orre LM, Pernemalm M, Lengqvist J et al (2007) Up-regulation, modification, and translocation of S100A6 induced by exposure to ionizing radiation revealed by proteomics profiling. Mol Cell Proteomics 6:2122–2131

4 Moving Aggressively: S100A4 and Tumor Invasion

111

88. Ostap EM (2008) Tropomyosins as discriminators of myosin function. Adv Exp Med Biol 644:273–282 89. Osterreicher CH, Penz-Osterreicher M, Grivennikov SI et al (2011) Fibroblast-specific protein 1 identifies an inflammatory subpopulation of macrophages in the liver. Proc Natl Acad Sci USA 108:308–313 90. Otterbein LR, Kordowska J, Witte-Hoffmann C et al (2002) Crystal structures of S100A6 in the Ca(2+)-free and Ca(2+)-bound states: the calcium sensor mechanism of S100 proteins revealed at atomic resolution. Structure 10:557–567 91. Pedersen KB, Andersen K, Fodstad O et al (2004) Sensitization of interferon-gamma induced apoptosis in human osteosarcoma cells by extracellular S100A4. BMC Cancer 4:52 92. Pedersen KB, Nesland JM, Fodstad O et al (2002) Expression of S100A4, E-cadherin, alphaand beta-catenin in breast cancer biopsies. Br J Cancer 87:1281–1286 93. Poincloux R, Collin O, Lizarraga F et al (2011) Contractility of the cell rear drives invasion of breast tumor cells in 3D Matrigel. Proc Natl Acad Sci USA 108:1943–1948 94. Pollard TD, Cooper JA (2009) Actin, a central player in cell shape and movement. Science 326:1208–1212 95. Potts BC, Smith J, Akke M et al (1995) The structure of calcyclin reveals a novel homodimeric fold for S100 Ca(2+)-binding proteins. Nat Struct Biol 2:790–796 96. Qian BZ, Pollard JW (2010) Macrophage diversity enhances tumor progression and metastasis. Cell 141:39–51 97. Rani SG, Mohan SK, Yu C (2010) Molecular level interactions of S100A13 with amlexanox: inhibitor for formation of the multiprotein complex in the nonclassical pathway of acidic fibroblast growth factor. Biochemistry 49:2585–2592 98. Ravasi T, Hsu K, Goyette J et al (2004) Probing the S100 protein family through genomic and functional analysis. Genomics 84:10–22 99. Reddy TR, Li C, Guo X et al (2011) Design, synthesis, and structure-activity relationship exploration of 1-substituted 4-aroyl-3-hydroxy-5-phenyl-1H-pyrrol-2(5H)-one analogues as inhibitors of the annexin A2-S100A10 protein interaction. J Med Chem 54:2080–2094 100. Rety S, Sopkova J, Renouard M et al (1999) The crystal structure of a complex of p11 with the annexin II N-terminal peptide. Nat Struct Biol 6:89–95 101. Rosty C, Ueki T, Argani P et al (2002) Overexpression of S100A4 in pancreatic ductal adenocarcinomas is associated with poor differentiation and DNA hypomethylation. Am J Pathol 160:45–50 102. Rudland PS, Platt-Higgins A, Renshaw C et al (2000) Prognostic significance of the metastasis-inducing protein S100A4 (p9Ka) in human breast cancer. Cancer Res 60:1595–1603 103. Rustandi RR, Drohat AC, Baldisseri DM et al (1998) The Ca(2+)-dependent interaction of S100B(beta beta) with a peptide derived from p53. Biochemistry 37:1951–1960 104. Santamaria-Kisiel L, Rintala-Dempsey AC, Shaw GS (2006) Calcium-dependent and -independent interactions of the S100 protein family. Biochem J 396:201–214 105. Schmidt-Hansen B, Klingelhofer J, Grum-Schwensen B et al (2004) Functional significance of metastasis-inducing S100A4(Mts1) in tumor-stroma interplay. J Biol Chem 279: 24498–24504 106. Schmidt-Hansen B, Ornas D, Grigorian M et al (2004) Extracellular S100A4(mts1) stimulates invasive growth of mouse endothelial cells and modulates MMP-13 matrix metalloproteinase activity. Oncogene 23:5487–5495 107. Schneider M, Hansen JL, Sheikh SP (2008) S100A4: a common mediator of epithelialmesenchymal transition, fibrosis and regeneration in diseases? J Mol Med 86:507–522 108. Semov A, Moreno MJ, Onichtchenko A et al (2005) Metastasis-associated protein S100A4 induces angiogenesis through interaction with Annexin II and accelerated plasmin formation. J Biol Chem 280:20833–20841 109. Steeg PS (2006) Tumor metastasis: mechanistic insights and clinical challenges. Nat Med 12:895–904 110. Stein U, Arlt F, Walther W et al (2006) The metastasis-associated gene S100A4 is a novel target of beta-catenin/T-cell factor signaling in colon cancer. Gastroenterology 131:1486–1500

112

R.P. House et al.

111. Stetler-Stevenson WG, Aznavoorian S, Liotta LA (1993) Tumor cell interactions with the extracellular matrix during invasion and metastasis. Annu Rev Cell Biol 9:541–573 112. Strutz F, Okada H, Lo CW et al (1995) Identification and characterization of a fibroblast marker: FSP1. J Cell Biol 130:393–405 113. Strynadka NC, James MN (1989) Crystal structures of the helix-loop-helix calcium-binding proteins. Annu Rev Biochem 58:951–998 114. Takenaga K, Nakamura Y, Endo H et al (1994) Involvement of S100-related calcium-binding protein pEL98 (or mts1) in cell motility and tumor cell invasion. Jpn J Cancer Res 85: 831–839 115. Takenaga K, Nakamura Y, Sakiyama S (1997) Expression of antisense RNA to S100A4 gene encoding an S100-related calcium-binding protein suppresses metastatic potential of highmetastatic Lewis lung carcinoma cells. Oncogene 14:331–337 116. Takenaga K, Nakamura Y, Sakiyama S et al (1994) Binding of pEL98 protein, an S100related calcium-binding protein, to nonmuscle tropomyosin. J Cell Biol 124:757–768 117. Thulin E, Kesvatera T, Linse S (2011) Molecular determinants of S100B oligomer formation. PLoS One 6:e14768 118. Timpson P, Mcghee EJ, Morton JP et al (2011) Spatial regulation of RhoA activity during pancreatic cancer cell invasion driven by mutant p53. Cancer Res 71:747–757 119. Vallely KM, Rustandi RR, Ellis KC et al (2002) Solution structure of human Mts1 (S100A4) as determined by NMR spectroscopy. Biochemistry 41:12670–12680 120. Van Der Valk MA (1981) Survival, tumor incidence, and gross pathology in 33 mouse strains. In: Hilgers J, Sluyser M (eds) Mammary tumors in the mouse. Elsevier, Amsterdam, pp 46–115 121. Van Dieck J, Brandt T, Teufel DP et al (2010) Molecular basis of S100 proteins interacting with the p53 homologs p63 and p73. Oncogene 29:2024–2035 122. Van Dieck J, Fernandez-Fernandez MR, Veprintsev DB et al (2009) Modulation of the oligomerization state of p53 by differential binding of proteins of the S100 family to p53 monomers and tetramers. J Biol Chem 284:13804–13811 123. Van Dieck J, Lum JK, Teufel DP et al (2010) S100 proteins interact with the N-terminal domain of MDM2. FEBS Lett 584:3269–3274 124. Vicente-Manzanares M, Ma X, Adelstein RS et al (2009) Non-muscle myosin II takes centre stage in cell adhesion and migration. Nat Rev Mol Cell Biol 10:778–790 125. Vicente-Manzanares M, Zareno J, Whitmore L et al (2007) Regulation of protrusion, adhesion dynamics, and polarity by myosins IIA and IIB in migrating cells. J Cell Biol 176:573–580 126. Wang W, Wyckoff JB, Goswami S et al (2007) Coordinated regulation of pathways for enhanced cell motility and chemotaxis is conserved in rat and mouse mammary tumors. Cancer Res 67:3505–3511 127. Wang Y, Ding SJ, Wang W et al (2007) Profiling signaling polarity in chemotactic cells. Proc Natl Acad Sci USA 104:8328–8333 128. Wang YY, Ye ZY, Zhao ZS et al (2010) High-level expression of S100A4 correlates with lymph node metastasis and poor prognosis in patients with gastric cancer. Ann Surg Oncol 17:89–97 129. Watanabe Y, Usada N, Minami H et al (1993) Calvasculin, as a factor affecting the microfilament assemblies in rat fibroblasts transfected by src gene. FEBS Lett 324:51–55 130. Watanabe Y, Usuda N, Tsugane S et al (1992) Calvasculin, an encoded protein from mRNA termed pEL-98, 18A2, 42A, or p9Ka, is secreted by smooth muscle cells in culture and exhibits Ca(2+)-dependent binding to 36-kDa microfibril-associated glycoprotein. J Biol Chem 267:17136–17140 131. Wilder PT, Charpentier TH, Liriano MA et al (2010) In vitro screening and structural characterization of inhibitors of the S100B-p53 interaction. Int J High Throughput Screen 2010:109–126 132. Wright NT, Varney KM, Ellis KC et al (2005) The three-dimensional solution structure of Ca(2+)-bound S100A1 as determined by NMR spectroscopy. J Mol Biol 353:410–426 133. Wyckoff J, Wang W, Lin EY et al (2004) A paracrine loop between tumor cells and macrophages is required for tumor cell migration in mammary tumors. Cancer Res 64: 7022–7029

4 Moving Aggressively: S100A4 and Tumor Invasion

113

134. Wyckoff JB, Pinner SE, Gschmeissner S et al (2006) ROCK- and myosin-dependent matrix deformation enables protease-independent tumor-cell invasion in vivo. Curr Biol 16:1515–1523 135. Wyckoff JB, Wang Y, Lin EY et al (2007) Direct visualization of macrophage-assisted tumor cell intravasation in mammary tumors. Cancer Res 67:2649–2656 136. Xue C, Plieth D, Venkov C et al (2003) The gatekeeper effect of epithelial-mesenchymal transition regulates the frequency of breast cancer metastasis. Cancer Res 63:3386–3394 137. Yamaguchi H, Condeelis J (2007) Regulation of the actin cytoskeleton in cancer cell migration and invasion. Biochim Biophys Acta 1773:642–652 138. Yammani RR, Carlson CS, Bresnick AR et al (2006) Increase in production of matrix metalloproteinase 13 by human articular chondrocytes due to stimulation with S100A4: role of the receptor for advanced glycation end products. Arthritis Rheum 54:2901–2911 139. Yammani RR, Long D, Loeser RF (2009) Interleukin-7 stimulates secretion of S100A4 by activating the JAK/STAT signaling pathway in human articular chondrocytes. Arthritis Rheum 60:792–800 140. Yonemura Y, Endou Y, Kimura K et al (2000) Inverse expression of S100A4 and E-cadherin is associated with metastatic potential in gastric cancer. Clin Cancer Res 6:4234–4242 141. Zeisberg EM, Potenta S, Xie L et al (2007) Discovery of endothelial to mesenchymal transition as a source for carcinoma-associated fibroblasts. Cancer Res 67:10123–10128 142. Zeisberg M, Yang C, Martino M et al (2007) Fibroblasts derive from hepatocytes in liver fibrosis via epithelial to mesenchymal transition. J Biol Chem 282:23337–23347

Chapter 5

Regulation of TGF-b Signaling and Metastatic Progression by Tumor Microenvironments Michael K. Wendt and William P. Schiemann

Abstract The TGF-b signaling system comprises a complex and dynamic cascade of molecular interactions that invoke a variety of intracellular and extracellular reactions that coalesce to maintain tissue homeostasis. A rapidly accumulating body of scientific literature clearly demonstrates a conversion in TGF-b function from that of a powerful tumor suppressor in normal epithelium and early-stage carcinomas to that of a prometastatic molecule in their late-stage counterparts. Collectively, this malicious switch in TGF-b behavior is termed the “TGF-b Paradox.” Historically, cell autonomous changes that transpire during tumor development and progression have been studied extensively as a means to decipher the “TGF-b Paradox.” Although highly informative and intriguing, these findings have yet to unravel the molecular underpinnings of the “TGF-b Paradox,” thereby suggesting involvement of additional signaling components and players that originate beyond the confines of developing carcinomas. Indeed, recent studies have been directed at interrogating the microenvironments of developing carcinomas and how changes within this unique cellular niche manifest the “TGF-b Paradox.” For instance, tumor microenvironments house an array of essential cellular, structural, and humoral factors that include stromal cells and altered elastic moduli, integrins and their engagement of matrix proteins, hypoxic zones, and a host of cytokines, growth factors, and chemokines that collectively influence the response of carcinoma cells to TGF-b. Here we review recent findings demonstrating the importance of the tumor microenvironment to regulate TGF-b signaling and its stimulation of metastatic progression. In addition, we also highlight recent in vitro and in vivo scientific advances capable of recapitulating various aspects of the metastatic process and its regulation by TGF-b.

M.K. Wendt • W.P. Schiemann (*) Division of General Medical Sciences-Oncology, Case Comprehensive Cancer Center, Case Western Reserve University, Wolstein Research Building, 2103 Cornell Road, Cleveland, OH 44106, USA e-mail: [email protected]

A. Fatatis (ed.), Signaling Pathways and Molecular Mediators in Metastasis, DOI 10.1007/978-94-007-2558-4_5, © Springer Science+Business Media B.V. 2012

115

116

M.K. Wendt and W.P. Schiemann

Indeed, incorporating and extending these novel systems to analyses of the “TGF-b Paradox” may offer new inroads in resolving this enigma and improving the overall survival of cancer patients.

Abbreviations 2D 3D CSF-1 Dab2 ECM EGF EMT ERK FAK Hgs IHC JNK LOX MAPK MEC MMP PTK RBM SARA TAK-1 TGF-b TbR-I TbR-II TbR-III VEGF

5.1

2-Dimensional 3-Dimensional Colony stimulating factor-1 Disabled-2 Extracellular matrix Epidermal growth factor Epithelial-mesenchymal transition Extracellular signal-regulated kinase Focal adhesion kinase Hepatocyte growth factor-regulated tyrosine kinase substrate Immunohistochemistry c-Jun N-terminal kinase Lysyl oxidase Mitogen-activated protein kinase Mammary epithelial cell Matrix metalloproteinase Protein tyrosine kinase Reconstituted basement membrane Smad anchor for receptor activation TGF-b-activated kinase 1 Transforming growth factor-b TGF-b type I receptor TGF-b type II receptor TGF-b type III receptor Vascular endothelial growth factor

Introduction

Transforming growth factor-b (TGF-b) is a ubiquitously expressed cytokine that regulates an assortment of biological activities in essentially all cell types and tissues. Besides its role in regulating cell development, differentiation, and survival, TGF-b also inhibits the proliferation of epithelial, endothelial, and hematopoietic cell lineages [1–4]. Interestingly, resistance to TGF-b-mediated cytostasis is a hallmark of neoplastic transformation, which ultimately transform the signals produced by this cytokine into oncogenic activities, particularly enhanced cancer cell invasion

5

Regulation of TGF-β Signaling and Metastatic Progression…

117

and metastasis [2, 5]. This conversion in TGF-b function is known as the “TGF-b Paradox” [5–7], which underlies the lethality of TGF-b in developing carcinomas, including their acquisition of EMT, metastatic, and chemoresistant phenotypes [5, 8, 9]. The molecular mechanisms whereby TGF-b mediates tumor suppression in normal and neoplastic cells remain to be fully elucidated. Likewise, how TGF-b promotes tumor progression in late-stage carcinomas is even more mysterious. Scientific investigations into TGF-b action have historically applied “cell-centric” approaches to interrogate the “TGF-b Paradox.” Indeed, these analyses have relied heavily on the differential expression of gene transcripts, proteins, and microRNAs between normal and malignant cells as a potential means to explain the dichotomous of TGF-b in responsive cells. Although these studies have yielded enormous volumes of data related to potential alterations in the coupling of TGF-b to its downstream effectors [10–12], they have failed to accurately recapitulate the “TGF-b Paradox” and unlock its ability to promote the development and progression of human tumors. In light of these limitations, recent studies are now actively investigating the impact that dynamic microenvironment alterations play in dictating the behaviors of TGF-b during metastatic progression. Here we review recent findings that elucidate novel mechanisms whereby tumor microenvironments manifest the “TGF-b Paradox.” Likewise, we also expound the virtues of novel 3-dimensional (3D)-organotypic culture techniques to recapitulate the changing microenvironments encountered by metastatic cancer cells as they disseminate throughout the body to distant locales.

5.2

The TGF-b Signaling System

Mammals express three genetically distinct TGF-b ligands (e.g., TGF-bs 1–3) whose mature and biologically active forms are ~97% identical and exhibit virtually indistinguishable actions in vitro [1, 13]. Interestingly, individual TGF-b ligands are expressed in a spatiotemporal manner during embryogenesis and tissue morphogenesis, which contributes to the array of diverse and nonredundant phenotypes displayed by mice lacking distinct TGF-b isoforms [14]. Transmembrane signaling by TGF-b is propagated by its binding to three high-affinity receptors, TGF-b type I (TbR-I), type II (TbR-II), and type III (TbR-III or betaglycan). When expressed, TbR-III is the most abundant TGF-b receptor present on the cell surface where it was originally believed to function solely as an accessory receptor that binds and modulates TGF-b function in responsive cells. However, recent findings implicate TbR-III as an essential mediator of the tumor suppressing activities of TGF-b. Indeed, loss of TbR-III expression associates with disease progression and poor overall survival in patients with cancers of the breast, ovary, prostate, lung, pancreas, kidney, and endometrium [15], suggesting that measures capable of elevating TbR-III expression or activity may circumvent the “TGF-b Paradox” and its initiation of oncogenic TGF-b signaling. Although TbR-III lacks intrinsic enzymatic activity, TbR-I and TbR-II both possess Ser/Thr protein kinases in their cytoplasmic

118

M.K. Wendt and W.P. Schiemann

domains that serve to initiate downstream signaling [2, 3, 16–18]. In fact, the binding of TGF-b to TbR-II enables this polypeptide to transphosphorylate and activate TbR-I, which subsequently binds, phosphorylates, and stimulates the latent transcription factors, Smad2 and Smad3 [2, 3, 19]. Once phosphorylated, Smads 2 and 3 rapidly form higher order complexes with the common Smad, Smad4, with the resulting heterotrimeric complexes that rapidly translocate and accumulate in the nucleus where they cooperate with an ever expanding list of transcriptional activators or repressors to govern gene expression in a gene- and cell-specific manner [2, 3, 16–18]. Changes in cell behavior regulated by the activation of Smad2/3 is referred to as “canonical TGF-b signaling” and is modulated in all subcellular compartments by numerous effector molecules. For example, the specific recruitment and phosphorylation of Smad2/3 by TbR-I is regulated by several adaptor proteins, including SARA (Smad Anchor for Receptor Activation; [20]), Hgs (Hepatocyte Growth factor-regulated tyrosine kinase Substrate; [21]), and Dab2 (Disabled-2; [22]). Incorporation of the inhibitory Smad, Smad7, into active TGF-b receptor complexes displaces Smad2/3 and prevents their phosphorylation and activation by TbR-I [23–25]. Elevated Smad7 expression also inhibits TGF-b signaling through its ability to recruit Smurf E3 ubiquitin ligases, which promote the ubiquitination and degradation of TbR-I [26, 27]. Smad2/3 signaling is also governed in the nucleus by their association with a variety of interacting proteins, including ATF-3, Sp-1, AP-1, Jun/Fos, Stats, and members of the Forkhead family of transcription factors [2, 3, 16–18]. Finally, Smad2/3 signaling is also subject to fine-tuning following their phosphorylation by cellular protein kinases, as well as to inactivation by their dephosphorylation and ubiquitination by phosphatases and E3 ubiquitin ligases, respectively [28]. Figure 5.1 depicts the TGF-b signaling system and the potential role of Smads 2 and 3 in regulating metastatic progression driven by TGF-b. Additional details linking Smad2/3 to tumor metastasis are discussed in the following sections. TGF-b also couples to a variety of noncanonical signaling systems (i.e., Smad2/3-independent), including ERK1/2, p38 MAPK, JNK, AKT, the small GTP-binding proteins Rho, Rac, and Cdc42 (see [5, 9, 17, 29]). The specific sequence of events and molecular mechanisms that couple TGF-b receptors to the activation of these pathways remain to be fully delineated. However, we recently defined a novel signaling axis comprised of avb3 integrin:Src:FAK:p130Cas:TbR-II:Grb2 that is critical for TGF-b stimulation of MAP kinases, EMT, and pulmonary metastasis of breast cancer cells [30–32]. Additional evidence suggests that the ability of TGF-b to activate MAPKs can also emanate from TbR-I and its physical interaction with the E3 ubiquitin ligase, TRAF6, which ubiquitinates itself and the MAPK kinase kinase TAK-1 (TGF-b-activated kinase 1) to facilitate JNK and p38 MAPK activation [33, 34]. TGF-b displays a dichotomous connection with NF-kB, whose transcriptional activity is inhibited by TGF-b in normal epithelial cells [35], but is robustly stimulated by this cytokine in their transformed counterparts [36–38]. The inverted coupling of TGF-b to NF-kB in normal and malignant cells may reflect a loss of TbR-III expression, which suppresses NF-kB activity through the formation of TbR-III:b-arrestin-2:NF-kB complexes [39]. Likewise, the coupling

5

Regulation of TGF-β Signaling and Metastatic Progression…

119

Fig. 5.1 The TGF-b signaling system during metastatic progression. TGF-b initiates transmembrane signaling by binding to its three cell surface receptors, TbR-I, TbR-II, and TbR-III. Once activated, TbR-I phosphorylates Smads 2 and 3, which form hetrotrimeric complexes with Smad4 that translocate to the nucleus to regulate gene expression. The transduction of messages via Smad2/3 constitutes the canonical arm of the TGF-b signaling system. Activated TGF-b receptors also stimulate a variety of non-Smad2/3-based messages that constitute the noncanonical arm of the TGF-b signaling system. Also shown are the influences and outcomes of TGF-b signaling within the primary tumor versus those at metastatic sites

of TGF-b to epithelial-mesenchymal transition (EMT; see below) is critically reliant on NF-kB activity [38]. We extended these findings to show that mammary tumorigenesis and EMT were both sufficient induce the formation of TbRI:xIAP:TAB1:TAK1:IKKb complexes operant in activating NF-kB [40–42], as well as its initiation of an autocrine Cox-2:PGE2:EP2 signaling axis that drives breast cancer metastasis [43]. Interestingly, IKKa also impacts canonical TGF-b signaling by interacting physically with Smad3 to enhance its binding at EMT-regulated

120

M.K. Wendt and W.P. Schiemann

promoters [44]. Finally, the ability of TGF-b to regulate cellular behaviors also transpires through its stimulation of a number of nonreceptor protein tyrosine kinases (PTKs), including focal adhesion kinase (FAK; [45–49]), c-Src [30–32, 50], and c-Abl [51–54]. Collectively, the preferential coupling of TGF-b to its noncanonical effectors appears inappropriately amplified in metastatic cancer cells, thereby generating a signaling imbalance that overrides and/or dampens the tumor suppressing messages transduced by Smad2/3 in human tumors ([46]; Fig. 5.1). Unfortunately, precisely how noncanonical effectors are activated by TGF-b in normal and malignant cells remains to be determined definitively, as does the manner in which these pathways become dysregulated and magnified during the acquisition of metastatic phenotypes by cancer cells. These relationships and their regulation by tumor microenvironments during metastatic progression driven by TGF-b are discussed below.

5.3

EMT

The impact of TGF-b on metastatic progression is strongly linked to EMT, which reflects the ability of immotile, polarized epithelial cells to acquire highly motile, apolar fibroblastoid-like phenotypes [8, 9, 55–57]. More specifically, epithelial cells undergoing EMT exhibit several distinct features, including (i) the loss of cell polarity due to downregulated expression of epithelial cell markers (e.g., E-cadherin, zona occluden-1, and b4 integrin); (ii) cytoskeletal architecture reorganization and intracellular organelle redistribution; (iii) upregulated expression of fibroblast markers (e.g., vimentin, N-cadherin, a-smooth muscle actin); and (iv) elevated invasion factors (e.g., MMP-9, fibronectin) [8, 9, 55–57]. In general, these steps underlie the pathophysiological reactions associated with EMT, which were recently categorized into three distinct subtypes: (i) Type 1 EMT, which reflects the epithelial plasticity associated with embryogenesis and tissue morphogenesis; (ii) Type 2 EMT, which reflects the epithelial plasticity of tissue regeneration and fibrotic reactions; and (iii) Type 3 EMT, which reflects the plasticity of carcinoma cells during metastatic progression [58]. Indeed, the initiation of Type 3 EMT confers carcinoma cells a selective invasive advantage to exit both the primary tumor [45] and the circulation at sites of dissemination [47, 59]. At first blush, this classification scheme acknowledges that the extent to which an EMT reaction transpires and/or resolves likely reflects the overall health and well-being of the epithelium and its immediate microenvironment. Unfortunately, the vast majority of EMT studies employ “cellcentric” approaches to assess the functional consequences of EMT in normal and malignant cells, and as such, the contributions of the microenvironment to regulating epithelial plasticity remains a critical and underexplored question. Despite these limitations and knowledge gaps, recent inroads into how microenvironments influence EMT have been gleaned through the use of novel 3D-organotypic culture systems capable of modeling distinct steps of the metastatic cascade. The impact of these cultures systems on the morphologies and phenotypes associated with Type 3

5

Regulation of TGF-β Signaling and Metastatic Progression…

121

EMT are discussed below. Readers desiring in-depth summaries pertaining to the molecular mechanisms whereby TGF-b promotes EMT are directed to several recent reviews [8, 9].

5.4

The “TGF-b Paradox” in Animal Models

TGF-b was originally defined and purified 30 years ago by its ability to promote the morphological transformation and anchorage-independent growth of normal rat kidney fibroblasts (NRK-49 cells; [60, 61]). However, ensuing examinations of TGF-b action in epithelial cells quickly established this cytokine as a potent inhibitor of cell cycle progression [62], thus providing the initial framework for the “TGF-b Paradox” in traditional culture systems. Extending these analyses to animal studies quickly confirmed the dichotomous nature of TGF-b in regulating tumorigenesis. Indeed, implantation of slow-release TGF-b pellets into the mammary gland significantly inhibited its growth and morphogenesis in a reversible manner [63]. In contrast, administering neutralizing TGF-b antibodies suppressed breast cancer development in mice in part by enhancing splenic natural killer (NK) cell activity [64]. Likewise, transgenic expression of dominant-negative TbR-II enhanced mammary tumorigenesis in carcinogen-treated mice [65], while transgenic expression [66] or systemic administration [67] of a soluble Fc:TbR-II fusion protein both inhibited metastatic progression in mice. In addition, canonical TGF-b signaling was shown to inhibit the tumorigenicity of normal, premalignant, and malignant breast cancer cells, while stimulating that of highly invasive and metastatic breast cancer cells [68–70]. These findings coalesce to support a model wherein TGF-b drives the metastatic progression of late-stage carcinomas, which can upregulate their secretion of TGF-b by as much as 40-fold as compared to their normal counterparts [71]. Accordingly, conditional expression of active TGF-b1 in established mammary tumors significantly enhanced their metastatic progression without affecting the proliferation or growth of the corresponding primary tumors [72]. Similarly, targeted expression of active TGF-b1 in mouse keratinocytes promoted their proliferation in response to carcinogens, as well as induced their progression to aggressive spindle cell carcinomas [73, 74]. Additional insights into the importance of TGF-b signaling during normal development were gleaned from genetically engineering mice to lack expression of individual TGF-b isoforms and their receptors. For example, homozygous deletion of TGF-b1 results in ~50% of the developing pups to die in utero at embryonic day E10.5 [75, 76], while those that survive to term typically succumb to massive inflammatory reactions that result in multifocal infiltration of lymphocytes and macrophages into the heart, lungs, and salivary glands [77]. Homozygous deletion of TGF-b2 elicited perinatal lethality due to multiple developmental defects indicative of aberrant Type I EMT reactions during organogenesis and tissue morphogenesis [78], while disruption of the TGF-b3 locus also produces perinatal lethality resulting from defective EMT that occurred during lung and palate development [79].

122

M.K. Wendt and W.P. Schiemann

Mice lacking TbR-II exhibit yolk sac hematopoiesis and vasculogenesis defects that phenocopy those observed in TGF-b1-deficient mice [80], while mice deficient in TbR-I expression die during midgestation due to placental and yolk sac vascular defects [81]. Collectively, these and numerous additional studies have helped to define the essential role of TGF-b signaling regulating organ development and immune surveillance and privilege [82–85]. Recent technological advancements have enabled the generation of mice that harbor alleles of the TGF-b signaling system that are flanked by LoxP sites (Floxed) as a means to circumvent the embryonic lethality associated with traditional gene targeting approaches. Indeed, conditional deletion of TbR-II in has provided new understanding of its role in regulating cardiac, skull, and palate development [86–88], and in mediating mammary gland involution and wound healing [89, 90]. It is important to note that the biology of TGF-b can largely be divided into two broad categories: regulation of cell cycling versus regulation of cell microenvironments (i.e., extracellular matrix (ECM) remodeling, angiogenesis, fibroblast activation, and immune cell infiltration). Thus, besides its ability to induce cytostasis in epithelial cells, TGF-b also governs the behaviors of adjacent fibroblasts and their synthesis and secretion of paracrine factors and ECM molecules that collectively suppress carcinoma development. Indeed, rendering fibroblasts deficient in TbR-II expression led to the formation of prostate intraepithelial neoplasia and invasive carcinoma of the forestomach [91]. Likewise, conditional deletion of TbR-II in mammary gland fibroblasts expanded their abundance and proliferative potential in a manner correlating with abnormal ductal development [92]. Moreover, these same TbR-II-deficient mammary fibroblasts greatly exaggerated the growth and invasion of breast cells simultaneously engrafted under the renal capsule in part through their upregulated expression of TGF-a, MSP (macrophage-stimulating protein), and HGF (hepatocyte growth factor) [91–94]. Even more remarkably, similar deletion of TbR-II in mammary carcinoma cells promoted the inappropriate activation of two distinct paracrine signaling axes – i.e., SDF-1:CXCR4 and CXCL5:CXCR2 – whose activation led to the recruitment of immature GR1+CD11b+ myeloid cells that drive breast cancer metastasis by inhibiting host tumor immunosurveillance, and by inducing MMP expression [95]. Interestingly, targeted deletion of Smad4 in T cells also elicited carcinoma formation within the gastrointestinal track (e.g., colon, rectum, intestine, and stomach) due to aberrant stromal expansion and signaling [96]. Finally, similar conditional deletion of TbR-I was targeted to the oral cavity of mice, which facilitated carcinogen-induced tumor formation in a manner correlated with constitutive PI3K/AKT activation [97]. Moreover, mice rendered heterozygous null for TbR-I displayed twice as many intestinal tumors as compared to their wild-type counterparts when crossed onto an Apc(Min/+) background [98]. Collectively, these findings highlight the importance of tissue homeostasis and cell microenvironments to facilitate the tumor suppressing activities of TGF-b, as well as establish the pathophysiological basis of the “TGF-b Paradox” in preclinical models of TGF-b-responsive carcinomas. In the succeeding sections, we present recent findings that frame our understanding of the “TGF-b Paradox” and its regulation by the diversity of microenvironments encountered by carcinoma cells during their acquisition of metastatic phenotypes.

5

Regulation of TGF-β Signaling and Metastatic Progression…

5.5

5.5.1

123

TGF-b Signaling Within Primary Tumor Microenvironments Mechanotransduction

The use of traditional 2-dimensional (2D) tissue culture systems to compare the behaviors and activities of normal versus malignant cells have proven to be wholly insufficient in recapitulating the “TGF-b Paradox.” To circumvent this experimental deficiency, researchers have begun employing novel 3D-organotypic cultures to more accurately model and assess how changes within tumor microenvironments impact the functions of TGF-b. For example, culturing normal and malignant mammary epithelial cells (MECs) on top of or embedded into cushions of reconstituted basement membranes (RBMs) has provided numerous insights into how malignant transformation alters acinar development [99]. Although the precise conditions employed in these analyses can impart dramatic alterations in organoid development and behavior [100], it is nonetheless widely accepted that these 3D-organotypic systems more accurately recapitulate the signaling dynamics experienced by MECs and their microenvironments during mammary gland development [101]. It is important to note that tumor development is typically accompanied by intense desmoplastic and fibrotic reactions that enable primary tumors to be palpable in the context of surrounding normal tissue architectures. Moreover, tumor fibrosis results in the formation of mechanically rigid tumor microenvironments that (i) enhance the selection and expansion of developing neoplasms, particularly that of late-stage metastatic tumors, and (ii) predict for poor clinical outcomes in patients with cancers of the colon, ovary, and breast [102–105]. Interestingly, these aberrant cellular activities are highly reminiscent of those attributed to TGF-b [1, 2, 6, 106], whose upregulated expression dictates the composition tumor reactive stroma. In a reciprocal manner, tumor reactive stroma plays an important role during cancer initiation and progression by determining whether TGF-b suppresses or promotes tumor formation [5, 91–94]. In fact, it has been argued that “phenotypes dominate genotypes,” a statement referring to the ability of the ECM and cell microenvironments to either suppress or promote tumorigenesis in a manner independent of genotypic alternations in MECs [107]. Dramatic evidence supporting this idea was provided by the findings that normal mice readily developed from (i) blastocysts injected with stable teratoma cells [108], and (ii) enucleated oocytes injected with nuclei isolated from melanoma cells [109]. Unfortunately, a role for TGF-b to impact these events remains surprisingly unexplored. However, infecting chicken embryos with RSV in ovo was shown to elicit cellular transformation and tumor formation only in response to tissue wounding, a reaction mediated by TGF-b [110, 111]. Collectively, these findings provide compelling evidence to support the notion that “phenotypes dominate genotypes,” as well as implicate TGF-b as a principal player operant in overseeing this phenomenon. As such, inclusion of collagen and other matrices to RBM preparations represented a significant scientific achievement to initiate mechanotransduction in

124

M.K. Wendt and W.P. Schiemann

modeled tumor microenvironments [112, 113]. For instance, elevated collagen concentrations results in integrin clustering and increased activation of focal adhesion complexes, growth factor receptors, and MAP kinase pathways [102–105, 113]. Along these lines, aberrant lysyl oxidase (LOX) activity also associates with cancer progression and the development of desmoplasia [104]. Indeed, elevated expression of LOX family members, particularly LOX, LOXL (LOX-like), and LOXL2, correlates with increased malignancy and the acquisition of invasive/metastatic phenotypes, and with the induction of EMT and the formation of the premetastatic niche [103, 113–118]. In particular, LOX expression is essential for hypoxia-induced metastasis of human MDA-MB-231 breast cancer cells in mice [115, 116]. Moreover, elevated LOX expression in breast cancer is observed most frequently in poorly differentiated, high grade tumors and, consequently, predicts for increased disease recurrence and decreased patient survival [115, 116]. Likewise, LOX expression has recently been validated as a prognostic marker for metastasis development in patients with head and neck cancers [119]. Interestingly, we observed TGF-b to be a potent inducer of LOX expression and activity, an event coupled to the acquisition of oncogenic TGF-b signaling in cancers of the breast [120]. Finally, two-photon intravital imaging analyses demonstrated that cancer cells attempting to exit the primary tumor utilize a paracrine signaling axis comprised of carcinoma-derived colony stimulating factor-1 (CSF-1) and macrophage-derived epidermal growth factor (EGF) that facilitates tumor cell migration and invasion [121–124]. Importantly, TGF-b plays a critical role in establishing this paracrine signaling network through its ability to (i) upregulate CSF-1 receptor expression [121]; (ii) stabilize EGF receptor (EGFR) expression at the cell surface [47]; and (iii) facilitate tumor-infiltration of macrophages via a FAK-dependent pathway [45]. The ability of mechanotransduction to alter function and composition of TGF-b signaling modules in normal and malignant cells has recently been linked to signals originating from integrins. For example, we observed TbR-II to interact physically with b3 integrin following its activation by vitronectin or in response to EMT induced by TGF-b [30–32]. Furthermore, the formation of b3 integrin:TbR-II complexes is critically dependent upon the expression and activity of FAK, such that depleting cells of FAK expression not only disrupts this interaction, but also markedly reduces the ability of TGF-b to stimulate breast cancer invasion and metastasis [45]. Along these lines, b1 integrin also interacts physically with TbR-II [30] and is required for its stimulation of EMT and p38 MAPK [125]. Additionally, the FAK effector, p130Cas, can bind and sequester Smad3 to prevent its phosphorylation by TbR-I and subsequent transit to the nucleus [126]. In extending these findings, we found elevated p130Cas expression to mark metastatic progression in breast cancers, as well as to distort the balance of TGF-b signaling from canonical to noncanonical effectors in metastatic MECs [46]. Indeed, genetic depletion of p130Cas was sufficient in drastically reducing the ability of TGF-b to promote breast cancer development and progression [46], presumably by enhancing their sensitivity to apoptotic stimuli [127]. Taken together, these findings highlight the direct influence integrins and focal adhesion complexes possess in promoting tumor progression,

5

Regulation of TGF-β Signaling and Metastatic Progression…

125

particularly that driven by TGF-b. These studies also emphasize the power of 3D-organotypic models to faithfully recapitulate these events in sensitive and rapid cell-based assays capable of interrogating the “TGF-b Paradox.” The use of collagen to increase matrix rigidity strongly suggests that integrins, particularly those that ligate collagen, have the ability to influence TGF-b signaling. This notion is supported by our recent findings demonstrating that propagating malignant MECs in compliant 3D-organotypic cultures restored the cytostatic activities of TGF-b in highly metastatic 4T1 cells, which normally fail to undergo growth arrest in response to TGF-b in traditional 2D-cultures [32, 41, 51, 128]. Importantly, inclusion of type I collagen in these 3D-organotypic cultures to increase their rigidity/tension and to initiate mechanotransduction dose-dependently uncoupled TGF-b from the regulation of cell cycle progression in 4T1 cells [51]. Conversely, treating 4T1 organoids with the small molecule TbR-I inhibitor, TbR-I Inhibitor II stimulated the growth of 4T1 cells in compliant 3D-organotypic cultures wherein TGF-b functions as a tumor suppressor. However, administering this same pharmacological treatment regimen to tense 3D-organotypic cultures wherein TGF-b functions as a tumor promoter greatly inhibited the growth of the resultant 4T1 organoids [120], thereby validating the first in vitro system that recapitulates the “TGF-b Paradox” solely by modulating the tension sensed by malignant MECs. Interestingly, a recent study suggests that dramatic decreases in matrix compliance (i.e., elastic modulus) can serve as the driving force to alter cell signaling. Indeed, increasing matrix rigidity not only induced dramatic differences in cell proliferation, but also suppressed E-cadherin expression in a manner associated with the exaggerated development of EMT phenotypes [129]. Analogous alterations in tumor microenvironments also enabled E-cadherin-negative breast cancer cells to reinitiate expression of this junctional protein during metastatic outgrowth in the lungs of mice [130]. Although a direct participatory role for TGF-b during these events was not addressed, these analyses have nonetheless linked mechanotransduction to the activation of known TGF-b effectors. Moreover, increased matrix rigidity elicits increased TGF-b production [120], presumably via augmented integrin-mediated activation of latent TGF-b complexes [131]. Taken together, these studies stress the importance of integrins and matrix rigidity in regulating TGF-b signaling modules, and in enhancing the release of TGF-b from inactive matrix depots.

5.5.2

Hypoxia

It has long been appreciated that tumor development and progression significantly deprives these neoplasms access to sufficient nutrients and oxygen that facilitates their unabated growth, leading to the hypothesis that chemotherapeutic targeting of tumor angiogenesis would prove effective in starving tumors to death [132, 133]. Unfortunately, administering angiostatic agents to cancer patients has offered little improvement in overall patient survival, presumably reflecting the fact that hypoxic tumors house significantly elevated quantities of oncogenic factors that collectively

126

M.K. Wendt and W.P. Schiemann

drive carcinoma cell dissemination to more hospitable environments [134]. Indeed, hypoxic culture conditions are sufficient to produce hyperinvasive and EMT phenotypes in carcinoma cells [135]. Moreover, hypoxia and TGF-b act in concert to induce the expression of vascular endothelial growth factor (VEGF) and the chemokine receptor CXCR4, two molecules essential to the metastatic process [136]. As expected, these oncogenic TGF-b activities are mediated by p38 MAPK [136], and by Smad7, which somewhat counterintuitively actually promoted carcinoma invasion under hypoxic conditions [137]. Collectively, these findings support a paradigm in which the activities of canonical and noncanonical TGF-b signaling systems are balanced under normal homeostatic conditions [138], but suddenly become unbalanced and slanted towards noncanonical TGF-b signaling under hypoxic tumor conditions. Indeed, these findings highlight the intimate relationship between hypoxia and oncogenic TGF-b signaling and their ability to be interrogated mechanistically via employment of 3D-organotypic culture systems.

5.5.3

Reactive Tumor Stroma

The cellular and structural composition of tumor microenvironments has a profound impact on tumor progression and TGF-b signaling. As mentioned above, targeted deletion of TbR-II in mammary carcinoma cells facilitated exaggerated infiltration of immature myeloid cells that greatly enhanced tumor invasion and metastasis in part via upregulated production of matrix metalloproteinases (MMPs) and TGF-b1 [95]. Interestingly, targeted inactivation of the TGF-b signaling system in fibroblasts [91] or the Smad4 pathway in T cells [96] both elicited carcinoma development due to disruptions of tumor suppressing paracrine signaling networks. Attempts to model these complex microenvironmental events in vitro have recently been initiated. For instance, the influence of CD4+ T cells and their production of cytokines, particularly TGF-b, to drive breast cancer metastasis was determined to be dependent upon the development of M2-type macrophages. Indeed, co-culturing M2 macrophages with mammary carcinoma cells in 3D-organotypic cultures led to the production of invasive structures from resulting organoids in a manner indicative of increased carcinoma malignancy [139]. Although the synthesis of protruding branched structures is believed to reflect the extent of metastatic progression, it should be noted that the validity of this assumption remains to be rigorously examined. Thus, while certain metastatic cells do in fact grow as branched structures [140, 141], others readily assume a dense spherical morphology when propagated in 3D-organotypic systems. Indeed, we recently observed nonmetastatic cells to display dysmorphic branching structures that readily disappear during metastatic progression and in response to EMT induced by TGF-b (Fig. 5.2), which also promoted luminal filling of established acinar structures [47, 140]. Collectively, these findings highlight the potential of 3D-organotypic cultures to model distinct cell morphologies and behaviors in a manner that recapitulates defined aspects of tumor:stromal interactions. Extending these systems to include additional stromal

5

Regulation of TGF-β Signaling and Metastatic Progression…

127

Fig. 5.2 The influence of TGF-b signaling on normal and malignant MEC acinar morphology. Photomicrographs of MECs grown in 3D-organotypic cultures (left to right): (i) normal MECs grown in the absence (a, 400×) or presence (b, 100×) of EGF; (ii) nonmetastatic breast cancer cells (c, 100×); and (iii) metastatic breast cancer cells (d, 100×). Accompanying schematic depicts these morphologies and how TGF-b signaling drives the acquisition of metastatic phenotypes

mediators (e.g., fibroblasts, endothelial, and immune cells) propagated under varying oxygen tensions and elastic moduli will greatly enhance the ability of science and medicine to decipher and manipulate the molecular underpinnings of the “TGF-b Paradox.”

5.5.4

Visualizing Metastatic Progression Stimulated by TGF-b

Immunohistochemistry (IHC) is one of the most powerful tools available to pathologists attempting the grade and subtype tumor biopsies. Moreover, as the molecular basis of tumorigenesis continues to unfold, these IHC analyses will become more sophisticated in their ability to accurately diagnose and prognose patient outcomes, as well as to monitor the effectiveness of specific therapeutic regimens. However, employing IHC to gauge the extent of TGF-b function in human tissues is fraught with caveats and misinterpretations because this technique only captures a small snap-shot of the dynamics built into the TGF-b signaling system. Powerful developments in intravital imaging have greatly accelerated our understanding of (i) “how,” “when,” and “where” cancer cells systemically disseminate, and (ii) the role of TGF-b in mediating discrete steps of the metastatic cascade [45–47]. By employing dual bioluminescent imaging techniques, Korpal et al. [142] demonstrated the essential importance of canonical TGF-b

128

M.K. Wendt and W.P. Schiemann

Fig. 5.3 Smad2/3 Signaling is decreased dramatically during the outgrowth of pulmonary metastases. (a) Metastatic 4T1 breast cancer cells stably expressing CMV-renilla luciferase in conjunction with either SBE (4T1-SBE)- or p3TP (4T1-p3TP)-driven firefly luciferase were propagated (±) TGF-b in 2D- or 3D-cultures. Data are the fold-induction of luciferase activity stimulated by TGF-b. (b) 4T1-SBE cells were engrafted onto the mammary fat pad of Balb/C mice. Primary tumor growth and metastasis was assessed 4 weeks later using dual bioluminescent imaging to visualize these events (CMV-renilla), as well as the extent of TGF-b-specific Smad2/3 signaling (SBE-firefly). Data are the mean luciferase ratios of SBE-firefly:CMV-renilla (±SD; n = 4)

signaling to promote the establishment of osteolytic bone lesions by metastatic breast cells. Importantly, bisphosphonate administration was shown to be more effective in suppressing osteolytic bone lesions and canonical TGF-b signaling early in the metastatic process as opposed to after these lesions were fully established [142]. These findings suggest that canonical TGF-b signaling is differentially regulated during specific stages of the metastatic cascade. Accordingly, transient activation of Smad2/3 by TGF-b was sufficient in converting the migration of breast cancer cells from cohesive to single cell programs

5

Regulation of TGF-β Signaling and Metastatic Progression…

129

[143, 144]. Remarkably, the ability of breast cancer metastases to resume proliferative programs within the pulmonary microenvironment required these cells to first inactivate Smad2/3 signaling [143, 144], and as such, it is tempting to speculate that altered elastic moduli govern the coupling of TGF-b to Smad2/3. In support of this supposition, Fig. 5.3 shows that TGF-b becomes selectively uncoupled from the activation of Smad2/3 in (i) compliant 3D-organotypic cultures relative to rigid tissue culture plastic (Fig. 5.3a), and (ii) pulmonary metastases relative to their site of origin (Fig. 5.3b). Collectively, these intriguing findings demonstrate the plasticity present in the TGF-b signaling system as carcinoma cells undergo EMT and metastatic outgrowth, presumably reflecting a shift from canonical (i.e., Smad2/3-based) to noncanonical (i.e., non-Smad2/3-based) signaling that originates from altered mechanotransduction within the tumor microenvironment [9, 46, 47]. Clearly, extending these novel intravital imaging techniques holds great promise to delineate and define the mechanics of the TGF-b signaling during distinct stages of metastatic progression.

5.6

TGF-b Signaling Within the Circulation

After exiting the confines of the primary tumor, disseminated cancer cells are confronted with the challenges of traversing and surviving the nonadherent microenvironments provided by the lymphatic and circulatory systems. In fact, the lack of ECM support, coupled with shear forces encountered in circulation, make this process of the metastatic cascade one of the harshest and deadliest faced by metastatic cells. This notion is bolstered by “experimental metastasis” studies which demonstrated that 95% of inoculated cancer cells die prior to entering a metastatic niche [145, 146]. At present, a direct role for TGF-b in regulating the behaviors of circulating tumor cells remains largely unexplored. However, TGF-b does promote the generation of “mammospheres,” which are nonadherent structures enriched for stem-like progenitor cells [147]. These data suggest that TGF-b and its induction of EMT functions to promote the selection and expansion of cancer initiating cells that are ideally suited for exiting the primary tumor and surviving nonadherent conditions. Anoikis is a specialized form of apoptosis exhibited by adherent cells when they are deprived of their normal cell: ECM interactions [148]. Interestingly, anoikis also plays a critical role during acinar development by promoting “nonadherent” luminal cells to undergo apoptosis during organoid hollowing [149]. As mentioned previously, EMT induced by TGF-b prevents acinar hollowing [47, 140], suggesting that TGF-b confers resistance to anoikis in post-EMT cell populations. Along these lines, myofibroblasts can be protected from anoikis following TGF-b-mediated activation of FAK and AKT [48]. Although additional studies are clearly warranted to solidify the relationship between TGF-b and anoikis, these findings do coalesce to support a model wherein circulating tumor cells activated by TGF-b are afforded a selective survival advantage during their rapid and violent transit to distant metastatic sites.

130

5.7 5.7.1

M.K. Wendt and W.P. Schiemann

TGF-b Signaling Within Metastatic Sites Pulmonary Microenvironment

As mentioned previously, FAK mediates the interaction between integrins and TGF-b receptors [9, 46], as well as promotes the metastatic outgrowth of breast cancer cells in pulmonary tissues [150]. Interestingly, recent evidence demonstrates that the success or failure of breast cancer cells grow in pulmonary microenvironments can be modeled and recapitulated by the sparse plating of cells in 3D-organotypic cultures [150–152]. Moreover, culturing normal MECs in these 3D-organotypic cultures is sufficient to induce acinar formation and milk production [153], strongly suggesting that these culture systems represent suitable surrogates to model the differentiation and development of normal mammary glands [99], as well as metastatic dormancy in isolated low-grade carcinoma cells [151]. The dichotomous nature of these two examples (i.e., acinar development versus metastatic dormancy) raises two important questions for future analyses. First, what are the underlying mechanisms that enable a single microenvironment to elicit such varying responses in MECs? And second, what role does TGF-b play in directing these disparate functions of MECs? Although both questions remain unanswered, a recent study described a metastatic gene signature associated with the ability of breast cancers to metastasize to the lungs [154]. Central players to this response were IDs (Inhibitor of Differentiation) 1 and 3, which promoted breast cancer cells to extravasate and colonize the lung [154]. Along these lines, we found TGF-b to strongly suppress the expression of IDs 1 and 3 in normal MECs, but to dramatically induce their expression in late-stage breast cancer cells [106]. Thus, future studies need to gauge the extent to which dysregulated coupling of TGF-b to ID1/3 expression manifests the “TGF-b Paradox,” as well as dictates the response of normal and malignant MECs to microenvironmental signals. TGF-b stimulation of angiopoietin-like 4 expression enabled breast cancer cells to specifically extravasate and colonize the lungs, but not the bone [59], suggesting that TGF-b signaling is critical in determining metastatic tropism. Likewise, activation of Smad2/3 by TGF-b was shown to be critical in promoting the invasive exodus of breast cancer cells out of the primary tumor; however, reinitiation of proliferation programs necessary for metastatic colonization of the lungs required the activity of Smad2/3 to be silenced (Fig. 5.3; [143, 144]). Unfortunately, these analyses only employed a “Smad2/3-centric” approach, and in doing so, failed to address the relative contribution of noncanonical TGF-b signaling inputs to the latter steps of metastasis. Given the established and essential function of noncanonical TGF-b signaling to EMT and metastasis [8, 9], we propose a provocative explanation that the inherent plasticity of the TGF-b signaling system enables metastatic cells to actively induce or repress specific branches of this unique signaling system metastatic progression via a microenvironment-dependent manner. Future studies need to address this hypothesis using an intricate combination of 3D-organotypic cultures and in vivo metastasis models in mice.

5

Regulation of TGF-β Signaling and Metastatic Progression…

5.7.2

131

Skeletal Microenvironment

Despite the fact that skeletal metastases pose a significant health concern for many cancers, the development of appropriate 3D-organotypic cultures to model the interactions of carcinoma cells with a bone microenvironment have yet to be fully exploited. However, a recent study did employ bone hydroxyapatite (HA) to mineralize polymeric scaffolds which supported the growth of breast cancer cells [155]. Importantly, mineralized microenvironments rendered breast cancer cells more competent to synthesize pro-osteoclastic interleukin-8, as well as more sensitive to bisphosphonate treatment [155]. Because TGF-b induces interleukin-8 expression [156], and bisphosphonates inhibit TGF-b signaling [142], it is tempting to speculate that mineralized microenvironments support oncogenic TGF-b. Accordingly, inhibiting TGF-b signaling via administration of the TbR-I inhibitor, SD-208 [157] or a soluble TbR-II molecule [158] both effectively reduced bone metastases. Moreover, Smad4-deficiency or expression of a dominant-negative TbR-II mutant both abrogated breast cancer metastasis to bone in part via diminished the expression of PTHrP, IL-11, and CTGF [10, 159, 160]. Collectively, these findings highlight the importance of TGF-b signaling in mediating bone metastases, and more importantly, the role of bone microenvironments to directly impact the pathophysiology of TGF-b. Indeed, the continued refinement of mineralized microenvironmental cultures clearly holds tremendous promise to elucidate the molecular mechanisms whereby skeletal-derived signals manifest the “TGF-b Paradox.”

5.8

Summary

Attempts to solve the “TGF-b Paradox” have yet to be actualized since the discovery of this phenomenon nearly 30 years ago. Indeed, the inability of science and medicine to solve the “TGF-b Paradox” reflects their collective failure to develop a sensitive and rapid cell-based assay that fully recapitulates this phenomenon in vitro. Here we discussed recent advances in our understanding of the “TGF-b Paradox” and its role in promoting metastatic progression, and in doing so, highlighted the importance of novel 3D-organotypic culture systems to deconstruct the metastatic cascade to better investigate the molecular connections between tumor microenvironments and the TGF-b signaling system. Collectively, these findings coalesce to support the model in Fig. 5.4 that depicts the potential role of ECM and microenvironmental rigidity to alter the response of carcinoma cells to TGF-b. Indeed, we propose that early-stage carcinomas evolve in compliant microenvironments that favor canonical Smad2/3 signaling stimulated by TGF-b. The continued growth of the developing neoplasm enhances ECM rigidity by upregulating TGF-b production, which further enhances the demsoplastic process and its inappropriate formation of receptor complexes comprised of integrins, growth factor receptors, and TGF-b receptors. Once formed, these hyperactivated signaling modules greatly

Fig. 5.4 Schematic depicting the role of microenvironmental and ECM rigidity in mediating oncogenic TGF-b signaling. Tumor-initiated carcinomas evolve in compliant microenvironments that favor canonical signaling by TGF-b. Neoplastic growth enhances microenvironmental rigidity, leading to the inappropriate formation of integrin:TbR-II and TGF-b receptor:EGFR receptor complexes and the amplification of noncanonical signaling TGF-b. These unbecoming events culminate in maximal noncanonical TGF-b signaling, leading to the acquisition of EMT, stem-like, and metastatic phenotypes. Deposition of metastatic cells at compliant micrometastatic niches partially restores the cytostatic function of TGF-b, resulting in tumor dormancy. Over time, this microenvironment cycle is repeated, leading to disease recurrence and poor clinical outcomes in patients harboring metastatic disease

132 M.K. Wendt and W.P. Schiemann

5

Regulation of TGF-β Signaling and Metastatic Progression…

133

amplify the activation of noncanonical effectors by TGF-b, which culminate in the acquisition of EMT, stem-like, and metastatic phenotypes. Arrival of disseminated carcinoma cells to metastatic niches once again places these cells in compliant microenvironments, which partially reinstates the cytostatic activities of TGF-b and initiates micrometastatic dormancy. Over time, this vicious microenvironment cycle is repeated, leading to disease recurrence and poor clinical outcomes in carcinoma patients harboring metastatic disease. The basic tenets of this model clearly are supported by the findings presented herein, and as such, this model should serve as a launching point for future studies aimed at identifying the individual effectors operant in regulating ECM tension and TGF-b function in distinct carcinoma subtypes. The use of advanced 3D-organotypic cultures, together with powerful intravital imaging techniques will undoubtedly delineate the complex interactions whereby tumor microenvironments govern oncogenic TGF-b signaling and its stimulation of metastatic progression. In doing so, we believe that it will one day be possible to manipulate the “TGF-b Paradox” and prevent its pathophysiological manifestations, thereby improving the prognosis and overall survival of patients with metastatic disease. Acknowledgements We thank members of the Schiemann Laboratory for critical comments and reading of the manuscript. W.P.S. was supported by grants from the National Institutes of Health (CA129359), the Komen Foundation (BCTR0706967), the Department of Defense (BC084651); and the Case Comprehensive Cancer Center and the University Hospitals Seidman Cancer Center. M.K.W. was supported by the American Cancer Society (PF-09-120-01-CS).

References 1. Blobe GC, Schiemann WP, Lodish HF (2000) Role of TGF-b in human disease. N Engl J Med 342:1350–1358 2. Galliher AJ, Neil JR, Schiemann WP (2006) Role of TGF-b in cancer progression. Future Oncol 2:743–763 3. Massague J, Gomis RR (2006) The logic of TGF-b signaling. FEBS Lett 580:2811–2820 4. Siegel PM, Massague J (2003) Cytostatic and apoptotic actions of TGF-b in homeostasis and cancer. Nat Rev Cancer 3:807–821 5. Tian M, Schiemann WP (2009) The TGF-b paradox in human cancer: an update. Future Oncol 5:259–271 6. Schiemann WP (2007) Targeted TGF-b chemotherapies: friend or foe in treating human malignancies? Expert Rev Anticancer Ther 7:609–611 7. Rahimi RA, Leof EB (2007) TGF-b signaling: a tale of two responses. J Cell Biochem 102:593–608 8. Taylor MA, Parvani JG, Schiemann WP (2010) The pathophysiology of epithelialmesenchymal transition induced by TGF-b in normal and malignant mammary epithelial cells. J Mammary Gland Biol Neoplasia 15:169–190 9. Wendt MK, Allington TM, Schiemann WP (2009) Mechanisms of the epithelial-mesenchymal transition by TGF-b. Future Oncol 5:1145–1168 10. Kang Y, Siegel PM, Shu W, Drobnjak M, Kakonen SM, Cordon-Cardo C, Guise TA, Massague J (2003) A multigenic program mediating breast cancer metastasis to bone. Cancer Cell 3:537–549

134

M.K. Wendt and W.P. Schiemann

11. Minn AJ, Kang Y, Serganova I, Gupta GP, Giri DD, Doubrovin M, Ponomarev V, Gerald WL, Blasberg R, Massague J (2005) Distinct organ-specific metastatic potential of individual breast cancer cells and primary tumors. J Clin Invest 115:44–55 12. Zavadil J, Bitzer M, Liang D, Yang YC, Massimi A, Kneitz S, Piek E, Bottinger EP (2001) Genetic programs of epithelial cell plasticity directed by TGF-b. Proc Natl Acad Sci U S A 98:6686–6691 13. Massague J (1998) TGF-b signal transduction. Annu Rev Biochem 67:753–791 14. Chang H, Brown CW, Matzuk MM (2002) Genetic analysis of the mammalian TGF-b superfamily. Endocr Rev 23:787–823 15. Gatza CE, Oh SY, Blobe GC (2010) Roles for the type III TGF-b receptor in human cancer. Cell Signal 22:1163–1174 16. Feng XH, Derynck R (2005) Specificity and versatility in TGF-b signaling through Smads. Annu Rev Cell Dev Biol 21:659–693 17. Moustakas A, Heldin CH (2005) Non-Smad TGF-b signals. J Cell Sci 118:3573–3584 18. Shi Y, Massague J (2003) Mechanisms of TGF-b signaling from cell membrane to the nucleus. Cell 113:685–700 19. Massague J, Seoane J, Wotton D (2005) Smad transcription factors. Genes Dev 19:2783–2810 20. Tsukazaki T, Chiang TA, Davison AF, Attisano L, Wrana JL (1998) SARA, a FYVE domain protein that recruits Smad2 to the TGF-b receptor. Cell 95:779–791 21. Miura S, Takeshita T, Asao H, Kimura Y, Murata K, Sasaki Y, Hanai JI, Beppu H, Tsukazaki T, Wrana JL, Miyazono K, Sugamura K (2000) Hgs (Hrs), a FYVE domain protein, is involved in Smad signaling through cooperation with SARA. Mol Cell Biol 20:9346–9355 22. Hocevar BA, Smine A, Xu XX, Howe PH (2001) The adaptor molecule Disabled-2 links the TGF-b receptors to the Smad pathway. EMBO J 20:2789–2801 23. Hayashi H, Abdollah S, Qiu Y, Cai J, Xu YY, Grinnell BW, Richardson MA, Topper JN, Gimbrone MAJ, Wrana JL, Falb D (1997) The MAD-related protein Smad7 associates with the TGF-b receptor and functions as an antagonist of TGF-b signaling. Cell 89: 1165–1173 24. Nakao A, Afrakht M, Moren A, Nakayama T, Christian JL, Heuchel R, Itoh S, Kawabata M, Heldin NE, Heldin CH, ten Dijke P (1997) Identification of Smad7, a TGF-b inducible antagonist of TGF-b signalling. Nature 389:631–635 25. Souchelnytskyi S, Nakayama T, Nakao A, Moren A, Heldin CH, Christian JL, ten Dijke P (1998) Physical and functional interaction of murine and xenopus Smad7 with bone morphogenetic protein receptors and TGF-b receptors. J Biol Chem 273:25364–25370 26. Ebisawa T, Fukuchi M, Murakami G, Chiba T, Tanaka K, Imamura T, Miyazono K (2001) Smurf1 interacts with TGF-b type I receptor through Smad7 and induces receptor degradation. J Biol Chem 276:12477–12480 27. Kavsak P, Rasmussen RK, Causing CG, Bonni S, Zhu H, Thomsen GH, Wrana JL (2000) Smad7 binds to Smurf2 to form an E3 ubiquitin ligase that targets the TGF-b receptor for degradation. Mol Cell 6:1365–1375 28. Itoh S, ten Dijke P (2007) Negative regulation of TGF-b receptor/Smad signal transduction. Curr Opin Cell Biol 19:176–184 29. Kang JS, Liu C, Derynck R (2009) New regulatory mechanisms of TGF-b receptor function. Trends Cell Biol 19:385–394 30. Galliher AJ, Schiemann WP (2006) b3 integrin and Src facilitate TGF-b mediated induction of epithelial-mesenchymal transition in mammary epithelial cells. Breast Cancer Res 8:R42 31. Galliher AJ, Schiemann WP (2007) Src phosphorylates Tyr284 in TGF-b type II receptor and regulates TGF-b stimulation of p38 MAPK during breast cancer cell proliferation and invasion. Cancer Res 67:3752–3758 32. Galliher-Beckley AJ, Schiemann WP (2008) Grb2 binding to Tyr284 in TbR-II is essential for mammary tumor growth and metastasis stimulated by TGF-b. Carcinogenesis 29: 244–251 33. Yamashita M, Fatyol K, Jin C, Wang X, Liu Z, Zhang YE (2008) TRAF6 mediates Smadindependent activation of JNK and p38 by TGF-b. Mol Cell 31:918–924

5

Regulation of TGF-β Signaling and Metastatic Progression…

135

34. Sorrentino A, Thakur N, Grimsby S, Marcusson A, von Bulow V, Schuster N, Zhang S, Heldin CH, Landstrom M (2008) The type I TGF-b receptor engages TRAF6 to activate TAK1 in a receptor kinase-independent manner. Nat Cell Biol 10:1199–1207 35. Azuma M, Motegi K, Aota K, Yamashita T, Yoshida H, Sato M (1999) TGF-b1 inhibits NF-kB activity through induction of IkB-a expression in human salivary gland cells: a possible mechanism of growth suppression by TGF-b1. Exp Cell Res 250:213–222 36. Arsura M, Panta GR, Bilyeu JD, Cavin LG, Sovak MA, Oliver AA, Factor V, Heuchel R, Mercurio F, Thorgeirsson SS, Sonenshein GE (2003) Transient activation of NF-kB through a TAK1/IKK kinase pathway by TGF-b1 inhibits AP-1/SMAD signaling and apoptosis: implications in liver tumor formation. Oncogene 22:412–425 37. Park J-I, Lee M-G, Cho K, Park B-J, Chae K-S, Byun D-S, Ryu B-K, Park Y-K, Chi S-G (2003) TGF-b1 activates interleukin-6 expression in prostate cancer cells through the synergistic collaboration of the Smad2, p38-NF-kB, JNK, and Ras signaling pathways. Oncogene 22:4314–4332 38. Huber MA, Azoitei N, Baumann B, Grunert S, Sommer A, Pehamberger H, Kraut N, Beug H, Wirth T (2004) NF-kB is essential for epithelial-mesenchymal transition and metastasis in a model of breast cancer progression. J Clin Invest 114:569–581 39. You HJ, How T, Blobe GC (2009) The type III TGF-b receptor negatively regulates NF-kB signaling through its interaction with b-arrestin2. Carcinogenesis 30:1281–1287 40. Neil JR, Johnson KM, Nemenoff RA, Schiemann WP (2008) Cox-2 inactivates Smad signaling and enhances EMT stimulated by TGF-b through a PGE2-dependent mechanisms. Carcinogenesis 29:2227–2235 41. Neil JR, Schiemann WP (2008) Altered TAB1:IkB kinase interaction promotes TGF-bmediated NF-kB activation during breast cancer progression. Cancer Res 68:1462–1470 42. Neil JR, Tian M, Schiemann WP (2009) X-linked inhibitor of apoptosis protein and its E3 ligase activity promote TGF-b-mediated NF-kB activation during breast cancer progression. J Biol Chem 284:21209–21217 43. Tian M, Schiemann WP (2010) PGE2 receptor EP2 mediates the antagonistic effect of COX-2 on TGF-b signaling during mammary tumorigenesis. FASEB J 24:1105–1116 44. Brandl M, Seidler B, Haller F, Adamski J, Schmid RM, Saur D, Schneider G (2010) IKK(alpha) controls canonical TGF-b-SMAD signaling to regulate genes expressing SNAIL and SLUG during EMT in panc1 cells. J Cell Sci 123:4231–4239 45. Wendt M, Schiemann W (2009) Therapeutic targeting of the focal adhesion complex prevents oncogenic TGF-b signaling and metastasis. Breast Cancer Res 11:R68 46. Wendt MK, Smith JA, Schiemann WP (2009) p130Cas is required for mammary tumor growth and TGF-b-mediated metastasis through regulation of Smad2/3 activity. J Biol Chem 284:34145–34156 47. Wendt MK, Smith JA, Schiemann WP (2010) TGF-b-induced epithelial-mesenchymal transition facilitates epidermal growth factor-dependent breast cancer progression. Oncogene 29: 6485–6498 48. Horowitz JC, Rogers DS, Sharma V, Vittal R, White ES, Cui Z, Thannickal VJ (2007) Combinatorial activation of FAK and AKT by TGF-b1 confers an anoikis-resistant phenotype to myofibroblasts. Cell Signal 19:761–771 49. Thannickal VJ, Lee DY, White ES, Cui Z, Larios JM, Chacon R, Horowitz JC, Day RM, Thomas PE (2003) Myofibroblast differentiation by TGF-b1 is dependent on cell adhesion and integrin signaling via focal adhesion kinase. J Biol Chem 278:12384–12389 50. Park SS, Eom YW, Kim EH, Lee JH, Min DS, Kim S, Kim SJ, Choi KS (2004) Involvement of c-Src kinase in the regulation of TGF-b1-induced apoptosis. Oncogene 23:6272–6281 51. Allington TM, Galliher-Beckley AJ, Schiemann WP (2009) Activated Abl kinase inhibits oncogenic TGF-b signaling and tumorigenesis in mammary tumors. FASEB J 23:4231–4243 52. Allington TM, Schiemann WP (2011) The Cain and Abl of epithelial-mesenchymal transition and TGF-b in mammary epithelial cells. Cells Tissues Organs 193:98–113 53. Wang S, Wilkes MC, Leof EB, Hirschberg R (2005) Imatinib mesylate blocks a non-Smad TGF-b pathway and reduces renal fibrogenesis in vivo. FASEB J 19:1–11

136

M.K. Wendt and W.P. Schiemann

54. Wilkes MC, Leof EB (2006) TGF-b activation of c-Abl is independent of receptor internalization and regulated by phosphatidylinositol 3-kinase and PAK2 in mesenchymal cultures. J Biol Chem 281:27846–27854 55. Moustakas A, Heldin CH (2007) Signaling networks guiding epithelial-mesenchymal transitions during embryogenesis and cancer progression. Cancer Sci 98:1512–1520 56. Thiery JP (2002) Epithelial-mesenchymal transitions in tumour progression. Nat Rev Cancer 2:442–454 57. Zavadil J, Bottinger EP (2005) TGF-b and epithelial-to-mesenchymal transitions. Oncogene 24:5764–5774 58. Kalluri R, Weinberg RA (2009) The basics of epithelial-mesenchymal transition. J Clin Invest 119:1420–1428 59. Padua D, Zhang XH, Wang Q, Nadal C, Gerald WL, Gomis RR, Massague J (2008) TGF-b primes breast tumors for lung metastasis seeding through angiopoietin-like 4. Cell 133:66–77 60. Roberts AB, Anzano MA, Lamb LC, Smith JM, Sporn MB (1981) New class of transforming growth factors potentiated by epidermal growth factor: isolation from non-neoplastic tissues. Proc Natl Acad Sci U S A 78:5339–5343 61. Moses HL, Branum EL, Proper JA, Robinson RA (1981) Transforming growth factor production by chemically transformed cells. Cancer Res 41:2842–2848 62. Carr BI, Hayashi I, Branum EL, Moses HL (1986) Inhibition of DNA synthesis in rat hepatocytes by platelet-derived type b transforming growth factor. Cancer Res 46:2330–2334 63. Silberstein GB, Daniel CW (1987) Reversible inhibition of mammary gland growth by TGF-b. Science 237:291–293 64. Arteaga CL, Hurd SD, Winnier AR, Johnson MD, Fendly BM, Forbes JT (1993) Anti-TGF-b antibodies inhibit breast cancer cell tumorigenicity and increase mouse spleen natural killer cell activity. Implications for a possible role of tumor cell/host TGF-b interactions in human breast cancer progression. J Clin Invest 92:2569–2576 65. Bottinger EP, Jakubczak JL, Haines DC, Bagnall K, Wakefield LM (1997) Transgenic mice overexpressing a dominant-negative mutant type II TGF-b receptor show enhanced tumorigenesis in the mammary gland and lung in response to the carcinogen 7,12-dimethylbenz-[a]anthracene. Cancer Res 57:5564–5570 66. Yang YA, Dukhanina O, Tang B, Mamura M, Letterio JJ, MacGregor J, Patel SC, Khozin S, Liu ZY, Green J, Anver MR, Merlino G, Wakefield LM (2002) Lifetime exposure to a soluble TGF-b antagonist protects mice against metastasis without adverse side effects. J Clin Invest 109:1607–1615 67. Muraoka RS, Dumont N, Ritter CA, Dugger TC, Brantley DM, Chen J, Easterly E, Roebuck LR, Ryan S, Gotwals PJ, Koteliansky V, Arteaga CL (2002) Blockade of TGF-b inhibits mammary tumor cell viability, migration, and metastases. J Clin Invest 109:1551–1559 68. Tian F, Byfield SD, Parks WT, Stuelten CH, Nemani D, Zhang YE, Roberts AB (2004) Smadbinding defective mutant of TGF-b type I receptor enhances tumorigenesis but suppresses metastasis of breast cancer cell lines. Cancer Res 64:4523–4530 69. Tian F, DaCosta BS, Parks WT, Yoo S, Felici A, Tang B, Piek E, Wakefield LM, Roberts AB (2003) Reduction in Smad2/3 signaling enhances tumorigenesis but suppresses metastasis of breast cancer cell lines. Cancer Res 63:8284–8292 70. Tang B, Vu M, Booker T, Santner SJ, Miller FR, Anver MR, Wakefield LM (2003) TGF-b switches from tumor suppressor to prometastatic factor in a model of breast cancer progression. J Clin Invest 112:1116–1124 71. Anzano MA, Roberts AB, De Larco JE, Wakefield LM, Assoian RK, Roche NS, Smith JM, Lazarus JE, Sporn MB (1985) Increased secretion of type b transforming growth factor accompanies viral transformation of cells. Mol Cell Biol 5:242–247 72. Muraoka-Cook RS, Kurokawa H, Koh Y, Forbes JT, Roebuck LR, Barcellos-Hoff MH, Moody SE, Chodosh LA, Arteaga CL (2004) Conditional overexpression of active TGF-b1 in vivo accelerates metastases of transgenic mammary tumors. Cancer Res 64:9002–9011

5

Regulation of TGF-β Signaling and Metastatic Progression…

137

73. Cui W, Fowlis DJ, Bryson S, Duffie E, Ireland H, Balmain A, Akhurst RJ (1996) TGF-b1 inhibits the formation of benign skin tumors, but enhances progression to invasive spindle carcinomas in transgenic mice. Cell 86:531–542 74. Fowlis DJ, Cui W, Johnson SA, Balmain A, Akhurst RJ (1996) Altered epidermal cell growth control in vivo by inducible expression of TGF-b1 in the skin of transgenic mice. Cell Growth Differ 7:679–687 75. Shull MM, Ormsby I, Kier AB, Pawlowski S, Diebold RJ, Yin M, Allen R, Sidman C, Proetzel G, Calvin D et al (1992) Targeted disruption of the mouse TGF-b1 gene results in multifocal inflammatory disease. Nature 359:693–699 76. Kulkarni AB, Huh CG, Becker D, Geiser A, Lyght M, Flanders KC, Roberts AB, Sporn MB, Ward JM, Karlsson S (1993) TGF-b1 null mutation in mice causes excessive inflammatory response and early death. Proc Natl Acad Sci U S A 90:770–774 77. Christ M, McCartney-Francis NL, Kulkarni AB, Ward JM, Mizel DE, Mackall CL, Gress RE, Hines KL, Tian H, Karlsson S et al (1994) Immune dysregulation in TGF-b1-deficient mice. J Immunol 153:1936–1946 78. Sanford LP, Ormsby I, Gittenberger-de Groot AC, Sariola H, Friedman R, Boivin GP, Cardell EL, Doetschman T (1997) TGF-b2 knockout mice have multiple developmental defects that are non-overlapping with other TGF-b knockout phenotypes. Development 124:2659–2670 79. Kaartinen V, Voncken JW, Shuler C, Warburton D, Bu D, Heisterkamp N, Groffen J (1995) Abnormal lung development and cleft palate in mice lacking TGF-b3 indicates defects of epithelial-mesenchymal interaction. Nat Genet 11:415–421 80. Oshima M, Oshima H, Taketo MM (1996) TGF-b receptor type II deficiency results in defects of yolk sac hematopoiesis and vasculogenesis. Dev Biol 179:297–302 81. Larsson J, Goumans MJ, Sjostrand LJ, van Rooijen MA, Ward D, Leveen P, Xu X, ten Dijke P, Mummery CL, Karlsson S (2001) Abnormal angiogenesis but intact hematopoietic potential in TGF-b type I receptor-deficient mice. EMBO J 20:1663–1673 82. Kuruvilla AP, Shah R, Hochwald GM, Liggitt HD, Palladino MA, Thorbecke GJ (1991) Protective effect of TGF-b1 on experimental autoimmune diseases in mice. Proc Natl Acad Sci U S A 88:2918–2921 83. Goey H, Keller JR, Back T, Longo DL, Ruscetti FW, Wiltrout RH (1989) Inhibition of early murine hemopoietic progenitor cell proliferation after in vivo locoregional administration of TGF-b1. J Immunol 143:877–880 84. Bartlett WC, Purchio A, Fell HP, Noelle RJ (1991) Cognate interactions between helper T cells and B cells. VI. TGF-b inhibits B cell activation and antigen-specific, physical interactions between Th and B cells. Lymphokine Cytokine Res 10:177–183 85. Kehrl JH, Wakefield LM, Roberts AB, Jakowlew S, Alvarez-Mon M, Derynck R, Sporn MB, Fauci AS (1986) Production of TGF-b by human T lymphocytes and its potential role in the regulation of T cell growth. J Exp Med 163:1037–1050 86. Langlois D, Hneino M, Bouazza L, Parlakian A, Sasaki T, Bricca G, Li JY (2010) Conditional inactivation of TGF-b type II receptor in smooth muscle cells and epicardium causes lethal aortic and cardiac defects. Transgenic Res 19:1069–1082 87. Seo HS, Serra R (2009) Tgfbr2 is required for development of the skull vault. Dev Biol 334:481–490 88. Ito Y, Yeo JY, Chytil A, Han J, Bringas P Jr, Nakajima A, Shuler CF, Moses HL, Chai Y (2003) Conditional inactivation of Tgfbr2 in cranial neural crest causes cleft palate and calvaria defects. Development 130:5269–5280 89. Bierie B, Gorska AE, Stover DG, Moses HL (2009) TGF-b promotes cell death and suppresses lactation during the second stage of mammary involution. J Cell Physiol 219:57–68 90. Martinez-Ferrer M, Afshar-Sherif AR, Uwamariya C, de Crombrugghe B, Davidson JM, Bhowmick NA (2010) Dermal TGF-b responsiveness mediates wound contraction and epithelial closure. Am J Pathol 176:98–107 91. Bhowmick NA, Neilson EG, Moses HL (2004) Stromal fibroblasts in cancer initiation and progression. Nature 432:332–337

138

M.K. Wendt and W.P. Schiemann

92. Cheng N, Bhowmick NA, Chytil A, Gorksa AE, Brown KA, Muraoka R, Arteaga CL, Neilson EG, Hayward SW, Moses HL (2005) Loss of TGF-b type II receptor in fibroblasts promotes mammary carcinoma growth and invasion through upregulation of TGF-a-, MSP- and HGFmediated signaling networks. Oncogene 24:5053–5068 93. Bhowmick NA, Moses HL (2005) Tumor-stroma interactions. Curr Opin Genet Dev 15:97–101 94. Bierie B, Moses HL (2006) Tumour microenvironment: TGF-b: the molecular Jekyll and Hyde of cancer. Nat Rev Cancer 6:506–520 95. Yang L, Huang J, Ren X, Gorska AE, Chytil A, Aakre M, Carbone DP, Matrisian LM, Richmond A, Lin PC, Moses HL (2008) Abrogation of TGF-b signaling in mammary carcinomas recruits Gr-1+CD11b+ myeloid cells that promote metastasis. Cancer Cell 13:23–35 96. Kim BG, Li C, Qiao W, Mamura M, Kasprzak B, Anver M, Wolfraim L, Hong S, Mushinski E, Potter M, Kim SJ, Fu XY, Deng C, Letterio JJ (2006) Smad4 signalling in T cells is required for suppression of gastrointestinal cancer. Nature 441:1015–1019 97. Bian Y, Terse A, Du J, Hall B, Molinolo A, Zhang P, Chen W, Flanders KC, Gutkind JS, Wakefield LM, Kulkarni AB (2009) Progressive tumor formation in mice with conditional deletion of TGF-b signaling in head and neck epithelia is associated with activation of the PI3K/Akt pathway. Cancer Res 69:5918–5926 98. Zeng Q, Phukan S, Xu Y, Sadim M, Rosman DS, Pennison M, Liao J, Yang GY, Huang CC, Valle L, Di Cristofano A, de la Chapelle A, Pasche B (2009) Tgfbr1 haploinsufficiency is a potent modifier of colorectal cancer development. Cancer Res 69:678–686 99. Weaver VM, Fischer AH, Peterson OW, Bissell MJ (1996) The importance of the microenvironment in breast cancer progression: recapitulation of mammary tumorigenesis using a unique human mammary epithelial cell model and a three-dimensional culture assay. Biochem Cell Biol 74:833–851 100. Benton G, George J, Kleinman HK, Arnaoutova IP (2009) Advancing science and technology via 3D culture on basement membrane matrix. J Cell Physiol 221:18–25 101. Kenny PA, Lee GY, Myers CA, Neve RM, Semeiks JR, Spellman PT, Lorenz K, Lee EH, Barcellos-Hoff MH, Petersen OW, Gray JW, Bissell MJ (2007) The morphologies of breast cancer cell lines in three-dimensional assays correlate with their profiles of gene expression. Mol Oncol 1:84–96 102. Butcher DT, Alliston T, Weaver VM (2009) A tense situation: forcing tumour progression. Nat Rev Cancer 9:108–122 103. Erler JT, Weaver VM (2009) Three-dimensional context regulation of metastasis. Clin Exp Metastasis 26:35–49 104. Paszek MJ, Weaver VM (2004) The tension mounts: mechanics meets morphogenesis and malignancy. J Mammary Gland Biol Neoplasia 9:325–342 105. Turley EA, Veiseh M, Radisky DC, Bissell MJ (2008) Mechanisms of disease: epithelialmesenchymal transition–does cellular plasticity fuel neoplastic progression? Nat Clin Pract Oncol 5:280–290 106. Tian M, Schiemann WP (2009) Preclinical efficacy of cystatin C to target the oncogenic activity of TGF-b in breast cancer. Transl Oncol 2:174–183 107. Bissell MJ (2007) Modelling molecular mechanisms of breast cancer and invasion: lessons from the normal gland. Biochem Soc Trans 35:18–22 108. Mintz B, Illmensee K (1975) Normal genetically mosaic mice produced from malignant teratocarcinoma cells. Proc Natl Acad Sci U S A 72:3585–3589 109. Hochedlinger K, Blelloch R, Brennan C, Yamada Y, Kim M, Chin L, Jaenisch R (2004) Reprogramming of a melanoma genome by nuclear transplantation. Genes Dev 18:1875–1885 110. Bissell MJ, Labarge MA (2005) Context, tissue plasticity, and cancer: are tumor stem cells also regulated by the microenvironment? Cancer Cell 7:17–23 111. Dolberg DS, Hollingsworth R, Hertle M, Bissell MJ (1985) Wounding and its role in RSVmediated tumor formation. Science 230:676–678

5

Regulation of TGF-β Signaling and Metastatic Progression…

139

112. Paszek MJ, Zahir N, Johnson KR, Lakins JN, Rozenberg GI, Gefen A, Reinhart-King CA, Margulies SS, Dembo M, Boettiger D, Hammer DA, Weaver VM (2005) Tensional homeostasis and the malignant phenotype. Cancer Cell 8:241–254 113. Levental KR, Yu H, Kass L, Lakins JN, Egeblad M, Erler JT, Fong SF, Csiszar K, Giaccia A, Weninger W, Yamauchi M, Gasser DL, Weaver VM (2009) Matrix crosslinking forces tumor progression by enhancing integrin signaling. Cell 139:891–906 114. Erler JT, Bennewith KL, Cox TR, Lang G, Bird D, Koong A, Le QT, Giaccia AJ (2009) Hypoxia-induced lysyl oxidase is a critical mediator of bone marrow cell recruitment to form the premetastatic niche. Cancer Cell 15:35–44 115. Erler JT, Bennewith KL, Nicolau M, Dornhofer N, Kong C, Le QT, Chi JT, Jeffrey SS, Giaccia AJ (2006) Lysyl oxidase is essential for hypoxia-induced metastasis. Nature 440:1222–1226 116. Erler JT, Giaccia AJ (2006) Lysyl oxidase mediates hypoxic control of metastasis. Cancer Res 66:10238–10241 117. Sion AM, Figg WD (2006) Lysyl oxidase (LOX) and hypoxia-induced metastases. Cancer Biol Ther 5:909–911 118. Payne SL, Hendrix MJ, Kirschmann DA (2007) Paradoxical roles for lysyl oxidases in cancer–a prospect. J Cell Biochem 101:1338–1354 119. Le QT, Harris J, Magliocco AM, Kong CS, Diaz R, Shin B, Cao H, Trotti A, Erler JT, Chung CH, Dicker A, Pajak TF, Giaccia AJ, Ang KK (2009) Validation of lysyl oxidase as a prognostic marker for metastasis and survival in head and neck squamous cell carcinoma: Radiation Therapy Oncology Group trial 90–03. J Clin Oncol 27:4281–4286 120. Taylor MA, Amin J, Kirschmann DA and Schiemann WP (2011). Lysyl oxidase and hydrogen peroxide promote oncogenic signaling by transforming growth factor-b in mammary epithelial cells. (submitted) 121. Patsialou A, Wyckoff J, Wang Y, Goswami S, Stanley ER, Condeelis JS (2009) Invasion of human breast cancer cells in vivo requires both paracrine and autocrine loops involving the colony-stimulating factor-1 receptor. Cancer Res 69:9498–9506 122. Wyckoff J, Wang W, Lin EY, Wang Y, Pixley F, Stanley ER, Graf T, Pollard JW, Segall J, Condeelis J (2004) A paracrine loop between tumor cells and macrophages is required for tumor cell migration in mammary tumors. Cancer Res 64:7022–7029 123. Wang W, Wyckoff JB, Goswami S, Wang Y, Sidani M, Segall JE, Condeelis JS (2007) Coordinated regulation of pathways for enhanced cell motility and chemotaxis is conserved in rat and mouse mammary tumors. Cancer Res 67:3505–3511 124. Roussos ET, Keckesova Z, Haley JD, Epstein DM, Weinberg RA, Condeelis JS (2010) AACR Special Conference on epithelial-mesenchymal transition and cancer progression and treatment. Cancer Res 70:7360–7364 125. Bhowmick NA, Ghiassi M, Bakin A, Aakre M, Lundquist CA, Engel ME, Arteaga CL, Moses HL (2001) TGF-b1 mediates epithelial to mesenchymal transdifferentiation through a RhoAdependent mechanism. Mol Biol Cell 12:27–36 126. Kim W, Seok Kang Y, Soo Kim J, Shin NY, Hanks SK, Song WK (2008) The integrin-coupled signaling adaptor p130Cas suppresses Smad3 function in TGF-b signaling. Mol Biol Cell 19:2135–2146 127. Cabodi S, Tinnirello A, Di Stefano P, Bisaro B, Ambrosino E, Castellano I, Sapino A, Arisio R, Cavallo F, Forni G, Glukhova M, Silengo L, Altruda F, Turco E, Tarone G, Defilippi P (2006) p130Cas as a new regulator of mammary epithelial cell proliferation, survival, and HER2-Neu oncogene-dependent breast tumorigenesis. Cancer Res 66:4672–4680 128. Lee YH, Albig AR, Regner M, Schiemann BJ, Schiemann WP (2008) Fibulin-5 initiates epithelial-mesenchymal transition (EMT) and enhances EMT induced by TGF-b in mammary epithelial cells via a MMP-dependent mechanism. Carcinogenesis 29:2243–2251 129. Tilghman RW, Cowan CR, Mih JD, Koryakina Y, Gioeli D, Slack-Davis JK, Blackman BR, Tschumperlin DJ, Parsons JT (2010) Matrix rigidity regulates cancer cell growth and cellular phenotype. PLoS One 5:e12905

140

M.K. Wendt and W.P. Schiemann

130. Chao YL, Shepard CR, Wells A (2010) Breast carcinoma cells re-express E-cadherin during mesenchymal to epithelial reverting transition. Mol Cancer 9:179 131. Wipff PJ, Rifkin DB, Meister JJ, Hinz B (2007) Myofibroblast contraction activates latent TGF-b1 from the extracellular matrix. J Cell Biol 179:1311–1323 132. Folkman J (1985). Angiogenesis and its inhibitors. Important Adv Oncol, 42–62 133. Folkman J (1971) Tumor angiogenesis: therapeutic implications. N Engl J Med 285:1182–1186 134. Paez-Ribes M, Allen E, Hudock J, Takeda T, Okuyama H, Vinals F, Inoue M, Bergers G, Hanahan D, Casanovas O (2009) Antiangiogenic therapy elicits malignant progression of tumors to increased local invasion and distant metastasis. Cancer Cell 15:220–231 135. Lester RD, Jo M, Montel V, Takimoto S, Gonias SL (2007) uPAR induces epithelial mesenchymal transition in hypoxic breast cancer cells. J Cell Biol 178:425–436 136. Dunn LK, Mohammad KS, Fournier PG, McKenna CR, Davis HW, Niewolna M, Peng XH, Chirgwin JM, Guise TA (2009) Hypoxia and TGF-b drive breast cancer bone metastases through parallel signaling pathways in tumor cells and the bone microenvironment. PLoS One 4:e6896 137. Heikkinen PT, Nummela M, Jokilehto T, Grenman R, Kahari VM, Jaakkola PM (2010) Hypoxic conversion of SMAD7 function from an inhibitor into a promoter of cell invasion. Cancer Res 70:5984–5993 138. Xu X, Han J, Ito Y, Bringas P Jr, Deng C, Chai Y (2008) Ectodermal Smad4 and p38 MAPK are functionally redundant in mediating TGF-b/BMP signaling during tooth and palate development. Dev Cell 15:322–329 139. DeNardo DG, Barreto JB, Andreu P, Vasquez L, Tawfik D, Kolhatkar N, Coussens LM (2009) CD4(+) T cells regulate pulmonary metastasis of mammary carcinomas by enhancing protumor properties of macrophages. Cancer Cell 16:91–102 140. Viloria-Petit AM, David L, Jia JY, Erdemir T, Bane AL, Pinnaduwage D, Roncari L, Narimatsu M, Bose R, Moffat J, Wong JW, Kerbel RS, O’Malley FP, Andrulis IL, Wrana JL (2009) A role for the TGF-b-Par6 polarity pathway in breast cancer progression. Proc Natl Acad Sci U S A 106:14028–14033 141. Sodunke TR, Turner KK, Caldwell SA, McBride KW, Reginato MJ, Noh HM (2007) Micropatterns of Matrigel for three-dimensional epithelial cultures. Biomaterials 28: 4006–4016 142. Korpal M, Yan J, Lu X, Xu S, Lerit DA, Kang Y (2009) Imaging TGF-b signaling dynamics and therapeutic response in breast cancer bone metastasis. Nat Med 15:960–966 143. Giampieri S, Manning C, Hooper S, Jones L, Hill CS, Sahai E (2009) Localized and reversible TGF-b signalling switches breast cancer cells from cohesive to single cell motility. Nat Cell Biol 11:1287–1296 144. Giampieri S, Pinner S, Sahai E (2010) Intravital imaging illuminates TGF-b signaling switches during metastasis. Cancer Res 70:3435–3439 145. Fidler IJ (1970) Metastasis: guantitative analysis of distribution and fate of tumor embolilabeled with 125 I-5-iodo-2¢-deoxyuridine. J Natl Cancer Inst 45:773–782 146. Wendt MK, Cooper AN, Dwinell MB (2008) Epigenetic silencing of CXCL12 increases the metastatic potential of mammary carcinoma cells. Oncogene 27:1461–1471 147. Mani SA, Guo W, Liao MJ, Eaton EN, Ayyanan A, Zhou AY, Brooks M, Reinhard F, Zhang CC, Shipitsin M, Campbell LL, Polyak K, Brisken C, Yang J, Weinberg RA (2008) The epithelial-mesenchymal transition generates cells with properties of stem cells. Cell 133:704–715 148. Nagaprashantha LD, Vatsyayan R, Lelsani PC, Awasthi S, Singhal SS (2011) The sensors and regulators of cell-matrix surveillance in anoikis resistance of tumors. Int J Cancer 128: 743–752 149. Debnath J (2008) Detachment-induced autophagy during anoikis and lumen formation in epithelial acini. Autophagy 4:351–353 150. Shibue T, Weinberg RA (2009) Integrin beta1-focal adhesion kinase signaling directs the proliferation of metastatic cancer cells disseminated in the lungs. Proc Natl Acad Sci U S A 106:10290–10295

5

Regulation of TGF-β Signaling and Metastatic Progression…

141

151. Barkan D, Kleinman H, Simmons JL, Asmussen H, Kamaraju AK, Hoenorhoff MJ, Liu ZY, Costes SV, Cho EH, Lockett S, Khanna C, Chambers AF, Green JE (2008) Inhibition of metastatic outgrowth from single dormant tumor cells by targeting the cytoskeleton. Cancer Res 68:6241–6250 152. Barkan D, El Touny LH, Michalowski AM, Smith JA, Chu I, Davis AS, Webster JD, Hoover S, Simpson RM, Gauldie J, Green JE (2010) Metastatic growth from dormant cells induced by a Col-I-enriched fibrotic environment. Cancer Res 70:5706–5716 153. Weigelt B, Bissell MJ (2008) Unraveling the microenvironmental influences on the normal mammary gland and breast cancer. Semin Cancer Biol 18:311–321 154. Gupta GP, Perk J, Acharyya S, de Candia P, Mittal V, Todorova-Manova K, Gerald WL, Brogi E, Benezra R, Massague J (2007) ID genes mediate tumor reinitiation during breast cancer lung metastasis. Proc Natl Acad Sci U S A 104:19506–19511 155. Pathi SP, Kowalczewski C, Tadipatri R, Fischbach C (2010) A novel 3-D mineralized tumor model to study breast cancer bone metastasis. PLoS One 5:e8849 156. Fong YC, Maa MC, Tsai FJ, Chen WC, Lin JG, Jeng LB, Yang RS, Fu WM, Tang CH (2008) Osteoblast-derived TGF-b stimulates IL-8 release through AP-1 and NF-kB in human cancer cells. J Bone Miner Res 23:961–970 157. Mohammad KS, Javelaud D, Fournier PG, Niewolna M, McKenna CR, Peng XH, Duong V, Dunn LK, Mauviel A, Guise TA (2011) The TGF-b receptor I kinase inhibitor SD-208 reduces the development and progression of melanoma bone metastases. Cancer Res 71(1):175–184 158. Hu Z, Zhang Z, Guise T, Seth P (2010) Systemic delivery of an oncolytic adenovirus expressing soluble TGF-b receptor II-Fc fusion protein can inhibit breast cancer bone metastasis in a mouse model. Hum Gene Ther 21:1623–1629 159. Kang Y, He W, Tulley S, Gupta GP, Serganova I, Chen CR, Manova-Todorova K, Blasberg R, Gerald WL, Massague J (2005) Breast cancer bone metastasis mediated by the Smad tumor suppressor pathway. Proc Natl Acad Sci U S A 102:13909–13914 160. Yin JJ, Selander K, Chirgwin JM, Dallas M, Grubbs BG, Wieser R, Massague J, Mundy GR, Guise TA (1999) TGF-b signaling blockade inhibits PTHrP secretion by breast cancer cells and bone metastases development. J Clin Invest 103:197–206

Chapter 6

Rho GTPases and Their Activators, Guanine Nucleotide Exchange Factors (GEFs): Their Roles in Glioma Cell Invasion Bo Hu, Marc Symons, Bodour Salhia, Shannon P. Fortin, Nhan L. Tran, James Rutka, and Shi-Yuan Cheng

Abstract Accumulated studies showed that constitutively activated Rho GTPases such as Rac1 in malignant human glioblastomas are responsible for the highly invasive phenotype observed in these aggressive brain cancers. Notably, no activating mutations of Rac1 have been reported in human glioblastomas or other types of cancers. Moreover, Rac1 and several GEFs have been implicated in cell invasion and metastasis of glioblastomas and other types of human cancers. Mechanistically, the functions of the Rho GTPases are regulated by three distinct classes of molecules. Among them, guanine nucleotide exchange factors (GEFs) are the activators for Rho GTPases. Here, we review the role of Rho GTPases, particularly Rac1 and

B. Hu Cancer Institute, University of Pittsburgh, School of Medicine, Pittsburgh, PA 15213-1862, USA Department of Medicine, University of Pittsburgh, School of Medicine, Pittsburgh, PA 15213-1862, USA M. Symons Center for Oncology and Cell Biology, The Feinstein Institute for Medical Research at North Shore-LIJ, 350 Community Drive, Manhasset, NY 11030, USA B. Salhia • S.P. Fortin • N.L. Tran Translational Genomics Research Institute, 445 N. Fifth Street, Phoenix, AZ 85004, USA J. Rutka The Arthur and Sonia Labatt Brain Tumor Research Center, Hospital for Sick Children, Toronto, ON M5G 1X8, Canada S.-Y. Cheng (*) Cancer Institute, University of Pittsburgh, School of Medicine, Pittsburgh, PA 15213-1862, USA Department of Pathology, University of Pittsburgh, School of Medicine, Pittsburgh, PA 15213-1862, USA e-mail: [email protected]

A. Fatatis (ed.), Signaling Pathways and Molecular Mediators in Metastasis, DOI 10.1007/978-94-007-2558-4_6, © Springer Science+Business Media B.V. 2012

143

144

B. Hu et al.

several GEFs including Trio, SWAP-70, Ect2, Vav3 and Dock180-ELMO1 in glioblastoma cell invasion. Since studies of Rho GTPases and their GEFs in glioblastomas are just emerging, we place specific emphasis on the current knowledge of their roles in cell motility and cancer cell invasion as well as potential response to extracellular stimuli that promotes glioma cell invasion. Additionally, we include studies of two membrane receptor proteins, Fn14 and TROY that promote glioma cell invasion through activation of Rho GTPases. Lastly, we discuss future directions for understanding the functions of Rac1 and GEFs in glioma cell invasion and implications to establish these key modulators as potential targets to inhibit diffusely invasive glioblastomas in the brain. The insight provided by this review will help to develop new therapeutic approaches to treat patients with malignant gliomas.

