Current Topics in Developmental Biology, Volume 66 (Current Topics in Developmental Biology)

262

Contents of Previous Volumes

Volume 66 1. Stepwise Commitment from Embryonic Stem to Hematopoietic and Endothelial Cells Changwon Park, Jesse J. Lugus, and Kyunghee Choi

2. Fibroblast Growth Factor Signaling and the Function and Assembly of Basement Membranes Peter Lonai

3. TGF-b Superfamily and Mouse Craniofacial Development: Interplay of Morphogenetic Proteins and Receptor Signaling Controls Normal Formation of the Face Marek Dudas and Vesa Kaartinen

4. The Colors of Autumn Leaves as Symptoms of Cellular Recycling and Defenses Against Environmental Stresses Helen J. Ougham, Phillip Morris, and Howard Thomas

5. Extracellular Proteases: Biological and Behavioral Roles in the Mammalian Central Nervous System Yan Zhang, Kostas Pothakos, and Styliana-Anna (Stella) Tsirka

6. The Genetic Architecture of House Fly Mating Behavior Lisa M. Meffert and Kara L. Hagenbuch

7. Phototropins, Other Photoreceptors, and Associated Signaling: The Lead and Supporting Cast in the Control of Plant Movement Responses Bethany B. Stone, C. Alex Esmon, and Emmanuel Liscum

8. Evolving Concepts in Bone Tissue Engineering Catherine M. Cowan, Chia Soo, Kang Ting, and Benjamin Wu

9. Cranial Suture Biology Kelly A Lenton, Randall P. Nacamuli, Derrick C. Wan, Jill A. Helms, and Michael T. Longaker

Volume 67 1. Deer Antlers as a Model of Mammalian Regeneration Joanna Price, Corrine Faucheux, and Steve Allen

Series Editor Gerald P. Schatten Director, PITTSBURGH DEVELOPMENTAL CENTER Deputy Director, Magee-Women’s Research Institute Professor and Vice-Chair of Ob-Gyn Reproductive Sci. & Cell Biol.-Physiology University of Pittsburgh School of Medicine Pittsburgh, Pennsylvania 15213

Editorial Board Peter Gru¨ss Max-Planck-Institute of Biophysical Chemistry Go¨ttingen, Germany

Philip Ingham University of Sheffield, United Kingdom

Mary Lou King University of Miami, Florida

Story C. Landis National Institutes of Health National Institute of Neurological Disorders and Stroke Bethesda, Maryland

David R. McClay Duke University, Durham, North Carolina

Yoshitaka Nagahama National Institute for Basic Biology, Okazaki, Japan

Susan Strome Indiana University, Bloomington, Indiana

Virginia Walbot Stanford University, Palo Alto, California

Founding Editors A. A. Moscona Alberto Monroy

Contributors Numbers in parentheses indicate the pages on which the authors’ contributions begin.

Kyunghee Choi (1), Developmental Biology Program and Molecular Cell Biology Program, Washington University School of Medicine, Department of Pathology and Immunology, St. Louis, Missouri 63110 Catherine M. Cowan (239), Department of Bioengineering, University of California Los Angeles, Los Angeles, California 90095 Marek Dudas (65), Developmental Biology Program at the Saban Research Institute of Children’s Hospital Los Angeles, Los Angeles, California 90027 and Department of Pathology, Keck School of Medicine, University of Southern California, Los Angeles, California 90089 C. Alex Esmon (215), University of Missouri–Columbia, Columbia, Missouri 65211 Kara L. Hagenbuch (189), Department of Ecology and Evolutionary Biology, Rice University, Houston, Texas 77251-1892 Jill A. Helms (287), Children’s Surgical Research Program, Division of Plastic and Reconstructive Surgery, Department of Surgery, Stanford University School of Medicine, Stanford, California 94305-5148 Vesa Kaartinen (65), Developmental Biology Program at the Saban Research Institute of Children’s Hospital Los Angeles, Los Angeles, California 90027 and Department of Pathology, Keck School of Medicine, University of Southern California, Los Angeles, California 90089 Kelly A Lenton (287), Children’s Surgical Research Program, Division of Plastic and Reconstructive Surgery, Department of Surgery, Stanford University School of Medicine, Standford, California 94305-5148 Emmanuel Liscum (215), University of Missouri–Columbia, Columbia, Missouri 65211 {

{

Peter Lonai (37), Department of Molecular Genetics, The Weizmann Institute of Science, Rehovot, Israel 76100

Deceased.

ix

x

Contributors

Michael T. Longaker (287), Children’s Surgical Research Program, Division of Plastic and Reconstructive Surgery, Department of Surgery, Stanford University School of Medicine, Stanford, California 94305-5148 Jesse J. Lugus (1), Molecular Cell Biology Program, Washington University School of Medicine, Department of Pathology and Immunology, St. Louis, Missouri 63110 Lisa M. Meffert (189), Department of Ecology and Evolutionary Biology, Rice University, Houston, Texas 77251-1892 Phillip Morris (135), Institute of Grassland and Environmental Research, Plas Gogerddan, Aberystwyth, Ceredigion, SY23 3EB, Wales, United Kingdom Randall P. Nacamuli (287), Children’s Surgical Research Program, Division of Plastic and Reconstructive Surgery, Department of Surgery, Stanford University School of Medicine, Stanford, California 94305-5148 Helen J. Ougham (135), Institute of Grassland and Environmental Research, Plas Gogerddan, Aberystwyth, Ceredigion, SY23 3EB, Wales, United Kingdom Changwon Park (1), Developmental Biology Program, Washington University School of Medicine, Department of Pathology and Immunology, St. Louis, Missouri 63110 Kostas Pothakos (161), Department of Pharmacological Sciences, State University of New York at Stony Brook, Stony Brook, New York 11794-8651 Chia Soo (239), Weintraub Center for Reconstructive Biotechnology, University of California Los Angeles, Los Angeles, California 90095 and University of Southern California, Keck School of Medicine, Division of Plastic Surgery, Los Angeles, Calfornia 90053 Bethany B. Stone (215), University of Missouri–Columbia, Columbia, Missouri 65211 Howard Thomas (135), Institute of Grassland and Environmental Research, Plas Gogerddan, Aberystwyth, Ceredigion, SY23 3EB, Wales, United Kingdom Kang Ting (239), Weintraub Center for Reconstructive Biotechnology, University of California Los Angeles, Los Angeles, California 90095 Styliana-Anna (Stella) Tsirka (161), Department of Pharmocological Sciences, State University of New York at Stony Brook, Stony Brook, New York 11794-8651

Contributors

xi

Derrick C. Wan (287), Children’s Surgical Research Program, Division of Plastic and Reconstructive Surgery, Department of Surgery, Stanford University School of Medicine, Stanford, California 94305-5148 Benjamin Wu (239), Department of Bioengineering, University of California Los Angeles, Los Angeles, California 90095 Yan Zhang (161), Department of Pharmocological Sciences, State University of New York at Stony Brook, Stony Brook, New York 11794-8651

In Memorium, Professor Peter Lonai

August 11, 1936–November 6, 2004

Professor Peter Lonai was a scientist with wide knowledge and perspectives in biology. He had the courage and vision to enter into new fields before his peers realized the importance of these topics. He had deep understanding in classical developmental biology, and many scientists from the two faculties of biology at the Weizmann Institute of Science consulted him on these topics. His work on early developmental steps in the mouse embryo was highly appreciated in the field. This includes his studies on epithelial– mesenchymal interactions, the FGF signaling, and most recently, the laminin-dependent mechanisms that regulate endodermal and ectodermal embryonic stem cell fates. Peter was a loved colleague in the Weizmann Institute’s Department of Molecular Genetics, an intellectual, and a real gentleman. He was among the founders of the transgenic/knock-out facilities in the Institute. In fact, his knowledge and expertise initiated this important operation that helped many groups at the Weizmann Institute. Science meant everything to him, and he continued to come to the lab until the last days of his life. His last review is the best documentation of his enormous dedication to science. Professor Adi Kimchi

xiii

1

____________________________________________________________________________

Stepwise Commitment from Embryonic Stem to Hematopoietic and Endothelial Cells Changwon Park,* Jesse J. Lugus,{ and Kyunghee Choi*,{ *Developmental Biology Program {

Molecular Cell Biology Program Washington University School of Medicine Department of Pathology and Immunology St. Louis, Missouri 63110

I. Embryonic Stem Cell A. Signaling Pathways Regulating Embryonic Stem Cell Self-Renewal B. Transcriptional Control of Embryonic Stem Cell Self-Renewal II. From ES to Hematopoietic Progenitors A. An Overview of Hematopoietic Development B. Hemangioblast C. The Identification of Blast Colony–Forming Cells from In Vitro DiVerentiated Embryonic Stem Cells D. From Flk-1-Expressing Mesoderm to Hematopoietic and Endothelial Cells E. Hematopoietic Inductive Signals F. Transcriptional Control of Hematopoietic and Endothelial Cell Lineage Commitment G. In Vivo Potential of Embryonic Stem Cell–Derived Hematopoietic Progenitors III. Conclusions and Future Directions Acknowledgments References

There is great excitement in generating diVerent types of somatic cells from in vitro diVerentiated embryonic stem (ES) cells, because they can potentially be utilized for therapies for human diseases for which there are currently no eVective treatments. Successful generation and application of ES-derived somatic cells requires better understanding of molecular mechanisms that regulate self-renewal and lineage commitment. Accordingly, many studies are aimed toward understanding mechanisms for maintaining the stem cell state and pathways leading to lineage specification. In this chapter we discuss recent studies that examine molecules that are critical for ES cell self-renewal, as well as hematopoietic and endothelial cell lineage diVerentiation from ES cells. C 2005, Elsevier Inc.

Current Topics in Developmental Biology, Vol. 66 Copyright 2005, Elsevier Inc. All rights reserved. 0070-2153/05 $35.00

1

2

Park et al.

I. Embryonic Stem Cell In 1981 investigators successfully derived pluripotent embryonic stem (ES) cells from blastocysts, the preimplantation stage of mouse embryos (Evans and Kaufman, 1981; Martin, 1981). Subsequently, ES cell lines from many diVerent species, including human, have been derived (Thomson et al., 1995, 1996, 1998). The derivation of ES cells is quite straightforward in that blastocysts are plated onto a feeder layer of fibroblasts and the inner cell mass of blastocysts ultimately gives rise to colonies of undiVerentiated cells (ES cell colonies), which are isolated and further expanded. Once established, ES cells can be maintained as pluripotent stem cells on a feeder layer of fibroblasts. When introduced back into a blastocyst, ES cells can contribute to all tissues with the exception of extraembryonic endoderm and trophoblast of the developing embryo (Beddington and Robertson, 1989; Bradley et al., 1984). It is this particular trait that makes ES cells a valuable tool for genetic engineering. In addition, ES cells can be diVerentiated in vitro into many diVerent somatic cell types (Bagutti et al., 1996; Bain et al., 1995; Buttery et al., 2001; Dani et al., 1997; Doetschman et al., 1985; Drab et al., 1997; Fraichard et al., 1995; Keller et al., 1993; Kramer et al., 2000; Liu et al., 2000; Maltsev et al., 1993, 1994; Nakano et al., 1994; Potocnik et al., 1994; Risau et al., 1988; Rohwedel et al., 1994; Strubing et al., 1995; Vittet et al., 1996; Wang et al., 1992; Wiles and Keller, 1991; Yamashita et al., 2000), opening up the possibility to utilize ES-derived cells as a potential source for cell transplantation or cell-based therapy (Fig. 1). Thus, ES cells have gained much scientific and general public attention. A. Signaling Pathways Regulating Embryonic Stem Cell Self-Renewal In 1987 Smith and Hooper discovered that buValo rat liver cells (BRLCs) secreted a substance into the media that could maintain ES cells as pluripotent in the absence of a feeder layer (Smith and Hooper, 1987). Previously, when removed from a feeder layer, ES cells diVerentiated into extraembryonic endoderm. However, the BRLC-conditioned media possessed a Self-renewal

Differentiation

ES cells Embryoid bodies Figure 1 Schematic diagram of in vitro diVerentiation of ES cells. Embryoid bodies (EBs) are composed of many diVerent cell types, as indicated by diVerent colors.

1. Hematopoietic and Endothelial Cells

3

diVerentiation-inhibiting activity (DIA) that prevented the diVerentiation of ES cells in the absence of a feeder layer. Furthermore, this activity could be dialyzed out from the media, showing that soluble factor(s) could maintain ES cell self-renewal without any interaction with feeder cells. The same year, Moreau and coworkers reported the discovery of a novel human interleukin (IL) that induced the proliferation of the murine DA-1 early myeloid cell line and was given the moniker HILDA for human interleukin DA responsive cytokine (Moreau et al., 1987b). Later work showed that HILDA was also an activator of eosinophils, as well as a potent inducer of burst-promoting activity on human marrow (Moreau et al., 1987a). Leukemia inhibitory factor (LIF) first emerged as a cytokine that was able to suppress the proliferation and drive the diVerentiation of M1 myeloid leukemia cells (Gearing et al., 1987). Subsequently, work emerged that tied LIF, DIA, HILDA, and the maintenance of ES cell pluripotency together. First, Williams and coworkers noted that partially purified DIA and LIF had a number of similarities (Williams et al., 1988). The group discovered that in the absence of feeder cells, LIF associated with the cell membranes of a number of ES and embryonal carcinoma (EC) cell lines and was able to maintain >95% of ES cells in ES cell colonies. Moreover, using ES cells maintained in purified, recombinant LIF for up to 22 passages, chimeric animals were successfully produced and some showed up to 90% chimerism. Smith and colleagues then showed that 10 ng/ml of either DIA or HILDA/ LIF was suYcient to suppress any type of diVerentiation of CP1 ES cells (Smith et al., 1988). The final piece of the puzzle came from Moreau and coworkers when they showed that the genes for LIF, HILDA, and DIA were one and the same (Moreau et al., 1988). LIF is a ligand for a heterodimeric receptor composed of LIF receptor (LIFR ) and the gp130 cytokine receptor (Davis et al., 1993). This ligand–receptor complex then activates Janus-associated tyrosine kinases (Jak), which phosphorylate the receptor chains. The phosphorylated receptors, in turn, serve as docking sites for Srchomology 2 (SH2) domains of additional proteins that may also be phosphorylated by the Jaks. A key substrate of the Jaks is the signal transducer and activator of transcription (Stat) family of transcription factors. Specifically, the LIFR –gp130 complex is an eVector of Stat3 phosphorylation and dimerization, inducing nuclear translocation and subsequent transcription (Boeuf et al., 1997; Niwa et al., 1998). Stat3 recruitment and activiation are integral to ES cell self-renewal, because a dominant negative Stat3 isoform rapidly induces ES cell diVerentiation (see the following). More recent work has given insight into other important mechanisms of ES cell self-renewal (Qi et al., 2004; Ying et al., 2003). Ying and coworkers, noting the antagonist eVects of bone morphogenetic proteins (BMPs) on neural diVerentiation, showed that in the absence of serum, BMP-4 provides a synergistic eVect to LIF in the maintenance of ES cell phenotype, because

4

Park et al.

the eVect of BMP-4 was additive to that of LIF alone (Ying et al., 2003). Moreover, E14Tg2a ES cells passaged six times in serum-free N2B27 media with LIF and BMP-4 could generate chimeric animals. The authors went on to show that BMP-4 does not signal through the LIFR –gp130 complex nor does it activate Stat3. Instead, BMP-4 activates the Smad pathway to induce ES cell self-renewal and activate members of the Id family. Additionally, in this study, the authors also showed that BMP-4 activated the Erk pathway. The Id genes encode helix-loop-helix (HLH) factors that antagonize transcriptional activation by basic helix-loop-helix (bHLH) transcription factors. The Ids bind bHLH transcription factors, preventing bHLH heterodimerization and DNA binding to suppress transcriptional activity. Gene expression showed that BMP-4 is capable of inducing up to tenfold increases in both Id1 and Id3 gene expression. The authors posit that the Ids are integral to the suppression of ES cell diVerentiation. For example, the Id-mediated suppression of ES cell diVerentiation can be overcome through expression of supra-physiological levels of E47. The authors theorize that high levels of E47 successfully out-compete the Ids to interact with Neurogenin2 and drive ES cells to a neurogenic fate. Similarly, Qi et al. (2004) reported the synergistic eVect of BMP-4 and LIF on ES cell self-renewal. However, in this study they showed that BMP-4 could keep the pluripotency of ES cells by inhibiting both Erk and p38 mitogen-activated protein kinases (MAPKs). Importantly, in the presence of MAPK inhibitors such as PD98059 and SB203580 (type IA receptor for BMPs), they were able to establish ES cell lines from Alk-3
/
blastocysts, which normally fail to expand and form ES colonies. Lastly, work from Anneren et al. has shown the importance of a nonreceptor tyrosine kinase in maintenance of ES cell pluripotency (Anneren et al., 2004). The authors examined the Src tyrosine kinase cYes, found to be expressed at high levels in multiple types of stem cells (Ivanova et al., 2002; Ramalho-Santos et al., 2002), for its role in ES cell self-renewal. The authors found that inhibiting cYes had no antagonistic eVect on Jak, Erk, or Stat3 phosphorylation, but expression of the ES cell marker Nanog was lost. Additionally, in cells treated with both a synthetic Src inhibitor (SU6656) and retinoic acid, an inducer of ES cell diVerentiation, ES cells displayed synergistic diVerentiation cues from the two molecules, demonstrating that cYes is an important molecule in maintaining ES cell self-renewal. B. Transcriptional Control of Embryonic Stem Cell Self-Renewal 1. Stat3 The third mammalian gene cloned of the signal transducers and activators of transcription (Stat) family was originally identified as an IL-6-activated transcription factor. All Stats share a phosphotyrosine-binding, SH2

1. Hematopoietic and Endothelial Cells

5

domain. Work by Boeuf et al. (1997) and Niwa et al. (1998) has shown that one of the most important ES cell self-renewal cytokines, LIF, signals through the LIFR –gp130 dimer complex to activate both the Shp2–Erk and Jak–Stat pathways. Through generation of chimeric receptors, Niwa et al. (1998) showed that Stat3 docking sites on the cytoplasmic region of gp130 are necessary to maintain ES cell self-renewal, implicating Stat3 in the modulation of transcriptional activity. Additionally, these chimeric receptors showed high levels of Erk activity upon receptor stimulation but that addition of the Mek inhibitor PD098059 did not inhibit stem cell colony formation, demonstrating that Erk pathway activation is not required for ES cell self-renewal. Lastly, use of the dominant negative mutant Stat3F, which has a tyrosine substitution at Tyr705 to phenylalanine (Y705F) and is incapable of phosphorylation, dimerization, and nuclear translocation, indicated that Stat3 is necessary for stem cell colony formation because no colonies could be generated from cells harboring a ‘‘supertransfected’’ episome expressing the Stat3F cDNA. 2. Nanog Nanog, or Enk, was originally identified in a screen for homeobox-containing transcripts via degenerate oligonucleotide polymerase chain reaction in a murine ES cell-derived cDNA library (Wang et al., 2003). The authors sought to identify transcripts of the Nk-2 family, which contains members such as the cardiogenic factor Nkx2–5. In this report the authors mapped the gene locus and determined the genetic architecture and showed that Nanog expression in developing embryos was first detected in compacted morulae, localized to interior cells, the future site of the inner cell mass (ICM). Nanog expression was further specified to the epiblast and absent from the primitive endoderm. Later, work from two groups (Chambers et al., 2003; Mitsui et al., 2003) further characterized the gene under the Nanog name. Several new observations came about from these two papers. Mitsui et al. (2003) used digital diVerential display to identify expressed sequence tags (ESTs) preferentially expressed in ES cells. Herein, they discovered that overexpression of Nanog under control of the chicken actin promoter/CMV-IE enhancer yielded ES cells that did not diVerentiate upon LIF withdrawal. The authors also demonstrated that Nanog expression is lost upon exposure to retinoic acid. Through targeted disruption of Nanog via introduction of a -Gal-Neo fusion gene, the authors demonstrated that Nanog
/
ES cells lost their pluripotency and went down an endodermal diVerentiation pathway, expressing markers of the pariental (LamininB1, Dab2) and visceral endoderm (Bmp2, Ihh), as well as endoderm-specific transcription factors (Hnf4, Gata6). The authors generated Nanog heterozygous animals through blastocyst injection of Nanogþ/
ES cells, but Nanog
/
embryos were found to be

6

Park et al.

lethal by embryonic day (E) 5.5. Lastly, the DNA recognition sequence of Nanog was determined and was shown to be highly divergent from the consensus DNA-recognition sequence of other murine Nk-2 factors, demonstrating that Nanog is indeed distinct from the Nk-2 family. Using a ‘‘supertransfected’’ episomal expression screen, Chambers et al. (2003) identified Nanog in parallel as a self-sustaining factor in ES cells, capable of bypassing the LIF requirement for self-renewal. The relationship between the Jak–Stat pathway and Nanog was examined by using the inhibitor of Jak activity D6665 to show that Nanog-mediated self-renewal of ES cells is not Jak–Stat dependent, and showing that addition of LIF to cells carrying a Nanog episome augmented their ability to form ES cell colonies. Additionally, because LIF signaling is enhanced by inhibition of the Erk mitogenactivated kinase pathway (Burdon et al., 1999), the Ras–Erk pathway was stimulated and there was no change in the Nanog expression upon Erk stimulation. Lastly, Nanog expression was examined in Oct4
/
embryos and it was determined that Nanog was expressed even in the absence of Oct4. 3. Oct3/4 To study transcription factors that are active in early mammalian development, Rosner and colleagues (1990) screened a cDNA library generated from F9 EC mRNA with a probe from the homeobox region of Oct2 at low stringency. All cDNAs isolated were from the same murine gene, which was given the name Oct3 due to the presence of an octamer sequence that was originally identified in the gene products of Oct1 and Oct2 (Singh et al., 1986; Staudt et al., 1986). Oct3 also turned out to be the gene that encodes the previously identified nuclear factor NF-A3 (Lenardo et al., 1989). Previously, Oct4 had been identified by Scholer and colleagues (1989) in a screen for octamer binding proteins. It was then determined that the two groups were working on the same gene, and the names Oct3, Oct4, and Oct3/4 have been used to identify the same gene and gene product. Oct3/4 is preferentially expressed in unfertilized oocytes, the ICM of the blastocyst, primitive ectoderm in egg-cylinder-stage embryos, and primordial germ cells, in addition to pluripotent cells in the mouse (Pesce et al., 1998). Rosner and colleagues (1990) showed that upon diVerentiation of embryonal carcinoma cells with retinoic acid, Oct3/4 expression decreases. These data from various expression patterns suggested that Oct3/4 plays an important role in maintaining cellular pluripotency. Further work demonstrated that Oct3/4 encodes a transcription factor that specifically binds an octamer motif and is capable of inducing transcription in in vitro assays. Through targeted gene ablation (Nichols et al., 1998), it was shown that Oct3/4 is required to initiate fetal development, because Oct3/4 deficiency leads to inability for founder cells of the ICM to acquire pluripotency. Instead, these cells become diverted

1. Hematopoietic and Endothelial Cells

7

into trophectoderm lineage. Conditional expression/repression of Oct3/4 showed that a precise level of Oct3/4 is required, because three diVerent fates are seen to be dependent upon Oct3/4 expression. ES cells require a critical level to maintain stem cell self-renewal, and a more than twofold increase causes diVerentiation into mesoderm and endoderm, whereas a reduction to less than 50% leads to induction of trophectoderm (Niwa et al., 2000). Collectively, models have emerged where ES cell fate (selfrenewal, endoderm/mesoderm, or trophectoderm) is regulated by Oct3/4 dosage, stoichiometric interactions with trans-acting factors, and the presence of LIF (Niwa, 2001). To date, a number of Oct3/4 target genes have been identified. Oct3/4 can bind a number of target sequences, including the consensus octamer motif ATGCAAAT and the AT-rich sequence (Okamoto et al., 1990; Saijoh et al., 1996). The variation in sequence means that Oct3/4 DNA binding is accomplished through both interactions with other sequence-specific trans-acting factors and homo- and heterodimerization (Tomilin et al., 2000). Identified by their stem-cell specific expression as Oct3/4 target genes were Fgf-4 (Yuan et al., 1995), the transcriptional coactivator Utf-1 (Nishimoto et al., 1999), the Zn-finger transcription factor Rex1 (Ben-Shushan et al., 1998), and the platelet-derived growth factor  receptor (PDGFR) (Kraft et al., 1996). Other genes have been identified, and the common trait that these genes all have is ES cell-specific expression. Importantly, Oct3/4 has also been shown to demonstrate transcriptional repression activity toward several other transcription factors, including the caudal-related homeobox transcription factor Cdx-2 and the cardiac-specific bHLH transcription factor eHand (Niwa et al., 2000). Surprisingly, very little is known about the function of Oct3/4 target genes. The most heavily studied, Fgf-4, has been shown to be required for peri-implantation development yet is dispensable for stem cell self-renewal (Feldman et al., 1995). Clearly, future work is needed to ascertain the other trans-acting molecules that interact with Oct3/4, as well as the essential self-renewal ES cell genes that it activates.

II. From ES to Hematopoietic Progenitors A. An Overview of Hematopoietic Development The production of blood cells takes place in several distinct anatomical sites during mouse embryogenesis. Morphologically distinct primitive blood cells are first identifiable in the blood islands of the yolk sac at E7.5 of gestation. The liver rudiment is colonized by hematopoietic cells by E10.5 of the 20-day murine gestation period and thereafter becomes the principal fetal hematopoietic organ (Houssaint, 1981). Beginning at birth, bone marrow is

8

Park et al.

colonized by hematopoietic stem cells (HSCs) originating from the fetal liver. From birth and throughout adult life, all mature blood cells are produced in the bone marrow. The term primitive hematopoiesis is given to the initial yolk sac-derived erythroid lineage, whereas definitive hematopoiesis is applied to all lineages other than primitive erythroid (Keller et al., 1999). The origin of the HSCs that colonize the fetal liver has remained controversial. There are currently two models concerning the spatial origin of HSCs. The first model is that the shift in the hematopoietic sites reflects the migration of HSCs from the yolk sac to the fetal liver and from the fetal liver to the bone marrow. According to this model, the microenvironment of the yolk sac, fetal liver, or bone marrow will determine the developmental potential of the HSCs to produce primitive versus definitive blood. The second model is that the HSCs that establish fetal liver hematopoiesis develop within the intraembryoic para-aortic-splanchnopleure (PAS)/ aorta-gonad-mesonephros (AGM) region. Support for the first model comes from studies showing that the yolk sac contains multiple definitive hematopoietic progenitors including long-term repopulating cells, even though in situ the yolk sac appears to have limited potential in generating mature blood cells (Cumano et al., 1993; Huang and Auerbach, 1993; Huang et al., 1994; Liu and Auerbach, 1991; Wong et al., 1986). Recent studies by Palis and colleagues (1999) demonstrate that definitive erythroid progenitors develop within the yolk sac even before the circulation is established. In addition, high proliferative potential colony-forming cells (HPP-CFCs), which can generate definitive erythroid cells and macrophages, were first detected exclusively in the yolk sac at early somite stages (E8.25) (Palis et al., 2001). Furthermore, when E7 mouse yolk sac and embryos were cultured separately, hematopoietic cells developed only from the yolk sac (Moore and Metcalf, 1970). Many studies reinforce the notion that the yolk sac contains hematopoietic stem cells. For example, Weissman et al. (1978) transplanted yolk sac cells from E8–10 mouse embryos into the E8–9 yolk sac of recipient embryos. The donor cells could be identified by an H-2 haplotype or Thy-1 marker. In this experiment the authors saw a low level of donor cell contribution to lymphoid cells in the recipients. Huang and Auerbach (1993) have taken AA4.1þWGAbright yolk sac cells and reconstituted the myeloid and lymphoid compartments of lethally irradiated adult mice. In this study, bone marrow cells depleted of long-term repopulating cells were coinjected. Finally, Yoder et al. used busulfan-myeloablated neonatal mice as recipients to demonstrate that E9 or E10 yolk sac cells could successfully reconstitute the hematopoietic system (Yoder and Hiatt, 1997; Yoder et al., 1997a,b). These findings suggest that primitive hematopoietic stem cells require an embryonic environment and that the adult microenvironment may not support the diVerentiation of the primitive HSCs. Consistent with this

1. Hematopoietic and Endothelial Cells

9

interpretation, yolk sac-derived cells can reconstitute the adult hematopoietic system when precultured on AGM-derived stromal cells (Matsuoka et al., 2001). In vitro progenitor assays and long-term repopulation studies utilizing yolk sac cells clearly suggest that HSCs are present within the yolk sac. However, the timing of the emergence of HSCs within the yolk sac is still unknown. HSC activity can be measured only after the emergence of primitive erythroid populations. Many investigators are actively pursuing the possibility that the primitive erythroid progenitor could develop directly from the mesodermal cells. The notion that the definitive hematopoietic system originates from intraembryonic progenitors initially derives from the studies of avian embryos. When yolk sac chimeras between a quail embryonic body (from the E2 embryo) and the extraembryonic area of a chick (from the E2 embryo) were generated and the chimeras analyzed between E5 and E13, the intraembryonic organs were always of quail origin (Dieterlen-Lievre, 1975; Dieterlen-Lievre and Martin, 1981). Similarly, when chick–chick chimeras were generated and blood cells analyzed, adult hematopoietic cells were shown to be of intraembryonic origin. In these chick–chick chimeras, sex chromosomes, immunoglobulin allotypes, and major histocompatibility complexes were used to distinguish extraembryonic versus intraembryonic origin (Lassila et al., 1978). Accumulating studies now support the notion that HSCs colonizing the fetal liver originate within the mouse embryo, in an area called the PAS/AGM. The AGM gives rise to vascular, excretory, and reproductive tissues of the embryo (reviewed in Dzierzak, 1999). The earliest tissues to form in the AGM of the E8 embryos are the paired dorsal aortas (Kaufman, 1992). The aortas become connected to the yolk sac vasculature via the vitelline (omphalomesenteric) artery by E8.5. The paired aorta will fuse to form a single dorsal aorta by E9 (Garcia-Porrero et al., 1995). The umbilical artery forms the connection between the dorsal aorta and the placenta (Garcia-Porrero et al., 1995). The urogenital and gonadal systems mature within the AGM soon after the vascular system is established (Kaufman, 1992). When the PAS/AGM region from an E8.5–9 embryo is isolated and grafted under the kidney capsule in SCID mice, serum immunoglobulin M (IgM), IgM-secreting plasma cells, and B cells of the B1a phenotype of donor origin can be detected 3–6 months after the engraftment (Godin et al., 1993). Furthermore, the PAS/AGM region gives rise to both myeloid and lymphoid cells when cultured in vitro (Cumano et al., 1996). More importantly, lymphoid potential of the PAS/AGM region is found even before the circulation is established (Godin et al., 1995). Finally, the PAS/AGM region contains spleen colony-forming cells (CFU-S) and the AGM from precirculation stage or E10 embryos contains long-term repopulating HSCs (Cumano et al., 2001; Medvinsky and Dzierzak, 1996;

10

Park et al.

Medvinsky et al., 1993). By further dissecting the AGM and testing for longterm repopulating HSC activity, the aorta has been shown to be the initial site of adult HSC emergence (E10.5), followed by the vitelline and umbilical arteries (de Bruijn et al., 2000). These observations strongly argue that cells that colonize the fetal liver originate within the embryo.

B. Hemangioblast The origin of the blood cells that establish the yolk sac blood islands and the AGM is another area of active investigation. Studies over the last 100 years have shown that blood cells within yolk sac blood islands and the AGM develop in close proximity to the vascular system. It has to be noted, however, that the vascular system does not necessarily associate with hematopoiesis. In the yolk sac, mesodermal cells, which have migrated from the primitive streak, form aggregates to establish blood islands at around E7. Over the next 12 hours, the central cells within the blood islands generate primitive blood cells while the peripheral cells diVerentiate into endothelial cells. These blood islands subsequently fuse to form the first extraembryonic vascular network. The close developmental association of the hematopoietic and endothelial cell lineages within the yolk sac blood islands of the developing embryo has led to the hypothesis that they arise from a common precursor, termed the hemangioblast (Murray, 1932; Sabin, 1920; Wagner, 1980). Similarly, blood cells of the embryo proper develop in close association with the endothelium of the dorsal aorta (Dieterlen-Lievre, 1997; Garcia-Porrero et al., 1995; Tavian et al., 1996, 1999). The major arteries such as the vitelline and umbilical arteries of embryos also have been reported to associate with emerging blood cells (de Bruijn et al., 2000; Garcia-Porrero et al., 1995). In contrast to the common progenitor concept in the yolk sac, blood cells in the embryo proper are believed to diVerentiate from the endothelium. For example, in the floor of the dorsal aorta of the chicken or quail (JaVredo et al., 1998), intra-aortic CD45þVEGFR-2
(Flk-1
) hematopoietic cells appear to develop from VEGFR-2þ (Flk-1þ) cells that take up DiI-conjugated acetylated low-density lipoprotein (DiI-acLDL). In mice, VE-cadherinþ, CD45
, TER119
cells (potentially endothelial cells) from E9.5 mouse embryos could generate hematopoietic cells, including lymphocytes (Nishikawa et al., 1998a). Similarly, the HSC activity in the AGM regions and vitelline and umbilical arteries of E11.5 Runx1þ/Lacz embryos was derived from cells in the endothelium that do not express CD45 but do express Runx1, a critical transcription factor for definitive hematopoietic development (North et al., 2002). Therefore, the term hemogenic endothelium is often used to describe the hematopoietic potential of presumably aortic endothelial cells. The precise relationship between the hemangioblast and hemogenic endothelial cells is currently not known.

1. Hematopoietic and Endothelial Cells

11

C. The Identification of Blast Colony–Forming Cells from In Vitro Differentiated Embryonic Stem Cells Even though there has been a great interest in identifying the hemangioblast in the developing embryo, the use of embryo-derived cells has proven diYcult because the developmental sequence occurs rapidly, the tissues are diYcult to access, and only a small number of cells can be obtained. An alternate source of embryonic cells for the studies of early embryonic events is the in vitro diVerentiated progeny of ES cells. ES cells diVerentiate eYciently in vitro and give rise to three-dimensional, diVerentiated cell masses called embryoid bodies (EBs) (Fig. 1; reviewed in Choi, 2002; Keller et al., 1999). ES cells can also be diVerentiated on stromal cells or type IV collagen without intermediate formation of the EB structure (Nakano et al., 1994; Nishikawa et al., 1998b). Many diVerent lineages have been reported to develop within EBs, including neuronal, muscle, endothelial, and hematopoietic lineages (reviewed in Choi, 2002). Of these, the hematopoietic lineage has been the most extensively characterized. The development of hematopoietic and endothelial cells within EBs mimics in vivo events such that yolk sac blood island-like structures with vascular channels containing hematopoietic cells can be found within cystic EBs (Doetschman et al., 1985). As in the developing embryo, the primitive erythroid cells develop prior to definitive hematopoietic populations (Keller et al., 1993; Palis et al., 1999). The developmental kinetics of various hematopoietic lineage precursors within EBs and molecular and cellular studies of these cells have demonstrated that the sequence of events leading to the onset of hematopoiesis within EBs is similar to that found within the normal mouse embryo. In addition, EBs provide a large number of cells representing an early or primitive stage of development that is otherwise diYcult to access in an embryo. Therefore, the in vitro diVerentiation model of ES cells is an ideal system for obtaining and studying primitive progenitors of all cell lineages. Using the in vitro ES diVerentiation model system, the blast colonyforming cell (BL-CFC) population present within day 2.5–3.5 EBs has been shown to represent the hemangioblast (Choi et al., 1998). BL-CFCs are transient and develop prior to the primitive erythroid population (Choi et al., 1998; Kennedy et al., 1997). BL-CFCs form blast colonies in response to vascular endothelial growth factor (VEGF), a ligand for the receptor tyrosine kinase, Flk-1 (Matthews et al., 1991; Millauer et al., 1993), in semi-solid media such as methylcellulose cultures. Gene expression analysis indicated that cells within the blast colonies (blast cells) express a number of genes common to both hematopoietic and endothelial lineages, including Scl, CD34, and the VEGF receptor, Flk-1 (Kennedy et al., 1997). Blast cells contain primitive and definitive hematopoietic as well as endothelial cell progenitors (Choi et al., 1998; Kennedy et al., 1997). Most importantly,

12

Park et al.

the hematopoietic and endothelial precursors present within the blast colonies are clonal, as demonstrated by cell-mixing studies of two diVerent ES lines (Choi et al., 1998). D. From Flk-1-Expressing Mesoderm to Hematopoietic and Endothelial Cells Great progress has been made in recent years toward delineating the cellular sequence leading to hematopoietic and endothelial cell development from mesoderm (Fig. 2). First, Fehling et al. (2003) examined Brachyuryþ/GFP ES cells and demonstrated that Brachyuryþ mesodermal cells develop first. Flk-1 is turned on within Brachyuryþ cells to form BrachyuryþFlk-1þ mesoderm. Cell-replating studies demonstrated that BrachyuryþFlk-1þ cells contained hemangioblasts. Second, Motoike et al. (2003) performed fatemapping studies of Flk-1þ cells by examining Flk-1þ/Cre knock-in mice. When crossed to Rosa-26 reporter (R26R) mice (Soriano, 1999), and when E8.5 embryos were stained for LacZ expression, Flk-1 expression was seen in all vascular endothelial cells and hematopoietic cells. When LacZ staining of Flk1þ/Cre; R26R to Flk-1þ/LacZ mice was compared, the authors observed that Cre excision occurs precisely at the time of endogenous Flk-1 expression. In addition to vascular and hematopoietic cells, Flk-1 expression was also detected in cardiac and skeletal muscles in E10.5 embryos. Third, Ema et al. (2003) generated Flk-1þ/Scl ES cells and demonstrated that they generated an increased number of blast colonies as compared with wild-type

Figure 2 Flk-1-expressing mesoderm is thought to generate the circulatory system, including blood, endothelial, skeletal, smooth muscle, and cardiac muscle cells.

1. Hematopoietic and Endothelial Cells

13


/Scl

ES cells. Flk-1 embryos were also generated and examined. Although blast colony generation was rescued in these ES cells, the hematopoietic and endothelial defects in the mice were not, likely due to the migratory defects of cells lacking Flk-1. Furthermore, Flk-1-expressing cells from Flk-1þ/Scl produced predominantly hematopoietic and endothelial cells in culture, whereas Flk-1 expressing cells from Scl
/
ES cells could not diVerentiate into endothelial cells. Instead, Scl-deficient Flk-1þ cells readily generated smooth muscle cells in vitro. These studies indicate that Scl expression is critical for hemangioblast development, and also suggest that the coordinated expression of Flk1 and Scl is critical for proper development of hematopoietic, endothelial, and smooth muscle cells. The contribution of Flk-1þ cells to pericytes was also shown by Yamashita et al. (2000). In this study, ES-derived Flk-1þ cells were able to generate endothelial and smooth muscle cells in vitro and in vivo. Finally, Flk-1 and Scl were shown to be molecular determinants of the ES-derived hemangioblast (Chung et al., 2002). In this study, a nonfunctional human CD4 (hCD4) was knocked into the Scl locus and cells expressing Flk-1 and hCD4 were sorted and shown to readily generate blast colonies. Moreover, the kinetic and cell-replating studies of Flk-1- and hCD4-expressing cells demonstrated that hematopoietic and endothelial cells developed via sequential generation of Flk-1 and Scl-expressing cells. Flk-1þ cells first arise in developing EBs, and the Scl gene is turned on within Flk-1þ cells to give rise to Flk-1þhCD4þ cells. Alternatively, a subpopulation of the initial Flk-1þhCD4
cells remains Scl negative. Within Flk-1þhCD4þ cells, Flk-1 is downregulated to generate Flk-1
hCD4þ cells. Replating studies demonstrate that hematopoietic progenitors are enriched within Flk-1þhCD4þ and Flk-1
hCD4þ cells, whereas endothelial cells develop from Flk-1þhCD4þ and Flk-1þhCD4
cell populations. These studies indicate that there are two populations of endothelial cell progenitors, Scl dependent and independent. The report suggests that Scl-dependent endothelial cells develop from the hemangioblast (Fig. 3).

Figure 3 Schematic diagram of the emergence of Flk-1- and Scl-expressing hemangioblast, angioblast, and hematopoietic progenitors from ES cells. Transcription factors that function at each developmental stage are also indicated.

14

Park et al.

E. Hematopoietic Inductive Signals One of the most heavily pursued areas of current research in developmental hematopoiesis has been to identify inductive signals that initiate the hematopoietic program. The fact that only the mesoderm that is situated adjacent to visceral endoderm generates both blood and endothelial cells in the extraembryonic yolk sac indicates that factors/signals from the visceral endoderm are critical for hematopoietic induction. Studies of quail–chick chimeras (Pardanaud et al., 1996) have shown that the endoderm can induce hematopoiesis from the somatopleural mesoderm, which only has angiogenic potential. Similarly, BelaoussoV et al. (1998) have shown that primitive endoderm can induce hematopoiesis from the anterior epiblast, which is prospective neural ectoderm. Xenopus animal cap cultures have been useful in identifying factors/signals that can induce hematopoietic diVerentiation. Animal caps, which normally diVerentiate into ectoderm, can generate mesoderm in the presence of several growth factors such as BMPs, Activin A, and basic fibroblast growth factor (bFGF). Given the fact that blood cells develop from mesoderm, these mesoderm-inducing factors could have a hematopoietic inductive role. Indeed, the formation of erythroid cells from the animal cap can be induced by BMP-4 and bFGF or by BMP-4 and Activin A, and the generation of erythroid cells by exogenously expressed Gata1 can be potentiated by bFGF (Huber et al., 1998). In addition, BMP-4 could induce generation of erythroid cells through upregulation of Gata2 (Maeno et al., 1996). Similarly, the formation of blood islands from quail epiblasts is dependent on bFGF (Flamme and Risau, 1992). In this system, bFGF-mediated blood island formation correlates with the induction of the Flk1 gene (Flamme et al., 1995), suggesting that bFGF is critical for the emergence of the hemangioblast. Furthermore, in quail embryos, bFGF, VEGF, and transforming growth factor (TGF)- 1 can induce hematopoietic diVerentiation from the somatopleural mesoderm (Pardanaud et al., 1996). Similarly in mice, both Activin A and BMP-4 can induce hematopoietic diVerentiation from the anterior headfold region (Kanatsu and Nishikawa, 1996), and Indian hedgehog can promote hematopoietic diVerentiation from the anterior epiblast (Dyer et al., 2001). The ES–EB system has proven valuable for identifying factors involved in hematopoietic induction. Ultimately, this information can be used to manipulate the system such that all the progeny of EB cells take a hematopoietic fate. Several studies have shown that mesoderm-inducing factors can aVect EB diVerentiation. For example, the addition of bFGF or Activin A to diVerentiatin EBs enhances Brachyury gene expression, a marker of mesodermal tissue (Yamada et al., 1994). This observation indicates that cells within EBs can respond to external signals and therefore are useful for

1. Hematopoietic and Endothelial Cells

15

examining factors involved in hematopoietic specification. Johansson and Wiles (1995) initiated studies investigating the role of mesoderm-inducing factors on hematopoietic development. When ES cells were diVerentiated in serum-free, chemically defined medium (CDM) in the presence of various mesoderm-inducing factors, BMP-4 could induce the expression of the embryonic globin gene, H1. Similarly, studies by Adelman et al. (2002) demonstrated that BMP-4 was a prerequisite for erythroid-lineage-specific gene expression in ES cells. In this study, BMP-4 was added to serum-free medium and the expression of the erythroid cell-specific genes, Eklf and Gata1, was examined to show that BMP-4 was important for Eklf and Gata1 gene induction. Recently, factors that can induce Flk-1- and Scl-expressing cells were examined by utilizing in vitro diVerentiation models of ES cells (Park et al., 2004). In serum-free conditions, BMP-4 was critical for Flk-1 induction. BMP-4 activated the Smad1/5 pathway, whereas inhibition of the Smad1/5 pathway resulted in a reduction of Flk-1þ cell generation. Consistent with the notion that BMP-4 is critical for the generation of Flk-1þ cells, Bmp4deficient mice die between E7.5 and E9.5 with defects in mesoderm formation and patterning. Those that survive up to E9.5 display severe defects in blood islands (Winnier et al., 1995). Additionally, mice lacking the type I BMP receptor (Alk-3), which binds BMP-2 and BMP-4, fail to complete gastrulation, and none survive up to E7.0 (Mishina et al., 1995). Mice deficient in Smad1 or Smad5, downstream signaling molecules of TGF- family members, display varying degrees of defects (or no obvious phenotype) in hematopoietic and vascular development. This variation may be due to overlapping function between Smad1, 5, and 8 (Tremblay et al., 2001). For example, Smad1-deficient mice display early embryonic lethality and die between E9.5 and E10.5 due to failure of chorioallantoic fusion (Lechleider et al., 2001; Tremblay et al., 2001). Smad5-deficient mice die between E9.5 and E11.5. Mutant embryos are anemic and have disorganized vessels, despite formation of the primitive plexus. There seemed to be more primitive blood cells in E8.5 mutant yolk sacs, although E9.5 mutant yolk sacs contained almost no blood cells (Chang et al., 1999). Subsequent studies demonstrated that Smad5-deficient yolk sacs contained a higher frequency of HPP-CFCs, and Smad5-deficient ES cells gave rise to increased hematopoietic progenitors, including blast colonies in vitro (Liu et al., 2003). For proper hematopoietic development, the expansion of Scl-expressing cells requires VEGF-mediated signaling (Park et al., 2004). VEGF has been shown to play a critical role in vasculogenesis, angiogenesis, and hematopoiesis during embryonic development. VEGF expression can be detected as early as E5.5 in the extraembryonic visceral and parietal endoderm during mouse embryogenesis (Miquerol et al., 1999). At E7.0–7.5, VEGF is expressed in the extraembryonic visceral endoderm and extraembryonic

16

Park et al.

mesoderm, but not in the embryo proper. VEGF expression within the embryo proper can be detected at E8.0 in the definitive endoderm. The dorsal aorta forms in close proximity to the embryonic endoderm. Mice heterozygous for Vegf (Vegf þ/
) are embryonic lethal due to defects in vascular development (Carmeliet et al., 1996; Ferrara et al., 1996). In these mice the production of hematopoietic cells is significantly reduced. Furthermore, mice with slightly higher levels of VEGF expression (two to threefold) display early embryonic lethality due to severe abnormalities in heart development (Miquerol et al., 2000). Finally, hypomorphic Vegf þ/lo animals are viable and normal, but Vegf lo/lo embryos die early due to abnormalities in yolk sac vasculature and from deficiencies in the development of the dorsal aorta (Damert et al., 2002). Recent studies demonstrate that VEGF production from the yolk sac visceral endoderm is suYcient and necessary for blood island formation and for vascular development. In these studies, chimeras between Vegf wild-type tetraploid embryos and diploid Vegf lo/lo embryos showed rescue in blood island formation and in vascular development (Damert et al., 2002). In these chimeras, yolk sac visceral endoderm and trophoblast tissue will develop from the tetraploid embryos. Moreover, the hematopoietic cell population in the embryo proper of these chimeras increased as the contribution of Vegf wildtype tetraploid cells to the yolk sac visceral endoderm was augmented. Importantly, chimeras generated between Vegf lo/lo tetraploid embryos with Vegf þ/þ ES cells showed defects in yolk sac vascular development. Collectively, these studies indicate that tight regulation of VEGF expression is crucial for correct vascular and hematopoietic development in the early embryo. Another important player in hematopoiesis is TGF- 1, which is expressed in yolk sac blood islands, mesodermal cells of the allantois, and cardiogenic mesoderm of the embryo (Akhurst et al., 1990). TGF- 1 binds its cognate receptors TGF- RII and TGF- RI. TGF- receptor II expression largely correlates with that of TGF- 1 (Lawler et al., 1994). Tgf 1
/
mice showed two distinct phenotypes; 50% were perinatal lethal, and the rest died around E10.5 (Dickson et al., 1995). The latter displayed yolk sac anemia due to a severe reduction of erythrocytes and a defect in endothelial cell diVerentiation. Mice deficient in Tgf rII showed defects in yolk sac hematopoiesis and vasculogenesis (Oshima et al., 1996). Tgf rI
/
mice displayed increased numbers of erythroid progenitors, whereas granulocyte-macrophage colony–forming cells (CFU-GM) and mixed colony-forming cells (CFU-Mix) appeared to be normal (Larsson et al., 2001). These studies suggest that TGF- 1 signaling is necessary for normal vascular development, but could be inhibitory to the growth and diVerentiation of hematopoietic progenitors. Consistent with these data, TGF- 1 inhibited BMP-4 and VEGF-mediated hematopoietic induction in the ES–EB system (Park et al., 2004). Hedgehog signaling is important in pattern formation and morphogenesis in a variety of developing embryos (McMahon et al., 2003). In mice, three

1. Hematopoietic and Endothelial Cells

17

members have been identified so far: Sonic hedgehog (Shh), Desert hedgehog (Dhh), and Indian hedgehog (Ihh). Among these, only Ihh is expressed in the visceral endoderm of the yolk sac (Dyer et al., 2001; Farrington et al., 1997) and its receptor–downstream molecules are expressed in the posterior epiblast, which is destined to form blood and endothelial cells during early gastrulation. Two reports have demonstrated that primitive endoderm or IHH itself can induce hematopoiesis from explanted epiblasts (BelaoussoV et al., 1998; Dyer et al., 2001). In addition, the anterior epiblast, which is fated to give rise to neuroectoderm, can generate hematopoietic cells upon IHH treatment. These findings clearly suggest that IHH-mediated signaling has an important role for hematopoiesis. Furthermore, IHH could also induce the expression of BMP-4 from anterior epiblast (Dyer et al., 2001). Given the role of BMP-4-mediated signaling in the upregulation of Brachyury expression and the generation of Flk-1þ cells from ES cells, it is possible that IHH acts as an upstream regulator of BMP-4. The Notch pathway has been considered a cell fate determinant of multipotent precursor cells (Artavanis-Tsakonas et al., 1999). During hematopoiesis, Notch-mediated signaling is required for T-cell commitment and has also been implicated in modulating the self-renewal capacity of hematopoietic stem cells in adult bone marrow (Ohishi et al., 2003). Recent work by Kumano et al. (2003) demonstrated new roles for the Notch pathway during early hematopoiesis. The PAS culture from Notch1
/
embryos displayed severely impaired hematopoiesis, although the yolk sac generated a similar number of hematopoietic colonies as compared with wild type. Notch2
/
embryos did not show any significant diVerence from wild-type embryos. However, when transplanted into conditioned neonatal mice, E9.5 Notch1
/
PAS or yolk sac-derived cells could not reconstitute the hematopoietic system, indicating an essential role for Notch1 in generating HSCs in the yolk sac and PAS. F. Transcriptional Control of Hematopoietic and Endothelial Cell Lineage Commitment 1. Scl Originally identified as a target of t(1;14) chromosomal translocations in Tcell acute lymphoblastic leukemia (Finger et al., 1989), Scl is a bHLH transcription factor that is required for the generation of all hematopoietic lineages in the mouse (Robb et al., 1995; Shivadasani et al., 1995). In the adult, Scl is expressed in hematopoietic and endothelial cells. In the developing embryo Scl is expressed as early as E7.5, and by E8.5 the expression is localized to the yolk sac blood progenitors and endothelial cells (Green et al., 1991). Scl is also expressed in the developing brain (Green

18

Park et al.

et al., 1992). Like many genes important for hematopoiesis, the role of Scl was delineated through generation of Scl-deficient mice. Scl-null animals die by E10.5 due to defective embryonic hematopoiesis (Robb et al., 1995; Shivdasani et al., 1995). Moreover, Scl-null ES cells failed to contribute to any hematopoietic tissues in chimeric animals. Subsequent studies showed that Scl is also required for endothelial cell development, because Scl
/
ES cells fail to contribute to the remodeling of the primary vascular plexus in the yolk sac (Visvader et al., 1998). Scl deficiency resembles loss of the erythroid transcription factor GATA-1 or the LIM protein Lmo2. At the molecular level, Scl-null ES cells do not express either major or minor isoforms of globin. Additionally, several other hematopoietic genes are downregulated in Scl
/
cells. Myb, Pu.1, and Nf-E2 all have diminished levels of expression as compared with wild type, and Gata1 and Eklf are nearly absent (Robb et al., 1996). Conversely, it has been demonstrated that not only does enforced expression of Scl strongly enhance the blood formation in embryos (Gering et al., 1998), but Scl is capable of converting somitic and pronephric duct tissues into hemangioblasts (Gering et al., 2003), indicating that Scl has a dominant role in the commitment to hemangioblasts, much like the ability for MyoD to commit cells to a myogenic fate. Several studies further support the notion that Scl is critical for hemangioblast specification. First, Scldeficient ES cells failed to give rise to blast colonies (Faloon et al., 2000; Robertson et al., 2000), the progeny of the hemangioblasts. Moreover, enhanced expression of Scl from the Flk-1 locus (Flk-1þ/Scl ES cells) produced a higher number of blast colonies compared with Flk-1þ/þ cells (Ema et al., 2003). The mechanisms of Scl-mediated transcription are still unclear. Studies by Porcher et al. (1999) indicate that mutant Scl unable to bind DNA can still rescue embryonic hematopoiesis and restore definitive hematopoiesis considerably in Scl
/
ES cells. Because the HLH domain was absolutely required, the emerging model for Scl initiating embryonic hematopoiesis is that it functions as a nucleating factor to bring transcription factors together. For example, Scl will recruit an E protein, such as E2A, to an E-box (CANNTG), bound to a bridging molecule that are LIM domain-containing molecules such as Lmo2 (see later) and Ldb1 (Wadman et al., 1997). This complex then links the E-box to a GATA motif bound by either GATA-1 or GATA-2 (more later). The precise architecture and role of this complex remains an attractive model, albeit needing further inquiry (Cantor and Orkin, 2002). 2. GATA-2 The second member cloned of the six-member GATA binding protein family, GATA-2, shows strong expression in pluripotent cells. All GATA proteins bind the consensus (T/A)GATA(G/A) DNA sequence (Orkin,

1. Hematopoietic and Endothelial Cells

19

1992). Additionally, each GATA protein contains two Zn-finger motifs. The amino-terminal finger is used for protein–protein interactions with other trans-acting factors, and the carboxy–terminal finger is required for DNA binding. GATA-1, GATA-2, and GATA-3 are considered to be the ‘‘hematopoietic’’ GATAs, whereas GATA-4, GATA-5, and GATA-6 are all considered to be ‘‘endodermal’’, because they are expressed in endodermal tissues and their absence leads to developmental defects in endodermally derived tissues such as the heart, gut, intestines, and lungs. Gata1 and Gata2 have been identified as early acting genes in hematopoiesis. Current models indicate that Gata2 is induced early in the extraembryonic yolk sac and induces expansion of hematopoietic progenitors, whereas Gata1 is induced later, by GATA-2 and functions in erythroid cell maturation (Ohneda and Yamamoto, 2002). Unlike many other hematopoietic transcription factors, GATA-2 is rarely associated with leukemias caused by aberrant expression. Gata2
/
animals are embryonically lethal at E10.5 due to severe anemia (Tsai et al., 1994). The Gata2
/
animals displayed a severe reduction in the number of primitive erythroid cells. When Gata2
/
ES cells were used to generate chimeric animals, there was no contribution to any hematopoietic compartments nor was -globin made by the Gata2
/
cells. Thus, early work defined GATA-2 as having a role in the expansion/proliferation of the early, primitive hematopoietic compartment but largely dispensable for the diVerentiation of the majority of hematopoietic lineages. Subsequent work using conditional induction of GATA-2 has helped to show that enforced GATA-2 expression can enhance the production of hematopoietic progenitors (Kitajima et al., 2002). This report showed that forced expression of GATA-2 suppressed expression of other later hematopoietic transcription factors, including Pu.1 and c-Myb. Additionally, from this same report, the link between GATA-2 and the formation of HSCs is examined as the authors see that enforced GATA-2 expression leads to a three-fold increase in the number of hematopoietic colonies and that upon FACS analysis through the presence of an internal ribosomal entry site-enhanced green fluorescent protein (IRES-EGFP) linked to the Gata2 artificial promoter, nearly all EGFPþ cells were also Scaþ and c-Kitþ, showing that GATA-2 directly augments the HSC pool. Lastly, recent work has demonstrated the mechanism by which GATA-2 induces transcription of Gata1 and GATA-1 in turn suppresses Gata2 transcription. Using a quantitative chromatin immunoprecipitative (ChIP) assay, Grass and colleagues (2003) show that GATA-1 binds a highly restricted upstream region of the approximately 70kb Gata2 domain. GATA-1 then rapidly displaces GATA-2, which is coupled to transcriptional repression. GATA-1 also displaces cAMP response element-bind protein (CREB) binding protein (CBP), despite the fact that GATA-1 binds CBP in other contexts. This repression correlates with reduced domain-wide histone acetylation. GATA-1 instigates Gata2

20

Park et al.

repression by means of disruption of positive autoregulation, followed by establishment of a domain-wide repressive chromatin structure (Grass et al., 2003). This GATA switch likely accounts for the change to a diVerentiation transcriptional program from a proliferative transcriptional program in hematopoietic progenitors. Collectively, GATA-2 appears to function in the expansion/proliferation of the early, primative hematopoietic compartment. 3. Lmo2 Lmo2 is a Zn-finger, LIM-only protein. Unlike other Zn-finger transcription factors, the LIM-only proteins contain only the -helical structure of other Zn-finger family members, thus enabling protein–protein interactions but making protein–DNA interactions impossible (Perez-Alvarado et al., 1994). Like many other hematopoietic transcription factors, aberrant expression of Lmo2 through chromosomal translocations can lead to leukemias. This occurs because Lmo2 acts like a transcription factor, serving to bridge complexes together to induce gene transcription with the aid of the basal transcription machinery. As described earlier, Lmo2 (and Ldb1) does not physically bind DNA, but directly interacts with transcription factors that do. Lmo2’s tight linkage with Scl-mediated transcription is further evidence that it behaves like a bona fide transcription factor, because loss of Lmo2 leads to impairment of Scl-mediated transcription (Larson et al., 1996; ValgeArcher et al., 1994). Accordingly, Lmo2
/
mice die due to failure of yolk sac erythropoiesis (Warren et al., 1994). In addition, because Lmo2 is posited to bridge Scl/E2A and GATA proteins, the phenotype of the Gata1-null mutant animal is similar to the phenotype of the Lmo2-null animal (Weiss et al., 1994). Finally, coexpression of GATA-1 with Scl and Lmo2 in Xenopus organisms at one-cell-stage embryo leads to ventralization, and blood cell formation in these embryos becomes obvious throughout the dorsal–ventral axis. Thus, it appears that a Scl–Lmo2–GATA-1 complex is critical for specifying mesoderm to become blood during Xenopus embryogenesis. 4. Runx1 Also known as Aml1 and Cbf2, Runx1 is a frequent target of translocations in leukemias. Runx1 shows homology to the Drosophila paired-rule gene Runt and binds to the TGT/cGGT DNA sequence (Daga et al., 1992; Meyers et al., 1993). To achieve basal transcriptional activity, Runx1 heterodimerizes with Cbf through an 118 amino acid Runt homology domain (RHD) (Meyers et al., 1993; Ogawa et al., 1993; Wang et al., 1993). The AML– ETO fusion is one of the most frequent translocations seen in acute myeloid leukemia (AML) caused by t(8;21) that fuses the DNA-binding domain of Runx1 (AML) to the activation moiety of the Zn-finger transcription factor

1. Hematopoietic and Endothelial Cells

21

Eto (Erickson et al., 1992; Miyoshi et al., 1993; Nisson et al., 1992). Understanding of the normal gene product came in 1996 when two groups ablated the gene encoding Runx1 (Okuda et al., 1996; Wang et al., 1996). Homozygous animals died between E12.5 and E13.5. Although the appearance and development of both the embryo and the extraembryonic yolk sac were normal, the null animals showed hemorrhage in the central nervous system. Moreover, despite normal yolk sac hematopoiesis and the presence of primitive nucleated erythrocytes, the animals lacked any definitive fetal-liverproduced hematopoietic tissue. Whereas normal animals will have primitive nucleated erythrocytes, immature granulocytes, and macrophages/monocytes, as well as numerous less diVerentiated cells, Runx1-null animals had only primitive nucleated erythrocytes present in their liver (Wang et al., 1996). Analysis of chimeric animals showed that although Runx1þ/
ES cells were capable of contributing to bone marrow, peripheral blood, thymus, and spleen, Runx1
/
ES cells could not contribute to any hematopoietic tissues despite contribution to other tissues (Okuda et al., 1996). Studies with Runx1þ/LacZ mice further support the role of Runx1 in definitive hematopoiesis (North et al., 1999, 2002). Runx1 is expressed in the ventral wall of the dorsal aorta, as well as in the vitelline and umbilical arteries. Its expression then is found in clusters of cells that are closely located to the lumen of these regions. These cells are hematopoietic cells, as shown by their round shape and expression of the pan-leukocyte marker CD45 (JaVredo et al., 1998; Tavian et al., 1996). Taking into account the study that the AGM, vitelline, and umbilical arteries harbor long-term repopulation hematopoietic cells (LTR-HSCs), it raises the possibility that LacZ (Runx1)-expressing round cells in these regions could be HSCs. Recently, North et al. (2002) demonstrated that Runx1þ cells are HSCs. In this study it was demonstrated that transplanted Runx1þ cells, not Runx1
cells from the AGM, vitelline, and umbilical arteries can reconstitute the hematopoietic system of irradiated recipient mice, suggesting that Runx1 is critical for HSC generation. In addition to its role in definitive hematopoiesis, several lines of evidence suggest that Runx1 could be involved in hemangioblast development. Runx1 is first detectable in the extraembryonic mesodermal cells (E7.25) and then in both primitive erythrocytes and endothelial cells in the yolk sac blood islands at E8.0 (Lacaud et al., 2002; North et al., 1999). A study of Runx1 expression kinetics showed that Runx1 is also expressed at the same time that Flk-1 and Scl are expressed. Moreover, blast colonies and BL-CFCs express Runx1, suggesting that Runx1 is an important regulator of hemangioblast formation. Consistent with this idea, Runx1
/
ES cells generated a significantly reduced number of blast colonies. Subsequent replating experiments revealed that blast colonies from Runx1
/
ES cells normally gave rise to primitive erythroid colonies, but failed to generate definitive hematopoietic

22

Park et al.

colonies. Collectively, these results indicate that Runx1 is required for both hemangioblast development and definitive hematopoiesis. Runx1-mediated gene transcription is achieved through cooperation with other trans-acting molecules such as Ets-1. For instance, the T-cell receptor promoter contains adjacent Runx1 and Ets binding motifs. Alone, each is suYcient to bind its cognate DNA. However, mutation of either single site disrupts transcription, although transcriptional activation is independent of the spacing of the sites (Wotton et al., 1993, 1994). This same site synergy is also seen with c-Myb (Hernandez-Munain and Krangel, 1995). More recent data indicate that Runx1 serves as an assembly factor, inducing conformation changes and recruiting additional trans-acting molecules to activate gene transcription, and may possess little trans-activating activity itself (Hernandez-Munain and Krangel, 2002). Yamaguchi and coworkers have shed further light on Runx1’s trans-activating capabilities, showing that acetylation of specific lysine residues (Lys-24 and Lys-43) by p300 is required to obtain maximal DNA binding and transcriptional capability (Yamaguchi et al., 2004).

G. In Vivo Potential of Embryonic Stem Cell–Derived Hematopoietic Progenitors The ability of ES cells to generate many diVerent somatic cells in vitro argues for their usage as a source for cell transplantation, provided that they can function in vivo. Accumulating studies demonstrate that in vitro generated hematopoietic progenitors can function in vivo, although the generation of HSCs from ES cells has not been firmly established. Muller and Dzierzak (1993) have utilized in vitro diVerentiated ES cells as donor cells in cell transfer studies using newborn Wv/Wv and SCID mice as recipients. In these studies, donor-derived cells were found only within the lymphoid cell lineage, although donor-derived cells persisted for longer than 6 months. Day 13 EB cells were found to be the most eYcient in repopulating the hematopoietic system. Because entire EBs without further enrichment of hematopoietic progenitor cells were used in this study, the findings that ES-derived cells had limited lymphoid potential could reflect the rarity of HSCs present within day 13 EB cell populations. Potocnik et al. (1997) have isolated AA4.1þB220þ and AA4.1þB220
cell populations and tested the in vivo reconstitution potential in Rag-1-deficient mice (4–6 weeks old). Both cell populations engrafted the recipients, but AA4.1þB220þ cells had limited life span and limited potential (i.e., they persisted for only up to 8 weeks in the recipients and could only give rise to the B-cell lineage): However, AA4.1þB220
cells survived longer in the recipients and could be found even 24 weeks after the transplantation. AA4.1þB220
cells were found to be

23

1. Hematopoietic and Endothelial Cells þ

þ

þ

more primitive compared with AA4.1 B220 cells, because AA4.1 B220
cells could give rise to both B and T cells. Donor-derived B220þc-Kitþ cells were also found in the recipient bone marrow, suggesting that AA4.1þB220
cells contained more primitive hematopoietic progenitors. However, the HSC activity was not examined in this study. The ability for ES-derived cells to repopulate the hematopoietic system was much lower, compared with fetal liver cells, arguing for a need to develop an optimal protocol for utilizing ES-derived cells for hematopoietic reconstitution. It is important to point out that ES cells derived from the 129-mouse strain were used in both studies to reconstitute C57 (i.e., C57B46) BL/6 mice, demonstrating that allogeneic transplantation works well. A study by Hole et al. (1996) showed multilineage (lymphoid and myeloid compartment) reconstitution by in vitro diVerentiated EBs. In this study, ES cells were diVerentiated for 4 days and then transplanted into lethally irradiated adult mice. It should be noted that the authors used the earliest time point at which primitive multilineage hematopoietic precursors can be detected. The recipient mice survived for more than 3 months and contained EB-derived lymphocytes and granulocytes. However, EB-mediated repopulation was lost by 6 months, indicating that the day 4 EB cells that they used could have contained short-term repopulating hematopoietic progenitors or committed multilineage precursors. Most recently, Burt et al. (2004) were able to isolate a c-KitþCD45þ population from diVerentiated ES cells. ES-derived c-KitþCD45þ cells could reconstitute hematopoietic compartments when transplanted into lethally irradiated recipients. The authors showed donor-derived contribution to lymphocytes, monocytes, and granulocytes in the peripheral blood of the recipients. Other hematopoietic organs were not examined in this study. This study showed that intrabone injection generated more eYcient engraftment of the ES-derived cells, compared with intravein injection. In addition to nonmanipulated EB cell transplantation, genetically modified ES cells have been used in several studies. Perlingeiro et al. (2001) have utilized the Bcr–Abl oncogene to transduce a blast cell population, containing both multipotential hematopoietic and endothelial cell progenitors (Choi et al., 1998). Blast cells transformed with the Bcr–Abl oncogene and cultured on OP9 cells, previously shown to support HSCs (Nakano et al., 1994), could repopulate sublethally irradiated, 8-week-old 129Sv–Ev or NODSCID mice. Consistent with previous studies showing that blast cells contain both primitive and definitive erythroid, and myeloid progenitors (Choi et al., 1998; Kennedy et al., 1997), the Bcr–Abl-transduced blast cells generated primitive and definitive erythroid, myeloid, and lymphoid cell lineages in the recipients. The caveat of this study is that the cell population used for transplantation harbors an oncogene. The recipients ultimately developed myeloproliferative disorders between 5 and 9 weeks after transplantation. HoxB4, one of the homeotic genes, is expressed and implicated in

24

Park et al.

self-renewal of definitive HSCs (Sauvageau et al., 1995). Kyba et al. (2002) generated ES cells overexpressing a tetracycline-inducible HoxB4. When induced, HoxB4-overexpressing EBs cultured on OP9 cells showed a higher percentage of cell population expressing HSC markers c-kit and CD31. Also, the HoxB4-induced EBs were able to reconstitute the hematopoietic system of irradiated recipients, including myeloid and lymphoid lineages. Importantly, the EB-derived bone marrow cells from the first recipient were detected in secondary recipients.

III. Conclusions and Future Directions During ES diVerentiation, Flk-1-expressing cells initially develop from the mesoderm. Flk-1þ mesoderm generates many types of cells of the circulatory system, including blood, endothelial, smooth muscle, cardiac muscle, and skeletal muscle cells. Scl expression will further specify Flk-1þ cells to the hemangioblast. Pending issues concerning hemangioblasts are as follows. First, ES-derived BL-CFCs fit the description of in vitro equivalent hemangioblasts of the yolk sac blood islands. Nevertheless, there is no definite proof that such a cell exists in the developing embryo, yolk sac, or AGM. Clearly, cell-marking experiments will be necessary to determine the existence of a common progenitor. Second, the existence of the hemangioblast in adults needs to be investigated. Although it is conceptually accepted that hemangioblasts develop during embryogenesis and produce hematopoietic and endothelial cells of adults, recent studies suggest that hemangioblasts exist in adult stages as well. For example, human AC133þ cells from granulocyte colony-stimulating factor mobilized peripheral blood can diVerentiate into both hematopoietic and endothelial cells in cultures. These AC133þ cells can form new blood vessels in vivo (Gehling et al., 2000). In addition, Pelosi et al. (2002) showed that single CD34þFlk-1þ cells from human bone marrow or cord blood can generate both hematopoietic and endothelial cells. These potential postnatal hemangioblasts exhibited long-term proliferative potential in culture. Furthermore, a population of cells enriched for hematopoietic stem cells, such as Sca-1þc-KitþLin
cells, could also contribute to new blood vessel formation at the single-cell level (Bailey et al., 2004; Grant et al., 2002). Third, the full developmental potential of the hemangioblast should be determined. BL-CFCs can generate both hematopoietic and endothelial cells, although their full potential has not been carefully examined. Recent studies suggest that smooth muscle cells can also diVerentiate from BL-CFCs (Ema et al., 2003). In this study, Ema and colleagues showed that blast cells generated smooth muscle cells in the absence of VEGF. Lastly, molecular regulatory mechanisms involved in hemangioblast

1. Hematopoietic and Endothelial Cells

25

specification and diVerentiation should be identified and investigated. This knowledge will be valuable for characterizing hemangioblast self-renewal and diVerentiation, modulating hemangioblast function, and isolating hemangioblast, hematopoietic, and endothelial progenitors for therapeutic purposes.

Acknowledgments We would like to thank many investigators for providing valuable reagents. We apologize to the many authors that we could not cite due to space constraints. This work was supported by grants from the National Institutes of Health, NHLBI, R01s HL63736 and HL55337 (to K.C.).

References Adelman, C. A., Chattopadhyay, S., and Bieker, J. J. (2002). The BMP/BMPR/Smad pathway directs expression of the erythroid-specific EKLF and GATA1 transcription factors during embryoid body diVerentiation in serum-free media. Development 129, 539–549. Akhurst, R. J., Lehnert, S. A., Faissner, A., and DuYe, E. (1990). TGF beta in murine morphogenetic processes: The early embryo and cardiogenesis. Development 108, 645–656. Anneren, C., Cowan, C. A., and Melton, D. A. (2004). The Src family of tyrosine kinases is important for embryonic stem cell self-renewal. J. Biol. Chem. 279, 31590–31598. Artavanis-Tsakonas, S., Rand, M. D., and Lake, R. J. (1999). Notch signaling: Cell fate control and signal integration in development. Science 284, 770–776. Bagutti, C., Wobus, A. M., Fassler, R., and Watt, F. M. (1996). DiVerentiation of embryonal stem cells into keratinocytes: Comparison of wild-type and beta 1 integrin-deficient cells. Dev. Biol. 179, 184–196. Bailey, A. S., Jiang, S., Afentoulis, M., Baumann, C. I., Schroeder, D. A., Olson, S. B., Wong, M. H., and Fleming, W. H. (2004). Transplanted adult hematopoietic stems cells diVerentiate into functional endothelial cells. Blood 103, 13–19. Bain, G., Kitchens, D., Yao, M., Huettner, J. E., and Gottlieb, D. I. (1995). Embryonic stem cells express neuronal properties in vitro. Dev. Biol. 168, 342–357. Beddington, R. S., and Robertson, E. J. (1989). An assessment of the developmental potential of embryonic stem cells in the midgestation mouse embryo. Development 105, 733–737. BelaoussoV, M., Farrington, S. M., and Baron, M. H. (1988). Hematopoietic induction and respecification of A-P identity by visceral endoderm signaling in the mouse embryo. Development 125, 5009–5018. Ben-Shushan, E., Thompson, J. R., Gudas, L. J., and Bergman, Y. (1998). Rex-1, a gene encoding a transcription factor expressed in the early embryo, is regulated via Oct 3/4 and Oct-6 binding to an octamer site and a novel protein, Rox-1, binding to an adjacent site. Mol. Cell. Biol. 18, 1866–1878. Boeuf, H., Hauss, C., Graeve, F. D., Baran, N., and Kedinger, C. (1997). Leukemia inhibitory factor-dependent transcriptional activation in embryonic stem cells. J. Cell Biol. 138, 1207–1217. Bradley, A., Evans, M., Kaufman, M. H., and Robertson, E. (1984). Formation of germ-line chimaeras from embryo-derived teratocarcinoma cell lines. Nature 309, 255–256.

26

Park et al.

Burdon, T., Stracey, C., Chambers, I., Nichols, J., and Smith, A. (1999). Suppression of SHP-2 and ERK signalling promotes self-renewal of mouse embryonic stem cells. Dev. Biol. 210, 30–43. Burt, R. K., Verda, L., Kim, D. A., Oyama, Y., Luo, K., and Link, C. (2004). Embryonic stem cells as an alternate marrow donor source: Engraftment without graft-versus-host disease. J. Exp. Med. 199, 895–904. Buttery, L. D., Bourne, S., Xynos, J. D., Wood, H., Hughes, F. J., Hughes, S. P., Episkopou, V., and Polak, J. M. (2001). DiVerentiation of osteoblasts and in vitro bone formation from murine embryonic stem cells. Tissue Eng. 7, 89–99. Cantor, A. B., and Orkin, S. H. (2002). Transcriptional regulation of erythropoiesis: An aVair involving multiple partners. Oncogene 21, 3368–3376. Carmeliet, P., Ferreira, V., Breier, G., Pollefeyt, S., Kiekens, L., Gertsenstein, M., Fahrig, M., Vandenhoeck, A., Harpal, K., Eberhardt, C., Declercq, C., Pawling, J., Moons, L., Collen, D., Risau, W., and Nagy, A. (1996). Abnormal blood vessel development and lethality in embryos lacking a single VEGF allele. Nature 380, 435–439. Chambers, I., Colby, D., Robertson, M., Nichols, J., Lee, S., Tweedie, S., and Smith, A. (2003). Functional expression cloning of Nanog, a pluripotency sustaining factor in embryonic stem cells. Cell 113, 643–655. Chang, H., Huylebroeck, D., Verschueren, K., Guo, Q., Matzuk, M., and Zwijsen, A. (1999). Smad5 knockout mice die at mid-gestation due to multiple embryonic and extraembryonic defects. Development 126, 1631–1642. Choi, K. (2002). The hemangioblast: A common progenitor of hematopoietic and endothelial cells. J. Hematother. Stem Cell Res. 11, 91–101. Choi, K., Kennedy, M., Kazarov, A., Papadimitriou, J. C., and Keller, G. (1998). A common precursor for hematopoietic and endothelial cells. Development 125, 725–732. Chung, Y. S., Zhang, W. J., Arentson, E., Kingsley, P. D., Palis, J., and Choi, K. (2002). Lineage analysis of the hemangioblast as defined by FLK1 and SCL expression. Development 129, 5511–5520. Cumano, A., Dieterlen-Lievre, F., and Godin, I. (1996). Lymphoid potential, probed before circulation in mouse, is restricted to caudal intraembryonic splanchnopleura. Cell 86, 907–916. Cumano, A., Ferraz, J. C., Klaine, M., Di Santo, J. P., and Godin, I. (2001). Intraembryonic, but not yolk sac hematopoietic precursors, isolated before circulation, provide long-term multilineage reconstitution. Immunity 15, 477–485. Cumano, A., Furlonger, C., and Paige, C. J. (1993). DiVerentiation and characterization of B-cell precursors detected in the yolk sac and embryo body of embryos beginning at the 10- to 12-somite stage. Proc. Natl. Acad. Sci. USA 90, 6429–6433. Daga, A., Tighe, J. E., and Calabi, F. (1992). Leukaemia/Drosophila homology. Nature 356, 484. Damert, A., Miquerol, L., Gertsenstein, M., Risau, W., and Nagy, A. (2002). InsuYcient VEGFA activity in yolk sac endoderm compromises haematopoietic and endothelial diVerentiation. Development 129, 1881–1892. Dani, C., Smith, A. G., Dessolin, S., Leroy, P., Staccini, L., Villageois, P., Darimont, C., and Ailhaud, G. (1997). DiVerentiation of embryonic stem cells into adipocytes in vitro. J. Cell Sci. 110, 1279–1285. Davis, S., Aldrich, T. H., Stahl, N., Pan, L., Taga, T., Kishimoto, T., Ip, N. Y., and Yancopoulos, G. D. (1993). LIFR beta and gp130 as heterodimerizing signal transducers of the tripartite CNTF receptor. Science 260, 1805–1808. de Bruijn, M. F., Speck, N. A., Peeters, M. C., and Dzierzak, E. (2000). Definitive hematopoietic stem cells first develop within the major arterial regions of the mouse embryo. EMBO J. 19, 2465–2474.

1. Hematopoietic and Endothelial Cells

27

Dickson, M. C., Martin, J. S., Cousins, F. M., Kulkarni, A. B., Karlsson, S., and Akhurst, R. J. (1995). Defective haematopoiesis and vasculogenesis in transforming growth factor-beta 1 knockout mice. Development 121, 1845–1854. Dieterlen-Lievre, F. (1975). On the origin of haemopoietic stem cells in the avian embryo: An experimental approach. J. Embryol. Exp. Morphol. 33, 607–619. Dieterlen-Lievre, F. (1997). Intaembryonic hematopoietic stem cells. Aplastic Anemia Stem Cell Biol. 11, 1149–1171. Dieterlen-Lievre, F., and Martin, C. (1981). DiVuse intraembryonic hemopoiesis in normal and chimeric avian development. Dev. Biol. 88, 180–191. Doetschman, T. C., Eistetter, H., Katz, M., Schmidt, W., and Kemler, R. (1985). The in vitro development of blastocyst-derived embryonic stem cell lines: Formation of visceral yolk sac, blood islands and myocardium. J. Embryol. Exp. Morphol. 87, 27–45. Drab, M., Haller, H., Bychkov, R., Erdmann, B., Lindschau, C., Haase, H., Morano, I., Luft, F. C., and Wobus, A. M. (1997). From totipotent embryonic stem cells to spontaneously contracting smooth muscle cells: A retinoic acid and db-cAMP in vitro diVerentiation model. FASEB J. 11, 905–915. Dyer, M. A., Farrington, S. M., Mohn, D., Munday, J. R., and Baron, M. H. (2001). Indian hedgehog activates hematopoiesis and vasculogenesis and can respecify prospective neurectodermal cell fate in the mouse embryo. Development 128, 1717–1730. Dzierzak, E. (1999). Embryonic beginnings of definitive hematopoietic stem cells. Ann. N. Y. Acad. Sci. 872, 256–262; discussion 262–264. Ema, M., Faloon, P., Zhang, W. J., Hirashima, M., Redi, T., Stanford, W., Choi, K., and Rossant, J. (2003). Combinatorial eVects of Flk-1 and Tal1 (SCL) on vascular and hematopoietic development in the mouse. Genes Dev. 17, 380–393. Erickson, P., Gao, J., Chang, K. S., Look, T., Whisenant, E., Raimondi, S., Lasher, R., Trujillo, J., Rowley, J., and Drabkin, H. (1992). Identification of breakpoints in t(8;21) acute myelogenous leukemia and isolation of a fusion transcript, AML1/ETO, with similarity to Drosophila segmentation gene, runt. Blood 80, 1825–1831. Evans, M. J., and Kaufman, M. H. (1981). Establishment in culture of pluripotential cells from mouse embryos. Nature 292, 154–156. Faloon, P., Arentson, E., Kazarov, A., Deng, C. X., Porcher, C., Orkin, S., and Choi, K. (2000). Basic fibroblast growth factor positively regulates hematopoietic development. Development 127, 1931–1941. Farrington, S. M., BelaoussoV, M., and Baron, M. H. (1997). Winged-helix, hedgehog and Bmp genes are diVerentially expressed in distinct cell layers of the murine yolk sac. Mech. Dev. 62, 197–211. Fehling, H. J., Lacaud, G., Kubo, A., Kennedy, M., Robertson, S., Keller, G., and KouskoV, V. (2003). Tracking mesoderm induction and its specification to the hemangioblast during embryonic stem cell diVerentiation. Development 130, 4217–4227. Feldman, B., Poueymirou, W., Papaioannou, V. E., DeChiara, T. M., and Goldfarb, M. (1995). Requirement of FGF-4 for postimplantation mouse development. Science 267, 246–249. Ferrara, N., Carver-Moore, K., Chen, H., Dowd, M., Lu, L., O’Shea, K. S., Powell-Braxton, L., Hillan, K. J., and Moore, M. W. (1996). Heterozygous embryonic lethality induced by targeted inactivation of the VEGF gene. Nature 380, 439–442. Finger, L. R., Kagan, J., Christopher, G., Kurtzberg, J., Hershfield, M. S., Nowell, P. C., and Croce, C. M. (1989). Involvement of the TCL5 gene on human chromosome 1 in T-cell leukemia and melanoma. Proc. Natl. Acad. Sci. USA 86, 5039–5043. Flamme, I., Breier, G., and Risau, W. (1995). Vascular endothelial growth factor (VEGF) and VEGF receptor 2 (flk-1) are expressed during vasculogenesis and vascular diVerentiation in the quail embryo. Dev. Biol. 169, 699–712.

28

Park et al.

Flamme, I., and Risau, W. (1992). Induction of vasculogenesis and hematopoiesis in vitro. Development 116, 435–439. Fraichard, A., Chassande, O., Bilbaut, G., Dehay, C., Savatier, P., and Samarut, J. (1995). In vitro diVerentiation of embryonic stem cells into glial cells and functional neurons. J. Cell Sci. 108, 3181–3188. Garcia-Porrero, J. A., Godin, I. E., and Dieterlen-Lievre, F. (1995). Potential intraembryonic hemogenic sites at pre-liver stages in the mouse. Anat. Embryol. (Berl). 192, 425–435. Gearing, D. P., Gough, N. M., King, J. A., Hilton, D. J., Nicola, N. A., Simpson, R. J., Nice, E. C., Kelso, A., and Metcalf, D. (1987). Molecular cloning and expression of cDNA encoding a murine myeloid leukaemia inhibitory factor (LIF). EMBO J. 6, 3995–4002. Gehling, U. M., Ergun, S., Schumacher, U., Wagener, C., Pantel, K., Otte, M., Schuch, G., Schafhausen, P., Mende, T., Kilic, N., Kluge, K., Schafer, B., Hossfeld, D. K., and Fiedler, W. (2000). In vitro diVerentiation of endothelial cells from AC133-positive progenitor cells. Blood 95, 3106–3112. Gering, M., Rodaway, A. R. F., Gottgens, B., Patient, R. K., and Green, A. R. (1998). The SCL gene specifies haemangioblast development from early mesoderm. EMBO J. 17, 4029–4045. Gering, M., Yamada, Y., Rabbitts, T. H., and Patient, R. K. (2003). Lmo2 and Scl/Tal1 convert non-axial mesoderm into haemangioblasts which diVerentiate into endothelial cells in the absence of Gata1. Development 130, 6187–6199. Godin, I., Dieterlen-Lievre, F., and Cumano, A. (1995). Emergence of multipotent hemopoietic cells in the yolk sac and paraaortic splanchnopleura in mouse embryos, beginning at 8.5 days postcoitus. Proc. Natl. Acad. Sci. USA 92, 773–777. Godin, I. E., Garcia-Porrero, J. A., Coutinho, A., Dieterlen-Lievre, F., and Marcos, M. A. (1993). Para-aortic splanchnopleura from early mouse embryos contains B1a cell progenitors. Nature 364, 67–70. Grant, M. B., May, W. S., Caballero, S., Brown, G. A., Guthrie, S. M., Mames, R. N., Byrne, B. J., Vaught, T., Spoerri, P. E., Peck, A. B., and Scott, E. W. (2002). Adult hematopoietic stem cells provide functional hemangioblast activity during retinal neovascularization. Nat. Med. 8, 607–612. Grass, J. A., Boyer, M. E., Pal, S., Wu, J., Weiss, M. J., and Bresnick, E. H. (2003). GATA-1dependent transcriptional repression of GATA-2 via disruption of positive autoregulation and domain-wide chromatin remodeling. Proc. Natl. Acad. Sci. USA 100, 8811–8816. Green, A. R., Lints, T., Visvader, J., Harvey, R., and Begley, C. G. (1992). SCL is coexpressed with GATA-1 in hemopoietic cells but is also expressed in developing brain. Oncogene 7, 653–660. Green, A. R., Salvaris, E., and Begley, C. G. (1991). Erythroid expression of the ‘‘helix-loophelix’’ gene, SCL. Oncogene 6, 475–479. Hernandez-Munain, C., and Krangel, M. S. (1995). c-Myb and core-binding factor/PEBP2 display functional synergy but bind independently to adjacent sites in the T-cell receptor delta enhancer. Mol. Cell. Biol. 15, 3090–3099. Hernandez-Munain, C., and Krangel, M. S. (2002). Distinct roles for c-Myb and core binding factor/polyoma enhancer-binding protein 2 in the assembly and function of a multiprotein complex on the TCR delta enhancer in vivo. J. Immunol. 169, 4362–4369. Hole, N., Graham, G. J., Menzel, U., and Ansell, J. D. (1996). A limited temporal window for the derivation of multilineage repopulating hematopoietic progenitors during embryonal stem cell diVerentiation in vitro. Blood 88, 1266–1276. Houssaint, E. (1981). DiVerentiation of the mouse hepatic primordium. II. Extrinsic origin of the haemopoietic cell line. Cell DiVer. 10, 243–252. Huang, H., and Auerbach, R. (1993). Identification and characterization of hematopoietic stem cells from the yolk sac of the early mouse embryo. Proc. Natl. Acad. Sci. USA 90, 10110–10114.

1. Hematopoietic and Endothelial Cells

29

Huang, H., Zettergren, L. D., and Auerbach, R. (1994). In vitro diVerentiation of B cells and myeloid cells from the early mouse embryo and its extraembryonic yolk sac. Exp. Hematol. 22, 19–25. Huber, T. L., Zhou, Y., Mead, P. E., and Zon, L. I. (1998). Cooperative eVects of growth factors involved in the induction of hematopoietic mesoderm. Blood 92, 4128–4137. Ivanova, N. B., Dimos, J. T., Schaniel, C., Hackney, J. A., Moore, K. A., and Lemischka, I. R. (2002). A stem cell molecular signature. Science 298, 601–604. JaVredo, T., Gautier, R., Eichmann, A., and Dieterlen-Lievre, F. (1998). Intraaortic hemopoietic cells are derived from endothelial cells during ontogeny. Development 125, 4575–4583. Johansson, B. M., and Wiles, M. V. (1995). Evidence for involvement of activin A and bone morphogenetic protein 4 in mammalian mesoderm and hematopoietic development. Mol. Cell. Biol. 15, 141–151. Kanatsu, M., and Nishikawa, S. I. (1996). In vitro analysis of epiblast tissue potency for hematopoietic cell diVerentiation. Development 122, 823–830. Kaufman, M. (1992). ‘‘The Atlas of Mouse Development’’ Academic Press, London. Keller, G., Kennedy, M., Papayannopoulou, T., and Wiles, M. V. (1993). Hematopoietic commitment during embryonic stem cell diVerentiation in culture. Mol. Cell. Biol. 13, 473–486. Keller, G., Lacaud, G., and Robertson, S. (1999). Development of the hematopoietic system in the mouse. Exp. Hematol. 27, 777–787. Kennedy, M., Firpo, M., Choi, K., Wall, C., Robertson, S., Kabrun, N., and Keller, G. (1997). A common precursor for primitive erythropoiesis and definitive haematopoiesis. Nature 386, 488–493. Kitajima, K., Masuhara, M., Era, T., Enver, T., and Nakano, T. (2002). GATA-2 and GATA2/ER display opposing activities in the development and diVerentiation of blood progenitors. EMBO J. 21, 3060–3069. Kraft, H. J., Mosselman, S., Smits, H. A., Hohenstein, P., Piek, E., Chen, Q., Artzt, K., and van Zoelen, E. J. (1996). Oct-4 regulates alternative platelet-derived growth factor alpha receptor gene promoter in human embryonal carcinoma cells. J. Biol. Chem. 271, 2873–12878. Kramer, J., Hegert, C., Guan, K., Wobus, A. M., Muller, P. K., and Rohwedel, J. (2000). Embryonic stem cell–derived chondrogenic diVerentiation in vitro: Activation by BMP-2 and BMP-4. Mech. Dev. 92, 193–205. Kumano, K., Chiba, S., Kunisato, A., Sata, M., Saito, T., Nakagami-Yamaguchi, E., Yamaguchi, T., Masuda, S., Shimizu, K., Takahashi, T., Ogawa, S., Hamada, Y., and Hirai, H. (2003). Notch1 but not Notch2 is essential for generating hematopoietic stem cells from endothelial cells. Immunity 18, 699–711. Kyba, M., Perlingeiro, R. C., and Daley, G. Q. (2002). HoxB4 confers definitive lymphoidmyeloid engraftment potential on embryonic stem cell and yolk sac hematopoietic progenitors. Cell 109, 29–37. Lacaud, G., Gore, L., Kennedy, M., KouskoV, V., Kingsley, P., Hogan, C., Carlsson, L., Speck, N., Palis, J., and Keller, G. (2002). Runx1 is essential for hematopoietic commitment at the hemangioblast stage of development in vitro. Blood 100, 458–466. Larson, R. C., Lavenir, I., Larson, T. A., Baer, R., Warren, A. J., Wadman, I., Nottage, K., and Rabbitts, T. H. (1996). Protein dimerization between Lmo2 (Rbtn2) and Tal1 alters thymocyte development and potentiates T cell tumorigenesis in transgenic mice. EMBO J. 15, 1021–1027. Larsson, J., Goumans, M. J., Sjostrand, L. J., van Rooijen, M. A., Ward, D., Leveen, P., Xu, X., ten Dijke, P., Mummery, C. L., and Karlsson, S. (2001). Abnormal angiogenesis but intact hematopoietic potential in TGF-beta type I receptor-deficient mice. EMBO J. 20, 1663–1673.

30

Park et al.

Lassila, O., Eskola, J., Toivanen, P., Martin, C., and Dieterlen-Lievre, F. (1978). The origin of lymphoid stem cells studied in chick yold sac-embryo chimaeras. Nature 272, 353–354. Lawler, S., Candia, A. F., Ebner, R., Shum, L., Lopez, A. R., Moses, H. L., Wright, C. V., and Derynck, R. (1994). The murine type II TGF-beta receptor has a coincident embryonic expression and binding preference for TGF-beta 1. Development 120, 165–175. Lechleider, R. J., Ryan, J. L., Garrett, L., Eng, C., Deng, C., Wynshaw-Boris, A., and Roberts, A. B. (2001). Targeted mutagenesis of Smad1 reveals an essential role in chorioallantoic fusion. Dev. Biol. 240, 157–167. Lenardo, M. J., Staudt, L., Robbins, P., Kuang, A., Mulligan, R. C., and Baltimore, D. (1989). Repression of the IgH enhancer in teratocarcinoma cells associated with a novel octamer factor. Science 243, 544–546. Liu, B., Sun, Y., Jiang, F., Zhang, S., Wu, Y., Lan, Y., Yang, X., and Mao, N. (2003). Disruption of Smad5 gene leads to enhanced proliferation of high-proliferative potential precursors during embryonic hematopoiesis. Blood 101, 124–133. Liu, C. P., and Auerbach, R. (1991). In vitro development of murine T cells from prethymic and preliver embryonic yolk sac hematopoietic stem cells. Development 113, 1315–1323. Liu, S., Qu, Y., Stewart, T. J., Howard, M. J., Chakrabortty, S., Holekamp, T. F., and McDonald, J. W. (2000). Embryonic stem cells diVerentiate into oligodendrocytes and myelinate in culture and after spinal cord transplantation. Proc. Natl. Acad. Sci. USA 97, 6126–6131. Maeno, M., Mead, P. E., Kelley, C., Xu, R. H., Kung, H. F., Suzuki, A., Ueno, N., and Zon, L. I. (1996). The role of BMP-4 and GATA-2 in the induction and diVerentiation of hematopoietic mesoderm in Xenopus laevis. Blood 88, 1965–1972. Maltsev, V. A., Rohwedel, J., Hescheler, J., and Wobus, A. M. (1993). Embryonic stem cells diVerentiate in vitro into cardiomyocytes representing sinusnodal, atrial and ventricular cell types. Mech. Dev. 44, 41–50. Maltsev, V. A., Wobus, A. M., Rohwedel, J., Bader, M., and Hescheler, J. (1994). Cardiomyocytes diVerentiated in vitro from embryonic stem cells developmentally express cardiac-specific genes and ionic currents. Circ. Res. 75, 233–244. Martin, G. R. (1981). Isolation of a pluripotent cell line from early mouse embryos cultured in medium conditioned by teratocarcinoma stem cells. Proc. Natl. Acad. Sci. USA 78, 7634–7638. Matsuoka, S., Tsuji, K., Hisakawa, H., Xu, M. J., Ebihara, Y., Ishii, T., Sugiyama, D., Manabe, A., Tanaka, R., Ikeda, Y., Asano, S., and Nakahata, T. (2001). Generation of definitive hematopoietic stem cells from murine early yolk sac and paraaortic splanchnopleures by aorta-gonad-mesonephros region-derived stromal cells. Blood 98, 6–12. Matthews, W., Jordan, C. T., Gavin, M., Jenkins, N. A., Copeland, N. G., and Lemischka, I. R. (1991). A receptor tyrosine kinase cDNA isolated from a population of enriched primitive hematopoietic cells and exhibiting close genetic linkage to c-kit. Proc. Natl. Acad. Sci. USA 88, 9026–9030. McMahon, A. P., Ingham, P. W., and Tabin, C. J. (2003). Developmental roles and clinical significance of hedgehog signaling. Curr. Top. Dev. Biol. 53, 1–114. Medvinsky, A., and Dzierzak, E. (1996). Definitive hematopoiesis is autonomously initiated by the AGM region. Cell 86, 897–906. Medvinsky, A. L., Samoylina, N. L., Muller, A. M., and Dzierzak, E. A. (1993). An early preliver intraembryonic source of CFU-S in the developing mouse. Nature 364, 64–67. Meyers, S., Downing, J. R., and Hiebert, S. W. (1993). Identification of AML-1 and the (8;21) translocation protein (AML-1/ETO) as sequence-specific DNA-binding proteins: The runt homology domain is required for DNA binding and protein-protein interactions. Mol. Cell. Biol. 13, 6336–6345.