Abbreviations BAI-1 Crk Dbl DD DH domain DHR domain Dock Ect2 EGFR ELMO1 Fn14 GAP GBM GDI GEF p130Cas PDGFR PH domain PtdIns Pyk RNAi ROCK SH2 or 3 SWAP TNF TNFRSF TRAF Trio

Brain-specific angiogenesis inhibitor-1 v-crk sarcoma virus CT10 oncogene homolog Diffuse B-cell lypmphoma Death domain Dbl homology (DH) domain Dock-homology region Dedicator of cytokinesis Epithelial cell transforming sequence 2 Epithelial growth factor receptor Engulfment and cell motility-1 Fibroblast growth factor inducible 14 GTPase-activating protein Glioblastoma multiforme Guanine nucleotide dissociation inhibitor Guanine nucleotide exchange factor (GDP–GTP exchange factor) crk-associate substrate 130 kDa protein Platelet-derived growth factor receptor Pleckstrin homology domain Phosphatidylinositol proline-rich tyrosine kinase RNA interference Rho kinase src homologous domain 2 or 3 switching B-cell complex 70 kDa subunit Tumor necrosis factor Tumor necrosis factor receptor superfamily Tumor necrosis factor receptor associated factor Triple function domains (PTPRF interacting)

6 Rac1 and its GEFs in Glioma Cell Invasion

TROY/Tnfrsf19 TSC TWEAK Vav3 VEGFR

6.1

145

Tumor necrosis factor receptor superfamily member 19 Triple sex combs Tumor necrosis factor-like weak inducer of apoptosis vav 3 guanine nucleotide exchange factor Vascular endothelial growth factor receptor

Introduction

Human glioblastoma multiforme (GBM) is the most common malignant brain cancer in adults and portends a dismal prognosis [1, 2]. A hallmark of human GBM is the intrinsic ability of single tumor cells to infiltrate throughout the brain, rendering these tumors virtually incurable by all existing therapies and also leading to their great propensity for recurrence [1, 2]. Genetic and clinical studies demonstrated that a number of signaling pathways are constitutively activated in actively infiltrating glioma cells [3–5]. A key group of molecules that are involved in glioma cell invasion are the Rho GTPases. Rho GTPases are molecular switches that are essential in controlling cell movement, morphology, and survival [6, 7]. As with other members of the Ras family of GTPases, most Rho GTPase cycle between an active GTPbound state and an inactive GDP-bound state. In their GTP-bound state, GTPases recognize downstream targets to elicit intracellular responses including the regulation of actin cytoskeleton dynamics [7, 8]. Biochemical and genetic studies have led to the identification of three classes of regulatory proteins that control the nucleotide exchange or degradation, and therefore activation state of Rho GTPases. These are the guanine nucleotide exchange factors (GDP–GTP exchange factors or GEFs), the GTPase-activating proteins (GAPs) and the guanine nucleotide dissociation inhibitors (GDIs) [8–10]. Rho GTPases are activated by GEFs that promote the exchange of GDP to GTP. Subsequently, Rho GTPases are inactivated by GTPase activating proteins (GAPs), which stimulate the intrinsic GTP hydrolyzing activity of the GTPases [11]. Rho GTPases are also controlled by Rho GDIs. Rho GDIs can regulate GTPases in both a negative fashion, by preventing the interaction of GTPases with GEFs for example, and a positive fashion, by regulating proper intracellular targeting of GTPases [12, 13]. Importantly, the activity of all three classes of regulatory proteins is controlled by signal transduction events [12–15]. Rho GTPases are ubiquitously expressed and 22 members have been described in humans [16]. Among them, Rac1 is the most studied and has been shown to play a central role in promoting cancer cell invasion [17, 18]. Rac1 causes lamellipodia formation thus participating in actin-driven cell migration [17, 19]. Rac1 relays signals from a wide variety of growth factors, cytokines and adhesion proteins to a large array of effector proteins [20, 21]. GEFs activate Rho GTPases such as Rac1 by catalyzing the exchange of GDP for GTP in response to various extracellular signals. Many GEFs were identified as oncogenes after transfection of immortalized fibroblast cell lines with cDNA expression libraries [8–10]. Moreover, only a few GEFs such as Vav and Tiam1 have been

146

B. Hu et al.

implicated in tumor cell migration and invasion [22–25]. Until recently, very little was known regarding the expression or function of GEFs in GBM. However, during the last several years, with the knowledge that Rac1 is a potent inducer of glioma cell migration and invasion [18, 26, 27], a number of investigators who study glioma biology and tumor invasion have used various approaches to study the role of Rho GEFs, especially those that can activate Rac1, in stimulating glioma cell invasion. In this review, we describe our current knowledge of the Rho GTPases including Rac1 and several Rac1 GEFs in cell motility and the role of these molecules in the modulation of human glioma cell invasion. We also present our studies of two members of the TNF receptor superfamily (TNFRSF), Fn14 and TROY in glioma cell invasion. Lastly, we will discuss futuristic implications of the function and regulation of Rac1 GEFs in glioma cell invasion.

6.2

6.2.1

The Role of Rho GTPases and Rac1 in Glioma Cell Invasion Rho GTPases in Glioma Cell Invasion

Rho GTPase family proteins are thought to regulate cell invasion in a large part via controlling cell adhesion, cell polarity and the organization of the actin cytoskeleton [6, 7, 20, 28]. In human cancers, tumor cells can employ different modes of invasive behavior, termed mesenchymal and amoeboid [29, 30]. Mesenchymal invasion relies on Rac-mediated signaling and extension of lamellar structures which attach and pull the cell forwards, whereas amoeboid invasion is mediated by the Rho-ROCK signaling axis and involves blebbing protrusions and cortical contractility to squeeze the cell through pores in the extracellular matrix [29, 31, 32]. Notably, during the process of invasion and metastasis, tumor cells can switch between the mesenchymal and amoeboid modes of invasion. The mechanisms that control this plasticity are under intense investigation [33, 34]. An important consequence of this dual migratory behavior is that it may allow tumor cells to escape therapeutics that specifically target either mode of migration and therefore combinations of drugs targeting the respective modes of invasion may be needed to achieve clinical benefit. There is mounting evidence that Rho GTPases are deregulated in human tumors [16, 35, 36]. Unlike Ras itself where activating mutations are commonplace, with the exception of RhoH [37], which is frequently rearranged in lymphomas, thus far no activating mutations of Rho proteins have been identified in tumors [38, 39]. Rac1, Cdc42, RhoA and RhoC, however, have been found to be frequently overexpressed at the protein level in a wide range of tumors and in some cases, expression levels have been correlated with tumor grade [16, 40]. Notably, in astrocytomas, both Rac1 and RhoA proteins have been found to be overexpressed in high-grade versus low-grade tumors [27, 41]. In addition, the marked plasma membrane

6 Rac1 and its GEFs in Glioma Cell Invasion

147

localization of Rac observed in a significant subset of GBM tumors, but not in low-grade astrocytomas or non-neoplastic brain, indicates that this GTPase is hyperactivated in GBMs [27].

6.2.2

Rac1 in Glioma Cell Invasion

A role for the Rac1 GTPase in the invasive behavior of GBM cells has been established in several studies, both using an in vitro Matrigel invasion assay and an ex vivo organotypic brain slice invasion assay [18, 26, 42, 43]. In gliomas, Rac1 mediates glioma cell invasion promoted by fibroblast growth factor-inducible 14 [43], Ephrin-B3 [44] or inhibition of ROCK [26]. One potential mechanism that mediates the role of Rac1 in invasion is the activation of NF-kB [43]. Another mechanism is that Rac1, via its effector synaptojanin 2 [18], controls the formation of invadopodia, which are specialized domains of the plasma membrane at which extracellular matrix degradation takes place [45]. Interestingly, siRNA-mediated depletion of Rac3, a GTPase that is highly homologous with Rac1, also potently inhibits GBM cell invasion through Matrigel in vitro [18]. This finding suggests non-redundant roles for Rac1 and Rac3 in glioma cell invasion. The observation that inhibition of ROCK, an effector of both RhoA and RhoC, stimulates the invasion of GBM cells through Matrigel, both in two-dimensional [26] and three-dimensional settings (Symons M et al, unpublished observations), indicates that in these conditions, GBM cells employ a mesenchymal mode of invasion. In line with this, siRNA-mediated depletion of Rac1 [18], or inhibition of matrix-degrading proteases (Symons M et al, unpublished observations) also inhibits glioblastoma cell invasion through Matrigel. Interestingly however, inhibition of ROCK strongly inhibits the invasive behavior of glioblastoma cells in organotypic brain slices [46]. Thus, these observations support the notion that GBM cells can adapt their invasive behavior to their environmental conditions. The roles of RhoA and RhoC, as well as other Rho GTPase family members, in GBM invasiveness remain to be explored.

6.3

6.3.1

The Role of Rho GEFs, Trio, SWAP-70, Vav3 and Ect2 and in Glioma Cell Invasion Guanine Nucleotide Exchange Factors (GEFs)

The first mammalian Rho GEF was identified as a transforming gene in diffuse B-cell lymphoma cells, and was therefore designated Dbl [9]. Since then 80 Rho GEFs were identified in humans and at least 27 GEFs stimulate Rac1 activity [8]. According to their protein structures Rho GEFs are divided into two distinct families.

148

B. Hu et al.

The classical GEF family contains the Dbl homology (DH) domain and an adjacent Pleckstrin homology (PH) domain [8–10]. DH domains are responsible for catalyzing the exchange of GDP for GTP within Rho GTPases by transiently promoting GTPase intermediates that are devoid of nucleotide and Mg2+ [8]. Cellular GTP is present in approximately tenfold molar excess relative to GDP, and binding of GTP stimulates the release of the protein from the GEF. DH-associated PH domains, by binding phosphoinositides, have been proposed to localize Dbl proteins to plasma membranes and to regulate their GEF activity through allosteric mechanisms [8]. This in turn renders the GTPase active and allows it to initiate a signaling complex with one of several effectors. Outside the DH-PH domains, Dbl-family proteins show significant divergence and typically contain other protein domains that underlie the unique cellular functions of the different family members [8]. Since the identification of Dbl, 69 distinct mammalian members of this Dbl family have been identified in humans [8, 47]. More recently, another smaller family of Rho GEFs has been identified related to the founding member Dock1 (dedicator of cytokinesis 1, also called Dock180) [48]. These are unrelated in primary sequence to the Dbl family proteins and instead are characterized by two regions of high sequence conservation that are designated Dock-homology region -1 and -2 (DHR-1 and DHR-2). The Dock family members are considered unconventional or atypical GEFs. Within the Dock family, 11 are found in humans [8, 14, 49]. In the case of Dock180, the prototypical Dock in this family, the DHR-1 domain was shown to mediate a specific interaction with phophatidylinositol (PtdIns)(3, 5)-biphosphate and PtdIns(3,4,5)P3 signals in vitro and in cells [50]. The DHR-1 domain was shown to have a critical role in Dock-Rac signaling by positioning the Dock GEF downstream of PI3K which produces PtdIns(3,4,5)P3 in response to various extracellular stimuli and Rac activation [51]. Similar to DH-PH modules in Dbl GEFs, the DHR-2 domain mediates interactions with the nucleotide-free form of the Rho GTPases that they catalytically target, leading to the exchange of GDP to GTP on Rho GTPases [52]. Several DHR-2 domains have been shown to be both necessary and sufficient to promote specific guanidine nucleotide exchange on various GTPases in vitro and in vivo [49, 51, 53]. Although several Dbl GEFs such as Vavs and Tiams are regulated by receptor tyrosine kinases (RTKs) through protein phosphorylation and direct interaction with the RTKs [54], little is known about how GEFs are regulated. Our understanding of the function and regulation of these GEFs will be central to determining the mechanisms that underlie the spatial, as well as the temporal, activation of GTPases during glioma cell invasion in response to outside influences. With the knowledge that Rac1 is a potent inducer of glioma cell migration and invasion [18, 26, 27], a gene expression array database was examined comparing human gliomas to non-neoplastic tissue to identify GEFs that act on Rac proteins and may be important for glioma invasion [27]. Indeed, targeting Rho GEFs is likely to generate less generalized toxicity than direct targeting of Rac. Based on increased expression when compared with non-neoplastic brain; association with higher tumor grade and poor patient outcome, we selected Trio, SWAP-70, Ect2 and Vav3 for further study [27, 55, 56]. Expression levels of these Rac GEFs were validated

6 Rac1 and its GEFs in Glioma Cell Invasion

149

using quantitative real-time polymerase chain reaction (PCR) and protein levels assessed using immunohistochemistry in independent glioma specimens [18, 26, 27]. Depletion of the respective GEFs using RNA interference significantly inhibited the migratory and invasive behavior of glioblastoma cells using radial cell and organotypic brain slice models [27, 55, 56].

6.3.2

Trio and SWAP-70

Most of the Rac GEF family members are expressed preferentially in specific tissues and at different developmental stages, suggesting involvement in specific events during embryogenesis [8]. The multidomain protein Trio is the founding member of an intriguing family of Rho-GEFs that contains two GEF domains, a kinase domain and numerous accessory domains [57, 58]. The first DH–PH domain module of Trio is N-terminally situated (TrioN or GEFD1) and activates the Rac pathway directly [59] or indirectly through RhoG [60]. The second GEF domain of Trio situated in the C-terminus (TrioC or GEFD2) acts on RhoA [57], suggesting that full-length Trio could activate several GTPases in vivo. The members of the Trio family play an important role in neurogenesis, phagocytosis, and myogenesis [61–63] and their structural organization is very well conserved through evolution [64]. Trio is a modulator of the cytoskeleton through the Rho and/or Rac pathways and is vital in axon guidance, neuronal cell migration [58], in regulation of focal adhesion dynamics [65] as well as in anchorage-independent growth [66]. In the adult mammalian CNS, Trio expression in the nervous system is limited to neurons, and neither mRNA nor protein levels were detected in blood vessel walls, the ependymal layer of the third ventricle or in glial cells [67, 68]. In addition to Trio being upregulated grade-dependently at both the mRNA and protein levels in tumor cells of human GBMs [27], high expression levels of Trio were also seen in breast tumors, especially in those with poor prognosis [69]. Tgat, an alternative Trio isoform derived from TrioC, was identified as a new oncogene in adult T-leukemia [70]. Accordingly, glial, mammary and T-cell transformation is somehow associated with increased expression of Trio. Trio is also involved in the invasive behavior of glioblastoma and breast cancers [27, 69]. In GBM, Trio is expressed at higher levels in the tumor rim versus the core lending further support to the importance of Trio during the invasive process (unpublished data). This suggests that aberrant regulation and activity of Trio may have an important prognostic value in breast cancer and leukemia as well as gliomas. Although considered to be promising therapeutic targets, the only GEF inhibitors identified so far act against Trio [10, 71–73]. To date, two types of Trio inhibitors have been characterized: Tripa and derived peptides, which associate with and inhibit TrioC [10, 72], and NPPD and ITX3, which are compounds biochemically active on TrioN. NPPD and ITX3 were identified from a screen of 3,500 chemical compounds using a GEF activity assay in a permeant yeast strain [71, 73]. NPPD

150

B. Hu et al.

is toxic in mammalian cells and blocked TrioN-mediated activity of RhoG and Rac1 thus disrupting various physiological processes [71]. ITX3 is a nontoxic inhibitor and in transfected mammalian cells, ITX3 blocks TrioN-mediated dorsal membrane ruffling and Rac1 activation while having no effect on other GEFmediated RhoA or Rac1 activation. Thus unlike NSC23766 which targets a Rac region involved in binding TrioN- and Tiam1 [74], ITX3 has no effect on Tiam1. ITX3 specifically inhibited endogenous TrioN activity, as evidenced by its ability to inhibit neurite outgrowth in nerve growth factor (NGF)–stimulated PC12 cells or C2C12 differentiation into myotubes [73]. The emergence of GEF compounds like ITX3 validate the notion that GEFs can specifically be targeted in cancer cells with minimal toxicity. SWAP-70 is a multifunctional signaling protein involved in membrane ruffling that works cooperatively with activated Rac [56, 75]. It has been suggested that SWAP-70 acts as an effector or adaptor in response to PI3-kinase activity and is important in regulating the actin cytoskeleton and cell motility by promoting PtdIns(3,4,5)P3–dependent activation of Rac [75, 76]. SWAP-70 is expressed in various tissues, such as heart, kidney, spleen, and liver [76] but not expressed to any significant degree in the brain [56]. Src-transformed mouse embryonic fibroblasts lacking SWAP-70 were found to be significantly less invasive than their parental counterparts, and expression of both SWAP-70 and Src induced constant membrane ruffling but not found with Src alone [77].

6.3.3

Ect2

Epithelial cell transforming sequence 2 (Ect2) was originally identified as a proto-oncogene capable of transforming NIH3T3 cells and causing tumors in nude mice when an N-terminally truncated Ect2 is expressed [78–80], suggesting a role of Ect2 overexpression in tumorigenesis. Paradoxically, full-length Ect2 is overexpressed in several human tumor types, including brain, lung, esophageal, pancreatic bladder and ovarian tumors [27, 55, 81, 82]. Moreover, Saito et al. demonstrated that both the removal of the negative regulatory domain and alteration of subcellular localization are required to induce the oncogenic activity of Ect2 [80]. Since the oncogenic activity of Ect2 is strongly inhibited by dominant negative Rho GTPases, Ect2 sequestered in the nucleus during interphase is thought to be inactive because nucleus-Ect2 is unable to activate Rho GTPases situated in the cytoplasm [80]. A study of GBMs using comparative genomic hybridization and a BAC DNA microarray of the human genome identified a chromosomal gain of 3q26.3, the ECT2 locus, in 3/10 GBMs [83]. Previous studies have also shown an important role for Ect2 in the regulation of cytokinesis [84]. A dominant-negative C-terminally truncated Ect2, or microinjection of anti-Ect2 antibody prevents contractile ring formation during cytokinesis and leads to the formation of multinucleated cells [84, 85]. This finding was validated in a study showing that inhibition of Ect2 by siRNA leads to formation of large multinucleated cells in human glioma SNB19 and U87MG cells [27].

6 Rac1 and its GEFs in Glioma Cell Invasion

151

Interestingly, whereas low-grade astrocytomas show predominantly nuclear Ect2 staining, GBMs display prominent staining of Ect2 in both the cytoplasm and nucleus [27]. It is highly probable that the sub-cellular localization of Ect2 in human astrocytomas dictates the functional significance and state of activation. This knowledge could be utilized in future studies for pre-screening patients, for example, for eligibility with anti-Ect2 therapy.

6.3.4

Vav3

In mammals, there are three Vav protein family members; Vav1 is specifically expressed in the hematopoietic system, whereas Vav2 and Vav3 are more ubiquitously expressed [86]. Vav proteins are evolutionarily conserved from nematodes to mammals and play a pivotal role in many aspects of cellular signaling, coupling cell surface receptors to various effector functions [86]. Vav proteins have several protein-protein interaction domains including a Src homology 2 (SH2) and two Src homology 3 (SH3) domains that can, respectively, interact with a large variety of tyrosine kinase receptors or adaptor proteins [87]. Vav proteins are regulated by tyrosine phosphorylation [86]. The participation of the Vav proteins in several processes that require cytoskeletal reorganization, such as the formation of the immunological synapse, phagocytosis, platelet aggregation, spreading, and transformation have been well documented [86]. Several lines of evidence show that Vav proteins are crucial for developmental, mitogenic, and pathological processes. Vav3 is expressed at low levels in nonneoplastic brain and low-grade gliomas [27] but expressed at high levels by immunohistochemistry in both the tumor cells and tumor-associated endothelial cells in all GBMs examined [27]. A number of growth factor receptors that can recruit Vav proteins, including EGFR, PDGFR, IGF1R, and TrkA are abundantly expressed in malignant glioblastomas, providing a potential mechanism for the role of Vav3 in glioblastoma cell migration and invasion [27]. Increased expression of Vav3 has also been reported in prostate cancer, whereby through the PI3KAkt pathway, Vav3 inappropriately activates androgen receptor signaling and stimulates growth in prostate cancer cells [88]. Vav3 is overexpressed in over 80% of human breast cancer specimens, particularly in poorly differentiated lesions [89]. Vav3 complexes with and activates estrogen receptor alpha partially via PI3K-Akt signaling leading to stimulated growth of breast cancer cells [89]. In anaplastic large cell lymphomas (ALCL), aberrant expression of the oncogenic fusion protein nucleophosmin-anaplastic lymphoma kinase (NPM-ALK) is present in 75% of cases. Vav3 is phosphorylated in NPM-ALK positive biopsies from patients suffering from ALCL, demonstrating the pathological relevance of this observation [90]. The use of Vav3-specific shRNA and a dominant negative Rac1 mutant inhibited the migration and invasion of NPM-ALK-expressing cells [90]. Hunter et al. demonstrated that overexpression of Vav2 or Vav3 in primary microvascular endothelial cells promotes Rac1 activation, cell migration and assembly

152

B. Hu et al.

in response to ephrin-A1 stimulation [91]. These findings collectively suggest that Vav3 overexpression may aberrantly enhance development and/or progression in various types of tumors including gliomas.

6.4

Dock180, a Rac1 GEF That Promotes Glioma Cell Invasion

6.4.1

The Dock Family Members

Based on the homologies in their protein sequences [48], the Dock family can be divided into four classes, DockA, which include Dock1, Dock2 and Dock5; DockB, which has Dock3 (modifier of cell adhesion, MOCA) and Dock4; DockC, which includes Dock6, Dock7 and Dock8 (Zir1, Zir2, and Zir3) and DockD, which has Dock9, Dock 10 and Dock11 (Zizimin1, Zizimin2 and Zizimin3) [49]. In addition to the well-studied Dock1, the roles of Dock2, Dock4, Dock7 and Dock9 in cell motility are better understood than other Docks. For the cellular functions of other Docks, please see more comprehensive reviews [47, 49, 51, 92].

6.4.1.1

Dock2

Dock2 was identified as a haematopoietic cell specific Dock essential for lymphocyte migration, interstitial motility and sphingosine-1-phosphate-mediated egress [93–95]. Dock2 associates with engulfment and cell motility 1 (ELMO1) through its Src-homology 3 (SH3) domain and CrkII, and regulates Rac1 activation of lymphocyte adhesion, motility and cytoskeletal re-organization [96–101]. Dock2 also regulates Rac1 and Rac2 stimulation of neutrophil chemotaxis [102, 103]. Moreover, intracellular Dock2 dynamics are sequentially regulated by distinct phospholipids to localize Rac activation during neutrophil chemotaxis [104]. Additionally, Dock2 is involved in Rac, Erk and other signals in activation of dendritic cells, T cells and cancer cells [105–107].

6.4.1.2

Dock4

Dock4 was initially cloned as a tumor suppressor gene that is homozygously deleted in a mouse NF2 and TP53 osteosarcoma model [108]. Dock4 mutations are present in a subset of human cancer cell lines and myelogenous leukemia, ovarian cancers, autism and dysplaxia [109–114]. Dock4 rescues deficiency caused by loss of a Dock1 homolog in C. elegans and suppresses tumor growth and invasion in Dock4deleted tumors [108]. Dock 4 mediates RhoG activation of Rap and Rac1 GTPases and promotes cell motility through interaction with ELMO1 or ELMO2 in cancer

6 Rac1 and its GEFs in Glioma Cell Invasion

153

cells and during dendritic development [115–118]. Additionally, Dock4 also enhances Wnt/beta-catenin activation of Rac1 and increases cell growth and transformation [119].

6.4.1.3

Dock7

Dock7 was identified as an activator of Rac GTPases through a yeast two-hybrid screening in the fetal brain. Dock7–Rac activation leads to phosphorylation and inactivation of the microtubule destabilizing proteins stathmin/Op18 thus regulating neuronal polarity during axonal development [120, 121]. In Schwann cells, stimulation of ErbB2 by neuregulin-1 that phosphorylates Dock7 at a tyrosine residue at its C-terminus, resulting in activation of Dock7-Rac1/Cdc42-JNK signaling and enhancement of cell migration [122]. Additionally, proteomic studies identified Dock7 as a binding partner for the TSC1 complex, suggesting its role in TSC1regulated tumorigenesis [123–125].

6.4.1.4

Dock9

A biochemical approach identified Dock9 (Zizimin1) as a specific activator for Cdc42 through direct interaction with a Cdc42-binding domain (CZH2). Dock9 activates Cdc42 and promotes cell motility ad dendrite growth in hippocampal neurons [49, 53, 126–128]. Dock9 also activates Rac2 through interacting with a specific region of Rac2 [129]. Recently, structural analysis identified a nucleotide sensor within the DHR2 domain that contributes to release of GDP and then to discharge of the activated GTP-bound Cdc42, thus providing the most complete picture of a GDP-GTP exchange cycle of modulation of Rho GTPase activation to date [130, 131].

6.4.2

Dock180-ELMO1 is a Bipartite GEF That Mediates Rac1-Promoted Cell Motility

Dock1 or Dock180 is the prototypical member of the Dock family [8, 47, 49, 51, 92]. Dock180 was first identified as a CrkII-binding protein that regulates NIH 3T3 cell morphology [132]. Dock180 interacts with Crk and p130Cas, leading to activation of Rac1 and stimulation of cell migration [133–136]. Genetic approaches identified a Dock180 homolog in C. elegans (CED-5) and Drosophila (myoblast city) that modulates cell body engulfment, cell migration, myoblast fusion, dorsal closure and cytoskeletal organization through interaction with CED-2/CrkII and CED-10/Rac [137–147]. Dock180 binds to PtdIns(3,4,5)P3 through its unconventional Docker DHR-1 domain, resulting in its membrane translocation [50, 148]. Dock180

154

B. Hu et al.

facilitates nucleotide exchange on Rac1 activation through its DHR-2 domain [48, 52, 145, 149]. Importantly, Dock180 forms a bipartite complex with engulfment and cell motility 1 (ELMO1) that is required for GDP/GTP exchange and stimulation of Rac1 activity and cell motility [52]. In C. elegans, Drosophila and mammalian cells, ELMO1 and its homolog, CED-12 enhances phagocytosis and cell migration by forming a bipartite complex with CED-5/Dock180 [150]. This bipartite GEF synergistically functions upstream of CED-10/Rac1 promoting Rac-dependent cell migration and cell corpse engulfment [151–153]. The Dock180-ELMO1 complex interacts with CrkII that in turn regulates the stability and function of this bipartite GEF [154]. Dock180-ELMO1 further forms a complex including paxillin-Crk-p130Cas and Rac1 that mediates early flow-induced polarization of actin-based protrusion of the lamellipodium in wounded monolayers and single cells [155]. Moreover, other evidence showed that Dock180-ELMO1 and CrkII might act independently for efficient removal of apoptotic cells [156]. During bacterial invasion into epithelial cells, IpgB1, an effector of Shigellar driver mimic the role of RhoG by interacting with the ELMO1-Dock180 complex, thus stimulating Rac1 and promoting membrane ruffle formation [157]. However, RhoG is not required for integrin-mediated ELMO1-Dock180-Rac1 signaling and cell migration [158]. In Dictyoselium, DdELMO1 associates with DockD but not DockA and activates both Rac1A and RacC, stimulating cell migration [159]. Interestingly, Dock180 itself is subjected to constant ubiquitin-mediated protein degradation and ELMO1 acts as an inhibitor of ubiquitination of Dock180 thus preventing Dock180 protein from degradation [160]. This also explains a constant association of ELMO1 with Dock180 in unstimulated cells [150].

6.4.3

Dock180-ELMO1 Couples Receptor Signaling to Rac

Similar to classical GEFs, the bipartite GEF Dock180-ELMO1 couples receptor signaling to its target Rho GTPase, Rac1, thus promoting receptor-induced cell motility, engulfment, myoblast fusion, dorsal closure and cytoskeletal organization [54]. In Drosophila, PDGFR and EGFR are essential for guidance of cell migration of embryonic precursors of blood cells and border cell clusters [161–166]. Significantly, Dock180 was found directly downstream of PDGFR/VEGFR- and EGFR- signaling during Drosophila development. Genetic evidence showed that myoblast city/Dock180 mediates two separate steps of PDGFR/VEGFR- and EGFR-induced blood cell precursor migration [163–166]. In mammalian cells, integrin receptors recruit the CrkII-Dock180-Rac1 complex to the membrane for phagocytosis of apoptotic cells and cell adhesion to laminin10/11 and fibronectin [136, 167–169]. Dock180-ELMO1 also mediates urokinase plasminogen activator receptor (uPAR)-induced cell migration [170]. Additionally, the p130Cas-CrkII-Dock180 complex was found to be involved in lysyl oxidase regulation of actin filament formation and Fcg receptor-mediated phagocytosis [171, 172].

6 Rac1 and its GEFs in Glioma Cell Invasion

155

In mice, Dock180 is essential during embryonic development and deletion of Dock180 impaired myoblast fusion during muscle development leading to postnatal death of pups [173], corroborating what had been observed for the role of myoblast city/Dock in Drosophila [161–164] and zebrafish [146]. The role of ELMO1 during development has recently been investigated. In zebrafish, ELMO1 is expressed in the vascular and neuronal system. Morpholino-knockdown of ELMO1 severely impaired the formation of vasculature and Netrin-1 and its receptor, Unc5B were identified as the upstream activator of the Dock180ELMO1 complex leading to activation of Rac1 in endothelial cells and vessel formation [174]. During engulfment of apoptotic cells, phosphatidylserine acts as an “eat-me” signal that binds to its receptor, brain-specific angiogenesis inhibitor (BAI). BAI cooperates with Dock180-ELMO1-Rac1 complex to promote maximal engulfment of apoptotic cells [175]. Moreover, ELMO1-deficient mice were unexpectedly viable and grossly normal. However, in the testis of ELMOnull mice, disrupted seminiferous epithelium, multinucleated giant cells, uncleaned apoptotic germ cells and decreased sperm production were found, indicating a selective requirement for ELMO1 in clearance of apoptotic germ cells in the testis [176, 177]

6.4.4

ELMO1-Dock180 Plays an Important Role in Promoting Glioma Cell Invasion

Although it has been established that the atypical bipartite Rac1 GEF Dock180ELMO1 plays critical roles in the activation of Rac1 signaling and stimulation of cell motility, studies of the role of Dock180-ELMO1 in human cancer invasion and metastasis are limited [8, 47, 49, 51, 92]. Increased expression or mutation of Dock180 has not been reported in human cancers and other diseases. However, genetic variations of ELMO1 were found to associate with type 1 and type 2 diabetes [178, 179–181]. In human gliomas, expression of ELMO1 is elevated in a subtype of clinical glioblastomas [182]. In Src-transformed fibroblasts, the p130Cas-Dock180 complex was found to mediate Src-FAK stimulation of cell motility and invasion [183]. In human ovarian cancer cells knockdown of Dock180 decreased expression of ELMO1 and Rac1 activity, changed cell morphology and reduced cell proliferation and migration [184]. Some of us have reported that the bipartite GEF Dock180-ELMO1 plays a critical role in glioma cell invasion [185]. Immunohistochemical analysis of primary human glioma specimens showed that Dock180 and ELOM1 are highly expressed in actively infiltrating glioma cells within the invasive regions along blood vessels, neuronal structures, and corpus callosum, but not in the central regions of these tumors. Elevated expression of ELMO1 and Dock180 was also found in various human glioma cell lines compared to normal human astrocytes. Furthermore, suppression of endogenous ELMO1 and Dock180 expression by each type of siRNA significantly impeded glioma cell migration and invasion

156

B. Hu et al.

in vitro, and in brain tissue slices with a concomitant reduction in Rac1 activation. Conversely, exogenous expression of ELMO1 and Dock180 in glioma cells with low levels of endogenous expression enhances their migratory and invasive capacities in vitro and in brain tissue. These data suggest that the bipartite GEF, ELMO1 and Dock180 play an important role in promoting cancer cell invasion and could be potential therapeutic targets for the treatment of diffuse malignant gliomas [185].

6.5

6.5.1

The Role of the TNFRSF Members, Fn14 and TROY in Glioma Cell Invasion Fn14

Cytokines are membrane bound or soluble proteins released from cells that act mechanistically via autocrine, paracrine, or endocrine signaling. These signals regulate important biological activities including reproduction, growth, development, homeostasis, injury response and repair, and inflammation [186]. The tumor necrosis factor (TNF) superfamily of ligands is a subgroup of cytokines heavily researched for their role in many disease states, including human cancers and glioblastomas. The TNF cytokines are type II proteins that can exist in either a membrane embedded or cleaved soluble form. The ligands can be active in either form, via assembly as a non-covalent homotrimer [187]. TNF receptor superfamily (TNFRSF) members are known to have a scaffold of disulfide bridges that gives rise to the cysteine rich domains required to be a family member. TNF receptors signal via protein-protein interactions to promote either cell death or cell survival [188]. Cytoplasmic domains may contain TNF receptor associated factors (TRAFs) and death domains (DDs); the subset of TNFRSF members which contain a DD are referred to as death receptors [187], and the DD signals for cell death via caspase dependent signaling and activation of JUN kinase pathways [188, 189]. TRAF mediated survival pathways have been associated with the activation of NFkB [189]. Of the TNF receptors characterized for their role in cancer progression, TNFRSF12a, also known as Fibroblast Growth Factor-Inducible 14 or Fn14, is one member of the TNF receptor superfamily that has been implicated in tumor progression [190, 191]. Fn14 was first identified in murine cells as an immediate-early response gene that encodes a cell surface localized type Ia transmembrane protein [192]. Fn14 is the smallest known TNF receptor family member and contains only one cysteine rich domain in the extracellular region. The Fn14 cytoplasmic domain lacks a death domain but contains TRAF binding sites specific for TRAFs 1, 2, 3, and 5 [190]. Fn14 serves as a cell surface receptor for the multifunctional cytokine

6 Rac1 and its GEFs in Glioma Cell Invasion

157

tumor necrosis factor-like weak inducer of apoptosis (TWEAK) [193]. In humans, Fn14 mRNA is expressed at the highest relative level in heart, placenta, and kidney, at an intermediate level in lung, skeletal muscle and pancreas, and is relatively low in brain and liver tissue. Expression of Fn14 was found to associate with liver regeneration and hepatocarcinogenesis [194], and endothelial growth and migration [195]. Fn14 expression is induced during nerve regeneration and tissue injury, and in several types of human cancers including brain, breast, cervical, esophageal, hepatocellular, and testicular carcinoma in situ [191]. In brain tumors, Fn14 expression level is elevated in advanced glial tumors and is also up-regulated in migrating glioma cells in vitro [196] and in invading GBM cells in vivo [43]. Some of us have reported that activation of Fn14 by TWEAK promotes glioma cell migration, invasion and survival [43, 196, 197]. Inhibition of TWEAK-Fn14 binding by the Fn14 decoy receptor abrogates TWEAK-induced glioma cell migration and invasion [43, 196], suggesting that TWEAK-stimulated glioma cell motility via activation of the Fn14 receptor. In addition, several groups, including ours, have reported that Fn14 may be able to signal independent of the ligand when expressed ectopically in vitro. Specifically, ectopic expression of Fn14 has been shown to induce the NFkB and Rac1 signaling pathways [43, 190, 198] in various cellular systems including PC12 cell neurite extension and outgrowth, migration of rat aortic smooth muscle cells [199], pro-apoptotic stimulation of glioma cells [197], glioma cell migration and invasion [43], and invasion of oesophageal adenocarcinoma cells [200]. To date, it is unknown whether TWEAK-independent Fn14 signaling occurs in vivo. In gliomas, Fn14 mRNA is highly expressed, whereas TWEAK mRNA expression is low [196], which may favor a TWEAK-independent Fn14 signaling event. Together, these findings stipulate that activation of Fn14 results from either TWEAK binding the receptor or under circumstances of Fn14 overexpression independent of TWEAK. Dysregulation of the actin cytoskeleton contributes to the migratory phenotype of cancer cells, and promotes the invasion so descriptive of GBM. Rho GTPases are a family of molecular switches that regulate the actin cytoskeleton, cell polarity, and microtubule dynamics [17]. The TWEAK-Fn14 signaling axis mediates glioma migration and invasion via the Rac1 GTPase, and fosters a self-promoting feedback loop whereby Rac1 mediated TWEAK-Fn14 signaling induces Fn14 gene expression via the NFkB pathway [43]. TWEAK stimulated glioma cells were found to induce rapid phosphorylation of IKKb and IkBa coincident with total IkBa protein degradation. Use of either an Fn14-Fc decoy receptor or transduction of glioma cells with a dominant negative Rac1N17 expressing adenovirus abrogates these events. Furthermore, the Fn14 promoter region contains an NF-kB consensus sequence that has been shown to be occupied by the NF-kB transcription factor upon TWEAK stimulation. siRNA-mediated depletion of Rac1 expression strongly suppresses TWEAK-induced glioma invasion and Fn14 mRNA transcript levels [43], suggesting that the TWEAK-Fn14 signaling axis may be one of the key inputs to Rac hyperactivation in GBM and consequently the malignant invasive behavior of glial tumors.

158

6.5.2

B. Hu et al.

TROY

Similar to Fn14, TROY (TNFRSF19) has been shown to play an important role in GBM invasion. TNFRSF19/TROY gene encodes a type I cell surface receptor that is expressed on migrating or proliferating progenitor cells of the hippocampus, thalamus, and cerebral cortex [201–206]. TROY is an orphan member of the TNF receptor superfamily that lacks a cytoplasmic death domain but contains a single TRAF binding site [201]. TROY mRNA expression directly correlates with increasing glial tumor grade, and among malignant gliomas TROY expression correlates inversely with overall patient survival. During embryonic development, TROY is widely expressed, but its expression in the postnatal organism is restricted [201– 203, 205, 206]. The strict control of TROY expression indicates that aberrant expression may be unfavorable. The aberrant re-expression of TROY in GBM may play an important role in GBM progression, specifically cell invasion. Some of us also recently reported that TROY is able to promote glioma cell migration and invasion; TROY signals via Rac1 to direct glial cell motility, and TROY is found localized to the leading edge of migrating cells, notably enriched in lamellipodia, when overexpressed in glioma cells [207]. siRNA mediated depletion of Rac1 results in decreased migration of glioma cells overexpressing TROY. Importantly, the non-receptor tyrosine kinase Pyk2 was shown to be a cytoplasmic binding partner of TROY via mass spectrometry analysis of proteins that immunoprecipitated with the TROY receptor [207]. Pyk2 is highly expressed in the nervous system is uniquely localized within the cells to transduce signal from the extracellular matrix and soluble mediators through cell surface proteins including integrins, receptor tyrosine kinases, and G-protein coupled receptors into activated intracellular signaling pathways that regulates cell growth, survival and migration [208, 209]. Up-regulation of Pyk2 expression has been noted in several human tumors including hepatocellular carcinoma [210, 211], NSCLC [82] and breast cancer [212] and correlates with increased metastatic potential and reduced survival. Pyk2 expression is elevated in glioma and expression is increased with advancing WHO grade [213] and is seen up-regulated in the invasive tumor rim cells versus tumor core of glioblastoma samples [214, 215]. Knockdown of Pyk2 expression via shRNA or inhibition of Pyk2 activity by kinase-deficient Pyk2 suppresses TROY-induced glioma cell migration and Rac1 activation, thus implicating Pyk2 as a proximal signaling effector for TROY [207]. Taken together, these data suggest the important role of TROY-Pyk2-Rac1 signaling axis in glioma progression. To date, it is unclear how Fn14 and TROY stimulate Rac1 activation to drive glioma migration and invasion. It is likely that Fn14 and TROY modulates cytoskeleton dynamics by influencing one or more of Rac GEF. Since several Rac GEFs, including Ect2, Vav3, Trio, and Dock180, have been shown to be overexpressed in GBM specimens and modulate Rac1-dependent glioma cell migration and invasion [27, 185], it is possible that Fn14 and TROY signal through one or more of these GEFs to drive the hyperactivation of Rac1 to enhance glioma cell migration and invasion.

6 Rac1 and its GEFs in Glioma Cell Invasion

159

6.6 Concluding Statements: Targeting GEFs as an Anti-invasive Strategy for Glioblastoma Treatments The stimulatory role of Rac1 and its cognate GEFs in promoting glioma cell invasion has been firmly established. Targeted therapy to block functions of Rac1 and its GEFs in gliomas is predicted to abrogate the highly invasive nature of glioma cells. Much work is still needed, however, to fully understand the activity of this large family of GEF proteins as it pertains to glioma biology. One of the most challenging tasks is to decipher the relative contribution of each GEF towards the activation of one GTPase versus another since GEFs commonly have exchange factor activity for more than one GTPase. This is true for Trio, Ect2, Vav3 and perhaps other GEFs that will be identified in regulating glioblastoma cell invasion. Such a phenomenon will factor strongly into the issue of molecular redundancy – that is with a relatively large number of GEFs that perform similar functions understanding the context of vulnerability will be an important component for targeting strategies. Work in this realm is currently ongoing to determine the signaling pathways that mediate the activation of different Rho GTPases and their downstream effectors. This will ultimately require the establishment of assays, such as the G-LISA (Cytoskeleton, Inc) assay that should allow us to do this, designed to measure the activity of Rac and other GTPases in human specimens. Also of importance is the heterogeneous nature of malignant glioblastomas, i.e. the tumor-to-tumor variability of GEF expression in clinical gliomas, whereby some of the tumors express GEFs at low levels and others express the same GEFs at high levels. In fact, decreased expression of SWAP-70 in clinical glioma samples was found to correlate with favorable outcome of patients with glioblastomas [56]. In the future, it is hoped that several targets will become available for halting the migration of GBM cells which will represent strategic inhibitory points along well characterized signaling pathways. Discovery and optimization of existing small molecule inhibitors (such as NS237600, an inhibitor that interferes with the interaction of Rac1 with a number of different GEFs, including Trio and Tiam1) against target GEFs are also among the next important steps to this emerging field. These and future studies hold great promise in identifying novel therapeutic approaches against the intrinsically invasive phenotype of diffuse gliomas based on inhibiting GEF function.

References 1. Wen PY, Kesari S (2008) Malignant gliomas in adults. N Engl J Med 359:492–507 2. Furnari FB, Fenton T, Bachoo RM et al (2007) Malignant astrocytic glioma: genetics, biology, and paths to treatment. Genes Dev 21:2683–2710 3. Giese A, Bjerkvig R, Berens ME, Westphal M (2003) Cost of migration: invasion of malignant gliomas and implications for treatment. J Clin Oncol 21:1624–1636 4. Lefranc F, Brotchi J, Kiss R (2005) Possible future issues in the treatment of glioblastomas: special emphasis on cell migration and the resistance of migrating glioblastoma cells to apoptosis. J Clin Oncol 23:2411–2422

160

B. Hu et al.

5. Tcga C (2008) Comprehensive genomic characterization defines human glioblastoma genes and core pathways. Nature 455:1061–1068 6. Burridge K, Wennerberg K (2004) Rho and Rac take center stage. Cell 116:167–179 7. Schmitz AA, Govek EE, Bottner B, Van Aelst L (2000) Rho GTPases: signaling, migration, and invasion. Exp Cell Res 261:1–12 8. Rossman KL, Der CJ, Sondek J (2005) GEF means go: turning on RHO GTPases with guanine nucleotide-exchange factors. Nat Rev Mol Cell Biol 6:167–180 9. Eva A, Vecchio G, Rao CD, Tronick SR, Aaronson SA (1988) The predicted DBL oncogene product defines a distinct class of transforming proteins. Proc Natl Acad Sci U S A 85:2061–2065 10. Schmidt S, Diriong S, Mery J, Fabbrizio E, Debant A (2002) Identification of the first RhoGEF inhibitor, TRIPalpha, which targets the RhoA-specific GEF domain of Trio. FEBS Lett 523:35–42 11. Moon SY, Zheng Y (2003) Rho GTPase-activating proteins in cell regulation. Trends Cell Biol 13:13–22 12. DerMardirossian C, Bokoch GM (2005) GDIs: central regulatory molecules in Rho GTPase activation. Trends Cell Biol 15:356–363 13. Dovas A, Couchman JR (2005) RhoGDI: multiple functions in the regulation of Rho family GTPase activities. Biochem J 390:1–9 14. Schmidt A, Hall A (2002) Guanine nucleotide exchange factors for Rho GTPases: turning on the switch. Genes Dev 16:1587–1609 15. Bernards A, Settleman J (2005) GAPs in growth factor signalling. Growth Factors 23:143–149 16. Gomez del Pulgar T, Benitah SA, Valeron PF, Espina C, Lacal JC (2005) Rho GTPase expression in tumourigenesis: evidence for a significant link. Bioessays 27:602–613 17. Etienne-Manneville S, Hall A (2002) Rho GTPases in cell biology. Nature 420:629–635 18. Chan AY, Coniglio SJ, Chuang YY et al (2005) Roles of the Rac1 and Rac3 GTPases in human tumor cell invasion. Oncogene 24:7821–7829 19. Raftopoulou M, Hall A (2004) Cell migration: Rho GTPases lead the way. Dev Biol 265:23–32 20. Sahai E, Marshall CJ (2002) RHO-GTPases and cancer. Nat Rev Cancer 2:133–142 21. Hall A (2005) Rho GTPases and the control of cell behaviour. Biochem Soc Trans 33:891–895 22. Fernandez-Zapico ME, Gonzalez-Paz NC, Weiss E et al (2005) Ectopic expression of VAV1 reveals an unexpected role in pancreatic cancer tumorigenesis. Cancer Cell 7:39–49 23. Bartolome RA, Molina-Ortiz I, Samaniego R, Sanchez-Mateos P, Bustelo XR, Teixido J (2006) Activation of Vav/Rho GTPase signaling by CXCL12 controls membrane-type matrix metalloproteinase-dependent melanoma cell invasion. Cancer Res 66:248–258 24. Leeuwen FN, Kain HE, Kammen RA, Michiels F, Kranenburg OW, Collard JG (1997) The guanine nucleotide exchange factor Tiam1 affects neuronal morphology; opposing roles for the small GTPases Rac and Rho. J Cell Biol 139:797–807 25. Minard ME, Ellis LM, Gallick GE (2006) Tiam1 regulates cell adhesion, migration and apoptosis in colon tumor cells. Clin Exp Metastasis 23:301–313 26. Salhia B, Rutten F, Nakada M et al (2005) Inhibition of Rho-kinase affects astrocytoma morphology, motility, and invasion through activation of Rac1. Cancer Res 65:8792–8800 27. Salhia B, Tran NL, Chan A et al (2008) The guanine nucleotide exchange factors trio, Ect2, and Vav3 mediate the invasive behavior of glioblastoma. Am J Pathol 173:1828–1838 28. Hall A (2009) The cytoskeleton and cancer. Cancer Metastasis Rev 28:5–14 29. Sahai E, Marshall CJ (2003) Differing modes of tumour cell invasion have distinct requirements for Rho/ROCK signalling and extracellular proteolysis. Nat Cell Biol 5:711–719 30. Wolf K, Mazo I, Leung H et al (2003) Compensation mechanism in tumor cell migration: mesenchymal-amoeboid transition after blocking of pericellular proteolysis. J Cell Biol 160:267–277 31. Sanz-Moreno V, Gadea G, Ahn J et al (2008) Rac activation and inactivation control plasticity of tumor cell movement. Cell 135:510–523 32. Friedl P, Wolf K (2003) Tumour-cell invasion and migration: diversity and escape mechanisms. Nat Rev Cancer 3:362–374

6 Rac1 and its GEFs in Glioma Cell Invasion

161

33. Friedl P, Wolf K (2010) Plasticity of cell migration: a multiscale tuning model. J Cell Biol 188:11–19 34. Sanz-Moreno V, Marshall CJ (2010) The plasticity of cytoskeletal dynamics underlying neoplastic cell migration. Curr Opin Cell Biol 22:690–696 35. Karlsson R, Pedersen ED, Wang Z, Brakebusch C (2009) Rho GTPase function in tumorigenesis. Biochim Biophys Acta 1796:91–98 36. Harding MA, Theodorescu D (2010) RhoGDI signaling provides targets for cancer therapy. Eur J Cancer 46:1252–1259 37. Preudhomme C, Roumier C, Hildebrand MP et al (2000) Nonrandom 4p13 rearrangements of the RhoH/TTF gene, encoding a GTP-binding protein, in non-Hodgkin’s lymphoma and multiple myeloma. Oncogene 19:2023–2032 38. Rihet S, Vielh P, Camonis J, Goud B, Chevillard S, de Gunzburg J (2001) Mutation status of genes encoding RhoA, Rac1, and Cdc42 GTPases in a panel of invasive human colorectal and breast tumors. J Cancer Res Clin Oncol 127:733–738 39. Fritz G, Brachetti C, Bahlmann F, Schmidt M, Kaina B (2002) Rho GTPases in human breast tumours: expression and mutation analyses and correlation with clinical parameters. Br J Cancer 87:635–644 40. Ridley AJ (2004) Rho proteins and cancer. Breast Cancer Res Treat 84:13–19 41. Iwadate Y, Sakaida T, Hiwasa T et al (2004) Molecular classification and survival prediction in human gliomas based on proteome analysis. Cancer Res 64:2496–2501 42. Chuang YY, Tran NL, Rusk N, Nakada M, Berens ME, Symons M (2004) Role of synaptojanin 2 in glioma cell migration and invasion. Cancer Res 64:8271–8275 43. Tran NL, McDonough WS, Savitch BA et al (2006) Increased fibroblast growth factorinducible 14 expression levels promote glioma cell invasion via Rac1 and nuclear factor{kappa}B and correlate with poor patient outcome. Cancer Res 66:9535–9542 44. Nakada M, Drake KL, Nakada S, Niska JA, Berens ME (2006) Ephrin-B3 ligand promotes glioma invasion through activation of Rac1. Cancer Res 66:8492–8500 45. Gimona M, Buccione R, Courtneidge SA, Linder S (2008) Assembly and biological role of podosomes and invadopodia. Curr Opin Cell Biol 20:235–241 46. Beadle C, Assanah MC, Monzo P, Vallee R, Rosenfeld SS, Canoll P (2008) The role of myosin II in glioma invasion of the brain. Mol Biol Cell 19:3357–3368 47. Bos JL, Rehmann H, Wittinghofer A (2007) GEFs and GAPs: critical elements in the control of small G proteins. Cell 129:865–877 48. Cote JF, Vuori K (2002) Identification of an evolutionarily conserved superfamily of DOCK180related proteins with guanine nucleotide exchange activity. J Cell Sci 115:4901–4913 49. Meller N, Merlot S, Guda C (2005) CZH proteins: a new family of Rho-GEFs. J Cell Sci 118:4937–4946 50. Cote JF, Motoyama AB, Bush JA, Vuori K (2005) A novel and evolutionarily conserved PtdIns(3,4,5)P3-binding domain is necessary for DOCK180 signalling. Nat Cell Biol 7:797–807 51. Cote JF, Vuori K (2007) GEF what? Dock180 and related proteins help Rac to polarize cells in new ways. Trends Cell Biol 17:383–393 52. Brugnera E, Haney L, Grimsley C et al (2002) Unconventional Rac-GEF activity is mediated through the Dock180-ELMO complex. Nat Cell Biol 4:574–582 53. Meller N, Irani-Tehrani M, Ratnikov BI, Paschal BM, Schwartz MA (2004) The novel Cdc42 guanine nucleotide exchange factor, zizimin1, dimerizes via the Cdc42-binding CZH2 domain. J Biol Chem 279:37470–37476 54. Schiller MR (2006) Coupling receptor tyrosine kinases to Rho GTPases–GEFs what’s the link. Cell Signal 18:1834–1843 55. Sano M, Genkai N, Yajima N et al (2006) Expression level of ECT2 proto-oncogene correlates with prognosis in glioma patients. Oncol Rep 16:1093–1098 56. Seol HJ, Smith CA, Salhia B, Rutka JT (2009) The guanine nucleotide exchange factor SWAP-70 modulates the migration and invasiveness of human malignant glioma cells. Transl Oncol 2:300–309

162

B. Hu et al.

57. Debant A, Serra-Pages C, Seipel K et al (1996) The multidomain protein Trio binds the LAR transmembrane tyrosine phosphatase, contains a protein kinase domain, and has separate racspecific and rho-specific guanine nucleotide exchange factor domains. Proc Natl Acad Sci U S A 93:5466–5471 58. Bateman J, Van Vactor D (2001) The Trio family of guanine-nucleotide-exchange factors: regulators of axon guidance. J Cell Sci 114:1973–1980 59. Blangy A, Vignal E, Schmidt S, Debant A, Gauthier-Rouviere C, Fort P (2000) TrioGEF1 controls Rac- and Cdc42-dependent cell structures through the direct activation of rhoG. J Cell Sci 113(Pt 4):729–739 60. Katoh H, Negishi M (2003) RhoG activates Rac1 by direct interaction with the Dock180binding protein Elmo. Nature 424:461–464 61. Briancon-Marjollet A, Ghogha A, Nawabi H et al (2008) Trio mediates netrin-1-induced Rac1 activation in axon outgrowth and guidance. Mol Cell Biol 28:2314–2323 62. Charrasse S, Comunale F, Fortier M, Portales-Casamar E, Debant A, Gauthier-Rouviere C (2007) M-cadherin activates Rac1 GTPase through the Rho-GEF trio during myoblast fusion. Mol Biol Cell 18:1734–1743 63. Lin MZ, Greenberg ME (2000) Orchestral maneuvers in the axon: trio and the control of axon guidance. Cell 101:239–242 64. Portales-Casamar E, Briancon-Marjollet A, Fromont S, Triboulet R, Debant A (2006) Identification of novel neuronal isoforms of the Rho-GEF Trio. Biol Cell 98:183–193 65. Medley QG, Buchbinder EG, Tachibana K, Ngo H, Serra-Pages C, Streuli M (2003) Signaling between focal adhesion kinase and trio. J Biol Chem 278:13265–13270 66. Medley QG, Serra-Pages C, Iannotti E et al (2000) The trio guanine nucleotide exchange factor is a RhoA target. Binding of RhoA to the trio immunoglobulin-like domain. J Biol Chem 275:36116–36123 67. Ma XM, Huang JP, Eipper BA, Mains RE (2005) Expression of Trio, a member of the Dbl family of Rho GEFs in the developing rat brain. J Comp Neurol 482:333–348 68. McPherson CE, Eipper BA, Mains RE (2002) Genomic organization and differential expression of Kalirin isoforms. Gene 284:41–51 69. Lane J, Martin TA, Mansel RE, Jiang WG (2008) The expression and prognostic value of the guanine nucleotide exchange factors (GEFs) Trio, Vav1 and TIAM-1 in human breast cancer. Int Semin Surg Oncol 5:23 70. Yoshizuka N, Moriuchi R, Mori T et al (2004) An alternative transcript derived from the trio locus encodes a guanosine nucleotide exchange factor with mouse cell-transforming potential. J Biol Chem 279:43998–44004 71. Blangy A, Bouquier N, Gauthier-Rouviere C et al (2006) Identification of TRIO-GEFD1 chemical inhibitors using the yeast exchange assay. Biol Cell 98:511–522 72. Bouquier N, Fromont S, Zeeh JC et al (2009) Aptamer-derived peptides as potent inhibitors of the oncogenic RhoGEF Tgat. Chem Biol 16:391–400 73. Bouquier N, Vignal E, Charrasse S et al (2009) A cell active chemical GEF inhibitor selectively targets the Trio/RhoG/Rac1 signaling pathway. Chem Biol 16:657–666 74. Gao Y, Dickerson JB, Guo F, Zheng J, Zheng Y (2004) Rational design and characterization of a Rac GTPase-specific small molecule inhibitor. Proc Natl Acad Sci U S A 101:7618–7623 75. Shinohara M, Terada Y, Iwamatsu A et al (2002) SWAP-70 is a guanine-nucleotide-exchange factor that mediates signalling of membrane ruffling. Nature 416:759–763 76. Hilpela P, Oberbanscheidt P, Hahne P et al (2003) SWAP-70 identifies a transitional subset of actin filaments in motile cells. Mol Biol Cell 14:3242–3253 77. Murugan AK, Ihara S, Tokuda E, Uematsu K, Tsuchida N, Fukui Y (2008) SWAP-70 is important for invasive phenotypes of mouse embryo fibroblasts transformed by v-Src. IUBMB Life 60:236–240 78. Miki T (2000) Malignant transformation and regulation of cell division by a Rho exchange factor ECT2. Seikagaku 72:1249–1253

6 Rac1 and its GEFs in Glioma Cell Invasion

163

79. Miki T, Smith CL, Long JE, Eva A, Fleming TP (1993) Oncogene ect2 is related to regulators of small GTP-binding proteins. Nature 362:462–465 80. Saito S, Liu XF, Kamijo K et al (2004) Deregulation and mislocalization of the cytokinesis regulator ECT2 activate the Rho signaling pathways leading to malignant transformation. J Biol Chem 279:7169–7179 81. Hirata D, Yamabuki T, Miki D et al (2009) Involvement of epithelial cell transforming sequence-2 oncoantigen in lung and esophageal cancer progression. Clin Cancer Res 15:256–266 82. Zhang ML, Lu S, Zhou L, Zheng SS (2008) Correlation between ECT2 gene expression and methylation change of ECT2 promoter region in pancreatic cancer. Hepatobiliary Pancreat Dis Int 7:533–538 83. Roversi G, Pfundt R, Moroni RF et al (2006) Identification of novel genomic markers related to progression to glioblastoma through genomic profiling of 25 primary glioma cell lines. Oncogene 25:1571–1583 84. Tatsumoto T, Xie X, Blumenthal R, Okamoto I, Miki T (1999) Human ECT2 is an exchange factor for Rho GTPases, phosphorylated in G2/M phases, and involved in cytokinesis. J Cell Biol 147:921–928 85. Saito S, Tatsumoto T, Lorenzi MV et al (2003) Rho exchange factor ECT2 is induced by growth factors and regulates cytokinesis through the N-terminal cell cycle regulator-related domains. J Cell Biochem 90:819–836 86. Hornstein I, Alcover A, Katzav S (2004) Vav proteins, masters of the world of cytoskeleton organization. Cell Signal 16:1–11 87. Bustelo XR (2000) Regulatory and signaling properties of the Vav family. Mol Cell Biol 20:1461–1477 88. Dong Z, Liu Y, Lu S et al (2006) Vav3 oncogene is overexpressed and regulates cell growth and androgen receptor activity in human prostate cancer. Mol Endocrinol 20:2315–2325 89. Lee K, Liu Y, Mo JQ, Zhang J, Dong Z, Lu S (2008) Vav3 oncogene activates estrogen receptor and its overexpression may be involved in human breast cancer. BMC Cancer 8:158 90. Colomba A, Courilleau D, Ramel D et al (2008) Activation of Rac1 and the exchange factor Vav3 are involved in NPM-ALK signaling in anaplastic large cell lymphomas. Oncogene 27:2728–2736 91. Hunter SG, Zhuang G, Brantley-Sieders D, Swat W, Cowan CW, Chen J (2006) Essential role of Vav family guanine nucleotide exchange factors in EphA receptor-mediated angiogenesis. Mol Cell Biol 26:4830–4842 92. Miyamoto Y, Yamauchi J (2010) Cellular signaling of Dock family proteins in neural function. Cell Signal 22:175–182 93. Nishihara H, Kobayashi S, Hashimoto Y et al (1999) Non-adherent cell-specific expression of DOCK2, a member of the human CDM-family proteins. Biochim Biophys Acta 1452: 179–187 94. Fukui Y, Hashimoto O, Sanui T et al (2001) Haematopoietic cell-specific CDM family protein DOCK2 is essential for lymphocyte migration. Nature 412:826–831 95. Nombela-Arrieta C, Mempel TR, Soriano SF et al (2007) A central role for DOCK2 during interstitial lymphocyte motility and sphingosine-1-phosphate-mediated egress. J Exp Med 204:497–510 96. Reif K, Cyster J (2002) The CDM protein DOCK2 in lymphocyte migration. Trends Cell Biol 12:368–373 97. Nishihara H, Maeda M, Oda A et al (2002) DOCK2 associates with CrkL and regulates Rac1 in human leukemia cell lines. Blood 100:3968–3974 98. Sanui T, Inayoshi A, Noda M et al (2003) DOCK2 regulates Rac activation and cytoskeletal reorganization through interaction with ELMO1. Blood 102:2948–2950 99. Janardhan A, Swigut T, Hill B, Myers MP, Skowronski J (2004) HIV-1 Nef binds the DOCK2ELMO1 complex to activate rac and inhibit lymphocyte chemotaxis. PLoS Biol 2:E6

164

B. Hu et al.

100. Shulman Z, Pasvolsky R, Woolf E et al (2006) DOCK2 regulates chemokine-triggered lateral lymphocyte motility but not transendothelial migration. Blood 108:2150–2158 101. Garcia-Bernal D, Sotillo-Mallo E, Nombela-Arrieta C et al (2006) DOCK2 is required for chemokine-promoted human T lymphocyte adhesion under shear stress mediated by the integrin alpha4beta1. J Immunol 177:5215–5225 102. Kunisaki Y, Nishikimi A, Tanaka Y et al (2006) DOCK2 is a Rac activator that regulates motility and polarity during neutrophil chemotaxis. J Cell Biol 174:647–652 103. Sai J, Raman D, Liu Y, Wikswo J, Richmond A (2008) Parallel phosphatidylinositol 3-kinase (PI3K)-dependent and Src-dependent pathways lead to CXCL8-mediated Rac2 activation and chemotaxis. J Biol Chem 283:26538–26547 104. Nishikimi A, Fukuhara H, Su W et al (2009) Sequential regulation of DOCK2 dynamics by two phospholipids during neutrophil chemotaxis. Science 324:384–387 105. Wang L, Nishihara H, Kimura T et al (2010) DOCK2 regulates cell proliferation through Rac and ERK activation in B cell lymphoma. Biochem Biophys Res Commun 395:111–115 106. El Haibi CP, Sharma PK, Singh R et al (2010) PI3Kp110-, Src-, FAK-dependent and DOCK2-independent migration and invasion of CXCL13-stimulated prostate cancer cells. Mol Cancer 9:85 107. Gotoh K, Tanaka Y, Nishikimi A et al (2010) Selective control of type I IFN induction by the Rac activator DOCK2 during TLR-mediated plasmacytoid dendritic cell activation. J Exp Med 207:721–730 108. Yajnik V, Paulding C, Sordella R et al (2003) DOCK4, a GTPase activator, is disrupted during tumorigenesis. Cell 112:673–684 109. Pagnamenta AT, Bacchelli E, de Jonge MV et al (2010) Characterization of a family with rare deletions in CNTNAP5 and DOCK4 suggests novel risk loci for autism and dyslexia. Biol Psychiatry 68:320–328 110. Poelmans G, Buitelaar JK, Pauls DL, Franke B (2011) A theoretical molecular network for dyslexia: integrating available genetic findings. Mol Psychiatry 16(4):365–382 111. Gadd S, Sredni ST, Huang CC, Perlman EJ (2010) Rhabdoid tumor: gene expression clues to pathogenesis and potential therapeutic targets. Lab Invest 90:724–738 112. Maestrini E, Pagnamenta AT, Lamb JA et al (2010) High-density SNP association study and copy number variation analysis of the AUTS1 and AUTS5 loci implicate the IMMP2LDOCK4 gene region in autism susceptibility. Mol Psychiatry 15:954–968 113. Kuo KT, Guan B, Feng Y et al (2009) Analysis of DNA copy number alterations in ovarian serous tumors identifies new molecular genetic changes in low-grade and high-grade carcinomas. Cancer Res 69:4036–4042 114. Liang H, Castro PD, Ma J, Nagarajan L (2005) Finer delineation and transcript map of the 7q31 locus deleted in myeloid neoplasms. Cancer Genet Cytogenet 162:151–159 115. Hiramoto K, Negishi M, Katoh H (2006) Dock4 is regulated by RhoG and promotes Rac-dependent cell migration. Exp Cell Res 312:4205–4216 116. Ueda S, Fujimoto S, Hiramoto K, Negishi M, Katoh H (2008) Dock4 regulates dendritic development in hippocampal neurons. J Neurosci Res 86:3052–3061 117. Pannekoek WJ, Kooistra MR, Zwartkruis FJ, Bos JL (2009) Cell-cell junction formation: the role of Rap1 and Rap1 guanine nucleotide exchange factors. Biochim Biophys Acta 1788:790–796 118. Hiramoto-Yamaki N, Takeuchi S, Ueda S et al (2010) Ephexin4 and EphA2 mediate cell migration through a RhoG-dependent mechanism. J Cell Biol 190:461–477 119. Upadhyay G, Goessling W, North TE, Xavier R, Zon LI, Yajnik V (2008) Molecular association between beta-catenin degradation complex and Rac guanine exchange factor DOCK4 is essential for Wnt/beta-catenin signaling. Oncogene 27:5845–5855 120. Watabe-Uchida M, John KA, Janas JA, Newey SE, Van Aelst L (2006) The Rac activator DOCK7 regulates neuronal polarity through local phosphorylation of stathmin/Op18. Neuron 51:727–739 121. Pinheiro EM, Gertler FB (2006) Nervous Rac: DOCK7 regulation of axon formation. Neuron 51:674–676

6 Rac1 and its GEFs in Glioma Cell Invasion

165

122. Yamauchi J, Miyamoto Y, Chan JR, Tanoue A (2008) ErbB2 directly activates the exchange factor Dock7 to promote Schwann cell migration. J Cell Biol 181:351–365 123. Guo L, Ying W, Zhang J et al (2010) Tandem affinity purification and identification of the human TSC1 protein complex. Acta Biochim Biophys Sin (Shanghai) 42:266–273 124. Rosner M, Hanneder M, Siegel N, Valli A, Hengstschlager M (2008) The tuberous sclerosis gene products hamartin and tuberin are multifunctional proteins with a wide spectrum of interacting partners. Mutat Res 658:234–246 125. Nellist M, Burgers PC, van den Ouweland AM, Halley DJ, Luider TM (2005) Phosphorylation and binding partner analysis of the TSC1-TSC2 complex. Biochem Biophys Res Commun 333:818–826 126. Meller N, Irani-Tehrani M, Kiosses WB, Del Pozo MA, Schwartz MA (2002) Zizimin1, a novel Cdc42 activator, reveals a new GEF domain for Rho proteins. Nat Cell Biol 4:639–647 127. Meller N, Westbrook MJ, Shannon JD, Guda C, Schwartz MA (2008) Function of the N-terminus of zizimin1: autoinhibition and membrane targeting. Biochem J 409:525–533 128. Kuramoto K, Negishi M, Katoh H (2009) Regulation of dendrite growth by the Cdc42 activator Zizimin1/Dock9 in hippocampal neurons. J Neurosci Res 87:1794–1805 129. Kwofie MA, Skowronski J (2008) Specific recognition of Rac2 and Cdc42 by DOCK2 and DOCK9 guanine nucleotide exchange factors. J Biol Chem 283:3088–3096 130. Yang J, Zhang Z, Roe SM, Marshall CJ, Barford D (2009) Activation of Rho GTPases by DOCK exchange factors is mediated by a nucleotide sensor. Science 325:1398–1402 131. Rittinger K (2009) Snapshots form a big picture of guanine nucleotide exchange. Sci Signal 2:63 132. Hasegawa H, Kiyokawa E, Tanaka S et al (1996) DOCK180, a major CRK-binding protein, alters cell morphology upon translocation to the cell membrane. Mol Cell Biol 16:1770–1776 133. Matsuda M, Ota S, Tanimura R et al (1996) Interaction between the amino-terminal SH3 domain of CRK and its natural target proteins. J Biol Chem 271:14468–14472 134. Kiyokawa E, Hashimoto Y, Kobayashi S, Sugimura H, Kurata T, Matsuda M (1998) Activation of Rac1 by a Crk SH3-binding protein, DOCK180. Genes Dev 12:3331–3336 135. Kiyokawa E, Hashimoto Y, Kurata T, Sugimura H, Matsuda M (1998) Evidence that DOCK180 up-regulates signals from the CrkII-p130(Cas) complex. J Biol Chem 273: 24479–24484 136. Cheresh DA, Leng J, Klemke RL (1999) Regulation of cell contraction and membrane ruffling by distinct signals in migratory cells. J Cell Biol 146:1107–1116 137. Erickson MR, Galletta BJ, Abmayr SM (1997) Drosophila myoblast city encodes a conserved protein that is essential for myoblast fusion, dorsal closure, and cytoskeletal organization. J Cell Biol 138:589–603 138. Nolan KM, Barrett K, Lu Y, Hu KQ, Vincent S, Settleman J (1998) Myoblast city, the Drosophila homolog of DOCK180/CED-5, is required in a Rac signaling pathway utilized for multiple developmental processes. Genes Dev 12:3337–3342 139. Wu YC, Horvitz HR (1998) C. elegans phagocytosis and cell-migration protein CED-5 is similar to human DOCK180. Nature 392:501–504 140. Reddien PW, Horvitz HR (2000) CED-2/CrkII and CED-10/Rac control phagocytosis and cell migration in Caenorhabditis elegans. Nat Cell Biol 2:131–136 141. Tosello-Trampont A-C, Brugnera E, Ravichandran KS (2001) Evidence for a Conserved Role for CrkII and Rac in Engulfment of Apoptotic Cells. J Biol Chem 276:13797–13802 142. Zhou Z, Caron E, Hartwieg E, Hall A, Horvitz HR (2001) The C. elegans PH domain protein CED-12 regulates cytoskeletal reorganization via a Rho/Rac GTPase signaling pathway. Dev Cell 1:477–489 143. Wang X, Wu YC, Fadok VA et al (2003) Cell corpse engulfment mediated by C. elegans phosphatidylserine receptor through CED-5 and CED-12. Science 302:1563–1566 144. Valles AM, Beuvin M, Boyer B (2004) Activation of Rac1 by paxillin-Crk-DOCK180 signaling complex is antagonized by Rap1 in migrating NBT-II cells. J Biol Chem 279: 44490–44496 145. Lu M, Kinchen JM, Rossman KL et al (2005) A Steric-inhibition model for regulation of nucleotide exchange via the Dock180 family of GEFs. Curr Biol 15:371–377

166

B. Hu et al.

146. Moore CA, Parkin CA, Bidet Y, Ingham PW (2007) A role for the Myoblast city homologues Dock1 and Dock5 and the adaptor proteins Crk and Crk-like in zebrafish myoblast fusion. Development 134:3145–3153 147. Pajcini KV, Pomerantz JH, Alkan O, Doyonnas R, Blau HM (2008) Myoblasts and macrophages share molecular components that contribute to cell-cell fusion. J Cell Biol 180:1005–1019 148. Kobayashi S, Shirai T, Kiyokawa E, Mochizuki N, Matsuda M, Fukui Y (2001) Membrane recruitment of DOCK180 by binding to PtdIns(3,4,5)P3. Biochem J 354:73–78 149. Cote JF, Vuori K (2006) In vitro guanine nucleotide exchange activity of DHR-2/DOCKER/ CZH2 domains. Methods Enzymol 406:41–57 150. Gumienny TL, Brugnera E, Tosello-Trampont AC et al (2001) CED-12/ELMO, a novel member of the CrkII/Dock180/Rac pathway, is required for phagocytosis and cell migration. Cell 107:27–41 151. Wu YC, Tsai MC, Cheng LC, Chou CJ, Weng NY (2001) C. elegans CED-12 acts in the conserved crkII/DOCK180/Rac pathway to control cell migration and cell corpse engulfment. Dev Cell 1:491–502 152. Grimsley CM, Kinchen JM, Tosello-Trampont AC et al (2004) Dock180 and ELMO1 proteins cooperate to promote evolutionarily conserved Rac-dependent cell migration. J Biol Chem 279:6087–6097 153. Geisbrecht ER, Haralalka S, Swanson SK, Florens L, Washburn MP, Abmayr SM (2008) Drosophila ELMO/CED-12 interacts with Myoblast city to direct myoblast fusion and ommatidial organization. Dev Biol 314:137–149 154. Akakura S, Kar B, Singh S et al (2005) C-terminal SH3 domain of CrkII regulates the assembly and function of the DOCK180/ELMO Rac-GEF. J Cell Physiol 204:344–351 155. Zaidel-Bar R, Kam Z, Geiger B (2005) Polarized downregulation of the paxillin-p130CASRac1 pathway induced by shear flow. J Cell Sci 118:3997–4007 156. Tosello-Trampont AC, Kinchen JM, Brugnera E, Haney LB, Hengartner MO, Ravichandran KS (2007) Identification of two signaling submodules within the CrkII/ELMO/Dock180 pathway regulating engulfment of apoptotic cells. Cell Death Differ 14:963–972 157. Handa Y, Suzuki M, Ohya K et al (2007) Shigella IpgB1 promotes bacterial entry through the ELMO-Dock180 machinery. Nat Cell Biol 9:121–128 158. Meller J, Vidali L, Schwartz MA (2008) Endogenous RhoG is dispensable for integrin-mediated cell spreading but contributes to Rac-independent migration. J Cell Sci 121:1981–1989 159. Para A, Krischke M, Merlot S et al (2009) Dictyostelium Dock180-related RacGEFs regulate the actin cytoskeleton during cell motility. Mol Biol Cell 20:699–707 160. Makino Y, Tsuda M, Ichihara S et al (2006) Elmo1 inhibits ubiquitylation of Dock180. J Cell Sci 119:923–932 161. Cho NK, Keyes L, Johnson E et al (2002) Developmental control of blood cell migration by the Drosophila VEGF pathway. Cell 108:865–876 162. Kraut R, Menon K, Zinn K (2001) A gain-of-function screen for genes controlling motor axon guidance and synaptogenesis in Drosophila. Curr Biol 11:417–430 163. Duchek P, Somogyi K, Jekely G, Beccari S, Rorth P (2001) Guidance of cell migration by the Drosophila PDGF/VEGF receptor. Cell 107:17–26 164. Fulga TA, Rorth P (2002) Invasive cell migration is initiated by guided growth of long cellular extensions. Nat Cell Biol 4:715–719 165. Ishimaru S, Ueda R, Hinohara Y, Ohtani M, Hanafusa H (2004) PVR plays a critical role via JNK activation in thorax closure during Drosophila metamorphosis. EMBO J 23:3984–3994 166. Bianco A, Poukkula M, Cliffe A et al (2007) Two distinct modes of guidance signalling during collective migration of border cells. Nature 448:362–365 167. Albert ML, Kim JI, Birge RB (2000) alphavbeta5 integrin recruits the CrkII-Dock180-rac1 complex for phagocytosis of apoptotic cells. Nat Cell Biol 2:899–905 168. Gu J, Sumida Y, Sanzen N, Sekiguchi K (2001) Laminin-10/11 and fibronectin differentially regulate integrin-dependent Rho and Rac activation via p130(Cas)-CrkII-DOCK180 pathway. J Biol Chem 276:27090–27097

6 Rac1 and its GEFs in Glioma Cell Invasion

167

169. Gustavsson A, Yuan M, Fallman M (2004) Temporal dissection of beta1-integrin signaling indicates a role for p130Cas-Crk in filopodia formation. J Biol Chem 279:22893–22901 170. Smith HW, Marra P, Marshall CJ (2008) uPAR promotes formation of the p130Cas-Crk complex to activate Rac through DOCK180. J Cell Biol 182:777–790 171. Lee WL, Cosio G, Ireton K, Grinstein S (2007) Role of CrkII in Fc{gamma} Receptormediated Phagocytosis. J Biol Chem 282:11135–11143 172. Payne SL, Hendrix MJ, Kirschmann DA (2006) Lysyl oxidase regulates actin filament formation through the p130(Cas)/Crk/DOCK180 signaling complex. J Cell Biochem 98: 827–837 173. Laurin M, Fradet N, Blangy A, Hall A, Vuori K, Cote JF (2008) The atypical Rac activator Dock180 (Dock1) regulates myoblast fusion in vivo. Proc Natl Acad Sci U S A 105(40): 15446–15451 174. Epting D, Wendik B, Bennewitz K, Dietz CT, Driever W, Kroll J (2010) The Rac1 regulator ELMO1 controls vascular morphogenesis in zebrafish. Circ Res 107:45–55 175. Park D, Tosello-Trampont AC, Elliott MR et al (2007) BAI1 is an engulfment receptor for apoptotic cells upstream of the ELMO/Dock180/Rac module. Nature 450:430–434 176. Elliott MR, Zheng S, Park D et al (2010) Unexpected requirement for ELMO1 in clearance of apoptotic germ cells in vivo. Nature 467:333–337 177. Elliott MR, Ravichandran KS (2010) ELMO1 signaling in apoptotic germ cell clearance and spermatogenesis. Ann N Y Acad Sci 1209:30–36 178. Shimazaki A, Kawamura Y, Kanazawa A et al (2005) Genetic variations in the gene encoding ELMO1 are associated with susceptibility to diabetic nephropathy. Diabetes 54:1171–1178 179. Pezzolesi MG, Katavetin P, Kure M et al (2009) Confirmation of genetic associations at ELMO1 in the GoKinD collection supports its role as a susceptibility gene in diabetic nephropathy. Diabetes 58:2698–2702 180. Leak TS, Perlegas PS, Smith SG et al (2009) Variants in intron 13 of the ELMO1 gene are associated with diabetic nephropathy in African Americans. Ann Hum Genet 73:152–159 181. Doria A (2010) Genetics of diabetes complications. Curr Diab Rep 10:467–475 182. Phillips HS, Kharbanda S, Chen R et al (2006) Molecular subclasses of high-grade glioma predict prognosis, delineate a pattern of disease progression, and resemble stages in neurogenesis. Cancer Cell 9:157–173 183. Hsia DA, Mitra SK, Hauck CR et al (2003) Differential regulation of cell motility and invasion by FAK. J Cell Biol 160:753–767 184. Wang H, Linghu H, Wang J et al (2010) The role of Crk/Dock180/Rac1 pathway in the malignant behavior of human ovarian cancer cell SKOV3. Tumour Biol 31:59–67 185. Jarzynka MJ, Hu B, Hui KM et al (2007) ELMO1 and Dock180, a bipartite Rac1 guanine nucleotide exchange factor, promote human glioma cell invasion. Cancer Res 67:7203–7211 186. Foster D, Parrish-Novak J, Fox B, Xu W (2004) Cytokine-receptor pairing: accelerating discovery of cytokine function. Nat Rev Drug Discov 3:160–170 187. Locksley RM, Killeen N, Lenardo MJ (2001) The TNF and TNF receptor superfamilies: integrating mammalian biology. Cell 104:487–501 188. Baker SJ, Reddy EP (1998) Modulation of life and death by the TNF receptor superfamily. Oncogene 17:3261–3270 189. Fesik SW (2000) Insights into programmed cell death through structural biology. Cell 103:273–282 190. Brown SA, Richards CM, Hanscom HN, Feng SL, Winkles JA (2003) The Fn14 cytoplasmic tail binds tumour-necrosis-factor-receptor-associated factors 1, 2, 3 and 5 and mediates nuclear factor-kappaB activation. Biochem J 371:395–403 191. Willis AL, Tran NL, Chatigny JM et al (2008) The fibroblast growth factor-inducible 14 receptor is highly expressed in HER2-positive breast tumors and regulates breast cancer cell invasive capacity. Mol Cancer Res 6:725–734 192. Meighan-Mantha RL, Hsu DK, Guo Y et al (1999) The mitogen-inducible Fn14 gene encodes a type I transmembrane protein that modulates fibroblast adhesion and migration. J Biol Chem 274:33166–33176

168

B. Hu et al.

193. Wiley SR, Winkles JA (2003) TWEAK, a member of the TNF superfamily, is a multifunctional cytokine that binds the TweakR/Fn14 receptor. Cytokine Growth Factor Rev 14:241–249 194. Feng SL, Guo Y, Factor VM et al (2000) The Fn14 immediate-early response gene is induced during liver regeneration and highly expressed in both human and murine hepatocellular carcinomas. Am J Pathol 156:1253–1261 195. Wiley SR, Cassiano L, Lofton T et al (2001) A novel TNF receptor family member binds TWEAK and is implicated in angiogenesis. Immunity 15:837–846 196. Tran NL, McDonough WS, Donohue PJ et al (2003) The human Fn14 receptor gene is upregulated in migrating glioma cells in vitro and overexpressed in advanced glial tumors. Am J Pathol 162:1313–1321 197. Tran NL, McDonough WS, Savitch BA, Sawyer TF, Winkles JA, Berens ME (2005) The tumor necrosis factor-like weak inducer of apoptosis (TWEAK)-fibroblast growth factorinducible 14 (Fn14) signaling system regulates glioma cell survival via NFkappaB pathway activation and BCL-XL/BCL-W expression. J Biol Chem 280:3483–3492 198. Han S, Yoon K, Lee K et al (2003) TNF-related weak inducer of apoptosis receptor, a TNF receptor superfamily member, activates NF-kappa B through TNF receptor-associated factors. Biochem Biophys Res Commun 305:789–796 199. Tanabe K, Bonilla I, Winkles JA, Strittmatter SM (2003) Fibroblast growth factor-inducible-14 is induced in axotomized neurons and promotes neurite outgrowth. J Neurosci 23: 9675–9686 200. Watts GS, Tran NL, Berens ME et al (2007) Identification of Fn14/TWEAK receptor as a potential therapeutic target in esophageal adenocarcinoma. Int J Cancer 121:2132–2139 201. Hu S, Tamada K, Ni J, Vincenz C, Chen L (1999) Characterization of TNFRSF19, a novel member of the tumor necrosis factor receptor superfamily. Genomics 62:103–107 202. Park JB, Yiu G, Kaneko S et al (2005) A TNF receptor family member, TROY, is a coreceptor with Nogo receptor in mediating the inhibitory activity of myelin inhibitors. Neuron 45:345–351 203. Pispa J, Mikkola ML, Mustonen T, Thesleff I (2003) Ectodysplasin, Edar and TNFRSF19 are expressed in complementary and overlapping patterns during mouse embryogenesis. Gene Expr Patterns 3:675–679 204. Hisaoka T, Morikawa Y, Kitamura T, Senba E (2004) Expression of a member of tumor necrosis factor receptor superfamily, TROY, in the developing olfactory system. Glia 45:313–324 205. Shao Z, Browning JL, Lee X et al (2005) TAJ/TROY, an orphan TNF receptor family member, binds Nogo-66 receptor 1 and regulates axonal regeneration. Neuron 45:353–359 206. Hisaoka T, Morikawa Y, Kitamura T, Senba E (2003) Expression of a member of tumor necrosis factor receptor superfamily, TROY, in the developing mouse brain. Brain Res Dev Brain Res 143:105–109 207. Paulino VM, Yang Z, Kloss JM et al (2010) TROY (TNFRSF19) is overexpressed in advanced glial tumors and promotes glioblastoma cell invasion via Pyk2-Rac1 signaling. Mol Cancer Res 8(11):1558–1567 208. Girault JA, Costa A, Derkinderen P, Studler JM, Toutant M (1999) FAK and PYK2/CAKbeta in the nervous system: a link between neuronal activity, plasticity and survival? Trends Neurosci 22:257–263 209. Avraham H, Park SY, Schinkmann K, Avraham S (2000) RAFTK/Pyk2-mediated cellular signalling. Cell Signal 12:123–133 210. Sun CK, Ng KT, Sun BS et al (2007) The significance of proline-rich tyrosine kinase2 (Pyk2) on hepatocellular carcinoma progression and recurrence. Br J Cancer 97:50–57 211. Sun CK, Man K, Ng KT et al (2008) Proline-rich tyrosine kinase 2 (Pyk2) promotes proliferation and invasiveness of hepatocellular carcinoma cells through c-Src/ERK activation. Carcinogenesis 29:2096–2105 212. Behmoaram E, Bijian K, Jie S et al (2008) Focal adhesion kinase-related proline-rich tyrosine kinase 2 and focal adhesion kinase are co-overexpressed in early-stage and invasive

6 Rac1 and its GEFs in Glioma Cell Invasion

169

ErbB-2-positive breast cancer and cooperate for breast cancer cell tumorigenesis and invasiveness. Am J Pathol 173:1540–1550 213. Gutenberg A, Bruck W, Buchfelder M, Ludwig HC (2004) Expression of tyrosine kinases FAK and Pyk2 in 331 human astrocytomas. Acta Neuropathol 108:224–230 214. Mariani L, Beaudry C, McDonough WS et al (2001) Glioma cell motility is associated with reduced transcription of proapoptotic and proliferation genes: a cDNA microarray analysis. J Neurooncol 53:161–176 215. Hoelzinger DB, Mariani L, Weis J et al (2005) Gene expression profile of glioblastoma multiforme invasive phenotype points to new therapeutic targets. Neoplasia 7:7–16

Chapter 7

Impact of Tumor Hypoxia, Src, and Met Signaling in the Dissemination of Tumor Cells Yao Dai, Wenyin Shi, Nikolett Molnar, and Dietmar Siemann

Abstract Metastasis is the major cause of therapeutic failure and high mortality in cancer patients. The metastatic cascade consists of a series of cellular processes including migration, invasion and survival. Hypoxia has been identified as a key contributor to critical functions associated with tumor cell dissemination. Hypoxiainduced alterations are primarily mediated by the transcription factor hypoxiainducible factor 1 (HIF-1), a factor that induces the expression of an array of genes associated with cell aggressiveness. In addition, altered intracellular signaling pathways, such as c-Src and c-Met, aid tumor cells to progress towards a metastatic phenotype and strongly correlates with poor prognosis. Given the pivotal role of hypoxia/HIF-1, c-Src and c-Met in the metastatic process, the potential for developing therapeutic strategies targeting these pathways is significant. Although the effects of HIF-1 inhibitors require further investigation, small molecule kinase inhibitors targeting Src or Met already have been shown to have considerable efficacy in metastatic diseases. Overall, the fact that cancer cell dissemination can be driven by both external (hypoxia) and internal features (c-Src, c-Met) suggests that the combined targeting of tumor microenvironment and intracellular signaling pathways may have particular utility in impeding cancer metastasis.

Y. Dai • D. Siemann (*) Department of Radiation Oncology, Shands Cancer Center, University of Florida, Gainesville, FL 32610, USA e-mail: [email protected] W. Shi Department of Radiation Oncology, Thomas Jefferson University, Philadelphia, PA 19107, USA N. Molnar Department of Radiation Oncology, Shands Cancer Center, University of Florida, Gainesville, FL 32610, USA Department of Pharmacology and Therapeutics, University of Florida, Gainesville, FL 32610, USA

A. Fatatis (ed.), Signaling Pathways and Molecular Mediators in Metastasis, DOI 10.1007/978-94-007-2558-4_7, © Springer Science+Business Media B.V. 2012

171

172

7.1

Y. Dai et al.

Introduction

Despite improvement in early detection, more refined diagnostic modalities, and a better understanding of the natural history of the disease, metastatic cancers remain largely incurable. The presence of metastases results in a substantial reduction of the quality of life and the drastic worsening of the prognosis. For example, though highly curable if locally confined, nearly 90% of prostate cancer patients with advanced diseases develop skeletal metastases, a result which lowers the 5-year survival rate to ~31% [1]. In colorectal cancer, the 5-year survival rate is less than 10% in patients with liver metastases [2], compared to nearly 90% in patients with early stages of the disease. The tumor microenvironment has long been considered a critical player facilitating cancer progression and dissemination [3]. Hypoxia, a consequence of inadequate vascular supply, is a common feature of most solid tumors detected in 50–60% of all human tumors [4]. From a therapeutic perspective, the existence of low oxygen tensions in tumors has been associated with increased resistance to cytotoxic therapies and elevated metastatic capacity [5, 6]. Acquisition of the metastatic phenotype is also the result of abnormal intracellular signaling pathways that enhance multiple functions associated with cell motility and invasiveness. Key among complex signaling networks, c-Src and HGF/c-Met trigger a variety of signaling cascades and facilitate the ability of tumor cells to metastasize. Indeed, their dysregulated expression has been strongly associated with cancer progression and metastasis in a variety of human cancers.

7.2

Hypoxia Is an Extrinsic Driving Force That Promotes Tumor Cell Dissemination

As tumors grow the associated blood vessel network often is inadequate to provide nutritional support to the rapidly expanding neoplastic cell population. As a consequence, regions within solid tumors experience oxygen deprivation and become hypoxic. Tumor cells acquire the ability to adapt to hypoxic environments by lowering oxygen consumption, switching to anaerobic glycolysis, co-opting blood vessel formation, and migrating toward blood vessels [7]. Using a variety of measurement techniques including microelectrodes, hypoxic cell markers, and hypoxia-associated molecules, hypoxia has been demonstrated to be a common feature in most solid human cancers and associated with poor prognoses [8, 9]. Importantly, cancer cells existing in low oxygen microenvironments not only contribute to therapeutic resistance [5] but also express a more metastatic phenotype [6].

7

Impact of Tumor Hypoxia, Src, and Met Signaling in the Dissemination…

173

Fig. 7.1 HIF-1 activation of selected genes involved in metastasis

7.2.1

Impact of Hypoxia on Metastasis

Increasingly laboratory [11] and clinical [10] evidence supports a role for hypoxia in the development of metastatic disease. In both pre-clinical models and patients, hypoxic tumors are generally characterized to have a higher tendency to metastasize Hypoxia regulates a myriad of genes predominantly via hypoxia-inducible factor 1 (HIF-1), a transcriptional factor that mediates hypoxia-induced alterations [9, 12]. These HIF-1-induced genes are widely involved in metastasis-associated phenotypes, such as epithelial-to-mesenchymal transition (EMT), migration, invasion, survival and angiogenesis (Fig. 7.1) [6, 9, 13]. For example, hypoxic exposure enhances the transition from epithelial characteristics to a dispersed morphology, a typical aspect of EMT, due to loss of E-cadherin and increased expression of EMT-associated genes [14]. Hypoxia also increases cancer cell invasiveness by activating several proteolytic enzymes responsible for extracellular matrix degradation, such as matrix metalloproteinase-2 (MMP-2) and urokinase plasminogen activator receptor (uPAR) and uPA; although it should be noted that the expression of some of these genes is independent of HIF-1 [13]. Vascular endothelial growth factor (VEGF) is another hypoxia mediated critical HIF-1 target gene. VEGF is known to stimulate endothelial cell proliferation and migration, as well as the breakdown of the endothelial cell barrier through disruption of adhesion junctions which increases blood vessel permeability [15]. Thus hypoxic exposure affects many of the key aspects of the metastatic cascade including angiogenesis, intravasation, extravasation, and secondary growth [13].

174

7.2.2

Y. Dai et al.

Acute Hypoxia and Chronic Hypoxia

Tumor hypoxia is generally classified into two types: chronic and acute hypoxia [16]. Tumor cells residing at the limits of oxygen diffusion from functional blood vessels may experience chronic or “diffusion-limited” hypoxia. Such hypoxic conditions typically last for relatively long periods of hours or days [16]. In contrast, some tumor cells may be exposed to short-term, transient hypoxia as a result of intermittent blood flow due to tumor vasculature abnormalities [17]. Such acute or “perfusion-limited” hypoxia is characterized by rapid reoxygenation and hypoxicoxic cycles [18] having periodicities of minutes to hours. However, it should be recognized that such temporal distinctions are somewhat artificial and it is most likely that there is no discrete temporal boundary between these two states in the tumor [7]. Although both types of hypoxia coexist in human tumors, their relative impacts on the metastatic dissemination of cancer cells may vary [19]. While some studies appear to favor chronic hypoxia as the key factor [20, 21], others suggest that acute hypoxia may be equally or more critical in mediating a cancer cell’s metastatic behavior [19, 22, 23]. For example, when tumor-bearing mice were exposed to repeated short-term intervals of hypoxia, the frequency of lung metastases was markedly increased [22, 24]. This complexity is further illustrated by reported conflicting findings often in the same tumor type. For example, in prostate cancer models, while one study proposed that chronic hypoxia may achieve a more aggressive phenotype than acute hypoxia [21], others reported that 48 h chronic hypoxic exposure resulted in decreased invasion [25] and increased apoptosis [26].

7.2.3

HIF-1 Signaling and Metastasis

HIF-1 over-expression has been implicated as a potential prognostic and predictive factor in human cancers. In the presence of ambient oxygen, the HIF-1a subunit is degraded by the ubiquitin-proteasome system via binding to the von Hippel-Lindau (VHL) protein. Under hypoxic conditions, HIF-1a is rapidly stabilized and functionally activated [27]. In some cancers, HIF-1 can be constitutively activated in an O2-independent manner under aerobic conditions as a consequence of dysregulated signaling pathways that involve over-activation of oncogenes (EGFR, Akt, Src) or inactivation of tumor suppressor genes (PTEN, VHL) [27]. The finding that HIF-1 can be activated through either oncogene gain of function or tumor suppressor loss of function suggests that increased HIF-1 signaling activity represents a final convergent pathway in cancer cell dissemination regardless of oxygen status. Since HIF-1 is believed to mediate a pleiotropic role under both aerobic and anaerobic conditions, it may serve as a surrogate marker of tumor oxygenation and response. Multiple experimental studies have linked HIF-1 activity with cancer progression. For example, immuno-histochemical analysis of human cancer biopsies displayed increased levels of HIF-1a protein in the majority of primary tumors and

7

Impact of Tumor Hypoxia, Src, and Met Signaling in the Dissemination…

175

their metastases [28]. In breast cancer, HIF-1 is suggested to contribute to lung [29] and bone [30] metastases. Furthermore, studies from early stage of breast, cervical, and endometrial cancers suggest that a subset of patients whose tumors exhibit high HIF-1a expression have a significantly increased mortality, indicating the important role of HIF-1 in metastasis irrespective of primary tumor status [31].

7.3

Src and Met Kinases Are Intrinsic Contributors in Facilitating Cancer Metastasis

Deregulated activation of oncogenic proteins has been widely accepted as a cause of cancer onset, progression and metastasis. Src non-receptor tyrosine kinase and Met receptor tyrosine kinase are two crucial signaling kinase families that orchestrate a plethora of cellular functions involved in the metastasis process. In human cancers, Src hyperactivation occurs frequently and aberrant Met activation, by both liganddependent and -independent mechanisms, is a typical event.

7.3.1

Src Signaling and Metastasis

Src tyrosine kinase family consists of nine members that share a unique N-terminal region, two Src homology domains, a highly conserved kinase domain, and a C-terminal tail containing a negative regulatory tyrosine residue [32]. c-Src is the prototypical member that plays a pleiotropic role in controlling multiple functions including proliferation, adhesion, migration, invasion, and angiogenesis. In cancer, Src family kinases generally lack mutation or gene amplification but their enzymatic activities frequently increase in primary tumors with further enhancement in metastatic lesions [33–36]. Aberrant expression and activation of c-Src have been correlated with aggressiveness and adverse outcome.

7.3.1.1

Role of Src in Metastasis-Associated Functions

Src signaling participates in multiple steps of tumor cell dissemination including angiogenesis, motility, invasion, and distant colonization [37–40]. Constitutively activated c-Src induces cadherin-catenin complex loss-of-function, thereby promoting cell differentiation and invasiveness [41]. Moreover, c-Src has long been known to regulate the dynamics of the cytoskeleton, resulting in increased cell motility and deceased cell-matrix adhesion [42]. In addition, c-Src is a pivotal contributor in the invasion process by up-regulating matrix proteases that lead to degradation of the basement membrane [43]. The function of c-Src in endothelial cells also has been extensively studied. c-Src mediates VEGF/VEGFR signaling

176

Y. Dai et al.

Fig. 7.2 c-Src promotes bone metastasis. Nude mice were intracardially injected with an MDA-MB-231-derived highly metastatic subline that was transfected with a control vector (Control), a c-Src shRNA vector (Src RNAi), or c-Src shRNA and shRNA-resistant c-Src expression vectors (Rescued). Bioluminescent, radiographic and H&E analysis of bone lesions from representative mice in each group were shown at the indicated times after inoculation. In the X-ray images, areas of bone lysis are indicated by dotted lines. In the H&E staining, asterisks indicate tumor (Data are modified from Zhang et al. [158], with permission)

and results in endothelial cell migration, invasion and tube formation, aspects of angiogenesis, a critical process that aids the dissemination of tumor cells [40]. Src can also phosphorylate VE-Cadherin, a cell adhesion molecule that mediates vascular cell-cell junction, leading to increased vascular permeability and facilitating entry of tumor cells into blood vessels [44]. c-Src is also a key mediator of bone resorption due to its essential role in osteoclast function [45]. Disruption of Src signaling in osteoclasts results in decreased cell motility and prevents the formation of ruffled membranes, a key step during bone resorption [46]. Mice injected with human breast cancer cells with a constitutively active Src developed increased osteolytic bone metastases [47]. In a highly metastatic breast cancer subline c-Src knockdown significantly decreased outgrowth of the bone lesions; this could be rescued by c-Src protein reintroduction (Fig. 7.2). Taken together, the role of c-Src in bone microenvironment makes it an attractive target for anti-metastatic therapy in cancers with osteolytic lesions such as breast, lung, prostate, and kidney.

7

Impact of Tumor Hypoxia, Src, and Met Signaling in the Dissemination…

7.3.1.2

177

Src Signaling Pathways

Src participates in the activation of various downstream pathways through molecular interactions with a host of proteins including trans-membrane receptors, such as growth factor receptors and G protein-coupled receptors, and other non-receptor tyrosine kinases including focal adhesion kinase (FAK) and signal transducer and activator of transcription proteins (STATs) [37, 48]. As a biological consequence of interactions with Src, each of these effectors functions in unique signaling pathways. FAK, a key protein in modulating the formation and turnover of focal adhesions, is perhaps the best known partner of c-Src. FAK and c-Src have been shown to form a transient, active complex following integrin engagement by extracellular matrix proteins. As a consequence both kinases are activated on at least four different tyrosine residues, leading to amplification of FAK kinase activity and triggering of other signaling molecules [49]. Upon association, the FAK/c-Src complex activates a cascade of downstream cytoskeletal components including paxillin and p130Cas, recruits ERK, Jun kinase (JNK) and Rho GTPase pathways, and triggers actin stress fiber rearrangement, focal adhesion and cell migration. Recent studies have also documented that both c-Src and FAK catalytic activities are important in promoting VEGF-associated angiogenesis and protease-associated invasion, suggesting that c-Src cooperating with FAK generates signals leading to tumor growth and metastasis [50]. In addition to VEGFR, c-Src has also been shown to be downstream of a number of other growth factor receptors, including epidermal growth factor receptor (EGFR) and platelet-derived growth factor receptor (PDGFR) [37]. Thus upon activation, c-Src can amplify a great number of critical signaling pathways and effector molecules that are responsible for migration, invasion, survival, and angiogenesis [37, 38, 51].

7.3.2

Met Signaling and Metastasis

Receptor tyrosine kinase c-Met is encoded by MET proto-oncogene. Oncogenic c-Met signaling is widely implicated in various human malignancies and the expression of both the hepatocyte growth factor (HGF) ligand and the c-Met receptor has been correlated with tumor progression and poor prognosis in patients. Although c-Met is normally activated by HGF under physiological conditions, abnormal c-Met activation is typically observed in human cancers [52, 53] and the HGF/c-Met axis has been linked to malignant transformation and metastasis [54]. In lung cancer for example, expression levels of both HGF and c-Met have been associated with advanced tumor stage and worse clinical outcome [55]. In prostate cancer, increased serum levels of HGF have served as an independent prognosticator in advanced disease [56]; also, Met is expressed more frequently in metastatic lesions than in primary tumors [57], with 100% of prostate cancer bone metastases being Met-positive [58]. These findings support the notion that elevated HGF and c-Met may be used as predictive markers for metastatic formation in cancer patients.

178

Y. Dai et al.

Table 7.1 Abnormal MET status and metastasis in human cancers MET status Cancer types Metastatic phenotypes HGF-dependent Paracrine loop SCC Scattering, lung metastasis [70] Thyroid cancer Morphology, invasion, adhesion [71] HCC Morphology, motility [72] Prostate cancer Scattering, motility, invasion [73] Autocrine loop NSCLC Spontaneous metastasis [74] Melanoma Liver metastasis [75] HGF-independent Gene amplification Activating mutation

Crosstalk with other proteins Overexpresison

CRC Multiple cancers HNSCC HNSCC NSCLC Breast cancer CRC Melanoma

Liver metastasis [65, 76] Migration, invasion, lymph node and pulmonary metastasis [77] Lymph node metastasis [63] Distant metastasis [78] Morphology, motility [79] Migration [80] Liver metastasis [81] Motility, invasion, lung metastasis [82]

HNSCC head and neck squamous cell carcinoma, HCC hepatocellular carcinoma, NSCLC nonsmall cell lung carcinoma, CRC colorectal cancer

7.3.2.1

c-Met Is Aberrantly Activated in Human Cancers

c-Met is one of the most common abnormally activated signaling molecules in human cancers [59]. Alterations in MET gene amplification as well as activating mutations and crosstalk between the c-Met receptor and other signaling molecules have been reported to result in constitutive activation of the c-Met kinase independent of the HGF ligand [60, 61]. These abnormalities have been found in papillary renal cell carcinoma [62], head and neck cancer [63], gastric cancer [64] and metastatic colorectal cancer [65]. In some cases, tumor cells express both HGF and c-Met, thus establishing an autocrine loop in which tumor cell secreted HGF binds to c-Met and causes its activation. Such HGF-dependent autocrine c-Met activation is considered to be a self-supportive mechanism involved in cancer cell transformation, proliferation and survival. It has been detected in a variety of human primary and metastatic tumors, including breast cancers [66], gliomas [67] and osteosarcomas [68]. Although aberrant MET activation occurs in only some cancer types, in a majority of human malignancies wild-type c-Met over-expression, due to MET mRNA up-regulation, is a predominant event [69]. In this circumstance, the c-Met receptor still depends on the HGF ligand for activation, but increased expression of c-Met on the cell surface would favor HGF-independent activation through spontaneous receptor dimerization [61]. The contribution of the abnormal MET status on tumor progression and response to therapy has been widely studied. In addition, there exists a growing body of preclinical and clinical evidence that

7

Impact of Tumor Hypoxia, Src, and Met Signaling in the Dissemination…

179

supports the notion that a tumor’s metastatic potential can be affected by HGF/c-Met axis abnormalities. Selected examples are shown in Table 7.1.

7.3.2.2

c-Met, Invasive Growth and Metastasis

Functionally, c-Met orchestrates invasive growth by activating multiple cell functions including proliferation, scattering, migration, invasion, branching morphogenesis, and angiogenesis [83, 84]. Under normal conditions the HGF/c-Met axis is essential for embryogenesis and tissue homeostasis, while in cancer, both the ligand (HGF) and the receptor (c-Met) have been closely linked to the activation of invasive growth and metastatic spread [83]. Activating somatic mutations of MET and associated with increased aggressiveness and extensive metastases [77]. Similarly, the induction of HGF/c-Met autocrine loops in cell lines or in transgenic animals can lead to the generation of invasive tumors [75]. HGF-binding enhances the potential of invasive growth that contributes to tumor cell dissemination in cancer cells with highly expressed c-Met [85]. For example, in PC-3 prostate cancer cells that over-express c-Met [73, 86], human HGF induces cell scatter, migration, invasion (Fig. 7.3a) and clonogenic growth (Fig. 7.3b). The process of cell scattering can be divided into three phases, namely cell-cell dissociation, cell migration, and cell spreading [87]. The scattered phenotype is typically induced by HGF (hence its original designation as scatter factor) [88]. Mechanistically, HGF/c-Met acts as a stimulator to increase cell migration and invasion (Table 7.1) by promoting cytoskeleton rearrangement and stimulating the production of proteases responsible for basement membrane degradation [87]. HGF also functions as a potent pro-angiogenic molecule that directly stimulates endothelial cell migration, invasion and tube formation [89]. Furthermore, HGF can promote the production of other angiogenic factors in tumor cells [90], thus amplifying tumor angiogenesis. MET activation may also lead to cell survival advantages as suggested by the finding of MET-induced protection from anoikis [91], an anchorage-independent cell death mode that negatively affects tumor cell dissemination [92].

7.3.2.3

HGF/c-Met Signaling Pathways

The c-Met protein is synthesized as a 170 kDa precursor which is glycosylated and proteolytically cleaved into a 50 kDa a-chain and a 140 kDa b-chain thus forming a heterodimeric molecule. Typically, c-Met is autophosphorylated by HGF at tyrosines 1234 and 1235 in the activation loop of the kinase domain in order to achieve full enzyme activation [51]. Phosphorylated c-Met further triggers activation of two main downstream signaling pathways, Ras-mitogenactivated protein kinase (MAPK) and phosphoinositide 3-kinases (PI3K)-Akt (Fig. 7.3c), via the adaptor proteins Gab-1 and Grb2 [85]. Both ERK and PI3K pathways have been found to be involved in cell spreading; HGF-induced cancer cell scattering is dependent on the activity of Rho family GTPases mediated by

180

Y. Dai et al.

Fig. 7.3 Effects of human HGF (25 ng/ml) on cellular functions and c-Met signaling pathways in PC-3 cells. (a) Cells were treated with HGF for 24 h. Cell scattering, motility and invasion were detected using a scatter assay, a wound-healing assay, and a Matrigel-based chamber assay, respectively. Magnification, ×5. (b) Cells (700 cells/well) were seeded into a 24-well plate and incubated with HGF for 10 days. Cell proliferation was assessed using an MTT assay. Columns, mean; bars, SD (n = 6). ***P < 0.001 (t-test). (c) Serum-starved cells were exposed to HGF for 10 min. c-Met signaling molecules were shown by Western blot analysis. C = control (Data are adapted from Dai and Siemann [73])

the PI3K pathway [93]. Met also activates the migration-associated components Ras, Rac1 and p21-activated kinase (PAK) and controls cytoskeletal rearrangement, cell adhesion and cell motility [94]. PI3K activation of AKT also promotes cell survival. c-Src is another downstream target of c-Met [95]. In addition, other signaling molecules, such as Jun amino-terminal kinase (JNK), nuclear factor-kB (NF-kB) and b-catenin, can participate in the HGF/c-Met pathway [52]. Overall, Met signaling is highly complex and involves distinct but interacting cascades [53]. In addition to ligand activation, c-Met can also be transactivated by other transmembrane receptors. For example, EGFR can physically interact with c-Met in tumor cells [96] such that c-Met is phosphorylated by an EGFR ligand in the absence of HGF [96]. Reciprocally, HGF/c-Met signaling may induce expression of EGFR ligands thus leading to EGFR signaling activation [97], suggesting a crosstalk

7

Impact of Tumor Hypoxia, Src, and Met Signaling in the Dissemination…

181

between these two receptor tyrosine kinases. In addition to EGFR, other receptors including HER2 [98] and Ron [99], another Met family member, have been shown to interact with c-Met to enhance the malignant phenotype.

7.4

Interplay Between HIF-1, c-Src and c-Met Signaling Pathways

The crosstalk between HIF-1 and other signaling pathways has been extensively studied [12]. Under hypoxic conditions HIF-1 up-regulates pro-survival and proinvasive signaling pathways such as Akt [100] and c-Met [101] while under aerobic conditions HIF-1 is regulated by a variety of genetic factors including SRC [12].

7.4.1

Hypoxia, HIF-1 and c-Src Signaling

Although data concerning the role of c-Src signaling under hypoxic conditions are somewhat limited, it has been suggested that Src activation may promote survival of hypoxic cells [102]. A recent study in pancreatic and cervical carcinoma xenografts also indicated that the expression levels of total Src protein were higher in the chronically hypoxic regions of the tumors [103]. Several other studies have reported that c-Src is required for increased expression of VEGF. In the presence of oxygen, consistent with the finding that constitutively activated c-Src leads to elevated HIF-1a expression [104], Jiang et al. reported that increased HIF-1a synthesis and VEGF gene activation are caused by SRC gene activation [105]. In contrast, oxygen deprivation can increase the kinase activity of c-Src leading to up-regulation of VEGF mRNA [106]. Furthermore, Src-dependent hypoxia-induced VEGF expression appears to be regulated by STAT3 activation and HIF-1a accumulation [107]. Clearly, definitive insights into the relationship between HIF-1 and c-Src signaling are lacking and continuing exploration of possible interactions of these pathways is needed.

7.4.2

Hypoxia, HIF-1 and HGF/c-Met Signaling

Hypoxia directly promotes motility and invasion by inducing MET transcription [101, 108, 109] and sensitizing cells to HGF stimulation [101]. This invasive phenotype can be reversed by MET [101] or HIF-1a [108] knockdown. Since the HGF/c-Met pathway induces both the MET gene and HIF-1 activity [13], amplification of HGF-induced c-Met signaling by hypoxia provides a molecular mechanism by which low oxygen tensions such as can occur in solid tumors could cause malignancies to become more aggressive, invasive and metastatic. Although this possibility is intriguing, it should be noted that the MET gene is not always found to be up-regulated by oxygen deprivation [110–112].

182

Y. Dai et al.

With regard to an association between HIF-1 and c-Met, immuno-histochemical studies in breast cancer patients showed HIF-1a and MET to be positively correlated, independent predictors for distant metastasis, and associated with poor survival [113]. These findings strengthen the possibility that tumor hypoxia promotes metastasis through MET induction. Similar relationships have been reported for pancreatic adenocarcinoma [114] and papillary thyroid carcinoma cells [115]. However, MET transcription and cell invasiveness can also be induced by the genetic inactivation of VHL, which leads to HIF-1a stabilization in the absence of hypoxia and suggests an aerobic regulation of c-Met by alterations of hypoxia-associated proteins [116]. Collectively, the current uncertainties surrounding the effects of hypoxia on MET expression may reflect the complexities of the crosstalk between c-Met receptor and HIF-1 signaling.

7.4.3

Interplay Between c-Src and the HGF/c-Met Axis

c-Src and c-Met are two critical signaling molecules that trigger multiple similar cell functions including migration, invasion, survival, and angiogenesis. Not surprisingly increased activation of both c-Src and c-Met have been associated with enhanced metastatic potential in cancer cells. Although a direct physical interaction between these two functional related proteins has not been reported, the molecules do display close relationships at least in a certain cancer types. In breast cancer, activation of c-Src results in elevated HGF expression and activation of autocrine HGF/c-Met signaling [117]. c-Src also is activated in response to HGF stimulation in both breast carcinoma cells and breast epithelial cells; corresponding to a coassociation between c-Met and c-Src [118]. In prostate cancer cells, c-Src activation of MAPKs plays a key role in HGF-mediated cell protection against chemotherapeutic agents [119]. In colorectal cancer, c-Met activation of c-Src was observed to be critical in mediating tumorigenesis and liver metastasis [120]. In addition, a majority of colorectal cell lines found to express high Src family kinase activity also demonstrated activated c-Met [121]. Moreover, inhibition of Src reduced c-Met phosphorylation in most cases, indicating a reversed signal flow from Src to c-Met [121]. All these findings indicate that c-Src can function both as an upstream initiator and a downstream effector of c-Met, clearly highlighting the potential association between these two key signaling molecules.

7.5

Therapeutic Interventions Targeting Hypoxia/HIF-1, Src, or Met Signaling to Inhibit Metastasis

The development of novel therapeutic strategies that impair the metastatic process is critical to improving cancer patient survival. Two approaches to interfere with the dissemination of cancer cells are to target the regulation of signaling pathways associated with invasiveness and angiogenesis, essential components of the

7

Impact of Tumor Hypoxia, Src, and Met Signaling in the Dissemination…

183

metastatic cascade. Given their pleiotropic roles in microenvironmental responsive signaling cascades, targeting HIF-1, c-Src and c-Met would be expected to have therapeutic efficacy in constraining tumor cell dissemination.

7.5.1

Targeting Hypoxia and HIF-1

Given its central role in tumor cell response to oxygen deprivation, HIF-1 has become an attractive target for drug development [27]. HIF-1 inhibitors can tentatively be divided into several categories that target different processes during HIF1a activation, including mRNA expression, transcriptional activity, protein translation, DNA binding, and protein degradation [146]. Interestingly, most of these inhibitors were not developed as HIF-1 inhibitors per se; belonging instead to a wide variety of molecules including signaling pathway inhibitors, chaperone inhibitors, chemotherapeutic agents and natural compounds [147]. Many of the targeted compounds have been found to have antiangiogenic effects likely due to the elimination of HIF-1 activity in tumor cells and/or endothelial cells. Such agents include gefitinib (EGFR inhibitor), trastuzumab (HER2 inhibitor), and rapamycin (mTOR inhibitor). Although the primary focus of such agents has been the assessment of their effects on primary tumor growth and angiogenesis, studies evaluating the impact of HIF-1 inhibitors on metastasis have begun (Table 7.2).

7.5.2

Targeting Src Kinase

The pivotal role of Src in neoplastic cell progression has made it a relevant target in cancer management. Currently, the three lead small molecule ATP competitors, dasatinib, saracatinib and bosutinib, are being evaluated in clinical trials for many different types of solid tumors [40]. Preclinical studies are promising and indicate that Src kinase inhibitors can affect not only primary tumor growth [148], but also the development of secondary metastases [127, 128, 149–151]. For example, in vitro, dasatinib inhibits PC-3 cell adhesion (Fig. 7.4a), migration (Fig. 7.4b) and invasion (Fig. 7.4c) as well as significantly reducing prostate cancer cell induced angiogenesis (Fig. 7.4d). Saracatinib and bosutinib also have been shown to effectively suppress the metastatic phenotype in vitro and impair the formation of metastases in vivo in a variety of cancer models (Table 7.2), further supporting the potential of Src inhibitors as anti-metastatic agents.

7.5.3

Targeting Met Kinase

Dysregulation of c-Met signaling is strongly associated with metastasis and poor prognosis in multiple cancers including carcinomas, sarcomas, and gliomas, making MET a potentially important prognostic indicator for cancer progression [87].

ATP-competitor

ATP-competitor

ATP-competitor

ATP-competitor

ATP-competitor ATP-competitor ATP-competitor ATP non-competitor

Src targeting Dasatinib

Saracatinib

Bosutinib

Met targeting SU11274

BMS-777607 PHA-665752 PF-2341066 ARQ197

Melanoma; Multiple cancers; Ovarian cancer Prostate cancer Multiple cancers Lung carcinoma Lung carcinoma; Colon carcinoma

Prostate cancer [135]; Breast cancer [136, 137]; Colon cancer and lung cancer [138]

Bladder cancer [129]; Sarcoma [130]; Prostate cancer [131, 132]; HSCC [133]; Breast cancer [134]

Pancreatic cancer [125]; Melanoma [126]; Prostate cancer [127, 128];

Migration, liver metastasis [139]; Scatter, invasion, tubulogenesis [140]; Motility, invasion [141] Scatter, motility, invasion [73] Morphology, motility, invasion [142] Invasion [143] Invasion [144]; Liver metastasis [145]

Migration [125, 128]; Invasion [125, 128]; Lung metastasis [126]; Secondary growth [127]; Lymph node metastasis [128] Migration [129, 130, 132, 134]; Invasion [129–131, 133, 134]; Scatter [129]; Adhesion [130]; Lymph node metastasis [129, 133]; Lung metastasis [130]; Bone metastasis [131] Migration [136]; Invasion [135, 136]; Extravasation [138]; Skeletonal metastasis [135]; Spontaneous metastasis [137]

Table 7.2 Selective small molecule inhibitors targeting HIF-1, Src or Met for the inhibition of metastasis in preclinical studies Compounds Targets Cancer models Disrupted functions HIF-1 targeting 2-methoxyestradiol HIF-1a translation Gastric adenocarcinoma Migration, invasion, adhesion, systemic tumor inhibitor spread [122] YC-1 Multiple mechanisms Hepatoma Migration, invasion, lung and liver metastasis [123] Vitexin Multiple mechanisms Multiple cancer cell lines Migration, invasion, angiogenesis [124]

184 Y. Dai et al.

7

Impact of Tumor Hypoxia, Src, and Met Signaling in the Dissemination…

185

Fig. 7.4 Effects of dasatinib on PC-3 cell functions. (a) Sub-confluent cells were treated with dasatinib for 24 h and floating cells were counted using a hemocytometer. (b) A confluent cell monolayer was scratched and exposed to dasatinib (10 nM) for 24 h. Magnification, ×5. (c) Cells were seeded into Matrigel-based chambers (104/well) using serum as a chemo-attractant and incubated with dasatinib for 24 h. Invaded cells were counted under a microscope. (d) Dasatinibpretreated (24 h) cells were injected (105 cells/site) intradermally in four locations on the chests of nude mice. Three days later the skin flaps were removed and the blood vessels grown to the tumor sites were counted using a dissecting microscope. (a and c): Columns, mean; bars, SD (n = 3); (d): Whiskers, 90th (above) and 10th (below) percentiles (n = 12). *P < 0.05; **P < 0.01 [compared to control, t-test for (a and c), Wilcoxon rank sum test for (d)]

The prevalence of HGF/c-Met pathway activation in human malignancies has led to rapid and extensive growth in the development of agents targeting this signaling pathway. Many intervention strategies have been developed including antagonistic compounds (competitors), monoclonal antibodies (mAbs) and small molecule kinase inhibitors [152]. Therapeutic mAbs specifically blocking c-Met activation are being evaluated in patients with advanced malignancies. These include mAbs

186

Y. Dai et al.

targeting both HGF (AMG102) and c-Met (MetMAb) [59]. Although neutralizing antibodies have shown promise when inhibiting HGF-stimulated functions, Met kinase inhibitors may have a broader range of application, having found to be effective in cancers driven by HGF [153], Met [154], or both [74, 155, 156]. To date, the primary clinical evaluations of these compounds have focused on their ability to inhibit primary tumor growth. However, preclinical studies clearly indicate that Met targeting agents may have considerable utility in interfering with the metastatic process (Table 7.2).

7.5.4

Combining Agents to Target Cancer Cell Dissemination

Metastasis is a complex process. Although targeted therapies have shown some promise in treating metastatic disease, successful therapeutic interventions are likely to require a combinatorial agent treatment approach. Given the heterogeneity of malignant cells [3] and the complexity and redundancy of molecular events involved in the metastatic process, eliminating one aspect of the metastatic cascade may still allow another to proceed uninterrupted [157]. Thus the simultaneous targeting of two or more signaling pathways involved in the metastatic process should provide superior outcomes than either intervention alone. For example, targeting both HIF-1 and c-Src may provide complimentary treatment strategies for controlling cancer cell spread from primary tumors containing hypoxic microenvironments. In cancers in which c-Src and c-Met are active, the combined targeting of both these two kinases should result in superior therapeutic efficacy. Ultimately a better understanding of the tumor microenvironment and molecular signatures in involved in cancer cell dissemination will offer opportunities to optimize the application of novel targeting agents alone or in combination to improve treatment outcomes.

7.6

Concluding Remarks

Hypoxia, a common feature of human cancers, is known to accelerate tumor cell dissemination. It is generally believed that in most cases HIF-1links the hypoxic cell response to metastatic progression. Indeed, positive correlations between tumor hypoxia and increased metastatic capacity have been suggested by both clinical and preclinical investigations. Altered intracellular signaling pathways such as c-Src and HGF/c-Met axis also have been identified as key contributors to the metastatic phenotype. Evidence exists to support the notion that such signaling cascades may be amplified by oxygen deprivation. Furthermore, the interrelated pathways of signaling molecules, HIF-1, c-Src and c-Met may cooperate to promote cell functions that lead to an elevated metastatic potential. The metastatic process of cancer cells may therefore be independently augmented by both extrinsic environmental factors (hypoxia) and intrinsic cellular features (dysregulated signaling pathways) (Fig. 7.5).

7

Impact of Tumor Hypoxia, Src, and Met Signaling in the Dissemination…

187

Fig. 7.5 Scheme of hypoxia/ HIF-1, c-Src kinase and the HGF/c-Met axis in the metastatic process

Given the pivotal role of hypoxia/HIF-1, c-Src and c-Met in tumor cell dissemination, small molecule therapeutic agents targeting these pathways have significant potential to interfere with the metastatic phenotype. While the impact of HIF-1 inhibitors remains to be investigated, small molecules targeting Src and Met kinases have been efficacious in impeding metastatic disease. Clearly it is of interest to explore the strategy of combining therapeutic interventions targeting both the tumor microenvironment and the intracellular signaling pathways associated with the spreading of cancer cells. Establishment of such combinatorial treatment modalities holds great promise to successfully interrupt cancer metastasis.

References 1. Jemal A, Siegel R, Xu J, Ward E (2010) Cancer statistics, 2010. CA Cancer J Clin 60:277–300 2. Manfredi S, Lepage C, Hatem C, Coatmeur O, Faivre J, Bouvier AM (2006) Epidemiology and management of liver metastases from colorectal cancer. Ann Surg 244:254–9 3. Talmadge JE, Fidler IJ (2010) AACR centennial series: the biology of cancer metastasis: historical perspective. Cancer Res 70:5649–69 4. Ljungkvist AS, Bussink J, Kaanders JH, Wiedenmann NE, Vlasman R, van der Kogel AJ (2006) Dynamics of hypoxia, proliferation and apoptosis after irradiation in a murine tumor model. Radiat Res 165:326–36 5. Bristow RG, Hill RP (2008) Hypoxia and metabolism. Hypoxia, DNA repair and genetic instability. Nat Rev Cancer 8:180–92 6. Chan DA, Giaccia AJ (2007) Hypoxia, gene expression, and metastasis. Cancer Metastasis Rev 26:333–9 7. Chaudary N, Hill RP (2007) Hypoxia and metastasis. Clin Cancer Res 13:1947–9 8. Vergis R, Corbishley CM, Norman AR et al (2008) Intrinsic markers of tumour hypoxia and angiogenesis in localised prostate cancer and outcome of radical treatment: a retrospective analysis of two randomised radiotherapy trials and one surgical cohort study. Lancet Oncol 9:342–51 9. Gort EH, Groot AJ, van der Wall E, van Diest PJ, Vooijs MA (2008) Hypoxic regulation of metastasis via hypoxia-inducible factors. Curr Mol Med 8:60–7 10. Fyles A, Milosevic M, Pintilie M et al (2006) Long-term performance of interstitial fluid pressure and hypoxia as prognostic factors in cervix cancer. Radiother Oncol 80:132–7 11. Zhang L, Hill RP (2007) Hypoxia enhances metastatic efficiency in HT1080 fibrosarcoma cells by increasing cell survival in lungs, not cell adhesion and invasion. Cancer Res 67:7789–97

188

Y. Dai et al.

12. Semenza GL (2003) Targeting HIF-1 for cancer therapy. Nat Rev Cancer 3:721–32 13. Sullivan R, Graham CH (2007) Hypoxia-driven selection of the metastatic phenotype. Cancer Metastasis Rev 26:319–31 14. Mak P, Leav I, Pursell B et al (2010) ERbeta impedes prostate cancer EMT by destabilizing HIF-1alpha and inhibiting VEGF-mediated snail nuclear localization: implications for Gleason grading. Cancer Cell 17:319–32 15. Dvorak HF, Nagy JA, Feng D, Brown LF, Dvorak AM (1999) Vascular permeability factor/ vascular endothelial growth factor and the significance of microvascular hyperpermeability in angiogenesis. Curr Top Microbiol Immunol 237:97–132 16. Brown JM, Wilson WR (2004) Exploiting tumour hypoxia in cancer treatment. Nat Rev Cancer 4:437–47 17. Fukumura D, Jain RK (2007) Tumor microenvironment abnormalities: causes, consequences, and strategies to normalize. J Cell Biochem 101:937–49 18. Kimura H, Braun RD, Ong ET et al (1996) Fluctuations in red cell flux in tumor microvessels can lead to transient hypoxia and reoxygenation in tumor parenchyma. Cancer Res 56:5522–8 19. Rofstad EK, Galappathi K, Mathiesen B, Ruud EB (2007) Fluctuating and diffusion-limited hypoxia in hypoxia-induced metastasis. Clin Cancer Res 13:1971–8 20. Thews O, Wolloscheck T, Dillenburg W et al (2004) Microenvironmental adaptation of experimental tumours to chronic vs. acute hypoxia. Br J Cancer 91:1181–9 21. Alqawi O, Wang HP, Espiritu M, Singh G (2007) Chronic hypoxia promotes an aggressive phenotype in rat prostate cancer cells. Free Radic Res 41:788–97 22. Cairns RA, Kalliomaki T, Hill RP (2001) Acute (cyclic) hypoxia enhances spontaneous metastasis of KHT murine tumors. Cancer Res 61:8903–8 23. Cairns RA, Hill RP (2004) Acute hypoxia enhances spontaneous lymph node metastasis in an orthotopic murine model of human cervical carcinoma. Cancer Res 64:2054–61 24. Hill RP, De Jaeger K, Jang A, Cairns R (2001) pH, hypoxia and metastasis. Novartis Found Symp 240:154–65; discussion 65–8 25. Ackerstaff E, Artemov D, Gillies RJ, Bhujwalla ZM (2007) Hypoxia and the presence of human vascular endothelial cells affect prostate cancer cell invasion and metabolism. Neoplasia 9:1138–51 26. McKenzie S, Sakamoto S, Kyprianou N (2008) Maspin modulates prostate cancer cell apoptotic and angiogenic response to hypoxia via targeting AKT. Oncogene 27:7171–9 27. Giaccia A, Siim BG, Johnson RS (2003) HIF-1 as a target for drug development. Nat Rev Drug Discov 2:803–11 28. Zhong H, De Marzo AM, Laughner E et al (1999) Overexpression of hypoxia-inducible factor 1alpha in common human cancers and their metastases. Cancer Res 59:5830–5 29. Liao D, Corle C, Seagroves TN, Johnson RS (2007) Hypoxia-inducible factor-1alpha is a key regulator of metastasis in a transgenic model of cancer initiation and progression. Cancer Res 67:563–72 30. Hiraga T, Kizaka-Kondoh S, Hirota K, Hiraoka M, Yoneda T (2007) Hypoxia and hypoxiainducible factor-1 expression enhance osteolytic bone metastases of breast cancer. Cancer Res 67:4157–63 31. Semenza GL (2010) Defining the role of hypoxia-inducible factor 1 in cancer biology and therapeutics. Oncogene 29:625–34 32. Kopetz S, Shah AN, Gallick GE (2007) Src continues aging: current and future clinical directions. Clin Cancer Res 13:7232–6 33. Trevino JG, Summy JM, Lesslie DP et al (2006) Inhibition of SRC expression and activity inhibits tumor progression and metastasis of human pancreatic adenocarcinoma cells in an orthotopic nude mouse model. Am J Pathol 168:962–72 34. Verbeek BS, Vroom TM, Adriaansen-Slot SS et al (1996) c-Src protein expression is increased in human breast cancer. An immunohistochemical and biochemical analysis. J Pathol 180:383–8 35. Tatarov O, Mitchell TJ, Seywright M, Leung HY, Brunton VG, Edwards J (2009) SRC family kinase activity is up-regulated in hormone-refractory prostate cancer. Clin Cancer Res 15:3540–9

7

Impact of Tumor Hypoxia, Src, and Met Signaling in the Dissemination…

189

36. Bolen JB, Veillette A, Schwartz AM, DeSeau V, Rosen N (1987) Activation of pp 60c-src protein kinase activity in human colon carcinoma. Proc Natl Acad Sci U S A 84:2251–5 37. Fizazi K (2007) The role of Src in prostate cancer. Ann Oncol 18:1765–73 38. Finn RS (2008) Targeting Src in breast cancer. Ann Oncol 19:1379–86 39. Chambers AF, Groom AC, MacDonald IC (2002) Dissemination and growth of cancer cells in metastatic sites. Nat Rev Cancer 2:563–72 40. Kim LC, Song L, Haura EB (2009) Src kinases as therapeutic targets for cancer. Nat Rev Clin Oncol 6:587–95 41. Irby RB, Yeatman TJ (2002) Increased Src activity disrupts cadherin/catenin-mediated homotypic adhesion in human colon cancer and transformed rodent cells. Cancer Res 62:2669–74 42. Frame MC (2002) Src in cancer: deregulation and consequences for cell behaviour. Biochim Biophys Acta 1602:114–30 43. Allgayer H, Wang H, Gallick GE et al (1999) Transcriptional induction of the urokinase receptor gene by a constitutively active Src. Requirement of an upstream motif (-152/-135) bound with Sp1. J Biol Chem 274:18428–37 44. Criscuoli ML, Nguyen M, Eliceiri BP (2005) Tumor metastasis but not tumor growth is dependent on Src-mediated vascular permeability. Blood 105:1508–14 45. Miyazaki T, Sanjay A, Neff L, Tanaka S, Horne WC, Baron R (2004) Src kinase activity is essential for osteoclast function. J Biol Chem 279:17660–6 46. Sanjay A, Houghton A, Neff L et al (2001) Cbl associates with Pyk2 and Src to regulate Src kinase activity, alpha(v)beta(3) integrin-mediated signaling, cell adhesion, and osteoclast motility. J Cell Biol 152:181–95 47. Myoui A, Nishimura R, Williams PJ et al (2003) C-SRC tyrosine kinase activity is associated with tumor colonization in bone and lung in an animal model of human breast cancer metastasis. Cancer Res 63:5028–33 48. Ishizawar R, Parsons SJ (2004) c-Src and cooperating partners in human cancer. Cancer Cell 6:209–14 49. Schlaepfer DD, Mitra SK (2004) Multiple connections link FAK to cell motility and invasion. Curr Opin Genet Dev 14:92–101 50. Mitra SK, Schlaepfer DD (2006) Integrin-regulated FAK-Src signaling in normal and cancer cells. Curr Opin Cell Biol 18:516–23 51. Wheeler DL, Iida M, Dunn EF (2009) The role of Src in solid tumors. Oncologist 14:667–78 52. Birchmeier C, Birchmeier W, Gherardi E, Vande Woude GF (2003) Met, metastasis, motility and more. Nat Rev Mol Cell Biol 4:915–25 53. Trusolino L, Bertotti A, Comoglio PM (2010) MET signalling: principles and functions in development, organ regeneration and cancer. Nat Rev Mol Cell Biol 11:834–48 54. Lesko E, Majka M (2008) The biological role of HGF-MET axis in tumor growth and development of metastasis. Front Biosci 13:1271–80 55. Siegfried JM, Weissfeld LA, Singh-Kaw P, Weyant RJ, Testa JR, Landreneau RJ (1997) Association of immunoreactive hepatocyte growth factor with poor survival in resectable nonsmall cell lung cancer. Cancer Res 57:433–9 56. Gupta A, Karakiewicz PI, Roehrborn CG, Lotan Y, Zlotta AR, Shariat SF (2008) Predictive value of plasma hepatocyte growth factor/scatter factor levels in patients with clinically localized prostate cancer. Clin Cancer Res 14:7385–90 57. Russo AL, Jedlicka K, Wernick M et al (2009) Urine analysis and protein networking identify met as a marker of metastatic prostate cancer. Clin Cancer Res 15:4292–8 58. Knudsen BS, Gmyrek GA, Inra J et al (2002) High expression of the Met receptor in prostate cancer metastasis to bone. Urology 60:1113–7 59. Liu X, Newton RC, Scherle PA (2010) Developing c-MET pathway inhibitors for cancer therapy: progress and challenges. Trends Mol Med 16:37–45 60. Comoglio PM, Giordano S, Trusolino L (2008) Drug development of MET inhibitors: targeting oncogene addiction and expedience. Nat Rev Drug Discov 7:504–16 61. Danilkovitch-Miagkova A, Zbar B (2002) Dysregulation of Met receptor tyrosine kinase activity in invasive tumors. J Clin Invest 109:863–7

190

Y. Dai et al.

62. Schmidt L, Duh FM, Chen F et al (1997) Germline and somatic mutations in the tyrosine kinase domain of the MET proto-oncogene in papillary renal carcinomas. Nat Genet 16:68–73 63. Di Renzo MF, Olivero M, Martone T et al (2000) Somatic mutations of the MET oncogene are selected during metastatic spread of human HNSC carcinomas. Oncogene 19:1547–55 64. Houldsworth J, Cordon-Cardo C, Ladanyi M, Kelsen DP, Chaganti RS (1990) Gene amplification in gastric and esophageal adenocarcinomas. Cancer Res 50:6417–22 65. Di Renzo MF, Olivero M, Giacomini A et al (1995) Overexpression and amplification of the met/HGF receptor gene during the progression of colorectal cancer. Clin Cancer Res 1:147–54 66. Tuck AB, Park M, Sterns EE, Boag A, Elliott BE (1996) Coexpression of hepatocyte growth factor and receptor (Met) in human breast carcinoma. Am J Pathol 148:225–32 67. Koochekpour S, Jeffers M, Rulong S et al (1997) Met and hepatocyte growth factor/scatter factor expression in human gliomas. Cancer Res 57:5391–8 68. Ferracini R, Di Renzo MF, Scotlandi K et al (1995) The Met/HGF receptor is over-expressed in human osteosarcomas and is activated by either a paracrine or an autocrine circuit. Oncogene 10:739–49 69. Benvenuti S, Comoglio PM (2007) The MET receptor tyrosine kinase in invasion and metastasis. J Cell Physiol 213:316–25 70. Dong G, Lee TL, Yeh NT, Geoghegan J, Van Waes C, Chen Z (2004) Metastatic squamous cell carcinoma cells that overexpress c-Met exhibit enhanced angiogenesis factor expression, scattering and metastasis in response to hepatocyte growth factor. Oncogene 23:6199–208 71. Mineo R, Costantino A, Frasca F et al (2004) Activation of the hepatocyte growth factor (HGF)-Met system in papillary thyroid cancer: biological effects of HGF in thyroid cancer cells depend on Met expression levels. Endocrinology 145:4355–65 72. Xie Q, Liu KD, Hu MY, Zhou K (2001) SF/HGF-c-Met autocrine and paracrine promote metastasis of hepatocellular carcinoma. World J Gastroenterol 7:816–20 73. Dai Y, Siemann DW (2010) BMS-777607, a small-molecule met kinase inhibitor, suppresses hepatocyte growth factor-stimulated prostate cancer metastatic phenotype in vitro. Mol Cancer Ther 9:1554–61 74. Navab R, Liu J, Seiden-Long I et al (2009) Co-overexpression of Met and hepatocyte growth factor promotes systemic metastasis in NCI-H460 non-small cell lung carcinoma cells. Neoplasia 11:1292–300 75. Otsuka T, Takayama H, Sharp R et al (1998) c-Met autocrine activation induces development of malignant melanoma and acquisition of the metastatic phenotype. Cancer Res 58:5157–67 76. Zeng ZS, Weiser MR, Kuntz E et al (2008) c-Met gene amplification is associated with advanced stage colorectal cancer and liver metastases. Cancer Lett 265:258–69 77. Lorenzato A, Olivero M, Patane S et al (2002) Novel somatic mutations of the MET oncogene in human carcinoma metastases activating cell motility and invasion. Cancer Res 62:7025–30 78. Ghadjar P, Blank-Liss W, Simcock M et al (2009) MET Y1253D-activating point mutation and development of distant metastasis in advanced head and neck cancers. Clin Exp Metastasis 26:809–15 79. Puri N, Salgia R (2008) Synergism of EGFR and c-Met pathways, cross-talk and inhibition, in non-small cell lung cancer. J Carcinog 7:9 80. Tuck AB, Elliott BE, Hota C, Tremblay E, Chambers AF (2000) Osteopontin-induced, integrin-dependent migration of human mammary epithelial cells involves activation of the hepatocyte growth factor receptor (Met). J Cell Biochem 78:465–75 81. Fujita S, Sugano K (1997) Expression of c-met proto-oncogene in primary colorectal cancer and liver metastases. Jpn J Clin Oncol 27:378–83 82. Yu Y, Merlino G (2002) Constitutive c-Met signaling through a nonautocrine mechanism promotes metastasis in a transgenic transplantation model. Cancer Res 62:2951–6 83. Boccaccio C, Comoglio PM (2006) Invasive growth: a MET-driven genetic programme for cancer and stem cells. Nat Rev Cancer 6:637–45 84. Eder JP, Vande Woude GF, Boerner SA, LoRusso PM (2009) Novel therapeutic inhibitors of the c-Met signaling pathway in cancer. Clin Cancer Res 15:2207–14

7

Impact of Tumor Hypoxia, Src, and Met Signaling in the Dissemination…

191

85. Migliore C, Giordano S (2008) Molecular cancer therapy: can our expectation be MET? Eur J Cancer 44:641–51 86. Humphrey PA, Zhu X, Zarnegar R et al (1995) Hepatocyte growth factor and its receptor (c-MET) in prostatic carcinoma. Am J Pathol 147:386–96 87. Maulik G, Shrikhande A, Kijima T, Ma PC, Morrison PT, Salgia R (2002) Role of the hepatocyte growth factor receptor, c-Met, in oncogenesis and potential for therapeutic inhibition. Cytokine Growth Factor Rev 13:41–59 88. Savagner P, Yamada KM, Thiery JP (1997) The zinc-finger protein slug causes desmosome dissociation, an initial and necessary step for growth factor-induced epithelial-mesenchymal transition. J Cell Biol 137:1403–19 89. Rosen EM, Lamszus K, Laterra J, Polverini PJ, Rubin JS, Goldberg ID (1997) HGF/SF in angiogenesis. Ciba Found Symp 212:215–26; discussion 27–9 90. Dong G, Chen Z, Li ZY, Yeh NT, Bancroft CC, Van Waes C (2001) Hepatocyte growth factor/ scatter factor-induced activation of MEK and PI3K signal pathways contributes to expression of proangiogenic cytokines interleukin-8 and vascular endothelial growth factor in head and neck squamous cell carcinoma. Cancer Res 61:5911–8 91. Tang MK, Zhou HY, Yam JW, Wong AS (2010) c-Met overexpression contributes to the acquired apoptotic resistance of nonadherent ovarian cancer cells through a cross talk mediated by phosphatidylinositol 3-kinase and extracellular signal-regulated kinase 1/2. Neoplasia 12:128–38 92. Simpson CD, Anyiwe K, Schimmer AD (2008) Anoikis resistance and tumor metastasis. Cancer Lett 272:177–85 93. Wells CM, Ahmed T, Masters JR, Jones GE (2005) Rho family GTPases are activated during HGF-stimulated prostate cancer-cell scattering. Cell Motil Cytoskeleton 62:180–94 94. Ridley AJ, Comoglio PM, Hall A (1995) Regulation of scatter factor/hepatocyte growth factor responses by Ras, Rac, and Rho in MDCK cells. Mol Cell Biol 15:1110–22 95. Zhu H, Naujokas MA, Fixman ED, Torossian K, Park M (1994) Tyrosine 1356 in the carboxyl-terminal tail of the HGF/SF receptor is essential for the transduction of signals for cell motility and morphogenesis. J Biol Chem 269:29943–8 96. Jo M, Stolz DB, Esplen JE, Dorko K, Michalopoulos GK, Strom SC (2000) Cross-talk between epidermal growth factor receptor and c-Met signal pathways in transformed cells. J Biol Chem 275:8806–11 97. Reznik TE, Sang Y, Ma Y et al (2008) Transcription-dependent epidermal growth factor receptor activation by hepatocyte growth factor. Mol Cancer Res 6:139–50 98. Khoury H, Naujokas MA, Zuo D et al (2005) HGF converts ErbB2/Neu epithelial morphogenesis to cell invasion. Mol Biol Cell 16:550–61 99. Maggiora P, Lorenzato A, Fracchioli S et al (2003) The RON and MET oncogenes are coexpressed in human ovarian carcinomas and cooperate in activating invasiveness. Exp Cell Res 288:382–9 100. Lau CK, Yang ZF, Ho DW et al (2009) An Akt/hypoxia-inducible factor-1alpha/plateletderived growth factor-BB autocrine loop mediates hypoxia-induced chemoresistance in liver cancer cells and tumorigenic hepatic progenitor cells. Clin Cancer Res 15:3462–71 101. Pennacchietti S, Michieli P, Galluzzo M, Mazzone M, Giordano S, Comoglio PM (2003) Hypoxia promotes invasive growth by transcriptional activation of the met protooncogene. Cancer Cell 3:347–61 102. Lluis JM, Buricchi F, Chiarugi P, Morales A, Fernandez-Checa JC (2007) Dual role of mitochondrial reactive oxygen species in hypoxia signaling: activation of nuclear factor-{kappa} B via c-SRC and oxidant-dependent cell death. Cancer Res 67:7368–77 103. Pham NA, Magalhaes JM, Do T et al (2009) Activation of Src and Src-associated signaling pathways in relation to hypoxia in human cancer xenograft models. Int J Cancer 124:280–6 104. Karni R, Dor Y, Keshet E, Meyuhas O, Levitzki A (2002) Activated pp 60c-Src leads to elevated hypoxia-inducible factor (HIF)-1alpha expression under normoxia. J Biol Chem 277:42919–25

192

Y. Dai et al.

105. Jiang BH, Agani F, Passaniti A, Semenza GL (1997) V-SRC induces expression of hypoxiainducible factor 1 (HIF-1) and transcription of genes encoding vascular endothelial growth factor and enolase 1: involvement of HIF-1 in tumor progression. Cancer Res 57:5328–35 106. Mukhopadhyay D, Tsiokas L, Zhou XM, Foster D, Brugge JS, Sukhatme VP (1995) Hypoxic induction of human vascular endothelial growth factor expression through c-Src activation. Nature 375:577–81 107. Gray MJ, Zhang J, Ellis LM et al (2005) HIF-1alpha, STAT3, CBP/p300 and Ref-1/APE are components of a transcriptional complex that regulates Src-dependent hypoxia-induced expression of VEGF in pancreatic and prostate carcinomas. Oncogene 24:3110–20 108. Hara S, Nakashiro K, Klosek SK, Ishikawa T, Shintani S, Hamakawa H (2006) Hypoxia enhances c-Met/HGF receptor expression and signaling by activating HIF-1alpha in human salivary gland cancer cells. Oral Oncol 42:593–8 109. Knudsen BS, Vande Woude G (2008) Showering c-MET-dependent cancers with drugs. Curr Opin Genet Dev 18:87–96 110. Toda S, Yamada S, Aoki S, Inokuchi A, Sugihara H (2005) Air-liquid interface promotes invasive growth of laryngeal squamous cell carcinoma with or without hypoxia. Biochem Biophys Res Commun 326:866–72 111. Eckerich C, Zapf S, Fillbrandt R, Loges S, Westphal M, Lamszus K (2007) Hypoxia can induce c-Met expression in glioma cells and enhance SF/HGF-induced cell migration. Int J Cancer 121:276–83 112. Wang T, Niki T, Goto A et al (2007) Hypoxia increases the motility of lung adenocarcinoma cell line A549 via activation of the epidermal growth factor receptor pathway. Cancer Sci 98:506–11 113. Chen HH, Su WC, Lin PW, Guo HR, Lee WY (2007) Hypoxia-inducible factor-1alpha correlates with MET and metastasis in node-negative breast cancer. Breast Cancer Res Treat 103:167–75 114. Ide T, Kitajima Y, Miyoshi A et al (2006) Tumor-stromal cell interaction under hypoxia increases the invasiveness of pancreatic cancer cells through the hepatocyte growth factor/cMet pathway. Int J Cancer 119:2750–9 115. Scarpino S, Cancellario d’Alena F, Di Napoli A, Pasquini A, Marzullo A, Ruco LP (2004) Increased expression of Met protein is associated with up-regulation of hypoxia inducible factor-1 (HIF-1) in tumour cells in papillary carcinoma of the thyroid. J Pathol 202:352–8 116. Koochekpour S, Jeffers M, Wang PH et al (1999) The von Hippel-Lindau tumor suppressor gene inhibits hepatocyte growth factor/scatter factor-induced invasion and branching morphogenesis in renal carcinoma cells. Mol Cell Biol 19:5902–12 117. Hung W, Elliott B (2001) Co-operative effect of c-Src tyrosine kinase and Stat3 in activation of hepatocyte growth factor expression in mammary carcinoma cells. J Biol Chem 276:12395–403 118. Rahimi N, Hung W, Tremblay E, Saulnier R, Elliott B (1998) c-Src kinase activity is required for hepatocyte growth factor-induced motility and anchorage-independent growth of mammary carcinoma cells. J Biol Chem 273:33714–21 119. Fan S, Meng Q, Laterra JJ, Rosen EM (2009) Role of Src signal transduction pathways in scatter factor-mediated cellular protection. J Biol Chem 284:7561–77 120. Herynk MH, Zhang J, Parikh NU, Gallick GE (2007) Activation of Src by c-Met overexpression mediates metastatic properties of colorectal carcinoma cells. J Exp Ther Oncol 6:205–17 121. Emaduddin M, Bicknell DC, Bodmer WF, Feller SM (2008) Cell growth, global phosphotyrosine elevation, and c-Met phosphorylation through Src family kinases in colorectal cancer cells. Proc Natl Acad Sci U S A 105:2358–62 122. Rohwer N, Lobitz S, Daskalow K et al (2009) HIF-1alpha determines the metastatic potential of gastric cancer cells. Br J Cancer 100:772–81 123. Shin DH, Kim JH, Jung YJ et al (2007) Preclinical evaluation of YC-1, a HIF inhibitor, for the prevention of tumor spreading. Cancer Lett 255:107–16 124. Choi HJ, Eun JS, Kim BG, Kim SY, Jeon H, Soh Y (2006) Vitexin, an HIF-1alpha inhibitor, has anti-metastatic potential in PC12 cells. Mol Cells 22:291–9

7

Impact of Tumor Hypoxia, Src, and Met Signaling in the Dissemination…

193

125. Morton JP, Karim SA, Graham K et al (2010) Dasatinib inhibits the development of metastases in a mouse model of pancreatic ductal adenocarcinoma. Gastroenterology 139:292–303 126. Fraser CK, Lousberg EL, Guerin LR et al (2010) Dasatinib alters the metastatic phenotype of B16-OVA melanoma in vivo. Cancer Biol Ther 10:715–27 127. Koreckij T, Nguyen H, Brown LG, Yu EY, Vessella RL, Corey E (2009) Dasatinib inhibits the growth of prostate cancer in bone and provides additional protection from osteolysis. Br J Cancer 101:263–8 128. Park SI, Zhang J, Phillips KA et al (2008) Targeting SRC family kinases inhibits growth and lymph node metastases of prostate cancer in an orthotopic nude mouse model. Cancer Res 68:3323–33 129. Green TP, Fennell M, Whittaker R et al (2009) Preclinical anticancer activity of the potent, oral Src inhibitor AZD0530. Mol Oncol 3:248–61 130. Dong M, Rice L, Lepler S, Pampo C, Siemann DW (2010) Impact of the Src inhibitor saracatinib on the metastatic phenotype of a fibrosarcoma (KHT) tumor model. Anticancer Res 30:4405–13 131. Yang JC, Bai L, Yap S, Gao AC, Kung HJ, Evans CP (2010) Effect of the specific Src family kinase inhibitor saracatinib on osteolytic lesions using the PC-3 bone model. Mol Cancer Ther 9:1629–37 132. Chang YM, Bai L, Liu S, Yang JC, Kung HJ, Evans CP (2008) Src family kinase oncogenic potential and pathways in prostate cancer as revealed by AZD0530. Oncogene 27:6365–75 133. Ammer AG, Kelley LC, Hayes KE et al (2009) Saracatinib impairs head and neck squamous cell carcinoma invasion by disrupting invadopodia function. J Cancer Sci Ther 1:52–61 134. Nautiyal J, Majumder P, Patel BB, Lee FY, Majumdar AP (2009) Src inhibitor dasatinib inhibits growth of breast cancer cells by modulating EGFR signaling. Cancer Lett 283:143–51 135. Rabbani SA, Valentino ML, Arakelian A, Ali S, Boschelli F (2010) SKI-606 (Bosutinib) blocks prostate cancer invasion, growth, and metastasis in vitro and in vivo through regulation of genes involved in cancer growth and skeletal metastasis. Mol Cancer Ther 9:1147–57 136. Vultur A, Buettner R, Kowolik C et al (2008) SKI-606 (bosutinib), a novel Src kinase inhibitor, suppresses migration and invasion of human breast cancer cells. Mol Cancer Ther 7:1185–94 137. Jallal H, Valentino ML, Chen G, Boschelli F, Ali S, Rabbani SA (2007) A Src/Abl kinase inhibitor, SKI-606, blocks breast cancer invasion, growth, and metastasis in vitro and in vivo. Cancer Res 67:1580–8 138. Weis S, Cui J, Barnes L, Cheresh D (2004) Endothelial barrier disruption by VEGF-mediated Src activity potentiates tumor cell extravasation and metastasis. J Cell Biol 167:223–9 139. Kenessey I, Keszthelyi M, Kramer Z et al (2010) Inhibition of c-Met with the specific small molecule tyrosine kinase inhibitor SU11274 decreases growth and metastasis formation of experimental human melanoma. Curr Cancer Drug Targets 10:332–42 140. Wang X, Le P, Liang C et al (2003) Potent and selective inhibitors of the Met [hepatocyte growth factor/scatter factor (HGF/SF) receptor] tyrosine kinase block HGF/SF-induced tumor cell growth and invasion. Mol Cancer Ther 2:1085–92 141. Koon EC, Ma PC, Salgia R et al (2008) Effect of a c-Met-specific, ATP-competitive smallmolecule inhibitor SU11274 on human ovarian carcinoma cell growth, motility, and invasion. Int J Gynecol Cancer 18:976–84 142. Christensen JG, Schreck R, Burrows J et al (2003) A selective small molecule inhibitor of c-Met kinase inhibits c-Met-dependent phenotypes in vitro and exhibits cytoreductive antitumor activity in vivo. Cancer Res 63:7345–55 143. Zou HY, Li Q, Lee JH et al (2007) An orally available small-molecule inhibitor of c-Met, PF-2341066, exhibits cytoreductive antitumor efficacy through antiproliferative and antiangiogenic mechanisms. Cancer Res 67:4408–17 144. Munshi N, Jeay S, Hill J, et al. ARQ 197, a highly selective small molecule inhibitor of c-Met, inhibits invasive and metastatic growth of cancer cells [abstract 2367]. Proceedings of the American Association for Cancer Research Annual Meeting, Los Angeles, CA 2007 145. Li Y, Zhou W, Chen D, et al. Anti-metastatic activity of ARQ 197, a highly selective oral small molecule inhibitor of c-Met, in experimental metastatic models of colon cancer [abstract

194

146. 147. 148. 149.

150.

151. 152. 153.

154.

155.

156.

157. 158.

Y. Dai et al. 2191]. Proceedings of the American Association for Cancer Research Annual Meeting, Los Angeles, CA 2007 Onnis B, Rapisarda A, Melillo G (2009) Development of HIF-1 inhibitors for cancer therapy. J Cell Mol Med 13:2780–6 Rapisarda A, Melillo G (2011) HIF-1 inhibitors for cancer therapy. In: Siemann DW (ed) Tumor Microenvironment, 1st edn. Wiley-Blackwell, UK Golas JM, Lucas J, Etienne C et al (2005) SKI-606, a Src/Abl inhibitor with in vivo activity in colon tumor xenograft models. Cancer Res 65:5358–64 Rucci N, Recchia I, Angelucci A et al (2006) Inhibition of protein kinase c-Src reduces the incidence of breast cancer metastases and increases survival in mice: implications for therapy. J Pharmacol Exp Ther 318:161–72 Nam JS, Ino Y, Sakamoto M, Hirohashi S (2002) Src family kinase inhibitor PP2 restores the E-cadherin/catenin cell adhesion system in human cancer cells and reduces cancer metastasis. Clin Cancer Res 8:2430–6 Nam S, Kim D, Cheng JQ et al (2005) Action of the Src family kinase inhibitor, dasatinib (BMS-354825), on human prostate cancer cells. Cancer Res 65:9185–9 Cecchi F, Rabe DC, Bottaro DP (2010) Targeting the HGF/Met signalling pathway in cancer. Eur J Cancer 46:1260–70 Qian F, Engst S, Yamaguchi K et al (2009) Inhibition of tumor cell growth, invasion, and metastasis by EXEL-2880 (XL880, GSK1363089), a novel inhibitor of HGF and VEGF receptor tyrosine kinases. Cancer Res 69:8009–16 Accornero P, Lattanzio G, Mangano T et al (2008) An in vivo model of Met-driven lymphoma as a tool to explore the therapeutic potential of Met inhibitors. Clin Cancer Res 14:2220–6 Hov H, Holt RU, Ro TB et al (2004) A selective c-met inhibitor blocks an autocrine hepatocyte growth factor growth loop in ANBL-6 cells and prevents migration and adhesion of myeloma cells. Clin Cancer Res 10:6686–94 Schroeder GM, An Y, Cai ZW et al (2009) Discovery of N-(4-(2-Amino-3-chloropyridin-4yloxy)-3-fluorophenyl)-4-ethoxy-1-(4-fluor ophenyl)-2-oxo-1,2-dihydropyridine-3-carboxamide (BMS-777607), a Selective and Orally Efficacious Inhibitor of the Met Kinase Superfamily. J Med Chem 52:1251–4 Gupta GP, Massague J (2006) Cancer metastasis: building a framework. Cell 127:679–95 Zhang XH, Wang Q, Gerald W, Hudis CA, Norton L, Smid M, Foekens JA, Massagué J (2009) Latent bone metastasis in breast cancer tied to Src-dependent survival signals. Cancer Cell 16:67–78

Part II

Homing and Colonization at Distant Sites

Chapter 8

Cellular and Molecular Biology of Cancer Cell Extravasation J. Matthew Barnes and Michael D. Henry

Abstract Hematogenous metastasis involves the entry of cancer cells into the circulation at a primary tumor site and the extravasation of those cells at a secondary organ which may ultimately support the growth of a metastatic tumor. Although extravasation is likely an obligate step in metastasis, it is relatively poorly understood in part due the difficulty studying this process in appropriate experimental models in vivo. Thus, there remain open questions about how cancer cells interact with the vascular wall during this process including the extent to which non-selective mechanisms such as size restriction versus specific adhesive interactions determine the behavior of extravasating cancer cells; how cancer cells cross the endothelium; and the degree to which extravasation limits the overall efficiency of metastasis. The answers to these questions are influenced by specific properties of both the cancer cells and the nature of the vascular bed involved. In this Chapter, we review our current understanding of the cellular and molecular biology of cancer cell extravasation and discuss how this knowledge impacts clinical issues related to the biology of circulating tumor cells and cancer therapy.

Abbreviations CTC DTC EMT

Circulating tumor cell Disseminated tumor cell Epithelial-to-mesenchymal transition

J.M. Barnes • M.D. Henry (*) Departments of Molecular Physiology and Biophysics and Pathology, Holden Comprehensive Cancer Center, University of Iowa Carver College of Medicine, Iowa City, IA, USA e-mail: [email protected]

A. Fatatis (ed.), Signaling Pathways and Molecular Mediators in Metastasis, DOI 10.1007/978-94-007-2558-4_8, © Springer Science+Business Media B.V. 2012

197

198

J.M. Barnes and M.D. Henry

EpCAM MET SCID TEM

8.1

Epithelial cell adhesion molecule Mesenchymal-to-epithelial transition Severe and combined immune deficient Trans-endothelial migration

Introduction

Metastasis, or the spread of cancer cells from a primary tumor to a distant tissue, is associated with the lethal consequences of cancer progression. Metastasis is comprised of a series of events, beginning with local tissue invasion, followed by entrance into the bloodstream or lymphatic system, survival within circulation, extravasation (exit from the vasculature), and finally colonization of and proliferation at a secondary organ site. The focus of this chapter is on cancer cell extravasation, a topic which remains poorly understood in part because tracking the fate of individual cancer cells from the circulation into the vascular beds of various organs remains a technically daunting task. Nonetheless both in vivo and ex vivo analyses have provided insight into the cellular and molecular mechanism of cancer cell extravasation. Although leukocyte diapedesis has provided an instructive paradigm for understanding cancer cell extravasation, the latter exhibits numerous differences and it is likely that the mechanism of cancer cell extravasation is both cell-type and organ-specific. Another aspect of this process which remains a matter of some debate is the extent to which metastatic efficiency is limited by cancer cell extravasation. Here, we discuss evidence from both sides of this issue. Finally, we discuss the relationship between epithelial-to-mesenchymal transition (EMT) and cancer cell extravasation. A growing body of evidence indicates that EMT is one mechanism driving the invasive phenotype of various cancers, but the role of EMT in extravasation has been unclear. Recent data suggests that EMT phenotype augments the ability of prostate cancer cells to extravasate, in part through modulating interactions between cancer cells and endothelial cells. A better understanding of the connection between EMT and extravasation capability may have an impact on the analysis of circulating tumor cells, as well as provide potential therapeutic targets to forestall metastatic disease.

8.2

Mechanisms of Cancer Cell Extravasation

In order for circulating tumor cells (CTCs) to form metastatic tumors, they must exit the bloodstream and enter the parenchyma of a secondary tissue. Extravasation involves passage of cells across the endothelium, termed transendothelial migration (TEM), as well as transit through the subendothelial basal lamina. Several mechanisms of cancer cell extravasation have been reported, and this variation may be due in part to differences in experimental models. Results from many groups support a model by which cancer cells adopt certain leukocyte-like properties allowing them

8 Cellular and Molecular Biology of Cancer Cell Extravasation

199

to adhere to endothelial cells by specific and dynamic cell surface interactions, thereby facilitating passage between or through the endothelial cells and through the underlying basement membrane. Others report static, nonselective mechanisms of cancer cell adhesion followed by either TEM, often involving retraction of endothelial cells, or proliferation within the lumen of the vessel.

8.2.1

Leukocyte Versus Cancer Cell Extravasation

Leukocyte extravasation is a highly regulated process which allows immune cells to home to lymphatic and other tissues throughout the body. The cellular and molecular biology of leukocyte extravasation is generally well understood and has been reviewed recently [1–3]. Each step of leukocyte extravasation is mediated by specific combinatorial interactions between a receptor and a ligand. Temporal and spatial regulation of some of these molecules dictates the selectivity of a leukocyte for a given tissue. Selectins are surface glycoproteins expressed on the endothelium which initiate rolling of leukocytes by binding to various selectin ligands on the surface of these cells. While rolling, leukocytes encounter a gradient of chemokines presented by endothelial cells. Subsequent activation of chemokine receptor signal transduction facilitates changes in the affinity and avidity of leukocyte integrins for their endothelial ligands, the intercellular adhesion molecules (ICAMs). This results in integrin-mediated firm adhesion of leukocytes which is followed by diapedesis. Cancer cells of various histological origins have been shown to express cell surface proteins known to be important in leukocyte extravasation, including cell adhesion and chemokine receptors which interact with corresponding ligands at the luminal surface of the endothelium, and this has been implicated in the organ tropism of metastasis as recently reviewed [4, 5]. As an example, bone marrow endothelial cells constitutively express the chemokine SDF-1 and the adhesion molecules VCAM-1 and E-selectin, which are used to facilitate adhesion of circulating leukocytes [6]. Numerous reports indicate that a variety of cancers express the SDF-1 receptor, CXCR4 [7]. Antibody neutralization of CXCR4 has been shown to greatly reduce metastatic colonization of breast cancer cells in mice [8]. In a panel of prostate cancer cells, CXCR4 expression levels were shown to correlate with metastatic aggressiveness, and targeting this receptor with a small molecule, T140, reduced chemotactic invasion of PC-3 cells towards SDF-1 [9]. Extravasation of PC-3 cells has also been shown to be facilitated by expression of monocyte chemoattractant protein-1 on bone marrow endothelial cells [10]. In vitro-derived bonetropic clones of the MDA.MB.231 cell line have been shown to overexpress CXCR4 as compared to the heterogeneous parent [11]. Recently Jamieson et al. detected overexpression of the monocyte chemokine receptor, CX3CR1, in human prostate cancer tissue samples and show that its ligand, fractalkine, is expressed on the plasma membrane of osteoblasts, mesenchymal, and other stromal cells of the bone marrow [12]. Furthermore, dihydrotestosterone induces cleavage of stromal, but not endothelial plasma membrane-associated fractalkine; thus establishing a soluble chemotactic gradient promoting prostate cancer cell homing to bone and ultimately

200

J.M. Barnes and M.D. Henry

adhesion to the bone marrow endothelium [12]. Forced expression of the E-selectin ligand, Sialyl Lewis X, in PC-3 prostate cancer cells has been shown to induce cell rolling and adhesion to IL-1b-stimulated bone marrow endothelial cells in vitro and this interaction is blocked with anti-E-selectin antibody [13]. These and other studies support a role for leukocyte extravasation molecules and activated endothelial cells in the bone tropism of cancers such as breast and prostate. However, chemokine signaling may play roles relevant to metastasis beyond extravasation [14, 15]. On the other hand, N-cadherin is an example of an adhesion molecule implicated in cancer cell extravasation, which is not known to be involved in leukocyte diapedesis. N-cadherin is well positioned to be involved in the association between cancer cells and endothelial cells as it is present on the luminal surface of endothelia and not enriched at cell junctions [16]. Melanoma cells have been shown to adhere to endothelial cells via homophilic N-cadherin binding which leads to retraction of the VE-cadherin and PECAM junctions between neighboring endothelial cells [17]. Engagement of melanoma cell N-cadherin leads to transactivation of Src kinase and nuclear localization of b-catenin, which facilitate the formation of pseudopodia into the disrupted endothelial cell junction and subsequent TEM to the basal side of the endothelium [18, 19]. Similarly, siRNA-mediated knockdown of N-cadherin in PC-3 prostate cancer cells was shown to reduce TEM [20]. Although the signaling consequences of N-cadherin binding were not addressed here, PC-3 TEM has been shown to involve Rac1-mediated invadapodia formation [10, 21]. Most of the in vitro studies of extravasation discussed above are performed by seeding cancer cells onto a monolayer of endothelial cells growing on a semipermeable transwell insert or on a layer of extracellular matrix proteins. While this provides a convenient model for extravasation, tumor cells in vivo are in the presence of flowing blood cells. To more accurately mimic physiological conditions, dynamic in vitro models of extravasation have been used in which mixtures of cancer cells and leukocytes are subjected to flow through chambers lined with endothelial monolayers. Using this approach, evidence has been provided for cooperation between leukocytes and cancer cells in extravasation. Strell et al. showed that breast cancer cells, which lack b2-integrin but express its ligand ICAM-1, bind to b2-expressing neutrophils and this hetero-cellular complex then binds to the endothelial cells via ICAM-1 [22]. These results are in agreement with other reports of neutrophilassisted extravasation of breast cancer cells [23] and melanoma cells [24]. Platelets have also been shown to bind to cancer cells of various types, often through selectin-mediated interactions, leading to formation of emboli which exhibit enhanced adhesion to the endothelium [25–28]. It may be the case that certain cancer cells are actually dependent upon leukocyte and/or platelet interactions in order to form metastatic tumors, which is interesting considering the growing understanding of the role of inflammation in cancer progression and metastasis [29]. Although leukocytes provide an attractive mechanistic paradigm for understanding cancer cell extravasation, the selective pressures which would drive epithelial cells to acquire detailed leukocyte gene expression patterns and cellular behavior are not obvious. As discussed in the following sections, cancer cell extravasation does not follow a strict paradigm and has been reported to occur by a number of mechanisms,

8 Cellular and Molecular Biology of Cancer Cell Extravasation

201

which appear to be both cell type- and organ-specific. One feature of cancer cell extravasation, observed primarily in vitro, is that cancer cells induce the retraction of endothelial cells, unlike leukocytes [30]. There are even alternatives to classical extravasation, such as expansive intraluminal proliferation, which have been proposed as a means for metastatic colonization.

8.2.2

Cancer Cell Adhesion to the Endothelium in Extravasation

Many studies on extravasation have focused on the mechanisms of tumor cell adhesion to the endothelium, as this is presumed to be the initiating event in this process. Here, intravital videomicroscopy and detailed accounting of cancer cell fate following systemic injection have provided important insights into the process of extravasation in vivo. These studies are summarized in Table 8.1. Two general mechanisms are thought to be involved: (1) non-selective and passive lodgment of cancer cells in the microvasculature by size restriction and (2) selective adhesion between cancer cells and the vasculature of particular organ sites, including larger diameter vessels. Although these mechanisms are often regarded to be in conflict, both size restriction and selective adhesion could influence the overall process of extravasation. Determining the relative contribution of these mechanisms will be important in potential therapeutic contexts.

8.2.2.1

Mechanical Arrest of Circulating Tumor Cells in the Microvasculature

Ewing first proposed that cancer cells become entrapped in the first capillary beds they encounter; thus contributing to organ-site preference for metastasis [31]. Intuitively, this mechanism derives from the fact that epithelial-derived cancer cells are typically larger than the largest of the white blood cells (human monocytes range from about 12–20 mm in diameter [32]). Although CTCs are deformable and may thus be able to negotiate this restrictive barrier to some degree, they may not be as successful as leukocytes [33, 34]. CTCs would thus seem susceptible to size restriction when encountering the microvasculature (human capillaries are as small as 7 mm in diameter [35]). Indeed, it has long been appreciated in experimental mouse models of metastasis that cancer cells injected into the tail vein accumulate predominantly in the lung [36]. Here, this is demonstrated utilizing bioluminescence imaging (Figs. 8.1 and 8.2). Early efforts aimed at observing the behavior of CTCs in the microvasculature of the rabbit ear described the association of cancer cells with the endothelium and subsequent extravasation, but did not conclude that size restriction was the sole means of cancer cell arrest at this site [37]. However, the pioneering work of Ann Chambers, Alan Groom, and colleagues utilizing in vivo video microscopy have provided abundant support for the concept of size restriction in the lung and other organs [38–43].

FVB/N mouse

Nude mouse

Nude and SCID mouse SD and Nude rat

C57BL/6 mouse

SD rat

C57BL/6 mouse

Human breast

Human lung and melanoma

Human breast

Mouse melanoma

Human and rat colon

Mouse mammary

Human colon

Host FVB/N mouse

Cell type Human breast

Mesenteric vein

Intra-arterial

Tail vein

Intra-arterial

Tail vein

Carotid artery

Spleen

Injection route Tail vein

Liver

Liver

Lung

Liver

Lung

Brain

Liver

Organ site Lung

Table 8.1 Selected references reporting cancer cell extravasation in vivo Observation technique and timeline 2PM –resected and perfused tissue At 2 h, 2% of cells had extravasated At 24 h, 22% of cells had extravasated 2PM – resected and perfused tissue At 2 h, 4% of cells had extravasated At 24 h, 56% of cells had extravasated MPLSM – living animals Extravasation occurred between days 1–9 >36% lung cancer cells extravasated >45% melanoma cells extravasated 2PM – resected and perfused tissue Extensive extravasation at 48 h IVVM – living animals 5–30 min (peak at 20 min) ESCM – fixed and embedded tissue At 5 h, nearly all cells had extravasated IVVM Most cells had extravasated at 30 min IVVM – living animals At day 3, >80% of cells had extravasated Size restriction

Specific adhesion

Not directly addressed

Specific adhesion

Not directly addressed

Size restriction

Specific adhesion

Primary method of CTC arrest Specific adhesion

Luzzi [41]

Haier [70]

Voura [17]

Schluter [51]

Gupta [78]

Kienast [44]

Martin [69]

Reference Martin [69]

202 J.M. Barnes and M.D. Henry

Chick embryo CAM

Nude mouse

Chick embryo CAM

Mouse melanoma

Mouse mammary

Mouse melanoma

CAM vein

Mesenteric vein

CAM vein

Injection route

CAM

Liver

CAM

Organ site IVVM Extravasation started at 2 h At 24 h, >80% of cells had extravasated IVVM – living animals At day 1, 51–63% of cells had extravasated At day 21, 97–100% of cells had extravasated IVVM After 8 h, 83% of cells had extravasated

Observation technique and timeline

Size restriction

Size restriction

Size restriction

Primary method of CTC arrest

Chambers [43]

Morris [39]

Koop [40]

Reference

MPLSM multi-photon laser scanning microscopy, 2PM two-photon-microscopy, IVVM in vivo videomicroscopy, ESCM emission scanning confocal microscopy, FM fluorescence microscopy, REF rat embryonic fibroblast, SCID severe and combined immune deficient, SD Sprague-Dawley, CAM chorioallantoic membrane

Host

Cell type

8 Cellular and Molecular Biology of Cancer Cell Extravasation 203

204

J.M. Barnes and M.D. Henry

Fig. 8.1 Long-term bioluminescence imaging of tail vein injected cancer cells. A luciferase expressing metastatic derivative of PC-3 prostate cancer cells was injected systemically into SCID mice via the tail vein route. In an image taken minutes after injection, these cells are observed arresting in the lung (a). At 2 months post-injection, colonization of various tissues is observed (b, dorsal image on top, ventral image on bottom). Note that not all mice develop tumors in the region of the lungs, despite their original arrest in this organ

8 Cellular and Molecular Biology of Cancer Cell Extravasation

205

Fig. 8.2 Short-term serial bioluminescence imaging of tail vein injected cancer cells. Luciferase expressing B16.F10 melanoma cancer cells were injected systemically into SCID mice via the tail vein route. At the indicated time points after injection, bioluminescence images were obtained. Note that by 48 h, nearly all of the signal seen at 4 h is gone, indicating clearance of cells from the lung vasculature. Between 120 and 168 h, bioluminescence signal steadily increases, indicating outgrowth of lung-colonized cancer cells. Graphical quantification of the bioluminescence images is shown at the bottom of the figure

Recently, monitoring circulating lung carcinoma and melanoma cells in the brains of living mice via multi-photon microscopy over an extended time course, Kienast et al. have stunningly visualized detailed, individual steps of extravasation [44]. Minutes after injection, all adherent cells were reported to be arrested by sizerestriction in vessels of similar diameter than the cancer cells or in vascular branch points. Some cells were observed dislodging from their original point of arrest only to arrest in another microvascular branch point for periods of up to 48 h. Both lung carcinoma and melanoma cells were observed extravasating into brain parenchyma at days 1–9 after injection. The process of TEM involved extension of invadopodia through the endothelium as described in the PC-3 and melanoma cell lines in

206

J.M. Barnes and M.D. Henry

Sect. 8.2.1. Intriguingly, intraluminal proliferation of lung carcinoma cells was only observed when host mice were treated with anti-VEGF therapy, indicating a critical role for neoangiogenesis in the metastatic colonization by this cell line. This finding is consistent with reports that VEGF production by cancer cells facilitates extravasation by activating Src-family kinases in endothelial cells, leading to disruption of their VE-cadherin junctions [45]. Although intraluminal proliferation was observed, colonies of cells were never witnessed rupturing through the vascular wall and often would regress [44]. It is possible that the architecture of the brain vasculature is nonpermissive to this type of metastatic colonization (differences in organ vasculature are the focus of Sect. 8.3).

8.2.2.2

Specific Cancer-Endothelial Cell Interactions

Although it can be experimentally demonstrated that the majority of cancer cells injected into the tail vein are retained in the lung following injection, extrapulmonary tumors do develop in some tumor models, as demonstrated in Fig. 8.1. Glinskii et al. show adherent DU-145 prostate cancer cells in the microvasculature of the lung, vertebrae, sternum, kidney, and thyroid gland after tail vein injection. Surprisingly, nearly equal numbers of cells were quantified in the lung and vertebrae [46]. Furthermore, treating Du-145 cells with blocking antibodies recognizing either Thomas Friedenreich (TF) antigen (expressed on DU-145 cells) or its receptor, galectin-3 (expressed on the surface of endothelial cells), greatly reduced CTC arrest in all organs. Treatment with E, P, or L-Selectin blocking antibodies, on the other hand, did not significantly affect tumor cell adhesion [46]. b3 integrins have been implicated in cancer cell transendothelial migration, although these are likely to engage ligands in the subendothelial matrix [47, 48]. Wang et al. found that blocking a3 and b1 integrin subunits on human fibrosarcoma cells leads to significant reduction of CTC arrest in the lung vasculature following renal injection in mice [49]. Similar results were obtained when infusing blocking antibodies to laminin-332 into mice prior to cancer cell injection. It was concluded that there are small, naturally occurring foci of exposed basement membrane throughout the lung vasculature, exposing laminin-332 to integrins on CTCs, facilitating adhesion [49]. However laminin-332 is not normally a component of the subendothelial basement membrane [50], so details of this mechanism remain to be explained. To determine if metastatic potential correlates with the ability of cancer cells to adhere to and extravasate through the endothelium, colon cancer cell lines of low, medium, and high aggressiveness were monitored via intravital fluorescence microscopy following systemic injection into Sprague-Dawley rats [51]. Regardless of metastatic potential, each cell line bound efficiently to the endothelium of the two most common tissues of colon cancer metastasis, the liver and lung. In less common sites of colon cancer metastasis, including kidney, intestine, skin, muscle, and spleen these cell lines all adhered poorly. Metastatic potential did, however, correlate with the frequency of extravasation and tissue invasion in the lung and liver. Surprisingly, mechanical lodging of cancer cells in the microvasculature was rare, and most adherent cells did not block blood flow through the lumen of the vessel. Schluter

8 Cellular and Molecular Biology of Cancer Cell Extravasation

207

et al. did not observe rolling of cancer cells on the endothelium prior to adhesion in vivo, although colon cancer cells have been observed rolling on endothelial cells in vitro under laminar flow [52]. These data suggest that adhesion to the endothelium occurs through cell- and organ-specific interactions and further molecular characterization may help explain the tropism of metastatic colon cancer to the liver and lung. Importantly, these data support the argument that metastatic tumor formation reflects the efficiency of extravasation [51]. Ex vivo fluorescence imaging of isolated perfused lung preparations from mice injected systemically with human and rat fibrosarcoma cells revealed attachment of cancer cells to the endothelial walls of the pre-capillary arterioles, which are much larger in diameter than the cancer cells themselves. Furthermore, these arterioles remained perfusable, showing that the cancer cells had not occluded the lumens of the microvasculature. These data provide support for a model in which specific interactions between cancer and endothelial cells facilitate adhesion [53]. An unexpected finding of this work was that adherent fibrosarcoma cells rarely extravasated and that most viable cells began to proliferate within the lumen of the blood vessel, leading to colony formation which caused enlargement of the capillaries. Additional support for intravascular proliferation was obtained using this same imaging approach with lungs from mice with subcutaneous breast cancer tumors [54]. These experimental data support a longstanding hypothesis that intraluminal proliferation and subsequent colony growth through the vessel wall can lead to metastatic colonization in humans [55–57].

8.3

Cancer Cell Extravasation and Vascular Architecture of Common Sites of Metastasis

The tendency of primary cancers to metastasize to certain organs was first observed by Paget in the late 1800s [58]. The ability of CTCs to colonize some tissues and not others has since been referred to as the “seed and soil” hypothesis, describing the metastatic compatibility of a cancer cell with a given tissue. Because it is assumed that CTCs must first extravasate in order to colonize a tissue, observed differences in tumor tropism may be due to the ability of these cells to extravasate. One factor which may contribute to site-variable tumor cell extravasation efficiency is differences in organ vascular structure. Here, we focus on four organs which are common target tissues of metastatic carcinoma: bone, brain, liver, and lung. This subject has recently been reviewed in brief [59].

8.3.1

Bone Marrow Microvasculature

The bone marrow consists of two compartments, the hematopoietic compartment, which produces blood cells, and the stroma which contains the vasculature and its supporting fibroblasts, as well as adipocytes and nerves. The bone marrow

208

J.M. Barnes and M.D. Henry

microvascular endothelial cells are fenestrated and arranged in discontinuous sinusoids with little to no basement membrane and are thus surrounded only by hematopoietic tissue [6]. Functionally, this allows easy passage of blood cells into and out of the bone marrow sinusoids. While it is possible that CTCs which are poor at TEM may favor colonization of bone due its vascular structure, there is molecular evidence for bone-specific adhesion and extravasation (discussed above) which may help explain why bone is a common site of metastasis for several cancers. In prostate cancer, the lumbar spine is a common site of metastasis which reflects the fact that this site encounters high numbers of cancer cells draining through the Batson’s venous plexus structure in the pelvic floor [60]. On a clinical note, the detection of disseminated prostate, breast, and colon tumor cells in bone marrow may have important prognostic and predictive utility [61].

8.3.2

Brain Microvasculature

At the microvascular level, the most unique architectural feature of the brain vasculature is the blood-brain-barrier (BBB). The BBB is formed as a consequence of specialized tight junctions between brain endothelial cells; the result is markedly low passive diffusion of polar solutes and macromolecules, making the BBB highly selective to gasses, nutrients, and drugs [62]. This vascular structure would certainly seem to pose a challenge to CTC extravasation. Indeed, it is known that leukocyte extravasation into brain parenchyma is reduced compared to other tissues [4, 62]. Furthermore, it has been shown that when monocytes traverse the brain vasculature, they do so by a specialized trans-cellular route (through the endothelial cell as opposed to common monocyte diapedesis between endothelial cells), thus leaving the BBB tight junctions intact [63]. The fidelity of the BBB thus remains high except for conditions of trauma, ischemia, and various disease states which can cause leakiness or breakdown of the BBB, resulting in increased transport of macromolecules and leukocytes into brain parenchyma. In light of this information, it is surprising that the brain is a common site for metastatic colonization by several cancers. Breast cancer metastasis to brain is common and has been reviewed [64]. It is possible that cancer cells utilize unique mechanisms which facilitate brain extravasation. In vitro TEM assays using BBB-forming brain microvascular endothelial cells (BMECs) have shown that MDA.MB.231 breast cancer cells are able to disrupt BBB adherens junctions and transmigrate through the monolayer. This process was shown to be enhanced by exogenous addition of VEGF, which led to increased cytoskeletal rearrangement and permeability of BMECs [65]. In support of these findings, brain metastatic derivatives of MDA. MB.231 cells were shown to overexpress VEGF-A as well as the angiogenic interleukin IL-8 [66]. Finally, oral administration of a VEGF receptor tyrosine kinase inhibitor, PTK787, led to a large reduction of brain tumor formation in mice [66]. These studies, along with those outlined in Sect. 8.2.2.1, highlight an important role for angiogenesis and VEGF signaling in brain metastasis. Many conventional therapeutic strategies to combat brain metastasis are excluded by the BBB. It is tempting

8 Cellular and Molecular Biology of Cancer Cell Extravasation

209

to speculate that VEGF signaling has a direct impact on the extravasation of CTCs in the brain. VEGF targeting may prevent extravasation of CTCs that have potential to form brain metastases and may improve patient survival.