1. Hematopoietic and Endothelial Cells

31

Millauer, B., Wizigmann-Voos, S., Schnurch, H., Martinez, R., Moller, N. P., Risau, W., and Ullrich, A. (1993). High aYnity VEGF binding and developmental expression suggest Flk-1 as a major regulator of vasculogenesis and angiogenesis. Cell 72, 835–846. Miquerol, L., Gertsenstein, M., Harpal, K., Rossant, J., and Nagy, A. (1999). Multiple developmental roles of VEGF suggested by a LacZ-tagged allele. Dev. Biol. 212, 307–322. Miquerol, L., Langille, B. L., and Nagy, A. (2000). Embryonic development is disrupted by modest increases in vascular endothelial growth factor gene expression. Development 127, 3941–3946. Mishina, Y., Suzuki, A., Ueno, N., and Behringer, R. R. (1995). Bmpr encodes a type I bone morphogenetic protein receptor that is essential for gastrulation during mouse embryogenesis. Genes Dev. 9, 3027–3037. Mitsui, K., Tokuzawa, Y., Itoh, H., Segawa, K., Murakami, M., Takahashi, K., Maruyama, M., Maeda, M., and Yamanaka, S. (2003). The homeoprotein Nanog is required for maintenance of pluripotency in mouse epiblast and ES cells. Cell 113, 631–642. Miyoshi, H., Kozu, T., Shimizu, K., Enomoto, K., Maseki, N., Kaneko, Y., Kamada, N., and Ohki, M. (1993). The t(8;21) translocation in acute myeloid leukemia results in production of an AML1-MTG8 fusion transcript. EMBO J. 12, 2715–2721. Moore, M. A., and Metcalf, D. (1970). Ontogeny of the haemopoietic system: Yolk sac origin of in vivo and in vitro colony forming cells in the developing mouse embryo. Br. J. Haematol. 18, 279–296. Moreau, J. F., Bonneville, M., Godard, A., Gascan, H., Gruart, V., Moore, M. A., and Soulillou, J. P. (1987a). Characterization of a factor produced by human T cell clones exhibiting eosinophil-activating and burst-promoting activities. J. Immunol. 138, 3844–3849. Moreau, J. F., Bonneville, M., Godard, A., Peyrat, M. A., Capron, M., Moore, M. A., and Soulillou, J. P. (1987b). Allogeneic T lymphocyte clones derived from rejected human kidney produce high levels of a new lymphokine active on murine interleukin 3-sensitive cell lines. Transplant Proc. 19, 300–302. Moreau, J. F., Donaldson, D. D., Bennett, F., Witek-Giannotti, J., Clark, S. C., and Wong, G. G. (1988). Leukaemia inhibitory factor is identical to the myeloid growth factor human interleukin for DA cells. Nature 336, 690–692. Motoike, T., Markham, D. W., Rossant, J., and Sato, T. N. (2003). Evidence for novel fate of Flk1þ progenitor: Contribution to muscle lineage. Genesis 35, 153–159. Muller, A. M., and Dzierzak, E. A. (1993). ES cells have only a limited lymphopoietic potential after adoptive transfer into mouse recipients. Development 118, 1343–1351. Murray, P. D. F. (1932). The development in vitro of the blood of the early chick embryo. Proc. Roy. Soc. London 11, 497–521. Nakano, T., Kodama, H., and Honjo, T. (1994). Generation of lymphohematopoietic cells from embryonic stem cells in culture. Science 265, 1098–1101. Nichols, J., Zevnik, B., Anastassiadis, K., Niwa, H., Klewe-Nebenius, D., Chambers, I., Scholer, H., and Smith, A. (1998). Formation of pluripotent stem cells in the mammalian embryo depends on the POU transcription factor Oct4. Cell 95, 379–391. Nishikawa, S. I., Nishikawa, S., Kawamoto, H., Yoshida, H., Kizumoto, M., Kataoka, H., and Katsura, Y. (1998a). In vitro generation of lymphohematopoietic cells from endothelial cells purified from murine embryos. Immunity 8, 761–769. Nishikawa, S. I., Nishikawa, S., Hirashima, M., Matsuyoshi, N., and Kodama, H. (1998b). Progressive lineage analysis by cell sorting and culture identifies FLK1þ VE-cadherinþ cells at a diverging point of endothelial and hemopoietic lineages. Development 125, 1747–1757. Nishimoto, M., Fukushima, A., Okuda, A., and Muramatsu, M. (1999). The gene for the embryonic stem cell coactivator UTF1 carries a regulatory element which selectively interacts with a complex composed of Oct-3/4 and Sox-2. Mol. Cell. Biol. 19, 5453–5465.

32

Park et al.

Nisson, P. E., Watkins, P. C., and Sacchi, N. (1992). Transcriptionally active chimeric gene derived from the fusion of the AML1 gene and a novel gene on chromosome 8 in t(8;21) leukemic cells. Cancer Genet. Cytogenet. 63, 81–88. Niwa, H. (2001). Molecular mechanism to maintain stem cell renewal of ES cells. Cell Struct. Funct. 26, 137–148. Niwa, H., Burdon, T., Chambers, I., and Smith, A. (1998). Self-renewal of pluripotent embryonic stem cells is mediated via activation of STAT3. Genes Dev. 12, 2048–2060. Niwa, H., Miyazaki, J., and Smith, A. G. (2000). Quantitative expression of Oct-3/4 defines diVerentiation, dediVerentiation or self-renewal of ES cells. Nat. Genet. 24, 372–376. North, T., Gu, T. L., Stacy, T., Wang, Q., Howard, L., Binder, M., Marin-Padilla, M., and Speck, N. A. (1999). Cbfa2 is required for the formation of intra-aortic hematopoietic clusters. Development 126, 2563–2575. North, T. E., de Bruijn, M. F., Stacy, T., Talebian, L., Lind, E., Robin, C., Binder, M., Dzierzak, E., and Speck, N. A. (2002). Runx1 expression marks long-term repopulating hematopoietic stem cells in the midgestation mouse embryo. Immunity 16, 661–672. Ogawa, E., Inuzuka, M., Maruyama, M., Satake, M., Naito-Fujimoto, M., Ito, Y., and Shigesada, K. (1993). Molecular cloning and characterization of PEBP2 beta, the heterodimeric partner of a novel Drosophila runt-related DNA binding protein PEBP2 alpha. Virology 194, 314–331. Ohishi, K., Katayama, N., Shiku, H., Varnum-Finney, B., and Bernstein, I. D. (2003). Notch signalling in hematopoiesis. Semin. Cell Dev. Biol. 14, 143–150. Ohneda, K., and Yamamoto, M. (2002). Roles of hematopoietic transcription factors GATA-1 and GATA-2 in the development of red blood cell lineage. Acta Haematol. 108, 237–245. Okamoto, K., Okazawa, H., Okuda, A., Sakai, M., Muramatsu, M., and Hamada, H. (1990). A novel octamer binding transcription factor is diVerentially expressed in mouse embryonic cells. Cell 60, 461–472. Okuda, T., van Deursen, J., Hiebert, S. W., Grosveld, G., and Downing, J. R. (1996). AML1, the target of multiple chromosomal translocations in human leukemia, is essential for normal fetal liver hematopoiesis. Cell 84, 321–330. Orkin, S. H. (1992). GATA-binding transcription factors in hematopoietic cells. Blood 80, 575–581. Oshima, M., Oshima, H., and Taketo, M. M. (1996). TGF-beta receptor type II deficiency results in defects of yolk sac hematopoiesis and vasculogenesis. Dev. Biol. 179, 297–302. Palis, J., Chan, R. J., Koniski, A., Patel, R., Starr, M., and Yoder, M. C. (2001). Spatial and temporal emergence of high proliferative potential hematopoietic precursors during murine embryogenesis. Proc. Natl. Acad. Sci. USA 98, 4528–4533. Palis, J., Robertson, S., Kennedy, M., Wall, C., and Keller, G. (1999). Development of erythroid and myeloid progenitors in the yolk sac and embryo proper of the mouse. Development 126, 5073–5084. Pardanaud, L., Luton, D., Prigent, M., Bourcheix, L. M., Catala, M., and Dieterlen-Lievre, F. (1996). Two distinct endothelial lineages in ontogeny, one of them related to hemopoiesis. Development 122, 1363–1371. Park, C., Afrikanova, I., Chung, Y. S., Zhang, W. J., Arentson, E., Fong, G.-H. G., Rosendahl, A., and Choi, K. (2004). A hierarchical order of factors in the generation of FLK1- and SCL-expressing hematopoietic and endothelial progenitors from embryonic stem cells. Development 131, 2749–2762. Pelosi, E., Valtieri, M., Coppola, S., Botta, R., Gabbianelli, M., Lulli, V., Marziali, G., Masella, B., Muller, R., Sgadari, C., Testa, U., Bonanno, G., and Peschle, C. (2002). Identification of the hemangioblast in postnatal life. Blood 100, 3203–3208.

1. Hematopoietic and Endothelial Cells

33

Perez-Alvarado, G. C., Miles, C., Michelsen, J. W., Louis, H. A., Winge, D. R., Beckerle, M. C., and Summers, M. F. (1994). Structure of the carboxy-terminal LIM domain from the cysteine rich protein CRP. Nat. Struct. Biol. 1, 388–398. Perlingeiro, R. C., Kyba, M., and Daley, G. Q. (2001). Clonal analysis of diVerentiating embryonic stem cells reveals a hematopoietic progenitor with primitive erythroid and adult lymphoid-myeloid potential. Development 128, 4597–4604. Pesce, M., Wang, X., Wolgemuth, D. J., and Scholer, H. (1998). DiVerential expression of the Oct-4 transcription factor during mouse germ cell diVerentiation. Mech. Dev. 71, 89–98. Porcher, C., Liao, E. C., Fujiwara, Y., Zon, L. I., and Orkin, S. H. (1999). Specification of hematopoietic and vascular development by the bHLH transcription factor SCL without direct DNA binding. Development 126, 4603–4615. Potocnik, A. J., Kohler, H., and Eichmann, K. (1997). Hemato-lymphoid in vivo reconstitution potential of subpopulations derived from in vitro diVerentiated embryonic stem cells. Proc. Natl. Acad. Sci. USA 94, 10295–10300. Potocnik, A. J., Nielsen, P. J., and Eichmann, K. (1994). In vitro generation of lymphoid precursors from embryonic stem cells. EMBO J. 13, 5274–5283. Qi, X., Li, T. G., Hao, J., Hu, J., Wang, J., Simmons, H., Miura, S., Mishina, Y., and Zhao, G. Q. (2004). BMP4 supports self-renewal of embryonic stem cells by inhibiting mitogen-activated protein kinase pathways. Proc. Natl. Acad. Sci. USA 101, 6027–6032. Ramalho-Santos, M., Yoon, S., Matsuzaki, Y., Mulligan, R. C., and Melton, D. A. (2002). ‘‘Stemness’’: Transcriptional profiling of embryonic and adult stem cells. Science 298, 597–600. Risau, W., Sariola, H., Zerwes, H. G., Sasse, J., Ekblom, P., Kemler, R., and Doetschman, T. (1988). Vasculogenesis and angiogenesis in embryonic stem cell–derived embryoid bodies. Development 102, 471–478. Robb, L., Elwood, N. J., Elefanty, A. G., Kontgen, F., Li, R., Barnett, L. D., and Begley, C. G. (1996). The scl gene product is required for the generation of all hematopoietic lineages in the adult mouse. EMBO J. 15, 4123–4129. Robb, L., Lyons, I., Li, R., Hartley, L., Kontgen, F., Harvey, R. P., Metcalf, D., and Begley, C. G. (1995). Absence of yolk sac hematopoiesis from mice with a targeted disruption of the scl gene. Proc. Natl. Acad. Sci. USA 92, 7075–7079. Robertson, S. M., Kennedy, M., Shannon, J. M., and Keller, G. (2000). A transitional stage in the commitment of mesoderm to hematopoiesis requiring the transcription factor SCL/tal-1. Development 127, 2447–2459. Rohwedel, J., Maltsev, V., Bober, E., Arnold, H. H., Hescheler, J., and Wobus, A. M. (1994). Muscle cell diVerentiation of embryonic stem cells reflects myogenesis in vivo: Developmentally regulated expression of myogenic determination genes and functional expression of ionic currents. Dev. Biol. 164, 87–101. Rosner, M. H., Vigano, M. A., Ozato, K., Timmons, P. M., Poirier, F., Rigby, P. W., and Staudt, L. M. (1990). A POU-domain transcription factor in early stem cells and germ cells of the mammalian embryo. Nature 345, 686–692. Sabin, F. R. (1920). Studies on the origin of blood vessels and of red corpuscles as seen in the living blastoderm of the chick during the second day of incubation. Contrib. Embryol. 9, 213–262. Saijoh, Y., Fujii, H., Meno, C., Sato, M., Hirota, Y., Nagamatsu, S., Ikeda, M., and Hamada, H. (1996). Identification of putative downstream genes of Oct-3, a pluripotent cell-specific transcription factor. Genes Cells 1, 239–252. Sauvageau, G., Thorsteinsdottir, U., Eaves, C. J., Lawrence, H. J., Largman, C., Lansdorp, P. M., and Humphries, R. K. (1995). Overexpression of HOXB4 in hematopoietic cells causes the selective expansion of more primitive populations in vitro and in vivo. Genes Dev. 9, 1753–1765.

34

Park et al.

Scholer, H. R., Hatzopoulos, A. K., Balling, R., Suzuki, N., and Gruss, P. (1989). A family of octamer-specific proteins present during mouse embryogenesis: Evidence for germlinespecific expression of an Oct factor. EMBO J. 8, 2543–2550. Shivdasani, R. A., Mayer, E. L., and Orkin, S. H. (1995). Absence of blood formation in mice lacking the T-cell leukaemia oncoprotein tal-1/SCL. Nature 373, 432–434. Singh, H., Sen, R., Baltimore, D., and Sharp, P. A. (1986). A nuclear factor that binds to a conserved sequence motif in transcriptional control elements of immunoglobulin genes. Nature 319, 154–158. Smith, A. G., Heath, J. K., Donaldson, D. D., Wong, G. G., Moreau, J., Stahl, M., and Rogers, D. (1988). Inhibition of pluripotential embryonic stem cell diVerentiation by purified polypeptides. Nature 336, 688–690. Smith, A. G., and Hooper, M. L. (1987). BuValo rat liver cells produce a diVusible activity which inhibits the diVerentiation of murine embryonal carcinoma and embryonic stem cells. Dev. Biol. 121, 1–9. Soriano, P. (1999). Generalized lacZ expression with the ROSA26 Cre reporter strain. Nat. Genet. 21, 70–71. Staudt, L. M., Singh, H., Sen, R., Wirth, T., Sharp, P. A., and Baltimore, D. (1986). A lymphoid-specific protein binding to the octamer motif of immunoglobulin genes. Nature 323, 640–643. Strubing, C., Ahnert-Hilger, G., Shan, J., Wiedenmann, B., Hescheler, J., and Wobus, A. M. (1995). DiVerentiation of pluripotent embryonic stem cells into the neuronal lineage in vitro gives rise to mature inhibitory and excitatory neurons. Mech. Dev. 53, 275–287. Tavian, M., Coulombel, L., Luton, D., Clemente, H. S., Dieterlen-Lievre, F., and Peault, B. (1996). Aorta-associated CD34þ hematopoietic cells in the early human embryo. Blood 87, 67–72. Tavian, M., Hallais, M. F., and Peault, B. (1999). Emergence of intraembryonic hematopoietic precursors in the pre-liver human embryo. Development 126, 793–803. Thomson, J. A., Itskovitz-Eldor, J., Shapiro, S. S., Waknitz, M. A., Swiergiel, J. J., Marshall, V. S., and Jones, J. M. (1998). Embryonic stem cell lines derived from human blastocysts. Science 282, 1145–1147. Thomson, J. A., Kalishman, J., Golos, T. G., Durning, M., Harris, C. P., Becker, R. A., and Hearn, J. P. (1995). Isolation of a primate embryonic stem cell line. Proc. Natl. Acad. Sci. USA 92, 7844–7848. Thomson, J. A., Kalishman, J., Golos, T. G., Durning, M., Harris, C. P., and Hearn, J. P. (1996). Pluripotent cell lines derived from common marmoset (Callithrix jacchus) blastocysts. Biol. Reprod. 55, 254–259. Tomilin, A., Remenyi, A., Lins, K., Bak, H., Leidel, S., Vriend, G., Wilmanns, M., and Scholer, H. R. (2000). Synergism with the coactivator OBF-1 (OCA-B, BOB-1) is mediated by a specific POU dimer configuration. Cell 103, 853–864. Tremblay, K. D., Dunn, N. R., and Robertson, E. J. (2001). Mouse embryos lacking Smad1 signals display defects in extra-embryonic tissues and germ cell formation. Development 128, 3609–3621. Tsai, F. Y., Keller, G., Kuo, F. C., Weiss, M., Chen, J., Rosenblatt, M., Alt, F. W., and Orkin, S. H. (1994). An early haematopoietic defect in mice lacking the transcription factor GATA-2. Nature 371, 221–226. Valge-Archer, V. E., Osada, H., Warren, A. J., Forster, A., Li, J., Baer, R., and Rabbitts, T. H. (1994). The LIM protein RBTN2 and the basic helix-loop-helix protein TAL1 are present in a complex in erythroid cells. Proc. Natl. Acad. Sci. USA 91, 8617–8621. Visvader, J. E., Fujiwara, Y., and Orkin, S. H. (1998). Unsuspected role for the T-cell leukemia protein SCL/tal-1 in vascular development. Genes Dev. 12, 473–479.

1. Hematopoietic and Endothelial Cells

35

Vittet, D., Prandini, M. H., Berthier, R., Schweitzer, A., Martin-Sisteron, H., Uzan, G., and Dejana, E. (1996). Embryonic stem cells diVerentiate in vitro to endothelial cells through successive maturation steps. Blood 88, 3424–3431. Wadman, I. A., Osada, H., Grutz, G. G., Agulnick, A. D., Westphal, H., Forster, A., and Rabbitts, T. H. (1997). The LIM-only protein Lmo2 is a bridging molecule assembling an erythroid, DNA-binding complex which includes the TAL1, E47, GATA1 and Ldb1/NLI proteins. EMBO J. 16, 3145–3157. Wagner, R. C. (1980). Endothelial cell embryology and growth. Adv. Microcirc. 9, 45–75. Wang, Q., Stacy, T., Binder, M., Marin-Padilla, M., Sharpe, A., and Speck, N. (1996). Disruption of the Cbfa2 gene causes necrosis and hemorrhaging in the central nervous system and blocks definitive hematopoiesis. Proc. Natl. Acad. Sci. USA 93, 3444–3449. Wang, R., Clark, R., and Bautch, V. L. (1992). Embryonic stem cell–derived cystic embryoid bodies form vascular channels: An in vitro model of blood vessel development. Development 114, 303–316. Wang, S., Wang, Q., Crute, B. E., Melnikova, I. N., Keller, S. R., and Speck, N. A. (1993). Cloning and characterization of subunits of the T-cell receptor and murine leukemia virus enhancer core-binding factor. Mol. Cell. Biol. 13, 3324–3339. Wang, S. H., Tsai, M. S., Chiang, M. F., and Li, H. (2003). A novel NK-type homeobox gene, ENK (early embryo specific NK), preferentially expressed in embryonic stem cells. Gene Expr. Patterns 3, 99–103. Warren, A. J., Colledge, W. H., Carlton, M. B., Evans, M. J., Smith, A. J., and Rabbitts, T. H. (1994). The oncogenic cysteine-rich LIM domain protein rbtn2 is essential for erythroid development. Cell 78, 45–57. Weiss, M. J., Keller, G., and Orkin, S. H. (1994). Novel insights into erythroid development revealed through in vitro diVerentiation of GATA-1 embryonic stem cells. Genes Dev. 8, 1184–1197. Weissman, I., Papaioannou, V., and Gardner, R. (1978). ‘‘Fetal Hematopoietic Origins of the Adult Hematolymphoid System.’’ Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. Wiles, M. V., and Keller, G. (1991). Multiple hematopoietic lineages develop from embryonic stem (ES) cells in culture. Development 111, 259–267. Williams, R. L., Hilton, D. J., Pease, S., Willson, T. A., Stewart, C. L., Gearing, D. P., Wagner, E. F., Metcalf, D., Nicola, N. A., and Gough, N. M. (1988). Myeloid leukaemia inhibitory factor maintains the developmental potential of embryonic stem cells. Nature 336, 684–687. Winnier, G., Blessing, M., Labosky, P. A., and Hogan, B. L. (1995). Bone morphogenetic protein-4 is required for mesoderm formation and patterning in the mouse. Genes Dev. 9, 2105–2116. Wong, P. M., Chung, S. W., Chui, D. H., and Eaves, C. J. (1986). Properties of the earliest clonogenic hemopoietic precursors to appear in the developing murine yolk sac. Proc. Natl. Acad. Sci. USA 83, 3851–3854. Wotton, D., Ghysdael, J., Wang, S., Speck, N. A., and Owen, M. J. (1994). Cooperative binding of Ets-1 and core binding factor to DNA. Mol. Cell. Biol. 14, 840–850. Wotton, D., Prosser, H. M., and Owen, M. J. (1993). Regulation of human T cell receptor beta gene expression by Ets-1. Leukemia 7, S55–S60. Yamada, G., Kioussi, C., Schubert, F. R., Eto, Y., Chowdhury, K., Pituello, F., and Gruss, P. (1994). Regulated expression of Brachyury(T), Nkx1.1 and Pax genes in embryoid bodies. Biochem. Biophys. Res. Commun. 199, 552–563. Yamaguchi, Y., Kurokawa, M., Imai, Y., Izutsu, K., Asai, T., Ichikawa, M., Yamamoto, G., Nitta, E., Yamagata, T., Sasaki, K., Mitani, K., Ogawa, S., Chiba, S., and Hirai, H. (2004).

36

Park et al.

AML1 is functionally regulated through p300-mediated acetylation on specific lysine residues. J. Biol. Chem. 279, 15630–15638. Yamashita, J., Itoh, H., Hirashima, M., Ogawa, M., Nishikawa, S., Yurugi, T., Naito, M., Nakao, K., and Nishikawa, S. (2000). Flk1-positive cells derived from embryonic stem cells serve as vascular progenitors. Nature 408, 92–96. Ying, Q. L., Nichols, J., Chambers, I., and Smith, A. (2003). BMP induction of Id proteins suppresses diVerentiation and sustains embryonic stem cell self-renewal in collaboration with STAT3. Cell 115, 281–292. Yoder, M. C., and Hiatt, K. (1997). Engraftment of embryonic hematopoietic cells in conditioned newborn recipients. Blood 89, 2176–2183. Yoder, M. C., Hiatt, K., Dutt, P., Mukherjee, P., Bodine, D. M., and Orlic, D. (1997a). Characterization of definitive lymphohematopoietic stem cells in the day 9 murine yolk sac. Immunity 7, 335–344. Yoder, M. C., Hiatt, K., and Mukherjee, P. (1997b). In vivo repopulating hematopoietic stem cells are present in the murine yolk sac at day 9.0 postcoitus. Proc. Natl. Acad. Sci. USA 94, 6776–6780. Yuan, H., Corbi, N., Basilico, C., and Dailey, L. (1995). Developmental-specific activity of the FGF-4 enhancer requires the synergistic action of Sox2 and Oct-3. Genes Dev. 9, 2635–2645.

2

____________________________________________________________________________

Fibroblast Growth Factor Signaling and the Function and Assembly of Basement Membranes Peter Lonai { Department of Molecular Genetics The Weizmann Institute of Science Rehovot, Israel 76100

I. Introduction II. Earlyl Embryogenesis, Growth Factors, Growth Factor Receptors, and the Basement Membrane A. Early Mammalian Development B. Embryoid Bodies: A Model for Early Mammalian Development C. Fibroblast Growth Factor Signaling During Early Embryogenesis D. Relative Localization of Growth Factors, Growth Factor Receptor, and the Basement Membrane E. AYnity of Growth Factors and Their Receptors to Heparin and Heparan Sulfates F. Basement Membranes and Their Network-Forming Elements III. Laminins and Basement Membrane–Mediated Signaling A. Genetic Analysis of Laminins B. Laminin-1 and Epiblast DiVerentiation C. Endoderm and Ectoderm DiVerentiation Follow DiVerent Pathways D. From Stem Cells to Pregastrulation Embryo: A Cascade of Cellular and Molecular Interactions E. Laminin Receptors and Anchorage Sites IV. Current Questions References

I. Introduction The subject of this chapter is the contribution of the extracellular matrix (ECM) to signal transduction during early development and epithelial diVerentiation. Special emphasis is given to basement membranes (BMs) and fibroblast growth factors (FGFs) as studied in embryoid body (EB) cultures. Extensive literature analyses the role of the ECM in cell physiology. This activity gained great impetus from recognizing, already more than 30 years {

Deceased

Current Topics in Developmental Biology, Vol. 66 Copyright 2005, Elsevier Inc. All rights reserved. 0070-2153/05 $35.00

37

38

Peter Lonai

ago, that epithelial mesenchymal interactions are required for morphogenesis and that many cells, and especially cell sheets, interact via the ECM (Kallman and Grobstein, 1965; Saxen and ThesleV, 1992; Wessels, 1970). The ensuing studies were summarized in excellent reviews on the protein chemistry and structural biology of BM (Colognato and Yurchenco, 2000; Hohenester et al., 1999a; Timpl and Brown, 1996), and on the biochemistry of intercellular haparan sulfate molecules (Bernfield et al., 1999) and their integrin receptors (Hynes, 2002). The role of the ECM, and especially that of the BM, in tubulogenesis (Hogan and Kolodziej, 2002; Lubarsky and Krasnow, 2003; Sottile, 2004), branching morphogenesis (Hogan and Yingling, 1998), cell migration, metastasis formation (Kalluri, 2003; Patarroyo et al., 2002), tumor angiogenesis (Folkman, 2002; Kalluri, 2003), and in tumor–stroma interactions (Fata et al., 2004) greatly advanced this field. This large body of evidence pointed to the great importance of the ECM in manifold physiological and pathological processes rather than explaining the mechanism of its contributions. Recent progress in understanding the connection between laminin isotypes and signal transduction in the EB system provided a specific if narrow insight into the mechanism of BM-mediated epithelialization in the early embryo. Our discussion pays special attention to these results.

II. Early Embryogenesis, Growth Factors, Growth Factor Receptors, and the Basement Membrane A. Early Mammalian Development One focus of our discussion concentrates on interactions between the BM separating the extraembryonic endoderm and the epiblast of the mammalian embryo. These early stages of development bridge the diVerentiation of the preimplantation blastocyst into the pregastrulation egg-cylinder embryo. Before analyzing the ECM’s role in these early processes of embryogenesis, a brief introduction into early mammalian development and into the relationship of growth factors with the ECM may be useful. Following fertilization, the mammalian embryo goes through a limited number of cleavage divisions during which the blastomers retain their pluripotency. At the eight-cell stage the mouse embryo undergoes compaction, when the hitherto loosely attached blastomers form a tightly packed mass, and the contour of the individual blastomers becomes barely detectable. The first cell fate decisions take place during this stage. At compaction, adhesion structures become expressed (Johnson et al., 1986) and, as shown by Graham and his colleagues, the external blastomers become committed to trophoblast, while the inner blastomers acquire intercellular matrix (ICM) fate (Graham, 1978). Major molecular elements of compaction are

2. Fibroblast Growth Factor Signaling

39

E-cadherin and the connected -catenin-APC-wnt pathway (Ohsugi et al., 1996). The earliest expressed FGF receptor, Fgfr2, is also expressed at this stage, when it is localized to the external blastomers and later to the trophectoderm lineage (HaVner-Krausz et al., 1999). As blastocyst development starts, a fluid-filled cavity forms in the morula and apical–basal polarity is established in the peripheral cells. Cells of the prospective ICM, some 20 in number, are the ancestors of all embryonic tissues. The ICM is localized at the apical pole of the blastocyst, and the rest of the blastocyst is covered by a polarized epithelium, the trophectoderm. An additional cell fate decision takes place in the late blastocyst [at embryonic day (E) 3–3.5 in mouse embryos]; ICM cells bordering the blastocyst cavity transform into primitive endoderm (Gardner, 1982). At this stage the ICM expresses Fgf4 (Niswander and Martin, 1992), the ligand of Fgfr2, which is localized in the adjacent trophectoderm (HaVner-Krausz et al., 1999). Trophoblast proliferation is promoted by Fgf4 and Fgfr2 (Tanaka et al., 1998), which are also required for endoderm diVerentiation (Chen et al., 2000; Feldman et al., 1995; Goldin and Papaioannou, 2003). The first BM components, laminin- 1 and laminin- 1, are expressed at the eight-cell stage. In the blastocyst, laminin-1 and 11 (Fig. 1A) (L. Sorokin, personal communication) and collagen IV (Fig. 1C) are expressed in the BM situated along the basal side of the trophectoderm and later also at the basal side of the primitive endoderm that separates the ICM from the endoderm (Fig. 1A–C). Laminin-1 and laminin-5 (laminin-10–11) are expressed together at this stage (Ekblom et al., 2003; Miner et al., 2004). After implantation, the ICM and the polar trophectoderm covering it undergo fast proliferation and intrude into the former blastocyst cavity, forming a conical structure, the so-called egg-cylinder. The primitive endoderm also proliferates and covers the entire cavity of the late blastocyst. At the eggcylinder stage, the endoderm covering the egg-cylinder becomes the visceral endoderm, whereas the layer associated with the trophectoderm becomes the parietal endoderm. The parietal endoderm synthesizes the thick BM of the Reichert membrane at the maternal–fetal interface, and the visceral endoderm synthesizes the subendodermal embryonic and extraembryonic BM (Fig. 1E). Initially the egg-cylinder contains round stem cells, which are similar to the cells of the ICM and form a solid epithelial bud. Later, close to gastrulation, (between E5.5 and 6.0 in the mouse embryo), a cavity in the proamniotic canal is formed and the stem cells of the primitive endoderm (ICM) surrounding it diVerentiate into a columnar monolayer, the pseudo-stratified columnar epithelium of the epiblast. With epithelialization of the primitive ectoderm into epiblast, the early postimplantation embryo is ready for gastrulation. During gastrulation the definitive germ layers, the neuroectoderm, the mesoderm, and the embryonic endoderm are formed; the segmented body plan is realized; and organ and limb development commences. An excellent short description of early embryogenesis and gastrulation can be found in Manipulating the

40

Peter Lonai

Figure 1 Distribution of basement membrane proteins in the early mouse embryo. (A)–(C) 3.5 days postcoitum (dpc) blastocysts; (D) 4-day-old embryoid body; (E) 7.5 dpc embryo. (A)–(D) Whole mount confocal images according to Li et al. (2001a). (E) Section, immunofluorescence. (A)–(D) Green: fluoresceinated phalloidin to detect fibrillar actin. Red: specific antibody. (A) 1LG4; (B) laminin- 1; (C) Collagen IV; (D) 1LG4; (E) 1LG4. Arrow in (A) and (B): ICM; arrowhead in (A) and (B): primitive endoderm. In (E) arrow: subendodermal BM; double arrow: extraembryonic subendodermal BM; arrowhead: Reichert’s membrane.

Mouse Embryo by Hogan et al. (1994), and a more detailed description is given by Tam and Behringer (1997).

B. Embryoid Bodies: A Model for Early Mammalian Development EB cultures constitute an ideal system for investigating pregastrulation development. Cultured ES cells diVerentiate into EBs when grown on bacteriological plates, a condition that does not support cell adhesion (Martin

2. Fibroblast Growth Factor Signaling

41

et al., 1977). On bacteriological plates floating ES cell aggregates form, which diVerentiate into round structures made of two epithelia, the outer endoderm and the inner ectoderm layer. The endoderm synthesizes laminin and type IV collagen isotypes, which form the network of the subendodermal BM and consist of vacuolated polar cells, which at their basal domain face the BM, whereas their external apical domain wears microvilli. The endoderm of the EB, both by its morphology and by virtue of the genes it expresses, resembles the primitive and later visceral endoderm of the late blastocyst and the pregastrulation egg-cylinder embryo. The inner ectoderm of the EB is a pseudo-stratified columnar epithelium. Its basal domain is attached to the subendodermal BM, and its apical domain faces the central cavity and is distinguished by an accumulation of actin fibers. The columnar epithelium of the EB is similar to the embryonic ectoderm or epiblast, an epithelium, which gives rise to all cell lineages of the embryo. Taken together, the EB contains two epithelia, each of which is attached in a reversed polarity to a common subendodermal BM at their basal domains. EBs are considered to be faithful models of the pregastrulation mouse embryo (Coucouvanis and Martin, 1995). Although the pregastrulation mouse embryo is exceedingly small and diYcult to isolate, EBs can be grown in large amounts with considerable ease. This is why EB cultures became the tool for recent research on the role of the BM in epithelial diVerentiation (Henry and Campbell, 1998; Li et al., 2001a, 2002, 2004; Murray and Edgar, 2000).

C. Fibroblast Growth Factor Signaling During Early Embryogenesis Of the 24 FGF genes, three, Fgf4, Fgf3, and Fgf5, are expressed during preimplantation embryogenesis. According to genetic evidence, Fgf5 is a regulator of the hair growth cycle (Hebert et al., 1994), Fgf3 is required for normal ear and tail development (Mansour et al., 1993), while Fgf4 is the only one that is by itself required for trophectoderm and primitive endoderm diVerentiation of the late preimplantation blastocyst and early postimplantation embryo (Feldman et al., 1995; Goldin and Papaioannou, 2003; Wilder et al., 1997), where it is expressed in the ICM (Niswander and Martin, 1992). None of the four FGF receptors (FGFRs) is individually required for early gastrulation or pregastrulation embryogenesis. Fgfr1 is required for development of the posterior mesoderm (Deng et al., 1994; Yamaguchi et al., 1994) by controlling the migration of the prospective mesoderm through the primitive streak (Ciruna and Rossant, 2001). Loss of Fgfr2 is lethal at midgestation due to severe defects of lung diVerentiation and causes the complete loss of limb outgrowth (Arman et al., 1999; Xu et al., 1998). It has

42

Peter Lonai

been shown that this Fgfr2 loss of function phenotype is due to the b transcriptional variant (De Moerlooze et al., 2000), whereas targeted mutagenesis of the Fgfr2c variant causes craniofacial and bone development anomalies due to defective osteoblast diVerentiation (Eswarakumar et al., 2002, 2004). Fgfr3, somewhat in contrast to Fgfr2c, exhibits negative control of chondrogenesis during endochondral bone formation (Colvin et al., 1996; Deng et al., 1996), whereas loss of Fgfr4 exhibits no independent phenotype and cooperates with Fgfr3 in the control of lung alveogenesis (Weinstein et al., 1998). The role of FGF signaling in early embryogenesis is suggested by the requirement of Fgf4 for trophectoderm and endoderm development (Goldin and Papaioannou, 2003) and by the finding that dominant negative FGFR mutation elicited by expressing truncated Fgfr2 cDNA in ES cells abolishes endoderm and ectoderm diVerentiation in the EB, where multiple Fgfr are expressed in synchrony (Chen et al., 2000). This latter result explains why a rearranged targeting vector of Fgfr2 induced peri-implantation lethality and loss of endoderm diVerentiation (Arman et al., 1998). Involvement of FGF signaling at stages preceding endoderm diVerentiation was reported by Chai et al. (1998), who found that dominant negative, truncated Fgfr4 cDNA disrupts embryogenesis in the early blastocyst after the fifth cleavage division. The early pattern of Fgfr2 expression suggests that this receptor, presumably together with other Fgfr, is important for preimplantation embryogenesis. HaVner-Krausz et al. (1999) reported that Fgfr2c is expressed maternally in the unfertilized egg and that both Fgfr2c and b are active in the external cells of the compacted morula and in the trophectoderm lineage that develops from them. According to the present consensus, Fgf4 in the ICM mediates early FGF signaling. Fgf4 signals are transmitted by Fgfr2 and other Fgfr localized in the trophectoderm and later in the primitive endoderm. Better understanding of the exact distribution of FGFR in the early embryo was hindered by the lack of splice variant–specific antibodies that are suitable for immunofluorescence detection; moreover, the resolution of in situ hybridization proved to be less than adequate for the resolution of the very small distances in the preimplantation embryo. New antibodies and double-mutant crosses should provide more definite understanding. Much is still to be learned about the gene network of preimplantation development. As mentioned before, the first cell fate decision is made at compaction, when the outer cells of the eight-cell mouse morula acquire trophectoderm fate while the inner cells acquire ICM fate (Graham, 1978). Compaction is achieved by E-cadherin (Johnson et al., 1986), required for trophectoderm diVerentiation (Larue et al., 1994). The earliest cell fate decisions, namely, the choice between pluripotent, trophectoderm, and endoderm cell fate, is regulated by the concentration of Oct3/4 (Niwa et al.,

2. Fibroblast Growth Factor Signaling

43

2000), whereas the pluripotency of ES cells and the primitive ectoderm is maintained by nanog (Chambers et al., 2003). A number of other genes, the expression level of which decreases with ES cell diVerentiation, may serve similar functions (Tanaka et al., 2002). Oct3/4 expression is in part controlled by COUP-TF1 and COUP-TFII (Ben-Shushan et al., 1995), which influence laminin-1 expression in the primitive endoderm (Murray and Edgar, 2001b). Endoderm diVerentiation is controlled by GATA-6 and GATA-4. These transcription factors are under FGF control. They transform ES cells into endoderm-like cells, which, among other endodermspecific proteins, express COUP-TF1 and the three chains of laminin-1 (Fujikura et al., 2002; Li et al., 2004). Akt/PKB of the PI3K pathway also contributes to the control of laminin expression (Li et al., 2001b). This eVect is regulated by FGF signaling (Chen et al., 2000).

D. Relative Localization of Growth Factors, Growth Factor Receptor, and the Basement Membrane Many interactions between growth factors and their receptors take place across BMs that separate them, since growth factors and growth factor receptors are frequently localized to cell layers adjacent to alternate sides of the BM (Lonai, 2003). The growth factor, kit, and its ligand, the stem cell factor, have been localized to neighboring epithelial and mesenchymal cell sheets (Keshet et al., 1991). A similar arrangement was found for the Pdgfra receptor and its PDGF-A ligand (Orr-Urtreger and Lonai, 1992). Outstanding examples are represented by FGFR isotypes. First it was observed that Fgfr1 is localized mainly to mesenchymal cells, whereas a structurally closely related isotype, Fgfr2, is mainly expressed in epithelia (Orr-Urtreger et al., 1991; Peters et al., 1992). More sophisticated regulation was found for localization of the transcriptional alternatives of Ffr2. The C-terminal half of the third immunoglobulin (Ig)-like loop in the ligand-binding domains of FGFR1, FGFR2, and FGFR3 are encoded by one of two mutually exclusive exons conferring diVerent ligand-binding specificity (Givol et al., 2003; Johnson and Williams, 1993). Two splice variants, b and c, are distinguished. It has been shown that Fgfr2b is mostly expressed in epithelia, whereas Fgfr2c is expressed in mesenchymes (Orr-Urtreger et al., 1993). For example, the epithelial Fgfr2b receptor binds mesenchymal FGF isotypes, such as Fgf10 (Ohuchi et al., 1997), whereas the mesenchymal Fgfr2c variant recognizes Fgf9, since it has been demonstrated for epithelia of the developing lung (Arman et al., 1999). General applicability of this mutual regulation was supported by a study of the mitogenicity of FGFR isotypes in BAF3 cells expressing various FGF isotypes. It was found that c- and b-type receptors recognize separate groups of FGF ligands, and, as far as it has

44

Peter Lonai

been determined, those that activate b-type receptors were expressed in mesenchymal tissues, whereas those that recognize c-type receptors were expressed in epithelial tissues (Ornitz et al., 1996). Numerous clinical studies indicate the prevalence of this mutual regulation. For example, misexpression of FGFR2 splice variants leads to defective osteogenesis, as observed in Apert syndrome (Hajihosseini et al., 2001; Ibrahimi et al., 2001), whereas its ligand-independent activation causes various craniosynostosis syndromes (Eswarakumar et al., 2004; Wilkie, 1997). It may not be far-fetched to suggest that this mutual control of transcriptional localization of FGF and FGFR developed to serve epithelial mesenchymal interactions.

E. Affinity of Growth Factors and Their Receptors to Heparin and Heparan Sulfates FGFs are distinguished by their aYnity to heparin and once were denominated as heparin-binding growth factors. By now most growth factors, or at least some of their variants, were shown to have aYnity to heparin or to heparan sulfates (HS) of the ECM (Turnbull et al., 2001). Because HS proteoglycans (HSPGs) are organic components of the ECM, this finding indicates that growth factors may be stored and concentrated by ECM components, such as syndecans and glypicans or the BM. The importance of cell surface HS molecules for FGF signaling was first shown by experiments with cells that carry null mutations of HS synthesis. Yayon et al. (1991) showed that FGFR expressed in mutant cells of this kind were incapable of FGF binding. HS are synthesized by a cascade of sugar transfer enzymes, and the variable sequence of their sugar moieties has the potential to create very high multiplicity (Turnbull et al., 2001). Actually it has been found that diVerent FGF isotypes have selective aYnity to diVerent HS variants (Allen and Rapraeger, 2003; Guimond and Turnbull, 2000; Ostrovsky et al., 2002). Specific contribution of heparin-like molecules to FGF signaling was demonstrated by crystallographic analysis of FGF–FGFR complexes in the presence of heparin. These studies demonstrated that the FGF–FGFR complex creates a groove, which accommodates heparin moieties contributing to the stability of the receptor–ligand complex and enhancing its dimerization (Pellegrini et al., 2000; Plotnikov et al., 1999; Schlessinger et al., 2000). Genetic analysis lent independent support to the structural and biochemical evidence for the role of HSPGs in developmental signaling. Mutant analysis in Drosophila organisms revealed their importance for early development (Perrimon and Bernfield, 2000). It has been shown that HSPG mutations phenocopy FGF and FGFR mutations in Drosophila organisms,

2. Fibroblast Growth Factor Signaling

45

which contributes to the importance of FGF–FGFR–HS linkage in FGF signaling (Lin et al., 1999). Mutations in vertebrates, including human orthologues of the genes active in HS synthesis (Bullock et al., 1998; Kurima et al., 1998; Lin et al., 2000), interfere with organogenesis at later stages than in Drosophila organisms, probably due to functional overlap in the complex vertebrate genome. Taken together, these data suggest that HSPGs of the matrix actively contribute to cell-to-cell signaling. Multiple mechanisms may be involved. HS may be instrumental in the storage and concentration of heparin-binding growth factors in the vicinity of cells that bear the relevant signaling receptor (Schlessinger et al., 1995). In the case of FGF signaling, the relationship between receptor tyrosine kinaseses, growth factors, and HS developed into a functional requirement in the sense that the signaling unit requires the contribution of specific HS components (Plotnikov et al., 1999). The specificity of receptor–matrix interactions contributes to the specificity of signal transduction, and the variability of HS molecules may have a decisive role here. Future research will have to clarify to what degree the specificity of growth factor signaling is influenced by the vast polymorphism of HS sequences.

F. Basement Membranes and Their Network-Forming Elements The role of the BM as an activator of epithelialization of the primitive endoderm is a central issue of this article. The BM and its components have been extensively reviewed (Colognato and Yurchenco, 2000; Ekblom et al., 2003; Timpl and Brown, 1996) and are only briefly introduced here. The BM is a thin, blanketlike modification of the ECM, which separates cell layers in most organs. As such it can reach large dimensions such as beneath the germinal layer of the skin, which covers the entire surface of the body. In certain organs the BM fulfills specific functions, such as in the glomerular membrane of the kidney, in synaptic membranes of neuromuscular junctions, and in the Schwann cell layers of myelinated nerves. The network-forming components of the BM are laminin and type IV collagen isotypes. Laminins are cross-shaped molecules that form an independent network (Yurchenco et al., 1992), whereas collagen IV, which is essential for the stability of the BM, is dispensable for its initiation (Poschl et al., 2004). Laminins form a flat polymer by associating through their N-terminal side chains (Cheng et al., 1997), whereas the C-terminal globular domains of the  chains provide anchoring sites to the cell membrane (Hohenester et al., 1999b). Laminin globular domains 1–3 (LG1–3) display aYnity to 6 1, 6 4, and 7 1 integrins, whereas the LG4–5 domains of the E3 peptic fragment exhibit aYnity for dystroglycan, heparin, and

46

Peter Lonai

sulfatides. Recent data separate the dystroglycan-binding site from the site that binds heparin and sulfatides (Wizemann et al., 2003). Fifteen laminin isotypes are distinguished. They use of one of five , one of three , and one of three chains (Table I). All  chains have cell-binding globular domains, and the globular domains of 2 (Talts et al., 1998) 4 (Talts et al., 2000), and 5 (Nielsen et al., 2000) were shown to have similar structure and binding aYnity. Although laminin-1 is the most characteristic isotype during early development (Ekblom et al., 2003), it is localized to specific sites in the adult as well (Virtanen et al., 2000). The exact localization of laminin isotypes is far from understood, and multiple isotypes are expressed in specific sites of certain organs throughout development and in the adult. Well-illustrated examples are the kidney (Miner et al., 1997) and the neuromuscular junction (Patton et al., 1997), which express multiple laminin isotypes at specific localizations. Inhibition with antibodies to specific peptic fragments of laminin chains revealed that the E3 fragment of the 1 chain, which contains the LG4–5 modules, is required for kidney or salivary gland diVerentiation in organ culture (Kadoya et al., 1995, 2003; Klein et al., 1988, 1990; Sorokin et al., 1992). These results, as well as the protein chemistry of laminin globular domains (Timpl et al., 2000), indicate that BM anchorage and BMmediated signaling resides in the LG modules of laminin  chains. Therefore, although many other BM components, such as agrin, perlecan, and Hsp-47, fulfill important functions (Yurchenco et al., 2004), in the following we concentrate on discussing the laminin family.

Table I Laminin Isotypes  chain 1 2

3

4

5

and chains

Denomination

1, 1 2, 1 1, 1 2, 1 1, 3 1, 1 2, 1 3, 2 1, 1 2, 1 2, 3 B1, 1 B2, 1 B2, 3

Laminin-1 Laminin-3 Laminin-2 Laminin-4 Laminin-12 Laminin-6 Laminin-7 Laminin-5 Laminin-8 Laminin-9 Laminin-14 Laminin-10 Laminin-11 Laminin-15

47

2. Fibroblast Growth Factor Signaling

III. Laminins and Basement Membrane–Mediated Signaling A. Genetic Analysis of Laminins Gene targeting revealed the delicate specificity of laminin isotypes. The mutant defects largely follow the specific gene expression patterns of the laminin isotypes (Table II). This is apparent in  chain mutations, where disruption of Lama2 (Miyagoe et al., 1997; Timpl et al., 2000) and Lama4 (Patton et al., 2001; Thyboll et al., 2002) expressed in the neuromuscular junction cause peripheral nervous system defects, whereas the Lama5 mutation is distinguished by defects in kidney glomeruli and kidney agenesis (Miner and Li, 2000; Miner et al., 1998). Moreover, in agreement with the early embryonic expression pattern of laminin-1, mutations of laminin-1, laminin- 1 (Miner et al., 2004), laminin- 1 (Smyth et al., 1999), and the deletion of its E3 fragment containing 1LG4–5 (P. Ekblom and S. Scheele, personal communication) aVect diVerentiation of the first epithelia of the

Table II Laminin Mutations AVecting Embryogenesis or Embryoid Body DiVerentiation Gene Lamb2 Lama2 Lama3 Lamc1 Lama5

Lama4

Lamb1 Lama1

1LG4–5

Phenotype Adult lethal. Neuromuscular junction (NMJ) and glomerular defects. Disrupted BM. Lethal around 5 weeks. Muscular dystrophy, peripheral neuropathy, disrupted BM. Neonatal lethal. Epithelial adhesion defects. Defective skin and teeth. Restricted BM defects E5.5 lethal. No epiblast diVerentiation. Endoderm retained. Failure of BM assembly. Late embryonic lethality; exencephaly, syndactyly, placental labyrinth defect, glomerular defects, sporadic kidney agenesis. BM discontinuous. Transient microvascular defect; NMJ defect, small blood vessels BM defective at birth, later becoming normal. Enhancement of vascularisation, enhanced tumor growth and metastasis. E5.5 lethal. Loss of epiblast diVerentiation. Failure of BM assembly. E5.5–6.5 lethal. Loss of epiblast diVerentiation. BM present. 5 chain replaces 1. Embryoid bodies not investigated. E5.5 lethal. Endoderm and BM retained no epiblast diVerentiation. Truncated 1 chain incorporated in BM.

Reference Noakes et al., 1995 Miyagoe et al., 1997 Ryan et al., 1999 Smyth et al., 1999 Miner et al., 1998

Patton et al., 2001; Thyboll et al., 2002; Zhou et al., 2004

Miner et al., 2004 Miner et al., 2004

P. Ekblom and S. Scheele, personal communication

48

Peter Lonai

mammalian embryo. The specificity of localization and severity of these mutant phenotypes suggest that laminin isotypes fulfill specific functions in the organs and tissues of their expression. Localized expression and localized function of individual members of a gene family reflect localized transcriptional regulation, that is, the availability of specific transcription factors at specific locations. Local regulation may or may not be associated with locally specific function. In the latter case an isotype expressed at a specific localization fulfills a unique function and is not replaced by another isotype, even if their localization partially overlaps. There are only few data in the laminin field that shed light on this question. Muscular dystrophy caused by laminin-2 defects is partially rescued by laminin-1 (Gawlik et al., 2004), suggesting that the two  chains display similar functions at the neuromuscular junction. In contrast, although the 1 chain of laminin-1 and the 5 chain of laminin-10 or laminin-11 are coexpressed at the subendodermal BM of the early embryo or EB, laminin5 does not rescue the loss of epiblast diVerentiation due to the deletion of 1LG4–5 (P. Ekblom and S. Scheele, personal communication). It follows that although the function of laminin-1 and laminin-2 may be similar at the neuromuscular junction, laminin-1 and laminin-5 fulfill diVerent roles in the subendodermal BM.

B. Laminin-1 and Epiblast Differentiation Genetic analysis of laminin function by gene targeting (see Table II) and especially the targeted disruption of the constituent loci encoding of the earliest expressed laminin isoform, laminin-1, revealed unexpected insights into its role in epithelial diVerentiation. Disruption of Lamc1 encoding laminin- 1, a laminin chain present in 10 out of 15 laminin isoforms (see Table I), resulted in defective blastocyst development and early postimplantation lethality (Smyth et al., 1999). The trophectodermal and subendodermal BM of the early postimplantation Lamc / embryo was absent, although endoderm development was observed and the internal ES cells of the EB did not develop beyond the stem cell stage. This phenotype was similar to the targeted disruption of 1-integrin (Fassler and Meyer, 1995), which is thought to result from defective laminin-1 synthesis (Aumailley et al., 2000). On the other hand, a third gene, dystroglycan, a membranebound protein in communication with dystrophin, was shown to be required for the stability of the BM (Henry and Campbell, 1998; Li et al., 2002). The Lamc1 mutation could be rescued by exogenous laminin-1, suggesting that laminin-1 is required for BM assembly and that the BM and its laminin-1 component are required for the epithelialization of ES cells (Murray and Edgar, 2000). The role of the cell-binding LG4–5 domains of the laminin-1

2. Fibroblast Growth Factor Signaling

49

chain was dissected by a series of experiments, where the rescue of the Lamc1, 1-integrin, and dystroglycan mutations by exogenous laminin-1 was inhibited by the addition of 1LG4–5 containing E3 fragments (Li et al., 2002). The results demonstrated that the cell-binding domain of laminin-1 is required for epithelialization of the primitive ectoderm and suggested that 1-integrin and dystroglycan, which were thought to act as laminin receptors, do not fulfill this role during epiblast diVerentiation. Independent results obtained while investigating the role of FGF signaling in early development support and extend these conclusions. FGFR monomers truncated at the cytoplasmic domain act as a dominant negative mutation. Heterodimerization among FGFR isotypes with a truncated monomer inactivates most FGFRs synthesized by the same cell. Indeed, ES cells expressing truncated Fgfr2 cDNA (abbreviated as dnFgfr) inhibited the diVerentiation of both epithelia of the EB although, as we have shown, EBs express multiple FGFR isotypes (Chen et al., 2000). Besides defective expression of numerous endoderm- and ectoderm-specific genes, such as Hnf4, vHnf1, Evx1, Gata4, and Gata6, as well as Eomes, the kit ligand and embryonic globin, dnFgfr ES cells failed to transcribe the polypeptide chains of laminin-1 and collagen IV. Significantly, this phenotype could be partially rescued by exogenous laminin-1 (Li et al., 2001a). These results establish the role of laminin-1 in the epithelialization of the primitive ectoderm and demonstrate that EB diVerentiation is initiated by FGF signaling. An important intermediate of FGF signaling was discovered when the GATA-4 and GATA-6 transcription factors were investigated. GATA-6 and GATA-4 have been shown to be required for endoderm diVerentiation (Morrisey et al., 1998) and to transform ES cells into endoderm-like cells (Fujikura et al., 2002). We showed that endoderm-like GATA transformed ES cells, similar to exogenous laminin-1 rescue of both laminin and collagen IV expression and epiblast diVerentiation when cocultivated with the dnFgfr ES cells (Li et al., 2004). Using the Lamc1 / mutation of Smyth et al. (1997), it was demonstrated that the active principle of this physiological cell-to-cell interaction is in fact laminin-1. This result suggested that once expressed, GATA-4 and GATA-6, which are downstream of FGF signaling, are suYcient to activate endoderm diVerentiation and to initiate epiblast diVerentiation independently from FGF signaling (Li et al., 2004).

C. Endoderm and Ectoderm Differentiation Follow Different Pathways Partial rescue of the dnFgfr mutant by exogenously added laminin, or cocultivation with GATA factor-transformed endoderm-like cells, revealed that laminin-1 is a specific inducer of epiblast epithelialization. In appropriate

50

Peter Lonai

concentrations laminin-1, or spent media of GATA-6 or GATA-4-transformed ES cells, induced epiblast diVerentiation without inducing endoderm diVerentiation in ES cells, thus separating endoderm and ectoderm diVerentiation into two consecutive but distinct pathways (Li et al., 2001a, 2004). Taken together, these observations suggest that although endoderm diVerentiation is induced by FGF signaling, epiblast diVerentiation is induced by laminin-1. Endoderm and ectoderm diVerentiation were also separated by their dependence on Lif (Murray and Edgar, 2001a). Murray and Edgar showed that although Lif inhibited the complete diVerentiation of the endoderm, its precursors still produced a BM and could activate epiblast diVerentiation. This result suggested that Lif inhibits only certain aspects of endoderm diVerentiation and does not interfere with epithelialization of the primitive ectoderm, although it failed to characterize the BM-producing primitive endoderm precursor in detail. Further issues separating endoderm and ectoderm diVerentiation were discovered when the downstream elements of laminin-induced epiblast diVerentiation were studied. A dominant negative mutation of the Rho kinase, ROCK, when expressed in ES cells, abolished epiblast epithelialization without aVecting endoderm diVerentiation or the deposition of the basement membrane. Moreover, when laminin binding after the addition of laminin containing spent media from GATA-transformed cultures was assessed by an 1LG4-specific antibody, only the ectoderm series, including undiVerentiated ES cells, exhibited laminin-binding receptors (Li et al., 2004). These results indicate, in addition to separating endoderm and ectoderm diVerentiation, that the two cell lineages originate in one undiVerentiated precursor pool. GATA-4 and GATA-6 activate endoderm diVerentiation when overexpressed in ES cells (Fujikura et al., 2002); moreover, epiblast diVerentiation is induced in pluripotential stem cells (Li et al., 2004). The most likely scenario for EB diVerentiation, therefore, is that external signals activate peripheral cells of the ES cell aggregate, which then undergo endoderm diVerentiation controlled by GATA factors. Early endoderm cells produce laminin and collagen IV chains and deposit the subendodermal BM. Subsequently, the laminin component of the BM induces cytoskeletal rearrangements in the remaining undiVerentiated stem cells through ROCK kinase, which is required for the epithelialization of the primitive ectoderm (Fig. 2). This suggests that neither endoderm nor epiblast diVerentiation requires diVerentiated precursors. Therefore, the most likely explanation is that both the extraembryonic ectoderm and the epiblast directly derive from the same pluripotential stem cells. This signifies that the role of laminin-1 is to confer epithelial cell fate on pluripotential precursors. Future studies will have to answer whether this role extends to the stem cells of other epithelia. Concomitant to epiblast diVerentiation, EBs undergo cavitation, similar to the formation of the so-called proallantoic canal of the later egg-cylinder

2. Fibroblast Growth Factor Signaling

51

Figure 2 From ES cells to endoderm and epiblast: a scheme of basement membrane-mediated interactions. This scheme depicts the most important cellular interactions and some of the molecular interactions that lead from diVerentiation of the ICM into embryonic ectoderm and and through deposition of the subendodermal basement membrane to the pregastrulation epiblast.

embryo. Cavitation, as suggested by Coucouvanis and Martin (1995), is based on the apoptotic death of cells that do not participate in endoderm or ectoderm diVerentiation. These authors proposed that cavitation is due to a signal emanating from the endoderm and that the diVerentiated ectoderm is rescued by its contact with the BM. Recent results tend to modify and extend this interpretation. Columnar ectoderm diVerentiation is induced independently from endoderm diVerentiation by laminin-1 (Li et al., 2001a) and specifies the endoderm-derived signal as the laminin-1 component of the subendodermal BM. The BM’s role in rescuing the diVerentiated

52

Peter Lonai

ectoderm seems to have little relevance if we consider that the ectoderm develops from pluripotent ES cells, which can survive for weeks in vitro without BM deposition; therefore, the remaining stem cells do not need to be rescued from apoptosis. Moreover, only those stem cells that have made contact with laminin-1 of the BM diVerentiate into columnar ectoderm. It is therefore possible that cavitation is due to mechanical separation between the two diVerentiated epithelia surrounding the BM and the remaining stem cells of the EB. Alternatively and in addition, anoxic necrosis or the apoptotic eVect of BMP-4 produced by the diVerentiating ectoderm (Coucouvanis and Martin, 1999) may be responsible for the local death of stem cells. Future results are expected to decide between these alternatives.

D. From Stem Cells to Pregastrulation Embryo: A Cascade of Cellular and Molecular Interactions Activation of endoderm diVerentiation by FGF signaling, which leads to BM deposition and ROCK-mediated cytoskeletal rearrangement of the epiblast, is only part of the interactions that take place during EB diVerentiation (see Fig. 2). As mentioned before, zygotic Fgfr2 is expressed in the peripheral blastomers of the compacted morula and later in the primitive endoderm (HaVner-Krausz et al., 1999). Its ligand, Fgf4, first appears in the ICM of the blastocyst. The spatial temporal regulation of this receptor– ligand pair coincides with the appearance of the first BMs, in the basal domain of the trophectoderm and at the primitive endoderm–ICM interface. There are few data to interpret the functional aspects of this expression pattern. Chai et al. (1998) reported that dominant negative Fgfr4 cDNA interferes with preimplantation embryogenesis at the fifth cleavage division, coinciding with early stages of blastocyst development. Future research will have to clarify the interactions between Fgf4 and Fgfr2 (and other Fgfr) and the BM, as exhibited during the preimplantation stages. FGF signaling creates multiple interactions with docking proteins, leading to Ras or PI3K activation. A frequent partner of FGF signaling is FRS2 (Lax et al., 2002), which interacts with Grb2. Targeted disruption of FRS2 is lethal at the gastrulation stage (Hadari et al., 2001), whereas Grb2 embryos die at implantation and Grb2 / ES cells form neither endoderm or ectoderm (Table III) (Cheng et al., 1998). Because FRS2 is upstream of Grb2, one could have expected a more severe phenotype. Therefore, it might be interesting to explore early FGF signaling further. Evidence for the involvement of additional signaling pathways in EB diVerentiation was indicated by the loss of Akt/PKB phosphorylation in dnFgfr EBs as compared with wild-type EBs (Chen et al., 2000). A tight connection between PI3K–Akt/PKB signaling and BM assembly was indicated by a

53

2. Fibroblast Growth Factor Signaling Table III Mutations AVecting Embryoid Body DiVerentiation Gene

Mutation

Cdh1

Targeted

Itgb1

Targeted

Grb2

Targeted

Dag1

Targeted

dnFgfr

Truncated cDNA; dominant negative mutation Rho binding negative, dominant negative mutation

Rock2

Phenotype

Reference

E-cadherin is maternally expressed. It is required for compaction of the morula; in its absence ES cells do not aggregate and EB diVerentiation is defective. Loss of 1-integrin is lethal in the preimplantation embryo. Interferes with BM assembly and laminin-1 expression. No epiblast epithelialization. Adaptor protein for receptor tyrosine kinase signaling E5.5 lethal. No endoderm or epiblast diVerentiation. Dystroglycan is required for stabilization of the BM. In EBs BM assembly and epiblast diVerentiation takes place. Truncated Fgfr2 cDNA expressed in ES cells. No endoderm or epiblast diVerentiation. Defective PI3K signaling. No laminin-1 or collagen IV synthesis.

Huber et al., 1996; Larue et al., 1994

Rho kinase 2. Mediator of Rho isoforms active in cytoskeletal rearrangements. Dominant negative mutation expressed in ES cells. No epiblast diVerentiation. Endoderm and BM retained.

Aumailley et al., 2000; Fassler and Meyer, 1995 Cheng et al., 1998

Henry and Campbell, 1998

Chen et al., 2000

Li et al., 2004

robust increase in laminin and collagen IV synthesis and BM assembly of EBs that derived from ES cells expressing constitutively active p110 or Akt (Li et al., 2001b). The first cytological change of the diVerentiating EB following ES cell aggregation is the formation of endoderm cells at the periphery of the aggregate. GATA-6 and, under its control, GATA-4 (Morrisey et al., 1998), are necessary and suYcient to induce endoderm diVerentiation (Fujikura et al., 2002). Next, GATA-6 is activated by FGF signaling and then activates both the cytoskeletal rearrangement and BM protein synthesis that are required for endoderm diVerentiation (Li et al., 2004). GATA-6, among other endoderm-specific genes, activates the COUP-TFI and COUPTFII transcription factors, which in turn can activate the expression of laminin isotypes (Murray and Edgar, 2001b). Endoderm diVerentiation takes place in cells along the periphery of the ES cell aggregate, and the BM deposited along their basal domain is not

54

Peter Lonai

required for their survival (Smyth et al., 1999), since these cells do not bind laminin (Li et al., 2004). It follows that BM proteins are synthesized by the endoderm but are bound by the ectoderm and its precursors, and not by the endoderm cells of their origin. BM anchorage defines the direction of BM-mediated signaling as leading from the endoderm to the ectoderm and the definite position of the epiblast. Hence, the BM contributes to the establishment of the simple patterns of the EB and the egg-cylinder embryo. Despite their incompleteness, these data allow us to delineate the main elements of an EB diVerentiation pathway (see Fig. 2). FGF and E-cadherin signaling activates the commitment of peripheral cells in the ES cell aggregate (or in the ICM) to primitive endoderm diVerentiation. An important arm of this interaction follows the PI3K–Akt/PKB pathway and culminates in the synthesis of laminin and collagen chains and in their deposition as the subendodermal BM. Laminin-1 of the BM, through the LG4–5 globular domains of its E3 fragment (Li et al., 2002), activates epithelial transformation of ES cells that form a columnar epithelium requiring the activity of ROCK and RhoC (Li et al., 2004). DiVerentiation of the primitive ectoderm of the ICM into columnar epithelium of the epiblast makes the egg-cylinder embryo ready for gastrulation. Gastrulation by forming the mesoderm and definitive endoderm and by defining the relative extent of the neuroectoderm puts down the basis for the development of the head, trunk, limbs, and visceral organs.

E. Laminin Receptors and Anchorage Sites BMs are anchored to the cell surface through the C-terminal end of their  chains containing five globular modules (Andac et al., 1999). Thus, the polymeric laminin network represents an organized lattice of ligands, which requires a complementary array of anchorage sites (Colognato et al., 1999). Recent research revealed that the C-terminal end of laminin 1 chains (or more accurately, their LG4–5 modules), besides anchoring the laminin network to the cell membrane, induces epiblast diVerentiation (Li et al., 2001a, 2002, 2004; Murray and Edgar, 2000). The question therefore arises whether the laminin anchorage and receptor sites are separate or bifunctional. This problem is under extensive investigation, but at the time being there are only few data to be discussed. The globular domains of laminin-1 bind numerous specific cell surface molecules. LG1–3 binds integrin 6 1, 6 4, and 7 1, whereas LG4–5 (or more accurately, the LG4 domain) binds dystroglycan, sulfatides, perlecan, fibulin, and heparin (Timpl et al., 2000). In a penetrating study, Li et al. (2002) investigated whether the E3 fragment of laminin-1 containing the 1LG4–5 modules can inactivate the rescue of EB diVerentiation by

2. Fibroblast Growth Factor Signaling

55

exogenous laminin-1. In contrast to previous consensus, this research revealed that although dystroglycan and 1-integrin are required for BM maintenance and for the continuous synthesis of the laminin 1 chain, they do not fulfill the role of an E3 receptor that could be responsible for ES cell epithelialization. These results are in good agreement with structural analysis, which separated the dystroglycan-, heparin-, and sulfatide-binding sites of 1LG4 (Wizemann et al., 2003). Although externally added heparin inhibits EB diVerentiation (Li et al., 2002), it is not clear whether LG4–5 behaves as a ligand for anchorage, a ligand for receptor signaling, or both. Deletion of the 1LG4–5 domain by gene targeting contributed to the clarification of this problem. 1LG4–5 / embryos die shortly after implantation, but, in contrast to other mutants of laminin-1, laminin- 1, and laminin- 1 (Miner et al., 2004; Smyth et al., 1999), these mutants develop a subendodermal BM, which exhibits the epitopes of the 1LG1–3 domain, suggesting that laminin-1 can retain its anchorage function in the absence of 1LG4–5 (P. Ekblom and S. Scheele, personal communication). According to this result, 1LG4–5 may be recognized by the signaling receptor that initiates stem cell epithelialization, although it cannot represent the single anchorage site for laminin-1.

IV. Current Questions The significance of 1LG4–5 as an inducer of epithelialization is in providing a definite molecular mechanism for a major ECM component. This insight contributes to a better understanding of the pregastrulation stage of mammalian development. However, it also holds promise for a broader view of epithelial transition and BM-mediated crosstalk in general. In the following section some of the emerging questions and experimental hypotheses are analyzed. The question arises whether, similar to the induction of primitive ectoderm epithelialization by the 1LG4–5 domain, other LG4–5 domains of other laminin isotypes are also involved with epithelial diVerentiation. According to structural analysis, laminin-1 and laminin-2 chains both bind perlecan, heparin, -dystroglycan, and sulfatides (Talts et al., 1999) and similar characteristics also have been established for laminin-4 (Talts et al., 2000) and laminin-5 (Nielsen et al., 2000), indicating that the globular domains of diVerent  chain variants display similar structures and similar binding aYnities. As the functional representation of structural similarity, exogenous laminin-1 can rescue the reversed polarity of Madine-Darby canine kidney epithelial cells induced by dominant negative Rac1 (O’Brien et al., 2001), or human mammary epithelia cultured in collagen I gels (Gudjonsson et al., 2002). These results suggest that the polarity-inducing

56

Peter Lonai

capacity of laminin-1 is not restricted to one aspect of early embryogenesis. A step further, Streuli et al. (1995) demonstrated that the E3 fragment of laminin-1 activates -casein expression in human mammary epithelia, and Slade et al. (1999) showed that the polarity of mammary luminal epithelia can be corrected by the E8 and E3 fragments of the laminin -1 chain. In addition, Kadoya et al. (2003) reported similar activity of another isotype, the LG4 module of laminin-5, which could restore salivary gland-branching morphogenesis. We conclude, therefore, that the diVerent G-terminal globular domains of laminin  chain isotypes exhibit similar structures and binding aYnities and that they are involved in the establishment of polarity not only in the first embryonic epithelia, but also during organogenesis and epithelialization in general. The most straightforward question regarding laminin-induced epithelial polarization relates to the signaling receptor recognizing the 1LG4–5 module. Taking the binding characteristics of the LG4 module in account (Hohenester et al., 1999b; Timpl et al., 2000), it is reasonable to assume that it should be a heparin or heparan sulfate-binding molecule. Discovery of the receptor will provide the key to unravel the pathway of laminin-activated signaling. Although the BM could anchor to both cell sheets surrounding it, epithelialization of the primitive endoderm is distinguished by its polarity. Although laminin and type IV collagen are synthesized by the endoderm, the 1LG4–5 modules bind exclusively to the ectoderm and its derivatives. Thus, the flow of signals from endoderm to ectoderm is established. Although the polarity and main elements of this pathway are known, it is not clear whether FGFRs are the only receptor tyrosine kinases involved in the process. Neither is the specific role of P13K and Akt/PKB in the control of laminin and collagen IV synthesis (Li et al., 2001b) known, and we know next to nothing, besides the role of Rho kinase (Li et al., 2004), of the pathway that starts with the laminin-induced initiation of diVerentiation and with the formation of the columnar ectoderm of the epiblast. It has been shown by Fujikura et al. (2002) and Li et al. (2004) that GATA-4 and GATA-6 transform ES cells to endoderm-like cell lines. Similarly, laminin-1 can induce epithelialization of pluripotent stem cells into a uniform columnar epithelium (Li et al., 2004). It follows that the target cells of FGF and GATA factor-induced epithelialization of the endoderm, and the subsequent laminin-induced epithelialization of the columnar epithelium, derive from a common stem cell pool. Does this mean that FGF signaling and GATA transcription factors have the propensity to induce endoderm diVerentiation and BM assembly, whereas laminin-1 is a specific inducer of multipotent columnar epithelia? Future experiments with diVerent laminin isotypes in appropriate in vitro systems will have to clarify this problem.

2. Fibroblast Growth Factor Signaling

57

Assuming that the role of laminin LG4–5 modules is connected to the polarization of epithelia, we also have to consider the role of integrin-binding sites of the LG1–3 and the N-terminal LN-VI domains of the  chain. Aumailley et al. (2000) have reported that 1-integrin is required for laminin-1 expression. Another study connects ILK, the integrin-like kinase, with cell spreading and the actin cytoskeleton (Sakai et al., 2003). Additional research is required to clarify the role of laminin-binding integrins, which themselves are signaling molecules (Hynes, 2002). Finally, we have to ask whether all or most BM-associated functions are connected to laminins. The answer must be no. Through their multiple heparin-binding moieties, ECM molecules may indeed store and present haparan sulfates. HS binding is not a unique characteristics of the BM, because intercellular ECM proteins, such as glypicans and syndecans, also bind them. The high sensitivity of developmental and organogenetic processes to loss of the enzymes for HS synthesis (Perrimon and Bernfield, 2000) points to their great importance in maintaining the cell to matrix crosstalk.

References Allen, B. L., and Rapraeger, A. C. (2003). Spatial and temporal expression of heparan sulfate in mouse development regulates FGF and FGF receptor assembly. J. Cell Biol. 163, 637–648. Andac, Z., Sasaki, T., Mann, K., Brancaccio, A., Deutzmann, R., and Timpl, R. (1999). Analysis of heparin, alpha-dystroglycan and sulfatide binding to the G domain of the laminin alpha1 chain by site-directed mutagenesis. J. Mol. Biol. 287, 253–264. Arman, E., HaVner-Krausz, R., Chen, Y., Heath, J. K., and Lonai, P. (1998). Targeted disruption of FGFR2 suggests a role for FGF signaling in pre-gastrulation mammalian development. Proc. Natl. Acad. Sci. USA 95, 5082–5087. Arman, E., HaVner-Krausz, R., Gorivodsky, M., and Lonai, P. (1999). Fgfr2 is required for limb outgrowth and lung-branching morphogenesis. Proc. Natl. Acad. Sci. USA 96, 11895–11899. Aumailley, M., Pesch, M., Tunggal, L., Gaill, F., and Fassler, R. (2000). Altered synthesis of laminin 1 and absence of basement membrane component deposition in 1 integrin-deficient embryoid bodies. J. Cell. Sci. 113, 259–268. Ben-Shushan, E., Sharir, H., Pikarsky, E., and Bergman, Y. (1995). A dynamic balance between ARP-1/COUP-TFII, EAR-3/COUP-TFI, and retinoic acid receptor: Retinoid X receptor heterodimers regulates Oct-3/4 expression in embryonal carcinoma cells. Mol. Cell. Biol. 15, 1034–1048. Bernfield, M., Gotte, M., Park, P. W., Reizes, O., Fitzgerald, M. L., Lincecum, J., and Zako, M. (1999). Functions of cell surface heparan sulfate proteoglycans. Annu. Rev. Biochem. 68, 729–777. Bullock, S. L., Fletcher, J. M., Beddington, R. S., and Wilson, V. A. (1998). Renal agenesis in mice homozygous for a gene trap mutation in the gene encoding heparan sulfate 2-sulfotransferase. Genes Dev. 12, 1894–1906. Chai, N., Patel, Y., Jacobson, K., McMahon, J., McMahon, A., and Rappolee, D. A. (1998). FGF is an essential regulator of the fifth cell division in preimplantation mouse embryos. Dev. Biol. 198, 105–115.

58

Peter Lonai

Chambers, I., Colby, D., Robertson, M., Nichols, J., Lee, S., Tweedie, S., and Smith, A. (2003). Functional expression cloning of nanog, a pluripotency sustaining factor in embryonic stem cells. Cell 113, 643–655. Chen, Y., Li, X., Eswarakumar, V. P., Seger, R., and Lonai, P. (2000). Fibroblast growth factor (FGF) signaling through PI 3-kinase and Akt/PKB is required for embryoid body diVerentiation. Oncogene 19, 3750–3756. Cheng, A. M., Saxton, T. M., Sakai, R., Kulkarni, S., Mbamalu, G., Vogel, W., Tortorice, C. G., CardiV, R. D., Cross, J. C., Muller, W. J., et al. (1998). Mammalian Grb2 regulates multiple steps in embryonic development and malignant transformation. Cell 95, 793–803. Cheng, Y. S., Champliaud, M. F., Burgeson, R. E., Marinkovich, M. P., and Yurchenco, P. D. (1997). Self-assembly of laminin isoforms. J. Biol. Chem. 272, 31525–31532. Ciruna, B., and Rossant, J. (2001). FGF signaling regulates mesoderm cell fate specification and morphogenetic movement at the primitive streak. Dev. Cell 1, 37–49. Colognato, H., Winkelmann, D. A., and Yurchenco, P. D. (1999). Laminin polymerization induces a receptor-cytoskeleton network. J. Cell. Biol. 145, 619–631. Colognato, H., and Yurchenco, P. D. (2000). Form and function: The laminin family of heterotrimers. Dev. Dyn. 218, 213–234. Colvin, J. S., Bohne, B. A., Harding, G. W., McEwen, D. G., and Ornitz, D. M. (1996). Skeletal overgrowth and deafness in mice lacking fibroblast growth factor receptor 3. Nat. Genet. 12, 390–397. Coucouvanis, A., and Martin, G. R. (1995). Signals for death and survival: A two step mechanism for cavitation in the vertebrate embryo. Cell 83, 279–287. Coucouvanis, E., and Martin, G. R. (1999). BMP signaling plays a role in visceral endoderm diVerentiation and cavitation in the early mouse embryo. Development 126, 535–546. De Moerlooze, L., Spencer-Dene, B., Revest, J., Hajihosseini, M., Rosewell, I., and Dickson, C. (2000). An important role for the IIIb isoform of fibroblast growth factor receptor 2 (FGFR2) in mesenchymal-epithelial signalling during mouse organogenesis. Development 127, 483–492. Deng, C., Wynshaw-Boris, A., Zhou, F., Kuo, A., and Leder, P. (1996). Fibroblast growth factor receptor-3 is a negative regulator of bone growth. Cell 84, 911–921. Deng, C.-X., Wynshaw-Boris, A., Shen, M. M., Daugherty, C., Ornitz, D. M., and Leder, P. (1994). Murine FGFR-1 is required for early postimplantation growth and axial organization. Genes Dev. 8, 3045–3057. Ekblom, P., Lonai, P., and Talts, J. F. (2003). Expression and biological role of laminin-1. Matrix Biol. 22, 35–47. Eswarakumar, V. P., Mark, C. Horowitz, Rachel, Locklin, Morriss-Kay, G. M., and Lonai, P. (2004). A gain of function mutation of Fgfr2c demonstrates the role of this receptor variant in osteogenesis. Proc. Natl. Acad. Sci. USA 101, 12555–12560. Eswarakumar, V. P., Monsonego-Ornan, E. M., Pines, M., Antonopoulou, I., Morriss-Kay, G. M., and Lonai, P. (2002). The IIIc alternative of Fgfr2 is a positive regulator of bone formation. Development 129, 3783–3793. Fassler, R., and Meyer, M. (1995). Consequences of lack of 1 integrin expression in mice. Genes Dev. 9, 1896–1908. Fata, J. E., Werb, Z., and Bissell, M. J. (2004). Regulation of mammary gland branching morphogenesis by the extracellular matrix and its remodeling enzymes. Breast Cancer Res. 6, 1–11. Feldman, B., Poueymirou, W., Papaioannou, V. E., DeChiare, T. M., and Goldfarb, M. (1995). Requirement of FGF4 for postimplantation mouse development. Science 267, 246–249. Folkman, J. (2002). Role of angiogenesis in tumor growth and metastasis. Semin. Oncol. 29, 15–18.

2. Fibroblast Growth Factor Signaling

59

Fujikura, J., Yamato, E., Yonemura, S., Hosoda, K., Masui, S., Nakao, K., Miyazaki Ji, J., and Niwa, H. (2002). DiVerentiation of embryonic stem cells is induced by GATA factors. Genes Dev. 16, 784–789. Gardner, R. L. (1982). Investigations of cell lineage and diVerentiation in the extraembryonic ectoderm of the mouse embryo. J. Embryol. Exp. Morphol. 68, 175–198. Gawlik, K., Miyagoe-Suzuki, Y., Ekblom, P., Takeda, S., and Durbeej, M. (2004). Laminin 1 chain reduces muscular dystrophy in laminin 2 chain deficient mice. Hum. Mol. Genet. 13, 1775–1784. Givol, D., Eswarakumar, V. P., and Lonai, P. (2003). ‘‘Molecular and Cellular Biology of FGF Signaling.’’ Oxford University Press, New York. Goldin, S. N., and Papaioannou, V. E. (2003). Paracrine action of FGF4 during periimplantation development maintains trophectoderm and primitive endoderm. Genesis 36, 40–47. Graham, C. F. (1978). Features of cell lineage in preimplantation mouse development. J. Embryol. Exp. Morphol. 48, 53–72. Gudjonsson, T., Ronnov-Jessen, L., Villadsen, R., Rank, F., Bissell, M. J., and Petersen, O. W. (2002). Normal and tumor-derived myoepithelial cells diVer in their ability to interact with luminal breast epithelial cells for polarity and basement membrane deposition. J. Cell. Sci. 115, 39–50. Guimond, S. E., and Turnbull, J. E. (2000). Fibroblast growth factor receptor signaling is dictated by specific heparan sulphate saccharides. Curr. Biol. 9, 1343–1346. Hadari, Y. R., Gotoh, N., Kouhara, H., Lax, I., and Schlessinger, J. (2001). Critical role for the docking-protein FRS2 alpha in FGF receptor-mediated signal transduction pathways. Proc. Natl. Acad. Sci. USA 98, 8578–8583. HaVner-Krausz, R., Gorivodsky, M., Chen, Y., and Lonai, P. (1999). Expression of FGFR2 during oogenesis, preimplantation and early postimplantation embryogenesis. Mech. Dev. 85, 167–172. Hajihosseini, M. K., Wilson, S., De Moerlooze, L., and Dickson, C. (2001). A splicing switch and gain-of-function mutation in FgfR2-IIIc hemizygotes causes Apert/PfeiVersyndrome-like phenotypes. Proc. Natl. Acad. Sci. USA 98, 3855–3860. Hebert, J. M., Rosenquist, T., Gotz, J., and Martin, G. R. (1994). FGF5 as a regulator of the hair growth cycle: Evidence from targeted and spontaneous mutations. Cell 78, 1017–1025. Henry, M. D., and Campbell, K. P. (1998). A role for dystroglycan in basement membrane assembly. Cell 95, 859–870. Hogan, B., Beddington, R., Costantini, F., and Lacy, E. (1994). ‘‘Manipulating the Mouse Embryo: A Laboratory Manual.’’ Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. Hogan, B. L., and Kolodziej, P. A. (2002). Organogenesis: Molecular mechanisms of tubulogenesis. Nat. Rev. Genet. 3, 513–523. Hogan, B. L., and Yingling, J. M. (1998). Epithelial/mesenchymal interactions and branching morphogenesis of the lung. Curr. Opin. Genet. Dev. 8, 481–486. Hohenester, E., Tisi, D., Talts, J. F., and Timpl, R. (1999a). The crystal structure of a laminin G-like module reveals the molecular basis of -dystroglycan binding to laminins, perlecan and agrin. Mol. Cell 4, 783–792. Hohenester, E., Tisi, D., Talts, J. F., and Timpl, R. (1999b). The crystal structure of a laminin G-like module reveals the molecular basis of alpha-dystroglycan binding to laminins, perlecan, and agrin. Mol. Cell 4, 783–792. Huber, O., Bierkamp, C., and Kemler, R. (1996). Cadherins and catenins in development. Curr. Opin. Cell Biol. 8, 685–691. Hynes, R. O. (2002). Integrins: Bidirectional, allosteric signaling machines. Cell 110, 673–687.

60

Peter Lonai

Ibrahimi, O. A., Eliseenkova, A. V., Plotnikov, A. N., Yu, K., Ornitz, D. M., and Mohammadi, M. (2001). Structural basis for fibroblast growth factor receptor 2 activation in Apert syndrome. Proc. Natl. Acad. Sci. USA 98, 7182–7187. Johnson, D. E., and Williams, L. T. (1993). Structural and functional diversity in the FGF receptor multigene family. Adv. Cancer Res. 60, 1–41. Johnson, M. H., Maro, B., and Takeichi, M. (1986). The role of cell adhesion in the synchronization and orientation of polarization in 8-cell mouse blastomers. J. Embryol. Exp. Morph. 93, 239–255. Kadoya, Y., Kadoya, K., Durbeej, M., Holmvall, K., Sorokin, L., and Ekblom, P. (1995). Antibodies against domain E3 of laminin-1 and integrin 6 subunit perturb branching epithelial morphogenesis of the embryonic submandibular gland, but by diVerent modes. J. Cell Biol. 129, 521–534. Kadoya, Y., Mochizuki, M., Nomizu, M., Sorokin, L., and Yamashina, S. (2003). Role for laminin-alpha5 chain LG4 module in epithelial branching morphogenesis. Dev. Biol. 263, 153–164. Kallman, F., and Grobstein, C. (1965). Source of collagen at epitheliomesenchymal interfaces during inductive interactions. Dev. Biol. 11, 169–183. Kalluri, R. (2003). Basement membranes: Structure, assembly and role in tumour angiogenesis. Nat. Rev. Cancer 3, 422–433. Keshet, E., Lyman, S. D., Williams, D. E., Anderson, D. M., Jenkins, N. A., Copeland, N. G., and Parada, L. F. (1991). Embryonic RNA expression patterns of the c-kit receptor and its cognate ligand suggest multiple functional roles in mouse development. EMBO J. 10, 2425–2435. Klein, G., Langegger, M., Timpl, R., and Ekblom, P. (1988). Role of laminin A chain in the development of epithelial cell polarity. Cell 55, 331–341. Klein, G., Zhang, H.-Y., Ekblom, M., Klein, G., Mugrauer, G., Fecker, L., Deutzmann, R., Timpl, R., and Ekblom, P. (1990). Transient and locally restricted expression of laminin A chain mRNA by developing epithelial cells during kidney organogenesis. Cell 60, 337–346. Kurima, K., Warman, M. L., Krishnan, S., Domowicz, M., Krueger, R. C., Jr., Deyrup, A., and Schwartz, N. B. (1998). A member of a family of sulfate-activating enzymes causes murine brachymorphism. Proc. Natl. Acad. Sci. USA 95, 8681–8685. Larue, L., Ohsugi, M., Hirchenhain, J., and Kemler, R. (1994). E-cadherin null mutant embryos fail to form a trophectoderm epithelium. Proc. Natl. Acad. Sci. USA 91, 8263–8267. Lax, I., Wong, A., Lamothe, B., Lee, A., Frost, A., Hawes, J., and Schlessinger, J. (2002). The docking protein FRS2alpha controls a MAP kinase-mediated negative feedback mechanism for signaling by FGF receptors. Mol. Cell 10, 709–719. Li, L., Arman, E., Ekblom, P., Edgar, D., Murray, P., and Lonai, P. (2004). Distinct GATA6and laminin-dependent mehanisms regulate endodermal and ectodermal embryonic stem cell fates. Development 131, 5277–5286. Li, S., Harrison, D., Carbonetto, S., Fassler, R., Smyth, N., Edgar, D., and Yurchenco, P. D. (2002). Matrix assembly, regulation, and survival functions of laminin and its receptors in embryonic stem cell diVerentiation. J. Cell Biol. 157, 1279–1290. Li, X., Chen, Y., Scheele, S., Arman, E., HaVner-Krausz, R., Ekblom, P., and Lonai, P. (2001a). FGF signaling and basement membrane assembly are connected during epithelial morphogenesis of the embryoid body. J. Cell Biol. 153, 811–822. Li, X., Talts, U., Talts, J. F., Arman, E., Ekblom, P., and Lonai, P. (2001b). Akt/PKB regulates laminin and collagen IV isotypes of the basement membrane. Proc. Natl. Acad. Sci. USA 98, 14416–14421. Lin, X., BuV, E. M., Perrimon, N., and Michelson, A. M. (1999). Heparan sulfate proteoglycans are essential for FGF receptor signaling during Drosophila embryonic development. Development 126, 3715–3723.

2. Fibroblast Growth Factor Signaling

61

Lin, X., Wei, G., Shi, Z., Dryer, L., Esko, J. D., Wells, D. E., and Matzuk, M. M. (2000). Disruption of gastrulation and heparan sulfate biosynthesis in EXT1-deficient mice. Dev. Biol. 224, 299–311. Lonai, P. (2003). Epithelial mesenchymal interactions, the ECM and limb development. J. Anat. 202, 43–50. Lubarsky, B., and Krasnow, M. A. (2003). Tube morphogenesis: Making and shaping biological tubes. Cell 112, 19–28. Mansour, S. L., Goddard, J. M., and Capecchi, M. R. (1993). Mice homozygous for a targeted disruption of the proto-oncogene int-2 have developmental defects in the tail and inner ear. Development 117, 13–28. Martin, G. R., Wiley, L. M., and Damjanov, I. (1977). The development of cystic embryoid bodies in vitro from clonal teratocarcinoma cells. Dev. Biol. 61, 230–244. Miner, J. H., Cunningham, J., and Sanes, J. R. (1998). Roles of laminin in embryogenesis: Exencephaly, syndactyly, and placenthopathy in mice lacking the laminin 5 chain. J. Cell Biol. 143, 1713–1723. Miner, J. H., and Li, C. (2000). Defective glomerulogenesis in the absence of laminin alpha5 demonstrates a developmental role for the kidney glomerular basement membrane. Dev. Biol. 217, 278–289. Miner, J. H., Li, C., Mudd, J. L., Go, G., and Sutherland, A. E. (2004). Compositional and structural requirements for laminin and basement membranes during mouse embryo implantation and gastrulation. Development 131, 2247–2256. Miner, J. H., Patton, B. L., Lentz, S. I., Gilbert, D. J., Snider, W. D., Jenkins, N. A., Copeland, N. G., and Sanes, J. R. (1997). The laminin alpha chains: Expression, developmental transitions, and chromosomal locations of alpha1–5, identification of heterotrimeric laminins 8–11, and cloning of a novel alpha3 isoform. J. Cell Biol. 137, 685–701. Miyagoe, Y., Hanaoka, K., Nonaka, I., Hayasaka, M., Nabeshima, Y., Arahata, K., and Takeda, S. (1997). Laminin alpha2 chain-null mutant mice by targeted disruption of the Lama2 gene: A new model of merosin (laminin 2)-deficient congenital muscular dystrophy. FEBS Lett. 415, 33–39. Morrisey, E. E., Tang, Z., Sigrist, K., Lu, M. M., Jiang, F., Ip, H. S., and Parmacek, M. S. (1998). GATA6 regulates HNF4 and is required for diVerentiation of visceral endoderm in the mouse embryo. Genes Dev. 12, 3579–3590. Murray, P., and Edgar, D. (2000). Regulation of programmed cell death by basement membranes in embryonic development. J. Cell Biol. 150, 1215–1221. Murray, P., and Edgar, D. (2001a). The regulation of embryonic stem cell diVerentiation by leukaemia inhibitory factor (LIF). DiVerentiation 68, 227–234. Murray, P., and Edgar, D. (2001b). Regulation of laminin and COUP-TF expression in extraembryonic endodermal cells. Mech. Dev. 101, 213–215. Nielsen, P. K., Gho, Y. S., HoVman, M. P., Watanabe, H., Makino, M., Nomizu, M., and Yamada, Y. (2000). Identification of a major heparin and cell binding site in the LG4 module of the laminin alpha 5 chain. J. Biol. Chem. 275, 14517–14523. Niswander, L., and Martin, G. R. (1992). Fgf-4 expression during gastrulation, myogenesis, limb and tooth development in the mouse. Development 114, 755–768. Niwa, H., Miyazaki, J., and Smith, A. G. (2000). Quantitative expression of Oct-3/4 defines diVerentiation, dediVerentiation or self-renewal of ES cells. Nat. Genet. 24, 372–376. Noakes, P. G., Gautam, M., Mudd, J., Sanes, J. R., and Merlie, J. P. (1995). Aberrant diVerentiation of neuromuscular junctions in mice lacking s-laminin/laminin beta 2. Nature 374, 258–262. O’Brien, L. E., Jou, T. S., Pollack, A. L., Zhang, Q., Hansen, S. H., Yurchenco, P., and Mostov, K. E. (2001). Rac1 orientates epithelial apical polarity through eVects on basolateral laminin assembly. Nat. Cell Biol. 3, 831–838.

62

Peter Lonai

Ohsugi, M., Hwang, S. Y., Butz, S., Knowles, B. B., Solter, D., and Kemler, R. (1996). Expression and cell membrane localization of catenins during mouse preimplantation development. Dev. Dyn. 206, 391–402. Ohuchi, H., Nakagawa, T., Yamamoto, A., Araga, A., Ohata, T., Ishimaru, Y., Yoshioka, H., Kuwana, T., Nohno, T., Yamasaki, M., et al. (1997). The mesenchymal factor, FGF10, initiates and maintains the outgrowth of the chick limb bud through interaction with FGF8, an apical ectodermal factor. Development 124, 2235–2244. Ornitz, D. M., Xu, J., Colvin, J. S., McEwen, D. G., MacArthur, C. A., Coulier, F., Gao, G., and Goldfarb, M. (1996). Receptor specificity of the fibroblast growth factor family. J. Biol. Chem. 271, 15292–15297. Orr-Urtreger, A., Bedford, M. T., Burakova, T., Arman, E., Zimmer, Y., Yayon, A., Givol, D., and Lonai, P. (1993). Developmental localization of the splicing alternatives of fibroblast growth factor receptor-2 (FGFR2). Dev. Biol. 158, 475–486. Orr-Urtreger, A., Givol, D., Yayon, A., Yarden, Y., and Lonai, P. (1991). Developmental expression of two murine fibroblast growth factor receptors, flg and bek. Development 113, 1419–1434. Orr-Urtreger, A., and Lonai, P. (1992). Platelet-derived growth factor-A and its receptor are expressed in separate, but adjacent cell layers of the mouse embryo. Development 115, 1045–1058. Ostrovsky, O., Berman, B., Gallagher, J., Mulloy, B., Fernig, D. G., Delehedde, M., and Ron, D. (2002). DiVerential eVects of heparin saccharides on the formation of specific fibroblast growth factor (FGF) and FGF receptor complexes. J. Biol. Chem. 277, 2444–2453. Patarroyo, M., Tryggvason, K., and Virtanen, I. (2002). Laminin isoforms in tumor invasion, angiogenesis and metastasis. Semin. Cancer Biol. 12, 197–207. Patton, B. L., Cunningham, J. M., Thyboll, J., Kortesmaa, J., Westerblad, H., Edstrom, L., Tryggvason, K., and Sanes, J. R. (2001). Properly formed but improperly localized synaptic specializations in the absence of laminin alpha4. Nat. Neurosci. 4, 597–604. Patton, B. L., Miner, J. H., Chiu, A. Y., and Sanes, J. R. (1997). Distribution and function of laminins in the neuromuscular system of developing, adult, and mutant mice. J. Cell Biol. 139, 1507–1521. Pellegrini, L., Burke, D. F., von Delft, F., Mulloy, B., and Blundell, T. L. (2000). Crystal structure of fibroblast growth factor receptor ectodomain bound to ligand and heparin. Nature 407, 1029–1034. Perrimon, N., and Bernfield, M. (2000). Specificities of heparan sulfate proteoglycans in developmental processes. Nature 404, 725–728. Peters, K. G., Werner, S., Chen, G., and Williams, L. T. (1992). Two Fgfr receptors genes are diVerentially expressed in epithelial and mesenchymal tissues during limb formation and organogenesis in the mouse. Development 114, 233–243. Plotnikov, A. N., Schlessinger, Y., Hubbard, S. R., and Mohammadi, M. (1999). Structural basis for FGF receptor dimerization and activation. Cell 98, 641–650. Poschl, E., Schlotzer-Schrehardt, U., Brachvogel, B., Saito, K., Ninomiya, Y., and Mayer, U. (2004). Collagen IV is essential for basement membrane stability but dispensable for initiation of its assembly during early development. Development 131, 1619–1628. Ryan, M. C., Lee, K., Miyashita, Y., and Carter, W. G. (1999). Targeted disruption of the LAMA3 gene in mice reveals abnormalities in survival and late stage diVerentiation of epithelial cells. J. Cell Biol. 145, 1309–1323. Sakai, T., Li, S., Docheva, D., GrashoV, C., Sakai, K., Kostka, G., Braun, A., Pfeifer, A., Yurchenco, P. D., and Fassler, R. (2003). Integrin-linked kinase (ILK) is required for polarizing the epiblast, cell adhesion, and controlling actin accumulation. Genes Dev. 17, 926–940. Saxen, L., and ThesleV, I. (1992). Epithelial-mesenchymal interactions in murine organogenesis. Ciba Found. Symp. 165, 183–193; discussion 193–198.

2. Fibroblast Growth Factor Signaling

63

Schlessinger, J., Lax, I., and Lemmon, M. (1995). Regulation of growth factor activation by proteoglycans: What is the role of the low aYnity receptor. Cell 83, 357–360. Schlessinger, J., Plotnikov, A. N., Ibrahimi, O. A., Eliseenkova, A. V., Yeh, B. K., Yayon, A., Linhardt, R. J., and Mohammadi, M. (2000). Crystal structure of a ternary FGF-FGFRheparin complex reveals a dual role for heparin in FGFR binding and dimerization. Mol. Cell 6, 743–750. Slade, M. J., Coope, R. C., Gomm, J. J., and Coombes, R. C. (1999). The human mammary gland basement membrane is integral to the polarity of luminal epithelial cells. Exp. Cell Res. 247, 267–278. Smyth, N., Vatansever, H. S., Murray, P., Meyer, M., Frie, C., Paulsson, M., and Edgar, D. (1999). Absence of basement membranes after targeting the LAMC1 gene results in embryonic lethality due to failure of endoderm diVerentiation. J. Cell Biol. 144, 151–160. Sorokin, L., Conzelman, S., Ekblom, P., Battaglia, C., Aumailley, M., and Timpl, R. (1992). Monoclonal antibodies against laminin A chain fragment E3 and their eVects on binding to cells and proteoglycan and on kidney development. Exp. Cell Res. 201, 137–144. Sottile, J. (2004). Regulation of angiogenesis by extracellular matrix. Biochim. Biophys. Acta 1654, 13–22. Streuli, C. H., Schmidhauser, C., Bailey, N., Yurchenco, P., Skubitz, A. P., Roskelley, C., and Bissell, M. J. (1995). Laminin mediates tissue-specific gene expression in mammary epithelia. J. Cell Biol. 129, 591–603. Talts, J. F., Andac, Z., Gohring, W., Brancaccio, A., and Timpl, R. (1999). Binding of the G domains of laminin alpha1 and alpha2 chains and perlecan to heparin, sulfatides, alphadystroglycan and several extracellular matrix proteins. EMBO J. 18, 863–870. Talts, J. F., Mann, K., Yamada, Y., and Timpl, R. (1998). Structural analysis and proteolytic processing of recombinant G domain of mouse laminin alpha2 chain. FEBS Lett. 426, 71–76. Talts, J. F., Sasaki, T., Miosge, N., Gohring, W., Mann, K., Mayne, R., and Timpl, R. (2000). Structural and functional analysis of the recombinant G domain of the laminin alpha4 chain and its proteolytic processing in tissues. J. Biol. Chem. 275, 35192–35199. Tam, P. P. L., and Behringer, R. R. (1997). Mouse gastrulation: The formation of a mammalian body plan. Mech. Dev. 68, 3–25. Tanaka, S., Kunath, T., Hadjantonakis, A. K., Nagy, A., and Rossant, J. (1998). Promotion of trophoblast stem cell proliferation by FGF4. Science 282, 2072–2075. Tanaka, T. S., Kunath, T., Kimber, W. L., Jaradat, S. A., Stagg, C. A., Usuda, M., Yokota, T., Niwa, H., Rossant, J., and Ko, M. S. (2002). Gene expression profiling of embryo-derived stem cells reveals candidate genes associated with pluripotency and lineage specificity. Genome Res. 12, 1921–1928. Thyboll, J., Kortesmaa, J., Cao, R., Soininen, R., Wang, L., Iivanainen, A., Sorokin, L., Risling, M., Cao, Y., and Tryggvason, K. (2002). Deletion of the laminin alpha4 chain leads to impaired microvessel maturation. Mol. Cell. Biol. 22, 1194–1202. Timpl, R., and Brown, J. C. (1996). Supramolecular assembly of basement membranes. BioEssays 18, 123–132. Timpl, R., Tisi, D., Talts, J. F., Andac, Z., Sasaki, T., and Hohenester, E. (2000). Structure and function of laminin LG modules. Matrix Biol. 19, 309–317. Turnbull, J., Powell, A., and Guimond, S. (2001). Heparan sulfate: Decoding a dynamic multifunctional cell regulator. Trends Cell Biol. 11, 75–82. Virtanen, I., Gullberg, D., Rissanen, J., Kivilaakso, E., Kiviluoto, T., Laitinen, L. A., Lehto, V. P., and Ekblom, P. (2000). Laminin alpha1-chain shows a restricted distribution in epithelial basement membranes of fetal and adult human tissues. Exp. Cell Res. 257, 298–309. Weinstein, M., Xu, X., Ohyama, K., and Deng, C.-X. (1998). FGFR3 and FGFR4 function cooperatively to direct alveogenesis in the murine lung. Development 125, 3615–3623.