8.3.3

Liver Microvasculature

The liver is unique in that it has both an arterial blood supply from the aorta as well as a portal venous supply, carrying blood from the gastrointestinal tract. These blood sources both drain into specialized liver sinusoids which deliver blood to hepatocytes and eventually to the large hepatic vein. These sinusoids are composed of extensively fenestrated endothelial cells and phagocytic Kupffer cells with no basal lamina [67]. Functionally, the sinusoids act as a sieve, allowing free passage of large macromolecules and blood cells. The fenestrations in hepatic circulation are both maintained and induced by VEGF signaling [67], which has been shown to augment extravasation and metastatic colonization of both the liver and of the lung, an organ which lacks fenestrations [45]. The liver undergoes constitutive leukocyte trafficking, and extravasation appears to occur in the sinusoids more often than in post-capillary venules [68]. It is thus believed by many that the discontinuous, fenestrated hepatic microcirculation is more permissive to tumor colonization than the vasculature of other organs [4]. In support of this hypothesis, Martin et al. showed that the same breast cancer cell line extravasates at a higher rate and frequency in the liver than in the lung [69]. The liver is the most common site of colon and pancreatic cancer metastasis. Here the liver is the first organ encountered by cancer cells via portal circulation. As discussed in Sect. 8.2.2.2 data from several labs have demonstrated adhesion of CTCs at pre-sinusoidal venules and in sinusoids of larger diameter than cancer cells, suggesting that colon cancer cell extravasation within the liver is initiated by specific interactions [51, 70]. Similarly, B16F1 melanoma cells injected into the portal veins of IL-1a-treated mice exhibited reduced velocity and increased retention in the liver vasculature compared to control mice, suggestive of cell rolling and adhesion to the activated endothelium [71]. Consistent with these findings are various studies showing that E-selectin ligands, expressed on colorectal cancer and melanoma cells, play an important role in CTC arrest and metastatic extravasation [52, 72]. Additionally, Rosenow et al. showed that extracellular matrix collagens and laminins, exposed in the numerous liver endothelial fenestrations, interact with integrin-a2b1 on colorectal cancer cells facilitating tumor cell adhesion and extravasation. Furthermore, blockage of this interaction with lebein-1 and rhodocetin reduced tumor formation [73].

8.3.4

Lung Microvasculature

The lung is a common metastatic site for various cancers, including breast, colon, and melanoma, and is one of the most studied organs in experimental metastasis. The microvasculature within the lung is smaller than many organs, with diameters as small

210

J.M. Barnes and M.D. Henry

as 7 mm [35], similar to the diameter of a red blood cell. The pulmonary capillaries are arranged in a dense, sheet-like network which covers the surface of alveoli, facilitating efficient gas exchange. These capillaries are composed of continuous, non-fenestrated endothelial cells attached firmly to a thin basement membrane [68]. The lung is an active site of leukocyte trafficking and extravasation, and the concentration of leukocytes in pulmonary blood is as much as 100-fold greater than in systemic blood [4, 35]. Leukocyte adhesion and extravasation in the pulmonary vasculature is thought to occur most frequently at the level of the capillaries, whereas in most systemic tissues this occurs in post-capillary venules [68, 74]. Furthermore, leukocyte adhesion in pulmonary capillaries appears to occur by size restriction and is not preceded by a rolling step as occurs in other tissues [75]. Once adherent, specific molecular interactions involving CD11/18, ICAM-1, and VCAM have been reported to be important in maintenance of adhesion and in leukocyte infiltration of lung parenchyma [76]. Evidence exists for specific molecular interactions between CTCs and the lung endothelium, which contribute to the observed lung tropism of some cancers. Using a phage-display library, Brown and Ruoslahti identified a surface protein, metadherin, which targets phage specifically to lung microvasculature [77]. After producing an antibody to metadherin, it was shown that this protein is upregulated in breast cancer samples when compared to healthy tissue. Both metadherin-neutralizing antibody and metadherin-siRNA treatment of metastatic 4T1 mammary carcinoma cells significantly reduced lung metastases in syngeneic hosts [77]. Minn et al. isolated a lung-tropic subpopulation of MDA.MB.231 breast carcinoma, named LM2, by serial tail vein passage in mice. RNA from this cell line and the parent were subjected to comparative microarray analysis, which revealed a lung-metastatic gene signature of 54 transcripts [11]. A follow-up to this work showed that four of the genes within this signature, including epiregulin, matrix metalloprotease 1 and 2, and COX2, contribute to vascular remodeling in the primary tumor, which promotes both tumor growth and dissemination of LM2 cells [78]. When these genes were knocked down simultaneously in LM2 cells, pulmonary extravasation and colonization were greatly reduced. Furthermore, small molecule inhibition of these four gene products led to reduced primary tumor growth as well as reduced pulmonary extravasation and tumor formation. These findings support a genetic basis for lung tropism and suggest that mutations acquired during tumorigenesis are co-opted for primary tumor growth as well as invasion, extravasation, and metastatic tumor formation [78] In support of this hypothesis, TGFb signaling at the site of a primary breast cancer tumor has been shown to induce production of angiopoietin-like 4 in invading cells, which facilitates retention of CTCs in the lungs of mice, but not bone, and enhances TEM in vitro [79].

8.4

Metastatic Inefficiency and Extravasation

As mentioned above, early studies on the fate of circulating tumor cells led to the observation that after systemic injection, cells became arrested at the first capillary bed encountered [36, 80]. For example, cells injected via the tail vein would arrest in the

8 Cellular and Molecular Biology of Cancer Cell Extravasation

211

lung, and those injected into the hepatic portal vein would arrest in the liver. A sharp drop in cell viability in these experiments occurred immediately after injection, with as many as 99% of cells dying within 24 h (see Fig. 8.2). It was concluded that the majority of these cells died due to hemodynamic shear stress and by mechanical rupture following size restriction in the microvasculature [81]. Of the small percentage of surviving cells, fewer yet gave rise to metastatic lesions. Thus, great emphasis was placed on the ability of CTCs to extravasate and subsequently proliferate in order to successfully colonize a tissue. These observations introduced the concept of metastatic inefficiency, for which there is direct clinical evidence [82]. To date, debate continues over which steps of metastasis contribute most to the inefficiency of this process. Koop et al. took a quantitative approach to measure the rate limiting steps of metastasis by adding inert microspheres to suspensions of B16 melanoma cells and injecting these systemically into chick embryos. The cell:sphere ratio was monitored over time to determine the percentage of viable cells at each step of metastasis. Twelve to 24 h post injection 85% of the melanoma cells survived circulation and had extravasated [40]. Strikingly similar results were recapitulated by using the same method of cell accounting in a mouse model following injection into the mesenteric vein [41]. Interestingly, this latter study showed that although the majority of injected cells extravasated, only 0.07% led to micrometastases and 0.02% gave rise to tumors. Because B16F10 cells are a highly aggressive cell line, the question of whether less metastatic cell lines could extravasaste was addressed. Using the chick chorioallantoic membrane model, ras-transformed or wild type NIH/3T3 cells, as well as wild type mouse embryonic fibrobalasts, were compared for their extravasation potential. No differences were found between the three cell lines, as all were able to successfully extravasate, but only the transformed cells were capable of tumor formation [42]. These are among the seminal reports arguing that although extravasation is necessary for metastatic tumor formation, it is an efficient step and not related to the overall aggressiveness of the cancer cell. These findings supported the idea that the ability to survive and proliferate in the parenchyma of a secondary tissue ultimately dictates the efficiency of metastasis. Recent support for the hypothesis that late steps of metastasis are the predominant contributors to metastatic inefficiency was provided by Podsypanina et al. Mice were injected systemically with primary mouse mammary cells containing a doxycycline-dependent promoter driving the expression of the oncogenes MYC and KrasD12. At 1.5, 8, and 17 weeks post-injection, mice were placed on either a control or doxycycline diet and in all cases, pulmonary tumor growth was observed in the doxycyline-fed animals using bioluminescence imaging. Intriguingly, although no tumors formed in the control-diet mice, histological analysis of lung tissue showed that in the absence of oncogene expression, these control cells had colonized the lung tissue, were viable, and mitotically active. These findings suggest that nontumorigenic cells in circulation are capable of extravasation and colonization but secondary tumor growth requires oncogenic transformation [83]. In contrast, recent studies demonstrate that extravasation of a particular breast cancer cell type occurs at different rates in lung versus liver. Six hours after injection, two-photon microscopy revealed that most cells were arrested in the lumen of the microvasculature in both organs. However, at 24 h post-injection, over half of the

212

J.M. Barnes and M.D. Henry

cancer cells had extravasated into liver parenchyma, whereas less than one quarter had in the lung [69]. Thus, experimental determination of the extent to which extravasation contributes to metastatic inefficiency may depend on: (1) the cancer cell type involved, as melanoma cells appear to be an overall more aggressive lung-metastatic cell than many carcinomas, (2) the organs in which extravasation is being studied, as structural differences at the vascular level likely have a large impact on the permissiveness of tumor cell adhesion and extravasation, and (3) the imaging technique being used to qualify extravasation, as even with high resolution fluorescence imaging, delineation of intravascular versus extravascular location of cancer cells can be very challenging, especially in capillary-dense tissues such as the lung.

8.5

Epithelial-to-Mesenchymal Transition and Cancer Cell Extravasation

EMT is a process by which a polarized epithelial cell undergoes morphological changes resulting in a mesenchymal phenotype. EMT is characterized by loss of functional epithelial markers, such as E-cadherin and cytokeratin, and gain of mesenchymal markers such as N-cadherin and vimentin. EMT is often accompanied with large changes in the production of extracellular matrix and basement membrane proteins, and increases in protease secretion. These changes result in loss of epithelial polarity and the acquisition of migratory and invasive characteristics. This process is highly regulated both spatially and temporally during embryogenesis [84] and also plays a role in the processes of inflammation and wound healing [85]. Recent evidence has shown that EMT contributes to the progression of solid tumors by permitting detachment of cells from their primary site and by inducing a migratory phenotype, allowing the cell to invade local tissue and enter the lymphatic system or bloodstream [86]. Despite a large number of publications supporting a role for EMT in tumor progression and metastasis, information linking EMT and cancer cell extravasation is lacking. To identify novel genes involved in cancer cell extravasation, Drake et al. serially passaged PC-3 cells through a monolayer of human lung endothelial cells seeded onto transwell inserts and isolated a subpopulation named TEM4–18. Compared to the PC-3 input, these cells were enhanced in TEM and were shown to colonize tissues at a significantly higher rate in SCID mice. Furthermore, TEM4–18 cells exhibited an EMT-like morphology and near complete loss of E-cadherin mRNA and protein. Microarray analysis showed upregulation of a zinc-finger homeodomain transcription factor, Zeb1, in TEM4–18 cells [20]. Zeb1 had previously been implicated in EMT in development and pathological contexts [ 87, 88]. shRNA-mediated knockdown of Zeb1 in TEM4–18 cells resulted in a partial reversal of EMT, including restoration of E-cadherin and epithelial morphology, and reduction of TEM [20]. These data provide evidence that repressors of an epithelial phenotype, such as Zeb1, facilitate extravasation of prostate cancer cells.

8 Cellular and Molecular Biology of Cancer Cell Extravasation

213

However, other studies suggest that cells bearing epithelial characteristics are capable of metastasis. To study the role of EMT in individual steps of metastasis in vivo, Tsuji et al. compared the invasive and metastatic properties of normal epithelial and EMT (p12CDK2-AP1-driven) hamster cheek pouch carcinoma cells. Both cell lines formed subcutaneous tumors at a similar rate, but only the EMT cells invaded local tissue and could be detected in the blood, consistent with previous reports of EMT-driven invasion [89]. When these cell lines were injected intravenously into mice, only the wild-type epithelial cells gave rise to lung tumors, suggesting that the EMT cells are incapable of metastatic colonization. When the two cell lines were differentially labeled, mixed, and injected subcutaneously, lung tumors formed which only consisted of wild type cells and, although EMT cell DNA was readily detectable in blood, none was detected in lung tumors. These data suggest that EMT cells are responsible for early events in metastasis, namely invasion and intravasation, and allow neighboring epithelial tumor cells to enter the circulation. During later steps in this model, it appears that only the epithelial cells are able to extravasate and proliferate at a secondary tissue, although extravasation, per se was not evaluated. These studies are in agreement with reports that epithelial-like MDA.MB.231 cells are enhanced in metastasis from a subcutaneous site when mixed with mesenchymal stem cells than on their own but that the resulting metastatic tumors do not contain detectable mesenchymal cells [90]. As described in Sect. 8.4, the work by Podsypanina et al. provides evidence that epithelial cells are capable of lung colonization [83], although it is difficult to ascertain whether these cells had, in fact, extravasated in the lung microvasculature. It is possible that epithelial cells have a propensity for intravascular colonization and proliferation, while the EMT phenotype facilitates trans-endothelial migration.

8.6

Clinical Aspects of Cancer Cell Extravasation

A more thorough understanding of cancer cell extravasation is likely to impact efforts to better predict the likelihood of occult metastasis and perhaps to inhibit metastasis. One area relates to the biology of CTCs. By definition, CTCs represent the precursor to extravasating cancer cells, and as the forgoing discussion illuminates, their biology is linked. The isolation and quantification of CTCs provides promise as a prognostic and predictive tool in the clinic. Early CTC studies successfully showed correlation of CTC number with poor prognosis in breast cancer patients [91, 92]. Longitudinal monitoring of CTCs during therapy of metastatic breast cancer patients was also shown to be prognostic of progression-free and overall survival [92]. These findings encourage clinicians that CTC quantification may be able to replace testing of disseminated tumor cells (DTCs). DTCs have been identified in the bone marrow of patients with breast, colon, and prostate cancer and have been shown to correlate with poor prognosis and cancer progression [93, 94], however isolation of bone marrow is painful and complicated

214

J.M. Barnes and M.D. Henry

compared to drawing a blood sample. Most current methods of CTC isolation rely on the physical selection of epithelial cells by exploiting their surface expression of epithelial cellular adhesion molecule (EpCAM) [95]. However, with growing support for EMT in tissue invasion and intravasation, it is possible that some CTCs have undergone at least a partial EMT. Such cells will have reduced expression or possible loss of EpCAM, thus potentially important CTCs will be missed entirely by standard isolation approaches. Isolation of CTCs by targeting mesenchymal markers, such as N-cadherin, alongside EpCAM may provide a more powerful prognostic outcome than standard approaches. Furthermore, molecular analysis of CTCs may provide information on the ability of these cells to extravasate and form metastatic tumors; such information would help tailor patient-specific therapies. Finally, one may question whether there is any therapeutic utility in blocking cancer cell extravasation. From this chapter it is apparent that a number of cell adhesion and intracellular signaling molecules may be viable targets for inhibiting cancer cell extravasation. To the extent that these targets overlap with those involved in leukocyte trafficking, one might anticipate side effects due to leukocyte dysfunction. In this regard, one particularly interesting target is N-cadherin which may not be involved in leukocyte trafficking and for which antibody and inhibitory peptides already exist [96, 97]. However, since metastatic dissemination may be an early event, possibly occurring prior to initial diagnosis, whether an opportunity exists to intervene in this step of metastasis is unclear. If secondary metastasis (metastasis not from the primary tumor) plays an important role clinically, blockade of extravasation may benefit patients who already have disseminated disease. Alternatively, there may be situations such as in the peri-operative state in which large numbers of cancer cells are liberated into the bloodstream for which temporary inhibition of cancer cell extravasation may be useful.

8.7

Conclusions

In order for a circulating tumor cell to form a metastatic tumor, at some point it must exit the vasculature and gain access to the parenchyma of a metastatic organ site. This is most often thought to occur by the process of extravasation. We have discussed many reported mechanisms of cancer cell extravasation, which have been derived from various cancer cell lines, in vitro models, and organs within several animal hosts. Although many differences in the cellular and molecular interactions between cancer cells and the endothelium have been observed, two general arguments prevail: (1) cancer cell extravasation is facilitated by specific molecular interactions and the ability to extravasate correlates with metastatic potential, and in some cases, organ tropism and, (2) extravasation of circulating tumor cells is non-selective, efficient, and unrelated to metastatic potential. As we have discussed throughout this chapter, these possibilities are not necessarily mutually exclusive. While size restriction is most certain to occur in some

8 Cellular and Molecular Biology of Cancer Cell Extravasation

215

organs, productive extravasation may yet depend on specific cell adhesion mechanisms. It is also clear that at least some cancer cells can transit the first restrictive vascular bed they encounter to gain access to other distal organs that may be more conducive to metastatic colonization. The mechanism by which a cancer cell extravasates likely depends on the tissue-derivation of the cancer cell, the vascular bed in which it encounters, and the expression of various proteins on both the cancer and endothelial cells. Many knowledge gaps in cancer cell extravasation are beginning to be filled, thanks to the persistent interest in this important aspect of tumor biology. As highlighted above, there is recent evidence for a role of EMT in cancer cell extravasation, and this has clinical implications on both the isolation and analysis of circulating tumor cells. Ongoing and future work will hope to clarify the genetic, molecular, and cellular basis for cancer cell extravasation and determine the feasibility of targeting this step of metastasis therapeutically. Acknowledgements We thank Jones Nauseef for critical reading of the manuscript. JMB was supported by a Department of Defense pre-doctoral fellowship, PC094754. Work on cancer cell extravasation in the Henry lab has been supported by a Grant-in-Aid from the American Heart Association.

References 1. Nourshargh S, Hordijk PL, Sixt M (2010) Breaching multiple barriers: leukocyte motility through venular walls and the interstitium. Nat Rev Mol Cell Biol 11(5):366–378 2. Friedl P, Weigelin B (2008) Interstitial leukocyte migration and immune function. Nat Immunol 9(9):960–969 3. Vestweber D (2007) Adhesion and signaling molecules controlling the transmigration of leukocytes through endothelium. Immunol Rev 218:178–196 4. Strell C, Entschladen F (2008) Extravasation of leukocytes in comparison to tumor cells. Cell Commun Signal 6:10 5. Miles FL et al (2008) Stepping out of the flow: capillary extravasation in cancer metastasis. Clin Exp Metastasis 25(4):305–324 6. Kopp HG et al (2005) The bone marrow vascular niche: home of HSC differentiation and mobilization. Physiology (Bethesda) 20:349–356 7. Zlotnik A (2006) Chemokines and cancer. Int J Cancer 119(9):2026–2029 8. Muller A et al (2001) Involvement of chemokine receptors in breast cancer metastasis. Nature 410(6824):50–56 9. Hart CA et al (2005) Invasive characteristics of human prostatic epithelial cells: understanding the metastatic process. Br J Cancer 92(3):503–512 10. van Golen KL et al (2008) CCL2 induces prostate cancer transendothelial cell migration via activation of the small GTPase Rac. J Cell Biochem 104(5):1587–1597 11. Minn AJ et al (2005) Distinct organ-specific metastatic potential of individual breast cancer cells and primary tumors. J Clin Invest 115(1):44–55 12. Jamieson WL et al (2008) CX3CR1 is expressed by prostate epithelial cells and androgens regulate the levels of CX3CL1/fractalkine in the bone marrow: potential role in prostate cancer bone tropism. Cancer Res 68(6):1715–1722 13. Barthel SR et al (2009) Alpha 1,3 fucosyltransferases are master regulators of prostate cancer cell trafficking. Proc Natl Acad Sci U S A 106(46):19491–19496

216

J.M. Barnes and M.D. Henry

14. Burger JA, Kipps TJ (2006) CXCR4: a key receptor in the crosstalk between tumor cells and their microenvironment. Blood 107(5):1761–1767 15. Luboshits G et al (1999) Elevated expression of the CC chemokine regulated on activation, normal T cell expressed and secreted (RANTES) in advanced breast carcinoma. Cancer Res 59(18):4681–4687 16. Salomon D et al (1992) Extrajunctional distribution of N-cadherin in cultured human endothelial cells. J Cell Sci 102(Pt 1):7–17 17. Voura EB, Sandig M, Siu CH (1998) Cell-cell interactions during transendothelial migration of tumor cells. Microsc Res Tech 43(3):265–275 18. Qi J et al (2005) Transendothelial migration of melanoma cells involves N-cadherin-mediated adhesion and activation of the beta-catenin signaling pathway. Mol Biol Cell 16(9):4386–4397 19. Qi J et al (2006) Involvement of Src family kinases in N-cadherin phosphorylation and betacatenin dissociation during transendothelial migration of melanoma cells. Mol Biol Cell 17(3):1261–1272 20. Drake JM et al (2009) ZEB1 enhances transendothelial migration and represses the epithelial phenotype of prostate cancer cells. Mol Biol Cell 20(8):2207–2217 21. Sequeira L et al (2008) Rho GTPases in PC-3 prostate cancer cell morphology, invasion and tumor cell diapedesis. Clin Exp Metastasis 25(5):569–579 22. Strell C et al (2007) Surface molecules regulating rolling and adhesion to endothelium of neutrophil granulocytes and MDA-MB-468 breast carcinoma cells and their interaction. Cell Mol Life Sci 64(24):3306–3316 23. Wu QD et al (2001) Human neutrophils facilitate tumor cell transendothelial migration. Am J Physiol Cell Physiol 280(4):C814–C822 24. Dong C et al (2005) Melanoma cell extravasation under flow conditions is modulated by leukocytes and endogenously produced interleukin 8. Mol Cell Biomech 2(3):145–159 25. Borsig L et al (2002) Synergistic effects of L- and P-selectin in facilitating tumor metastasis can involve non-mucin ligands and implicate leukocytes as enhancers of metastasis. Proc Natl Acad Sci U S A 99(4):2193–2198 26. Camerer E et al (2004) Platelets, protease-activated receptors, and fibrinogen in hematogenous metastasis. Blood 104(2):397–401 27. Crissman JD et al (1988) Morphological study of the interaction of intravascular tumor cells with endothelial cells and subendothelial matrix. Cancer Res 48(14):4065–4072 28. Warren BA, Vales O (1972) The adhesion of thromboplastic tumour emboli to vessel walls in vivo. Br J Exp Pathol 53(3):301–313 29. Grivennikov SI, Greten FR, Karin M (2010) Immunity, inflammation, and cancer. Cell 140(6):883–899 30. Kramer RH, Nicolson GL (1979) Interactions of tumor cells with vascular endothelial cell monolayers: a model for metastatic invasion. Proc Natl Acad Sci U S A 76(11):5704–5708 31. Ewing, J., Neoplastic Diseases. 1928. 32. van der Meer JW et al (1982) Characteristics of human monocytes cultured in the Teflon culture bag. Immunology 47(4):617–625 33. Vona G et al (2000) Isolation by size of epithelial tumor cells: a new method for the immunomorphological and molecular characterization of circulating tumor cells. Am J Pathol 156(1):57–63 34. Zheng S et al (2007) Membrane microfilter device for selective capture, electrolysis and genomic analysis of human circulating tumor cells. J Chromatogr A 1162(2):154–161 35. Doerschuk CM et al (1993) Comparison of neutrophil and capillary diameters and their relation to neutrophil sequestration in the lung. J Appl Physiol 74(6):3040–3045 36. Fidler IJ (1970) Metastasis: quantitative analysis of distribution and fate of tumor embolilabeled with 125 I-5-iodo-2¢-deoxyuridine. J Natl Cancer Inst 45(4):773–782 37. Wood S Jr (1958) Pathogenesis of metastasis formation observed in vivo in the rabbit ear chamber. AMA Arch Pathol 66(4):550–568 38. Morris VL et al (1993) Early interactions of cancer cells with the microvasculature in mouse liver and muscle during hematogenous metastasis: videomicroscopic analysis. Clin Exp Metastasis 11(5):377–390

8 Cellular and Molecular Biology of Cancer Cell Extravasation

217

39. Morris VL et al (1994) Mammary carcinoma cell lines of high and low metastatic potential differ not in extravasation but in subsequent migration and growth. Clin Exp Metastasis 12(6):357–367 40. Koop S et al (1995) Fate of melanoma cells entering the microcirculation: over 80% survive and extravasate. Cancer Res 55(12):2520–2523 41. Luzzi KJ et al (1998) Multistep nature of metastatic inefficiency: dormancy of solitary cells after successful extravasation and limited survival of early micrometastases. Am J Pathol 153(3):865–873 42. Koop S et al (1996) Independence of metastatic ability and extravasation: metastatic rastransformed and control fibroblasts extravasate equally well. Proc Natl Acad Sci U S A 93(20): 11080–11084 43. Chambers AF et al (1992) Early steps in hematogenous metastasis of B16F1 melanoma cells in chick embryos studied by high-resolution intravital videomicroscopy. J Natl Cancer Inst 84(10):797–803 44. Kienast Y et al (2010) Real-time imaging reveals the single steps of brain metastasis formation. Nat Med 16(1):116–122 45. Weis S et al (2004) Endothelial barrier disruption by VEGF-mediated Src activity potentiates tumor cell extravasation and metastasis. J Cell Biol 167(2):223–229 46. Glinskii OV et al (2005) Mechanical entrapment is insufficient and intercellular adhesion is essential for metastatic cell arrest in distant organs. Neoplasia 7(5):522–527 47. Wang X et al (2005) Beta3 integrins facilitate matrix interactions during transendothelial migration of PC3 prostate tumor cells. Prostate 63(1):65–80 48. Bauer K, Mierke C, Behrens J (2007) Expression profiling reveals genes associated with transendothelial migration of tumor cells: a functional role for alphavbeta3 integrin. Int J Cancer 121(9):1910–1918 49. Wang H et al (2004) Tumor cell alpha3beta1 integrin and vascular laminin-5 mediate pulmonary arrest and metastasis. J Cell Biol 164(6):935–941 50. Hallmann R et al (2005) Expression and function of laminins in the embryonic and mature vasculature. Physiol Rev 85(3):979–1000 51. Schluter K et al (2006) Organ-specific metastatic tumor cell adhesion and extravasation of colon carcinoma cells with different metastatic potential. Am J Pathol 169(3):1064–1073 52. Tremblay PL, Huot J, Auger FA (2008) Mechanisms by which E-selectin regulates diapedesis of colon cancer cells under flow conditions. Cancer Res 68(13):5167–5176 53. Al-Mehdi AB et al (2000) Intravascular origin of metastasis from the proliferation of endothelium-attached tumor cells: a new model for metastasis. Nat Med 6(1):100–102 54. Wong CW et al (2002) Intravascular location of breast cancer cells after spontaneous metastasis to the lung. Am J Pathol 161(3):749–753 55. Iwasaki T (1915) Histological and experimental observations on the destruction of tumor cells in the blood vessels. J Pathol Bacteriol 20:85–105 56. Takahashi M (1915) An experimental study of metastasis. J Pathol Bacteriol 20:1–13 57. Crissman JD et al (1985) Arrest and extravasation of B16 amelanotic melanoma in murine lungs. A light and electron microscopic study. Lab Invest 53(4):470–478 58. Paget S (1989) The distribution of secondary growths in cancer of the breast. 1889. Cancer Metastasis Rev 8(2):98–101 59. Nguyen DX, Bos PD, Massague J (2009) Metastasis: from dissemination to organ-specific colonization. Nat Rev Cancer 9(4):274–284 60. Geldof AA (1997) Models for cancer skeletal metastasis: a reappraisal of Batson’s plexus. Anticancer Res 17(3A):1535–1539 61. Riethdorf S, Wikman H, Pantel K (2008) Review: Biological relevance of disseminated tumor cells in cancer patients. Int J Cancer 123(9):1991–2006 62. Abbott NJ et al (2010) Structure and function of the blood-brain barrier. Neurobiol Dis 37(1):13–25 63. Wolburg H, Wolburg-Buchholz K, Engelhardt B (2005) Diapedesis of mononuclear cells across cerebral venules during experimental autoimmune encephalomyelitis leaves tight junctions intact. Acta Neuropathol 109(2):181–190

218

J.M. Barnes and M.D. Henry

64. Weil RJ et al (2005) Breast cancer metastasis to the central nervous system. Am J Pathol 167(4):913–920 65. Lee TH et al (2003) Vascular endothelial growth factor modulates the transendothelial migration of MDA-MB-231 breast cancer cells through regulation of brain microvascular endothelial cell permeability. J Biol Chem 278(7):5277–5284 66. Kim LS et al (2004) Vascular endothelial growth factor expression promotes the growth of breast cancer brain metastases in nude mice. Clin Exp Metastasis 21(2):107–118 67. Reichen J (1999) The role of the sinusoidal endothelium in liver function. News Physiol Sci 14:117–121 68. Aird WC (2007) Phenotypic heterogeneity of the endothelium: II. Representative vascular beds. Circ Res 100(2):174–190 69. Martin MD et al (2010) Rapid extravasation and establishment of breast cancer micrometastases in the liver microenvironment. Mol Cancer Res 8(10):1319–1327 70. Haier J et al (2003) An intravital model to monitor steps of metastatic tumor cell adhesion within the hepatic microcirculation. J Gastrointest Surg 7(4):507–514; discussion 514–5 71. Scherbarth S, Orr FW (1997) Intravital videomicroscopic evidence for regulation of metastasis by the hepatic microvasculature: effects of interleukin-1alpha on metastasis and the location of B16F1 melanoma cell arrest. Cancer Res 57(18):4105–4110 72. Gout S, Tremblay PL, Huot J (2008) Selectins and selectin ligands in extravasation of cancer cells and organ selectivity of metastasis. Clin Exp Metastasis 25(4):335–344 73. Rosenow F et al (2008) Integrins as antimetastatic targets of RGD-independent snake venom components in liver metastasis [corrected]. Neoplasia 10(2):168–176 74. Downey GP et al (1993) Neutrophil sequestration and migration in localized pulmonary inflammation. Capillary localization and migration across the interalveolar septum. Am Rev Respir Dis 147(1):168–176 75. Doerschuk CM (2001) Mechanisms of leukocyte sequestration in inflamed lungs. Microcirculation 8(2):71–88 76. Doerschuk CM (2000) Leukocyte trafficking in alveoli and airway passages. Respir Res 1(3):136–140 77. Brown DM, Ruoslahti E (2004) Metadherin, a cell surface protein in breast tumors that mediates lung metastasis. Cancer Cell 5(4):365–374 78. Gupta GP et al (2007) Mediators of vascular remodelling co-opted for sequential steps in lung metastasis. Nature 446(7137):765–770 79. Padua D et al (2008) TGFbeta primes breast tumors for lung metastasis seeding through angiopoietin-like 4. Cell 133(1):66–77 80. Weiss L et al (1992) Lethal deformation of cancer cells in the microcirculation: a potential rate regulator of hematogenous metastasis. Int J Cancer 50(1):103–107 81. Weiss L (1991) Deformation-driven, lethal damage to cancer cells. Its contribution to metastatic inefficiency. Cell Biophys 18(2):73–79 82. Tarin D et al (1984) Clinicopathological observations on metastasis in man studied in patients treated with peritoneovenous shunts. Br Med J (Clin Res Ed) 288(6419):749–751 83. Podsypanina K et al (2008) Seeding and propagation of untransformed mouse mammary cells in the lung. Science 321(5897):1841–1844 84. Thiery JP et al (2009) Epithelial-mesenchymal transitions in development and disease. Cell 139(5):871–890 85. Thiery JP (2003) Epithelial-mesenchymal transitions in development and pathologies. Curr Opin Cell Biol 15(6):740–746 86. Thiery JP (2002) Epithelial-mesenchymal transitions in tumour progression. Nat Rev Cancer 2(6):442–454 87. Spaderna S et al (2008) The transcriptional repressor ZEB1 promotes metastasis and loss of cell polarity in cancer. Cancer Res 68(2):537–544 88. Vandewalle C, Van Roy F, Berx G (2009) The role of the ZEB family of transcription factors in development and disease. Cell Mol Life Sci 66(5):773–787

8 Cellular and Molecular Biology of Cancer Cell Extravasation

219

89. Tsuji T et al (2008) Epithelial-mesenchymal transition induced by growth suppressor p12CDK2-AP1 promotes tumor cell local invasion but suppresses distant colony growth. Cancer Res 68(24):10377–10386 90. Karnoub AE et al (2007) Mesenchymal stem cells within tumour stroma promote breast cancer metastasis. Nature 449(7162):557–563 91. Cristofanilli M et al (2004) Circulating tumor cells, disease progression, and survival in metastatic breast cancer. N Engl J Med 351(8):781–791 92. Hayes DF et al (2006) Circulating tumor cells at each follow-up time point during therapy of metastatic breast cancer patients predict progression-free and overall survival. Clin Cancer Res 12(14 Pt 1):4218–4224 93. Fehm T et al (2002) Cytogenetic evidence that circulating epithelial cells in patients with carcinoma are malignant. Clin Cancer Res 8(7):2073–2084 94. Morgan TM et al (2009) Disseminated tumor cells in prostate cancer patients after radical prostatectomy and without evidence of disease predicts biochemical recurrence. Clin Cancer Res 15(2):677–683 95. Fehm T et al (2008) Micrometastatic spread in breast cancer: detection, molecular characterization and clinical relevance. Breast Cancer Res 10(Suppl 1):S1 96. Li H, Price DK, Figg WD (2007) ADH1, an N-cadherin inhibitor, evaluated in preclinical models of angiogenesis and androgen-independent prostate cancer. Anticancer Drugs 18(5):563–568 97. Tanaka H et al (2010) Monoclonal antibody targeting of N-cadherin inhibits prostate cancer growth, metastasis and castration resistance. Nat Med 16(12):1414–1420

Chapter 9

Role of CXCL12 and CXCR4 in Tumor Biology and Metastasis Ramesh K. Ganju, Yadwinder S. Deol, and Mohd W. Nasser

Abstract Chemokine receptor CXCR4 and its cognate ligand CXCL12, also known as stromal derived factor-1 (SDF-1), have been shown to play an important role in growth and metastasis of various tumors. CXCR4 is the most common chemokine receptor that has been demonstrated to be overexpressed in several cancers. Its overexpression is also correlated with poor clinical outcomes and survival in various cancers, including breast, prostate, and lung. CXCR4/CXCL12 signaling axis plays an important role in organ selective metastasis since CXCR4 overexpressing cancer cells have been shown to metastasize to organs such as bone, lymph nodes, liver, lung, and brain, which produce high amounts of CXCL12 and thus provide a favorable microenvironment. CXCR4/CXCL12 axis has been shown to enhance tumor growth and metastasis by maintaining cancer stem cells and modulating tumor stroma through activation of cancer-associated fibroblasts and recruitment of CXCR4+ endothelial precursor cells, thereby enhancing angiogenesis. CXCR4/CXCL12 axis has been shown to mediate both pro-tumorigenesis and metastasis by activating multiple signaling pathways, including protein tyrosine pathways through activation of Pyk2. Several CXCR4/CXCL12 antagonists/agonists have been shown to have potential therapeutic effects as they significantly inhibit tumor growth and metastasis of various cancers in both in vitro and in vivo mouse models. These studies suggest that the CXCR4/CXCL12 signaling axis is an important target for developing innovative therapies against various cancers.

R.K. Ganju (*) • Y.S. Deol • M.W. Nasser Department of Pathology, The Ohio State University Medical Center, Columbus OH, USA e-mail: [email protected]

A. Fatatis (ed.), Signaling Pathways and Molecular Mediators in Metastasis, DOI 10.1007/978-94-007-2558-4_9, © Springer Science+Business Media B.V. 2012

221

222

9.1

R.K. Ganju et al.

CXCR4/CXCL12 Signaling Pathways in Cancer

CXCR4/CXCL12 axis has been shown to mediate pro-tumorigenic effects and metastasis by activating multiple signaling pathways [1–3]. CXCR4 is a seventransmembrane receptor that belongs to the G-protein coupled receptor (GPCR) family. Therefore, CXCR4 has been shown to partly mediate its functional effects through GPCR signaling pathways [4]. GPCRs have been shown to bind to heterotrimeric G-protein composed of Ga, Gb and Gg subunits [5–7]. CXCR4-Gprotein complex in basal state is bound to GDP. However, upon binding its ligand CXCL12, GDP is replaced by GTP which disrupts the G-protein into a bg dimer and the a-monomer. Based on the sequence similarity, the Ga subunit is divided into four families: Gas, Gai, Gaq, and Ga12. However, CXCR4 has been shown to mainly mediate its signaling through Gai. Gai modulates the downstream signaling by inhibiting adenyl cyclase [8, 9]. Gai signaling has also been associated with CXCR4 signaling through activation of Src related kinases (Src), Erk 1/2 and NF-kB [4, 9–12]. ERK can phosphorylate and activate other cellular proteins as well as translocate into the nucleus and phosphorylate and activate transcription factors, leading to changes in gene expression and cell cycle progression. CXCL12-mediated activation of MAPK extracellular signal regulated kinases (MEK) has been shown to inactivate BCL-2 which is pro-apoptotic and thus can inhibit apoptosis of cells [10, 13]. Therefore, CXCR4/CXCL12-axis may promote cell survival by posttranslational inactivation of the cell death machinery as well as by increased transcription of cell survival-related genes. CXCR4/CXCL12-mediated chemotaxis and proliferation is also mediated by PI3 Kinase which can be activated both by Gbg and Ga subunits [10, 14]. PI3K then can activate the serine-threonine kinase AKT that plays a key role in tumor cell survival, proliferation, and chemotaxis [15]. In addition to GPCR signaling pathways, CXCR4 also has been shown to activate protein kinase pathways, especially focal adhesion tyrosine kinases [16, 17]. We have shown that focal adhesion kinase (FAK) and related adhesion focal tyrosine kinase (RAFTK), also known as Pyk2, plays an important role in CXCR4-mediated signaling in breast cancer, Kaposi sarcoma, and T-cells [18, 19]. CXCR4-mediated signaling was also shown to enhance motility and invasion in breast cancer cells [18]. In addition, CXCR4 has also been shown to activate components of focal adhesion complexes such as Crc and paxilin [18] (Fig. 9.1). CXCR4 also mediates its signaling through tyrosine phosphatases SHP2 and an adaptor-ubiquitin ligase like protein Cbl [4, 18, 20, 21]. Cbl protein was tyrosine-phosphorylated upon CXCL12 treatment in breast cancer cells [18]. Cbl, a 120-kDa protein that contains a tyrosine-kinase-binding (TKB) domain, can function as an adaptor molecule by binding to various proteins [22, 23]. It also contains a RING finger domain and negatively regulates signaling by directing the ubiquitination and degradation of activated receptor tyrosine kinases [24, 25]. The C-terminal region of Cbl contains a proline-rich domain that binds to the SH3-domain containing proteins. It also contains a cluster of tyrosine residues that bind to SH2-domain containing proteins [26–28]. Some of the domain-SH3 containing proteins that bind to Cbl may also

9

Role of CXCL12 and CXCR4 in Tumor Biology and Metastasis

223

Fig. 9.1 Schematic representation of CXCL12-induced and CXCR4-mediated signaling mechanisms that regulate chemotaxis and proliferation in tumor cells

bind to components of lipid rafts, which play an important role in chemotaxis [29–31]. The Cbl protein has been shown to regulate cell spreading in response to integrin engagement and is involved in the functional organization of the actin cytoskeleton [32, 33]. CXCL12 treatment also activated PI3 kinase increasing its association with Cbl and SHP2. CXCL12-induced chemotaxis and chemoinvasion was significantly inhibited by PI3K, RAFTK/Pyk2 and tyrosine phosphatase inhibitors [4, 18, 20]. It has been proposed that CXCL12-induced chemotaxis may be mediated through the activation and formation of multimeric-signaling complex comprised of Pyk2, SHP2 and PI3 kinase [4, 18, 20]. The complex then results in cytoskeletal changes and activation of MAP kinases and transcription factors. CXCL12 treatment of PC-3 cells led to MEK, IKK and IkBa phosphorylation leading to nuclear localization of NF-kB [34]. These transcription factors may in turn enhance expression of metallo-proteinases and other proteins. In this regard, CXCL12 has also been shown to activate MMP2 and MMP9 in breast and prostrate cancer cells [35]. Further, CXCR4 dependent signaling has been shown to decrease TIMP2 expression which in combination with enhanced matrix metalloproteinase secretion may increase the invasiveness of the prostate cancer cells [36]. Serratì et al. showed that CXCL12 produced by endothelial as well as epithelial cells induces the expression of urokinase-type plasminogen activator receptor (uPAR) in the CXCR4-expressing breast cancer cells [37]. Expression of uPAR has been linked to transition from dormant to metastatic status of tumor cells in vivo [38, 39]. In addition, CXCL12 can up-regulate the expression of adhesion molecules like VLA-4 which are involved

224

R.K. Ganju et al.

in tumor cell invasion [40, 41]. CXCL12/CXCR4 axis has also been shown to up-regulate the expression of b3 integrin leading to the activation of avb3 receptors. Activated avb3 integrins have been shown to mediate adhesion of prostrate cancer cells to bone marrow epithelium [42, 43]. These processes may regulate CXCL12-mediated chemoinvasion and chemotaxis that may lead to the development of metastasis. CXCR4 has been shown to contain a short C-terminal domain (CTD) that contains tyrosine residues, which are phosphorylated upon ligand activation. CTD plays a major role in regulating CXCR4 receptor desensitization and down-regulation [44, 45]. CXCR4 CTD has been shown to play an important role in the process of epithelial to mesenchymal transition (EMT). Aberrant CXCR4 chemokine receptor function due to c-terminal truncation mutation has been shown to be present in various diseases, including warts, hypogammaglobulinemia, immunodeficiency, and myelokathexis syndromes. In the case of breast cancer cells, it has been shown that overexpression of CXCR4 receptor containing c-terminal truncated cytoplasmic tails results in altered morphology as indicated by EMT and higher growth rates [44]. Breast cancer MCF-7 cells expressing truncated CXCR4 (CXCR4-ΔCTD) exhibited higher growth rate and EMT transition compared to wild type CXCR4 expressing cells. Furthermore, cells expressing truncated CXCR4 showed decreased E-cadherin expression and increased ERK activation [44]. These studies indicate that CTD region of CXCR4 is important for its regulation, expression and recycling. Various factors have been shown to regulate CXCR4 including tumor suppressor protein p53. p53 negatively regulates the expression of CXCR4 in breast cancer cells as downregulation of wild type p53 was shown to increase endogenous CXCR4 expression in breast cancer cells [46]. PRIMA-1 and CP-31398, which have been shown to rescue p53, were shown to reduce expression of CXCR4, at both mRNA and cell-surface levels [46]. These studies indicate that p53 targeting drugs may also reduce CXCL12/CXCR4-mediated cell proliferation and metastasis. In addition, p53 has also been shown to inhibit CXCL12 expression in fibroblasts, thereby modulating adjacent cancer cell migration and invasion [47]. Recently, it was shown that wt p53 reduces expression of CXCL12 in stromal fibroblasts [48]. Furthermore, p53 mutant expressing stromal fibroblasts were shown to express high amounts of CXCL12, thereby augmenting the tumor growth in prostrate cancer. CXCR4 has also been shown to cross-talk with other receptors such as epidermal growth factor receptor (EGFR), human epidermal growth factor receptor 2 (HER2), and transforming growth factor beta (TGF-b), thereby regulating tumor growth and metastasis. In ovarian cancer cells, CXCR4 activation was shown to induce EGFR trans-activation by enhancing EGFR phosphorylation [49, 50]. Furthermore, it was shown that CXCL12 induced c-Src activation that may enhance EGFR phosphorylation and trans-activation [49]. CXCL12 has also been shown to trans-activate HER2/neu through Src kinase activation in breast cancer cells [11]. Furthermore, HER2 has been shown to enhance CXCR4 expression, thereby promoting growth and metastasis in lung and breast cancer cells [11, 51, 52]. EGFR has also been shown to enhance CXCR4 expression in non-small lung cancer cells through

9

Role of CXCL12 and CXCR4 in Tumor Biology and Metastasis

225

increased expression and activation of HIF-1 alpha [53]. Breast cancer patients showing poor overall survival rate has been shown to co-express CXCR4 and EGFR/ HER2 [51]. In breast cancer cells, CXCR4 was also shown to form a complex with insulin like growth factor-1 receptor (IGF-1R) [54]. Furthermore, IGF was shown to enhance migration and chemotaxis in these cells by activating CXCR4-mediated signaling pathways [54]. In addition, cross-talk between CXCL12/CXCR4 and TGF-b1 signaling was shown to initiate and maintain the differentiation of fibroblasts into myofibroblasts, thereby promoting tumor growth and metastasis in breast cancer [55]. Recently, there have been several reports suggesting cross-talk between estrogen receptor and CXCR4 signaling. It was shown that enhanced CXCR4 signaling drives estrogen receptor positive breast cancer to endocrine therapy resistant phenotype [56]. Furthermore, CXCR4 over-expression in ER positive MCF-7 cells was shown to enhance hormone independence [57].

9.2

CXCR4/CXCL12 in Breast Cancer Progression and Metastasis

CXCR4 is expressed in a wide variety of tissues including bone marrow, blood, spleen, thymus, lymph nodes, pituitary and adrenal glands [3, 58]. However, CXCR4 was recently shown to be one of the most common chemokine receptors to be overexpressed in several human cancers such as breast, ovarian, melanoma, prostrate, etc. [59]. Figure 9.2 shows the CXCR4 expression in a variety of human and neoplastic tissues as detected by immunohistochemical staining using anti-CXCR4 antibody UMB-2 from Dr. Schulz’s laboratory [59]. It has been demonstrated that CXCR4 levels measured by immunohistochemistry are extremely low or absent in normal breast epithelium, while >90% of specimens with atypical ductal hyperplasia showed positive staining for CXCR4 [60, 61]. CXCR4 is also present in ductal carcinoma in situ (DCIS). Moreover, CXCR4-expression is observed in about 75% of biopsy specimens of invasive ductal carcinoma and high expression of CXCR4 is correlated with decreased overall survival of patients in breast cancer [61, 62]. Similarly, increased expression levels of CXCR4 have been linked to cancer progression from atypical hyperplasia to invasive cancer [63]. Recently, it was shown that poor clinical outcome in triple negative breast cancer patients was shown to be co-related with high CXCR4 expression [64]. The CXCR4/CXCL12 pathway is involved in several aspects of breast tumor progression, including metastasis [65–67]. Metastatic cells characteristically lose growth inhibitory responses, undergo alterations in adhesiveness, and migrate to other distant organs by invading the blood and lymphatic vessels [66, 67]. This migration leads to secondary tumor formation that represents the most devastating feature of breast and other cancers. One of the important steps in the intricate process leading to the formation of metastases is tumor cell motility [68, 69]. It has been observed that more motile tumor cells may have increased metastatic potential [70]. We have shown that CXCL12 induces chemotaxis, chemoinvasion and

226

R.K. Ganju et al.

Fig. 9.2 CXCR4 immunohistochemical staining in human formalin-fixed and paraffin-embedded tissues. (a–k) CXCR4 immunohistochemical staining of a variety of human normal and neoplastic tissues using anti-CXCR4 antibody UMB-2. Sections were dewaxed, microwaved in citric acid and incubated with UMB-2 at a dilution of 110. Sections were sequentially treated with biotinylated anti-rabbit IgG and AB solution. Sections were then developed in diaminobenzidine and lightly counterstained with hematoxylin. Note that UMB-2 detected CXCR4 receptors at the plasma membrane of a subset of tumor cells in a variety of human tumors including mammary carcinoma (a), endometrial carcinoma (b), cervical carcinoma (c), ovarian carcinoma (d), gastric carcinoma (e), pancreatic carcinoma (f), colonic carcinoma (g) and malignant pheochromocytoma (j). In nearly all of these cases UMB-2 immunoreactive tumor cells exhibited a heterogeneous distribution throughout the tumor. UMB-2 revealed predominant cytoplasmic staining in prostate carcinoma (h). In glioblastoma, UMB-2 detected CXCR4 receptors predominantly on the plasma membrane of endothelial cell of tumoral blood vessels (i). UMB-2 also detected CXCR4 receptors in the zona fasciculata of the adrenal cortex (k). Inset in (c), peptide adsorption control. Arrowheads in (i), tumoral blood vessels. C capsule; Z.g. zona glomerulosa, Z.f. zona fasciculata, Z.r. zona reticularis. Scale bars, A=B=C=D=E=F=G=H=I=J=K=50 mm, K=100 mm (Reprinted by permission of Dr. Schulz [59])

9

Role of CXCL12 and CXCR4 in Tumor Biology and Metastasis

227

adhesion of CXCR4 positive breast cancer cells in vitro. Molecules, which enhance the motility of tumor cells, have been implicated to play a role in the development and progression of tumors to the metastatic step [68, 70–73]. Metastasis is a frequent cause of death in patients with malignant neoplasms. Increased expression of CXCR4 has also been correlated with lymph node metastasis in breast cancer [74]. It is known that during metastasis, tumor cells employ several mechanisms to regulate the trafficking of tumor cells [66, 67]. It is now evident that CXCL12/CXCR4 plays a significant role in organ-selective cancer metastasis in breast cancer [75]. In particular, breast cancer cells were shown to express the chemokine receptors, CXCR4, whereas organs such as lymph nodes, bones and lungs that represent important sites of breast cancer metastasis were shown to express ligands of these receptors [72, 76] (Fig. 9.3). Furthermore, in vivo and in vitro neutralization of the CXCL12/CXCR4 interaction leads to a marked inhibition of metastasis [72, 77, 78]. CXCR4 was recently shown to be abundantly expressed by a subpopulation of breast cancer cells that have enhanced metastatic activity to bone or the adrenal medulla [79]. Moreover, breast cancer cells isolated from mammary fat pad xenografts were observed to have high surface expression of CXCR4 and enhanced CXCL12-induced chemotaxis [12]. Also, cancer cells that metastasize to lungs were found to have increased expression of CXCR4 and migration towards CXCL12 [12]. These studies suggest that the CXCR4 receptor and its cognate ligand CXCL12 are involved in regulating the trafficking and metastasis of these tumor cells to various tissues. Muller et al. reported that CXCR4 and CXCL12 are central players in regulating metastasis [72, 76]. Furthermore, using a mouse model of breast cancer, they have shown that neutralizing antibodies to CXCR4 significantly limited metastases to lymph nodes and lung [72]. These data were the first to identify a key function for CXCL12/CXCR4 in metastatic breast cancer. We have also shown that CXCL12-induced and CXCR4-mediated signaling plays an important role in breast cancer migration in vitro [18, 20]. These results were supported by Liang et al., who showed that blocking CXCR4 expression at the mRNA level by siRNAs decreased breast cancer cell invasion in in vitro assays and inhibited metastasis in an animal model [80]. CXCR4 has been shown to be one of the few genes that are enriched in a subpopulation of metastatic breast cancer cells [81]. Over-expression of CXCR4 alone has been shown to significantly increase the bone metastases in vivo. Li et al. demonstrated that HER2/neu enhances the expression and function of CXCR4 by inhibiting CXCR4 degradation [52]. Potentially, co-expression of CXCR4 with EGFR/HER2-neu may help to select for a metastatic clone with a particular propensity to metastases to the bone marrow. Further studies have demonstrated that elevated levels of pAkt and CXCR4 are induced by factors that are both dependent and independent of HER-2, and the activation of HER-2/ CXCR4/Akt signaling pathway in primary breast tumors may contribute to the formation of bone metastasis in breast cancer [82]. CXCL12/CXCR4 axis is also implicated in the bone metastasis of prostrate cancers [83]. CXCR4/CXCL12 signaling has been shown to stimulate production of matrix metalloproteinases and increase integrin activity [35, 43, 84]. Further, metastasis triggering mechanisms like hypoxia that enhance expression of hypoxia-inducible factor-1 (HIF-1) have

228

R.K. Ganju et al.

Fig. 9.3 Model of chemokine regulation of breast-cancer metastasis. Metastasis is an orderly, multistep process involving the movement of cancer cells from the primary tumor to specific organs under the guidance of specific chemokines. First, cancerous mammary epithelial cells undergo clonal proliferation, invade local tissue, induce angiogenesis, and express CXC chemokine receptor 4 (CXCR4) on their surface. Then, cancer cells detach from the primary tumor, migrate across lymphatic and vascular walls in the tumor, and enter the systemic circulation. Cancer cells are arrested in vascular beds in organs that produce high levels of the CXCR4 ligand (CXCL12), which is expressed on the surface of vascular endothelial cells. Binding of CXCL12 to CXCR4 induces the migration of cancer cells into normal tissue, where the cells proliferate, induce angiogenesis, and form metastatic tumors. Breast-cancer cells do not usually metastasize to organs that produce low levels of CXCL12, such as the kidney (Reprinted by permission of Dr. Murphy [76])

also been shown to induce CXCR4 expression [85]. It has been shown that inactivating mutation of von Hippel Lindau (VHL) tumor suppressive gene, which normally targets HIF-1 for degradation, increases CXCR4 expression in adrenal cell carcinomas [86]. VEGF has also been shown to enhance CXCR4 expression during breast cancer progression and metastasis [12, 87]. Oncoproteins, such as RET/PTC,

9

Role of CXCL12 and CXCR4 in Tumor Biology and Metastasis

229

have been shown to enhance the transforming ability of breast cancer cells by enhancing CXCR4 expression [88]. In addition, the Pax3-FKHR has been shown to increase CXCR4 expression of rhabdomyosarcoma that leads to enhanced migration and adhesion of these cells [89].

9.3

CXCR4/CXCL12 in Lung Cancer Growth and Metastasis

There are some reports which suggest a correlation between CXCR4 and clinical outcomes in non-small cell lung cancer (NSCLC). It has been suggested that CXCR4 expression may be increased in patients with metastatic disease [90]. CXCR4 expression has also been shown to be in both cytoplasm and nucleus; however, there are conflicting reports regarding the clinical relevance of such differential expression [91, 92]. It has also been shown that nuclear expression of CXCR4 in NSCLC tumor cells significantly correlates to lymph node metastasis [91]. In contrast, Wagner et al. showed nuclear staining to be significantly associated with improved disease free survival whereas cell-membrane staining was associated with distant metastasis and decreased disease-free survival in lung adenocarcinoma patients [92]. Interestingly, it has also been shown previously that differential expression of CXCR4 correlated with metastatic potential in vitro and in vivo [93]. This suggests that the movement of NSCLC from primary sites to metastatic nodes might be dependent on the level of CXCR4 [93]. It has been shown that high expression levels of CXCR4 are correlated with metastasis to the brain in NSCLC that have shorter survival rates [94]. Further, since CXCR4 expression is inducible by hypoxia, Liu et al. showed that when HIF-1a and HIF-2a were knocked down using RNAi approach in NSCLC cells, it reduced the invasion, adhesion and migration in response to CXCL12 [95]. Burger et al. showed high levels of functional CXCR4 receptors in small cell lung cancer (SCLC) cells [96, 97]. They further showed that CXCR4 activation induces migratory and invasive responses and adhesion to marrow stromal cells in a CXCR4 and integrin-dependent fashion [96, 97]. CXCR4/CXCL12 axis is an important mediator for the adherence of SCLC cells to extra cellular matrix (ECM) proteins like fibronectin, collagen I and VCAM-1 [98]. CXCR4-induced adhesion of SCLC cells to marrow stromal cells protected them against etoposide-induced apoptosis [98]. The CXCR4 antagonist T140 and its derivatives have been shown to block the CXCR4-mediated adhesion to ECM components and thereby affect SCLC chemo-sensitivity [98]. Collectively, these studies indicate that expression of CXCR4 by SCLC cells facilitates the activation of integrins that regulate the adhesion of tumor cells to the bone marrow microenvironment, which in turn confers drug resistance and tumor cell growth. Moreover, hypoxia is able to induce a significant increase in the expression level of CXCR4 on NSCLC cells through the VHLHIF-1a pathway, supporting the notion that HIF-1a-mediated up-regulation of CXCR4 may be a common response of tumor cells to hypoxia [53, 99]. Recent studies also implicate that activation of EGFR by EGF increases expression of CXCR4 and chemotactic potential of lung cancer cell lines. It has been suggested that EGFR

230

R.K. Ganju et al.

within the primary tumors of NSCLC provides the microenvironment and signals necessary to up-regulate CXCR4 expression and promote metastasis [100]. A series of targets and therapeutic strategies for the treatment of lung cancer are currently being investigated [53]. However, all patients ultimately develop resistance against these agents, including chemotherapy, possibly due to abnormal signal transduction and high EGFR expression level. CXCL12 has been shown to modulate EGFRmediated signaling pathways [100].

9.4

CXCL12/CXCR4 and Other Cancers

Several other neoplasias are known to express CXCR4 chemokine receptors. Most notably, CXCR4 activation by CXCL12 induces migration and/or survival of neuronal and glial tumors, neuroblastoma cells, colorectal cancer, esophageal cancer, melanoma, renal cell cancer, hepatocellular carcinoma, ovarian cancer and rhabdomyosarcoma [101]. In other cancers, such as prostate and pancreatic, the CXCL12/ CXCR4 axis was shown to promote tumor cell trans-endothelial chemotaxis [102, 103]. In patients with colorectal cancer and melanoma, CXCR4 expression of primary tumor cells correlates with recurrence, metastasis, and survival [104–106]. Interestingly, in these cancer patients, rate of bone metastasis is not as high as that of prostate and breast cancers indicating that CXCL12/CXCR4 may not be the only mechanism required for the establishment and localization of tumors to the bone marrow. Expression of CXCR4 in hepatocellular carcinoma has been shown to decrease 3-year survival rates and correlates to local tumor progression, lymphatic and distant metastasis in these patients [107]. In addition, increased levels of nuclear expression of CXCR4 and VEGF-C was shown to be correlated with lymph node metastasis and worse prognosis in these patients [107]. This finding has also been observed in colorectal cancer patients where nuclear expression of CXCR4 is related to poor outcome [108]. In renal cell carcinomas, CXCR4 expression has been found to localize in cytoplasm of metastatic renal cell carcinomas whereas in the case of primary non-metastatic carcinomas, its expression was observed in the nucleus [109–111]. In melanoma cancer cells, increased metastasis of CD133+ cells was shown to be mediated through CXCR4/CXCL12 axis [106]. Direct co-relation has been observed between grade and CXCR4 expression in prostate cancer [112]. CXCR4 neutralizing antibody and a blocking peptide have been shown to decrease the metastasis of prostrate cancer to bone [113]. A positive co-relation has been observed between tissue levels of CXCL12 and metastatic lesions [114]. Signaling through CXCR4 could alter the ability of cancer cell lines to adhere to endothelium and invade through extracellular matrix components in the bone marrow [42, 115]. CXCR4/CXCL12 has also been shown to play an important role in the development of hematopoietic malignancies such as acute leukemias, multiple myeloma, B-cell chronic lymphocytic leukemia, and acute myelogenous leukemias [116]. CXCL12 produced by bone marrow stromal cells has been implicated in providing growth and survival signals through activation of CXCR4 receptors present on various hematopoietic malignant cells (91).

9

Role of CXCL12 and CXCR4 in Tumor Biology and Metastasis

9.5

231

Role of CXCR4/CXCL12 Axis in Enhancing Growth Through Modulation of Tumor Stroma

CXCL12/CXCR4 axis has also been shown to regulate growth and metastasis by enhancing development of carcinoma-associated fibroblasts (CAFs), thereby modulating tumor stroma, an active element of tumor microenvironment [117, 118]. In tumors, cancer cells coexist with other cell stypes such as fibroblast, endothelial, immune cells, and extracellular matrix (ECM) that form the tumor stroma [2, 118, 119]. Higher incidence of tumor development has been observed in tissues with chronically inflamed stroma in which resting fibroblasts get activated [120, 121]. Such activated subsets of fibroblasts are CAFs or myofibroblasts and have been observed in the stroma of invasive human breast cancers [118, 119]. Myofibroblasts possess increased contractile ability, promote angiogenesis and stimulate epithelial cell growth through production of ECM and secretion of growth factors, cytokines, and chemokines [122, 123]. Recently, it was shown that CAFs derived from breast tumors secrete high levels of CXCL12 which enhance tumor growth through paracrine mechanism by activating CXCR4 receptor present on tumor cells [117]. In addition, CXCL12 has also been shown to increase angiogenesis through endocrine mechanisms by recruiting CXCR4+ endothelial pre-cursor cells to the tumor stroma [5, 117, 118] (Fig. 9.4). Similarly, CXCL12, secreted by CAFs, stimulates the in vivo growth of benign prostate hyperplasia (BPH) that expresses CXCR4 [124, 125]. In addition, TGF-b and CXCL12 autocrine signaling mechanisms have been shown to enhance tumor progression by development of tumor-promoting CAF myofibroblasts from preexisting stromal fibroblasts [55]. It was shown that TGF-b released by carcinoma cells can elicit enhanced endogenous TGF-b and CXCL12 production via TGF-b receptor-Smad signaling pathway [55]. Furthermore, TGF-b was also shown to induce CXCR4 expression in stromal fibroblasts and tumor cells, thereby facilitating the generation of two autocrine signaling loops, mediated by TGF-b and CXCL12, acting in a positive feedback manner [55]. Such autocrine signaling loops self stimulate and cross-communicate with each other to maintain the myofibroblastic phenotype. CAFs have also been shown to promote the proliferation of breast cancer CD44+CD24− stem cells through their ability to secrete CXCL12. Furthermore, CXCL12/CXCR4 has been shown to play an important role in breast and lung cancer stem cells survival and drug resistance [117, 118, 126]. CXCL12/CXCR4 axis may also stimulate tumor growth by enhancing recruitment of Gr1+CD11b+ myeloid derived suppressor cells (MDSCs) to the tumor stroma [127]. It was shown recently that in Wnt signaling-induced mammary gland tumorigenesis, CXCL12 produced by myoepithelial cells and stromal fibroblasts supports tumorigenesis by involving Gr1+CD11b+ MDSCs [128]. MDSCs have been shown to enhance tumor growth by inhibiting tumor immune response [127, 128]. Recently, Hiratsuka et al., using CXCR4 conditional knockout mice, showed that inhibition of CXCR4 in myeloid cells suppress tumor growth, angiogenesis and metastasis by decreasing recruitment of Gr-1+CD11b+ bone marrow derived myeloid cells (BMDCs) to the primary tumors [129]. Thus, targeting CXCL12/CXCR4

232

R.K. Ganju et al.

Fig. 9.4 Schematic representation of tumor-promoting effects provoked by CXCL12 released by stomal fibroblasts in invasive human mammary carcinomas. CXCL12 released by stomal fibroblasts facilitates tumorigenesis by two different mechanisms. Stromal fibroblast-derived CXCL12 enhances tumor growth by stimulating angiogenesis through recruiting circulating EPCs into the tumor mass (endocrine effect) and by direct paracrine stimulation of tumor cells through CXCR4 expressed on carcinoma cells (paracrine effect) (Reprinted by permission of Dr. Weinberg and the Cell Press, as published in Cell. 2005;121(3):335–48)

signaling axis will prevent the development and maintenance of tumor promoting myofibroblast and therefore, may prove to be a useful antitumor therapeutic strategy in the future.

9.6

CXCR4/CXCL12 and Angiogenesis

CXCL12/CXCR4 axis has been shown to play an important role in vasculature and angiogenesis. CXCR4 or CXCL12 knockout mice are embryonically lethal due to disruptions in the development of hematopoietic, vasculature and neuronal network in many organs [105, 130]. During tumor development, CXCL12 produced by stromal cells recruits CXCR4 positive endothelial precursor cells to tumors, thereby enhancing endothelial cell growth and vasculature [117, 118]. The ability to recruit and sustain an active vasculature is critical for the survival of tumors. Recently, prostate cancer tumor-associated blood vessels and basal cell hyperplasia were shown to express CXCL12 [131]. Tumors derived from CXCR4 over-expressing

9

Role of CXCL12 and CXCR4 in Tumor Biology and Metastasis

233

PC3 cells show significantly higher blood vessel density, functionality and invasiveness [132]. CXCL12/CXCR4 has been shown to enhance angiogenesis at metastatic sites by down-regulating glycolytic enzyme phospho-glycerate kinase (PGK) [133]. PGK has angiostatic properties as it cleaves extracellular plasminogen to produce angiostatin that releases the angiostatic clamp [133]. This angiogenetic switch has been shown to play an important role for enhancing growth of metastasized colonies and thereby enhance metastasis. Vascular endothelial growth factor (VEGF), which plays an important role in angiogenesis and the survival of metastatic breast carcinoma cells, was recently shown to induce expression of the CXCR4 receptor in carcinoma cells, which promotes their chemotaxis [87]. Tumor angiogenesis can effect the tumor growth and neo-vascularization by several ways [117, 134]. New blood vessels arise from pre-existing vessels through the dissociation, migration, and division of differentiated endothelial cells. Further, these blood vessels can also be synthesized de novo from bone marrow-derived cells (BMDCs) [135]. BMDCs, such as endothelial progenitor cells, tumor-associated macrophages, Tie2-expressing monocytes, and myeloid progenitor cells, have also been shown to enhance tumor angiogenesis, thereby facilitating tumor growth [135–139]. Some BMDCs express high levels of CXCR4 receptor and can be mobilized from the bone marrow to sites producing the chemokine ligand, CXCL12 [117, 118, 129]. The importance of the stromal derived factor-1 (SDF-1)-CXCR4 axis in angiogenesis is apparent from the lack of gastrointestinal blood vessels in CXCR4−/− mice [140].

9.7

Inhibitors of CXCR4/CXCL12 Signaling Axis

CXCR4/CXCL12 has been shown to be a potential therapeutic target because targeting CXCR4/CXCL12 axis in cancer will block migration and dissemination of tumor cell metastasis, paracrine growth and survival signals and pro-angiogenesis effects. Currently, several types of CXCR4/CXCL12 antagonist/agonists are being used to inhibit this axis, such as bicyclams (AMD3100), T22, TN14003, CTCE9908, and ALX40–4C, which are analogs and peptides based on the amino-terminal region of the chemokine, CXCL12 [141]. AMD3100 is a specific antagonist of CXCL12 binding to CXCR4, inhibiting CXCL12-mediated calcium mobilization, chemotaxis and GTP binding, and does not cross-react with other chemokine receptors [142, 143]. AMD3100 antagonists are macrocyclic polyamines which were shown to inhibit X4 strain of HIV binding to CXCR4 which acts as co-receptor for HIV [142, 143]. Functional studies have shown that a strong and direct interaction occurs between CXCR4 and bicyclams without receptor internalization [144]. AMD3100 (recently re-named as Plerixafor or Mozobil) was shown to inhibit CXCR4/CXCL12 mediated tumor growth and metastasis both in vitro and in vivo in various types of cancers [144]. Recently, radiolabeled AMD3100 was developed to selectively target and monitor CXCR4 positive tumor growth and metastasis [145]. CXCR4 antagonists, such as Plerixafor (AMD3100), have been shown to sensitize

234

R.K. Ganju et al.

SCLC cells to cytotoxic drugs, such as etoposide [146], suggesting the CXCR4/ CXCL12 signaling axis as a novel target in drug-resistant SCLC therapy. Combining CXCR4 antagonist with cytotoxic chemotherapy may be an attractive strategy to overcome chemo-resistance and relapse in lung cancer. Another CXCR4 inhibitor, TN14003, derived from 14-mer peptide antagonists of CXCR4 (T140) has been used in various preclinical studies, including in vivo models for breast cancer [147]. These studies show that in addition to limiting metastasis of breast cancer via inhibiting migration, TN14003 may also be useful as a diagnostic tool to identify CXCR4 receptor positive tumor cells in culture and tumor samples [148]. TN14003 has also been shown to be effective in other cancers including small cell lung cancer (SCLC) [146], malignant melanoma and pancreatic cancer [141]. Peptide analogs of CXCL12 (CTCE-9908 and CTCE-0214), which have shown tumor and metastatic inhibiting activity, recently received orphan drug status by the Food and Drug Administration for the treatment of osteogenic sarcoma [149, 150]. CTCE-9908 has shown both inhibition of the primary tumor and anti-metastatic effects in animal models of melanoma, osteosarcoma, breast, and prostate tumors [149, 151–153]. In prostrate cancer, CTCE-9908 delivered intraperitoneally was also associated with inhibition of VEGF and angiogenesis and reduced recruitment of myeloid host cells [152]. Recently, using a transgenic model for breast cancers, the administration of CTCE-9908 resulted in a 30–40% decrease in lung metastasis [154]. Chemotherapy drugs or an anti-angiogenic agent in combination with CTCE-9908 have been shown to possess enhanced anti-tumorigenic and metastatic effects [154]. Currently, CTCE-9908 is undergoing Phase 1 and Phase II clinical trials in hepatocellular carcinoma [149]. Other molecules, such as CD26/dipeptidyl IV (DPPIV), MMPs and serine proteases, have been shown to regulate CXCL12/CXCR4 pathway by cleaving CXCL12 rendering it inactive [155]. Recent studies from our group suggest that synthetic non-psycho active cannabinoids inhibit breast or lung cancer growth and metastasis. Synthetic cannabinoids CP55,940 and WIN55,212–2, which bind to cannabinoid receptor CB1 and cannabinoid receptor CB2, as well as the CB2-selective agonist JWH-015, inhibited CXCL12-induced and CXCR4-mediated chemotaxis and invasion of Jurkat T cells [156]. These synthetic cannabinoids have also been shown to inhibit tumor growth in transgenic breast cancer tumor model systems [157, 158]. We have also recently shown that Slit-2, which is hyper methylated and inactivated in breast and lung cancer cells, inhibits CXCL12/CXCR4-induced breast cancer cell chemotaxis, chemoinvasion, and adhesion, the fundamental components that promote metastasis [20, 159, 160]. Slit, which has been reported to play an important role in axonal movement [161–165], was recently shown to modulate CXCR4-induced leukocyte trafficking [166]. Slit is a ~200 kDa secretory protein that consists of a family of three genes: Slit 1, Slit 2 and Slit 3. Slit, which mediates its function by binding to the Roundabout (Robo) was shown to inhibit CXCL12induced tyrosine phosphorylation of focal adhesion components such as RAFTK/ Pyk2 at residues 580 and 881, focal adhesion kinase at residue 576, and paxillin [20]. Recent studies from Hinck’s group found that loss of Slits in murine mammary gland or human breast carcinoma cells result in up-regulation of the CXCL12 and

9

Role of CXCL12 and CXCR4 in Tumor Biology and Metastasis

235

CXCR4 signaling axis, specifically within mammary epithelium [167]. In addition, inverse correlation between Slit and CXCR4 expression is observed in human breast tumor tissues [167]. Furthermore, Slit over-expression in breast cancer cell line MDA-MB-231, was shown to significantly suppress tumor growth in in vivo mouse models [167]. Furthermore, they have shown that Slit2 may inhibit tumor growth by down-modulating CXCR4 expression [167]. We have also shown that Slit2 inhibits tumor growth in MCF-7 cells by modulating b-catenin/TCF pathway [159]. Our group also showed that Slit2 inhibited CXCL12/CXCR4-mediated chemotaxis, chemoinvasion and adhesion by decreasing the activation of RAFTK/Pyk2, FAK and paxillin [160]. These findings suggest that Slits may act as a negative regulator of CXCL12/CXCR4 axis in tumor development.

9.8

Conclusive Remarks

Chemokine receptor CXCR4 and its ligand CXCL12 have been shown to be crucial for tumor proliferation, migration, adhesion, angiogenesis, and metastasis. CXCR4/CXCL12 axis has been shown to possess a multi-faceted role in diverse processes of tumor progression and metastasis through regulation of cancer stem cells, modulation of tumor stroma, enhancing angiogenesis, formation of premetastatic niche and organ selective metastasis. Therefore, it is of utmost importance to understand the mechanisms by which CXCL12/CXCR4 modulate microenvironment to enhance growth and help in homing and progression of cancer cells to different organs. This will help to identify novel targets for development of therapies to inhibit both growth and metastasis against various types of cancers, including breast, liver, prostate, pancreas, and leukemia. This is especially imperative considering the resistance of various cancers to chemotherapy and the poor prognosis of CXCR4 positive cancer patients. Furthermore, identification of CXCR4 as an early novel biomarker and therapeutic target may lead to the improvement of the overall survival of cancer patients.

References 1. Otsuka S, Bebb G (2008) The CXCR4/SDF-1 chemokine receptor axis: a new target therapeutic for non-small cell lung cancer. J Thorac Oncol 3:1379–1383 2. Richmond A (2010) Chemokine modulation of the tumor microenvironment. Pigment Cell Melanoma Res 23:312–313 3. Teicher BA, Fricker SP (2010) CXCL12 (SDF-1)/CXCR4 pathway in cancer. Clin Cancer Res 16:2927–2931 4. Ganju RK, Brubaker SA, Meyer J et al (1998) The alpha-chemokine, stromal cell-derived factor-1alpha, binds to the transmembrane G-protein-coupled CXCR-4 receptor and activates multiple signal transduction pathways. J Biol Chem 273:23169–23175 5. Balkwill F (2004) Cancer and the chemokine network. Nat Rev Cancer 4:540–550 6. Lagerstrom MC, Schioth HB (2008) Structural diversity of G protein-coupled receptors and significance for drug discovery. Nat Rev Drug Discov 7:339–357

236

R.K. Ganju et al.

7. Viola A, Luster AD (2008) Chemokines and their receptors: drug targets in immunity and inflammation. Annu Rev Pharmacol Toxicol 48:171–197 8. Goldsmith ZG, Dhanasekaran DN (2007) G protein regulation of MAPK networks. Oncogene 26:3122–3142 9. Mellado M, Rodriguez-Frade JM, Manes S et al (2001) Chemokine signaling and functional responses: the role of receptor dimerization and TK pathway activation. Annu Rev Immunol 19:397–421 10. Bendall LJ, Baraz R, Juarez J et al (2005) Defective p38 mitogen-activated protein kinase signaling impairs chemotaxic but not proliferative responses to stromal-derived factor-1alpha in acute lymphoblastic leukemia. Cancer Res 65:3290–3298 11. Cabioglu N, Summy J, Miller C et al (2005) CXCL-12/stromal cell-derived factor-1alpha transactivates HER2-neu in breast cancer cells by a novel pathway involving Src kinase activation. Cancer Res 65:6493–6497 12. Helbig G, Christopherson KW 2nd, Bhat-Nakshatri P et al (2003) NF-kappaB promotes breast cancer cell migration and metastasis by inducing the expression of the chemokine receptor CXCR4. J Biol Chem 278:21631–21638 13. Suzuki Y, Rahman M, Mitsuya H (2001) Diverse transcriptional response of CD4(+) T cells to stromal cell-derived factor (SDF)-1: cell survival promotion and priming effects of SDF-1 on CD4(+) T cells. J Immunol 167:3064–3073 14. Ward SG (2006) T lymphocytes on the move: chemokines, PI 3-kinase and beyond. Trends Immunol 27:80–87 15. Barbero S, Bonavia R, Bajetto A et al (2003) Stromal cell-derived factor 1alpha stimulates human glioblastoma cell growth through the activation of both extracellular signal-regulated kinases 1/2 and Akt. Cancer Res 63:1969–1974 16. Wang JF, Park IW, Groopman JE (2000) Stromal cell-derived factor-1alpha stimulates tyrosine phosphorylation of multiple focal adhesion proteins and induces migration of hematopoietic progenitor cells: roles of phosphoinositide-3 kinase and protein kinase C. Blood 95:2505–2513 17. Zhang XF, Wang JF, Matczak E et al (2001) Janus kinase 2 is involved in stromal cellderived factor-1alpha-induced tyrosine phosphorylation of focal adhesion proteins and migration of hematopoietic progenitor cells. Blood 97:3342–3348 18. Fernandis AZ, Prasad A, Band H et al (2004) Regulation of CXCR4-mediated chemotaxis and chemoinvasion of breast cancer cells. Oncogene 23:157–167 19. Ganju RK, Hatch WC, Avraham H et al (1997) RAFTK, a novel member of the focal adhesion kinase family, is phosphorylated and associates with signaling molecules upon activation of mature T lymphocytes. J Exp Med 185:1055–1063 20. Prasad A, Fernandis AZ, Rao Y et al (2004) Slit protein-mediated inhibition of CXCR4induced chemotactic and chemoinvasive signaling pathways in breast cancer cells. J Biol Chem 279:9115–9124 21. Chernock RD, Cherla RP, Ganju RK (2001) SHP2 and cbl participate in alpha-chemokine receptor CXCR4-mediated signaling pathways. Blood 97:608–615 22. Smit L, van der Horst G, Borst J (1996) Formation of Shc/Grb2- and Crk adaptor complexes containing tyrosine phosphorylated Cbl upon stimulation of the B-cell antigen receptor. Oncogene 13:381–389 23. Thien CB, Langdon WY (2001) Cbl: many adaptations to regulate protein tyrosine kinases. Nat Rev Mol Cell Biol 2:294–307 24. Peschard P, Park M (2003) Escape from Cbl-mediated downregulation: a recurrent theme for oncogenic deregulation of receptor tyrosine kinases. Cancer Cell 3:519–523 25. Thien CB, Walker F, Langdon WY (2001) RING finger mutations that abolish c-Cbl-directed polyubiquitination and downregulation of the EGF receptor are insufficient for cell transformation. Mol Cell 7:355–365 26. Chiang SH, Baumann CA, Kanzaki M et al (2001) Insulin-stimulated GLUT4 translocation requires the CAP-dependent activation of TC10. Nature 410:944–948 27. Hartley D, Corvera S (1996) Formation of c-Cbl.phosphatidylinositol 3-kinase complexes on lymphocyte membranes by a p56lck-independent mechanism. J Biol Chem 271:21939–21943

9

Role of CXCL12 and CXCR4 in Tumor Biology and Metastasis

237

28. Kimura A, Baumann CA, Chiang SH et al (2001) The sorbin homology domain: a motif for the targeting of proteins to lipid rafts. Proc Natl Acad Sci USA 98:9098–9103 29. Gomez-Mouton C, Abad JL, Mira E et al (2001) Segregation of leading-edge and uropod components into specific lipid rafts during T cell polarization. Proc Natl Acad Sci USA 98:9642–9647 30. Krawczyk C, Bachmaier K, Sasaki T et al (2000) Cbl-b is a negative regulator of receptor clustering and raft aggregation in T cells. Immunity 13:463–473 31. Manes S, Mira E, Gomez-Mouton C et al (1999) Membrane raft microdomains mediate front-rear polarity in migrating cells. EMBO J 18:6211–6220 32. Meng F, Lowell CA (1998) A beta 1 integrin signaling pathway involving Src-family kinases, Cbl and PI-3 kinase is required for macrophage spreading and migration. EMBO J 17:4391–4403 33. Scaife RM, Langdon WY (2000) c-Cbl localizes to actin lamellae and regulates lamellipodia formation and cell morphology. J Cell Sci 113(Pt 2):215–226 34. Kukreja P, Abdel-Mageed AB, Mondal D et al (2005) Up-regulation of CXCR4 expression in PC-3 cells by stromal-derived factor-1alpha (CXCL12) increases endothelial adhesion and transendothelial migration: role of MEK/ERK signaling pathway-dependent NF-kappaB activation. Cancer Res 65:9891–9898 35. Chinni SR, Sivalogan S, Dong Z et al (2006) CXCL12/CXCR4 signaling activates Akt-1 and MMP-9 expression in prostate cancer cells: the role of bone microenvironment-associated CXCL12. Prostate 66:32–48 36. Wang J, Sun Y, Song W et al (2005) Diverse signaling pathways through the SDF-1/CXCR4 chemokine axis in prostate cancer cell lines leads to altered patterns of cytokine secretion and angiogenesis. Cell Signal 17:1578–1592 37. Serrati S, Margheri F, Fibbi G et al (2008) Endothelial cells and normal breast epithelial cells enhance invasion of breast carcinoma cells by CXCR-4-dependent up-regulation of urokinase-type plasminogen activator receptor (uPAR, CD87) expression. J Pathol 214:545–554 38. Ahmad A, Kong D, Wang Z et al (2009) Down-regulation of uPA and uPAR by 3,3¢diindolylmethane contributes to the inhibition of cell growth and migration of breast cancer cells. J Cell Biochem 108:916–925 39. Dass K, Ahmad A, Azmi AS et al (2008) Evolving role of uPA/uPAR system in human cancers. Cancer Treat Rev 34:122–136 40. Ngo HT, Leleu X, Lee J et al (2008) SDF-1/CXCR4 and VLA-4 interaction regulates homing in Waldenstrom macroglobulinemia. Blood 112:150–158 41. Sanz-Rodriguez F, Hidalgo A, Teixido J (2001) Chemokine stromal cell-derived factor1alpha modulates VLA-4 integrin-mediated multiple myeloma cell adhesion to CS-1/ fibronectin and VCAM-1. Blood 97:346–351 42. Engl T, Relja B, Blumenberg C et al (2006) Prostate tumor CXC-chemokine profile correlates with cell adhesion to endothelium and extracellular matrix. Life Sci 78:1784–1793 43. Engl T, Relja B, Marian D et al (2006) CXCR4 chemokine receptor mediates prostate tumor cell adhesion through alpha5 and beta3 integrins. Neoplasia 8:290–301 44. Ueda Y, Neel NF, Schutyser E et al (2006) Deletion of the COOH-terminal domain of CXC chemokine receptor 4 leads to the down-regulation of cell-to-cell contact, enhanced motility and proliferation in breast carcinoma cells. Cancer Res 66:5665–5675 45. Rubin JB (2009) Chemokine signaling in cancer: one hump or two? Semin Cancer Biol 19:116–122 46. Mehta SA, Christopherson KW, Bhat-Nakshatri P et al (2007) Negative regulation of chemokine receptor CXCR4 by tumor suppressor p53 in breast cancer cells: implications of p53 mutation or isoform expression on breast cancer cell invasion. Oncogene 26:3329–3337 47. Moskovits N, Kalinkovich A, Bar J et al (2006) p53 Attenuates cancer cell migration and invasion through repression of SDF-1/CXCL12 expression in stromal fibroblasts. Cancer Res 66:10671–10676 48. Addadi Y, Moskovits N, Granot D et al (2010) p53 status in stromal fibroblasts modulates tumor growth in an SDF1-dependent manner. Cancer Res 70:9650–9658

238

R.K. Ganju et al.

49. Porcile C, Bajetto A, Barbieri F et al (2005) Stromal cell-derived factor-1alpha (SDF-1alpha/ CXCL12) stimulates ovarian cancer cell growth through the EGF receptor transactivation. Exp Cell Res 308:241–253 50. Hernandez L, Smirnova T, Kedrin D et al (2009) The EGF/CSF-1 paracrine invasion loop can be triggered by heregulin beta1 and CXCL12. Cancer Res 69:3221–3227 51. Cabioglu N, Yazici MS, Arun B et al (2005) CCR7 and CXCR4 as novel biomarkers predicting axillary lymph node metastasis in T1 breast cancer. Clin Cancer Res 11:5686–5693 52. Li YM, Pan Y, Wei Y et al (2004) Upregulation of CXCR4 is essential for HER2-mediated tumor metastasis. Cancer Cell 6:459–469 53. Phillips RJ, Mestas J, Gharaee-Kermani M et al (2005) Epidermal growth factor and hypoxiainduced expression of CXC chemokine receptor 4 on non-small cell lung cancer cells is regulated by the phosphatidylinositol 3-kinase/PTEN/AKT/mammalian target of rapamycin signaling pathway and activation of hypoxia inducible factor-1alpha. J Biol Chem 280:22473–22481 54. Akekawatchai C, Holland JD, Kochetkova M et al (2005) Transactivation of CXCR4 by the insulin-like growth factor-1 receptor (IGF-1R) in human MDA-MB-231 breast cancer epithelial cells. J Biol Chem 280:39701–39708 55. Kojima Y, Acar A, Eaton EN et al (2010) Autocrine TGF-beta and stromal cell-derived factor-1 (SDF-1) signaling drives the evolution of tumor-promoting mammary stromal myofibroblasts. Proc Natl Acad Sci USA 107:20009–20014 56. Hall JM, Korach KS (2003) Stromal cell-derived factor 1, a novel target of estrogen receptor action, mediates the mitogenic effects of estradiol in ovarian and breast cancer cells. Mol Endocrinol 17:792–803 57. Rhodes LV, Short SP, Neel NF et al (2011) Cytokine receptor CXCR4 mediates estrogenindependent tumorigenesis, metastasis, and resistance to endocrine therapy in human breast cancer. Cancer Res 71:603–613 58. Sun X, Cheng G, Hao M et al (2010) CXCL12/CXCR4/CXCR7 chemokine axis and cancer progression. Cancer Metastasis Rev 29:709–722 59. Fischer T, Nagel F, Jacobs S et al (2008) Reassessment of CXCR4 chemokine receptor expression in human normal and neoplastic tissues using the novel rabbit monoclonal antibody UMB-2. PLoS One 3:e4069 60. Schmid BC, Rudas M, Rezniczek GA et al (2004) CXCR4 is expressed in ductal carcinoma in situ of the breast and in atypical ductal hyperplasia. Breast Cancer Res Treat 84:247–250 61. Luker KE, Luker GD (2006) Functions of CXCL12 and CXCR4 in breast cancer. Cancer Lett 238:30–41 62. Kato M, Kitayama J, Kazama S et al (2003) Expression pattern of CXC chemokine receptor-4 is correlated with lymph node metastasis in human invasive ductal carcinoma. Breast Cancer Res 5:R144–R150 63. Allinen M, Beroukhim R, Cai L et al (2004) Molecular characterization of the tumor microenvironment in breast cancer. Cancer Cell 6:17–32 64. Chu QD, Panu L, Holm NT et al (2010) High chemokine receptor CXCR4 level in triple negative breast cancer specimens predicts poor clinical outcome. J Surg Res 159:689–695 65. Kang Y (2005) Functional genomic analysis of cancer metastasis: biologic insights and clinical implications. Expert Rev Mol Diagn 5:385–395 66. Chiang AC, Massague J (2008) Molecular basis of metastasis. N Engl J Med 359:2814–2823 67. Yang J, Mani SA, Weinberg RA (2006) Exploring a new twist on tumor metastasis. Cancer Res 66:4549–4552 68. Kim H, Muller WJ (1999) The role of the epidermal growth factor receptor family in mammary tumorigenesis and metastasis. Exp Cell Res 253:78–87 69. Price JT, Bonovich MT, Kohn EC (1997) The biochemistry of cancer dissemination. Crit Rev Biochem Mol Biol 32:175–253

9

Role of CXCL12 and CXCR4 in Tumor Biology and Metastasis

239

70. van Golen KL (2003) Inflammatory breast cancer: relationship between growth factor signaling and motility in aggressive cancers. Breast Cancer Res 5:174–179 71. Harmey JH, Bucana CD, Lu W et al (2002) Lipopolysaccharide-induced metastatic growth is associated with increased angiogenesis, vascular permeability and tumor cell invasion. Int J Cancer 101:415–422 72. Muller A, Homey B, Soto H et al (2001) Involvement of chemokine receptors in breast cancer metastasis. Nature 410:50–56 73. Verbeek BS, Adriaansen-Slot SS, Vroom TM et al (1998) Overexpression of EGFR and c-erbB2 causes enhanced cell migration in human breast cancer cells and NIH3T3 fibroblasts. FEBS Lett 425:145–150 74. Kang H, Watkins G, Douglas-Jones A et al (2005) The elevated level of CXCR4 is correlated with nodal metastasis of human breast cancer. Breast 14:360–367 75. Hinton CV, Avraham S, Avraham HK (2010) Role of the CXCR4/CXCL12 signaling axis in breast cancer metastasis to the brain. Clin Exp Metastasis 27:97–105 76. Murphy PM (2001) Chemokines and the molecular basis of cancer metastasis. N Engl J Med 345:833–835 77. Chen Y, Stamatoyannopoulos G, Song CZ (2003) Down-regulation of CXCR4 by inducible small interfering RNA inhibits breast cancer cell invasion in vitro. Cancer Res 63:4801–4804 78. Tamamura H, Hori A, Kanzaki N et al (2003) T140 analogs as CXCR4 antagonists identified as anti-metastatic agents in the treatment of breast cancer. FEBS Lett 550:79–83 79. Kang Y, Siegel PM, Shu W et al (2003) A multigenic program mediating breast cancer metastasis to bone. Cancer Cell 3:537–549 80. Liang Z, Wu H, Reddy S et al (2007) Blockade of invasion and metastasis of breast cancer cells via targeting CXCR4 with an artificial microRNA. Biochem Biophys Res Commun 363:542–546 81. Cabioglu N, Sahin AA, Morandi P et al (2009) Chemokine receptors in advanced breast cancer: differential expression in metastatic disease sites with diagnostic and therapeutic implications. Ann Oncol 20:1013–1019 82. Kim R, Arihiro K, Emi M et al (2006) Potential role of HER-2; in primary breast tumor with bone metastasis. Oncol Rep 15:1477–1484 83. Hirbe AC, Morgan EA, Weilbaecher KN (2010) The CXCR4/SDF-1 chemokine axis: a potential therapeutic target for bone metastases? Curr Pharm Des 16:1284–1290 84. Struckmann K, Mertz K, Steu S et al (2008) pVHL co-ordinately regulates CXCR4/CXCL12 and MMP2/MMP9 expression in human clear-cell renal cell carcinoma. J Pathol 214:464–471 85. Schioppa T, Uranchimeg B, Saccani A et al (2003) Regulation of the chemokine receptor CXCR4 by hypoxia. J Exp Med 198:1391–1402 86. Staller P, Sulitkova J, Lisztwan J et al (2003) Chemokine receptor CXCR4 downregulated by von Hippel-Lindau tumour suppressor pVHL. Nature 425:307–311 87. Bachelder RE, Wendt MA, Mercurio AM (2002) Vascular endothelial growth factor promotes breast carcinoma invasion in an autocrine manner by regulating the chemokine receptor CXCR4. Cancer Res 62:7203–7206 88. Castellone MD, Guarino V, De Falco V et al (2004) Functional expression of the CXCR4 chemokine receptor is induced by RET/PTC oncogenes and is a common event in human papillary thyroid carcinomas. Oncogene 23:5958–5967 89. Tarnowski M, Grymula K, Reca R et al (2010) Regulation of expression of stromal-derived factor-1 receptors: CXCR4 and CXCR7 in human rhabdomyosarcomas. Mol Cancer Res 8:1–14 90. Su L, Zhang J, Xu H et al (2005) Differential expression of CXCR4 is associated with the metastatic potential of human non-small cell lung cancer cells. Clin Cancer Res 11:8273–8280

240

R.K. Ganju et al.