64

Peter Lonai

Wessels, N. K. (1970). Mammalian lung development: Interactions in formation and morphogenesis of tracheal buds. J. Exp. Zool. 175, 455–466. Wilder, P. J., Kelly, D., Brigman, K., Peterson, C. L., Nowling, T., Gao, Q. S., McComb, R. D., Capecchi, M. R., and Rizzino, A. (1997). Inactivation of the FGF-4 gene in embryonic stem cells alters the growth and/or the survival of their early diVerentiated progeny. Dev. Biol. 192, 614–629. Wilkie, A. O. (1997). Craniosynostosis: Genes and mechanisms. Hum. Mol. Gen. 6, 1647–1656. Wizemann, H., Garbe, J. H., Friedrich, M. V., Timpl, R., Sasaki, T., and Hohenester, E. (2003). Distinct requirements for heparin and alpha-dystroglycan binding revealed by structure-based mutagenesis of the laminin alpha2 LG4-LG5 domain pair. J. Mol. Biol. 332, 635–642. Xu, X., Weinstein, M., Li, C., Naski, M., Cohen, R. I., Ornitz, D. M., Leder, P., and Deng, C. (1998). Fibroblast growth factor receptor 2 (FGFR2)-mediated reciprocal regulation loop between FGF8 and FGF10 is essential for limb induction. Development 125, 753–765. Yamaguchi, T. P., Harpal, K., Henkemeyer, M., and Rossant, J. (1994). Fgr-1 is required for embryonic growth and mesodermal patterning during mouse gastrulation. Genes Dev. 8, 3032–3044. Yayon, A., Klagsbrun, M., Esko, J. D., Leder, P., and Ornitz, D. M. (1991). Cell surface, heparin-like molecules are required for binding of basic fibroblast growth factor to its high aYnity receptor. Cell 64, 841–848. Yurchenco, P. D., Amenta, P. S., and Patton, B. L. (2004). Basement membrane assembly, stability and activities observed through a developmental lens. Matrix Biol. 22, 521–538. Yurchenco, P. D., Cheng, Y. S., and Colognato, H. (1992). Laminin forms an independent network in basement membranes. J. Cell Biol. 117, 1119–1133. Zhou, Z., Doi, M., Wang, J., Cao, R., Liu, B., Chan, K. M., Kortesmaa, J., Sorokin, L., Cao, Y., and Tryggvason, K. (2004). Deletion of laminin-8 results in increased tumor neovascularization and metastasis in mice. Cancer Res. 64, 4059–4063.

3

____________________________________________________________________________

TGF- Superfamily and Mouse Craniofacial Development: Interplay of Morphogenetic Proteins and Receptor Signaling Controls Normal Formation of the Face Marek Dudas and Vesa Kaartinen Developmental Biology Program at the Saban Research Institute of Children’s Hospital Los Angeles, Los Angeles, California 90027 and Department of Pathology, Keck School of Medicine University of Southern California Los Angeles, California 90089

I. Introduction II. TGF- Superfamily Signaling A. Bone Morphogenetic Proteins and Related Growth Factors B. Structure of TGF- Family Ligands C. Ligand Antagonists and Ligand Heterodimers Increase the Signaling Complexity D. Receptors Do Not Make Our Understanding of the TGF- System Easier E. Signaling Convergence by Type I Receptors and Smad Proteins F. Unconventional Receptors and Alternative TGF- Signaling Pathways III. Craniofacial Phenotypes in Mutants of TGF- Superfamily Ligands and Receptors A. BMP and GDF Signaling B. TGF- Signaling C. Activin and Inhibin Signaling IV. Head Organizers and Early Anterior Development A. Anterior Visceral Endoderm Acts Synergistically with Derivatives of the Gastrula Organizer B. The Future Head Location: Prechordal Plate Mesenchyme with BMP Downregulation/Nodal Upregulation C. Paraxial Mesoderm, Neural Crest Segregation, and Possible Involvement of BMP-4 Signaling V. Neural Crest in Early Craniofacial Development A. Neural Crest Cells Migrate to Multiple Sites of the Developing Embryo B. Cranial Neural Crest Is the Major Player in Head Development C. BMP Signaling and the Induction of Neural Crest D. BMP Signaling and Delamination of Neural Crest Cells VI. Facial Prominences and Formation of the Face A. Facial Development Is Based on Fusion of Several Regions of Tissue VII. Mandibular Development A. Identity of the First Branchial Arch B. TGF- /Smad Signaling Regulates Growth of Meckel’s Cartilage Current Topics in Developmental Biology, Vol. 66 Copyright 2005, Elsevier Inc. All rights reserved. 0070-2153/05 $35.00

65

66

Dudas and Kaartinen C. BMP Signaling Is Critical in the Rostral Part of Developing Lower Jaw D. Intramembranous Ossification of the Mandible

VIII. Palatal Development and Cleft Palate A. Palatogenesis in Mice as a Model for Human Development and Disease B. Apoptosis C. Alternative Fates of the MEE: Migration, Epithelial to Mesenchymal TransdiVerentiation, or Both? D. The Role of TGF- Superfamily Signaling in Palatogenesis E. Epithelial–Mesenchymal Interactions, Interactive Signaling Pathways, and Morphogenesis of the Prefusion Palatal Shelves IX. Clinical Research and Applications A. Craniofacial Fracture Healing B. Prevention of Heterotopic Bone Formation C. Teeth and Periodontal Regeneration X. Conclusions Acknowledgments References

I. Introduction The combination of brain, sensory organs, craniofacial skeleton, and cephalic musculature within the head makes it a uniquely complex structure. The sensory organs of the head are far more intricate than in the rest of the body, and originate from neurogenic placodes, structures found only in the embryonic head region. The head muscles, with the exception of the tongue musculature, are formed from unsegmented paraxial and prechordal mesoderm, in striking contrast to somatic muscles, which are derived from epithelial somites of a segmental nature. Interspecies comparisons show that facial bones are the most variable parts of the skeleton, contributing to formation of such a complex phenotypic feature as facial expression, one of the strongest visual stimuli used for the recognition of individuals, especially among primates (Kendrick et al., 2001). Skeletal components such as teeth, membrane bone, and secondary cartilage are tissue types located exclusively in the head (clavicles being the only exception). One of the greatest discoveries in developmental biology was the finding that the craniofacial skeleton is intimately connected with neural tissue—the vast majority of craniofacial bones and cartilages are, in fact, derivatives of neural crest cells (Northcutt and Gans, 1983). Why ectodermal cells that usually give rise to peripheral nervous system, neural ganglia, neuroendocrine cells, and melanocytes can also form structures that are derived strictly from the mesoderm elsewhere in the body skeleton is a challenging developmental and evolutionary question. The fact that the craniofacial skeleton is a mixture of bones and cartilage originating from the cranial neural crest and those originating from the

3. TGF- Signaling and Facial Development

67

mesoderm brings new insights to craniofacial development, since it is now clear that molecular processes involved in chondrogenesis, osteogenesis, and fracture healing are remarkably diVerent between these two groups of skeletal compounds (Helms and Schneider, 2003). Furthermore, birth defects with a craniofacial component belong to the most frequent malformations in humans, thus representing a considerable health, psychological, and economic burden to aVected families, as well as to society. The treatment of these disorders is often impossible, or represents a painful, stressful, and lengthy multistep procedure. Thus, an understanding of the biological processes underlying craniofacial development and physiology can bring a substantial contribution to current medical knowledge. Recent advances in genetics and molecular biology (e.g., vertebrate genome projects, new bioinformatic tools, microarray assays, gene knockout technology, tissue-specific gene targeting, in vivo imaging and microimaging) have produced an exponential growth of knowledge of embryonic development, and a substantial amount of data on head morphogenesis has accumulated over the past decade. Unlike ever before, the nature of these new techniques has opened windows into cellular and molecular mechanisms underlying the body plan creation, resulting in frequent redefinitions and updates in embryology, as well as in classification of developmental diseases. Inherently, this information boom results in increased branching and complexity in the scientific literature, raising the need to consolidate new data into logical and vital blocks of knowledge. Bone morphogenetic proteins and related peptides from the transforming growth factor beta (TGF- ) superfamily represent a distinct group of growth factors involved in head embryogenesis. They play roles in processes essential for craniofacial development: neural crest formation and migration, and cartilage and bone physiology. Their involvement in embryonic angiogenesis further underlines their importance in normal morphogenesis and related developmental diseases. This chapter is a compilation of the current knowledge of the role of members of the TGF- superfamily and their signaling pathways in facial development.

II. TGF- Superfamily Signaling A. Bone Morphogenetic Proteins and Related Growth Factors The discovery of bone morphogenetic proteins (BMPs) was instigated by the observation that ectopic bone was formed in fascia that had been used during surgery to bridge large gaps in the urinary bladder (Neuhof, 1917). This was followed by the discovery that, in addition to urinary epithelium, demineralized bone also possesses osteogenic capability when transplanted

Table I TGF- Superfamily Ligands, Their Corresponding Type I Receptors with Downstream Smads, and Type II Receptors*

3. TGF- Signaling and Facial Development

69

into connective tissues (Senn, 1989; Urist, 1965; Van de Putte and Urist, 1965). Osteogenic tissues were purported to produce osteogenic factors, which were later isolated and characterized (Reddi, 1997). Independently from this, a growth factor capable of supporting anchorage-independent growth of nonneoplastic fibroblasts in culture was described (Roberts et al., 1981; Tucker et al., 1984). This factor was named transforming growth factor beta (TGF- ) and was later shown to inhibit or promote cell growth, depending on the cell type studied, and on the presence of other growth factors (Massague, 1990; Moses et al., 1987; Sporn and Roberts, 1987). The osteogenic activity that gave BMPs their name is just one specialized application of their broad-spectrum physiological functions, namely, the induction of bone diVerentiation. Many BMPs have nothing to do with osteogenesis at all, but play important morphogenetic roles during the embryonic development of internal organs, skin, and nervous system. Based on sequence and structural homology (Chang et al., 2002; de Caestecker, 2004; Zhao, 2003), BMPs, together with growth and diVerentiation factors, TGF- s, activins, and inhibins, are now united in a group of around 40 evolutionary conserved, small signaling peptides, called the TGF- superfamily (Table I). They share a small group of receptors with common downstream signaling pathways, which directly aVect the regulation of transcription. Resulting signaling eVects diverge in diVerent cell types into a broad spectrum of physiological changes, usually aVecting the cell cycle, cell survival, and cell diVerentiation. It has been shown that the same growth factors may induce opposite eVects in diVerent concentrations, and these eVects diVer in diVerent cell types, which means that we cannot define their function simply in physiological terms. For the purpose of this chapter, we consider TGF- family members to be morphogens used by the developing embryo to transduce the spatial and/or temporal tissue-specific information. This information is usually evaluated together with many other signals and handled in a cell-specific manner, making the dissection of the role of TGF- signaling diYcult. Herein, we review situations where the role of TGF- signaling is so critical that its abrogation enables study of the aVected developmental process, with the focus on the craniofacial structures. We hope that this chapter will provide useful information that contributes to the unfolding story of ontogenesis.

*Type I receptors ALK-1 to ALK-8 have been grouped into three categories based on structural and functional similarity (highlighted with yellow, green, and blue). R-Smads activated by individual type I receptors are marked with a dot (). Bone morphogenetic proteins (BMP) and growth and diVerentiation factors (GDF) that show a remarkable similarity have been indicated with the same color font. Other TGF- superfamily ligands are grouped together by the name. y (‘‘yes’’) indicates known ligand-receptor interaction; n (‘‘no’’) indicates no interaction.

70

Dudas and Kaartinen

B. Structure of TGF- Family Ligands All members of the TGF- superfamily share several common features: they are secreted into the extracellular space after intracellular proteolytic cleavage of large inactive dimeric precursor molecules by subtilisin-like pro-protein convertases such as SPC1/Furin, SPC4/PACE4, and SPC6A (Constam and Robertson, 2000; Cui et al., 1998; Sha et al., 1989). The carboxy-terminal products of this cleavage are receptor-binding molecules capable of direct signaling. In the case of TGF- s and growth and diVerentiation factor (GDF)-8, the N-terminal inactive part remains noncovalently attached as latency-associated peptide (LAP) to the C-terminal part, so the secreted products are inactive (latent ligands). An extracellular proteolytic action (e.g., by thrombospondin 1, plasminogen system) is required to release signaling dimers (mature ligands) from these large complexes (Chen et al., 2000; Rifkin et al., 1997, 1999). For example, latent TGF- 1 is stored in large amounts in the secretory  granules of circulating platelets and is released during platelet activation, together with thrombospondin 1 (Assoian et al., 1983; Chen et al., 2000). Furthermore, the inactive precursor complexes may interact with specific components of the extracellular matrix [e.g., latent TGF- binding proteins (LTBPs)], and accumulate as an extracellular pool, further complicating our understanding of their life cycle and signaling logic (Kaartinen and Warburton, 2003). TGF- superfamily members are also characterized by six intramolecular disulfide bridges that form the cysteine knot, a folding structure important for interactions with receptors (Sun and Davies, 1995). The seventh conserved cysteine is responsible for covalent dimerization in order to form active ligands, the exceptions being GDF-3, GDF-9, BMP-15, lefty1, and lefty2, which dimerize noncovalently, since the cysteine is substituted with serine (Chang et al., 2002). It should be noted here that BMP-1, unlike the other BMPs, is not a member of the TGF- superfamily, and thus its name is confusing, although still used in the literature (Rattenholl et al., 2002). BMP-1 is a proteinase involved in the processing of many biologically important molecules, including procollagens, laminin, and, interestingly, Chordin, an inhibitor of TGF- signaling (Scott et al., 1999). BMP-1 is a product of the same gene as the proteinase tolloid of the metzincin family and results from alternative mRNA splicing (Takahara et al., 1994).

C. Ligand Antagonists and Ligand Heterodimers Increase the Signaling Complexity Several secreted peptides are known to have an antagonistic eVect on TGF- superfamily signaling. This heterogenous group of signaling modulators is

3. TGF- Signaling and Facial Development

71

still growing, and currently comprises Noggin, Chordin/SOG, Follistatin, Follistatin-related protein (FSRP), DAN/Cerberus protein family, Sclerostin, decorin, 2-macroglobulin, and connective tissue growth factor (CTGF) (Kusu et al., 2003; Shi and Massague, 2003). These antagonists directly interact with ligands and/or with ligand-receptor binding and block signal transmission. Structural studies provide increasing evidence that some of these inhibitors share the same cysteine knot structural feature of TGF- growth factors, and can also form dimers as well, suggesting that they may compete for receptor binding with TGF- growth factors, and/or that they may have evolved from a common ancestor (Groppe et al., 2002; Shi and Massague, 2003). Otherwise, the antagonists share only slight sequence homology, and it is generally assumed that they have evolved independently from each other (Balemans and Van Hul, 2002). Accumulated knowledge shows that nothing is black and white in TGF- signaling. For example, the ‘‘antagonist’’ CTGF interferes with TGF- 1 and BMP-4 signaling, but with opposite eVects—it acts as a signaling activator in one case and as an antagonists in the other (Abreu et al., 2002). Signaling interactions within the TGF- ligand family are even more complex, and not only because some ligands show a partial aYnity to receptors for other ligands, but also because several family members act as antagonists of other ligands (e.g., inhibins vs activins; Lefty vs Nodal; activin, Nodal, or GDF-8 vs BMPs) (de Caestecker, 2004). Furthermore, in addition to forming homodimers, many TGF- members can form heterodimers with other members, resulting in mixed ligands with unexpected aYnities for diVerent receptors. In addition, the same ligand dimer can give rise to diVerent signaling outcomes upon interaction with diVerent receptors, ranging from normal binding with subsequent receptor activation, through binding only with no downstream action, to functioning as a signaling inhibitor (Israel et al., 1996; Sampath et al., 1990). Rather than exceptional, this behavior seems to be typical for the TGF- group of growth factors, and a complete description of new findings with all relevant contradictions in the current literature would be worth a separate extensive review. More details can be found in recent literature (de Caestecker, 2004; Shi and Massague, 2003).

D. Receptors Do Not Make Our Understanding of the TGF- System Easier The ‘‘canonical’’ and mostly studied transduction node of TGF- signaling is the activation of the receptor complex, composed of two diVerent types of transmembrane receptors. Type II receptors (five members known: Tbr2, Actr2, Bmpr1a, Bmpr1b, Misr2) are constitutively active Ser/Thr kinases capable of phosphorylating and thus activating type I receptors (Mehra and Wrana, 2002; Moustakas et al., 1993). Type I receptors (eight members

72

Dudas and Kaartinen

Table II Interactions of TGF- Superfamily Ligands with Type I and Type II Receptors or Receptor Complexes, and Subsequent Downstream Signal Transmission*

*Data in Tables I and II have been collected and combined from multiple literature sources (Attisano et al., 1993, 1996; Chapman et al., 2002; Cheng et al., 2003; de Caestecker, 2004; Derynck and Feng, 1997; Ebisawa et al., 1999; Erlacher et al., 1998; Franzen et al., 1993; Gouedard et al., 2000; Hogan, 1996a,b; Ikeda et al., 1996; Jamin et al., 2002; Lebrun and Vale, 1997; Lee and McPherron, 2001; Liu et al., 1995; Macias-Silva et al., 1998; Miettinen et al.,

3. TGF- Signaling and Facial Development

73

identified so far—activin receptor-like kinase ALK-1 to ALK-8) are inactive proteins that must be phosphorylated to acquire Ser/Thr kinase activity, resulting in phosphorylation of specific downstream messengers. Current understanding is that downstream signaling occurs only when two type I receptors form a complex with two type II receptors in the presence of a dimeric ligand. Only in such a complex are type I receptors activated by type II receptors, and the signal is transmitted into the cytoplasm. Everything else is variable: the strength and order of ligand binding (either to type I or first to type II) is diVerent for diVerent ligand–receptor combinations (Hart et al., 2002; Kirsch et al., 2000; Massague, 1998; Shi and Massague, 2003); both type I and type II receptors may or may not form homodimers, which can be ligand-dependent as well as ligand-independent (de Caestecker, 2004; Gilboa et al., 1998; Nohe et al., 2002). In addition, type I or type II heterodimers (or at least their mutual interactions) have been described (de Caestecker, 2004; Goumans et al., 2003; Matsuyama et al., 2003; Oh et al., 2000; ten Dijke and Hill, 2004; Ward et al., 2002). Finally, if the same ligand finds receptors preassembled in dimers, the signaling results may diVer from the situation where dimerization occurs only upon ligand binding (de Caestecker, 2004; Nohe et al., 2002). Ligand–receptor interactions are summarized in Tables I and II. In addition to type I and II receptors, so-called accessory receptors, coreceptors, or type III receptors have been described: betaglycan, endoglin/CD105, Cripto, Cryptic, and Nma/BAMBI (SchiVer et al., 2001; Shen and Schier, 2000; Shi and Massague, 2003; Yeo and Whitman, 2001). These accessory receptors probably do not transfer any signal, but may play important roles in processes such as ligand attraction to type I or II receptors, receptor–ligand complex stabilization or destabilization, type I– type II receptor complex stabilization or destabilization, or interactions with other regulatory molecules, as suggested by several studies and reviews (de Caestecker, 2004; Shen and Schier, 2000).

E. Signaling Convergence by Type I Receptors and Smad Proteins Extensive studies and numerous reviews indicate that type I receptors are the main transmitters of the TGF- signal from the cell surface into the cell. The list of currently known intracellular downstream targets is probably 1994; Moore et al., 2003; Nishitoh et al., 1996; Oh et al., 2000, 2002; Reissmann et al., 2001; Rosenzweig et al., 1995; ten Dijke et al., 1994; Visser et al., 2001; Wiater and Vale, 2003; Wrana et al., 1992; Yamashita et al., 1995; Yeo and Whitman, 2001; Zhao and Hogan, 1996). Observations in live organisms are sometimes diVerent from conclusions based on experiments in vitro. Thus, despite our best eVort, this compilation cannot be perfect, and furthermore, updates occur in the literature practically on a monthly basis.

74

Dudas and Kaartinen

incomplete, but, based on the structural homology, it can be assumed that individual type I receptors will use similar downstream signaling mechanisms. The mainstream signaling occurs via phosphorylation of a small set of Smad proteins (Smad 1, 2, 3, 5, and 8, called receptor-regulated, or R-Smads). Smad4 cooperates with all R-Smads after their phosphorylation, and is also called cooperative or common partner Smad (co-Smad). Smad6 and Smad7 bind to intracellular domains of type I receptors and function as inhibitors of R-Smad phosphorylation (I-Smads). Smad9 is in fact Smad8, and Smad10 is a novel amphibian molecule similar to Smad4 (Howell et al., 1999; LeSueur et al., 2002). In addition to I-Smads, downstream signaling is also controlled by intracellular regulation of receptor activation, or by regulation of the receptor access to downstream messengers by regulatory proteins such as FKBP12, SARA, Tob, and others (Attisano and Wrana, 1996; de Caestecker, 2004; Huse et al., 1999; Massague and Chen, 2000; Miyazono et al., 2001; Yoshida et al., 2000). Type I receptors show a strict substrate specificity for their downstream Smads, which is defined by the L45 loop in the kinase domain. According to their sequence homology and kinase activity, type I receptors can be divided into three groups, as distinguished with color in Table I (Chen et al., 1997; de Caestecker, 2004): (1) ALK-1, -2, and -8 phosphorylate Smads 1 and 5 (although in amphibians ALK-2 has also been shown to phosphorylate Smad8); (2) ALK-3 and 6 phosphorylate Smads 1, 5, and 8; (3) ALK-4, -5, and -7 phosphorylate Smad2 and Smad3. All phosphorylated R-Smads form complexes with their common partner, Smad4, and are subsequently translocated to the nucleus, where they act as transcriptional regulators. Taken together, the type I receptors are the key node, where the various actions of multiple ligand/receptor/antagonist/coreceptor combinations (40 dimerizing ligands/8  5 dimerizing receptors, etc.) are finally translated into three phosphorylation patterns of five proteins. Thus, experimental abrogation of type I and II receptors provides a tool to better understand the role of TGF- signaling in developmental processes, and the ability to precisely dissect involved downstream events.

F. Unconventional Receptors and Alternative TGF- Signaling Pathways In addition to Smad pathways, several other downstream signaling mechanisms have been proposed for type I and/or II receptors, including mitogenactivated protein kinase (MAPK) pathways (ERK, JNK, p38 MAPK, MAPKKK TAK1, TAB1), PI3 kinase, protein kinase C (PKC), PP2A, small

3. TGF- Signaling and Facial Development

75

Rho-related guanosine triphosphatases (GTP-ases), XIAP, LIM kinase 1 (LIMK1), and p70 S6K (Bakin et al., 2002; de Caestecker, 2004; Itoh et al., 2003; Mulder, 2000; Petritsch et al., 2000; Takekawa et al., 2002). However, the exact connection of these pathways with TGF- receptors is not fully known, and the possibility of their secondary activation/inhibition by Smaddependent processes is frequently discussed in literature. Existence of Smad-independent pathways is strongly supported by a recent work showing that Smad4 (the universal downstream signaler for all ALK receptors) is dispensable in certain TGF- signaling processes in early embryos (Chu et al., 2004). Also, it has been known that some tumor cells lack functional Smad4, but still respond to TGF- 1 by growth arrest in the same manner as cells that have this tumor suppressor intact (Giehl et al., 2000; Sheppard, 2001; Simeone et al., 2000). Likewise, TGF- was shown to stimulate fibronectin expression in a N-terminal Jun kinasedependent, but Smad4-independent manner in human fibrosarcoma-derived cells (Hocevar et al., 1999). The molecular mechanisms underlying these findings are not clear, but several studies show possible directions for future research. For example, in platelets, the fibrinogen receptor integrin IIb 3 has been shown to respond to TGF- 1, which improves its fibrinogenbinding properties and influences the signaling via PKC (Hoying et al., 1999). Because platelets do not have a nucleus, all these eVects must be based on nontranscriptional interactions. Interestingly, integrin v 6 has been previously known to bind and proteolytically activate latent TGF- 1 through local release of MT1 matrix melalloproteinase (Mu et al., 2002; Munger et al., 1999). These findings bring a new aspect into understanding the local nature of TGF- action in tissues, and raise the possibility that diVerent integrins may functionally cooperate in ligand attraction and activation, resulting in alternative signal transmission. Another interesting finding comes from the search for the molecules interacting with LIMK1, the key negative regulator of actin depolymerization. The cytoplasmic tail of Bmpr2 appeared among the hit clones in a yeast two-hybrid screen, and has been shown to bind and inactivate LIMK1 only in the absence of BMP-4 ligand (Foletta et al., 2003). This work shows for the first time that type II receptors are involved in downstream signaling through physical interaction with downstream targets other than their natural substrates, type I receptors. In summary, all of these alternative pathways act in a nontranscriptional manner (at least in the initial steps), in contrast to the ‘‘canonical’’ pathways employing Smads, which are direct members of transcriptionregulating complexes. None of the alternative signaling has been shown to play a role in embryonic development, thus representing a challenge for future research.

76

Dudas and Kaartinen

III. Craniofacial Phenotypes in Mutants of TGF- Superfamily Ligands and Receptors A. BMP and GDF Signaling Based on their abundance and expression pattern, BMPs and GDFs represent a heterogeneous group of growth factors sharing the same receptors in a complicated manner that is poorly understood. Based on the sequence homology and similarities in physiological behavior, these ligands are currently sorted into eight groups, but this sorting should be considered approximate and not final (see Table I) (Hogan, 1996a,b; Zhao and Hogan, 1996). The BMP–GDF signaling logic during development is diYcult to understand, and our present knowledge is very fragmented, composed of hundreds of mutually unconnected observations coming from multiple animal species, developmental stages, and practically all organ systems. The summary overview given in Table III shows that except for BMP-4, the deletion of any single ligand or receptor has practically no eVect on craniofacial development in viable mutants, or else the deletion causes early lethality prior to the beginning of cranial development. On the other hand, expression patterns suggest that the importance of BMP–GDF signaling in the head region should be comparable to the rest of the body (Mina, 2001). New important knowledge in this field comes from conditional inactivation of individual receptors in a tissue-specific manner utilizing the Cre–IoxP system. This allows deletion of early lethal genes later in development, and only in certain cell lineages and tissue types. Thus, vitally important early developmental processes are often not aVected, and the later eVects of every mutation can be studied in detail. Briefly, in this binary transgenic system, two diVerent transgenic strains are crossed with each other. One strain expresses the Cre recombinase under the control of a tissue-specific promoter, and another strain is genetically manipulated to possess a so-called floxed allele, in which two loxP sites flank a functionally essential segment of the gene. When the Cre recombinase is expressed in a cell harboring the floxed allele, a piece of DNA flanked by the loxP sites is spliced oV by Cre-recombinase activity, leading to a loss of an essential function of the targeted gene. Recently, Dudas et al. analyzed the eVect of abrogation of Alk2 in neural crest cells using this powerful system. These Alk2/Wnt1-Cre mice display severe defects both in the calvaria and in facial structures (Dudas et al., 2004b). As can be seen in Fig. 3 frontal bones show poor ossification when compared with a control littermate. Moreover, zygomatic arches display deficient posterior regions. The mandible in Alk2/Wnt1-Cre mice is hypoplastic, and Meckel’s cartilages fail to fuse in the midline. These mice also display cleft palate, which appears to result from the defective elevation of

Table III Craniofacial Phenotypes of Animals with Mutations in TGF- Superfamily Members and Related Signaling Molecules

(Continued )

Table III

Continued

(Continued )

Table III Continued

(Continued )

Table III Continued

84

Dudas and Kaartinen

palatal shelves. Some of these findings are described later in paragraphs discussing individual developmental processes in more detail.

B. TGF- Signaling There are three isoforms of TGF- in mammals (TGF- 1, 2, and 3). They all display unique expression patterns during mouse development, both temporally and spatially (Akhurst et al., 1990a,b; Fitzpatrick et al., 1990; Hogan et al., 1994; Pelton et al., 1990a; Wall and Hogan, 1994). Still, it is somehow surprising that mice lacking one particular TGF- isoform do not share phenotypic features with mice lacking another isoform, nor do they demonstrate obvious functional redundancy. Tgf- 1 / mice exhibit two diVerent phenotypes (Shull et al., 1992) without craniofacial involvement (Kulkarni et al., 1993). About 50% of them die around embryonic day 8.5 (E8.5), with severe defects in yolk sac vasculogenesis and hematopoiesis, whereas the remaining mice survive beyond birth, but then develop severe multifocal inflammatory disease. Mice lacking Tgf- 2 display several diVerent developmental defects with variable penetrance, and die soon after birth. These phenotypes include cleft palate. Interestingly, mice deficient in Tgf- 3 display nonsyndromic cleft palate with 100% penetrance. Although the palatal defect in Tgf- 2 / mice is likely caused by poor mesenchymal proliferation leading to a growth retardation of prefusion palatal shelves (Sanford et al., 1997), the palatal shelves in Tgf- 3-null mutants grow normally, become adherent, but still fail to fuse (Kaartinen et al., 1995, 1997; Proetzel et al., 1995; Taya et al., 1999). Mice deficient in Tgfbr2 display a phenotype identical to that seen in the most severely aVected Tgf- 1-null mutants (Oshima et al., 1996), whereas Alk5 / mice die at midgestation with defects in angiogenesis, but not in hematopoiesis (Larsson et al., 2001). Craniofacial defects of mice lacking Tgfbr2 specifically in neural crest cells (NCCs) were recently described (Ito et al., 2003). In these mice, cranial NCCs migrate normally. However, aVected mice display cleft palate, mandibular hypoplasia, and severe defects in calvaria development, which result from defects in cell proliferation, both in the dura mater and in the NCC-derived palatal mesenchyme, respectively.

C. Activin and Inhibin Signaling Activin A and B subunits are widely expressed during development, as well as during adulthood (Feijen et al., 1994; Roberts and Barth, 1994). These proteins can form three dimeric ligands: activin A ( A + A), activin B ( B + B), and activin AB ( A + B). Dimerization with the distantly related inhibin-specific subunit  yields two diVerent inhibins, inhibin A ( + A) and

3. TGF- Signaling and Facial Development

85

inhibin B ( + B). Two more activin chains, C and E, have been cloned in mammals (Fang et al., 1997; Schmitt et al., 1996), whereas D has so far been identified only in Xenopus laevis (Oda et al., 1995). Activins C and E seem to be specific for the adult liver with dispensable function, and no role during embryogenesis has been revealed by gene targeting (Lau et al., 2000). Mice deficient in A subunit are viable, but die within 24 hours with partially penetrant hypotrophic, missing, or cleft secondary palate. Mutants always lack lower (mandibular) incisors and mandibular molars, which do not develop beyond a rudimentary bud (Ferguson et al., 1998); all other mandibular and maxillar teeth develop normally, despite the fact that A expression is equal in all tooth types. Act B deficiency leads only to impaired eyelid development and female infertility, and defect of both subunits leads simply to a combined phenotype (Matzuk et al., 1995a,b). Inhibin  chain deletion does not cause any craniofacial abnormalities (Matzuk et al., 1992), whereas mice deficient in activin antagonist Follistatin display several pathological phenotypes, including cleft palate (Matzuk et al., 1995c). In conclusion, from the activin family only activin subunit A plays a critical role in craniofacial development. The role of other activin subunits, if any, remains unclear. The possibility that other ligands compensate for their function in activin knockouts without penetrating phenotypes must be experimentally addressed. Combined deletion of receptors Actr2A and Actr2B results in a lethal defect in mesoderm formation, and all embryos die at the gastrulation stage (Song et al., 1999). Keeping in mind that activin double mutants passed this developmental stage normally, it is clear that activin type II receptors must mediate other than activin A, B, or AB signal during this process, which is just one of many examples of receptor sharing in TGF- signaling. Individual inactivation of Actr2A leads to partially penetrant cleft palate and a lack of incisors, a defect similar to those seen in A mutants, whereas 10% of embryos showed mild eyelid defects similar to B mutants (Matzuk et al., 1995a; Song et al., 1999). Around 80% of embryos developed normally with no malformations detected. In contrast to the ligand mutants, Actr2A inactivation also caused mandibular hypoplasia with variable, often severe dysmorphism of Meckel’s cartilage and its derivatives. No craniofacial defects were detected in Actr2B mutants (Oh and Li, 1997), except low-penetrant cleft palate dependent on the genetic background (Ferguson et al., 2001). It is remarkable that deletions of the individual activin type II receptors only partially overlap with ligand-dependent malformations, which can be explained by mutual functional compensation between Actr2A and 2B. Because double mutants do not survive beyond gastrulation, this question must be addressed with appropriate tools (e.g., by utilizing conditional gene knockout technology). The fact that new phenotypical features appeared in Actr2A mutants in comparison to ligand mutants suggests that ligands other than activins must be involved (based on known receptor aYnities, candidates may be inhibins,

86

Dudas and Kaartinen

nodal, BMP-2, -6, -7, and GDF-1, -5, -8, -9b, -11). The nature of these defects (hypotrophic mandible, defective Meckel’s cartilage) suggest that they may be connected with neural crest development, which can be addressed in future by tissue-specific deletion. Partial penetrance of defects suggests that this will not be a simple task, because receptor–ligand promiscuity is likely. Another unresolved question is: which type I receptor passes the activin signal during the developmental processes aVected by ligand and type II receptor inactivation? Homozygous Smad2 knockouts die at embryonic day E10 (Nomura and Li, 1998), but heterozygotes survive with a range of variable phenotypes. Around 3% of them lack mandibular incisors and molars, suggesting the involvement of one of the Smad2-signaling receptors. Among them, ALK-4 has been shown to bind activins, but further research is necessary, because Alk4 knockouts die before gastrulation (Gu et al., 1998). Current knowledge suggests that the most typical craniofacial malformations caused by the activin family signaling are a lack of mandibular incisors and molars, and cleft palate. These features depend strongly on the Actr2A receptor and on an unknown type I receptor, and other unidentified receptors and compensating ligands may be involved. Furthermore, the Actr2A receptor is involved in the mandible and Meckel’s cartilage development through an activin-independent pathway.

IV. Head Organizers and Early Anterior Development To keep this review straightforward and focused, we think the best starting point is the formation of anterior identity, prechordal mesoderm (the future source of the most head musculature and of some connective tissue and skeletal elements), and delamination of cranial NCCs (future source of the majority of skeletal and connective tissues of the head).

A. Anterior Visceral Endoderm Acts Synergistically with Derivatives of the Gastrula Organizer In mammals, extraembryonal tissues have been shown to play a key role in establishing the body plan (Beddington and Robertson, 1998). The anterior visceral endoderm (AVE), which overlies the prospective anterior side of the epiblast and defines initial anterior identity of the mouse embryo, acts synergistically with the prechordal plate and paraxial mesoderm (derivatives of the gastrula organizer) to induce development of the head (Tam and Steiner, 1999). Nodal, expressed from the epiblast, has been shown to play an important role in initial establishment of the AVE. During normal gastrulation, the AVE is gradually displaced by the definitive endoderm, whereas in mice lacking the

3. TGF- Signaling and Facial Development

87

BMP type I receptor Alk3 (Bmpr1a) in the epiblast, the AVE appears expanded, suggesting that BMPs provide inhibitory signals for the AVE (Davis et al., 2004). Secreted BMP antagonists Chordin and Noggin, expressed by the mouse node, are not a requisite for establishing the AVE, but are required for the subsequent maintenance and further expansion of the anterior pattern (Bachiller et al., 2000). To conclude, the current view favors a model where (1) anterior identity is first displayed by the AVE and is established before initiation of gastrulation, and (2) the AVE provides signaling activity to promote anterior and suppress posterior patterning (Robb and Tam, 2004).

B. The Future Head Location: Prechordal Plate Mesenchyme with BMP Downregulation/Nodal Upregulation The prechordal plate is a region located immediately rostral to the notochord, just under the developing forebrain. This region is sometimes called the head organizer, and, in fact, in some vertebrates such as amphibians and fish, this region originates in part from the Spemann’s gastrula organizer (Foley et al., 1997; Pera and Kessel, 1997; Zoltewicz and Gerhart, 1997). The formation of the prechordal mesoderm is dependent on intricate changes in TGF- superfamily signaling. First, inhibition of BMP signaling has been shown to be critical for successful prechordal gastrulation. This involves the expression of secreted multifunctional BMP/Nodal/Wnt inhibitor Cerberus (Glinka et al., 1997; Piccolo et al., 1999). Inactivation of other BMP antagonists, Noggin and Chordin, results in forebrain defects and cyclopia (Bachiller et al., 2000). On the other hand, activin and nodal signaling must be switched on to allow formation of the prechordal mesoderm and its subsequent maintenance. Their inactivation leads to similar developmental defects with cyclopia (Agius et al., 2000; Bisgrove et al., 1999; Conlon et al., 1994; Feldman et al., 1998; Gritsman et al., 1999, 2000; Meno et al., 1999; Sampath et al., 1998; Thisse and Thisse, 1999; Zhou et al., 1993). Chimeric embryos composed of cells deficient in Smad2 (downstream of Nodal) and wild-type cells, as well as Nodal–Smad2 heterozygotes, are cyclopic (Heyer et al., 1999; Nomura and Li, 1998), and inactivation of Smad2 in the epiblast has been shown to result in failure to specify prechordal plate progenitors (Vincent et al., 2003). The idea that BMP downregulation and simultaneous Nodal upregulation are both required for prechordal mesoderm specification is further supported by observations of cyclopia in one-eyedpinhead (Oep) factor mutants in fish (Schier et al., 1997; Zhang et al., 1998). This EGF-CFC-related protein has been shown to function as a cofactor of Nodal and inhibitor of BMP signaling (Gritsman et al., 1999; Kiecker et al., 2000; SchiVer et al., 2001; Whitman, 2001; Yeo and Whitman, 2001). The fact that Nodal inhibitor Cerberus, as well as a high level of Nodal itself, are both required for

88

Dudas and Kaartinen

proper prechordal mesenchyme development suggests the involvement of complex spatio-temporal relations in this region that await more detailed investigation. The complexity of tissue–tissue interactions will further increase later, after the immigration of NCCs with their own expression pattern of TGF- –BMP ligands and receptors. C. Paraxial Mesoderm, Neural Crest Segregation, and Possible Involvement of BMP-4 Signaling The paraxial mesenchyme represents another important precursor of the head tissues. It is located laterally to the prechordal plate, and ventrally/ laterally from the developing cranial portion of the neural tube. Although this region is frequently described as unsegmented, a transient formation of seven loose aggregations (somitomeres) of its cells has been observed during development (Fig. 1). Cells of these repetitive structures migrate into various parts of the head and contribute to the formation of muscles, skeletal components, and endothelium of arteries of all branchial arches (Jacobson, 1988; Meier and Tam, 1982; Tam et al., 1982). One of the best-documented roles of the paraxial mesoderm in patterning of the head is the influence on the flow of NCCs into specific destinations (see section V). As demonstrated in Fig. 1, cranial NCCs in vertebrates migrate ventrally as three distinct streams: trigeminal stream from rhombomeres 1 and 2 (R1, R2) populating the head, including the first pharyngeal arch; hyoid stream from R4, populating the second branchial arch; and postotic stream from R6 and R7, populating the third and more distal branchial arches. In mice (and chick), there are no streams arising from R3 and R5, although their NCCs originate from these segments. Most of these NCCs die by apoptosis in chickens probably due to increased BMP-4 signaling, whereas the NCCs from the neighboring even-numbered rhombomeres express BMP antagonist Noggin, protecting them from death (Smith and Graham, 2001); so far, no similar mechanism has been identified in mice. The surviving minority of NCCs joins the more rostral or caudal adjacent stream. Migration of NCCs derived from the R4 transplanted to the R3 region copies the migration pattern of original R3-derived NCCs, and pieces of R3 transplanted into the R6 region generated areas with no NCC formation around them (Farlie et al., 1999; Kuratani and Eichele, 1993; Niederlander and Lumsden, 1996). These observations imply that both the rhombomeres and somitomeres carry molecular determinants, and that mutual interaction controls the proper segmental behavior of NCC migration. EphrinB2 and ephrinA5 are involved in attraction–repulsion regulation of cranial NCC migration in mice, and proper segmental guidance has been shown to be

Figure 1 A schematic presentation of cranial neural crest cell (NCC) migration, and contribution of cranial neural crest cells and somitomeres to craniofacial structures. Lateral and dorsal views (left and right) show the approximate positions of the segmented neural tube and the neighboring paraxial mesoderm in the embryo. Blue arrows indicate the origin and the migration direction of cranial NCCs. Final destinations of NCCs and cells derived from the somitomeres of the paraxial mesoderm are indicated with lines. F, Forebrain; ggl., ganglion; MA, anterior midbrain; MP, posterior midbrain; opt., optic vesicle; OV, otic vesicle; pom, periotic mesenchyme; R1 to R7, rhombomeres 1 to 7; S1 to S3, somites 1 to 3; SM1 to SM7, somitomeres 1 to 7.

90

Dudas and Kaartinen

critical for the proper patterning of the mandible and pharyngeal arch derivatives (Francis-West et al., 2003).

V. Neural Crest in Early Craniofacial Development A. Neural Crest Cells Migrate to Multiple Sites of the Developing Embryo Described for the first time in 1868 by the Swiss embryologist His as a band of cells between the surface epithelium and the neural tube in chicken embryos, NCCs became one of the most studied cells during embryogenesis (Hall, 1999; Trainor et al., 2003). They appear at the dorsolateral edge of the closing neural folds, along practically the entire length of the vertebrate neuraxis. This is the line generally referred to as the neural plate border, where the surface ectodermal epithelium functionally splits into two areas: neuroectoderm, or neural plate, which forms the neural tube and completely invaginates into the embryonic body, and the rest, which continues to be the body surface epithelium and becomes the epidermis. Cell fate tracing experiments were extraordinarily fruitful in neural crest research and revealed that NCCs originate by delamination from the epithelium, migrate into the mesenchyme, and populate multiple distant sites in the embryo, giving rise to multiple cell types. The cell types described were the neurons of peripheral nerves and neural ganglia, Schwann cells forming the myelin, various chromaYn and neuroendocrine cells, melanocytes, and glial cells.

B. Cranial Neural Crest Is the Major Player in Head Development During the 1890s, Julia Platt demonstrated in the mud puppy Necturus sp. that the visceral cartilages of the head and dentinoblasts also originate from the neural crest (NC) (Platt, 1897; Trainor et al., 2003). This was a very interesting observation opposing the contemporary classical knowledge on the origin of tissue types from the three germ layers. Indeed, future research provided substantial evidence that in the head region, and only there, most bone, cartilage, and connective tissue does not arise from the mesoderm, but from the NC. As demonstrated by grafting experiments, the cranial portion of the NC maintains the osteogenic activity when transplanted into the body trunk and forms ectopic hard-tissue nodules, whereas the trunk NC transplanted to the head region cannot contribute to the formation of the craniofacial bones, meninges, and smooth muscle layer of blood vessels (Etchevers et al., 1999; Le Douarin et al., 1974, 1977; Nakamura, 1982; Nakamura and Ayer-le Lievre, 1982). These and other functional diVerences contributed to the definition of the cranial neural crest (CNC) as a distinct part of the NC.

3. TGF- Signaling and Facial Development

91

It comprises the entire NC from the forebrain to caudal hindbrain, where the trunk NC begins. The segmentation and cell fate of cranial NCCs are summarized in Fig. 1. C. BMP Signaling and the Induction of Neural Crest Contact-dependent interactions between the surface ectoderm and neuroepithelium are a trigger for NCC specification, and both epithelial cell types contribute to the neural crest population (Selleck and Bronner-Fraser, 1995). In preneural avian embryos, the entire surface ectoderm expresses BMP-4 and BMP-7, whereas at later stages this expression ceases. The exception is the dorsal neural tube, where the expression continues, suggesting that BMP signaling plays a role in positioning the neural plate borders (the dorsal part of the tube is formed by fusion of the left and right borders of the neural plate during the final stage of invagination). Cell fate tracing demonstrated that daughter cells of a single cell from this dorsal part of the neural tube are later found in both the neural tube and the NC, suggesting that BMP signaling is involved in generating of at least a part of the NCC population. In in vivo and in vitro experiments with avian embryos, BMP-4 and BMP-7 can substitute for the missing surface epithelium in NC induction assays, and BMP-4 is suYcient to induce the expression of NC-specific zincfinger transcription factor Slug, followed by NC segregation from the neural tube (Liem et al., 1995). Other neural tube genes shown to be expressed under the control of BMP-4 are cadherin 6b, RhoB, Msx1, Msx2, and Pax3. Implantation of cells expressing the Noggin inhibitor into the closing neural tube inhibits NC formation and migration. Taken together, the spatiotemporal expression pattern of BMP-4 correlates well with functional studies showing its role in NC induction by the neural tube. Important questions remain: how important is BMP-4 expression in the surface epithelium, and what role is played by BMP-7? The weak part of the studies described earlier is that serum-enriched culture media were used in some experiments, and several results are not reproducible with chemically defined substrates (Garcia-Castro et al., 2002). Because this topic is not directly related to craniofacial development, we refer the reader to reviews describing this controversial issue in more detail (Gammill and Bronner-Fraser, 2002, 2003; Santagati and Rijli, 2003; Trainor et al., 2003). D. BMP Signaling and Delamination of Neural Crest Cells Cells from the neural plate and the marginal ectoderm that become NCCs undergo a specific process of phenotypical transformation. Cells of epithelial origin transdiVerentiate into mesenchyme, while they completely rebuild

92

Dudas and Kaartinen

their cell–cell contacts, acquire motility, and change their interactions with the extracellular matrix (Duband et al., 1995; Monier-Gavelle and Duband, 1995). Transformed cells must destroy or pass the barrier represented by the basal lamina, and migrate in the appropriate direction. This process, known as epithelial-to-mesenchymal transdiVerentiation (EMT), occurs at many places in the developing embryo, and it occurs during tissue remodeling and/or at the fusion of two structures covered by epithelium. In adults, EMT probably does not occur under physiological conditions, but has been implicated as a cancer invasion mechanism (Ellenrieder et al., 2001). Guanosine triphosphate (GTP)-binding protein RhoB from the Ras superfamily is expressed in the dorsal part of the neural tube under the control of BMP-4 (and possibly other BMPs). This protein is also transiently expressed in new NCCs, and its inhibition prevents NC delamination but not the specification of premigratory NCCs (Liu and Jessell, 1998). The exact mechanism of action of RhoB during the NCC delamination remains unclear. However, closer examination reveals that inhibition of RhoB causes a lack of actin stress fibers and prevents morphological changes of prospective NCCs. This is consistent with the known role of Rho proteins in cytoskeletal rearrangements during EMT, especially in reorganization of the actin cytoskeleton (Kaartinen et al., 2002; Nobes and Hall, 1995; Ridley and Hall, 1992). Studies on several cell types suggest that RhoB, but not RhoA or RhoC, is the rapidly inducible source of Rho activity in the cell, which is rapidly and strongly expressed after exposure to growth factors such as BMPs (Jahner and Hunter, 1991; Liu and Jessell, 1998; Nobes et al., 1995). This can explain in part how the TGF- superfamily signaling influences the cell fate of responsive cells.

VI. Facial Prominences and Formation of the Face A. Facial Development Is Based on Fusion of Several Regions of Tissue Early steps of facial development are astonishingly similar between diVerent vertebrate species. In mice, cranial neural crest cells (CNCs) start to migrate soon after gastrulation, and the entire delamination takes place between the stages of 3 to 16 somites (Francis-West et al., 2003). CNCs populate the five facial prominences: the frontonasal prominence, the paired maxillary, and the paired mandibular prominences (Fig. 2), which form superior, inferior, and lateral boundaries of the future oral cavity, stomodeum. These facial primordia are covered by a thin epithelium derived from the ectoderm. As outlined previously, the mesenchyme is largely composed of the NC, whereas the core of the facial prominences contains some mesodermal cells (Francis-West et al., 2003). DiVerential proliferation of facial processes is

3. TGF- Signaling and Facial Development

93

Figure 2 Initial phases of facial formation. (A) Frontal views demonstrate the development of the facial region by fusion of various prominences. The frontonasal mass (FNM, yellow) gives rise to medial and lateral nasal processes (MNP, orange; LNP, red), which fuse together to form the nostrils. In addition, FNM fuses in the midline with maxillary prominences (MXP, green) of the first branchial arch to form the upper lip, maxilla, and palate. The lower portions of the first branchial arches (BA1, blue) fuse in the midline and give rise to the mandible and lower lip. (B) Fate mapping of NCCs using the Wnt1-Cre/R26R reporter assay (Ito et al., 2003) shows that facial bones and cartilage, and frontal bones, stain blue and are derived from the neural crest.

94

Dudas and Kaartinen

believed to define the final shape of the embryonic face (Gui et al., 1993; McGonnell et al., 1998; MinkoV, 1980, 1991). As soon as facial processes reach the appropriate form and size, they fuse with neighboring primordia (Johnston and Bronsky, 1995). For instance, shortly after NCC migration, the ectoderm of frontonasal processes forms two thickened epithelial regions, the nasal placodes, which subsequently curl outward to give rise to the lateral and medial nasal processes (LNP and MNP). LNP and MNP grow further and eventually fuse with the maxillary processes of the first branchial arches, forming the upper lip and the primary palate. As delineated earlier, the role of TGF- –BMP signaling during the induction and migration of NCCs has been intensively studied; in contrast, little is known about the role of TGF- s and BMPs in other aspects of facial formation, particularly in fusion of facial primordia. Based on the expression patterns of BMP-2 and BMP-4 in the developing chick face, it has been suggested that these growth factors play a role in outgrowth of the primordia (Francis-West et al., 1994). Consistent with this, Ashique et al. (2002) demonstrated that the BMP antagonist Noggin reduced proliferation and outgrowth of the frontonasal mass and maxillary prominences. Interestingly, it also has been shown in chick embryos that Noggin, in conjunction with retinoids, can induce a duplicate set of frontonasal mass skeletal elements in place of maxillary bones, suggesting that BMP signaling, together with retinoids, is involved in specification of tissue identity, at least in avian facial prominences (Lee et al., 2001). Furthermore, it has been shown that BMPs can both induce and maintain expression of their bona fide eVectors Msx1 and Msx2. Both these genes display very strong and characteristic patterns of epithelial expression in the first pharyngeal arch, and mice deficient in these closely related homeobox genes display severe developmental defects in derivatives of the first pharyngeal arch, including the mandible, palate, and frontal bones (Satokata et al., 2000).

VII. Mandibular Development Jaws, especially teeth, are remarkable anatomical structures. Currently, it is highly controversial regarding which emerged first during phylogeny (Butler, 1995; Smith and Coates, 1998), and also, information about the contribution of the embryonic germ layers to developing teeth is being frequently refined. The largest portion of present knowledge on teeth development comes from studies in rodents that possess monophyodont dentition with continuously growing incisors, and missing canine and premolar teeth. Because this knowledge is not directly applicable as a model for human diseases, and because this topic is already well covered in the present literature, we refer the reader to the latest research and review articles by specialists in this field (Chai et al.,

3. TGF- Signaling and Facial Development

95

2000; Cobourne and Sharpe, 2003; Jernvall and ThesleV, 2000; Jung et al., 2003; Laurikkala et al., 2003; Slavkin et al., 1992; ThesleV, 2003; ThesleV and Mikkola, 2002; Xu et al., 2003; Zhang et al., 2003). Here we concentrate on the lower jaw development, a derivative of the first branchial arch. A. Identity of the First Branchial Arch Branchial arches are metameric, segmental structures of the embryonic neck that give rise to gills in some groups of vertebrates (e.g., fish), while they are transformed into many unsimilar derivatives in others. Similarly, as in other repetitive structures (somites, vertebrae, limb parts), branchial arch identity is given by expression of an unique combination of homeobox genes (GendronMaguire et al., 1993; Mallo and Brandlin, 1997; Rijli et al., 1993; Schneider and Helms, 2003; Trainor and Krumlauf, 2000, 2001). The subepithelial mass of the first branchial arch is populated by NCCs delaminating from the midbrain to the rhombomere 2. These cells do not show homeobox expression (Francis-West et al., 2003; Hunt and Krumlauf, 1991; Hunt et al., 1998; Richman and Lee, 2003). This represents an exception from other branchial arches expressing Hox genes, and may be a result of distinct regulation of the corresponding NC by the isthmus (Hunt et al., 1998; Irving and Mason, 2000; Noden, 1983; Prince and Lumsden, 1994; Trainor et al., 2002). After immigration of NCCs, the first branchial arch consists of the ectodermal epithelium of stomodeum (primitive oral cavity), followed by the primitive gut endoderm after the pharyngeal membrane, surface ectodermal epithelium (future skin) and ectomesenchyme (i.e., cells derived from NCCs that lie peripherally under the epithelium) (Chai et al., 2000; Noden, 1986, 1988; Trainor and Tam, 1995), and a small portion of central mesenchymal core, probably derived from the original mesodermal cells of the first branchial arch (Francis-West et al., 2003). Both mesenchymal components combine during further development of the skeleton, musculature, and other connective tissues, with involvement of many signaling pathways setting the axial polarity and/or segmentation of developing tissues. B. TGF- /Smad Signaling Regulates Growth of Meckel’s Cartilage A key event in the development of derivatives of the first branchial arch is the formation of Meckel’s cartilage, which acts as an axis along which the mandible, nerves, and vessels develop. Meckel’s cartilage appears for the first time in the molar region as a cell condensate. Modern molecular tools for cell lineage labeling provided a new insight into composition of Meckel’s cartilage (Ito et al., 2002): NCCs initiate its formation and form two chondrogenic fronts responsible for anterior and posterior growth.

96

Dudas and Kaartinen

Later, middle parts contain a major portion of non-NCCs. Speculations are held about the origin of these non-NCCs; they may be of mesodermal origin from the branchial arch mesenchymal core, or migrate from the neural tube as so-called VENT cells (Chai et al., 2000; Sohal et al., 1999). During further development, Meckel’s cartilage is mostly resorbed, with the exception of the anterior part forming the mandibular symphysis, and the posterior part giving rise to the middle ear ossicles malleus and stapes, and to anterior malleolar and sphenomandibular ligaments. The proliferation and chondrogenic diVerentiation of cells in Meckel’s cartilage is under the direct control of TGF- –Smad2 signaling (Chai et al., 1994). In this study, R-Smads Smad2 and Smad3 have been detected in the cartilage and its perichondrium, where they show phosphorylation and nuclear localization, whereas inhibitory Smad7 is localized predominantly in the perichondrium. Exogenous TGF- 1 selectively increased the proliferation of NC-derived cartilage cells and promoted formation of type II collagen in chondrocytes and type I collagen in the perichondrium. This eVect was diminished in Smad2+/ explants, whereas Smad3 haploinsuYciency had no eVect. The exact role of particular receptors, as well as the involvement of other TGF- superfamily ligands, is still controversial. For example, mice deficient in Tgfbr2 (the canonical partner of ALK-5 in mediation of TGF- signals) in NCCs develop the rostral process of Meckel’s cartilage, and the length of the cartilage seems to be proportional to the length of the lower jaw (Ito et al., 2003). Although that study did not describe whether Meckel’s cartilages fuse normally in the midline, the findings in the Tgfbr2 mutant raise questions about whether TGF- /Smad2 signaling is dominant in regulation of the growth of Meckel’s cartilage as proposed earlier and/or whether the signaling cascade works in a canonical manner. Ongoing experiments suggest that the signaling network is not simple, because inactivation of BMP receptor ALK-2 severely aVects mandibular morphology and growth of the anterior pole of Meckel’s cartilage (see later).

C. BMP Signaling Is Critical in the Rostral Part of Developing Lower Jaw The specific role of TGF- superfamily signaling in NC-derived cells of Meckel’s cartilage has been definitely confirmed by recent experiments using conditional gene knockout techniques. In these studies, the Cre recombinase was driven by the NC-specific Wnt1-Cre promoter, which was used to inactivate several TGF- superfamily receptors. Inactivation of Alk2 (Fig. 3) did not aVect NCC delamination and migration, but resulted in fully penetrant mandibular hypoplasia by approximately 40% (Dudas et al., 2004b). Detailed analysis at E13 revealed that the anterior

3. TGF- Signaling and Facial Development

97

Figure 3 Tissue-specific deletion of Alk2 receptor in neural crest cells leads to numerous developmental defects in the craniofacial region. (A, B) Micro-CT (computerized tomography) of the skull, frontal superior aspect; (C, D) micro-CT, frontal inferior aspect; (E, F) alizarin red/alcian blue staining of the skull, lateral view; (G, H) staining of the mandibles, upper view.

98

Dudas and Kaartinen

part of Meckel’s cartilage, containing the chondrogenic front, was missing. Further experiments showed that cell proliferation was noticeably reduced in the anterior region, which can explain shortening of the cartilage. Intermediate parts of Meckel’s cartilage appeared normal in whole-mount stainings, but histological evaluation revealed smaller and rounded diameter of cartilage and lower mitotic index of chondrocytes. This may be a consequence of insuYcient performance of the anterior growth front, or of continuing insuYciency of cartilage cells deficient in ALK-2 to respond to growth factors. Posterior parts of Meckel’s cartilages were not aVected, and their derivatives (middle ear ossicles) developed normally. This can serve as one of many examples of functional regionalization in the first branchial arch, when cells of the same origin and morphology acquire diVerent biological responsiveness, depending on their anteroposterior location. The phenotype of newborn Alk2 mutants in the NC was fully consistent with the previously described findings. Rostral processes of Meckel’s cartilages were missing (i.e., did not reach the rostral midline, and, subsequently, the mental symphysis was completely absent). The mandibles themselves were more or less normally shaped, but substantially smaller than in controls. Because it is known that the mandible is formed from NC-derived as well as non-NC-derived cells (Ito et al., 2002), it is possible that NCCs play a passive, osteogenic role in the developing mandible, without a substantial morphogenetic function. Striking exceptions are secondary cartilages of the condylar and angular processes, which were completely missing (see Fig. 3). It is not clear whether this defect is a consequence of failed chondrogenic induction in the performed mandible or of a failure of NCCs to contribute to these structures. Addressing this question may help to further understand specific aspects of BMP signaling in craniofacial bone development in general, because similar defects were found in several other bones, including complete absence of the temporomandibular joint and calvaria defects. For more details and for a complete phenotype description in all aVected systems, see the original articles (Dudas et al., 2004b; Kaartinen et al., 2004). Almost exactly opposite eVects on mandibular development have been described after inactivation of FGF-8 in the epithelium of the first branchial arch. This deletion resulted in a complete loss of mandibular skeletal structures, with the exception of mandibular midline with incisors and the rostral process of Meckel’s cartilage, which developed normally (Trumpp et al., 1999). Taken together with the ALK-2 study, it is tempting to conclude that the anterior mandible and Meckel’s cartilage represent a separate Micro-CT, as well as bone and cartilage staining (red and blue, respectively), in mutant mouse newborns reveals impaired ossification in frontal bones, short mandible, cleft palate (not shown), and absence of several structures in the temporal region (dotted square in E, F), including completely missing temporomandibular joint (arrow).

3. TGF- Signaling and Facial Development

99

developmental unit. This part of the lower jaw may be under the control of signals from the ectodermal epithelium, such as BMP-4, which is a potential ligand for ALK-2 and is expressed in the rostral tips of mandibular arch epithelium (Bennett et al., 1995; Francis-West et al., 1994). Other possible signaling candidates are BMP-7, expressed also by the ectoderm of the first branchial arch (Lyons et al., 1995), and BMP-2, present in the arch ectomesenchyme (Bennett et al., 1995; Francis-West et al., 1994). Further support for the idea that BMPs are predominantly involved in the development of the anterior part of the lower jaw comes from the observation that abrogation of TGF- signaling (Tgfbr2) in NCCs seems to have little eVect on the rostral process of Meckel’s cartilage (Ito et al., 2003), in contrast to inactivation of the BMP type I receptor ALK-2. On the other hand, the mandible is smaller, similar to Alk2 mutants, and, in addition, the angular process is completely missing. This suggests that both BMP and TGF- signaling act simultaneously and synergistically in the NC-derived cells of the developing mandible, but in some of its regions, one of them plays a more dominant or critical role (BMP signaling in rostral Meckel’s cartilage and secondary processal cartilages, TGF- signaling in the angular process, Fig. 4).

D. Intramembranous Ossification of the Mandible Ossification of the head bones is diVerent from endochondral ossification seen in the body. So-called intramembranous ossification starts as condensations of the mesenchyme into nodules. Some of them turn into capillaries, whereas others give rise to osteoblasts producing osteoid matrix, capable of binding and depositing calcium (Cohen, 2000b). The mandibular bone is formed from the ectomesenchyme, which is derived from NCCs. The head epidermis was proposed as a source of proossification instruction signals; BMP-2, BMP-4, and BMP-7 are the candidate signalers that instruct NCCs to diVerentiate into bone by inducing the expression of CBFA1/RUNX2, and core binding factor A1/runt homeodomain protein 2 (Ducy et al., 1997; Harada et al., 1999). The role of CBFA1 is to upregulate the expression of osteocalcin, osteopontin, and several other extracellular matrix proteins characteristic of bone. Mice homozygous for the knockout CBFA1 allele completely lack bone, whereas the cartilagenous skeleton is fully formed (Komori et al., 1997; Otto et al., 1997). In the mandibular region of newborns, the mandible and teeth are missing, but Meckel’s cartilage is present in an intact form. CBFA1 links BMP signaling to cleidocranial dysplasia in humans, because heterozygous mice showed a very similar phenotype, and all aVected human individuals are heterozygous for mutations in the CBFA1 gene (Lee et al., 1997; Mundlos et al., 1997).

100

Dudas and Kaartinen

Figure 4 Early development of the mandible. (A) Ossification of the future mandible starts in mesenchymal condensations around Meckel’s cartilage (blue), that becomes partly embedded into the bony tissue (B), while other parts disappear. The posterior part of Meckel’s cartilage gives rise to the middle ear ossicles malleus and incus, the anterior malleal ligament, and the sphenomandibular ligament (B, C, D). Rostral tip of Meckel’s cartilage persists as the cartilage of the mandibular symphysis (D). In the posterior end, secondary cartilages appear on the condylar and angular processes (green). (C) A schematic representation of diVerential eVects of BMP and TGF- signalings on the mandibular development.

VIII. Palatal Development and Cleft Palate A. Palatogenesis in Mice as a Model for Human Development and Disease Cleft palate is one of the most common congenital birth defects in humans. It aVects approximately 1 out of 700 individuals with some variations in all ethnic groups all over the world. During recent decades, palatal fusion

3. TGF- Signaling and Facial Development

101

(palatogenesis) has been very intensely studied, and, in conjunction with significant improvements in many diVerent disciplines of biomedical research (e.g., molecular biology, mouse and human genetics and bioimaging), has produced a wealth of information about the physiology of normal palatogenesis, as well as information about the pathogenetic mechanism of cleft palate. Because palatogenesis in mice is remarkably similar to that in humans, the mouse has become a laboratory animal of choice for studies on palatal fusion (Abbott and Birnbaum, 1991; Blavier et al., 2001; Kosazuma et al., 2004; Schutte and Murray, 1999; Taya et al., 1999). In mice, the palate is formed around E12 from outgrowths of maxillary processes of the mandibular arch. Proliferation of the mostly NC-derived mesenchyme of palatal shelves leads to their rapid growth, which initially takes place vertically along the sides of the tongue (Fig. 5A, B). Elongation of the lower jaw and other morphogenetic events direct rapid expansion of the oral cavity, which allows descent of the tongue and subsequent elevation of the palatal shelves (Fig. 5C, D). In mice this takes place around the E14. Soon after the elevation, the epithelium in tips of the apposing palatal shelves [so-called medial edge epithelium (MEE)] becomes adherent (Taya et al., 1999). MEE cells intercalate and form the so-called palatal epithelial midline seam (Martinez-Alvarez et al., 2000b), and eventually disappear (Fig. 5E–H). Cleft palate can result from a failure in any one of these steps (see the schemes on Fig. 5). The critical step in palatal fusion is removal of MEE cells from the midline seam. Three diVerent fates have been hypothesized to account for the disappearance of the MEE from the palatal midline: (1) programmed cell death (apoptosis), (2) epithelial to mesenchymal transdiVerentiation, and (3) migration.

B. Apoptosis The original hypothesis of apoptosis of the MEE was presented several decades ago by numerous investigators (Farbman, 1968; Glucksmann, 1965; Pourtois, 1966; Saunders, 1966; Shuler, 1995). During recent years, development of new analytical tools has played an instrumental role in revitalizing studies on the role of apoptosis in midline palatal fusion. First, the TUNEL assay was used to verify that there actually are positively staining cells in the midline seam, particularly in the epithelial triangles (Mori et al., 1994; Taniguchi et al., 1995). These studies were more recently extended by Martinez-Alvarez et al. (2000b), who also suggested that TGF- 3 has a role as an inducer of apoptosis during palatogenesis, and by Cuervo and colleagues, who showed that retinoids play a key role in induction of apoptosis in the MEE (Cuervo et al., 2002; Cuervo and Covarrubias, 2004). Additional in vivo evidence to support an idea that apoptosis plays a key role in defining the fate

102

Dudas and Kaartinen

Figure 5 Palatal shelf growth, elevation, and fusion. Schemes on the left side show cross sections through the embryonic palate during various developmental stages. Drawings on the right side demonstrate the corresponding macroscopic views. (A, B) Palatal shelves appear as short protrusions from the maxillary region and grow vertically by the sides of the tongue.

3. TGF- Signaling and Facial Development

103

of the MEE was provided by Yoshida et al. (1998), who demonstrated that mice deficient in Apaf-1 (mammalian homologue of C. elegans CED-4) display a palatal phenotype very similar to that of TGF- 3 / mice, in which fully grown palatal shelves fail to fuse due to a failure of MEE cells to die. An alternative view presented by Nawshad et al. (2004) suggests that only the outer layer of the MEE, called the periderm, undergoes apoptosis, whereas the basal MEE cells undergo EMT. C. Alternative Fates of the MEE: Migration, Epithelial to Mesenchymal Transdifferentiation, or Both? Lineage tracing using membrane-intercalating vital dye, DiI, was originally used to study the fates of migrating NCCs (Serbedzija et al., 1989). Subsequently, this method was used to study MEE fate in mouse palates both in vitro (Carette and Ferguson, 1992; Fitchett and Hay, 1989; Shuler et al., 1991) and in vivo (Shuler et al., 1992). Although Carette and Ferguson concluded that MEE cells migrate to oral and nasal epithelial triangles, Fitchett and Hay, and Shuler et al., showed that a large portion of MEE cells transdiVerentiate to mesenchymal cells during palatal fusion (EMT). These conclusions were also supported by immunostaining for epithelial and mesenchymal markers (Shuler et al., 1991, 1992). Subsequent lineage-tracing studies using green fluorescent protein in conjunction with retroviral or adenoviral gene transduction have been consistent with these original findings, showing that the basal layer of the MEE undergoes EMT (Cuervo et al., 2002; Martinez-Alvarez et al., 2000b). However, neither the molecular mechanism of induction of this process during palatogenesis nor how it is coordinated with apoptotic cell death is currently known. D. The Role of TGF- Superfamily Signaling in Palatogenesis All three mammalian TGF- isoforms are expressed in the palatal region before and during the fusion. TgfF- 3 expression can first be seen in the

Growth defects occurring at this stage in the palatal mesenchyme lead to wide palatal clefts. (C, D) Subsequently, palatal shelves elevate to a horizontal position. Again, defects in palatal growth and/or elevation defects lead to wide clefts. Moreover, unrelated disorders accompanied with reduced size of the oral cavity may mechanically prevent palatal elevation by blocking the descent of the tongue. (E, F) Palatal shelves continue to grow horizontally until they meet and adhere to each other in the midline. In addition to insuYcient mesenchymal growth, epithelial dysfunction preventing disappearance of the midline epithelial seam, and subsequent fusion of mesenchymal palatal masses, can lead to complete, partial, or submucous palatal clefts. (G, H) Under physiological conditions, palatal shelves fuse in the midline forming the secondary palate, the upper wall of the mouth consisting of the hard and soft palate.

104

Dudas and Kaartinen

epithelium of tips of vertically growing shelves. Its expression is very strong in the MEE of apposing shelves and in the midline seam, but ceases simultaneously with the disappearance of the midline seam. In contrast, Tgf- 2 is expressed in the palatal mesenchyme during growth of palatal shelves, as well as during elevation and fusion, whereas the pattern of Tgf- 1 is more diVuse both in the mesenchyme and, later, in the epithelium (Fitzpatrick et al., 1990; Pelton et al., 1990a,b). As described before, mice deficient in TGF- 3 suVer from isolated cleft palate without any other craniofacial symptoms (Kaartinen et al., 1995; Proetzel et al., 1995). In these mice, fully grown palatal shelves fail to fuse, and therefore the role of TGF- 3, specifically in the MEE, has been a subject of intense study. Some studies have suggested that TGF- 3 specifically induces EMT (Kaartinen et al., 1997; Sun et al., 1998a,b), and other investigations have shown that TGF- 3 induces specific morphological changes in the MEE (Gato et al., 2002; Martinez-Alvarez et al., 2000a; Taya et al., 1999; Tudela et al., 2002). These include the formation of long filopodia on the apical surface of the apposing epithelia, expression of chondroitin sulfate proteoglycan on the apical surface of the MEE, and emergence of bulging or protruding cells, which were postulated to be critical for palatal adhesion and intercalation of apposing shelves, and for subsequent apoptosis (Gato et al., 2002; Taya et al., 1999; Tudela et al., 2002). Other studies have shown that TGF- 3 regulates expression of matrix metalloproteinases (MMPs), particularly MMP-13, which likely plays an important role in the remodeling of the basement membrane during epithelial fusion (Blavier et al., 2001). Recently, Dudas et al. (2004a) demonstrated that TGF- 3 signaling in the MEE is mediated predominantly by the TGF- type I receptor, ALK-5, which subsequently activates the intracellular signal transducer Smad2. In addition, it was recently reported that TGF- 3 signaling is capable of activating the LEF1 gene in the MEE (Nawshad and Hay, 2003). These authors demonstrated that the activation of LEF1 was Smad2-dependent, but did not involve -catenin. This is rather unexpected, because TCF/LEF1 transcription factors are usually activated by the canonical Wnt/ -catenin pathway, which is the only well-characterized signaling system involved in the induction of EMT at the time of this writing.

E. Epithelial–Mesenchymal Interactions, Interactive Signaling Pathways, and Morphogenesis of the Prefusion Palatal Shelves As outlined earlier, many details are now known about TGF- signaling and its biological role in the MEE. In addition, recent studies have started to address the complex but important questions of how TGF- signaling

3. TGF- Signaling and Facial Development

105

Figure 6 A model of signaling interactions during palatogenesis. A schematic representation of a transversal section through the palatal shelf. Light yellow, nasal (up) and oral (down) epithelium; strong yellow, medial edge epithelium; tan, palatal mesenchyme. TGF- signaling molecules are in green, BMPs in yellow, FGF signaling in orange. Green arrows represent stimulatory or upregulatory eVects; red pointers represent inhibitory eVects or negative regulation. (A) TGF- 3 knockout mice have cleft palate caused by a failure of fully grown palatal shelves to fuse, because the midline epithelium fails to disappear. ALK-5 is the canonical

106

Dudas and Kaartinen

interacts with other indispensable signal transduction pathways and how the coordinated reciprocal interactions between the palatal epithelium and the underlying mesenchyme take place during palatal shelf growth, elevation, and fusion. These novel findings have been summarized in the schematic presentation in Fig. 6. Zhang et al. (2002) first showed that the bona fide TGF- –BMP eVector Msx1 is expressed in the palatal mesenchyme and controls a genetic hierarchy of BMPs and sonic hedgehog (Shh). These investigators proposed that in the anterior palate, Msx1 expression induced by BMP-4 is also required to maintain steady levels of BMP-4. Furthermore, BMP-4 is required to induce Shh expression in the anterior palatal epithelium, which in turn signals back to the mesenchyme to induce cell proliferation and palatal growth. Because Alk2/Wnt1-Cre mutants display defective palatal shelf elevation (Dudas et al., 2004b), it is conceivable that at least some of these mesenchymal BMP signals are mediated via ALK-2 (however, mechanical prevention of shelf elevation by reduced size of the oral cavity cannot be excluded, as discussed in the original paper). Rice et al. (2004) recently showed that also Fgf10 (expressed in the mesenchyme), which signals via Fgfr2b (expressed in the epithelium), is required in induction of epithelial expression of Shh and that this network, in conjunction with the BMP signaling described earlier, is needed both in growth and in appropriate morphogenesis (shaping) of palatal shelves. In addition, it was recently reported that Tgf- 3 / mice display high levels of TGF- 1 in the palatal mesenchyme (Martinez-Alvarez et al., 2004). This in turn was postulated to lead to aberrant epithelial expression of the zinc-finger transcriptional repressor Snail and subsequent promotion of cell survival in the MEE. Interestingly, NCC specific ALK-5 abrogation leads to severe facial clefting, including cleft palate, which underlines the importance of mesenchymal TGF- signaling in palatogenesis (M. Dudas and V. Kaartinen, unpublished results).

receptor for TGF- 3 that has been shown to mediate downstream signaling required for successful palatal fusion via Smad2 (Cui et al., 2003; Dudas et al., 2004a; Kaartinen et al., 1995, 1997; Proetzel et al., 1995). (B) It has been suggested that PI-3 kinase is one of the downstream eVectors of TGF- 3 signaling (Kang and Svoboda, 2002, 2003). (C) Snail is normally expressed in a small subgroup of midline epithelial cells during a physiological fusion. However, in the absence of TGF- 3, aberrant activation of epithelial Snail by pathologically elevated levels of TGF- 1 in the mesenchyme promotes cell survival (Martinez-Alvarez et al., 2004). (D) FGF-10 expressed in the palatal mesenchyme has been shown to induce sonic hedgehog (Shh) in the oral epithelium of the developing palate (Rice et al., 2004). Interactions between hedgehog and BMP-2 and BMP-4 signaling are intriguing and may represent important regulatory mechanisms for palatal development (Murray and Schutte, 2004; Rice et al., 2004). (E) Downstream hedgehog signaling through patched (Ptc), smoothened (Smo), and mammalian Gli probably contribute to upregulation of BMP-2 expression in the palatal mesenchyme (Lum and Beachy, 2004; Rice et al., 2004).

3. TGF- Signaling and Facial Development

107

IX. Clinical Research and Applications Germline mutations, somatic mutations, and polymorphisms of genes related to TGF- signaling lead to various diseases in humans. In addition to developmental malformations, a large portion of TGF- superfamily signaling research is devoted to cancer (Siegel and Massague, 2003; Waite and Eng, 2003). Some human ‘‘TGF- ’’ diseases have phenotypic features similar to those described in rodent knockout models, but many others are quite diVerent (Akhurst, 2004; Chang et al., 2002; Kosaki et al., 1999; Mizuguchi et al., 2004). These diVerences often prevent a direct application of research observations into treatment of human diseases. The most promising approach for the practical use in the near future is the delivery of individual signaling ligands or antagonists into tissue in order to induce or facilitate healing and regeneration, or to block excessive production of hard tissues.

A. Craniofacial Fracture Healing Bone morphogenetic proteins have been shown to play a role not only during skeletal development, but also in bone injury healing (Bostrom, 1998). BMP-2 is currently one of the most intensively studied regenerative proteins with the most promising properties. For example, a mesenchymal cell line from calvariae of newborn mice responds to BMP-2 by osteoblastic diVerentiation and was successfully used to repair experimental craniotomy defects (Kadowaki et al., 2004). Similarly, bone marrow cells induced with basic FGF (bFGF) and BMP-2 diVerentiate into mature bone and are capable to heal cranial defects in rats (Akita et al., 2004). Recently, BMP-7 was clinically used in a human patient to induce bone formation from the bone marrow, in order to repair a large defect in the mandible (Gronthos, 2004; Warnke et al., 2004). TGF- signaling is also involved in membranous bone healing and may become important in mandibular fracture repair (Steinbrech et al., 2000). TGF- s also have positive eVects on the joint cartilage—they can promote cartilage anabolism and osteochondrogenesis, and are currently being experimentally tested for the treatment of osteoarthrosis (Grimaud et al., 2002), which may have implications for diseases of the temporomandibular joint.

B. Prevention of Heterotopic Bone Formation BMP-signaling antagonists have been tested in animals for use in prevention of heterotopic ossification, which is a frequent complication in

108

Dudas and Kaartinen

patients following head or neck traumas, or after bone transplantations (Weber et al., 2003). Another use for BMP antagonists would be in the prevention of the developmental defects of craniosynostoses—an idea based on recent findings that BMP signaling and BMP inhibitor Noggin are involved in the process of cranial suture ossification (Cohen, 2000a; Levine et al., 1998; Liu et al., 1999; Mavrogiannis et al., 2001; Satokata et al., 2000; Warren et al., 2003).

C. Teeth and Periodontal Regeneration A lot of enthusiasm has recently been expressed in the area of tooth biology after the identification of stem cells in the enamel organ epithelium, dental papilla, and dental pulp mesenchyme, and in late cap-stage and bell-stage tooth organs. Adult odontogenic stem cells are responsive to various biological, or even mechanical, diVerentiation stimuli and can potentially lead to new tooth repair technologies (Chai and Slavkin, 2003) or peridontal tissue regeneration. For example, human recombinant BMP-2 is able to promote osteogenesis as well as cementogenesis and is being investigated for the use in periodontal regeneration (King, 2001).

X. Conclusions During recent years, much has been learned about TGF- superfamily signaling in facial morphogenesis. In this progress, genetically manipulated mice in conjunction with new state-of-the-art methods of developmental, cell, and molecular biology have been critically important. Using these techniques, the function of individual components of specific TGF- signaling pathways in vivo can be analyzed and better understood. Future challenges will be in understanding the role of TGF- signaling in interactions between diVerent cell types, such as epithelial, mesenchymal, and neural crest cells, and particularly the relationship between TGF- and other morphogenetic signaling pathways.

Acknowledgments We thank S. Buckley for comments on the manuscript. This work was supported by CHLA RCDF Award (to M.D.), and the E. Schneider Foundation and the NIH grant DE13085 (to V.K.).

3. TGF- Signaling and Facial Development

109

References Abbott, B. D., and Birnbaum, L. S. (1991). TCDD exposure of human embryonic palatal shelves in organ culture alters the diVerentiation of medial epithelial cells. Teratology 43, 119–132. Abreu, J. G., Ketpura, N. I., Reversade, B., and De Robertis, E. M. (2002). Connective-tissue growth factor (CTGF) modulates cell signalling by BMP and TGF-beta. Nat. Cell Biol. 4, 599–604. Agius, E., Oelgeschlager, M., Wessely, O., Kemp, C., and De Robertis, E. M. (2000). Endodermal Nodal-related signals and mesoderm induction in Xenopus. Development 127, 1173–1183. Ahn, K., Mishina, Y., Hanks, M. C., Behringer, R. R., and Crenshaw, E. B., III (2001). BMPRIA signaling is required for the formation of the apical ectodermal ridge and dorsal-ventral patterning of the limb. Development 128, 4449–4461. Akhurst, R. J. (2004). TGF beta signaling in health and disease. Nat. Genet. 36, 790–792. Akhurst, R. J., Fitzpatrick, D. R., Gatherer, D., Lehnert, S. A., and Millan, F. A. (1990a). Transforming growth factor betas in mammalian embryogenesis. Prog. Growth Factor Res. 2, 153–168. Akhurst, R. J., Lehnert, S. A., Gatherer, D., and DuYe, E. (1990b). The role of TGF beta in mouse development. Ann. N. Y. Acad. Sci. 593, 259–271. Akita, S., Fukui, M., Nakagawa, H., Fujii, T., and Akino, K. (2004). Cranial bone defect healing is accelerated by mesenchymal stem cells induced by coadministration of bone morphogenetic protein-2 and basic fibroblast growth factor. Wound. Repair Regen. 12, 252–259. Arango, N. A., Lovell-Badge, R., and Behringer, R. R. (1999). Targeted mutagenesis of the endogenous mouse Mis gene promoter: In vivo definition of genetic pathways of vertebrate sexual development. Cell 99, 409–419. Ashcroft, G. S., Yang, X., Glick, A. B., Weinstein, M., Letterio, J. L., Mizel, D. E., Anzano, M., Greenwell-Wild, T., Wahl, S. M., Deng, C., and Roberts, A. B. (1999). Mice lacking Smad3 show accelerated wound healing and an impaired local inflammatory response. Nat. Cell Biol. 1, 260–266. Ashique, A. M., Fu, K., and Richman, J. M. (2002). Endogenous bone morphogenetic proteins regulate outgrowth and epithelial survival during avian lip fusion. Development 129, 4647–4660. Assoian, R. K., Komoriya, A., Meyers, C. A., Miller, D. M., and Sporn, M. B. (1983). Transforming growth factor-beta in human platelets. Identification of a major storage site, purification, and characterization. J. Biol. Chem. 258, 7155–7160. Attisano, L., Carcamo, J., Ventura, F., Weis, F. M., Massague, J., and Wrana, J. L. (1993). Identification of human activin and TGF beta type I receptors that form heteromeric kinase complexes with type II receptors. Cell 75, 671–680. Attisano, L., and Wrana, J. L. (1996). Signal transduction by members of the transforming growth factor-beta superfamily. Cytokine Growth Factor Rev. 7, 327–339. Attisano, L., Wrana, J. L., Montalvo, E., and Massague, J. (1996). Activation of signalling by the activin receptor complex. Mol. Cell. Biol. 16, 1066–1073. Bachiller, D., Klingensmith, J., Kemp, C., Belo, J. A., Anderson, R. M., May, S. R., McMahon, J. A., McMahon, A. P., Harland, R. M., Rossant, J., and De Robertis, E. M. (2000). The organizer factors Chordin and Noggin are required for mouse forebrain development. Nature 403, 658–661. Bakin, A. V., Rinehart, C., Tomlinson, A. K., and Arteaga, C. L. (2002). p38 mitogen-activated protein kinase is required for TGFbeta-mediated fibroblastic transdiVerentiation and cell migration. J. Cell. Sci. 115, 3193–3206.

110

Dudas and Kaartinen

Balemans, W., and Van Hul, W. (2002). Extracellular regulation of BMP signaling in vertebrates: A cocktail of modulators. Dev. Biol. 250, 231–250. Baur, S. T., Mai, J. J., and Dymecki, S. M. (2000). Combinatorial signaling through BMP receptor IB and GDF5: Shaping of the distal mouse limb and the genetics of distal limb diversity. Development 127, 605–619. Beddington, R. S., and Robertson, E. J. (1998). Anterior patterning in mouse. Trends Genet. 14, 277–284. Behringer, R. R., Cate, R. L., Froelick, G. J., Palmiter, R. D., and Brinster, R. L. (1990). Abnormal sexual development in transgenic mice chronically expressing mullerian inhibiting substance. Nature 345, 167–170. Behringer, R. R., Finegold, M. J., and Cate, R. L. (1994). Mullerian-inhibiting substance function during mammalian sexual development. Cell 79, 415–425. Belo, J. A., Bachiller, D., Agius, E., Kemp, C., Borges, A. C., Marques, S., Piccolo, S., and De Robertis, E. M. (2000). Cerberus-like is a secreted BMP and nodal antagonist not essential for mouse development. Genesis 26, 265–270. Belville, C., Van Vlijmen, H., Ehrenfels, C., Pepinsky, B., Rezaie, A. R., Picard, J. Y., Josso, N., di Clemente, N., and Cate, R. L. (2004). Mutations of the anti-mullerian hormone gene in patients with persistent mullerian duct syndrome: Biosynthesis, secretion, and processing of the abnormal proteins and analysis using a three-dimensional model. Mol. Endocrinol. 18, 708–721. Bennett, J. H., Hunt, P., and Thorogood, P. (1995). Bone morphogenetic protein-2 and -4 expression during murine orofacial development. Arch. Oral Biol. 40, 847–854. Beppu, H., Kawabata, M., Hamamoto, T., Chytil, A., Minowa, O., Noda, T., and Miyazono, K. (2000). BMP type II receptor is required for gastrulation and early development of mouse embryos. Dev. Biol. 221, 249–258. Berk, M., Desai, S. Y., Heyman, H. C., and Colmenares, C. (1997). Mice lacking the ski protooncogene have defects in neurulation, craniofacial, patterning, and skeletal muscle development. Genes Dev. 11, 2029–2039. Bisgrove, B. W., Essner, J. J., and Yost, H. J. (1999). Regulation of midline development by antagonism of lefty and nodal signaling. Development 126, 3253–3262. Blavier, L., Lazaryev, A., GroVen, J., Heisterkamp, N., DeClerck, Y. A., and Kaartinen, V. (2001). TGF-beta3-induced palatogenesis requires matrix metalloproteinases. Mol. Biol. Cell 12, 1457–1466. Borges, A. C., Marques, S., and Belo, J. A. (2001). The BMP antagonists cerberus-like and noggin do not interact during mouse forebrain development. Int. J. Dev. Biol. 45, 441–444. Borges, A. C., Marques, S., and Belo, J. A. (2002). Goosecoid and cerberus-like do not interact during mouse embryogenesis. Int. J. Dev. Biol. 46, 259–262. Borton, A. J., Frederick, J. P., Datto, M. B., Wang, X. F., and Weinstein, R. S. (2001). The loss of Smad3 results in a lower rate of bone formation and osteopenia through dysregulation of osteoblast diVerentiation and apoptosis. J. Bone Miner. Res. 16, 1754–1764. Bostrom, M. P. (1998). Expression of bone morphogenetic proteins in fracture healing. Clin. Orthop. 355, S116–S123. Brennan, J., Lu, C. C., Norris, D. P., Rodriguez, T. A., Beddington, R. S., and Robertson, E. J. (2001). Nodal signalling in the epiblast patterns the early mouse embryo. Nature 411, 965–969. Brown, C. W., Houston-Hawkins, D. E., WoodruV, T. K., and Matzuk, M. M. (2000). Insertion of Inhbb into the Inhba locus rescues the Inhba-null phenotype and reveals new activin functions. Nat. Genet. 25, 453–457. Brown, D. J., Kim, T. B., Petty, E. M., Downs, C. A., Martin, D. M., Strouse, P. J., Moroi, S. E., Milunsky, J. M., and Lesperance, M. M. (2002). Autosomal dominant stapes ankylosis with broad thumbs and toes, hyperopia, and skeletal anomalies is caused by heterozygous

3. TGF- Signaling and Facial Development

111

nonsense and frameshift mutations in NOG, the gene encoding noggin. Am. J. Hum. Genet. 71, 618–624. Brunet, L. J., McMahon, J. A., McMahon, A. P., and Harland, R. M. (1998). Noggin, cartilage morphogenesis, and joint formation in the mammalian skeleton. Science 280, 1455–1457. Butler, P. M. (1995). Ontogenetic aspects of dental evolution. Int. J. Dev. Biol. 39, 25–34. Cacalano, G., Farinas, I., Wang, L. C., Hagler, K., Forgie, A., Moore, M., Armanini, M., Phillips, H., Ryan, A. M., Reichardt, L. F., Hynes, M., Davies, A., and Rosenthal, A. (1998). GFRalpha1 is an essential receptor component for GDNF in the developing nervous system and kidney. Neuron 21, 53–62. Carabatsos, M. J., Elvin, J., Matzuk, M. M., and Albertini, D. F. (1998). Characterization of oocyte and follicle development in growth diVerentiation factor-9-deficient mice. Dev. Biol. 204, 373–384. Carette, M. J., and Ferguson, M. W. (1992). The fate of medial edge epithelial cells during palatal fusion in vitro: An analysis by Dil labelling and confocal microscopy. Development 114, 379–388. Chai, Y., Jiang, X., Ito, Y., Bringas, P., Jr., Han, J., Rowitch, D. H., Soriano, P., McMahon, A. P., and Sucov, H. M. (2000). Fate of the mammalian cranial neural crest during tooth and mandibular morphogenesis. Development 127, 1671–1679. Chai, Y., Mah, A., Crohin, C., GroV, S., Bringas, P., Jr., Le, T., Santos, V., and Slavkin, H. C. (1994). Specific transforming growth factor-beta subtypes regulate embryonic mouse Meckel’s cartilage and tooth development. Dev. Biol. 162, 85–103. Chai, Y., and Slavkin, H. C. (2003). Prospects for tooth regeneration in the 21st century: A perspective. Microsc. Res. Tech. 60, 469–479. Chang, H., Brown, C. W., and Matzuk, M. M. (2002). Genetic analysis of the mammalian transforming growth factor-beta superfamily. Endocr. Rev. 23, 787–823. Chang, H., Huylebroeck, D., Verschueren, K., Guo, Q., Matzuk, M. M., and Zwijsen, A. (1999). Smad5 knockout mice die at mid-gestation due to multiple embryonic and extraembryonic defects. Development 126, 1631–1642. Chang, H., and Matzuk, M. M. (2001). Smad5 is required for mouse primordial germ cell development. Mech. Dev. 104, 61–67. Chang, H., Zwijsen, A., Vogel, H., Huylebroeck, D., and Matzuk, M. M. (2000). Smad5 is essential for left-right asymmetry in mice. Dev. Biol. 219, 71–78. Chapman, S. C., Bernard, D. J., Jelen, J., and WoodruV, T. K. (2002). Properties of inhibin binding to betaglycan, InhBP/p120 and the activin type II receptors. Mol. Cell. Endocrinol. 196, 79–93. Chen, H., Herndon, M. E., and Lawler, J. (2000). The cell biology of thrombospondin-1. Matrix Biol. 19, 597–614. Chen, X. D., Allen, M. R., Bloomfield, S., Xu, T., and Young, M. (2003). Biglycan-deficient mice have delayed osteogenesis after marrow ablation. Calcif. Tissue Int. 72, 577–582. Chen, X. D., Shi, S., Xu, T., Robey, P. G., and Young, M. F. (2002). Age-related osteoporosis in biglycan-deficient mice is related to defects in bone marrow stromal cells. J. Bone Miner. Res. 17, 331–340. Chen, Y., Bhushan, A., and Vale, W. (1997). Smad8 mediates the signaling of the ALK-2 [corrected] receptor serine kinase. Proc. Natl. Acad. Sci. USA 94, 12938–12943. Cheng, S. K., Olale, F., Bennett, J. T., Brivanlou, A. H., and Schier, A. F. (2003). EGF-CFC proteins are essential coreceptors for the TGF-beta signals Vg1 and GDF1. Genes Dev. 17, 31–36. Chu, G. C., Dunn, N. R., Anderson, D. C., Oxburgh, L., and Robertson, E. J. (2004). DiVerential requirements for Smad4 in TGF -dependent patterning of the early mouse embryo. Development 131, 3501–3512.

112

Dudas and Kaartinen

Cobourne, M. T., and Sharpe, P. T. (2003). Tooth and jaw: Molecular mechanisms of patterning in the first branchial arch. Arch. Oral Biol. 48, 1–14. Cohen, M. M., Jr. (2000a). Craniofacial disorders caused by mutations in homeobox genes MSX1 and MSX2. J. Craniofac. Genet. Dev. Biol. 20, 19–25. Cohen, M. M., Jr. (2000b). Merging the old skeletal biology with the new. I. Intramembranous ossification, endochondral ossification, ectopic bone, secondary cartilage, and pathologic considerations. J. Craniofac. Genet. Dev. Biol. 20, 84–93. Collignon, J., Varlet, I., and Robertson, E. J. (1996). Relationship between asymmetric nodal expression and the direction of embryonic turning. Nature 381, 155–158. Colmenares, C., Heilstedt, H. A., ShaVer, L. G., Schwartz, S., Berk, M., Murray, J. C., and Stavnezer, E. (2002). Loss of the SKI proto-oncogene in individuals aVected with 1p36 deletion syndrome is predicted by strain-dependent defects in Ski / mice. Nat. Genet. 30, 106–109. Conlon, F. L., Lyons, K. M., Takaesu, N., Barth, K. S., Kispert, A., Herrmann, B., and Robertson, E. J. (1994). A primary requirement for nodal in the formation and maintenance of the primitive streak in the mouse. Development 120, 1919–1928. Constam, D. B., and Robertson, E. J. (2000). SPC4/PACE4 regulates a TGFbeta signaling network during axis formation. Genes Dev. 14, 1146–1155. Crawford, S. E., Stellmach, V., Murphy-Ullrich, J. E., Ribeiro, S. M., Lawler, J., Hynes, R. O., Boivin, G. P., and Bouck, N. (1998). Thrombospondin-1 is a major activator of TGF-beta1 in vivo. Cell 93, 1159–1170. Cuervo, R., and Covarrubias, L. (2004). Death is the major fate of medial edge epithelial cells and the cause of basal lamina degradation during palatogenesis. Development 131, 15–24. Cuervo, R., Valencia, C., Chandraratna, R. A., and Covarrubias, L. (2002). Programmed cell death is required for palate shelf fusion and is regulated by retinoic acid. Dev. Biol. 245, 145–156. Cui, X. M., Chai, Y., Chen, J., Yamamoto, T., Ito, Y., Bringas, P., and Shuler, C. F. (2003). TGF-beta3-dependent SMAD2 phosphorylation and inhibition of MEE proliferation during palatal fusion. Dev. Dyn. 227, 387–394. Cui, Y., Jean, F., Thomas, G., and Christian, J. L. (1998). BMP-4 is proteolytically activated by furin and/or PC6 during vertebrate embryonic development. EMBO J. 17, 4735–4743. Daluiski, A., Engstrand, T., Bahamonde, M. E., Gamer, L. W., Agius, E., Stevenson, S. L., Cox, K., Rosen, V., and Lyons, K. M. (2001). Bone morphogenetic protein-3 is a negative regulator of bone density. Nat. Genet. 27, 84–88. Datto, M. B., Frederick, J. P., Pan, L., Borton, A. J., Zhuang, Y., and Wang, X. F. (1999). Targeted disruption of Smad3 reveals an essential role in transforming growth factor betamediated signal transduction. Mol. Cell. Biol. 19, 2495–2504. Davis, S., Miura, S., Hill, C., Mishina, Y., and Klingensmith, J. (2004). BMP receptor IA is required in the mammalian embryo for endodermal morphogenesis and ectodermal patterning. Dev. Biol. 270, 47–63. de Caestecker, M. (2004). The transforming growth factor-beta superfamily of receptors. Cytokine Growth Factor Rev. 15, 1–11. Derynck, R., and Feng, X. H. (1997). TGF-beta receptor signaling. Biochim. Biophys. Acta 1333, F105–F150. Dickson, M. C., Martin, J. S., Cousins, F. M., Kulkarni, A. B., Karlsson, S., and Akhurst, R. J. (1995). Defective haematopoiesis and vasculogenesis in transforming growth factor-beta 1 knock out mice. Development 121, 1845–1854. Dickson, M. C., Slager, H. G., DuYe, E., Mummery, C. L., and Akhurst, R. J. (1993). RNA and protein localisations of TGF beta 2 in the early mouse embryo suggest an involvement in cardiac development. Development 117, 625–639.

3. TGF- Signaling and Facial Development

113

Dionne, M. S., Brunet, L. J., Eimon, P. M., and Harland, R. M. (2002). Noggin is required for correct guidance of dorsal root ganglion axons. Dev. Biol. 251, 283–293. Dionne, M. S., Skarnes, W. C., and Harland, R. M. (2001). Mutation and analysis of Dan, the founding member of the Dan family of transforming growth factor beta antagonists. Mol. Cell. Biol. 21, 636–643. Dong, J., Albertini, D. F., Nishimori, K., Kumar, T. R., Lu, N., and Matzuk, M. M. (1996). Growth diVerentiation factor-9 is required during early ovarian folliculogenesis. Nature 383, 531–535. Duband, J. L., Monier, F., Delannet, M., and Newgreen, D. (1995). Epithelium-mesenchyme transition during neural crest development. Acta Anat. (Basel) 154, 63–78. Ducy, P., Zhang, R., GeoVroy, V., Ridall, A. L., and Karsenty, G. (1997). Osf2/Cbfa1: A transcriptional activator of osteoblast diVerentiation. Cell 89, 747–754. Dudas, M., Nagy, A., Laping, N. J., Moustakas, A., and Kaartinen, V. (2004a). Tgf-beta3induced palatal fusion is mediated by Alk-5/Smad pathway. Dev. Biol. 266, 96–108. Dudas, M., Sridurongrit, S., Nagy, A., Okazaki, K., and Kaartinen, V. (2004b). Craniofacial defects in mice lacking BMP type I receptor Alk2 in neural crest cells. Mech. Dev. 121, 173–182. Dudley, A. T., Lyons, K. M., and Robertson, E. J. (1995). A requirement for bone morphogenetic protein-7 during development of the mammalian kidney and eye. Genes Dev. 9, 2795–2807. Dudley, A. T., and Robertson, E. J. (1997). Overlapping expression domains of bone morphogenetic protein family members potentially account for limited tissue defects in BMP7 deficient embryos. Dev. Dyn. 208, 349–362. Dunn, N. R., Winnier, G. E., Hargett, L. K., Schrick, J. J., Fogo, A. B., and Hogan, B. L. (1997). HaploinsuYcient phenotypes in Bmp4 heterozygous null mice and modification by mutations in Gli3 and Alx4. Dev. Biol. 188, 235–247. Ebisawa, T., Tada, K., Kitajima, I., Tojo, K., Sampath, T. K., Kawabata, M., Miyazono, K., and Imamura, T. (1999). Characterization of bone morphogenetic protein-6 signaling pathways in osteoblast diVerentiation. J. Cell Sci. 112, 3519–3527. Ellenrieder, V., Hendler, S. F., Boeck, W., SeuVerlein, T., Menke, A., Ruhland, C., Adler, G., and Gress, T. M. (2001). Transforming growth factor beta1 treatment leads to an epithelialmesenchymal transdiVerentiation of pancreatic cancer cells requiring extracellular signalregulated kinase 2 activation. Cancer Res. 61, 4222–4228. Elvin, J. A., Yan, C., Wang, P., Nishimori, K., and Matzuk, M. M. (1999). Molecular characterization of the follicle defects in the growth diVerentiation factor 9-deficient ovary. Mol. Endocrinol. 13, 1018–1034. Enomoto, H., Araki, T., Jackman, A., Heuckeroth, R. O., Snider, W. D., Johnson, E. M., Jr., and Milbrandt, J. (1998). GFR alpha1-deficient mice have deficits in the enteric nervous system and kidneys. Neuron 21, 317–324. Erlacher, L., McCartney, J., Piek, E., ten Dijke, P., Yanagishita, M., Oppermann, H., and Luyten, F. P. (1998). Cartilage-derived morphogenetic proteins and osteogenic protein-1 diVerentially regulate osteogenesis. J. Bone Miner. Res. 13, 383–392. Etchevers, H. C., Couly, G., Vincent, C., and Le Douarin, N. M. (1999). Anterior cephalic neural crest is required for forebrain viability. Development 126, 3533–3543. Fang, J., Wang, S. Q., Smiley, E., and Bonadio, J. (1997). Genes coding for mouse activin beta C and beta E are closely linked and exhibit a liver-specific expression pattern in adult tissues. Biochem. Biophys. Res. Commun. 231, 655–661. Farbman, A. I. (1968). Electron microscope study of palate fusion in mouse embryos. Dev. Biol. 18, 93–116. Farlie, P. G., Kerr, R., Thomas, P., Symes, T., Minichiello, J., Hearn, C. J., and Newgreen, D. (1999). A paraxial exclusion zone creates patterned cranial neural crest cell outgrowth adjacent to rhombomeres 3 and 5. Dev. Biol. 213, 70–84.

114

Dudas and Kaartinen

Feijen, A., Goumans, M. J., and van den Eijnden-van Raaij, A. J. (1994). Expression of activin subunits, activin receptors and follistatin in postimplantation mouse embryos suggests specific developmental functions for diVerent activins. Development 120, 3621–3637. Feldman, B., Gates, M. A., Egan, E. S., Dougan, S. T., Rennebeck, G., Sirotkin, H. I., Schier, A. F., and Talbot, W. S. (1998). Zebrafish organizer development and germ-layer formation require nodal-related signals. Nature 395, 181–185. Ferguson, C. A., Tucker, A. S., Christensen, L., Lau, A. L., Matzuk, M. M., and Sharpe, P. T. (1998). Activin is an essential early mesenchymal signal in tooth development that is required for patterning of the murine dentition. Genes Dev. 12, 2636–2649. Ferguson, C. A., Tucker, A. S., Heikinheimo, K., Nomura, M., Oh, P., Li, E., and Sharpe, P. T. (2001). The role of eVectors of the activin signalling pathway, activin receptors IIA and IIB, and Smad2, in patterning of tooth development. Development 128, 4605–4613. Fitchett, J. E., and Hay, E. D. (1989). Medial edge epithelium transforms to mesenchyme after embryonic palatal shelves fuse. Dev. Biol. 131, 455–474. Fitzpatrick, D. R., Denhez, F., Kondaiah, P., and Akhurst, R. J. (1990). DiVerential expression of TGF beta isoforms in murine palatogenesis. Development 109, 585–595. Foletta, V. C., Lim, M. A., Soosairajah, J., Kelly, A. P., Stanley, E. G., Shannon, M., He, W., Das, S., Massague, J., Bernard, O., and Soosairaiah, J. (2003). Direct signaling by the BMP type II receptor via the cytoskeletal regulator LIMK1. J. Cell Biol. 162, 1089–1098. Foley, A. C., Storey, K. G., and Stern, C. D. (1997). The prechordal region lacks neural inducing ability, but can confer anterior character to more posterior neuroepithelium. Development 124, 2983–2996. Francis-West, P. H., Robson, L., and Evans, D. J. (2003). Craniofacial development: The tissue and molecular interactions that control development of the head. Adv. Anat. Embryol. Cell Biol. 169, 1–138. Francis-West, P. H., Tatla, T., and Brickell, P. M. (1994). Expression patterns of the bone morphogenetic protein genes Bmp-4 and Bmp-2 in the developing chick face suggest a role in outgrowth of the primordia. Dev. Dyn. 201, 168–178. Franzen, P., ten Dijke, P., Ichijo, H., Yamashita, H., Schulz, P., Heldin, C. H., and Miyazono, K. (1993). Cloning of a TGF beta type I receptor that forms a heteromeric complex with the TGF beta type II receptor. Cell 75, 681–692. Galloway, S. M., McNatty, K. P., Cambridge, L. M., Laitinen, M. P., Juengel, J. L., Jokiranta, T. S., McLaren, R. J., Luiro, K., Dodds, K. G., Montgomery, G. W., Beattie, A. E., Davis, G. H., and Ritvos, O. (2000). Mutations in an oocyte-derived growth factor gene (BMP 15) cause increased ovulation rate and infertility in a dosage-sensitive manner. Nat. Genet. 25, 279–283. Galvin, K. M., Donovan, M. J., Lynch, C. A., Meyer, R. I., Paul, R. J., Lorenz, J. N., Fairchild-Huntress, V., Dixon, K. L., Dunmore, J. H., Gimbrone, M. A., Jr., Falb, D., and Huszar, D. (2000). A role for smad6 in development and homeostasis of the cardiovascular system. Nat. Genet. 24, 171–174. Gammill, L. S., and Bronner-Fraser, M. (2002). Genomic analysis of neural crest induction. Development 129, 5731–5741. Gammill, L. S., and Bronner-Fraser, M. (2003). Neural crest specification: Migrating into genomics. Nat. Rev. Neurosci. 4, 795–805. Garcia-Castro, M. I., Marcelle, C., and Bronner-Fraser, M. (2002). Ectodermal Wnt function as a neural crest inducer. Science 297, 848–851. Gato, A., Martinez, M. L., Tudela, C., Alonso, I., Moro, J. A., Formoso, M. A., Ferguson, M. W., and Martinez-Alvarez, C. (2002). TGF-beta(3)-induced chondroitin sulphate proteoglycan mediates palatal shelf adhesion. Dev. Biol. 250, 393–405. Gaussin, V., Van de, P. T., Mishina, Y., Hanks, M. C., Zwijsen, A., Huylebroeck, D., Behringer, R. R., and Schneider, M. D. (2002). Endocardial cushion and myocardial defects

3. TGF- Signaling and Facial Development

115

after cardiac myocyte-specific conditional deletion of the bone morphogenetic protein receptor ALK3. Proc. Natl. Acad. Sci. USA 99, 2878–2883. Geiser, A. G., Letterio, J. J., Kulkarni, A. B., Karlsson, S., Roberts, A. B., and Sporn, M. B. (1993). Transforming growth factor beta 1 (TGF-beta 1) controls expression of major histocompatibility genes in the postnatal mouse: Aberrant histocompatibility antigen expression in the pathogenesis of the TGF-beta 1 null mouse phenotype. Proc. Natl. Acad. Sci. USA 90, 9944–9948. Gendron-Maguire, M., Mallo, M., Zhang, M., and Gridley, T. (1993). Hoxa-2 mutant mice exhibit homeotic transformation of skeletal elements derived from cranial neural crest. Cell 75, 1317–1331. Giehl, K., Seidel, B., Gierschik, P., Adler, G., and Menke, A. (2000). TGFbeta1 represses proliferation of pancreatic carcinoma cells which correlates with Smad4-independent inhibition of ERK activation. Oncogene 19, 4531–4541. Gilboa, L., Wells, R. G., Lodish, H. F., and Henis, Y. I. (1998). Oligomeric structure of type I and type II transforming growth factor beta receptors: Homodimers form in the ER and persist at the plasma membrane. J. Cell Biol. 140, 767–777. Glinka, A., Wu, W., Onichtchouk, D., Blumenstock, C., and Niehrs, C. (1997). Head induction by simultaneous repression of Bmp and Wnt signalling in Xenopus. Nature 389, 517–519. Glucksmann, A. (1965). Cell death in normal development. Arch. Biol. (Liege) 76, 419–437. Gong, Y., Krakow, D., Marcelino, J., Wilkin, D., Chitayat, D., Babul-Hirji, R., Hudgins, L., Cremers, C. W., Cremers, F. P., Brunner, H. G., Reinker, K., Rimoin, D. L., Cohn, D. H., Goodman, F. R., Reardon, W., Patton, M., Francomano, C. A., and Warman, M. L. (1999). Heterozygous mutations in the gene encoding noggin aVect human joint morphogenesis. Nat. Genet. 21, 302–304. Gouedard, L., Chen, Y. G., Thevenet, L., Racine, C., Borie, S., Lamarre, I., Josso, N., Massague, J., and di Clemente, N. (2000). Engagement of bone morphogenetic protein type IB receptor and Smad1 signaling by anti-Mullerian hormone and its type II receptor. J. Biol. Chem. 275, 27973–27978. Goumans, M. J., Valdimarsdottir, G., Itoh, S., Lebrin, F., Larsson, J., Mummery, C., Karlsson, S., and ten Dijke, P. (2003). Activin receptor-like kinase (ALK)1 is an antagonistic mediator of lateral TGFbeta/ALK5 signaling. Mol. Cell 12, 817–828. Granholm, A. C., Srivastava, N., Mott, J. L., Henry, S., Henry, M., Westphal, H., Pichel, J. G., Shen, L., and HoVer, B. J. (1997). Morphological alterations in the peripheral and central nervous systems of mice lacking glial cell line-derived neurotrophic factor (GDNF): Immunohistochemical studies. J. Neurosci. 17, 1168–1178. Green, E. L. (1954). Quantitative genetics of skeletal variations in the mouse. I. Crosses between three short-ear strains (P, NB, SEC/2). J. Natl. Cancer Inst. 15, 609–627. Green, M. C. (1968). Mechanism of the pleiotropic eVects of the short-ear mutant gene in the mouse. J. Exp. Zool. 167, 129–150. Grimaud, E., Heymann, D., and Redini, F. (2002). Recent advances in TGF-beta eVects on chondrocyte metabolism. Potential therapeutic roles of TGF-beta in cartilage disorders. Cytokine Growth Factor Rev. 13, 241–257. Gritsman, K., Talbot, W. S., and Schier, A. F. (2000). Nodal signaling patterns the organizer. Development 127, 921–932. Gritsman, K., Zhang, J., Cheng, S., Heckscher, E., Talbot, W. S., and Schier, A. F. (1999). The EGF-CFC protein one-eyed pinhead is essential for nodal signaling. Cell 97, 121–132. Gronthos, S. (2004). Reconstruction of human mandible by tissue engineering. Lancet 364, 735–736. Groppe, J., Greenwald, J., Wiater, E., Rodriguez-Leon, J., Economides, A. N., Kwiatkowski, W., AVolter, M., Vale, W. W., Belmonte, J. C., and Choe, S. (2002). Structural basis of BMP signalling inhibition by the cystine knot protein Noggin. Nature 420, 636–642.

116

Dudas and Kaartinen

Gu, Z., Nomura, M., Simpson, B. B., Lei, H., Feijen, A., van den Eijnden-van Raaij, Donahoe, P. K., and Li, E. (1998). The type I activin receptor ActRIB is required for egg cylinder organization and gastrulation in the mouse. Genes Dev. 12, 844–857. Gu, Z., Reynolds, E. M., Song, J., Lei, H., Feijen, A., Yu, L., He, W., MacLaughlin, D. T., van den Eijnden-van Raaij, Donahoe, P. K., and Li, E. (1999). The type I serine/threonine kinase receptor ActRIA (ALK2) is required for gastrulation of the mouse embryo. Development 126, 2551–2561. Gui, T., Osumi-Yamashita, N., and Eto, K. (1993). Proliferation of nasal epithelial and mesenchymal cells during primary palate formation. J. Craniofac. Genet. Dev. Biol. 13, 250–258. Hall, B. K. (1999). ‘‘The Neural Crest in Development and Evolution.’’ Springer-Verlag, New York. Hamrick, M. W., McPherron, A. C., and Lovejoy, C. O. (2002). Bone mineral content and density in the humerus of adult myostatin-deficient mice. Calcif. Tissue Int. 71, 63–68. Harada, H., Tagashira, S., Fujiwara, M., Ogawa, S., Katsumata, T., Yamaguchi, A., Komori, T., and Nakatsuka, M. (1999). Cbfa1 isoforms exert functional diVerences in osteoblast diVerentiation. J. Biol. Chem. 274, 6972–6978. Hart, P. J., Deep, S., Taylor, A. B., Shu, Z., Hinck, C. S., and Hinck, A. P. (2002). Crystal structure of the human TbetaR2 ectodomain–TGF-beta3 complex. Nat. Struct. Biol. 9, 203–208. Helms, J. A., and Schneider, R. A. (2003). Cranial skeletal biology. Nature 423, 326–331. Heuckeroth, R. O., Enomoto, H., Grider, J. R., Golden, J. P., Hanke, J. A., Jackman, A., Molliver, D. C., Bardgett, M. E., Snider, W. D., Johnson, E. M., Jr., and Milbrandt, J. (1999). Gene targeting reveals a critical role for neurturin in the development and maintenance of enteric, sensory, and parasympathetic neurons. Neuron 22, 253–263. Heuckeroth, R. O., Kotzbauer, P., Copeland, N. G., Gilbert, D. J., Jenkins, N. A., Zimonjic, D. B., Popescu, N. C., Johnson, E. M., Jr., and Milbrandt, J. (1997). Neurturin, a novel neurotrophic factor, is localized to mouse chromosome 17 and human chromosome 19p13.3. Genomics 44, 137–140. Heyer, J., Escalante-Alcalde, D., Lia, M., Boettinger, E., Edelmann, W., Stewart, C. L., and Kucherlapati, R. (1999). Postgastrulation Smad2-deficient embryos show defects in embryo turning and anterior morphogenesis. Proc. Natl. Acad. Sci. USA 96, 12595–12600. Hocevar, B. A., Brown, T. L., and Howe, P. H. (1999). TGF-beta induces fibronectin synthesis through a c-Jun N-terminal kinase-dependent, Smad4-independent pathway. EMBO J. 18, 1345–1356. Hogan, B. L. (1996a). Bone morphogenetic proteins in development. Curr. Opin. Genet. Dev. 6, 432–438. Hogan, B. L. (1996b). Bone morphogenetic proteins: Multifunctional regulators of vertebrate development. Genes Dev. 10, 1580–1594. Hogan, B. L., Blessing, M., Winnier, G. E., Suzuki, N., and Jones, C. M. (1994). Growth factors in development: The role of TGF-beta related polypeptide signalling molecules in embryogenesis. Development 120(Suppl), 53–60. Honma, Y., Araki, T., Gianino, S., Bruce, A., Heuckeroth, R., Johnson, E., and Milbrandt, J. (2002). Artemin is a vascular-derived neurotropic factor for developing sympathetic neurons. Neuron 35, 267–282. Hoodless, P. A., Pye, M., Chazaud, C., Labbe, E., Attisano, L., Rossant, J., and Wrana, J. L. (2001). FoxH1 (Fast) functions to specify the anterior primitive streak in the mouse. Genes Dev. 15, 1257–1271. Howell, M., Itoh, F., Pierreux, C. E., Valgeirsdottir, S., Itoh, S., ten Dijke, P., and Hill, C. S. (1999). Xenopus Smad4beta is the co-Smad component of developmentally regulated transcription factor complexes responsible for induction of early mesodermal genes. Dev. Biol. 214, 354–369.

3. TGF- Signaling and Facial Development

117

Hoying, J. B., Yin, M., Diebold, R., Ormsby, I., Becker, A., and Doetschman, T. (1999). Transforming growth factor beta1 enhances platelet aggregation through a non-transcriptional eVect on the fibrinogen receptor. J. Biol. Chem. 274, 31008–31013. Hsiao, E. C., Koniaris, L. G., Zimmers-Koniaris, T., Sebald, S. M., Huynh, T. V., and Lee, S. J. (2000). Characterization of growth-diVerentiation factor 15, a transforming growth factor beta superfamily member induced following liver injury. Mol. Cell. Biol. 20, 3742–3751. Huang, S., Tang, B., Usoskin, D., Lechleider, R. J., Jamin, S. P., Li, C., Anzano, M. A., Ebendal, T., Deng, C., and Roberts, A. B. (2002). Conditional knockout of the Smad1 gene. Genesis 32, 76–79. Hunt, P., Clarke, J. D., Buxton, P., Ferretti, P., and Thorogood, P. (1998). Stability and plasticity of neural crest patterning and branchial arch Hox code after extensive cephalic crest rotation. Dev. Biol. 198, 82–104. Hunt, P., and Krumlauf, R. (1991). Deciphering the Hox code: Clues to patterning branchial regions of the head. Cell 66, 1075–1078. Huse, M., Chen, Y. G., Massague, J., and Kuriyan, J. (1999). Crystal structure of the cytoplasmic domain of the type I TGF beta receptor in complex with FKBP12. Cell 96, 425–436. Ikeda, T., Takahashi, H., Suzuki, A., Ueno, N., Yokose, S., Yamaguchi, A., and Yoshiki, S. (1996). Cloning of rat type I receptor cDNA for bone morphogenetic protein-2 and bone morphogenetic protein-4, and the localization compared with that of the ligands. Dev. Dyn. 206, 318–329. Irving, C., and Mason, I. (2000). Signalling by FGF8 from the isthmus patterns anterior hindbrain and establishes the anterior limit of Hox gene expression. Development 127, 177–186. Israel, D. I., Nove, J., Kerns, K. M., Kaufman, R. J., Rosen, V., Cox, K. A., and Wozney, J. M. (1996). Heterodimeric bone morphogenetic proteins show enhanced activity in vitro and in vivo. Growth Factors 13, 291–300. Ito, Y., Bringas, P., Jr., Mogharei, A., Zhao, J., Deng, C., and Chai, Y. (2002). Receptorregulated and inhibitory Smads are critical in regulating transforming growth factor betamediated Meckel’s cartilage development. Dev. Dyn. 224, 69–78. Ito, Y., Yeo, J. Y., Chytil, A., Han, J., Bringas, P., Jr., Nakajima, A., Shuler, C. F., Moses, H. L., and Chai, Y. (2003). Conditional inactivation of Tgfbr2 in cranial neural crest causes cleft palate and calvaria defects. Development 130, 5269–5280. Itoh, S., Thorikay, M., Kowanetz, M., Moustakas, A., Itoh, F., Heldin, C. H., and ten Dijke, P. (2003). Elucidation of Smad requirement in transforming growth factor-beta type I receptorinduced responses. J. Biol. Chem. 278, 3751–3761. Jacobson, A. G. (1988). Somitomeres: Mesodermal segments of vertebrate embryos. Development 104(Suppl), 209–220. Jahner, D., and Hunter, T. (1991). The ras-related gene rhoB is an immediate-early gene inducible by v-Fps, epidermal growth factor, and platelet-derived growth factor in rat fibroblasts. Mol. Cell. Biol. 11, 3682–3690. Jamin, S. P., Arango, N. A., Mishina, Y., Hanks, M. C., and Behringer, R. R. (2002). Requirement of Bmpr1a for Mullerian duct regression during male sexual development. Nat. Genet. 32, 408–410. Jernvall, J., and ThesleV, I. (2000). Reiterative signaling and patterning during mammalian tooth morphogenesis. Mech. Dev. 92, 19–29. Johnston, M. C., and Bronsky, P. T. (1995). Prenatal craniofacial development: New insights on normal and abnormal mechanisms. Crit Rev. Oral Biol. Med. 6, 368–422. Jorgez, C. J., Klysik, M., Jamin, S. P., Behringer, R. R., and Matzuk, M. M. (2004). Granulosa cell-specific inactivation of follistatin causes female fertility defects. Mol. Endocrinol. 18, 953–967.

118

Dudas and Kaartinen

Josso, N., Picard, J. Y., Imbeaud, S., di Clemente, N., and Rey, R. (1997). Clinical aspects and molecular genetics of the persistent mullerian duct syndrome. Clin. Endocrinol. (Oxf). 47, 137–144. Jung, H. S., Hitoshi, Y., and Kim, H. J. (2003). Study on tooth development, past, present, and future. Microsc. Res. Tech. 60, 480–482. Kaartinen, V., Cui, X. M., Heisterkamp, N., GroVen, J., and Shuler, C. F. (1997). Transforming growth factor-beta3 regulates transdiVerentiation of medial edge epithelium during palatal fusion and associated degradation of the basement membrane. Dev. Dyn. 209, 255–260. Kaartinen, V., Dudas, M., Nagy, A., Sridurongrit, S., Lu, M. M., and Epstein, J. A. (2004). Cardiac outflow tract defects in mice lacking ALK2 in neural crest cells. Development 131, 3481–3490. Kaartinen, V., Haataja, L., Nagy, A., Heisterkamp, N., and GroVen, J. (2002). TGFbeta3induced activation of RhoA/Rho-kinase pathway is necessary but not suYcient for epitheliomesenchymal transdiVerentiation: Implications for palatogenesis. Int. J. Mol. Med. 9, 563–570. Kaartinen, V., Voncken, J. W., Shuler, C., Warburton, D., Bu, D., Heisterkamp, N., and GroVen, J. (1995). Abnormal lung development and cleft palate in mice lacking TGF-beta 3 indicates defects of epithelial-mesenchymal interaction. Nat. Genet. 11, 415–421. Kaartinen, V., and Warburton, D. (2003). Fibrillin controls TGF-beta activation. Nat. Genet. 33, 331–332. Kadowaki, A., Tsukazaki, T., Hirata, K., Shibata, Y., Okubo, Y., Bessho, K., Komori, T., Yoshida, N., and Yamaguchi, A. (2004). Isolation and characterization of a mesenchymal cell line that diVerentiates into osteoblasts in response to BMP-2 from calvariae of GFP transgenic mice. Bone 34, 993–1003. Kang, P., and Svoboda, K. K. (2002). PI-3 kinase activity is required for epithelialmesenchymal transformation during palate fusion. Dev. Dyn. 225, 316–321. Kang, P., and Svoboda, K. K. (2003). Nicotine inhibits palatal fusion and modulates nicotinic receptors and the PI-3 kinase pathway in medial edge epithelia. Orthod. Craniofac. Res. 6, 129–142. Katagiri, T., Boorla, S., Frendo, J. L., Hogan, B. L., and Karsenty, G. (1998). Skeletal abnormalities in doubly heterozygous Bmp4 and Bmp7 mice. Dev. Genet. 22, 340–348. Kendrick, K. M., da Costa, A. P., Leigh, A. E., Hinton, M. R., and Peirce, J. W. (2001). Sheep don’t forget a face. Nature 414, 165–166. Kiecker, C., Muller, F., Wu, W., Glinka, A., Strahle, U., and Niehrs, C. (2000). Phenotypic eVects in Xenopus and zebrafish suggest that one-eyed pinhead functions as antagonist of BMP signalling. Mech. Dev. 94, 37–46. Kim, R. Y., Robertson, E. J., and Solloway, M. J. (2001). Bmp6 and Bmp7 are required for cushion formation and septation in the developing mouse heart. Dev. Biol. 235, 449–466. King, G. N. (2001). The importance of drug delivery to optimize the eVects of bone morphogenetic proteins during periodontal regeneration. Curr. Pharm. Biotechnol. 2, 131–142. King, J. A., Marker, P. C., Seung, K. J., and Kingsley, D. M. (1994). BMP5 and the molecular, skeletal, and soft-tissue alterations in short ear mice. Dev. Biol. 166, 112–122. King, J. A., Storm, E. E., Marker, P. C., DiLeone, R. J., and Kingsley, D. M. (1996). The role of BMPs and GDFs in development of region-specific skeletal structures. Ann. N. Y. Acad. Sci. 785, 70–79. Kingsley, D. M., Bland, A. E., Grubber, J. M., Marker, P. C., Russell, L. B., Copeland, N. G., and Jenkins, N. A. (1992). The mouse short ear skeletal morphogenesis locus is associated with defects in a bone morphogenetic member of the TGF beta superfamily. Cell 71, 399–410.

3. TGF- Signaling and Facial Development

119

Kirsch, T., Sebald, W., and Dreyer, M. K. (2000). Crystal structure of the BMP-2-BRIA ectodomain complex. Nat. Struct. Biol. 7, 492–496. Komori, T., Yagi, H., Nomura, S., Yamaguchi, A., Sasaki, K., Deguchi, K., Shimizu, Y., Bronson, R. T., Gao, Y. H., Inada, M., Sato, M., Okamoto, R., Kitamura, Y., Yoshiki, S., and Kishimoto, T. (2000). Targeted disruption of Cbfa1 results in a complete lack of bone formation owing to maturational arrest of osteoblasts. Cell 89, 755–764. Kosaki, R., Gebbia, M., Kosaki, K., Lewin, M., Bowers, P., Towbin, J. A., and Casey, B. (1999). Left-right axis malformations associated with mutations in ACVR2B, the gene for human activin receptor type IIB. Am. J. Med. Genet. 82, 70–76. Kosazuma, T., Hashimoto, S., Ohno, H., Chou, M. J., and Shiota, K. (2004). Organ culture of the fetal mouse palate for screening the developmental toxicity of chemicals: A validation study. Congenit. Anom. (Kyoto) 44, 60–71. Kulkarni, A. B., Huh, C. G., Becker, D., Geiser, A., Lyght, M., Flanders, K. C., Roberts, A. B., Sporn, M. B., Ward, J. M., and Karisson, S. (1993). Transforming growth factor beta 1 null mutation in mice causes excessive inflammatory response and early death. Proc. Natl. Acad. Sci. USA 90, 770–774. Kung, A. L., Rebel, V. I., Bronson, R. T., Ch’ng, L. E., SieV, C. A., Livingston, D. M., and Yao, T. P. (2000). Gene dose-dependent control of hematopoiesis and hematologic tumor suppression by CBP. Genes Dev. 14, 272–277. Kuratani, S. C., and Eichele, G. (1993). Rhombomere transplantation repatterns the segmental organization of cranial nerves and reveals cell-autonomous expression of a homeodomain protein. Development 117, 105–117. Kusu, N., Laurikkala, J., Imanishi, M., Usui, H., Konishi, M., Miyake, A., ThesleV, I., and Itoh, N. (2003). Sclerostin is a novel secreted osteoclast-derived bone morphogenetic protein antagonist with unique ligand specificity. J. Biol. Chem. 278, 24113–24117. Larrain, J., Bachiller, D., Lu, B., Agius, E., Piccolo, S., and De Robertis, E. M. (2000). BMPbinding modules in chordin: A model for signalling regulation in the extracellular space. Development 127, 821–830. Larsson, J., Goumans, M. J., Sjostrand, L. J., van Rooijen, M. A., Ward, D., Leveen, P., Xu, X., ten Dijke, P., Mummery, C. L., and Karlsson, S. (2001). Abnormal angiogenesis but intact hematopoietic potential in TGF-beta type I receptor-deficient mice. EMBO J. 20, 1663–1673. Lau, A. L., Kumar, T. R., Nishimori, K., Bonadio, J., and Matzuk, M. M. (2000). Activin betaC and betaE genes are not essential for mouse liver growth, diVerentiation, and regeneration. Mol. Cell. Biol. 20, 6127–6137. Laurikkala, J., Kassai, Y., Pakkasjarvi, L., ThesleV, I., and Itoh, N. (2003). Identification of a secreted BMP antagonist, ectodin, integrating BMP, FGF, and SHH signals from the tooth enamel knot. Dev. Biol. 264, 91–105. Lawler, J., Sunday, M., Thibert, V., Duquette, M., George, E. L., Rayburn, H., and Hynes, R. O. (1998). Thrombospondin-1 is required for normal murine pulmonary homeostasis and its absence causes pneumonia. J. Clin. Invest. 101, 982–992. Lawson, K. A., Dunn, N. R., Roelen, B. A., Zeinstra, L. M., Davis, A. M., Wright, C. V., Korving, J. P., and Hogan, B. L. (1999). Bmp4 is required for the generation of primordial germ cells in the mouse embryo. Genes Dev. 13, 424–436. Le Douarin, N., Fontaine, J., and Le Lievre, C. (1974). New studies on the neural crest origin of the avian ultimobranchial glandular cells—interspecific combinations and cytochemical characterization of C cells based on the uptake of biogenic amine precursors. Histochemistry 38, 297–305. Le Douarin, N. M., Teillet, M. A., and Le Lievre, C. (1977). Influence of the tissue environment on the diVerentiation of neural crest cells. Soc. Gen. Physiol Ser. 32, 11–27.

120

Dudas and Kaartinen

Lebrun, J. J., and Vale, W. W. (1997). Activin and inhibin have antagonistic eVects on liganddependent heteromerization of the type I and type II activin receptors and human erythroid diVerentiation. Mol. Cell Biol. 17, 1682–1691. Lechleider, R. J., Ryan, J. L., Garrett, L., Eng, C., Deng, C., Wynshaw-Boris, A., and Roberts, A. B. (2001). Targeted mutagenesis of Smad1 reveals an essential role in chorioallantoic fusion. Dev. Biol. 240, 157–167. Lee, B., Thirunavukkarasu, K., Zhou, L., Pastore, L., Baldini, A., Hecht, J., GeoVroy, V., Ducy, P., and Karsenty, G. (1997). Missense mutations abolishing DNA binding of the osteoblast-specific transcription factor OSF2/CBFA1 in cleidocranial dysplasia. Nat. Genet. 16, 307–310. Lee, K. J., Mendelsohn, M., and Jessell, T. M. (1998). Neuronal patterning by BMPs: A requirement for GDF7 in the generation of a discrete class of commissural interneurons in the mouse spinal cord. Genes Dev. 12, 3394–3407. Lee, S. H., Fu, K. K., Hui, J. N., and Richman, J. M. (2001). Noggin and retinoic acid transform the identity of avian facial prominences. Nature 414, 909–912. Lee, S. J., and McPherron, A. C. (2001). Regulation of myostatin activity and muscle growth. Proc. Natl. Acad. Sci. USA 98, 9306–9311. LeSueur, J. A., Fortuno, E. S., III, McKay, R. M., and GraV, J. M. (2002). Smad10 is required for formation of the frog nervous system. Dev. Cell 2, 771–783. Levine, J. P., Bradley, J. P., Roth, D. A., McCarthy, J. G., and Longaker, M. T. (1998). Studies in cranial suture biology: Regional dura mater determines overlying suture biology. Plast. Reconstr. Surg. 101, 1441–1447. Li, D. Y., Sorensen, L. K., Brooke, B. S., Urness, L. D., Davis, E. C., Taylor, D. G., Boak, B. B., and Wendel, D. P. (1999). Defective angiogenesis in mice lacking endoglin. Science 284, 1534–1537. Liem, K. F., Jr., Tremml, G., Roelink, H., and Jessell, T. M. (1995). Dorsal diVerentiation of neural plate cells induced by BMP-mediated signals from epidermal ectoderm. Cell 82, 969–979. Liu, F., Ventura, F., Doody, J., and Massague, J. (1995). Human type II receptor for bone morphogenic proteins (BMPs): Extension of the two-kinase receptor model to the BMPs. Mol. Cell. Biol. 15, 3479–3486. Liu, J. P., and Jessell, T. M. (1998). A role for rhoB in the delamination of neural crest cells from the dorsal neural tube. Development 125, 5055–5067. Liu, Y., Festing, M., Thompson, J. C., Hester, M., Rankin, S., El Hodiri, H. M., Zorn, A. M., and Weinstein, M. (2004). Smad2 and Smad3 coordinately regulate craniofacial and endodermal development. Dev. Biol. 270, 411–426. Liu, Y. H., Tang, Z., Kundu, R. K., Wu, L., Luo, W., Zhu, D., Sangiorgi, F., Snead, M. L., and Maxson, R. E. (1999). Msx2 gene dosage influences the number of proliferative osteogenic cells in growth centers of the developing murine skull: A possible mechanism for MSX2mediated craniosynostosis in humans. Dev. Biol. 205, 260–274. Lowe, L. A., Yamada, S., and Kuehn, M. R. (2001). Genetic dissection of nodal function in patterning the mouse embryo. Development 128, 1831–1843. Lum, L., and Beachy, P. A. (2004). The Hedgehog response network: Sensors, switches, and routers. Science 304, 1755–1759. Luo, G., Hofmann, C., Bronckers, A. L., Sohocki, M., Bradley, A., and Karsenty, G. (1995). BMP-7 is an inducer of nephrogenesis, and is also required for eye development and skeletal patterning. Genes Dev. 9, 2808–2820. Lyons, K. M., Hogan, B. L., and Robertson, E. J. (1995). Colocalization of BMP 7 and BMP 2 RNAs suggests that these factors cooperatively mediate tissue interactions during murine development. Mech. Dev. 50, 71–83.

3. TGF- Signaling and Facial Development

121

Macias-Silva, M., Hoodless, P. A., Tang, S. J., Buchwald, M., and Wrana, J. L. (1998). Specific activation of Smad1 signaling pathways by the BMP7 type I receptor, ALK2. J. Biol. Chem. 273, 25628–25636. Mallo, M., and Brandlin, I. (1997). Segmental identity can change independently in the hindbrain and rhombencephalic neural crest. Dev. Dyn. 210, 146–156. Martinez-Alvarez, C., Blanco, M. J., Perez, R., Rabadan, M. A., Aparicio, M., Resel, E., Martinez, T., and Nieto, M. A. (2004). Snail family members and cell survival in physiological and pathological cleft palates. Dev. Biol. 265, 207–218. Martinez-Alvarez, C., Bonelli, R., Tudela, C., Gato, A., Mena, J., O’Kane, S., and Ferguson, M. W. (2000a). Bulging medial edge epithelial cells and palatal fusion. Int. J. Dev. Biol. 44, 331–335. Martinez-Alvarez, C., Tudela, C., Perez-Miguelsanz, J., O’Kane, S., Puerta, J., and Ferguson, M. W. (2000b). Medial edge epithelial cell fate during palatal fusion. Dev. Biol. 220, 343–357. Massague, J. (1990). The transforming growth factor-beta family. Annu. Rev. Cell Biol. 6, 597–641. Massague, J. (1998). TGF-beta signal transduction. Annu. Rev. Biochem. 67, 753–791. Massague, J., and Chen, Y. G. (2000). Controlling TGF-beta signaling. Genes Dev. 14, 627–644. Matsuyama, S., Iwadate, M., Kondo, M., Saitoh, M., Hanyu, A., Shimizu, K., Aburatani, H., Mishima, H. K., Imamura, T., Miyazono, K., and Miyazawa, K. (2003). SB-431542 and Gleevec inhibit transforming growth factor-beta-induced proliferation of human osteosarcoma cells. Cancer Res. 63, 7791–7798. Matzuk, M. M., Finegold, M. J., Mather, J. P., Krummen, L., Lu, H., and Bradley, A. (1994). Development of cancer cachexia-like syndrome and adrenal tumors in inhibin-deficient mice. Proc. Natl. Acad. Sci. USA 91, 8817–8821. Matzuk, M. M., Finegold, M. J., Su, J. G., Hsueh, A. J., and Bradley, A. (1992). Alpha-inhibin is a tumour-suppressor gene with gonadal specificity in mice. Nature 360, 313–319. Matzuk, M. M., Kumar, T. R., and Bradley, A. (1995a). DiVerent phenotypes for mice deficient in either activins or activin receptor type II. Nature 374, 356–360. Matzuk, M. M., Kumar, T. R., Shou, W., Coerver, K. A., Lau, A. L., Behringer, R. R., and Finegold, M. J. (1996). Transgenic models to study the roles of inhibins and activins in reproduction, oncogenesis, and development. Recent Prog. Horm. Res. 51, 123–157. Matzuk, M. M., Kumar, T. R., Vassalli, A., Bickenbach, J. R., Roop, D. R., Jaenisch, R., and Bradley, A. (1995b). Functional analysis of activins during mammalian development. Nature 374, 354–356. Matzuk, M. M., Lu, N., Vogel, H., Sellheyer, K., Roop, D. R., and Bradley, A. (1995c). Multiple defects and perinatal death in mice deficient in follistatin. Nature 374, 360–363. Mavrogiannis, L. A., Antonopoulou, I., Baxova, A., Kutilek, S., Kim, C. A., Sugayama, S. M., Salamanca, A., Wall, S. A., Morriss-Kay, G. M., and Wilkie, A. O. (2001). HaploinsuYciency of the human homeobox gene ALX4 causes skull ossification defects. Nat. Genet. 27, 17–18. McGonnell, I. M., Clarke, J. D., and Tickle, C. (1998). Fate map of the developing chick face: Analysis of expansion of facial primordia and establishment of the primary palate. Dev. Dyn. 212, 102–118. McPherron, A. C., Lawler, A. M., and Lee, S. J. (1999). Regulation of anterior/posterior patterning of the axial skeleton by growth/diVerentiation factor 11. Nat. Genet. 22, 260–264. McPherron, A. C., and Lee, S. J. (1997). Double muscling in cattle due to mutations in the myostatin gene. Proc. Natl. Acad. Sci. USA 94, 12457–12461. McPherron, A. C., and Lee, S. J. (2002). Suppression of body fat accumulation in myostatindeficient mice. J. Clin. Invest. 109, 595–601. Mehra, A., and Wrana, J. L. (2002). TGF-beta and the Smad signal transduction pathway. Biochem. Cell Biol. 80, 605–622.

122

Dudas and Kaartinen

Meier, S., and Tam, P. P. (1982). Metameric pattern development in the embryonic axis of the mouse. I. DiVerentiation of the cranial segments. DiVerentiation 21, 95–108. Meno, C., Gritsman, K., Ohishi, S., Ohfuji, Y., Heckscher, E., Mochida, K., Shimono, A., Kondoh, H., Talbot, W. S., Robertson, E. J., Schier, A. F., and Hamada, H. (1999). Mouse Lefty2 and zebrafish antivin are feedback inhibitors of nodal signaling during vertebrate gastrulation. Mol. Cell 4, 287–298. Meno, C., Ito, Y., Saijoh, Y., Matsuda, Y., Tashiro, K., Kuhara, S., and Hamada, H. (1997). Two closely related left-right asymmetrically expressed genes, lefty-1 and lefty-2: Their distinct expression domains, chromosomal linkage and direct neuralizing activity in Xenopus embryos. Genes Cells 2, 513–524. Meno, C., Shimono, A., Saijoh, Y., Yashiro, K., Mochida, K., Ohishi, S., Noji, S., Kondoh, H., and Hamada, H. (1998). Lefty-1 is required for left-right determination as a regulator of lefty-2 and nodal. Cell 94, 287–297. Miettinen, P. J., Ebner, R., Lopez, A. R., and Derynck, R. (1994). TGF-beta induced transdiVerentiation of mammary epithelial cells to mesenchymal cells: involvement of type I receptors. J. Cell Biol. 127, 2021–2036. Mina, M. (2001). Morphogenesis of the medial region of the developing mandible is regulated by multiple signaling pathways. Cells Tissues Organs 169, 295–301. MinkoV, R. (1980). Regional variation of cell proliferation within the facial processes of the chick embryo: A study of the role of ‘merging’ during development. J. Embryol. Exp. Morphol. 57, 37–49. MinkoV, R. (1991). Cell proliferation during formation of the embryonic facial primordia. J. Craniofac. Genet. Dev. Biol. 11, 251–261. Mishina, Y., Crombie, R., Bradley, A., and Behringer, R. R. (1999). Multiple roles for activinlike kinase-2 signaling during mouse embryogenesis. Dev. Biol. 213, 314–326. Mishina, Y., Hanks, M. C., Miura, S., Tallquist, M. D., and Behringer, R. R. (2002). Generation of Bmpr/Alk3 conditional knockout mice. Genesis 32, 69–72. Mishina, Y., Rey, R., Finegold, M. J., Matzuk, M. M., Josso, N., Cate, R. L., and Behringer, R. R. (1996). Genetic analysis of the Mullerian-inhibiting substance signal transduction pathway in mammalian sexual diVerentiation. Genes Dev. 10, 2577–2587. Mishina, Y., Suzuki, A., Gilbert, D. J., Copeland, N. G., Jenkins, N. A., Ueno, N., and Behringer, R. R. (1995a). Genomic organization and chromosomal location of the mouse type I BMP-2/4 receptor. Biochem. Biophys. Res. Commun. 206, 310–317. Mishina, Y., Suzuki, A., Ueno, N., and Behringer, R. R. (1995b). Bmpr encodes a type I bone morphogenetic protein receptor that is essential for gastrulation during mouse embryogenesis. Genes Dev. 9, 3027–3037. Miyazono, K., Kusanagi, K., and Inoue, H. (2001). Divergence and convergence of TGF-beta/ BMP signaling. J. Cell Physiol. 187, 265–276. Mizuguchi, T., Collod-Beroud, G., Akiyama, T., Abifadel, M., Harada, N., Morisaki, T., Allard, D., Varret, M., Claustres, M., Morisaki, H., Ihara, M., Kinoshita, A., Yoshiura, K., Junien, C., Kajii, T., Jondeau, G., Ohta, T., Kishino, T., Furukawa, Y., Nakamura, Y., Niikawa, N., Boileau, C., and Matsumoto, N. (2004). Heterozygous TGFBR2 mutations in Marfan syndrome. Nat. Genet. 36, 855–860. Monier-Gavelle, F., and Duband, J. L. (1995). Control of N-cadherin-mediated intercellular adhesion in migrating neural crest cells in vitro. J. Cell Sci. 108, 3839–3853. Moore, M. W., Klein, R. D., Farinas, I., Sauer, H., Armanini, M., Phillips, H., Reichardt, L. F., Ryan, A. M., Carver-Moore, K., and Rosenthal, A. (1996). Renal and neuronal abnormalities in mice lacking GDNF. Nature 382, 76–79. Moore, R. K., Otsuka, F., and Shimasaki, S. (2003). Molecular basis of bone morphogenetic protein-15 signaling in granulosa cells. J. Biol. Chem. 278, 304–310.

3. TGF- Signaling and Facial Development

123

Mori, C., Nakamura, N., Okamoto, Y., Osawa, M., and Shiota, K. (1994). Cytochemical identification of programmed cell death in the fusing fetal mouse palate by specific labelling of DNA fragmentation. Anat. Embryol. (Berl) 190, 21–28. Moses, H. L., CoVey, R. J., Jr., Leof, E. B., Lyons, R. M., and Keski-Oja, J. (1987). Transforming growth factor beta regulation of cell proliferation. J. Cell Physiol. Suppl 5, 1–7. Moustakas, A., Lin, H. Y., Henis, Y. I., Plamondon, J., O’Connor-McCourt, M. D., and Lodish, H. F. (1993). The transforming growth factor beta receptors types I, II, and III form hetero-oligomeric complexes in the presence of ligand. J. Biol. Chem. 268, 22215–22218. Mu, D., Cambier, S., Fjellbirkeland, L., Baron, J. L., Munger, J. S., Kawakatsu, H., Sheppard, D., Broaddus, V. C., and Nishimura, S. L. (2002). The integrin alpha(v)beta8 mediates epithelial homeostasis through MT1-MMP-dependent activation of TGF-beta1. J. Cell Biol. 157, 493–507. Mulder, K. M. (2000). Role of Ras and Mapks in TGFbeta signaling. Cytokine Growth Factor Rev. 11, 23–35. Mundlos, S., Otto, F., Mundlos, C., Mulliken, J. B., Aylsworth, A. S., Albright, S., Lindhout, D., Cole, W. G., Henn, W., Knoll, J. H., Owen, M. J., Mertelsmann, R., Zabel, B. U., and Olsen, B. R. (1997). Mutations involving the transcription factor CBFA1 cause cleidocranial dysplasia. Cell 89, 773–779. Munger, J. S., Huang, X., Kawakatsu, H., GriYths, M. J., Dalton, S. L., Wu, J., Pittet, J. F., Kaminski, N., Garat, C., Matthay, M. A., Rifkin, D. B., and Sheppard, D. (1999). The integrin alpha v beta 6 binds and activates latent TGF beta 1: A mechanism for regulating pulmonary inflammation and fibrosis. Cell 96, 319–328. Murray, J. C., and Schutte, B. C. (2004). Cleft palate: Players, pathways, and pursuits. J. Clin. Invest. 113, 1676–1678. Nakamura, H. (1982). Mesenchymal derivatives from the neural crest. Arch. Histol. Jpn. 45, 127–138. Nakamura, H., and Ayer-le Lievre, C. S. (1982). Mesectodermal capabilities of the trunk neural crest of birds. J. Embryol. Exp. Morphol. 70, 1–18. Nawshad, A., and Hay, E. D. (2003). TGFbeta3 signaling activates transcription of the LEF1 gene to induce epithelial mesenchymal transformation during mouse palate development. J. Cell Biol. 163, 1291–1301. Nawshad, A., LaGamba, D., and Hay, E. D. (2004). Transforming growth factor beta (TGFbeta) signalling in palatal growth, apoptosis and epithelial mesenchymal transformation (EMT). Arch. Oral Biol. 49, 675–689. Neuhof, H. (1917). Title unavailable—historical document. Surg. Gynec. Obst. 24, 199–383. Niederlander, C., and Lumsden, A. (1996). Late emigrating neural crest cells migrate specifically to the exit points of cranial branchiomotor nerves. Development 122, 2367–2374. Nishitoh, H., Ichijo, H., Kimura, M., Matsumoto, T., Makishima, F., Yamaguchi, A., Yamashita, H., Enomoto, S., and Miyazono, K. (1996). Identification of type I and type II serine/threonine kinase receptors for growth/diVerentiation factor-5. J. Biol. Chem. 271, 21345–21352. Nobes, C. D., and Hall, A. (1995). Rho, rac, and cdc42 GTPases regulate the assembly of multimolecular focal complexes associated with actin stress fibers, lamellipodia, and filopodia. Cell 81, 53–62. Nobes, C. D., Hawkins, P., Stephens, L., and Hall, A. (1995). Activation of the small GTPbinding proteins rho and rac by growth factor receptors. J. Cell Sci. 108, 225–233. Noden, D. M. (1983). The role of the neural crest in patterning of avian cranial skeletal, connective, and muscle tissues. Dev. Biol. 96, 144–165. Noden, D. M. (1986). Origins and patterning of craniofacial mesenchymal tissues. J. Craniofac. Genet. Dev. Biol. Suppl. 2, 15–31.

124

Dudas and Kaartinen

Noden, D. M. (1988). Interactions and fates of avian craniofacial mesenchyme. Development 103(Suppl), 121–140. Nohe, A., Hassel, S., Ehrlich, M., Neubauer, F., Sebald, W., Henis, Y. I., and Knaus, P. (2002). The mode of bone morphogenetic protein (BMP) receptor oligomerization determines diVerent BMP-2 signaling pathways. J. Biol. Chem. 277, 5330–5338. Nomura, M., and Li, E. (1998). Smad2 role in mesoderm formation, left-right patterning and craniofacial development. Nature 393, 786–790. Northcutt, R. G., and Gans, C. (1983). The genesis of neural crest and epidermal placodes: A reinterpretation of vertebrate origins. Q. Rev. Biol. 58, 1–28. Oda, S., Nishimatsu, S., Murakami, K., and Ueno, N. (1995). Molecular cloning and functional analysis of a new activin beta subunit: A dorsal mesoderm-inducing activity in Xenopus. Biochem. Biophys. Res. Commun. 210, 581–588. Oh, S. P., and Li, E. (1997). The signaling pathway mediated by the type IIB activin receptor controls axial patterning and lateral asymmetry in the mouse. Genes Dev. 11, 1812–1826. Oh, S. P., Seki, T., Goss, K. A., Imamura, T., Yi, Y., Donahoe, P. K., Li, L., Miyazono, K., ten Dijke, P., Kim, S., and Li, E. (2000). Activin receptor-like kinase 1 modulates transforming growth factor-beta 1 signaling in the regulation of angiogenesis. Proc. Natl. Acad. Sci. USA 97, 2626–2631. Oh, S. P., Yeo, C. Y., Lee, Y., Schrewe, H., Whitman, M., and Li, E. (2002). Activin type IIA and IIB receptors mediate Gdf11 signaling in axial vertebral patterning. Genes Dev. 16, 2749–2754. Oike, Y., Hata, A., Mamiya, T., Kaname, T., Noda, Y., Suzuki, M., Yasue, H., Nabeshima, T., Araki, K., and Yamamura, K. (1999a). Truncated CBP protein leads to classical RubinsteinTaybi syndrome phenotypes in mice: Implications for a dominant-negative mechanism. Hum. Mol. Genet. 8, 387–396. Oike, Y., Takakura, N., Hata, A., Kaname, T., Akizuki, M., Yamaguchi, Y., Yasue, H., Araki, K., Yamamura, K., and Suda, T. (1999b). Mice homozygous for a truncated form of CREBbinding protein exhibit defects in hematopoiesis and vasculo-angiogenesis. Blood 93, 2771–2779. Oshima, M., Oshima, H., and Taketo, M. M. (1996). TGF-beta receptor type II deficiency results in defects of yolk sac hematopoiesis and vasculogenesis. Dev. Biol. 179, 297–302. Otto, F., Thornell, A. P., Crompton, T., Denzel, A., Gilmour, K. C., Rosewell, I. R., Stamp, G. W., Beddington, R. S., Mundlos, S., Olsen, B. R., Selby, P. B., and Owen, M. J. (1997). Cbfa1, a candidate gene for cleidocranial dysplasia syndrome, is essential for osteoblast diVerentiation and bone development. Cell 89, 765–771. Pelton, R. W., Dickinson, M. E., Moses, H. L., and Hogan, B. L. (1990a). In situ hybridization analysis of TGF beta 3 RNA expression during mouse development: Comparative studies with TGF beta 1 and beta 2. Development 110, 609–620. Pelton, R. W., Hogan, B. L., Miller, D. A., and Moses, H. L. (1990b). DiVerential expression of genes encoding TGFs beta 1, beta 2, and beta 3 during murine palate formation. Dev. Biol. 141, 456–460. Pera, E. M., and Kessel, M. (1997). Patterning of the chick forebrain anlage by the prechordal plate. Development 124, 4153–4162. Petritsch, C., Beug, H., Balmain, A., and Oft, M. (2000). TGF-beta inhibits p70 S6 kinase via protein phosphatase 2A to induce G(1) arrest. Genes Dev. 14, 3093–3101. Piccolo, S., Agius, E., Leyns, L., Bhattacharyya, S., Grunz, H., Bouwmeester, T., and De Robertis, E. M. (1999). The head inducer Cerberus is a multifunctional antagonist of Nodal, BMP and Wnt signals. Nature 397, 707–710. Pichel, J. G., Shen, L., Sheng, H. Z., Granholm, A. C., Drago, J., Grinberg, A., Lee, E. J., Huang, S. P., Saarma, M., HoVer, B. J., Sariola, H., and Westphal, H. (1996a). Defects in enteric innervation and kidney development in mice lacking GDNF. Nature 382, 73–76.

3. TGF- Signaling and Facial Development

125

Pichel, J. G., Shen, L., Sheng, H. Z., Granholm, A. C., Drago, J., Grinberg, A., Lee, E. J., Huang, S. P., Saarma, M., HoVer, B. J., Sariola, H., and Westphal, H. (1996b). GDNF is required for kidney development and enteric innervation. Cold Spring Harb. Symp. Quant. Biol. 61, 445–457. Piek, E., Ju, W. J., Heyer, J., Escalante-Alcalde, D., Stewart, C. L., Weinstein, M., Deng, C., Kucherlapati, R., Bottinger, E. P., and Roberts, A. B. (2001). Functional characterization of transforming growth factor beta signaling in Smad2- and Smad3-deficient fibroblasts. J. Biol. Chem. 276, 19945–19953. Platt, J. B. (1897). The development of the cartilaginous skull and of the branchial and hypoglossal musculature in Necturus. Morphol Jb. 25, 377–464. Pourtois, M. (1966). Onset of the acquired potentially for fusion in the palatal shelves of rats. J. Embryol. Exp. Morphol. 16, 171–182. Prince, V., and Lumsden, A. (1994). Hoxa-2 expression in normal and transposed rhombomeres: Independent regulation in the neural tube and neural crest. Development 120, 911–923. Proetzel, G., Pawlowski, S. A., Wiles, M. V., Yin, M., Boivin, G. P., Howles, P. N., Ding, J., Ferguson, M. W., and Doetschman, T. (1995). Transforming growth factor-beta 3 is required for secondary palate fusion. Nat. Genet. 11, 409–414. Rankin, C. T., Bunton, T., Lawler, A. M., and Lee, S. J. (2000). Regulation of left-right patterning in mice by growth/diVerentiation factor-1. Nat. Genet. 24, 262–265. Rattenholl, A., Pappano, W. N., Koch, M., Keene, D. R., Kadler, K. E., Sasaki, T., Timpl, R., Burgeson, R. E., Greenspan, D. S., and Bruckner-Tuderman, L. (2002). Proteinases of the bone morphogenetic protein-1 family convert procollagen VII to mature anchoring fibril collagen. J. Biol. Chem. 277, 26372–26378. Reddi, A. H. (1997). Bone morphogenetic proteins: An unconventional approach to isolation of first mammalian morphogens. Cytokine Growth Factor Rev. 8, 11–20. Reissmann, E., Jornvall, H., Blokzijl, A., Andersson, O., Chang, C., Minchiotti, G., Persico, M. G., Ibanez, C. F., and Brivanlou, A. H. (2001). The orphan receptor ALK7 and the Activin receptor ALK4 mediate signaling by Nodal proteins during vertebrate development. Genes Dev. 15, 2010–2022. Rice, R., Spencer-Dene, B., Connor, E. C., Gritli-Linde, A., McMahon, A. P., Dickson, C., ThesleV, I., and Rice, D. P. (2004). Disruption of Fgf10/Fgfr2b-coordinated epithelialmesenchymal interactions causes cleft palate. J. Clin. Invest. 113, 1692–1700. Richman, J. M., and Lee, S. H. (2003). About face: Signals and genes controlling jaw patterning and identity in vertebrates. Bioessays 25, 554–568. Ridley, A. J., and Hall, A. (1992). The small GTP-binding protein rho regulates the assembly of focal adhesions and actin stress fibers in response to growth factors. Cell 70, 389–399. Rifkin, D. B., Gleizes, P. E., Harpel, J., Nunes, I., Munger, J., Mazzieri, R., and Noguera, I. (1997). Plasminogen/plasminogen activator and growth factor activation. Ciba Found. Symp. 212, 105–118. Rifkin, D. B., Mazzieri, R., Munger, J. S., Noguera, I., and Sung, J. (1999). Proteolytic control of growth factor availability. APMIS 107, 80–85. Rijli, F. M., Mark, M., Lakkaraju, S., Dierich, A., Dolle, P., and Chambon, P. (1993). A homeotic transformation is generated in the rostral branchial region of the head by disruption of Hoxa-2, which acts as a selector gene. Cell 75, 1333–1349. Robb, L., and Tam, P. P. (2004). Gastrula organiser and embryonic patterning in the mouse. Semin. Cell Dev. Biol. 15, 543–554. Roberts, A. B., Anzano, M. A., Lamb, L. C., Smith, J. M., and Sporn, M. B. (1981). New class of transforming growth factors potentiated by epidermal growth factor: Isolation from nonneoplastic tissues. Proc. Natl. Acad. Sci. USA 78, 5339–5343.

126

Dudas and Kaartinen

Roberts, V. J., and Barth, S. L. (1994). Expression of messenger ribonucleic acids encoding the inhibin/activin system during mid- and late-gestation rat embryogenesis. Endocrinology 134, 914–923. Roebroek, A. J., Umans, L., Pauli, I. G., Robertson, E. J., van Leuven, F., Van de Ven, W. J., and Constam, D. B. (1998). Failure of ventral closure and axial rotation in embryos lacking the proprotein convertase Furin. Development 125, 4863–4876. Rosenzweig, B. L., Imamura, T., Okadome, T., Cox, G. N., Yamashita, H., ten Dijke, P., Heldin, C. H., and Miyazono, K. (1995). Cloning and characterization of a human type II receptor for bone morphogenetic proteins. Proc. Natl. Acad. Sci. USA 92, 7632–7636. Rossi, J., Luukko, K., Poteryaev, D., Laurikainen, A., Sun, Y. F., Laakso, T., Eerikainen, S., Tuominen, R., Lakso, M., Rauvala, H., Arumae, U., Pasternack, M., Saarma, M., and Airaksinen, M. S. (1999). Retarded growth and deficits in the enteric and parasympathetic nervous system in mice lacking GFR alpha2, a functional neurturin receptor. Neuron 22, 243–252. Saika, S., Saika, S., Liu, C. Y., Azhar, M., Sanford, L. P., Doetschman, T., Gendron, R. L., Kao, C. W., and Kao, W. W. (2001). TGFbeta2 in corneal morphogenesis during mouse embryonic development. Dev. Biol. 240, 419–432. Sampath, K., Rubinstein, A. L., Cheng, A. M., Liang, J. O., Fekany, K., Solnica-Krezel, L., Korzh, V., Halpern, M. E., and Wright, C. V. (1998). Induction of the zebrafish ventral brain and floorplate requires cyclops/nodal signalling. Nature 395, 185–189. Sampath, T. K., Coughlin, J. E., Whetstone, R. M., Banach, D., Corbett, C., Ridge, R. J., Ozkaynak, E., Oppermann, H., and Rueger, D. C. (1990). Bovine osteogenic protein is composed of dimers of OP-1 and BMP-2A, two members of the transforming growth factorbeta superfamily. J. Biol. Chem. 265, 13198–13205. Sanchez, M. P., Silos-Santiago, I., Frisen, J., He, B., Lira, S. A., and Barbacid, M. (1996). Renal agenesis and the absence of enteric neurons in mice lacking GDNF. Nature 382, 70–73. Sanford, L. P., Ormsby, I., Gittenberger-de Groot, A. C., Sariola, H., Friedman, R., Boivin, G. P., Cardell, E. L., and Doetschman, T. (1997). TGFbeta2 knockout mice have multiple developmental defects that are non-overlapping with other TGFbeta knockout phenotypes. Development 124, 2659–2670. Santagati, F., and Rijli, F. M. (2003). Cranial neural crest and the building of the vertebrate head. Nat. Rev. Neurosci. 4, 806–818. Satokata, I., Ma, L., Ohshima, H., Bei, M., Woo, I., Nishizawa, K., Maeda, T., Takano, Y., Uchiyama, M., Heaney, S., Peters, H., Tang, Z., Maxson, R., and Maas, R. (2000). Msx2 deficiency in mice causes pleiotropic defects in bone growth and ectodermal organ formation. Nat. Genet. 24, 391–395. Saunders, J. W., Jr. (1966). Death in embryonic systems. Science 154, 604–612. Schier, A. F., Neuhauss, S. C., Helde, K. A., Talbot, W. S., and Driever, W. (1997). The oneeyed pinhead gene functions in mesoderm and endoderm formation in zebrafish and interacts with no tail. Development 124, 327–342. SchiVer, S. G., Foley, S., KaVashan, A., Hronowski, X., Zichittella, A. E., Yeo, C. Y., Miatkowski, K., Adkins, H. B., Damon, B., Whitman, M., Salomon, D., Sanicola, M., and Williams, K. P. (2001). Fucosylation of Cripto is required for its ability to facilitate nodal signaling. J. Biol. Chem. 276, 37769–37778. Schmitt, J., Hotten, G., Jenkins, N. A., Gilbert, D. J., Copeland, N. G., Pohl, J., and Schrewe, H. (1996). Structure, chromosomal localization, and expression analysis of the mouse inhibin/activin beta C (Inhbc) gene. Genomics 32, 358–366. Schneider, R. A., and Helms, J. A. (2003). The cellular and molecular origins of beak morphology. Science 299, 565–568. Schrewe, H., Gendron-Maguire, M., Harbison, M. L., and Gridley, T. (1994). Mice homozygous for a null mutation of activin beta B are viable and fertile. Mech. Dev. 47, 43–51.

3. TGF- Signaling and Facial Development

127

Schuchardt, A., D’Agati, V., Larsson-Blomberg, L., Costantini, F., and Pachnis, V. (1994). Defects in the kidney and enteric nervous system of mice lacking the tyrosine kinase receptor Ret. Nature 367, 380–383. Schuchardt, A., D’Agati, V., Larsson-Blomberg, L., Costantini, F., and Pachnis, V. (1995). RET-deficient mice: An animal model for Hirschsprung’s disease and renal agenesis. J. Intern. Med. 238, 327–332. Schutte, B. C., and Murray, J. C. (1999). The many faces and factors of orofacial clefts. Hum. Mol. Genet. 8, 1853–1859. Scott, I. C., Blitz, I. L., Pappano, W. N., Imamura, Y., Clark, T. G., Steiglitz, B. M., Thomas, C. L., Maas, S. A., Takahara, K., Cho, K. W., and Greenspan, D. S. (1999). Mammalian BMP-1/Tolloid-related metalloproteinases, including novel family member mammalian Tolloid-like 2, have diVerential enzymatic activities and distributions of expression relevant to patterning and skeletogenesis. Dev. Biol. 213, 283–300. Selleck, M. A., and Bronner-Fraser, M. (1995). Origins of the avian neural crest: The role of neural plate-epidermal interactions. Development 121, 525–538. Senn, N. (1989). On the healing of aseptic bone cavities by implantation of antiseptic decalcified bone. Am. J. Med. Sci. 98, 219–243. Serbedzija, G. N., Bronner-Fraser, M., and Fraser, S. E. (1989). A vital dye analysis of the timing and pathways of avian trunk neural crest cell migration. Development 106, 809–816. Settle, S., Marker, P., Gurley, K., Sinha, A., Thacker, A., Wang, Y., Higgins, K., Cunha, G., and Kingsley, D. M. (2001). The BMP family member Gdf7 is required for seminal vesicle growth, branching morphogenesis, and cytodiVerentiation. Dev. Biol. 234, 138–150. Settle, S. H., Jr., Rountree, R. B., Sinha, A., Thacker, A., Higgins, K., and Kingsley, D. M. (2003). Multiple joint and skeletal patterning defects caused by single and double mutations in the mouse Gdf6 and Gdf5 genes. Dev. Biol. 254, 116–130. Sha, X., Brunner, A. M., Purchio, A. F., and Gentry, L. E. (1989). Transforming growth factor beta 1: Importance of glycosylation and acidic proteases for processing and secretion. Mol. Endocrinol. 3, 1090–1098. Shawlot, W., Min, D. J., Wakamiya, M., and Behringer, R. R. (2000). The cerberus-related gene, Cerr1, is not essential for mouse head formation. Genesis 26, 253–258. Shen, M. M., and Schier, A. F. (2000). The EGF-CFC gene family in vertebrate development. Trends Genet. 16, 303–309. Sheppard, D. (2001). Integrin-mediated activation of transforming growth factor-beta(1) in pulmonary fibrosis. Chest 120, 49S–53S. Shi, W., Heisterkamp, N., GroVen, J., Zhao, J., Warburton, D., and Kaartinen, V. (1999). TGF-beta3-null mutation does not abrogate fetal lung maturation in vivo by glucocorticoids. Am. J. Physiol. 277, L1205–L1213. Shi, Y., and Massague, J. (2003). Mechanisms of TGF-beta signaling from cell membrane to the nucleus. Cell 113, 685–700. Shinagawa, T., Nomura, T., Colmenares, C., Ohira, M., Nakagawara, A., and Ishii, S. (2001). Increased susceptibility to tumorigenesis of ski-deficient heterozygous mice. Oncogene 20, 8100–8108. Shou, W., Aghdasi, B., Armstrong, D. L., Guo, Q., Bao, S., Charng, M. J., Mathews, L. M., Schneider, M. D., Hamilton, S. L., and Matzuk, M. M. (1998). Cardiac defects and altered ryanodine receptor function in mice lacking FKBP12. Nature 391, 489–492. Shuler, C. F. (1995). Programmed cell death and cell transformation in craniofacial development. Crit Rev. Oral Biol. Med. 6, 202–217. Shuler, C. F., Guo, Y., Majumder, A., and Luo, R. Y. (1991). Molecular and morphologic changes during the epithelial-mesenchymal transformation of palatal shelf medial edge epithelium in vitro. Int. J. Dev. Biol. 35, 463–472.

128

Dudas and Kaartinen

Shuler, C. F., Halpern, D. E., Guo, Y., and Sank, A. C. (1992). Medical edge epithelium fate traced by cell lineage analysis during epithelial-mesenchymal transformation in vivo. Dev. Biol. 154, 318–330. Shull, M. M., and Doetschman, T. (1994). Transforming growth factor-beta 1 in reproduction and development. Mol. Reprod. Dev. 39, 239–246. Shull, M. M., Ormsby, I., Kier, A. B., Pawlowski, S., Diebold, R. J., Yin, M., Allen, R., Sidman, C., Proetzel, G., Calvin, D., Annunziata, N., and Doetschman, T. (1992). Targeted disruption of the mouse transforming growth factor-beta 1 gene results in multifocal inflammatory disease. Nature 359, 693–699. Siegel, P. M., and Massague, J. (2003). Cytostatic and apoptotic actions of TGF-beta in homeostasis and cancer. Nat. Rev. Cancer 3, 807–821. Simeone, D. M., Pham, T., and Logsdon, C. D. (2000). Disruption of TGFbeta signaling pathways in human pancreatic cancer cells. Ann. Surg. 232, 73–80. Simpson, E. H., Johnson, D. K., Hunsicker, P., SuVolk, R., Jordan, S. A., and Jackson, I. J. (1999). The mouse Cer1 (Cerberus related or homologue) gene is not required for anterior pattern formation. Dev. Biol. 213, 202–206. Sirard, C., de la Pompa, J. L., Elia, A., Itie, A., Mirtsos, C., Cheung, A., Hahn, S., Wakeham, A., Schwartz, L., Kern, S. E., Rossant, J., and Mak, T. W. (1998). The tumor suppressor gene Smad4/Dpc4 is required for gastrulation and later for anterior development of the mouse embryo. Genes Dev. 12, 107–119. Slavkin, H. C., Hu, C. C., Sakakura, Y., Diekwisch, T., Chai, Y., Mayo, M., Bringas, P., Jr., Simmer, J., Mak, G., Sasano, Y., and David, S. (1992). Gene expression, signal transduction and tissue-specific biomineralization during mammalian tooth development. Crit Rev. Eukaryot. Gene Expr. 2, 315–329. Smith, A., and Graham, A. (2001). Restricting Bmp-4 mediated apoptosis in hindbrain neural crest. Dev. Dyn. 220, 276–283. Smith, M. M., and Coates, M. I. (1998). Evolutionary origins of the vertebrate dentition: Phylogenetic patterns and developmental evolution. Eur. J. Oral Sci. 106(Suppl), 482–500. Sohal, G. S., Ali, M. M., Ali, A. A., and Dai, D. (1999). Ventrally emigrating neural tube cells contribute to the formation of Meckel’s and quadrate cartilage. Dev. Dyn. 216, 37–44. Solloway, M. J., Dudley, A. T., BikoV, E. K., Lyons, K. M., Hogan, B. L., and Robertson, E. J. (1998). Mice lacking Bmp6 function. Dev. Genet. 22, 321–339. Solloway, M. J., and Robertson, E. J. (1999). Early embryonic lethality in Bmp5;Bmp7 double mutant mice suggests functional redundancy within the 60A subgroup. Development 126, 1753–1768. Song, J., Oh, S. P., Schrewe, H., Nomura, M., Lei, H., Okano, M., Gridley, T., and Li, E. (1999). The type II activin receptors are essential for egg cylinder growth, gastrulation, and rostral head development in mice. Dev. Biol. 213, 157–169. Song, X. J., Li, D. Q., Farley, W., Luo, L. H., Heuckeroth, R. O., Milbrandt, J., and Pflugfelder, S. C. (2003). Neurturin-deficient mice develop dry eye and keratoconjunctivitis sicca. Invest Ophthalmol. Vis. Sci. 44, 4223–4229. Sorensen, L. K., Brooke, B. S., Li, D. Y., and Urness, L. D. (2003). Loss of distinct arterial and venous boundaries in mice lacking endoglin, a vascular-specific TGFbeta coreceptor. Dev. Biol. 261, 235–250. Sporn, M. B., and Roberts, A. B. (1987). Peptide growth factors: Current status and therapeutic opportunities. In ‘‘Important Advances in Oncology’’ (V. T. Devita, Jr., S. Hellman, and S. A. Rosenberg, Eds.), Vol. 1987, pp. 75–86. Lippincott Williams & Wilkins, Philadelphia. Stanley, E. G., Biben, C., Allison, J., Hartley, L., Wicks, I. P., Campbell, I. K., McKinley, M., Barnett, L., Koentgen, F., Robb, L., and Harvey, R. P. (2000). Targeted insertion of a lacZ reporter gene into the mouse Cer1 locus reveals complex and dynamic expression during embryogenesis. Genesis 26, 259–264.

3. TGF- Signaling and Facial Development

129

Steinbrech, D. S., Mehrara, B. J., Rowe, N. M., Dudziak, M. E., Luchs, J. S., Saadeh, P. B., Gittes, G. K., and Longaker, M. T. (2000). Gene expression of TGF-beta, TGF-beta receptor, and extracellular matrix proteins during membranous bone healing in rats. Plast. Reconstr. Surg. 105, 2028–2038. Stenvers, K. L., Tursky, M. L., Harder, K. W., Kountouri, N., Amatayakul-Chantler, S., Grail, D., Small, C., Weinberg, R. A., Sizeland, A. M., and Zhu, H. J. (2003). Heart and liver defects and reduced transforming growth factor beta2 sensitivity in transforming growth factor beta type III receptor-deficient embryos. Mol. Cell. Biol. 23, 4371–4385. Storm, E. E., Huynh, T. V., Copeland, N. G., Jenkins, N. A., Kingsley, D. M., and Lee, S. J. (1994). Limb alterations in brachypodism mice due to mutations in a new member of the TGF beta-superfamily. Nature 368, 639–643. Storm, E. E., and Kingsley, D. M. (1996). Joint patterning defects caused by single and double mutations in members of the bone morphogenetic protein (BMP) family. Development 122, 3969–3979. Storm, E. E., and Kingsley, D. M. (1999). GDF5 coordinates bone and joint formation during digit development. Dev. Biol. 209, 11–27. Stottmann, R. W., Anderson, R. M., and Klingensmith, J. (2001). The BMP antagonists Chordin and Noggin have essential but redundant roles in mouse mandibular outgrowth. Dev. Biol. 240, 457–473. Sun, D., McAlmon, K. R., Davies, J. A., Bernfield, M., and Hay, E. D. (1998a). Simultaneous loss of expression of syndecan-1 and E-cadherin in the embryonic palate during epithelialmesenchymal transformation. Int. J. Dev. Biol. 42, 733–736. Sun, D., Vanderburg, C. R., Odierna, G. S., and Hay, E. D. (1998b). TGFbeta3 promotes transformation of chicken palate medial edge epithelium to mesenchyme in vitro. Development 125, 95–105. Sun, P. D., and Davies, D. R. (1995). The cystine-knot growth-factor superfamily. Annu. Rev. Biophys. Biomol. Struct. 24, 269–291. Takahara, K., Lyons, G. E., and Greenspan, D. S. (1994). Bone morphogenetic protein-1 and a mammalian tolloid homologue (mTld) are encoded by alternatively spliced transcripts which are diVerentially expressed in some tissues. J. Biol. Chem. 269, 32572–32578. Takaku, K., Oshima, M., Miyoshi, H., Matsui, M., Seldin, M. F., and Taketo, M. M. (1998). Intestinal tumorigenesis in compound mutant mice of both Dpc4 (Smad4) and Apc genes. Cell 92, 645–656. Takekawa, M., Tatebayashi, K., Itoh, F., Adachi, M., Imai, K., and Saito, H. (2002). Smaddependent GADD45beta expression mediates delayed activation of p38 MAP kinase by TGF-beta. EMBO J. 21, 6473–6482. Tam, P. P., Meier, S., and Jacobson, A. G. (1982). DiVerentiation of the metameric pattern in the embryonic axis of the mouse. II. Somitomeric organization of the presomitic mesoderm. DiVerentiation 21, 109–122. Tam, P. P., and Steiner, K. A. (1999). Anterior patterning by synergistic activity of the early gastrula organizer and the anterior germ layer tissues of the mouse embryo. Development 126, 5171–5179. Taniguchi, K., Sato, N., and Uchiyama, Y. (1995). Apoptosis and heterophagy of medial edge epithelial cells of the secondary palatine shelves during fusion. Arch. Histol. Cytol. 58, 191–203. Taya, Y., O’Kane, S., and Ferguson, M. W. (1999). Pathogenesis of cleft palate in TGF-beta3 knockout mice. Development 126, 3869–3879. ten Dijke, P., and Hill, C. S. (2004). New insights into TGF-beta-Smad signalling. Trends Biochem. Sci. 29, 265–273.

130

Dudas and Kaartinen

ten Dijke, P., Yamashita, H., Sampath, T. K., Reddi, A. H., Estevez, M., Riddle, D. L., Ichijo, H., Heldin, C. H., and Miyazono, K. (1994). Identification of type I receptors for osteogenic protein-1 and bone morphogenetic protein-4. J. Biol. Chem. 269, 16985–16988. ThesleV, I. (2003). Developmental biology and building a tooth. Quintessence. Int. 34, 613–620. ThesleV, I., and Mikkola, M. (2002). The role of growth factors in tooth development. Int. Rev. Cytol. 217, 93–135. Thisse, C., and Thisse, B. (1999). Antivin, a novel and divergent member of the TGFbeta superfamily, negatively regulates mesoderm induction. Development 126, 229–240. Tomac, A. C., Agulnick, A. D., Haughey, N., Chang, C. F., Zhang, Y., Backman, C., Morales, M., Mattson, M. P., Wang, Y., Westphal, H., and HoVer, B. J. (2002). EVects of cerebral ischemia in mice deficient in Persephin. Proc. Natl. Acad. Sci. USA 99, 9521–9526. Trainor, P., and Krumlauf, R. (2000). Plasticity in mouse neural crest cells reveals a new patterning role for cranial mesoderm. Nat. Cell Biol. 2, 96–102. Trainor, P. A., Ariza-McNaughton, L., and Krumlauf, R. (2002). Role of the isthmus and FGFs in resolving the paradox of neural crest plasticity and prepatterning. Science 295, 1288–1291. Trainor, P. A., and Krumlauf, R. (2001). Hox genes, neural crest cells and branchial arch patterning. Curr. Opin. Cell Biol. 13, 698–705. Trainor, P. A., Melton, K. R., and Manzanares, M. (2003). Origins and plasticity of neural crest cells and their roles in jaw and craniofacial evolution. Int. J. Dev. Biol. 47, 541–553. Trainor, P. A., and Tam, P. P. (1995). Cranial paraxial mesoderm and neural crest cells of the mouse embryo: Co-distribution in the craniofacial mesenchyme but distinct segregation in branchial arches. Development 121, 2569–2582. Tremblay, K. D., Dunn, N. R., and Robertson, E. J. (2001). Mouse embryos lacking Smad1 signals display defects in extra-embryonic tissues and germ cell formation. Development 128, 3609–3621. Tremblay, K. D., Hoodless, P. A., BikoV, E. K., and Robertson, E. J. (2000). Formation of the definitive endoderm in mouse is a Smad2-dependent process. Development 127, 3079–3090. Trumpp, A., Depew, M. J., Rubenstein, J. L., Bishop, J. M., and Martin, G. R. (1999). Cremediated gene inactivation demonstrates that FGF8 is required for cell survival and patterning of the first branchial arch. Genes Dev. 13, 3136–3148. Tucker, R. F., Branum, E. L., Shipley, G. D., Ryan, R. J., and Moses, H. L. (1984). Specific binding to cultured cells of 125I-labeled type beta transforming growth factor from human platelets. Proc. Natl. Acad. Sci. USA 81, 6757–6761. Tudela, C., Formoso, M. A., Martinez, T., Perez, R., Aparicio, M., Maestro, C., Del Rio, A., Martinez, E., Ferguson, M., and Martinez-Alvarez, C. (2002). TGF-beta3 is required for the adhesion and intercalation of medial edge epithelial cells during palate fusion. Int. J. Dev. Biol. 46, 333–336. Umans, L., Vermeire, L., Francis, A., Chang, H., Huylebroeck, D., and Zwijsen, A. (2003). Generation of a floxed allele of Smad5 for cre-mediated conditional knockout in the mouse. Genesis 37, 5–11. Urist, M. R. (1965). Bone: Formation by autoinduction. Science 150, 893–899. Urness, L. D., Sorensen, L. K., and Li, D. Y. (2000). Arteriovenous malformations in mice lacking activin receptor-like kinase-1. Nat. Genet. 26, 328–331. Van de Putte, K. A., and Urist, M. R. (1965). Osteogenesis in the interior of intramuscular implants of decalcified bone matrix. Clin. Orthop. 43, 257–270. Varani, S., Elvin, J. A., Yan, C., DeMayo, J., DeMayo, F. J., Horton, H. F., Byrne, M. C., and Matzuk, M. M. (2002). Knockout of pentraxin 3, a downstream target of growth diVerentiation factor-9, causes female subfertility. Mol. Endocrinol. 16, 1154–1167.

3. TGF- Signaling and Facial Development

131

Varlet, I., Collignon, J., and Robertson, E. J. (1997). Nodal expression in the primitive endoderm is required for specification of the anterior axis during mouse gastrulation. Development 124, 1033–1044. Vassalli, A., Matzuk, M. M., Gardner, H. A., Lee, K. F., and Jaenisch, R. (1994). Activin/ inhibin beta B subunit gene disruption leads to defects in eyelid development and female reproduction. Genes Dev. 8, 414–427. Vincent, S. D., Dunn, N. R., Hayashi, S., Norris, D. P., and Robertson, E. J. (2003). Cell fate decisions within the mouse organizer are governed by graded Nodal signals. Genes Dev. 17, 1646–1662. Visser, J. A., Olaso, R., Verhoef-Post, M., Kramer, P., Themmen, A. P., and Ingraham, H. A. (2001). The serine/threonine transmembrane receptor ALK2 mediates Mullerian inhibiting substance signaling. Mol. Endocrinol. 15, 936–945. Waite, K. A., and Eng, C. (2003). From developmental disorder to heritable cancer: It’s all in the BMP/TGF-beta family. Nat. Rev. Genet. 4, 763–773. Waldrip, W. R., BikoV, E. K., Hoodless, P. A., Wrana, J. L., and Robertson, E. J. (1998). Smad2 signaling in extraembryonic tissues determines anterior-posterior polarity of the early mouse embryo. Cell 92, 797–808. Wall, N. A., and Hogan, B. L. (1994). TGF-beta related genes in development. Curr. Opin. Genet. Dev. 4, 517–522. Wang, S., Wu, Z., Sorenson, C. M., Lawler, J., and Sheibani, N. (2003). Thrombospondin-1deficient mice exhibit increased vascular density during retinal vascular development and are less sensitive to hyperoxia-mediated vessel obliteration. Dev. Dyn. 228, 630–642. Ward, S. M., Desgrosellier, J. S., Zhuang, X., Barnett, J. V., and Galper, J. B. (2002). Transforming growth factor beta (TGFbeta) signaling via diVerential activation of activin receptor-like kinases 2 and 5 during cardiac development. Role in regulating parasympathetic responsiveness. J. Biol. Chem. 277, 50183–50189. Warnke, P. H., Springer, I. N. G., Wiltfang, J., Acil, Y., Eufinger, H., Wehmoller, M., Russo, P. A. J., Bolte, H., Sherry, E., Behrens, E., and Terheyden, H. (2004). Growth and transplantation of a custom vascularised bone graft in a man. Lancet 364, 766–770. Warren, S. M., Brunet, L. J., Harland, R. M., Economides, A. N., and Longaker, M. T. (2003). The BMP antagonist noggin regulates cranial suture fusion. Nature 422, 625–629. Wawersik, S., Purcell, P., Rauchman, M., Dudley, A. T., Robertson, E. J., and Maas, R. (1999). BMP7 acts in murine lens placode development. Dev. Biol. 207, 176–188. Weber, F. E., Schmokel, H., Oelgeschlager, M., Nickel, J., Maly, F. E., Hortschansky, P., and Gratz, K. W. (2003). Deletion mutants of BMP folding variants act as BMP antagonists and are eYcient inhibitors for heterotopic ossification. J. Bone Miner. Res. 18, 2142–2151. Weinstein, M., Yang, X., and Deng, C. (2000). Functions of mammalian Smad genes as revealed by targeted gene disruption in mice. Cytokine Growth Factor Rev. 11, 49–58. Weinstein, M., Yang, X., Li, C., Xu, X., Gotay, J., and Deng, C. X. (1998). Failure of egg cylinder elongation and mesoderm induction in mouse embryos lacking the tumor suppressor smad2. Proc. Natl. Acad. Sci. USA 95, 9378–9383. Whitman, M. (2001). Nodal signaling in early vertebrate embryos: Themes and variations. Dev. Cell 1, 605–617. Wiater, E., and Vale, W. (2003). Inhibin is an antagonist of bone morphogenetic protein signaling. J. Biol. Chem. 278, 7934–7941. Winnier, G., Blessing, M., Labosky, P. A., and Hogan, B. L. (1995). Bone morphogenetic protein-4 is required for mesoderm formation and patterning in the mouse. Genes Dev. 9, 2105–2116. Wrana, J. L., Attisano, L., Carcamo, J., Zentella, A., Doody, J., Laiho, M., Wang, X. F., and Massague, J. (1992). TGF beta signals through a heteromeric protein kinase receptor complex. Cell 71, 1003–1014.

132

Dudas and Kaartinen

Wu, X. B., Li, Y., Schneider, A., Yu, W., Rajendren, G., Iqbal, J., Yamamoto, M., Alam, M., Brunet, L. J., Blair, H. C., Zaidi, M., and Abe, E. (2003). Impaired osteoblastic diVerentiation, reduced bone formation, and severe osteoporosis in noggin-overexpressing mice. J. Clin. Invest. 112, 924–934. Xu, J., and Attisano, L. (2000). Mutations in the tumor suppressors Smad2 and Smad4 inactivate transforming growth factor beta signaling by targeting Smads to the ubiquitinproteasome pathway. Proc. Natl. Acad. Sci. USA 97, 4820–4825. Xu, T., Bianco, P., Fisher, L. W., Longenecker, G., Smith, E., Goldstein, S., Bonadio, J., Boskey, A., Heegaard, A. M., Sommer, B., Satomura, K., Dominguez, P., Zhao, C., Kulkarni, A. B., Robey, P. G., and Young, M. F. (1998). Targeted disruption of the biglycan gene leads to an osteoporosis-like phenotype in mice. Nat. Genet. 20, 78–82. Xu, X., Brodie, S. G., Yang, X., Im, Y. H., Parks, W. T., Chen, L., Zhou, Y. X., Weinstein, M., Kim, S. J., and Deng, C. X. (2000). Haploid loss of the tumor suppressor Smad4/Dpc4 initiates gastric polyposis and cancer in mice. Oncogene 19, 1868–1874. Xu, X., Jeong, L., Han, J., Ito, Y., Bringas, P., Jr., and Chai, Y. (2003). Developmental expression of Smad1-7 suggests critical function of TGF-beta/BMP signaling in regulating epithelial-mesenchymal interaction during tooth morphogenesis. Int. J. Dev. Biol. 47, 31–39. Yamashita, H., ten Dijke, P., Huylebroeck, D., Sampath, T. K., Andries, M., Smith, J. C., Heldin, C. H., and Miyazono, K. (1995). Osteogenic protein-1 binds to activin type II receptors and induces certain activin-like eVects. J. Cell Biol. 130, 217–226. Yan, C., Wang, P., DeMayo, J., DeMayo, F. J., Elvin, J. A., Carino, C., Prasad, S. V., Skinner, S. S., Dunbar, B. S., Dube, J. L., Celeste, A. J., and Matzuk, M. M. (2001). Synergistic roles of bone morphogenetic protein 15 and growth diVerentiation factor 9 in ovarian function. Mol. Endocrinol. 15, 854–866. Yang, X., Castilla, L. H., Xu, X., Li, C., Gotay, J., Weinstein, M., Liu, P. P., and Deng, C. X. (1999a). Angiogenesis defects and mesenchymal apoptosis in mice lacking SMAD5. Development 126, 1571–1580. Yang, X., Letterio, J. J., Lechleider, R. J., Chen, L., Hayman, R., Gu, H., Roberts, A. B., and Deng, C. (1999b). Targeted disruption of SMAD3 results in impaired mucosal immunity and diminished T cell responsiveness to TGF-beta. EMBO J. 18, 1280–1291. Yang, X., Li, C., Xu, X., and Deng, C. (1998). The tumor suppressor SMAD4/DPC4 is essential for epiblast proliferation and mesoderm induction in mice. Proc. Natl. Acad. Sci. USA 95, 3667–3672. Yang, Y. A., Tang, B., Robinson, G., Hennighausen, L., Brodie, S. G., Deng, C. X., and Wakefield, L. M. (2002). Smad3 in the mammary epithelium has a nonredundant role in the induction of apoptosis, but not in the regulation of proliferation or diVerentiation by transforming growth factor-beta. Cell Growth DiVer. 13, 123–130. Yao, T. P., Oh, S. P., Fuchs, M., Zhou, N. D., Ch’ng, L. E., Newsome, D., Bronson, R. T., Li, E., Livingston, D. M., and Eckner, R. (1998). Gene dosage-dependent embryonic development and proliferation defects in mice lacking the transcriptional integrator p300. Cell 93, 361–372. Yeo, C., and Whitman, M. (2001). Nodal signals to Smads through Cripto-dependent and Cripto-independent mechanisms. Mol. Cell 7, 949–957. Yi, S. E., Daluiski, A., Pederson, R., Rosen, V., and Lyons, K. M. (2000). The type I BMP receptor BMPRIB is required for chondrogenesis in the mouse limb. Development 127, 621–630. Yi, S. E., LaPolt, P. S., Yoon, B. S., Chen, J. Y., Lu, J. K., and Lyons, K. M. (2001). The type I BMP receptor BmprIB is essential for female reproductive function. Proc. Natl. Acad. Sci. USA 98, 7994–7999.

3. TGF- Signaling and Facial Development

133

Ying, Y., Liu, X. M., Marble, A., Lawson, K. A., and Zhao, G. Q. (2000). Requirement of Bmp8b for the generation of primordial germ cells in the mouse. Mol. Endocrinol. 14, 1053–1063. Ying, Y., and Zhao, G. Q. (2001). Cooperation of endoderm-derived BMP2 and extraembryonic ectoderm-derived BMP4 in primordial germ cell generation in the mouse. Dev. Biol. 232, 484–492. Yoshida, H., Kong, Y. Y., Yoshida, R., Elia, A. J., Hakem, A., Hakem, R., Penninger, J. M., and Mak, T. W. (1998). Apaf1 is required for mitochondrial pathways of apoptosis and brain development. Cell 94, 739–750. Yoshida, Y., Tanaka, S., Umemori, H., Minowa, O., Usui, M., Ikematsu, N., Hosoda, E., Imamura, T., Kuno, J., Yamashita, T., Miyazono, K., Noda, M., Noda, T., and Yamamoto, T. (2000). Negative regulation of BMP/Smad signaling by Tob in osteoblasts. Cell 103, 1085–1097. Young, M. F., Bi, Y., Ameye, L., and Chen, X. D. (2002). Biglycan knockout mice: New models for musculoskeletal diseases. Glycoconj. J. 19, 257–262. Zhang, H., and Bradley, A. (1996). Mice deficient for BMP2 are nonviable and have defects in amnion/chorion and cardiac development. Development 122, 2977–2986. Zhang, J., Talbot, W. S., and Schier, A. F. (1998). Positional cloning identifies zebrafish oneeyed pinhead as a permissive EGF-related ligand required during gastrulation. Cell 92, 241–251. Zhang, Y., Wang, S., Song, Y., Han, J., Chai, Y., and Chen, Y. (2003). Timing of odontogenic neural crest cell migration and tooth-forming capability in mice. Dev. Dyn. 226, 713–718. Zhang, Z., Song, Y., Zhao, X., Zhang, X., Fermin, C., and Chen, Y. (2002). Rescue of cleft palate in Msx1-deficient mice by transgenic Bmp4 reveals a network of BMP and Shh signaling in the regulation of mammalian palatogenesis. Development 129, 4135–4146. Zhao, G. Q. (2003). Consequences of knocking out BMP signaling in the mouse. Genesis 35, 43–56. Zhao, G. Q., Chen, Y. X., Liu, X. M., Xu, Z., and Qi, X. (2001). Mutation in Bmp7 exacerbates the phenotype of Bmp8a mutants in spermatogenesis and epididymis. Dev. Biol. 240, 212–222. Zhao, G. Q., Deng, K., Labosky, P. A., Liaw, L., and Hogan, B. L. (1996). The gene encoding bone morphogenetic protein 8B is required for the initiation and maintenance of spermatogenesis in the mouse. Genes Dev. 10, 1657–1669. Zhao, G. Q., and Hogan, B. L. (1996). Evidence that mouse Bmp8a (Op2) and Bmp8b are duplicated genes that play a role in spermatogenesis and placental development. Mech. Dev. 57, 159–168. Zhao, G. Q., Liaw, L., and Hogan, B. L. (1998). Bone morphogenetic protein 8A plays a role in the maintenance of spermatogenesis and the integrity of the epididymis. Development 125, 1103–1112. Zhao, R., Lawler, A. M., and Lee, S. J. (1999). Characterization of GDF-10 expression patterns and null mice. Dev. Biol. 212, 68–79. Zhou, X., Sasaki, H., Lowe, L., Hogan, B. L., and Kuehn, M. R. (1993). Nodal is a novel TGF-beta-like gene expressed in the mouse node during gastrulation. Nature 361, 543–547. Zhu, Y., Richardson, J. A., Parada, L. F., and GraV, J. M. (1998). Smad3 mutant mice develop metastatic colorectal cancer. Cell 94, 703–714. Zoltewicz, J. S., and Gerhart, J. C. (1997). The Spemann organizer of Xenopus is patterned along its anteroposterior axis at the earliest gastrula stage. Dev. Biol. 192, 482–491.

4

____________________________________________________________________________

The Colors of Autumn Leaves as Symptoms of Cellular Recycling and Defenses Against Environmental Stresses Helen J. Ougham, Phillip Morris, and Howard Thomas Institute of Grassland and Environmental Research Plas Gogerddan, Aberystwyth, Ceredigion, SY23 3EB, Wales, United Kingdom

I. Introduction II. The Role of Chlorophyll in Protein Recycling A. The Biochemistry of Chlorophyll Degradation in Senescing Leaves B. Genes and Genetic Variation for Chlorophyll Degradation C. Chlorophyll as a Regulator of Protein Metabolism in Senescing Cells D. Senescence in Relation to Programmed Death of Green Plant Cells III. Non-Green Pigments in Senescing Leaves A. Revelation of Autumn Colors B. Carotenoids C. Anthocyanins and Other Flavonoids IV. Pigments and Stress Defenses in Senescing Leaves A. Color Changes in Senescence as Signals B. Is Autumn Color a Costly Signal? C. Possible Functions of Leaf Color D. Does Dishonesty Pay? E. Insect Preference for Green Leaves F. Visual and Olfactory Signals V. Conclusions Acknowledgments References

The color changes that occur during foliar senescence are directly related to the regulation of nutrient mobilization and resorption from leaf cells, often under conditions of biotic and abiotic stress. Chlorophyll is degraded through a metabolic pathway that becomes specifically activated in senescence. Chlorophyll catabolic enzymes and genes have been identified and characterized and aspects of their regulation analyzed. Particular genetic interventions in the pathway lead to disruptions in protein mobilization and increased sensitivity to light-dependent cell damage and death. The chemistry and metabolism of carotenoid and anthocyanin pigments in senescing leaves are considered. Bright autumn colors observed in the foliage of some woody species have been hypothesized to act as a defense signal to potential insect herbivores. Critical consideration of the biochemical and physiological Current Topics in Developmental Biology, Vol. 66 Copyright 2005, Elsevier Inc. All rights reserved. 0070-2153/05 $35.00

135

136

Ougham et al.

features of normal leaf senescence leads to the conclusion that accumulating or unmasking compounds with new colors are unlikely to represent a costly investment on the part of the tree. The influences of human evolutionary and social history on our own perception of autumn coloration are discussed. The possibility that insect herbivores may respond to volatiles emitted during leaf senescence, rather than to bright colors, is also presented. Finally, some new approaches to the analysis of protein recycling in senescence are briefly considered. C 2005, Elsevier Inc.

I. Introduction This discussion considers the significance of the striking changes in pigmentation that occur when green plant tissues undergo senescence. The metabolic events underlying the highly visible symptoms of senescence are directly concerned with the functional and structural transdiVerentiation (Thomas et al., 2003) of cells, from units with a primary assimilation role into centers of nutrient mobilization and recovery. Plants, as sessile organisms, experience nonoptimal environments as a way of life. A period of remodeling, such as occurs in senescence, is a potentially vulnerable time for tissues, organs, and the whole plant, and this is reflected in physiological changes accompanying the developing nutrient-recycling function that serve to defend against the intrusion of abiotic and biotic challenges. Here again, pigments are diagnostic of defenses against stress. In this review, we focus on two specific aspects of recycling and stress resistance: recent developments in understanding the molecular and cellular control of chlorophyll degradation, and autumn colors as potential signals in biotic interactions between plants and animals.