91. Na IK, Scheibenbogen C, Adam C et al (2008) Nuclear expression of CXCR4 in tumor cells of non-small cell lung cancer is correlated with lymph node metastasis. Hum Pathol 39:1751–1755 92. Wagner PL, Hyjek E, Vazquez MF et al (2009) CXCL12 and CXCR4 in adenocarcinoma of the lung: association with metastasis and survival. J Thorac Cardiovasc Surg 137:615–621 93. Su LP, Zhang JP, Xu HB et al (2005) The role of CXCR4 in lung cancer metastasis and its possible mechanism. Zhonghua Yi Xue Za Zhi 85:1190–1194 94. Chen G, Wang Z, Liu XY et al (2011) High-level CXCR4 expression correlates with brainspecific metastasis of non-small cell lung cancer. World J Surg 35:56–61 95. Liu YL, Yu JM, Song XR et al (2006) Regulation of the chemokine receptor CXCR4 and metastasis by hypoxia-inducible factor in non small cell lung cancer cell lines. Cancer Biol Ther 5:1320–1326 96. Burger JA, Kipps TJ (2006) CXCR4: a key receptor in the crosstalk between tumor cells and their microenvironment. Blood 107:1761–1767 97. Burger M, Glodek A, Hartmann T et al (2003) Functional expression of CXCR4 (CD184) on small-cell lung cancer cells mediates migration, integrin activation, and adhesion to stromal cells. Oncogene 22:8093–8101 98. Hartmann TN, Burger JA, Glodek A et al (2005) CXCR4 chemokine receptor and integrin signaling co-operate in mediating adhesion and chemoresistance in small cell lung cancer (SCLC) cells. Oncogene 24:4462–4471 99. Kijima T, Maulik G, Ma PC et al (2002) Regulation of cellular proliferation, cytoskeletal function, and signal transduction through CXCR4 and c-Kit in small cell lung cancer cells. Cancer Res 62:6304–6311 100. Molina JR, Adjei AA, Jett JR (2006) Advances in chemotherapy of non-small cell lung cancer. Chest 130:1211–1219 101. Kryczek I, Wei S, Keller E et al (2007) Stroma-derived factor (SDF-1/CXCL12) and human tumor pathogenesis. Am J Physiol Cell Physiol 292:C987–C995 102. Koshiba T, Hosotani R, Miyamoto Y et al (2000) Expression of stromal cell-derived factor 1 and CXCR4 ligand receptor system in pancreatic cancer: a possible role for tumor progression. Clin Cancer Res 6:3530–3535 103. Taichman RS, Cooper C, Keller ET et al (2002) Use of the stromal cell-derived factor-1/ CXCR4 pathway in prostate cancer metastasis to bone. Cancer Res 62:1832–1837 104. Schimanski CC, Schwald S, Simiantonaki N et al (2005) Effect of chemokine receptors CXCR4 and CCR7 on the metastatic behavior of human colorectal cancer. Clin Cancer Res 11:1743–1750 105. Tachibana K, Hirota S, Iizasa H et al (1998) The chemokine receptor CXCR4 is essential for vascularization of the gastrointestinal tract. Nature 393:591–594 106. Kim M, Koh YJ, Kim KE et al (2010) CXCR4 signaling regulates metastasis of chemoresistant melanoma cells by a lymphatic metastatic niche. Cancer Res 70:10411–10421 107. Xiang Z, Zeng Z, Tang Z et al (2009) Increased expression of vascular endothelial growth factor-C and nuclear CXCR4 in hepatocellular carcinoma is correlated with lymph node metastasis and poor outcome. Cancer J 15:519–525 108. Speetjens FM, Liefers GJ, Korbee CJ et al (2009) Nuclear localization of CXCR4 determines prognosis for colorectal cancer patients. Cancer Microenviron 2:1–7 109. Wang L, Wang Z, Yang B et al (2009) CXCR4 nuclear localization follows binding of its ligand SDF-1 and occurs in metastatic but not primary renal cell carcinoma. Oncol Rep 22:1333–1339 110. Wang L, Yang B, Yang Q et al (2009) Strong expression of chemokine receptor CXCR4 by renal cell carcinoma cells correlates with metastasis. Clin Exp Metastasis 26:1049–1054 111. Wang LH, Liu Q, Xu B et al (2010) Identification of nuclear localization sequence of CXCR4 in renal cell carcinoma by constructing expression plasmids of different deletants. Plasmid 63:68–72 112. Sun YX, Wang J, Shelburne CE et al (2003) Expression of CXCR4 and CXCL12 (SDF-1) in human prostate cancers (PCa) in vivo. J Cell Biochem 89:462–473

9

Role of CXCL12 and CXCR4 in Tumor Biology and Metastasis

241

113. Sun YX, Schneider A, Jung Y et al (2005) Skeletal localization and neutralization of the SDF-1(CXCL12)/CXCR4 axis blocks prostate cancer metastasis and growth in osseous sites in vivo. J Bone Miner Res 20:318–329 114. Gladson CL, Welch DR (2008) New insights into the role of CXCR4 in prostate cancer metastasis. Cancer Biol Ther 7:1849–1851 115. Kaplan RN, Psaila B, Lyden D (2006) Bone marrow cells in the ‘pre-metastatic niche’: within bone and beyond. Cancer Metastasis Rev 25:521–529 116. Raman D, Baugher PJ, Thu YM et al (2007) Role of chemokines in tumor growth. Cancer Lett 256:137–165 117. Orimo A, Gupta PB, Sgroi DC et al (2005) Stromal fibroblasts present in invasive human breast carcinomas promote tumor growth and angiogenesis through elevated SDF-1/CXCL12 secretion. Cell 121:335–348 118. Orimo A, Weinberg RA (2006) Stromal fibroblasts in cancer: a novel tumor-promoting cell type. Cell Cycle 5:1597–1601 119. Ostman A, Augsten M (2009) Cancer-associated fibroblasts and tumor growth–bystanders turning into key players. Curr Opin Genet Dev 19:67–73 120. Coussens LM, Werb Z (2002) Inflammation and cancer. Nature 420:860–867 121. Jacobs TW, Byrne C, Colditz G et al (1999) Radial scars in benign breast-biopsy specimens and the risk of breast cancer. N Engl J Med 340:430–436 122. Mueller MM, Fusenig NE (2004) Friends or foes – bipolar effects of the tumour stroma in cancer. Nat Rev Cancer 4:839–849 123. Serini G, Gabbiani G (1999) Mechanisms of myofibroblast activity and phenotypic modulation. Exp Cell Res 250:273–283 124. Kalluri R, Zeisberg M (2006) Fibroblasts in cancer. Nat Rev Cancer 6:392–401 125. Olumi AF, Grossfeld GD, Hayward SW et al (1999) Carcinoma-associated fibroblasts direct tumor progression of initiated human prostatic epithelium. Cancer Res 59:5002–5011 126. Sheridan C, Kishimoto H, Fuchs RK et al (2006) CD44+/CD24- breast cancer cells exhibit enhanced invasive properties: an early step necessary for metastasis. Breast Cancer Res 8:R59 127. Mantovani A, Allavena P, Sica A et al (2008) Cancer-related inflammation. Nature 454:436–444 128. Liu BY, Soloviev I, Chang P et al (2010) Stromal cell-derived factor-1/CXCL12 contributes to MMTV-Wnt1 tumor growth involving Gr1+CD11b+ cells. PLoS One 5:e8611 129. Hiratsuka S, Duda DG, Huang Y et al (2011) C-X-C receptor type 4 promotes metastasis by activating p38 mitogen-activated protein kinase in myeloid differentiation antigen (Gr-1)positive cells. Proc Natl Acad Sci USA 108:302–307 130. Nagasawa T, Hirota S, Tachibana K et al (1996) Defects of B-cell lymphopoiesis and bonemarrow myelopoiesis in mice lacking the CXC chemokine PBSF/SDF-1. Nature 382:635–638 131. Darash-Yahana M, Pikarsky E, Abramovitch R et al (2004) Role of high expression levels of CXCR4 in tumor growth, vascularization, and metastasis. FASEB J 18:1240–1242 132. Beider K, Abraham M, Begin M et al (2009) Interaction between CXCR4 and CCL20 pathways regulates tumor growth. PLoS One 4:e5125 133. Wang J, Dai J, Jung Y et al (2007) A glycolytic mechanism regulating an angiogenic switch in prostate cancer. Cancer Res 67:149–159 134. Yang L, DeBusk LM, Fukuda K et al (2004) Expansion of myeloid immune suppressor Gr+CD11b+ cells in tumor-bearing host directly promotes tumor angiogenesis. Cancer Cell 6:409–421 135. Lyden D, Hattori K, Dias S et al (2001) Impaired recruitment of bone-marrow-derived endothelial and hematopoietic precursor cells blocks tumor angiogenesis and growth. Nat Med 7:1194–1201 136. De Palma M, Venneri MA, Galli R et al (2005) Tie2 identifies a hematopoietic lineage of proangiogenic monocytes required for tumor vessel formation and a mesenchymal population of pericyte progenitors. Cancer Cell 8:211–226

242

R.K. Ganju et al.

137. Lin EY, Li JF, Gnatovskiy L et al (2006) Macrophages regulate the angiogenic switch in a mouse model of breast cancer. Cancer Res 66:11238–11246 138. Purhonen S, Palm J, Rossi D et al (2008) Bone marrow-derived circulating endothelial precursors do not contribute to vascular endothelium and are not needed for tumor growth. Proc Natl Acad Sci USA 105:6620–6625 139. Wels J, Kaplan RN, Rafii S et al (2008) Migratory neighbors and distant invaders: tumorassociated niche cells. Genes Dev 22:559–574 140. Guleng B, Tateishi K, Ohta M et al (2005) Blockade of the stromal cell-derived factor-1/ CXCR4 axis attenuates in vivo tumor growth by inhibiting angiogenesis in a vascular endothelial growth factor-independent manner. Cancer Res 65:5864–5871 141. Burger JA, Peled A (2009) CXCR4 antagonists: targeting the microenvironment in leukemia and other cancers. Leukemia 23:43–52 142. De Clercq E, Yamamoto N, Pauwels R et al (1992) Potent and selective inhibition of human immunodeficiency virus (HIV)-1 and HIV-2 replication by a class of bicyclams interacting with a viral uncoating event. Proc Natl Acad Sci USA 89:5286–5290 143. De Clercq E, Yamamoto N, Pauwels R et al (1994) Highly potent and selective inhibition of human immunodeficiency virus by the bicyclam derivative JM3100. Antimicrob Agents Chemother 38:668–674 144. De Clercq E (2003) The bicyclam AMD3100 story. Nat Rev Drug Discov 2:581–587 145. Nimmagadda S, Pullambhatla M, Stone K et al (2010) Molecular imaging of CXCR4 receptor expression in human cancer xenografts with [64Cu]AMD3100 positron emission tomography. Cancer Res 70:3935–3944 146. Burger JA, Stewart DJ (2009) CXCR4 chemokine receptor antagonists: perspectives in SCLC. Expert Opin Investig Drugs 18:481–490 147. Yoon Y, Liang Z, Zhang X et al (2007) CXC chemokine receptor-4 antagonist blocks both growth of primary tumor and metastasis of head and neck cancer in xenograft mouse models. Cancer Res 67:7518–7524 148. Liang Z, Yoon Y, Votaw J et al (2005) Silencing of CXCR4 blocks breast cancer metastasis. Cancer Res 65:967–971 149. Wong D, Korz W (2008) Translating an Antagonist of Chemokine Receptor CXCR4: from bench to bedside. Clin Cancer Res 14:7975–7980 150. Wong RS, Bodart V, Metz M et al (2008) Comparison of the potential multiple binding modes of bicyclam, monocylam, and noncyclam small-molecule CXC chemokine receptor 4 inhibitors. Mol Pharmacol 74:1485–1495 151. Kim SY, Lee CH, Midura BV et al (2008) Inhibition of the CXCR4/CXCL12 chemokine pathway reduces the development of murine pulmonary metastases. Clin Exp Metastasis 25:201–211 152. Porvasnik S, Sakamoto N, Kusmartsev S et al (2009) Effects of CXCR4 antagonist CTCE9908 on prostate tumor growth. Prostate 69:1460–1469 153. Richert MM, Vaidya KS, Mills CN et al (2009) Inhibition of CXCR4 by CTCE-9908 inhibits breast cancer metastasis to lung and bone. Oncol Rep 21:761–767 154. Hassan S, Buchanan M, Jahan K et al (2011) CXCR4 peptide antagonist inhibits primary breast tumor growth, metastasis and enhances the efficacy of anti-VEGF treatment or docetaxel in a transgenic mouse model. Int J Cancer 129(1):225–232 155. Sun YX, Pedersen EA, Shiozawa Y et al (2008) CD26/dipeptidyl peptidase IV regulates prostate cancer metastasis by degrading SDF-1/CXCL12. Clin Exp Metastasis 25:765–776 156. Ghosh S, Preet A, Groopman JE et al (2006) Cannabinoid receptor CB2 modulates the CXCL12/CXCR4-mediated chemotaxis of T lymphocytes. Mol Immunol 43:2169–2179 157. Qamri Z, Preet A, Nasser MW et al (2009) Synthetic cannabinoid receptor agonists inhibit tumor growth and metastasis of breast cancer. Mol Cancer Ther 8:3117–3129 158. Caffarel MM, Andradas C, Mira E et al (2010) Cannabinoids reduce ErbB2-driven breast cancer progression through Akt inhibition. Mol Cancer 9:196 159. Prasad A, Paruchuri V, Preet A et al (2008) Slit-2 induces a tumor-suppressive effect by regulating beta-catenin in breast cancer cells. J Biol Chem 283:26624–26633

9

Role of CXCL12 and CXCR4 in Tumor Biology and Metastasis

243

160. Prasad A, Qamri Z, Wu J et al (2007) Slit-2/Robo-1 modulates the CXCL12/CXCR4-induced chemotaxis of T cells. J Leukoc Biol 82:465–476 161. Bashaw GJ, Goodman CS (1999) Chimeric axon guidance receptors: the cytoplasmic domains of slit and netrin receptors specify attraction versus repulsion. Cell 97:917–926 162. Brose K, Tessier-Lavigne M (2000) Slit proteins: key regulators of axon guidance, axonal branching, and cell migration. Curr Opin Neurobiol 10:95–102 163. Li HS, Chen JH, Wu W et al (1999) Vertebrate slit, a secreted ligand for the transmembrane protein roundabout, is a repellent for olfactory bulb axons. Cell 96:807–818 164. Ringstedt T, Braisted JE, Brose K et al (2000) Slit inhibition of retinal axon growth and its role in retinal axon pathfinding and innervation patterns in the diencephalon. J Neurosci 20:4983–4991 165. Wu W, Wong K, Chen J et al (1999) Directional guidance of neuronal migration in the olfactory system by the protein Slit. Nature 400:331–336 166. Wu JY, Feng L, Park HT et al (2001) The neuronal repellent Slit inhibits leukocyte chemotaxis induced by chemotactic factors. Nature 410:948–952 167. Marlow R, Strickland P, Lee JS et al (2008) SLITs suppress tumor growth in vivo by silencing Sdf1/Cxcr4 within breast epithelium. Cancer Res 68:7819–7827

Chapter 10

Transcription Factors Stat5a/b and Stat3 in Prostate Cancer Growth and Metastases Tuomas K. Mirtti, Pooja Talati, and Marja T. Nevalainen

Abstract The majority of prostate cancer fatalities are caused by development of castration-resistant growth and metastatic spread of the primary tumor. There are currently no effective treatments for advanced prostate cancer and therefore it is imperative to develop more effective therapies for prostate cancer. Transcription factors Stat5a/b and Stat3 both represent potential therapeutic target proteins for advanced prostate cancer. Stat5 is the major signal transducer downstream of Prolactin (Prl) receptors in human prostate cancer cells, while Stat3 is known as the key mediator of the IL-6 effects. Stat5a/b and Stat3 both are constitutively active in clinical prostate cancers of high histological grades and in distant prostate cancer metastases. In addition, Stat5a/b and Stat3 both functionally interact with the androgen receptor in human prostate cancer cells. While both Stat5a/b and Stat3 bind to the same DNA consensus sequence, Stat5a/b regulates largely different genes from those that are regulated by Stat3 in human prostate cancer cells. Importantly, Stat5a/b

Tuomas K. Mirtti and Pooja Talati contributed equally to this chapter. T.K. Mirtti Department of Pathology, Haartman Institute, University of Helsinki, Helsinki 00014, Finland P. Talati Department of Cancer Biology, Kimmel Cancer Center, Thomas Jefferson University, Philadelphia, PA 19107, USA M.T. Nevalainen (*) Department of Cancer Biology, Kimmel Cancer Center, Thomas Jefferson University, Philadelphia, PA 19107, USA Department of Medical Oncology, Kimmel Cancer Center, Thomas Jefferson University, Philadelphia, PA 19107, USA Department of Urology, Kimmel Cancer Center, Thomas Jefferson University, Philadelphia, PA 19107, USA e-mail: [email protected]; [email protected]

A. Fatatis (ed.), Signaling Pathways and Molecular Mediators in Metastasis, DOI 10.1007/978-94-007-2558-4_10, © Springer Science+Business Media B.V. 2012

245

246

T.K. Mirtti et al.

has a preferential role over Stat3 in the promotion of prostate cancer cell viability and tumor growth. At the same time, both Stat5 and Stat3 stimulate metastatic behavior of human prostate cancer cells in vitro and in vivo. The Janus kinase family proteins (Jaks) are primary activators of Stat proteins. Inhibitors of Jak2 would target both transcription factors and theoretically provide control of prostate tumor growth as well as metastatic potential.

10.1

Introduction

Prostate cancer is the most common non-skin malignancy in the US and the second leading cause of cancer related deaths in males in the western countries [1]. The majority of prostate cancer fatalities are caused by metastatic spread of the primary tumor and development of castration resistant growth. The complex molecular mechanisms underlying progression of prostate cancer from organ-confined cancer to metastatic disease are largely unknown and, currently, there are no effective therapies for disseminated castration-resistant prostate cancer. Identification of the molecular changes associated with the progression of prostate cancer will provide a rational basis for the development of improved therapies for advanced disease. Currently, there is a lack of reliable markers that would identify those prostate cancers that are most likely to progress to aggressive disease and would benefit from extensive treatment. Here, we review the recent published data on transcription factors Stat3 and Stat5a/b in the regulation of prostate cancer growth and metastasis formation, as well as their interaction with the androgen receptor. The clinical importance of these transcription factors, as prognostic markers and therapeutic targets, will be discussed.

10.2

Structure and Function of STAT3 and STAT5a/b

Signal Transducer and Activator of Transcription (Stat)-3 and Stat5 belong to the seven member family of transcription factors (Stat1, Stat2, Stat3, Stat4, Stat5a, Stat5b and Stat6)[2]. The homology between the protein structures suggests that all Stats have a common genetic origin and have later diverged through duplications into three separate genetic loci: Stat1 and Stat4 in chromosome 2, Stat2 and Stat6 in chromosome 12 and Stat3, and Stat5a and Stat5b in chromosome 17 [3–7]. All the Stat proteins are coded by separate genes, even the highly homologous isoforms Stat5a and Stat5b [2]. The common genetic locus of Stat3 and Stat5a/b is chromosome 17 (bands q11–1 to q22), homologous to mouse chromosome 11[7–9]. The major structural features of all Stat proteins are similar. The N-terminal domain is involved in stabilizing interactions between two Stat dimers to form tetramers, thus facilitating the transcriptional activity of weak promoters. The coiled-coil domain facilitates protein-protein interactions by chaperone interactions

10

Transcription Factors Stat5a/b and Stat3 in Prostate Cancer Growth and Metastases

247

[10] and co-activator recruitment [11]. The DNA-binding domain interacts with the Gamma-interferon Activation Sequence (GAS) motifs in the target gene regulatory elements [12]. The linker domain stabilizes DNA-binding [13]. The SH2 (Src homology 2) domain mediates Stat dimerization and receptor-specific recruitment. During dimerization, a phosphorylated tyrosine residue of one Stat subunit binds to the SH2 domain of another Stat molecule [14]. In the C-terminal part of Stat proteins there is a transcriptional activation (TA) domain which interacts with critical co-activators and is directly involved in the initiation of transcription [15, 16]. Additionally, Stat5 proteins are post-translationally glycosylated on threonine 92 near the N-terminus, which allows interaction with the CREB-binding protein [17]. The highly homologous isoforms Stat5a and Stat5b differ in their C-terminal amino acid sequence. Stat5a (94-kDa) has 20 unique C-terminal amino acids while Stat5b (92 kDa) has eight distinct C-terminal amino acids in its sequence. Stat3 shares sequence homology with Stat5a/b, but the main differences are in the length and sequence of the TA domains [15]. Stat3 is activated by phosphorylation of tyrosine residue Y705 whereas Stat5a and Stat5b are activated at tyrosines Y694 and Y699, respectively [7, 18, 19]. Serine 727 phosphorylation supplements the activation of Stat3, whereas phosphorylation of serine 725 and serine 730 may negatively control the transcriptional activity of Stat5a and Stat5b, respectively [18, 20]. While the nuclear translocation of Stat dimers is an energy-dependent process [21, 22], there is recent evidence that non-phosphorylated Stat3 could regulate gene expression also [23, 24]. Ligands activating different Stat proteins vary considerably. Stat5a is the predominant signal transmitter of prolactin receptor activation, while Stat5b mediates growth hormone response [16]. Stat3 mediates primarily the action of interleukin-6 (IL-6) family proteins [25]. As the cytokine receptors that activate Stats usually lack tyrosine kinase activity, this activity has to be provided by other cytoplasmic kinases. The predominant kinases in Stat signaling are the Janus kinase family proteins (Jaks), Jak2 being the primary activator of both Stat5a/b and Stat3 [26, 27]. The cytokine Prl is locally expressed at high levels in high-grade prostate cancers [28], distant prostate cancer metastases, and castration-resistant prostate cancers [29]. In the canonical Prl-Jak2-Stat5a/b signaling pathway (Fig. 10.1), ligand binding-induced conformational changes in the prolactin receptor (PrlR) activate Jak2 molecules, which then phosphorylate tyrosine motifs in PrlR. Stat5a/b SH2 domains recognize these phosphorylated motifs, come close to their docking sites, and are subsequently phosphorylated by Jak2[2]. The IL-6 receptor (IL-6R) has two subunits: gp80 or IL-6Ra, which is cytokinespecific; and gp130, which is shared by IL-6 and related cytokines [30]. IL-6 binds first to IL-6Ra [31], which subsequently recruits the signal-transducing subunit gp130 (Fig. 10.1). The association of gp130 with IL-6 and IL-6Ra leads to the formation of the high-affinity IL-6R complex [32, 33] and phosphorylation of Jak2 preassociated with the cytoplasmic domain of gp130[34]. Jak2 phosphorylates Stat3 at Tyr705[25, 35], and phosphorylated Stat3 dimers translocate to nucleus where they bind to GAS elements of target genes. In addition to Stat3, IL-6 also can activate Ras and induce its translocation to the cell membrane, where Ras activates Raf,

248

T.K. Mirtti et al.

Fig. 10.1 The Janus kinase family proteins (Jaks) are primary activators of Stat proteins. In the canonical Prl-Jak2-Stat5a/b signaling pathway ligand binding induces conformational changes in the prolactin receptor (PrlR) and activates Jak2 molecules. Jak2 molecules then phosphorylate tyrosine motifs in PrlR, which are recognized by the SH2 domain on Stat5a/b and allow subsequent recruitment and phosphorylation of Stat5a/b by Jak2 [2]. Phosphorylated Stat5a/b dimerizes, undergoes nuclear translocation and binds to GAS elements of target genes. The IL-6 receptor (IL6R) has two subunits: gp80, or IL-6Ra, and gp130. IL-6 binds first to IL-6Ra [31], which subsequently recruits the signal-transducing subunit gp130. The association of gp130 with IL-6 and IL-6Ra leads to phosphorylation of Jak2 preassociated with the cytoplasmic domain of gp130 [34]. Jak2 phosphorylates and activates Stat3. Stat3, similar to Stat5, dimerizes, nuclear translocates and binds to GAS elements on the DNA. In addition to Stat3, IL-6 also activates Ras which activates Raf, leading to activation of mitogen-activated protein kinase kinase (MEK) and MAPK (Erk1/2) [36]. The third signaling pathway activated by IL-6 through a Jak2-mediated pathway is activation of the phosphoinositol 3 kinase (PI3K)-protein kinase B –Akt pathway [37, 38]

leading to activation of mitogen-activated protein kinase kinase (MEK) and MAPK (Erk1/2)[36]. The third signaling pathway activated by IL-6 via Jak2 is phosphoinositol 3 kinase (PI3K)-protein kinase B –Akt pathway [37, 38] (Fig. 10.1).

10

Transcription Factors Stat5a/b and Stat3 in Prostate Cancer Growth and Metastases

10.3

249

Regulators of Stat Signaling

The physiological actions of Stats are regulated by interactions with different cellular proteins and dephosphorylation [15]. In cancer, however, the various mechanisms that lead to persistent activity of Stat5a/b and Stat3 are under extensive research. Although Stat3 and Stat5a/b both recognize the same GAS consensus sequence TTCN2–4GAA, the gene expression regulatory functions are largely different for Stat5a/b versus Stat3, probably because of different co-activator, co-repressor, and chaperone expression profiles and interactions [39]. In prostate cancer, the positive regulatory mechanisms related to Stat activation may involve autocrine prolactin production activating the Jak2-Stat5a/b signaling [28, 29], gain-of-function mutations of Jak2, or somatic changes of Stat5a/b genes. Other tyrosine kinases, such as Src and Fer, have also been suggested to activate Stat5a/b in prostate cancer cells [40, 41]. In addition, receptor tyrosine kinases of EGFR family have been shown to activate Stat3 signaling [42]. Caspase-3, an apoptosis inhibitor, was found to be the probable effector of EGF/Stat3 action in DU145 prostate cancer cells [42]. In addition, there is evidence of somatic missense mutations in EGFR of prostate cancer patients, which leads to phosphorylation of Stat3. However, these effects were ligand-independent and phosphorylation of Stat3 disappeared upon addition of EGF [43]. Signaling pathways that negatively regulate Stat3 and Stat5a/b function may affect prostate cancer growth. The most studied of these regulatory mechanisms are Protein Inhibitors of Activated Stat proteins (PIAS), cytoplasmic and nuclear Protein Tyrosine Phosphatases (PTP), and Suppressors of Cytokine Signaling (SOCS) proteins. PIAS family of inhibitors includes PIAS1, PIAS3, PIASx, and PIASy, and alternative splicing variants of PIASx [44]. These proteins are located in the nucleus and repress Stat3 and Stat5a/b activity by a direct interaction. PIAS3 negatively regulates Stat3 DNA binding and represses the PRL-induced transcriptional activity of Stat5a/b [45, 46]. PTPs regulate the intranuclear dephosphorylation of the Jak2/ Stat5 signaling components and one of the PTP member, SHP-2, has been shown to interact directly with Stats [47]. Recently, another member of PTP family, SHP-1, has been shown to suppress IL-6-induced, Stat3-mediated prostate cancer cell proliferation in hormone-dependent LNCaP cancer cell model [48]. The third established down-regulatory mechanism of Stat signaling encompasses the suppressors of cytokine signaling -protein family. This family has eight members, SOCS1–7 and CIS (cytokine-inducible SH2 domain protein), and they block the upstream Jak activators by direct inhibition and/or by competitive binding to receptors [49]. Alternatively, SOCS proteins may bind to phosphorylated receptors competing for Stat binding [50, 51]. The SOCS1, SOCS3, SOCS5 and CIS protein levels are lower in hormone-independent cell lines DU145 and PC-3, but higher in wild type, hormone-dependent LNCaP cells as compared to normal epithelial cells [52]. Others have recently demonstrated that inhibition of SOCS-3 leads to reactivation of Stat1 and Stat3, but causes activation of extrinsic and intrinsic apoptosis pathways [53]. In castration-refractory clinical cancers, SOCS-3 protein levels are elevated [53].

250

10.4

T.K. Mirtti et al.

Stat5a/b and Stat3 Expression Patterns in Clinical Prostate Cancer and as Biomarkers of the Disease Progression

The findings on the constitutively active autocrine Prl-Jak2-Stat5a/b and IL-6Jak2-Stat3 pathways in prostate tissue [54–56] have led to studies focused on the role of Stat3 and Stat5a/b in prostate cancer progression. Li et al. (2004) showed that nuclear Stat5a/b expression was associated with high Gleason grade in clinical prostate cancer specimens [28]. In addition to the association with the loss of prostate cancer differentiation, nuclear Stat5a/b and prolactin expression, both were elevated in distant prostate cancer metastases and castration-resistant prostate cancers [29, 57]. Li et al. (2005) further showed that activated nuclear Stat5a/b predicted early prostate cancer recurrence in a univariate analysis [58]. Furthermore, multivariate analysis using Cox proportional hazard regression showed that Stat5a/b activation remained an independent prognostic factor (HR 1.63, 95% CI 0.99–2.69) of short progression-free survival of prostate cancer. However, the patient cohort in this study was heterogenous in terms of treatment modalities and therefore the significance of active Stat5a/b in predicting the clinical outcome should be addressed in a more homogenous group of patients with disease-specific mortality as an additional end-point. It has also been shown that Stat5a/b is constitutively active in over 60% of clinical prostate cancer metastasis [59], and another study indicated a higher frequency of Stat5a/b activation in castration-resistant prostate cancers treated by an androgen blockade vs. hormone naïve prostate cancers [57]. IL-6 is an autocrine cytokine in prostate cancer [55], and Stat3 is constitutively activated in human prostate cancer tissue [56]. Activation of Stat3 has been associated with high Gleason score in immunohistochemical and proteomic studies [56, 60–62]. The study by Abdulghani et al. (2008) showed that Stat3 is activated in the majority of clinical prostate cancer metastases [63]. In the study by Horinaga et al. (2005), including 92 patients, the increased phospho-Stat3 immunoreactivity was associated with shorter time to biochemical recurrence after radical prostatectomy [64]. In contrast, the nuclear expression of Stat3 correlated inversely with distant metastases formation in another study, and low Stat3 expression was an independent predictor of metastasis in patients treated with radiotherapy or combined radio- and chemotherapy [65]. In a large scale prognostic analysis by Li et al. (2005) including 357 primary prostate cancer patients treated with radical prostatectomy, active Stat3 did not predict prostate cancer recurrence (unpublished data), while active Stat5 was a marker of poor clinical outcome in the same cohort [58]. A recent study reported that Stat3associated regulatory protein SHP-1 inversely correlated with disease progression and biochemical recurrence after radical prostatectomy [66]. Considering the abundance of Stat3 in later stages of prostate cancer, i.e., metastasis and castration resistance, the prognostic significance of Stat3 should be evaluated with disease-specific mortality as the end-point.

10

Transcription Factors Stat5a/b and Stat3 in Prostate Cancer Growth and Metastases

10.5

251

Stat5a/b and Stat3 in Growth Promotion of Prostate Cancer

Both Stat5a/b and Stat3 have been shown to be involved in the regulation of viability and proliferation of human prostate cancer cells. The crucial role of Stat5a/b in promotion of prostate cancer cell viability and tumor growth has been demonstrated in several different studies. Inhibition of Stat5a/b transcriptional activity by adenoviral delivery of a dominant negative (DN) mutant of Stat5a/b (lacking the TA domain) results in apoptotic death of a battery of human prostate cancer cell lines in culture [41, 67, 68]. In addition, inhibition of Stat5 markedly decreased the growth of human prostate xenograft tumors in nude mice [41, 67, 68], whereas normal prostate epithelial cells did not respond to Stat5a/b inhibition [41]. Blocking the protein expression of Stat5a/b by RNA interference or antisense oligonucleotides resulted in a rapid cell death of human prostate cancer cells [41, 68]. Moreover, Stat5b siRNAs, but few Stat5a siRNAs, decreased the viability of DU145 prostate cancer cells, suggesting that Stat5b might have the predominant role over Stat5a in regulating prostate cancer cell growth [41]. Inhibition of Stat5a/b also blocked the progression of prostate cancer in the TRAMP mouse prostate cancer model [69]. Mediators of Stat5a/b effects on prostate cancer cell viability include Cyclin D1 and BclXL proteins [68], and additional targets were recently identified by gene expression profiling [41, 59]. The role of Stat3 in prostate cancer growth is more controversial. Induction of transcription factor Stat3 expression has been reported to induce a change of normal prostate epithelial cells to a malignant phenotype [70]. In addition, blocking Stat3 function has been shown to induce apoptosis of DU145 prostate cancer cells in vitro [60, 71]. There is evidence that active Stat3 binds to DNA in benign tissue adjacent to prostate cancer in whole tissue sections suggesting that Stat3 activation may occur before detectable histological alteration of the prostate [56]. On the contrary, others have shown that LNCaP cells that express a dominant-negative form of Stat3 exhibit a proliferative response to IL-6 stimulation [72], indicating that Stat3, in fact, mediates growth inhibition of human prostate cancer cells. In addition, IL-6 produced by prostate stromal cells derived from benign hyperplasia had a growth inhibitory effect on LNCaP cells [73]. When Stat5a/b and Stat3 were compared side-by-side for their significance on prostate cancer cell viability and survival, it became evident that Stat5a/b has the predominant role in growth promotion of prostate cancer over Stat3. Specifically, siRNA inhibition of Stat5a/b resulted in 70% decrease in the viability of DU145 human prostate cancer cells, while Stat3-inhibition reduced the DU145 cell viability only by 20%[41]. Similarly, inhibition of transcriptional activity of Stat3 by adenoviral delivery of a DN form of Stat3 decreased the number of viable DU145 cancer cells by 30% as compared to 87% decrease induced by DNStat5a/b [41]. Both the DNStat5a/b and DNStat3 constructs lack the C-terminal transactivation domains, resulting in Stat5 and Stat3 proteins able to bind to DNA but unable to initiate transcription. In the same study, the inhibition of Stat3 activity by DNStat3

252

T.K. Mirtti et al.

did not have any significant effect on the growth of orthotopic or subcutaneous prostate tumors in nude mice, whereas DNStat5a/b reduced both the growth rates and the volumes of the tumors robustly [41].

10.6

Stat5a/b and Stat3 in Promotion of Metastatic Behavior of Prostate Cancer

Both Stat5a/b and Stat3 induce metastatic behavior of human prostate cancer cells in vitro and in vivo [41, 63]. Adenoviral expression, of wild-type (WT) Stat3 in DU145 cells led to 33–50-fold increase in metastases formation in the lungs of nude mice after tail-vein injections of the cells in comparison to the control DU145 cells. Stat3 increased migration of prostate cancer cells, in addition to reorganization of actin cytoskeleton and the microtubule network [63]. Furthermore, Abdulghani et al. (2008) showed that Jak2 inhibition by a dominant-negative mutant of Jak2 (DNJak2) blocked tyrosine phosphorylation of Stat3 and led to similar morphological changes as Stat3 inhibition, whereas overexpression of WTJak2 increased Stat3 activity and cell migration [63]. Similarly to Stat3, Stat5a/b is activated in the majority of distant human prostate cancer metastases [41]. Introduction of an active form of Stat5 promoted DU145 prostate cancer cell migration and invasion in vitro, and increased the in vivo lung metastasis formation by 11-fold in nude mice after tail-vein injections [41, 59]. Surprisingly, when wild-type Stat5, not the active form of Stat5, was introduced to DU145 prostate cancer cells, it increased metastases formation only by four to fivefold [59]. These data indicate that the autocrine Stat5-activating signaling pathway is not fully functional in DU145 cells, since wild-type Stat5 was not able affect the metastatic potential of the cells to the same extent as the active form of Stat5. Disrupted Prl-Jak2-Stat5 signaling pathway in DU145 cells may be due to low level expression of Prl-receptors in these cells [74]. Metastases-related mechanisms associated with active Stat5a/b include increased migration, decreased cellsurface E-cadherin levels, and increased heterotypic adhesion of prostate cancer cells to endothelial cells [59]. Gene expression profiling indicated that the majority of Stat5a/b regulated genes were metastases-related [41, 59].

10.7

Molecular Mechanisms Underlying Stat5/Stat3 -Induced Metastatic Behavior of Human Prostate Cancer Cells

Prostate cancer metastases formation is a multi-step process. First, prostate cancer cells have to move from the primary tumor mass into the bloodstream where they must endure the hemodynamic stresses, evade the natural host immune response, and arrest at the secondary site. Once at the secondary site, the metastatic prostate

10

Transcription Factors Stat5a/b and Stat3 in Prostate Cancer Growth and Metastases

253

cancer cells need to adapt to the local microenvironment, proliferate, and induce angiogenesis to form successful secondary colonies. Failure to complete any step in this cascade prevents formation of distant metastases. Both Stat5a/b and Stat3 have been shown to promote metastasis formation of blood-borne prostate tumor cells in vivo after tail-vein injections [41, 59]. Hematogenous metastasis is believed to be a relatively inefficient process. This is due to the destruction of cells in the bloodstream by shear stress, the immune system, and a slow rate of extravasation. In addition, one key contributor for metastatic potential of cancer cells is the capability for anchorage-independent growth and survival in the bloodstream and at the distant site. While both Stat5a/b and Stat3 promoted metastases formation in the in vivo experimental metastasis assay, only Stat5a/b, not Stat3, had a major role in induction of prostate cancer cell survival [41, 68]. In addition, Stat5a/b regulated a largely different set of genes from those of Stat3 in human prostate cancer cells [41]. This suggests that the two transcription factors likely affect different steps of the metastases process (i.e., survival, arrest at the secondary site, extravasation, invasion, migration, colonization) and have different mechanisms of action to induce colonization of the circulating tumor cells. Both Stat5a/b and Stat3 induce migration of human prostate cancer cells [41, 59, 63]. Acquisition of a migratory phenotype is concurrent with the epithelial-mesenchymal transition (EMT), whereby epithelial cells lose polarity and homotypic adhesion and undergo dramatic remodeling of the cytoskeleton. Both Stat5a/b and Stat3 were shown to affect the cellular organization of the microtubule network while only Stat3 induced polarization of the actin cytoskeleton [41, 63]. Ng et al. has shown that Stat3 interacts with stathmin, a small cytosolic phosphoprotein that binds a/b-tubulin heterodimers, and facilitates cell migration [75]. The interactions of Stat5a/b or Stat3 with stathmin in human prostate cancer cells remain to be investigated. In general, hallmarks of EMT in prostate cancer cells include loss of epithelial markers, such as E-cadherin, concomitant with up-regulation of mesenchymal markers, such as N-cadherin, also known as “cadherin switching”[76, 77]. E-cadherin, the major component of epithelial adherens junctions, is a single-span transmembrane glycoprotein with a cytosolic domain that binds to a, b -catenin and p120 proteins, which interact with the cytoskeleton and facilitate cell-cell adhesion [78–81]. Importantly, Stat5 activation induced down-regulation of cell surface E-cadherin in human prostate cancer cells in vitro [59], and low cell surface E-cadherin expression correlated with high nuclear Stat5 levels in clinical prostate cancers [58]. Decreased homotypic adhesion of prostate cancer cells was induced by constitutively active Stat3 shown in another study [82]. In functional E-cadherin complexes, which are located in adherens junctions, the cytosolic domains of E-cadherins bind to a and b- catenin which interact with the microtubule network, thus the loss of homotypic adhesion may initiate depolymerization of microtubules, promoting prostate cancer cell motility. Recently, decreased b-catenin expression was shown to predict early prostate cancer-specific death [83]. However, b-catenin expression alone is not able to induce EMT of cancer cells [84]. Also, decreased p120 expression, which stabilizes cadherins, has been associated with a shorter, recurrence-free survival of prostate cancer [83]. In summary, the individual and

254

T.K. Mirtti et al.

detailed roles of Stat5a/b and Stat3 in EMT of human prostate cancer cells need to be determined. Loss of E-cadherin can be achieved by mutation, DNA methylation, or silencing of the promoter region. Genes that suppress E-cadherin expression are zinc finger proteins that bind to the E-cadherin gene (CDH1) promoter and include Snail, Slug, SIP1, and TWIST [85–89]. TWIST, a basic helix-loop-helix transcription factor, is a crucial modulator of a prostate-specific metastatic phenotype. This phenotype is achieved through TWIST-induced loss of E-cadherin-mediated cell-cell adhesion and expression of mesenchymal markers [84, 87, 90–92]. In prostate cancer, TWIST expression is correlated with high Gleason grade and TWIST is expressed at high levels in metastatic human prostate cancer cell lines, DU145 and PC-3[90, 91]. Two groups have independently demonstrated that active Stat3 binds to the TWIST promoter and directly regulates the transcriptional activity of TWIST [93, 94]. Regulation of TWIST expression by Stat5a/b and Stat3 in human prostate cancer has yet to be determined. In order to escape into the extravascular tissue to establish metastases, circulating tumor cells arrested in capillary beds of different organs must invade the endothelial cell lining of blood vessels and degrade its underlying basement membrane. Tumor cell metastasis is facilitated by formation of “premetastatic niches” in the destination organs [95]. At premetastatic sites, unidentified tumor-secreted factors are thought to induce elevated fibronectin expression [95] which may serve as an anchor to arrest circulating tumor cells. The key integrin receptor subunit binding to fibronectin is beta1. Integrins are transmembrane heterodimer receptors consisting of an alpha and beta subunit [96]. By binding to extracellular matrix (ECM) components, such as laminin, collagen, and fibronectin, they convey heterotypic signals and interactions. There are at least 24 members in the integrin family, 18 alpha and 8 beta subunits. Beta 1, 3 and 6 are upregulated in human prostate cancer – they are known to localize in focal contacts and to mediate spreading and cytoskeletal rearrangements in normal cells [97, 98]. There are 5 different cytoplasmic variants of the human beta1 subunit: beta 1A, beta 1B, beta 1C, beta1C-2 and betaD, of which only b1A and b1C are expressed in prostate epithelium [99, 100]. b1A is upregulated in prostate cancer, stimulates proliferation, and allows for anchorageindependent growth [100]. The splice variant b1C mRNA and protein levels are reduced in prostate cancer, compared to levels in benign prostate epithelial cells [101, 102], and integrin b1C is an inhibitor of prostate cancer cell proliferation. In prostate cancer, active Stat5a has been shown to promote heterotypic adhesion of prostate cancer cells to endothelial cells [59]. Constitutively active Stat3 induces transcriptional up-regulation of b6 bound to fibronectin in human prostate cancer cell lines [82]. There is no information in the literature on Stat5 regulation of integrins in human prostate cancer cells or xenograft tumors. Finally, prostate cancer migration and invasion may result from secretion of proteases by the cancer cells. Matrix metalloproteases (MMPs), a multigene family of zinc-dependent endopeptidases, degrade ECM components and have been implicated in EMT by cleavage of E-cadherin [103–105]. MMP-9 and MMP-2 expression positively correlate with Gleason score and promote development of metastases

10

Transcription Factors Stat5a/b and Stat3 in Prostate Cancer Growth and Metastases

255

of prostate cancer cells [106–108]. Stat3, in turn, has been shown to regulate MMP-2 and MMP-1 expression by direct promoter binding [109, 110]. In summary, both Stat5a/b and Stat3 are involved in the development of disseminated prostate cancer. Little is known about the molecular mechanisms mediating the metastases process of Stat5a/b vs. Stat3, which are likely to differ and therefore require further studies. Both Stat5a/b and Stat3 may provide effective therapeutic targets for metastatic prostate cancer. Acknowledgments We would like to thank Ms Elyse Amico for her editorial contributions to the chapter.

References 1. Jemal A, Siegel R, Xu J, Ward E (2010) Cancer statistics, 2010. CA Cancer J Clin 60(5):277–300 2. Ihle JN (2001) The Stat family in cytokine signaling. Curr Opin Cell Biol 13(2):211–217 3. Haddad B, Pabon-Pena CR, Young H, Sun WH (1998) Assignment1 of STAT1 to human chromosome 2q32 by FISH and radiation hybrids. Cytogenet Cell Genet 83(1–2):58–59 4. Yamamoto K, Kobayashi H, Arai A, Miura O, Hirosawa S, Miyasaka N (1997) cDNA cloning, expression and chromosome mapping of the human STAT4 gene: both STAT4 and STAT1 genes are mapped to 2q32.2–
256

T.K. Mirtti et al.

15. Levy DE, Darnell JE Jr (2002) Stats: transcriptional control and biological impact. Nat Rev Mol Cell Biol 3(9):651–662 16. Darnell JE Jr (1997) STATs and gene regulation. Science 277(5332):1630–1635 17. Gewinner C, Hart G, Zachara N, Cole R, Beisenherz-Huss C, Groner B (2004) The coactivator of transcription CREB-binding protein interacts preferentially with the glycosylated form of Stat5. J Biol Chem 279(5):3563–3572 18. Becker S, Groner B, Muller CW (1998) Three-dimensional structure of the Stat3beta homodimer bound to DNA. Nature 394(6689):145–151 19. Gouilleux F, Wakao H, Mundt M, Groner B (1994) Prolactin induces phosphorylation of Tyr694 of Stat5 (MGF), a prerequisite for DNA binding and induction of transcription. EMBO J 13(18):4361–4369 20. Yamashita H, Nevalainen MT, Xu J, LeBaron MJ, Wagner KU, Erwin RA, Harmon JM, Hennighausen L, Kirken RA, Rui H (2001) Role of serine phosphorylation of Stat5a in prolactin-stimulated beta- casein gene expression. Mol Cell Endocrinol 183(1–2):151–163 21. Iyer J, Reich NC (2008) Constitutive nuclear import of latent and activated STAT5a by its coiled coil domain. FASEB J 22(2):391–400 22. Reich NC, Liu L (2006) Tracking STAT nuclear traffic. Nat Rev Immunol 6(8):602–612 23. Yang J, Chatterjee-Kishore M, Staugaitis SM, Nguyen H, Schlessinger K, Levy DE, Stark GR (2005) Novel roles of unphosphorylated STAT3 in oncogenesis and transcriptional regulation. Cancer Res 65(3):939–947 24. Yang J, Liao X, Agarwal MK, Barnes L, Auron PE, Stark GR (2007) Unphosphorylated STAT3 accumulates in response to IL-6 and activates transcription by binding to NFkappaB. Genes Dev 21(11):1396–1408 25. Lutticken C, Wegenka UM, Yuan J, Buschmann J, Schindler C, Ziemiecki A, Harpur AG, Wilks AF, Yasukawa K, Taga T et al (1994) Association of transcription factor APRF and protein kinase Jak1 with the interleukin-6 signal transducer gp130. Science 263(5143):89–92 26. Ihle JN (1995) The Janus protein tyrosine kinase family and its role in cytokine signaling. Adv Immunol 60:1–35 27. Darnell JE Jr (1996) The JAK-STAT pathway: summary of initial studies and recent advances. Recent Prog Horm Res 51:391–403; discussion 403–394 28. Li H, Ahonen TJ, Alanen K, Xie J, LeBaron MJ, Pretlow TG, Ealley EL, Zhang Y, Nurmi M, Singh B, Martikainen PM, Nevalainen MT (2004) Activation of signal transducer and activator of transcription 5 in human prostate cancer is associated with high histological grade. Cancer Res 64(14):4774–4782 29. Dagvadorj A, Collins S, Jomain JB, Abdulghani J, Karras J, Zellweger T, Li H, Nurmi M, Alanen K, Mirtti T, Visakorpi T, Bubendorf L, Goffin V, Nevalainen MT (2007) Autocrine prolactin promotes prostate cancer cell growth via Janus kinase-2-signal transducer and activator of transcription-5a/b signaling pathway. Endocrinology 148(7):3089–3101 30. Hirano T, Matsuda T, Nakajima K (1994) Signal transduction through gp130 that is shared among the receptors for the interleukin 6 related cytokine subfamily. Stem Cells 12(3):262–277 31. Hibi M, Murakami M, Saito M, Hirano T, Taga T, Kishimoto T (1990) Molecular cloning and expression of an IL-6 signal transducer, gp130. Cell 63(6):1149–1157 32. Heinrich PC, Behrmann I, Muller-Newen G, Schaper F, Graeve L (1998) Interleukin-6-type cytokine signalling through the gp130/Jak/STAT pathway. Biochem J 334(Pt 2):297–314 33. Carbia-Nagashima A, Arzt E (2004) Intracellular proteins and mechanisms involved in the control of gp130/JAK/STAT cytokine signaling. IUBMB Life 56(2):83–88 34. Heinrich PC, Behrmann I, Haan S, Hermanns HM, Muller-Newen SF (2003) Principles of interleukin-6 (IL-6)-type cytokine signalling and its regulation. Biochem J 374:1–20 35. Darnell JE Jr, Kerr IM, Stark GR (1994) Jak-STAT pathways and transcriptional activation in response to IFNs and other extracellular signaling proteins. Science 264(5164):1415–1421 36. Hu L, Shi Y, Hsu JH, Gera J, Van Ness B, Lichtenstein A (2003) Downstream effectors of oncogenic ras in multiple myeloma cells. Blood 101(8):3126–3135 37. Zhang J, Li Y, Shen B (2003) PI3-K/Akt pathway contributes to IL-6-dependent growth of 7TD1 cells. Cancer Cell Int 3(1):1

10

Transcription Factors Stat5a/b and Stat3 in Prostate Cancer Growth and Metastases

257

38. Jee SH, Chu CY, Chiu HC, Huang YL, Tsai WL, Liao YH, Kuo ML (2004) Interleukin-6 induced basic fibroblast growth factor-dependent angiogenesis in basal cell carcinoma cell line via JAK/STAT3 and PI3-kinase/Akt pathways. J Invest Dermatol 123(6):1169–1175 39. Ehret GB, Reichenbach P, Schindler U, Horvath CM, Fritz S, Nabholz M, Bucher P (2001) DNA binding specificity of different STAT proteins. Comparison of in vitro specificity with natural target sites. J Biol Chem 276(9):6675–6688 40. Zoubeidi A, Rocha J, Zouanat FZ, Hamel L, Scarlata E, Aprikian AG, Chevalier S (2009) The Fer tyrosine kinase cooperates with interleukin-6 to activate signal transducer and activator of transcription 3 and promote human prostate cancer cell growth. Mol Cancer Res 7(1):142–155 41. Gu L, Dagvadorj A, Lutz J, Leiby B, Bonuccelli G, Lisanti MP, Addya S, Fortina P, Dasgupta A, Hyslop T, Bubendorf L, Nevalainen MT (2010) Transcription factor Stat3 stimulates metastatic behavior of human prostate cancer cells in vivo, whereas Stat5b has a preferential role in the promotion of prostate cancer cell viability and tumor growth. Am J Pathol 176(4):1959–1972 42. Zhou W, Grandis JR, Wells A (2006) STAT3 is required but not sufficient for EGF receptor-mediated migration and invasion of human prostate carcinoma cell lines. Br J Cancer 95(2):164–171 43. Cai CQ, Peng Y, Buckley MT, Wei J, Chen F, Liebes L, Gerald WL, Pincus MR, Osman I, Lee P (2008) Epidermal growth factor receptor activation in prostate cancer by three novel missense mutations. Oncogene 27(22):3201–3210 44. Schmidt D, Muller S (2003) PIAS/SUMO: new partners in transcriptional regulation. Cell Mol Life Sci 60(12):2561–2574 45. Chung CD, Liao J, Liu B, Rao X, Jay P, Berta P, Shuai K (1997) Specific inhibition of Stat3 signal transduction by PIAS3. Science 278(5344):1803–1805 46. Rycyzyn MA, Clevenger CV (2002) The intranuclear prolactin/cyclophilin B complex as a transcriptional inducer. Proc Natl Acad Sci U S A 99(10):6790–6795 47. Chen Y, Wen R, Yang S, Schuman J, Zhang EE, Yi T, Feng GS, Wang D (2003) Identification of Shp-2 as a Stat5A phosphatase. J Biol Chem 278(19):16520–16527 48. Tassidis H, Culig Z, Wingren AG, Harkonen P (2010) Role of the protein tyrosine phosphatase SHP-1 in Interleukin-6 regulation of prostate cancer cells. Prostate 70(14):1491–1500 49. Starr R, Willson TA, Viney EM, Murray LJ, Rayner JR, Jenkins BJ, Gonda TJ, Alexander WS, Metcalf D, Nicola NA, Hilton DJ (1997) A family of cytokine-inducible inhibitors of signalling. Nature 387(6636):917–921 50. Alexander WS, Hilton DJ (2004) The role of suppressors of cytokine signaling (SOCS) proteins in regulation of the immune response. Annu Rev Immunol 22:503–529 51. Croker BA, Kiu H, Nicholson SE (2008) SOCS regulation of the JAK/STAT signalling pathway. Semin Cell Dev Biol 19(4):414–422 52. Evans MK, Yu CR, Lohani A, Mahdi RM, Liu X, Trzeciak AR, Egwuagu CE (2007) Expression of SOCS1 and SOCS3 genes is differentially regulated in breast cancer cells in response to proinflammatory cytokine and growth factor signals. Oncogene 26(13):1941–1948 53. Puhr M, Santer FR, Neuwirt H, Susani M, Nemeth JA, Hobisch A, Kenner L, Culig Z (2009) Down-regulation of suppressor of cytokine signaling-3 causes prostate cancer cell death through activation of the extrinsic and intrinsic apoptosis pathways. Cancer Res 69(18):7375–7384 54. Ahonen TJ, Harkonen PL, Rui H, Nevalainen MT (2002) PRL signal transduction in the epithelial compartment of rat prostate maintained as long-term organ cultures in vitro. Endocrinology 143(1):228–238 55. Giri D, Ozen M, Ittmann M (2001) Interleukin-6 is an autocrine growth factor in human prostate cancer. Am J Pathol 159(6):2159–2165 56. Dhir R, Ni Z, Lou W, DeMiguel F, Grandis JR, Gao AC (2002) Stat3 activation in prostatic carcinomas. Prostate 51(4):241–246 57. Tan SH, Dagvadorj A, Shen F, Gu L, Liao Z, Abdulghani J, Zhang Y, Gelmann EP, Zellweger T, Culig Z, Visakorpi T, Bubendorf L, Kirken RA, Karras J, Nevalainen MT (2008) Transcription factor Stat5 synergizes with androgen receptor in prostate cancer cells. Cancer Res 68(1):236–248

258

T.K. Mirtti et al.

58. Li H, Zhang Y, Glass A, Zellweger T, Gehan E, Bubendorf L, Gelmann EP, Nevalainen MT (2005) Activation of signal transducer and activator of transcription-5 in prostate cancer predicts early recurrence. Clin Cancer Res 11(16):5863–5868 59. Gu L, Vogiatzi P, Puhr M, Dagvadorj A, Lutz J, Ryder A, Addya S, Fortina P, Cooper C, Leiby B, Dasgupta A, Hyslop T, Bubendorf L, Alanen K, Mirtti T, Nevalainen MT (2010) Stat5 promotes metastatic behavior of human prostate cancer cells in vitro and in vivo. Endocr Relat Cancer 17(2):481–493 60. Mora LB, Buettner R, Seigne J, Diaz J, Ahmad N, Garcia R, Bowman T, Falcone R, Fairclough R, Cantor A, Muro-Cacho C, Livingston S, Karras J, Pow-Sang J, Jove R (2002) Constitutive activation of Stat3 in human prostate tumors and cell lines: direct inhibition of Stat3 signaling induces apoptosis of prostate cancer cells. Cancer Res 62(22):6659–6666 61. Campbell CL, Jiang Z, Savarese DM, Savarese TM (2001) Increased expression of the interleukin-11 receptor and evidence of STAT3 activation in prostate carcinoma. Am J Pathol 158:25–32 62. Grubb RL, Deng J, Pinto PA, Mohler JL, Chinnaiyan A, Rubin M, Linehan WM, Liotta LA, Petricoin EF, Wulfkuhle JD (2009) Pathway biomarker profiling of localized and metastatic human prostate cancer reveal metastatic and prognostic signatures. J Proteome Res 8(6):3044–3054 63. Abdulghani J, Gu L, Dagvadorj A, Lutz J, Leiby B, Bonuccelli G, Lisanti MP, Zellweger T, Alanen K, Mirtti T, Visakorpi T, Bubendorf L, Nevalainen MT (2008) Stat3 promotes metastatic progression of prostate cancer. Am J Pathol 172(6):1717–1728 64. Horinaga M, Okita H, Nakashima J, Kanao K, Sakamoto M, Murai M (2005) Clinical and pathologic significance of activation of signal transducer and activator of transcription 3 in prostate cancer. Urology 66(3):671–675 65. Torres-Roca JF, DeSilvio M, Mora LB, Khor LY, Hammond E, Ahmad N, Jove R, Forman J, Lee RJ, Sandler H, Pollack A (2007) Activated STAT3 as a correlate of distant metastasis in prostate cancer: a secondary analysis of Radiation Therapy Oncology Group 86-10. Urology 69(3):505–509 66. Tassidis H, Brokken LJ, Jirstrom K, Ehrnstrom R, Ponten F, Ulmert D, Bjartell A, Harkonen P, Wingren AG (2010) Immunohistochemical detection of tyrosine phosphatase SHP-1 predicts outcome after radical prostatectomy for localized prostate cancer. Int J Cancer 126(10):2296–2307 67. Ahonen TJ, Xie J, LeBaron MJ, Zhu J, Nurmi M, Alanen K, Rui H, Nevalainen MT (2003) Inhibition of transcription factor Stat5 induces cell death of human prostate cancer cells. J Biol Chem 278(29):27287–27292 68. Dagvadorj A, Kirken RA, Leiby B, Karras J, Nevalainen MT (2008) Transcription factor signal transducer and activator of transcription 5 promotes growth of human prostate cancer cells in vivo. Clin Cancer Res 14(5):1317–1324 69. Kazansky AV, Spencer DM, Greenberg NM (2003) Activation of signal transducer and activator of transcription 5 is required for progression of autochthonous prostate cancer: evidence from the transgenic adenocarcinoma of the mouse prostate system. Cancer Res 63(24):8757–8762 70. Huang HF, Murphy TF, Shu P, Barton AB, Barton BE (2005) Stable expression of constitutively-activated STAT3 in benign prostatic epithelial cells changes their phenotype to that resembling malignant cells. Mol Cancer 4(1):2 71. Barton BE, Karras JG, Murphy TF, Barton A, Huang H (2004) Signal transducer and activator of transcription (STAT3) activation in prostate cancer: direct STAT3 inhibition induces apoptosis in prostate cancer lines. Mol Cancer Ther 3:11–20 72. Spiotto MT, Chung TD (2000) STAT3 mediates IL-6-induced growth inhibition in the human prostate cancer cell line LNCaP. Prostate 42(2):88–98 73. Degeorges A, Tatoud R, Fauvel Lafeve F, Podgorniak MP, Millot G, de Cremoux P, Calvo F (1996) Stromal cells from human benign prostate hyperplasia produce a growth-inhibitory factor for LNCaP prostate cancer cells, identified as interleukin-6. Int J Cancer 68(2):207–214 74. Peirce SK, Chen WY, Chen WY (2001) Quantification of prolactin receptor mRNA in multiple human tissues and cancer cell lines by real time RT-PCR. J Endocrinol 171(1):R1–R4

10

Transcription Factors Stat5a/b and Stat3 in Prostate Cancer Growth and Metastases

259

75. Ng DC, Lin BH, Lim CP, Huang G, Zhang T, Poli V, Cao X (2006) Stat3 regulates microtubules by antagonizing the depolymerization activity of stathmin. J Cell Biol 172(2):245–257 76. Gravdal K, Halvorsen OJ, Haukaas SA, Akslen LA (2007) A switch from E-cadherin to N-cadherin expression indicates epithelial to mesenchymal transition and is of strong and independent importance for the progress of prostate cancer. Clin Cancer Res 13(23):7003–7011 77. Tomita K, van Bokhoven A, van Leenders GJ, Ruijter ET, Jansen CF, Bussemakers MJ, Schalken JA (2000) Cadherin switching in human prostate cancer progression. Cancer Res 60(13):3650–3654 78. Huber AH, Weis WI (2001) The structure of the beta-catenin/E-cadherin complex and the molecular basis of diverse ligand recognition by beta-catenin. Cell 105(3):391–402 79. Ireton RC, Davis MA, van Hengel J, Mariner DJ, Barnes K, Thoreson MA, Anastasiadis PZ, Matrisian L, Bundy LM, Sealy L, Gilbert B, van Roy F, Reynolds AB (2002) A novel role for p120 catenin in E-cadherin function. J Cell Biol 159(3):465–476 80. Ishiyama N, Lee SH, Liu S, Li GY, Smith MJ, Reichardt LF, Ikura M (2010) Dynamic and static interactions between p120 catenin and E-cadherin regulate the stability of cell-cell adhesion. Cell 141(1):117–128 81. Hirano S, Kimoto N, Shimoyama Y, Hirohashi S, Takeichi M (1992) Identification of a neural alpha-catenin as a key regulator of cadherin function and multicellular organization. Cell 70(2):293–301 82. Azare J, Leslie K, Al-Ahmadie H, Gerald W, Weinreb PH, Violette SM, Bromberg J (2007) Constitutively activated Stat3 induces tumorigenesis and enhances cell motility of prostate epithelial cells through integrin beta 6. Mol Cell Biol 27(12):4444–4453 83. van Oort IM, Tomita K, van Bokhoven A, Bussemakers MJ, Kiemeney LA, Karthaus HF, Witjes JA, Schalken JA (2007) The prognostic value of E-cadherin and the cadherin-associated molecules alpha-, beta-, gamma-catenin and p120ctn in prostate cancer specific survival: a long-term follow-up study. Prostate 67(13):1432–1438 84. Onder TT, Gupta PB, Mani SA, Yang J, Lander ES, Weinberg RA (2008) Loss of E-cadherin promotes metastasis via multiple downstream transcriptional pathways. Cancer Res 68(10):3645–3654 85. Savagner P, Yamada KM, Thiery JP (1997) The zinc-finger protein slug causes desmosome dissociation, an initial and necessary step for growth factor-induced epithelial-mesenchymal transition. J Cell Biol 137(6):1403–1419 86. Giroldi LA, Bringuier PP, de Weijert M, Jansen C, van Bokhoven A, Schalken JA (1997) Role of E boxes in the repression of E-cadherin expression. Biochem Biophys Res Commun 241(2):453–458 87. Yang J, Mani SA, Donaher JL, Ramaswamy S, Itzykson RA, Come C, Savagner P, Gitelman I, Richardson A, Weinberg RA (2004) Twist, a master regulator of morphogenesis, plays an essential role in tumor metastasis. Cell 117(7):927–939 88. Comijn J, Berx G, Vermassen P, Verschueren K, van Grunsven L, Bruyneel E, Mareel M, Huylebroeck D, van Roy F (2001) The two-handed E box binding zinc finger protein SIP1 downregulates E-cadherin and induces invasion. Mol Cell 7(6):1267–1278 89. Guaita S, Puig I, Franci C, Garrido M, Dominguez D, Batlle E, Sancho E, Dedhar S, De Herreros AG, Baulida J (2002) Snail induction of epithelial to mesenchymal transition in tumor cells is accompanied by MUC1 repression and ZEB1 expression. J Biol Chem 277(42):39209–39216 90. Wallerand H, Robert G, Pasticier G, Ravaud A, Ballanger P, Reiter RE, Ferriere JM (2010) The epithelial-mesenchymal transition-inducing factor TWIST is an attractive target in advanced and/or metastatic bladder and prostate cancers. Urol Oncol 28(5):473–479 91. Kwok WK, Ling MT, Lee TW, Lau TC, Zhou C, Zhang X, Chua CW, Chan KW, Chan FL, Glackin C, Wong YC, Wang X (2005) Up-regulation of TWIST in prostate cancer and its implication as a therapeutic target. Cancer Res 65(12):5153–5162 92. Alexander NR, Tran NL, Rekapally H, Summers CE, Glackin C, Heimark RL (2006) N-cadherin gene expression in prostate carcinoma is modulated by integrin-dependent nuclear translocation of Twist1. Cancer Res 66(7):3365–3369

260

T.K. Mirtti et al.

93. Lo HW, Hsu SC, Xia W, Cao X, Shih JY, Wei Y, Abbruzzese JL, Hortobagyi GN, Hung MC (2007) Epidermal growth factor receptor cooperates with signal transducer and activator of transcription 3 to induce epithelial-mesenchymal transition in cancer cells via up-regulation of TWIST gene expression. Cancer Res 67(19):9066–9076 94. Cheng GZ, Chan J, Wang Q, Zhang W, Sun CD, Wang LH (2007) Twist transcriptionally up-regulates AKT2 in breast cancer cells leading to increased migration, invasion, and resistance to paclitaxel. Cancer Res 67(5):1979–1987 95. Kaplan RN, Riba RD, Zacharoulis S, Bramley AH, Vincent L, Costa C, MacDonald DD, Jin DK, Shido K, Kerns SA, Zhu Z, Hicklin D, Wu Y, Port JL, Altorki N, Port ER, Ruggero D, Shmelkov SV, Jensen KK, Rafii S, Lyden D (2005) VEGFR1-positive haematopoietic bone marrow progenitors initiate the pre-metastatic niche. Nature 438(7069):820–827 96. Hynes RO (2002) Integrins: bidirectional, allosteric signaling machines. Cell 110(6): 673–687 97. Fornaro M, Manes T, Languino LR (2001) Integrins and prostate cancer metastases. Cancer Metastasis Rev 20(3–4):321–331 98. Goel HL, Alam N, Johnson IN, Languino LR (2009) Integrin signaling aberrations in prostate cancer. Am J Transl Res 1(3):211–220 99. Fornaro M, Tallini G, Bofetiado CJ, Bosari S, Languino LR (1996) Down-regulation of beta 1C integrin, an inhibitor of cell proliferation, in prostate carcinoma. Am J Pathol 149(3): 765–773 100. Goel HL, Breen M, Zhang J, Das I, Aznavoorian-Cheshire S, Greenberg NM, Elgavish A, Languino LR (2005) beta1A integrin expression is required for type 1 insulin-like growth factor receptor mitogenic and transforming activities and localization to focal contacts. Cancer Res 65(15):6692–6700 101. Moro L, Perlino E, Marra E, Languino LR, Greco M (2004) Regulation of beta1C and beta1A integrin expression in prostate carcinoma cells. J Biol Chem 279(3):1692–1702 102. Perlino E, Lovecchio M, Vacca RA, Fornaro M, Moro L, Ditonno P, Battaglia M, Selvaggi FP, Mastropasqua MG, Bufo P, Languino LR (2000) Regulation of mRNA and protein levels of beta1 integrin variants in human prostate carcinoma. Am J Pathol 157(5):1727–1734 103. Koshikawa N, Giannelli G, Cirulli V, Miyazaki K, Quaranta V (2000) Role of cell surface metalloprotease MT1-MMP in epithelial cell migration over laminin-5. J Cell Biol 148(3):615–624 104. Maretzky T, Reiss K, Ludwig A, Buchholz J, Scholz F, Proksch E, de Strooper B, Hartmann D, Saftig P (2005) ADAM10 mediates E-cadherin shedding and regulates epithelial cell-cell adhesion, migration, and beta-catenin translocation. Proc Natl Acad Sci U S A 102(26):9182–9187 105. Lochter A, Galosy S, Muschler J, Freedman N, Werb Z, Bissell MJ (1997) Matrix metalloproteinase stromelysin-1 triggers a cascade of molecular alterations that leads to stable epithelial-to-mesenchymal conversion and a premalignant phenotype in mammary epithelial cells. J Cell Biol 139(7):1861–1872 106. Sehgal G, Hua J, Bernhard EJ, Sehgal I, Thompson TC, Muschel RJ (1998) Requirement for matrix metalloproteinase-9 (gelatinase B) expression in metastasis by murine prostate carcinoma. Am J Pathol 152(2):591–596 107. Sauer CG, Kappeler A, Spath M, Kaden JJ, Michel MS, Mayer D, Bleyl U, Grobholz R (2004) Expression and activity of matrix metalloproteinases-2 and -9 in serum, core needle biopsies and tissue specimens of prostate cancer patients. Virchows Arch 444(6):518–526 108. Morgia G, Falsaperla M, Malaponte G, Madonia M, Indelicato M, Travali S, Mazzarino MC (2005) Matrix metalloproteinases as diagnostic (MMP-13) and prognostic (MMP-2, MMP-9) markers of prostate cancer. Urol Res 33(1):44–50 109. Itoh M, Murata T, Suzuki T, Shindoh M, Nakajima K, Imai K, Yoshida K (2006) Requirement of STAT3 activation for maximal collagenase-1 (MMP-1) induction by epidermal growth factor and malignant characteristics in T24 bladder cancer cells. Oncogene 25(8): 1195–1204 110. Xie TX, Wei D, Liu M, Gao AC, Ali-Osman F, Sawaya R, Huang S (2004) Stat3 activation regulates the expression of matrix metalloproteinase-2 and tumor invasion and metastasis. Oncogene 23(20):3550–3560

Chapter 11

Survival and Growth of Prostate Cancer Cells in the Bone: Role of the Alpha-Receptor for Platelet-Derived Growth Factor in Supporting Early Metastatic Foci Qingxin Liu, Yun Zhang, Danielle Jernigan, and Alessandro Fatatis

Abstract Skeletal dissemination is by far the most feared complication of prostate cancer and the main cause of patients’ demise. Today, 85% of patients are diagnosed with prostate cancer in the absence of clinically evident secondary tumors. However, a significant number of these patients will eventually develop the advanced form of the disease and present with skeletal metastases. A definitive, curative treatment for bone metastatic disease is still an unmet clinical need and the current standard of care relies on merely palliative measures. There is vast consensus on the fact that cancer cells that spread to the skeleton need to find favorable local conditions to survive and grow. Indeed, those cells that fail to receive appropriate support will either remain dormant or undergo cell death, thereby exerting a negligible clinical impact on patients’ quality of life and overall survival. Significant efforts have been made to identify the cellular events as well as signaling molecules and mechanisms implicated in the growth of cancer cells in the bone marrow. Pre-clinical and clinical studies have investigated the role exerted by the receptors for Platelet-Derived Growth Factor (PDGFRs) in promoting skeletal metastases. Interestingly, the alpha isoform of PDGFR (PDGFRa) appears to be directly implicated in the initial colonization of the skeleton, a step of metastatic progression that might have been significantly overlooked.

Q. Liu • Y. Zhang • D. Jernigan Department of Pharmacology and Physiology, Drexel University, College of Medicine, 245 N. 15th Street, MS 488, Philadelphia, PA 19102, USA e-mail: [email protected] A. Fatatis (*) Department of Pharmacology and Physiology, Drexel University, College of Medicine, 245 N. 15th Street, MS 488, Philadelphia, PA 19102, USA Department of Pathology and Laboratory Medicine, Drexel University, College of Medicine, Philadelphia, PA, USA e-mail: [email protected]

A. Fatatis (ed.), Signaling Pathways and Molecular Mediators in Metastasis, DOI 10.1007/978-94-007-2558-4_11, © Springer Science+Business Media B.V. 2012

261

262

11.1

Q. Liu et al.

The Metastatic Process and Bone-Tropism

Metastasis is a process by which cancer cells leaving a primary tumor must successfully complete a series of sequential steps before colonizing in distant organs [1, 2]. After invading the basement membrane and gaining access to blood vessels these Circulating Tumor Cells (CTC) must avoid anoikis and immune destruction to effectively travel through the hematogenous system [3, 4]. In a minority of circumstances, CTC may become physically trapped in capillaries because of size restriction, eventually growing intravascularly before invading the surrounding stroma [5, 6]. However, it is becoming increasingly clear that cells expressing the appropriate arsenal of adhesion molecules and receptors on their plasma membrane can preferentially adhere to the endothelial wall of capillary beds in specific organs [7]. Once firmly attached to the endothelium, cancer cells can then extravasate in response to chemoattractant cues produced by the parenchyma. These two abilities are responsible, at least in part, for the propensity that many solid tumors have to target some organs more than others in the body, a feature defined as organ-tropism [8]. Although several molecules with adhesive and chemoattractant properties for cancer cells have been identified, only a small number of studies have been conducted to conclusively prove that counteracting the completion of these two events by CTC can reduce experimental metastases in pre-clinical models. This paucity of efforts translates into a general indifference for preventive anti-metastatic interventions in the clinical trial setting, a viewpoint also supported by the common belief that occult metastatic foci already exist at diagnosis in the majority of cancer patients. However, CTC are detected with very high frequency in the blood of patients with metastatic tumors after the eradication of the primary neoplasia. Therefore, secondary tumors are a key source of CTC, which plausibly seed additional lesions throughout the duration of the disease. Targeted therapies against adhesive and chemoattractant molecules could block these showers of secondary metastases [9], and ultimately mitigate the explosive expansion of metastatic disease frequently observed in the clinic for several solid tumors. Once arrested at secondary organs, CTC may transition into Disseminated Cancer Cell (DTC) by invading the parenchyma, as described above. This new cell-status does not guarantee progression into a metastatic lesion. In fact, the vast majority of DTC succumb to the new microenvironment [10]. To avoid death, DTC must rapidly adapt to survive in the foreign parenchyma and either proliferate or become dormant. The latter condition may last for several years before cells eventually resume growth and provide clinical evidence of their presence [11]. Indeed, if DTC remain dormant for as long as the patient’s lifetime, their detection might be psychologically unsettling for both the patient and caring physician while exerting an overall insignificant clinical impact. In contrast, the proliferation of DTC into small malignant foci with the ability to progress into macroscopic lesions leads to the inevitable metastatic disease and life-threatening complications [12].

11

Survival and Growth of Prostate Cancer Cells in the Bone…

263

As the difference between dormancy and successful adaptation to the new microenvironment may determine the balance between life and death for the patient, significant work is being conducted to understand the impact that local factors exert in supporting the survival of DTC [13].

11.2

Support Offered by the Bone Marrow Microenvironment to Cancer Cells Disseminated to the Skeleton

The dramatic alterations in bone tissue architecture of metastatic lesions affecting the skeleton of prostate cancer patients points towards recognizable alterations of the mechanisms of physiological bone turnover, implicating bone-matrix synthesizing cells (osteoblasts) and bone-matrix degrading cells (osteoclasts). These studies have therefore delineated a scenario in which, when cancer cells grow into large tumor masses, they can alter the surrounding marrow and compact bone in their favor, establishing plethoric interactions with bone-residing cells and ensuring their progression into fullblown metastases [14, 15]. For instance, malignant phenotypes can produce molecules such as parathyroid-hormone-related peptide (PTHrP) or interleukin-6 (IL-6), which stimulate osteoblasts to produce the Receptor Activator of NF-kB ligand (RANKL). Both soluble and membrane-bound RANKL bind to RANK on the surface of osteoclast-precursor cells leading to their differentiation into mature osteoclasts, which are able to cause degradation of bone matrix and release trophic factors that sustain the proliferation of cancer cell [16]. Progressively more cancer cells will participate in attracting circulating progenitors and generate activated osteoclasts, in a self-feeding vicious-cycle that promotes the survival and expansion of the tumor mass within the bone tissue [17]. It should be emphasized that prostate cancer metastases, generally considered osteoblastic, commonly display large areas of pronounced osteolysis, either isolated or mixed together with pure osteoblastic areas [18]. Thus, as the expansion of the prostate metastatic tumors occurs at the expenses of the bone matrix, this must be degraded to free space that will be taken by the proliferating cancer cells [19]. The possibility that the growth factors supporting tropism and proliferation of cancer cells with demonstrated bone-tropism, such as breast and prostate cancer cells could be released from the bone-matrix degrading activity of osteoclasts led to the idea that the inhibition of these cells would negatively affect the progression of metastatic lesions [20].

11.3 Inhibition of Osteoclast Bone-Resorption Activity to Counteract Skeletal Metastases in Animals and Humans The bone-resorption activity of osteoclasts can be impaired by different compounds. Osteoprotegerin (OPG) is a soluble decoy receptor for RANKL, which prevents RANK activation and subsequent stimulation of osteoclasts differentiation,

264

Q. Liu et al.

activation and survival [21]. Bisphosphonates are non-hydrolysable analogs of pyrophosphates that are endocytosed by osteoclasts during their osteolytic activity and hamper their cellular metabolism, thereby reducing bone-matrix resorption [22]. Denosumab is a humanized antibody directed against RANKL, which in contrast to bisphosphonates does not accumulate in the bone and shows a long half-life in circulation [23]. Administration of either OPG or bisphosphonates has been shown to limit the growth of cancer cells in the skeleton of animal models [24–29]. These studies, conducted implanting cancer cells directly into long bones or inoculating the animal via the hematogenous route, have provided useful information concerning the role of osteolysis in the progression of metastases and created a solid experimental ground for testing drugs counteracting this specific event. For practical reasons, including the need for detecting tumor foci by radiography or bioluminescence, osteoclast-inhibitors have been consistently tested in animals bearing skeletal lesions of significant size. As a result, these studies translate to a clinical setting pertaining to human subjects with large bone tumors associated with substantial pain, risk for pathological fractures and spinal cord compression. Indeed, bisphosphonates currently play an integral role in the treatment regimen of patients with advanced bone-metastatic disease [22]. Zoledronic acid (ZA) is a bisphosphonate showing a potent analgesic effect that can significantly delay the time to Skeletal Related Events (SREs) [30–32]. However, the palliative role of bisphophonates is unfortunately not associated with curative properties. A recent clinical trial in which ZA was compared to placebo in 422 advanced prostate cancer patients failed to show significant differences in disease progression, performance status and quality of life among the groups [33]. In addition, the MRC PR04 and Zometa 704 studies were designed to evaluate the efficacy of clodronate or ZA, respectively, to prevent the emergence of clinically evident skeletal metastases in prostate cancer patients. Both drugs showed death rates, overall survival and time to first bone metastasis similar to placebo groups [20]. Although comparative studies have recently demonstrated a superior effect of denosumab in delaying the occurrence of SREs, this feature was not associated with significant effects on tumor burden and survival [24, 34]. In line with these results, pre-clinical models have show that the progression of the bone metastatic disease from breast cancer cells was transiently delayed, at later stages the total tumor burden per animal became equivalent to that in vehicle-treated animals [29, 35]. Some evidence is emerging about a direct anti-tumor effect of both bisphosphonates and denosumab, which may also counteract the growth of cancer cells in soft-tissue organs. However, additional studies are needed to confirm this possibility [20]. Since the administration of different therapeutics targeting osteoclasts is ineffective in reducing tumor burden and increasing patients’ survival, it seems likely that additional and more critical trophic support for skeletal metastatic lesions is provided by molecules and events somehow unrelated to bone resorption. However, the cellular source and chemical identity of this support are still

11

Survival and Growth of Prostate Cancer Cells in the Bone…

265

unclear and the eradication of cancer cells from fully established metastatic lesions in the skeleton must be considered still unattainable with the therapies currently available. Thus, one could envision that less-established skeletal lesions could be more easily treated due to still incomplete symbiotic interactions between cancer cells and surrounding stroma. However, studies focused on early bone metastatic foci have been rarely conducted. This is probably based on the assumption that the support of bone-matrix degradation to skeletal colonization by DTC could be invoked from the very initial stages of prostate cancer spreading to the skeleton. In addition, bone metastases are commonly studied either examining autoptic human specimens from late-stage lesions or using animal models bearing skeletal tumors sufficiently large to be easily imaged and harvested. These approaches inherently provide very limited information on isolated DTC and small foci that cannot be detected by bioluminescence and are incapable of producing enough skeletal destruction to be visualized by radiographic analyses. There is, however, emerging evidence for a minimal involvement of osteoclasts and the lack of histopathological evidence for osteolysis in early human metastatic lesions [36]. Recent data generated in our lab using a pre-clinical model of metastasis also confirm this observation. In these studies, human prostate cancer cells were engineered to stably express green fluorescent protein (eGFP) and inoculated in the blood circulation of mice. The metastases generated by these cells homing to the skeleton were detected at 1–3 weeks following inoculation. The metastatic tumors with a cross-section area of 28 × 103 mm2 or larger, usually detected 2 weeks after cell inoculation were surrounded by a continuous layer of osteoclasts, identified by TRAcP staining (Fig. 11.1). However, the smaller foci that developed during the first 2 weeks after the arrival of cancer cells to the bone were spatially unrelated to osteoclasts (Fig. 11.1). These results indicate a late involvement of osteoclasts in the progression of bone metastatic foci and further suggest the existence of mechanisms alternative to bone degradation to support the initial survival of prostate cancer cells in early skeletal micro-metastases [37]. As previously emphasized, the potential for therapeutic intervention on DTC and microscopic skeletal lesions should not be underestimated. There is a strong likelihood that initial skeletal foci can subsequently seed secondary sites and that this process leads to a continuous amplification of the number of metastatic lesions [3]. The dramatic increase in bone lesions detected with consecutive radiographs and scintiscans in prostate cancer patients [38], often attributed to the emergence of pre-existing and previously undetected foci, could in fact be due to the new seeding of the skeleton. In both circumstances, the ability to effectively treat these initial lesions to counteract their growth and further establishment in the bone tissue should provide superior therapeutic outcomes than the current strategies against advanced, macroscopic metastases. This would be particularly true for prostate cancer, in which a negative correlation between the rate of increase of the number of bone lesions and survival time has been reported [38].

266

Q. Liu et al.

Fig. 11.1 Metastatic progression from initial homing in the bone marrow to macroscopic skeletal tumors and spatial relationship with osteoclasts. Green-fluorescent human prostate cancer cells were inoculated in the blood circulation of immunocompromised mice via left cardiac ventricle. The presence of active osteoclasts in the bone marrow regions colonized by cancer cells was histologically established by TRAcP staining (lower insets). Metastases with cross-section area larger than 28 × 103 mm2 – indicated by the green fluorescent signal – were surrounded by an evident layer of active osteoclasts, as shown in the magnified panel. In contrast, smaller metastases were spatially unrelated to osteoclasts, which appear sparsely distributed (black arrows). These observation lead to hypothesize that the initial stages of bone colonization by prostate cancer cells do not require the intervention of osteoclasts

11.4

Which Factors Allow Cancer Cells to Survive the Early Stages of Metastatic Colonization of the Bone?

Immediately after their homing into the bone marrow, DTC are found as single cells or small foci and considered vulnerable to the new microenvironment [14, 39]. A lack of osteoclasts involvement at these stages, as proposed above, should be considered likely. In fact, the recruitment of osteoclast-progenitors from the blood circulation to the bone-seeded sites by DTC-derived soluble factors might be very difficult unless a tumor mass of sufficient size is first attained. During these initial steps, DTC may rather rely on local factors readily available in the surrounding environment as a result of normal metabolic activities, possibly related to hematopoiesis and not necessarily involving bone turnover [40–42]. As an alternative, DTC could coopt nearby bone stromal cells to produce survival factors, in response to cytokines produced in a paracrine fashion [43]. This latest scenario

11

Survival and Growth of Prostate Cancer Cells in the Bone…

267

could also contribute to conditioning the bone microenvironment to make it more receptive to successive waves of metastatic cells [44]. Finally, the preparation of different organ microenvironments for the successive colonization of DTC could derive from the establishment of a pre-metastatic niche [45, 46]. Accordingly, the expression of appropriate receptors for these trophic factors would provide selected DTC phenotypes with a significant survival advantage during these early metastatic stages [47].

11.5

Receptors for Platelet Derived Growth Factor in Primary and Metastatic Prostate Cancer

Studies conducted by us and others show that primary prostate cancers and their skeletal metastases are positive for PDGFR expression, with the PDGFRa isoform being the most represented [37, 48]. Normal prostate glands express low levels of PDGFRa, which increase upon malignant transformation [37, 49]. Interestingly, heterogeneous expression of PDGFRa within the same malignant prostate tissue can be observed. This suggests that cancer phenotypes with different expression patterns for PDGFRa coexist in the same gland, a scenario reproduced by the human prostate cancer cell lines currently available. Testing these cell lines, we found that among prostate cells originally obtained from different metastatic sites, the expression of the PDGFRa is detected only in the bone metastasis-derived cells, whereas cells obtained from patients with lymph node or brain metastases fail to express PDGFRs [50]. A demonstrated instrumental role of PDGFRa in sustaining the survival and growth of disseminated prostate cancer cells would provide an attractive target for therapy. This is particularly true in light of the partially overlapping role that the alpha and beta isoforms of PDGFR play in the physiology of the adult organism, as described below. Oncology has one of the poorest records for investigational drugs [51]. A significant problem is that the majority of proto-oncogenes exert functions that are essential for both malignant and normal cells. Obviously, drugs targeting essential proteins present high risk of significant side effects, which provide physicians with narrow therapeutic windows and often leads to ineffective tumor eradication or suspension of the therapeutic regimen due to intolerable toxicity. Identifying and targeting molecules more relevant for malignant than normal cells seems therefore a rational strategy and PDGFRa appears to be an extremely viable candidate. We have recently shown that, in addition to the PDGFs normally produced in the bone microenvironment [40, 52], PDGFRa in prostate cancer cells can be also recruited in a ligand-independent fashion (trasactivation) by the soluble fraction of the human bone marrow [53, 54]. In addition, the exogenous overexpression of this receptor can confer bone-metastatic potential to prostate phenotypes that would normally fail to grow after extravasating at the skeleton [37]. Taken together, these observations provide convincing experimental ground to implicate PDGFRa in skeletal metastasis.

268

11.6

Q. Liu et al.

PDGFR Structure and Signaling

PDGFRs are tyrosine kinases and some of the best-studied growth factor receptors. Two structurally related forms of PDGFRs are known, PDGFRa and PDGFRb. Their extracellular portion contains five immunoglobulin-like domains whereas the intracellular part of the molecule contains a kinase domain [55]. Five PDGF ligands have been to date identified, PDGF-AA, -BB, -AB, -CC and –DD, which display different binding affinities for the different receptors [56]. Since PDGF ligands are dimeric molecules, they bind two receptors simultaneously. Upon binding, the receptor molecules also dimerize and this process triggers the reciprocal phosphorylation at tyrosine residues located at specific sites on the intracellular portion of each receptor [57–59]. This trans phosphorylation of PDGFR upon ligand binding serves two important purposes. First, phosphorylation of a tyrosine residue in the kinase domain increases its catalytic efficiency. In addition, phosphorylation of tyrosine residues outside the kinase domain creates docking sites for signaling molecules. Some of these molecules can function as enzymes, such as phosphatidylinositol 3¢-kinase (PI3K), phospholipase C (PLC)-g, the Src family of tyrosine kinases, the tyrosine phosphatase SHP-2 and a GTPase activating protein (GAP) for Ras. Additional signaling molecules lack enzymatic activity and function as adaptors, such as Grb2, Nck, Shc and others [55]. The biological functions of some of these signaling molecules have been characterized and are fundamental for cellular homeostasis. PI3K activates the downstream kinase Akt, which is strongly involved in promoting cellular survival [60–63]. The recruitment of Ras by GAP will activate the MAP kinase pathway, which is determinant for cell proliferation [64, 65]. The PLC family of enzymes mobilizes intracellular calcium, which is a crucial second messenger for all metazoan cells [66]. Finally, Src is important for mitogenic responses induced by PDGF as well as other growth factors [67–70]. The binding of ligands to PDGFRs induces internalization, which can be followed by either recycling of the receptor to the cell surface or its degradation. Interestingly, the rate of PDGFR internalization is an important parameter in the regulation of signaling from this receptor [55]. Both PDGFRa and PDGFRb are implicated in development, however PDGFR plays a more prominent role during embryogenesis [71, 72]. In the adult organism, both receptors cooperate in controlling cellular and physiological processes that largely overlap, such as angiogenesis, wound healing and tissue homeostasis [71, 73, 74]. However, PDGFRb plays a definitely predominant role, as shown by experiments in mice in which the cytoplasmic domains between PDGFRa and –b were swapped. Whereas the intracellular domain of PDGFRb could fully substitute for the PDGFRa, the replacement of the PDGFRb cytoplasmic domain with that of PDGFRa caused abnormalities in vascular smooth muscle cell development and function [75].

11

Survival and Growth of Prostate Cancer Cells in the Bone…

11.7

269

PDGFRa as Therapeutic Target in Bone-Metastatic Cancer

Initial studies investigating the effect of PDGFR inhibition on the growth of prostate cancer cells have used the small-molecule inhibitor Imatinib mesylate (Gleevec). This compound had been originally developed to block the BRC-ABL tyrosine kinase in chronic myeloid leukemia and successively it was found to target also the KIT kinase and PDGFR kinase activities in gastrointestinal stromal tumors [76]. Because of the remarkable effects shown in these particular forms of tumor, Imatinib is considered as one of the best examples of targeted therapeutics [77]. When tested in animal models of prostate cancer skeletal metastasis, Imatinib was reported as being able to reduce the expansion of cancer cells within the bone [78]. However, its effects seemed exerted prevalently on the PDGFRb expressed by the endothelial cells of the tumor vasculature rather than affecting directly the survival and growth of prostate cancer cells [79]. A similar mechanism for PDGFR blockade was also recently proposed to impair lung cancer skeletal metastasis. This study reported large areas of necrosis and hemorrhage at the skeletal levels due to the significant, albeit transient apoptosis induced in the bone marrow stromal cells of mice treated with the multi kinase inhibitor sunatinib [80]. However, while the effect described could reduce the dissemination of cancer cells to the skeleton in patients, one is left to wonder whether the toxic side effects procured could possibly make this approach clinically unsound. In addition, when the blockade of PDGFRs involved both alpha and beta isoforms with comparable efficacy, as observed with Imatinib, a predictable consequence was the toxicity reported in phase I and II clinical trials conducted with this compound, which in most cases had to be interrupted [81, 82]. It should be emphasized that the pre-clinical studies supporting a role of Imatinib in impairing the survival of prostate cancer cells were almost exclusively conducted using animal models in which bone tumors were produced by directly implanting cancer cells using an intra-osseous route. Although this approach significantly shortens the duration of each experiment, it also bypasses the initial stages of homing and colonization of the bone environment. The discrepancies between the histopathological features produced by the intra-osseous approach and the alterations generated by cancer cells that reach the bone via the hematogenous route and grow in a progressive fashion in the marrow might easily explain the disappointing effects of Imatinib in the treatment of advanced prostate cancer. In light of this evidence, it seems plausible that the selective inactivation of PDGFRa – using a monoclonal antibody rather than a broad-range inhibitor such as Imatinib – could limit the survival of malignant cells that depend on it while causing limited side effects, due to the largely duplicate role exerted by PDGFRb [83].

270

11.8

Q. Liu et al.

Targeting PDGFRa Counteracts the Early Stages of Skeletal Metastasis by Prostate Cancer Cells

Initial studies employing human glioblastoma and leiomyosarcoma xenografts in athymic mice tested the antitumor effect of the IMC-3G3 antibody (Olaratumab). This monoclonal, fully humanized antibody was found to specifically blocks PDGFRa signaling and significantly inhibit tumor growth [84] and is currently being evaluated in patients with several forms of solid tumors [83]. When IMC-3G3 was administered to mice inoculated with prostate cancer cells using either a preventive or curative protocol, the bone-metastatic tumor burden was drastically reduced and, in additional studies, the overall survival significantly prolonged [37, 85]. More importantly, these effects seem exerted on the early stages of skeletal colonization by prostate cancer cells. This conclusion could be drawn from experiments in which mice inoculated with cancer cells were treated with IMC-3G3 and sacrificed after just 1 week of skeletal colonization. When bone metastatic lesions were compared to those detected in the animals of the saline control group, the PDGFRatargeted antibody showed a dramatic effect in counteracting prostate cancer cells growth. More importantly, mice treated with ZA showed bone tumors that were indistinguishable in size from those detected in control animals, in line with the idea that the initial stages of bone colonization by prostate cancer cells may not require osteoclasts’ involvement and matrix degradation to support metastatic progression. However, these early colonization stages, while refractory to the effects of bisphosphonates, are susceptible to PDGFRa blockade (Fig. 11.2).

11.9

Concluding Remarks

The process of bone marrow colonization by prostate cancer cells during skeletal metastasis is exceedingly complex and likely undergoing multiple phases of cross talk between malignant epithelial cells and stromal components. To date, most energy has been spent in investigating the late stages of metastatic progression, in an effort to characterizing the histopathological features and identifying molecular tissue factors characteristic of advanced, clinically symptomatic skeletal metastases. Although our understanding of these advanced lesions has significantly improved, we have also learned that late metastatic disease is remarkably difficult to treat and a curative outcome almost impossible to achieve. The fact that currently only palliative care can be effectively provided to patients with advanced, metastatic prostate cancer should not deter us from additional studies centered on fully established, macroscopic skeletal tumors. However, this approach should be combined with the investigation of the early metastatic stages, which must be developing even when still clinically undetectable, can be now effectively studies in animal models and include cancer cells that are

11

Survival and Growth of Prostate Cancer Cells in the Bone…

271

Fig. 11.2 The blockade of PDGFRa impairs the early stages of bone-metastatic tumor growth. Mice were inoculated with green-fluorescent human prostate cancer cells via the hematogenous route, treated with either the bisphosphonate Zoledronic Acid or the monoclonal anti-PDGFRa antibody IMC-3G3 (Olaratumab) and sacrificed 1 week later. Early bone-metastatic lesions were identified by fluorescence stereomicroscopy. Tumor foci in the two groups of animals were compared to those detected in control, saline-treated animals. Mice in the IMC-3G3-treated groups showed lesions significantly reduced in size, whereas bone metastatic tumors detected in mice that received ZA were comparable in size to those measured in control animals (*p < 0.0001) (See also Russell et al. [85])

most likely significantly more vulnerable to therapeutic intervention. In this context, here we have provided evidence that the possibility of targeting PDGFRa in prostate cancer cells should be definitely taken into serious consideration.