II. The Role of Chlorophyll in Protein Recycling A. The Biochemistry of Chlorophyll Degradation in Senescing Leaves Protein mobilization in senescence is regulated by a network of processes (Dangl et al., 2000; Ho¨ rtensteiner and Feller, 2002; Thomas and Donnison, 2000), among which the induction of chlorophyll degradation is an early and, for plastid membrane polypeptides, essential event (Thomas et al., 2002). Net loss of chlorophyll from green tissues during senescence and other terminal developmental events culminates in the accumulation of colorless products (nonfluorescent chlorophyll catabolites, or NCCs) (Mu¨ hlecker and Kra¨ utler, 1996). The enzymic pathway of NCC formation from chlorophyll (Ho¨ rtensteiner, 2004) commences with chlorophyllase, which dephytylates

4. Leaf Senescence, Nutrient Recycling, and Stress Defenses

137

chlorophyll a. Magnesium (Mg) is removed from chlorophyllide a by a dechelatase activity. The tetrapyrrole macrocycle of the product of Mg removal is opened oxygenolytically by pheophorbide a oxygenase (PaO), producing a red bilin, RCC. A reductase immediately converts RCC into a colorless fluorescent product, FCC. Further enzymic and nonenzymic reactions metabolize FCC to NCCs in a species-specific manner (Ho¨rtensteiner and Feller, 2002; Thomas et al., 2001). Catabolites of chlorophyll b are not normally observed in senescing tissues, leading to the notion that there is interconversion between chlorophyll(ide) a and b and catabolism exclusively by the a-specific pathway. An enzymic activity capable of converting chlorophyllide b to a has been shown to become elevated during senescence (Scheumann et al., 1999). Terminal catabolites are sequestered in the cell vacuole. There is no evidence that the N of the chlorophyll ring is exported from the cell during senescence. Chlorophyll catabolism is summarized in Fig. 1.

B. Genes and Genetic Variation for Chlorophyll Degradation Genes for most of the steps in the chlorophyll catabolism pathway have been cloned (Gray et al., 2002; Jakob-Wilk et al., 1999; Pruzˇ inska´ et al., 2003; Tanaka et al., 2003; Tommasini et al., 1998; Tsuchiya et al., 1999; Wu¨ thrich et al., 2000). A number of mutations, genetic variants, and transgenics modifying chlorophyll catabolism have been described (Pruzˇ inska´ et al., 2003; Thomas and Howarth, 2000; Thomas et al., 2001). In general, they fall into two main categories: 1. Stay-greens are genetic variants in which the yellowing of senescing leaves is delayed, or slowed, or both, relative to comparable normal genotypes. Stay-greens have been diVerentiated in turn into two kinds—functional and cosmetic (Thomas and Smart, 1993). In functional stay-greens, the link between enhanced stability of chlorophyll, retention of photosynthetic capacity, and delayed protein mobilization is maintained. In cosmetic stay-greens, yellowing is disabled but photosynthetic rate usually declines over a similar time-course to normally senescing yellowing genotypes, and there is partial stabilization of protein. Section II.C discusses this further. 2. Photosensitive genotypes. As described in more detail in Section II.D, variants with deficiencies in particular steps of tetrapyrrole metabolism display pathological symptoms that often mimic disease lesions and are consistent with the accumulation of photodynamic intermediates upstream of the metabolic blockage (Ho¨ rtensteiner, 2004). Interestingly, in those cosmetic stay-greens in which the location of the metabolic

Figure 1 The chlorophyll degradation pathway. 1, Chlorophyllase; 2, magnesium dechelatase; 3, pheophorbide a oxygenase; 4, RCC reductase; 5, species-specific enzymic and nonenzymic conversion of FCC to nonfluorescent terminal catabolites.

4. Leaf Senescence, Nutrient Recycling, and Stress Defenses

139

deficiency has been studied, the evidence points to disruption of one or the other step in the chlorophyll catabolic pathway but no corresponding increase in photosensitivity (Bachmann et al., 1994; Roca et al., 2004). Conversely, and contrary to expectation, photosensitive chlorophyll catabolism variants seem not to behave as cosmetic stay-greens when induced to senesce in the absence of light (Tanaka et al., 2003).

C. Chlorophyll as a Regulator of Protein Metabolism in Senescing Cells As long as the nitrogen requirements of growing tissues and organs can be met by uptake from the soil and assimilation in roots and leaves, senescence does not usually make a major contribution to the plant’s internal nitrogen cycle. However, if sink demand cannot be met by current assimilation— as may happen when development switches from the vegetative to reproductive phase, for example—nitrogen reserves become remobilized. First, lowmolecular-weight sources, such as free amino acids and vacuolar nitrate, are drained from the system, then polymers begin to be catabolized. Chloroplasts are protein storage bodies as well as photosynthetic organelles. The onset of senescence marks the functional transition of plastids from assimilation to remobilization, of which chlorophyll catabolism is the visible symptom. Yellowing and protein nitrogen remobilization are generally quite well correlated (Thomas et al., 2002). Genetic and environmental factors that interfere with chlorophyll degradation during senescence also modify protein degradation. For example, a mutant gene that confers cosmetic staygreenness in Festuca and Lolium species has the eVect of stabilizing chloroplast membrane proteins during senescence (Roca et al., 2004; Thomas et al., 2002). On the evidence of mutants and in vitro reconstitution experiments, pigment-binding proteins must be properly complexed with chlorophyll if they are to fold correctly, otherwise they are vulnerable to proteolytic attack (Thomas, 1997). Because pigment proteolipids have both a photosynthetic function and a role in thylakoid structure (Allen and Forsberg, 2001), stabilizing chlorophyll–protein complexes in senescence confers durability on chloroplast membranes and reduces the lability of membrane-associated components that are not themselves directly stabilized by chlorophyll, such as cytochrome f (Bachmann et al., 1994; Davies et al., 1990). Conversely, an analysis of the behavior of the light-harvesting chlorophyll complex of photosystem 2 during senescence of a stay-green Festuca mutant revealed that part of an otherwise deeply-buried thylakoid intrinsic protein that extends into the stroma may be subject to proteolysis, just like soluble-phase plastid proteins (Thomas and Howarth, 2000).

140

Ougham et al.

D. Senescence in Relation to Programmed Death of Green Plant Cells In the present discussion, the term senescence is used in the specialized context of the controlled recovery of nutrients from green tissues and is associated with the transdiVerentiation of cells and organelles from centers of primary photoassimilation into remobilizing storage structures. Senescence, as the name implies, usually occurs at the end of the life of the leaf and is often classified as an aspect of programmed cell death (Dangl et al., 2000; van Doorn and Woltering, 2004). Nevertheless, senescence in the context of green cell transdiVerentiation has features that distinguish it from other cell death processes in a fundamental way. Particularly significant is its reversibility. In many, perhaps most, species it is possible to induce regreening of senescent leaves by interfering with source-sink or hormonal status, or both. Zavaleta-Mancera et al. (1999a,b) showed that during regreening of tobacco leaves, gerontoplasts were rediVerentiated into chloroplasts, senescence-enhanced genes and their products were turned oV, and components of the plastid assembly machinery were reactivated. Thomas et al. (2003) argued that the reversibility of senescence, among other characteristics, classifies the process as a diVerentiation event and not an aspect of programmed cell death. In some ways, it is unfortunate that history has left us with the term senescence to describe a phenomenon that, mechanistically, is better understood in developmental rather than deteriorative terms. It may be significant that scientists in the field of animal aging and cell death have moved away from employing the word senescence in recent years because of its imprecise, ambiguous associations (Gordon Lithgow, personal communication). We suggest that we accept the inappropriate etymology and choose to define senescence in our own pragmatic way (in the words of Humpty Dumpty, ‘‘When I use a word . . . it means just what I choose it to mean— neither more nor less.’’) (Carroll, 1872), thereby avoiding the distractions of semantics (van Doorn and Woltering, 2004). If senescence is accepted as being functionally distinct from, rather than a form of, programmed cell death, a fruitful area of study opens up concerned with the relationship between the two phenomena in plant development and survival. Ho¨ rtensteiner (2004) has expressed this most dramatically in terms of the obligate requirement for correct expression of the senescence syndrome to avoid the pathological consequences of cell death. In other words, programmed senescence and (programmed) cell death are mutually antagonistic. This has become increasingly clear from recent studies in which the molecular genetics of lesion formation and programmed cell death has converged with the identification and cloning of genes for the enzymes of the chlorophyll catabolism pathway. Mach et al. (2001) cloned the ACD2 (accelerated cell death 2) locus of Arabidopsis and found it to be identical with the gene encoding the chlorophyll catabolic enzyme RCC reductase

4. Leaf Senescence, Nutrient Recycling, and Stress Defenses

141

(Wu¨ thrich et al., 2000). The knockout has a phenotype that takes the form of light-dependent spreading lesions. Subsequently, the ACD1 (accelerated cell death 1) gene of Arabidopsis, a mutation of which also gives a photosensitive cell death phenotype, was shown to encode PaO, the enzyme that opens the chlorophyll macrocycle during senescence (Pruzˇ inska´ et al., 2003; Tanaka et al., 2003). Abnormalities of tetrapyrrole metabolism are well known to lead to pathological photosensitivity, and not just in plants (Ho¨ rtensteiner, 2004). There remains the enigma of the contrasting senescence and lightresponse behavior of photosensitive mutants and cosmetic stay-greens as described earlier in Section II.B. It will be necessary to understand in much greater detail the mechanisms by which senescing leaves respond to light and other abiotic stresses, and the control mechanisms by which these stresses are resisted or avoided, before this paradox can be resolved.

III. Non-Green Pigments in Senescing Leaves A. Revelation of Autumn Colors Removal of chlorophyll is a defining feature of leaf senescence in all higher plant species. In contrast, the coloration remaining, after the chlorophyll has been catabolized and before tissue death, is much more variable, depending on both genetic background—diVerences can be found within as well as between species—and environmental factors, particularly stresses due to low temperature, high light, drought, and so on. Optical brighteners synthesized in senescing leaves can enhance the color contributed by other pigments; a striking example is the compound 6-hydroxykynurenic acid, which, by reinforcing carotenoid coloration, imparts the brilliant golden shade characteristic of senescent Ginkgo biloba leaves (Matile, 1994). However, the range of leaf colors from yellow through orange to red, pink, and purple is mainly due to two classes of compound: carotenoids and anthocyanins (Matile, 2000).

B. Carotenoids Leaf carotenoids are highly hydrophobic, and most are yellow or orange in color. The most abundant carotenoids in the chloroplast of a green leaf are typically beta-carotene and alpha-carotene, and the xanthophylls (oxygenated carotenoids) violaxanthin, neoxanthin, antheraxanthin, zeaxanthin, and lutein. The proportions vary with species, leaf age, and environmental conditions (reviewed in Biswal, 1995). In green mesophyll cells in the light, carotenoids function as accessory pigments in the photosynthetic apparatus

142

Ougham et al.

and as protectants against photooxidative damage. During leaf senescence, disappearance of the chlorophyll reveals these previously masked compounds. Thus, in the many plant species that do not accumulate hydrophilic pigments such as anthocyanins, the senescing leaf appears yellow or orangish. Although red carotenoids do exist, they are more common in reproductive and dispersal structures, the best-known example being lycopene, which is synthesized in solanaceous fruits as they ripen. One unusual example of red carotenoids in tree leaves was observed by Hormaetxe et al. (2004), who found eschscholtzxanthin and derivatives in the foliage of box trees under photoinhibitory conditions encountered during winter acclimation. Like senescent chloroplasts (see Section II.D), and in contrast to fruit chromoplasts, the red plastids in these leaves are able to rediVerentiate to green chloroplasts when environmental conditions change; unlike many of the phenylpropanoid compounds described in the next section, the red carotenoids are not terminal metabolites. In general, remobilization of carotenoids during leaf senescence does not appear to be a consistently high priority for the plant. Being composed almost entirely of carbon and hydrogen (Fig. 2), they do not contain elements that are in short supply at this stage. Chlorophyll and its derivatives, the breakdown products of which are also not exported from the senescing leaf, present a potential hazard to leaf cells as photosynthesis declines and are therefore catabolized by the detoxification process described in Section II, but carotenoids represent no such threat. Rather, as antioxidants they may have a role to play in protecting against photooxidative damage during a vulnerable phase of the life of the leaf. For example, in leaves of the mastic tree Pistacia lentiscus, lutein and neaxanthin levels remained constant during the early stages of senescence (up to 20% chlorophyll loss) while beta-carotene levels increased by 9%; only once chlorophyll had largely disappeared did carotenoid levels decline (neoxanthin by approximately 20%; lutein and betacarotene by approximately 35%) (Munne-Bosch and Penuelas, 2003). Taken in conjunction with similar behavior by other antioxidant compounds, including alpha-tocopherol and ascorbate, the authors inferred a photoprotective role for carotenoids during the chlorophyll catabolism phase of leaf senescence. Similarly, Merzlyak and Gitelson (1995) considered that the retention of carotenoids responsible for the intense yellow color of Acer platanoides leaves in autumn was required for protection against blue light irradiation. Whatever the extent of their participation in protection against light damage, it is certainly the case that in many plant species, ranging from beech trees (Garcia-Plazaola and Becerril, 2001) to temperate grasses (Biswal et al., 1994), the disappearance of carotenoids is retarded relative to that of chlorophylls during leaf senescence (reviewed in Biswal, 1995). In green leaves, the carotenoids are almost exclusively localized in the thylakoid membranes, where they form part of the photosynthetic pigment–protein

4. Leaf Senescence, Nutrient Recycling, and Stress Defenses

Figure 2 leaves.

143

Some of the carotenoid pigments that contribute to autumn colors in senescing

complexes. During senescence, they relocate mainly to the lipid-rich spherical bodies known as plastoglobuli, which are a striking feature of the plastids in senescent tissues (Steinmu¨ller and Tevini, 1985; Tevini and Steinmu¨ ller, 1985). In this respect, senescing leaf chloroplasts resemble the specialized chromoplasts of flower petals and fruit, in which plastoglobuli also often contain the carotenoid pigments. In addition to the change in cellular compartmentation, the complement of carotenoid compounds also often alters during leaf senescence. Frequently, the proportion of esterified carotenoids increases at the expense of unesterified forms (Biswal et al., 1994; Garcia-Plazaola and Becerril, 2001; Tevini and Steinmuller, 1985; Young et al., 1991). Some or all of the xanthophylls zeaxanthin, violaxanthin, antheraxanthin, and lutein often become more abundant relative to other carotenoids (Afitlhile et al., 1993; Garcia-Plazaola

144

Ougham et al.

et al., 2003). Both esterification and changes in relative proportions of these pigments can result in subtle alterations to the yellow-orange coloration of leaves during senescence. Carotenoids, therefore, can undergo degradation, relocation, and chemical modification during leaf senescence, but there is little evidence for de novo synthesis of carotenoids in senescing leaves; genes encoding enzymes of the carotenoid biosynthesis pathway, such as geranyl-geranyl pyrophosphate synthase, phytoene synthase, and phytoene desaturase, are not reported among the wide range of genes whose expression is upregulated during leaf senescence (Andersson et al., 2004; Buchanan-Wollaston et al., 2003). It appears, therefore, that any contribution that carotenoids make to the colors of senescent leaves depends mainly on their preexistence in those leaves before the onset of senescence.

C. Anthocyanins and Other Flavonoids The major classes of flavonoid polyphenols contributing to the color of flowers, leaves, and fruits are the anthocyanins, flavonols, chalcones, and aurones (Fig. 3). The anthocyanins are the most widespread and recognizable group of plant pigments after chlorophyll, because these water-soluble compounds are responsible for nearly all of the red, pink, mauve, violet, blue, and purple colors in the petals, leaves, stems, and fruits of plants. Anthocyanins are present in nature as heterosides whose aglycone (or anthocyanidin) is a derivative of the flavylium ion. This combination with sugars is important in the case of flower pigments in providing solubility and stability to light and may be important in leaf development as a way of keeping potentially toxic metabolites in an inactive form within the cell. The three most common anthocyanidins are cyanidin (magenta), pelargonidin (orange-red, with one less hydroxyl group than cyanidin), and delphinidin (purple-blue, with one more hydroxyl than cyanidin). Ji et al. (1992) surveyed the leaf pigments of 119 taxa within the aceraceae and found that cyanidin and delphinidin glycosides accounted for most of the anthocyanins in the highly-colored autumnal foliage of these species. Cyanidin, pelagonidin, and delphinidin correspond to the three main flavonols (quercetin, kaempferol, and myricetin) in order of increasing B-ring hydroxylation. Three anthocyanidin methyl esters are also quite common, peonidin derived from cyanidin and petunidin and malvidin derived from delphinidin. Each of these anthocyanidins occurs in plants with various sugar attachments as a range of O-glycosides (anthocyanins) rather than as the aglycones (anthocyanidins). The main variation is in the type of sugar (glucose, galactose, rhamnose, xylose, or arabinose), the number of sugars (mono, di, or triglycosides) and the attachment of the sugar (usually at the 3 OH or the

4. Leaf Senescence, Nutrient Recycling, and Stress Defenses

145

Figure 3 Generalized biosynthetic pathway of flavonoids leading to anthocyanin and related pigments in plants. Key enzymes: CHI, chalcone isomerase; CHS, chalcone synthase; DFR, dihydroflavonol reductase; F3OH, flavanone-3-hydroxylase; PAL, phenylalanine ammonia lyase.

146

Ougham et al.

3-and 5-OH). Anthocyanins with acyl attachments are also quite common in some genera such as the solanaceae. The anthocyanins are therefore all based on the single aromatic structure of cyanidin (see Fig. 3), and their structural diversity is derived by modification of hydroxyl groups either by methylation or by glycosylation, or by acylation with diVerent organic acids. These compounds are nearly universal in vascular plants except for the centrospermae, where they are replaced by a chemically distinct group of pigments, the betacyanins, derived from the amino acid l-dopa, and typified by the major pigment of beetroot (Beta vulgaris). In summary, anthocyanin biosynthesis proceeds via p-coumaroyl-CoA, derived from l-phenylalanine in general phenylpropanoid metabolism, which enters a condensation reaction with three molecules of malonyl-CoA to form a C15 tetrahydroxychalcone intermediate. In the subsequent flavonoid pathway (see Fig. 3), this cyclizes to the corresponding flavanone and is hydroxylated and then reduced to produce the flavan-3-4-cis diol precursors of anthocyanins and also of condensed tannins. The flavan3-4-cis diols then undergo dehydration to form the flavylium cation, although the details of the latter steps of anthocyanin biosynthesis are still incomplete. Important regulatory steps in the metabolic sequence leading to anthocyanidin synthesis occur at phenylalanine ammonia lyase, chalcone synthase, and dihydroflavonol reductase. In general, these activities, and transcription of the genes that encode them, are sustained or even increased with organ age, but there seems to be no clear obligate relationship with leaf senescence. For example, Romero-Puertas and Delledonne (2003) have described how nitrous oxide delays leaf senescence but also activates the expression of phenylalanine ammonia lyase and chalcone synthase genes as part of a disease resistance/cell death mechanism. On the other hand, Kannangara and Hansson (1998) observed an interaction between anthocyanin and chlorophyll metabolism in young Euphorbia leaves, where several enzymes of chlorophyll biosynthesis sharply decreased in abundance at the onset of red anthocyanin accumulation. It would be interesting to know if a comparable metabolic relationship exists to enhance chlorophyll catabolism in leaves that turn red during senescence. In considering the metabolic costs of biosynthesising anthocyanins de novo, it should be kept in mind that the fluxes through the phenylpropanoid pathway do not need to be particularly large to result in a significant observable change. For example, even in Acer rubrum, which had the most intensely colored leaves analyzed by Lee et al. (2003), anthocyanins accounted for less than 6 g cm
2; by contrast, chlorophyll in presenescent leaves of this species amounted to more than five times this value. The final step in anthocyanin biosynthesis is glycosylation, which has the eVect of stabilizing the molecule. Further modifications then include additional methylations, glycosylations, and acylations of hydroxyl groups to

4. Leaf Senescence, Nutrient Recycling, and Stress Defenses

147

produce the great range of anthocyanin colours present in the plant kingdom. One of the largest monomeric anthocyanins known is the heavenly blue complex from the petals of Ipomoea tricolor (morning glory) and Commelina (Hondo et al., 1992). This has a molecular weight for the flavylium cation of 1759 (Goto et al., 1987), comprises peonidin with six molecules of glucose and three molecules of caVeic acid, and serves to illustrate the complexity that can arise in naturally occurring anthocyanins. Anthocyanins are mainly found in flower petals and in developing fruits, where they impart a broad spectrum of colors to these tissues. However, they may also accumulate in roots, leaves, bracts, seeds, and stems of both developing seedlings and mature plants. Anthocyanin accumulation is very sensitive to climatic conditions and is often associated with stress responses, particularly to low temperatures, and this is generally under tight genetic control, often mediated by MYB domain and basic helix-loop-helix transcription factors (Koes et al., 1994; Weisshaar and Jenkins, 1998). Anthocyanins accumulate in the cell vacuole (Alfenito et al., 1998), within which they are often located in spherical organelles known as anthocyanoplasts (Pecket and Small, 1980). There is a suggestive parallel here with aspects of chlorophyll metabolism in senescence, where conjugation and translocation of glycosides across the tonoplast is the ultimate metabolic fate, followed in some cases by nonenzymic chemical modifications within the acid milieu of the vacuolar sap (Thomas et al., 2001). The tissue distribution of anthocyanins in the autumnal foliage of a range of woody species has been comprehensively surveyed by Lee et al. (2003). With a few exceptions, in which the pigment was partly or entirely confined to cells of the adaxial epidermis (e.g., Acer spicatum, Euonymus atropurporeus, some Prunus spp.), anthocyanin concentrates in palisade parenchyma cells. This observation, taken with the complete absence of pigmentation from lower layers of mesophyll and the abaxial epidermis, is consistent with a function in light interception. Anthocyanins are compounds that readily alter their structures and, hence, color through the action of diVerent agents. The stability of anthocyanins increases with the number of methoxyl groups on the B ring and decreases with decreasing hydroxylation, and in general they are less labile at acid pH. In aqueous solution, anthocyanins are found equilibrated in four basic structures (the flavylium cation, quinonoidal base, carbinol base, and chalcone base), and the proportions of these forms (and hence the color) are determined principally by pH, with the red flavylium ion predominating in acid solution (Strack and Wray, 1989). At higher pH the color changes to anthocyanic forms that may be colored (bluish in the case of the quinonoidal base and yellow to orange for the chalcone base) or colorless (in the case of the carbinol base), depending on whether the A and B rings are conjugated. Hence, the relative amounts of the structural forms that coexist in equilibrium is a function of pH and the extent of addition of functional groups to the

148

Ougham et al.

basic anthocyanin structure. In acid solution, the colors range from orangered (pelargonidin) through magenta (cyanidin) to mauve (delphinidin). If the pH is raised to near pH 7.0, solutions become colorless due to the formation of the pseudobase, and above pH 7.0 the bluer anhydrobases are formed, whereas at very high pH irreversible changes occur following ionization of the phenolic hydroxyls. Age-related acidification of the vacuole may well account for intensification of the red color of preexisting blue or leuko molecular species. Although glycosylation of anthocyanidins at C3 to produce anthocyanins results in a marked shift in color, the amount of anthocyanin present in the tissue, which can vary from 0.01% to 15.0% of dry weight, has a much more marked eVect. For example, in normal blue cornflowers the anthocyanin concentration is 0.05%, whereas in the deep purple varieties it is 13–14% (Goodwin and Mercer, 1972). Blueness of flowers can also be due to copigmentation between an anthocyanin and a flavonol. An example of this is in maroon and mauve Primula species, where the anthocyanin is malvidin-3glucoside in both cases, the diVerence in color being due to copigmentation with high concentrations of kempferol glucosides in the mauve variety. Spectral shifts due to copigmentation occur at pH values of 1–7, but these are not limited to polyphenolics and can also occur with purines and alkaloids, resulting in bathochromic shifts in the visible region of the adsorption spectra. Anthocyanins can also form complexes with several divalent or trivalent metals such as copper (Cu), aluminum (AI), or iron (Fe), leading to changes in color and varying as a function of pH. For example, Alþþþ ions bond with anthocyanins with ortho-dihydroxyl groups on the B ring, causing a bathochromic shift to give more blue coloration. This is best illustrated by comparing the blue color of cornflowers (Centaurea cyanus) with the red color of roses: the anthocyanin is cyanidin in both cases, but the blue color of cornflowers is due to a combined eVect of metal chelation with iron and copigmentation with apigenin diglycoside to form procyanin, a blue crystalline iron complex. It is also well known that if the mineral balance of Hydrangea species is correct, aluminum is easily accumulated and the petals turn blue, otherwise they are red. Many mineral elements, including Fe, are extensively mobilized from senescing leaves (Himelblau and Amasino, 2001), making it unlikely that concentration of metal ions by normal physiological mechanisms plays a part in color intensification, although toxic accumulation may be significant under some circumstances. The brown color in some petals (e.g., wallflowers) is due to a combination of the magenta anthocyanin in the vacuole with the yellow carotenoids in the chromoplast. The yellows and browns of autumn foliage occurring at the end of the senescence period are accounted for mainly by the presence of carotenoids (see Section III.B.), and by the formation of dark oxidation

4. Leaf Senescence, Nutrient Recycling, and Stress Defenses

149

products of polyphenols such as condensed or hydrolyzable tannins as subcellular compartmentation collapses. Loss of anthocyanin pigments during this period may also occur by intramolecular rearrangements, resulting in the formation of the colorless pseudo-bases. A recently discovered fate for anthocyanins is via reduction to colorless 2,3-cis-flavan-3-ols (e.g., epicatechin), mediated by a newly discovered enzyme anthocyanin reductase, which has been proposed to be involved in synthesis of condensed tannins (Xie et al., 2003). The factors aVecting the determination of the color of plant tissues as a result of anthocyanin accumulation are therefore a combination of the extent of glycosylation and acylation, the pH in the vacuole, and the presence of metal ions and copigments such as flavonols and flavones (Mol et al., 1998). The perception of anthocyanin color in vivo can also be appreciably altered by cell structure. For example, Noda et al. (1994) showed that a transcription factor regulating the intensity of Antirrhinum flower color does so via control of cell shape. It is clear, therefore, that the progression of autumn colors shown by leaves of many trees may not be due solely to the loss of chlorophyll revealing the underlying colors of anthocyanins. Senescence-induced changes in vacuolar pH, increased or decreased levels of metal ions, the degradation of copigments such as flavonols and carotenoids, and the polymerization and oxidation of condensed and hydrolyzable tannins may well combine to produce the progression of color changes, from reds, oranges, yellows, and finally browns, independent of levels of de novo synthesis of anthocyanins. Evidence for de novo synthesis of anthocyanins in leaves during senescence is currently weak, and, in fact, in senescing leaves of the copper varieties of beech and hazel, senescence is preceded by the loss of anthocyanin so that, for a while, the foliage turns as green as in wild-type trees (Matile, 2000). Furthermore, although condensed tannins (or proanthocyanidins) give rise to anthocyanidins on acid hydrolysis, there is no evidence that this occurs in planta even during senescence. The wide range of potential modifications to which anthocyanins are subject cautions against assumptions that enhanced coloration in autumn must be the result solely of net synthesis of these compounds and therefore metabolically costly.

IV. Pigments and Stress Defenses in Senescing Leaves A. Color Changes in Senescence as Signals We have seen that the recycling function of leaf senescence is potentially vulnerable to disruption by light stress, and that pigment metabolism is normally organized and controlled in senescence to minimize photodamage. Moreover, the relationship between, on the one hand, pigment metabolism

150

Ougham et al.

and, on the other, the formation of lesions mimicking the symptoms of fungal and bacterial disease, emphasizes the importance of senescence processes in plant responses to biotic stress. Developing the theme of pigments and biotic interactions further, the case of herbivorous predation on plant tissues illustrates the ecological significance of cellular events in foliar senescence. Archetti (2000) and Hamilton and Brown (2001) have elaborated the autumn signaling hypothesis, which raises interesting questions about the origins and functions of plant pigments and the way the animal eye responds to them. The hypothesis proposes that the autumn coloration observed in many tree species acts as an honest (handicap) signal to potential insect predators about the tree’s investment in defense—defensively committed and vigorous trees should produce the most intense coloration and, hence, the greatest deterrent to insects. It is suggested that the signaling mechanism in trees, and the insects’ avoidance response, are features that have coevolved. Although some subsequent publications have lent support to the hypothesis (e.g., Hagen et al., 2003), others have provided experimental evidence (Holopainen and Peltonen, 2002) or theoretical grounds (Wilkinson et al., 2002) to refute it. The majority of work in this area has so far been published by ecologists. Here we examine some aspects of the hypothesis and the results presented to date in the context of our knowledge about leaf senescence and plant physiology.

B. Is Autumn Color a Costly Signal? A concept widely used in the autumn signaling theory is that autumn coloration is costly for the tree to produce (and, by implication, must therefore have some definite purpose). This idea probably arises by analogy with animal metabolism, in which any biosynthetic process has requirements for both energy and carbon that must be met from food intake or by breaking down some of the animal’s own tissue to generate fuel and raw materials. However, plants are autotrophic, and, provided they are exposed to adequate light, air, and water, carbon and energy are not limiting factors as they are for heterotrophs. Indeed, it has been argued that terrestrial plants have evolved physically and physiologically to dump excess carbon captured through promiscuous assimilation (Thomas and Sadras, 2001). According to this proposition, the general overabundance of carbon and energy with which plants are cursed has resulted in proliferation of the huge range of carbon-rich secondary compounds that are unique to the plant kingdom (Hadacek, 2002; Pichersky and Gang, 2000) as a consequence of a kind of speculative metabolic doodling that occasionally pays oV in terms of improved fitness. In any event, it is questionable to assume that activities requiring carbon and energy are necessarily costly to a plant in the same

4. Leaf Senescence, Nutrient Recycling, and Stress Defenses

151

sense that they would be for an animal or other heterotroph. This is particularly true of autumnal processes occurring in vegetation. Growth in plants is more sensitive to reduced temperature than is photosynthesis (Hjelm and Ogren, 2003); hence, as the temperature declines in autumn, the tree stops growing, further reducing the need for carbon and energy, which it can easily obtain again when required. The recycling function of autumnal senescence is part of a strategy to safeguard winter survival and resumption of growth in spring and culminates in the discarding of foliar skeletons consisting largely of carbon, oxygen, and hydrogen in various combinations. This throw-away residue may include two classes of pigment containing no elements of great reclamation value: carotenoids, which, as we have seen, can be unmasked, transformed, or relocated, but not in general synthesized de novo during senescence; and anthocyanins, which often are newly synthesized during the senescence process (Ishikura, 1972), although, as discussed, the argument that to do so is metabolically costly is at best questionable.

C. Possible Functions of Leaf Color The question as to why trees produce bright colors in autumn, if they are not an honest signal about defense capability, has been reviewed by Wilkinson et al. (2002). Briefly, plant biologists have two main hypotheses to explain the synthesis of anthocyanins. They may have a role in defense against abiotic stress (Steyn et al., 2002), protecting against potentially damaging forms of oxygen and chemical radicals. As photosynthesis declines during foliar senescence, light energy must be dissipated in alternative ways, some of which lead to the generation of reactive oxygen species (Feild et al., 2001). EYcient recapture of nutrients exported from the senescing leaf requires protection against photooxidative damage. Hoch et al. (2003) demonstrated the capacity of anthocyanins to facilitate nutrient recovery during leaf senescence in three deciduous woody species; Lee et al. (2003) similarly inferred a correlation between anthocyanin production and eYciency of nitrogen resorption in a number of deciduous forest species. It is noteworthy that many plant species also synthesize anthocyanins in the leaf in response to stresses such as cold, drought, or very high light intensity, when again carbon assimilation and demand are unmatched and there may be an increased risk of free radical production (Hoch et al., 2001). The coevolution theory suggests that vigorous and defensively committed trees can ‘‘aVord’’ the loss of photosynthate resulting from early senescence, whereas less vigorous trees need to continue photosynthesizing for longer—but, as pointed out earlier, during the autumn period it is not carbon and energy that are at a premium, but nitrogen and other nutrients. A tree lacking vigor because of nutrient deficiency or abiotic stress cannot make productive use

152

Ougham et al.

of photosynthesis and is more, rather than less, likely to undergo early senescence. Interestingly, Schaberg et al. (2003) found that the extent and earliness of onset of red coloration in maple leaves was positively correlated with foliar nitrogen deficiency, an observation that contradicts the hypothesis that it is vigorous and healthy trees that initiate leaf senescence early but is entirely in accord with plant scientists’ observations of many species under many conditions. Other suggested functions of phenylpropanoid pigments in protecting against abiotic stresses include roles as antioxidants (Tsuda et al., 1994) and osmolytes (Chalker-Scott, 2002). An additional role not often considered is suggested by the striking fact that intense pigmentation is a characteristic of deciduous species. It may be that colored secondary compounds benefit the plant by contributing to the allelopathic properties of leaf litter (Wardle et al., 1998). Alternatively, the anthocyanins may simply represent a convenient dumping ground for excess carbon, in a form that is not metabolizable by, or attractive to, potential predators and pathogens. The fact that leaves are colored may be coincidental; our own evolutionary and social history has led human beings to attribute great significance to pigments in the wavelength range we can perceive, but a compound’s color may not have any particular correlation with its function (there is, for example, no particular reason why the human gall bladder needs to be green!). The comparative cell and molecular biology of foliar senescence supports the view that the senescing leaf is the evolutionary progenitor of brightly colored floral and reproductive structures attractive to animal pollinators and dispersers (Matile et al., 1999; Thomas et al., 2003). It follows that the heightened physiological and psychological sensitivity of humans to the colors of autumn foliage may not have any direct biological meaning. It may, rather, be a secondary consequence of spectral tuning by fruit and leaf color during evolution of the trichromatic primate visual system (Surridge et al., 2003). The connection between the colors of fruits and autumn leaves has been considered by Stiles (1982), who suggested that trees bearing colored fruit in fall may have evolved synchronization between fruit ripening and leaf coloration as an additional signal to seed-dispersing birds.

D. Does Dishonesty Pay? Even if the red coloration is not costly to produce, could the signaling hypothesis still hold good? Subsequent authors (Lachmann et al., 2001; Wilkinson et al., 2002) have pointed out that it is not essential that an honest signal be costly provided that dishonesty is penalized. In the case of autumn colors, dishonesty would consist in a poorly defended tree producing bright colors as a misleading deterrent. However, as indicated earlier, it is precisely

4. Leaf Senescence, Nutrient Recycling, and Stress Defenses

153

stressed, and therefore generally poorly defended, trees that do produce more of the bright autumn colors. If the coloration were indeed a deterrent, far from being a costly misdirection, it could in fact act in the tree’s favor. Therefore, neither is the ‘‘signal’’ costly to produce, nor does its dishonest production penalize the tree. However, an alternative explanation should also be considered. The signaling theory would still be valid if trees that are poor in nutrients in the autumn subsequently defend their (limited) resources more heavily in the following spring, compared to ‘‘richer’’ individuals that may simply outgrow their pests. In this scenario, the signal becomes an honest and unfakeable indicator of resources, since early bright leaves represent low resources. If, for a given plant species, low nonrenewable nutrient resource is associated with increased defenses in spring, then insects would be expected to make appropriate evasive action in autumn. Future experimental work on the relationships between resource status and leaf coloration in autumn, insect responses, and defensive commitment the following spring will be necessary to clarify this issue.

E. Insect Preference for Green Leaves Moving from the reason why trees develop the coloration in the first place to a consideration of the reasons why insect predators avoid brightly colored leaves, the composition of these leaves in comparison with green foliage should be taken into account. By the time a leaf is orange or red, it will have broken down and exported a high proportion of its total protein. Photosynthesis will have ceased, remaining low-molecular-weight carbohydrates will have been removed, and, in general, it will have a much lower content of nutrients that an insect could digest than would a green leaf on the same tree. Anthocyanins that have accumulated will not be digestible by insects; they, or other, colorless, secondary products accumulating at the same time, may even be unpalatable and act as antifeedants. Furthermore, antinutritional factors such as inhibitors of digestive tract proteases are prominent among the genes and gene products upregulated in senescence and cell death (e.g., Huang et al., 2001). It is therefore quite feasible that insects initially land on green and red/yellow leaves in equal numbers but quickly vacate the latter after an initial sampling—to assess this possibility would require more detailed observation of insect behavior than has been presented in any of the studies to date. Alternatively, the predators may indeed be responding to the visual signal, but its significance is not ‘‘this is a vigorous tree able to withstand your attack’’ but simply ‘‘this is a leaf with poor nutritional quality, possibly unpalatable.’’ Wilkinson et al. (2002) point out that because an individual tree may simultaneously bear green, yellow, and red leaves, the hypothesis that leaf color signals the overall vigor or defensive

154

Ougham et al.

capability of the tree is suspect. However, it could feasibly signal the nutritional value of the individual leaf; because studies to date (Hagen et al., 2003; Holopainen and Peltonen, 2002) have measured insect colonization or insect damage relative to the mean proportion of brightly colored leaves on a tree, rather than on an individual leaf basis, the question remains open. To complicate the issue further, Holopainen and Peltonen (2002) point out that birch aphids preferentially land on yellow rather than on green or red leaves, and propose that such leaves, which are actively exporting lowmolecular-weight nitrogenous compounds such as amino acids during early senescence, are a rich source of accessible nutrient for the insects. A diYculty in interpreting the data on insect behavior as a whole is that most studies (e.g., Hagen et al., 2003) did not look at red and yellow leaves separately; it may indeed be the case that some insect species are attracted to yellowing leaves more than green leaves, but to red leaves least of all. This possibility would resolve some of the inconsistencies in the story so far, including the fact pointed out by Wilkinson et al. (2002) that yellow is normally an attractive color to aphids. An interesting perspective on this issue is provided by the common observation that the proportion of infertile individuals in a plant population increases under stress. In species with strongly expressed monocarpic or reproduction-associated senescence patterns, barren plants may remain green while fertile individuals degrade their chlorophyll normally. Such barren plants have been reported to benefit the population by acting as decoys, reducing herbivory pressures on individuals destined to produce the next generation (Thomas and Sadras, 2001).

F. Visual and Olfactory Signals The color changes observed in tree leaves in autumn can be so spectacular that it is easy for humans to overlook the possibility that the predators may be responding to diVerent signals entirely, arising from other biochemical changes that the tree may be undergoing at the same time. It is possible that the insects are deterred not by color at all, but by the volatile substances emanating from plant leaves during senescence. Most land plants emit significant amounts of volatiles such as aldehydes and isoprenes during natural or wound-induced senescence (de Gouw et al., 1999; Fall et al., 1999), and it is known that insects can perceive and respond to quite low concentrations of these compounds (Ruther et al., 2002). There is also evidence that the volatiles may act as defense agents by attracting insect parasitoids (Hoballah and Turlings, 2001). Peak emission of such volatiles would be expected to coincide with the phase when the leaf was yellow or red but still alive. It would be informative to measure volatile emissions during the color change period and correlate the results with levels of insect colonization.

4. Leaf Senescence, Nutrient Recycling, and Stress Defenses

155

V. Conclusions In recent years, a fairly complete picture has emerged of the metabolic molecular and subcellular networks responsible for pigmentation changes during the growth and senescence of foliage. This has allowed the development of a model of leaf senescence in which pigmentation changes partially set the pace for proteolysis and nitrogen recycling at the same time as they play a critical role in sustaining cell viability under conditions of abiotic and biotic challenge. Although rapid progress has been made in understanding pigment metabolism in senescence, the mechanism of protein degradation and its control remains poorly understood. The interconversions and relocations of the amino acid products of proteolysis in leaf tissues are quite well established (Dangl et al., 2000), but the step between the intact protein and its hydrolysis products continues to elude definitive analysis. Many senescenceassociated and upregulated protease genes have been described (Bhalerao et al., 2003; Buchanan-Wollaston, 1997; Buchanan-Wollaston et al., 2003), but it is unclear how many, if any at all, are necessary for normal protein breakdown. Some new developments may help to introduce much-needed innovative ideas into the field of proteolysis and its control in senescence. Improved microscopy techniques are beginning to provide evidence for traYc between plastids (which contain most of the mobilizable protein in senescing cells) and vacuoles, which have long been considered to be the main sites of intracellular proteolysis (Chiba et al., 2003; Guiame´ t et al., 1999). Cascades of caspase proteases are characteristic of programmed cell death in animal systems, but plant genomes seem not to include orthologues of caspase genes; nevertheless, recent sequence searches and functional analyses have revealed families of so-called metacaspases in plants that may fulfil various signaling and proteolytic roles in terminal and pathological plant processes, including senescence (Watanabe and Lam, 2004). Another potentially fruitful development is the application of quantitative trait mapping to test the relative map positions of genetic loci for, on the one hand, nitrogen assimilation and reallocation traits in crop development and, on the other, protease enzyme activities and gene sequences (Andreas Fischer, unpublished results). Much work needs to be done before protein recycling in plant senescence could be said to be a well-understood process, but new tools and approaches are being applied and rapid progress can be expected in the near future.

Acknowledgments The authors’ research on pigment metabolism and leaf senescence is supported by the UK Biotechnology and Biological Sciences Research Council and the Department of Environment, Food, and Rural AVairs.

156

Ougham et al.

References Afitlhile, M. M., Dent, R. M., and Cowan, A. K. (1993). Changes in carotenoid composition in senescing leaves of Hordeum-vulgare L Cv Dyan. J. Plant Physiol. 142, 43–49. Alfenito, M. R., Souer, E., Goodman, C. D., Buell, R., Mol, J., Koes, R., and Walbot, V. (1998). Functional complementation of anthocyanin sequestration in the vacuole by widely divergent glutathione S-transferases. Plant Cell 10, 1135–1149. Allen, J. F., and Forsberg, J. (2001). Molecular recognition in thylakoid structure and function. Trends Plant Sci. 6, 317–326. Andersson, A., Keskitalo, J., Sjo¨ din, A., Bhalerao, R., Sterky, F., Wissel, K., Tandre, K., Aspeborg, H., Moyle, R., Ohmiya, Y., Bhalerao, R., Brunner, A., Gustafsson, P., Karlsson, J., Lundeberg, J., Nilsson, O., Sandberg, G., Strauss, S., Sundberg, B., Uhlen, M., Jansson, S., and Nilsson, P. (2004). A transcriptional timetable of autumn senescence. Genome Biol. 5, R24. Archetti, M. B. (2000). The origin of autumn colours by coevolution. J. Theor. Biol. 205, 625–630. Bachmann, A., Ferna´ ndez-Lo´ pez, J., Ginsburg, S., Thomas, H., Bouwkamp, J. C., Solomos, T., and Matile, P. B. (1994). Stay-green genotypes of Phaseolus vulgaris. Chloroplast proteins and chlorophyll catabolites during foliar senescence. New Phytol. 126, 593–600. Bhalerao, R., Keskitalo, J., Sterky, F., Erlandsson, R., Bjorkbacka, H., Birve, S. J., Karlsson, J., Gardestrom, P., Gustafsson, P., Lundeberg, J., and Jansson, S. (2003). Gene expression in autumn leaves. Plant Physiol. 131, 430–442. Biswal, B. (1995). Carotenoid catabolism during leaf senescence and its control by light. J. Photoch. Photobiol. B. 30, 3–13. Biswal, B., Rogers, L. J., Smith, A. L., and Thomas, H. (1994). Carotenoid composition and its relationship to chlorophyll and DI protein during leaf development in a normally senescing cultivar and a stay-green mutant of Festuca pratensis. Phytochemistry 37, 1257–1262. Buchanan-Wollaston, V. (1997). The molecular biology of leaf senescence. J. Exp. Bot. 48, 181–199. Buchanan-Wollaston, V., Earl, S., Harrison, E., Mathas, E., Navabpour, S., Page, T., and Pink, D. (2003). The molecular analysis of leaf senescence—a genomics approach. Plant Biotechnol. J. 1, 3–22. Carroll, L. (1872). ‘‘Through the Looking-Glass and What Alice Found There.’’ Macmillan and Co., London. Chalker-Scott, L. (2002). Do anthocyanins function as osmoregulators in leaf tissues? Adv. Bot. Res. 37, 103–127. Chiba, A., Ishida, H., Nishizawa, N. K., Makino, A., and Mae, T. (2003). Exclusion of ribulose-1,5-bisphosphate carboxylase/oxygenase from chloroplasts by specific bodies in naturally senescing leaves of wheat. Plant Cell Physiol. 44, 914–921. Dangl, J. L., Dietrich, R. A., and Thomas, H. (2000). Senescence and programmed cell death. In ‘‘Biochemistry and Molecular Biology of Plants’’ (B. Buchanan, W. Gruissem, and R. Jones, Eds.), pp. 1044–1100. ASPP, Rockville, Md. Davies, T. G. E., Thomas, H., Thomas, B. J., and Rogers, L. J. (1990). Leaf senescence in a nonyellowing mutant of Festuca pratensis: Metabolism of cytochrome f. Plant Physiol. 93, 588–595. de Gouw, J. A., Howard, C. J., Custer, T. G., and Fall, R. (1999). Emissions of volatile organic compounds from cut grass and clover are enhanced during the drying process. Geophys. Res. Lett. 26, 811–814. Fall, R., Karl, T., Hansel, A., Jordan, A., and Lindinger, W. (1999). Volatile organic compounds emitted after leaf wounding: On-line analysis by proton-transfer-reaction mass spectrometry. J. Geophys. Res. Atmos. 104, 15963–15974.

4. Leaf Senescence, Nutrient Recycling, and Stress Defenses

157

Feild, T. S., Lee, D. W., and Holbrook, N. M. (2001). Why leaves turn red in autumn. The role of anthocyanins in senescing leaves of red-osier dogwood. Plant Physiol. 127, 566–574. Garcia-Plazaola, J., and Becerril, J. M. (2001). Seasonal changes in photosynthetic pigments and antioxidants in beech (Fagus sylvatica) in a Mediterranean climate: Implications for tree decline diagnosis. Aus. J. Plant. Physiol. 28, 225–232. Garcia-Plazaola, J. I., Hernandez, A., and Becerril, J. M. (2003). Antioxidant and pigment composition during autumnal leaf senescence in woody deciduous species diVering in their ecological traits. Plant Biol. 5, 557–566. Goodwin, T. W., and Mercer, E. I. (1972). ‘‘Introduction to Plant Biochemistry.’’ Pergamon Press, Oxford, England. Goto, T. (1987). Structure, stability, and color variation of natural anthocyanins. Progr. Chem. Org. Nat. P. 52, 113–158. Gray, J., Janick-Bruckner, D., Bruckner, B., Close, P. S., and Johal, G. S. (2002). Lightdependent death of maize Ils1 cells is mediated by mature chloroplasts. Plant Physiol. 130, 1894–1907. Guiame´ t, J. J., Pichersky, E., and Noode´ n, L. D. (1999). Mass exodus from senescing soybean chloroplasts. Plant Cell Physiol. 40, 986–992. Hadacek, F. (2002). Secondary metabolites as plant traits: Current assessment and future perspectives. Crit. Rev. Plant Sci. 21, 273–322. Hagen, S. B., Folstad, I., and Jakobsen, S. W. (2003). Autumn colouration and herbivore resistance in mountain birch (Betula pubescens). Ecol. Lett. 6, 807–811. Hamilton, W. D., and Brown, S. P. (2001). Autumn tree colours as a handicap signal. Philos. R. Soc. Lond., B, Biol. Sci. 268, 1489–1493. Himelblau, E., and Amasino, R. M. (2001). Nutrients mobilized from leaves of Arabidopsis thaliana during leaf senescence. J. Plant Physiol. 158, 1317–1323. Hjelm, U., and Ogren, E. (2003). Is photosynthetic acclimation to low temperature controlled by capacities for storage and growth at low temperature? Results from comparative studies of grasses and trees. Physiol. Plant 119, 113–120. Hoballah, M. E. F., and Turlings, T. C. J. (2001). Experimental evidence that plants under caterpillar attack may benefit from attracting parasitoids. Evol. Ecol. Res. 3, 553–565. Hoch, W. A., Singsaas, E. L., and McCown, B. H. (2003). Resorption protection. Anthocyanins facilitate nutrient recovery in autumn by shielding leaves from potentially damaging light levels. Plant Physiol. 133, 1296–1305. Hoch, W. A., Zeldin, E. L., and McCown, B. H. (2001). Physiological significance of anthocyanins during autumnal leaf senescence. Tree Physiol. 21, 1–8. Holopainen, J. K., and Peltonen, P. (2002). Bright autumn colours of deciduous trees attract aphids: Nutrient retranslocation hypothesis. Oikos 99, 184–188. Hondo, T., Yoshida, K., Nakagawa, A., Kawai, T., Tamura, H., and Goto, T. (1992). Structural basis of blue-colour development in flower petals from Commelina communis. Nature 358, 515–518. Hormaetxe, K., Hernandez, A., Becerril, J. M., and Garcia-Plazaola, J. I. (2004). Role of red carotenoids in photoprotection during winter acclimation in Buxus sempervirens leaves. Plant Biol. 6, 325–332. Ho¨ rtensteiner, S. (2004). The loss of green color during chlorophyll degradation—a prerequisite to prevent cell death? Planta 219, 191–194. Ho¨ rtensteiner, S., and Feller, U. (2002). Nitrogen metabolism and remobilization during senescence. J. Exp. Bot. 53, 927–937. Huang, Y.-J., To, K.-Y., Yap, M.-N., Chiang, W.-J., Suen, D.-F., and Chen, S.-C. G. (2001). Cloning and characterization of leaf senescence up-regulated genes in sweet potato. Physiol. Plant 113, 384–391.

158

Ougham et al.

Ishikura, N. (1972). Autumnal reddening of leaves. 1. Anthocyanins and other phenolics in autumn leaves. Phytochemistry 11, 2555–2558. Jakob-Wilk, D., Holland, D., Goldschmidt, E. E., Riov, J., and Eyal, Y. (1999). Chlorophyll breakdown by chlorophyllase: Isolation and functional expression of the Chlase 1 gene from ethylene-treated citrus fruit and its regulation during development. Plant J. 20, 653–661. Ji, S. B., Yokoi, M., Saito, N., and Mao, L. S. (1992). Distribution of anthocyanins in aceraceae leaves. Biochem. Systemat. Ecol. 20, 771–781. Kannangara, C. G., and Hansson, M. (1998). Arrest of chlorophyll accumulation prior to anthocyanin formation in Euphorbia pulcherrima. Plant Physiol. Biochem. 36, 843–848. Koes, R. E., Quattrocchio, F., and Mol, J. N. M. (1994). The flavonoid biosynthetic-pathway in plants—function and evolution. Bioessays 16, 123–132. Lachmann, M., Szamado, S., and Bergstrom, C. T. (2001). Cost and conflict in animal signals and human language. Proc. Natl. Acad. Sci. USA 98, 13189–13194. Lee, D. W., O’Keefe, J., Holbrook, N. M., and Feild, T. S. (2003). Pigment dynamics and autumn leaf senescence in a New England deciduous forest, eastern USA. Ecol. Res. 18, 677–694. Mach, J. M., Castillo, A. R., Hoogstraten, R., and Greenberg, J. T. (2001). The Arabidopsisaccelerated cell death gene ACD2 encodes red chlorophyll catabolite reductase and suppresses the spread of disease symptoms. Proc. Natl. Acad. Sci. USA 98, 771–776. Matile, P. (1994). Fluorescent idioblasts in autumn leaves of Ginkgo-biloba. Bot. Helv. 104, 87–92. Matile, P. (2000). Biochemistry of Indian summer: Physiology of autumnal leaf coloration. Exp. Gerontol. 35, 145–158. Matile, P., Ho¨ rtensteiner, S., and Thomas, H. (1999). Chlorophyll degradation. Annu. Rev. Plant Phys. 50, 67–95. Merzlyak, M. N., and Gitelson, A. (1995). Why and what for the leaves are yellow in autumn— on the interpretation of optical-spectra of senescing leaves (Acer platanoides L). J. Plant. Physiol. 145, 315–320. Mol, J., Grotewold, E., and Koes, R. (1998). How genes paint flowers and seeds. Trends Plant Sci. 3, 212–217. Mu¨ hlecker, W., and Kra¨ utler, B. (1996). Breakdown of chlorophyll: Constitution of nonfluorescing chlorophyll catabolites from senescent cotyledons of the dicot rape. Plant Physiol. Biochem. 34, 61–75. Munne-Bosch, S., and Penuelas, J. (2003). Photo- and antioxidative protection during summer leaf senescence in Pistacia lentiscus L. grown under Mediterranean field conditions. Ann. Bot. 92, 385–391. Noda, K.-J., Glover, B., Linstead, P., and Martin, C. (1994). Flower colour intensity depends on specialized cell shape controlled by a Myb-related transcription factor. Nature 369, 661–664. Pecket, R. C., and Small, C. J. (1980). Occurrence, location and development of anthocyanoplasts. Phytochemistry 19, 2571–2576. Pichersky, E., and Gang, D. R. (2000). Genetics and biochemistry of secondary metabolites in plants: An evolutionary perspective. Trends Plant Sci. 5, 439–445. Pruzˇ inska´ , A., Tanner, G., Anders, I., Roca, M., and Ho¨ rtensteiner, S. (2003). Chlorophyll breakdown: Pheophorbide a oxygenase is a Rieske-type iron-sulfur protein, encoded by the accelerated cell death 1 gene. Proc. Natl. Acad. Sci. USA 100, 15259–15264. Roca, M., James, C., Pruzˇ inska´ , A., Ho¨ rtensteiner, S., Thomas, H., and Ougham, H. (2004). Analysis of the chlorophyll catabolism pathway in leaves of an introgression senescence mutant of Lolium temulentum. Phytochemistry 65, 1231–1238. Romero-Puertas, M. C., and Delledonne, M. (2003). Nitric oxide signaling in plant-pathogen interactions. IUBMB Life 55, 579–583.

4. Leaf Senescence, Nutrient Recycling, and Stress Defenses

159

Ruther, J., Reinecke, A., and Hilker, M. (2002). Plant volatiles in the sexual communication of Melolontha hippocastani: Response towards time-dependent bouquets and novel function of (Z)-3-hexen-1-ol as a sexual kairomone. Ecol. Entomol. 27, 76–83. Schaberg, P. G., Van den Berg, A. K., Murakami, P. F., Shane, J. B., and Donnelly, J. R. (2003). Factors influencing red expression in autumn foliage of sugar maple trees. Tree Physiol. 23, 325–333. Scheumann, V., Schoch, S., and Ru¨ diger, W. (1999). Chlorophyll b reduction during senescence of barley seedlings. Planta 209, 364–370. Steinmuller, D., and Tevini, M. (1985). Composition and function of plastoglobuli. I. Isolation and purification from chloroplasts and chromoplasts. Planta. 163, 201–207. Steyn, W. J., Wand, S. J. E., Holcroft, D. M., and Jacobs, G. (2002). Anthocyanins in vegetative tissues: A proposed unified function in photoprotection. New Phytol. 155, 349–361. Stiles, E. W. (1982). Fruit flags—2 hypotheses. Am. Nat. 120, 500–509. Strack, D., and Wray, V. (1989). Anthocyanins. In ‘‘Methods in Plant Biochemistry’’ (P. M. Dey and J. B. Harborne, Eds.), Vol. 1, pp. 326–352. Academic Press, London. Surridge, A. K., Osorio, D., and Mundy, N. I. (2003). Evolution and selection of trichromatic vision in primates. Trends Ecol. Evol. 18, 198–205. Tanaka, R., Hirashima, M., Satoh, S., and Tanaka, A. (2003). The Arabidopsis-accelerated cell death gene ACD1 is involved in oxygenation of pheophorbide a: Inhibition of the pheophorbide a oxygenase activity does not lead to the ‘‘stay-green’’ phenotype in Arabidopsis. Plant Cell Physiol. 44, 1266–1274. Tevini, M., and Steinmuller, D. (1985). Composition and function of plastoglobuli. II. Lipid composition of leaves and plastoglobuli during beech leaf senescence. Planta 163, 91–96. Thomas, H. (1997). Chlorophyll: A symptom and a regulator of plastid development. New Phytol. 136, 163–181. Thomas, H., and Donnison, I. (2000). Back from the brink: Plant senescence and its reversibility. In ‘‘Programmed Cell Death in Animals and Plants’’ (J. Bryant, S. G. Hughes, and J. M. Garland, Eds.), pp. 149–162. Bios, Oxford, UK. Thomas, H., and Howarth, C. J. (2000). Five ways to stay green. J. Exp. Bot. 51, 329–337. Thomas, H., Ougham, H., Canter, P., and Donnison, I. (2002). What stay-green mutants tell us about nitrogen remobilization in leaf senescence. J. Exp. Bot. 53, 801–808. Thomas, H., Ougham, H., and Ho¨ rtensteiner, S. (2001). Recent advances in the cell biology of chlorophyll catabolism. Adv. Bot. Res. 35, 1–52. Thomas, H., Ougham, H. J., WagstaV, C., and Stead, A. J. (2003). Defining senescence and death. J. Exp. Bot. 54, 1127–1132. Thomas, H., and Sadras, V. O. (2001). The capture and gratuitous disposal of resources by plants. Funct. Ecol. 15, 3–12. Thomas, H., and Smart, C. M. (1993). Crops that stay green. Ann. Appl. Biol. 123, 193–219. Tommasini, R., Vogt, E., Fromenteau, M., Ho¨ rtensteiner, S., Matile, P., Amrhein, N., and Martinoia, E. (1998). An ABC transporter of Arabidopsis thaliana has both glutathioneconjugate and chlorophyll catabolite transport activity. Plant J. 13, 773–780. Tsuchiya, T., Ohta, H., Okawa, K., Iwamatsu, A., Shimada, H., Matsuda, T., and Takamiya, K. (1999). Cloning of chlorophyllase, the key enzyme in chlorophyll degradation: Finding of a lipase motif and the induction by methyl jasmonate. Proc. Natl. Acad. Sci. USA 96, 15362–15367. Tsuda, T., Watanabe, M., Ohshima, K., Norinobu, S., Cho, S. W., Kawakishi, S., and Osawa, T. (1994). Antioxidative activity of the anthocyanin pigments cyanidin 3-O- -D-glucoside and cyanidin. J. Ag. Food Chem. 42, 2407–2410. van Doorn, W. G., and Woltering, E. J. (2004). Senescence and cell death: Substance or semantics? J. Exp. Bot. 55, 2147–2153.

160

Ougham et al.

Wardle, D. A., Nilsson, M. C., Gallet, C., and Zackrisson, O. (1998). An ecosystem-level perspective of allelopathy. Biol. Rev. 73, 305–319. Watanabe, N., and Lam, E. (2004). Recent advance in the study of caspase-like proteases and Bax inhibitor-1 in plants: Their possible roles as regulator of programmed cell death. Mol. Plant Pathol. 5, 65–70. Weisshaar, B., and Jenkins, G. I. (1998). Phenylpropanoid biosynthesis and its regulation. Curr. Opin. Plant Biol. 1, 251–257. Wilkinson, D. M., Sherratt, T. N., Phillip, D. M., Wratten, S. D., Dixon, A. F. G., and Young, A. J. (2002). The adaptive significance of autumn leaf colours. Oikos 99, 402–407. Wu¨ thrich, K. L., Bovet, L., Hunziker, P. E., Donnison, I. S., and Ho¨ rtensteiner, S. (2000). Molecular cloning, functional expression and characterisation of RCC reductase involved in chlorophyll catabolism. Plant J. 21, 189–198. Xie, D. Y., Sharma, S. B., Paiva, N. L., Ferreira, D., and Dixon, R. A. (2003). Role of anthocyanidin reductase, encoded by BANYULS in plant flavonoid biosynthesis. Science 299, 396–399. Young, A. J., Wellings, R., and Britton, G. (1991). The fate of chloroplast pigments during senescence of primary leaves of Hordeum-vulgare and Avena-sativum. J. Plant Physiol. 137, 701–705. Zavaleta-Mancera, H. A., Franklin, K. A., Ougham, H. J., Thomas, H., and Scott, I. M. (1999a). Regreening of Nicotiana leaves. I. Reappearance of NADPH-protochlorophyllide oxidoreductase and light-harvesting chlorophyll a/b-binding protein. J. Exp. Bot. 50, 1677–1682. Zavaleta-Mancera, H. A., Thomas, B. J., Thomas, H., and Scott, I. M. (1999b). Regreening of Nicotiana leaves. II. RediVerentiation of plastids. J. Exp. Bot. 50, 1683–1689.

5

____________________________________________________________________________

Extracellular Proteases: Biological and Behavioral Roles in the Mammalian Central Nervous System Yan Zhang, Kostas Pothakos, and Styliana-Anna (Stella) Tsirka Department of Pharmacological Sciences State University of New York at Stony Brook Stony Brook, New York 11794-8651

I. Introduction II. Plasminogen Activators A. Tissue-Type Plasminogen Activator B. Urokinase-Type Plasminogen Activator III. IV. V. VI. VII.

Plasmin(ogen) Inhibitors of Plasminogen Activators and Plasmin tPA EVects on Rodent Behavior uPA and Plasminogen EVects on Rodent Behavior Conclusions Acknowledgments References

Extracellular proteases and their inhibitors have been implicated in both physiological and pathological states in the central nervous system (CNS). Given the presence of several classes of proteases, it is believed that each enzyme may undertake distinct biological roles. Some are indispensable for neuronal migration, neurite outgrowth and pathfinding, and synaptic plasticity. Others are required for neuronal death and tumor growth and invasion. Furthermore, studies from transgenic animals lacking or overexpressing one or more of the proteases have suggested that functional compensation and redundance among diVerent members do exist. Normally, protease activity is tightly regulated by specific inhibitors to prevent disastrous proteolysis. Various insults can disrupt the fine control of proteolysis and cause pathological changes. Novel strategies have been attempted to maintain or restore proteaseinhibitor homeostasis, thus minimizing damages to the CNS. They may provide us with eVective therapeutic tools for fighting certain neurological disorders. C 2005, Elsevier Inc.

Current Topics in Developmental Biology, Vol. 66 Copyright 2005, Elsevier Inc. All rights reserved. 0070-2153/05 $35.00

161

162

Zhang et al.

I. Introduction There are several major classes of proteases present in the CNS, which are grouped according to diVerences in the composition of their catalytic sites. These are the families of (1) serine proteases, with the most well-known members such as thrombin, tissue-type plasminogen activator (tPA), and plasmin; (2) matrix metalloproteinases (MMPs), which consist of more than 20 members identified to date that all require Zn2þ for their enzymatic activity; (3) cysteine proteases, which include 14 caspases involved in distinct steps of the apoptotic pathway; and (4) aspartic proteases, such as the lysosomal peptidases cathepsins. Proteases are considered to be key players in the maintenance of CNS homeostasis. They are implicated in almost every aspect of normal developmental processes, such as cell proliferation, cell migration, apoptosis, axonal growth, and synaptogenesis. They also contribute to a wide variety of neuropathology, including neurodegenerative diseases (e.g., Alzheimer’s disease, Parkinson’s disease), demyelination diseases (e.g., multiple sclerosis), brain tumors, and cerebral ischemia. For each family of proteases, there exist specific endogenous inhibitors whose basic function is to restrain protease activities. Imbalance between the strength of proteases and their inhibitors compromises neural homeostasis and results in neuropathological changes. In this chapter, we provide an overview of recent research progresses on plasminogen activators, plasmin(ogen), and their inhibitors in the CNS with specific emphasis on (1) their unique roles in neurophysiology and behavior, (2) factors that regulate their functions, and (3) what is known about the potential mechanisms through which these proteases and their inhibitors contribute to several CNS diseases.

II. Plasminogen Activators Plasminogen activators (PAs) are best known as thrombolytics, because they dissolve blood clots in the vasculature. They specifically cleave the Arg–Val bond in the zymogen plasminogen to generate the active protease plasmin. Plasmin is then able to digest the fibrin polymers in the blood clots. In the last few decades, with the generation of gene-targeted mice, deficient in one or more components of the fibrinolytic system (Bugge et al., 1995a,b, 1996; Carmeliet et al., 1994; Dewerchin et al., 1996; Ploplis et al., 1995), a rapidly growing list of cellular functions outside of the bloodstream has been attributed to these proteins, especially in the CNS. For example, in addition to its traditional substrate fibrin, plasmin has a wide spectrum of substrates, including most of the extracellular matrix (ECM) proteins in the CNS, such as fibronectin and laminin. Therefore, activation of plasminogen by PAs serves to initiate a potent proteolytic cascade leading to ECM degradation

5. Extracellular Proteases

163

and remodeling. There are two types of immunologically distinct PAs, tPA, and urokinase-type plasminogen activator (uPA). They are encoded by diVerent genes and share similar enzymology, but diVer in their domain organization and properties of the noncatalytical regions (Vassalli et al., 1991). It has been shown by others and us that tPA is the predominant PA constitutively expressed in normal CNS (Carroll et al., 1994; Qian, 1993; Sappino et al., 1993; Tsirka et al., 1997; Ware et al., 1995), whereas uPA in the CNS may be more relevant in the context of brain tumor biology (Levicar et al., 2003; Mohanam et al., 1994).

A. Tissue-Type Plasminogen Activator tPA belongs to the serine protease family of the ECM proteases. It has a catalytic site consisting of a serine, histidine, and an aspartic acid residue, and uses the serine residue for nucleophilic catalysis. Structurally, tPA starts with an N-terminal finger domain homologous to the fibrin-binding fingers of fibronectin, followed by a domain homologous to the epidermal growth factor domain (EGF) and two triple-disulfide structures called kringle domains, and ends with a C-terminal catalytic domain homologous to trypsin-like proteases (Dobrovolsky and Titaeva, 2002) (Fig. 1). The presence of multiple noncatalytic domains of tPA is critical in mediating

Figure 1 Domain structure of tPA, uPA, and plasminogen. tPA consists of the finger-like domain of fibronectin, epidermal growth factor domain, two kringles, and the catalytic domain. uPA consists of the epidermal growth factor domain, followed by the kringle and catalytic domains. Plasminogen consists of five kringle domains and the catalytic domain. Human plasminogen is cleaved at the Arg561–Val562 peptide bond by tPA or uPA to generate the active protease plasmin.

164

Zhang et al.

protein–protein and cell–cell interactions. In the mature brain parenchyma, tPA is produced by neurons and microglia with the most prominent expression in the hippocampus and hypothalamus (Carroll et al., 1994; Qian, 1993; Sappino et al., 1993; Tsirka et al., 1997; Ware et al., 1995); it is stored intracellularly and can be secreted upon membrane depolarization (Gualandris et al., 1996). tPA has been implicated in normal CNS physiology, such as neuronal migration, long-term potentiation associated with learning and memory, and neurite outgrowth. In 1981, Krystosek and Seeds first reported the implication of t-PA in neurite growth using diVerentiated lines of neuroblastoma cells. A fibrin overlay assay revealed that the predominant site of t-PA activity was on the growth cones. During embryogenesis, increased t-PA expression coincides with extensive cell migration, proliferation, and tissue remodeling in the CNS (Friedman and Seeds, 1995). The generation of mice in which the tPA gene (tPA/) was rendered nonfunctional has opened the door to new studies on the tPA-dependent physiological and pathological events in the CNS (Carmeliet et al., 1994). tPA/ mice showed delayed migration of cerebral granule neurons in the developing cerebellum (Seeds et al., 1999). Later, when these mice reached adulthood, they exhibited significantly impaired cerebellar motor learning similar to that caused by specific tPA inhibitors such as endogenous plasminogen activator inhibitor (PAI), type 1 plasminogen activator inhibitor (PAI-1), or a synthetic t-PA inhibitor, t-PA stop (Seeds et al., 2003). tPA/ mice also displayed a selective reduction in the late phase of the phenomenon of long-term potentiation (Huang et al., 1996). Similarly, the administration of inhibitors of t-PA proteolytic activity inhibits the long-term potentiation induced in the mossy fibers of the hippocampus (Baranes et al., 1998). A recent report from our group described the direct evidence of tPA’s involvement in hippocampal mossy fiber outgrowth in vivo, a commonly used animal model for human temporal lobe epilepsy (Wu et al., 2000). tPA/ mice presented with decreased and disarrayed sprouts, due to the lack of processing of an ECM protein called DSD-1-PG/phosphacan. This protein is a potent regulator of neurite outgrowth in the CNS. The accumulation of uncleaved DSD-1-PG/ phosphacan in tPA/ mice interferes with the appropriate neurite extension and termination. In a PC12 cell culture model, it was demonstrated that the expression level of neuroserpin, a serine protease inhibitor widely expressed in developing and mature brain, inversely correlates with the number and length of neurites extending from the cells upon growth factor treatment (Parmar et al., 2002). In summary, the available evidence indicates that t-PA is central to the regulation of neuroplasticity in both the developing and adult brain. The actions of tPA could be aVecting several pathways, such as degradation of matrix and cell adhesions via the plasmin–metalloproteases cascade to facilitate movement, activation of a cell-signaling event via

5. Extracellular Proteases

165

ligand–receptor binding, activation of pro-growth factors, liberation of matrix-associated cytokines, and promotion of cell adhesion mediated via PA–PAI interactions and integrin (Seeds et al., 1997). On the other hand, exaggerated tPA activity contributes to pathological processes such as neurodegeneration and inflammation (Siao and Tsirka, 2002b). Expression of tPA is quickly upregulated in the CNS in response to excitotoxic insults that mimic several clinical conditions, including stroke, Alzheimer’s disease, Parkinson’s disease, multiple sclerosis, and amyotrophic lateral sclerosis. tPA/ mice are more resistant to excitotoxin-induced neuronal death in the hippocampus (Tsirka et al., 1995), and the neuronal sensitivity can be restored when exogenous tPA protein is infused back into the brains of these mice prior to the delivery of excitotoxin (Tsirka et al., 1997). Plasminogen-deficient (plg/) mice exhibit resistance to excitotoxicity similar to that of tPA/ mice (Tsirka et al., 1997), suggesting that the tPA–plasmin proteolytic cascade promotes neuronal death acutely. Destruction of the supportive ECM substratum appears as one potential mechanism, because plasmin has been shown to degrade ECM laminin in a time-dependent manner preceding neuronal death (Chen and Strickland, 1997) (Fig. 2). Another mechanism is proposed as tPA potentiates signaling through glutamatergic receptors by proteolytically regulating the function of N-methyl-D-aspartate (NMDA) receptor and thus increasing calcium influx (Nicole et al., 2001) (see Fig. 2). tPA-mediated neurodegeneration also occurs through mechanisms independent of its proteolytic activity, such as through microglial activation (Tsirka, 2002). Microglia are CNS immunocompetent cells of the monocyte lineage. Upon insults, they are stimulated from their usual ramified-shaped resting state to undergo the process of activation (Giulian and Baker, 1986). The state of microglial activation is presented by morphological changes to ameboid shape, cell migration toward injury site followed by local proliferation, and upregulation of gene expression, all of which lead to increased capacity of antigen presentation and phagocytosis. Activated microglia increase the production of cytokines, proteases, reactive oxygen species, and nitric oxide (Kreutzberg, 1996; Lipton and Rosenberg, 1994). Blocking excessive microglial activation can confer protection against neurotoxicity in diVerent injury models (Rogove and Tsirka, 1998; Thanos et al., 1993), suggesting that overly activated microglia are detrimental to neuronal wellbeing. tPA/ mice exhibit attenuated microglial activation in response to excitotoxic stimuli (Tsirka et al., 1997); lipopolysaccharide-induced activation of cultured tPA/ microglia is dampened as well (Siao and Tsirka, 2002a). However, microglial activation in plg/ mice is comparable to that of wild-type animals (Tsirka et al., 1997). In addition, catalytically inactive tPA (S478A tPA, recombinant human tPA with a S478 ) A478 mutation at the active site) can activate microglia just as the wild-type protein does,

166

Zhang et al.

Figure 2 Pro-survival and anti-survival properties of tPA. tPA modulates neuronal well-being via distinct but intercalated pathways. It promotes neuronal survival via adequate microglial activation, and counteracting zinc toxicity. Meanwhile, it precipitates neuronal death via disruption of ECM substratum, potentiation of glutamatergic toxicity, and exaggerated microglial activation. Balance between the two opposing forces determines the final fate of neurons.

indicating that tPA activates microglia via a nonproteolytic mechanism (Rogove et al., 1999). It turned out that tPA mediates microglial activation via its finger domain through interaction with annexin II on the microglial cell surface (Siao and Tsirka, 2002a). How this interaction is translated into intracellular signaling events is currently under intensive investigation. Because tPA comes from both neurons and microglia, one important issue is to assess how tPA produced from distinct cellular origins is coordinated in the context of microglial activation and neurodegeneration. By introducing

5. Extracellular Proteases

167

tPA transgenes under the control of neuronal- or microglial-specific promoters into tPA/ mice, Siao et al. (2003) compared the outcome of excitotoxic damage in transgenic mice with selective tPA expression in neurons or microglia only. The neuronal tPA-expressing mice exhibit accelerated microglial activation but slower progression of neuronal death, whereas microglial tPA-expressing mice exhibit greater neurodegeneration. Based on these findings, a working hypothesis was put forward that tPA, initially released from injured neurons, acts as a cytokine to activate microglia at the site of injury, perhaps with the intention to restrain and clear tissue damage. These activated microglia then secrete additional tPA, which promotes microglial activation and ECM degradation beyond controllable/ regulatable level, and eventually promotes neurodegeneration (Siao et al., 2003) (see Fig. 2). Surprisingly, the story between tPA and neuronal integrity did not end with tPA’s characterization as the accomplice of neuron killing. Using a cell culture system, Kim et al. (1999) reported a neuroprotective role of tPA against zinc-induced toxicity through a nonproteolytic mechanism. Later, this eVect was confirmed in vivo as well (Siddiq and Tsirka, 2004). Zinc is an abundant trace element in the CNS, required by many proteins for their normal functions (such as MMPs). It exists in two interconvertible states, a protein-bound form and a free form. High levels of free zinc can cause neurotoxicity via modulating postsynaptic glutamate receptors (Choi and Koh, 1998), which is similar to that caused by excitotoxins (Olney, 1986). Despite the fact that tPA and zinc both contribute to neurodegeneration when acting separately, they attenuate each other’s toxicity when working together (Kim et al., 1999; Siddiq and Tsirka, 2004). Further exploration of this complex relationship was attempted (Siddiq and Tsirka, 2004). Zinc has been shown to inhibit tPA’s enzymatic activity. On the other hand, tPA can prevent the accumulation of extracellular free zinc by two ways. When zinc concentration is still low, tPA can bind to it and counteract its toxicity. When there is not enough tPA to buVer the rising level of zinc, tPA then facilitates zinc import into the cells, where zinc can be sequestered, although little is known about the intracellular destination of zinc. Considering the coexistence of tPA and zinc in both normal and diseased brains, there seems to be a point at which the balance between the two eVectors of cell death is finely tuned to optimize neuronal survival. Approaches to keep this delicate balance intact or restore its normal function may be of great benefit in preventing pathological complications after CNS injury (see Fig. 2). Recently, one plasminogen activator isolated from the saliva of vampire bat Desmodus rotundus (DSPA1) (Kratzschmar et al., 1991), which shares over 72% amino acid sequence identity with human tPA, was found to be free of neurotoxicity while retaining full strength of fibrinolytic capacity

168

Zhang et al.

(Liberatore et al., 2003). When DSPA1 was intracerebrally infused into tPA/ mice, no excitotoxin-induced neurodegeneration was observed as seen in tPA-infused animals, even when DSPA1 was used at 10-fold the concentration of infused tPA. DSPA1 did not restore microglial activation in tPA/ mice either. In addition, DSPA1 did not exacerbate NMDAmediated neuronal death in wild-type mice like tPA did. The unique property of DSPA1 is likely due to its peculiar dependence on fibrin (Bringmann et al., 1995; Toschi et al., 1998). Its catalytic eYciency increases 102,000-fold in the presence of fibrin yet only 8-fold by fibrinogen, whereas the catalytic eYciency of tPA is specifically enhanced only 72-fold by fibrin (Bringmann et al., 1995). Therefore, DSPA1 will be specifically activated at the site of fibrin deposition without causing generalized proteolytic tissue damage. Because of this major advantage, there is an ongoing clinical trial in Europe using DSPA1 in patients suVering from acute cerebral ischemia (Liberatore et al., 2003).

B. Urokinase-Type Plasminogen Activator uPA is a trypsin-like protease first isolated from the urine. It exists either in a 54-kD single-chain form or as a two-chained protein linked by an interchain disulfide bond. The domain structure of uPA is quite similar to that of tPA, consisting of an N-terminal EGF-like domain followed by a kringle domain, and a C-terminal domain homologous to trypsin-like proteases (Dobrovolsky and Titaeva, 2002) (see Fig. 1). Histidine, aspartic acid, and serine residues form the catalytic site of uPA. Nascently synthesized and secreted uPA is a proenzyme with little or no activity, which can be activated after being cleaved by various proteases, including its immediate substrate plasmin, thereby generating a positive feedback loop of selfactivation (Levicar et al., 2003). In contrast to the ambiguous identity of a cell surface-binding partner for tPA, it has been clearly shown that a specific receptor for uPA (uPAR) is expressed on the surface of many cell types (Roldan et al., 1990). uPAR is a 65-kD glycoprotein consisting of three homologous extracellular domains (D1–3) and is covalently bound to the cell membrane at the C-terminus via a glycosylphosphatidylinositol (GPI) anchor (Mondino et al., 1999). The uPA–uPAR interaction occurs between the EGF-like domain of uPA and the N-terminus of the receptor with high specificity and aYnity. This ligand–receptor binding induces conformational changes in uPAR, which is then able to interact with integrins and ECM protein vitronectin. Therefore, uPA–uPAR interaction results in the formation of a multiprotein complex that facilitates local proteolysis, cell adhesion, and migration. Considering the necessity of cell movement and tissue invasion during tumor progression, it is not surprising to see that uPA–uPAR is closely associated with the growth and dissemination of

5. Extracellular Proteases

169

various tumors, including those that originate in the brain (Mohanam et al., 1994; Schmitt et al., 1997). It was first noted that mRNA and protein levels of uPA and uPAR are dramatically elevated in malignant brain tumors compared with normal brain tissues and low-grade tumors (Caccamo et al., 1994; Kinder et al., 1993; Landau et al., 1994; Yamamoto et al., 1994a). The expression is localized in tumor cells and in endothelial cells of the surrounding neovasculature. In addition, higher expression is consistently present at the invasive edge of tumors. Later, a positive correlation was observed between uPA–uPAR expression levels and tumor invasiveness and recurrence, which are associated with worse prognosis (Hsu et al., 1995; Zhang et al., 2000). The causal relationship between high uPA–uPAR level and tumor invasiveness was confirmed in culture using several brain tumor cell lines of diVerent grades (MacDonald et al., 1998). The cells were evaluated for surface uPAR expression, endogenous uPA activity, and capacity to degrade ECM judged by migration on Transwell membranes and invasion of Matrigel. High levels of uPAR and uPA activity correlate with cellular degradation of ECM, cell migration, and Matrigel invasion. Cell migration and invasion were enhanced by exogenously added uPA in a dose-dependent manner. These eVects can be abolished by disruption of uPA–uPAR interaction at the cell surface with removal of membrane-bound uPAR. A working model was therefore formed, because uPA–uPAR binding on the surface of malignant cells is directly involved in the activation of ECM proteolytic cascades responsible for the invasiveness of those cells. Thereafter, blocking uPA–uPAR interaction is being actively pursued as a novel strategy to inhibit growth and spread of malignant brain tumors that are refractory to conventional therapies (Reuning et al., 2003). Downregulation of uPAR levels by antisense RNA inhibits glioblastoma cell migration and invasion in vitro (Mohan et al., 1999; Mohanam et al., 1997); it also reduces glioblastoma formation and causes regression of preexisting tumor in vivo (Mohan et al., 1999). Ligands that specifically target the overexpressed uPAR on glioblastoma multiforme can cause regression in tumor growth by blocking angiogenesis, decreasing tumor cell proliferation, and increasing tumor cell apoptosis (Bu et al., 2004; Mohanam et al., 2002; Rustamzadeh et al., 2003). These seemingly promising strategies need yet to be tested in a larger scale of animals and examined for their long-term eYcacy and side eVects. Given the fact that uPA and uPAR are synthesized by many cell types throughout the body and are critical in maintaining ECM turnover, it is worrisome that nonspecific interruption of their physiological functions might cause harmful consequences. In addition, information about tumor behaviors in mice deficient in uPA, uPAR, or both is not yet available; a thorough evaluation of these knockout mice will enrich our knowledge of the uPA–uPAR system in tumorigenesis and may disclose compensatory mechanisms in promoting tumor growth, invasion, and metastasis in the absence of uPA and uPAR.

170

Zhang et al.

III. Plasmin(ogen) Plasminogen is the precursor of serine protease plasmin. It is a 92-kD glycoprotein consisting of seven structural domains: a N-terminal preactivation peptide, five kringle domains, and a C-terminal catalytic domain (see Fig. 1). The multiple kringle domains are important for substrate recognition, membrane association, and inhibitor binding (Syrovets and Simmet, 2004). Plasminogen can be activated by both tPA and uPA, as well as by kallikrein and several coagulation factors. The activation of plasminogen involves the cleavage of the Arg561–Val562 peptide bond, generating plasmin in the form of a two-chain protein linked by a disulfide bond (Dobrovolsky and Titaeva, 2002). In contrast to the dual cellular sources of tPA in the CNS, plasminogen is exclusively synthesized by neurons, with high expression in the hippocampus, cerebellum, and cerebral cortex (Sappino et al., 1993; Tsirka et al., 1997; Zhang et al., 2002a). The separate origins of tPA and plasminogen are considered to be a self-protecting mechanism to avoid intracellular proteolysis (Sandgren et al., 1991; Tsirka et al., 1997). Plasmin has a broad spectrum of substrates, including the ECM components laminin and fibronectin. More importantly, plasmin also activates MMPs, which can degrade other ECM proteins. Therefore, the two protease systems work in concert to promote ECM proteolytic degradation and remodeling (Lijnen, 2001a). Plasmin has been shown to play an important role in both physiological and pathological processes in the CNS (Syrovets and Simmet, 2004). Plasmin is one major prosecutor in tPA-mediated neurodegeneration. Plg/ mice are resistant to excitotoxic injuries, because in these mice ECM substratum for neuronal survival is not disturbed due to the lack of plasmin activity (Chen and Strickland, 1997; Tsirka et al., 1997). Some recent studies on prion diseases demonstrate that the pathological form of prion protein can bind to both tPA and plasminogen, and it stimulates tPA-catalyzed plasmin generation (Epple et al., 2004; Fischer et al., 2000); in turn, tPA accelerates the cleavage of prion protein by plasmin (Kornblatt et al., 2003). The significance of these interactions is still under investigation. It was speculated that ECM degradation initiated by the tPA–plasmin system may contribute to the pathogenesis of this group of degenerative diseases (Bass and Ellis, 2002), or elimination of the infectious moiety of prion protein via plasmin cleavage may prevent disease propagation (Kornblatt et al., 2003). On the other hand, plasmin appears to be neuroprotective in the case of Alzheimer’s disease (AD). Plasmin has been shown to degrade amyloid protein and block its neurotoxicity, and there are decreased levels of plasmin in the brain of AD patients (Ledesma et al., 2000; Melchor et al., 2003; Tucker et al., 2000). Plasmin enriched in lipid rafts of neurons can clear excessive amyloid deposition. However, such rafts are disorganized in AD

5. Extracellular Proteases

171

patients; therefore, the plasmin-mediated amyloid clearance pathway is crippled (Ledesma et al., 2003). The seemingly opposing roles of plasmin in neuronal survival versus death emphasize the functional complexity of this protease that is determined by particular cellular settings and various stimuli. Plasmin has also been shown to work as a downstream eVector of tPA to promote neurite outgrowth. This function is attributed to the remodeling of ECM substratum by tPA–plasmin-initiated proteolytic cascade. Previously we reported that the tPA/plasmin system is involved in the regulation of hippocampal mossy fiber outgrowth after excitotoxin-stimulation of amygdala (Wu et al., 2000). This regulation consists of two diVerent events: neurite pathfinding through the supragranular layer, which is independent of the activation of plasminogen by tPA; and the termination of neurite outgrowth, which is mediated by proteolyic cleavage of the chondroitin sulfate proteoglycan phosphacan by plasmin. Unprocessed phosphacan is a potent repellent of mossy fiber outgrowth in culture, presumably by opposing cell–cell and cell–matrix interactions mediated by neural cell adhesion molecules. Recent studies showed that another chrondroitin sulfate proteoglycan, NG2, binds to the kringle domains of plasminogen and enhances plasmin generation by uPA (Goretzki et al., 2000) and tPA (our unpublished observations). NG2 is inhibitory to neurite outgrowth both in vivo and in vitro. By analogy with phosphacan cleavage by plasmin, one can speculate that plasmin may process NG2 to alter its suppressive eVect on neurite outgrowth as well.

IV. Inhibitors of Plasminogen Activators and Plasmin Serpins (serine protease inhibitor) are suicide substrate-like inhibitors of serine proteases (Silverman et al., 2001). The family includes PAIs, nexin-1 (PN-1), neuroserpin, and 2 antiplasmin. They act by binding to the active site of target proteases, which involves docking of the inhibitor to the target protease, cleavage of the reactive center loop, and rapid translocation of the protease to the opposite pole of the inhibitor (Silverman et al., 2001). PN-1 is the first identified neural serpin, expressed by both neurons and glia (Gloor et al., 1986; Guenther et al., 1985; Mansuy et al., 1993; Reinhard et al., 1988). It is the most potent inhibitor of thrombin, but it can also inhibit tPA, uPA, and plasmin to a lesser extent. Transgenic mice with neuronal PN-1 overexpression show increased long-term potentiation (LTP) in the hippocampus and develop progressive disturbances of motor behavior and sensorimotor integration, whereas PN-1-deficient mice have decreased hippocampal LTP. Both overexpression and lack of PN-1 cause epileptic activity in vivo and in vitro due to an imbalance between the

172

Zhang et al.

excitatory and inhibitory synaptic transmission (Luthi et al., 1997; Meins et al., 2001). Neuroserpin is an axonally secreted inhibitor used preferentially against tPA, and less eVectively toward uPA and plasmin (Hastings et al., 1997; Osterwalder et al., 1996, 1998). Intracranial distribution of neuroserpin coincides well with that of tPA, suggesting that it is likely to be a critical regulator of tPA activity-mediated events in the CNS (Hastings et al., 1997). Mice with neuroserpin overexpression have decreased tPA enzymatic activity (Cinelli et al., 2001), which remains unchanged in neuroserpin-deficient mice (Madani et al., 2003). This phenotype could be explained by compensation from other tPA inhibitors for the lack of neuroserpin or undetected subtle changes of tPA activity at the synaptic level. Just as in the on-demand secretion of intracellularly stored tPA (Gualandris et al., 1996), neuroserpin is also rapidly released from the cells upon depolarization (Berger et al., 1999), making the pair of protease and inhibitor readily available in response to neuronal activity. Compared to the extensively studied functions of tPA in CNS physiology and pathology, there is limited information regarding the role of neuroserpin as the endogenous regulator of tPA’s activity in these processes (Yepes and Lawrence, 2004). Expression of neuroserpin is most prominent during neurogenesis in embryos and correlates with synaptic activity in adult brains (Muller and Griesinger, 1998; Osterwalder et al., 1996). It decreases the overall length and number of extending neurites during neuronal diVerentiation in culture (Parmar et al., 2002), suggesting that neuroserpin may modulate neuroplasticity by counterbalancing the action of tPA. Mice overexpressing neuroserpin and mice lacking neuroserpin both exhibit neophobic phenotype in explorative behaviors, suggesting that neuroserpin may modify emotional behaviors independent of tPA’s catalytic activity (Madani et al., 2003). Neuroserpin overexpression has been detected by DNA microarray in patients with chronic schizophrenia (Hakak et al., 2001), a condition often associated with emotional instabilities and behavioral disturbances (Lewis and Lieberman, 2000). On the other hand, mice overexpressing neuroserpin suVer from smaller infarcts after induction of ischemic stroke, along with an attenuation of microglial activation (Cinelli et al., 2001). In addition, treatment with neuroserpin alone or in combination with tPA significantly reduces brain lesions associated with solo tPA treatment (Yepes et al., 2000; Zhang et al., 2002b). This is consistent with earlier reports on decreased infarct volumes in tPA/ mice subjected to cerebral ischemia (Nagai et al., 1999; Wang et al., 1998). Therefore, neuroserpin assumes a neuroprotective role against ischemiainduced brain damage by blocking extravascular tPA proteolytic activity and microglial activation. Neuroserpin is also implicated in the prevention of seizure propagation (Yepes et al., 2002). Intrahippocampal delivery of neuroserpin markedly delays seizure progression to a similar level as seen

173

5. Extracellular Proteases /

/

in tPA mice, whereas plg mice develop seizure comparable to that seen in wild-type animals. This suggests that neuroserpin regulates tPAmediated seizure spreading via mechanisms independent of plasminogen. Neuroserpin’s interference with tPA-modulated glutamatergic synaptic transmission is certainly one interpretation for its antiseizure activity, because NR1 subunit of NMDA receptor is a newly discovered nonplasminogen substrate for tPA (Nicole et al., 2001). Mutations of the human neuroserpin gene have been linked to a newly recognized familial encephalopathy with neuroserpin inclusion bodies, an autosomal-dominantly inherited progressive dementia accompanied by myoclonus epilepsy (Davis et al., 1999; Takao et al., 2000). These mutations not only confer abnormal polymerization of neuroserpin but also compromise its inhibition of tPA (Belorgey et al., 2002). However, absence of seizure or any other signs of altered excitability in neuroserpin-null mice suggests that other tPA inhibitors may be involved and compensate for the defect, which is in line with the normal level of tPA activity in these animals (Madani et al., 2003). In summary, many roles of tPA within the CNS seem to bypass the activation of plasminogen, and neuroserpin is a key tPA opponent in these events. However, neuroserpin does share functional redundancy with other tPA inhibitors in certain context. There are three distinct PAIs, PAI-1 (previously known as endothelial inhibitor), PAI-2 (placental or monocyte/macrophage-derived inhibitor), and PAI-3 (urine-derived inhibitor) (Sprengers and Kluft, 1987). PAI-1 and PAI-2 both are specific inhibitors for tPA and uPA, but PAI-1 is the primary physiological regulator of tPA and uPA activity (Wind et al., 2002). Besides the inhibition of tPA and uPA, PAI-1 also binds to the ECM protein vitronectin with high aYnity, whereas PA-complexed PAI-1 does not. Lowdensity lipoprotein receptor-related protein (LRP) is another binding partner for PAI-1. However, the high-aYnity binding site for LRP is cryptic in free PAI-1, and it can be exposed upon PA–PAI-1 interaction. Formation of PA–PAI-1–LRP multiprotein complex is believed to facilitate endocytosis and degradation of PA and PAI-1 (Nykjaer et al., 1992; Stefansson et al., 1998). It is interesting to note that there is an upregulation of PAI-1 mRNA after restraint stress (Yamamoto et al., 2002), which may function to prevent anxiety development by blocking tPA activity, suggesting a role for PAI-1 in the regulation of stress-induced neuronal rewiring (Pawlak et al., 2003). Working against tPA-mediated neurodegeneration, PAI-1’s neuroprotective role was best studied in the context of transforming growth factor beta 1 (TGF 1) signaling (Buisson et al., 2003). It has been shown that TGF 1 protects cultured neurons from tPA-mediated NMDA excitotoxicity via a mechanism involving upregulation of astrocyte-derived PAI-1 driven by transcription factor Smad3 and ERK kinase activation (Buisson et al., 1998; Docagne et al., 2002; Gabriel et al., 2003). The same mechanism was

174

Zhang et al.

proven to be eVective in rescuing neuronal damage in animal models of cerebral ischemia as well (Docagne et al., 1999; Zhu et al., 2002); whereas deficiency in the PAI-1 gene was associated with a significant increase in infarct size (Nagai et al., 1999). Thus, restoration of tPA–PAI-1 balance merits further attention in battling excitotoxic neuronal injury. There is also a growing body of evidence that suggests a causal role for PAI-1 in tumor growth, vascularization, and metastasis (Wind et al., 2002). It is known that pericellular plasmin activity generated by the uPA–uPAR system is decisive for the degradation of ECM during tumor invasion (Levicar et al., 2003). It was therefore puzzling to see the presence of high PAI-1 amounts in human malignant astrocytic tumors rather than in the corresponding normal tissues (Sandstrom et al., 1999; Yamamoto et al., 1994b). In addition, high levels of tumor-associated PAI-1 were found to be correlated with a poor prognosis (Schmitt et al., 1997). This unexpected phenomenon led to a debate on whether PAI-1 is proinvasive or antiinvasive during tumorigenesis (Andreasen et al., 2000; Rakic et al., 2003; Wind et al., 2002). Accumulating evidence suggests that it may function both ways, being antiinvasive by blocking uPA–PAR-mediated ECM degradation, and being proinvasive by antiproteolytic and nonproteolytic mechanisms (Rakic et al., 2003; Wind et al., 2002). It has been shown that malignant cell invasion and angiogenesis are severely impaired in PAI-1/ mice, which can be restored to control levels by applying exogenous PAI-1 (Bajou et al., 1998), whereas angiogenesis is increased in mice overexpressing PAI-1 (McMahon et al., 2001). However, these eVects are critically dependent on the concentration of PAI1, which is proangiogenesis at low concentrations, but antiangiogenesis at supraphysiological concentrations (Bajou et al., 2004). The eVects also vary depending on the cellular source of PAI-1 (tumor cells vs host cells); tumorderived PAI-1 even at high levels cannot change the course of angiogenesis and tumor progression in PAI-1/ mice (Bajou et al., 2004). Several mechanisms underlying the multiple functionality of PAI-1 have been put forward (Levicar et al., 2003; Rakic et al., 2003; Wind et al., 2002). By antagonizing the PA–plasmin system-mediated matrix degradation, PAI-1 may preserve ECM integrity that serves as a supporting substratum for colonization by vascular endothelial cell and subsequent neovasculature assembly. Meanwhile, tumor cells may be trapped by the intact basement barrier in the ECM, leading to failure of migration. On the other hand, complex interactions between PAI-1 and multiple ECM proteins (vitronectin and integrins) control cell migration through a nonproteolytic pathway. Binding of PAI-1 and vitronectin blocks cell-attachment sites between vitronectin and its receptor integrin that are required for cell motility (Stefansson and Lawrence, 1996). uPA forms a complex with PAI-1, making vitronectin accessible to integrin, and cell migration is restored (Deng et al., 1996; Kjoller et al., 1997). Thereby, PAI-1 can control cell–matrix

5. Extracellular Proteases

175

interaction by regulating the accessibility of specific cell-adhesion sites. Together, these findings suggest that a precise balance between proteases and inhibitors may be essential for tumor growth and invasion. 2 Antiplasmin is the main physiological inhibitor of plasmin activity. It is secreted as a 70-kDa single-chain protein by hepatocytes (Dobrovolsky and Titaeva, 2002). It rapidly scavenges free uncomplexed plasmin by formation of an inactive 1:1 stoichiometric complex, ensuring a short half-life of plasmin in blood and tissue (Lijnen, 2001b). Although 2 antiplasmin expression in the CNS has not been documented, it is likely to be produced solely by brain resident cells in the absence of a compromised blood–brain barrier (BBB). With BBB breakdown, there may be an influx from the bloodstream. The main role of 2 antiplasmin in the CNS is in the regulation of plasmin activity to modulate neuronal survival in various injury paradigms. It has been shown that intracerebral infusion of 2 antiplasmin confers neuroprotection to excitotoxic damage in the mouse hippocampus (Tsirka et al., 1997) and rat striatum (Campbell et al., 2004). Infiltration of inflammatory cells into the lesioned site is greatly attenuated in the presence of 2 antiplasmin (Campbell et al., 2004). Paradoxically, it was reported that 2 antiplasmin-deficient mice display decreased infarct volume after focal cerebral ischemia, whereas plg/ mice suVer from larger infarcts (Nagai et al., 1999). Despite the excitotoxic modality in ischemic cell damage, it was speculated that these phenotypes are the result of an alternative pathway other than tPA–plasmin-initiated ECM degradation. This pathway involves exacerbated vascular occlusion due to lack of plasmin-mediated fibrin clearance, leading to expansion of the infarct area (Nagai et al., 1999).

V. tPA Effects on Rodent Behavior Following the initial reports that tPA is expressed in the mammalian rodent brain, and that neuronal activity in the rat and mouse hippocampus and cerebellum induces tPA mRNA expression (Carroll et al., 1994; Qian et al., 1993; Seeds et al., 1995), studies were initiated to investigate the role of tPA in tasks associated with hippocampal or cerebellar function, namely, spatial learning, and consolidation of memory and motor learning, respectively. The results do not delineate an undisputed role for tPA as far as higher cognitive functions are concerned (learning and memory), but associate tPA to structural and electrophysiological changes related to learning and occurring in relevant brain areas (Table I). Furthermore, little attention has been paid to more basic behavioral attributes of the tPA/ mouse. Specifically, it has been shown that tPA aVects the learning of tasks dependent on the cerebellum and the striatum. Following the learning of a complex motor task where the rats needed to transverse a runway by placing

Table I

Studies That Have Examined the Relationship of tPA to Learning, LTP, and mRNA Levels tPA/ Performance

tPAþ/þ Performance

LTP

Centonze et al., 2002 #

Qian et al., 1993 Increased mRNA 9 Baranes et al., 1998> = Frey et al., 1996 " > Huang et al., 1996 ; L-LTP

Morris water maze

Huang et al., 1996 No diVerences Huang et al., 1996 No diVerences Horwood et al., 2004 No diVerences Pawlak et al., 2002 #

Tasks and LTP

Barnes maze Radial arm maze (spatial version) Step down inhibitory avoidance Context conditioning

Striatum-related tasks Operant conditioning tasks

Madani et al., 1999 "

Calabresi et al., 2000 # Huang et al., 1996 No diVerences

Cerebellum-related tasks

tPA Overexpression Performance

Seeds et al., 1995 Increased mRNA Calabresi et al., 2000 # Huang et al., 1996 Horwood et al., 2001 # Horwood et al., 2004 Ripley et al., 2001

Arrow pointing down signifies decrease; arrow pointing up signifies increase.

5. Extracellular Proteases

177

their paws on pegs (a cerebellum-dependent task), tPA mRNA levels were increased in the Purkinje neurons of the cerebellum (Seeds et al., 1995). When a two-way active avoidance task was used (a striatum-dependent task where the mouse learns to move from one chamber to another to avoid a mild shock) tPA/ mice were shown to have impaired performance (Calabresi et al., 2000; Huang et al., 1996). tPA/ mice have also been found to have impaired inhibitory behavior in operant conditioning tasks where the mouse learns that a certain behavior on its part will be followed by a reward after varying intervals (Horwood et al., 2001, 2004; Ripley et al., 2001). Interestingly PAI-1/ mice showed deficits in acquiring the operant task similar to the ones of the tPA/ mice (Horwood et al., 2001). Furthermore, activation of the mossy fiber pathway in the hippocampal region of tPA/ mice resulted in aberrant mossy fiber outgrowth (Wu et al., 2000). Treatment of hippocampal cells with tPA aVects their morphological characteristics. When tPA inhibitors were used, the formation of perforated synapses (via an increased activation of NMDA receptors) was significantly impaired (NeuhoV et al., 1999). Treatment with tPA was also responsible for elongated mossy fibers, as well as the formation of synaptic varicosities in hippocampal cell cultures (Baranes et al., 1998). The same researchers also found that tPA enhances the late phase of long-term potentiation (L-LTP, a molecular phenomenon long believed to underlie learning and memory) in the hippocampal mossy fiber pathway. Similar eVects on the L-LTP of the hippocampal SchaVer collateral pathway were reported earlier in other studies (Frey et al., 1996; Huang et al., 1996). Impaired LTP has also been reported in the corticostriatal pathway of tPA/ mice (Centonze et al., 2002). Mice with a disruption in the tPA gene also appear to have impaired performance in learning tasks that are hippocampus related. For example, when such mice were trained in a context conditioning task (where the mouse is exposed to a mild shock in an experimental chamber on the first day of testing followed by the measurement of time, and the mouse remains ‘‘frozen’’ in the same environment 24 h later), their performance was inferior compared with that of wild-type mice (Calabresi et al., 2000). Notably, the same task did not result in significant diVerences in the hands of another group (Huang et al., 1996). It is possible that the discrepancy is due to genetic background diVerences between the two strains of mice. Another study used the step-down inhibitory avoidance task (where the mouse learns to associate the floor of the chamber with a mild shock as it steps down from a platform, and, following a delay, it is measured how long it takes for the mouse to step down from the same platform) to find that the tPA/ mice performance was significantly worse than that of the wild-type ones, as early as 90 min after training, and 1, 2, and 7 days later (Pawlak et al., 2002). In a study that used mice that were overexpressing tPA, the results

178

Zhang et al. /

were consistent with those of the tPA mouse studies. Specifically, these mice performed significantly better in the Morris water maze (where the mouse learns to find a hidden platform underneath the opaque water surface of a pool) and the homing holeboard task (where the mouse learns to find an escape hole on a maze that leads to its home cage) compared with wild-type mice; furthermore, the same animals exhibited increased hippocampal LTP potentiation at levels proportional to the tPA overexpression (Madani et al., 1999). On the other hand, two studies found no diVerences between tPA/ and wild-type mice in tasks testing spatial learning. The first study reported that tPA/ mice did not diVer from wild-type mice in the Morris water maze and the Barnes maze (where the mouse finds a hole out of 18 that leads to a safe box) tasks (Huang et al., 1996). However, in this case the extent to which the animals were back-crossed has been uncertain (and the variability of genetic material may contribute to the results). The second study using a spatial version of the radial arm maze (where the mouse learns to explore and consume a reward in each of the eight maze arms by visiting them for the fewest possible times) found no diVerences between tPA/ mice and wild-type mice (Horwood et al., 2004). Nonetheless, it seems that the increased interest shown for the learning and memory tasks resulted in the sporadic and nonsystematic testing of the fundamental motor, exploratory, and anxiety-related behaviors of tPA/ mice. The degree of exploration of an open field was shown to be the same between tPA/ mice and controls, except for the females of both groups, which had a lower number of movements (Huang et al., 1996). To determine if there were any biases present in the step-down avoidance task, one study examined the overall locomotion of the mice in their cage and anxiety-like behaviors in the elevated-plus maze (where the amount of time the mouse spends in open unprotected arms as opposed to closed ones is a measure of its anxiety) and found no diVerences between the two groups (Pawlak et al., 2002). A more detailed characterization of the tPA/ mice was made when horizontal and vertical activities in an open field were found to be similar and less than that of the wild-type mice, respectively (Calabresi et al., 2000). The same study reported that the tPA/ mice showed lower habituation and reactivity to spatial change rates in an open field, but did not diVer from wild-type mice when the reaction to change of an object was measured. Another notable finding was that following the application of chronic restraint stress, tPA/ mice exhibited less anxiety-like behaviors on the elevated-plus maze compared with the wild-type mice (Pawlak et al., 2003). It is becoming apparent from the previous studies that tPA has a role in learning and memory, but probably not a central one. Its eVects have been more prominent in tasks associated with the striatum and the cerebellum, but not the hippocampus. The tPA mRNA levels are increased following the

5. Extracellular Proteases

179

learning of diVerent tasks, indicating that tPA is released according to its characterization as an immediate early gene and the process for it to be replenished is getting underway. It is possible that tPA facilitates, via its substrates, the changes in the extracellular matrix necessary to accommodate the formation of new long-term memories. In this context, its facilitation of LTP would serve the same purpose. Specifically, the impaired L-LTP in the SchaVer collateral synapse of the hippocampus has been associated with deficits in hippocampus-related tasks. That would contradict the unimpaired performance of tPA/ in hippocampus-related tasks reported in the same studies (e.g., Baranes et al., 1998; Frey et al., 1996; Huang et al., 1996). However, first it is possible that experimental conditions (such as the temperature in an intact animal could compensate for such a defect) could have influenced the electrophysiological results. Second, L-LTP was not completely absent in the tPA/ slices, which could lead to the conclusion that it is present in the intact animal at a threshold high enough to produce the needed eVects for the formation of memories (Huang et al., 1996). Finally, the findings that tPA/ mice exhibit impaired behavioral inhibition while learning an operational task (Ripley et al., 2001) and limited anxietylike behavior following chronic constraint (Pawlak et al., 2003) warrant further study.

VI. uPA and Plasminogen Effects on Rodent Behavior Research examining a possible role for uPA and plasminogen in mouse behavior has been even more sporadic than that examining tPA. It has been reported that mice overexpressing uPA consume less food, weigh less, and live longer (Miskin and Masos, 1997). The same research group had reported earlier that these transgenic animals were impaired in their learning of the Morris water maze, as well as in a conditioned taste aversion task, where the mice learn to associate a novel taste with malaise (Meiri et al., 1994). uPA/ mice have been reported to not diVer from wild-type mice in acquiring an operant conditioning task (Horwood et al., 2001). Plasminogendeficient mice exhibit increased stress response of grooming in an open field task and impaired response in the acoustic startle response task, where the mouse’s flinch in response to a loud noise is measured (Hoover-Plow et al., 2001). The same study did not find any diVerences between the groups in the acquisition of the Morris water maze task. A study that examined mice that either overexpress or are deficient of neuroserpin, an inhibitor of tPA and uPA, reported that both groups displayed anxiety-like behavior in an open field task and neophobic responses toward novel objects (Madani et al., 2003). Clearly, more studies are warranted that will examine in a systematic manner the behavioral phenotypes of uPA and plasminogen-deficient mice,

180

Zhang et al.

as well as those of their inhibitors. The previously mentioned studies are hard to interpret in their isolation, but even with them the theme that seems to emerge is that performance in learning and memory tasks related to the hippocampus is not impaired.

VII. Conclusions It is evident that proteases and their inhibitors play a critical role in mammalian neurophysiology and neuropathology. Tightly regulated balance between proteases and their inhibitors is vital to normal CNS function and survival, whereas pathological conditions are inevitable when control is lost over various CNS injuries. A thorough understanding of regulatory mechanisms that maintain or disrupt this delicate balance will provide us with codes to decipher its role in CNS activity and novel therapeutic strategies for related CNS diseases.

Acknowledgments The authors thank members of the Tsirka laboratory for helpful discussions. This work was supported by an NIH grant to S. E. T.

References Andreasen, P., Egelund, R., and Petersen, H. (2000). The plasminogen activation system in tumor growth, invasion, and metastasis. Cell. Mol. Life Sci. 57, 25–40. Bajou, K., Maillard, C., Jost, M., Lijnen, R., Gils, A., Declerck, P., Carmeliet, P., Foidart, J., and Noel, A. (2004). Host-derived plasminogen activator inhibitor-1 (PAI-1) concentration is critical for in vivo tumoral angiogenesis and growth. Oncogene 23, 6986–6990. Bajou, K., Noel, A., Gerard, R., Masson, V., Brunner, N., Holst-Hansen, C., Skobe, M., Fusenig, N., Carmeliet, P., Collen, D., and Foidart, J. (1998). Absence of host plasminogen activator inhibitor 1 prevents cancer invasion and vascularization. Nat. Med. 4, 923–928. Baranes, D., Lederfein, D., Huang, Y., Chen, M., Bailey, C., and Kandel, E. (1998). Tissue plasminogen activator contributes to the late phase of LTP and to synaptic growth in the hippocampal mossy fiber pathway. Neuron 21, 813–825. Bass, R., and Ellis, V. (2002). Cellular mechanisms regulating non-haemostatic plasmin generation. Biochem. Soc. Trans. 30, 189–194. Belorgey, D., Crowther, D. C., Mahadeva, R., and Lomas, D. A. (2002). Mutant neuroserpin (S49P) that causes familial encephalopathy with neuroserpin inclusion bodies is a poor proteinase inhibitor and readily forms polymers in vitro. J. Biol. Chem. 277, 17367–17373. Berger, P., Kozlov, S., Cinelli, P., Kruger, S., Vogt, L., and Sonderegger, P. (1999). Neuronal depolarization enhances the transcription of the neuronal serine protease inhibitor neuroserpin. Mol. Cell. Neurosci. 14, 455–467.

5. Extracellular Proteases

181

Bringmann, P., Gruber, D., Liese, A., Toschi, L., Kra¨ tzschmar, J., Schleuning, W.-D., and Donner, P. (1995). Structural features mediating fibrin selectivity of vampire bat plasminogen activators. J. Biol. Chem. 270, 25596–25603. Bu, X., Khankaldyyan, V., Gonzales-Gomez, I., Groshen, S., Ye, W., Zhuo, S., Pons, J., Stratton, J., Rosenberg, S., and Laug, W. (2004). Species-specific urokinase receptor ligands reduce glioma growth and increase survival primarily by an antiangiogenesis mechanism. Lab. Invest. 84, 667–678. Bugge, T., Flick, M., Danton, M., Daugherty, C., Romer, J., Dano, K., Carmeliet, P., Collen, D., and Degen, J. (1996). Urokinase-type plasminogen activator is eVective in fibrin clearance in the absence of its receptor or tissue-type plasminogen activator. Proc. Natl. Acad. Sci. USA 93, 5899–5904. Bugge, T., Flick, M., Daugherty, C., and Degen, J. (1995a). Plasminogen deficiency causes severe thrombosis but is compatible with development and reproduction. Genes Dev. 9, 794–807. Bugge, T. H., Suh, T. T., Flick, M. J., Daugherty, C. C., Romer, J., Solberg, H., Ellis, V., Dano, K., and Degen, J. L. (1995b). The receptor for urokinase-type plasminogen activator is not essential for mouse development or fertility. J. Biol. Chem. 270, 16886–16894. Buisson, A., Lesne, S., Docagne, F., Ali, C., Nicole, O., MacKenzie, E., and Vivien, D. (2003). Transforming growth factor-beta and ischemic brain injury. Cell. Mol. Neurobiol. 23, 539–550. Buisson, A., Nicole, O., Docagne, F., Sartelet, H., Mackenzie, E. T., and Vivien, D. (1998). Upregulation of a serine protease inhibitor in astrocytes mediates the neuroprotective activity of transforming growth factor 1. FASEB J. 12, 683–1691. Caccamo, D., Keohane, M., and McKeever, P. (1994). Plasminogen activators and inhibitors in gliomas: An immunohistochemical study. Mod. Pathol. 7, 99–104. Calabresi, P., Napolitano, M., Centonze, D., Marfia, G., Gubellini, P., Teule, M., Berretta, N., Bernardi, G., Frati, L., Tolu, M., and Gulino, A. (2000). Tissue plasminogen activator controls multiple forms of synaptic plasticity and memory. Eur. J. Neurosci. 12, 1002–1012. Campbell, S. J., Finlay, M., Clements, J. M., Wells, G., Miller, K. M., Perry, V. H., and Anthony, D. C. (2004). Reduction of excitotoxicity and associated leukocyte recruitment by a broad-spectrum matrix metalloproteinase inhibitor. J. Neurochem. 89, 1378–1386. Carmeliet, P., Schoonjans, L., Kieckens, L., Ream, B., Degen, J., Bronson, R., Vos, R. S., Oord, S. V. D., Collen, D., and Mulligan, R. (1994). Physiological consequences of loss of plasminogen activator gene function in mice. Nature 368, 419–424. Carroll, P., Tsirka, S., Richards, W., Frohman, M., and Strickland, S. (1994). The mouse tissue plasminogen activator gene 50 flanking region directs appropriate expression in development and a seizure-enhanced response in the CNS. Development 120, 3173–3183. Centonze, D., Napolitano, M., Saulle, E., Gubellini, P., Picconi, B., Martorana, A., Pisani, A., Gulino, A., Bernardi, G., and Calabresi, P. (2002). Tissue plasminogen activator is required for corticostriatal long-term potentiation. Eur. J. Neurosci. 16, 713–721. Chen, Z., and Strickland, S. (1997). Neuronal death in the hippocampus is promoted by plasmincatalyzed degradation of laminin. Cell 91, 917–925. Choi, D. W., and Koh, J. Y. (1998). Zinc and brain injury. Ann. Rev. Neurosci. 21, 347–375. Cinelli, P., Madani, R., Tsuzuki, N., Vallet, P., Arras, M., Zhao, C., Osterwalder, T., Rulicke, T., and Sonderegger, P. (2001). Neuroserpin, a neuroprotective factor in focal ischemic stroke. Mol. Cell. Neurosci. 18, 443–457. Davis, R., Shrimpton, A., Holohan, P., Bradshaw, C., Feiglin, D., Collins, G., Sonderegger, P., Kinter, J., Becker, L., Lacbawan, F., Krasnewich, D., Muenke, M., Lawrence, D., Yerby, M., Shaw, C., Gooptu, B., Elliott, P., Finch, J., Carrell, R., and Lomas, D. (1999). Familial dementia caused by polymerization of mutant neuroserpin. Nature 401, 376–379. Deng, G., Curriden, S., Wang, S., Rosenberg, S., and LoskutoV, D. (1996). Is plasminogen activator inhibitor-1 the molecular switch that governs urokinase receptor-mediated cell adhesion and release? J. Cell Biol. 134, 1563–1571.

182

Zhang et al.

Dewerchin, M., NuVelen, A. V., Wallays, G., Bouche, A., Moons, L., Carmeliet, P., Mulligan, R. C., and Collen, D. (1996). Generation and characterization of urokinase receptordeficient mice. J. Clin. Invest. 97, 870–878. Dobrovolsky, A., and Titaeva, E. (2002). The fibrinolysis system: Regulation of activity and physiologic functions of its main components. Biochemistry Mosc. 67, 99–108. Docagne, F., Nicole, O., Marti, H., MacKenzie, E., Buisson, A., and Vivien, D. (1999). Transforming growth factor- 1 as a regulator of the serpins/t-PA axis in cerebral ischemia. FASEB J. 13, 1315–1324. Docagne, F., Nicole, O., Gabriel, C., Fernandez-Monreal, M., Lesne, S., Ali, C., Plawinski, L., Carmeliet, P., MacKenzie, E., Buisson, A., and Vivien, D. (2002). Smad3-dependent induction of plasminogen activator inhibitor-1 in astrocytes mediates neuroprotective activity of transforming growth factor-beta 1 against NMDA-induced necrosis. Mol. Cell. Neurosci. 21, 634–644. Epple, G., Schleuning, W., Kettelgerdes, G., Kottgen, E., Gessner, R., and Praus, M. (2004). Prion protein stimulates tissue-type plasminogen activator-mediated plasmin generation via a lysine-binding site on kringle 2. J. Thromb. Haemost. 2, 962–968. Fischer, M., Roeckl, C., Parizek, P., Schwarz, H., and Aguzzi, A. (2000). Binding of diseaseassociated prion protein to plasminogen. Nature 408, 479–483. Frey, U., Muller, M., and Kuhl, D. (1996). A diVerent form of long-lasting potentiation revealed in tissue plasminogen activator mutant mice. J. Neurosci. 16, 2057–2063. Friedman, G. C., and Seeds, N. (1995). Tissue plasminogen activator mRNA expression in granule neurons coincides with their migration in the developing cerebellum. J. Comp. Neurol. 360, 658–670. Gabriel, C., Ali, C., Lesne, S., Fernandez-Monreal, M., Docagne, F., Plawinski, L., MacKenzie, E., Buisson, A., and Vivien, D. (2003). Transforming growth factor -induced expression of type 1 plasminogen activator inhibitor in astrocytes rescues neurons from excitotoxicity. FASEB J. 17, 277–279. Giulian, D., and Baker, T. (1986). Characterization of ameboid microglia isolated from developing mammalian brain. J. Neurosci. 6, 2163–2178. Gloor, S., Odink, K., Guenther, J., Nick, H., and Monard, D. (1986). A glia-derived neurite promoting factor with protease inhibitory activity belongs to the protease nexins. Cell 47, 687–693. Goretzki, L., Lombardo, C. R., and Stallcup, W. B. (2000). Binding of the NG2 proteoglycan to kringle domains modulates the functional properties of angiostatin and plasmin(ogen). J. Biol. Chem. 275, 28625–28633. Gualandris, A., Jones, T., Strickland, S., and Tsirka, S. (1996). Membrane depolarization induces calcium-dependent secretion of tissue plasminogen activator. J. Neurosci. 16, 2220–2225. Guenther, J., Nick, H., and Monard, D. (1985). A glia-derived neurite-promoting factor with protease inhibitory activity. EMBO J. 4, 1963–1966. Hakak, Y., Walker, J. R., Li, C., Wong, W. H., Davis, K. L., Buxbaum, J. D., Haroutunian, V., and Fienberg, A. A. (2001). Genome-wide expression analysis reveals dysregulation of myelination-related genes in chronic schizophrenia. PNAS 98, 4746–4751. Hastings, G. A., Coleman, T. A., Haudenschild, C. C., Stefansson, S., Smith, E. P., Barthlow, R., Cherry, S., Sandkvist, M., and Lawrence, D. A. (1997). Neuroserpin, a brain-associated inhibitor of tissue plasminogen activator is localized primarily in neurons. Implications for the regulation of motor learning and neuronal survival. J. Biol. Chem. 272, 33062–33067. Hoover-Plow, J., Skomorovska-Prokvolit, O., and Welsh, S. (2001). Selective behaviors altered in plasminogen-deficient mice are reconstituted with intracerebroventricular injection of plasminogen. Brain Res. 898, 256–264. Horwood, J., Ripley, T., and Stephens, D. (2001). DRL performance in mice with deletion of tPA, uPA or PAI-1 genes. Behav. Pharmacol. 12, 487–496.

5. Extracellular Proteases

183

Horwood, J., Ripley, T., and Stephens, D. (2004). Evidence for disrupted NMDA receptor function in tissue plasminogen activator knockout mice. Behav. Brain Res. 150, 127–138. Hsu, D., Efird, J., and Hedley-Whyte, E. (1995). Prognostic role of urokinase-type plasminogen activator in human gliomas. Am. J. Pathol. 147, 114–123. Huang, Y., Bach, M., Lipp, H., Zhuo, M., Wolfer, D., Hawkins, R., Schoonjans, L., Kandel, E., Godfraind, J., Mulligan, R., Collen, D., and Carmeliet, P. (1996). Mice lacking the gene encoding tissue-type plasminogen activator show a selective interference with late-phase longterm potentiation in both SchaVer collateral and mossy fiber pathways. Proc. Natl. Acad. Sci. USA 93, 8699–8704. Kim, Y.-H., Park, J.-H., Hong, S. H., and Koh, J.-Y. (1999). Nonproteolytic neuroprotection by human recombinant tissue plasminogen activator. Science 284, 647–650. Kinder, D., Berger, M., Mueller, B., and Silber, J. (1993). Urokinase plasminogen activator is elevated in human astrocytic gliomas relative to normal adjacent brain. Oncol. Res. 5, 409–414. Kjoller, L., Kanse, S., Kirkegaard, T., Rodenburg, K., Ronne, E., Goodman, S., Preissner, K., Ossowski, L., and Andreasen, P. (1997). Plasminogen activator inhibitor-1 represses integrinand vitronectin-mediated cell migration independently of its function as an inhibitor of plasminogen activation. Exp. Cell Res. 232, 420–429. Kornblatt, J., Marchal, S., Rezaei, H., Kornblatt, M., Balny, C., Lange, R., Debey, M., Hui Bon Hoa, G., Marden, M., and Grosclaude, J. (2003). The fate of the prion protein in the prion/plasminogen complex. Biochem. Biophys. Res. Commun. 305, 518–522. Kratzschmar, J., Haendler, B., Langer, G., Boidol, W., Bringmann, P., Alagon, A., Donner, P., and Schleuning, W. (1991). The plasminogen activator family from the salivary gland of the vampire bat Desmodus rotundus: Cloning and expression. Gene 105, 229–237. Kreutzberg, G. (1996). Microglia: A sensor for pathological events in the CNS. Trends Neurosci. 19, 312–318. Landau, B., Kwaan, H., Verrusio, E., and Brem, S. S. (1994). Elevated levels of urokinase-type plasminogen activator and plasminogen activator inhibitor type-1 in malignant human brain tumors. Cancer Res. 54, 1105–1108. Ledesma, M., Abad-Rodriguez, J., Galvan, C., Biondi, E., Navarro, P., Delacourte, A., Dingwall, C., and Dotti, C. (2003). Raft disorganization leads to reduced plasmin activity in Alzheimer’s disease brains. EMBO Rep. 4, 1190–1196. Ledesma, M., Da Silva, J., Crassaerts, K., Delacourte, A., De Strooper, B., and Dotti, C. (2000). Brain plasmin enhances APP alpha-cleavage and Abeta degradation and is reduced in Alzheimer’s disease brains. EMBO Rep. 1, 530–535. Levicar, N., Nuttall, R., and Lah, T. (2003). Proteases in brain tumour progression. Acta Neurochir. (Wien) 145, 825–838. Lewis, D., and Lieberman, J. (2000). Catching up on schizophrenia: Natural history and neurobiology. Neuron 28, 325–334. Liberatore, G. T., Samson, A., Bladin, C., Schleuning, W.-D., and Medcalf, R. L. (2003). Vampire bat salivary plasminogen activator (desmoteplase): A unique fibrinolytic enzyme that does not promote neurodegeneration. Stroke 34, 537–543. Lijnen, H. (2001a). Plasmin and matrix metalloproteinases in vascular remodeling. Thromb. Haemost. 86, 324–333. Lijnen, H. (2001b). Gene targeting in hemostasis. Alpha2-antiplasmin. Front. Biosci. 6, D239–247. Lipton, S. A., and Rosenberg, P. A. (1994). Excitatory amino acids as a final common pathway for neurologic disorders. N. Engl. J. Med. 330, 613–622. Luthi, A., Putten, H. V. D., Botteri, F. M., Mansuy, I. M., Meins, M., Frey, U., Sansig, G., Portet, C., Schmutz, M., Schroder, M., Nitsch, C., Laurent, J.-P., and Monard, D. (1997). Endogenous serine protease inhibitor modulates epileptic activity and hippocampal longterm potentiation. J. Neurosci. 17, 4688–4699.

184

Zhang et al.

MacDonald, T., DeClerck, Y., and Laug, W. E. (1998). Urokinase induces receptor mediated brain tumor cell migration and invasion. J. Neurooncol. 40, 215–226. Madani, R., Hulo, S., Toni, N., Madani, H., Steimer, T., Muller, D., and Vassalli, J. D. (1999). Enhanced hippocampal long-term potentiation and learning by increased neuronal expression of tissue-type plasminogen activator in transgenic mice. EMBO J. 18, 3007–3012. Madani, R., Kozlov, S., Akhmedov, A., Cinelli, P., Kinter, J., Lipp, H., Sonderegger, P., and Wolfer, D. (2003). Impaired explorative behavior and neophobia in genetically modified mice lacking or overexpressing the extracellular serine protease inhibitor neuroserpin. Mol. Cell. Neurosci. 23, 473–494. Mansuy, I., van der Putten, H., Schmid, P., Meins, M., Botteri, F., and Monard, D. (1993). Variable and multiple expression of protease nexin-1 during mouse organogenesis and nervous system development. Development 119, 1119–1134. McMahon, G. A., Petitclerc, E., Stefansson, S., Smith, E., Wong, M. K. K., Westrick, R. J., Ginsburg, D., Brooks, P. C., and Lawrence, D. A. (2001). Plasminogen activator inhibitor-1 regulates tumor growth and angiogenesis. J. Biol. Chem. 276, 33964–33968. Meins, M., Piosik, P., Schaeren-Wiemers, N., Franzoni, S., Troncoso, E., Kiss, J. Z., Brosamle, C., Schwab, M. E., Molnar, Z., and Monard, D. (2001). Progressive neuronal and motor dysfunction in mice overexpressing the serine protease inhibitor protease nexin-1 in postmitotic neurons. J. Neurosci. 21, 8830–8841. Meiri, N., Masos, T., Rosenblum, K., Miskin, R., and Dudai, Y. (1994). Overexpression of urokinase-type plasminogen activator in transgenic mice is correlated with impaired learning. Proc. Natl. Acad. Sci. USA 91, 3196–3200. Melchor, J., Pawlak, R., Chen, Z., and Strickland, S. (2003). The possible role of tissue-type plasminogen activator (tPA) and tPA blockers in the pathogenesis and treatment of Alzheimer’s disease. J. Mol. Neurosci. 20, 287–289. Miskin, R., and Masos, T. (1997). Transgenic mice overexpressing urokinase-type plasminogen activator in the brain exhibit reduced food consumption, body weight and size, and increased longevity. J. Gerontol. Biol. Sci. Med. Sci. 52, B118–B124. Mohan, P. M., Chintala, S. K., Mohanam, S., Gladson, C. L., So Kim, E., Gokaslan, Z. L., Lakka, S. S., Roth, J. A., Fang, B., Sawaya, R., Kyritsis, A. P., and Rao, J. S. (1999). Adenovirus-mediated delivery of antisense gene to urokinase-type plasminogen activator receptor suppresses glioma invasion and tumor growth. Cancer Res. 59, 3369–3373. Mohanam, S., Chandrasekar, N., Yanamandra, N., Khawar, S., Mirza, F., Dinh, D., Olivero, W., and Rao, J. (2002). Modulation of invasive properties of human glioblastoma cells stably expressing amino-terminal fragment of urokinase-type plasminogen activator. Oncogene 21, 7824–7830. Mohanam, S., Chintala, S., Go, Y., Bhattacharya, A., Venkaiah, B., Boyd, D., Gokaslan, Z., Sawaya, R., and Rao, J. (1997). In vitro inhibition of human glioblastoma cell line invasiveness by antisense uPA receptor. Oncogene 14, 1351–1359. Mohanam, S., Sawaya, R., Yamamoto, M., Bruner, J., Nicholson, G., and Rao, J. (1994). Proteolysis and invasiveness of brain tumors: Role of urokinase-type plasminogen activator receptor. J. Neurooncol. 22, 153–160. Mondino, A., Resnati, M., and Blasi, F. (1999). Structure and function of the urokinase receptor. Thromb. Haemost. 82, 19–22. Muller, C., and Griesinger, C. (1998). Tissue plasminogen activator mediates reverse occlusion plasticity in visual cortex. Nat. Neurosci. 1, 47–53. Nagai, N., De Mol, M., Lijnen, H. R., Carmeliet, P., and Collen, D. (1999). Role of plasminogen system components in focal cerebral ischemic infarction: A gene targeting and gene transfer study in mice. Circulation 99, 2440–2444. NeuhoV, H., Roeper, J., and Schweizer, M. (1999). Activity-dependent formation of perforated synapses in cultured hippocampal neurons. Eur. J. Neurosci. 11, 4241–4250.

5. Extracellular Proteases

185

Nicole, O., Docagne, F., Ali, C., Margaill, I., Carmeliet, P., MacKenzie, E., Vivien, D., and Buisson, A. (2001). The proteolytic activity of tissue-plasminogen activator enhances NMDA receptor-mediated signaling. Nat. Med. 7, 59–64. Nykjaer, A., Petersen, C., Moller, B., Jensen, P., Moestrup, S., Holtet, T., Etzerodt, M., Thogersen, H., Munch, M., and Andreasen, P. (1992). Purified alpha 2-macroglobulin receptor/LDL receptor-related protein binds urokinase plasminogen activator inhibitor type-1 complex. Evidence that the alpha 2-macroglobulin receptor mediates cellular degradation of urokinase receptor-bound complexes. J. Biol. Chem. 267, 14543–14546. Olney, J. (1986). Inciting excitotoxic cytocide among central neurons. Adv. Exp. Med. Biol. 203, 631–645. Osterwalder, T., Cinelli, P., Baici, A., Pennella, A., Krueger, S. R., Schrimpf, S. P., Meins, M., and Sonderegger, P. (1998). The axonally secreted serine proteinase inhibitor, neuroserpin, inhibits plasminogen activators and plasmin but not thrombin. J. Biol. Chem. 273, 2312–2321. Osterwalder, T., Contartese, J., Stoeckli, E., Kuhn, T., and Sonderegger, P. (1996). Neuroserpin, an axonally secreted serine protease inhibitor. EMBO J. 15, 2944–2953. Parmar, P., Coates, L., Pearson, J., Hill, R., and Birch, N. (2002). Neuroserpin regulates neurite outgrowth in nerve growth factor-treated PC12 cells. J. Neurochem. 82, 1406–1415. Pawlak, R., Magarinos, A., Melchor, J., McEwen, B., and Strickland, S. (2003). Tissue plasminogen activator in the amygdala is critical for stress-induced anxiety-like behavior. Nat. Neurosci. 6, 168–174. Pawlak, R., Nagai, N., Urano, T., Napiorkowska-Pawlak, D., Ihara, H., Takada, Y., Collen, D., and Takada, A. (2002). Rapid, specific and active site-catalyzed eVect of tissue-plasminogen activator on hippocampus-dependent learning in mice. Neuroscience 113, 995–1001. Ploplis, V. A., Carmeliet, P., Vazirzadeh, S., Van Vlaenderen, I., Moons, L., Plow, E. F., and Collen, D. (1995). EVects of disruption of the plasminogen gene on thrombosis, growth, and health in mice. Circulation 92, 2585–2593. Qian, Z., Gilbert, M., Colicos, M., Kandel, E., and Kuhl, D. (1993). Tissue-plasminogen activator is induced as an immediate-early gene during seizure, kindling and long-term potentiation. Nature 361, 453–457. Rakic, J., Maillard, C., Jost, M., Bajou, K., Masson, V., Devy, L., Lambert, V., Foidart, J., and Noel, A. (2003). Role of plasminogen activator-plasmin system in tumor angiogenesis. Cell. Mol. Life Sci. 60, 463–473. Reinhard, E., Meier, R., Halfter, W., Rovelli, G., and Monard, D. (1988). Detection of gliaderived nexin in the olfactory system of the rat. Neuron 1, 387–394. Reuning, U., Sperl, S., Kopitz, C., Kessler, H., Kruger, A., Schmitt, M., and Magdolen, V. (2003). Urokinase-type plasminogen activator (uPA) and its receptor (uPAR): Development of antagonists of uPA/uPAR interaction and their eVects in vitro and in vivo. Curr. Pharm. Des. 9, 1529–1543. Ripley, T., Horwood, J., and Stephens, D. (2001). Evidence for impairment of behavioural inhibition in performance of operant tasks in tPA/ mice. Behav. Brain Res. 125, 215–227. Rogove, A., Siao, C., Keyt, B., Strickland, S., and Tsirka, S. (1999). Activation of microglia reveals a non-proteolytic cytokine function for tissue plasminogen activator in the central nervous system. J. Cell Sci. 112, 4007–4016. Rogove, A. D., and Tsirka, S. E. (1998). Neurotoxic responses by microglia elicited by excitotoxic injury in the mouse hippocampus. Curr. Biol. 8, 19–25. Roldan, A., Cubellis, M., Masucci, M., Behrendt, N., Lund, L., Dano, K., Appella, E., and Blasi, F. (1990). Cloning and expression of the receptor for human urokinase plasminogen activator, a central molecule in cell surface, plasmin dependent proteolysis. EMBO J. 9, 467–474.

186

Zhang et al.

Rustamzadeh, E., Li, C., Doumbia, S., Hall, W., and Vallera, D. (2003). Targeting the overexpressed urokinase-type plasminogen activator receptor on glioblastoma multiforme. J. Neurooncol. 65, 63–75. Sandgren, E., Palmiter, R., Heckel, J., Daugherty, C., Brinster, R., and Degen, J. (1991). Complete hepatic regeneration after somatic deletion of an albumin-plasminogen activator transgene. Cell 66, 245–256. Sandstrom, M., Johansson, M., Sandstrom, J., Bergenheim, A., and Henriksson, R. (1999). Expression of the proteolytic factors, tPA and uPA, PAI-1 and VEGF during malignant glioma progression. Int. J. Dev. Neurosci. 17, 473–481. Sappino, A., Madani, R., Huarte, J., Belin, D., Kiss, J., Wohlwend, A., and Vassalli, J. (1993). Extracellular proteolysis in the adult murine brain. J. Clin. Invest. 92, 679–685. Schmitt, M., Harbeck, N., Thomssen, C., Wilhelm, O., Magdolen, V., Reuning, U., Ulm, K., Hofler, H., Janicke, F., and GraeV, H. (1997). Clinical impact of the plasminogen activation system in tumor invasion and metastasis: Prognostic relevance and target for therapy. Thromb. Haemost. 78, 285–296. Seeds, N., Basham, M., and Ferguson, J. (2003). Absence of tissue plasminogen activator gene or activity impairs mouse cerebellar motor learning. J. Neurosci. 23, 7368–7375. Seeds, N., Basham, M., and HaVke, S. (1999). Neuronal migration is retarded in mice lacking the tissue plasminogen activator gene. Proc. Natl. Acad. Sci. USA 96, 14118–14123. Seeds, N., Siconolfi, I., and HaVke, S. (1997). Neuronal extracellular proteases facilitate cell migration, axonal growth, and pathfinding. Cell Tissue Res. 290, 367–370. Seeds, N., Williams, B., and Bickford, P. (1995). Tissue plasminogen activator induction in Purkinje neurons after cerebellar motor learning. Science 270, 1992–1994. Siao, C., Fernandez, S., and Tsirka, S. (2003). Cell type-specific roles for tissue plasminogen activator released by neurons or microglia after excitotoxic injury. J. Neurosci. 23, 3234–3242. Siao, C., and Tsirka, S. E. (2002a). Tissue plasminogen activator mediates microglial activation via its finger domain through annexin II. J. Neurosci. 22, 3352–3358. Siao, C., and Tsirka, S. (2002b). Extracellular proteases and neuronal cell death. Cell Mol. Biol.(Noisy-le-grand) 48, 151–161. Siddiq, M., and Tsirka, S. (2004). Modulation of zinc toxicity by tissue plasminogen activator. Mol. Cell. Neurosci. 25, 162–171. Silverman, G. A., Bird, P. I., Carrell, R. W., Church, F. C., Coughlin, P. B., Gettins, P. G. W., Irving, J. A., Lomas, D. A., Luke, C. J., Moyer, R. W., Pemberton, P. A., Remold-O’Donnell, E., Salvesen, G. S., Travis, J., and Whisstock, J. C. (2001). The serpins are an expanding superfamily of structurally similar but functionally diverse proteins. Evolution, mechanism of inhibition, novel functions, and a revised nomenclature. J. Biol. Chem. 276, 33293–33296. Sprengers, E., and Kluft, C. (1987). Plasminogen activator inhibitors. Blood 69, 381–387. Stefansson, S., and Lawrence, D. (1996). The serpin PAI-1 inhibits cell migration by blocking integrin alpha V beta 3 binding to vitronectin. Nature 383, 441–443. Stefansson, S., Muhammad, S., Cheng, X.-F., Battey, F. D., Strickland, D. K., and Lawrence, D. A. (1998). Plasminogen activator inhibitor-1 contains a cryptic high aYnity binding site for the low density lipoprotein receptor-related protein. J. Biol. Chem. 273, 6358–6366. Syrovets, T., and Simmet, T. (2004). Novel aspects and new roles for the serine protease plasmin. Cell. Mol. Life Sci. 61, 873–885. Takao, M., Benson, M., Murrell, J., Yazaki, M., Piccardo, P., Unverzagt, F., Davis, R., Holohan, P., Lawrence, D., Richardson, R., Farlow, M., and Ghetti, B. (2000). Neuroserpin mutation S52R causes neuroserpin accumulation in neurons and is associated with progressive myoclonus epilepsy. J. Neuropathol. Exp. Neurol. 59, 1070–1086.

5. Extracellular Proteases

187

Thanos, S., Mey, J., and Wild, M. (1993). Treatment of the adult retina with microgliasuppressing factors retards axotomy-induced neuronal degradation and enhances axonal regeneration in vivo and in vitro. J. Neurosci. 13, 455–466. Toschi, L., Bringmann, P., Petri, T., Donner, P., and Schleuning, W. (1998). Fibrin selectivity of the isolated protease domains of tissue-type and vampire bat salivary gland plasminogen activators. Eur. J. Biochem. 252, 108–112. Tsirka, S. (2002). Tissue plasminogen activator as a modulator of neuronal survival and function. Biochem. Soc. Trans. 30, 222–225. Tsirka, S. E., Gualandris, A., Amaral, D. G., and Strickland, S. (1995). Excitotoxin-induced neuronal degeneration and seizure are mediated by tissue-type plasminogen activator. Nature 377, 340–344. Tsirka, S. E., Rogove, A. D., Bugge, T. H., Degan, J. A., and Strickland, S. (1997). An extracellular proteolytic cascade promotes neural degeneration in the mouse hippocampus. J. Neurosci. 17, 543–552. Tucker, H. M., Kihiko, M., Caldwell, J. N., Wright, S., Kawarabayashi, T., Price, D., Walker, D., ScheV, S., McGillis, J. P., Rydel, R. E., and Estus, S. (2000). The plasmin system is induced by and degrades amyloid-beta aggregates. J. Neurosci. 20, 3937–3946. Vassalli, J., Sappino, A., and Berlin, D. (1991). The plasminogen activator/plasmin system. J. Clin. Invest 88, 1067–1072. Wang, Y. F., Tsirka, S. E., Strickland, S., Stieg, P. E., Soriano, S. G., and Lipton, S. A. (1998). Tissue plasminogen activator (tPA) increases neuronal damage after focal cerebral ischemia in wild-type and tPA-deficient mice. Nat. Med. 4, 228–231. Ware, J., Dibenedetto, A., and Pittman, R. (1995). Localization of tissue plasminogen activator mRNA in adult rat brain. Brain Res. Bull. 37, 275–281. Wind, T., Hansen, M., Jensen, J., and Andreasen, P. (2002). The molecular basis for antiproteolytic and non-proteolytic functions of plasminogen activator inhibitor type-1: Roles of the reactive centre loop, the shutter region, the flexible joint region and the small serpin fragment. Biol. Chem. 383, 21–36. Wu, Y. P., Siao, C.-J., Lu, W., Sung, T.-C., Frohman, M. A., Milev, P., Bugge, T. H., Degen, J. L., Levine, J., Margolis, R. U., and Tsirka, S. E. (2000). The tPA/plasmin extracellular proteolytic system contributes to hippocampal mossy fiber reorganization through a novel proteoglycan substrate. J. Cell Biol. 148, 1295–1304. Yamamoto, K., Takeshita, K., Shimokawa, T., Yi, H., Isobe, K.-I., LoskutoV, D. J., and Saito, H. (2002). Plasminogen activator inhibitor-1 is a major stress-regulated gene: Implications for stress-induced thrombosis in aged individuals. PNAS 99, 890–895. Yamamoto, M., Sawaya, R., Mohanam, S., Rao, V., Bruner, J., Nicolson, G., and Rao, J. S. (1994a). Expression and localization of urokinase-type plasminogen activator receptor in human gliomas. Cancer Res. 54, 5016–5020. Yamamoto, M., Sawaya, R., Mohanam, S., LoskutoV, D., Bruner, J., Rao, V., Oka, K., Tomonaga, M., Nicolson, G., and Rao, J. (1994b). Expression and cellular localization of messenger RNA for plasminogen activator inhibitor type 1 in human astrocytomas in vivo. Cancer Res. 54, 3329–3332. Yepes, M., and Lawrence, D. (2004). Neuroserpin: A selective inhibitor of tissue-type plasminogen activator in the central nervous system. Thromb. Haemost. 91, 457–464. Yepes, M., Sandkvist, M., Coleman, T. A., Moore, E., Wu, J.-Y., Mitola, D., Bugge, T. H., and Lawrence, D. A. (2002). Regulation of seizure spreading by neuroserpin and tissue-type plasminogen activator is plasminogen-independent. J. Clin. Invest. 109, 1571–1578. Yepes, M., Sandkvist, M., Wong, M. K. K., Coleman, T. A., Smith, E., Cohan, S. L., and Lawrence, D. A. (2000). Neuroserpin reduces cerebral infarct volume and protects neurons from ischemia-induced apoptosis. Blood 96, 569–576.

188

Zhang et al.

Zhang, L., SeiVert, D., Fowler, B., Jenkins, G., Thinnes, T., LoskutoV, D., Parmer, R., and Miles, L. (2002a). Plasminogen has a broad extrahepatic distribution. Thromb. Haemost. 87, 493–501. Zhang, X., Fei, Z., Bu, X., Zhen, H., Zhang, Z., Gu, J., and Chen, Y. (2000). Expression and significance of urokinase type plasminogen activator gene in human brain gliomas. J. Surg. Oncol. 74, 90–94. Zhang, Z., Zhang, L., Yepes, M., Jiang, Q., Li, Q., Arniego, P., Coleman, T. A., Lawrence, D. A., and Chopp, M. (2002b). Adjuvant treatment with neuroserpin increases the therapeutic window for tissue-type plasminogen activator administration in a rat model of embolic stroke. Circulation 106, 740–745. Zhu, Y., Yang, G.-Y., Ahlemeyer, B., Pang, L., Che, X.-M., Culmsee, C., Klumpp, S., and Krieglstein, J. (2002). Transforming growth factor-beta 1 increases bad phosphorylation and protects neurons against damage. J. Neurosci. 22, 3898–3909.

6

____________________________________________________________________________

The Genetic Architecture of House Fly Mating Behavior Lisa M. Meffert and Kara L. Hagenbuch Department of Ecology and Evolutionary Biology Rice University Houston, Texas 77251-1892

I. Introduction II. Evolutionary Dynamics of Quantitative Genetic Interactions A. Dominance B. Epistasis C. Genotype-by-Environment Interactions and Learning D. Pleiotropy III. Courtship in the House Fly IV. Evidence of Quantitative Genetic Interactions in House Flies A. Assays of Additive Genetic Variances in Bottlenecked Populations B. Line Cross Analyses C. Repeatability Assays V. Prevalence of Nonadditive Genetic EVects in Animal Behavior VI. Future Directions VII. Summary Acknowledgments References

This chapter summarizes several experimental approaches used to identify the eVects of dominance, epistasis, and genotype-by-environment interactions in the genetic architecture of the mating behavior of the common house fly (Musca domestica L.). Quantitative genetic investigations of mating behavior hold special intrigue for unraveling the complexities of fitness traits, with applications to theory on sexual selection and speciation. Besides being well suited to large-scale quantitative genetic protocols, the house fly has a remarkably complex courtship repertoire, aVording special opportunities for studies on communication, social interactions, and learning. Increased additive genetic variances for the courtship repertoire of experimentally bottlenecked populations provided evidence for the presence of dominance and/or epistasis. Negative genetic variances in these populations suggested genotype-by-environment interactions, where the environment is the mating partner. Line cross assays of populations that had been subjected to selection for divergent courtship repertoire confirmed that both dominance and epistasis have significant eVects. These crosses also showed more directly that Current Topics in Developmental Biology, Vol. 66 Copyright 2005, Elsevier Inc. All rights reserved. 0070-2153/05 $35.00

189

190

MeVert and Hagenbuch

the expression of the male’s genotype is dependent upon the preferences of his mating partner. Repeatability studies also detailed how males alter their courtship performances with successive encounters within and across females, such that the males learn to improve their techniques in securing copulations. A review of 41 animal behavior studies found that a wide range of traits and taxa have dominance, epistasis, and genotype-y-environment interactions, although house fly courtship may remain a unique model where learning is an intersexually selected trait. Future development of more sophisticated molecular techniques for the M. domestica genome will help unravel the underlying biochemical and developmental pathways of these quantitative genetic interactions for a more complete understanding of the processes of inbreeding depression, outbreeding depression, and pleiotropy. C 2005, Elsevier Inc.

I. Introduction The most fundamental motive for the genetic dissection of house fly mating behavior is exploiting a model experimental system for understanding the architecture of fitness traits. Mating behavior, in particular, is intrinsically tied to overall reproductive success and thus serves as a major fitness component. Natural selection (along with genetic drift) is expected to erode the additive genetic variance of such fitness traits, resulting in characteristically low heritabilities (Fisher, 1958). Additionally, dominance and epistasis are expected to camouflage the residual components of additive genetic variance (Goodnight, 1988; Willis and Orr, 1993; and see later). Thus, the level of additive genetic variation can serve as a measure of historical selection pressure and future evolutionary potential (RoV and Mousseau, 1987). Moreover, the relative contributions of dominance and epistasis indicate the potential for the release of sequestered additive genetic variance by bottlenecks (see MeVert, 2000 for a review). The house fly is especially amenable to the quantitative genetic manipulations required for investigating the architecture of fitness traits. It shares many of the advantages of the well-studied Drosophila melanogaster, such as fast generation length, large family sizes, and simple culturing techniques. Its comparatively large body size, however, facilitates the large-scale parent– oVspring analyses, selection experiments, and repeatability assays necessary to partition genetic variance components. For example, the technical eVorts to sex virgins, videotape courtship, and collect oVspring from individual male– female pairs are less demanding when stereoscopes and transportation pooters are unnecessary. More importantly, the intricate male–female interactions in house fly courtship (see later) provide excellent opportunities for evaluating the genetics of particularly complex fitness traits, especially in terms of the evolutionary roots of communication, social behavior, and learning.

6. Quantitative Genetic Interactions in House Fly Courtship

191

House fly courtship also serves as a model system for more specific experimental tests of sexual selection theory. Sexual selection theory seeks to explain the evolution of traits that confer mating advantages even when the traits have negative consequences for other aspects of fitness (e.g., see Andersson, 1994; Price et al., 1993). In intersexual selection, male traits evolve when females discriminate among potential mates, driving the coordinated evolution of the female preferences themselves (e.g., Fisher, 1958; Lande, 1981). In this basic model, a genetic correlation generated as a daughter with its mother’s preference carries alleles for the desired male trait, while a son with the trait also carries alleles for the female preference (Lande, 1981). The reinforcing coevolution of preferences and male traits then proceeds until the additive genetic variance is exhausted or some equilibrium is achieved with antagonistic natural selection pressures (Lande, 1981). Much of the intrigue of intersexual selection involves the evolution of sexual dimorphisms, whereby the genome component expressed in one gender (typically the female) acts as a selective pressure on the other. Whether the sexual selection process is cooperative or antagonistic (termed run-away and chase-away, respectively) is very controversial, especially since both features can operate in the same system (e.g., Hicks et al., 2004). Again, the experimental advantages of the house fly oVer special opportunities to unravel the genetic underpinnings and evolutionary consequences of sexual selection. Theoretical work suggests that sexual selection can drive the formation of new species. In his landmark paper, Lande (1981) examined how divergent sexual selection responses (sensu Fisher, 1958) can generate reproductive isolation, and thus form new species. Moreover, founder-flush events have been proposed to open evolutionary avenues for such divergence by releasing additive genetic variance from nonadditive genetic structure (see MeVert, 1999, 2000 for reviews). Part of the extensive species radiation of the Hawaiian Drosophila, for example, has been attributed to founder-induced stimulation of divergent sexual selection process (e.g., Kaneshiro, 1980). There has been some debate about the likelihood and stability of such behavioral divergence (e.g., see Iwasa and Pomiankowski, 1995; Price, 1998). For example, Boake et al. (1997) noted that the species recognition pattern found in the D. heteroneura–D. silvestris system was unrelated to a sexually selected head-width trait. They contended that this was negative evidence for a continuum in the processes of sexual selection and speciation. Aspi (2000), however, held that the sexual selection–species recognition continuum does occur within the D. montana–D. littoralis system (see also Aspi and Hoikkala, 2000). These kinds of sexual selection studies rely on post-hoc interpretations of extant species, but the experimental accessibility of the house fly permits investigations of the initial stages of speciation via sexual selection and founder events (e.g., MeVert, 1999; MeVert and Regan, 2002).

192

MeVert and Hagenbuch

In this chapter, we first discuss the evolutionary significance of the quantitative genetic interactions of dominance, epistasis, genotype-by-environment interactions, and pleiotropy. We then detail the courtship repertoire of the house fly and summarize various experiments that have identified all of these dynamics, including a special form of genotype-by-environment interaction: learning. We also summarize the animal behavior literature to show that such complexities are common in a wide range of taxa and traits. Finally, we provide predictions for future explorations with this model system.

II. Evolutionary Dynamics of Quantitative Genetic Interactions Figure 1 depicts phenotypic deviations for the quantitative genetic models of pure additivity, pure dominance, additive-by-additive epistasis, and genotype-by-environment interactions. In each case, two representative loci are modeled (i.e., ‘‘X’’ and ‘‘Y’’), with each locus having two alleles (i.e., X–x and Y–y, respectively). The surfaces represent the phenotypic values for the nine possible genotypes under this simplified two-locus, two-allele system. Under pure additivity (i.e., no interaction), the phenotypic values simply summate across alleles and across loci (Fig. 1A). With pure dominance (i.e., one allele completely masks the eVects of another allele at the same locus), phenotypic plateaus are apparent across the wide range of genotypic values, compounded by the additive eVects across loci (Fig. 2B). Under epistasis, the interlocus eVects act synergistically, yielding like phenotypes for disparate genetic combinations (i.e., compare genotypes XXYY and xxyy in Fig. 1C). Figure 1C depicts additive-by-additive epistasis, which is the simplest form of the three major components of digenic epistasis (i.e., additive-by-additive, additive-by-dominant, and dominant-by-dominant). The representative genotype-by-environment case here (Fig. 1D) simply reverses the eVects of the second locus from the additive model in Fig. 1A, due to reversed expression in a diVerent environment. Pleiotropy occurs when one or more loci influence the expression of more than one trait. Most experimental studies oVer univariate analyses, but the coordination of elements of courtship repertoire are especially illuminating under multivariate analyses.

A. Dominance Natural selection should work eYciently in culling out detrimental dominant alleles, but detrimental recessive alleles will remain in large populations, hidden in the pool of heterozygotes (Falconer, 1989). Mutation contributes to this deleterious recessive load (Frankham, 1995; Lande, 1994, 1995; Lynch

6. Quantitative Genetic Interactions in House Fly Courtship

193

Figure 1 Phenotypic values for four models on two-locus genetic interactions: (A) pure additivity (i.e., no interaction), (B) pure dominance, (C) additive-by-additive epistasis, and (D) genotype-by environment interaction. For each panel, the surfaces represent the phenotypic eVects for the nine possible genotypes for two loci (denoted ‘‘X’’ and ‘‘Y’’), each with two alleles (i.e., X–x and Y–y, respectively). For the additive (A) and dominance (B) panels, the eVects of the two loci are summed to obtain the resultant phenotypic value. In the epistatic panel (C), the eVects are multiplicative (by definition). The representative genotype-by-environment panel (D) depicts the additive model (A) where the influence of the ‘‘B’’ locus is reversed, due to an interaction with a diVerent environment.

et al., 1995; Schultz and Lynch, 1997), and severe population bottlenecks can promote the fixation of old mutations, simulating many generations of input by new deleterious mutations (Lande, 1994). More generally, inbreeding can expose the additive genetic variance that is hidden by dominance, as rare recessive alleles increase in frequency (Hill and Caballero, 1992; Lo´pez-Fanjul and Villaverde, 1989; Whitlock et al., 1993; Willis and Orr, 1993). The widespread occurrence of inbreeding depression attests to the ubiquity of deleterious load and, thus, the importance of dominance (e.g., see reviews by Charlesworth and Charlesworth, 1987; Frankham, 1995;

194

MeVert and Hagenbuch

Thornhill, 1993). The reduction in average heterozygosity for traits with heterozygote advantage is also a factor in inbreeding depression (e.g., Brewer et al., 1990), although simple dominance has been suggested to be more common (Charlesworth and Charlesworth, 1987). With such structuring, dominance can be responsible for heterosis (i.e., the increased fitness of hybrids), along with the decreased fitness of inbred individuals (Falconer, 1989). Thus, dominance structures the ability of populations to withstand population bottlenecks and to recover from extinction threats. Multiple experiments have identified the house fly as being an excellent model for investigating the consequences of inbreeding and the evolutionary significance of dominance (see later).

B. Epistasis Epistasis plays a central role in Wright’s shifting balance theory (Wright, 1969). In contrast to the Fisherian view of large panmictic populations with negligible epistasis, Wright’s model relies on epistasis to generate the amongdeme variation of structured populations (Goodnight and Wade, 2000). In particular, the sorting of adaptive among-locus interactions within a deme can create incompatibility of the genomes of diVerent demes (Lo´ pez-Fanjul et al., 2000; Lynch, 1991; Wolf et al., 2000). More specifically, outbreeding depression can result when interdemic hybrids suVer from the reshuZing of coadapted epistatic complexes (Aspi, 2000; Lynch, 1991; Wolf et al., 2000). Consequently, epistasis is thought to drive the evolution of reproductive incompatibility and, thus, the process of speciation (Aspi, 2000). The prevalence of epistatic variation within species remains controversial (e.g., see Coyne et al., 1997; Goodnight and Wade, 2000), although both the Wrightian and Fisherian views hold that epistasis critically influences reproductive isolation. Within populations, traits structured by some forms of epistasis can also manifest inbreeding depression (Charlesworth, 1998; Falconer, 1989). Thus, even though the cascading eVects of interlocus interactions are commonly recognized in developmental biology, the relative importance in quantitative genetic architecture needs to be explored. Importantly, the logistical demands of identifying epistasis relegate the work to relatively few study systems (see later)—the house fly being one.

C. Genotype-by-Environment Interactions and Learning Genotype-by-environment interactions can distort the mapping of genotype onto phenotype, and thus diminish the ability to predict evolutionary trajectories through heritability coeYcients (Via and Lande, 1985, 1987).

6. Quantitative Genetic Interactions in House Fly Courtship

195

Indeed, genotype-by-environment eVects can result in negative selection responses (Via and Lande, 1987). Behavior traits, in particular, are susceptible to two special forms of genotype-by-environment eVects: learning and social interactions, which are not mutually exclusive. Learning constitutes a general form of genotype-by-environment interaction. With social interactions, however, the ‘‘environment’’ is the interacting conspecific (Boake and Hoikkala, 1995), creating genotype-by-genotype interactions as a special form of epistasis (MeVert, 1995; Wolf et al., 1998). These kinds of behavioral modifications assess the relative sensitivity or canalization of the genetic architecture to environmental pressures and risks (Lynch and Walsh, 1998). In a behavioral ecology sense, the relative influence of genotype-byenvironment interactions dictate the level of adaptive plasticity available for optimal expression, such as in house fly mating strategies.

D. Pleiotropy Pleiotropy is also a critical influence on development and evolution, and is thought to be very widespread due to the complexity of biochemical and developmental webs (Lynch and Walsh, 1998). The genetic and phenotypic intercorrelations generated by pleiotropy can directly influence selection responses and thus dictate the amount of evolutionarily accessible space (Gromko, 1987; Lynch and Walsh, 1998). Importantly, inbreeding or selection can alter the pleiotropic architecture of traits structured by dominance or epistasis (Cheverud and Routman, 1996; Goodnight, 1988; Hansen and Wagner, 2001). In particular, the conversion of additive genetic variance from the nonadditive components (Cheverud and Routman, 1996; Willis and Orr, 1993; see MeVert, 2000 for a review) can restructure genetic covariances/correlations across traits (Bryant and MeVert, 1988; Shaw et al., 1995). This restructuring can open up or close down evolutionary pathways (e.g., see Regan et al., 2003). Dipteran mating behavior is controlled by the numerous aspects of ambulatory activity and sensory capabilities that are integrated through metabolic and neurological webs (Faugeres et al., 1971; Markow, 1981; Sharp, 1984; Taylor, 1975). Thus, the influences of dominance and epistasis found for house fly courtship (e.g., MeVert, 2000 and later) need to be explored on the level of the multivariate phenotype.

III. Courtship in the House Fly There are two basic stages in housefly courtship: a stalking type of behavior by the male, followed by his mounted displays and attempted copulation. In the stalking phase, the male makes creeping movements toward the female,

196

MeVert and Hagenbuch

at which point he might stand next to her for brief episodes (see MeVert and Bryant, 1991). During this premounting phase, the male often taps the female with his forelegs, often stimulating a fending-oV behavior from the female (MeVert and Bryant, 1991). Figure 2A–E depicts five stereotypical displays that are performed once the male mounts the female (male performance of BUZZ, LUNGE, HOLD, and LIFT and the female’s WING OUT). At mounting, the male will buzz his wings (BUZZ), ostensibly producing auditory and/or tactile cues (see arrow in Fig. 2A). The male then orients toward the head of the female and lunges over her head (LUNGE, Fig. 2B). During LUNGE, the male periodically stops BUZZ and holds his wings either in their normal resting position (horizontally along the dorsal plane) or over the female’s head (HOLD, Fig. 2B, C). During LUNGE, the male can also periodically attempt to lift the female’s forelegs with his own forelegs (LIFT; see arrow in Fig. 2B). At any point during mounting, the female thrusts her wings out perpendicular to her body to kick her hind legs up and over her wings (WING OUT; see arrow in Fig. 2C). The male then attempts copulation (Fig. 2E), during which time the female stops WING OUT (see arrow in Fig. 2D). Figure 2F–J shows the same relative sequences for a mating pair that omits HOLD, LIFT, and WING OUT, using characteristically less LUNGE. Table I quantifies the intercorrelations among the five courtship traits (see MeVert and Regan, 2002). LUNGE, HOLD, LIFT, and WING OUT are positively intercorrelated, with BUZZ being negatively correlated with HOLD. This general relationship is also identified in the principal component solution (Table II). In standard morphometrics, the first principal component, which summarizes the major positive intercorrelations, is termed the size axis (Pimentel, 1979). For these behavior analyses, the first principal component (PC1) explains the gradation from more complex (i.e., larger ‘‘size’’) to simpler (i.e., smaller) courtships. Figure 2A–E thus portrays a characteristically complex courtship, with a strong positive loading on this axis, whereas Fig. 2F–J portrays the characteristically simple repertoire: one with a negative loading. This classification of the continuum of courtship complexity largely reflects the level of female reluctance to mate, in that populations with negative loadings have females that are significantly more receptive than those with positive loadings (MeVert and Bryant, 1991; MeVert and Regan, 2002). To a lesser extent, this size continuum also explains diVerences in the males’ ability to successfully acquire copulations, such that the more aggressive males are found in populations with positive loadings on this size axis (MeVert and Bryant, 1991; MeVert and Regan, 2002). Earlier studies suggested that the male forces the execution of WING OUT (Tobin and StoVolano, 1978), but we have found that the female is directly involved in the display. Sacca (1964) reported that the female leg kick serves to mutilate the male’s wings, but we

Figure 2 Qualitative comparison of the complex (A–E) and simple (F–J) courtship repertoires of the house fly. See ‘‘Courtship in the House Fly’’ for a description of the figure.

198

MeVert and Hagenbuch

Table I Pearson Correlation CoeYcients Among the Five Courtship Displays Measured in 160 Matings

BUZZ LUNGE HOLD LIFT

BUZZ

LUNGE

HOLD

1.000 — — —

0.245{ (0.002) 1.000 — —


0.370{ (<0.0001) 0.177* (0.025) 1.000 —

LIFT

WING OUT

0.113 (0.155)
0.082 (0.304) 0.580{ (<0.0001) 0.482{ (<0.0001) 0.052 (0.513) 0.173* (0.029) 1.000 0.376{ (<0.0001)

See MeVert and Regan, 2002, for details. Significance values are given as *, p < 0.05; , p < 0.01; {, p < 0.001. Exact p-values are given in parentheses below each correlation coeYcient. All of the significant eVects remain after correction for multiple testing (the step-down Finner modification of the standard Bonferroni; see Brown and Russell, 1997). {

Table II Correlations of Courtship Traits with the First Principal Component (PC1) of the Phenotypic Intercorrelation Structure for 160 Courtships Trait

PC1

BUZZ LUNGE HOLD LIFT WING OUT

0.116 0.614 0.173 0.573 0.501

See MeVert and Regan, 2002, for details. Percentage of variance explained is 40.8%.

have witnessed countless courtships where the female’s legs do not even reach the male’s wings. Nevertheless, the female behaviors do appear to serve to dislodge the mounted male, whereas the male displays serve to thwart her rejection eVorts. Thus, this size continuum explains the relative ability of the male to conquer more reluctant females. Importantly, isolated virgin male– female pairs will eventually mate (e.g., within 24 hours). Although some have suggested that ‘‘rape’’ is involved in this phenomenon, learning by males may play a greater role in this process (see later).

IV. Evidence of Quantitative Genetic Interactions in House Flies In this section, we summarize prior bottleneck experiments that provided indirect evidence of dominance, epistasis, and genotype-by-environment interactions. We then extend prior analyses on line crosses to further

6. Quantitative Genetic Interactions in House Fly Courtship

199

examine the roles of nonadditive eVects on the pleiotropic components of the courtship repertoire. We also present a repeatability assay for more direct identification of the role of learning in males’ eVorts to achieve copulations. We find that dominance, epistasis, and learning (i.e., genotype-byenvironment interactions) all strongly influence the genetic architecture of house fly mating behavior.

A. Assays of Additive Genetic Variances in Bottlenecked Populations In eVorts to simulate speciation via founder–flush events (e.g., see MeVert, 1999), bottleneck experiments have provided indirect evidence that both dominance and epistasis structure house fly mating behavior. The common phenomenon of reduced overall mating propensity in inbred laboratory lines implicated the influence of dominance (e.g., MeVert and Bryant, 1991; MeVert et al., 1999). In contrast, cases of increased mating propensity after inbreeding suggested the purge of genetic load (e.g., MeVert and Bryant, 1991; MeVert et al., 1999). Indirect support of more complex nonadditive eVects was revealed when bottlenecked populations exhibited increased additive genetic variances (and heritabilities) for specific components of the courtship repertoire (MeVert, 1995; MeVert et al., 2002). In particular, purely additive genetic models predict reduced evolutionary potential in inbred lines (Falconer, 1989), but theory on dominance and epistasis shows how additive genetic variances (and heritabilities) can increase after a founder event (Cheverud and Routman, 1996; Goodnight, 1988; Willis and Orr, 1993; see MeVert, 2000 for a review). Thus, the increased additive genetic variances in bottlenecked populations (MeVert, 1995; MeVert et al., 2002) reflected the influence of dominance and/or epistasis, although their relative influences were undecipherable. Line cross analyses in a selection experiment (see later) were more convincing in demonstrating both dominance and epistasis in house fly courtship, as well as the evolutionary consequences of outbreeding and inbreeding depression. The bottleneck experiments also provided indirect evidence of genotypeby-environment interactions in courtship. More specifically, MeVert (1995) used a genotype-by-environment model, where the environment is the mating partner, to explain the paradoxical negative estimates of additive genetic variances for courtship traits (see also MeVert et al., 2002). In theory, negative additive genetic variances are not possible because they imply the negative sign of a squared term. However, many additive genetic variances, as estimated by parent–oVspring covariances (see Falconer, 1989), were significantly negative, as tested by bootstrap sampling. Adding a genotypeby-environment component in a simulation of parent–oVspring covariances did indeed yield significant negative variance estimates. In this simulation,

200

MeVert and Hagenbuch

the resultant courtship performance was influenced by the male’s vigor, the female’s preferences, and the interaction between the two. More specifically, the bias for more negative estimates in the parent–son covariances, as compared with the parent–daughter covariances, suggested that the male house fly was relatively more plastic in modulating his behavior to meet a female’s preferences (MeVert, 1995; MeVert et al., 2002). The female’s behavior, in contrast, was more stereotypical. The bottleneck experiments thus identified complex social interactions in a relatively simple organism. By being supported through bootstrap simulations, however, this kind of evidence of genotype-by-environment eVects was admittedly tenuous. Subsequent selection and repeatability studies (see later) were more compelling in rendering these conclusions.

B. Line Cross Analyses Line cross analyses on populations that had undergone artificial selection for divergent courtship repertoire yielded more direct support of both dominance and epistasis (MeVert et al., 2002). This experiment was designed to test how incipient speciation can arise through divergent sexual selection trajectories (see MeVert and Regan, 2002; MeVert et al., 2002). In this study, four populations were subjected to selection for divergence along the first principal component (see Table II): two replicate lines that had undergone selection for more extreme positive scores (see Fig. 2A–E) and two lines that had undergone selection for negative scores (see Fig. 2F–J; see MeVert and Regan, 2002 for experimental details). After eight generations of selection, diallel crosses were performed between the lines that had evolved along opposite trajectories (i.e., females from a positive trajectory line mating with males from a negative trajectory line, and vice versa for the other replicate). Using univariate analyses on these crosses, MeVert et al. (2002) found significant dominance and epistatic eVects for LUNGE, HOLD, LIFT, and WING OUT. Our analyses presented later, however, will address the pleiotropic nature of the integrated repertoire through tests on the multivariate loadings onto the first principal component (see Table II). We used joint-scaling tests (Hayman, 1960; Lynch and Walsh, 1998) to identify dominance and epistasis for the loadings on the size axis of courtship integration by testing the six means (P1, P2, F1, F2, and the two backcrosses) for goodness of fit with the expected line means. Thus, three quantitative genetic models: (1) purely additive, (2) additive with dominance, and (3) additive with both dominance and additive-by-additive epistasis were tested in a hierarchical manner. A significant deviation from the model (tested by chi-square) suggests rejection in favor of the next model in the hierarchy.

6. Quantitative Genetic Interactions in House Fly Courtship

201

Figure 3 Line cross assays for strains subjected to artificial selection for divergence in courtship. Separate panels are shown for each of the two replicate blocks. The x-axis represents the genomic representation for each line mean (i.e., 0 and 1 for the parental lines, 0.5 for the F1 and F2 hybrids, and 0.25 and 0.75 for the appropriate backcrosses). The y-axis represents the trait value for the intensity of the repertoire. The open circles indicate the F2 means. The straight lines identify the expected means based upon pure additivity, and the bars represent the 95% confidence intervals for the means. Each panel also provides the chi-square significance values for the hierarchical fits to the diallel models: additivity, additivity with dominance, and additivity with dominance and additive-by-additive epistasis. A significant deviation calls for rejection of the model in favor of the next one in the hierarchy. Significance values are given as *, p < 0.001. PC1 refers to the first principal component.

Figure 3 depicts the data from the line crosses (with 95% confidence intervals), along with their expectations for pure additivity (see Stevens, 1994). In the first block of line crosses (Fig. 3A), the models of pure additivity and dominance are inadequate to explain the data, although the model that includes additive-by-additive epistasis is suYcient (see Fig. 1). In the second block (Fig. 3B), the model of additivity with both dominance

202

MeVert and Hagenbuch

and additive-by-additive epistasis is rejected (see Fig. 1), suggesting even higher-order epistasis (i.e., dominance-by-additive and/or dominance-bydominance epistasis), for loadings on this size axis. Heterosis (i.e., the F1 mean lying outside of the parental phenotypes; see Falconer, 1989) is indicated in the second block (Fig. 3B), and both blocks have evidence of outbreeding depression (i.e., breakdown in the F2 or backcrosses; see Fig. 3). This experiment thus lent credence to the models of speciation via sexual selection by generating the emergent stages of genomic incompatibility. As noted earlier, the presence of dominance is hardly controversial, because inbreeding depression is a widespread phenomenon (e.g., Thornhill, 1993). Indeed, dominance was predictable in house fly mating behavior, based upon the prevalence (although not ubiquity) of inbreeding depression in experimentally bottlenecked populations (e.g., MeVert and Bryant, 1991, 1992; MeVert et al., 1999). Epistasis is considerably more controversial (e.g., Coyne et al., 1997; Goodnight and Wade, 2000) and more diYcult to detect (see MeVert, 2000). These line cross analyses did yield highly significant eVects of epistatic outbreeding depression (see Fig. 3). Importantly, the eVects were seen in crosses between parental lines that were not significantly diVerent from each other (Fig. 3A), identifying the essence of epistasis. That is, two populations can evolve the same phenotype with materially diVerent genetic underpinnings (e.g., see Fig. 1C), resulting in genetic architectures that are incompatible with each other (e.g., Thornhill, 1993). This kind of data on epistasis is rather rare, although similar results were found for Ewing’s (1967) study on D. melanogaster locomotion. These data show how the eVects of inbreeding depression, heterosis, and outbreeding depression may operate simultaneously in a complex phenotype. This experiment also confirmed the genotype-by-environment interactions that were proposed in the prior bottleneck study (MeVert, 1995). By videotaping the successful courtships between males and females from the divergent populations (i.e., P1 and P2 in Fig. 3), MeVert and Regan (2002) found that males can successfully accommodate the preferences of a ‘‘foreign’’ female (i.e., one from a divergent population). More specifically, males from a population with characteristically simple courtship (i.e., negative loadings on PC1; see Fig. 2F–J) performed the complex repertoire when paired with females from a population at the other end of the size continuum (positive loadings on PC1; see Fig. 2A–E), and vice versa (MeVert and Regan, 2002). In accordance with the prior inference that the females are less plastic than the males (MeVert, 1995; see also MeVert et al., 2002), the female was consistent in the intensity of her WING OUT display, regardless of the type of male that was courting her (see Fig. 2 and MeVert and Regan, 2002). These assays strongly suggested that the male house fly can identify the cryptic preferences of his mating partner in order to secure a copulation. Repeatability assays were thus critical for evaluating the specific nature of these learning curves.

6. Quantitative Genetic Interactions in House Fly Courtship

203

C. Repeatability Assays Aragaki and MeVert (1998) assayed the repeatability of a male’s performance of BUZZ (see Fig. 2) within and across females (i.e., two courtships per female). They found that the males systematically reduced the intensity of BUZZ between the first and second unsuccessful courtships with the same female, resetting BUZZ to high intensity with the first courtship for the second female (Aragaki and MeVert, 1998). In contrast, female performance of WING OUT was consistent within and across courting males (Aragaki and MeVert, 1998). As in the other studies, the male appeared to be responding to rigid demands of his mating partner. This study thus identified shortterm learning curves in male house flies, but did not evaluate the potential long-term adaptive value of this phenomenon. For the repeatability study presented here, we started with a new population that had been in the laboratory for approximately five generations. On the sixth day after eclosion (i.e., the age of peak sexual activity, see MeVert and Bryant, 1991), we videotaped a random virgin male with a single random virgin female for 30 min or until copulation. Later in the day, but no less than 4 h after the first session, the same male was videotaped with another virgin female. We continued this procedure for each male over a total of 3 days for a total of six females per male, testing a total of 121 males (726 females). No female was used more than once. In these analyses, we simply counted the number of courtship encounters until copulation. We tested for repeatability across males by ANOVA (analysis of variance), using the identities of the males as the grouping parameter (Boake, 1989). The number of courtships performed before copulation or within the 30-minute time period was highly repeatable among males (F(120, 605) ¼ 2.35, p < 0.0001), indicating significant genetic variance for the level of mating activity. For the analyses on only those females that allowed copulation (i.e., 257/726 ¼ 35%), there was a significant trend for linear functions of improved copulation success with exposure to serial females, such that the males used progressively fewer courtships to achieve copulation (Fig. 4). In these latter analyses, we have randomized the possible influences of age (as much as possible) by analyzing only those cases when the females allowed copulation. More specifically, if males simply performed fewer and fewer courtships with the six serial females as they aged, we could have found a spurious negative regression. The linear eVects of the 3-day trials were thus randomized when the cases of unreceptive females were ignored. Another possible source of a spurious negative regression would be having the most fit males, in terms of requiring fewer courtship encounters for female acceptance, represent an increasingly larger fraction of the serial copulations (i.e., females 1 through 5; see Fig. 4). Indeed, only three males achieved five copulations (resulting in the particularly large 95% confidence interval in Fig. 4). Thus, a more conservative

204

MeVert and Hagenbuch

Figure 4 Repeatability analyses on the number of courtship encounters for a male mating with five serial females. Bars depict 95% confidence intervals. The least-squares regression line is highly significant ( p < 0.0001). More conservative tests on the nine males that mated with exactly four females still reveal a significant negative regression ( p ¼ 0.04; see text for details).

approach would be to evaluate only males with the same level of vigor, even though it would reduce the power of the test. Nevertheless, using the nine males that gained exactly four matings still yielded a significant negative regression (F(1, 34) ¼ 4.46; p ¼ 0.04), with a slope comparable to that in the pooled analyses. Thus, male house flies exhibit adaptive genotype-by-environment interactions in courtship not only through their short-term encounters with specific females (MeVert and Regan, 2002), but also through longterm processes of improved technique (see Fig. 4). We are unaware of another animal model where learning the preferences of a mating partner is a sexually selected trait. The majority of studies on animal learning address positive and negative feedback loops for pain and nutritional rewards. For example, associative conditioning to food rewards has been identified in blow flies (McGuire and Tully, 1987) and honey bees (Brandes, 1988). Moreover, avoidance behavior to shock treatments has been well studied in D. melanogaster (e.g., Hewitt et al., 1983) and the house mouse (e.g., Oliverio, 1971). The literature on optimal male mate choice (e.g., Siegel and Hall, 1979) does not apply to this system because male house flies will persistently court unreceptive (i.e., mated) females, other males, dead conspecifics, and even heterospecifics. At present, therefore, it seems that the house fly model may be unique in that associative learning is enforced by the positive feedback loop of female courtship acceptance, in accordance with basic models of sexual selection. If our inferences are correct, then the house fly females should be acting as a selective pressure on the males’ competence to learn.

6. Quantitative Genetic Interactions in House Fly Courtship

205

V. Prevalence of Nonadditive Genetic Effects in Animal Behavior Related animal behavior studies have also revealed significant eVects of dominance, epistasis, and genotype-by-environment interactions. Table III summarizes a review of 41 studies on nonadditive eVects in animal behavior, excluding any involving house flies and using a stricter definition of genotypeby-environment eVects for the studies on learning (see also MeVert et al., 2002). Most of the studies focused on traits relevant to issues of communication, social behavior, sexual selection, and speciation, as with the house fly courtship model detailed in this chapter. Even those traits with apparently less direct relevance to these topics, such as locomotion, are likely to have indirect eVects due to pleiotropic eVects on mating success. Over 60% of the studies had identified dominance or genotype-by-environment eVects, and greater than 30% showed some combined eVect (Fig. 5). Significant data on epistasis was relatively scarce, but few systems are amenable to the necessary protocols (e.g., see MeVert, 2000). In our search, we found only four studies that directly refuted the influence of some nonadditive eVect (Brandes, 1991; Cohan et al., 1989; Hedrick, 1994; Lynch, 1994), yet each one confirmed the significance of some other nonadditive component. Thus, we conclude that the house fly is not particularly unusual in exhibiting nonadditive genetic eVects for behavior traits, especially those involved with fitness and mate choice. Other studies have provided direct evidence for the complex evolutionary processes predicted by models of dominance and epistasis. For example, nest-building (Bult and Lynch, 2000) and wheel-running (Bruell, 1964; Dohm et al., 1996) of the house mouse exhibited heterosis. Heterosis was also identified for learning (Hewitt et al., 1983), knockdown resistance (Cohan et al., 1989), and larval feeding rates (Sewell et al., 1975) in D. melanogaster. In addition, McGuire and Tully (1987) found both heterosis and the breakdown of epistatic interactions for learning in blowflies. Inbreeding and outbreeding depression in the D. montana courtship song was revealed by line cross analyses (Aspi, 2000). Thus, house fly experiments contribute to the growing body of literature on the consequences of dominance and epistasis, particularly for traits that could be important to speciation.

VI. Future Directions Clearly, the house fly has served as an excellent quantitative genetic model for identifying the importance of dominance, epistasis, pleiotropy, and genotype-by-environment interactions. The next logical development would be to integrate assays of the molecular genetic underpinnings of such quantitative genetic phenomena. Unfortunately, the linkage map for M. domestica

206

MeVert and Hagenbuch

Table III Sources of Literature (Source) for Evidence of Nonadditive Genetic EVects in Animal Behavior (for Fig. 5) Animal Armyworm moth Blow fly C. elegans D. littoralis D. melanogaster

D. montana D. persimilis D. silvestris

D. simulans Flour beetle Garter snake Guppy Honey bee House mouse

Japanese quail Milkweed bug Molly Mosquito Moth Vole

Behavior Migration Learning Locomotion Courtship Foraging Knockdown resistance Learning Locomotion Mate recognition Mating propensity Mating propensity Mating propensity Mating propensity Mating propensity Olfaction Olfaction Territoriality Courtship Geotaxis Phototaxis Aggression Courtship Courtship Geotaxis Cannibalism Foraging Courtship Defense Learning Learning Learning Locomotion Locomotion Nest-building Aggression Migration Courtship Photoperiodism Courtship Phonotaxis Migration

DOM

EPI

X X

X X

X X

X No

GXE X

X X

X X X X X X X X

X X X

X X X X No X X X X X

X X

X X X

X

X X X

X X X X X X

X

No

X X X X

X

X X X

X

Source Dingle (1994); Gatehouse (1986) McGuire and Tully (1987) Park and Horvitz (1986) Aspi and Hoikkala (1993) Sewell et al. (1975) Cohan et al. (1989) Hewitt et al. (1983) Weber (1996) Finley et al. (1997) Carracedo et al. (1995); Casares et al. (1993) Fulker (1966) Kessler (1969) Manning (1961) Sharp (1984) Fedorowicz et al. (1998) MacKay et al. (1996) HoVman (1994) Aspi (2000); Aspi and Hoikkala (1993) Polivanov (1975) Polivanov (1975) Boake and Konigsberg (1998) Boake and Hoikkala (1995) Boake and Konigsberg (1998) Ringo and Wood (1983) Stevens (1994) Arnold (1981a,b) Farr (1983); Farr and Peters (1984) Hunt et al. (1998) Brandes (1988, 1991) Henderson (1968a,b) Oliverio (1971); Oliverio et al. (1972) Bruell (1964) Dohm et al. (1996) Bult and Lynch (2000); Lynch (1994) Nol et al. (1996) Caldwell and Hegmann (1969); Dingle (1994) Travis (1994) Hard et al. (1993) Collins et al. (1999); Jia et al. (2000) Jang and Greenfield (2000) Dingle (1994); Rasmuson et al. (1977)

The table gives the study organism (Animal), followed by the general type of behavior (Behavior) in the study. No house fly experiments are included in this review. Studies were combined when it was apparent that one was directly derived from another. The ‘‘X’’ denotes the presence of the nonadditive genetic eVect, coded DOM, EPI, and GXE for dominance, epistasis, and genotype-by-environment interactions, respectively. The ‘‘No’’ indicates a definitive conclusion about the absence of the particular genetic interaction.

6. Quantitative Genetic Interactions in House Fly Courtship

207

Figure 5 Summary of nonadditive eVects found in 41 animal behavior studies (excluding all research on the house fly). The percentage of studies found in the sample are partitioned by the independent or combined eVects of the three major sources of nonadditivity: DOM (dominance), EPI (epistasis), and GXE (genotype-by-environment interactions). ALL indicates that all three types of nonadditive eVects were found in the same experimental model.

is not well developed at present. Techniques applying allozyme or visible markers have serious shortcomings. Recent work has identified microsatellite markers (Endsley et al., 2002), and amplified fragment length polymorphism (AFLP)-linkage mapping is underway. With such molecular sophistication, M. domestica could become an ideal model for investigating the molecular basis and developmental consequences of genetic load. Its relative rarity in the ability to detect epistasis can be exploited for quantitative trait loci (QTL) mapping of the genomic regions involved with hybrid breakdown (e.g., see Wang et al., 1999), especially at the level of premating isolating characters. As an ultimate goal, understanding pleiotropic cascades in the expression of major loci will define explicit evolutionary constraints on the evolution of multivariate phenotypes. Finally, this system may be unique in its accessibility to investigation of the genetics of social traits and learning.

VII. Summary 1. The house fly, M. domestica L., is particularly amenable to large-scale quantitative genetic investigations of fitness traits, sexual selection, and speciation. Bottleneck simulations, selection experiments, and repeatability assays are apparently more feasible than in related systems. The complex

208

MeVert and Hagenbuch

male–female interactions in house fly courtship aVord special opportunities for studies on communication, social interactions, and learning. 2. Basic evolutionary theory holds that the additive genetic component can serve as a metric for historical selection pressure and evolutionary potential, with pleiotropic architecture defining the multivariate limits of evolutionarily accessible space. Additive genetic variance is expected to be especially low for mating behavior traits, with dominance, epistasis, and genotype-by-environment interactions concealing residual components of additive genetic variance. 3. Bottleneck experiments have provided indirect evidence of dominance and epistasis in house fly mating behavior through increased additive genetic variances in inbred populations. A genotype-by-environment model of male– female interactions was used to explain the paradoxical negative estimates of additive genetic variances. 4. Taking pleiotropic eVects into consideration, line cross analyses on populations that had been selected for divergent courtship repertoire directly identified the eVects of dominance and epistasis. This experiment also corroborated previous inferences of males modulating their behavior to accommodate diVerent female preferences. 5. A prior repeatability study showed how the male house fly exhibits adaptive short-term learning curves in identifying the cryptic preferences of his mating partner. The novel repeatability study presented here showed long-term processes of improved techniques with serial matings. M. domestica may be distinctive by serving as a model wherein the male’s capacity to learn is a sexually selected trait driven by female preferences. 6. A literature review of 41 animal behavior studies showed that dominance, epistasis, and genotype-by-environment interactions are common across a wide range of traits and taxa. The house fly model remains unusual in the ability to identify epistatic structure and adaptive learning curves in the ability to satisfy female preferences. 7. Further development of molecular methods for the M. domestica genome would unravel the underlying biochemical and developmental pathways of these quantitative genetic interactions for a more thorough understanding of the processes of inbreeding depression, outbreeding depression, and pleiotropy.

Acknowledgments This research was funded by grants from the National Science Foundation (DEB-9408004, DEB-9726667, and DEB-0196101). Many thanks to A. Danielson-Franc¸ ois and A. Swann for their valuable comments on an earlier draft of the manuscript.

6. Quantitative Genetic Interactions in House Fly Courtship

209

References Andersson, M. (1994). ‘‘Sexual Selection.’’ Princeton University Press, Princeton, NJ. Aragaki, D. L. R., and MeVert, L. M. (1998). A test of how well the repeatability of courtship predicts its heritability. Anim. Behav. 55, 1141–1150. Arnold, S. J. (1981a). Behavioral variation in natural populations I. Phenotypic, genetic and environmental correlations between chemoreceptive responses to prey in the garter snake, Thamnophis elegans. Evolution 35, 489–509. Arnold, S. J. (1981b). Behavioral variation in natural populations II. The inheritance of a feeding response in crosses between geographic races of the garter snake, Thamnophis elegans. Evolution 35, 510–515. Aspi, J. (2000). Inbreeding and outbreeding depression in male courtship song characters in Drosophilia montana. J. Hered. 84, 273–282. Aspi, J., and Hoikkala, A. (1993). Laboratory and natural heritabilities of male courtship song characters in Drosophila montana and D. littoralis. J. Hered. 70, 400–406. Aspi, J., and Hoikkala, A. (2000). Male mating success and survival in the field with respect to size and courtship song characters in Drosophila littoralis and D. montana (Diptera: Drosophilidae). J. Ins. Behav. 8, 67–87. Boake, C. R. B. (1989). Repeatability: Its role in evolutionary studies of mating behavior. Evol. Ecol. 3, 173–182. Boake, C. R. B., DeAngelis, M. P., and Andreadis, D. K. (1997). Is sexual selection and species recognition a continuum? Mating behavior of the stalk-eyed fly Drosophila heteroneura. Proc. Nat. Acad. Sci. USA 94, 12442–12445. Boake, C. R. B., and Hoikkala, A. (1995). Courtship behaviour and mating success of wildcaught Drosophila silvestris males. Anim. Behav. 49, 1303–1313. Boake, C. R. B., and Konigsberg, L. (1998). Inheritance of male courtship behavior, aggressive success, and body size in Drosophila silvestris. Evolution 52, 1487–1492. Brandes, Ch. (1988). Estimation of heritability of learning behavior in honeybees (Apis mellifera capensis). Behav. Genet. 18, 119–132. Brandes, Ch. (1991). Genetic diVerences in learning behavior in honeybees (Apis mellifera capensi). Behav. Genet. 21, 271–294. Brewer, B. A., Lacy, R. C., Foster, M. L., and Alaks, G. (1990). Inbreeding depression in insular and central populations of Peryomyscus mice. J. Hered. 81, 257–266. Brown, B. B., and Russell, K. (1997). Methods correcting multiple testing: Operating characteristics. Stat. Med. 16, 2511–2528. Bruell, J. H. (1964). Heterotic inheritance of wheel-running in mice. J. Comp. Physiol. Psychol. 58, 159–163. Bryant, E. H., and MeVert, L. M. (1988). EVect of an experimental bottleneck on morphological integration in the housefly. Evolution 42, 698–707. Bult, A., and Lynch, C. B. (2000). Breaking through artificial selection limits of an adaptive behavior in mice and the consequences for correlated responses. Behav. Genet. 30, 193–206. Caldwell, R. L., and Hegmann, J. P. (1969). Heritability of flight duration in the milkweed bug, Lygaeus kalmii. Nature 223, 91–92. Carracedo, M. A., Pin˜ eiro, R., and Cesares, P. (1995). Chromosomal substitution analysis of receptivity and sexual isolation in Drosophila melanogaster females. J. Hered. 75, 541–546. Casares, P., Carracedo, M. C., San Miguel, E., Pin˜ eiro, R., and Garcia- Florez, L. (1993). Male mating speed in Drosophila melanogaster: DiVerences in genetic architecture and in the relative performance according to female genotype. Behav. Genet. 23, 349–358.

210

MeVert and Hagenbuch

Charlesworth, B. (1998). The eVect of synergistic epistasis on the inbreeding load. Genet. Res. 71, 85–89. Charlesworth, B., and Charlesworth, D. (1987). Inbreeding depression and its evolutionary consequences. Annu. Rev. Ecol. Syst. 18, 237–268. Cheverud, J. E., and Routman, E. J. (1996). Epistasis as a source of increased additive genetic variance at population bottlenecks. Evolution 50, 1042–1051. Cohan, F. M., HoVman, A. A., and Gayley, T. W. (1989). A test of the role of epistasis in divergence under uniform selection. Evolution 43, 766–774. Collins, R. D., Jang, Y., Reinhold, K., and Greenfield, M. D. (1999). Quantitative genetics of ultrasonic advertisement signaling in the lesser waxmoth Achroia grisella (Lepidoptera: Pyralidae). J. Hered. 83, 644–651. Coyne, J. A., Barton, N. H., and Turelli, M. (1997). Perspective: A critique of Sewall Wright’s shifting balance theory of evolution. Evolution 51, 643–671. Dingle, H. (1994). Genetic analysis of animal migration. In ‘‘Quantitative Genetic Studies of Behavioral Evolution’’ (C. R. B. Boake, Ed.), pp. 145–164. University of Chicago Press, Chicago. Dohm, M. R., Hayes, J. P., and Garland, Jr., T. (1996). Quantitative genetics of sprint running speed and swimming endurance in laboratory house mice (Mus domesticus). Evolution 50, 1688–1701. Endsley, M. A., Baker, M. D., and Krafsur, E. S. (2002). Microsatellite loci in the house fly, Musca domestica L. (Diptera: Muscidae). Mol. Ecol. Notes 2, 72–74. Ewing, A. W. (1967). Genetics and activity in Drosophila melanogaster. Experientia 15, 330–332. Falconer, D. S. (1989). ‘‘Introduction to Quantitative Genetics.’’ Longman, New York. Farr, J. A. (1983). The inheritance of quantitative fitness traits in guppies, Poecilia reticulata (Pices: Peociliidae). Evolution 37, 1193–1209. Farr, J. A., and Peters, K. (1984). The inheritance of quantitative fitness traits in guppies, Poecilia reticulata (Pisces: Poeciliidae). 2. Tests for inbreeding eVects. J. Hered. 52, 285–296. Faugeres, A., Petit, C., and Thibout, E. (1971). The components of sexual selection. Evolution 25, 265–275. Fedorowicz, G. M., Fry, J. D., Anholt, R. R. H., and Mackay, T. F. C. (1998). Epistatic interactions between smell-impaired loci in Drosophilia melanogaster. Genetics 148, 1885–1891. Finley, K. D., Taylor, B. J., Milstein, M., and McKeown, M. (1997). Dissatisfaction, a gene involved in sex-specific behavior and neural development of Drosophilia melangogaster. Genetics 94, 913–918. Fisher, R. A. (1958). ‘‘The Genetical Theory of Natural Selection.’’ Dover, New York. Frankham, R. (1995). Inbreeding and extinction: A threshold eVect. Cons. Biol. 9, 792–799. Fulker, D. W. (1966). Mating speed in male Drosophila melanogaster; A psychogenic analysis. Science 153, 203–205. Gatehouse, A. G. (1986). Migration in the African armyworm, Spodoptera exempta: Genetic determination of migratory capacity and a new synthesis. In ‘‘Insect Flight, Dispersal, and Migration’’ (W. Danthanarayana, Ed.), pp. 128–144. Springer-Verlag, Berlin. Goodnight, C. J. (1988). Epistasis and the eVect of founder events on the additive genetic variance. Evolution 42, 441–454. Goodnight, C. J., and Wade, M. J. (2000). The ongoing synthesis: A reply to Coyne, Barton, and Turelli. Evolution 54, 317–324. Gromko, M. H. (1987). Genetic constraint on the evolution of courtship behaviour in Drosophila melanogaster. J. Hered. 58, 435–441. Hansen, T. F., and Wagner, G. P. (2001). Modeling genetic architecture: A multilinear theory of gene interaction. Theor. Pop. Biol. 59, 61–86.

6. Quantitative Genetic Interactions in House Fly Courtship

211

Hayman, B. I. (1960). The separation of epistasis from additive and dominance variation in generation means. Genetica 31, 133–146. Hard, J. J., Bradshaw, W. E., and Holzapfel, C. M. (1993). The genetic basis of photoperiodism and its evolutionary divergence among populations of the pitcher-plant mosquito, Wyeomyia smithii. Amer. Natur. 142, 457–473. Hedrick, A. V. (1994). The heritability of mate-attractive traits: A case study on field crickets. In ‘‘Quantitative Genetic Studies of Behavioral Evolution’’ (C. R. B. Boake, Ed.), pp. 228–250. University of Chicago Press, Chicago. Henderson, N. D. (1968a). Genetic analysis of acquisition and retention of a conditioned fear in mice. J. Comp. Physiol. Psychol. 65, 325–330. Henderson, N. D. (1968b). The confounding eVects of genetic variables in early experience research: Can we ignore them? Dev. Psychobiol. 1, 146–152. Hewitt, J. K., Fulker, D. W., and Hewitt, C. A. (1983). Genetic architecture of olfactory discriminative avoidance conditioning in Drosophila melanogaster. J. Comp. Physiol. 97, 52–58. Hicks, S. K., Hagenbuch, K. L., and MeVert, L. M. (2004). Variable costs of mating, longevity, and starvation resistance in the housefly Musca domestica (Diptera: Muscidae). Env. Entomol. 33, 779–786. Hill, W. G., and Caballero, A. (1992). Artificial selection experiments. Annu. Rev. Ecol. Syst. 23, 287–310. HoVman, A. A. (1994). Genetic analysis of territory in Drosophila melanogaster. In ‘‘Quantitative Genetic Studies of Behavioral Evolution’’ (C. R. B. Boake, Ed.), pp. 188–205. University of Chicago Press, Chicago. Hunt, G. J., Guzman-Novoa, E., Fondrk, M. K., and Page, Jr., R. E. (1998). Quantitative trait loci for honey bee stinging behavior and body size. Genetics 148, 1203–1213. Iwasa, Y., and Pomiankowski, A. (1995). Continual changes in mate preferences. Nature 377, 420–422. Jang, Y. W., and Greenfield, M. D. (2000). Quantitative genetics of female choice in an ultrasonic pyralid moth, Achroia grisella: Variation and evolution of performance along multiple dimensions of the male advertisement signal. J. Hered. 84, 73–80. Jia, F.-Y., Greenfield, M. D., and Collins, R. D. (2000). Genetic variance of sexually selected traits in waxmoths: Maintenance by genotype X environment interactions. Evolution 54, 953–967. Kaneshiro, K. Y. (1980). Sexual selection, speciation, and the direction of evolution. Evolution 34, 437–444. Kessler, S. (1969). The genetics of Drosophila mating behavior. II. The genetic architecture of mating speed in Drosophila pseudoobscura. Genetics 62, 421–433. Lande, R. (1981). Models of speciation and sexual isolation. Proc. Nat. Acad. Sci. USA 78, 3721–3725. Lande, R. (1994). Risk of population extinction from new deleterious mutations. Evolution 48, 1460–1469. Lande, R. (1995). Mutation and conservation. Cons. Biol. 9, 782–791. Lo´ pez-Fanjul, C., Fernandez, A., and Toro, M. A. (2000). Epistasis and the conversion of nonadditive genetic variance at population bottlenecks. Theor. Pop. Biol. 58, 49–59. Lo´ pez-Fanjul, C., and Villaverde, A. (1989). Inbreeding increases genetic variance for viability in Drosophila melanogaster. Evolution 43, 1800–1804. Lynch, C. B. (1994). Evolutionary inferences from genetic analyses of cold adaptation in laboratory and wild populations of the house mouse. In ‘‘Quantitative Genetic Studies of Behavioral Evolution’’ (C. R. B. Boake, Ed.), pp. 278–301. University of Chicago Press, Chicago. Lynch, M. (1991). The genetic interpretation of inbreeding depression and outbreeding depression. Evolution 45, 622–629. Lynch, M., Connery, J., and Bu¨ rger, R. (1995). Mutational meltdowns in sexual populations. Evolution 49, 1067–1080.