References 1. Gupta GP, Massagué J (2006) Cancer metastasis: building a framework. Cell 127:679–695 2. Coghlin C, Murray GI (2010) Current and emerging concepts in tumour metastasis. J Pathol 222:1–15 3. Chaffer CL, Weinberg RA (2011) A perspective on cancer cell metastasis. Science 331:1559–1564 4. Gay LJ, Felding-Habermann B (2011) Contribution of platelets to tumour metastasis. Nat Rev Cancer 11(2):123–134. 5. Al-Mehdi AB, Tozawa K, Fisher AB et al (2000) Intravascular origin of metastasis from the proliferation of endothelium-attached tumor cells: a new model for metastasis. Nat Med 6:100–102

272

Q. Liu et al.

6. Alix-Panabières C, Riethdorf S, Pantel K (2008) Circulating tumor cells and bone marrow micrometastasis. Clin Cancer Res: an official journal of the American Association for Cancer Research 14:5013–5021 7. Glinskii O, Huxley V, Glinsky G et al (2005) Mechanical entrapment is insufficient and intercellular adhesion is essential for metastatic cell arrest in distant organs. Neoplasia 7:522–527 8. Nguyen DX, Bos PD, Massagué J (2009) Metastasis: from dissemination to organ-specific colonization. Nat Rev Cancer 9:274–284 9. Hanahan D, Weinberg RA (2011) Hallmarks of cancer: the next generation. Cell 144:646–674 10. Shibue T, Weinberg RA (2011) Metastatic colonization: settlement, adaptation and propagation of tumor cells in a foreign tissue environment. Semin Cancer Biol 21:99–106 11. Weinberg RA (2008) The many faces of tumor dormancy. APMIS: acta pathologica, microbiologica, et immunologica Scandinavica 116:548–551 12. Lin H, Balic M, Zheng S et al (2011) Disseminated and circulating tumor cells: role in effective cancer management. Crit Rev Oncol Hematol 77:1–11 13. Cawthorn TR, Amir E, Broom R et al (2009) Mechanisms and pathways of bone metastasis: challenges and pitfalls of performing molecular research on patient samples. Clin Exp Metastasis 26:935–943 14. Roodman GD (2007) Bone-breaking cancer treatment. Nat Med 13:25–26 15. Buijs JT, van der Pluijm G (2009) Osteotropic cancers: from primary tumor to bone. Cancer Lett 273:177–193 16. Roodman GD (2004) Mechanisms of bone metastasis. N Engl J Med 350:1655–1664 17. Kingsley LA, Fournier PG, Chirgwin JM et al (2007) Molecular biology of bone metastasis. Mol Cancer Ther 6:2609–2617 18. Cheville JC, Tindall D, Boelter C et al (2002) Metastatic prostate carcinoma to bone: clinical and pathologic features associated with cancer-specific survival. Cancer 95:1028–1036 19. Keller ET, Brown J (2004) Prostate cancer bone metastases promote both osteolytic and osteoblastic activity. J Cell Biochem 91:718–729 20. Lee RJ, Saylor PJ, Smith MR (2011) Treatment and prevention of bone complications from prostate cancer. Bone 48:88–95 21. Jones DH, Nakashima T, Sanchez OH et al (2006) Regulation of cancer cell migration and bone metastasis by RANKL. Nature 440:692–696 22. Saad F, Schulman CC (2004) Role of bisphosphonates in prostate cancer. Eur Urol 45:26–34 23. Bouganim N, Clemons MJ (2011) Bone-targeted agents in the treatment of bone metastases: RANK outsider or new kid on the block? Future Oncol 7:381–383 24. Nyambo R, Cross N, Lippitt J et al (2004) Human bone marrow stromal cells protect prostate cancer cells from TRAIL-induced apoptosis. J Bone Miner Res 19:1712–1721 25. Corey E, Brown LG, Kiefer JA et al (2005) Osteoprotegerin in prostate cancer bone metastasis. Cancer Res 65:1710–1718 26. Canon JR, Roudier M, Bryant R et al (2008) Inhibition of RANKL blocks skeletal tumor progression and improves survival in a mouse model of breast cancer bone metastasis. Clin Exp Metastasis 25:119–129 27. Lee Y-P, Schwarz EM, Davies M et al (2002) Use of zoledronate to treat osteoblastic versus osteolytic lesions in a severe-combined-immunodeficient mouse model. Cancer Res 62:5564–5570 28. Corey E, Brown LG, Quinn JE et al (2003) Zoledronic acid exhibits inhibitory effects on osteoblastic and osteolytic metastases of prostate cancer. Clin Cancer Res: an official journal of the American Association for Cancer Research. 9:295–306 29. Thudi NK, Martin CK, Nadella MVP et al (2008) Zoledronic acid decreased osteolysis but not bone metastasis in a nude mouse model of canine prostate cancer with mixed bone lesions. The Prostate 68:1116–1125 30. Costa L, Major PP (2009) Effect of bisphosphonates on pain and quality of life in patients with bone metastases. Nat Clin Pract Oncol 6:163–174 31. Bagi CM (2005) Targeting of therapeutic agents to bone to treat metastatic cancer. Adv Drug Deliv Rev 57:995–1010

11

Survival and Growth of Prostate Cancer Cells in the Bone…

273

32. Coleman RE, Guise T, Lipton A et al (2008) Advancing treatment for metastatic bone cancer: consensus recommendations from the Second Cambridge Conference. Clin Cancer Res 14:6387–6395 33. Saad F, Gleason DM, Murray R et al (2002) A randomized, placebo-controlled trial of zoledronic acid in patients with hormone-refractory metastatic prostate carcinoma. J Natl Cancer Inst 94:1458–1468 34. Stopeck AT, Lipton A, Body J-J et al (2010) Denosumab compared with zoledronic acid for the treatment of bone metastases in patients with advanced breast cancer: a randomized, double-blind study. J Clin Oncol 28:5132–5139 35. van der Pluijm G, Que I, Sijmons B et al (2005) Interference with the microenvironmental support impairs the de novo formation of bone metastases in vivo. Cancer Res 65:7682–7690 36. Clarke NW, Hart CA, Brown MD (2009) Molecular mechanisms of metastasis in prostate cancer. Asian J Androl 11:57–67 37. Russell M, Jamieson W, Dolloff N et al (2009) The alpha-receptor for platelet-derived growth factor as a target for antibody-mediated inhibition of skeletal metastases from prostate cancer cells. Oncogene 28:412–421 38. Ohmori K, Matsui H, Yasuda T et al (1997) Evaluation of the prognosis of cancer patients with metastatic bone tumors based on serial bone scintigrams. Jpn J Clin Oncol 27:263–267 39. Fidler I (2003) The pathogenesis of cancer metastasis: the ‘seed and soil’ hypothesis revisited. Nat Rev Cancer 3:453–458 40. Mundy GR (2000) Structure and physiology of the normal skeleton. In: Robert D, Rubens GRM (ed) Cancer and the skeleton, 3rd edn. London, UK: Informa Healthcare, pp 1–19 41. Zhang S, Wang J, Bilen M et al (2009) Modulation of prostate cancer cell gene expression by cell-to-cell contact with bone marrow stromal cells or osteoblasts. Clin Exp Metastasis 26(8):993–1004 42. Shiozawa Y, Pedersen EA, Havens AM et al (2011) Human prostate cancer metastases target the hematopoietic stem cell niche to establish footholds in mouse bone marrow. J Clin Invest 121(4):1298–1312 43. Bussard KM, Venzon DJ, Mastro AM (2010) Osteoblasts are a major source of inflammatory cytokines in the tumor microenvironment of bone metastatic breast cancer. J Cell Biochem 111(5):1138–1148 44. Bidard F-C, Pierga J-Y, Vincent-Salomon A et al (2008) A “class action” against the microenvironment: do cancer cells cooperate in metastasis? Cancer Metastasis Rev 27:5–10 45. Kaplan RN, Rafii S, Lyden D (2006) Preparing the “soil”: the premetastatic niche. Cancer Res 66:11089–11093 46. Kaplan RN, Psaila B, Lyden D (2007) Niche-to-niche migration of bone-marrow-derived cells. Trends Mol Med 13:72–81 47. Fidler I, Poste G (2008) The “seed and soil” hypothesis revisited. Lancet Oncol 9:808 48. Chott A, Sun Z, Morganstern D et al (1999) Tyrosine kinases expressed in vivo by human prostate cancer bone marrow metastases and loss of the type 1 insulin-like growth factor receptor. Am J Pathol 155:1271–1279 49. Fudge K, Wang CY, Stearns ME (1994) Immunohistochemistry analysis of platelet-derived growth factor A and B chains and platelet-derived growth factor alpha and beta-receptor expression in benign prostatic hyperplasias and Gleason-graded human prostate adenocarcinomas. Mod Pathol 7:549–554 50. Dolloff N, Shulby S, Nelson A et al (2005) Bone-metastatic potential of human prostate cancer cells correlates with Akt/PKB activation by alpha platelet-derived growth factor receptor. Oncogene 24:6848–6854 51. Kamb A, Wee S, Lengauer C (2007) Why is cancer drug discovery so difficult? Nat Rev Drug Discov 6:115–120 52. Yang D (2000) Platelet-derived growth factor (PDGF)-AA: a self-imposed cytokine in the proliferation of human fetal osteoblasts. Cytokine 12:1271–1274

274

Q. Liu et al.

53. Dolloff N, Russell MR, Loizos N et al (2007) Human bone marrow activates the Akt pathway in metastatic prostate cells through transactivation of the alpha-platelet-derived growth factor receptor. Cancer Res 67:555–562 54. Russell MR, Liu Q, Lei H et al (2010) The alpha-receptor for platelet-derived growth factor confers bone-metastatic potential to prostate cancer cells by ligand- and dimerizationindependent mechanisms. Cancer Res 70:4195–4203 55. Heldin CH, Westermark B (1999) Mechanism of action and in vivo role of platelet-derived growth factor. Physiol Rev 79:1283–1316 56. Fredriksson L, Li H, Eriksson U (2004) The PDGF family: four gene products form five dimeric isoforms. Cytokine Growth Factor Rev 15:197–204 57. Kazlauskas A, Cooper JA (1989) Autophosphorylation of the PDGF receptor in the kinase insert region regulates interactions with cell proteins. Cell 58:1121–1133 58. Bishayee S, Majumdar S, Khire J et al (1989) Ligand-induced dimerization of the plateletderived growth factor receptor. Monomer-dimer interconversion occurs independent of receptor phosphorylation. J Biol Chem 264:11699–11705 59. Baxter RM, Secrist JP, Vaillancourt RR et al (1998) Full activation of the platelet-derived growth factor beta-receptor kinase involves multiple events. J Biol Chem 273:17050–17055 60. Datta SR, Brunet A, Greenberg ME (1999) Cellular survival: a play in three Akts. Genes Dev 13:2905–2927 61. Vivanco I, Sawyers CL (2002) The phosphatidylinositol 3-Kinase AKT pathway in human cancer. Nat Rev Cancer 2:489–501 62. Luo J, Manning B, Cantley L (2003) Targeting the PI3K-Akt pathway in human cancer Rationale and promise. Cancer Cell 4:257–262 63. Chang F, Lee JT, Navolanic PM et al (2003) Involvement of PI3K/Akt pathway in cell cycle progression, apoptosis, and neoplastic transformation: a target for cancer chemotherapy. Leukemia 17:590–603 64. Schaeffer HJ, Weber MJ (1999) Mitogen-activated protein kinases: specific messages from ubiquitous messengers. Mol Cell Biol 19:2435–2444 65. Chang F, Steelman LS, Lee JT et al (2003) Signal transduction mediated by the Ras/Raf/MEK/ ERK pathway from cytokine receptors to transcription factors: potential targeting for therapeutic intervention. Leukemia 17:1263–1293 66. Berridge M, Bootman M, Roderick HL (2003) Calcium signalling: dynamics, homeostasis and remodelling. Nat Rev Mol Cell Biol 4:517–529 67. Gelderloos JA, Rosenkranz S, Bazenet C et al (1998) A role for Src in signal relay by the platelet-derived growth factor alpha receptor. J Biol Chem 273:5908–5915 68. DeMali KA, Godwin SL, Soltoff SP et al (1999) Multiple roles for Src in a PDGF-stimulated cell. Exp Cell Res 253:271–279 69. Yeatman TJ (2004) A renaissance for SRC. Nat Rev Cancer 4:470–480 70. Im E, Kazlauskas A (2007) Src family kinases promote vessel stability by antagonizing the Rho/ROCK pathway. J Biol Chem 282:29122–29129 71. Betsholtz C (2004) Insight into the physiological functions of PDGF through genetic studies in mice. Cytokine Growth Factor Rev 15:215–228 72. Tallquist M, Kazlauskas A (2004) PDGF signaling in cells and mice. Cytokine Growth Factor Rev 15:205–213 73. Hoch RV (2003) Roles of PDGF in animal development. Dev Suppl 130:4769–4784 74. Andrae J, Gallini R, Betsholtz C (2008) Role of platelet-derived growth factors in physiology and medicine. Genes Dev 22:1276–1312 75. Yarden Y, Escobedo JA, Kuang WJ et al (1986) Structure of the receptor for platelet-derived growth factor helps define a family of closely related growth factor receptors. Nature 323:226–232 76. Von Mehren M (2004) Targeted therapy with imatinib: hits and misses? J Clin Oncol 23:8–10 77. George DJ (2002) Receptor tyrosine kinases as rational targets for prostate cancer treatment: platelet-derived growth factor receptor and imatinib mesylate. Urology 60:115–121; discussion 122

11

Survival and Growth of Prostate Cancer Cells in the Bone…

275

78. Uehara H, Kim S, Karashima T et al (2003) Effects of blocking platelet-derived growth factor-receptor signaling in a mouse model of experimental prostate cancer bone metastases. J Natl Cancer Inst 95:458 79. Kim SJ, Uehara H, Yazici S et al (2006) Targeting platelet-derived growth factor receptor on endothelial cells of multidrug-resistant prostate cancer. J Natl Cancer Inst 98:783–793 80. Catena R, Luis-Ravelo D, Antón I et al (2011) PDGFR signaling blockade in marrow stroma impairs lung cancer bone metastasis. Cancer Res 71:164–174 81. Rao K, Goodin S, Levitt M et al (2005) A phase II trial of imatinib mesylate in patients with prostate specific antigen progression after local therapy for prostate cancer. Prostate 62:115–122 82. Lin A, Rini B, Weinberg V et al (2006) A phase II trial of imatinib mesylate in patients with biochemical relapse of prostate cancer after definitive local therapy. BJU Int 98:763–769 83. Shah GD, Loizos N, Youssoufian H et al (2010) Rationale for the development of IMC-3G3, a fully human immunoglobulin G subclass 1 monoclonal antibody targeting the platelet-derived growth factor receptor a. Cancer 116:1018–1026 84. Loizos N, Xu Y, Huber J et al (2005) Targeting the platelet-derived growth factor receptor alpha with a neutralizing human monoclonal antibody inhibits the growth of tumor xenografts: implications as a potential therapeutic target. Mol Cancer Ther 4:369–379 85. Russell MR, Liu Q, Fatatis A (2010) Targeting the alpha receptor for platelet-derived growth factor as a primary or combination therapy in a preclinical model of prostate cancer skeletal metastasis. Clin Cancer Res 16:5002–5010

Chapter 12

Androgen Receptor Regulation of Prostate Cancer Progression and Metastasis R.S. Schrecengost, M.A. Augello, and Karen E. Knudsen

Abstract Prostate cancer (PCa) is the most commonly diagnosed non-cutaneous malignancy and the second most lethal cancer in men amongst the United States. Localized tumors are effectively treated via radical prostatectomy and/or radiation therapy, however disseminated disease contributes greatly to patient morbidity. At present, therapeutic intervention for metastatic disease capitalizes on the addiction of these tumors to the androgen receptor (AR). The AR signaling axis regulates cell growth, proliferation, and migration of cancer cells, which makes understanding the contribution of AR toward these signaling pathways integral for PCa eradication. Several key signaling molecules are currently known to be controlled by AR and promote migration and invasion in castrate-resistant PCa (CRPC). This concept will be explored with regard to the interplay between AR and chemokine receptors, chromosomal fusions, oncogenes, and microRNAs. Additionally, current and preclinical AR-directed therapeutics used for treatment of metastatic PCa, which impinge on these pathways, and overall AR activity, will be discussed.

R.S. Schrecengost • M.A. Augello Department of Cancer Biology, Thomas Jefferson University, Philadelphia, PA, USA Kimmel Cancer Center, Thomas Jefferson University, 233 10th St., BLSB 1008, Philadelphia, PA 19107, USA K.E. Knudsen (*) Department of Cancer Biology, Thomas Jefferson University, Philadelphia, PA, USA Kimmel Cancer Center, Thomas Jefferson University, 233 10th St., BLSB 1008, Philadelphia, PA 19107, USA Department of Urology, Thomas Jefferson University, Philadelphia, PA, USA Department of Radiation Oncology, Thomas Jefferson University, Philadelphia, PA, USA e-mail: [email protected]

A. Fatatis (ed.), Signaling Pathways and Molecular Mediators in Metastasis, DOI 10.1007/978-94-007-2558-4_12, © Springer Science+Business Media B.V. 2012

277

278

R.S. Schrecengost et al.

Abbreviations Abiraterone ADT AR BC CDKS CML CRPC CTCs DHT E-Box ETS FAK FISH Flutamide GnRH HATs HSPs IHC KLF5 LBD MMPs miRNA MYC NCoR1 PIN PcG PCa PSA SDF-1 SFK SRC TRAMP TMPRSS2 VCaP

12.1

Abiraterone acetate Androgen deprivation therapy Androgen receptor Bicalutamide (Casodex) Cyclin-dependent kinases Chronic myelogenous leukemia Castration resistant prostate cancer Circulating tumor cells 5α-dihydrotestosterone Enhancer box Erythroblast transformation specific transcription factor Focal adhesion kinase Florescence in-situ hybridization Hydroxyflutamide Gonadotropin releasing hormone Histone acetyltransferases Heat shock proteins Immunohistochemistry Kruppel-like-factor 5 Ligand binding domain Matrix metalloproteinases MicroRNA c-Myc Nuclear Co-Repressor 1 Prostatic intraepithelial neoplasia Polycomb-group Prostate cancer Prostate specific antigen Stromal cell-derived factor 1 Macrophage inflammatory protein Steroid receptor coactivator Transgenic adenocarcinoma of the mouse prostate Transmembrane protease serine 2 Vertebral-Cancer of the Prostate

Part I: Clinical Prostate Cancer

Prostate cancer (PCa) is the most commonly diagnosed non-cutaneous malignancy and the second most lethal cancer in men amongst the United States [1]. Early detection is essential for positive prognosis, as localized disease is effectively treated

12

Androgen Receptor Regulation of Prostate Cancer Progression and Metastasis

279

through prostate resection or radiation therapy [2]. However, advanced PCa, which has disseminated to local or distant tissue, proves a significant clinical challenge. An overwhelming majority of these tumors are intrinsically refractory to standard chemotherapeutic treatments. Therefore, the first line clinical target for this stage of disease is the androgen receptor (AR), as AR activity has been determined to be essential for PCa development and progression. Extensive molecular, animal, and clinical models of PCa have demonstrated that tumors are addicted to AR signaling for both growth and survival at all stages of disease. As such, standard-of-care for non-organ confined tumors targets the AR signaling axis, whereby patients are subjected to hormone therapy, biochemically intended to ablate AR signaling [3–5]. This regimen results in an initial regression of the tumor and short-term remission. However, remission is transient, as recurrent tumors invariably arise, initiating the formation of castration resistant prostate cancers (CRPCs) [6]. In the development of CRPC, the AR signaling axis has been inappropriately reactivated, which promotes not only growth of the tumor, but can confer enhanced metastatic potential [7, 8]. These traits ultimately facilitate colonization of distant tissues (mainly bone), an event that promotes morbidity [9]. Consequently, there is an intense effort to understand the molecular pathways that govern how the AR signaling axis is reactivated in CRPC, and determine how reactivation contributes to metastatic disease.

12.1.1

The Role of AR in Prostate Cancer

AR is a member of the steroid nuclear receptor family, and functions as a liganddependent transcription factor. Expression of AR is detectable in a wide variety of tissues in both sexes, but has important clinical implications for the development and progression of prostate cancer [4]. In normal prostate tissue, AR plays a crucial role in maintaining architecture and function of the glandular epithelial tissue by regulating programs of differentiation and controlled growth [10]. Pathway activation is initiated via entry of circulating testosterone into prostatic epithelia, wherein it is converted into the potent AR ligand 5a-dihydrotestosterone (5a-DHT) [11]. This conversion event is influential on AR activation, as the affinity for testosterone by AR is nearly 100 times weaker than that of DHT [11]. Consequently, activation of AR is initiated upon DHT binding to the ligand binding domain (LBD). Upon binding, AR is freed from its inhibitory heat shock proteins (HSP) and translocates into the nucleus, which initiates AR homo-dimerization, posttranslational modifications, and association with DNA [12]. Therein, AR homodimers localize to specific sequences in the DNA (direct AR binding occurs at specific DNA sequences named androgen response elements (AREs)), and therein recruit co-modulators (co-activators, co-repressors, and transcriptional machinery) responsible for eliciting programs of proliferation and differentiation within the prostate [13–15] (Fig. 12.1). AR activity can be monitored clinically through induction of the well-defined AR target gene KLK3. KLK3 encodes a secreted prostate specific serine/threonine protease termed prostate specific antigen, (PSA),

280

R.S. Schrecengost et al.

Fig. 12.1 AR regulation in prostate cancer. In the absence of ligand, AR is sequestered in the cytoplasm by heat shock proteins (HSPs). Testosterone is converted into DHT by 5a-reductase and binds to AR. Upon activation by ligand, AR dissociates from HSPs, translocates to the nucleus, and binds at specific sites with the genome. Activation of transcriptional machinery results in expression of target genes. The consequential gene expression programs induce differentiation, survival, migration and proliferation. Clinically, AR activation (and tumor burden) can be monitored through PSA detection in serum

which can be measured in patient serum and used as a metric to determine relative AR activity [16]. As such, a dramatic or sustained increase in serum PSA is used as a marker of disease development and progression in both early and late stage disease [17].

12.1.2

Targeting the Androgen Receptor in Advanced Disease

The ability of AR to regulate gene expression (i.e., PSA) is governed by its ability to bind ligand, localize to specific sites on DNA, and recruit necessary transcriptional machinery [13, 15, 18]. As this process has been shown to be crucial for the growth of prostate cancer at all stages, it is not surprising that the AR pathway is the main target of most therapeutics used clinically to treat non-localized disease [4, 18, 19]. Typically, disseminated prostate adenocarcinomas present as androgen-dependent, and the goal of first line treatment is to deprive AR of ligand. Such strategies are referred to as androgen depletion therapy (ADT) and take advantage of gonadotropinreleasing hormone (GnRH) agonists, which suppress secretion of the pituitary hormone luteinizing hormone (LH), thereby inhibiting the synthesis of testicular androgens [4, 18, 19]. After an initial surge of testosterone production [20], circulating

12

Androgen Receptor Regulation of Prostate Cancer Progression and Metastasis

281

levels drop to those that mimic castrated males (<0.2 ng/dL) [21]. To maximize AR inhibition, combination therapy is often utilized, where patients are treated with ADT and direct AR antagonists (i.e., bicalutamide), which function as competitive inhibitors (and inverse agonists) to limit latent AR activity [4, 18, 19]. ADT proves exceedingly efficient in most patients, as tumor regression is preceded by a decline in serum PSA levels, indicative of molecular inhibition of the AR axis [4, 18]. Unfortunately, treatment failure often occurs after 2–3 years, wherein recurrent tumor formation is heralded by rising levels of PSA (biochemical recurrence) [4, 18, 19]. This failure is thought to be attributed to the selective pressure induced during ADT for cells that have inappropriately reactivated AR. Subsequent molecular modeling of human disease uncovered a variety of different mechanisms by which this process can occur that are of significant clinical relevance [18]. As a result, it is widely believed that it is through these biological alterations that PCa progresses to CRPC and contribute to lethal metastatic phenotypes.

12.1.3

Adaptive Mechanisms of AR Reactivation in CRPC

CRPC is characterized by the ability of the tumor to maintain a functional AR signaling axis, even in the absence of ligand and/or presence of antagonists [4, 18]. Such adaptations have been linked to changes in AR itself, as well as in alterations of key regulators in the AR signaling network [18, 22, 23] (Fig. 12.2). Overall, the most common clinical observation of AR reactivation involves direct modulation of AR levels. A significant fraction of CRPC specimens show elevated levels of both AR mRNA as well as protein compared to normal tissue [22, 23]. Chromosomal amplification of the AR locus has been detected in clinical samples, indicating that chromosomal instability can play a role in CRPC, but there are undoubtedly other mechanisms that exist to maintain enhanced AR levels [24, 25]. At present, surprisingly little is known about the molecular pathways that regulate AR expression or stability in normal or diseased tissue. Recently has it been demonstrated that the retinoblastoma tumor suppressor (RB) plays a key role in regulating AR expression through transcriptional mechanisms. In vitro, in vivo, and clinical models demonstrated that a significant fraction of CRPC samples fail to maintain a functional RB pathway, resulting in enhanced E2F1 transcriptional activity. Chromatin binding and functional analyses showed that the AR gene is a direct E2F1 target, whereby E2F1 binding to AR regulatory regions leads to increased transcriptional activity, AR mRNA induction, and elevated AR protein expression [26]. Importantly, xenograft studies modeling disease progression from hormone sensitive to CRPC showed that one of the major molecular changes observed in recurrent tumors was increased total AR levels, providing in vivo evidence that elevated AR is sufficient to promote CRPC. Additionally, clinical samples convincingly demonstrated that in patients with castration resistant metastatic disease, high levels of nuclear AR were associated with a greater risk of a short time to PCa related death [27]. Such evidence clearly implicates the role of enhanced AR levels in lethal cancer phenotypes.

282

R.S. Schrecengost et al.

Fig. 12.2 AR reactivation and progression to CRPC. PCa cells are maintained by AR activation, which can be monitored by PSA expression. Therapeutic treatment by testicular androgen depletion (GnRH) and AR antagonists target AR signaling, and result in tumor cell death or cell cycle arrest. Despite a short-term period of disease remission, PCa cells alter signaling that promotes castrate-resistant PCa (CRPC). AR reactivation represents the predominant means for recurrence, and is mediated by several mechanisms. These include AR deregulation (amplification or overexpression), cofactor alteration, gain-of-function mutations of AR, intracrine androgen synthesis, alternative splicing events that result in constitutively active receptors, and/or aberrant AR posttranslational modifications

In addition to elevation of AR levels, alterations in the AR protein itself have been attributed to CRPC-like behavior [4, 18]. A number of somatic point mutations have been observed in clinical specimens which alter/enhance the activity of AR. A large majority of these mutations have been mapped to the ligand binding domain, where key amino acid substitutions stabilize specific conformations that facilitate binding of non-canonical ligands, effectively converting other steroids and/or AR antagonists into agonists [6, 28, 29]. Among the most common LBD mutations is the T877A point-mutation (observed in human CRPC), which has the ability to not only utilize non-canonical ligands for activation, but can use hydroxyflutamide (flutamide) as an agonist [18, 30]. Although the frequency of these mutations is still

12

Androgen Receptor Regulation of Prostate Cancer Progression and Metastasis

283

a point of debate (estimates from 8% to 25% of metastatic tumors), they have potential to drive the transition to CRPC [18]. Mutations are not the only alteration to AR that can induce inappropriate receptor activation. Recent studies revealed the presence of AR transcript splice variants, which have bypassed the necessity for ligand-mediated transactivation potential. By utilizing exon-exclusion splicing events, the majority of these variants are truncated at the 3¢-terminus. This results in elimination of the LBD, which is crucial for holding AR inactive in the absence of ligand [31–33]. Consequently, these variants remain constitutively active, having maintained the elements necessary for their transcriptional regulatory function (N-terminal AF1 and DNA binding domain) [34]. Although variants retain the elements necessary for AR function, it is unclear as to whether the ability to dimerize with full length AR in vivo is affected. Additionally, based on their recent discovery, the consequence of the novel C-termini of these variants on AR target gene expression programs remain uncertain. Although AR splice variants maintain the ability to induce classical AR target genes (like KLK3/PSA), it is unknown what other potential oncogenic targets are induced by these proteins which may help to promote metastatic events. Additionally, as most current AR antagonists target the LBD of AR, the presence of these variants renders the cell refractory to the inhibitory growth effects of ADT. Thus, there is a critical need to elucidate the transcriptional network that is controlled by these variants in order to more effectively combat these clinically challenging tumors. In addition to direct alterations of AR, other major players in AR signaling axis (which are deregulated in disseminated disease) can act to promote aberrant activity during ADT [35, 36]. Specifically, AR cofactor alterations (consisting of a class of proteins that work in concert with AR to recruit essential transcriptional regulatory elements necessary for either repressing or enhancing transcription) have been shown to play a role in CRPC development. It is sensible to consider that either gain of coactivators, or loss of corepressors would help promote AR activity, and there is evidence to support both. Although there are many coactivators that have been implicated in regulating AR signaling, only a few have shown relevance to human disease. The most well defined AR coactivators, the steroid receptor coactivator (SRC) family members (1, 2, and 3), are known to be elevated in CRPC specimens, and function at the chromatin level to recruit/modulate chromatin modifiers and transcription initiation complexes [35–37]. When levels of these coactivators are elevated, cells become hypersensitive to low levels of active AR, increasing the basal level of AR signaling within the cell (contributing to CRPC development). Additionally, in vitro modeling suggests that coactivator overexpression is sufficient to convert AR antagonists into partial agonists, inducing AR activity [38]. Elevated levels of the known AR coactivator CBP have been shown to enhance AR activity in the presence of the AR antagonist flutamide, but not bicalutamide, indicating that such effects on AR signaling appear to be cofactor and drug specific (although more data is needed to confirm these effects clinically) [39]. Additionally, as coactivators are known to modulate the local chromatin environment, overexpression could induce AR mediated transcription of a unique gene subset. Although this notion is currently under investigation, it will be interesting to see if novel AR targets are

284

R.S. Schrecengost et al.

induced in CRPC and contribute to metastatic phenotypes in late stage disease. Inversely, corepressors function in opposition to that of coactivators and play key roles in the ability for AR antagonists to limit AR signaling [4]. Antagonist binding to AR induces a conformational change that favors the recruitment of corepressors rather than coactivators upon binding to DNA [40]. Downregulation or expulsion of corepressors like Nuclear Co-Repressor 1 (NCoR1) from DNA-bound AR can promote enhanced AR activity in the presence of antagonists or low level androgen [41]. Additionally, AR-cell cycle cross talk influences AR signaling though well established mechanisms involving cyclin dependent kinase (cdk)-independent functions of the cell cycle regulatory protein cyclin D1. After induction, cyclin D1 acts to inhibit AR signaling through recruitment of corepressor proteins like HDAC3 [15, 42]. Extensive studies in clinically relevant prostate cancer samples have shown that AR/cyclin D1 signaling axis is often disrupted through mislocalization, downregulation, and/or alternative splicing of cyclin D1, which has major implications for enhanced AR activity in late stage disease [15, 43]. Regardless of the mechanism, cofactor deregulation has important implications for not only restoring latent AR activity but also molding the chromatin environment to favor transcriptional programs that drive advanced disease. Perhaps the most inventive way CRPC tumors reactivate AR signaling axis is via autocrine androgen production. Multiple studies examining the enzymatic components necessary for androgen synthesis in PCa specimens have demonstrated that not only are all of the components for this process present, they are elevated in metastatic or recurrent tumors. Key enzymes responsible for the conversion of cholesterol to androgen precursors (CYP17A1 & HSDD3B2) were shown to be nearly tenfold higher in metastatic prostate cancer, while reduction enzymes, AKR1C3 and SRD5A1/2, responsible for the conversion of precursors to androgen, were elevated 8 and 9.2-fold respectively [44]. Such a robust increase in androgen production machinery likely increases local concentrations of available androgens during CRPC, reactivating the AR signaling cascade. Clinical trials are currently underway for molecular inhibitors of both AKR1C3 and SRD5A1/2 to try and combat this adaptive mechanism of ADT resistance. Irrespective of mechanism, CRPC clearly adapts means to sustain AR activity despite therapeutic intervention, and it appears that the transcriptional program controlled by AR could be altered in metastatic disease. Specifically, it appears there is a “CRPC specific AR program” which enriches AR occupancy at genes that drive phenotypes associated with late stage disease (i.e., mitosis and metastasis) [18]. For example, genome wide analysis of AR occupied chromatin regions in both androgen sensitive and CRPC cell lines uncovered CRPC specific AR occupancy enriched at a gene involved in mitotic progression (UBE2C). Clinical follow-up confirmed that UBE2C is upregulated in CRPC specimens, and is vital for mitotic progression in late stage disease [45, 46]. Unfortunately, the number of genes known to be regulated by this CRPC specific AR program are limited. As it has been well established that AR is essential for the metastatic progression of CRPC, there is currently a massive effort underway to uncover the identity of these AR driven

12

Androgen Receptor Regulation of Prostate Cancer Progression and Metastasis

Table 12.1 General characteristics of commonly used PCa cell lines Androgen Metastatic required for ADT potential: sub Q or Cell line AR status growth responsive prostatic injection LNCaP T877A Limited (Lymph node)

285

Notable features High AKT activity

LAPC-4

Wild Type

Limited (Lymph node, Lung)

Mutated p53

VCaP

Wild Type

N/A

Amplification of AR gene locus

C4–2

T877A

Limited (Bone)

22Rv-1

Splice Variant

PC3

Negative

DU145

Negative

Derivative of LNCaP line (CRPC) Absent CRPC Line (H874Y AR mutant) High (Lymph node, Truncated p53 Lung, Bone) High (Lymph node, P233L, V274F Lung, Liver) p53 mutation

All lines discussed above refer to parental cell lines isolated from patients with metastatic prostate cancer from different tissue types, with the exception of C4–2 (a derivative of the LNCaP line derived in culture) and 22RV-1 (a derivative of the cell line CWR22R from mouse xenografts). Of note, there are currently over 63 derived lines from the parent LNCaP line and over 20 from the parent PC3 line [47]

pathways, as they likely hold the answer to the molecular ontologies responsible for patient mortality. Studies investigating signaling pathways driving migration and metastasis of PCa have implicated both AR-dependent and AR-independent scenarios. Unfortunately, a major caveat among the field is that a large percentage of reports have utilized model systems that do not reflect clinical disease (Table 12.1). For example, AR-negative cells PC3 and DU145 are commonly the cell model of choice for assessing PCa migration/invasion signaling. While these cell lines are very aggressive and metastatic, they do not reflect the state of AR expression and activation (as mentioned above) as a hallmark of advanced disease [47]. An additional hindrance to understanding the impact of androgen signaling on PCa metastasis is the challenge of acquiring metastatic PCa specimens. The remainder of this chapter will focus on the established role of CRPC specific AR programs in metastatic progression with emphasis on: (1) the interplay between AR and chemokine receptors, (2) AR driven chromosomal fusions, (3) androgen controlled microRNAs and their downstream consequences, (4) oncogene activation and the impact on AR signaling, and (5) the current and future directions of CRPC directed therapeutics.

286

12.2

R.S. Schrecengost et al.

Part II: Interplay Between AR and Chemokine Receptors

Chemokine receptors belong to the large G-protein coupled receptor (GPCR) family of proteins. The chemokine system itself is a subclass of this superfamily with more than 20 currently known members. Each member maintains the same general structure of a seven-span transmembrane monomer and are subdivided into four groups based on the position of their first two cysteine residues (CXC, CC, C, and CX3C) [48, 49]. Chemokine receptors are found in a variety of cell types within normal tissues, and play significant roles in immune response, development, inflammation, and angiogenesis. Functional activation occurs through binding of soluble chemokine ligands, which initiates signal transduction pathways utilizing downstream mediators. As most of these processes are often disrupted during cancer development and progression, chemokine receptors have become attractive targets of study as mediators of metastatic phenotypes [49]. A specific member of the chemokine receptor family, CXCR4, is one of the best-characterized receptors due to well defined roles of this receptor in mediating metastasis of many different tumor types. Interestingly, a growing body of literature now suggests that CXCR4 is a crucial determinant of metastatic progression in many prostate cancers, and there is great interest in dissecting its role in promoting metastatic events as well as the mechanisms of its regulation.

12.2.1

CXCR4 Signaling

CXCR4 is a 352 amino acid GPCR that selectively binds a specific chemokine, stromal cell-derived factor 1 (SDF-1, also referred to as CXCL12) [49, 50]. Microarray and tissue analysis revealed that SDF-1 is expressed in a myriad of tissue and cell types (including bone), often acting as a chemoattractant for cells expressing CXCR4 at their surface [48, 51]. This migratory homing event involves complex intramolecular signaling events that are initiated by activation of the receptor and culminate with rearrangements in the actin cytoskeleton in addition to the induction of specific transcriptional programs associated with growth and survival [52–54]. Such cytoskeletal rearrangements generate the necessary force for the cell to translocate towards the source of the attractant (here SDF-1), localizing and adhering the cell to tissue specific sites. Although varying with cell type, CXCR4 activates this event via indirect activation of several kinases, which appear to be cell type specific [52, 55]. In vitro models of receptor activation indicate that CXCR4 can indirectly initiate the phosphorylation and activation of focal adhesion kinase (FAK), which can promote cytoskeletal rearrangements [56]. In parallel, the transcriptional programs initiated upon receptor activation act to ensure cell survival during the migratory event. Signaling cascades discovered by in vitro models of prostate cancer suggest that this process is mediated through MEK/ERK activation, leading to nuclear accumulation of the NF-kB transcription factor, and initiation of pro-growth and survival programs [54]. Unfortunately, there is still some debate as

12

Androgen Receptor Regulation of Prostate Cancer Progression and Metastasis

287

to the clinical relevance of these cascades in advanced prostate cancer, due to the lack of clinically available specimens for study. Irrespective of downstream signaling mediators, the biological consequence of CXCR4 expression in CRPC is appreciated, and has important implications for metastatic phenotypes observed in clinically relevant models of PCa metastasis.

12.2.2

CXCR4 in Prostate Cancer

The prostate gland is one of the few soft tissues within the body that express detectable levels of CXCR4 (although relatively low). CXCR4 is known to have important functions during organogenesis, but it is unclear if this represents a vital role in normal prostate function [57]. However, during prostate tumorigenesis, there is a clear selection for cells that induce expression of this receptor. Transcript analysis along with immunohistochemical data of benign, prostate adenocarcinoma, and metastatic PCa demonstrate a clear induction of CXCR4 mRNA and protein in both local and metastatic disease, with the highest protein levels occurring in bone metastases [51]. Importantly, no difference was detected between mRNA levels of precancerous lesions, local disease, and metastatic tissue, indicating that other regulatory mechanisms exist to maintain high CXCR4 expression in late disease [51]. Interestingly, tissue wide analysis of relative SDF-1 levels in both mice and humans revealed that preferable sites of PCa metastasis (bone and liver) showed higher levels of SDF-1 than did sites which rarely support PCa metastasis (eye and lung) [58]. Thus, it is reasonable to hypothesize that the induction of CXCR4 in metastatic disease acts to home cells to distant metastatic sites via its attraction to tissues which express large amounts of SDF-1. Extensive in vitro modeling of this event using clinically relevant models of PCa demonstrate that CXCR4 is elevated in many prostate cancer cell lines, and that inhibition of the receptor via pharmaceutical or nullifying antibodies inhibits migration by over 50% [59, 60]. There is currently conflicting data concerning whether or not induction of the CXCR4 pathway can promote cell growth in a prostate specific background. Different in vitro models have demonstrated that the effects of CXCR4 signaling on cell proliferation are cell type specific, and thus a more rigorous approach is needed to clarify the impact of CXCR4 on cellular growth [51, 59]. Taken together, such data imply that CXCR4 signaling plays an important role in PCa migration, and (as AR is essential for metastatic progression), interplay between the CXCR4 and the AR axis likely promotes lethal phenotypes.

12.2.3

AR and CXCR4 Interplay

The use of similar model systems that were used to demonstrate the role of CXCR4 in PCa has shown that PCa lines gain the ability to migrate toward a gradient generated by SDF-1, an event that was only possible in the presence of AR agonist R1881

288

R.S. Schrecengost et al.

(a non-hydrolyzable DHT analog). Moreover, similar migratory phenotypes are seen when cells are presented with a gradient generated only by R1881 [59]. Given the dependence on AR for CXCR4-mediated migration, research efforts have focused on the mechanisms through which these pathways could potentially overlap. Microarray evidence demonstrated that CXCR4 mRNA levels were enhanced in the presence of a functional AR signaling axis, and administration of ADT drastically inhibited this induction [61]. However, mapping of the CXCR4 gene locus showed no distinct regions for canonical AR binding, indicating AR regulation of CXCR4 was likely indirect [59]. This posit was further confirmed by experiments wherein activated AR showed no increase of CXCR4 transcript in the presence of protein synthesis inhibitors. These data imply that AR functions through intermediate pathways to regulate CXCR4. Analysis of potential transcription factors that are regulated by AR and promote migratory phenotypes in the prostate uncovered Krü ppel-like-factor 5 (KLF5) as a potential bridge between AR and CXCR4 [59]. Transcript analysis from cell lines demonstrated that KLF5 was quickly induced by AR signaling (within 1 h). Concordantly, overexpression of KLF5 was sufficient to induce CXCR4 expression and promoted cellular migration in the absence of AR signaling [59]. Based on these findings, a potential pathway emerges wherein AR reactivation may induce high levels of CXCR4 through the KLF5 mediator. It will be of significant interest to determine if the CRPC-specific AR program functions to enhance KLF5 expression in metastatic disease, thus accounting for the induction of CXCR4, and potentially facilitating enhanced disease progression. Additionally, it has been observed that increased migratory potential afforded by SDF1 (in the presence of R1881) was not as robust as the response to serum alone. Since in vitro models indicate that androgen signaling is crucial for migratory events, it is likely that other serum-derived factors act in concert with CXCR4 and/or AR to promote such metastatic events. Recent analysis of the effects of ADT on different tissues have demonstrated that AR signaling may also play a role in chemokine expression and cleavage in bone, potentially influencing PCa cell localization, adhesion, and extravazation, though the clinical implications for such regulation are still unclear [62]. A more rigorous approach aimed at deciphering clinically relevant pathways that are regulated by, or work in concert with, the AR signaling axis are needed to fully understand the molecular basis behind these events.

12.3

Part III: AR-Dependent Chromosomal Translocations

Biomarkers to stratify tumor subtype and stage of disease progression are at the current forefront of translational research. Recently, a fusion between transmembrane protease serine 2 (TMPRSS2), a well established androgen-regulated gene, and erythroblast transformation specific (ETS) transcription factor family members was discovered. Applying novel analysis of DNA microarray data, oncogene amplification/rearrangement of ERG and ETV1 (two ETS family members) was recognized [63]. This discovery led to the detection of a fusion product between the

12

Androgen Receptor Regulation of Prostate Cancer Progression and Metastasis

289

5¢ untranslated region of TMPRSS2 and the 3¢ exon of ERG, which occurred only under conditions of ERG gene overexpression. TMPRSS2 and ERG (the most commonly rearranged ETS member) are separated by approximately 3 megabases on chromosome 21. Rearrangement occurs either through interstitial deletion, which excises the region between the two genes [64, 65], or by chromosomal rearrangements between two independent chromosomes 21 to alter gene composition. The second mechanism of rearrangement is particularly relevant to AR-mediated processes, because acute DHT stimulation shortens the distance between TMPRSS2 and ERG in androgen-responsive cell lines, [66] and AR has been shown to be involved in chromosome proximity and double strand breaks [67]. The resultant protein product from TMPRSS2:ERG fusion is a truncated ERG protein due to the untranslated coding region of TMPRSS2.

12.3.1

TMPRSS2:ERG Fusions are Regulated by AR

As TMPRSS2 is a direct target of AR, the fused TMPRSS2:ERG gene was suspected to be directly regulated by AR. In support of this hypothesis, heightened AR activity (achieved via androgen stimulation) increased fusion product expression in non-malignant prostate epithelial cells, androgen-sensitive PCa cells, and tumor samples, [63, 68, 69] while castration of mice with TMPRSS2:ERG fusion-positive xenografts decreased fusion prevalence [70]. Given this dependence on AR signaling, it was initially proposed that TMPRSS2:ERG fusions might be directly involved in PCa progression. However, numerous in vitro and clinical studies analyzing the impact and incidence of fusion in PCa implicated TMPRSS2:ERG fusions in the promotion of invasion and invasive disease. In vitro analyses utilized VertebralCancer of the Prostate (VCaP) cells, which endogenously express the TMPRSS2:ERG fusion and are androgen-sensitive (Table 12.1). Modulation of fusion expression, either by overexpression of truncated ERG protein (DN-ERG) that is identical to the predicted TMPRSS2:ERG fusion product or by RNAi-mediated depletion have been instrumental for elucidating the biological consequence of fusion expression. For example, DN-ERG expression in VCaP and immortalized prostate epithelial cells, did not alter cell growth, tumor formation, or migration, but did increase invasion of cells in modified Boyden chamber assays [71, 72]. The mechanism of increased invasion was addressed through ectopic DN-ERG overexpression in combination with inhibitors of serine proteases and matrix metalloproteinases (MMPs). MMP inhibition did not alter invasive capacity, but inhibition of serine proteases with plasminogen activator inhibitor restored invasion to control conditions [72]. Another identified pathway of ERG-mediated invasion is through altered CXCR4 signaling. CXCR4 expression was influenced by androgen-mediated ERG expression, which promoted invasion of VCaP cells [73]. ERG overexpression corresponded with an upregulated gene signature profile that was similar to those associated with invasion gene networks [71]. Analysis of the gene signature following siRNA-mediated ERG depletion showed an increase in archetypical prostate

290

R.S. Schrecengost et al.

Fig. 12.3 TMPRSS2:ERG fusions and variants. TMPRSS2 and ERG fusions can occur by interstitial deletion or chromosomal rearrangement. (a) The transcript detected most frequently is a fusion between TMPRSS2 exon 1 and ERG exon 4 (TMPRSS2:ERGa or Type VI), which results in a truncated protein product (DN-ERG) [63]. (b) One alternative fusion is created between TMPRSS2 exon 2 and ERG exon 4 [71, 75]. This protein product incorporates 5 amino acids from TMPRSS2 into the amino terminus of ERG. (c) TMPRSS2:ERG fusions can also vary due to alternative splicing of transcripts. For example, ERG splicing can add 72 base pairs to the carboxy terminus of TMPRSS2:ERG fusion transcript (ERG + 72 bp). This transcript is associated with increased invasion and aggressive disease [76]. Light shaded boxes represent untranslatable coding regions, dashed-lines indicate a predicted protein product

epithelial markers, such as PSA, NKX3.1, and TMPRSS2 [71]. All of these data suggest that TMPRSS2:ERG may maintain or promote PCa into a de-differentiated state that could contribute to a migratory, mesenchymal cell phenotype. In addition to in vitro cell models, mouse models that utilized tissue regeneration with gene modulation have contributed to the current understanding of how AR regulates ERG-mediated PCa [74]. ERG overexpression alone was found to be insufficient for transformation of mouse prostate epithelial cells. However, when combined with enhanced AR signaling and AKT/PI3K pathway perturbation, invasive PCa was observed [74]. Similar to in vitro results, preclinical studies suggest that AR regulation of TMPRSS2:ERG fusion is not able to promote tumorigenesis of prostate epithelial cells but does cooperate with additional oncogenic signals to mediate invasiveness. ERG consists of 17 exons and generates nine different splice variants, seven of which form protein products. Most studies reporting TMPRSS2:ERG fusions have analyzed Type III fusion transcript (Fig. 12.3). However, since many splice variant products have alternative signaling and/or activity, such as is seen with the constitutively active AR splice variant, it is important to analyze additional fusions and resultant ERG products that are present in patient samples. While the vast majority of

12

Androgen Receptor Regulation of Prostate Cancer Progression and Metastasis

291

fusions occur between TMPRSS2 (exon 1) and ERG (exon 4), there is evidence that other variants can alter invasion and metastatic disease [75, 76]. TMPRSS2:ERG + 72 bp is a transcript containing an ERG splice variant that adds 72 base pairs from exon 11 to the carboxy terminus of the fusion [75, 76]. This fusion represents one detected alternative fusion that has been reported to promote cell invasion and is associated with aggressive disease. Additionally, fusion of ERG with exon 2 of TMPRSS2 results in a protein product with a unique N-terminus, which may contribute to altered biological response [71]. The detection of various fusion combinations and variants underscore the growing complexity as new information is constantly being learned about androgen-induced TMPRSS2:ERG fusions.

12.3.2

Clinical Implications of TMPRSS2:ERG Fusion

The discovery of TMPRSS2:ERG fusions has rapidly established that roughly 40–80% of PCa contain a variation of the fusion [77, 78]. TMPRSS2:ERG fusions also appear to be PCa-specific, as they have not been detected in any other commonly occurring tumor [79]. The ability of ERG to participate with non-ETS-family members has been detected, such as EWS-ERG and TLS-ERG in Ewing’s Sarcoma and acute megakaryoblastic leukemia [80, 81]. The use of clinical specimens has significantly aided the understanding that TMPRSS2:ERG fusions control PCa initiation and progression to metastatic CRPC. PCa is a multifocal cancer composed of multiple hyperplastic foci within an individual tumor mass. Typically, biopsies are taken from a single dominant focus. A study that analyzed multiple foci biopsies from a single tumor led to the discovery that fusion status is heterogeneous with an individual tumor [64]. Thus, if the biopsied focus is not representative of the entire tumor, the percentage of fusion-positive PCa could be hypothetically higher than currently reported. Although there are contradicting studies, many clinical studies report that TMPRSS2:ERG fusions correspond with an overall poor prognosis. For example, patients that presented with TMPRSS2:ERG-positive disease and who participated in a watchful waiting treatment regimen had more aggressive disease and a poorer prognosis, compared to fusion negative patients [82]. The analysis of metastatic tumor samples has given great insight into the relative timing and effect of fusion formation. One report indicated that 100% of evaluated metastases were created by the same process of interstitial deletion, and when these metastases were compared to matched primary samples, the metastasis and primary tumor contained TMPRSS2:ERG fusions generated by conserved means [64]. These observations suggest that the aggressiveness of a tumor could be impacted and/or predicted based on the process of TMPRSS2:ERG establishment in the primary tumor. Tumor samples of localized disease, hormonally naïve PCa, and CRPC have also demonstrated that the percentage of TMPRSS2:ERG fusions is equivalent, which suggests that selection is not occurring over the course of tumor progression to metastasis [64]. The hypothesis that one focus is seeding incurable, metastatic cells is supported by clinical data reporting conservation of TMPRSS2:ERG fusion incidence between multiple sites and detection of homogeneous

292

R.S. Schrecengost et al.

fusion products. Detecting and removing the fusion-positive foci may represent a therapeutic approach for inhibiting metastatic PCa and reducing poor prognosis. One recent approach has involved a non-invasive detection of fusion products in urine of patients following digital rectal exam to screen for at risk patients [83–85]. However, contrary to data from tumor samples, multiple fusion transcripts were detected in the urine from a single patient [85]. TMPRSS2:ERG fusions have gained significant attention because of the high incidence in both primary and metastatic tumors. However, additional study is needed to rigorously determine if the fusion is a consequence of PCa progression or promotes metastatic dissemination.

12.4

Part IV: Androgen-Regulated miRNAS

The discovery of micro RNAs (miRNAs) has added significant complexity to understanding gene regulation, expression, and function. miRNAs are endogenous noncoding RNAs that alter gene expression by binding to complementary sequences within target mRNAs, which results in disruption of gene translation and RNA degradation if sequence complementarity is high [86]. These 18–25 nucleotide RNA fragments can be transcribed from an independent promoter or, as is most often the case, are found in gene introns and are co-transcribed, then spliced to appropriate lengths [86]. A single miRNA can control anywhere from a few genes to hundreds of genes, making miRNAs an intricate and diverse mechanism of gene regulation that influences all biological processes [79, 87]. Although miRNAs regulate gene translation within the context of physiologic processes, they can also impact cancer suppression and promotion [87]. Large-scale analysis of miRNA detected a general decrease in miRNA expression in solid tumors, when compared to normal tissue [88]. Conversely, a miRNA signature for solid tumors has also been reported that contained 36 upregulated and six repressed miRNAs. Many miRNAs from this signature were found to be located in cancer-related chromosomal regions, which increases the likelihood that they are altered in cancer [89, 90].

12.4.1

Androgen-Regulated miRNAS in Prostate Cancer

Similar to other solid tumors, there is an ever-growing list of miRNAs deregulated in PCa that influence epigenetic alterations, cell cycle and apoptosis, castration resistant growth, and migration/metastasis [86]. In murine and in vitro models, reduction of androgen levels reduced the level of a large group of miRNAs in prostate tissue and androgen-responsive LNCaP cells, which was reversed by DHT treatment [91]. Furthermore, greater than 50 miRNAs are deregulated in PCa tumors, compared to normal tissue. Many are not regulated by AR activity, but in response to androgen stimulation, at least 18 different miRNAs are upregulated in androgen-sensitive cell lines. As expected, many of these participate in cell cycle

12

Androgen Receptor Regulation of Prostate Cancer Progression and Metastasis

293

regulation and a few are implicated in cell migration and invasive disease [92–94]. Comparison of cell lines established from advanced PCa (Gleason >8) and lowgrade disease (Gleason <5) revealed that a set of miRNAs may be enriched in aggressive tumors. These miRNAs include miR-222, miR-221, miR-331–3p, miR-16, miR-145, miR-9*, and miR-551a. Therefore, it is possible that PCa initiation and progression to advanced disease is mediated by androgen-regulated miRNAs. miR-101 is a miRNA that is lost during disease progression of gastric and prostate cancer [95, 96]. In vitro measurement indicated an increase in miR-101 in LNCaP cells, compared to prostate epithelial and AR-negative PCa cell lines [97]. Ectopic expression of miR-101 in androgen-sensitive cells reduced migratory capacity [97]. In these studies migration was influenced by EZH2, a Polycomb-group (PcG) member involved in applying repressive trimethylation marks on histones, which was found to be negatively regulated by miR-101 [95, 97]. EZH2 expression is increased in many cancer types and is overexpressed in the majority of metastatic PCa [95, 98]. A similar connection between miR-101 and EZH2 was identified in gastric cancer [96], thus suggesting that either miR-101 can act in a cell-type specific manner or that the AR-mediated regulation and EZH2 control are independent. Based on current data, miR-101 appears to be acting as a tumor suppressor by directly inhibiting EZH2, which is a differentiation and metastasis factor. One of the most studied and commonly upregulated miRNA is the ‘oncogenic’ miR-21. miR-21 is found to be upregulated in non-solid tumors [99, 100] as well as numerous solid tumor types, such as lung, breast, colon, and prostate [90, 101]. Within the context of PCa, androgen stimulation elevated miR-21 expression in multiple AR-positive androgen-sensitive and CRPC cell lines following. mRNA analysis of human PCa found miR-21 significantly upregulated in 60% of tumors [93]. miR-21 upregulation promoted anchorage independent growth, increased tumor growth, and led to castration resistance in xenograft mouse models [93]. In AR-negative cell lines, repression of miR-21 inhibited migration and invasion [102]. The results from in vitro models suggest that it is possible for miR-21 to regulate critical steps of metastatic progression in androgen-sensitive PCa or CRPC. Several miRNAs regulated by androgens are associated with the progression of aggressive CRPC, but there is not direct evidence for a role in cell motility. miR-125b is elevated in androgen-sensitive cell lines, compared to benign prostate epithelial cells and AR-negative cells, and is a direct AR target [94]. miR146a is downregulated in AR-negative cell lines, compared to androgen sensitive models [103]. These studies indicate that there are many factors not directly regulated by AR that contribute to the aggressive and incurable stage of metastatic PCa. In addition to the androgen-regulated miRNAs mentioned above, there are also many miRNAs that may play significant roles in PCa metastasis that are not regulated by AR. For example, comparison of cell lines from aggressive PCa (Gleason >8) with indolent disease (Gleason <5) revealed that a set of miRNAs are implicated in aggressive PCa. These miRNAs include miR-222, miR-221, miR-331–3p, miR-16, miR-145, miR-9*, and miR-551a [104]. A major barrier toward increasing our understanding of how AR influences miRNAs in metastatic progression is the assessment of miRNAs from

294

R.S. Schrecengost et al.

metastatic CRPC samples. Once obtained, current in vitro models of AR regulated miRNAs that influence migration and invasion of PCa can be expanded to clinical specimens, helping to illuminate the impact miRNAs have on metastatic progression and potential treatment options.

12.5 12.5.1

Part V: AR, Oncogenes, and Metastasis c-MYC (MYC) in PCA

Oncogenic transformation of prostate cancer cells is disparate from other solid tumors because there are only a limited number of bona fide oncogenes implicated in PCa. The proto-oncogene Myc and AR are tightly regulated, and each impact one another in regards to migration and invasion. Myc is a transcription factor that regulates expression of genes regulating many biological processes. Upon heterodimerization with Max, Myc binds to Enhancer Box (E-box) sequences, recruits histone acetyltransferases (HATs), and activates gene transcription. However, Myc is a proto-oncogene upregulated in most solid cancers, which is most often due to amplification of chromosome 8q24, the MYC gene-containing region. In normal cells, Myc expression is tightly controlled as increased expression drives transcription of genes that promote growth and proliferation. If Myc becomes highly expressed, cells can senesce or undergo apoptosis. Amplification of MYC is detected in roughly 30% of PCa patient tumors but there remains some discrepancy as to the whether Myc is overexpressed or amplified in primary tumors [105, 106]. Regardless, Myc is one of the most implicated genetic aberrations in aggressive and metastatic PCa, as Myc amplification is detected in 50% of clinically presented metastatic tumors [107]. Overall, the presence of amplification correlates with high tumor grade and poor prognosis [108, 109]. One of the most useful tools for understanding how Myc function regulates PCa in the context of AR-positive disease has been the generation of transgenic Myc mice. Transgenic mice were designed for Myc to be expressed under the control of probasin promoter (a murine AR target gene), which results in low level, prostatespecific Myc upregulation (Lo-Myc) dependent on androgen levels. Additionally, to further modulate Myc levels, expression was driven by a modified probasin promoter, ARR2Pb, that results in a high level of androgen-dependent Myc expression (Hi-Myc). More recently, Super-Lo-Myc mice were developed, placing Myc expression under control of the Nkx3.1 promoter, which resulted in expression levels below that of the Lo-Myc model [110]. All three models were found to develop prostatic intraepithelial neoplasia (PIN; precancerous) lesions spontaneously, and all except Super-Lo-Myc model progressed to invasive disease [111]. Overall, Myc-driven murine PIN and invasive prostate adenocarcinoma were similar to human tumors in morphology, invasiveness, and gene-expression profile. These findings are important because there are currently no other murine models of PCa that maintain AR expression and spontaneously form PIN lesions that could

12

Androgen Receptor Regulation of Prostate Cancer Progression and Metastasis

295

progress to invasive disease. Transgenic mice also supplied Myc-Cap cells, which are mouse cells generated from a Hi-Myc mouse tumor. These cells were found to have AR amplification despite never being exposed to ADT [112]. This observation was noteworthy, because most AR amplifications are detected in human tumors only after ADT [113], and because it suggests that Myc alone can regulate AR expression. Myc overexpression in vitro was found to bypass the need for androgens with respect to anchorage-independent growth; however, this did not translate to in vivo studies as xenografts tumors with Myc overexpression did not grow under castrated conditions. There have been relatively few in vitro studies that have directly assessed the relationship between AR and Myc. Those that have, however, clearly emphasize the interwoven relationship between these two transcriptional regulators. Stimulation with the synthetic androgen, R1881, increased c-Myc mRNA expression, which could be inhibited with bicalutamide [114]. In an ex vivo experiment, DHT stimulation increased Myc mRNA expression in prostatic tissue [115]. AR can also activate Myc indirectly by regulating TMPRSS2:ERG fusion expression, which has been found to control Myc activity [74, 116]. Conversely, Myc was shown to regulate AR expression through the binding of a consensus E-box sequence in the AR coding region [117]. One explanation of how AR and Myc regulate PCa migration is through the signaling molecule Ezrin, which links actin filaments with the plasma membrane and is a regulator of metastasis of several cancer types [118, 119]. Within the context of androgen therapy-sensitive cell lines, androgen-mediated activation of AR increased Ezrin mRNA and protein expression, which was reversed by bicalutamide [120]. Ezrin overexpression was capable of increasing cell migration, while siRNA-mediated depletion inhibited DHT-stimulated migration. In tumor samples, based on immunohistochemistry (IHC) Ezrin expression decreased following ADT (cyproterone acetate). Subsequently, Ezrin was found to be a direct target of Myc, which regulated AR-mediated migration [114]. Therefore, AR activation is capable of regulating Myc expression and/or activity to elevate Ezrin expression, which results in increased migration. Thus, in vitro and mouse model studies have established that Myc and AR are key regulators of PCa growth and invasiveness. Numerous studies assessed the relationship of Myc expression to localized or metastatic disease. Chromosome 8q24 amplification, which encodes the MYC gene, has been reported amplified in up to 50% of PIN and localized tumors by FISH, but amplified in 73–92% in matched metastatic PCa tumors [108, 121]. Circulating tumor cells (CTCs) were also used for cytogenetic analysis of amplified genes from CRPC tumors. CTCs (metastatic cells shed into circulation) represent an FDA approved source of tissue for prognostic clinical biomarker [122]. FISH analysis on CTCs collected from men with CRPC detected Myc and AR amplified in a large percentage of specimens. The number of CTCs/ml of blood can be correlated with disease burden and aggressiveness. When the CTC count was >10/ml, Myc was amplified 63% of the time, compared to 30% when CTC detection was <10/ml [107]. These data collected from human tissue demonstrate that Myc amplification is enriched in metastatic disease. Although it is not clear if Myc amplification causes CRPC or is selected for during ADT, one study concluded that Myc amplification

296

R.S. Schrecengost et al.

was observed as a consequence of endocrine therapy [123]. In addition to expression levels, intracellular localization of Myc may be a prognostic indicator. Under conditions of elevated Myc expression, nuclear Myc protein was upregulated in PIN and prostate adenocarcinoma in the absence of amplification [124]. Additionally, high Myc expression in primary PCa predicted recurrence and aggressive disease [125]. These data suggest that mechanisms apart from gene amplification can influence PCa progression. Although there is a large amount of research implicating Myc in poor prognosis and a metastatic phenotype [126], this pathway cannot currently be targeted pharmacologically. Myc amplification and/or expression status may be most useful for stratifying patients for optimizing therapeutic outcome.

12.5.2

c-SRC in PCa

AR-mediated migration of PCa can also be influenced by transduction of signals mediated by c-Src. c-Src, the prototypical member of Src-family kinases (SFK), was initially discovered as the human homolog of the oncogene v-src, which was found to be the transforming agent of Rous sarcoma [127]. c-Src has been found to control several biological processes, including proliferation, angiogenesis, migration, and metastasis in numerous model systems [128]. Although overexpression of c-Src in human cell lines is insufficient for cellular transformation, it is deregulated in several cancers, including lung, colon, breast, and prostate. One predominate function of c-Src in PCa is to transduce extracellular stimuli from growth factor receptors and integrins to AR signaling axis for promotion of growth and proliferation. For example, c-Src phosphorylates AR on tyrosine 534 and initiates nuclear translocation, DNA binding, and increases activation of target genes in the absence of androgens. Inhibition of c-Src attenuates AR phosphorylation and xenograft tumor growth under CRPC conditions [129, 130]. There are a limited number of reports evaluating the role of AR and c-Src within the context of androgen stimulation and proliferation. This remains to be an area with many questions remaining. The following discussion focuses on research that demonstrates how c-Src impinges on AR signaling to promote migration and invasive PCa. c-Src plays an established role in regulating intracellular signaling critical for cell migration. While many studies have focused on how c-Src regulates integrins, nonreceptor tyrosine kinases, and Rho-family GTPases, there is literature to implicate c-Src in controlling AR-dependent migration in PCa cells and driving aggressive, castration-resistant disease. Biochemical inhibition of c-Src inhibited invasion of both androgen-sensitive PCa cells [120] and CRPC cells [131]. Inhibition was also coupled with decreased AR recruitment to the AR target gene PSA. Conversely, c-Src upregulation was detected in C4–2 CRPC cells, in comparison with prostate epithelial cells [131]. These findings were recapitulated in vivo using the transgenic adenocarcinoma of the mouse prostate (TRAMP) model, which spontaneously develops PIN and can progress to CRPC [132]. Using this tool, it was demonstrated that disease progression was associated with elevated c-Src activity, based on monitoring

12

Androgen Receptor Regulation of Prostate Cancer Progression and Metastasis

297

tyrosine 418 phosphorylation. However, this study did not characterize AR expression or nuclear localization in the presence or absence of endogenous androgens. An additional caveat from the majority of studies is that overall c-Src activity is evaluated rather than deciphering the specific mechanism and target of c-Src activity. It is enticing to speculate that c-Src inhibitors may not only suppress migratory phenotypes, but may alter AR phosphorylation state, recruitment of cofactors, or protein stability. The in vitro findings have also been tested using human tumor samples. Heightened SFK activity was detected in CRPC tumors, compared to matched prostate tumors taken prior to ADT, and this was associated with significantly decreased survival [133]. Due to in vitro and preclinical data implicating a role for c-Src in prostate cancer progression and aggressiveness, the efficacy of c-Src inhibitors are currently being explored in clinical trials [128]. Dasatinib, a broad non-receptor tyrosine kinase inhibitor, is currently approved for use in Chronic myelogenous leukemia (CML) (second line therapy) and Philadelphia chromosome-positive acute lymphoblastic leukemia [134]. Orthotopic prostate tumors in nude mice that were treated with Dasatinib showed a decreased tumor volume and a reduction in the number of lymph node metastases [135]. In a phase II trial that evaluated Dasatinib for metastatic CRPC in chemotherapy-naïve, 43% achieved lack of progress after 3 months. Dasatinib was also found to have biological activity in the bone of patients, 81% of whom had bone metastases. Currently, Dasatinib is being evaluated in additional phase II trials (NCT00439270, NCT00385580, NCT00570700) and one phase III clinical trial for CRPC patients both as a monotherapy and in combination with docetaxel (NCT00744497). c-Src is one of the few targetable molecules, other than AR, that is being evaluated clinically. Based on the phase II trial results, Dasatinib could theoretically be more effective when used in combination with AR antagonists or used on a yet to be determined subclass of PCa patients.

12.6

Part VI: Targeting AR: Putative New Avenues for Therapeutic Intervention

PCa treatment regimens vary based on the invasiveness of disease. Organ-confined PCa is effectively treated by surgical resection or radiation treatment. Unfortunately, non-organ-confined tumors are not as easily eradicated and present a clinical challenge. Due in part to the pathways described in this chapter, treatments for invasive PCa are focused on targeting AR activity. As summarized herein, the standardof-care for invasive PCa is ADT, which involves treatment with GnRH agonists and bicalutamide (Casodex). GnRH agonists suppress AR activity through the disruption of testicular androgen synthesis. However, due to locally produced androgens at the tumor site, significant levels of androgens are ultimately available to activate AR signaling. For this reason, combined treatment with AR antagonists is necessary. Antagonists, such as flutamide and bicalutamide, function by binding to the ligand binding pocket of AR and blocking DHT binding. Although AR still localizes to the nucleus, antagonists recruit repressive complexes to suppress AR-mediated gene

298

R.S. Schrecengost et al.

Fig. 12.4 AR-directed therapies for metastatic PCa. Management of metastatic PCa is dependent on controlling AR activation and suppressing aberrant reactivation. Current standard of care involves treatment with AR antagonists (bicalutamide or flutamide). Other therapeutic approaches are currently being assessed in preclinical and clinical trials. Inhibition of CYP17A1 with abiraterone or TOK001 blocks synthesis of testosterone, thereby limiting AR ligand. TOK001 can also influence AR protein stability. EPI-001 is a small peptide that uniquely targets the amino terminus of AR and inhibits activity, which could be integral for CRPC tumors that often have AR carboxy terminus mutations that render antagonists ineffective. MDV3100 is a pure AR antagonist that works similarly to bicalutamide but binds with greater affinity, blocks AR nuclear translocation, and has no agonist functions

transcription. For several decades, combined androgen blockade has been the standard-of-care for treating locally invasive and metastatic PCa [136] (Fig. 12.4). Despite the initial effectiveness of endocrine therapy, a large percentage of patients will have tumor recurrence (CRPC) in 2–3 years. Remission is mediated by re-activation of AR signaling axis (Fig. 12.2) [18], however the only approved therapy for CRPC is docetaxel, which confers limited survival benefit. Therefore, the development of additional therapeutic options to directly modulate AR activity represents a current need in the clinical management of PCa. There are several drugs undergoing pre-clinical and clinical screening that target AR stability, activity, and/or nuclear localization that present very promising options for the future of PCa treatments.

12

Androgen Receptor Regulation of Prostate Cancer Progression and Metastasis

12.6.1

299

Indirect AR Targeting

As detailed in Part I, intracrine androgen synthesis occurs in PCa, resulting in aberrant AR activity under ADT conditions. Therefore, targeting the upregulated steroid metabolism machinery could serve clinical benefit. Abiraterone acetate (abiraterone) irreversibly binds CYP17A1, an enzyme containing C17, 20-lyase activity that is critical for production of testosterone precursors, which prevents the formation of androgens [137]. While there are other compounds capable of inhibiting this process, such as ketoconazole, abiraterone has higher specificity. Previous studies confirmed that abiraterone was effective at inhibiting androgen biosynthesis in vitro and in vivo [138], and this led to numerous clinical trials evaluating the viability of abiraterone as a clinical agent (more than 15 active phase I, II, and III trials at present). A completed phase I dose-escalation trial with CRPC patients that had failed previous therapies found that abiraterone rapidly decreased serum testosterone and additional precursor levels [139]. Unexpectedly, more than 50% of patients exhibited decreased PSA expression that lasted more than 3 months. Two phase II trials have also confirmed these findings. These trials enrolled chemotherapy-naïve patients or those previously treated with chemotherapy [140]. The findings indicated that a biochemical response was attainable in patients previously treated with docetaxel. There are two active phase III trials currently ongoing further assessing abiraterone. These two trials are double blind, placebo-controlled treatments of abiraterone plus prednisone in asymptomatic or mildly symptomatic metastatic PCa patients. The difference between trials is based on patient population. One trial enrolls patients previously treated with only AR antagonists and not chemotherapy or ketoconazole while the other requires one or two prior chemotherapies. These advanced trials will determine the ideal patient population for abiraterone treatment and help shape the understanding of whether patients previously treated with chemotherapies are amenable to hormonal manipulation.

12.6.2

Direct AR Targeting

The darylthiohydantoin compound MDV3100 is a pure AR antagonist that inhibits AR activity and AR-mediated tumorigenesis more effectively than bicalutamide [141]. MDV3100 acts by binding to the C-terminal end of AR (LBD) and blocking androgen binding. Although the method of AR inhibition is similar to bicalutamide, there are important differences that distinguish the two. Many of the experiments that initially characterized MDV3100 in comparison with bicalutamide were performed in LNCaP/AR cells that ectopically express higher amounts of wild-type AR. This aided in addressing the effectiveness of new compounds in a clinically relevant model. MDV3100 was determined to bind AR with at least fivefold higher affinity than bicalutamide and effectively inhibited AR-dependent gene expression. In fact, bicalutamide has been found to act as an agonist when AR is overexpressed

300

R.S. Schrecengost et al.

[142], which was recapitulated in this model. Another major difference between these AR-directed therapies was that MDV3100 blocked nuclear localization of AR after binding, which impeded the binding of AR to DNA [141]. In vivo studies also determined that MDV3100 caused reduced proliferation and regression of tumor xenografts, in comparison to bicalutamide-treated tumors that only had a slower growth rate and no regression. These preclinical studies resulted in a phase I/II clinical trial of men with CRPC who had progressed with first-line AR antagonists. In this study, MDV3100 was found to have antitumor activity that was independent of any previous hormonal or chemotherapeutic treatments [143]. Approximately 43% of patients with CRPC experienced a significant reduction in PSA, indicating that the agent successfully suppressed recurrent AR activity. Currently, a phase III trial is underway to determine the efficacy of MDV3100, compared to placebo, for men with metastatic CRPC that have either previously been treated with docetaxel or are chemotherapy-naïve (NCT01212991). Based on several differences in mechanism of action between bicalutamide and MDV3100, MDV3100 offers a potentially useful alternative for ADT that could eliminate or delay tumor recurrence. While MDV3100 has shown clinical effectiveness, there are additional drugs targeting AR that have exciting preclinical data to suggest they may be beneficial for therapy. TOK001, for example, is an analog of abiraterone acetate that was found to inhibit both CYP17 and AR directly. Cell models with both wild-type and mutant AR were found to have reduced AR activity and protein expression following TOK001 treatment [144, 145]. If the in vitro findings hold true in the clinic, use of this agent could result in dual inhibition of intracrine androgen synthesis and AR expression. At present, it is unknown how TOK001 functions to deplete AR protein levels in vivo [61]. Additionally, TOK001 has anti-proliferative activity in AR-negative cells due to an endoplasmic reticulum stress response, which may raise questions regarding off-target effects and toxicity. To determine toxicity, dosing, and efficacy, the ARMOR1 phase I clinical trial evaluating TOK001 in men with CRPC who are chemotherapy-naïve is currently enrolling patients (ARMOR1 NCT00959959). AR antagonists share the similar feature of binding to the LBD on AR and altering AR activity. However, the recent discovery of AR splice variants, which are constitutively active due to lack of LBD, in CRPC emphasize the importance of developing agents to target the N-terminal transactivation domain. One novel molecule recently tested is the small molecule EPI-001, which binds to AR N-terminal transactivation domain, thus inhibiting critical intramolecular interactions supported by this region, and suppressing AR association with chromatin-bound regulatory elements [146]. Biological outcomes included inhibition of in vitro androgen-induced proliferation and regression of in vivo androgen dependent and CPRC xenografts without host toxicity. These findings were the first to demonstrate feasibility for in vivo suppression of AR via targeting of N-terminal domain functions. While EPI001 has not been used in a clinical trial, the preclinical data represent a potential therapy that addresses a need for CRPC patients. The therapies discussed here are similar in that they work to inhibit AR activity. Given that atypical AR activation is paramount for driving the formation of

12

Androgen Receptor Regulation of Prostate Cancer Progression and Metastasis

301

metastatic CRPC, these treatment options are viable options for inhibiting PCa metastasis. However, most studies focus on time to recurrence and disease stabilization, and are not able to directly assess the biochemical impact of AR inhibition on metastasis. Elucidating AR-mediated migration and metastasis, and acquiring metastatic tumor samples, will enable a greater understanding of how AR inhibition can directly inhibit the incurable metastatic PCa.

12.7

Part VII: Conclusion and Perspectives

The abundance of in vivo and clinical data has clearly demonstrated how crucial a functional AR signaling axis is in inducing pathways necessary for metastatic progression. However, the current lack of relevant in vitro and in vivo model systems to dissect these pathways has been one of the greatest barriers towards significant pre-clinical progress. The majority of advanced PCa models prove extremely aggressive and metastatic [47], yet they lack a functional AR signaling pathway, which is weakly represented in advanced PCa. Interestingly, attempts at restoring AR activity in such systems to mimic clinical samples has demonstrated that AR function is altered in these systems. For example, rather than promoting cell growth, AR re-expression in PC3 cells is inhibitory [147]. Thus, pathways that regulate metastatic properties in these models are likely independent of the AR signaling axis. Consequently, it is difficult to predict how, if at all, findings in these models can be used to elucidate the true mechanisms behind CRPC mediated metastatic events. Additionally, such problems are compounded by the lack of available metastatic patient tissues currently available for research purposes. As a result, corroborating in vitro and in vivo findings in clinical samples is extremely difficult and greatly hinders research progress. Therefore the successful development of relevant models of metastatic prostate cancer is absolutely essential for the advancement of PCa research. Once accomplished, several key questions necessary for understanding advanced disease can be addressed. First, are the events that underlie progression to CRPC mutually exclusive, or do combinations of these adaptations exist in clinical tumors? As AR is the key driver of PCa at all disease stages, it is likely that recurrent tumors evolve multiple mechanisms to enhance and maintain AR activity. If true, could such combinations be responsible not only for tumor survival and growth in the presence of ADT but for the CRPC specific AR programs seen in metastatic disease? It would be of potential benefit to determine if specific adaptations (either individually or in combination) drive distinct genetic metastatic programs, and if profiling tumors for these signatures could be useful for specific therapeutic intervention. For example, if a tumor profile consists of both AR splice variants and AR mutations, traditional ADT therapy would likely prove inefficient. Instead, a more efficacious regimen might include drugs that inhibit AR activity independent of its LBD (like EPI001). Advances along this line of thinking have lead to the development of drugs

302

R.S. Schrecengost et al.

such as (3-hydroxy-17-(1 H-benzimidiazole-1-yl)androsta-5,16-diene) which not are not only capable of inhibiting local androgen production, but are potent antiandrogens as well [145]. Such profiling strategies have proven effective for other hormone driven cancers (like breast cancer), and insights into the effectiveness of tumor molecular profiling might prove to have diagnostic and therapeutic value in treating metastatic disease. Although the importance of AR signaling in PCa is increasingly well understood, less is known about transcriptional and post-transcriptional control of AR. Since all currently utilized therapeutics are directed at a single distinct domain within the receptor itself, and few have shown long term clinical efficacy, targeting AR from a different perspective might prove to be an effective means of limiting AR activity in advanced disease. Unfortunately, research into such avenues has not been extensive, and thus, agents that regulate AR transcription and stability are limited. The majority of therapeutics in development for CRPC are aimed at proteins that contain ligand binding or catalytic domains (e.g., AR or c-Src, respectively), which provide a pocket for small molecules to bind. Basic research identifying novel factors that regulate AR posttranslational modifications represent potential targets that could be exploited to disrupt AR protein stability. Likewise, additional pharmacological targets could be uncovered through assessment of factors that regulate AR and AR-target gene transcription. One example could be the E3 ubiquitin ligase RNF6, which has been demonstrated to ubiquitinate AR and promote transcriptional activity [148]. Interestingly, RNF6 is required for growth in androgen-depleted conditions and is overexpressed in hormone refractory PCa. A growing body of literature suggests that modifications such as this could contribute to CRPC specific AR targets, and inhibiting such proteins could represent a novel approach to indirectly regulate AR activity; potentially repressing CRPC specific AR transcriptional programs that contribute to metastatic disease. TMPRSS2:ERG fusions represent a burgeoning research field that has been studied in in vitro, pre-clinical, and clinical settings. As discussed in Part III, there are several studies that have linked TMPRSS2:ERG fusions with invasion. These studies were performed with rapid autopsy-accumulated metastatic samples. However, the surprisingly high incidence of fusion positive primary and metastatic tumors raises questions about the possible cooperating factors that may influence the proinvasive effects of the fusion proteins. Does the TMPRSS2:ERG fusion indicate a new era of PCa biomarker discovery? If androgen-mediated TMPRSS2:ERG fusions drive invasiveness and metastases, how will therapeutic options be tailored differently than the current regimen? Will fusion positive tumors be divided into a subgroup for treatment? Future studies will undoubtedly assess the clinical implication(s) of the fusions on the metastatic phenotype. In summary, it is clear that the AR signaling axis is crucial for both the development and progression of early and late stages of prostate cancer. Surprisingly, current knowledge of the specific roles that this pathway plays in promoting lethal metastatic phenotypes is extremely limited. Extensive research into the mechanisms behind CRPC specific AR programs involved in metastasis will likely illuminate novel pathways responsible for patient morbidity, and promote the development of targeted therapeutics that enhance patient survival.

12

Androgen Receptor Regulation of Prostate Cancer Progression and Metastasis

303

Acknowledgements We would like to thank all the members of the K. Knudsen lab, especially M. Schiewer, for insightful feedback and critical reading of this chapter.

References 1. Jemal A et al (2008) Cancer statistics, 2008. CA Cancer J Clin 58(2):71–96 2. Klein EA et al (2009) Outcomes for intermediate risk prostate cancer: are there advantages for surgery, external radiation, or brachytherapy? Urol Oncol 27(1):67–71 3. Klotz L (2006) Combined androgen blockade: an update. Urol Clin North Am 33(2): 161–166, v–vi 4. Knudsen KE, Scher HI (2009) Starving the addiction: new opportunities for durable suppression of AR signaling in prostate cancer. Clin Cancer Res 15(15):4792–4798 5. Loblaw DA et al (2007) Initial hormonal management of androgen-sensitive metastatic, recurrent, or progressive prostate cancer: 2006 update of an American Society of Clinical Oncology practice guideline. J Clin Oncol 25(12):1596–1605 6. Yuan X, Balk SP (2009) Mechanisms mediating androgen receptor reactivation after castration. Urol Oncol 27(1):36–41 7. Lawton CA et al (2001) Updated results of the phase III Radiation Therapy Oncology Group (RTOG) trial 85–31 evaluating the potential benefit of androgen suppression following standard radiation therapy for unfavorable prognosis carcinoma of the prostate. Int J Radiat Oncol Biol Phys 49(4):937–946 8. Bolla M et al (2002) Long-term results with immediate androgen suppression and external irradiation in patients with locally advanced prostate cancer (an EORTC study): a phase III randomised trial. Lancet 360(9327):103–106 9. Bubendorf L et al (2000) Metastatic patterns of prostate cancer: an autopsy study of 1,589 patients. Hum Pathol 31(5):578–583 10. Cunha GR et al (2004) Hormonal, cellular, and molecular regulation of normal and neoplastic prostatic development. J Steroid Biochem Mol Biol 92(4):221–236 11. Penning TM et al (2008) Pre-receptor regulation of the androgen receptor. Mol Cell Endocrinol 281(1–2):1–8 12. Centenera MM et al (2008) The contribution of different androgen receptor domains to receptor dimerization and signaling. Mol Endocrinol 22(11):2373–2382 13. Chmelar R et al (2007) Androgen receptor coregulators and their involvement in the development and progression of prostate cancer. Int J Cancer 120(4):719–733 14. Agoulnik IU, Weigel NL (2008) Androgen receptor coactivators and prostate cancer. Adv Exp Med Biol 617:245–255 15. Balk SP, Knudsen KE (2008) AR, the cell cycle, and prostate cancer. Nucl Recept Signal 6:e001 16. Beekman KW, Hussain M (2008) Hormonal approaches in prostate cancer: application in the contemporary prostate cancer patient. Urol Oncol 26(4):415–419 17. Ryan CJ et al (2006) Persistent prostate-specific antigen expression after neoadjuvant androgen depletion: an early predictor of relapse or incomplete androgen suppression. Urology 68(4):834–839 18. Knudsen KE, Penning TM (2010) Partners in crime: deregulation of AR activity and androgen synthesis in prostate cancer. Trends Endocrinol Metab 21(5):315–324 19. Chen Y, Sawyers CL, Scher HI (2008) Targeting the androgen receptor pathway in prostate cancer. Curr Opin Pharmacol 8(4):440–448 20. van Poppel H, Nilsson S (2008) Testosterone surge: rationale for gonadotropin-releasing hormone blockers? Urology 71(6):1001–1006 21. Oefelein MG (1998) Time to normalization of serum testosterone after 3-month luteinizing hormone-releasing hormone agonist administered in the neoadjuvant setting: implications for dosing schedule and neoadjuvant study consideration. J Urol 160(5):1685–1688 22. Linja MJ et al (2001) Amplification and overexpression of androgen receptor gene in hormone-refractory prostate cancer. Cancer Res 61(9):3550–3555

304

R.S. Schrecengost et al.

23. Latil A et al (2001) Evaluation of androgen, estrogen (ER alpha and ER beta), and progesterone receptor expression in human prostate cancer by real-time quantitative reverse transcriptionpolymerase chain reaction assays. Cancer Res 61(5):1919–1926 24. Ford OH 3rd et al (2003) Androgen receptor gene amplification and protein expression in recurrent prostate cancer. J Urol 170(5):1817–1821 25. Edwards J et al (2003) Androgen receptor gene amplification and protein expression in hormone refractory prostate cancer. Br J Cancer 89(3):552–556 26. Sharma A et al (2010) The retinoblastoma tumor suppressor controls androgen signaling and human prostate cancer progression. J Clin Invest 120(12):4478–4492 27. Donovan MJ et al (2010) Androgen receptor expression is associated with prostate cancerspecific survival in castrate patients with metastatic disease. BJU Int 105(4):462–467 28. Taplin ME (2007) Drug insight: role of the androgen receptor in the development and progression of prostate cancer. Nat Clin Pract Oncol 4(4):236–244 29. Brooke GN, Bevan CL (2009) The role of androgen receptor mutations in prostate cancer progression. Curr Genomics 10(1):18–25 30. Middleman MN, Lush RM, Figg WD (1996) The mutated androgen receptor and its implications for the treatment of metastatic carcinoma of the prostate. Pharmacotherapy 16(3):376–381 31. Dehm SM et al (2008) Splicing of a novel androgen receptor exon generates a constitutively active androgen receptor that mediates prostate cancer therapy resistance. Cancer Res 68(13):5469–5477 32. Hu R et al (2009) Ligand-independent androgen receptor variants derived from splicing of cryptic exons signify hormone-refractory prostate cancer. Cancer Res 69(1):16–22 33. Guo Z et al (2009) A novel androgen receptor splice variant is up-regulated during prostate cancer progression and promotes androgen depletion-resistant growth. Cancer Res 69(6):2305–2313 34. Jenster G et al (1995) Identification of two transcription activation units in the N-terminal domain of the human androgen receptor. J Biol Chem 270(13):7341–7346 35. Xu J, Wu RC, O’Malley BW (2009) Normal and cancer-related functions of the p160 steroid receptor co-activator (SRC) family. Nat Rev Cancer 9(9):615–630 36. Gregory CW et al (2001) A mechanism for androgen receptor-mediated prostate cancer recurrence after androgen deprivation therapy. Cancer Res 61(11):4315–4319 37. Agoulnik IU et al (2005) Role of SRC-1 in the promotion of prostate cancer cell growth and tumor progression. Cancer Res 65(17):7959–7967 38. Agoulnik IU, Weigel NL (2009) Coactivator selective regulation of androgen receptor activity. Steroids 74(8):669–674 39. Comuzzi B et al (2003) The transcriptional co-activator cAMP response element-binding protein-binding protein is expressed in prostate cancer and enhances androgen- and antiandrogen-induced androgen receptor function. Am J Pathol 162(1):233–241 40. Shang Y, Myers M, Brown M (2002) Formation of the androgen receptor transcription complex. Mol Cell 9(3):601–610 41. Burd CJ, Morey LM, Knudsen KE (2006) Androgen receptor corepressors and prostate cancer. Endocr Relat Cancer 13(4):979–994 42. Knudsen KE (2006) The cyclin D1b splice variant: an old oncogene learns new tricks. Cell Div 1:15 43. Comstock CE et al (2009) Cyclin D1 splice variants: polymorphism, risk, and isoformspecific regulation in prostate cancer. Clin Cancer Res 15(17):5338–5349 44. Montgomery RB et al (2008) Maintenance of intratumoral androgens in metastatic prostate cancer: a mechanism for castration-resistant tumor growth. Cancer Res 68(11):4447–4454 45. Wang Q et al (2009) Androgen receptor regulates a distinct transcription program in androgenindependent prostate cancer. Cell 138(2):245–256 46. Wang Q et al (2007) A hierarchical network of transcription factors governs androgen receptordependent prostate cancer growth. Mol Cell 27(3):380–392

12

Androgen Receptor Regulation of Prostate Cancer Progression and Metastasis

305

47. Sobel RE, Sadar MD (2005) Cell lines used in prostate cancer research: a compendium of old and new lines–part 1. J Urol 173(2):342–359 48. Furusato B et al (2010) CXCR4 and cancer. Pathol Int 60(7):497–505 49. Sun X et al (2010) CXCL12/CXCR4/CXCR7 chemokine axis and cancer progression. Cancer Metastasis Rev 29(4):709–722 50. Vindrieux D, Escobar P, Lazennec G (2009) Emerging roles of chemokines in prostate cancer. Endocr Relat Cancer 16(3):663–673 51. Sun YX et al (2003) Expression of CXCR4 and CXCL12 (SDF-1) in human prostate cancers (PCa) in vivo. J Cell Biochem 89(3):462–473 52. Fernandis AZ et al (2004) Regulation of CXCR4-mediated chemotaxis and chemoinvasion of breast cancer cells. Oncogene 23(1):157–167 53. Prasad A et al (2004) Slit protein-mediated inhibition of CXCR4-induced chemotactic and chemoinvasive signaling pathways in breast cancer cells. J Biol Chem 279(10):9115–9124 54. Kukreja P et al (2005) Up-regulation of CXCR4 expression in PC-3 cells by stromal-derived factor-1alpha (CXCL12) increases endothelial adhesion and transendothelial migration: role of MEK/ERK signaling pathway-dependent NF-kappaB activation. Cancer Res 65(21):9891–9898 55. Wang JF, Park IW, Groopman JE (2000) Stromal cell-derived factor-1alpha stimulates tyrosine phosphorylation of multiple focal adhesion proteins and induces migration of hematopoietic progenitor cells: roles of phosphoinositide-3 kinase and protein kinase C. Blood 95(8):2505–2513 56. Mitra SK, Schlaepfer DD (2006) Integrin-regulated FAK-Src signaling in normal and cancer cells. Curr Opin Cell Biol 18(5):516–523 57. McGrath KE et al (1999) Embryonic expression and function of the chemokine SDF-1 and its receptor, CXCR4. Dev Biol 213(2):442–456 58. Sun YX et al (2005) Skeletal localization and neutralization of the SDF-1(CXCL12)/CXCR4 axis blocks prostate cancer metastasis and growth in osseous sites in vivo. J Bone Miner Res 20(2):318–329 59. Frigo DE et al (2009) Induction of Kruppel-like factor 5 expression by androgens results in increased CXCR4-dependent migration of prostate cancer cells in vitro. Mol Endocrinol 23(9):1385–1396 60. Mochizuki H et al (2004) Interaction of ligand-receptor system between stromal-cell-derived factor-1 and CXC chemokine receptor 4 in human prostate cancer: a possible predictor of metastasis. Biochem Biophys Res Commun 320(3):656–663 61. Kazmin D et al (2006) Linking ligand-induced alterations in androgen receptor structure to differential gene expression: a first step in the rational design of selective androgen receptor modulators. Mol Endocrinol 20(6):1201–1217 62. Jamieson WL et al (2008) CX3CR1 is expressed by prostate epithelial cells and androgens regulate the levels of CX3CL1/fractalkine in the bone marrow: potential role in prostate cancer bone tropism. Cancer Res 68(6):1715–1722 63. Tomlins SA et al (2005) Recurrent fusion of TMPRSS2 and ETS transcription factor genes in prostate cancer. Science 310(5748):644–648 64. Mehra R et al (2008) Characterization of TMPRSS2-ETS gene aberrations in androgenindependent metastatic prostate cancer. Cancer Res 68(10):3584–3590 65. Iljin K et al (2006) TMPRSS2 fusions with oncogenic ETS factors in prostate cancer involve unbalanced genomic rearrangements and are associated with HDAC1 and epigenetic reprogramming. Cancer Res 66(21):10242–10246 66. Mani RS et al (2009) Induced chromosomal proximity and gene fusions in prostate cancer. Science 326(5957):1230 67. Lin C et al (2009) Nuclear receptor-induced chromosomal proximity and DNA breaks underlie specific translocations in cancer. Cell 139(6):1069–1083 68. Mertz KD et al (2007) Molecular characterization of TMPRSS2-ERG gene fusion in the NCIH660 prostate cancer cell line: a new perspective for an old model. Neoplasia 9(3):200–206

306

R.S. Schrecengost et al.