212

MeVert and Hagenbuch

Lynch, M., and Walsh, B. (1998). ‘‘Genetics and Analysis of Quantitative Traits.’’ Sinauer Associates, Sanderland, MD. MacKay, T. F. C., Hackett, J. B., Lyman, R. F., Wayne, M. L., and Anholt, R. R. H. (1996). Quantitative genetic variation of odor-guided behavior in a natural population of Drosophila melanogaster. Genetics 144, 727–735. Manning, A. (1961). The eVects of artificial selection for mating speed in Drosophila melanogaster. Anim. Behav. 9, 89–92. Markow, T. A. (1981). Courtship behavior and control of reproductive isolation between Drosophila mojavensis and Drosophila arizonensis. Evolution 35, 1022–1026. McGuire, T. R., and Tully, T. (1987). Characterization of genes involved with classical conditioning that produce diVerences between bidirectionally selected strains of the blow fly Phormia regina. Behav. Genet. 17, 97–107. MeVert, L. M. (1995). Bottleneck eVects on genetic variance for courtship repertoire. Genetics 139, 365–374. MeVert, L. M. (1999). How speciation experiments relate to conservation biology. BioScience 49, 701–715. MeVert, L. M. (2000). The evolutionary potential of morphology and mating behavior: The role of epistasis in bottlenecked populations. In ‘‘Epistasis and the Evolutionary Process’’ (J. B. Wolf, E. D. Brodie, and M. J. Wade, Eds.), pp. 177–193. Oxford University Press, Oxford, U.K. MeVert, L. M., and Bryant, E. H. (1991). Mating propensity and courtship behavior in serially bottlenecked lines of the housefly. Evolution 45, 293–306. MeVert, L. M., and Bryant, E. H. (1992). Divergent ambulatory and grooming behavior in serially bottlenecked lines of the housefly. Evolution 46, 1399–1407. MeVert, L. M., Hicks, S. K., and Regan, J. L. (2002). Nonadditive genetic eVects in animal behavior. Amer. Natur. 160S, S198–213. MeVert, L. M., and Regan, J. L. (2002). A test of speciation via sexual selection on female preferences. Anim. Behav. 64, 955–965. MeVert, L. M., Regan, J. L., and Brown, B. W. (1999). Convergent evolution of the mating behaviour of founder-flush populations of the housefly. J. Evol. Biol. 12, 859–868. Nol, E., Cheng, K., and Nichols, C. (1996). Heritability and phenotypic correlations of behavior and dominance rank of Japanese quail. Anim. Behav. 52, 813–820. Oliverio, A. (1971). Genetic variations and heredity in a measure of avoidance learning in mice. J. Comp. Physiol. Psychol. 74, 390–397. Oliverio, A., Catellano, C., and Messeri, P. (1972). A genetic analysis of avoidance, maze and wheel running behaviors in the mouse. J. Comp. Physiol. Psychol. 79, 459–473. Park, E.-C., and Horvitz, H. R. (1986). Mutations with dominant eVects on the behavior and morphology of the nematode Caenorhabditis elegans. Genetics 113, 821–852. Pimentel, R. A. (1979). ‘‘Morphometrics. The Multivariate Analysis of Biological Data.’’ Kendall/Hunt Publishing Company, Dubuque, IA. Polivanov, S. (1975). Response of Drosophila persimilis to phototactic and geotactic selection. Behav. Genet. 5, 255–267. Price, T. (1998). Sexual selection and natural selection in bird speciation. Philos. Trans. R. Soc. Lond., B, Biol. Sci. 353, 251–260. Price, T., Schluter, D., and Heckman, N. E. (1993). Sexual selection when the female directly benefits. Biol. J. Linn. Soc. 48, 187–211. Rasmuson, B., Rasmuson, M., and Nygren, J. (1977). Genetically controlled diVerences in behavior between cycling and noncycling populations of field voles (Microtus agrestis). Hereditas 87, 33–42. Regan, J. L., MeVert, L. M., and Bryant, E. H. (2003). A direct experimental test of founderflush eVects on the evolutionary potential for assortative mating. J. Evol. Biol. 16, 302–312.

6. Quantitative Genetic Interactions in House Fly Courtship

213

Ringo, J., and Wood, D. (1983). Pupation site selection in Drosophila simulans. Behav. Genet. 13, 17–27. RoV, D. A., and Mousseau, T. A. (1987). Quantitative genetics and fitness: Lessons from Drosophila. J. Hered. 58, 103–118. Sacca, G. (1964). Comparative bionomics of the genus. Musca Ann. Rev. Entomol. 9, 341–358. Sewell, D., Burnet, B., and Connolly, K. (1975). Genetic analysis of larval feeding behaviour in Drosophila melanogaster. Genet. Res. Camb. 24, 163–173. Schultz, S., and Lynch, M. (1997). Mutation and extinction: The role of variable mutational eVects, synergistic epistasis, beneficial mutations, and degree of outcrossing. Evolution 51, 1363–1371. Sharp, P. M. (1984). The eVect of inbreeding on competitive male-mating ability in Drosophila melanogaster. Genetics 106, 601–612. Shaw, F. H., Shaw, R. G., Wilkinson, G. S., and Turelli, M. (1995). Changes in genetic variances and covariances: G whiz!. Evolution 49, 1260–1267. Siegel, R. W., and Hall, J. C. (1979). Conditioned responses in courtship of normal and mutant Drosophila. Proc. Natl. Acad. Sci. 76, 3430–3434. Stevens, L. (1994). Genetic analysis of cannibalism behavior in Tribolium flour beetles. In ‘‘Quantitative Genetic Studies of Behavioral Evolution’’ (C. R. B. Boake, Ed.), pp. 206–227. University of Chicago Press, Chicago. Taylor, C. E. (1975). DiVerences in mating propensities: Some models for examining the genetic consequences. Behav. Genet. 5, 381–393. Thornhill, N. W. (1993). ‘‘The Natural History of Inbreeding and Outbreeding: Theoretical and Empirical Perspectives.’’ University of Chicago Press, Chicago. Tobin, E. N., and StoVolano, Jr., J. G. (1978). The courtship of Musca species found in North America. 1. The House Fly, Musca domestica. Ann. Entomol. Soc. Amer. 66, 1249–1257. Travis, J. (1994). Size-dependent behavioral variation and its genetic control within and among populations. In ‘‘Quantitative Genetic Studies of Behavioral Evolution’’ (C. R. B. Boake, Ed.), pp. 165–187. University of Chicago Press, Chicago. Via, S., and Lande, R. (1985). Genotype-environment interaction and the evolution of phenotypic plasticity. Evolution 39, 505–523. Via, S., and Lande, R. (1987). Evolution of genetic variability in a spatially heterogeneous environment: EVects of genotype- environment interaction. Genet. Res. Camb. 49, 147–156. Wang, D. L., Zhu, J., Li, Z. K., and Paterson, A. H. (1999). Mapping QTLs with epistatic eVects and QTL X environment interactions by mixed linear model approaches. Theor. Appl. Genet. 99, 1255–1264. Weber, K. E. (1996). Large genetic change at small fitness cost in large populations of Drosophila melanogaster selected for wind tunnel flight: Rethinking fitness surfaces. Genetics 144, 205–213. Whitlock, M. C., Phillips, P. C., and Wade, M. J. (1993). Gene interaction aVects the additive genetic variance in subdivided populations with migration and extinction. Evolution 47, 1758–1769. Willis, J. H., and Orr, H. A. (1993). Increased heritable variation following population bottlenecks: The role of dominance. Evolution 47, 949–956. Wolf, J. B. Brodie, E. D. and Wade, M. J. (Eds.) (2000) ‘‘Epistasis and the Evolutionary Process.’’ Oxford University Press, Oxford, U.K. Wolf, J. B., Brodie III, E. D., Cheverud, J. M., Moore, A. J., and Wade, M. J. (1998). Evolutionary consequences of indirect genetic eVects. Tr. Ecol. Evol. 13, 64–69. Wright, S. (1969). ‘‘Evolution and the Genetics of Populations.’’ University of Chicago Press, Chicago.

7

____________________________________________________________________________

Phototropins, Other Photoreceptors, and Associated Signaling: The Lead and Supporting Cast in the Control of Plant Movement Responses Bethany B. Stone, C. Alex Esmon, and Emmanuel Liscum University of Missouri–Columbia Columbia, Missouri 65211

I. Introduction II. Phototropins: The Primary Receptors in Light-Induced Plant Movement Responses III. Cryptochromes and Phytochromes: Modulatory Receptors in Light-Induced Plant Movement Responses IV. Phototropism: Plant Movements of Entire Organs V. Stomatal Opening: Plant Movements at the Cellular Level VI. Chloroplast Relocation: Plant Movements at the Subcellular Level VII. Conclusion References

I. Introduction One common feature of all organisms is their ability to perceive and respond to changes in the environment. For animals this might mean seeking out shelter when the weather threatens or gathering food as winter approaches. However, plants are stationary, confined, and left to adjust using other means. Although they cannot move in a metazoan sense, plants do adjust by various minor, but vital, changes in growth, positioning of subcellular components, intake of resources, and developmental timing. In order to trigger these responses, plants must sort through multiple signals from their surroundings, perhaps the most vital of which is the light environment. Because light is essential for plant growth and reproduction, evolution has given plants multiple means for sensing and responding to alterations in light quality, quantity, and direction, such that photosynthetic eYciency is maximized while simultaneously minimizing photooxidative damage. In this chapter we focus on light-driven plant movements that occur in response to perception of changes in light direction, intensity, and/or duration within the near-ultraviolet (UV), wavelengths between 320–390 nm, and blue (390 and 500 nm) regions of the electromagnetic spectrum. The most obvious Current Topics in Developmental Biology, Vol. 66 Copyright 2005, Elsevier Inc. All rights reserved. 0070-2153/05 $35.00

215

216

Stone et al.

plant movement induced by this portion of the light spectrum is phototropism, or directional growth response to directional light cues (Liscum, 2002). Plants also respond to changes in the intensity of incident near-UV or blue light by adjusting the position of chloroplasts (organelles that contain light-harvesting complexes required for photosynthesis) within cells (Wada et al., 2003). The ‘‘movement’’ (swelling and shrinking) of guard cells, a specialized pair of epidermal cells on stems and leaves that form stomata—or pores— through which gas exchange necessary for photosynthesis is controlled, is also regulated by the light quality and intensity, with pore size increasing in bluerich environments of moderate intensity (Assmann, 1993; Schroeder et al., 2001). Although research in these areas has a long history, this chapter focuses on recent developments, providing just enough background to bring the reader up to speed. Many of the recent discoveries have come from the application of molecular genetics in the model plant Arabidopsis thaliana (Liscum, 2002; Ma¨ ser et al., 2003). Therefore, unless stated otherwise, the reader may assume that all work reported in this chapter on Arabidopsis. In order to sample changes in their light environment, plants utilize no fewer than nine characterized photoreceptors, each with specific sensitivity to diVerent wavelengths and intensities of light. These light sensors are classified into three categories: phytochromes (phyA, phyB, phyC, phyD, and phyE) (Gyula et al., 2003; Parks, 2003; Quail, 2002), cryptochromes (cry1 and cry2) (Lin and Shalitin, 2003; Liscum et al., 2003), and phototropins (phot1 and phot2) (Briggs and Christie, 2002; Liscum et al., 2003). Members of all three of these categories of light receptors are used in varying degrees to adjust plant movements, but the last mentioned, the phototropins, are the primary receptors for the movement responses to be discussed here (phototropism, stomatal aperture control, and chloroplast relocation). The cryptochromes and phytochromes represent modulatory photoreceptors for the phototropin-dependent movement responses. In the next two sections, some of the basic properties of each of these receptor classes are discussed, followed by several sections dedicated to the responses themselves and signaling events associated with the receptors and those responses.

II. Phototropins: The Primary Receptors in Light-Induced Plant Movement Responses All of the plant movements reviewed in this chapter require light perception by at least one of two phototropin (phot) light sensors. The first phototropin, PHOT1 (see Table I for nomenclature), was identified genetically in a screen for Arabidopsis mutants with altered phototropic responses in low-intensity, directional blue light (Liscum and Briggs, 1995). The second phototropin, PHOT2, was identified initially by its sequence homology to

217

7. Plant Photoreceptors and Associated Signaling Table I Light Receptor Nomenclature Designations Molecular Form

Phototropins

Cryptochromes

Phytochromes

Wild-type genes Mutant genes Apoprotein Holoproteina Photochemical form Red light absorbing Far-red light absorbing

PHOT1, PHOT2 phot1, phot2 PHOT1, PHOT2 phot1, phot2

CRY1, CRY2 cry1, cry2 CRY1, CRY2 cry1, cry2

PHYA, PHYB, etc. phyA, phyB, etc. PHYA, PHYB, etc. phyA, phyB, etc.

a

PrA, PrB, etc. PfrA, PfrB, etc.

Apoprotein with associated chromophore.

PHOT1 (58% identity over the entire length of the proteins) (Jarillo et al., 1998). Subsequently, phot2 mutants were identified in a screen for plants that lacked a chloroplast avoidance responses in high-intensity light (Kagawa et al., 2001). Despite initial genetic association of a single phototropin with a single response, a number of studies have now demonstrated that phot1 and phot2 more often than not function as partially redundant receptors for multiple responses, as will be discussed throughout this chapter. Phototropins are members of a larger family of proteins known as the LOV domain family (Crosson et al., 2003). Although the biochemical functions of members of this protein family vary, each utilizes at least one conserved approximately 110-amino acid motif, the LOV domain, to regulate its function. The LOV domain derives its name from the fact that it acts as a light, oxygen, or voltage sensor domain (Crosson et al., 2003; Huala et al., 1997; Taylor and Zhulin, 1999; Zhulin and Taylor, 1997). Within the LOV domain family, the phototropins are somewhat atypical because they contain two LOV domains (designated LOV1 and LOV2) rather than one (Crosson et al., 2003; Huala et al., 1997) (Fig. 1). The ability of the phototropins to function as near-UV–blue-light receptors is derived from the association (noncovalent in a dark state) of a single flavin mononucleotide (FMN) molecule with each LOV domain, and the unique photocycle that occurs within this chromopeptide (Crosson et al., 2003; Kennis et al., 2003, 2004; Liscum et al., 2003). Although a detailed description of the LOV domain photocycle is beyond the scope of this chapter, it is important to note that light drives the formation of a covalent adduct between the FMN and a critical cysteine residue. This cysteinyl–FMN adduct is proposed to represent the ‘‘active state’’ of phototropins, one that results in the derepression of the carboxyl-terminal serine/threonine protein kinase domain of the phototropins (see Fig. 1) to allow downstream signaling (Christie et al., 2002; Harper et al., 2003, 2004).

218

Stone et al.

Figure 1 Domain organization of higher plant phototropins, phytochromes, cryptochromes, NPH3, NPH4/ARF7, and MSG2/IAA19. Phototropins (phot) have two light, oxygen, and voltage (LOV) domains, each with a single flavin mononucleotide (FMN) binding site in their N-termini and a Serine/Threonine (S/T) kinase domain in their C-termini. Higher plant phytochromes (phy) bind phytochromobilin (PfB) in a bilin lyase domain (BLD) and have an S/T kinase domain in their C-termini. Cryptochromes (cry) consist of a photolyase homology region (PHR) with deazaflavin or pterin (D/P) and flavin adenine dinucleotide (FAD) chromophore binding sites, as well as a cryptochrome C-terminal (CCT) domain. NPH3 has five sequence conserved domains (across the NPH3/RPT2 protein family) with a BTB/POZ (Broad-Complex, Tramtrack, and Bric-a-brac/poxvirus and zinc finger) domain spanning domains Ia and Ib, and a coiled-coil C-terminal to domain IV. NPH4/ARF7 has a DNA-binding domain (DBD) and a C-terminal domain (CTD) that is very similar to the CTD in MSG2/IAA19.

Interestingly, in Arabidopsis and rice, the eYciency of the ‘‘active state’’ formation and dark recovery have been shown to diVer between phot1 and phot2 (Kasahara et al., 2002). These distinct kinetic properties between the phototropins are consistent with slightly divergent amino acid sequences (Kagawa et al., 2001) and, when combined with diVerences in gene expression patterns (Jarillo et al., 2001; Kanegae et al., 2000), help explain why phot1 and phot2 exhibit light intensity-specific divergence in physiological function (discussed in later sections). There are also diVerences in the functional properties of the LOV1 and LOV2 domains within a given phototropin. The clearest example of this comes from a study by Christie and colleagues (2002) in which amino acid replacements were made to demonstrate that formation of the cysteinyl–FMN adduct in LOV2, but not LOV1, is necessary for the phototropic function of phot1. Although the role of LOV1 in phot1 function has not been clearly elucidated, it has been proposed to function as a dimerization motif (Christie et al., 2002). How is an ‘‘active state’’ of a phototropin converted to a meaningful cellular signal that initiates downstream biochemical events? As introduced

7. Plant Photoreceptors and Associated Signaling

219

previously, it is thought that absorption of near-UV/blue-light photons by the LOV2 photosensory domain of the phototropins leads to a derepression of the carboxyl-terminal protein kinase domain (Christie et al., 2002; Harper et al., 2003, 2004). Implicit in this hypothesis is the expectation that the kinase domain of the phototropins somehow functions in transduction of the ‘‘activated state’’ signal. A recently identified phot1 missense allele provides compelling support for this hypothesis. This allele, phot1–7, conditions an aphototropic phenotype indistinguishable from that of the phot1–5-null allele, and, while producing wild-type levels of PHOT1 protein, lacks lightinduced autophosphorylation activity (E. Liscum, unpublished observations). The aspartate806 that is converted to asparagine in the phot1–7 protein is normally necessary for Mg2þ chelation and phosphotransfer activity of protein kinases (Hanks and Hunter, 1995). The same mutation—when engineered into phot1—also eliminates light-dependent autophosphorylation activity in an insect cell assay system (Christie et al., 2002). At present there are no known phosphorylation substrates for the phototropins except the phototropins themselves. Thus, what role the phosphorylation activity of the phototropins plays in signal propagation remains an open question. A recent study by Salomon and colleagues (2003) has demonstrated that autophosphorylation of oat phot 1a is hierarchical: four serine residues (two amino-terminal and two slightly carboxyl-terminal to LOV1) being phosphorylated in low light (1
M m2) and four serines (all carboxylterminal to the previous four but still amino-terminal to LOV2) being phosphorylated at higher intensities (10
M m2). At least two hypotheses seem plausible to explain why such a hierarchy might exist and how it could be linked to signaling. First, phosphorylation of the low-light sights might influence protein–protein interactions between a phototropin and an interacting partner via changes in electrostatic properties (Liscum et al., 2003). Second, although the high-light phosphorylation events are clearly not necessary for primary signaling because they occur at light intensities 2 to 3 orders of magnitude greater than those required to induce phototropism itself (Liscum, 2002), they might be prerequisite for receptor desensitization and recycling under constant illumination conditions (Liscum, 2002; Salomon et al., 2003).

III. Cryptochromes and Phytochromes: Modulatory Receptors in Light-Induced Plant Movement Responses The cryptochrome (cry) class of photoreceptors are primarily involved in regulating the transition from etiolated (dark-grown) to de-etiolated (light-grown) developmental states and the timing of floral induction (Lin and Shalitin, 2003). The first developmental blue-light receptor identified in

220

Stone et al.

any organism was cryptochrome 1 of Arabidopsis (Ahmad and Cashmore, 1993). Like the phototropins, the cryptochromes are represented by a small family of proteins, in this case three: CRY1, CRY2, and CRY3 (Liscum et al., 2003; see Table I for nomenclature). Cryptochromes exhibit significant amino acid homology to DNA photolyases (Ahmad and Cashmore, 1993), enzymes that catalyze the blue-light-dependent repair of UV-damaged DNA (Sancar, 2003). Yet, cryptochromes exhibit no DNA repair activity (Cashmore et al., 1999). Why not? When the amino acid sequences of the photolyases and cryptochromes are compared, one immediately notices that photolyases are essentially a photosensory domain capable of binding two chromophores, an invariant catalytic flavin adenine dinucleotide (FAD) and a light-harvesting deazaflavin or pterin (Sancar, 2003). Cry1 and cry2 also are composed of a photolyaselike photosensory domain and an additional carboxyl-terminal extension (designated the CCT) conserved within the cryptochromes but absent in the photolyases (Lin and Shalitin, 2003; Liscum et al., 2003) (see Fig. 1). Cry3 represents a unique example altogether because it lacks the CCT and is targeted to the chloroplast rather than to the nucleus like cry1, cry2, and the photolyases (Brudler et al., 2003; Cashmore et al., 1999; Kleine et al., 2003; Sancar, 2003). Given these unique properties and the lack of evidence that it aVects any plant movement response, cry3 will not be discussed further here. A number of recent studies indicate that the CCT is essential for cry1 and cry2 function (Lin and Shalitin, 2003; Shalitin et al., 2002; Wang et al., 2001; Yang et al., 2001), providing one point for divergence in function from the photolyases. There is also structural evidence that the fundamental photochemical and photophysical events associated with photolyase action and cryptochrome signaling are diVerent (Brudler et al., 2003; Giovani et al., 2003). However, at present the precise mechanism(s) by which cryptochromes transduce perceived light signals is not understood. As was observed with phot1 and phot2, light intensity-dependent functional divergence has also been observed with cry1 and cry2. For example, with respect to the hypocotyl growth inhibition that is part of the suite of de-etiolation responses, cry1 is required for high-intensity light perception, whereas cry2 primarily functions in low-insensity light (Ahmad et al., 1998; Guo et al., 1999; Lin et al., 1998). Although the cryptochromes are not required for perception of directional light cues that induce phototropism (Lasce`ve et al., 1999; Liscum and Briggs, 1995), recent results indicate that cry1 and cry2 exhibit diVerent light dependencies in their ability to modulate phototropin-dependent phototropism. In particular, Whippo and Hangarter (2003) have demonstrated that cry1 is necessary for the attenuation of phototropism under high-intensity light conditions, whereas cry2 plays only a minor role.

7. Plant Photoreceptors and Associated Signaling

221

What properties of cry1 and cry2 allow them to discriminate among light conditions? At present there is no evidence that the photosensory properties of cryptochromes are suYciently diVerent to explain the often high-versus low-intensity functions of cry1 and cry2, respectively. DiVerences in protein stability between these receptors might provide some explanation for their diVering physiology roles. In particular, although cry1 appears stable in both darkness and light (Ahmad et al., 1998), cry2 exhibits light-dependent degradation that increases with increasing light intensity (Guo et al., 1999; Shalitin et al., 2002). These results suggest that cry2 should be of little physiological consequence under high-intensity light conditions, because it will be virtually absent under those conditions. Why cry2 functions more strongly than cry1 in low light is less clear. However, it is interesting to note that in contrast to cry2, which exhibits apparent constitutive nuclear localization (Guo et al., 1999; Kleiner et al., 1999), cry1 might be excluded from the nucleus in response to light (Guo et al., 1999; Yang et al., 2000). Thus, it is possible that cry1 and cry2 signaling might be discriminatory in whichever compartment of the cell is utilized, or that both function in the nucleus but in diVerent fashions (Lin and Shalitin, 2003; Liscum et al., 2003). The phytochrome family of primarily red/far-red light (600–800-nm region of the spectrum)-absorbing photoreceptors represents a second class of receptors involved in the developmental transition from etiolated to deetiolated growth and flowering in plants (Fankhauser, 2001; Gyula et al., 2003; Quail, 2002). Phytochromes also quite uniquely mediate seed germination and neighbor-sensing responses (Casal, 2000). One of the most striking photophysical characteristics of the phytochromes—the property that allowed for their biochemical isolation [see Sage (1992) for a wonderful story on the history of phytochrome research]—is that they can exist in one of two photo-interconvertable forms (see Table I for nomenclature): an inactive red light-absorbing form (Pr) and an active far-red light-absorbing form (Pfr) (Parker et al., 1952). Under natural light conditions, phytochromes establish a photoequilibrium between the two forms that is regulated by the ratio of red to far-red light present (Casal, 2000; Fankhauser, 2001). In doing so, the phytochromes are able to regulate light responses not only in an intensity-dependent fashion, but also in a quality-dependent manner. Like the phototropins, the phytochromes are light-responsive protein kinases (Quail, 2002). Although phytochromes are histidine kinase-like in ancestry (Bhoo et al., 2001; Davis et al., 1999; Yeh and Lagarias, 1998), higher plant versions function as serine/threonine kinases (Yeh and Lagarias, 1998). The phytochrome kinase domain is located in the carboxyl-terminal portion of the protein, and the photosensory chromophore-bearing region is found in the amino-terminal region (Fankhauser, 2001) (see Fig. 1). Unlike the phototropins and cryptochromes that utilize flavin-based chromophores, which impart near-UV/blue-light absorptive properties to those receptors, the

222

Stone et al.

phytochromes use a single covalently bound linear tetrapyrrole, phytochromobilin, as their light-absorbing cofactor (Fankhauser, 2001; Gyula et al., 2003). This chromopeptide configuration results in the now-classic two photosensory-state absorption spectra for the phytochromes that exhibit near-UV and red maxima for the Pr form, and blue and far-red maxima for the Pfr form (see Sage, 1992). Z/E isomerization of photochromobilin and presumed associated protein structural changes allow the phytochromes to be photo-interconverted between the two forms, such that absorption of red light by the Pr form results in formation of the Pfr form, which can then be converted into the Pr from by absorption of far-red light, and so on (Ru¨ diger and Thu¨ mmler, 1994). This unique photochemical property of the phytochromes allows them to function as eYcient molecular switches. Although the precise biochemical and structural mechanism(s) by which phytochromes transduce signals is currently unknown, considerable progress has been made in the identification of phytochrome-interacting proteins (Quail, 2000; Wang and Deng, 2002). A majority of these proteins have been found to be nuclear residents, and several act as transcriptional regulators, suggesting a simple mechanism by which phytochromes could influence a variety of morphological and developmental responses via changes in gene expression (Shimizu-Sato et al., 2002). Indeed, recent microarray studies show that phytochrome action regulates the expression of a large number of genes (Ma et al., 2001; Tepperman et al., 2001, 2004; Wang et al., 2002). It is interesting to note that in many cases the interaction of a phytochrome with a partner protein is not light regulated (Wang and Deng, 2002). How then is light influencing the function of such interacting proteins? The findings that phytochromes exhibit clear changes in intracellular localization in responses to light provide one likely explanation (Nagy et al., 2000, 2001). For example, although a majority of phyB is cytoplasmic in etiolated seedlings, exposure to red light results in its rapid translocation to the nucleus, where it forms nuclear speckles with increased time of exposure to light, far-red light being inhibitory to this process (Gil et al., 2000; Kircher et al., 1999). Once in the nucleus, the activated phyB can physically interact with transcriptional regulators such as phytochrome-interacting transcription factor (PIF3) and lead to alterations in gene expression (Martinez-Garcia et al., 2000; Shimizu-Sato et al., 2002).

IV. Phototropism: Plant Movements of Entire Organs Phototropism in plants requires perception of and response to directional blue light. As an example, let us first consider the positive phototropic response of seedling stems. If light is perceived from a direction not parallel to the main axis of growth (i.e., the stem), the plant will reorient its growth

7. Plant Photoreceptors and Associated Signaling

223

axis by initiating cell expansion in the flank of the stem opposite the incident light, while cell expansion will simultaneously be inhibited in the irradiated flank. This diVerential growth response results in bending of the stem toward the light (Liscum, 2002). In the roots, the opposite occurs, eVectively redirecting growth of the root away from the light (Okada and Shimura, 1992). The capability of the plants to reorient growth in this way has long been thought to represent an adaptive mechanism to maximize quality and quantity of light being absorbed for photosynthesis, as well as nutrient and water acquisition in the soil (see Iino, 1990; Liscum, 2002). A recent study by Galen and colleagues (2004) provides the first direct demonstration that genetic loci necessary for seedling phototropism, including phot1, provide a fitness advantage to plants in the natural environment. Although phototropism is a simple process from a physiological perspective, it is really a complex system of light perception and signal transduction that has evolved to allow eVective ‘‘sampling’’ of the light environment in order to alter morphology in an adaptive fashion. At least six separate photoreceptors, including phot1 (Liscum and Briggs, 1995), phot2 (Sakai et al., 2000), phyA (Parks et al., 1996), phyB (Janoudi et al., 1997), and cry1 and cry2 (Whippo and Hangarter, 2003), participate in these events. As introduced earlier, the phototropins function as the primary phototropic receptors, whereas the phytochromes and cryptochromes apparently modulate the responses initiated by the former. Given that phot1 was first identified in a screen for nonphototropic hypocotyls (stems in dicotyledonous seedlings) in low-intensity light, it is not surprising that phot1 is the primary photoreceptor in low-intensity blue light, both for positive hypocotyl phototropism and negative root phototropism (Liscum and Briggs, 1995). Although roots of phot1 mutants remain aphototropic in high-intensity light conditions, hypocotyls retain partial ability to bend toward the light (Sakai et al., 2000). These results indicate the function of at least one additional photoreceptor. The lack of hypocotyl phototropism in phot1phot2 double mutants in high-intensity light indicates that phot2 acts redundantly with phot1 in hypocotyl phototropic responses under these conditions (Sakai et al., 2001). Phototropins are the ‘‘on’’ and ‘‘oV’’ switch for phototropism, but other photoreceptors are the ‘‘dimmer’’ switch, controlling the degree of phototropic response. Irradiation with red light prior to exposure of seedlings to directional blue light enhances phototropic curvature, an enhancement lacking in phyA mutants (Parks et al., 1996). Whippo and Hangarter (2003) used time-lapse imaging and mathematical techniques to closely analyze phototropism in blue-light receptor single and double mutants. Using these methods, the authors were able to present the most convincing evidence yet that cryptochromes work synergistically with phototropins to orchestrate the degree of phototropic curvature. Interestingly, the data suggest that

224

Stone et al.

the two types of blue-light photoreceptors limit the phototropic response in high-intensity light (>100
M m2 s1) and enhance the response in low-intensity light (<1.0
M m2 s1). Research continues to explore possible events downstream of photoperception by phototropins. One likely component, NPH3, was isolated in the same mutant screen that identified PHOT1. Once cloned, it was discovered that NPH3 is a member of a 33-member gene family, designated the NRL (NPH3/RPT2-LIKE) family (Motchoulski and Liscum, 1999; E. Liscum, unpublished observations). Examination of the NPH3 protein sequence does not suggest any obvious biochemical function; however, several individual regions of the protein might provide some clues. For example, within the fourth conserved sequence domain of NPH3 (relative to other members of the NRL family) there is a highly conserved metazoan-type tyrosine phosphorylation site. Interestingly, an in-frame deletion of the target tyrosine results in the nph3–2 allele results in an aphototropic response indistinguishable from that of the nph3–6-null allele (Motchoulski and Liscum, 1999), indicating that the tyrosine residue (and possibly its phosphorylation) is required for NPH3 function. Two additional regions of interest are a BTB/ POZ (broad complex, tramtrack, bric a` brac/pox virus and zinc finger) domain in the amino terminus and a coiled-coil domain in the carboxyl terminus (see Fig. 1)—both being candidates for protein–protein interaction sites (Motchoulski and Liscum, 1999). With respect to the relationship between NPH3 and phot1, aphototropic nph3 mutants are phenotypically similar to phot1 mutants (Liscum and Briggs, 1995, 1996; Motchoulski and Liscum, 1999). Yeast two-hybrid and in vitro pull-down experiments revealed that the coiled-coil region of NPH3 interacts with the amino-terminal LOV domain-containing portion of phot1 (Motchoulski and Liscum, 1999). Although both proteins associate with the plasma membrane (Motchoulski and Liscum, 1999; Sakamoto and Briggs, 2002), neither contains an obvious membrane-localization sequence (Huala et al., 1997; Motchoulski and Liscum, 1999). It has been proposed that NPH3 forms a signaling complex with phot1 and other currently unidentified proteins at the plasma membrane (Liscum, 2002; Motchoulski and Liscum, 1999). There is also evidence that the integrity of this complex (e.g., whether phot1 and NPH3 are physically interacting) may aVect signaling (Liscum, 2002; Motchoulski and Liscum, 1999). Future biochemical studies of the phot1–NPH3 complex should clarify the mechanism(s) by which this complex functions in phototropism. NPH3 is just one of 33 genes that form the NRL family. Given the high degree of homology between members of this family, it remains possible that additional members of the NRL family also function as phototropininteracting proteins and participate in phototropic signaling. This idea turns out to have merit, because the RPT2 locus, identified by the phototropic

7. Plant Photoreceptors and Associated Signaling

225

defects of a mutant allele of this locus (Okada and Shimura, 1992, 1994), encodes a member of the NRL family (as should be obvious from the name of the family) (Sakai et al., 2000) and has been shown to interact with phot1 (Inada et al., 2004). Cell fractionation and precipitation experiments demonstrated that RPT2, like phot1 and NPH3, is present in the plasma membrane fraction (Inada et al., 2004). It is intriguing to note that RPT2 and NPH3 appear to form heterodimers (Inada et al., 2004), and NPH3 is capable of forming homodimers (A. Motchoulski and E. Liscum, unpublished observations), suggesting that the phot1 complex could be a very dynamic one. Although there is no evidence that either NPH3 or RPT2 interact with phot2, the homology between phot1 and phot2 suggests that members of the NRL family are likely to play a role in phot2 signaling. It remains untested which, if any, of the remaining NRL family members interact with phot2 and if a precise combination of phototropin and NRL protein(s) lends specificity to the cascade of signaling events that results in the variety of phototropin-dependent responses. One signal amplification process that may result from the activity of a phot1 signaling complex is modulation cytosolic calcium concentration ([Ca2þ]cyt). For example, the increase in [Ca2þ]cyt that occurs in wild-type, cry1, and cry2 seedlings within 20 sec of blue-light stimulation is dramatically reduced in phot1 single mutant seedlings (Baum et al., 1999) and is completely eliminated in phot1phot2 double mutants (Babourina et al., 2002; Stoelzle et al., 2003). Electrophysiological studies of mesophyll protoplasts demonstrated that the blue-light responsiveness of plasma membranelocalized calcium channels is abrogated by the presence of the protein kinase inhibitor K252a (Stoelzle et al., 2003). Presumably K252a is preventing the light activation of the phot1 and phot2 kinase domains and thus signaling processing. Unfortunately, no concrete relationship between Ca2þ signaling and phototropism has been demonstrated to date. Instead, recent work suggests that although Ca2þ signaling is involved in phototropin-driven hypocotyl growth inhibition, it is not required for phototropism (Folta et al., 2003). Another small molecule historically associated with phototropism is the plant hormone auxin, more specifically indole-3-acetic acid, or IAA (Went and Thimann, 1937). For over half a century scientists have provided compelling, yet circumstantial, evidence that the formation of a lateral auxin gradient across an organ might allow for polar changes in growth and development (see Trewavas et al., 1992). Auxin influences diVerent plant tissues and organs in diVerent ways. For example, in the stem auxin is responsible for cell elongation and increases in cell division, while it inhibits growth in the root. Based on these properties, the Cholodny–Went theory (Went and Thimann, 1937) was put forth nearly 70 years ago to explain the role of auxin in tropic growth responses. This theory holds that increases of

226

Stone et al.

auxin concentration along the shaded side of a phototropically stimulated plant would result in a shoot that bends toward the light because of auxininduced growth (Cholodny, 1927; Went and Thimann, 1937). In the root auxin would also increase on the side away from the incident light, but in this case growth would be inhibited, causing the root to bend away from the light. Although numerous studies have been performed to test the Cholodny–Went theory, none have provided a direct cause-and-eVect relationship between changes in auxin concentration and phototropism. Yet, recent mutant analyses have brought us closer to discerning the relationship between auxin transport and tropic growth responses (Chen et al., 1998; Friml et al., 2002; Luschnig et al., 1998; Marchant et al., 1999; Mu¨ ller et al., 1998; Rashotte et al., 2001; Watahiki and Yamamoto, 1997). With respect to the mechanisms that link phototropins to changes in auxin distribution, several possibilities have been suggested (Liscum, 2002). First, phot1 is ideally located at the plasma membrane, where it might directly interact with proteins responsible for a change in auxin transport, possibly through a phosphorylation cascade. Second, we have already discussed the hypothesis that phototropism-dependent increases in [Ca2þ]cyt might be an early component of signaling, and these ion changes might lead to activation of a Ca2þ-dependent protein kinase and indirectly cause the phosphorylation of an auxin transporter. It is also possible that phototropin-dependent signaling results in relocation of auxin eZux complexes from the basal end of the cell to the sides, redirecting the usual polar shoot-to-root flow of auxin to a lateral flow. This scenario is particularly attractive given the recent characterization of PIN3, a facilitator of auxin eZux. It was found that in addition to being required for stem and root tropic responses, PIN3 exhibits higher lateral than basal wall localization in stems, and its localization in a root columella cell can change within 2 min of gravitropic stimulation (Friml et al., 2002). In another study it was found that the predominantly basal cell localization of another facilitator of auxin transport, PIN1, is disrupted by exposure to unidirectional blue light and that this delocalization does not occur in phot1-null mutants (Blakeslee et al., 2004). Although intriguing, these results raise additional questions. For instance, how does phot1 influence these relocalization events? Does phot2 also play a role but under high-intensity light conditions? Does relocation of auxin eZux facilitator proteins result in the formation of a lateral gradient of auxin as predicted by the Cholodny–Went theory? Let us accept the strongly supported thesis that blue-light stimulation of phototropins does trigger the formation of an auxin gradient. How is such a gradient in morphogen interpreted to give rise to the observed diVerential growth response? There are two obvious possibilities, and neither is mutually exclusive of the other. First, accumulation of IAA, an acidic compound, in one flank of a stem might lead to the pH-dependent activation

7. Plant Photoreceptors and Associated Signaling

227

of enzymes, such as expansins (Cosgrove, 2000; Lee et al., 2001; Li et al., 2004), that promote cell wall loosening, allowing for simple turgor-driven increases in cellular expansion (Hager et al., 1971). Although this ‘‘acidgrowth hypothesis’’ (Hager et al., 1971) requires no new synthesis of RNAs or proteins for its truism, an alternative but equally plausible hypothesis exists that does require changes in gene expression and protein synthesis, as discussed later. In the same mutant screen that identified phot1 and nph3 mutants, a third aphototropic mutant, nph4, was isolated (Liscum and Briggs, 1995, 1996) that exhibits dramatically altered auxin responsiveness (Stowe-Evans et al., 1998; Watahiki and Yamamoto, 1997). When cloned, it was found that NPH4 encodes the auxin-responsive transcriptional activator ARF7 (Harper et al., 2000), a finding consistent with the severely impaired auxin-dependent gene expression responses of nph4-null mutant seedlings (Stowe-Evans et al., 1998). The finding that an auxin-responsive transcription factor (ARF) is necessary for proper phototropic curvature gives credence to the long-held notion that the phototropic response is based on an auxin gradient and further suggests that changes in gene expression and presumed abundance of the encoded proteins are components of the phototropic response system. NPH4/ARF7 is one member of a large family (as many as 23 members) of ARF proteins in Arabidopsis (Liscum and Reed, 2002). These transcriptional regulators are characterized by the presence of a DNA-binding domain (DBD) in their amino-terminal region, a carboxyl-terminal dimerization motif (CTD), and a variable middle region (MR), which functions as either an activation or repression domain depending upon the amino acid sequence (see Fig. 1) (Hagen and Guilfoyle, 2002; Liscum and Reed, 2002; Tiwari et al., 2003). Consistent with the impaired auxin-induced gene expression in nph4 loss-of-function mutants, ARF7 contains a glutamine-rich MR, which has been shown in transfected protoplast assays to function as a strong trans-activator domain (Tiwari et al., 2003). The ability of ARF7 to activate transcription also depends upon the ability of the DBD to bind to auxin response elements (AuxREs), which are simple TGTCTC repeats, within the promoter regions of a number of genes. Some examples of genes whose promoters include such an AuxRE motif and are regulated by ARFs include the Small Auxin-Upregulated RNAs (SAURs), GH3s, and AUX/IAAs (Hagen and Guilfoyle, 2002). The latter genes encode proteins that have been shown to heterodimerize with ARF proteins, via a conserved CTD motif (see Fig. 1) (Kim et al., 1997; Ulmasov et al., 1997, 1999). As discussed later, the Aux/IAA proteins function as repressors of ARF activity. What makes ARF proteins responsive to auxin? One current model (Liscum and Reed, 2002; Tatematsu et al., 2004) predicts that at low auxin concentrations, as would exist in hypocotyls of etiolated seedlings before

228

Stone et al.

exposure to a phototropic stimulus, ARFs are bound to AuxREs within specific target genes as inactive heterodimers with Aux/IAA proteins. When the concentration of auxin rises, as would occur in the shaded flank of a phototropically-stimulated hypocotyl, SCF-TIR1 (a multisubunit enzyme with three characterized subunits of SKP1, Cullin 1, and the TIR1 F-boxcontaining protein)-dependent proteolysis of the Aux/IAA proteins is stimulated (Gray et al., 2001; Kepinski and Leyser, 2004; Ramos et al., 2001; Zenser et al., 2001, 2003) and ARF–ARF homodimers are allowed to form, resulting in an active complex. In the case of activator ARFs, like NPH4/ARF7, transcription of target genes is thus stimulated in the presence of auxin (Liscum and Reed, 2002; Tiwari et al., 2003). Interestingly, NPH4/ARF7 may target its own repressor for transcriptional upregulation in response to auxin. Approximately 60% of the auxin inducibility of AUX/IAA19, whose encoded protein has been shown to heterodimerize with NPH4/ARF7, appears to be under the control of ARF7 (Tatematsu et al., 2004). What makes this finding particularly interesting is the additional observation that dominant mutations in IAA19 that are predicted to result in increased stability of the encoded mutant iaa19 protein condition an aphototropic response similar to that of arf7-null mutants, suggesting that the stabilized iaa19 protein continues to suppress ARF7 activity, even in the presence of increased auxin (Tatematsu et al., 2004). Although significant strides have been made in our understanding of how auxin is involved in the development of phototropic curvatures, much remains unresolved. As one glaring example, why are the target genes for regulation by NPH4/ARF7 that encode proteins necessary, not for additional transcriptional control, as is the case with Aux/IAA19, but rather for direct influence on the growth properties of the cells in question?

V. Stomatal Opening: Plant Movements at the Cellular Level Stomata are small pores found on leaf and stem surfaces that open and close in response to several environmental cues—including light—to control gas exchange and water loss via transpiration (Assmann, 1993; Schroeder et al., 2001; Zeiger, 1983). They are formed by two specialized epidermal cells, the guard cells, that are fixed together at the terminal ends of their long axes and control the aperture of the pore (Esau, 1977). In order for the stomata to ‘‘open,’’ the guard cells swell, pushing them out at the center of their long axes (because the ends are fixed in space) and widening the space, or pore, between them. ‘‘Closure’’ of stomata results from water loss from the guard cells. Stomata open in response to light in order to increase the influx of

7. Plant Photoreceptors and Associated Signaling

229

carbon dioxide destined for photosynthetic carbon fixation (Ma¨ ser et al., 2003). Although stomatal opening is sensitive to many external stimuli, including light, temperature, and humidity, only the light receptors have been identified. In 2001, Kinoshita and colleagues presented the first evidence of phototropin involvement in stomatal opening. Single mutations in either phot1 or phot2 retained normal stomatal control in both low- and highintensity blue light; however, phot1phot2 double mutants were unresponsive in both light intensities, indicating that phot1 and phot2 redundantly control blue-light-stimulated stomatal opening. As with other plant movement responses, downstream signaling events remain a primary area of focus. Yet, many mechanisms involved in stomatal aperture control have been elucidated, giving us a fairly developed model of likely subcellular events mediating this response. First, it is known that blue light triggers a proton pump (Hþ-ATPase) in the plasma membrane of guard cells, creating a membrane potential that drives a passive influx of Kþ ions through voltage-regulated channels (Schroeder et al., 1994). The increased concentration of Kþ raises the osmotic potential, resulting in an increase in turgor pressure and swelling of the guard cells (Schroeder et al., 2001). Second, the activity of the Hþ-ATPase is regulated by phosphorylation, the pump being active in a phosphorylated state (Kinoshita and Shimazaki, 1999). This observation makes direct regulation by a phototropin an attractive model. Recent findings that both the Hþ-ATPase and phototropins interact with a 14-3-3 protein in guard cells in a blue-lightdependent fashion make this possibility even more attractive (Emi et al., 2001; Kinoshita et al., 2003). Although an interaction between the phototropins and the Hþ-ATPase has not been described, it would appear that both are present in a complex that could allow such an interaction to occur.

VI. Chloroplast Relocation: Plant Movements at the Subcellular Level Perhaps the finest and most eloquent adjustment plants can make in response to light quality and quantity is the relocation of chloroplasts within a mesophyll cell (Wada et al., 2003). When light is optimal, chloroplasts are scattered within the cell; however, if the light intensity changes, the chloroplasts are relocated to adjust to that change. When exposed to highintensity blue light, chloroplasts exhibit an ‘‘avoidance response’’ in which they move to the anticlinal walls (i.e., edges of the cell that are parallel with the direction of incident light and perpendicular to the surface of the leaf) to minimize potential photooxidative damage. In contrast, when

230

Stone et al.

blue-light intensity is low, chloroplasts exhibit an ‘‘accumulation response’’ in which they move to the periclinal walls (i.e., cell edge parallel to the leaf surface and perpendicular to the vector of incident light) to maximize photosynthetic light capture. Chloroplast avoidance responses can also be triggered by mechanical stimuli, as Sato and others (1999) demonstrated in fern cells. This response suppresses any influence of low light that would normally result in chloroplast accumulation (Sato et al., 2001a). However, it remains unclear if this response is naturally occurring or an artifact caused by touch-stimulated increases in a transient signal that is also shared by the naturally occurring light-driven pathway. The first molecular breakthrough on the chloroplast movement responses with respect to light signaling came when Kagawa and Wada (2000) reported that phot1–5 mutants were slightly impaired in the accumulation response. Conclusive evidence in support of a role for phototropins in the chloroplast movement responses came when phot2 mutants were examined and found to essentially lack the avoidance response (Jarillo et al., 2001; Kagawa et al., 2001). Sakai and colleagues (2001) subsequently showed that both chloroplast movements are absent in phot1phot2 double mutants, suggesting that phot1 and phot2 act redundantly to mediate the accumulation response, whereas phot2 functions as the primary receptor for the avoidance response. Some lower plants, such as ferns and mosses, utilize red and blue light to modulate chloroplast positioning (Kadota et al., 2000; Sato et al., 2001a), and the process can be reversed by the phytochrome-dependent action of far-red light (Augustynowicz and Gabrys, 1999). However, in flowering plants it was believed that this process is exclusively modulated by blue light—until recently. DeBlasio and others (2003) reported that phytochromes play a role in chloroplast relocation; specifically, phyA and phyB modulate the transition between low and high fluence in the chloroplast avoidance response. In this study, they measured light transmittance through leaves at various fluence rates in single, double, or triple phyA, phyB, and phyD mutants and found that phyA and phyB exhibited enhanced chloroplast avoidance. These data suggest that phytochromes might influence the transition between the low-fluence response (modulated by both phot1 and phot2) and the high-fluence response (mediated exclusively by phot2). As with the other blue-light-dependent responses discussed in this chapter, the mechanisms downstream from photoperception have not been fully elucidated, but small pieces to the puzzle are being revealed. Use of bluelight microbeam irradiations with a fern system suggests that the signal for chloroplast avoidance is highly localized and cell-autonomous, whereas the chloroplast accumulation signal can travel through multiple cells and is sustained for longer periods (Kagawa and Wada, 1999; Wada et al., 2003).

231

7. Plant Photoreceptors and Associated Signaling 2þ

There are data that suggest the concentrations of Ca might regulate the type (accumulation or avoidance) and duration of chloroplast movement responses; however, the strongest evidence for Ca2þ-dependent chloroplast movement is as a result of mechanical stimuli, not light (Sato et al., 2001a). In this latter study, the ‘‘mechano-relocation’’ of chloroplasts within isolated cells required an external source of Ca2þ, whereas the photo-relocation did not. Furthermore, light-induced relocation of chloroplasts was not blocked by Ca2þ channel blockers, La3þ and Gd3þ, but mechanically stimulated chloroplast relocation was. These results indicate that Ca2þ movement across the plasma membrane is required for the mechanical stimulation of chloroplast avoidance and that other Ca2þ sources, such as vacuolar Ca2þ stores, or lower [Ca2þ]cyt might be used in photo-stimulated chloroplast movements. The idea that chloroplast relocation requires increases in [Ca2þ]cyt is consistent with evidence that downstream phototropin signaling might rely on changes in cytoplasmic Ca2þ concentrations. In plants, actin filaments provide a structural highway for organelle movements, including chloroplast relocation (Kandasamy and Meagher, 1999). This was recently demonstrated through the isolation of the chloroplast unusual positioning 1 (chup 1) mutant that is phenotypically defective in chloroplast positioning and genotypically defective in a gene that likely encodes part of the actin machinery (Oikawa et al., 2003). A study employing cytoskeletal inhibitors in the moss Physcomitrella patens suggested that signals derived from a blue-light receptor (presumably a phototropinrelated protein) might impact actin filaments for chloroplast movement, whereas those from a red-light receptor (presumably a phytochrome relative) might influence microtubules (Sato et al., 2001b). Although this type of subcellular highway discrimination might only occur in nonflowering plant species, it represents a testable model for the phototropin- and phytochrome-dependent chloroplast movement responses of species in this latter class.

VII. Conclusion Isolation of mutants in all phototropins, cryptochromes, and phytochromes have led to a cascade of discoveries and provided powerful tools for elucidating the molecular processes involved in blue light–driven plant movement responses. Future research will likely focus on events that occur downstream of photoperception and the crosstalk that apparently occurs between the multiple photoreceptor pathways. The long-term goal of this work is to understand the intricate processes that allow plants to adapt to—and thrive—in almost any environment.

232

Stone et al.

References Ahmad, M., and Cashmore, A. R. (1993). HY4 gene of A. thaliana encodes a protein with characteristics of a blue-light photoreceptor. Nature 366, 162–166. Ahmad, M., Jarillo, J. A., and Cashmore, A. R. (1998). Chimeric proteins between cry1 and cry2 Arabidopsis blue light photoreceptors indicate overlapping functions and varying protein stability. Plant Cell 10, 197–207. Assmann, S. M. (1993). Signal transduction in guard cells. Annu. Rev. Cell Biol. 9, 345–375. Augustynowicz, J., and Gabrys, H. (1999). Chloroplast movements in fern leaves: Correlation of movement dynamics and environmental flexibility of the species. Plant Cell Environ. 22, 1239–1248. Babourina, O., Newman, I., and Shabala, S. (2002). Blue light-induced kinetics of Hþ and Ca2þ fluxes in etiolated wild-type and phototropin-mutant Arabidopsis seedlings. Proc. Natl. Acad. Sci. USA 99, 2433–2438. Baum, G., Long, J. C., Jenkins, G. I., and Trewavas, A. J. (1999). Stimulation of the blue light phototropic receptor NPH1 causes a transient increase in cytosolic Ca2þ. Proc. Natl. Acad. Sci. USA 96, 13554–13559. Bhoo, S. H., Davis, S. J., Walker, J., Karniol, B., and Vierstra, R. D. (2001). Bacteriophytochromes are photochromic histidine kinases using a biliverdin chromphore. Nature 141, 776–779. Blakeslee, J. J., Bandyopadhyay, A., Peer, W. A., Makam, S. N., and Murphy, A. S. (2004). Relocalization of the PIN1 auxin eZux facilitation plays a role in phototropic responses. Plant Physiol. 134, 23–31. Briggs, W. R., and Christie, J. M. (2002). Phototropin 1 and phototropin 2: Two versatile plant blue-light receptors. Trends Plant Sci. 7, 204–209. Brudler, R., Hitomi, K., Daiyasu, H., Toh, H., Kucho, K., Ishiura, M., Kanehisa, M., Roberts, V. A., Todo, T., Tainer, J. A., and GetzoV, E. D. (2003). Identification of a new cryptochrome class. Structure, function and evolution. Mol. Cell 11, 59–67. Casal, J. J. (2000). Phytochromes, cryptochromes, phototropin: Photoreceptor interactions in plants. Photochem. Photobiol. 71, 1–11. Cashmore, A. R., Jarillo, J. A., Wu, Y. J., and Liu, D. (1999). Cryptochromes: Blue light receptors for plants and animals. Science 284, 760–765. Chen, R., Hilson, P., Sedbrook, J., Rosen, E., Caspar, T., and Masson, P. H. (1998). The Arabidopsis thaliana AGRAVITROPIC 1 gene encodes a component of the polarauxin-transport eZux carrier. Proc. Natl. Acad. Sci. USA 95, 15112–15117. Cholodny, N. (1927). Wuchshormone and tropismen bei den pflanzen. Biol. Zentralbl. 47, 604–626. Christie, J. M., Swartz, T. E., Bogomolni, R. A., and Briggs, W. R. (2002). Phototropin LOV domains exhibit distinct roles in regulating photoreceptor function. Plant J. 32, 205–219. Cosgrove, D. J. (2000). New genes and new biological roles for expansins. Curr. Opin. Plant Biol. 3, 73–78. Crosson, S., Rajagopal, S., and MoVat, K. (2003). The LOV domain family: Photoresponsive signaling modules coupled to diverse output domains. Biochemistry 42, 2–10. Davis, S. J., Vener, A. V., and Vierstra, R. D. (1999). Bacteriophytochromes: Phytochrome-like photoreceptors from nonphotosynthetic eubacteria. Science 286, 2517–2520. DeBlasio, S. L., Mullen, J. L., Luesse, D. R., and Hangarter, R. P. (2003). Phytochrome modulation of blue light-induced chloroplast movements in Arabidopsis. Plant Physiol. 133, 1471–1479.

7. Plant Photoreceptors and Associated Signaling

233

Emi, T., Kinoshita, T., and Shimazaki, K. (2001). Specific binding of a vf14-3-3a isoform to the plasma membrane Hþ-ATPase in response to blue light and fusicoccin in guard cells of broad bean. Plant Physiol. 125, 1115–1125. Esau, K. (1977). ‘‘Anatomy of Seed Plants,’’ pp. 91–96. John Wiley & Sons, New York. Fankhauser, C. (2001). The phytochromes, a family of red-far-red absorbing photoreceptors. J. Biol. Chem. 276, 11453–11456. Folta, K. M., Lieg, E. J., Durham, T., and Spalding, E. P. (2003). Primary inhibition of hypocotyl growth and phototropism depend diVerently on phototropin-mediated increases in cytoplasmic calcium induced by blue light. Plant Physiol. 113, 1464–1470. Friml, J., Wisniewska, J., Benkova, E., Mendgen, K., and Palme, K. (2002). Lateral relocation of auxin eZux regulator PIN3 mediates tropism in Arabidopsis. Nature 415, 806–809. Galen, C., Huddle, J., and Liscum, E. (2004). An experimental test of the adaptive evolution of phototropins: Blue-light receptors controlling phototropism in Arabidopsis thaliana. Evolution 58, 515–523. Gil, P., Kircher, S., Adam, E., Bury, E., Kozma-Bognar, L., Schafer, E., and Nagy, F. (2000). Photocontrol of subcellular partitioning of phytochrome-B: GFP fusion protein in tobacco seedlings. Plant J. 22, 135–145. Giovani, B., Byrdin, M., Ahmad, M., and Brettel, K. (2003). Light-induced electron transfer in a cryptochrome blue-light photoreceptor. Nature Struc. Biol. 10, 489–490. Gray, W. M., Kepinski, S., Rouse, D., Leyser, O., and Estelle, M. (2001). Auxin regulates SCFTIR1-dependent degradation of AUX/IAA proteins. Nature 414, 271–276. Guo, H., Doung, H., Ma, N., and Lin, C. (1999). The Arabidopsis blue light receptor cryptochrome 2 is a nuclear protein regulated by a blue light-dependent post transcriptional mechanism. Plant J. 19, 279–287. Gyula, P., Schafer, E., and Nagy, F. (2003). Light perception and signalling in higher plants. Curr. Opin. Plant Biol. 6, 446–452. Hagen, G., and Guilfoyle, T. (2002). Auxin-responsive gene expression: Genes, promoters and regulatory factors. Plant Mol. Biol. 49, 373–385. Hager, A., Menzel, H., and Krauss, A. (1971). Versuche und hypothese zur primarwirkung des auxins beim shtrekungswachstum. Planta 100, 47–75. Hanks, S. K., and Hunter, T. (1995). Protein Kinases 6: The eukaryotic protein kinase superfamily: Kinase (catalytic) domain structure and classification. FASEB J. 9, 576–596. Harper, S. M., Neil, L. C., Day, I. J., Hore, P. J., and Gardner, K. H. (2004). Conformational changes in a photosensory LOV domain monitored by time-resolved NMR spectroscopy. J. Am. Chem. Soc. 126, 3390–3391. Harper, S. M., Neil, L. C., and Gardner, K. H. (2003). Structural basis of a phototropin light switch. Science 301, 1541–1544. Harper, R. M., Stowe-Evans, E. L., Luesse, D. R., Muto, H., Tatematsu, K., Watahiki, M. K., Yamamoto, K., and Liscum, E. (2000). The NPH4 locus encodes the auxin response factor ARF7, a conditional regulator of diVerential growth in aerial Arabidopsis tissue. Plant Cell 12, 757–770. Huala, E., Oeller, P. W., Liscum, E., Han, I.-S., Larsen, E., and Briggs, W. (1997). Arabidposis NPH1: A protein kinase with a putative redox-sensing domain. Science 278, 2021–2023. Iino, M. (1990). Phototropism: Mechanisms and ecological implications. Plant Cell Environ. 13, 633–650. Inada, S., Ohgishi, M., Mayama, T., Okada, K., and Sakai, T. (2004). RPT2 is a signal transducer involved in phototropic response and stomatal opening by association with phototropin 1 in Arabidopsis thaliana. Plant Cell 16, 887–896.

234

Stone et al.

Janoudi, A.-K., Konjevic, R., Whitelam, G., Gordon, W., and PoV, K. L. (1997). Both phytochrome A and phytochrome B are required for the normal expression of phototropism in Arabidopsis thaliana seedlings. Physiol. Plant 101, 278–282. Jarillo, J. A., Ahmad, M., and Cashmore, A. R. (1998). NPL1 (Accession No AF053941): A second member of the NPH1 serine/threonine kinase family of Arabidopsis. Plant Physiol. 117, 719. Jarillo, J. A., Gabrys, H., Capel, J., Alonso, J. M., Ecker, J. R., and Cashmore, A. R. (2001). Phototropin-related NPL1 controls chlorplast relocation induced by blue light. Nature 410, 952–954. Kadota, A., Sato, Y., and Wada, M. (2000). Intracellular chloroplast photorelocation in the moss Physcomitrella patens is mediated by phytochrome as well as by a blue-light receptor. Planta 210, 932–937. Kagawa, T., and Wada, M. (1999). Chloroplast avoidance response induced by high-fluence blue light in prothallial cell of the fern Adiantum capillus-veneris as analyzed by microbeam irradiation. Plant Physiol. 119, 917–923. Kagawa, T., and Wada, M. (2000). Blue light-induced chloroplast relocation in Arabidopsis thaliana as analyzed by microbeam irradiation. Plant Cell Physiol. 41, 84–93. Kagawa, T., Sakai, T., Suetsugu, N., Oikawa, K., Ishiguro, S., Kato, T., Tabata, S., Okada, K., and Wada, M. (2001). Arabidopsis NPL1: A phototropin homolog controlling the chloroplast high-light avoidance response. Science 291, 2138–2141. Kandasamy, M. K., and Meagher, R. B. (1999). Actin-organelle interaction: Association with chloroplast in Arabidopsis leaf mesophyll cells. Cell Motil. Cytoskeleton 44, 110–118. Kanegae, H., Tahir, M., Savazzini, F., Yamamoto, K., Yano, M., Sasaki, T., Kanegae, T., Wada, M., and Takano, M. (2000). Rice homologues OsNPH1a and OsNPH1b are diVerentially photoregulated. Plant Cell Physiol. 41, 415–523. Kasahara, M., Swartz, T. E., Olney, M. A., Onodera, A., Mochizuki, N., Fukuzawa, H., Asamizu, E., Tabata, S., Kanegae, H., Takano, M., Christie, J. M., Nagatani, A., and Briggs, W. R. (2002). Photochemical properties of the flavin mononucleotide-binding domains of the phototropins from Arabidopsis, rice and Chlamydomonas reinhardtii. Plant Physiol. 129, 762–773. Kennis, J. T. M., Crosson, S., Gauden, M., van Stokkum, I. H. M., MoVat, K., and van Grondelle, R. (2003). Primary reactions of the LOV2 domain of phototropin, a plant bluelight photoreceptor. Biochemistry. 42, 3385–3392. Kepinski, S., and Leyser, O. (2004). Auxin-induced SCFTIR1-AUX/IAA interaction involves stable modification of the SCFTIR1 complex. Proc. Natl. Acad. Sci. USA 101, 12381–12386. Kim, J., Harter, K., and Theologis, T. (1997). Protein-protein interaction among the AUX/IAA proteins. Proc. Natl. Acad. Sci. USA 94, 11786–11791. Kinoshita, T., Doi, M., Suetsugu, N., Kagawa, T., Wada, M., and Shimazaki, K.-I. (2001). PHOT1 and PHOT2 mediate blue light regulation of stomatal opening. Nature 414, 656–660. Kinoshita, T., and Shimazaki, T.-I. (1999). Blue light activates the plasma membrane Hþ-ATPase by phosphorylation of the C-terminus in stomatal guard cells. EMBO J. 18, 5548–5558. Kinoshita, T., Emi, T., Tominaga, M., Sakamoto, K., Shigenaga, A., Doi, M., and Shimazaki, K.-I. (2003). Blue-light- and phosphorylation-dependent binding of a 14-3-3 protein to phototropins in stomatal guard cells of broad bean. Plant Physiol. 133, 1453–1463. Kircher, S., Kozma-Bognar, L., Kim, L., Adam, E., Harter, K., Schafer, E., and Nagy, F. (1999). Light quality-dependent nuclear import of the plant photoreceptors phytochrome A and B. Plant Cell 11, 1445–1456. Kleine, T., Lockhart, P., and Batschauer, A. (2003). An Arabidopsis protein closely related to Synechocystis cryptochrome is targeted to organelles. Plant J. 35, 93–103.

7. Plant Photoreceptors and Associated Signaling

235

Kleiner, O., Kircher, S., Harter, K., and Batschauer, A. (1999). Nuclear localization of the Arabidopsis blue light receptor cryptochrome 2. Plant J. 19, 289–296. Lasce`ve, G., Leymaire, J., Olney, M. A., Liscum, E., Christie, J. M., Vavasseur, A., and Briggs, W. R. (1999). Arabidopsis contains at least four independent blue light-activated signal transduction pathways. Plant Physiol. 120, 605–614. Lee, Y., Choi, D., and Kende, H. (2001). Expansins: Ever-expanding numbers and functions. Curr. Opin. Plant Biol. 4, 527–532. Li, H., Johnson, P., Stepanova, A., Alonso, J. M., and Ecker, J. R. (2004). Convergence of signaling pathways in the control of diVerential cell growth in Arabidopsis. Dev. Cell 7, 193–204. Lin, C., and Shalitin, D. (2003). Cryptochrome structure and signal transduction. Annu. Rev. Plant Biol. 54, 469–496. Lin, C. T., Yang, H. Y., Guo, H. W., Mockler, T., Chen, J., and Cashmore, A. R. (1998). Enhancement of blue-light sensitivity of Arabidopsis seedlings by a blue light receptor cryptochrome 2. Proc. Natl. Acad. Sci. USA 95, 2686–2690. Liscum, E. (2002). Phototropism: Mechanisms and outcomes. In ‘‘The Arabidopsis Book’’ (C. R. Somerville and E. M. Meyerowitz, Eds.). American Society of Plant Biologists, Rockville, MD, doi/10.1199/tab.0042, http://www.aspb.org/publications/arabidopsis/. Liscum, E., and Briggs, W. R. (1995). Mutations in the NPH1 locus of Arabidopsis disrupt the perception of phototropic stimuli. Plant Cell 7, 473–485. Liscum, E., and Briggs, W. R. (1996). Mutations in Arabidopsis in potential transduction and response components of the phototropic signaling pathway. Plant Physiol. 112, 291–296. Liscum, E., Hodgson, D. W., and Campbell, T. J. (2003). Blue light signaling through the cryptochromes and phototropins. So that’s what the blues is all about. Plant Physiol. 133, 1429–1436. Liscum, E., and Reed, J. (2002). Genetics of AUX/IAA and ARF action in plant growth and development. Plant Mol. Biol. 49, 387–400. Luschnig, C., Gaxiola, R. A., Grisafi, P., and Fink, G. R. (1998). EIR1, a root-specific protein involved in auxin transport, is required for gravitropism in Arabidopsis thaliana. Genes Dev. 12, 2175–2187. Ma, L. G., Li, J. M., Qu, L. J., Hager, J., Chen, Z. L., Zhao, H. Y., and Deng, X. W. (2001). Light control of Arabidopsis development entails coordinated regulation of genome expression and cellular pathways. Plant Cell 13, 2589–2607. Marchant, A., Kargul, J., May, S. T., Muller, P., Delbarre, A., Perrot-Rechenmann, C., and Bennett, M. J. (1999). AUX1 regulates root gravitropism in Arabidopsis by facilitating auxin uptake within root apical tissues. EMBO J. 18, 2066–2073. Martinez-Garcia, J. F., Huq, E., and Quail, P. H. (2000). Direct targeting of light signals to a promoter element-bound transcription factor. Science 288, 859–863. Ma¨ ser, P., Leonhardt, N., and Schroeder, J. I. (2003). The clickable guard cell: Electronically linked model of guard cell signal transduction pathways. In ‘‘The Arabidopsis Book’’ (C. R. Somerville and E. M. Meyerowitz, Eds.). American Society of Plant Biologists, Rockville, MD, doi/10.1199/tab.0099, http://www.aspb.org/publications/arabidopsis/. Motchoulski, A., and Liscum, E. (1999). Arabidopsis NPH3: A NPH1 photoreceptor-interacting protein essential for phototropism. Science 286, 961–964. Mu¨ ller, A., Guan, C., Galweiler, L., Tanzler, P., Huijser, P., Marchant, A., Parry, G., Bennett, M., Wisman, E., and Palme, K. (1998). AtPIN2 defines a locus of Arabidopsis for root gravitropism control. EMBO J. 17, 6903–6911. Nagy, F., Kircher, S., and Schafer, E. (2000). Nucleo-cytoplasmic partitioning of the plant photoreceptors phytochormes. Semin. Cell Dev. Biol. 11, 505–510. Nagy, F., Kircher, S., and Schafer, E. (2001). Intracellular traYcking of photoreceptors during light-induced signal transduction in plants. J. Cell Sci. 114, 475–480.

236

Stone et al.

Oikawa, K., Kasahara, M., Kiyosue, T., Kagawa, T., Suetsugu, N., Takahashi, F., Kanegae, T., Niwa, Y., Kadota, A., and Wada, M. (2003). CHLOROPLAST UNUSUAL POSITIONING1 is essential for proper chloroplast positioning. Plant Cell 15, 2815–2825. Okada, K., and Shimura, Y. (1992). Mutational analysis of root gravitropism and phototropism of Arabidopsis thaliana seedlings. Aust. J. Plant Physiol. 19, 439–448. Okada, K., and Shimura, Y. (1994). Genetic analyses of signaling in flower development using Arabidopsis. Plant Mol. Biol. 26, 1357–1377. Parker, M. W., Hendricks, S. B., Borthwick, H. A., and Jenner, C. E. (1952). Photoperiodic responses of plants and animals. Nature 169, 242–243. Parks, B. M. (2003). The red side of photomorphogenesis. Plant Physiol. 133, 1437–1444. Parks, B. M., Quail, P. H., and Hangarter, R. P. (1996). Phytochrome A regulates redlight induction of phototropic enhancement in Arabidopsis. Plant Physiol. 110, 155–162. Quail, P. H. (2000). Phytochrome-interacting factors. Semin. Cell Dev. Biol. 11, 457–466. Quail, P. H. (2002). Photosensory perception and signalling in plant cells: New paradigms? Curr. Op. Cell Biol. 14, 180–188. Ramos, J. A., Zenser, N., Leyser, O., and Callis, J. (2001). Rapid degradation of AUX/IAA proteins requires conserved amino acids of domain II and is proteosome-dependent. Plant Cell 15, 2349–2360. Rashotte, A. M., DeLong, A., and Muday, G. K. (2001). Genetic and chemical reductions in protein phosphatase activity alter auxin transport, gravity response, and lateral root growth. Plant Cell 13, 1683–1697. Ru¨ diger, W., and Thu¨ mmler, F. (1994). The phytochrome chromophore. In ‘‘Photomorphogenesis in Plants’’ (R. E. Kendrick and G. H. M. Kronenberg, Eds.), 2nd ed., pp. 51–69. Kluwer Academic Publishers, The Netherlands. Sage, L. C. (1992). ‘‘Pigment of the Imagination. A History of Phytochrome Research.’’ Academic Press, San Diego. Sakai, T., Wada, T., Ishiguro, S., and Okada, K. (2000). RPT2: A signal transducer of the phototropic response in Arabidopsis. Plant Cell 12, 225–236. Sakai, T., Kagawa, T., Kasahara, M., Swartz, T. E., Christie, J. M., Briggs, W. R., Wada, M., and Okada, K. (2001). Arabidopsis nph1 and np11: Blue light receptors that mediate both phototropism and chloroplast relocation. Proc. Natl. Acad. Sci. USA 98, 6969–6974. Sakamoto, K., and Briggs, W. R. (2002). Cellular and subcellular localization of phototropin 1. Plant Cell 14, 1723–1735. Salomon, M., Knieb, E., von Zeppelin, T., and Ru¨ diger, W. (2003). Mapping of low- and highfluence autophosphorylation sites in Phototropin 1. Biochemistry 42, 4217–4225. Sancar, A. (2003). Structure and function of DNA photolyase and cryptochrome blue-light photoreceptors. Chemical Reviews 103, 2203–2237. Sato, Y., Kadota, A., and Wada, M. (1999). Mechanically induced avoidance response of chloroplasts in fern protonemal cells. Plant Physiol. 121, 37–44. Sato, Y., Wada, M., and Kadota, A. (2001a). External Ca2þ is essential for chloroplast movement induced by mechanical stimulation but not by light stimulation. Plant Physiol. 127, 497–504. Sato, Y., Wada, M., and Kadota, A. (2001b). Choice of tracks, microtubules and/or actin filaments for chloroplast photo-movement is diVerentially controlled by phytochrome and a blue light receptor. J. Cell Sci. 120, 269–279. Schroeder, J. I., Ward, J. M., and Gassmann, W. (1994). Perspectives on the physiology and structure of inward-rectifying Kþ channels in higher plants: Biophysical implications for Kþ uptake. Annu. Rev. Biophys. Biomol. Struct. 23, 441–471. Schroeder, J. I., Allen, G. J., Hugouvieux, V., Kwak, J. M., and Waner, D. (2001). Guard cell signal transduction. Annu. Rev. Plant Physiol. Plant Mol. Biol. 52, 627–658.

7. Plant Photoreceptors and Associated Signaling

237

Shalitin, D., Yang, H. Y., Mockler, T. C., Maymon, M., Guo, H. W., Whitelam, G. C., and Lin, C. T. (2002). Regulation of Arabidopsis cryptochrome 2 by blue-light-dependent phosphorylation. Nature 417, 763–767. Shimizu-Sato, S., Huq, E., Tepperman, J. M., and Quail, P. H. (2002). A light-switchable gene promoter system. Nature Biotech. 20, 1041–1044. Stoelzle, S., Kagawa, T., Wada, M., Hedrich, R., and Dietrich, P. (2003). Blue light activates calcium-permeable channels in Arabidopsis mesophyll cells via the phototropin signaling pathway. Proc. Natl. Acad. Sci. USA 100, 1456–1461. Stowe-Evans, E. L., Harper, R. M., Motchoulski, A. V., and Liscum, E. (1998). NPH4, a conditional modulator of auxin-dependent diVerential growth responses in Arabidopsis. Plant Physiol. 118, 1265–1275. Tatematsu, K., Kumagai, S., Muto, H., Sato, A., Watahiki, M. K., Harper, R. M., Liscum, E., and Yamamoto, K. T. (2004). MASSUGU2 encodes Aux/IAA19, an auxin-regulated protein that functions together with the transcriptional activator NPH4/ARF7 to regulate diVerential growth responses of hypocotyl and formation of lateral roots in Arabidopsis thaliana. Plant Cell 16, 379–393. Taylor, B. L., and Zhulin, I. B. (1999). PAS domains: Internal sensors of oxygen, redox potential, and light. Microbiol. Mol. Biol. Rev. 63, 479–506. Tepperman, J. M., Zhu, T., Chang, H. S., Wang, X., and Quail, P. H. (2001). Multiple transcription-factor genes are early targets of phytochrome A signaling. Proc. Natl. Acad. Sci. USA 98, 9437–9442. Tepperman, J. M., Hudson, M. E., Khanna, R., Zhu, T., Chang, S. H., Wang, X., and Quail, P. H. (2004). Expression profiling of phyB mutant demonstrates substantial contribution of other phytochromes to red-light-regulated gene expression during seed de-etiolation. Plant J. 38, 725–739. Tiwari, S. B., Hagen, G., and Guilfoyle, T. (2003). The roles of auxin response factor domains in auxin-responsive transcription. Plant Cell 15, 533–543. Trewavas, T., Briggs, W. R., Bruinsma, J., Evan, M. L., Firn, R., Hertel, R., Iino, M., Jones, A. M., Leopold, A. C., Pilet, P. E., PoV, K. L., Roux, S. J., Salibury, F. B., Scott, T. K., Sievers, A., Zieschang, H. E., and Wayne, R. (1992). Forum: What remains of the CholodnyWent theory? Plant Cell Environ. 15, 759–794. Ulmasov, T., Hagen, G., and Guilfoyle, T. (1997). ARF1, a transcription factor that binds to auxin responsive elements. Science 276, 1865–1868. Ulmasov, T., Hagen, G., and Guilfoyle, T. (1999). Dimerization and DNA binding of auxin response factors. Plant J. 19, 1–11. Wada, M., Kagawa, T., and Sato, Y. (2003). Chloroplast movement. Annu. Rev. Plant Biol. 54, 455–468. Wang, H., and Deng, X. W. (2002). Phytochrome signaling mechanism. In ‘‘The Arabidopsis Book’’ (C. R. Somerville and E. M. Meyerowitz, Eds.). American Society of Plant Biologists, Rockville, MD, doi/10.1199/tab.0074.1, http://www.aspb.org/publications/arabidopsis/. Wang, H. Y., Ma, L. G., Habashi, J., Li, J. M., Zhao, H. Y., and Deng, X. W. (2002). Analysis of far-red light-regulated genome expression profiles of phytochrome A pathway mutants in Arabidopsis. Plant J. 32, 723–733. Wang, H., Ma, L.-G., Li, J.-M., Zhao, H.-Y., and Deng, X.-W. (2001). Direct interaction of Arabidopsis cryptochromes with COP1 in light control development. Science 294, 154–158. Watahiki, M. K., and Yamamoto, K. T. (1997). The massugul mutation of Arabidposis identified with failure of auxin-induced growth curvature of hypocotyl confers auxin insensitivity to hypocotyl and leaf. Plant Physiol. 115, 419–426. Went, F. W., and Thimann, K. V. (1937). ‘‘Phytohormones.’’ Macmillan, New York. Whippo, C. W., and Hangarter, R. P. (2003). Second positive phototropism results from coordinated co-action of the phototropins and cryptochromes. Plant Physiol. 132, 1–9.

238

Stone et al.

Yang, H. Q., Tang, R. H., and Cashmore, A. R. (2001). The signaling mechanism of Arabidopsis CRY1 involves direct interaction with COP1. Plant Cell 13, 2573–2587. Yang, H. Q., Wu, Y. J., Tang, R. H., Liu, D., Liu, Y., and Cashmore, A. R. (2000). The C termini of Arabidopsis cryptochromes mediate a constitutive light response. Cell. 103, 815–827. Yeh, K. C., and Legarias, J. C. (1998). Eukaryotic phytochromes: Light-regulated serinethreonine protein kinases with histidine kinase ancestry. Proc. Natl. Acad. Sci. USA 95, 13976–13981. Zeiger, E. (1983). The biology of stomatal guard cells. Ann. Rev. Plant Physiol. 34, 441–475. Zenser, N., Dreher, K. A., Edwards, S. R., and Callis, J. (2003). Acceleration of AUX/IAA proteolysis is specific for auxin and independent of AXR1. Plant J. 35, 285–294. Zenser, N., Ellsmore, A., Leasure, C., and Callis, J. (2001). Auxin modulates the degredation of AUX/IAA proteins. Proc. Natl. Acad. Sci. USA 98, 11795–11800. Zhulin, L. B., and Taylor, B. (1997). PAS domain S-boxes in Archea, bacteria and sensors for oxygen and redox. Trends Biochem. Sci. 22, 331–333.

8

____________________________________________________________________________

Evolving Concepts in Bone Tissue Engineering Catherine M. Cowan,* Chia Soo,{,{ Kang Ting,{ and Benjamin Wu* *Department of Bioengineering University of California Los Angeles Los Angeles, California 90095 { Weintraub Center for Reconstructive Biotechnology University of California Los Angeles Los Angeles, California 90095 { University of Southern California Keck School of Medicine Division of Plastic Surgery Los Angeles, California 90053

I. Introduction II. Bone Morphogenetic Proteins A. The BMP Family B. Osteogenic BMPs C. BMP-2-Induced Osteogenesis D. Increased Potency of BMP Heterodimers III. Bone Tissue Engineering A. Bone Tissue Engineering Approaches B. Lessons Learned from Nature C. Microenvironment D. Implantable ScaVolds E. Osteoprogenitor Cells F. Bone Tissue Engineering Using BMP-2, BMP-4, or BMP-7 G. BMP–ScaVold Interactions H. Translating BMP-2 and BMP-7 Research to Clinical Trials IV. Conclusions and Future Directions References

The field of tissue engineering integrates the latest advances in molecular biology, biochemistry, engineering, material science, and medical transplantation. Researchers in the developing field of regenerative medicine have identified bone tissue engineering as an attractive translational target. Clinical problems requiring bone regeneration are diverse, and no single regeneration approach will likely resolve all defects. Recent advances in the field of tissue engineering have included the use of sophisticated biocompatible scaVolds, new postnatal multipotent cell populations, and the appropriate cellular stimulation. In particular, synthetic polymer scaVolds allow for fast and reproducible construction, while still retaining Current Topics in Developmental Biology, Vol. 66 Copyright 2005, Elsevier Inc. All rights reserved. 0070-2153/05 $35.00

239

240

Cowan et al.

biocompatible characteristics. These criteria relate to the immediate goal of determining the ideal implant. The search is becoming a reality with widespread availability of biocompatible scaVolds; however, the desired parameters have not been clearly defined. Currently, most research focuses on the use of bone morphogenetic proteins (BMPs), specifically BMP-2 and BMP-7. These proteins induce osteogenic diVerentiation in vitro, as well as bone defect healing in vivo. Protein–scaVold interactions that enhance BMP binding are of the utmost importance, since prolonged BMP release creates the most osteogenic microenvironment. Transition into clinical studies has had only mild success and relies on large doses of BMPs for bone formation. Advances within the field of bone tissue engineering will likely overcome these challenges and lead to more clinically relevant therapies. C 2005, Elsevier Inc.

I. Introduction The field of tissue engineering has been evolving since the early 1960s. Today, it integrates the latest advances in molecular biology, biochemistry, engineering, material science, and medical transplantation (Langer and Vacanti, 1993). Researchers in the developing field of regenerative medicine have identified bone tissue engineering as an attractive translational target. Clinical problems requiring bone regeneration are diverse, and no single regeneration approach will likely resolve all injuries. Causes for deficient bone are attributed to nonunion fractures, craniofacial reconstruction, trauma, surgical resection, dental reconstruction, degenerative disorders, and augmentation of bone around hip implant and replacement. In particular, craniofacial, long bone, and spinal procedures constitute the three major challenges in skeletal reconstruction and represent a substantial biomedical burden. Recent advances in the field of tissue engineering have included the use of sophisticated biocompatible scaVolds, new postnatal multipotent cell populations, and appropriate cellular stimulation for creation of an appropriate osteogenic microenvironment. In particular, synthetic polymer scaVolds allow for fast and reproducible construction, while still retaining biocompatible characteristics. Additionally, the identification of multipotent mesenchymal precursor cells within adipose tissue has stimulated a new surge of research. Finally, many growth factors have been tested for bone regeneration, but the bone morphogenetic proteins (BMPs) demonstrate the most promise. Although the fundamentals of an ideal bone tissue engineering implant have made much progress in animal models, additional clinically relevant questions must be answered before clinical bone regeneration will attain widespread success. For example, why are recombinant human BMPs less active than corresponding naturally occurring BMPs? Also, what is the

8. Evolving Concepts in Bone Tissue Engineering

241

optimal BMP dose and treatment duration, and how do we engineer scaVolds with these release kinetics? The success of bone tissue engineering relies on these answers and more. Eventually tissue engineering applications will be administered to all tissue types; however, skeletal defects promise the most success in the immediate future. Many reviews have comprehensively outlined bone healing, BMP signaling, implantable scaVolds, and various growth factors (Table I) (Ambrose et al., 2003; Baltzer and Lieberman, 2004; Bauer and Muschler, 2000; Bonadio, 2000; Bonadio et al., 1999; Derubeis and Cancedda, 2004; Einhorn, 2003; Fleming et al., 2000; Gamradt and Lieberman, 2004; Geiger et al., 2003; Goldstein, 2002; Hollinger et al., 2000; Kawabata et al., 1998; Langer and Vacanti, 1993; Orban et al., 2002; Parikh, 2002; Prockop et al., 2003; Ramoshebi et al., 2002; Rengachary, 2002; Rose and OreVo, 2002; Rosso et al., 2004; ten Dijke et al., 2003; Warren et al., 2003; Winn et al., 2000; Zhu and Emerson, 2004). This chapter further investigates the roles of BMP-2, BMP-4, and BMP-7 in bone formation, BMP–scaVold interactions, and clinical relevance.

II. Bone Morphogenetic Proteins A. The BMP Family The bone morphogenetic protein (BMP) family has 15 members (BMP-1 through BMP-15) and belongs to the transforming growth factor-beta (TGF-
) superfamily. They were originally discovered by Urist in extracts of bovine bone that induced ectopic bone formation subcutaneously in rats (Urist, 1965). Although first identified because of their ability to induce ectopic bone formation in vivo (Hogan, 1996; Urist, 1965; Wozney and Rosen, 1998; Wozney et al., 1988), BMPs have quickly become known as multifunctional regulators of morphogenesis during embryonic development (Hogan, 1996). In terms of bone formation, BMPs regulate intramembranous as well as endochondral ossification through chemotaxis and mitosis of mesenchymal cells, induction of mesenchymal commitment to osteoblasts or chondrocytes, promotion of further osteoblast or chondrocyte diVerentiation, and programmed cell death (Reddi, 2001). BMPs transduce signals upon binding to a heteromeric complex of type I and type II serine/threonine kinase receptors on the cell surface. The type II receptors are constitutively active kinases that phosphorylate the type I receptors upon ligand binding (Shi and Massague, 2003). Of the type I receptors, BMPR1A is ubiquitously expressed in embryonic mesenchyme, whereas BMPR1B is more specifically expressed in cartilage condensations. Interestingly, mice deficient in BMPRIA result in early embryonic lethality (Mishina et al., 1995), whereas mice deficient in the BMPR1B gene exhibit failed mesenchymal condensation

Table I

Bone Morphogenetic Protein and Bone Tissue Engineering Reviews*

Review Articles (First Author & Year) Langer and Vacanti (1993) Kawabata et al. (1998) Bonadio et al. (1999) Bauer and Muschler (2000) Bonadio (2000) Fleming et al. (2000) Hollinger et al. (2000) Winn et al. (2000) Goldstein (2002) Orban et al. (2002) Parikh (2002) Ramoshebi et al. (2002) Rengachary (2002) Rose and OreVo (2002) Ambrose et al. (2003) ten Dijke et al. (2003) Einhorn (2003) Geiger et al. (2003) Prockop et al. (2003) Warren et al. (2003) Baltzer and Lieberman (2004) Derubeis and Cancedda (2004) Gamradt and Lieberman (2004) Rosso et al. (2004) Zhu and Emerson (2004)

Bone Healing

Tissue Engineering

ScaVolds/ Delivery

Osteoinductive Signals










Growth Factors

BMP Receptors and Signaling

Cell Delivery

Gene Therapy

Recombinant Protein Therapy

Clinical Applications



































































































































*Review papers describing current research in the fields of bone morphogenetic proteins and bone tissue engineering.














8. Evolving Concepts in Bone Tissue Engineering

243

within forming digits (Baur et al., 2000; Storm and Kingsley, 1999; Yi et al., 2000). Upon binding, an intracellular signal is transmitted through activation of SMAD (signaling, mothers against decapentaplegic) proteins, which translocate to the nucleus and alter gene expression. In addition to the SMAD signaling pathway, BMPs have been shown to signal through mitogen-activated protein kinase (MAPK) pathways. In the MAPK pathway, BMP-2 upregulates ERK and p38 cascades within diVerent cell types (Gallea et al., 2001; Noth et al., 2003). The existence of so many BMP members and tissues expressing these growth factors indicates the complexity and redundancy of the BMP signaling pathways. Thus, although almost all of the BMPs signal through the SMAD pathway, it is still unclear how osteogenic BMPs diVer from nonosteogenic BMPs in terms of SMAD signaling or direct downstream targets (Peng et al., 2003).

B. Osteogenic BMPs BMP-2, BMP-4, BMP-6, BMP-7, and BMP-9 have been deemed the most potent osteogenic BMPs (Hogan, 1996). They stimulate osteogenic diVerentiation in a variety of cell types, including fibroblasts, chondrocytes, osteoprogenitor cells, calvarial cells, periosteal cells, bone marrow stromal (BMS) cells, muscle-derived stem cells (MDSCs), and adipose-derived adult stromal (ADAS) cells (Boden et al., 1996; Cheng et al., 2003; Dragoo et al., 2003; Goldstein, 2001; Harland, 1994; Hughes et al., 1995; Krebsbach et al., 1998; Lee et al., 2001, 2002; Mayer et al., 1996a; Peng et al., 2002; Winn et al., 1998; Yamaguchi et al., 1996; Zegzula et al., 1997). Mechanistically, osteogenic BMPs may regulate osteoblast diVerentiation through increased transcription of core-binding factor 1/Runt-related family 2 (Cbfa1/Runx2)—a molecule deemed essential, but not necessarily suYcient for osteoblast commitment and diVerentiation—as well as other modulators in proliferation and diVerentiation such as Id helix-loop-helix and distal-less homeobox proteins (Enomoto et al., 2000; Gu et al., 2004; Maeda et al., 2004b; Peng et al., 2003; Tou et al., 2003). It is clear that osteogenic BMPs can also be chondrogenic. BMP-2, BMP-4, and BMP-7 have all demonstrated an ability to induce both chondrogenic and osteogenic diVerentiation of various cell types in vitro (Abu-Serriah et al., 2003; Lengner et al., 2004; Maeda et al., 2004b; Sampath et al., 1992; Shea et al., 2003). Developmentally, BMP-2 diVerentiates mesenchymal cells in the apical ridge at the tip of the limb bud (Lyons et al., 1990; Niswander and Martin, 1993), as well as chondrocyte diVerentiation and morphogenesis (Lyons et al., 1989), suggesting an endogenous method of bone tissue engineering. Additionally, the chemotactic eVects of BMP-2 are demonstrated by its role in the migration of cranial neural crest (CNC)

244

Cowan et al.

cells in craniofacial morphogenesis. Although BMP-2 knockout mice result in early embryonic death (Zhang and Bradley, 1996), heterozygous knockouts survive, but lack brachial arches and detectable migratory CNC cells (Kanzler et al., 2000). In vitro, early BMP-7 treatment (days 1–7) of newborn rat calvarial cells elicits the greatest chondrogenic response (i.e., increased Alcian blue staining and collagen type II expression), whereas late BMP-7 treatment elicits the greatest osteoblastic response (Asahina et al., 1993). Interestingly, mesenchymal precursor cells require 7–8 days of BMP-7 stimulation to maintain osteogenic diVerentiation (Shea et al., 2003), whereas 14–20 days of continuous BMP-2 or BMP-4 stimulation are required (Puleo, 1997; Yamaguchi et al., 1996), suggesting a certain sequential regulation among BMPs (Puleo, 1997; Shea et al., 2003). A possible osteogenic hierarchy among the BMPs with BMP2, BMP-6, and BMP-9 influencing earlier and BMP-2, BMP-4, BMP-7, and BMP-9 influencing later osteogenic events has also been proposed (Cheng et al., 2003). The observations underscore the hypothesis that although certain BMPs are ‘‘osteogenic,’’ optimal BMP-mediated osteoinduction (as opposed to chondroinduction) depends on specific BMP parameters such as type, dose, administration sequence, and duration and nonspecific BMP parameters such as cell type, cellular diVerentiation state, and microenvironment.

C. BMP-2-Induced Osteogenesis BMP-2, in particular, has been extensively studied for its ability to induce many of the events necessary for both intramembranous and endochondral ossification, often times more potently than recombinant human BMP (rhBMP)-4 or rhBMP-7 (Kang et al., 2004). In vitro, rhBMP-2 (25–400 ng/ ml) upregulates genes associated with osteogenic diVerentiation and downregulates genes associated with myogenic diVerentiation (Peng et al., 2003) as early as the first 24 h following stimulation (de Jong et al., 2004). Genes upregulated threefold to sixfold include alkaline phosphatase, collagen I, osteopontin, osteocalcin, junB, transcription factor GIF, latent TGF
binding protein 2 (TGF
BP2), plasminogen activator inhibitor (PAI-1), and Cbfa1/Runx2 (Hughes et al., 1995; Mundy et al., 1995; Peng et al., 2003; Rickard et al., 1994; Thies et al., 1992; Wang et al., 1993). BMS cells display maximal alkaline phosphatase expression when exposed to BMP-2 for at least 14 days, as opposed to only 7 days (Puleo, 1997). BMP-2 also induces bone nodule formation and calcium deposition in BMS cell cultures (Hanada et al., 1997). Interestingly, ADAS cells under BMP-2 stimulation produce more bone precursor cells than either osteoblasts or BMS cell cultures under the same conditions (De Ugarte et al., 2003) and accelerate ADAS

8. Evolving Concepts in Bone Tissue Engineering

245

cell-mediated bone formation in vivo (Cowan et al., 2005). Additionally, BMP-2 stimulates prechondroblasts to downregulate the expression of chondrogenic genes and upregulate the expression of osteogenic genes (Rosen et al., 1994). BMP-2 promotes cartilage formation through Sonic hedgehog (Shh) or Indian hedgehog (Ihh) signaling (Enomoto-Iwamoto et al., 2000). It stimulates both stages of chondrogenic cell diVerentiation, proliferating chondrocytes expressing collagen II and prehypertrophic chondrocytes expressing collagen X (Shukunami et al., 1998). Interestingly, BMP-2 transcription is regulated by retinoids, Cbfa1/Runx2 (positive feedback from osteoblasts), Sox-9 (positive feedback chondrocytes), and others (Helvering et al., 2000). In vivo, the microenvironment is critically important in determining either intramembranous or endochondral ossification. Although endochondral bone formation is usually accompanied by some intramembranous bone formation, the inverse is not true. In a femur ectopic bone formation assay, BMP-2 implantation induces both mesenchymal cell proliferation and chondroblast diVerentiation (Wang et al., 1990). The chondroblasts secrete a cartilaginous matrix within which they become progressively embedded. Further diVerentiation results in hypertrophic chondrocytes that calcify the cartilaginous matrix. As vascularization occurs in the calcified hypertrophic cartilage, mesenchymal cells invade and diVerentiate into osteoblasts. Additionally, within the femur model as well as craniofacial models, BMP-2 stimulates mesenchymal cells to directly diVerentiate into osteoblasts, which lay down the bone extracellular matrix (ECM) (Kenley et al., 1994; Marden et al., 1994). BMP-2 has regenerated critical size defects in rat calvaria (Kenley et al., 1994; Marden et al., 1994) and long bone (Stevenson et al., 1994; Yasko et al., 1992); rabbit ulna (Cook et al., 1994) and radius (Winn et al., 1998; Zegzula et al., 1997); sheep long bone (Gerhart et al., 1993), Mandible (Toriumi et al., 1991), and premaxilla (Mayer et al., 1996b); and African green monkey ulna (Cook et al., 1995). BMP-2 has also successfully induced spinal fusion in rats (Wang et al., 2003), rabbits (Liao et al., 2003), dogs (Muschler et al., 1994), and rhesus nonhuman primates (Boden et al., 1995; Schimandle et al., 1995). A dose-response eVect for BMP-2 in large-animal models may provide useful translational information for humans. In a dog spine model, rhBMP-2coated polylactic acid (PLA) scaVolds form bone with a threshold dose of 230 g, after which the rate or quality of bone formation is unchanged up to 40 times the amount of protein (Sandhu et al., 1996, 1997). The successful repetition of BMP-2-induced bone formation in so many animal models makes it a suitable growth factor for translational studies in humans, although increased dose or potency is necessary for more evolved animals.

246

Cowan et al.

D. Increased Potency of BMP Heterodimers BMP-2, BMP-4, BMP-6, and BMP-7 show remarkable overlap during development (Lyons et al., 1995; Solloway et al., 1998), indicating that these proteins may work cooperatively to influence tissue formation. Additionally, natural bone extracts containing BMP-2, BMP-3, BMP-4, and BMP-7 are tenfold more active than rhBMP-2 alone (Wang et al., 1990; Wozney et al., 1988). Additional studies have demonstrated that the BMP proteins found in the bone matrix are heterodimeric in nature (Sampath et al., 1990; Wozney et al., 1988). Published reports suggest that heterodimeric BMP proteins stimulate greater in vitro and in vivo activity than corresponding homodimers (Aono et al., 1995; Israel et al., 1996; Lyons et al., 1995; Suzuki et al., 1997). Transient transfection of chinese hamster ovary (CHO) cells with various combinations of BMP-2 through BMP-7 expression vectors result in the secretion of heterodimeric BMPs (Israel et al., 1996). In vitro, cells secreting BMP-2/6, BMP-2/7, or BMP-4/7 heterodimers demonstrate maximal alkaline phosphatase activity as compared with other heterodimeric combinations or corresponding homodimers (Aono et al., 1995; Israel et al., 1996). BMP-2/7 heterodimers induce maximal alkaline phosphatase activity when BMP-2 and BMP-7 are infected at a 1:1 ratio (Israel et al., 1996). These data indicate that BMP-2/6, BMP-2/7, and BMP-4/7 heterodimers may have unique properties and provide an additional element of complexity. In vivo, recombinant human BMP-2/6, BMP-2/7, and BMP-4/7 heterodimers are considerably more potent in inducing ectopic bone formation than are any of the corresponding homodimers, with BMP-2/7 heterodimers being the most potent (Israel et al., 1996). Interestingly, BMP-2/6 and BMP-2/7 heterodimers induced endochondral bone formation (Israel et al., 1996), whereas BMP-4/7 heterodimers induced intramembranous bone formation (Aono et al., 1995). The underlying reasons for the diVerences in bone formation pathways are unknown, but may reflect the use of diVerent carriers and implantation sites, with BMP-2/6 and BMP-2/7 on a demineralized bone carrier implanted into the ventral thorax as opposed to BMP-4/7 on a collagen matrix implanted subcutaneously. Additionally, BMP-2/7 and BMP-4/7 heterodimers have been implicated in several developmental mechanisms (Nishimatsu and Thomsen, 1998; Schmid et al., 2000; Suzuki et al., 1997). In an Xenopus embryology model, BMP-4/7 heterodimers induced mesoderm formation at doses 20 times below that of BMP-4 or BMP-7 homodimers (Suzuki et al., 1997). Although complexes of type I and II BMP receptors have been described to bind BMP-2, BMP-4, and BMP-7 homodimers (Liu et al., 1995), the mechanism by which BMP heterodimers might interact with monomeric or multimeric receptor complexes to trigger an enhanced biological response

8. Evolving Concepts in Bone Tissue Engineering

247

has yet to be elucidated. The increased activity and low dose requirements demonstrated by BMP heterodimers make them attractive for future bone tissue engineering research.