69. Bastus CN et al (2010) Androgen-induced TMPRSS2:ERG fusion in non-malignant prostate epithelial cells. Cancer Res 70(23):9544–9548 70. Hendriksen PJ et al (2006) Evolution of the androgen receptor pathway during progression of prostate cancer. Cancer Res 66(10):5012–5020 71. Tomlins SA et al (2008) Role of the TMPRSS2-ERG gene fusion in prostate cancer. Neoplasia 10(2):177–188 72. Klezovitch O et al (2008) A causal role for ERG in neoplastic transformation of prostate epithelium. Proc Natl Acad Sci USA 105(6):2105–2110 73. Cai J et al (2010) Androgens induce functional CXCR4 through ERG factor expression in TMPRSS2-ERG fusion-positive prostate cancer cells. Transl Oncol 3(3):195–203 74. Zong Y et al (2009) ETS family transcription factors collaborate with alternative signaling pathways to induce carcinoma from adult murine prostate cells. Proc Natl Acad Sci USA 106(30):12465–12470 75. Wang J et al (2006) Expression of variant TMPRSS2/ERG fusion messenger RNAs is associated with aggressive prostate cancer. Cancer Res 66(17):8347–8351 76. Wang J et al (2008) Pleiotropic biological activities of alternatively spliced TMPRSS2/ERG fusion gene transcripts. Cancer Res 68(20):8516–8524 77. Mehra R et al (2007) Comprehensive assessment of TMPRSS2 and ETS family gene aberrations in clinically localized prostate cancer. Mod Pathol 20(5):538–544 78. Mosquera JM et al (2008) Characterization of TMPRSS2-ERG fusion high-grade prostatic intraepithelial neoplasia and potential clinical implications. Clin Cancer Res 14(11): 3380–3385 79. Selbach M et al (2008) Widespread changes in protein synthesis induced by microRNAs. Nature 455(7209):58–63 80. Dastugue N et al (2002) Cytogenetic profile of childhood and adult megakaryoblastic leukemia (M7): a study of the Groupe Francais de Cytogenetique Hematologique (GFCH). Blood 100(2):618–626 81. Sorensen PH et al (1994) A second Ewing’s sarcoma translocation, t(21;22), fuses the EWS gene to another ETS-family transcription factor, ERG. Nat Genet 6(2):146–151 82. Demichelis F et al (2007) TMPRSS2:ERG gene fusion associated with lethal prostate cancer in a watchful waiting cohort. Oncogene 26(31):4596–4599 83. Laxman B et al (2008) A first-generation multiplex biomarker analysis of urine for the early detection of prostate cancer. Cancer Res 68(3):645–649 84. Laxman B et al (2006) Noninvasive detection of TMPRSS2:ERG fusion transcripts in the urine of men with prostate cancer. Neoplasia 8(10):885–888 85. Rostad K et al (2009) TMPRSS2:ERG fusion transcripts in urine from prostate cancer patients correlate with a less favorable prognosis. APMIS 117(8):575–582 86. Coppola V, De Maria R, Bonci D (2010) MicroRNAs and prostate cancer. Endocr Relat Cancer 17(1):F1–F17 87. Garzon R, Marcucci G, Croce CM (2010) Targeting microRNAs in cancer: rationale, strategies and challenges. Nat Rev Drug Discov 9(10):775–789 88. Lu J et al (2005) MicroRNA expression profiles classify human cancers. Nature 435(7043):834–838 89. Calin GA et al (2004) Human microRNA genes are frequently located at fragile sites and genomic regions involved in cancers. Proc Natl Acad Sci USA 101(9):2999–3004 90. Volinia S et al (2006) A microRNA expression signature of human solid tumors defines cancer gene targets. Proc Natl Acad Sci USA 103(7):2257–2261 91. Narayanan R et al (2010) MicroRNAs are mediators of androgen action in prostate and muscle. PLoS One 5(10):e13637 92. Ambs S et al (2008) Genomic profiling of microRNA and messenger RNA reveals deregulated microRNA expression in prostate cancer. Cancer Res 68(15):6162–6170 93. Ribas J et al (2009) miR-21: an androgen receptor-regulated microRNA that promotes hormone-dependent and hormone-independent prostate cancer growth. Cancer Res 69(18):7165–7169

12

Androgen Receptor Regulation of Prostate Cancer Progression and Metastasis

307

94. Shi XB et al (2007) An androgen-regulated miRNA suppresses Bak1 expression and induces androgen-independent growth of prostate cancer cells. Proc Natl Acad Sci USA 104(50):19983–19988 95. Varambally S et al (2008) Genomic loss of microRNA-101 leads to overexpression of histone methyltransferase EZH2 in cancer. Science 322(5908):1695–1699 96. Wang HJ et al (2010) MicroRNA-101 is down-regulated in gastric cancer and involved in cell migration and invasion. Eur J Cancer 46(12):2295–2303 97. Cao P et al (2010) MicroRNA-101 negatively regulates Ezh2 and its expression is modulated by androgen receptor and HIF-1alpha/HIF-1beta. Mol Cancer 9:108 98. Schulz WA, Hatina J (2006) Epigenetics of prostate cancer: beyond DNA methylation. J Cell Mol Med 10(1):100–125 99. Lawrie CH (2007) MicroRNA expression in lymphoma. Expert Opin Biol Ther 7(9):1363–1374 100. Navarro A et al (2008) MicroRNA expression profiling in classic Hodgkin lymphoma. Blood 111(5):2825–2832 101. Wang G et al (2008) Transcription factor and microRNA regulation in androgen-dependent and -independent prostate cancer cells. BMC Genomics 9(Suppl 2):S22 102. Li T et al (2009) MicroRNA-21 directly targets MARCKS and promotes apoptosis resistance and invasion in prostate cancer cells. Biochem Biophys Res Commun 383(3):280–285 103. Lin SL et al (2008) Loss of mir-146a function in hormone-refractory prostate cancer. RNA 14(3):417–424 104. Wang L et al (2009) Gene networks and microRNAs implicated in aggressive prostate cancer. Cancer Res 69(24):9490–9497 105. Gallucci M et al (2006) Cytogenetic profiles as additional markers to pathological features in clinically localized prostate carcinoma. Cancer Lett 237(1):76–82 106. Gallucci M et al (2009) Genetic profile identification in clinically localized prostate carcinoma. Urol Oncol 27(5):502–508 107. Leversha MA et al (2009) Fluorescence in situ hybridization analysis of circulating tumor cells in metastatic prostate cancer. Clin Cancer Res 15(6):2091–2097 108. Jenkins RB et al (1997) Detection of c-myc oncogene amplification and chromosomal anomalies in metastatic prostatic carcinoma by fluorescence in situ hybridization. Cancer Res 57(3):524–531 109. Sato K et al (1999) Clinical significance of alterations of chromosome 8 in high-grade, advanced, nonmetastatic prostate carcinoma. J Natl Cancer Inst 91(18):1574–1580 110. Iwata T et al (2010) MYC overexpression induces prostatic intraepithelial neoplasia and loss of Nkx3.1 in mouse luminal epithelial cells. PLoS One 5(2):e9427 111. Ellwood-Yen K et al (2003) Myc-driven murine prostate cancer shares molecular features with human prostate tumors. Cancer Cell 4(3):223–238 112. Watson PA et al (2005) Context-dependent hormone-refractory progression revealed through characterization of a novel murine prostate cancer cell line. Cancer Res 65(24):11565–11571 113. Visakorpi T et al (1995) In vivo amplification of the androgen receptor gene and progression of human prostate cancer. Nat Genet 9(4):401–406 114. Chuan YC et al (2010) Ezrin mediates c-Myc actions in prostate cancer cell invasion. Oncogene 29(10):1531–1542 115. Silva IS et al (2001) Androgen-induced cell growth and c-myc expression in human nontransformed epithelial prostatic cells in primary culture. Endocr Res 27(1–2):153–169 116. Sun C et al (2008) TMPRSS2-ERG fusion, a common genomic alteration in prostate cancer activates C-MYC and abrogates prostate epithelial differentiation. Oncogene 27(40):5348–5353 117. Grad JM et al (1999) Multiple androgen response elements and a Myc consensus site in the androgen receptor (AR) coding region are involved in androgen-mediated up-regulation of AR messenger RNA. Mol Endocrinol 13(11):1896–1911 118. Khanna C et al (2004) The membrane-cytoskeleton linker ezrin is necessary for osteosarcoma metastasis. Nat Med 10(2):182–186

308

R.S. Schrecengost et al.

119. Yu Y et al (2004) Expression profiling identifies the cytoskeletal organizer ezrin and the developmental homeoprotein Six-1 as key metastatic regulators. Nat Med 10(2):175–181 120. Chuan YC et al (2006) Androgen induction of prostate cancer cell invasion is mediated by ezrin. J Biol Chem 281(40):29938–29948 121. Nupponen NN et al (1998) Genetic alterations in hormone-refractory recurrent prostate carcinomas. Am J Pathol 153(1):141–148 122. de Bono JS et al (2008) Phase I pharmacokinetic and pharmacodynamic study of LAQ824, a hydroxamate histone deacetylase inhibitor with a heat shock protein-90 inhibitory profile, in patients with advanced solid tumors. Clin Cancer Res 14(20):6663–6673 123. Kaltz-Wittmer C et al (2000) FISH analysis of gene aberrations (MYC, CCND1, ERBB2, RB, and AR) in advanced prostatic carcinomas before and after androgen deprivation therapy. Lab Invest 80(9):1455–1464 124. Gurel B et al (2008) Nuclear MYC protein overexpression is an early alteration in human prostate carcinogenesis. Mod Pathol 21(9):1156–1167 125. Hawksworth D et al (2010) Overexpression of C-MYC oncogene in prostate cancer predicts biochemical recurrence. Prostate Cancer Prostatic Dis 13(4):311–315 126. Wolfer A et al (2010) MYC regulation of a “poor-prognosis” metastatic cancer cell state. Proc Natl Acad Sci USA 107(8):3698–3703 127. Martin GS (1970) Rous sarcoma virus: a function required for the maintenance of the transformed state. Nature 227(5262):1021–1023 128. Saad F, Lipton A (2010) SRC kinase inhibition: targeting bone metastases and tumor growth in prostate and breast cancer. Cancer Treat Rev 36(2):177–184 129. Guo Z et al (2006) Regulation of androgen receptor activity by tyrosine phosphorylation. Cancer Cell 10(4):309–319 130. Kraus S et al (2006) Receptor for activated C kinase 1 (RACK1) and Src regulate the tyrosine phosphorylation and function of the androgen receptor. Cancer Res 66(22):11047–11054 131. Asim M et al (2008) Src kinase potentiates androgen receptor transactivation function and invasion of androgen-independent prostate cancer C4-2 cells. Oncogene 27(25):3596–3604 132. Gingrich JR et al (1997) Androgen-independent prostate cancer progression in the TRAMP model. Cancer Res 57(21):4687–4691 133. Tatarov O et al (2009) SRC family kinase activity is up-regulated in hormone-refractory prostate cancer. Clin Cancer Res 15(10):3540–3549 134. Araujo J, Logothetis C (2010) Dasatinib: a potent SRC inhibitor in clinical development for the treatment of solid tumors. Cancer Treat Rev 36(6):492–500 135. Park SI et al (2008) Targeting SRC family kinases inhibits growth and lymph node metastases of prostate cancer in an orthotopic nude mouse model. Cancer Res 68(9):3323–3333 136. Labrie F et al (2005) Gonadotropin-releasing hormone agonists in the treatment of prostate cancer. Endocr Rev 26(3):361–379 137. Shepard DR, Raghavan D (2010) Innovations in the systemic therapy of prostate cancer. Nat Rev Clin Oncol 7(1):13–21 138. Haidar S et al (2003) Effects of novel 17alpha-hydroxylase/C17, 20-lyase (P450 17, CYP 17) inhibitors on androgen biosynthesis in vitro and in vivo. J Steroid Biochem Mol Biol 84(5):555–562 139. Attard G et al (2008) Phase I clinical trial of a selective inhibitor of CYP17, abiraterone acetate, confirms that castration-resistant prostate cancer commonly remains hormone driven. J Clin Oncol 26(28):4563–4571 140. Reid AH et al (2010) Significant and sustained antitumor activity in post-docetaxel, castrationresistant prostate cancer with the CYP17 inhibitor abiraterone acetate. J Clin Oncol 28(9):1489–1495 141. Tran C et al (2009) Development of a second-generation antiandrogen for treatment of advanced prostate cancer. Science 324(5928):787–790 142. Culig Z et al (1999) Switch from antagonist to agonist of the androgen receptor bicalutamide is associated with prostate tumour progression in a new model system. Br J Cancer 81(2):242–251

12

Androgen Receptor Regulation of Prostate Cancer Progression and Metastasis

309

143. Scher HI et al (2010) Antitumour activity of MDV3100 in castration-resistant prostate cancer: a phase 1–2 study. Lancet 375(9724):1437–1446 144. Handratta VD et al (2005) Novel C-17-heteroaryl steroidal CYP17 inhibitors/antiandrogens: synthesis, in vitro biological activity, pharmacokinetics, and antitumor activity in the LAPC4 human prostate cancer xenograft model. J Med Chem 48(8):2972–2984 145. Vasaitis T et al (2008) Androgen receptor inactivation contributes to antitumor efficacy of 17{alpha}-hydroxylase/17,20-lyase inhibitor 3beta-hydroxy-17-(1H-benzimidazole-1-yl) androsta-5,16-diene in prostate cancer. Mol Cancer Ther 7(8):2348–2357 146. Andersen RJ et al (2010) Regression of castrate-recurrent prostate cancer by a smallmolecule inhibitor of the amino-terminus domain of the androgen receptor. Cancer Cell 17(6):535–546 147. Shen R et al (2000) Androgen-induced growth inhibition of androgen receptor expressing androgen-independent prostate cancer cells is mediated by increased levels of neutral endopeptidase. Endocrinology 141(5):1699–1704 148. Xu K et al (2009) Regulation of androgen receptor transcriptional activity and specificity by RNF6-induced ubiquitination. Cancer Cell 15(4):270–282

Chapter 13

Integrin Signaling in Angiogenesis and Metastatic Cancer Progression in the Brain Mihaela Lorger and Brunhilde Felding-Habermann

Abstract Integrins have been identified as major contributors to the progression of brain cancer. This includes the development and spreading of primary brain tumors and, more recently, also metastatic disease originating from solid tumors outside the brain. In these processes, integrin functions affect not only the tumor cells themselves but also the reactions of host cells that respond to the presence of developing lesions within the brain microenvironment. Integrin functions can promote cell survival, proliferation, invasion and regulatory adjustments of signaling cascades triggered by other receptor types, including those for growth factors. Supported by the promise of ongoing clinical trials, specific targeting of certain integrins and their functions could thus prove successful for new therapeutic approaches against brain cancer and metastatic disease in the central nervous system.

13.1 13.1.1

Introduction Integrins

Integrins are a family of heterodimeric cell surface receptors consisting of a and b subunits (Fig. 13.1a). Eighteen different a and eight different b subunits combine to give rise to at least 24 different integrin heterodimers with distinct and partially overlapping ligand binding specificities (Fig. 13.1b). The integrin subunits consist of extracellular domains which combine to form the ligand binding site, transmembrane domains, and cytoplamic domains. Integrins bind to proteins within the extracellular matrix (ECM) and physically connect them to the cytoskeleton. This is mediated by recruiting cytoskeleton-binding proteins such as talin, paxillin and

M. Lorger • B. Felding-Habermann (*) The Scripps Research Institute, La Jolla, CA, USA e-mail: [email protected]

A. Fatatis (ed.), Signaling Pathways and Molecular Mediators in Metastasis, DOI 10.1007/978-94-007-2558-4_13, © Springer Science+Business Media B.V. 2012

311

Fig. 13.1 (a) Integrin heterodimers bind distinct classes of ligands through their extracellular domains. Integrin ligands include extracellular matrix proteins and soluble forms of matrix proteins, but also certain growth factors and proteases. Depending on the ligand, integrin ligation induces recruitment of cytoskeleton-binding proteins and/or signaling molecules to the integrin cytoplasmic domains (blue arrows). In malignancies, integrins often cross-talk with receptor tyrosine kinases (RTK), resulting in activation of downstream signaling pathways (green arrows). These complex and diverse interactions regulate cellular functions such as adhesion, migration, invasion, proliferation, as well as survival. Thereby, integrins ultimately co-determine the malignant potential of cancer cells, the extent of tumor angiogenesis, as well as the nature of inflammatory and immune responses to cancer. ECM extracellular matrix, FAK focal adhesion kinase, SFK Src family kinases, PI3K phosphatidylinositol-3 kinase, ILK integrin linked kinase, ERK extracellular signal regulated kinase, p130CAS Crk-associated substrate, RhoB ras homolog gene family, member B (b) Schematic representation of known heterodimeric interactions between integrin a and b subunits (Adapted from Hynes [7])

13

Integrin Signaling in Angiogenesis and Metastatic Cancer Progression in the Brain

313

vinculin to the cytoplasmic integrin domain. Furthermore, integrins transduce bidirectional signals across the plasma membrane. Integrin binding to the ECM can result in a conformational change and integrin activation. ECM binding induces integrin clustering into so-called focal adhesions and results in the recruitment of signaling proteins to the cytoplasmic domain of integrin b subunits. These signaling proteins include Focal Adhesion Kinase (FAK), Integrin Linked Kinase (ILK), Src Family Kinases (SFK), and p130CAS. The interaction triggers activation of downstream signaling pathways such as PI3K/Akt and ERK signaling. By mediating cell-matrix interactions and recruiting signaling and adapter molecules, integrins regulate critical cellular functions including adhesion, migration, invasion, proliferation and survival. In addition to transducing signals from the extracellular space (outside-in signaling), integrins can also become activated by binding of signaling molecules to their cytoplasmic tails (inside-out signaling). This can occur in response to cell stimulation by agonists that engage other receptors. The activation response triggers changes in integrin affinity for ligand binding and directly affects cellular functions. Once activated, high affinity integrins may bind soluble ligands, e.g., certain plasma proteins, and in doing so perceive signals that support survival of cells in suspension, as for instance in the blood stream. In addition to recognizing their cognate matrix or plasma protein ligands, integrins can also interact with other classes of ligands such as growth factors and hormones [1, 2]. Different ligand proteins can induce diverse molecular and cellular responses (Fig. 13.1). Through their complex ligand interaction profiles, integrins were also shown to cross-talk with growth factor receptors and oncogenes. This cross-talk is required for many of the cancer-promoting effects of integrins [3–8]. Different cell types are characterized by distinct integrin repertoires. Integrins are expressed on tumor cells, endothelial cells, stromal cells, bone marrow-derived cells, and immune cells. The expression pattern of integrins and their downstream cellular functions therefore codetermine the malignant potential of cancer cells (migration, invasion, proliferation, survival), tumor angiogenesis, as well as immune responses to cancer. In this chapter, we will summarize some examples of integrin involvement in these processes in the context of brain malignancies. Our main focus will be on integrins avb3, avb5 and b1-containing integrins, because they are among the most extensively studied integrins in glioma tumors and brain metastases.

13.1.2

Integrating Knowledge from Experimental Models and Clinical Studies

Literature on the involvement of integrins in brain malignancies reports clinical data from human patients and data obtained with experimental preclinical models. Clinical data focus on integrin expression in patient biopsies based on immunohistochemical and gene profiling analyses, as well as clinical trials with integrin antagonists. Experimental preclinical studies include orthotopic animal models

314

M. Lorger and B. Felding-Habermann

where tumor cells are studied within the brain microenvironment (intracranial), non-orthotopic animal models (mostly subcutaneous), and in vitro studies on cultured tumor cells. In line with the increasing recognition of the importance of the tumor microenvironment in cancer progression and the complexity of the in vivo situation as compared to simplified in vitro conditions, the three experimental settings often result in distinct observations. For example, the vasculature within the central nervous system has been shown to be significantly different from the vasculature in the subcutaneous space [9, 10]. Growth factor-dependent as well as integrin-dependent growth and angiogenesis of cancer cells differ significantly when the cells are implanted orthotopically into the brain versus subcutaneously [9–12]. It is therefore crucial to consider the specific conditions under which the corresponding observations have been made.

13.2

Expression of Integrins and Their Ligands Is Altered in Brain Malignancies

Integrin expression is frequently altered in cancer cells as well as in the tumor vasculature compared to normal cells and tissues. This is prominently observed in brain malignancies [13, 14]. Integrin expression in brain tumors is also temporospatially regulated and can change with disease progression and responses within the tumor microenvironment. For example, integrin avb3 is strikingly upregulated in glioma progression. avb3 has been shown to increase with a higher glioma grade and upon stimulation with TGF-b1 and TGF-b2 [15, 16], uPAR [17] or upon brain exposure to radiation. The latter induces avb3 expression in cancer cells, as well as in endothelial cells of the tumor microvasculature [18–20]. Hypoxia and PDGF, for example, recruit avb3 to the cell membrane and focal adhesions, thereby altering downstream integrin signaling [21, 22]. Thus, integrin expression patterns and their activation states are subject to change during disease progression and/or treatment. In addition to alterations in the cellular integrin repertoire in brain tumors and metastatic lesions in the brain, the distribution of ECM components within the tumor microenvironment is a key parameter that controls integrin-mediated cellular functions. This control involves integrin activation in response to binding ligands within the ECM. The central nervous system has a unique composition of ECM principally consisting of hyaluronane and proteoglycans, which can inhibit cell adhesion and migration [23, 24]. ECM of the normal brain lacks matrix substances that provide strength to the extracellular space in other organs, such as fibrillar collagen, fibronectin, laminin and vitronectin. Within the brain, these matrix components are restricted to perivascular regions and the glia limitans [3, 25]. In contrast, primary brain tumors show altered expression of certain ECM components, most notably vitronectin and tenascin-C, whose expression positively correlates with the tumor grade [3, 4, 26–30]. Both vitronectin and tenascin are deposited by cancer cells at the invasive edge of the tumor. Laminin and fibronectin have also been reported to be expressed in gliomas, mostly in association with blood vessels [29, 31–34].

13

Integrin Signaling in Angiogenesis and Metastatic Cancer Progression in the Brain

315

In contrast to primary brain tumors, no substantial analysis of ECM in brain metastasis has been reported in the literature.

13.3

Role of Integrins in Metastatic Colonization of the Brain

Approximately 30% of all patients with advanced cancer present with brain metastases. The clinical incidence of brain metastasis is increasing, likely due to improved treatments for extracranial metastatic lesions. The most common sources of brain metastasis are lung cancer, breast cancer and melanoma [35, 36]. To successfully colonize the brain, cancer cells need to arrest and survive within the brain microvasculature, extravasate through the vessel wall into the brain parenchyma, and resume proliferation to give rise to macroscopic brain lesions. Our knowledge about the involvement of integrins in these individual steps of brain metastasis is limited to only a few reports. A prominent integrin recognized as a potential key player in primary brain tumors is avb3. However, its role in the initial steps of brain colonization by circulating tumor cells is largely unknown. Studies of avb3 in experimental melanoma metastasis to the brain in immunodeficient mice indicated that the pattern of brain metastasis (leptomeningeal versus parenchymal) did not depend on avb3 expression [37]. In a similar model, hematogenous brain colonization [38] as well as intracranial growth of some breast cancer cells [12] was not altered by avb3 expression in the experimental models investigated. However, in breast cancer cells that naturally express avb3, the ability of the tumor cells to grow within the brain microenvironment strongly depended on the activation state of avb3 integrin as discussed below (see Sect. 13.5.2) [12]. During the initial growth phase after brain colonization, metastatic cancer cells expand around preexisting brain vessels (vessel cooption). The interaction between breast cancer cells and the vascular basement membrane in the brain seems to be mediated by b1 integrins expressed by the tumor cells. The evidence is based on the ability of a function-blocking anti-b1 integrin antibody to inhibit cancer cell attachment to and growth on human brain slices in vitro. It was suggested that this response is a consequence of disturbed interactions between tumor cells and the vascular basement membrane, which serves as a “soil” in brain metastasis [39]. Targeting of integrin a3b1 with a3-blocking antibody also significantly decreased experimental metastasis of lung cancer cells to the brain [40]. Apart from avb3 and b1-containing integrins, additional integrins likely contribute to different steps of brain metastasis, possibly with variations depending on the cancer cell type.

13.4

Role of Integrins in Cancer Cell Migration and Invasion Within the Brain

Invasive growth is a hallmark of malignant progression in primary brain tumors and poses a major problem in their clinical treatment. For example, dissemination of glioma cells within the brain follows myelinated fiber tracts and extracellular matrix

316

M. Lorger and B. Felding-Habermann

containing structures, especially the basement membrane of blood vessels and the glia limitans externa. Integrins are major regulators of glioma cell interactions with the extracellular matrix, and therefore play an important role in glioma invasion. Numerous in vitro studies demonstrated that a variety of purified matrix proteins including laminin, collagen IV, fibronectin, vitronectin and tenascin support the adhesion and migration of different glioma cell lines as well as that of primary glioma cultures to a varying degree [26, 28, 33, 41–45]. In vivo, ECM consists of a complex assembly of different matrix proteins that form fibrillar networks of specific compositions and molecular interactions [46]. Even small amounts of nonpermissive ligands have been reported to significantly alter the permissiveness of ligands that otherwise support adhesion and migration [25]. Because of this reason and the complexity and dynamic structure of the ECM in vivo, purified matrix proteins cannot accurately emulate the permissiveness of matrices for cellular interactions within tissues. To more closely mimic the in vivo situation, experiments with reconstituted basement membrane (matrigel) or brain slices have been performed. Glioma cell adhesion to the basement membrane of blood vessels on brain slices was shown to be mediated mainly by b1 integrins and avb3 [47]. These findings suggest that these receptors might be the main players under complex conditions as found in vivo.

13.4.1

b1-Containing Integrins

b1 integrin heterodimerizes with 12 different a subunits and therefore participates in binding to a variety of different ECM proteins. These include laminin, collagen and fibronectin, three major components of the vascular basement membrane. They also include vitronectin and tenascin, the most commonly up-regulated matrix proteins in high-grade gliomas. The b1 integrin subunit is frequently up-regulated in primary brain tumors. Numerous in vitro studies indicated a role of b1 integrins in the adhesion, migration and invasion of glioma and brain metastatic cells, most notably in combination with the a3 subunit, but also in combination with a2, a5, a8 and a9 [1, 3, 4, 25, 26, 39, 47–54]. b1 Integrin over-expression in an orthotopic glioma xenograft model was shown to increase glioma invasiveness [52]. In human patients, a9b1 expression correlated with an increased glioma grade, and in vitro a9b1 mediated NGF-induced glioma cell migration [1]. Integrin a9b1 and its ligand tenascin C were found highly up-regulated in leptomeningeal extensions of medulloblastoma compared to the primary tumors, suggesting that a9b1 might be involved in medulloblastoma spreading. Supporting this hypothesis, leptomeningeal spreading and integrin a9b1 expression both predicted a poor patient outcome. In line with the clinical data, medulloblastoma cells in vitro preferentially used a9b1 to adhere to a tenascin-rich ECM produced by glioma cells. This adhesion increased the survival and proliferation of the medulloblastoma cells in an a9b1 dependent manner [26].

13

Integrin Signaling in Angiogenesis and Metastatic Cancer Progression in the Brain

13.4.2

317

avb3 and avb5 Integrins

Like certain other integrin family members, integrins avb3 and avb5 primarily recognize an arginine-glycine-aspartic acid (RGD) peptide motif within their ligands. avb3 binds to several different matrix proteins, including vitronectin, fibronectin, osteopontin and tenascin, while avb5 binds only vitronectin within this group of ligands [4]. avb3 is expressed by the vasculature of brain tumors [55–57]. Importantly, avb3 and its ligand vitronectin were also found associated with glioma cells at the invasive margin of high-grade tumors (grade IV and III), while lower grade astrocytomas do not express vitronectin [27, 28, 57, 58]. Thus, avb3 and its ligand vitronectin correlate with progression of aggressive brain tumors. This finding was demonstrated in human tumor biopsies as well as in xenograft models, suggesting an involvement of avb3 in glioma invasion. Notably, expression of vitronectin in experimental glioma models was induced only when cancer cells were growing orthotopically in the brain but not in the subcutaneous space or in vitro [28]. This demonstrates the importance of the tumor microenvironment in the induction of vitronectin expression by glioma cells. In vitro experiments showed that integrins avb3 and avb5 both promote glioma cell migration and adhesion on vitronectin, as well as glioma invasion [15, 16, 27, 28, 54, 59]. These integrin-mediated processes can be regulated at different levels. For example, avb3-dependent adhesion of glioma cells to vitronectin and FGF-2 was inhibited by PEX, an auto-proteolytic fragment of MMP-2 [58]. Furthermore, migration of glioblastoma cells on vitronectin depends on PDGF only when avb3 is engaged, but not when avb5 is involved. Thus, for avb3, tumor cell migration involves cross-talk between the integrin and the PDGF receptor. This receptor supports recruitment of avb3 to focal adhesions within the leading edge of the migratory tumor cells. PDGF-induced glioblastoma cell migration required association of avb3 with the activated Src family kinase Lyn [21]. In addition to mediating migration on vitronectin, avb3 has also been postulated to increase glioma cell invasiveness by regulating the activation of proteases involved in the degradation and remodeling of the ECM [60, 61].

13.5

The Complex Role of Integrins in Regulating Growth and Angiogenesis of Brain Tumors

In animal models, primary brain tumors and brain metastases are initially growing without neovascularization by coopting the pre-existing brain microvasculature [38, 39, 62]. Once cancer lesions become larger, their growth depends on the recruitment of new blood vessels. This is achieved through sprouting of the preexisting vessels (angiogenesis) or recruitment of vascular progenitor cells that incorporate into new vessels (vasculogenesis). The process of angiogenesis is

318

M. Lorger and B. Felding-Habermann

controlled by integrins expressed by endothelial cells and cancer cells, most notably by integrins avb3 and avb5. In addition, avb3 on tumor cells can regulate growth of glioma and brain metastases in vitro by mechanisms that do not require angiogenesis [63, 64].

13.5.1

Impact of Different avb3 Ligands on Brain Tumor Growth and Angiogenesis

Integrin engagement with their ligands is crucially involved in the regulation of endothelial cell proliferation and survival. Integrin avb3 is the most studied integrin on endothelial cells and a marker of endothelial cell proliferation and angiogenesis. Unligated avb3 on endothelial cells triggers the so-called integrin-mediated cell death response by inducing apoptosis via caspase-8 [5]. The engagement of avb3 with different ligands can have both positive and negative effects on the survival of endothelial cells and consequently on angiogenesis. avb3 on endothelial cells can bind factors that are thought to enhance angiogenesis. These include VEGF-R2, vitronectin, fibronectin and thrombin. But, endothelial avb3 can also interact with factors thought to induce anti-angiogenic effects like thrombospondin, endostatin, angiostatin and tumstatin [65]. For example, binding of tumstatin to avb3 on blood vessels in primary brain tumors induced apoptosis of endothelial cells by inhibition of the FAK/PI3K/Akt/mTOR pathway and resulted in vessel regression [66]. Similarly to avb3 on endothelial cells, avb3 on tumor cells can cause opposing effects depending on the tumor cell type and the ligands bound to the receptor (summarized in Fig. 13.2). For example, tumstatin inhibited the growth of glioma cells expressing functional PTEN, functional mTOR signaling, and low levels of Akt, by binding to avb3 on the tumor cells [11]. However, gliomas with a dysfunctional PTEN pathway were not sensitive to tumstatin binding. In contrast to tumstatin, several avb3 ligands were shown to exert a positive effect on glioma growth. For example, in vitro studies using avb3 antagonists, or antisense RNA against the av subunit gene, demonstrated that attachment of brain tumor cells to vitronectin and tenascin promotes cell proliferation and protects against apoptosis [63, 64]. Another example includes thyroid hormones. Integrin avb3 was recently shown to bind triiodo-L-thyronine (T3) and L-thyroxine (T4). Both T3 and T4 readily penetrate the blood-brain barrier and enter the brain. It has been suggested that avb3 harbors two distinct T3 binding sites, one of which is shared with T4. T3 and T4 both induced ERK activation and proliferation of glioma cells in vitro [2]. In addition T3, but not T4, activated Src and PI3K, leading to the up-regulation of HIF-1a expression and to translocation of Thyroid receptor a (TRa) into the nucleus [2] (Fig. 13.2).

13

Integrin Signaling in Angiogenesis and Metastatic Cancer Progression in the Brain

319

Fig. 13.2 Different ligands induce distinct avb3-mediated responses in glioma cells. avb3 engagement by tumstatin negatively regulates glioma growth by inhibiting PI3K/Akt survival pathway and thereby inducing apoptosis (blue). Binding of avb3 to ECM proteins vimentin and tenascin promotes glioma adhesion, migration, invasion, proliferation, and inhibition of druginduced apoptosis (green). Binding of thyroid hormone T3 to avb3 induces the expression of HIF-1a via activation of Src and PI3K (red), while binding of T4 induces ERK activation (violet)

13.5.2

The Role of avb3 in Brain Tumor Biology Strongly Depends on the Tumor Microenvironment and the Genetic Background

In vitro, integrin avb3 antagonists were shown to cause glioma cell apoptosis mediated by caspase 3 and caspase 9. In vivo, avb3 inhibitors can lead to brain tumor regression [67]. However, over-expression of avb3 in animal models can also lead to the inhibition of tumor growth [68]. This apparent discrepancy may have a number of underlying reasons. It could be due to the complexity of targeting avb3 expressed by the tumor cells, avb3 of the vasculature, or both. It may also be due to the fact that distinct glioma cell models were used (different tumor grades and distinct genetic backgrounds), and that the tumor cells were implanted at different sites (mostly non-orthotopically into the subcutis versus orthotopically into the brain). The complexity of integrin avb3 biology is nicely exemplified in studies on

320

M. Lorger and B. Felding-Habermann

Fig. 13.3 Distinct roles of tumor and host associated integrins in glioma. Integrin avb3 can be expressed by glioma cells as well as by angiogenic endothelial cells and macrophages. When experimentally overexpressed in low-grade glioma, avb3 leads to increased hypoxia and reduced tumor growth. avb3 expressed by endothelial cells promotes tumor angiogenesis and thereby increases tumor growth. avb3 on macrophages can mediate their homing to brain lesions and promote secretion of TNFa, thereby inducing apoptosis in the cancer cells

orthotopic glioma animal models that dissect the contribution of avb3 expressed by the host vasculature and tumor-infiltrating cells (host-avb3) versus avb3 expressed by the cancer cells (tumor-avb3) [68, 69] (summarized in Fig. 13.3). To study the role of tumor cell avb3, low-grade glioma cells were modified to overexpress b3 and implanted into the brain of wild type mice. This resulted in reduced tumor growth and in vascular defects that lead to increased hypoxia. The apparent suppressive effect of tumor cell avb3 was overcome by experimental activation of the Akt and VEGF pathways in the tumor cells, as commonly seen in clinical samples of avb3-overexpressing high grade glioblastoma multiforme [69]. Thus, the role of tumor-avb3 in brain cancer biology strongly depends on the glioma grade and its genetic background. To assess a role of host avb3, glioma growth was compared in wild type versus b3 knock-out mice. After implantation of avb3 positive glioma cells, host-avb3 on endothelial cells lead to increased angiogenesis. Nevertheless, overall glioma lesion growth was similar in b3 positive and b3 null mice. This apparent discrepancy was caused by the fact that the growing tumors were infiltrated by a larger number of TNFa secreting and apoptosis inducing avb3-positive macrophages in wild type mice as opposed to b3 null mice [68]. Thus, avb3 integrin expressed on host and tumor cells can exert opposing functions on glioma progression depending on the glioma grade, its genetic background, and the host immune status (Fig. 13.3). In contrast to low grade glioma where no growth differences were observed in tumor cells after experimental expression of wild type versus constitutively activated avb3, growth and angiogenesis of breast cancer brain metastases in an animal

13

Integrin Signaling in Angiogenesis and Metastatic Cancer Progression in the Brain

321

model was highly dependent on the activation state of tumor cell-avb3 [12]. Activated avb3 resulted in a strong hyper-phosphorylation and thereby inactivation of the translational repressor 4E-BP1. This lead to up-regulation of VEGF expression at the post-transcriptional level. Consequently, brain lesions expressing activated avb3 displayed increased angiogenesis and strongly decreased hypoxia, providing a significant advantage for tumor growth. Notably, avb3 integrin function is highly dependent on the tumor microenvironment, as opposing effects of avb3 on the growth of both glioma and breast cancer cells were demonstrated in the subcutaneous versus the intracranial setting [12, 68]. Among other factors, this might be related to the fact that the expression of avb3 ligand vitronectin in glioma cells was induced only within the cerebral microenvironment and not within the subcutaneous space or in vitro [28]. Interestingly, cilengitide, a cyclic RGD peptide that blocks both avb3 and avb5, also inhibited glioma growth only when the tumors were growing orthotopically in the brain, but not in the subcutis [70]. In summary, this demonstrates that it is crucial to study the role of integrins in glioma biology as well as in brain metastasis in an appropriate microenvironmental setting.

13.5.3

Other Integrins Involved in Glioma Angiogenesis

Integrin avb8 has been reported to be strongly upregulated in glioma [71] and was recently identified as a regulator of glioma angiogenesis in an experimental animal model [72]. Integrin avb8 is a known receptor for the latent forms of TGF-b 1 and 3, and it mediates their bioactivation [73]. TGF-b activation and signaling promote blood vessel growth and sprouting. avb8 in an astrocytoma model was important for the induction of endothelial TGF-b signaling. At the same time, this integrin positively regulated VEGF expression at the transcriptional level. Phenotypically, down-regulation of avb8 in an astrocytoma model resulted in increased angiogenesis that was associated with strong vascular defects and increased hemorrhage [72].

13.6 13.6.1

Relevance of Integrins for Glioma Treatment Integrins Mediate Glioma Resistance to Radiation and Chemotherapy

Ionizing radiation, one of the standard therapies for primary brain tumors, induces unwanted pro-angiogenic effects and glioma radioresistance. This effect is thought to enable tumor recurrence after irradiation and surgery. This seems to be at least

322

M. Lorger and B. Felding-Habermann

partially due to radiation-induced expression of avb3 integrin on endothelial cells as well as on glioma cells [18–20]. Radiation-induced expression of avb3 on endothelial cells was shown to increase the Akt survival pathway, cell proliferation and migration. This escape mechanism from radiation damage resulted in increased angiogenesis that could be counteracted by inhibiting the integrin [18]. The resistance of glioma cells to ionizing radiation is promoted by avb3 and avb5. Combined radiation and inhibition of the two integrins with specific small molecule antagonists or siRNA showed synergistic effects in reducing glioma growth in vitro and in animal models compared to the individual therapies [18, 74, 75]. In vitro, avb3 and avb5 integrins have been shown to increase the transcription level of RhoB in ILK-independent manner, as well as RhoB activation through ILK, thereby reducing apoptosis of glioma cells and increasing their radioresistance [75]. Radiation-induced expression of avb3 was also shown to correlate with an increased invasiveness of glioma cells [20]. This could be counteracted by Temozolomide, which was able to suppress radiation-induced upregulatin of avb3 and, at the same time, to inhibit glioma invasiveness by inactivating FAK, a signaling regulator downstream of avb3 [20]. Binding of avb3 and avb5 integrins to vitronection was shown to inhibit druginduced apoptosis in glioma cells. This was associated with increased expression of anti-apoptotic proteins Bcl-2 and Bcl-XL [76]. Integrin b1 was also shown to protect glioma cells from radiation-induced cell death in vitro by activating the PI3K/ Akt-mediated pro-survival signaling [19, 49].

13.6.2

Targeting Brain Tumors with avb3/avb5 Integrin Antagonists

Due to its expression specifically on angiogenic blood vessels and its up-regulation in cancer, integrin avb3 has become the main therapeutic integrin target in cancer. Many different avb3 antagonists have been described in the literature, including peptides and function blocking antibodies [77]. Cilengitide (EMD 121974), a cyclic RGD peptide that blocks both avb3 and avb5 integrins recently received the most attention. Cilengitide induced apoptosis in glioma cells grown on vitronectin or tenascin [64]. In preclinical animal models of glioma, cilengitide and other avb3 antagonists increased survival by reducing tumor growth. This was associated with a decrease in tumor cell proliferation and angiogenesis, and an increase in tumor cell apoptosis mediated by caspases-3 and -9 [67, 70, 78]. Phase I and II clinical trials administering cilengitide to glioma patients demonstrated that cilengitide is well tolerated, apparently penetrates the blood-brain barrier, and is clinically active [79–81]. Further clinical studies with newly diagnosed glioma patients investigated a potential synergy of adding cilengitide to standard temozolomide/external beam radiotherapy. The results demonstrated that adding cilengitide was beneficial to patients with a low level of methylguanine methyltransferase

13

Integrin Signaling in Angiogenesis and Metastatic Cancer Progression in the Brain

323

(MGMT) expression in the tumor, as determined by gene promoter methylation [80]. Based on the promising phase I and II trials, an international randomized phase III clinical study (CENTRIC) was initiated and is expected to be completed by summer of 2011 [81]. Preclinical studies demonstrated that blocking integrin avb3 in glioma results in reduced hypoxia and vessel normalization [22]. This should lead to improved blood perfusion and thereby improved drug delivery. Cilengitide was also shown to sensitize angiogenic endothelial cells to apoptosis induced by ionizing radiation. Thereby cilengitide enhances the therapeutic damage to the tumor neovasculature. In preclinical models, cilengitide decreased the expression of pAkt in glioma cells, suggesting that its synergistic effect with radiation treatment might be due to the inhibition of pro-survival signaling after irradiation [74]. Thus, in clinical settings cilengitide might reduce the resistance to radiation and/or chemotherapy. Therefore, synergistic therapies are being considered [79, 81].

13.6.3

a5b1 Integrin Antagonists as Potential Therapeutics for Targeting Glioma Progression

The integrin a5 subunit was shown to be upregulated in glioma tissue compared to normal brain [82]. Perturbation of integrin a5b1 function with specific antagonists in vitro triggered cell cycle arrest in glioma cells [83, 84]. Integrin a5b1 antagonists also sensitized glioma cells to chemotherapy by reducing chemotherapy-induced expression of p53 and p21, and by favoring cell apoptosis over premature senescence, demonstrated in vitro [85]. Phase I clinical studies for solid malignancies administering Volociximab, an antibody that targets a5b1, and ATN-161, an a5b1targeting peptide, are currently ongoing [86].

13.7

Role of Integrins in Glioma Cancer Stem Cell Function

Small subpopulations of stem-like cells, so-called cancer stem cells (CSCs), have been identified in solid tumors including glioma [87, 88]. Most notable characteristics of CSCs are their capability to initiate tumor growth from a small number of cells, generation of diverse progeny and thereby heterogeneous cancer cell population, as well as resistance to chemotherapy and radiation. As such, CSCs are thought to be responsible for tumor recurrence, which poses major problems in the clinic [89–91]. Following a6 integrins with antibodies against the a6 subunit recently demonstrated their role in regulating a cancer stem cell population in glioma [92]. Glioma stem cells were shown to predominantly localize close to the vasculature in a socalled vascular niche [93]. Accordingly, the majority of a6 integrin positive glioma

324

M. Lorger and B. Felding-Habermann

cells in patient samples was associated with the vasculature and co-expressed the glioma stem cell marker CD133. In vitro, glioma cells expressing high levels of a6 integrin more efficiently formed tumorospheres, which is a measure of stem cell self-renewal and proliferation. Integrin a6high cells were also capable of differentiating into different cell lineages of the central nervous system and showed increased tumorigenicity. ShRNA-mediated down-regulation of a6 integrins resulted in reduced tumorosphere formation in vitro and reduced tumor growth in vivo. In summary, these data implicate a6 integrins as positive regulators of the cancer stem cell phenotype in glioma [92].

13.8

Outlook

Tumor host interactions are important determinants of brain tumor development. Such interactions are also critical and perhaps decisive for the establishment and progression of metastatic brain disease that originates from solid tumors outside the central nervous system. Both tumor cells and responding host cells rely on integrins for many functions that are emerging as important for metastasis within the brain microenvironment. Specific targeting of such integrin contributions may help improve treatments of primary and metastatic cancer in the brain.

References 1. Brown MC, Staniszewska I, Lazarovici P, Tuszynski GP, Del VL, Marcinkiewicz C (2008) Regulatory effect of nerve growth factor in alpha9beta1 integrin-dependent progression of glioblastoma. Neuro Oncol 10:968–980 2. Lin HY, Sun M, Tang HY, Lin C, Luidens MK, Mousa SA, Incerpi S, Drusano GL, Davis FB, Davis PJ (2009) L-Thyroxine vs. 3,5,3¢-triiodo-L-thyronine and cell proliferation: activation of mitogen-activated protein kinase and phosphatidylinositol 3-kinase. Am J Physiol Cell Physiol 296:C980–C991 3. Bellail AC, Hunter SB, Brat DJ, Tan C, Van Meir EG (2004) Microregional extracellular matrix heterogeneity in brain modulates glioma cell invasion. Int J Biochem Cell Biol 36:1046–1069 4. D’Abaco GM, Kaye AH (2007) Integrins: molecular determinants of glioma invasion. J Clin Neurosci 14:1041–1048 5. Desgrosellier JS, Cheresh DA (2010) Integrins in cancer: biological implications and therapeutic opportunities. Nat Rev Cancer 10:9–22 6. Guo W, Giancotti FG (2004) Integrin signalling during tumour progression. Nat Rev Mol Cell Biol 5:816–826 7. Hynes RO (2002) Integrins: bidirectional, allosteric signaling machines. Cell 110:673–687 8. Mitra SK, Schlaepfer DD (2006) Integrin-regulated FAK-Src signaling in normal and cancer cells. Curr Opin Cell Biol 18:516–523 9. Blouw B, Song H, Tihan T, Bosze J, Ferrara N, Gerber HP, Johnson RS, Bergers G (2003) The hypoxic response of tumors is dependent on their microenvironment. Cancer Cell 4:133–146 10. Guo P, Xu L, Pan S, Brekken RA, Yang ST, Whitaker GB, Nagane M, Thorpe PE, Rosenbaum JS, Su Huang HJ et al (2001) Vascular endothelial growth factor isoforms display distinct

13

11.

12.

13.

14. 15.

16.

17.

18.

19.

20.

21.

22.

23. 24. 25. 26.

27.

Integrin Signaling in Angiogenesis and Metastatic Cancer Progression in the Brain

325

activities in promoting tumor angiogenesis at different anatomic sites. Cancer Res 61: 8569–8577 Kawaguchi T, Yamashita Y, Kanamori M, Endersby R, Bankiewicz KS, Baker SJ, Bergers G, Pieper RO (2006) The PTEN/Akt pathway dictates the direct alphaVbeta3-dependent growthinhibitory action of an active fragment of tumstatin in glioma cells in vitro and in vivo. Cancer Res 66:11331–11340 Lorger M, Krueger JS, O’Neal M, Staflin K, Felding-Habermann B (2009) Activation of tumor cell integrin alphavbeta3 controls angiogenesis and metastatic growth in the brain. Proc Natl Acad Sci U S A 106:10666–10671 Gingras MC, Roussel E, Bruner JM, Branch CD, Moser RP (1995) Comparison of cell adhesion molecule expression between glioblastoma multiforme and autologous normal brain tissue. J Neuroimmunol 57:143–153 Paulus W, Baur I, Schuppan D, Roggendorf W (1993) Characterization of integrin receptors in normal and neoplastic human brain. Am J Pathol 143:154–163 Baumann F, Leukel P, Doerfelt A, Beier CP, Dettmer K, Oefner PJ, Kastenberger M, Kreutz M, Nickl-Jockschat T, Bogdahn U et al (2009) Lactate promotes glioma migration by TGFbeta2-dependent regulation of matrix metalloproteinase-2. Neuro Oncol 11:368–380 Platten M, Wick W, Wild-Bode C, Aulwurm S, Dichgans J, Weller M (2000) Transforming growth factors beta(1) (TGF-beta(1)) and TGF-beta(2) promote glioma cell migration via Up-regulation of alpha(V)beta(3) integrin expression. Biochem Biophys Res Commun 268:607–611 Adachi Y, Lakka SS, Chandrasekar N, Yanamandra N, Gondi CS, Mohanam S, Dinh DH, Olivero WC, Gujrati M, Tamiya T et al (2001) Down-regulation of integrin alpha(v)beta(3) expression and integrin-mediated signaling in glioma cells by adenovirus-mediated transfer of antisense urokinase-type plasminogen activator receptor (uPAR) and sense p16 genes. J Biol Chem 276:47171–47177 Abdollahi A, Griggs DW, Zieher H, Roth A, Lipson KE, Saffrich R, Grone HJ, Hallahan DE, Reisfeld RA, Debus J et al (2005) Inhibition of alpha(v)beta3 integrin survival signaling enhances antiangiogenic and antitumor effects of radiotherapy. Clin Cancer Res 11:6270–6279 Cordes N, Hansmeier B, Beinke C, Meineke V, van Beuningen D (2003) Irradiation differentially affects substratum-dependent survival, adhesion, and invasion of glioblastoma cell lines. Br J Cancer 89:2122–2132 Wick W, Wick A, Schulz JB, Dichgans J, Rodemann HP, Weller M (2002) Prevention of irradiation-induced glioma cell invasion by temozolomide involves caspase 3 activity and cleavage of focal adhesion kinase. Cancer Res 62:1915–1919 Ding Q, Stewart J Jr, Olman MA, Klobe MR, Gladson CL (2003) The pattern of enhancement of Src kinase activity on platelet-derived growth factor stimulation of glioblastoma cells is affected by the integrin engaged. J Biol Chem 278:39882–39891 Skuli N, Monferran S, Delmas C, Favre G, Bonnet J, Toulas C, Cohen-Jonathan ME (2009) Alphavbeta3/alphavbeta5 integrins-FAK-RhoB: a novel pathway for hypoxia regulation in glioblastoma. Cancer Res 69:3308–3316 Perris R, Johansson S (1990) Inhibition of neural crest cell migration by aggregating chondroitin sulfate proteoglycans is mediated by their hyaluronan-binding region. Dev Biol 137:1–12 Snow DM, Lemmon V, Carrino DA, Caplan AI, Silver J (1990) Sulfated proteoglycans in astroglial barriers inhibit neurite outgrowth in vitro. Exp Neurol 109:111–130 Giese A, Westphal M (1996) Glioma invasion in the central nervous system. Neurosurgery 39:235–250 Fiorilli P, Partridge D, Staniszewska I, Wang JY, Grabacka M, So K, Marcinkiewicz C, Reiss K, Khalili K, Croul SE (2008) Integrins mediate adhesion of medulloblastoma cells to tenascin and activate pathways associated with survival and proliferation. Lab Invest 88:1143–1156 Gladson CL, Cheresh DA (1991) Glioblastoma expression of vitronectin and the alpha v beta 3 integrin. Adhesion mechanism for transformed glial cells. J Clin Invest 88:1924–1932

326

M. Lorger and B. Felding-Habermann

28. Gladson CL, Wilcox JN, Sanders L, Gillespie GY, Cheresh DA (1995) Cerebral microenvironment influences expression of the vitronectin gene in astrocytic tumors. J Cell Sci 108(Pt 3):947–956 29. Higuchi M, Ohnishi T, Arita N, Hiraga S, Hayakawa T (1993) Expression of tenascin in human gliomas: its relation to histological malignancy, tumor dedifferentiation and angiogenesis. Acta Neuropathol 85:481–487 30. Zagzag D, Friedlander DR, Miller DC, Dosik J, Cangiarella J, Kostianovsky M, Cohen H, Grumet M, Greco MA (1995) Tenascin expression in astrocytomas correlates with angiogenesis. Cancer Res 55:907–914 31. Ljubimova JY, Fugita M, Khazenzon NM, Das A, Pikul BB, Newman D, Sekiguchi K, Sorokin LM, Sasaki T, Black KL (2004) Association between laminin-8 and glial tumor grade, recurrence, and patient survival. Cancer 101:604–612 32. Ljubimova JY, Lakhter AJ, Loksh A, Yong WH, Riedinger MS, Miner JH, Sorokin LM, Ljubimov AV, Black KL (2001) Overexpression of alpha4 chain-containing laminins in human glial tumors identified by gene microarray analysis. Cancer Res 61:5601–5610 33. Ohnishi T, Hiraga S, Izumoto S, Matsumura H, Kanemura Y, Arita N, Hayakawa T (1998) Role of fibronectin-stimulated tumor cell migration in glioma invasion in vivo: clinical significance of fibronectin and fibronectin receptor expressed in human glioma tissues. Clin Exp Metastasis 16:729–741 34. Oz B, Karayel FA, Gazio NL, Ozlen F, Balci K (2000) The distribution of extracellular matrix proteins and CD44S expression in human astrocytomas. Pathol Oncol Res 6:118–124 35. Al-Shamy G, Sawaya R (2009) Management of brain metastases: the indispensable role of surgery. J Neurooncol 92:275–282 36. Santarelli JG, Sarkissian V, Hou LC, Veeravagu A, Tse V (2007) Molecular events of brain metastasis. Neurosurg Focus 22:E1 37. Kusters B, Westphal JR, Smits D, Ruiter DJ, Wesseling P, Keilholz U, de Waal RM (2001) The pattern of metastasis of human melanoma to the central nervous system is not influenced by integrin alpha(v)beta(3) expression. Int J Cancer 92:176–180 38. Lorger M, Felding-Habermann B (2010) Capturing changes in the brain microenvironment during initial steps of breast cancer brain metastasis. Am J Pathol 176(6):2958–2971 39. Carbonell WS, Ansorge O, Sibson N, Muschel R (2009) The vascular basement membrane as “soil” in brain metastasis. PLoS One 4:e5857 40. Yoshimasu T, Sakurai T, Oura S, Hirai I, Tanino H, Kokawa Y, Naito Y, Okamura Y, Ota I, Tani N et al (2004) Increased expression of integrin alpha3beta1 in highly brain metastatic subclone of a human non-small cell lung cancer cell line. Cancer Sci 95:142–148 41. Belot N, Rorive S, Doyen I, Lefranc F, Bruyneel E, Dedecker R, Micik S, Brotchi J, Decaestecker C, Salmon I et al (2001) Molecular characterization of cell substratum attachments in human glial tumors relates to prognostic features. Glia 36:375–390 42. Berens ME, Rief MD, Loo MA, Giese A (1994) The role of extracellular matrix in human astrocytoma migration and proliferation studied in a microliter scale assay. Clin Exp Metastasis 12:405–415 43. Giese A, Loo MA, Rief MD, Tran N, Berens ME (1995) Substrates for astrocytoma invasion. Neurosurgery 37:294–301 44. Giese A, Rief MD, Loo MA, Berens ME (1994) Determinants of human astrocytoma migration. Cancer Res 54:3897–3904 45. Merzak A, Koochekpour S, Pilkington GJ (1995) Adhesion of human glioma cell lines to fibronectin, laminin, vitronectin and collagen I is modulated by gangliosides in vitro. Cell Adhes Commun 3:27–43 46. Singh P, Carraher C, Schwarzbauer JE (2010) Assembly of fibronectin extracellular matrix. Annu Rev Cell Dev Biol 26:397–419 47. Giese A, Laube B, Zapf S, Mangold U, Westphal M (1998) Glioma cell adhesion and migration on human brain sections. Anticancer Res 18:2435–2447 48. Chintala SK, Sawaya R, Gokaslan ZL, Rao JS (1996) Modulation of matrix metalloprotease-2 and invasion in human glioma cells by alpha 3 beta 1 integrin. Cancer Lett 103:201–208

13

Integrin Signaling in Angiogenesis and Metastatic Cancer Progression in the Brain

327

49. Cordes N, Seidler J, Durzok R, Geinitz H, Brakebusch C (2006) beta1-integrin-mediated signaling essentially contributes to cell survival after radiation-induced genotoxic injury. Oncogene 25:1378–1390 50. Hu B, Jarzynka MJ, Guo P, Imanishi Y, Schlaepfer DD, Cheng SY (2006) Angiopoietin 2 induces glioma cell invasion by stimulating matrix metalloprotease 2 expression through the alphavbeta1 integrin and focal adhesion kinase signaling pathway. Cancer Res 66:775–783 51. Muir D, Johnson J, Rojiani M, Inglis BA, Rojiani A, Maria BL (1996) Assessment of lamininmediated glioma invasion in vitro and by glioma tumors engrafted within rat spinal cord. J Neurooncol 30:199–211 52. Paulus W, Baur I, Beutler AS, Reeves SA (1996) Diffuse brain invasion of glioma cells requires beta 1 integrins. Lab Invest 75:819–826 53. Paulus W, Tonn JC (1994) Basement membrane invasion of glioma cells mediated by integrin receptors. J Neurosurg 80:515–519 54. VanMeter TE, Rooprai HK, Kibble MM, Fillmore HL, Broaddus WC, Pilkington GJ (2001) The role of matrix metalloproteinase genes in glioma invasion: co-dependent and interactive proteolysis. J Neurooncol 53:213–235 55. Gladson CL (1996) Expression of integrin alpha v beta 3 in small blood vessels of glioblastoma tumors. J Neuropathol Exp Neurol 55:1143–1149 56. Lim M, Guccione S, Haddix T, Sims L, Cheshier S, Chu P, Vogel H, Harsh G (2005) alpha(v) beta(3) Integrin in central nervous system tumors. Hum Pathol 36:665–669 57. Schnell O, Krebs B, Wagner E, Romagna A, Beer AJ, Grau SJ, Thon N, Goetz C, Kretzschmar HA, Tonn JC et al (2008) Expression of integrin alphavbeta3 in gliomas correlates with tumor grade and is not restricted to tumor vasculature. Brain Pathol 18:378–386 58. Bello L, Lucini V, Carrabba G, Giussani C, Machluf M, Pluderi M, Nikas D, Zhang J, Tomei G, Villani RM et al (2001) Simultaneous inhibition of glioma angiogenesis, cell proliferation, and invasion by a naturally occurring fragment of human metalloproteinase-2. Cancer Res 61:8730–8736 59. Tonn JC, Wunderlich S, Kerkau S, Klein CE, Roosen K (1998) Invasive behaviour of human gliomas is mediated by interindividually different integrin patterns. Anticancer Res 18: 2599–2605 60. Brooks PC, Stromblad S, Sanders LC, von Schalscha TL, Aimes RT, Stetler-Stevenson WG, Quigley JP, Cheresh DA (1996) Localization of matrix metalloproteinase MMP-2 to the surface of invasive cells by interaction with integrin alpha v beta 3. Cell 85:683–693 61. Wei Y, Lukashev M, Simon DI, Bodary SC, Rosenberg S, Doyle MV, Chapman HA (1996) Regulation of integrin function by the urokinase receptor. Science 273:1551–1555 62. Holash J, Maisonpierre PC, Compton D, Boland P, Alexander CR, Zagzag D, Yancopoulos GD, Wiegand SJ (1999) Vessel cooption, regression, and growth in tumors mediated by angiopoietins and VEGF. Science 284:1994–1998 63. MacDonald TJ, Ladisch S (2001) Antisense to integrin alpha v inhibits growth and induces apoptosis in medulloblastoma cells. Anticancer Res 21:3785–3791 64. Taga T, Suzuki A, Gonzalez-Gomez I, Gilles FH, Stins M, Shimada H, Barsky L, Weinberg KI, Laug WE (2002) alpha v-Integrin antagonist EMD 121974 induces apoptosis in brain tumor cells growing on vitronectin and tenascin. Int J Cancer 98:690–697 65. Hodivala-Dilke K (2008) alphavbeta3 integrin and angiogenesis: a moody integrin in a changing environment. Curr Opin Cell Biol 20:514–519 66. Maeshima Y, Colorado PC, Torre A, Holthaus KA, Grunkemeyer JA, Ericksen MB, Hopfer H, Xiao Y, Stillman IE, Kalluri R (2000) Distinct antitumor properties of a type IV collagen domain derived from basement membrane. J Biol Chem 275:21340–21348 67. Chatterjee S, Matsumura A, Schradermeier J, Gillespie GY (2000) Human malignant glioma therapy using anti-alpha(v)beta3 integrin agents. J Neurooncol 46:135–144 68. Kanamori M, Kawaguchi T, Berger MS, Pieper RO (2006) Intracranial microenvironment reveals independent opposing functions of host alphaVbeta3 expression on glioma growth and angiogenesis. J Biol Chem 281:37256–37264

328

M. Lorger and B. Felding-Habermann

69. Kanamori M, Vanden Berg SR, Bergers G, Berger MS, Pieper RO (2004) Integrin beta3 overexpression suppresses tumor growth in a human model of gliomagenesis: implications for the role of beta3 overexpression in glioblastoma multiforme. Cancer Res 64:2751–2758 70. MacDonald TJ, Taga T, Shimada H, Tabrizi P, Zlokovic BV, Cheresh DA, Laug WE (2001) Preferential susceptibility of brain tumors to the antiangiogenic effects of an alpha(v) integrin antagonist. Neurosurgery 48:151–157 71. Riemenschneider MJ, Mueller W, Betensky RA, Mohapatra G, Louis DN (2005) In situ analysis of integrin and growth factor receptor signaling pathways in human glioblastomas suggests overlapping relationships with focal adhesion kinase activation. Am J Pathol 167:1379–1387 72. Tchaicha JH, Mobley AK, Hossain MG, Aldape KD, McCarty JH (2010) A mosaic mouse model of astrocytoma identifies alphavbeta8 integrin as a negative regulator of tumor angiogenesis. Oncogene 29:4460–4472 73. Cambier S, Gline S, Mu D, Collins R, Araya J, Dolganov G, Einheber S, Boudreau N, Nishimura SL (2005) Integrin alpha(v)beta8-mediated activation of transforming growth factor-beta by perivascular astrocytes: an angiogenic control switch. Am J Pathol 166:1883–1894 74. Mikkelsen T, Brodie C, Finniss S, Berens ME, Rennert JL, Nelson K, Lemke N, Brown SL, Hahn D, Neuteboom B et al (2009) Radiation sensitization of glioblastoma by cilengitide has unanticipated schedule-dependency. Int J Cancer 124:2719–2727 75. Monferran S, Skuli N, Delmas C, Favre G, Bonnet J, Cohen-Jonathan-Moyal E, Toulas C (2008) Alphavbeta3 and alphavbeta5 integrins control glioma cell response to ionising radiation through ILK and RhoB. Int J Cancer 123:357–364 76. Uhm JH, Dooley NP, Kyritsis AP, Rao JS, Gladson CL (1999) Vitronectin, a glioma-derived extracellular matrix protein, protects tumor cells from apoptotic death. Clin Cancer Res 5:1587–1594 77. Hsu AR, Veeravagu A, Cai W, Hou LC, Tse V, Chen X (2007) Integrin alpha v beta 3 antagonists for anti-angiogenic cancer treatment. Recent Pat Anticancer Drug Discov 2:143–158 78. Yamada S, Bu XY, Khankaldyyan V, Gonzales-Gomez I, McComb JG, Laug WE (2006) Effect of the angiogenesis inhibitor Cilengitide (EMD 121974) on glioblastoma growth in nude mice. Neurosurgery 59:1304–1312 79. Reardon DA, Fink KL, Mikkelsen T, Cloughesy TF, O’Neill A, Plotkin S, Glantz M, Ravin P, Raizer JJ, Rich KM et al (2008) Randomized phase II study of cilengitide, an integrin-targeting arginine-glycine-aspartic acid peptide, in recurrent glioblastoma multiforme. J Clin Oncol 26:5610–5617 80. Reardon DA, Nabors LB, Stupp R, Mikkelsen T (2008) Cilengitide: an integrin-targeting arginine-glycine-aspartic acid peptide with promising activity for glioblastoma multiforme. Expert Opin Investig Drugs 17:1225–1235 81. Tabatabai G, Weller M, Nabors B, Picard M, Reardon D, Mikkelsen T, Ruegg C, Stupp R (2010) Targeting integrins in malignant glioma. Target Oncol 5:175–181 82. Kita D, Takino T, Nakada M, Takahashi T, Yamashita J, Sato H (2001) Expression of dominant-negative form of Ets-1 suppresses fibronectin-stimulated cell adhesion and migration through down-regulation of integrin alpha5 expression in U251 glioma cell line. Cancer Res 61:7985–7991 83. Maglott A, Bartik P, Cosgun S, Klotz P, Ronde P, Fuhrmann G, Takeda K, Martin S, Dontenwill M (2006) The small alpha5beta1 integrin antagonist, SJ749, reduces proliferation and clonogenicity of human astrocytoma cells. Cancer Res 66:6002–6007 84. Martin S, Cosset EC, Terrand J, Maglott A, Takeda K, Dontenwill M (2009) Caveolin-1 regulates glioblastoma aggressiveness through the control of alpha(5)beta(1) integrin expression and modulates glioblastoma responsiveness to SJ749, an alpha(5)beta(1) integrin antagonist. Biochim Biophys Acta 1793:354–367 85. Martinkova E, Maglott A, Leger DY, Bonnet D, Stiborova M, Takeda K, Martin S, Dontenwill M (2010) alpha5beta1 integrin antagonists reduce chemotherapy-induced premature senescence and facilitate apoptosis in human glioblastoma cells. Int J Cancer 127:1240–1248

13

Integrin Signaling in Angiogenesis and Metastatic Cancer Progression in the Brain

329

86. Carter A (2010) Integrins as target: first phase III trial launches, but questions remain. J Natl Cancer Inst 102:675–677 87. Singh SK, Clarke ID, Terasaki M, Bonn VE, Hawkins C, Squire J, Dirks PB (2003) Identification of a cancer stem cell in human brain tumors. Cancer Res 63:5821–5828 88. Singh SK, Hawkins C, Clarke ID, Squire JA, Bayani J, Hide T, Henkelman RM, Cusimano MD, Dirks PB (2004) Identification of human brain tumour initiating cells. Nature 432:396–401 89. Blazek ER, Foutch JL, Maki G (2007) Daoy medulloblastoma cells that express CD133 are radioresistant relative to CD133- cells, and the CD133+ sector is enlarged by hypoxia. Int J Radiat Oncol Biol Phys 67:1–5 90. Dave B, Chang J (2009) Treatment resistance in stem cells and breast cancer. J Mammary Gland Biol Neoplasia 14:79–82 91. Ischenko I, Seeliger H, Schaffer M, Jauch KW, Bruns CJ (2008) Cancer stem cells: how can we target them? Curr Med Chem 15:3171–3184 92. Lathia JD, Gallagher J, Heddleston JM, Wang J, Eyler CE, Macswords J, Wu Q, Vasanji A, McLendon RE, Hjelmeland AB et al (2010) Integrin alpha 6 regulates glioblastoma stem cells. Cell Stem Cell 6:421–432 93. Calabrese C, Poppleton H, Kocak M, Hogg TL, Fuller C, Hamner B, Oh EY, Gaber MW, Finklestein D, Allen M et al (2007) A perivascular niche for brain tumor stem cells. Cancer Cell 11:69–82

Chapter 14

Tumor and Host Determinants of Pulmonary Metastasis in Bladder Cancer Neveen Said and Dan Theodorescu

Abstract Despite the recent advances in the diagnosis of bladder cancer, recurrence after surgical intervention for muscle invasive disease is still problematic as nearly half of patients harbor occult distant metastases. Clinical data from human disease revealed that, invasive and metastatic bladder cancer cells can metastasize to lungs, and this in turn is associated with poor 5-year survival rate. Experimental rodent models of carcinogenesis and metastasis are available to study this phenomenon. Comparative gene expression profiling, proteomic and computational studies identified an intertwined network of metastasis promoters and suppressors that modulate the interactions between the components of the pulmonary milieu and cancer cells inflammatory mediators, ECM molecules, as well as peptide hormones. In this chapter we provide select exemplar of some of the molecular mechanisms underlying lung colonization by bladder cancer.

14.1

Introduction

Urothelial cancer (UC) is the most common malignancy affecting the urinary system. When it affects the bladder, it leads to an estimated 70,530 new cases in the United States, with a male to female ratio of 3:1 and approximately 14,680 deaths, expected in 2010 [1]. UC arises from the mucosal lining and is frequently multifocal.

N. Said Department of Urology, Paul Mellon Urologic Cancer Institute, University of Virginia, Charlottesville, VA 22908, USA e-mail: [email protected] D. Theodorescu (*) Departments of Surgery and Pharmacology, University of Colorado Comprehensive Cancer Center, Aurora, CO 80045, USA e-mail: [email protected]

A. Fatatis (ed.), Signaling Pathways and Molecular Mediators in Metastasis, DOI 10.1007/978-94-007-2558-4_14, © Springer Science+Business Media B.V. 2012

331

332

N. Said and D. Theodorescu

Numerous factors, including chromosomal markers, genetic polymorphisms, and genetic and epigenetic alterations may be involved in tumorigenesis, progression and metastasis. At initial presentation, 70% of patients with UC present with nonmuscle invasive (formerly known as “superficial”), and 30% present with muscle invasive disease [2]. Despite a good prognosis for patients with the former, recurrence is common and is associated with development of muscle invasive disease in up to 30%. Also, 50% of patients presenting with muscle invasive UC develop distant metastases leading to poor 5-year survival rate [1]. Clinical data from human disease as well as experimental rodent models of carcinogenesis and metastasis reveals that when it occurs, metastasis of this tumor is commonly found in the regional lymph node metastasis and the lungs. Given that metastases are responsible for most of the deaths from this disease [3, 4], understanding of the process will aid in the development of new approaches for treatment.

14.2

Multistep Process of Metastasis

The high incidence of pulmonary metastases in cancer patients was initially believed to be a random process based on blood flow predominance as lungs receive significant cardiac output [5, 6]. This was supported by the observation that metastases often initiate in pulmonary arterioles and later traverse the basement membrane into the lung parenchyma [6, 7]. However, growing body of literature supports the patterns of metastases is also a consequence of the “seed and soil” theory put forth by Stephen Paget in 1889 [8]. Hence, the development of lung metastasis is likely an active highly selective process instigated by tumor cells, and is strongly influenced by the interactions between tumor and host cells, and by both the immediate and extended tumor microenvironments [3, 7–14].

14.2.1

The Lung Microenvironment as a Host for Metastatic Tumor Cells

Barriers created by the lung microenvironment include physical barriers such as vascular endothelium, extracellular matrix (ECM) components and basement membranes, as well as physiologic barriers such as limited oxygen (hypoxia) and nutrients, and immunologic barriers by the immune system [9, 15]. The vascular endothelium is a critical regulator of pulmonary function and its phenotype is modulated by myriad stimuli such as proinflammatory cytokines and hemodynamic forces thereby impacting the development of vascular disease states [15]. Metastatic progression also depends on cancer cell secretome as both the growing primary tumor and shed tumor cells in the circulation (Fig. 14.1), secrete

14

Tumor and Host Determinants of Pulmonary Metastasis in Bladder Cancer

333

Primary tumor Intravasation Circulating Tumor Cells

Cancer Secretome

+ +

M

Migration/chemotaxis

Cancer Secretome M secretome

+ Colonization

Premetastatic

Metastatic

Fig. 14.1 A view of the multistep process of metastasis: Disseminated cancer cells in the circulation arise from highly invasive cell populations in the primary tumor that invade the blood vessel wall, intravasate and circulate in the blood stream. Secreted tumor factors “cancer secretome” form invasive cancer cells in the primary tumors and circulating tumor cells recruit, bone marrowderived myeloid cells, to the prospective metastatic site “lungs”. In turn, myeloid cells acquire an inflammatory macrophage phenotype, secreting factors “macrophage secretome” and contribute to the construction of a pre-metastatic niche favorable for implantation and growth of disseminating tumor cells forming micro- and macro-metastases

inflammatory mediators that triggers vascular inflammation in the lungs that is characterized by recruitment of bone marrow derived myeloid cells, margination, extravasation, and activation of circulating mononuclear cells into vessel walls to lung parenchyma [10, 12–14, 16–19]. This multistep process is the result of local and systemic secretion of chemotactic factors, macrophage-activating cytokines, endothelin-1, reactive oxygen species (ROS), growth factors, upregulation of vascular adhesion molecules, and production of MMPs. The buildup of an inflammatory microenvironment, generates a premetastatic niche with progressive pulmonary vascular inflammation hospitable for metastatic growth [10, 12–14, 18, 19]. The activation of endothelial cells (ECs) is a crucial step in pulmonary vascular inflammation, and can result in subsequent increase in surface expression of cell adhesion molecules, such as vascular cell adhesion molecule (VCAM)-1, intercellular adhesion molecule (ICAM)-1, and selectins, which contribute to the recruitment of inflammatory cells to ECs and their transmigration across the vascular wall [18, 20]. In turn, activated ECs secrete cytokines and chemokines such

334

N. Said and D. Theodorescu

as monocyte chemotactic protein 1 (MCP-1), which is a potent inducer for monocyte attachment to ECs and migration into subendothelial space [21–24]. Mechanisms leading to proinflammatory cell recruitment to the lung are exploited by circulating cancer cells to extravasate from circulation [20]. These mechanisms include common cell adhesion molecule interactions as well as the expression of inflammatory cytokines/chemokines leading to recruitment and activation of macrophages. Persistent activation of ECs and macrophages, with reciprocal persistent production of cytokines and ROS cause local chronic inflammation that promotes increased homing of tumor cells to the lungs [21–24] as well as activation of non-tumor cells in tumor lesions stimulating tumor angiogenesis, invasiveness and intravasation/ extravasation. Interestingly, a strong body of literature highlighted the central importance of nuclear factor-kB (NF-kB) in the activation of and cross-talk between all the cell types involved [22–24]. The premetastatic niche may also be responsible for metastases to specific organs. The key tumor-secreted factors that determine metastatic sites and mediate premetastatic niche formation have yet to be identified, although the roles for cytokines/chemokines, MMPs and growth factors’ signaling in pulmonary endothelial cells and macrophages have been reported [25, 26]. These make the lung microenvironment more receptive to cancers cells. We have recently reported the roles of endothelin-1 in as an important mediator of lung inflammation and colonization of bladder cancer cells (discussed below) [14]. Elevated fibronectin expression by fibroblasts and fibroblast-like cells resident at premetastatic sites seems to be an important factor in the development of the premetastatic niche [12]. In addition, the secretion of versican by tumor cells and subsequently by inflammatory macrophages has been recently found to play a role in lung metastasis [10, 16, 27]. Overexpression of tumor-derived immunosuppressive factors such as TGFb, VEGF, IL-6, MCP1/CCL2, MMPs and Cox-2/prostanoids in the inflammatory premetastatic niche dampens the tumor-suppressing activities of cells responsible of immune surveillance as natural killer cells and antigen-presenting cells and supports an inflammatory phenotype of tumor-associated macrophages, TAM; promote tumor progression by secreting growth factors and cytokines that stimulate tumor metastases [16, 17, 28]. TAMs can facilitate tumor formation and progression by activation of NF-kB and AP-1 [16, 17, 29], and are especially attracted to regions of hypoxia, where they secrete angiogenic inducers and proteases, and express high levels of the HIF-1 and -2 transcription factors [18, 30–32] that further augment macrophage recruitment and activation as well as tumor cell invasiveness. Furthermore, TAMs can release growth factors such as PDGF, EGF, cytokines/ chemokines, and MMPs which enhance proliferation, survival, and invasion of tumor cells as well as secrete and induce secretion of ECM proteoglycans such as versican from tumor cells and fibroblasts in the lung milieu further augmenting invasiveness, inflammation and metastasis [10, 23, 33–38]. Our laboratory has identified two major molecules implicated in bladder cancer metastasis to the lungs, the metastasis suppressor RhoGDI2 and the pro-metastatic Ral A/B.

14

Tumor and Host Determinants of Pulmonary Metastasis in Bladder Cancer

14.2.2

335

The Metastasis Suppressor RhoGDI2

Rho family GDP Dissociation Inhibitor 2 (RhoGDI2) protein has been identified as functional metastasis suppressor during studies of the differential invasive and metastatic properties of isogenic human bladder carcinoma cell lines, T24 (nonmetastatic), and T24T (highly invasive and metastatic) using experimental metastasis models and comparative genomic studies [39]. RhoGDI2 was also found to be a prognostic marker in patients after cystectomy that its reduced expression was associated with decreased patient survival. However, approximately 35% of patients with moderate or high levels of RhoGDI2 protein-developed metastatic disease suggesting that not only the expression level, but other mechanisms might regulate the metastasis suppressor effect of RhoGDI2 [40]. Phosphorylation, binding to specific partners, truncations, proteolytic cleavage, or change in subcellular localization was considered given the existing data on the other members of the RhoGDI family (reviewed in [41]). Comparative gene expression profiling of human bladder cancer tissues and cell lines identified that the down-stream effectors downregulated by RhoGDI2 are molecules that have long been implicated in pulmonary vascular diseases: Endothelin-1 (ET-1) [42] and versican [14, 27, 43].

14.2.3

Endothelin Axis

ET-1, an endothelial cell-derived vasoconstrictor peptide is an important member of the endothelin axis with myriad developmental, physiological and pathological functions [29, 44–46]. The “Endothelin Axis” consists of three similar small peptides, ET-1, ET-2 and ET-3, two G-protein-coupled receptors, ETAR and ETBR, and two membrane-bound proteases, the ET-converting enzymes, ECE-1 and ECE-2 [44], that activate the secreted pro forms of the peptide. ET-1 production is stimulated by a variety of cytokines and growth factors, hypoxia, and shear stress, while ETAR activation triggers signaling networks involved in cell proliferation, new vessel formation, invasion, inflammation, and metastatic spread [44, 47–49]. In benign and malignant diseases, ET-1 has been shown to activate the pro-inflammatory transcriptional factors AP-1 and NF-kB in human monocytes and cancer cells (Fig. 14.2) and to stimulate the production of inflammatory cytokines IL-6, MCP-1 and Cox2, as well as MMP activity, the key orchestrators of inflammation-mediated cancer cell invasiveness and metastasis [50–53]. Recently, gene expression data and immunostaining of from human bladder cancer has revealed that ET-1 and ETAR are overexpressed in muscle invasive disease, and that the expression level is associated with reduced patient survival [14]. Experimental and spontaneous metastasis models revealed that tumor ET-1 triggers inflammation in the lungs wherein inflammatory cells, primarily macrophages, enhance and facilitate the process of metastatic colonization [14]. The magnitude of this early inflammatory response determines the kinetics of subsequent development of lung metastases. This inflammatory response is likely triggered by tumor secretion of ET-1 and the pro-inflammatory

336

N. Said and D. Theodorescu

ETAR Tumor cell

ETAR

ET-1 ET-1

M

ETBR

Migration/chemotaxis

Cox-2 Survival Invasion

ET-1, Proinvasive/ inflammatory cytokines IL-6, MCP-1, VEGF

PGE-2

ET-1, IL-6, MCP-1, MMPs, Cox2

ET1-inducing cytokines and chemokines EGF, IL-6, MCP-1, VEGF, TGFβ

Chemotaxis/ invasion MMPs

Inflammatory TME Tumor cell colonization, survival, proliferation, invasion, angiogenesis, metastasis

Metastatic colonization

Premetastatic Phase Primary/circulating tumor cells

macrometastases

ET-1

ETAR

Primary/circulating Tumor cell

micrometastases

ET-1 ET-2 VEGF MCP-1 VEGFR1 ETBR CCR2

M M

M

M

ET-1, EGF, IL-6, VEGF, MCP-1, Cox2, PGE2

Inflammatory M

ETAR

Recruitment of BDMCs to lung

Bone marrow-derived myeloid cells (BDMCs)

phenotype

Fig. 14.2 Multiple pathways of endothelin involvement in lung metastasis: (a) Binding of ET-1 to the ETA receptor in the plasma membrane triggers signal-transduction pathways in both cancer cells and macrophages that converge in activation of NF-kB and AP-1 transcription factors. (b) Schematic illustration of the roles of ET-1 in bladder cancer metastasis to the lungs

14

Tumor and Host Determinants of Pulmonary Metastasis in Bladder Cancer

Ig domain

Tandem repeats

EGF repeats

αGAG

CRD

337

CBP

βGAG

CS GAG Chains Signal peptide

G1 Hyaluronan Binding

CD44 L-, P-selectin LDL Glycolipids Chemokines

Fibulin-1 Fibrillin-1

Tenascins

Fibronectin

G3 β1 integrin

Fig. 14.3 Structure of versican and its interactions: The N-terminal, G1 domain is composed of an immunoglobulin (Ig)-like motif, followed by two proteoglycan tandem repeats that bind to hyaluronan. The C-terminal, G3 domain, consists of two epidermal growth factor (EGF)-like repeats, a carbohydrate recognition domain (CRD), and a complement binding protein (CBP). Chondroitin sulfate (CS) chain binding regions exist between G1 and G3 consist of a protein core and one or two glycosaminoglycans (a and b GAG) that vary amongst tissues and specific molecules. Different domains of versican bind to and interact with a wide variety of molecules, such as growth factors (TGFb, PDGF-BB), cytokines, chemokines (MCP-1), adhesion/ECM molecules (selectins, HA, CD44, fibrinogen/fibrin, fibronectin, thrombospondin, collagens), MMPs and LDL

cytokines IL-6 and MCP-1. Interestingly, targeting ET-1 gene in tumor cells or pharmacologic inhibition of ETAR prior to injection of tumor cells reduced the early inflammatory response, both tumor and host MCP-1/CCL2, IL-6 and Cox-2 in the lung microenvironment and subsequent metastatic colonization. In contrast, when the early inflammatory response was allowed to develop upon injection of tumor cells, then later followed by ETAR blockade, the reduction of lung inflammation and clinical lung metastases was not as dramatic. Depletion of macrophages also significantly reduced the inflammatory response and the ensuing lung metastases.

14.2.4

Versican

Versican, a chondroitin sulfate proteoglycan, is a major structural component of the ECM (Fig. 14.3). Differential RNA splicing gives rise to four isoforms of versican (V0, V1, V2, and V3), which vary by the presence or absence of two glycosaminoglycan (GAG) binding domains named aGAG and bGAG. The various domains of versican form highly hydrated complexes that trap cytokines, enzymes, growth factors, lipoproteins, other extracellular matrix molecules, and signaling receptors. It is implicated in the pathogenesis of pulmonary vascular diseases and atherosclerosis, causing an ECM expansion, fostering pro-inflammatory cytokines and cells and dramatically influencing the inflammatory phenotype of cellular components of these diseases [54].

338

N. Said and D. Theodorescu

The association between increased levels of versican and progression of cancer to disseminated disease suggests that versican is important in promoting cancer cell proliferation, motility, invasion and metastasis [10, 16, 23, 33–38, 55–65]. More recent studies [10] identified cancer cell-secreted versican, as a macrophage activator that acts through TLR2/6 inducing TNF-alpha secretion by myeloid cells, generating an inflammatory microenvironment hospitable for metastatic growth. In vitro functional studies, revealed versican is a key mediator of cancer cellmacrophages cross-talk and significantly inhibited their adhesion to and invasion of ECM molecules as well as pulmonary microvascular endothelial cells [27, 66]. Interestingly, comparative gene expression profiling of invasive/metastatic bladder cancer cells with clinical bladder cancer specimens identified versican as the top upregulated gene, positively correlated with bladder cancer stage and poor prognosis, and in vitro with invasiveness of bladder cancer cell lines [27, 43]. It remains to be seen if genetic manipulation of versican expression in bladder cancer cells, either by stable overexpression or knockdown demonstrates a role of tumor versican in metastasis.

14.2.5

Ral GTPases

The Ras-like family of monomeric G proteins (Ral GTPases), RalA and RalB are paralogs (85% amino acid identity) that participate in diverse cellular functions [67] (Fig. 14.4). These signaling proteins, regulated by multiple pathways including Ras (reviewed in [68]), have been shown to be necessary for cellular phenotypes associated with cancer progression and metastasis. Phosphorylation and subsequent activation, of RalA and RalB is paralog-specific as they are phosphorylated at specific conserved sites by different kinases [69]. PKC phosphorylation of RalB at S198 in bladder cancer cell lines has been recently reported to be necessary for cytoskeletal organization, anchorage-independent growth, and cell migration in vitro and for subcutaneous tumor growth and lung metastasis in vivo [69]. In addition, Ral effector, RalBP1 was found to be highly expressed in bladder cancer patients and necessary for metastasis of human bladder cancer cell lines [70]. In addition RalA and/or RalB have also been shown to regulate transcription factors such as NF-kB and AP-1 [71, 72] and play a role in the regulation of expression of key molecules, including the prognostic marker and cell surface GPI-linked glycoprotein, CD24 [68].

14.2.6

CD24

CD24, is a glycosyl phosphatidyl inositol–linked surface protein [73], that has been identified as a downstream target of Ral signaling by profiling the expression of RalA/B–depleted bladder carcinoma cells [67, 68]. Originally identified as a

14

Tumor and Host Determinants of Pulmonary Metastasis in Bladder Cancer

339

PLASMA MEMBRANE Effectors:

GTP

• RalBP1 • PLD1 • Others

Activators • Oxidative Stress/Hypoxia • Serum Starvation • Heat shock • Intracellular Ca2+ • Androgen withdrawal • Ras-activated GEFs • Ras-independent GEFs

GDP

Ral

• Migration/Metastatic Phenotype: CD24 • Transcription: NFκB, AP1 • Stress response • Vesicle trafficking

Fig. 14.4 The regulation of Ral and downstream signaling: RalGTPases are activated by a variety of stimuli, including guanidine nucleotide exchange factors (RalGEFs), Ras or independent of Ras activation. Several other stimuli, known to be active in cancer cells, have been identified to activate Ral, including oxidative damage, hypoxia, calcium signaling, and androgen withdrawal. Once activated by exchange of GTP for GDP, Ral GTPases are able to associate physically with several known (and likely other uncharacterized) effector proteins, to mediate diverse cellular effects

surface differentiation marker of peripheral B-lymphocytes [73], CD24 received recent attention as a marker of stem cell populations of a variety of cancers [74–87]. CD24 overexpression has been associated with lung metastasis [88] as it promoted cancer cell rolling and adhesion to P-selectin in murine lung vasculature and platelets [76, 78–80, 85, 89–95]. Consistently, CD24 has been recently reported to be a hypoxia-regulated gene in some cancer cell lines [96, 97] that was downregulated by Cox-2 inhibitor, Celecoxib, suggesting role in inflammation associated with metastasis [98–100]. CD24 is highly expressed in urothelial as well as other cancers. Immunohistological localization of CD24 revealed high levels of expression correlated with the stromal/muscle invasion, stage, grade and shorter patient disease-free survival [67, 101]. Depletion of CD24 function in bladder cancer cell lines was found to be associated with decreased cell proliferation and anchorage-independent growth, changes in the actin cytoskeleton, and induction of apoptosis [67]. In addition, monoclonal antibodies against CD24 that were efficacious in inhibiting in vitro and in vivo growth of colorectal and pancreatic cancer cells as well as sensitizing them to conventional adjuvant chemotherapeutics [98– 100, 102–104], significantly reduced the number of lung metastases that developed after tail vein injection of the metastatic UMUC3 cell line in an experimental model of bladder cancer [105]. However, the exact mechanism of CD24 involvement in lung metastasis is still elusive.

340

N. Said and D. Theodorescu

References 1. Jemal A, Siegel R, Xu J, Ward E (2010) Cancer statistics, 2010. CA Cancer J Clin 60:277–300 2. Wu XR (2005) Urothelial tumorigenesis: a tale of divergent pathways. Nat Rev Cancer 5:713–725 3. Rocken M (2010) Early tumor dissemination, but late metastasis: insights into tumor dormancy. J Clin Invest 120:1800–1803 4. Steeg PS, Theodorescu D (2008) Metastasis: a therapeutic target for cancer. Nat Clin Pract Oncol 5:206–219 5. Ewing J (1928) A treatise on tumors. In: Ewing J (ed) Neoplastic disease, 3rd edn. WB Saunders, Philadelphia, pp 77–89 6. Zetter BR (1990) The cellular basis of site-specific tumor metastasis. N Engl J Med 322:605–612 7. Luo JL, Maeda S, Hsu LC, Yagita H, Karin M (2004) Inhibition of NF-kappaB in cancer cells converts inflammation- induced tumor growth mediated by TNFalpha to TRAIL-mediated tumor regression. Cancer Cell 6:297–305 8. Paget S (1989) The distribution of secondary growths in cancer of the breast. 1889. Cancer Metastasis Rev 8:98–101 9. Gupta GP, Massague J (2006) Cancer metastasis: building a framework. Cell 127:679–695 10. Kim S, Takahashi H, Lin WW, Descargues P, Grivennikov S, Kim Y et al (2009) Carcinomaproduced factors activate myeloid cells through TLR2 to stimulate metastasis. Nature 457: 102–106 11. Luo JL, Tan W, Ricono JM, Korchynskyi O, Zhang M, Gonias SL et al (2007) Nuclear cytokine-activated IKKalpha controls prostate cancer metastasis by repressing Maspin. Nature 446:690–694 12. Kaplan RN, Rafii S, Lyden D (2006) Preparing the “soil”: the premetastatic niche. Cancer Res 66:11089–11093 13. Kaplan RN, Riba RD, Zacharoulis S, Bramley AH, Vincent L, Costa C et al (2005) VEGFR1positive haematopoietic bone marrow progenitors initiate the pre-metastatic niche. Nature 438:820–827 14. Said N, Smith SC, Sanchez-Carbayo M, Theodorescu D (2011) Tumor Endothelin-1 contributes to macrophage infiltration and metastatic colonization of murine lungs by human cancer cells. J Clin Invest 121(1):132–147 15. SenBanerjee S, Lin Z, Atkins GB, Greif DM, Rao RM, Kumar A et al (2004) KLF2 is a novel transcriptional regulator of endothelial proinflammatory activation. J Exp Med 199:1305–1315 16. Mantovani A (2009) Cancer: inflaming metastasis. Nature 457:36–37 17. Bidard FC, Pierga JY, Vincent-Salomon A, Poupon MF (2008) A “class action” against the microenvironment: do cancer cells cooperate in metastasis? Cancer Metastasis Rev 27:5–10 18. Tieu BC, Lee C, Sun H, Lejeune W, Recinos A 3rd, Ju X et al (2009) An adventitial IL-6/ MCP1 amplification loop accelerates macrophage-mediated vascular inflammation leading to aortic dissection in mice. J Clin Invest 119:3637–3651 19. Qian B, Deng Y, Im JH, Muschel RJ, Zou Y, Li J et al (2009) A distinct macrophage population mediates metastatic breast cancer cell extravasation, establishment and growth. PLoS One 4:e6562 20. Taranova AG, Maldonado D 3rd, Vachon CM, Jacobsen EA, Abdala-Valencia H, McGarry MP et al (2008) Allergic pulmonary inflammation promotes the recruitment of circulating tumor cells to the lung. Cancer Res 68:8582–8589 21. Qi Y, Liang J, She ZG, Cai Y, Wang J, Lei T et al (2010) MCP-induced protein 1 suppresses TNFalpha-induced VCAM-1 expression in human endothelial cells. FEBS Lett 584(14): 3065–3072

14

Tumor and Host Determinants of Pulmonary Metastasis in Bladder Cancer

341

22. Rollins BJ (1996) Monocyte chemoattractant protein 1: a potential regulator of monocyte recruitment in inflammatory disease. Mol Med Today 2:198–204 23. Hess S, Methe H, Kim JO, Edelman ER (2009) NF-kappaB activity in endothelial cells is modulated by cell substratum interactions and influences chemokine-mediated adhesion of natural killer cells. Cell Transplant 18:261–273 24. Collins T, Cybulsky MI (2001) NF-kappaB: pivotal mediator or innocent bystander in atherogenesis? J Clin Invest 107:255–264 25. Hiratsuka S, Nakamura K, Iwai S, Murakami M, Itoh T, Kijima H et al (2002) MMP9 induction by vascular endothelial growth factor receptor-1 is involved in lung-specific metastasis. Cancer Cell 2:289–300 26. Hiratsuka S, Watanabe A, Aburatani H, Maru Y (2006) Tumour-mediated upregulation of chemoattractants and recruitment of myeloid cells predetermines lung metastasis. Nat Cell Biol 8:1369–1375 27. Said N, and Theodorescu, D. Role of Versican in Bladder Cancer Metastasis to Lungs. AACRJoint Tumor Microenvironment and Metastasis Research Society Meeting; 2010 September, 2010; Philadelphia, PA; 2010. 28. Joyce JA, Pollard JW (2009) Microenvironmental regulation of metastasis. Nat Rev Cancer 9:239–252 29. Hagemann T, Balkwill F, Lawrence T (2007) Inflammation and cancer: a double-edged sword. Cancer Cell 12:300–301 30. Corzo CA, Condamine T, Lu L, Cotter MJ, Youn JI, Cheng P et al (2010) HIF-1alpha regulates function and differentiation of myeloid-derived suppressor cells in the tumor microenvironment. J Exp Med 207:2439–2453 31. Imtiyaz HZ, Williams EP, Hickey MM, Patel SA, Durham AC, Yuan LJ et al (2010) Hypoxiainducible factor 2alpha regulates macrophage function in mouse models of acute and tumor inflammation. J Clin Invest 120:2699–2714 32. Doedens AL, Stockmann C, Rubinstein MP, Liao D, Zhang N, DeNardo DG et al (2010) Macrophage expression of hypoxia-inducible factor-1 alpha suppresses T-cell function and promotes tumor progression. Cancer Res 70:7465–7475 33. Ricciardelli C, Sakko AJ, Ween MP, Russell DL, Horsfall DJ (2009) The biological role and regulation of versican levels in cancer. Cancer Metastasis Rev 28:233–245 34. Lee JH, Kim SH, Lee ES, Kim YS (2009) CD24 overexpression in cancer development and progression: a meta-analysis. Oncol Rep 22:1149–1156 35. Yee AJ, Akens M, Yang BL, Finkelstein J, Zheng PS, Deng Z et al (2007) The effect of versican G3 domain on local breast cancer invasiveness and bony metastasis. Breast Cancer Res 9:R47 36. Wang W, Xu GL, Jia WD, Ma JL, Li JS, Ge YS et al (2009) Ligation of TLR2 by versican: a link between inflammation and metastasis. Arch Med Res 40:321–323 37. Chiodoni C, Colombo MP, Sangaletti S (2010) Matricellular proteins: from homeostasis to inflammation, cancer, and metastasis. Cancer Metastasis Rev 29:295–307 38. Folberg R, Arbieva Z, Moses J, Hayee A, Sandal T, Kadkol S et al (2006) Tumor cell plasticity in uveal melanoma: microenvironment directed dampening of the invasive and metastatic genotype and phenotype accompanies the generation of vasculogenic mimicry patterns. Am J Pathol 169:1376–1389 39. Gildea JJ, Seraj MJ, Oxford G, Harding MA, Hampton GM, Moskaluk CA et al (2002) RhoGDI2 is an invasion and metastasis suppressor gene in human cancer. Cancer Res 62:6418–6423 40. Theodorescu D, Sapinoso LM, Conaway MR, Oxford G, Hampton GM, Frierson HF Jr (2004) Reduced expression of metastasis suppressor RhoGDI2 is associated with decreased survival for patients with bladder cancer. Clin Cancer Res 10:3800–3806 41. Said N, Theodorescu D (2009) Pathways of metastasis suppression in bladder cancer. Cancer Metastasis Rev 28:327–333 42. Titus B, Frierson HF Jr, Conaway M, Ching K, Guise T, Chirgwin J et al (2005) Endothelin axis is a target of the lung metastasis suppressor gene RhoGDI2. Cancer Res 65:7320–7327

342

N. Said and D. Theodorescu

43. Wu Y, Siadaty MS, Berens ME, Hampton GM, Theodorescu D (2008) Overlapping gene expression profiles of cell migration and tumor invasion in human bladder cancer identify metallothionein 1E and nicotinamide N-methyltransferase as novel regulators of cell migration. Oncogene 27:6679–6689 44. Kedzierski RM, Yanagisawa M (2001) ENDOTHELIN SYSTEM: the double-edged sword in health and disease. Annu Rev Pharmacol Toxicol 41:851–876 45. Herrmann E, Tiemann A, Eltze E, Bolenz C, Bremer C, Persigehl T et al (2009) EndothelinA-receptor antagonism with atrasentan exhibits limited activity on the KU-19-19 bladder cancer cell line in a mouse model. J Cancer Res Clin Oncol 135:1455–1462 46. Herrmann E, Bogemann M, Bierer S, Eltze E, Toma MI, Kopke T et al (2007) The role of the endothelin axis and microvessel density in bladder cancer – correlation with tumor angiogenesis and clinical prognosis. Oncol Rep 18:133–138 47. Bagnato A, Spinella F, Rosano L (2005) Emerging role of the endothelin axis in ovarian tumor progression. Endocr Relat Cancer 12:761–772 48. Giaid A, Hamid QA, Springall DR, Yanagisawa M, Shinmi O, Sawamura T et al (1990) Detection of endothelin immunoreactivity and mRNA in pulmonary tumours. J Pathol 162:15–22 49. Rosano L, Spinella F, Di Castro V, Nicotra MR, Dedhar S, de Herreros AG et al (2005) Endothelin-1 promotes epithelial-to-mesenchymal transition in human ovarian cancer cells. Cancer Res 65:11649–11657 50. Kandalaft LE, Facciabene A, Buckanovich RJ, Coukos G (2009) Endothelin B receptor, a new target in cancer immune therapy. Clin Cancer Res 15(14):4521–4528 51. Sutcliffe AM, Clarke DL, Bradbury DA, Corbett LM, Patel JA, Knox AJ (2009) Transcriptional regulation of monocyte chemotactic protein-1 release by endothelin-1 in human airway smooth muscle cells involves NF-kappaB and AP-1. Br J Pharmacol 157:436–450 52. Spinella F, Rosano L, Di Castro V, Natali PG, Bagnato A (2004) Endothelin-1-induced prostaglandin E2-EP2, EP4 signaling regulates vascular endothelial growth factor production and ovarian carcinoma cell invasion. J Biol Chem 279:46700–46705 53. Spinella F, Rosano L, Di Castro V, Nicotra MR, Natali PG, Bagnato A (2004) Inhibition of cyclooxygenase-1 and -2 expression by targeting the endothelin a receptor in human ovarian carcinoma cells. Clin Cancer Res 10:4670–4679 54. Wight TN (2002) Versican: a versatile extracellular matrix proteoglycan in cell biology. Curr Opin Cell Biol 14:617–623 55. Domenzain C, Docampo MJ, Serra M, Miquel L, Bassols A (2003) Differential expression of versican isoforms is a component of the human melanoma cell differentiation process. Biochim Biophys Acta 1642:107–114 56. Ricciardelli C, Russell DL, Ween MP, Mayne K, Suwiwat S, Byers S et al (2007) Formation of hyaluronan- and versican-rich pericellular matrix by prostate cancer cells promotes cell motility. J Biol Chem 282:10814–10825 57. Lin HM, Chatterjee A, Lin YH, Anjomshoaa A, Fukuzawa R, McCall JL et al (2007) Genome wide expression profiling identifies genes associated with colorectal liver metastasis. Oncol Rep 17:1541–1549 58. Skandalis SS, Kletsas D, Kyriakopoulou D, Stavropoulos M, Theocharis DA (2006) The greatly increased amounts of accumulated versican and decorin with specific post-translational modifications may be closely associated with the malignant phenotype of pancreatic cancer. Biochim Biophys Acta 1760:1217–1225 59. Hirakawa S, Hong YK, Harvey N, Schacht V, Matsuda K, Libermann T et al (2003) Identification of vascular lineage-specific genes by transcriptional profiling of isolated blood vascular and lymphatic endothelial cells. Am J Pathol 162:575–586 60. Malemud CJ (2006) Matrix metalloproteinases (MMPs) in health and disease: an overview. Front Biosci 11:1696–1701 61. Ricciardelli C, Brooks JH, Suwiwat S, Sakko AJ, Mayne K, Raymond WA et al (2002) Regulation of stromal versican expression by breast cancer cells and importance to relapsefree survival in patients with node-negative primary breast cancer. Clin Cancer Res 8:1054–1060

14

Tumor and Host Determinants of Pulmonary Metastasis in Bladder Cancer

343

62. Domenzain-Reyna C, Hernandez D, Miquel-Serra L, Docampo MJ, Badenas C, Fabra A et al (2009) Structure and regulation of the versican promoter: the versican promoter is regulated by AP-1 and TCF transcription factors in invasive human melanoma cells. J Biol Chem 284:12306–12317 63. Brown LF, Guidi AJ, Schnitt SJ, Van De Water L, Iruela-Arispe ML, Yeo TK et al (1999) Vascular stroma formation in carcinoma in situ, invasive carcinoma, and metastatic carcinoma of the breast. Clin Cancer Res 5:1041–1056 64. Kischel P, Waltregny D, Dumont B, Turtoi A, Greffe Y, Kirsch S et al (2010) Versican overexpression in human breast cancer lesions: known and new isoforms for stromal tumor targeting. Int J Cancer 126:640–650 65. Paris S, Sesboue R, Chauzy C, Maingonnat C, Delpech B (2006) Hyaluronectin modulation of lung metastasis in nude mice. Eur J Cancer 42:3253–3259 66. Said N, Theodorescu D (2009) The metastasis suppressor RhoGDI2 suppresses expression of Versican, an invasion associated and macrophage stimulatory molecule. In: Proceedings of the American Association for Cancer Research, 2009 Apr 18–22, Denver, CO. AACR, Philadelphia (PA), 2009. Abstract nr 4810 67. Smith SC, Oxford G, Wu Z, Nitz MD, Conaway M, Frierson HF et al (2006) The metastasisassociated gene CD24 is regulated by Ral GTPase and is a mediator of cell proliferation and survival in human cancer. Cancer Res 66:1917–1922 68. Smith SC, Theodorescu D (2009) The Ral GTPase pathway in metastatic bladder cancer: key mediator and therapeutic target. Urol Oncol 27:42–47 69. Wang H, Owens C, Chandra N, Conaway MR, Brautigan DL, Theodorescu D (2010) Phosphorylation of RalB is important for bladder cancer cell growth and metastasis. Cancer Res 70:8760–8769 70. Wu Z, Owens C, Chandra N, Popovic K, Conaway M, Theodorescu D (2010) RalBP1 is necessary for metastasis of human cancer cell lines. Neoplasia 12(12):969–979 71. Henry DO, Moskalenko SA, Kaur KJ, Fu M, Pestell RG, Camonis JH et al (2000) Ral GTPases contribute to regulation of cyclin D1 through activation of NF-kappaB. Mol Cell Biol 20:8084–8092 72. Chien Y, Kim S, Bumeister R, Loo YM, Kwon SW, Johnson CL et al (2006) RalB GTPasemediated activation of the IkappaB family kinase TBK1 couples innate immune signaling to tumor cell survival. Cell 127:157–170 73. Pirruccello SJ, LeBien TW (1986) The human B cell-associated antigen CD24 is a single chain sialoglycoprotein. J Immunol 136:3779–3784 74. Kristiansen G, Pilarsky C, Pervan J, Sturzebecher B, Stephan C, Jung K et al (2004) CD24 expression is a significant predictor of PSA relapse and poor prognosis in low grade or organ confined prostate cancer. Prostate 58:183–192 75. Fogel M, Friederichs J, Zeller Y, Husar M, Smirnov A, Roitman L et al (1999) CD24 is a marker for human breast carcinoma. Cancer Lett 143:87–94 76. Aigner S, Ramos CL, Hafezi-Moghadam A, Lawrence MB, Friederichs J, Altevogt P et al (1998) CD24 mediates rolling of breast carcinoma cells on P-selectin. FASEB J 12: 1241–1251 77. Lopes EC, Ernst G, Aulicino P, Vanzulli S, Garcia M, Alvarez E et al (2002) Dissimilar invasive and metastatic behavior of vincristine and doxorubicin-resistant cell lines derived from a murine T cell lymphoid leukemia. Clin Exp Metastasis 19:283–290 78. Liu W, Vadgama JV (2000) Identification and characterization of amino acid starvationinduced CD24 gene in MCF-7 human breast cancer cells. Int J Oncol 16:1049–1054 79. Ma YQ, Geng JG (2002) Obligatory requirement of sulfation for P-selectin binding to human salivary gland carcinoma Acc-M cells and breast carcinoma ZR-75-30 cells. J Immunol 168:1690–1696 80. Kristiansen G, Sammar M, Altevogt P (2004) Tumour biological aspects of CD24, a mucinlike adhesion molecule. J Mol Histol 35:255–262

344

N. Said and D. Theodorescu

81. Ikenaga N, Ohuchida K, Mizumoto K, Yu J, Kayashima T, Hayashi A et al (2010) Characterization of CD24 expression in intraductal papillary mucinous neoplasms and ductal carcinoma of the pancreas. Hum Pathol 41:1466–1474 82. Lee HJ, Choe G, Jheon S, Sung SW, Lee CT, Chung JH (2010) CD24, a novel cancer biomarker, predicting disease-free survival of non-small cell lung carcinomas: a retrospective study of prognostic factor analysis from the viewpoint of forthcoming (seventh) new TNM classification. J Thorac Oncol 5:649–657 83. Mani SA, Guo W, Liao MJ, Eaton EN, Ayyanan A, Zhou AY et al (2008) The epithelialmesenchymal transition generates cells with properties of stem cells. Cell 133:704–715 84. Dontu G, Liu S, Wicha MS (2005) Stem cells in mammary development and carcinogenesis: implications for prevention and treatment. Stem Cell Rev 1:207–213 85. Lim SC, Oh SH (2005) The role of CD24 in various human epithelial neoplasias. Pathol Res Pract 201:479–486 86. Liu AY, Roudier MP, True LD (2004) Heterogeneity in primary and metastatic prostate cancer as defined by cell surface CD profile. Am J Pathol 165:1543–1556 87. Al-Hajj M, Wicha MS, Benito-Hernandez A, Morrison SJ, Clarke MF (2003) Prospective identification of tumorigenic breast cancer cells. Proc Natl Acad Sci U S A 100:3983–3988 88. Baumann P, Cremers N, Kroese F, Orend G, Chiquet-Ehrismann R, Uede T et al (2005) CD24 expression causes the acquisition of multiple cellular properties associated with tumor growth and metastasis. Cancer Res 65:10783–10793 89. Gassmann P, Kang ML, Mees ST, Haier J (2010) In vivo tumor cell adhesion in the pulmonary microvasculature is exclusively mediated by tumor cell–endothelial cell interaction. BMC Cancer 10:177 90. Kohler S, Ullrich S, Richter U, Schumacher U (2010) E-/P-selectins and colon carcinoma metastasis: first in vivo evidence for their crucial role in a clinically relevant model of spontaneous metastasis formation in the lung. Br J Cancer 102:602–609 91. McDonald B, Spicer J, Giannais B, Fallavollita L, Brodt P, Ferri LE (2009) Systemic inflammation increases cancer cell adhesion to hepatic sinusoids by neutrophil mediated mechanisms. Int J Cancer 125:1298–1305 92. Schmidmaier R, Baumann P (2008) ANTI-ADHESION evolves to a promising therapeutic concept in oncology. Curr Med Chem 15:978–990 93. Laubli H, Stevenson JL, Varki A, Varki NM, Borsig L (2006) L-selectin facilitation of metastasis involves temporal induction of Fut7-dependent ligands at sites of tumor cell arrest. Cancer Res 66:1536–1542 94. Fritzsche J, Hunerbein I, Schumacher G, Alban S, Ludwig R, Gille J et al (2005) In vitro investigation on the selectin binding mechanisms in tumor cell metastasis and their inhibition by heparin. Int J Clin Pharmacol Ther 43:570–572 95. Ludwig RJ, Boehme B, Podda M, Henschler R, Jager E, Tandi C et al (2004) Endothelial P-selectin as a target of heparin action in experimental melanoma lung metastasis. Cancer Res 64:2743–2750 96. Louie E, Nik S, Chen JS, Schmidt M, Song B, Pacson C et al (2010) Identification of a stemlike cell population by exposing metastatic breast cancer cell lines to repetitive cycles of hypoxia and reoxygenation. Breast Cancer Res 12:R94 97. Storci G, Sansone P, Mari S, D’Uva G, Tavolari S, Guarnieri T et al (2010) TNFalpha upregulates SLUG via the NF-kappaB/HIF1alpha axis, which imparts breast cancer cells with a stem cell-like phenotype. J Cell Physiol 225:682–691 98. Shpitz B, Giladi N, Sagiv E, Lev-Ari S, Liberman E, Kazanov D et al (2006) Celecoxib and curcumin additively inhibit the growth of colorectal cancer in a rat model. Digestion 74:140–144 99. Sagiv E, Arber N (2008) The novel oncogene CD24 and its arising role in the carcinogenesis of the GI tract: from research to therapy. Expert Rev Gastroenterol Hepatol 2:125–133 100. Sagiv E, Starr A, Rozovski U, Khosravi R, Altevogt P, Wang T et al (2008) Targeting CD24 for treatment of colorectal and pancreatic cancer by monoclonal antibodies or small interfering RNA. Cancer Res 68:2803–2812

14

Tumor and Host Determinants of Pulmonary Metastasis in Bladder Cancer

345

101. Choi YL, Lee SH, Kwon GY, Park CK, Han JJ, Choi JS et al (2007) Overexpression of CD24: association with invasiveness in urothelial carcinoma of the bladder. Arch Pathol Lab Med 131:275–281 102. Sagiv E, Memeo L, Karin A, Kazanov D, Jacob-Hirsch J, Mansukhani M et al (2006) CD24 is a new oncogene, early at the multistep process of colorectal cancer carcinogenesis. Gastroenterology 131:630–639 103. Sagiv E, Kazanov D, Arber N (2006) CD24 plays an important role in the carcinogenesis process of the pancreas. Biomed Pharmacother 60:280–284 104. Sagiv E, Rozovski U, Kazanov D, Liberman E, Arber N (2007) Gene expression analysis proposes alternative pathways for the mechanism by which celecoxib selectively inhibits the growth of transformed but not normal enterocytes. Clin Cancer Res 13:6807–6815 105. Overdevest JB, Smith SC, Thomas S, Theodorescu D (2009) CD24: target for a novel antimetastatic therapeutic. In: Proceedings of the American Association for Cancer Research, 2009 Apr 18–22, Denver, CO. AACR, Philadelphia (PA), 2009. Abstract nr 4941

Chapter 15

The Influence of Metastatic Breast Cancer on the Bone Microenvironment Andrea M. Mastro, Donna M. Sosnoski, Venkatesh Krishnan, and Karen M. Bussard

Abstract The bone is a welcoming microenvironment for metastases from many cancers including breast cancer. Cancer cells can easily enter the bone marrow cavity due to its vasculature; there they are exposed to the many growth factors and cytokines that are part of the continuous bone remodeling process. In addition, the cancer cells interact with the resident bone cells, osteoblasts and osteoclasts, to modify the microenvironment resulting in cancer cell colonization and eventually bone degradation. Osteoclasts release numerous molecules such as insulin-like growth factors (IGFs) and transforming growth factor beta (TGF-b) from the matrix. Osteoblasts undergo an inflammatory stress response and produce a set of cytokines that are osteoclastogenic. These cytokines include IL-6, IL-8 (murine homologue MIP-2), VEGF, MCP-1, MIG and LIX. Several of these molecules also are secreted by cancer cells. One notable exception is MCP-1 which is made only in small quantities. During the course of metastasis, osteoblast differentiation is suppressed and the prevalence of osteoblast apopotosis increases. Thus the bone microenvironment is modified by the interaction of cancer cells with both osteoblasts and osteoclasts. Bone loss results from hyperactive osteoclasts and hypoactive osteoblasts. This chapter summarizes the results of studies of osteoblast and breast cancer cell interactions that have been performed in cell culture, in mice and in a novel three dimensional culture bioreactor.

A.M. Mastro (*) • D.M. Sosnoski • V. Krishnan Department of Biochemistry, 431 S. Frear Building, University Park, PA 16802, USA e-mail: [email protected]; [email protected]; [email protected] K.M. Bussard Mammary Biology and Tumorigenesis Laboratory, NCI, NIH, Room 1112, Building 37, 37 Convent Drive, Bethesda, MD 20892, USA e-mail: [email protected]

A. Fatatis (ed.), Signaling Pathways and Molecular Mediators in Metastasis, DOI 10.1007/978-94-007-2558-4_15, © Springer Science+Business Media B.V. 2012

347

348

A.M. Mastro et al.

Abbreviations COX-2 DAB ECM G-CSF GM-CSF GFP GRO-a IFN IGFs IL KC LIX M-CSF MCP-1 MIG MIP-2 MMP OPG PDGF PGE PTHrP RANKL TGF-b TNF TRAP TUNEL VEGF

15.1

Cyclooxygenase-2 3,3′- Diaminobenzidine Extracellular matrix Granulocyte-colony stimulating factor Granulocyte macrophage-colony stimulating factor Green fluorescent protein Growth related oncogene alpha Interferon Insulin-like growth factors Interleukin Keratinocyte chemoattractant LPS induced CXC chemokine Macrophage-colony stimulating factor Monocyte chemotactic protein 1 Monokine induced by interferon gamma Macrophage inflammatory protein Matrix metalloproteinase Osteoprotogerin Platelet derived growth factor Prostaglandin Parathyroid hormone related peptide Receptor activator of NFkB ligand Transforming growth factor beta Tumor necrosis factor Tartrate resistant acid phosphatase Terminal deoxynucleotidyl transferase dUTP nick end labeling Vasular endothelial growth factor

Introduction

Late stage metastasis to the skeleton is common for many solid cancers including breast, prostate, lung, renal, and melanoma. This striking predilection for certain cancers to metastasize to the bone approximately 70–80% of the time [1] led to the hypothesis that the bone provides a unique microenvironment which supports the colonization and growth of these tumor cells (reviewed by [2]). So “What is special about bone?” [3]. The bone has the unique combinations of a rich blood supply, a vascular endothelium that permits easy cellular transit, numerous growth factors stored in its matrix, and constant turnover which releases these growth factors. Furthermore, the bone marrow hematopoietic stem cells and mesenchymal stem cells provide a supportive niche for cancer cells as well as a dynamic mixture of cytokines, chemokines, and other growth factors [3].

15

The Influence of Metastatic Breast Cancer on the Bone Microenvironment

349

In the skeleton, cancer cells tend to accumulate in the ribs, in the vertebrae and in the metaphyseal region of long bone. These areas in particular are composed of a meshwork of boney trabeculae which are rich in red marrow and are sites of active bone metabolism. The vascular bed in the trabecular regions consists of voluminous sinusoids compared to capillaries found in most other tissues. Consequently, blood flow through the sinusoids is sluggish. Furthermore, the sinusoids are closely juxtaposed to the trabecular bone. Thus cancer cells have an easy access to the bone marrow and are in close contact with the osteoblasts that line trabecular bone and other cells that occupy the marrow cavity [2]. Nonetheless, entrance into a supportive environment is not sufficient. Not all of the cancer cells that enter the bone marrow cavity will colonize it [4]. Most of the cells evacuate this space and re-enter the bloodstream. Some of the cancer cells in the bone may remain dormant for many years. It is very difficult to follow the progression of metastatic colonization from the time the cancer cells enter the bone marrow until obvious colonies and lesions occur. In an in vivo mouse study designed to follow tumor cells spatially and temporally, green fluorescent protein (GFP) labeled MDA-MB-231 human breast cancer cells were detected in the femurs within 1 h following intracardiac injection [5]. It was a rare event, as might be expected since cells are distributed throughout the body following injection into the arterial circulation. Less than 0.01% of the total injected cells were detected in a femur. However, >90% of the cells detected in the femur localized to the metaphyses after about 24 h. Furthermore, the cells that remained in the metaphyses were found close to the bone surface in the endosteal region. The numbers of cancer cells present in the bone decreased in the first 72 h; this finding is consistent with other investigations on cancer cell trafficking in nonskeletal tissues [6]. It was only after 72 h that there was evidence of colony formation from single cancer cells. Does this lag time indicate a dormancy period? Are the tumor cells somehow preparing the bone environment to optimize cancer growth?

15.1.1

Bone Remodeling

The skeleton constantly undergoes remodeling even in adults. The process of bone degradation by osteoclasts and bone deposition by osteoblasts is well balanced and coordinated by various cytokines and factors (Fig. 15.1a). Many cytokines and growth factors have been detected in the bone marrow [7]. Which of these are responsible for permitting metastatic colonization? Cytokines tend to act in an autocrine manner or a paracrine fashion with neighboring cells. They also are short-lived, act at very low concentrations and are specific to cells that express the corresponding receptor. Many are part of a cascade of other stimulatory or inhibitory factors. Thus, in spite of their overall expression, they are well regulated under normal physiological conditions. Trauma, infection and metastatic cancer cells can lead to an increase or decrease in levels and changes in specific types of cytokines. [8–10]. Among the cytokines involved in bone remodeling are osteoblast-derived interleukins; IL-1, IL-6, IL-8, IL-11, monocyte chemotactic protein (MCP-1),

350

A.M. Mastro et al.

Fig. 15.1 The bone microenvironment under conditions of (a) normal bone remodeling and in the presence of (b) osteolytic bone metastases. (a) The bone remodeling unit consists of osteoblasts that produce osteoid, bone matrix, and of osteoclasts, that degrade mineralized bone. Osteoblasts derive from mesenchymal stem cells in the marrow under control of Runx 2, a key osteoblastic transcription factor. Osteoclasts derive from mononuclear myeloid precursors which fuse to form pre-osteoclasts. Under the influence of M-CSF and RANKL produced by osteoblasts and other cells in the microenvironment, pre-osteoclasts differentiate into multicellular, activated osteoclasts that adhere to the bone and begin matrix degradation. Osteoblasts also produce osteoprotegerin, OPG, a decoy receptor to RANKL. The ratio of RANKL to OPG determines the extent of the osteoclast activity and bone degradation. Other cells of the osteoblastic lineage include bone lining cells and osteocytes. (b) Metastatic breast cancer cells in the bone microenvironment secrete PTHrP, cytokines and growth factors that negatively impact osteoblast function. RANKL and other pro-osteoclastogenic cytokines are increased with a concomitant reduction in OPG resulting in more osteoclast formation and bone degradation. Osteoblast differentiation is suppressed; new osteoid production is no longer able to keep pace with bone resorption. Current therapeutic targets are indicated in green. Bisphosphonates bind to hydroxyapatite, are ingested by osteoclasts and cause their apoptosis. These drugs may also cause cancer cell death; however, they may also negatively affect osteoblasts. Denosumab is an antibody directed to RANKL. It prevents osteoclast differentiation. Teriparatide is a recombinant peptide of parathyroid hormone that stimulates osteoblast activity and bone formation. In addition, pre-clinical trials with agents that target cathepsin K, certain MMPs, and TGF-b are underway (Reprinted with permission from Chen et al. Breast Cancer Research 2010, 12: 215)

cyclo-oxygenase 2 (COX-2), prostaglandins (PGE) and parathyroid hormone related peptide (PTHrP). Osteoblasts also produce RANKL (receptor activator of NFkB ligand) and its decoy receptor osteoprotogerin (OPG). These molecules tend to be osteoclastogenic. RANKL binds to RANK on pre-osteoclasts which are attracted to the site and differentiate into active osteoclasts, bone degrading cells. Degradation of

15

The Influence of Metastatic Breast Cancer on the Bone Microenvironment

351

bone releases some of the matrix-bound factors such as insulin- like growth factors (IGF1 and IGF2) and transforming growth factor beta (TGF-b) which can affect other cells. It has been noted [11] that many of the same molecules involved in remodeling also are expressed by osteoblasts as part of an inflammatory immune response. Both osteoblasts and osteoclasts express toll-like receptors and respond to trauma or bacterial infection with a set of inflammatory stress molecules that include IL-6, IL-8, MCP-1 and COX-2 [10, 12]. Many of these same molecules are produced by osteoblasts in the presence of metastatic cancer cells [9, 13].

15.1.2

The “Vicious Cycle” of Metastatic Tumor Growth in the Bone

An imbalance between osteoblasts and osteoclasts can lead to either excess bone breakdown or excess bone formation. The former process is classified as osteolytic which is characteristic of breast cancer. The latter is osteoblastic and is characteristic of prostate cancer (reviewed by [14]). Most cancers involve both aspects. Over a decade ago, the laboratory of Greg Mundy [15] formulated a paradigm to describe how breast cancers colonize, grow and degrade the bone (Fig. 15.1b) The cycle begins when breast cancer cells enter the marrow cavity and produce parathyroid hormone related peptide (PTHrP). This molecule activates osteoblasts to secrete RANKL which, in turn, binds to RANK on pre-osteoclasts. The osteoclasts differentiate and become active, bone resorbing cells. As the bone matrix is degraded, growth factors such as IGF1, IGF2 and TGF-b are released and can stimulate the cancer cells to produce more PTHrP, thus establishing a positive feedback loop. This cycle drives the osteoclasts to degrade more bone than can be replaced leading to bone loss. The approach to treating metastases in bone has been to break the vicious cycle by inhibiting the osteoclasts. For many years, patients have been treated with drugs of the bisphosphonate family as the standard of care [16]. These molecules bind to hydroxyapatite in the bone matrix and are ingested by the osteoclasts during the process of bone resorption causing osteoclast apoptosis. The third generation of these drugs, the nitrogen containing bisphosphonates such as zoledronic acid, inhibits farnesyl diphosphate synthase, an enzyme critical for protein prenylation. There is evidence that they directly or indirectly elicit tumor cell death as well. However, these drugs are not curative. Their use can reduce the rate of bone loss and improve the quality of life but not reduce mortality [17]. Another osteoclast targeted therapy is denosumab (Prolia™), a monoclonal antibody to RANKL, which inhibits the differentiation of osteoclasts through the RANK-RANKL pathway. It has been used to treat osteoporosis, and recently it has been approved under the name XGEVA™ for treatment of bone metastasis. While these drugs help reduce skeletal related events and improve quality of life, the bone lesions do not heal. Why? Some years ago an oncologist commented that under conditions of bone metastasis the osteoblasts appear to be functionally paralyzed. Unlike the situation in normal bone remodeling, they were not activated to repair the bone degraded by osteoclasts in the presence of cancer cells. This insight led to the theory that some of the bone loss

352

A.M. Mastro et al.

could be attributed to a decrease in bone deposition by the osteoblasts as well as an increase in bone degradation by osteoclasts. Furthermore, the osteoblasts may be unable to repair bone due to either premature apoptosis or to the inability to differentiate into mature osteoblasts and produce the proteins required for matrix deposition.

15.2

15.2.1

The Role of the Osteoblast in Osteolytic Breast Cancer Metastasis Breast Cancer and Osteoblast Apoptosis

Given the fact that bone lesions do not heal following the inhibition of osteoclasts, we hypothesized that osteoblast apoptosis increased, and/or the osteoblast differentiation decreased. The discovery was made first with in vitro studies and then with in vivo experiments that exposure of osteoblasts to breast cancer cells or breast cancer cell conditioned medium increased the number of apoptotic osteoblasts [18, 19]. In the in vitro study, two metastatic human breast cancer cell lines, MDA-MB-231 and MDA-MB-435 and one non-metastatic line MDA-MB-468 were used. Co-culture of a human osteoblast line, hFOB1.19 with the metastatic cells or their conditioned media resulted in approximately a 10% increase in apoptotic osteoblasts. The non-metastatic cells had no effect [18]. In a mouse study, MDA-MB-435 or MDA-MB-231 human cancer cells were inoculated into the left ventricle of the heart to permit bone colonization. Femurs from the mice were removed as early as 1 h after inoculation and as late as 6 weeks. TUNEL assays of fixed bone sections permitted visualization of apoptotic osteoblasts [19]. The numbers of apoptotic osteoblasts increased with time following inoculation of the cells. By histomorphometry it was determined that the total numbers of osteoblasts decreased with time. These findings appear to challenge the paradigm of the vicious cycle. If osteoblasts are critical for the stimulation of the differentiation and activation of osteoclasts and the osteoblasts themselves are victims of the process, how is the vicious cycle able to continue? Perhaps in later stage metastatic disease a RANK-RANKL independent pathway via IL-8 is responsible for osteoclasts activation [20]. In the case of very late stage disease, where both osteoblasts and osteoclasts are absent from the lesion, the cancer cells alone may contribute to matrix degradation [21]. Further studies were carried out to determine if osteoblast differentiation and production of matrix proteins was affected by metastatic breast cancer cells. In in vitro studies, MC3T3-E1 cells, (a murine osteoblast line), as well as primary murine calvarial osteoblasts did not differentiate or mineralize when exposed to conditioned medium from MDA-MB-231 cells [13]. Differentiation was assayed by expression of characteristic osteoblast differentiation genes e.g. osteocalcin, bone sialoprotein and alkaline phosphatase. Interestingly, this block in differentiation could be overcome with the addition of a neutralizing antibody to TGF-b. In addition, the conditioned medium led to marked changes in the osteoblast morphology and

15

The Influence of Metastatic Breast Cancer on the Bone Microenvironment

353

a decrease in adhesion. These morphological changes were not reversed solely by anti-TGF-b. A cocktail of antibodies to platelet derived growth factor (PDGF), IGF-2 and TGF-b was required to permit the focal adhesion plaques and actin stress fiber formation. These cytokines were found to signal through PI3 kinase and rac to bring about changes in osteoblast morphology [22].

15.2.2

The Osteoblast Inflammatory Response

While differentiation was suppressed in osteoblasts in the presence of metastatic breast cancer cells, the expression of several inflammatory cytokines was increased [9]. When conditioned medium from metastatic MDA-MB-231 cells was added to human hFOB1.19 osteoblasts, to murine MC3T3-E1 osteoblasts or to primary murine calvarial osteoblasts, the osteoblasts produced increased levels of IL-6, IL-8 (murine homologue MIP-2) and MCP-1. These cytokines are known to attract, differentiate and activate osteoclasts. Under conditions of normal bone remodeling, they play an important part in the cross talk between osteoblasts and osteoclasts. Under circumstances that change this interaction, these cytokines are part of what has been described as an osteoblast stress response [12]. Further experiments with mouse osteoblasts revealed increases in IL-6, MIP-2 (the mouse homolog of human IL-8), KC (the mouse homologue of GROa), MCP-1, MIP-1g, RANTES, LIX and VEGF [9, 23]. Neutralization of TGF-b in the cancer cell conditioned medium depressed but did not prevent the stress response by the osteoblasts. Other proinflammatory stimulators such as IL-1 and PTHrP were present in the cancer cell conditioned medium but not at high enough concentrations to cause an increase in the other osteoblast molecules. Interestingly, the stage of differentiation of the osteoblasts affected the degree of cytokine response. The more mature osteoblasts gave a greater response to the breast cancer cells than less mature ones [23]. In further experiments, MC3T3-E1 cells were treated with conditioned media from five variants of human MDA-MB-231 cells including a bone seeking and a brain seeking variant. Using a mouse multiplex cytokine assay, it was found that the osteoblasts secreted IL-6, KC, MIP-2, MCP-1 and VEGF. All of these levels were increased in the presence of the cancer cells. There was no significant difference among the cancer cell variants but, in general, conditioned medium from the parental lines of MDA-MB-231 brought about the greatest increase in osteoblast cytokines when placed on differentiated MC3T3-E1 cells [23]. What is the role of these cytokines in the metastatic process? It is known that conditioned media from cultured osteoblast is a strong chemoattractant for cancer cells [24]. Chemoattraction assays were run using each of the five cytokines found in bone culture supernatants; however, none exhibited chemoattractant properties. It is possible that the cytokines produced when osteoblasts are exposed to cancer cells act as mediators for osteoclastogenesis. In fact, the conditioned medium of osteoblasts that have been in the presence of breast cancer cells will trigger bone marrow cells to become osteoclasts in vitro (Fig. 15.2).

354

A.M. Mastro et al.

Fig. 15.2 The differentiation of osteoclasts by conditioned medium from osteoblasts exposed to breast cancer cell conditioned medium. Bone marrow non-adherent cells were isolated from mouse femurs and incubated in alpha-MEM medium with 10% fetal bovine serum alone (a) or (b) with a 1:1 mixture of medium and conditioned medium from MC3T3-E1 osteoblasts that had been exposed to conditioned medium from MDA-MB-231 cells for 24 h. The marrow cultures were carried out for 10 days at which time the cells were fixed and stained for TRAP. Large multinucleated osteoclasts differentiated in the presence of the osteoblast conditioned medium

15.3 15.3.1

Osteoblast Cytokine Responses In Vivo Cytokines in the Presence of Metastases

While cell culture can provide important clues as to how cells interact, the ultimate test is the response in vivo. A xenograft model of human metastatic breast cancer cells inoculated into athymic mice provided a useful tool to examine breast cancer cell interactions with cytokines in the bone metastatic microenvironment. Two parental types of MDA-MB-231 breast cancer cells plus a bone seeking and a metastasis suppressed variant were independently injected into the left cardiac ventricle of 6-week-old female athymic mice [25]. Four weeks later, mice were euthanized, femurs harvested, and cut into ends and shafts. Individual bone portions were crushed and incubated with serum free, alpha-MEM to collect the cytokines released within the bone microenvironment ex vivo [23]. The cytokines were quantified using species-specific antibody arrays to detect mouse cytokines released by cells of the bone, and human cytokines released by breast cancer cells that homed to the bone. A number of femurs were also fixed, decalcified, sectioned, and stained by immunohistochemistry to determine the location of human and murine cytokines present in the femurs [24]. Quantitative cytokine analysis showed that five cytokines tested, IL-6, KC, MCP-1, MIP-2 and VEGF were expressed in the shaft and the ends of the

15

The Influence of Metastatic Breast Cancer on the Bone Microenvironment

355

Table 15.1 Cytokine expressions of MDA-MB-231 cells recovered from the bone marrow Cytokines (pg/ml) MDA-MB-231 cell lines Il-6 IL-8 MCP-1 GRO-a VEGF W Pre 42 66 1 100 901 Post 58 50 2 40 369 PY Pre 8 20 2 13 540 1 Post 5 30 0 0 692 2 Post 10 40 4 2 1302 BO Pre 15 116 2 70 741 1 Post 4 37 2 26 554 2 Post 15 180 3 60 374 MDA-MB-231 cells were assayed for cytokine expression before (pre) intracardiac inoculation into athymic mice or following (post) isolation from the bone marrow 3 weeks later. Three variations of MDA-MB-231 human metastatic breast cancer cells were tested (W, PY, BO). W and PY were parental lines from Dan Welch (W) or Toshi Yoneda (PY). BO was a bone seeking variant derived from the parental PY line from Yoneda. Values were assayed using a Bioplex human cytokine array and were normalized to pg/ml per 1 × 106 cells

bones. The ends of the bones contained significantly more cytokines than the shafts. These findings were true for femurs from all mice regardless of the cancer cell variant injected. It was of vital importance to consider the cytokines produced by the cancer cells themselves. Cancer cells were found to secrete IL-6, IL-8, GROa and VEGF. Interestingly they produced no or only small amounts of MCP-1. MCP-1 is an inducible peptide that recruits monocytes to a site of injury, or pre-osteoclasts (monocyte lineage) to a site of bone resorption. In the case of osteolytic metastasis, MCP-1 is important for the attraction and activation of osteoclasts. It appears that the cancer cells co-opt the osteoblasts to produce more of this cytokine and thus attract osteoclast precursors to the site of bone degradation. This premise raises an additional question. Does the cytokine profile of the cancer cells change after growing in the bone? Perhaps the levels of MCP-1 increase after growth in vivo? Metastatic cells were recovered and cultured from the bones of mice afflicted with bone metastases. Culture supernatants were assayed for cytokine expression. Interestingly, the cytokine expression profile remained essentially the same after growth in vivo (Table 15.1). Immunohistochemical studies were conducted to localize MCP-1, IL-6 and VEGF in the femurs of mice inoculated with human breast cancer cells [24]. We hypothesized that MCP-1, VEGF and IL-6 would be localized in the metaphyseal region of the femurs, an area to which breast cancer cells preferentially migrate. Three weeks following intracardiac inoculation of MDA-MB-231GFP cells into athymic mice, mice were euthanized, the femurs removed, fixed, decalcified and prepared for cryosectioning. The femur sections were processed for immunohistochemistry with antibodies to murine MCP-1 and IL-6 and antibodies to human and murine VEGF (Fig. 15.3). Murine VEGF and MCP-1 were easily detectable in the trabecular bone matrix regions of the proximal and distal femoral ends. Both murine MCP-1 and murine VEGF were present in a strip of about 10–50 mm along

356

A.M. Mastro et al.

Fig. 15.3 Localization of murine MCP-1, IL-6, and VEGF in the femur of mice inoculated with metastatic breast cancer. Athymic mice were inoculated in the left ventricle of the heart with MDA-MB-231GFPcells. The femurs were harvested after 4 weeks, fixed, decalcified and cryosectioned [24]. The sections were stained for murine VEGF, MCP-1, or IL-6 and visualized using a brown DAB chomogen stain. Slides were counterstained using Gill’s hematoxylin. VEGF (a and c) and MCP-1 (b and d) were localized in the trabecular bone of the proximal and distal femur. VEGF (e and g) and MCP-1 (f and h) were localized in cortical bone of the proximal and distal metaphyses. Neither MCP-1(i) nor VEGF(j) were found in the cortical bone of the diaphysis. IL-6 (k) was localized throughout the bone marrow. IL-6 was not present in the trabecular bone or the cortical bone matrix. At least three sections were stained per bone and three bones examined. Shown are representative sections (Reprinted with permission Bussard et al. [24])

the edge of the cortical bone near the metaphyses. However, these cytokines were not present in the matrix of the cortical shaft. Human VEGF was found associated with the cancer cells as expected. Murine IL-6 was not present in the bone shaft but associated with hematopoietic cells throughout the marrow. In addition, murine cytokine expression of MCP-1 and VEGF were not detected in areas immediately adjacent to breast cancer cell colonies [24]. This lack of murine cytokines adjacent to cancer cell colonies may reflect an increase in apoptosis in nearby osteoblasts as previously discussed [18].

15

The Influence of Metastatic Breast Cancer on the Bone Microenvironment

Table 15.2 Mouse cytokines from femurs detected with a multiplex array Cytokines detected in pg/ml None detected 1–10 10–100 IFN-g IL-2 Eotaxin IP-10 IL-3 IL-17 GM-CSF LIF IL-4 M-CSF IL-1a LIX IL-5 RANTES IL-1b MIP-1a IL-7 IL-9 MIP-1b IL-10 IL-12p70 MIP-2 IL-12p40 IL-13 IL-15 TNF-a

357

100–2500 G-CSF IL-6 KC MCP-1 MIG VEGF

Femurs were removed from mice 3 weeks following inoculation of PBS, MDA-MB-231GFP or MDA-MB-231GFP cells into the left ventricle of the heart. Femurs were separated into ends and shafts, crushed and incubated for 24 h in 1 ml of serum free alpha-MEM. The media were assay with a Milliplex™ mouse 32 cytokine array and analyzed with a BioplexTM cytokine assay system. Three mice (6 femurs) were assayed for each group for a total of 18 bones

15.3.2

The Bone Marrow Microenvironment Over Time

Clearly, the bone microenvironment is altered during skeletal metastasis. However, metastasis is a dynamic process and the changes have not been examined over time, beginning from the arrival of metastatic cancer cells in the bone to their colonization, growth, and continued spread. There are a variety of cell types that contribute to the metastatic microenvironment including cancer cells, stromal cells, osteoblasts, vascular endothelial cells, circulating macrophages and/or other immune cells all of which may produce cytokines/chemokines and other growth factor. These molecules indicate that bone marrow microenvironment is in a constant state of flux during breast cancer cell metastasis. Experiments were carried out to determine how the expression pattern of cytokines/chemokines in the bone changed following the establishment and growth of metastatic and metastasis suppressed breast cancer cells. It was anticipated that the cytokines expressed in the bone metaphysis (the ends, areas to which breast cancer cells initially traffic) would be different from the bone diaphysis (shaft). In addition, breast cancer cells that localize in the bone but do not grow would likely exhibit a different cytokine profile than breast cancer cells that both localize and grow in the bone [26]. Two human breast cancer cell lines used for this study: the metastatic line, MDAMB-231GFP and a metastasis suppressed variant, MDA-MB-231BRMS1GFP, which is found in the bone following intracardiac injection but does not grow there [27]. The cancer cells were inoculated into the left cardiac ventricle of athymic mice. PBS inoculations served as a control for the procedure. Femurs were harvested at various times, examined by fluorescent microscopy for the presence of cancer cells and separated into ends (metaphysis) and shafts (diaphysis). Bone sections were crushed with the marrow intact and cultured for 24 h. The culture supernatant was assayed for various cytokines. An initial screen consisted of a 32-cytokine multiplex array. Of the 32 cytokines, nine were below the level of detection (Table 15.2).

358

A.M. Mastro et al.

These included IFN-g, Interleukins 3, 4, 5, 7, 10, 15 and IL-12p40, and TNFa. The others were detected in amounts ranging from 1 to 2,500 pg/ml depending on the cytokine. IL-2, IL17, M-CSF and RANTES were present at £1–10 pg/ml. IL-6, KC, MCP-1, MIG (Monokine Induced by Interferon-g), VEGF, LIX, MIP-2 and eotaxin were selected to assay in further experiments. These eight cytokines were of particular interest because of their roles in bone and cancer biology. IL-6 is an osteoblast inflammatory stress cytokines. This pro-inflammatory cytokine is present in many tissues and produced by many cells. In bone, osteoblasts secrete increasing amounts of IL-6 as they differentiate [28]. MCP-1, a member of the CC chemokine family also is produced by osteoblasts as part of the bone remodeling process. It stimulates the migration of monocyte-osteoclast progenitor cells from the blood or marrow to the bone. It is also associated with increased angiogenesis and tumor cell survival. Interestingly, MCP-1 is secreted in only very small concentrations by MDA-MB-231GFP cells and MDA-MB-231BRMS1GFP cells [23]. KC is produced by osteoblasts, stimulates angiogenesis, neutrophil chemotaxis and activation [29]. VEGF is a growth and angiogenic cytokine that promotes angiogenesis and vascular permeability [30]. MIG is a CXC chemokine that also has angiogenic properties. It is produced by cells of the bone marrow [31]. Initial experiments were conducted using post injection harvest time points of 1, 3, 5, 7, 14, 21 and 28 days. However, at 1 day post injection, it was discovered that cytokine expression was drastically reduced in all mice, including the control. In subsequent experiments, the first harvest time was delayed to 3 days to allow mice to recover from the inoculation procedure. The number of points (3, 11, 19, 27 days) was also reduced to allow for a greater number of animals per treatment group (n = 8). Examination of the femurs by stereo fluorescent microscopy revealed that, on average, the incidence of visible cancer cells in the femur was twice as high in the mice injected with MDA-MB-231GFP (47%) than in the mice injected with MDA-MB-231BRMS1GFP (26%). Colonies in the bones of mice with MDA-MB231BRMS1GFP injections tended to be much smaller than those inoculated with MDA-MB-231GFP cells. Following microscopic examination, the femurs were cut into shafts and ends, crushed and incubated for 24 h in serum free alpha-MEM. The culture supernatants were assayed for the presence of eight murine cytokines using species-specific antibodies: KC, MIP-2, VEGF, MIG, MCP-1, IL-6, LIX, eotaxin. Of these LIX and MIP-2 were barely detectable and were eliminated from further analyses. As reported previously [23], the ends of the bone contained significantly more cytokines than the cortical region (Fig. 15.4). This was true of all the cytokines tested except for MIG which was similar in concentration in both regions of the bone. The amount of cytokines derived from each femur varied widely. Even femurs from the same mouse inoculated with PBS were often very different. Even so, at 27 days after inoculation of the cancer cells, femurs from mice inoculated with either MDA-MB-231GFP or MDA-MB-231BRMS1 GFP cells contained significantly more MCP-1, IL-6, VEGF, MIG and eotaxin than those from PBS injected mice (Fig. 15.5). MCP-1 gradually increased in both MDA-MB-231GFP and MDA-MB-231BRMS1GFP

15

The Influence of Metastatic Breast Cancer on the Bone Microenvironment

359

Fig. 15.4 Comparison of cytokines in the shafts and ends of femurs of mice. Mice were treated as described in the legend to Table 15.2. Femurs were taken 3 weeks post inoculation of cancer cells. Shown are the averages of four bones for PBS treatment and eight bones for inoculation with MDA-MB-231 (231), and eight with MDA-MB-231BRMS1 (BRMS). *** indicates P < 0.001 ends > shaft

injected mice (Fig. 15.6). MIG levels the femurs of mice inoculated with human breast cancer cells were lower at day 19 compared to femurs of mice inoculated with PBS. However, by day 27, the metaphyses of femurs from mice inoculated with human breast cancer cells expressed larger amounts of MIG than the femurs of mice inoculated with PBS. It was generally seen that cytokine concentration in the femurs of mice inoculated with MDA-MB-231BRMS1 cells were expressed in similar quantities to the femurs from mice inoculated with MDA-MB-231 cells.

360

A.M. Mastro et al.

Fig. 15.5 Comparison of cytokines detected in the metaphyses of femurs of mice following intracardiac inoculation of metastatic and metastasis suppressed breast cancer cells. Femurs were removed from mice 3 weeks following inoculation of MDA-MB-231GFP cells, MDA-MB231BRMS1GFP cells or PBS into the left ventricle of the heart. The femurs were treated as described in the legend to Table 15.2. Femurs from eight mice (16 femurs) were assayed. Shown is the mean log of the cytokine concentration. The results were analyzed by a statistical model in which femurs from the same mouse were considered repeated measurements and the mice were grouped according to treatments with PBS, MDA-MB-231 (231) or MDA-MB-231BRMS1 (BRMS). *P < 0.05, **P < 0.01, ***P < 0.001

IL-6 increased earlier; by day 11 it was higher in the cancer- injected mice. Eotaxin showed no significant differences among groups until day 27. For the most part the cytokines from the MDA-MB-231BRMS1 GFP inoculated mice followed the same expression pattern as the MDA-MB-231GFP cells. The fact that the metastatic and non-metastatic cell lines elicit the same cytokine responses in the bone implies that the local cytokine environment is not responsible for the inability of the BRMS1 cells to proliferate in bone.

15.4

A Three-Dimensional Model of the Bone Microenvironment

Due to the sheer expanse of the human skeleton as well as difficulty in accessing the bone marrow cavity in the clinic, tumors and osteolytic lesions are usually not detected until late in the metastatic process. Lack of easy access makes it very difficult to study cellular and molecular mechanisms involved in cancer colonization of bone and hampers rapid development of therapeutic interventions. Although excised tissue captures end stages of metastases in bone, the critical initial stages of disease remain substantially inaccessible. Bioengineers have been attempting to develop

15

The Influence of Metastatic Breast Cancer on the Bone Microenvironment

361

Fig. 15.6 Changes in MCP-1 and MIG concentrations over time in the metaphyses of femurs following intracardiac inoculation of MDA-MB-231GFP cells, MDA-MB-231BRMS1GFP cells or PBS into the left ventricle of the heart. The femurs were treated as described in the legend to Table 15.2. Shown are the median log transformed values ± standard deviations for MCP-1 (left) and MIG (right). The numbers indicate that the values on a particular day are statistically different (P < 0.5) from the days listed above the box

cell culture systems that mimic the in vivo system [32–34]. These culture constructs, often referred to as bioreactors, attempt to bridge the complexity from tissue culture polystyrene to an animal model. Ideally, a bioreactor allows a rapid and orderly development of tissue starting from single cells. A novel, compartmentalized bioreactor based on the principle of simultaneous growth and dialysis [35] has been developed which allows this process to occur. A dialysis membrane separates a cell growth space from a larger volume medium reservoir. Cells are inoculated into the growth chamber in complete medium including serum. The reservoir is filled only with basal medium. During culture, cells are bathed in pH-equilibrated and oxygenated medium continuously dialyzing from the reservoir. At the same time, metabolic waste products such as lactic acid dialyze out of the growth compartment, maintaining low pericellular concentrations. Serum constituents or macromolecules synthesized by cells with molecular weights in excess of the dialysis membrane cutoff (typically 6–8KD) are retained and concentrated within the growth compartment. The pericellular environment is unperturbed by refreshing only the media reservoir (not cell compartment). Thus, the culture environment is very stable; various chemical gradients that guide

362

A.M. Mastro et al.

Fig. 15.7 Morphological progression of MC3T3-E1 cells over 10 months of bioreactor culture. Cultures were monitored by confocal microscopy of Alexa Fluor 568 phalloidin stained cells (a–d). Osteoblasts progress from rapidly proliferating cells (a and b) to ‘cobble stone’ shaped cells (c) to stellate shaped osteocyte-like cells (d). Metastatic cancer cells co-cultured with 2 month old, osteoblasts (red, stained with Cell Tracker Orange) for a period of 7 days (e–h). MDA-MB-231 cells (e and f) adhere, proliferate and form colonies and invade the multi-layered osteoblast tissue (f). The MDA-MB-231 cells form single files of cells (arrows f), and also cause a change in osteoblast morphology from cuboidal (c) to spindle shaped. In contrast, MDA-MB-231-BRMS1 cells (g), a metastasis suppressed variant of MDA-MB-231, adhere lightly to and proliferates on the osteoblast tissue. By day 7 of co-culture, the BRMS1 cells show signs of apoptosis (arrows g). Multiple myeloma cells (h) appear to aggregate in groups that undergo rapid proliferation and eventually cover over the osteoblast tissue

osteoblast maturation can develop. As a consequence, a thick and highly mineralized osteoblastic tissue can be grown and maintained for long terms [36, 37]. The bioreactor has allowed direct observation of osteoblast development; i.e. proliferation of pre-osteoblasts and differentiation into mineralizing osteoblasts that become engulfed in a thick, cell-secreted mineralized matrix that completely surrounds the cells (Fig. 15.7) [38]. Development and maturation of cell-derived matrix is important because it has been shown that interaction of cancer cells with reconstituted collagen layers is fundamentally different than interaction with authentic cell-produced matrix [39]. Phenotypic maturation of osteoblastic tissue was measured as a function of time in the bioreactor (Fig. 15.7a–d). In some cultures, after 5 months, contiguous, macroscopic mineral deposits were formed that proved consistent with bone hydroxyapatite by X-ray scattering and IR spectroscopy [38]. In addition, the gene expression profiles of characteristic osteoblastic proteins such as type 1 collagen, alkaline phosphatase, osteonectin, osteopontin and osteocalcin, mirrored that observed in vivo. After several months, cell morphology strongly resembled osteocytes (Fig. 15.7d) and cultures expressed molecules indicative of osteocytes;

15

The Influence of Metastatic Breast Cancer on the Bone Microenvironment

363

i.e. E11, DMP1, and sclerostin (Fig. 15.7a–d). The compartmentalized bioreactor represents a valid in vitro model for studies of bone biology and pathology. The introduction of MDA-MB-231GFP human metastatic breast cancer cells (known to invade the murine skeleton) [40] onto MC3T3-E1 osteoblast tissue grown in this bioreactor to various stages of phenotypic maturity allowed for observation of early stages of cancer cell colonization in real time by confocal microscopy (Fig. 15.7e, f) [41]. These stages included breast cancer cell to osteoblast tissue adhesion, cancer cell proliferation, tissue penetration, formation of non-vascularized colonies of tumor cells, and degradation of osteoblast derived extracellular matrix (ECM). Cancer cells proliferated and organized into single cell files (Fig. 15.7f, arrows) as described for infiltrating lobular or metaplastic breast carcinomas [42]. Migration of cancer cells along tracks of remodeled ECM produced by preceding invading cell(s) is likely responsible for the characteristic cell alignment patterns [43]. Invasion by chains of tumor cells linked together by cell-cell contacts is considered to be an effective penetration mechanism, [44] conferring high metastatic capacity and commensurately poor prognosis when observed in patients [42, 43]. Observation of single-cell filing in the osteoblast tissue model indicated a considerable degree of physiological relevance. Single-cell filing is common in soft tissue but also found in bone metastasis [45]. Also, the osteoblasts increased their production of inflammatory cytokines, i.e. IL-6, KC, MIP-2, MCP-1, VEGF. Concomitantly, production of collagen, osteocalcin and other differentiation proteins responsible for bone accretion was reduced. To clarify the effect of metastatic cancer cells on the osteoblast tissue, metastasis-suppressed MDA-MB-231-BRMS1 cells were cocultured with the mature osteoblast tissue (Fig. 15.7g). MDA-MB-231-BRMS1 cells interacted with the tissue in a very different manner than the metastatic MDA-MB-231 cells. BRMS1 cells formed small colonies on osteoblasts but did not penetrate the tissue, or form micro-tumors. Furthermore, the metastasis suppressed cells did not cause changes in osteoblast morphology. However, they showed marked signs of apoptosis (Fig. 15.7g, arrow). Thus we have demonstrated that 3D, multiple-cell-layer osteoblast tissue grown in the simultaneous-growth-and-dialysis bioreactor is a clinically relevant in vitro model in which to study osteoblastic aspects of metastatic breast cancer in bone. This bioreactor tissue model is an ideal tool to study various other cancer types, such as multiple myeloma, and prostate cancer that preferentially metastasize to bone but behave differently than breast cancer [46, 47]. This model allows one to control the bone microenvironment while varying types of cancer. Multiple myeloma arises from a hematopoietic lineage. The cells are found in close association with sites of active bone resorption; their ability to stimulate osteoclast formation and activity has been well characterized [48]. In the early stages of multiple myeloma, bone formation is actually increased. This increase is thought to reflect a compensation for the increase in osteoclastic resorption [49]. However, as the disease progresses, bone formation is rapidly reduced and as a consequence, the osteolytic aspect of multiple myeloma increases, resulting in the release of a number of pro-tumorogenic factors that promote cancer cell growth and survival. The bioreactor was employed to co-culture multiple myeloma cells with osteoblast tissue

364

A.M. Mastro et al.

Fig. 15.8 Three dimensional reconstruction of confocal Z-stack images of multiple-myeloma cells on 2 month old osteoblast tissue after 7 days of co-culture. Multiple-myeloma cells adhere to the top layer of osteoblast tissue and proliferate to from a confluent layer (~11–12 mm thick optical sections), but fail to penetrate through to the bottom layer of osteoblasts

(Figs. 15.7h and 15.8). Myeloma cells were observed to adhere to the osteoblasts and eventually formed a confluent monolayer. Unlike breast cancer cells that quickly penetrated the osteoblastic tissue, the myeloma cells remained in a layer over the osteoblasts. It has long been appreciated that the extracellular matrix (ECM) is a primary determinant of cell phenotype [50]. In recent years, the role of ECM in controlling malignancy and the remodeling of ECM that occurs during cancer cell colonization of tissue has become a focus of research [51]. A pantheon of important adhesins and proteases have been identified [52], and the overall structure of the ECM has been shown to control cancer cell migration [53]. For example, the interaction of cancer cells with reconstituted collagen layers is fundamentally different than interaction with authentic cell-produced matrix [39]. In an effort to investigate changes induced in the osteoblast ECM by the MDA-MB-231 cells, osteoblasts were exposed to conditioned medium from the cancer cells. After 4 h of incubation with MDA-MB-231 breast cancer conditioned medium, osteoblast RNA was extracted and analyzed

15

The Influence of Metastatic Breast Cancer on the Bone Microenvironment

365

Table 15.3 Gene expression analysis of 3 month old osteoblasts treated with breast cancer cell conditioned media (CM) in bioreactor culture system Fold change when treated with CM Up regulated genes Procollagen, type IV, alpha3 (Col4a3) 4.0 Elastin microfibril interfacer 1 (Emilin1) 4.0 Ectonucleoside triphosphate 5.0 diphhosphohydrolase 1 (Entpd1) Integrin alpha 2 (Itga2) 3.0 Integrin alpha 5 (Itga5) 3.6 Integrin beta 3 (Itga3) 4.0 Matrix metallopeptidase 1a (MMP1a) 4.5 Matrix metallopeptidase 3 (MMP3) 3.0 Down regulated genes Fibulin 1 (Fbln1) Hemolytic Complement (Hc) Synaptotagmin 1 (Syt1) A disintegrin-like and metallopeptidase with thrombospondin type1 motif, 5 (Adamts5) Cadherin 1 (Cdh1)

−7.0 −7.2 −4.0 −3.1 −4.2

MC3T3-E1 osteoblasts grown in the bioreactor for 3 months were treated with 50% MDA-MB-231 conditioned medium for 4 h. RNA was collected and profiled for expression of 84 genes involved in cell-cell and cell-matrix interaction using an SABiosciences gene expression array. Genes that exhibited a threefold or greater change are listed

with a gene array specific for expression of ECM related genes. Of the 84 genes on the array, three to fivefold increases were detected in several integrins, MMPs, and procollagen type IV (Table 15.3). Furthermore, several proteases and adhesion molecules were decreased. These preliminary findings along with observed morphology changes in direct co-cultures suggest that the osteoblast matrix was remodeled as a result of molecules secreted by the osteoblasts and the cancer cells and by cell-cell contact. The matrix that forms in the 3D culture is critical for an understanding of the cancer cell colonization of the host tissue. In order to increase the biological complexity of the system and to begin to model the vicious cycle, osteoclasts have also been added to the system. The aim is to create a tri-culture system containing the key cells that define the vicious cycle (osteoblasts, osteoclasts and metastatic cancer cells) and monitor their interaction. This model could be an invaluable tool in targeted therapeutic studies. In summary, the 3D in vitro culture system enables both tumor and non-tumor cells to grow and organize in a more biologically relevant manner, enhancing our ability to observe and to manipulate cancer progression. Three dimensional model systems that more accurately mimic in vivo phenotypes, such as the cell perfusion culture system in microfluidic channels, are being designed to advance drug toxicity, metabolism, and stem cell differentiation studies [54]. Most importantly, cell-cell interactions can be closely monitored in a 3D culture system, without specialized microscopy or sacrifice of animals.

366

15.5

A.M. Mastro et al.

Conclusions

A variety of systems described in this chapter and ranging from standard cell culture to three-dimensional bioreactors to mouse models have been used to study the interaction of breast cancer cells with the bone microenvironment. These observations, taken together, support the hypothesis that normal osteoblast function (i.e. deposition of matrix) is impaired in the presence of breast cancer cells. Furthermore, osteoblasts undergo a stress response and secrete increased levels of inflammatory cytokines that create an environment conducive to cancer cell propagation and to osteoclast activation. Thus the metastatic bone environment results from the synergy among the resident bone marrow cells and the invading cancer cells. While current therapies are directed towards activated osteoclasts and are effective in slowing lesion progression, new therapeutic strategies should focus beyond inhibition of osteoclasts and include enhancement or restoration of osteoblast function. Further understanding of the cellular and molecular basis for breast cancer interactions with bone requires a complete and in-depth understanding of the bone microenvironment including cell-cell interaction, cell-matrix interaction and cytokine crosstalk as they occur over the time of cancer colonization. Models that increase the biological complexity by inclusion of other cells of the metastatic bone environment could reveal unknown interactions and possible new treatment options. Acknowledgements This work was supported through grants from the U.S. Army Medical and Materiel Command Breast Cancer Research Program (WX81XWH-06-1-0432 and WX81XWH-0801-0448), The Susan G. Koman Breast Cancer Foundation (BCTR 0601044), with help from The National Foundation for Cancer Research, and The Penn State Hershey Cancer Institute. We would like to thank Erwin Vogler, Carol Gay, Ravi Dhurjati, and Laurie Shuman for their work with various aspects of the projects.

References 1. Rubens RD, Mundy GR (2000) Cancer and the skeleton. Martin Dunitz, London 2. Mastro AM, Gay CV, Welch DR (2003) The skeleton as a unique environment for breast cancer cells. Clin Exp Metastasis 20:275–284 3. Bussard KM, Gay CV, Mastro AM (2008) The bone microenvironment in metastasis; what is special about bone? Cancer Metastasis Rev 27:41–55 4. Fidler IJ (1970) Metastasis: quantitative analysis of distribution and fate of tumor embolilabeled with 125 I-5-iodo-2¢-deoxyuridine. J Natl Cancer Inst 45:773–782 5. Phadke PA, Mercer RR, Harms JF et al (2006) Kinetics of metastatic breast cancer cell trafficking in bone. Clin Cancer Res 12:1431–1440 6. Nash KT, Phadke PA, Navenot JM et al (2007) Requirement of KISS1 secretion for multiple organ metastasis suppression and maintenance of tumor dormancy. J Natl Cancer Inst 99:309–321 7. Roodman GD (2004) Mechanisms of bone metastasis. N Engl J Med 350:1655–1664 8. Fritz E, Glant T, Vermes C et al (2005) Chemokine gene activation in human bone marrowderived osteoblasts following exposure to particulate wear debris. J Biomed Mater Res A 77:192–201

15

The Influence of Metastatic Breast Cancer on the Bone Microenvironment

367

9. Kinder M, Chislock E, Bussard KM et al (2008) Metastatic breast cancer induces an osteoblast inflammatory response. Exp Cell Res 314:173–183 10. Marriott I (2004) Osteoblast responses to bacterial pathogens: a previously unappreciated role for bone-forming cells in host defense and disease progression. Immunol Res 30:291–308 11. Rodan GA (2003) The development and function of the skeleton and bone metastases. Cancer 97:726–732 12. Fritz EA, Glant TT, Vermes C et al (2002) Titanium particles induce the immediate early stress responsive chemokines IL-8 and MCP-1 in osteoblasts. J Orthop Res 20:490–498 13. Mercer R, Miyasaka C, Mastro AM (2004) Metastatic breast cancer cells suppress osteoblast adhesion and differentiation. Clin Exp Metastasis 21:427–435 14. Casimiro S, Guise TA, Chirgwin J (2009) The critical role of the bone microenvironment in cancer metastases. Mol Cell Endocrinol 310:71–81 15. Guise TA, Yin JJ, Taylor SD et al (1996) Evidence for a causal role of parathyroid hormonerelated protein in the pathogenesis of human breast-cancer-mediated osteolysis. J Clin Invest 98:1544–1549 16. Lipton A (2010) Bone continuum of cancer. Am J Clin Oncol 33:S1–S7 17. Coleman R, Gnant M (2009) New results from the use of bisphosphonates in cancer patients. Curr Opin Support Palliat Care 3:213–218 18. Mastro AM, Gay CV, Welch DR et al (2004) Breast cancer cells induce osteoblast apoptosis: a possible contributor to bone degradation. J Cell Biochem 91:265–276 19. Phadke PA, Mercer RR, Harms JF, Jia Y, Kappes JC, Frost AR, Jewell JL, Bussard KM, Nelson S, Moore C, Gay CV, Mastro AM, Welch DR (2006) Kinetics of metastatic breast cancer cell trafficking in bone. Clin Cancer Res 12:1431–1440 20. Bendre M, Gaddy-Kurten D, Foote-Mon T et al (2002) Expression of interleukin 8 and not parathyroid hormone-related protein by human breast cancer cells correlates with bone metastasis in vivo. Cancer Res 62:5571–5579 21. Orr FW, Wang HH, Lafrenie RM et al (2000) Interactions between cancer cells and the endothelium in metastasis. J Pathol 190:310–329 22. Mercer RM, Mastro AM (2005) Cytokines secreted by bone-metastatic breast cancer cells alter the expression pattern of f-actin and reduce focal adhesion plaques in osteoblasts through PI3K. Exp Cell Res 310:270–281 23. Bussard KM, Venzon DJ, Mastro AM (2010) Osteoblasts are a major source of inflammatory cytokines in the tumor microenvironment of bone metastatic breast cancer. J Cell Biochem 111(5):1138–1148 24. Bussard KM, Okita N, Sharkey N et al (2010) Localization of osteoblast inflammatory cytokines MCP-1 and VEGF to the matrix of the trabecula of the femur, a target area for metastatic breast cancer cell colonization. Clin Exp Metastasis 27:331–340 25. Yoneda T, Williams PJ, Hiraga T et al (2001) A bone-seeking clone exhibits different biological properties from the MDA-MD-231 parental human breast cancer cells and a brain-seeking clone in vivo and in vitro. J Bone Miner Res 16:1486–1495 26. Sosnoski DM, Krishnan V, Kraemer WJ et al (in press) Changes in cytokines of the bone microenvironment during breast cancer metastasis. Int J Breast Cancer 27. Meehan WJ, Welch DR (2003) Breast cancer metastasis suppressor 1: update. Clin Exp Metastasis 20:45–50 28. Kurihara N, Bertolini D, Suda T et al (1990) IL-6 stimulates osteoclast-like multinucleated cell formation in long term human marrow cultures by inducing IL-1 release. J Immunol 144:4226–4230 29. Bischoff D, Zhu J, Makhijani N et al (2005) KC chemokine expression by TGF-beta in C3H10T1/2 cells induced towards osteoblasts. Biochem Biophys Res Commun 326:364–370 30. Fitzgerald K, O’Neill L, Gearing A et al (2001) The cytokine facts book. Academic, New York 31. Schwartz GN, Liao F, Gress RE et al (1997) Suppressive effects of recombinant human monokine induced by IFN-gamma (rHuMig) chemokine on the number of committed and primitive hemopoietic progenitors in liquid cultures of CD34+ human bone marrow cells. J Immunol 159:895–904

368

A.M. Mastro et al.

32. Abousleiman RI, Sikavitsas VI, Fisher JP (2007) Bioreactors for tissues of the musculoskeletal system. Tissue engineering, vol 585. J. Fisher, Springer US, pp 243–259 33. Martin I, Wendt D, Heberer M (2004) The role of bioreactors in tissue engineering. Trends Biotechnol 22:80–86 34. Maxson S, Orr D, Burg KJL et al (2011) Bioreactors for tissue engineering. Tissue engineering, vol 1. C.V. Suschek and N. Pallua, Springer Berlin Heidelberg, pp 179–197 35. Vogler EA (1989) A compartmentalized device for the culture of animal cells. Biomater Artif Cells Artif Organs 17:597–610 36. Krishnan V, Shuman L, Sosnoski D et al (2010) Comparison of metastatic breast cancer colonization of osteoblastic tissue grown in two- and three-dimensional tissue culture. J Cell Physiol in press 37. Dhurjati R, Liu X, Gay CV et al (2006) Extended-term culture of bone cells in a compartmentalized bioreactor. Tissue Eng 12:3045–3054 38. Krishnan V, Dhurjati R, Vogler EA et al (2010) Osteogenesis in vitro: from pre-osteoblasts to osteocytes. In Vitro Cell Dev Biol Anim 97:11511–11515 39. Sabeh F, Shimizu-Hirota R, Weiss SJ (2009) Protease-dependent versus -independent cancer cell invasion programs: three-dimensional amoeboid movement revisited: jcb.200807195 40. Rusciano D, Burger M (2000) In vivo cancer metastasis assays. Cancer metastasis: experimental approaches, vol 29. D. Welch, Elsevier, Amsterdam, pp 207–242 41. Dhurjati R, Krishnan V, Shuman LA et al (2008) Metastatic breast cancer cells colonize and degrade three-dimensional osteoblastic tissue in vitro. Clin Exp Metastasis 25:741–752 42. Page DL, Anderson TJ (1987) Diagnostic histopathology of the breast. D.L. Page and T.J. Anderson, Churchill-Livingstone, Edinburgh, UK, pp 219–222 43. Friedl P, Wolf K (2003) Tumour-cell invasion and migration: diversity and escape mechanisms. Nat Rev Cancer 3:362–374 44. Friedl P, Wolf K (2008) Tube travel: the role of proteases in individual and collective cancer cell invasion. Cancer Res 68:7247–7249 45. Mastro AM, Vogler EA (2009) A three-dimensional osteogenic tissue model for the study of metastatic tumor cell interactions with bone. Cancer Res 69:4097–4100 46. Logothetis CJ, Lin SH (2005) Osteoblasts in prostate cancer metastasis to bone. Nat Rev Cancer 5:21–28 47. Patten RM, Shuman WP, Teefey S (1990) Metastases from malignant melanoma to the axial skeleton: a CT study of frequency and appearance. AJR Am J Roentgenol 155:109–112 48. Mundy GR, Raisz LG, Cooper RA et al (1974) Evidence for the secretion of an osteoclast stimulating factor in myeloma. N Engl J Med 291:1041–1046 49. Bataille R, Chappard D, Marcelli C et al (1991) Recruitment of new osteoblasts and osteoclasts is the earliest critical event in the pathogenesis of human multiple myeloma. J Clin Invest 88:62–66 50. Hay ED (1991) Cell biology of extracellular matrix. Plenum Press, New York 51. Wall SJ, Jiang Y, Muschel RJ et al (2003) Meeting report: proteases extracellular matrix, and cancer: an AACR special conference in cancer research. Cancer Res 63:4750–4755 52. Robert L (2002) Cell-matrix interactions in cancer spreading–effect of aging: an introduction. Semin Cancer Biol 12:157–163 53. Alexander NR, Branch KM, Parekh A et al (2008) Extracellular matrix rigidity promotes invadopodia activity. Current biology: CB 18:1295–1299 54. Toh YC, Zhang C, Zhang J et al (2007) A novel 3D mammalian cell perfusion-culture system in microfluidic channels. Lab Chip 7:302–309

Index

A Androgen depletion therapy (ADT), 280–281 Androgen receptor (AR), oncogenes, and metastasis c-MYC (MYC) CTCs, 295–296 Ezrin expression, 295 Lo-Myc model, 294 proto-oncogene, 294 transgenic mice, 294–295 c-SRC, 296–297 Androgen response elements (AREs), 279

B Benign prostate hyperplasia (BPH), 231 Biomarker-Integrated Approaches of Targeted Therapy for Lung Cancer Elimination (BATTLE), 7 Biomechanical extracellular matrix switches basement membrane, 81–83 biological information, 72 embryonic development and tumor progression, 71 integrins, 72 interstitial matrix, 79–81 physiochemical and mechanical properties, 78 therapeutic and imaging targets, 83–84 tumor metastasis (see Tumor metastasis) Bladder cancer CD24, 338–339 endothelin axis, 335–337 lung microenvironment ECs, 333 premetastatic niche, 334

primary tumor and shed tumor cells, 332–333 TAMs, 334 vascular endothelium, 332 metastasis suppressor RhoGDI2, 335 Ral GTPases, 338 UC, 331–332 versican, 337–338 Blood-brain-barrier (BBB), 208 Bone marrow derived myeloid cells (BMDCs), 231, 233 Bone marrow microvasculature, 208 Bone microenvironment bioreactor, 361–362 BRMS1 cells, 363 confocal microscopy, 363 confocal Z-stack images, 364 3D culture system, 365 ECM, 364 gene expression, 365 myeloma cells, 364 single-cell filing, 363 Brain cancer cancer cell migration and invasion b1-containing integrins, 316 avb3 and avb5 integrins, 317 ECM components, 314–315 experimental models and clinical data, 313–314 glioma cancer stem cell function, 323–324 glioma treatment a5b1 integrin antagonists, 323 radiation and chemotherapy, 321–322 avb3/avb5 integrin antagonists, 322–323

A. Fatatis (ed.), Signaling Pathways and Molecular Mediators in Metastasis, DOI 10.1007/978-94-007-2558-4, © Springer Science+Business Media B.V. 2012

369

370 Brain cancer (cont.) integrins a and b subunits, 311–312 cell types, 313 ECM binding, 313 metastatic colonization, 315 regulating growth and angiogenesis animal models, 319, 320 glioma cells, 320–321 TGF-b activation, 321 avb3 ligands, 318–319 VEGF expression, 321 avb3 expression, 314 Brain microvascular endothelial cells (BMECs), 208

C Cancer androgen suppression therapy, 4 biochemical progression, 4 bladder (see Bladder cancer) brain (see Brain cancer) breast cancer, 5 chemotherapy and combination therapy, 7 cytoreductive nephrectomy, 5 DMFS, 6–7 malignancy ranges, 3 monoclonal antibody Cetuximab, 5 oligometastatic disease, 7 post treatment prostate biopsy, 4 primary tumor treatment, 5–6 radiation dose escalation, 4 radiation therapy, 3–4 stereotactic body radiation therapy, 7 tumor self seeding hypothesis, 7 Cancer cell extravasation cellular and molecular interactions, 214 clinical aspects, 213–214 EMT and, 212–213 mechanisms endothelium, 201–207 leukocyte vs. cancer cell extravasation, 199–201 metastatic inefficiency and extravasation, 210–212 metastatic organ, 214 vascular architecture bone marrow microvasculature, 208 brain microvasculature, 208–209 liver microvasculature, 209 lung microvasculature, 209–210 Cancer cell migration and invasion, 315–317 Cancer stem cells (CSCs), 323

Index Carbohydrate recognition domain (CRD), 337 Carcinoma-associated fibroblasts (CAFs), 231 Castration resistant prostate cancers (CRPCs), 279 Chondroitin sulfate (CS), 337 Circulating tumor cells (CTC), 198, 262, 295 Colony stimulating factor-1 (CSF-1), 124 Complement binding protein (CBP), 337 C-terminal domain (CTD), 224 CXCL12 and CXCR4 angiogenesis, 232–233 breast cancer progression and metastasis chemokine regulation, 227, 228 EGFR/HER2, 227 immunohistochemistry, 225, 226 lymph node metastasis, 227 hepatocellular carcinoma, 230 inhibitors AMD3100, 233 CTCE-9908, 234 Slit, 234–235 TN14003, 234 lung cancer growth and metastasis, 229–230 renal cell carcinomas, 230 signaling pathways breast cancer cells, 224, 225 Cbl protein, 222–223 EGFR phosphorylation, 224–225 GPCR, 222 p53, 224 prostate cancer cells, 223–224 tumor stroma, 231–232 Cyclooxygenase 2 (COX-2), 350

D Death domains (DDs), 156 5a-Dihydrotestosterone (5a-DHT), 279 Disseminated tumor cells (DTCs), 213, 262 Distant metastases free survival (DMFS), 6–7 DNA methyltransferases (DNMTs), 42 Ductal carcinoma in situ (DCIS), 225

E Early Breast Cancer Trialists’ Collaborative Group (EBCTCG), 5 E-cadherin, 18–19 ECM. See Extracellular matrix (ECM) EMT. See Epithelial-mesenchymal transition (EMT) Endothelial cells (ECs), 333

Index Endothelium circulating tumor cells, 201, 204–206 extravasation in vivo, 201–203 specific cancer-endothelial cell interactions, 206–207 Epidermal growth factor (EGF) and Ras cofilin, 27 ErbB family, 26 human prostate and breast cancer cells, 27 plasma membrane, 28 Raf-MAPK pathway, 29 signaling pathways, 26–27 Epidermal growth factor receptor (EGFR), 224 Epithelial cell transforming sequence 2 oncogene (Ect2), 150–151 Epithelial cellular adhesion molecule (EpCAM), 214 Epithelial-mesenchymal transition (EMT) basement membrane, 14 biomarkers and functional significance cytoskeleton dynamics changes, 19–21 ECM remodeling (see Extracellular matrix (ECM)) epithelial cell-cell junctions disruption, 18–19 cell signaling networks EGF and Ras, 26–29 HGF and FGF, 29–30 receptor kinases, 23 TGF-beta, 24–26 Wnt, Notch and Hedgehog, 30–33 cellular phenotype, 14 cobblestone network and apical-basal polarity, 15 CXCR4 CTD, 224 epithelial cells, 14–15 metastatic cancer, 15 metastatic cells, 13 MET conversions, 17 migratory phenotype and invasion, 16 miRNAs (see MicroRNAs (miRNAs)) N-cadherin and vimentin, 212 polarized epithelial cell, 212 prostate cancer cells, 253 solid human tumors, 15 squamous cell carcinomas, 16 transcriptional regulators HMGA2 and PcG (see High mobility group A2 protein (HMGA2) and Polycomb-Group Proteins (PcG)) homeodomain proteins, 39–41 Snail and Slug, 33–35 Twist1 and Twist2, 37–39 Zeb1 and Zeb2, 35–37

371 tumor cell invasion, 75 tumor metastasis, 13 type-3, 16 type-1 and type-2, 16 Erythroblast transformation specific (ETS), 288 Express green fluorescent protein (eGFP), 265 Extracellular matrix (ECM), 363 alpha5beta1 integrin, 21–22 CXCL12/CXCR4, 231 epithelial tissues, 21 fibronectin, 21 integrins bind, 311 laminin, collagen, and fibronectin, 254 lung microenvironment, 332 metastatic progression, 23 MMPs, 22–23 type-1 and type-2 EMT, 22

F Fibroblast growth factor (FGF), 29–30 Fibroblast growth factor inducible 14 (Fn14), 156–157 Fibroblast Specific Protein-1 (FSP1), 19–21 Fibronectin (FN), 78–79 Focal adhesion kinase (FAK), 177, 222, 313

G Gamma-interferon Activation Sequence (GAS), 247 Glioblastoma multiforme (GBM), 145 Glioma cancer stem cell function, 323–324 Glioma cell invasion Dock180-ELMO1 bipartite Rac1 GEF, 155 cancer cell invasion, 156 couples receptor signaling-rac, 154–155 human gliomas, 155 immunohistochemical analysis, 155 Rac1-promoted cell motility, 153–154 Dock family members, 152–153 Ect2, 150–151 Fn14, 156–157 GEFs, 147–148 glioblastoma treatments, 159 Rac1, 147 Rho GTPases, 145–147 Trio and SWAP-70, 149–150 TROY, 158 Vav3, 151–152 Glycosaminoglycan (GAG), 337 Gonadotropinreleasing hormone (GnRH), 280

372 G-protein coupled receptor (GPCR), 222 Green fluorescent protein (GFP), 349 GTPase activating protein (GAP), 268 Guanine nucleotide exchange factors (GEFs), 147–149

H Heat shock proteins (HSPs), 279, 280 Hepatocyte growth factor (HGF), 29–30, 177 High mobility group A2 protein (HMGA2) and Polycomb-Group Proteins (PcG) Bmi-1, 42 DNA and chromatin structure, 41 DNMTs, 42 E-cadherin, 43 ectopic expression, 41 Ezh2, 43 human cancers, 41 mesenchymal cell phenotype, 41 PRC1 genes, 42 protein–protein interactions, 41 Histone acetyltransferases (HATs), 294 Human epidermal growth factor receptor 2 (HER2), 224 Hypoxia acute and chronic, 174 HIF-1 signaling and metastasis, 174–175 metastasis impacts, 173 Hypoxia-inducible factor 1 (HIF-1), 173, 227

I Immunohistochemistry (IHC), 292 Insulin like growth factor-1 receptor (IGF-1R), 225 Integrin linked kinase (ILK), 313 Intercellular adhesion molecule (ICAM)-1, 199, 333

K Krüppel-like-factor 5 (KLF5), 288

L Leukocyte vs. cancer cell extravasation, 199–201 Lewis Lung Carcinoma (LLC) cells, 96 Ligand binding domain (LBD), 279 Liver microvasculature, 209 Low expression regions (LER), 76

Index Lung microvasculature, 209–210 Luteinizing hormone (LH), 280 Lysyl oxidase (LOX), 124

M Mammary epithelial cells (MECs), 123 Matrix metalloproteinases (MMPs), 22–23, 254, 289 Mesenchymal-epithelial transition (MET), 15–16 Metastatic breast cancer bone remodeling, 349–350 osteoblast apoptosis, 352 osteoblast cytokine responses bone marrow microenvironment, 357–360 cells interact, 354 MCP-1, 355 murine VEGF, 355–356 osteoblast inflammatory response, 353–354 skeleton, 349 three-dimensional model bioreactor, 361–362 BRMS1 cells, 363 confocal microscopy, 363 confocal Z-stack images, 364 3D culture system, 365 ECM, 364 gene expression, 365 myeloma cells, 364 single-cell filing, 363 vicious cycle, 351 Methylguanine methyltransferase (MGMT), 322–323 MicroRNAs (miRNAs) aberrant activation, 45 AP-1 and Zeb1 transcription factors, 49 DICER, 44 Drosha microprocessor complex, 45 gene-regulatory molecules, 43 human genome, 43 Let-7/miR-98 family, 47–48 Lin28A, Lin28B and RPS2, 45 mammary epithelial cells via transcriptional activation, 48 miRNA-200 family, 46–47 miR-21 transcription, 49 Myc expression, 45 oncogenic and tumor suppressor pathways, 44 precursor miRNAs, 44 primary transcripts, 44

Index RISC, 44 TGF-beta stimulation, 48 tumorigenesis, 45 untranslated regions, 44 Monocyte chemotactic protein 1 (MCP-1), 334, 349 Myeloid derived suppressor cells (MDSCs), 231

N Non-small cell lung cancer (NSCLC), 229 Nuclear Co-Repressor 1 (NCoR1), 284 Nuclear factor-kB (NF-kB), 334

O Osteoblast cytokine responses bone marrow microenvironment breast cancer cells, 357 MDA-MB231BRMS1 GFP, 358 mice femurs, 359–360 MIG levels, 359 mouse cytokines, 357 cells interact, 354 MCP-1, 355 murine VEGF, 355–356 Osteoprotegerin (OPG), 263, 350

P Parathyroid hormone related peptide (PTHrP), 263, 350 Phosphatidylinositol 3’-kinase (PI3K), 268 Phospho-glycerate kinase (PGK), 233 Phospholipase C (PLC), 268 Platelet derived growth factor (PDGF), 353 Polycomb-group (PcG), 293 Polycomb repressive complex 1 and 2 (PRC1 and PRC2), 42 Primary tumor microenvironments hypoxia, 125 mechanotransduction CSF-1, 124 3D-organotypic systems, 123 FAK expression, 124 LOX, 124 MECs, 123 phenotypes dominate genotypes, 123 TGF-b Paradox, 125 tumor fibrosis, 123 metastatic progression stimulation, 127–129 reactive tumor stroma, 126–127

373 Prolactin receptor (PrlR), 247 Prostate cancer cells bone marrow microenvironment, 263 bone-metastatic cancer, 269 early stages, 266–267 metastatic process and bone-tropism, 262–263 osteoclast bone-resorption activity bisphosphonates, 264 bone metastases, 265 OPG, 263 RANK activation, 263–264 skeletal foci, 265 ZA, 264 PDGFR expression, 267–268 PDGFR structure and signaling, 268 skeletal metastasis, 270 Prostate cancer progression and metastasis AKR1C3 and SRD5A1/2, 284 androgen-regulated miRNAS, 292 AR, 279–280 AR and chemokine receptors, 286–288 AR-dependent chromosomal translocations clinical implications, 291–292 TMPRSS2, 289–291 AR, oncogenes, and metastasis c-MYC (MYC), 294–296 c-SRC, 296–297 characteristics, 285 CRPC, 281–282 cyclin D1, 284 LBD mutations, 283 metastatic disease, 301–302 therapeutic intervention direct AR targeting, 299–301 indirect AR targeting, 299 treatment, 280–281 Prostate specific antigen, (PSA), 280 Protein Inhibitors of Activated Stat proteins (PIAS), 249 Protein tyrosine phosphatases (PTP), 249

R Reactive oxygen species (ROS), 333 Receptor Activator of NF-kB ligand (RANKL), 263 Receptor tyrosine kinases (RTK), 33, 312 Related adhesion focal tyrosine kinase (RAFTK), 222 Rho family GDPD issociation inhibitor 2 (RhoGDI2), 335 RNA-induced silencing complex (RISC), 44

374 S S100A4 and tumor invasion cancer prognostic marker, 94–95 extracellular annexin A2, 104 corneal vascularization model, 103 human plasma and synovial fluid, 104 oligomerization, 104 tumor interstitial fluid, 103 intracellular targets cytoskeletal elements, 101–102 myosin-II, 99–101 p53 expression, 102–103 LLC, 96 mammary epithelium, 96 metastatic process, 95 molecule-induced oligomers, 105–106 pharmaceutical industry, 105 S100 protein family Ca 2+-binding proteins, 91 calcium, 94 calmodulin and troponin C, 94 human chromosome, 91 human pathologies, 92 monomer, 92–93 pseudo EF-hand, 92, 94 tumor cell motility, invasion, and angiogenesis, 91–92 target protein interactions, 98–99 therapeutic intervention, 105 transgenic mice, 96 tumor cell proliferation, 107 tumor microenvironment, 96–98 Skeletal related events (SREs), 264 Small cell lung cancer (SCLC), 229, 234 Src family kinases (SFK), 296, 313 Stat5a/b and Stat3, prostate cancer cadherin switching, 253 castration resistant growth, 246 disease progression, 250 E-cadherin, 254 growth promotion, 251–252 metastatic behavior, 252 MMPs, 254–255 promote metastasis formation, 261–253 stat signaling regulators, 249 structure and function, 246–248 transcription factors, 246 TWIST expression, 254 Steroid receptor coactivator (SRC), 283 Stromal derived factor-1 (SDF-1), 233 Suppressors of Cytokine Signaling (SOCS), 249

Index T TGF-b. See Transforming Growth Factor-b (TGF-b) Thomas Friedenreich (TF) antigen, 206 Thyroid receptor a (TRa ), 318 Transcriptional activation (TA), 247 Transendothelial migration (TEM), 198 Transforming growth factor-b (TGF-b) animal models, 121–122 canonical TGF-b signaling, 118 carcinoma cells, 131–132 3D-organotypic models, 123, 125 E-cadherin and tight junction proteins, 25 EMT, 120–121 human tumors, 117 lymphatic and circulatory systems, 129 mammospheres, 129 matrix-bound factors, 351 matrix remodeling, 25 metastatic, EMT and chemoresistant phenotypes, 117 metastatic progression, 118–120 multifunctional cytokin, 24 neoplastic transformation, 116 noncanonical signaling systems, 118 ovarian cancer cells, 25 oxygen tensions and elastic moduli, 127 primary tumor microenvironments (see Primary tumor microenvironments) pulmonary microenvironment, 130 skeletal microenvironment, 131 Smad proteins, 24 Smad2/3 signaling, 118 Smurf E3 ubiquitin ligases, 118 TbR-III expression, 117 TGF-b ligands, 117 transgenic mouse models, 26 transmembrane signaling, 117 tumor modeling, 26 Transgenic adenocarcinoma of the mouse prostate (TRAMP) model, 296 Tumor cells cell-cell adhesion molecules, 76 cellular and non-cellular components, 73 collagen and fibronectin, 74 E-cadherin, 73 EMT, 75 FN, 78–79 heterodimeric transmembrane receptors, 77 HIF-1, c-Src and c-Met signaling pathways, 181–182 hypoxia acute hypoxia and chronic hypoxia, 174

Index HIF-1 signaling and metastasis, 174–175 metastasis impacts, 173–174 LER, 76 malignant tumors, 73 matrix-modifying enzymes, 77 melanoma and endothelial cells, 79 metastatic cascade, 73–74 metastatic phenotype, 171 metastatic process, 186–187 met signaling and metastasis c-Met, invasive growth and metastasis, 179 HGF/c-Met signaling pathways, 179–181 human cancers, 178 receptor tyrosine kinase c-Met, 177 pre-metastatic niches, 76 proteolytic enzymes, 73 signaling cascades, 79 Src signaling and metastasis metastasis-associated functions, 175–176 Src signaling pathways, 177 stromal fibroblasts and inflammatory infiltrates, 73 therapeutic interventions cancer cell dissemination, 186 hypoxia and HIF-1, 183–184

375 Met Kinase, 183–186 Src Kinase, 183–185 tumor microenvironment, 75, 171 Tumor necrosis factor (TNF), 156 Tumor necrosis factor-like weak inducer of apoptosis (TWEAK), 157 Tumor necrosis factor receptor associated factor (TRAF), 156 Tumor necrosis factor receptor superfamily (TNFRSF) members, 156 Tyrosine-kinase-binding (TKB), 222

U urokinase-type plasminogen activator receptor (uPAR), 223 Urothelial cancer (UC), 331

V Vascular cell adhesion molecule (VCAM)-1, 333 Vascular endothelial growth factor (VEGF), 233 Vertebral-cancer of the prostate (VCaP), 289 von Hippel Lindau (VHL), 174, 228

Z Zoledronic acid (ZA), 264