III. Bone Tissue Engineering A. Bone Tissue Engineering Approaches Successful functional bone tissue engineering requires understanding the biomechanical properties of the native tissue for scaVold design and the biophysical microenvironment, as well as relevant stimulation and metabolic needs of the engineered tissue. Although bone tissue can self-repair, there are many cases where the damage is too severe or the local environment is suboptimal for adequate self-healing. These non-healing defects are termed critical size defects (CSDs); unfortunately, their meaning has been poorly defined. Technically, CSDs are known as defects that will not heal during the lifetime of the animal, although experimentally a 12-week healing period is acceptable. Should we then presume that a defect that heals only 50% of the defect area within a 12-week time period is a CSD? A review of the literature suggests that models investigating CSD healing observe about 10% healing in control defects during the study period and up through at least 12 weeks (Table II) (Abu-Serriah et al., 2003; Chang et al., 2003; Cook et al., 1995; den Boer et al., 2003; Gerhart et al., 1993; Hollinger and Kleinschmidt, 1990; Kruyt et al., 2004; Miki et al., 2000; Petite et al., 2000; Schlegel et al., 2004; Shang et al., 2001; Warnke et al., 2004). Interestingly, CSDs are acquired with age, because infants and juvenile animals can regenerate bone in large skeletal defects deemed critical sized in adults (Aalami et al., 2004). Critical size defect experiments attempt to answer additional questions that ectopic, subcutaneous, and intramuscular implantation models cannot. In essence, CSDs are functional in vivo bone regeneration assays that seek to recreate clinically relevant bone defects and then ask clinically applicable questions such as: (1) what is the rate and degree of bone healing for a given scaVold–growth factor combination; (2) is there a dose-response relationship for a given combination; (3) how do diVerent scaVold–growth factor combinations aVect the rate and degree of healing; (4) what is the maximal defect size that can be regenerated for a given scaVold–growth factor combination; and (5) how does the addition of cells influence the previous four questions? A variety of CSD locations have been documented, including the calvaria, mandible, and long bones. In particular, calvaria defects have a high complication rate, because of the morbidity associated with the underlying brain tissue, but make an excellent model for bone regeneration, because they are non-load-bearing and easily accessible.

248

Cowan et al.

Current bone reconstruction techniques include a combination of autogenous, allogeneic, and prosthetic materials. Although autografts are the ‘‘gold standard,’’ they are clinically limited by low donor supply and the associated harvesting morbidity (Schlegel et al., 2004; Wiltfang et al., 2004). Although acceptable, allografts harbor risks, including disease transmission, immunogenicity, loss of biological and mechanical properties, increased cost, and unavailability (Friedlaender, 1983). Unfortunately, autografts for craniofacial defects have an unacceptably high failure rate (13–30%) (Gregory, 1972), and allografts have an even higher failure rate (20–35%) (Enneking and Mindell, 1991). Prosthetic materials avoid some of these issues, but their eVectiveness is limited by unpredictable graft resorption, infection, structural failure, and unsatisfactory aesthetic outcomes (Bostrom and Mikos, 1997; Fong et al., 2003; Mulliken and Glowacki, 1980). The search for a reliable implantable material has spurred a new line of research on biocompatible implantable scaVolds. The ideal implant should be widely available and at a low cost. Realistically, one implant will not suYce for all situations, and a basic scaVold that can be customized will prevail (Lindsey, 2001). Most bone tissue engineering strategies are multicomponent, so for clarity we will separately highlight the lessons learned from nature, microenvironment, characteristics of the ideal implantable scaVold, osteoprogenitor cells, and BMP-induced tissue engineering.

B. Lessons Learned from Nature Bone formation can occur through two distinct pathways: intramembranous and endochondral ossification. Although the cranial structures and mandible undergo intramembranous bone formation, the appendicular skeleton forms through endochondral bone formation. Current research has not identified the characteristics of the microenvironment responsible for one bone formation process over another, although osteoblasts harvested from diVerent bones within the body vary in their expression levels of bone markers and proliferation rates, presumably related to their history, since bones in certain locations turn over less frequently than other bones. Thus, not all osteoblasts are created equal. During endochondral ossification, mesenchymal stem cells condense and diVerentiate into chondrocytes, which subsequently mature, undergo hypertrophy, mineralize, and are invaded by osteoprogenitor cells (Owen, 1980). In contrast, the flat bones utilize intramembranous bone formation, which involves direct diVerentiation of mesenchymal cells to osteoblasts (Marvaso and Bernard, 1977). The intramembranous bone development pathway is reported to decrease the time required for the completion of bone formation, because it eliminates the cartilaginous intermediate phase (Cunningham

8. Evolving Concepts in Bone Tissue Engineering

249

et al., 2002). From the perspective of optimal functional bone regeneration, the intramembranous pathway of bone formation may be an ideal strategy for tissue engineering, because it accelerates bone formation. Factors governing intramembranous versus endochondral bone formation remain largely unknown. For instance, it is well documented that embryonic calvarial cells can diVerentiate into multiple cell lineages, including osteoblasts, adipocytes, myoblasts, and chondrocytes in the proper permissive environment, but a distinct endochondral phase is never observed during the development of most calvarial and facial bones (Toma et al., 1997). In addition, calvarial cell cultures from 12-day chicken embryos are more chondrogenic than 17-day embryos when grown using serum substitute (Toma et al., 1997). Ultimately, both intrinsic cellular factors and extrinsic factors specific to the local craniofacial microenvironment may dictate intramembranous versus endochondral ossification. Consistent with this is the observation that absence of movement inhibits secondary cartilage formation in the articulating surfaces of the maxillary process (Toma et al., 1997), and the observation that nonstabilized fractures favor healing through endochondral ossification, whereas stabilized fractures heal more directly through intramembranous ossification (Le et al., 2001). Thus, it appears that given a similar osteoprogenitor cell and growth factor condition, the microenvironment can directly influence intramembranous or endochondral bone formation. With respect to growth factors, BMP-2, BMP-4, and BMP-7 expressions have been documented during both intramembranous and endochondral bone formation processes (Fig. 1). Because the BMPs are key regulators of bone formation, their expression generally precedes histologically or radiographically confirmed bone. Additionally, their expression profile is regulated both temporally and spatially during bone development and fracture healing. For example, the developing mouse embryo regulates the expression of these BMPs within the cranium. BMP-2 is ubiquitously expressed within most tissues. BMP-4, on the other hand, is expressed during the beginning of mesenchymal diVerentiation [embryonic day (E)6–E16] in the dermis, meninges, cranial base, and undiVerentiated cells lining the bone, whereas BMP-7 is located within the ectoderm and osteogenic fronts (Holleville et al., 2003). BMP-4 was also localized at the tip of the nose, where BMP-7 was not identified. At later stages in the mouse embryo (E17–E21), BMP-7 expression is found in developing bone of precartilaginous condensations and perichondrium but not in cartilage (Lyons et al., 1995), as well as in craniofacial regions, including tooth buds, precartilaginous mesenchyme of the mandible, and mesenchyme of the nasal epithelium (Lyons et al., 1995). BMP-7 is also detected in some regions of the brain known to induce the formation of specific cranial skeletal elements, in the condensing mesenchyme of developing craniofacial bones, and in perichondrial regions

250

Cowan et al.

Figure 1 Representation of BMP-2, BMP-4, and BMP-7 expression during intramembranous and endochondral ossification. Three-dimensional reconstruction of mouse cranium (upper left) and tibia (lower left) from microCT. Corresponding cartoon represents BMP-2, BMP-4, and BMP-7 expression patterns during cranial development (upper right) or long bone fracture healing (lower right).

of developing axial structures (Schowing, 1968). Interestingly, this expression profile localizes BMP-4 and BMP-7 to diVerent tissue domains during development and may play a role in diVerential bone formation pathways. Unfortunately, a BMP profile has not been elucidated for the developing long bones; however, during endochondral fracture healing, BMPs display a diVerent expression profile (Ferguson et al., 1999; Le et al., 2001). In a fracture healing model of the rat femur, BMP-7 expression was dramatically elevated in the periosteum at day 3, whereas the expression of BMP-2, BMP-4, and BMP-7 were identified in osteoblasts undergoing intramembranous ossification and fibroblasts surrounding cartilage islands by day 7 (Onishi et al., 1998). At this time, BMP-7, but not BMP-2 or BMP-4, was also localized to hypertrophic chondrocytes. BMP-4 stimulation is involved in callus formation after fracture (Yaoita et al., 2000), hypertrophic chondrocyte diVerentiation (Anderson et al., 2000), and endochondral bone formation (Shen et al., 2004). By day 14, BMP-7 expression was identified minimally within osteoblasts; however, BMP-2 and BMP-4 expressions were elevated in hypertrophic chondrocytes and osteoblasts. This expression profile indicates a sequential regulation with BMP-7 expression early in osteoblasts and hypertrophic chondrocytes, followed by later BMP-2 and BMP-4 expressions. Distraction osteogenesis (DO) is another example of fracture healing in which the two ends of healing bone are mechanically distracted apart in a gradual fashion to stimulate increased bone formation between the defect

8. Evolving Concepts in Bone Tissue Engineering

251

edges. Distraction generally occurs over a period of weeks, during which time the microenvironment is mechanically stimulated, resulting in endochondral ossification (Jazrawi et al., 1998; Rauch et al., 2000). During the latency period within the first week after fracture, the expression of BMP-7 is elevated in mesenchymal cells and osteoblasts, but decreases at the onset of distraction, 3–7 days after fracture (Spector et al., 2001). Expression of BMP-2 and BMP-4 correlates with initiation of distraction and decreases upon the cessation of distraction (Sato et al., 1999). BMP-2 and BMP-4 expression is coordinately elevated within cells lining bony islands (Spector et al., 2001), as well as osteoblasts and chondrocytes (Bostrom, 1998; Bostrom et al., 1995) of primitive woven bone. These data suggest that mechanical stimulation encountered during distraction regulates the expression of these BMPs. Additionally, a hypoxic environment is suYcient to induce the expression of BMP-2 in endothelial cells found at a fracture site (Bouletreau et al., 2002). Thus, the microenvironment conditions, including elevated mechanical stimulation, hypoxia, elevated expression of BMP-2 and BMP-4, or decreased expression of BMP-7 within diVerentiating mesenchymal cells, have been demonstrated during endochondral bone formation and can potentially influence bone tissue engineering. These hints from nature indicate that, within the bone healing microenvironment, BMP-7 may act to initiate and pattern osteoblastic diVerentiation, whereas BMP-2 and BMP-4 are necessary for further maturation. Overall, much like the naturally present BMP heterodimers mentioned previously, these findings suggest that the most successful BMP strategies may be to sequentially apply diVerent osteogenic BMPs at diVerent repair time points coordinated with the microenvironment.

C. Microenvironment As previously suggested, the microenvironment plays a large role in predicating endochondral versus intramembranous ossification. Depending on the microenvironment, a given mesenchymal cell can be ‘‘programmed’’ to diVerentiate into an osteoblast or chondroblast (Le et al., 2001; Toma et al., 1997). Hence, the importance of the microenvironment to tissue engineering cannot be overemphasized. In fact, successful ‘‘tissue engineering’’ essentially results from successful ‘‘microenvironment engineering.’’ Many elements contribute to the microenvironment experienced by cells, including intrinsic cellular factors, such as cell lineage and diVerentiation state, as well as extrinsic cellular factors, such as the ECM, soluble factors within the ECM, cell–cell interactions, and environmental factors (Bottaro et al., 2002). Importantly, scaVold material and architecture also determine

252

Cowan et al.

the mechanical properties of the three-dimensional structure (Bottaro et al., 2002). Growth and diVerentiation of most tissues depend on the complex interplay of events in which cells under the influence of various intrinsic and extrinsic factors continually elaborate and remodel an ECM that, in turn, can influence the intrinsic factors within the cell (i.e., cell–matrix interactions). For example, chondrocytes seeded onto collagen II maintain their phenotype, but, when seeded onto collagen I, they become fibroblastic (Nehrer et al., 1997), indicating the importance of the ECM in determining cellular fate and diVerentiation. For bone tissue engineering, the implant must create a microenvironment favorable for osteogenesis. ScaVolds not only provide an adhesive substrate and physical support, but also provide the surface properties important for cellular signaling. Surface properties such as texture, roughness, hydrophobicity, charge, and chemical composition are known to aVect cell adhesion and function (Bottaro et al., 2002). Recent studies suggest that osteoblasts prefer to attach to surfaces with a positive charge (Schneider et al., 2004), moderate hydrophobicity with increased surface roughness and irregularity (Geckeler et al., 2003), and are coated with proteins such as collagen and hydroxyapatite (Chou et al., 2004; Yamamoto et al., 1997). The presence of collagen in the majority of tissues points to its importance in cellular proliferation and diVerentiation. Clues like these help us to understand why some materials are inert, while others are extremely osteoinductive. Indeed, in our laboratory we have demonstrated that osteoconductive and osteoinductive scaVolds induce ADAS cells to diVerentiate into osteoblast—in the absence of growth factor stimulation (Cowan et al., 2004). Additionally, cell–ECM interactions through integrin receptors allow for the signal transduction of mechanical stimulation. Internally, these signals activate various pathways, including MAPK, phospholipids-C gamma, and phosphatidyl inositol 3-kinase pathways, which lead to changes in cellular migration, proliferation, transcription, and diVerentiation (Schwartz, 1997). Importantly, scaVolds also determine the mechanical properties of the three-dimensional structure (Bottaro et al., 2002) and may also influence the release pharmacokinetics of various soluble factors. In addition, various proteins, small molecules, or ions (i.e., calcium or zinc) in the ECM also contribute to the microenvironment (Bottaro et al., 2002). Proteins such as BMPs, acting in an autocrine or paracrine fashion, and androgens and parathyroid hormone (PTH) are some of the important endocrine factors that contribute to skeletal homeostasis (Harada and Rodan, 2003). At a deeper level, proteins must have the correct conformation and be present in an active form, as opposed to a latent form. Cell–cell interactions are another critical component of the microenvironment. Specifically, epithelial–mesenchymal induction is critical to the formation of mesenchymal condensations that represent the first sign of skeletogenic diVerentiation.

8. Evolving Concepts in Bone Tissue Engineering

253

Following this, a certain degree of cellular density or confluence within the condensation is required before further chondroblastic or osteoblastic diVerentiation can occur (Langille, 1994). This same phenomenon is also observed in cultured chondroblasts or osteoblasts (Purpura et al., 2004; Schulze-Tanzil et al., 2002). Lastly, environmental factors such as dynamic or static mechanical forces, shear forces, pH eVects, and oxygen tension can also directly impact the microenvironment (Bancroft et al., 2002). As previously mentioned, non-stabilizing fractures are associated with endochondral rather than intramembranous bone formation (Le et al., 2001), although the eVects of hypoxia on chondrogenic diVerentiation is still unclear (Malda et al., 2003). Beyond the scope of this present discussion, various host response factors such as age, sex, and disease states can also significantly influence the microenvironment (Cowan et al., 2003).

D. Implantable Scaffolds The terms ‘‘scaVold,’’ ‘‘delivery vehicle,’’ and ‘‘carrier’’ are often used interchangeably. They usually denote the same meanings; however, a scaVold per se contains engineered microporosity that fosters cellular ingrowth as opposed to simply having pharmacological eVects. ScaVold architecture and surface topography are critical to cell–matrix interactions and constitute a key component of the microenvironment in tissue engineering. The scaVold material not only defines the surface properties in terms of adhesive substrate and physical support, but also provides the surface properties important for cellular signaling (Bottaro et al., 2002). Biomaterials for bone regeneration can be classified as inert, not stimulating bone formation; osteoinductive, able to stimulated undiVerentiated cells into osteoblasts; or osteoconductive, a material that serves as a scaVold on which bone cells can attach, migrate, and proliferate. Historically, the first evidence of bone tissue engineering using an implantable scaVold occurred in 1969, when Winter et al. implanted a synthetic polyhydroxyethylmethacrylate (poly-HEMA) sponge under the skin of young pigs and demonstrated intramembranous ossification (Winter and Simpson, 1969). Since then, the development of ‘‘smart biomaterials’’ has been one of the most exciting emerging fields. This concept relies on the ability of an implanted biomaterial to exhibit the following properties: (1) biocompatibility (i.e., nontoxic, nonimmunogenic), (2) osteoconductive, (3) osteoinductive, (4) structural stability (i.e., mechanical properties like native bone), and (5) bioresponsiveness (i.e., match bone regeneration with scaVold degradation). Many scaVold materials have been developed and extensively covered in other reviews and thus are not covered here (Geiger et al., 2003; Rosso et al., 2004).

254

Cowan et al.

In general, the future of scaVolding is directed toward natural and synthetic degradable polymers (Chen et al., 2001; Chou et al., 2004; Hollinger and Battistone, 1986; Winn et al., 1998). Biodegradable scaVolds have a distinct advantage over non-biodegradable prosthetic implants, because they allow for bone regeneration, functionality, and eventual elimination of the implant. These biologically compatible and degradable implants allow for reproducible and customized therapy that minimizes many of the foreign body response problems associated with alloplasts (Dupoirieux et al., 1994). In special circumstances, such as weight-bearing total hip replacements and spinal fusion, metal implants are still necessary. These implants require additional nonmetal components (i.e., ceramics) or growth factors (i.e., BMP-2) for eVective osseous integration with the natural bone (Bragdon et al., 2003). Interestingly, BMP-2 stimulation is able to accelerate bone formation when coated on inert, osteoinductive, or osteoconductive materials. However, long-term adverse eVects are still uncertain and may include material wear (Slonaker and Goswami, 2004) and increased bone resorption at both increased stress sites (i.e., previous osseous integration sites) and nonstressed sites (i.e., within titanium cages for spinal fusion) (Wang et al., 1999). In addition, although many materials reproduce some structural properties of the ECM such as metals, ceramics, natural and synthetic polymers, and organic material such as collagen (Geiger et al., 2003; Langer and Vacanti, 1993; Orban et al., 2002; Parikh, 2002; Ramoshebi et al., 2002; Rose and OreVo, 2002), all are still rather rudimentary when compared with the ‘‘real thing.’’ Thus, the immediate challenge of bone tissue engineering is to design a scaVold structure that mimics the mechanical, structural, osteoconductive, and osteoinductive properties of natural bone extracellular matrix. The ideal scaVold should exploit the inherent regenerative capability of osseous tissue by allowing bone marrow, periosteum, or mesenchymederived osteoprogenitor cells to populate and mineralize the scaVold. Foremost, a clinically relevant carrier system must be safe and exhibit properties that are well characterized and reproducible. The ideal scaVold material will be osteoconductive and osteoinductive with three-dimensional interconnected pores that support cellular ingrowth, communication, and bone formation. Osteoconductive matrices use biological materials (i.e., collagen or demineralized bone) and/or nonbiological materials, including various metals (i.e., titanium), degradable polymers, and bioglass. Their advantages include controlled construction, no or decreased immunogenicity, and biocompatibility. The ultimate scaVold design and fabrication should attempt to recreate the physical environment detected in vivo by cells for tissue specific biocompatibility. Often a scaVold itself may not be suYcient for bone regeneration; thus, enhancing cellular migration, proliferation, or diVerentiation on the scaVold

8. Evolving Concepts in Bone Tissue Engineering

255

may help overcome insuYciencies. The use of numerous scaVold materials indicates the large volume of research conducted in this field. Unfortunately, no definitive answer to the optimal scaVold type has been reached, nor has one implant strategy been confirmed in numerous animal models. Integration of matrix proteins, such as fibronectin, enhances cellular attachment and migration (Nuttelman et al., 2001; Rowley and Mooney, 2002), whereas osteogenic coatings such as hydroxyapatite or collagen, progenitor cells such as BMS or ADAS cells, or additional stimulants such as growth factors may be needed for a successful treatment. Today, the ideal bone tissue engineering strategy incorporates not only a biodegradable extracellular matrix, but also osteoinductive factors or cells.

E. Osteoprogenitor Cells Bone tissue engineering research has demonstrated the therapeutic advantages of delivering osteoprogenitor cells in scaVolds. Intrinsic cellular factors such as cell lineage and diVerentiation status are additional key determinants of microenvironment. Consequently, implantation of cells that are already committed or diVerentiated along the osteoblastic lineage may significantly shift the microenvironment to favor and/or maintain osteoinduction. In addition, an exogenous influx of already committed and diVerentiated osteoblastic cells may significantly accelerate bone formation by truncating or bypassing altogether the usual steps of osteoblast commitment, proliferation, and diVerentiation. The ideal source for autologous osteoblastic cells is still not well established. Osteoblasts derived from bone and periosteum directly participate in bone healing, but are less feasible for bone tissue engineering strategies, which require the use of large numbers of cells for successful healing. Harvesting cells from these tissues contributes to the problem of missing or damaged bone. Many other cell types have demonstrated osteogenic properties and the ability to form bone in vivo. These include embryonic stem cells and adult-derived stem cells. The use of embryonic stem cells, which are capable of forming bone nodules (Buttery et al., 2001), has declined because of ethical controversy. Adult-derived MDSCs, fibroblasts, BMS cells, and ADAS cells have demonstrated a reproducible ability to diVerentiate down the osteogenic lineage and form bone in vivo (Bosch et al., 2000; Krebsbach et al., 2000; Lee et al., 2001, 2002; Peng et al., 2002; Wright et al., 2002). MDSCs and fibroblasts oVer the advantage of being widely available; however, unpleasant aesthetic outcomes may temporarily occur. Most cell-based therapies for bone regeneration have focused on the active multipotent precursor cells found in bone marrow (BMS cells) (Gundle et al., 1995; Haynesworth et al., 1992; Im et al., 2001; Johnstone et al., 1998;

256

Cowan et al.

Toquet et al., 1999; Wakitani et al., 1995). Although donating bone marrow may be an unpleasant experience for patients, the richness of the extracted cells aids in many bone tissue engineering strategies, making them a popular choice. Pittenger et al. (1999) defined BMS cells as the ‘‘gold standard,’’ because under the appropriate conditions they are able to diVerentiate along three lineages: osteoblastic, adipocytic, and chondrocytic. In vivo, implanted BMS cells have aided in the healing of critical-sized skeletal defects in many animal models, including mice (Cowan et al., 2004, 2005; Krebsbach et al., 1998), rats (Blum et al., 2003; Lieberman et al., 1999), rabbits (Chang et al., 2004; Dean et al., 2003), swine (Chang et al., 2003), goats (Kruyt et al., 2004), and sheep (Petite et al., 2000; Shang et al., 2001). Although some research has demonstrated successful bone defect healing with unstimulated BMS cells, most research has emphasized the need for additional growth factor stimulation within the microenvironment for timely healing. This is due to the low number of committed osteoprogenitor cells, usually 0.0001% of a bone marrow harvest (Bruder et al., 1997; Haynesworth et al., 1992). Accordingly, growth factor stimulation is required to recruit and expand the number of osteoprogenitor cells in harvested marrow specimens. The reduction of healthy bone marrow cells in aging or diseased adults also adds to the diminished number of osteoprogenitor cells (Egrise et al., 1992; Quarto et al., 1995). Because the success of using bone marrow in vivo is critically dependent on implantation of suYcient numbers of cells, this approach may be less applicable than others. More recently, adipose tissue has received much attention for its source of multipotent mesenchymal cells. In vitro, ADAS cells can diVerentiate down osteogenic, chondrogenic, myogenic, adipogenic and even neuronal pathways (Ashjian et al., 2003; Erickson et al., 2002; Halvorsen et al., 2001a,b; Mizuno et al., 2002; SaVord et al., 2002; Tholpady et al., 2003; Wickham et al., 2003; Zuk et al., 2001, 2002). The greatest advantage of these cells is that they are readily available in large numbers with minimal donor morbidity (i.e., cosmetic liposuction) and attach and proliferate rapidly in culture, making them a very attractive cell source for tissue engineering. ADAS cells under rhBMP-2 stimulation produce more in vitro bone nodules than either osteoblasts or BMS cell cultures under the same conditions (Ashjian et al., 2003; De Ugarte et al., 2003; Halvorsen et al., 2001a,b; Lee et al., 2003; Ogawa et al., 2004; Tholpady et al., 2003; Zuk et al., 2001, 2002). The in vivo osteogenic capability of ADAS cells placed extraskeletally (Dragoo et al., 2003; Lee et al., 2003) or within skeletal defects (Cowan et al., 2004, 2005) has equaled that of BMS cells and nearly reached that of committed osteoblasts. Local application of ADAS osteoprogenitor cells has successfully aided in the eVort to heal skeletal defects; however, as is the case for BMS cells, it may benefit from additional growth factor stimulation within the microenvironment for timely bone formation (Cowan et al., 2005). Clinically, ADAS

8. Evolving Concepts in Bone Tissue Engineering

257

cells may become more popular than BMS cells based on the notion that patients may be more willing to donate adipose tissue over bone marrow.

F. Bone Tissue Engineering Using BMP-2, BMP-4, or BMP-7 Soluble factors and appropriate vehicles or carriers for their controlled delivery constitute a critical element of the microenvironment. BMPs are water-soluble, relatively low-molecular-weight molecules that diVuse very easily in body fluids; thus, a suitable carrier or delivery vehicle is necessary to prevent undesirable migration and unintended rapid diVusion of BMPs (Rengachary, 2002). BMP release extremes (bolus injections or prolonged low-level release) are not beneficial to bone induction (Li and Wozney, 2001). Thus, BMP delivery vehicles need to supply the factor at a defined ‘‘optimal’’ level or temporal period to attain maximal osteogenic induction (Woo et al., 2001). In some instances, implantation of osteogenic cell-seeded scaVolds in the absence of growth factors is suYcient to induce bone formation (Cowan et al., 2004; Dean et al., 2003; den Boer et al., 2003; Krebsbach et al., 1998; Kruyt et al., 2004; Petite et al., 2000; Shang et al., 2001); however, markedly increased and timely bone formation generally requires growth factor (i.e., BMP) accompaniment. Furthermore, implantation of osteoinductive material/factors is very attractive, because it can be used in addition to or in the absence of implanted cells for fast and convenient clinical treatments. Although BMPs are clearly the most osteoinductive factors described to date, other factors important to bone formation include TGF-
, fibroblast growth factors (FGFs), vascular endothelial growth factor (VEGF), platelet-derived growth factor (PDGF), and human growth factor (hGF). In particular, BMP-2 and BMP-7 are potent osteogenic inducers and have been used more than any other factor in bone tissue engineering research. Table II outlines bone tissue engineering experiments utilizing CSDs. Because these are nonhealing defects, the addition of osteogenic factors and/or cells is required for bone formation. A closer evaluation of these CSD experiments demonstrates that BMP-2 and BMP-7 tissue engineer intramembranous bone in areas normally associated with either intramembranous (calvarial) or endochondral (long bone) bone formation. These analyses demonstrate the power of BMP-2 and BMP-7 to overcome endogenous signaling and redirect the bone formation process by eliminating the formation of cartilaginous intermediate tissue. Interestingly, lessons learned from nature are not always recapitulated in engineered situations of bone formation. In these CSD studies, BMP-4 stimulation induced endochondral ossification in tissues normally associated with either intramembranous

Table II Bone Healing Models in Critical Size Defects*

First Author (Year)

Animal

Cowan et al. (2004)

Mouse

Cowan et al. (2005)

Mouse

Model Calvaria

Calvaria

Size 4 mm

4 mm

Krebsbach et al. (1998) Mouse

Calvaria

5 mm

Wright et al. (2002)

Mouse

Calvaria

5 mm

SCID

Calvaria

5 mm

SCID

Calvaria

5 mm

Lee et al. (2001)

Biomaterial

Surface Modification

I/E

Percentage of Area with Bone

None

I

<10% at 12 weeks

None None None None None rhBMP-2 (20 ng/ml) þ RA (10 M) Osteoblasts rhBMP-2 (20 ng/ml) þ RA (10 M) BMS cells rhBMP-2 (20 ng/ml) þ RA (10 M) ADAS cells rhBMP-2 (20 ng/ml) þ RA (10 M) None None Spleen None stromal BMS cells None MDSC None MDSC BMP-4 retrovirus MDSC None MDSC BMP-4 retrovirus None None None None MDSC None MDSC BMP-2 adenovirus

I I I I I I

<10% at 12 weeks 90% at 12 weeks 40% at 12 weeks 90% at 12 weeks 85% at 12 weeks <10% at 12 weeks

I

80% at 2 weeks

I

80% at 4 weeks

I

80% at 2 weeks

I I

<10% at 12 weeks 15% at 4 weeks

I E E E E I I I I

99% at 2 weeks <10% at 3 weeks 100% at 2 weeks <10% at 3 weeks 100% at 2 weeks 30% at 4 weeks 25% at 4 weeks 30% at 4 weeks 90% at 4 weeks

Cells

Poly(glycolic acid-co-lactic acid) – (PLGA) PLGA PLGA PLGA PLGA PLGA PLGA

None

None

Apatite Apatite Apatite Apatite Apatite Apatite

None Osteoblasts Dura mater BMS cells ADAS cells None

PLGA

Apatite

PLGA

Apatite

PLGA

Apatite

Gelatin sponge Gelatin sponge

None None

Gelatin sponge Collagen I sponge Collagen I sponge Collagen I sponge Collagen I sponge None Collagen I sponge Collagen I sponge Collagen I sponge

None None None None None None None None None

Growth Factors

Lee et al. (2002)

SCID

Calvaria

5 mm

Peng et al. (2002)

Mouse

Calvaria

6 mm

Zanchetta et al. (2003)

Rat

Calvaria

5 mm

Woo et al. (2001)

Rat

Calvaria

5.6 mm

Miki et al. (2000)

Rat

Calvaria

8 mm

Sweeney et al. (1995)

Rat

Calvaria

8 mm

Gysin et al. (2002)

Rat

Calvaria

8 mm

Lutolf et al. (2003)

Rat

Calvaria

8 mm

Schmoekel et al. (2004) Rat

Calvaria

8 mm

None Collagen I sponge Collagen I sponge Collagen I sponge Gelfoam disk

None None None None None

None None hBMS cells hBMS cells MDSC

I I I I E

30% at 4 weeks 25% at 4 weeks 30% at 4 weeks 90% at 4 weeks <10% at 3 weeks

E E

40% at 3 weeks 60% at 3 weeks

I I I I

15% at 3 weeks 90% at 3 weeks <10% at 6 weeks 45% at 6 weeks

I

75% at 6 weeks

None None None None None

None None None BMP-2 adenovirus LacZ þ VEGF retrovirus BMP-4 retrovirus BMP-4 þ VEGF retrovirus None None None rhBMP-2 (29 g) for 7 days rhBMP-2 (35 g) for 21 days None rhBMP-2 (10 g) None None None

Gelfoam disk Gelfoam disk

None None

MDSC MDSC

None HE 800 (expolysaccharide) PLGA microspheres PLGA microspheres

None None None None

None None None None

PLGA microspheres

None

None

PLGA PLGA None Collagen I gel Reconstituted basement membrane Laminin gel Gelatin matrix Gelatin matrix Gelatin matrix MMP-sensitive hydrogels MMP-insensitive hydrogels MMP-sensitive hydrogels Collagen I gel None Fibrin matrix Fibrin matrix Fibrin matrix Fibrin matrix

Hydroxyapatite Hydroxyapatite None None None

I I I I I

<10% at 4 weeks 100% at 4 weeks <10% at 12 weeks 90% at 12 weeks 56% at 12 weeks

None None None None PEG PEG PEG None None None None None None

None None BMS cells BMS cells None None None None None None None None None

None None None BMP-4 adenovirus None rhBMP-2 (5 g) rhBMP-2 (5 g) rhBMP-2 (5 g) None None nglBMP-2 (1 g) nglBMP-2 (5 g) nglBMP-2 (20 g)

I I I I I I I I I I I I I

46% at 12 weeks <10% at 4 weeks 20% at 4 weeks 100% at 4 weeks 20% at 5 weeks 30% at 5 weeks 90% at 5 weeks 90% at 5 weeks 12% at 3 weeks 20% at 3 weeks 80% at 3 weeks 95% at 3 weeks 100% at 3 weeks

(Continued)

Table II Continued

First Author (Year)

Animal

Model

Size

Blum et al. (2003)

Rat

Calvaria

8 mm

Toung et al. (1998)

Rat

Nasalis bone

5
15 mm

Lieberman et al. (1999)

Rat

Femur

23
4 mm

Fang et al. (1996)

Rat

Femur

5 mm

Ono et al. (2004)

Rabbit

Calvaria

1.2 mm

Chang et al. (2004)

Rabbit

Calvaria

1.2 mm

Dean et al. (2003)

Rabbit

Calvaria

1.5 cm

Hong et al. (2000)

Rabbit

Calvaria

2 mm

Surface Modification

Biomaterial Scintered titanium fiber Scintered titanium fiber Scintered titanium fiber Scintered titanium fiber None Collagen I gel Collagen I gel Demineralized bone Demineralized bone Demineralized bone Demineralized bone Demineralized bone Collagen I sponge Collagen I sponge Collagen I sponge

mesh mesh mesh mesh

Cells

None None None None None None None None None None None None None None None

BMS cells BMS cells BMS cells BMS cells None None None None BMS cells BMS cells BMS cells None None None None

Hydroxyapatite pellets Hydroxyapatite pellets Hydroxyapatite pellets Hydroxyapatite pellets Alginate polymer Alginate polymer Poly(propylene fumarate)/
-tricalcium phosphate PPF/
-TCP PPF/
-TCP PPF/
-TCP with channels None None

None None None None None None None

None None None None BMS cells BMS cells BMS cells

PPF foam PLGA foam PLGA foam None None

BMS cells BMS cells BMS cells None None

Gelatin hydrogel Gelatin hydrogel

None None

None None

Growth Factors I/E

Percentage of Area with Bone

None hBMP-2 liposome hBMP-2 retrovirus hBMP-2 adenovirus None None hIGF-1 (3 g) None None
-gal adenovirus hBMP-2 adenovirus rhBMP-2 (20 g) BMP-4 plasmid PTH plasmid BMP-4 þ PTH plasmids None BMP-2 liposomes None BMP-2 liposomes
-gal adenovirus BMP-2 adenovirus None

I I I I I I I I I I I I E E E

30% at 4 weeks 31% at 4 weeks 35% at 4 weeks 43% at 4 weeks <10% at 4 weeks 91% at 4 weeks 84% at 4 weeks <10% at 8 weeks <10% at 8 weeks <10% at 8 weeks 60% at 8 weeks 28% at 8 weeks 100% by 18 weeks 100% by 18 weeks 100% at 4 weeks

I I I I I I I

38% 68% 30% 64% 47% 79% 57%

None None None None free rhTGF-
1 (500 ng) None rhTGF-
1 (500 ng)

I I I I I

25% at 20 100% at 6 47% at 20 50% at 16 60% at 16

I I

60% at 16 weeks 60% at 16 weeks

at at at at at at at

9 weeks 9 weeks 9 weeks 9 weeks 12 weeks 12 weeks 20 weeks weeks weeks weeks weeks weeks

Moghadam et al. (2004)

Clokie et al. (2002)

Rabbit

Rabbit

Calvaria

Calvaria

15
17 mm None

13–15 mm

Elshahat et al. (2004)

Rabbit

Calvaria

20 mm

Bourgeois (2003)

Rabbit

Femur

6 mm

Zegzula et al. (1997)

Rabbit

Radius

20 mm

Wiltfang et al. (2004)

Mini-pig Calvaria

CSD

None

Demineralized bone gel None Demineralized bone gel Calcium hydroxide None None Plaster of Paris (Calcium sulfate) None Poloxamer None Poloxamer Demineralized bone Norian CRS (calcium phosphate) None Bone source (calcium phosphate) None None None Bioglass None Bioglass Demineralized bone Apatite cement None Biphasic calcium phosphates None Calcium-deficient apatite None Calcium-deficient apatite/ None mannitol None None Autogenous bone None Polylactic acid None Polylactic acid None Polylactic acid None Polylactic acid None Autogenous bone None Autogenous bone PRP 1 Autogenous bone PRP 2 Cerasorb None Cerasorb PRP 1 Cerasorb PRP 2 Bio-Oss None Bio-Oss PRP 1 Bio-Oss PRP 2 Colloss None Colloss PRP 1

None

None

I

13% at 12 weeks

None None None None None None

None None None None None None

I I I I I I

45% 41% 13% 32% 15% 96%

None None None None None

None None None None None

I I I I I

12% at 12 weeks 17% at 12 weeks <10% at 8 weeks 50% at 8 weeks 75% at 8 weeks

None None None None

None None None None

I I I I

100% at 8 weeks 19% at 3 weeks 24% at 3 weeks 28% at 3 weeks

None None None None None None None None None None None None None None None None None

None None None rhBMP-2 (17 g) rhBMP-2 (35 g) rhBMP-2 (70 g) None None None None None None None None None None None

I I I I I I I I I I I I I I I I I

<10% at 8 weeks 90% at 2 weeks <10% at 8 weeks 90% at 6 weeks 97% at 4 weeks 100% at 4 weeks 58% at 12 weeks 57% at 12 weeks 60% at 12 weeks 15% at 12 weeks <10% at 12 weeks 12% at 12 weeks 43% at 12 weeks 48% at 12 weeks 47% at 12 weeks 60% at 12 weeks 70% at 12 weeks

at at at at at at

12 12 12 12 12 12

weeks weeks weeks weeks weeks weeks

(Continued)

Table II Continued

First Author (Year)

Toriumi et al. (1991)

Animal

Dog

Model

Mandible

Size

3 cm

Biomaterial Colloss None Steel reconstruction plates

Dog

Maxilla

1 cm

Chang et al. (2003)

Swine

Calvaria

5 cm

Schlegel et al. (2004)

Pig

Calvaria

6 mm

Kruyt et al. (2004)

Goat

Iliac wing

17 mm

Abu-Serriah et al. (2003) Shang et al. (2001)

Sheep

Mandible

35 mm

Sheep

Calvaria

20 mm

Petite et al. (2000)

Sheep

Metatarsals

25 mm

Cells

Growth Factors I/E

Percentage of Area with Bone

None None None

None None None

I I I

67% at 12 weeks <10% at 24 weeks <10% at 24 weeks

None

rhBMP-2 (250 g)

I

68% at 10 weeks

None Autogenous bone Polylactic acid Polylactic acid Collagen I gel Collagen I gel Autogenous bone Autogenous bone Colloss Colloss Colloss Calcium phosphate ceramic Calcium phosphate ceramic Collagen I sponge

PRP 2 None Demineralized bone Demineralized bone None None None None None None None None None None None None None None

None None None None BMS cells BMS cells None None None None None None BMS cells None

None None None rhBMP-2 (200 g)
-gal adenovirus hBMP-2 adenovirus None PRP None PRP (low dose) PRP (high dose) None None rhBMP-7 (1 mg/cm3)

I I I I I I I I I I I I I I

35% at 16 weeks 40% at 16 weeks 20% at 16 weeks 55% at 16 weeks 57% at 12 weeks 87% at 12 weeks 100% at 12 weeks 100% at 12 weeks 60% at 12 weeks 69% at 12 weeks 63% at 12 weeks 51% at 12 weeks 54% at 12 weeks 85% at 8 weeks

Calcium alignate gel Calcium alginate gel Coral

None None None

None BMS cells None

None None None

E E I

<10% at 18 weeks 72% at 18 weeks 36% at 16 weeks

Steel reconstruction plates Mayer et al. (1996b)

Surface Modification

Boyne et al. (1997)

Human

Maxilla sinus 8.51 mm

Collagen I sponge (Helistat)

None

None

Warnke et al. (2004)

Human

Mandible

>7 cm

Human

Tibia

15 mm

Demineralized bone None None Demineralized bone None

BMS cells

Geesink et al. (1999)

Titanium mesh and Collagen I sponge None Collagen I sponge None

None None None None rhBMP-7 (2.5 mg) None None None rhBMP-2 (1.5 mg) None None rhBMP-7 (1 mg) None rhBMP-7 (250 g–2 mg) rhBMP-2 (1.77–3.4 mg) rhBMP-7 (7 mg)

None None None

None None None

– – –

None

rhBMP-7 (2.5 mg)

–

den Boer et al. (2003)

Sheep

Tibia

3 cm

Gerhart et al. (1993)

Sheep

Femur

2.5 cm

Cook et al. (1995)

Monkey Ulna

2 cm

Tibia

Coral None Autogenous bone Hydroxyapatite Hydroxyapatite Hydroxyapatite None Inactive bone matrix Inactive bone matrix Autogenous bone Collagen I sponge Collagen I sponge Collagen I sponge Collagen I sponge

None None None None None None None None None None None None None None

BMS cells None None None None BMS cells None None None None None None None None

Collagen I sponge

I I I I I I I I I I I I I I

71% at 16 weeks 33% at 12 weeks 38% at 12 weeks 13% at 12 weeks 100% at 12 weeks 90% at 12 weeks <10% at 12 weeks <10% at 12 weeks 100% at 12 weeks 100% at 12 weeks <10% at 20 weeks 100% at 6–8 weeks <10% at 20 weeks 100% at 6–8 weeks

I

100% at 16 weeks (8 of 11) 100% at 7 weeks

–

<10% at 1 year <10% at 1 year 100% at 10 weeks (4 of 6) 100% at 10 weeks (5 of 6)

*Original research papers describing animal models where bone tissue engineering was conducted in critical size defects. ADAS, Adipose-derived adult stromal; BMS, bone marrow stromal; E, endochondral ossification; I, intramembranous ossification; MDSC, musclederived stromal cell; MMP, matrix metalloproteinase; PEG, polyethylene glycolide; PRP, platelet-rich plasma; RA, retinoic acid.

264

Cowan et al.

(calvarial) or endochondral (long bone) bone formation (Fang et al., 1996; Peng et al., 2002; Wright et al., 2002), as well as one study in which intramembranous ossification was detected in the rat calvaria when implanted on a gelatin matrix (Gysin et al., 2002). Interestingly, no discrepancies were noted for the times needed to heal defects through an intramembranous versus endochondral pathway. DiVerences in bone formation pathways are likely due to bone microenvironment, implanted scaVold, implanted cells or growth factors, and the pharmacokinetics of the implant. Although there is a large volume of research in this field, the lack of consistency in animal models, defect location, and BMP doses makes most cross-comparisons diYcult. Within the category of small animals, mice, rats, and rabbits have similar popularity, and sheep are the most popular largeanimal model. Additionally, the two most popular locations for defects are the calvaria and the femur, representing intramembranous and endochondral pathways, respectively. BMP doses vary significantly based upon the diVerential needs for diVerent animal models; however, sustained release is unanimously preferred over a bolus release and large doses are preferred over small doses (Hollinger et al., 1998). Unfortunately, large doses of BMPs are diYcult to deliver over prolonged periods of time, due to insuYcient knowledge on how to manipulate BMP–scaVold interactions to maintain sustained release into the microenvironment.

G. BMP–Scaffold Interactions Ideally, pharmacokinetics (PK) analyses facilitate our understanding of how factors are released, absorbed, stored, and metabolized in the body. Understanding these processes allows researchers to better predict signaling eVects and improve outcomes of treatments. As expected, the concentration of rhBMP-2 greatly aVects cellular processes, with femtomolar concentrations inducing chemotaxis, nanomolar concentrations supporting mesenchymal cell proliferation, and micromolar concentrations promoting bone diVerentiation (Reddi, 1981, 1994). This profile would predict the need for lower initial rhBMP-2 concentrations in bone defects, with increasing concentrations over time for an eVective bone tissue engineering strategy. Despite the availability of rhBMPs (Wozney et al., 1988), their clinical use has been hampered by the lack of detailed PK analyses with regard to various delivery systems. Although BMPs are extremely osteoinductive molecules, their release into the microenvironment must be regulated for optimal bone formation. Thus, the challenge still remains to determine the optimal mix of BMPs, dosage, release dynamics, and matrix carrier that will result in a clinically therapeutic treatment without utilization of superphysiological doses (see next section) (Rose and OreVo, 2002). In addition

8. Evolving Concepts in Bone Tissue Engineering

265

to the influence of rhBMP-2, structural features of biomaterials have also been shown to influence rhBMP-2 pharmacokinetics. Delivery systems evaluated in animal models include hydroxyapatite (Levine et al., 1997), collagen (Yasko et al., 1992), PLA (Zegzula et al., 1997), poly(glycolic acid-co-lactic acid) (PLGA) (Woo et al., 2001), and demineralized bone (DMB) (Schwartz et al., 1998). Binding and interactions between BMPs and carriers play important roles and may contribute to the sustainability of the growth factor. Further understanding of BMP-2–scaVold interactions will aid the progression of research in this field. Clinically, absorbable collagen sponges (ACSs) have a long-standing safety record as hemostatic agents and wound coverings (Chvapil, 1977; Geiger et al., 2003). Collagen is known to promote cellular invasion and wound healing, has excellent biocompatibility, degrades into nontoxic end-products, and has favorable interactions with cells and factors. ACS has repeatedly been observed to be an eVective carrier for rhBMPs, independent of site of implantation (Winn et al., 2000). Bovine type I collagen has been used almost exclusively in clinical settings, because it binds BMPs very well. ACS has displayed diVerential binding aYnity to BMP-2 and BMP-7. This diVerence is best illustrated by the U.S. Food and Drug Administration (FDA) requirement of BMP-2 to be used in conjunction with a metal cage that prevents load-induced displacement of BMP-2 from the intended space of ACS (Rengachary, 2002). Generally, cages are not required for BMP-7 delivery, because the binding aYnity of ACS to BMP-7 is significantly stronger and can apparently resist the loss of BMP during similar loads (see next section). The release of rhBMP-2 or other growth factors from ACS requires knowledge of protein–collagen interactions, loading, eYcacy, and PK. A positive correlation is demonstrated, with slower rhBMP-2 release dictating a greater osteoinductive activity. Current rhBMP-2 release kinetics from various materials follow a biphasic release profile, with an early rapid rhBMP-2 loss (half-life of 10 min to several hours), followed by a more gradual rhBMP-2 loss with a half-life of 1 to 10 days (Uludag et al., 1999a; Winn et al., 1999). Thus, the ideal rhBMP-2 concentrations would be released at a rate opposite of what has been demonstrated and be present for prolonged periods of time. Release rates of rhBMP-2 vary greatly between materials, with ACS retaining 59–74%, rabbit demineralized bone matrix (rDBM) retaining 30–50%, and hydroxyapatite retaining 11% after 3 h (Uludag et al., 1999a). Experimentally, it is not possible to completely rule out the importance of the initial burst, but the local sustained concentration of rhBMP-2 seems more critical for overall osteoinductive activity in a dosedependent manner (Kenley et al., 1993; Winn et al., 1998). These results suggest that the osteopotency of bone-regeneration devices can be improved by using engineered BMPs with superior aYnity to scaVold surfaces.

266

Cowan et al.

The isoelectric points (pIs) of scaVolds and growth factors are significant determinants of binding and release characteristics (Uludag et al., 1999b). Depending on the manufacturing process, collagen exhibits a pI in the neutral or slightly acidic pH range, whereas rhBMP-2 has a pI of approximately 9. Thus, at a physiological pH collagen and rhBMP-2 have opposite charges and yield the greatest binding (0.1–0.2 mg rhBMP-2 per mg collagen) (Friess et al., 1999). Thus, electrostatic attractive forces between rhBMP-2 and collagen are believed to be major factors controlling the protein–matrix interactions. Upon submersion, a collagen scaVold typically  absorbs about 90% of a 1.5-mg/ml rhBMP-2 solution at 37 C (Uludag et al., 1999b). Within PLGA microspheres, growth factors can become encapsulated; however, the acidic microclimate is a source of protein instability that can be overcome using water-soluble basic salts to neutralize the polymer microclimate pH (Zhu et al., 2000). The pI of a protein can diVer depending on its source and augmentation. Two common sources for rhBMP-2 proteins are CHO cells and Escherichia coli, producing a glycosylated protein or an unglycosylated protein, respectively (Uludag et al., 1999b). Although rhBMP-2 derived from CHO cells has a pI of 9.0, E. coli-derived rhBMP-2 has a pI of 8.5. The initial burst eVect is dependent on the protein’s pI. For example, within collagen sponges succinylated rhBMP-2 with a pI of approximately 3 results in a 99% burst release by day 1, whereas native or unmodified rhBMP-2 protein with a pI of approximately 9 exhibits a burst release near 70% by day 1 (Winn et al., 2000). In vivo, the protein with the higher pI (CHO-derived rhBMP-2) is retained on the collagen scaVold the most, whereas the proteins with lower pI (acetylated/succinylated rhBMP-2) are retained the least. The higher retention could relate to a higher binding or entrapment of rhBMP-2 within the sponge, solubility profile in biological milieu, or interaction with other biomolecules at the implant site. Further studies investigated factors influencing rhBMP-2 pharmacokinetics in a rat ectopic assay by quantifying rhBMP-2 retention in the local scaVold area (Uludag et al., 1999a). Interestingly, these results were not influenced by pI, because the clearance rates of rhBMP-2 from the body for CHO and E. coli-derived proteins were similar (Uludag et al., 1999b). Furthermore, long-term (i.e., 144 days) results demonstrate that rhBMP-2 loss from the area is not aVected by the percent released. Various techniques can be utilized to enhance rhBMP-2 binding to bone minerals, such as heparin conjugates (Gittens et al., 2004). Heparin binds growth factors via electrostatic interactions between its negatively charged sulfate groups and the protein’s positively charged amino acid residues. It also increases growth factor stability (i.e., biological half-life). Unlike collagenous carriers, mineral-based carriers appeared to bind a fraction of rhBMP-2 (typically 5–10% of implanted dose) irreversibly (i.e., without

8. Evolving Concepts in Bone Tissue Engineering

267

release). The biological consequence of the tightly bound BMP fraction is currently not known. Additional strategies manipulate scaVold ingredients to finesse the BMP binding capacity. For example, the addition of glucose to collagen minipellets alters the rate of rhBMP-2 release as desired (Maeda et al., 2004a). Gelatin hydrogels released rhBMP-2 in vivo with a release profile that could be changed by altering the water content of BMP-2-incorporated hydrogels (Yamamoto et al., 2003). Other scaVold alterations have attempted to improve and sustain rhBMP2 release without success. Succinylated collagen sponges have negatively charged residues on their surface to aid in binding to rhBMP-2 and decrease release rate; unfortunately, these sponges did not alter the rhBMP-2 release profile. In another study, cross-linking of collagen led to reduced rhBMP2 incorporation through physical hindrance and a reduction of the collagen’s ability to swell upon soaking (Friess et al., 1999; Uludag et al., 1999b, 2001). The rhBMP-2 on cross-linked collagen scaVolds exhibited low initial retention (47.8%) relative to non-cross-linked collagen scaVolds (87.1%). Additionally, rhBMP-2 exhibited a biexponential release pattern with t1/2a at 10 min followed by t1/2b at 89 and 51 h for cross-linked and non-cross-linked collagen scaVolds, respectively. Interestingly, plasmin-cleaved rhBMP-2 resulted in a truncated form, resulting in a pI shift from 9.0 to 6.5 (Uludag et al., 2000). The active site of BMPs, within the cystine knot, was unaVected by the cleavage (Kirsch et al., 2000). This resulted in an increased release rate where 63% of the dose was eliminated in 4.2 days as compared with 4.3 days for uncleaved rhBMP-2. Because of the importance of controlling protein–material binding interactions, certain materials (i.e., ceramics) with high binding aYnity for proteins may not be the optimal carriers. In fact, calcium phosphate composite was found to be a suboptimal carrier (Ruhe et al., 2003). Released rhBMP2 binds to calcium phosphate cement, resulting in delayed release from the composite. Thus, the nanoporosity of the calcium phosphate cement not only did not facilitate the release of rhBMP-2, but may have further limited it because of the protein adsorption on the ceramic. Hyaluronic acid scaVolds have been used as delivery vehicles for bioactive rhBMP-2 in bone repair therapies (Hunt et al., 2001; Kim and Valentini, 2002). Hyaluronic acid is a natural polyanionic polysaccharide with structural roles in organizing the cartilage ECM, cell motility, and wound healing (Laurent and Fraser, 1992). Modified hyaluronic acid demonstrates altered physical characteristics such as decreased solubility or the need for chemical reactions undesirable under physiological conditions (Hubbell et al., 1995). The addition of functional amine groups allows for coupling under physiological conditions. Local delivery of rhBMP-2 is achieved by physical or chemical incorporation into the porous hyaluronic acid hydrogels, yielding sustained release (Bulpitt and Aeschlimann, 1999). Interestingly, the presence

268

Cowan et al.

of collagen was required for the bioactivity of rhBMP-2. They released low levels of rhBMP-2 in a sustained manner (one third of initially loaded rhBMP-2 over 28 days) and stimulated pluripotent stem cell diVerentiation into the osteoblast lineage in vitro and bone formation in vivo (Hunt et al., 2001). The need for reliable and clinically relevant delivery vehicles for BMPs has initiated a surge of research investigating both BMP and scaVold biochemistry. The clinical success of current BMP therapies will undoubtedly motivate further research in the elucidation of the ideal binding strategy. The success thus far has been suYcient for the transition into clinical trials and promises great success in the near future, when scaVold–BMP interactions are better understood.

H. Translating BMP-2 and BMP-7 Research to Clinical Trials Pivotal preclinical studies designed to demonstrate eYcacy involve a transition from the use of small animals to large animals to human studies. Preclinical studies typically evaluate a dose profile, which is significantly dependent on choosing the appropriate endpoint measures (Goldstein, 2002). The animal model chosen should be evaluated for its ability to simulate human bone chemistry, metabolic turnover, and morphology. Rationale for moving from small to large animals involves recognition that the physiology of rat bone diVers from human bone in its nutritional makeup and its remodeling capacity. This knowledge begs the question of how much clinical relevance can be interpreted from animal studies. A largeanimal (canine) model can be used to investigate tibia defect healing that is very diYcult to heal, therefore providing a robust test for new technology. For spinal fusion, sheep are used because their spines are similar to those of humans in size and morphology. Required BMP concentrations depend on the state of the organism in the evolutionary scale, with highly evolved animals requiring more BMP for the fusion of comparable defects induced in lower mammals, and the type of defect with simple fractures where bone fronts are close, not needing as much BMP (Rengachary, 2002). Only four clinical studies have been performed: (1) maxillary sinus floor augmentation (Boyne et al., 1997), (2) alveolar ridge preservation (Howell et al., 1997), (3) fibular defect (Geesink et al., 1999), and (4) spinal fusion (Boden et al., 2002), and superphysiological doses of rhBMP-2 and rhBMP7 were needed to induce an eVect in each case (Boyne et al., 1997; Geesink et al., 1999; Howell et al., 1997). Although BMP-2 and BMP-7 seem to be excellent choices for bone tissue engineering stimulation, the transition from an animal model to humans has been only mildly successful. Advances in delivery systems should greatly improve clinical outcomes and eventually

8. Evolving Concepts in Bone Tissue Engineering

269

allow for the development of ‘‘oV-the-shelf’’ devices for clinical application in osseous repair. Currently, two devices have been approved by the FDA for clinical orthopedic applications. One using rhBMP-7/rhOP-1 (Osigraft, Howmedica International S. de R. L., Raheen, Limerick, Ireland) and another using rhBMP-2 (INFUSE Bone Graft and LT-CAGE Lumbar Tapered Fusion Device, Medtronic Sofamor Danek, Memphis, TN). Osigraft has been approved for the nonunion treatment of tibia of at least 9 months’ duration, secondary to trauma, in skeletally mature patients (Biotech, 1999). This device is to be used in cases where previous treatment with autograft has failed or use of autograft is unfeasible. Osigraft combines the active ingredient rhBMP-7/ rhOP-1 with a bovine type I collagen matrix for in vivo bone formation. Preliminary studies in rats and nonhuman primates demonstrated that 25% of rhBMP-7/rhOP-1 was released within 3 h. Preclinical subcutaneous implantation studies in rats demonstrated that within 24 h only 20–25% of the rhBMP-7/rhOP-1 remained in the product and only 6% remained after 6 days, with no substantial or prolonged tissue or organ uptake. Long-term toxicology studies (28 days) in rats with injected rhBMP-7/rhOP-1 using doses of 0.035, 0.35, and 3.5 mg/kg/day (0.23–23 times that used in humans) demonstrated dose-dependent localized eVects, including reduced ovary weight, reduced thymus weight, and increased adrenal weight with no changes in the kidney. Interestingly, the main finding was inflammation, ossification, and necrosis at the injection site, but not in other tissues. The InFUSE Bone Graft/LT-CAGE Lumbar Tapered Fusion Device is approved for spinal fusion procedures in skeletally mature patients with degenerative disc disease (DDD) at one level from L4–S1 (Medtronic Sofamor Danek, 2002). For this device, rhBMP-2 is soaked onto an ACS and placed within a metal device before implantation. Preclinical studies in rats and dogs demonstrated no related toxicities, as well as an absence of remote site bone formation with suprapharmacological doses. Injected rhBMP-2 cleared rapidly from systemic circulation, with none remaining after 2 h. Subcutaneous rhBMP-2 implanted on the ACS resulted in slow release from implant with a mean residence time of 8 days and only 0.1% of the implanted dose found within the blood. Nonhuman primates and humans required doses ranging from 0.4 to 1.5 mg/ml rhBMP-2, with increasing doses and length of time at implant site resulting in more bone formation. Because such high doses of BMP-2 and BMP-7 are necessary for significant bone formation clinically, a variety of concerns have arisen. For example, the promotion of uncontrolled bone growth around the implantation site or even cancer would make them unsafe for patients with a history of cancer. This concern is legitimate based on reports linking osteosarcomas with BMP activity (Ravel et al., 1996; Yoshikawa et al., 1994); however, no evidence supports BMP’s role as an oncogenic factor. Instead, BMPs

270

Cowan et al.

play a greater role in recruitment and diVerentiation rather than in promoting malignancy. At times, BMP-2 has also been shown to elicit undesirable heterotropic bone formation away from sites of administration (Valentin-Opran et al., 2002). Second, although the BMPs are extremely conserved, the possibility of a large immune/autoimmune reaction is still questionable. Both InFUSE Bone Graft/LT-CAGE Lumbar Tapered Fusion Device and Osigraft initiate a positive antibody response to rhBMP-2 and rhBMP-7/rhOP-1, respectively, as well as bovine type I collagen in both treatments (Biotech, 1999; Medtronic Sofamor Danek, 2002). Some studies have even described high blood pressure and even heart attacks in animal models, making this strategy bad for sensitive humans. Several factors led to the need for superphysiological doses of BMPs. First, cytokines produced ex vivo generally are not associated with binding proteins, possibly resulting in errors in protein assembly or a loss of functional activity. Second is the insuYcient delivery of growth factors, relating to the limited knowledge of BMP–scaVold interactions (Centrella et al., 1994; Cook et al., 1994). The use of viral delivery can generate large quantities of BMPs; however, in vivo complications have hampered their use (Lee et al., 2001; Peng et al., 2002; Wright et al., 2002). Additionally, suboptimal BMP protein–peptide binding and immobilization has challenged the ability to delivery substantial quantities, which is coupled by the fact that recombinant cytokines are quickly turned over (30 min) (Giannobile, 1996) in the harsh environment of a fracture site undergoing repair and thus are degraded before cells are stimulated by them (Ham, 1930; Urist, 1965). Each of these observations adds to the limited success of clinical therapies thus far. Future studies will need to elucidate whether BMPs are required throughout the bone regeneration and what the optimal doses are.

IV. Conclusions and Future Directions The field of regenerative medicine is constantly evolving and benefits from advances in cross-disciplinary specialties. From the mouse with a human ear on its back (Uludag et al., 1999a) to the current human mandibular prefabrication study (Warnke et al., 2004), widespread interest in tissue engineering has grown. Bone tissue engineering specifically benefits from further knowledge in the fields of bone biology, protein chemistry, material science, organ transplantation, cell-to-surface interfaces, and more. Advances in tissue engineering techniques using sophisticated biocompatible scaVolds, osteoprogenitor cell populations, and optimal cellular stimulation have multifaceted applications in regenerative medicine. These criteria relate to the immediate goal of determining the ideal implant. The search is becoming a reality with so many biocompatible scaVolds

8. Evolving Concepts in Bone Tissue Engineering

271

available; however, the desired parameters have not been clearly defined. Currently, most research has focused on natural and synthetic polymers because of their availability, reproducibility, and ease of handling and sculpting. Furthermore, research on the microenvironment of osteogenic coatings will greatly aid this field and reveal parameters necessary for the transition to human clinical studies. Apatites seem to be a natural choice because they are formed along with bone in vivo and have demonstrated encouraging results in animal studies (Chou et al., 2004). The addition of autologous osteogenic cells is advantageous for bone formation, although time consuming for clinical practice. The ability to easily harvest and rapidly accumulate large numbers of multipotent cells from adipose tissue for autologous transplantation has launched research within this specialty. Finally, growth factor stimulation is necessity, although the optimal cocktail and dose still need to be determined. BMP-2 and BMP-7 are popular choices and have led to very exciting results in both animal and clinical studies. Further analysis of BMP pharmacokinetics is necessary for advancements within this field. The discovery of additional osteoinductive growth factors, such as Nell-1 (a protein strongly expressed in neural tissues and containing epidermal growth factor (EGF)-like domains, type 1), represents an exciting alternative to BMP-based strategies. Nell-1 accelerates osteogenic diVerentiation in vitro, bone formation in vivo (Ting et al., 1999; Zhang et al., 2002, 2003), and has shown recent successes in rat calvarial healing models (unpublished data). Alternative stimulants include the use of oxycholesterol or BMP-2 oligopeptides (Chatzinikolaidou et al., 2003). This 20-amino acid-long sequence contained within a hydrogel enhances cell attachment in vitro and ectopic bone formation in vivo (Suzuki et al., 2000). Unfortunately, the same challenges facing BMPs may potentially face other growth factors or stimulants. The downside to current strategies is the variability between research projects with regard to BMP source, dosage, carrier, and animal models. The search continues for the most appropriate osteogenic cell type(s) and delivery method(s) that are speedy, safe, and clinically relevant. Thus, the optimal bone tissue engineering strategy is still evolving. Although a combination of scaVolds, cells, and growth factors provides for a microenvironment that leads to successful defect healing, this practice may diminish clinically, when time in the surgery room becomes of greater importance in treating a patient. Surgeons would prefer an implant that does not require additional preparation, such as harvesting and seeding multipotent cells onto scaVolds. If precursor cells are not implanted, then the scaVold itself must be responsible for inducing cellular invasion and bone formation. In conclusion, advances in osteogenic implants and bone tissue engineering strategies have already begun to pave the way for future human treatments and show promising results for the future of bone tissue engineering.

272

Cowan et al.

References Aalami, O. O., Nacamuli, R. P., Lenton, K. A., Cowan, C. M., Fang, T. D., Fong, K. D., Shi, Y. Y., Song, H. M., Sahar, D. E., and Longaker, M. T. (2004). Applications of a mouse model of calvarial healing: DiVerences in regenerative abilities of juveniles and adults. Plast. Reconstr. Surg. 114, 713–720. Abu-Serriah, M., Ayoub, A., Boyd, J., Paterson, C., and Wray, D. (2003). The role of ultrasound in monitoring reconstruction of mandibular continuity defects using osteogenic protein-1 (rhOP-1). Int. J. Oral Maxillofac. Surg. 32, 619–627. Ambrose, C. G., Gogola, G. R., Clyburn, T. A., Raymond, A. K., Peng, A. S., and Mikos, A. G. (2003). Antibiotic microspheres: Preliminary testing for potential treatment of osteomyelitis. Clin. Orthop. 1, 279–285. Anderson, H. C., Hodges, P. T., Aguilera, X. M., Missana, L., and Moylan, P. E. (2000). Bone morphogenetic protein (BMP) localization in developing human and rat growth plate, metaphysis, epiphysis, and articular cartilage. J. Histochem. Cytochem. 48, 1493–1502. Aono, A., Hazama, M., Notoya, K., Taketomi, S., Yamasaki, H., Tsukuda, R., Sasaki, S., and Fujisawa, Y. (1995). Potent ectopic bone-inducing activity of bone morphogenetic protein-4/7 heterodimer. Biochem. Biophys. Res. Commun. 210, 670–677. Asahina, I., Sampath, T. K., Nishimura, I., and Hauschka, P. V. (1993). Human osteogenic protein-1 induces both chondroblastic and osteoblastic diVerentiation of osteoprogenitor cells derived from newborn rat calvaria. J. Cell Biol. 123, 921–933. Ashjian, P. H., Elbarbary, A. S., Edmonds, B., DeUgarte, D., Zhu, M., Zuk, P. A., Lorenz, H. P., Benhaim, P., and Hedrick, M. H. (2003). In vitro diVerentiation of human processed lipoaspirate cells into early neural progenitors. Plast. Reconstr. Surg. 111, 1922–1931. Baltzer, A. W., and Lieberman, J. R. (2004). Regional gene therapy to enhance bone repair. Gene Ther. 11, 344–350. Bancroft, G. N., Sikavitsas, V. I., van den Dolder, J., SheYeld, T. L., Ambrose, C. G., Jansen, J. A., and Mikos, A. G. (2002). Fluid flow increases mineralized matrix deposition in 3D perfusion culture of marrow stromal osteoblasts in a dose-dependent manner. Proc. Natl. Acad. Sci. USA 99, 12600–12605. Bauer, T. W., and Muschler, G. F. (2000). Bone graft materials. An overview of the basic science. Clin. Orthop. 1, 10–27. Baur, S. T., Mai, J. J., and Dymecki, S. M. (2000). Combinatorial signaling through BMP receptor IB and GDF5: Shaping of the distal mouse limb and the genetics of distal limb diversity. Development 127, 605–619. Biotech, S. (1999). Osigraft Scientific Discussion. http://www.emea.eu.int/humandocs/PDFs/EPAR/osigraft/039301en6.pdf. Blum, J. S., Barry, M. A., Mikos, A. G., and Jansen, J. A. (2003). In vivo evaluation of gene therapy vectors in ex vivo-derived marrow stromal cells for bone regeneration in a rat critical-size calvarial defect model. Hum. Gene Ther. 14, 1689–1701. Boden, S. D., Kang, J., Sandhu, H., and Heller, J. G. (2002). Use of recombinant human bone morphogenetic protein-2 to achieve posterolateral lumbar spine fusion in humans: A prospective, randomized clinical pilot trial: 2002 Volvo Award in clinical studies. Spine 27, 2662–2673. Boden, S. D., McCuaig, K., Hair, G., Racine, M., Titus, L., Wozney, J. M., and Nanes, M. S. (1996). DiVerential eVects and glucocorticoid potentiation of bone morphogenetic protein action during rat osteoblast diVerentiation in vitro. Endocrinology 137, 3401–3407. Boden, S. D., Schimandle, J. H., and Hutton, W. C. (1995). Lumbar intertransverse-process spinal arthrodesis with use of a bovine bone-derived osteoinductive protein. A preliminary report. J. Bone Joint Surg. Am. 77, 1404–1417.

8. Evolving Concepts in Bone Tissue Engineering

273

Bonadio, J. (2000). Tissue engineering via local gene delivery: Update and future prospects for enhancing the technology. Adv. Drug Deliv. Rev. 44(2–3), 185–194. Bonadio, J., Smiley, E., Patil, P., and Goldstein, S. (1999). Localized, direct plasmid gene delivery in vivo: Prolonged therapy results in reproducible tissue regeneration. Nat. Med. 5, 753–759. Bosch, P., Musgrave, D. S., Lee, J. Y., Cummins, J., Shuler, T., Ghivizzani, T. C., Evans, T., Robbins, T. D., and Huard, J. (2000). Osteoprogenitor cells within skeletal muscle. J. Orthop. Res. 18, 933–944. Bostrom, M. P. (1998). Expression of bone morphogenetic proteins in fracture healing. Clin. Orthop. 1, S116–S123. Bostrom, M. P., Lane, J. M., Berberian, W. S., Missri, A. A., Tomin, E., Weiland, A., Doty, S. B., Glaser, D., and Rosen, V. M. (1995). Immunolocalization and expression of bone morphogenetic proteins 2 and 4 in fracture healing. J. Orthop. Res. 13, 357–367. Bostrom, R., Mikos, A., Eds. (1997). Tissue Engineering of Bone, Vol. 1. Boston, Birkhauser. Bourgeois, B., Laboux, O., Obadia, L., Gauthier, O., Betti, E., Aguado, E., Daculsi, G., and Bouler, J. M. (2003). Calcium-deficient apatite: A first in vivo study concerning bone ingrowth. J. Biomed. Mater. Res. 65A, 402–408. Bottaro, D. P., Liebmann-Vinson, A., and Heidaran, M. A. (2002). Molecular signaling in bioengineered tissue microenvironments. Ann. NY Acad. Sci. 961, 143–153. Bouletreau, P. J., Warren, S. M., Spector, J. A., Peled, Z. M., Gerrets, R. P., Greenwald, J. A., and Longaker, M. T. (2002). Hypoxia and VEGF up-regulate BMP-2 mRNA and protein expression in microvascular endothelial cells: Implications for fracture healing. Plast. Reconstr. Surg. 109, 2384–2397. Boyne, P. J., Marx, R. E., Nevins, M., Triplett, G., Lazaro, E., Lilly, L. C., Alder, M., and Nummikoski, P. (1997). A feasibility study evaluating rhBMP-2/absorbable collagen sponge for maxillary sinus floor augmentation. Int. J. Periodontics Restorative Dent. 17, 11–25. Bragdon, C. R., Doherty, A. M., Rubash, H. E., Jasty, M., Li, X. J., Seeherman, H., and Harris, W. H. (2003). The eYcacy of BMP-2 to induce bone ingrowth in a total hip replacement model. Clin. Orthop. 1, 50–61. Bruder, S. P., Jaiswal, N., and Haynesworth, S. E. (1997). Growth kinetics, self-renewal, and the osteogenic potential of purified human mesenchymal stem cells during extensive subcultivation and following cryopreservation. J. Cell. Biochem. 64, 278–294. Bulpitt, P., and Aeschlimann, D. (1999). New strategy for chemical modification of hyaluronic acid: Preparation of functionalized derivatives and their use in the formation of novel biocompatible hydrogels. J. Biomed. Mater. Res. 47, 152–169. Buttery, L. D., Bourne, S., Xynos, J. D., Wood, H., Hughes, F. J., Hughes, S. P., Episkopou, V., and Polak, J. M. (2001). DiVerentiation of osteoblasts and in vitro bone formation from murine embryonic stem cells. Tissue Eng. 7, 89–99. Centrella, M., Horowitz, M. C., Wozney, J. M., and McCarthy, T. L. (1994). Transforming growth factor-beta gene family members and bone. Endocr. Rev. 15, 27–39. Chang, S. C., Chuang, H., Chen, Y. R., Yang, L. C., Chen, J. K., Mardini, S., Chung, H. Y., Lu, Y. L., Ma, W. C., Lou, J., Mardini, S., and Mardinis, S. (2004). Cranial repair using BMP-2 gene engineered bone marrow stromal cells. J. Surg. Res. 119, 85–91. Chang, S. C., Chuang, H. L., Chen, Y. R., Chen, J. K., Chung, H. Y., Lu, Y. L., Lin, H. Y., Tai, C. L., and Lou, J. (2003). Ex vivo gene therapy in autologous bone marrow stromal stem cells for tissue-engineered maxillofacial bone regeneration. Gene Ther. 10, 2013–2019. Chatzinikolaidou, M., Zumbrink, T., et al. (2003). Stability of surface-enhanced ultrahydrophilic metals as a basis for bioactive rhBMP-2 surfaces. Mat.-wiss u. WerkstoVtech. 34, 1106–1112. Chen, G., Ushida, T., and Tateishi, T. (2001). Poly(dl-lactic-co-glycolic acid) sponge hybridized with collagen microsponges and deposited apatite particulates. J. Biomed. Mater. Res. 57, 8–14.

274

Cowan et al.

Cheng, H., Jiang, W., Phillips, F. M., Haydon, R. C., Peng, Y., Zhou, L., Luu, H. H., An, N., Breyer, B., Vanichakarn, P., Szatkowski, J. P., Park, J. Y., and He, T. C. (2003). Osteogenic activity of the fourteen types of human bone morphogenetic proteins (BMPs). J. Bone Joint Surg. Am. 85-A, 1544–1552. Chou, Y. F., Chiou, W. A., Xu, Y., Dunn, J. C., and Wu, B. M. (2004). The eVect of pH on the structural evolution of accelerated biomimetic apatite. Biomaterials 25, 5323–5331. Chvapil, M. (1977). Collagen sponge: Theory and practice of medical applications. J. Biomed. Mater. Res. 11, 721–741. Clokie, C. M., Moghadam, H., Jackson, M. T., and Sandor, G. K. (2002). Closure of critical sized defects with allogenic and alloplastic bone substitutes. J. Craniofac. Surg. 13, 111–121; discussion 122–123. Cook, S. D., BaVes, G. C., Wolfe, M. W., Sampath, T. K., Rueger, D. C., and Whitecloud, T. S., III. (1994). The eVect of recombinant human osteogenic protein-1 on healing of large segmental bone defects. J. Bone Joint Surg. Am. 76, 827–838. Cook, S. D., Wolfe, M. W., Salkeld, S. L., and Rueger, D. C. (1995). EVect of recombinant human osteogenic protein-1 on healing of segmental defects in non-human primates. J. Bone Joint Surg. Am. 77, 734–750. Cowan, C. M., Aalami, O., et al. (2005). Bone morphogenetic protein-2 and retinoic acid accelerate in vivo bone formation, osteoclast recruitment, and bone turnover. Tissue Eng. under review. Cowan, C. M., Shi, Y. Y., Aalami, O. O., Chou, Y. F., Mari, C., Thomas, R., Quarto, N., Contag, C. H., Wu, B., and Longaker, M. T. (2004). Adipose-derived adult stromal cells heal critical-sized mouse calvarial defects. Nat. Biotechnol. 22, 560–567. Cowan, C. M., Quarto, N., Warren, S. M., Salim, A., and Longaker, M. T. (2003). Age-related changes in the biomolecular mechanisms of calvarial osteoblast biology aVect fibroblast growth factor-2 signaling and osteogenesis. J. Biol. Chem. 278, 32005–32013. Cunningham, B. W., Shimamoto, N., Sefter, J. C., Dmitriev, A. E., Orbegoso, C. M., McCarthy, E. F., Fedder, I. L., and McAfee, P. C. (2002). Osseointegration of autograft versus osteogenic protein-1 in posterolateral spinal arthrodesis: Emphasis on the comparative mechanisms of bone induction. Spine J. 2, 11–24. de Jong, D. S., Vaes, B. L., Dechering, K. J., Feijen, A., Hendriks, J. M., Wehrens, R., Mummery, C. L., van Zoelen, E. J., Olijve, W., and Steegenga, W. T. (2004). Identification of novel regulators associated with early-phase osteoblast diVerentiation. J. Bone Miner. Res. 19, 947–958. De Ugarte, D. A., Morizono, K., Elbarbary, A., Alfonso, Z., Zuk, P. A., Zhu, M., Dragoo, J. L., Ashjian, P., Thomas, B., Benhaim, P., Chen, I., Fraser, J., and Hedrick, M. H. (2003). Comparison of multi-lineage cells from human adipose tissue and bone marrow. Cells Tissues Organs 174, 101–109. Dean, D., Topham, N. S., Meneghetti, S. C., Wolfe, M. S., Jepsen, K., He, S., Chen, J. E., Fisher, J. P., Cooke, M., Rimnac, C., and Mikos, A. G. (2003). Poly(propylene fumarate) and poly(dl-lactic-co-glycolic acid) as scaVold materials for solid and foamcoated composite tissue-engineered constructs for cranial reconstruction. Tissue Eng. 9, 495–504. den Boer, F. C., Wippermann, B. W., Blokhuis, T. J., Patka, P., Bakker, F. C., and Haarman, H. J. (2003). Healing of segmental bone defects with granular porous hydroxyapatite augmented with recombinant human osteogenic protein-1 or autologous bone marrow. J. Orthop. Res. 21, 521–528. Derubeis, A. R., and Cancedda, R. (2004). Bone marrow stromal cells (BMSCs) in bone engineering: Limitations and recent advances. Ann. Biomed. Eng. 32, 160–165. Dragoo, J. L., Choi, J. Y., Lieberman, J. R., Huang, J., Zuk, P. A., Zhang, J., Hedrick, M. H., and Benhaim, P. (2003). Bone induction by BMP-2 transduced stem cells derived from human fat. J. Orthop. Res. 21, 622–629.

8. Evolving Concepts in Bone Tissue Engineering

275

Dupoirieux, L., Costes, V., Jammet, P., and Souyris, F. (1994). Experimental study on demineralized bone matrix (DBM) and coral as bone graft substitutes in maxillofacial surgery. Int. J. Oral Maxillofac. Surg. 23, 395–398. Egrise, D., Martin, D., Vienn, A., Neve, P., and Schoutens, A. (1992). The number of fibroblastic colonies formed from bone marrow is decreased and the in vitro proliferation rate of trabecular bone cells increased in aged rats. Bone 13, 355–361. Einhorn, T. A. (2003). Clinical applications of recombinant human BMPs: Early experience and future development. J. Bone Joint Surg. Am. 85-A(Suppl. 3), 82–88. Elshahat, A., Shermak, M. A., Inoue, N., Chao, E. Y., and Manson, P. (2004). The use of Novabone and Norian in cranioplasty: A comparative study. J. Craniofac. Surg. 15, 483–489. Enneking, W. F., and Mindell, E. R. (1991). Observations on massive retrieved human allografts. J. Bone Joint Surg. Am. 73, 1123–1142. Enomoto, H., Enomoto-Iwamoto, M., Iwamoto, M., Nomura, S., Himeno, M., Kitamura, Y., Kishimoto, T., and Komori, T. (2000). Cbfa1 is a positive regulatory factor in chondrocyte maturation. J. Biol. Chem. 275, 8695–8702. Enomoto-Iwamoto, M., Nakamura, T., Aikawa, T., Higuchi, Y., Yuasa, T., Yamaguchi, A., Nohno, T., Noji, S., Matsuya, T., Kurisu, K., Koyama, E., Pacifici, M., and Iwamoto, M. (2000). Hedgehog proteins stimulate chondrogenic cell diVerentiation and cartilage formation. J. Bone Miner. Res. 15, 1659–1668. Erickson, G. R., Gimble, J. M., Franklin, D. M., Rice, H. E., Awad, H., and Guilak, F. (2002). Chondrogenic potential of adipose tissue-derived stromal cells in vitro and in vivo. Biochem. Biophys. Res. Commun. 290, 763–769. Fang, J., Zhu, Y. Y., Smiley, E., Bonadio, J., Rouleau, J. P., Goldstein, S. A., McCauley, L. K., Davidson, B. L., and Roessler, B. J. (1996). Stimulation of new bone formation by direct transfer of osteogenic plasmid genes. Proc. Natl. Acad. Sci. USA 93, 5753–5758. Ferguson, C., Alpern, E., Miclau, T., and Helms, J. A. (1999). Does adult fracture repair recapitulate embryonic skeletal formation? Mech. Dev. 87, 57–66. Fleming, J. E., Jr., Cornell, C. N., and Muschler, G. F. (2000). Bone cells and matrices in orthopedic tissue engineering. Orthop. Clin. North Am. 31, 357–374. Fong, K. D., Nacamuli, R. P., Song, H. M., Warren, S. M., Lorenz, H. P., and Longaker, M. T. (2003). New strategies for craniofacial repair and replacement: A brief review. J. Craniofac. Surg. 14, 333–339. Friedlaender, G. E. (1983). Immune responses to osteochondral allografts. Current knowledge and future directions. Clin. Orthop. 1, 58–68. Friess, W., Uludag, H., Foskett, S., Biron, R., and Sargeant, C. (1999). Characterization of absorbable collagen sponges as recombinant human bone morphogenetic protein-2 carriers. Int. J. Pharm. 185, 51–60. Gallea, S., Lallemand, F., Atfi, A., Rawadi, G., Ramez, V., Spinella-Jaegle, S., Kawai, S., Faucheu, C., Huet, L., Baron, R., and Roman-Roman, S. (2001). Activation of mitogenactivated protein kinase cascades is involved in regulation of bone morphogenetic protein-2induced osteoblast diVerentiation in pluripotent C2C12 cells. Bone 28, 491–498. Gamradt, S. C., and Lieberman, J. R. (2004). Genetic modification of stem cells to enhance bone repair. Ann. Biomed. Eng. 32, 136–147. Geckeler, K., Wacker, R., Martini, F., Hack, A., and Aicher, W. (2003). Enhanced biocompatibility for SAOS-2 osteosarcoma cells by surface coating with hydrophobic epoxy resins. Cell Physiol. Biochem. 13, 155–164. Geesink, R. G., Hoefnagels, N. H., and Bulstra, S. K. (1999). Osteogenic activity of OP-1 bone morphogenetic protein (BMP-7) in a human fibular defect. J. Bone Joint Surg. Br. 81, 710–718. Geiger, M., Li, R. H., and Friess, W. (2003). Collagen sponges for bone regeneration with rhBMP-2. Adv. Drug Deliv. Rev. 55, 1613–1629.

276

Cowan et al.

Gerhart, T. N., Kirker-Head, C. A., Kriz, M. J., Holtrop, M. E., Hennig, G. E., Hipp, J., Schelling, S. H., and Wang, E. (1993). Healing segmental femoral defects in sheep using recombinant human bone morphogenetic protein. Clin. Orthop. 1, 317–326. Giannobile, W. V. (1996). Periodontal tissue engineering by growth factors. Bone 19, 23S–37S. Gittens, S. A., Bagnall, K., Matyas, J. R., Lobenberg, R., and Uludag, H. (2004). Imparting bone mineral aYnity to osteogenic proteins through heparin-bisphosphonate conjugates. J. Control Release 98, 255–268. Goldstein, J. A. (2001). The use of bioresorbable material in craniofacial surgery. Clin. Plast. Surg. 28, 653–659. Goldstein, S. A. (2002). Tissue engineering: Functional assessment and clinical outcome. Ann. NY Acad. Sci. 961, 183–192. Gregory, C. F. (1972). The current status of bone and joint transplants. Clin. Orthop. 87, 165–166. Gu, K., Zhang, L., Jin, T., and Rutherford, R. B. (2004). Identification of potential modifiers of Runx2/Cbfa1 activity in C2C12 cells in response to bone morphogenetic protein-7. Cells Tissues Organs 176, 28–40. Gundle, R., Joyner, C. J., and TriYtt, J. T. (1995). Human bone tissue formation in diVusion chamber culture in vivo by bone-derived cells and marrow stromal fibroblastic cells. Bone 16, 597–601. Gysin, R., Wergedal, J. E., Sheng, M. H., Kasukawa, Y., Miyakoshi, N., Chen, S. T., Peng, H., Lau, K. H., Mohan, S., and Baylink, D. J. (2002). Ex vivo gene therapy with stromal cells transduced with a retroviral vector containing the BMP4 gene completely heals critical size calvarial defect in rats. Gene Ther. 9, 991–999. Halvorsen, Y. D., Bond, A., Sen, A., Franklin, D. M., Lea-Currie, Y. R., Sujkowski, D., Ellis, P. N., Wilkison, W. O., and Gimble, J. M. (2001). Thiazolidinediones and glucocorticoids synergistically induce diVerentiation of human adipose tissue stromal cells: Biochemical, cellular, and molecular analysis. Metabolism 50, 407–413. Halvorsen, Y. D., Franklin, D., Bond, A. L, Hitt, D. C., Auchter, C., Boskey, A. L., Paschalis, E. P., Wilkison, W. O., and Gimble, J. M. (2001). Extracellular matrix mineralization and osteoblast gene expression by human adipose tissue-derived stromal cells. Tissue Eng. 7, 729–741. Ham, A. (1930). A histological study of the early phases of bone repair. J. Bone Joint Surg. Am. 12, 827–844. Hanada, K., Dennis, J. E., and Caplan, A. I. (1997). Stimulatory eVects of basic fibroblast growth factor and bone morphogenetic protein-2 on osteogenic diVerentiation of rat bone marrow-derived mesenchymal stem cells. J. Bone Miner. Res. 12, 1606–1614. Harada, S., and Rodan, G. A. (2003). Control of osteoblast function and regulation of bone mass. Nature 423, 349–355. Harland, R. M. (1994). The transforming growth factor beta family and induction of the vertebrate mesoderm: Bone morphogenetic proteins are ventral inducers. Proc. Natl. Acad. Sci. USA 91, 10243–10246. Haynesworth, S. E., Goshima, J., Goldberg, V. M., and Caplan, A. I. (1992). Characterization of cells with osteogenic potential from human marrow. Bone 13, 81–88. Helvering, L. M., Sharp, R. L., Ou, X., and Geiser, A. G. (2000). Regulation of the promoters for the human bone morphogenetic protein 2 and 4 genes. Gene 256, 123–138. Hogan, B. L. (1996). Bone morphogenetic proteins: multifunctional regulators of vertebrate development. Genes Dev. 10, 1580–1594. Holleville, N., Quilhac, A., Bontoux, M., and Monsoro-Burq, A. H. (2003). BMP signals regulate Dlx5 during early avian skull development. Dev. Biol. 257, 177–189. Hollinger, J. O., and Battistone, G. C. (1986). Biodegradable bone repair materials. Synthetic polymers and ceramics. Clin Orthop. 1, 290–305.

8. Evolving Concepts in Bone Tissue Engineering

277

Hollinger, J. O., and Kleinschmidt, J. C. (1990). The critical size defect as an experimental model to test bone repair materials. J. Craniofac. Surg. 1, 60–68. Hollinger, J. O., Uludag, H., and Winn, S. R. (1998). Sustained release emphasizing recombinant human bone morphogenetic protein-2. Adv. Drug Deliv. Rev. 31, 303–318. Hollinger, J. O., Winn, S., and Bonadio, J. (2000). Options for tissue engineering to address challenges of the aging skeleton. Tissue Eng. 6, 341–350. Howell, T. H., Fiorellini, J., Jones, A., Alder, M., Nummikoski, P., Lazaro, M., Lilly, L., and Cochran, D. (1997). A feasibility study evaluating rhBMP-2/absorbable collagen sponge device for local alveolar ridge preservation or augmentation. Int. J. Periodontics Restorative Dent. 17, 124–139. Hubbell, J. A., Pathak, C. P., et al. (1995). Photopolymerizable biodegradable hydrogels as tissue contacting materials and controlled-release carriers. US patent 5,410,016. Hughes, F. J., Collyer, J., Stanfield, M., and Goodman, S. A. (1995). The eVects of bone morphogenetic protein-2, -4, and -6 on diVerentiation of rat osteoblast cells in vitro. Endocrinology 136, 2671–2677. Hong, L., Tabata, Y., Miyamoto, S., Yamamoto, M., Yamada, K., Hashimoto, N., and Ikada, Y. (2000). Bone regeneration at rabbit skull defects treated with transforming growth factorbeta1 incorporated into hydrogels with diVerent levels of biodegradability. J. Neurosurg. 92, 315–325. Hunt, D. R., Jovanovic, S. A., Wikesjo, U. M., Wozney, J. M., and Bernard, G. W. (2001). Hyaluronan supports recombinant human bone morphogenetic protein-2 induced bone reconstruction of advanced alveolar ridge defects in dogs. A pilot study. J. Periodontol. 72, 651–658. Im, G. I., Kim, D. Y., Shin, J. H., Hyun, C. W., and Cho, W. H. (2001). Repair of cartilage defect in the rabbit with cultured mesenchymal stem cells from bone marrow. J. Bone Joint Surg. Br. 83, 289–294. Israel, D. I., Nove, J., Kerns, K. M., Kaufman, R. J., Rosen, V., Cox, K. A., and Wozney, J. M. (1996). Heterodimeric bone morphogenetic proteins show enhanced activity in vitro and in vivo. Growth Factors 13, 291–300. Jazrawi, L. M., Majeska, R. J., Klein, M. L., Kagel, E., Stromberg, L., and Einhorn, T. A. (1998). Bone and cartilage formation in an experimental model of distraction osteogenesis. J. Orthop. Trauma 12, 111–116. Johnstone, B., Hering, T. M., Caplan, A. I., Goldberg, V. M., and Yoo, J. U. (1998). In vitro chondrogenesis of bone marrow-derived mesenchymal progenitor cells. Exp. Cell Res. 238, 265–272. Kang, Q., Sun, M. H., Cheng, H., Peng, Y., Montag, A. G., Deyrup, A. T., Jiang, W., Luu, H. H., Luo, J., Szatkowski, J. P., Vanichakarn, P., Park, J. Y., Li, Y., Haydon, R. C., and He, T. C. (2004). Characterization of the distinct orthotopic bone-forming activity of 14 BMPs using recombinant adenovirus-mediated gene delivery. Gene Ther. 11, 1312–1320. Kanzler, B., Foreman, R. K., Labosky, P. A., and Mallo, M. (2000). BMP signaling is essential for development of skeletogenic and neurogenic cranial neural crest. Development 127, 1095–1104. Kawabata, M., Imamura, T., and Miyazono, K. (1998). Signal transduction by bone morphogenetic proteins. Cytokine Growth Factor Rev. 9, 49–61. Kenley, R., Marden, L., Turek, T., Jin, L., Ron, E., and Hollinger, J. O. (1994). Osseous regeneration in the rat calvarium using novel delivery systems for recombinant human bone morphogenetic protein-2 (rhBMP-2). J. Biomed. Mater. Res. 28, 1139–1147. Kenley, R. A., Yim, K., Abrams, J., Ron, E., Turek, T., Marden, L. J., and Hollinger, J. O. (1993). Biotechnology and bone graft substitutes. Pharm. Res. 10, 1393–1401. Kim, H. D., and Valentini, R. F. (2002). Retention and activity of BMP-2 in hyaluronic acidbased scaVolds in vitro. J. Biomed. Mater. Res. 59, 573–584.

278

Cowan et al.

Kirsch, T., Sebald, W., and Dreyer, M. K. (2000). Crystal structure of the BMP-2-BRIA ectodomain complex. Nat. Struct. Biol. 7, 492–496. Krebsbach, P. H., Gu, K., Franceschi, R. T., and Rutherford, R. B. (2000). Gene therapydirected osteogenesis: BMP-7-transduced human fibroblasts form bone in vivo. Hum. Gene Ther. 11, 1201–1210. Krebsbach, P. H., Mankani, M. H., Satomura, K., and Kuznetsov, S. A, and Robey, P. G. (1998). Repair of craniotomy defects using bone marrow stromal cells. Transplantation 66, 1272–1278. Kruyt, M. C., Dhert, W. J., Yuan, H., Wilson, C. E., van Blitterswijk, C. A., Verbout, A. J., and de Bruijn, J. D. (2004). Bone tissue engineering in a critical size defect compared to ectopic implantations in the goat. J. Orthop. Res. 22, 544–551. Langer, R., and Vacanti, J. P. (1993). Tissue engineering. Science 260, 920–926. Langille, R. M. (1994) DiVerentiation of craniofacial mesenchyme. In ‘‘DiVerentiation and Morphogenesis of Bone,’’ (B. K. Hall, Ed.) Vol. 9, pp. 1–63. Boca Raton, FL, CRC Press, Inc. Laurent, T. C., and Fraser, J. R. (1992). Hyaluronan. FASEB J. 6, 2397–2404. Le, A. X., Miclau, T., Hu, D., and Helms, J. A. (2001). Molecular aspects of healing in stabilized and non-stabilized fractures. J. Orthop. Res. 19, 78–84. Lee, J. A., Parrett, B. M., Conejero, J. A., Laser, J., Chen, J., Kogon, A. J., Nanda, D., Grant, R. T., and Breitbart, A. S. (2003). Biological alchemy: Engineering bone and fat from fatderived stem cells. Ann. Plast. Surg. 50, 610–617. Lee, J. Y., Musgrave, D., Pelinkovic, D., Fukushima, K., Cummins, J., Usas, A., Robbins, P., Fu, F. H., and Huard, J. (2001). EVect of bone morphogenetic protein-2-expressing musclederived cells on healing of critical-sized bone defects in mice. J. Bone Joint Surg. Am. 83-A, 1032–1039. Lee, J. Y., Peng, H., Usas, A., Musgrave, D., Cummins, J., Pelinkovic, D., Jankowski, R., Ziran, B., Robbins, P., and Huard, J. (2002). Enhancement of bone healing based on ex vivo gene therapy using human muscle-derived cells expressing bone morphogenetic protein 2. Hum. Gene Ther. 13, 1201–1211. Lengner, C. J., Lepper, C., van Wijnen, A. J., Stein, J. L., Stein, G. S., and Lian, J. B. (2004). Primary mouse embryonic fibroblasts: A model of mesenchymal cartilage formation. J. Cell Physiol. 200, 327–333. Levine, J. P., Bradley, J., Turk, A. E., Ricci, J. L., Benedict, J. J., Steiner, G., Longaker, M. T., and McCarthy, J. G. (1997). Bone morphogenetic protein promotes vascularization and osteoinduction in preformed hydroxyapatite in the rabbit. Ann. Plast. Surg. 39, 158–168. Li, R. H., and Wozney, J. M. (2001). Delivering on the promise of bone morphogenetic proteins. Trends Biotechnol. 19, 255–265. Liao, S. S., Guan, K., Cui, F. Z., Shi, S. S., and Sun, T. S. (2003). Lumbar spinal fusion with a mineralized collagen matrix and rhBMP-2 in a rabbit model. Spine 28, 1954–1960. Lieberman, J. R., Daluiski, A., Stevenson, S., Wu, L., McAllister, P., Lee, Y. P., Kabo, J. M., Finerman, G. A., Berk, A. J., and Witte, O. N. (1999). The eVect of regional gene therapy with bone morphogenetic protein-2-producing bone-marrow cells on the repair of segmental femoral defects in rats. J. Bone Joint Surg. Am. 81, 905–917. Lindsey, W. H. (2001). Osseous tissue engineering with gene therapy for facial bone reconstruction. Laryngoscope 111, 1128–1136. Liu, F., Ventura, F., Doody, J., and Massague, J. (1995). Human type II receptor for bone morphogenic proteins (BMPs), extension of the two-kinase receptor model to the BMPs. Mol. Cell. Biol. 15, 3479–3486. Lutolf, M. P., Lauer-Fields, J. L., Schmoekel, H. G., Metters, A. T., Weber, F. E., Fields, G. B., and Hubbell, J. A. (2003). Synthetic matrix metalloproteinase-sensitive hydrogels for the conduction of tissue regeneration: Engineering cell-invasion characteristics. Proc. Natl. Acad. Sci. USA. 100, 5413–5418.

8. Evolving Concepts in Bone Tissue Engineering

279

Lyons, K. M., Hogan, B. L., and Robertson, E. J. (1995). Colocalization of BMP 7 and BMP 2 RNAs suggests that these factors cooperatively mediate tissue interactions during murine development. Mech. Dev. 50, 71–83. Lyons, K. M., Pelton, R. W., and Hogan, B. L. (1989). Patterns of expression of murine Vgr-1 and BMP-2a RNA suggest that transforming growth factor-beta-like genes coordinately regulate aspects of embryonic development. Genes Dev. 3, 1657–1668. Lyons, K. M., Pelton, R. W., and Hogan, B. L. (1990). Organogenesis and pattern formation in the mouse: RNA distribution patterns suggest a role for bone morphogenetic protein-2A (BMP-2A). Development 109, 833–844. Maeda, H., Sano, A., and Fujioka, K. (2004a). Controlled release of rhBMP-2 from collagen minipellet and the relationship between release profile and ectopic bone formation. Int. J. Pharm. 275, 109–122. Maeda, S., Hayashi, M., Komiya, S., Imamura, T., and Miyazono, K. (2004b). Endogenous TGF-beta signaling suppresses maturation of osteoblastic mesenchymal cells. EMBO J. 23, 552–563. Malda, J., Martens, D. E., Tramper, J., van Blitterswijk, C. A., and Riesle, J. (2003). Cartilage tissue engineering: Controversy in the eVect of oxygen. Crit. Rev. Biotechnol. 23, 175–194. Marden, L. J., Hollinger, J. O., Chaudhari, A., Turek, T., Schaub, R. G., and Ron, E. (1994). Recombinant human bone morphogenetic protein-2 is superior to demineralized bone matrix in repairing craniotomy defects in rats. J. Biomed. Mater. Res. 28, 1127–1138. Marvaso, V., and Bernard, G. W. (1977). Initial intramembraneous osteogenesis in vitro. Am. J. Anat. 149, 453–468. Mayer, H., Scutt, A. M., and Ankenbauer, T. (1996a). Subtle diVerences in the mitogenic eVects of recombinant human bone morphogenetic proteins 2 to 7 on DNA synthesis on primary bone-forming cells and identification of BMP-2/4 receptor. Calcif. Tissue Int. 58, 249–255. Mayer, M., Hollinger, J., Ron, E., and Wozney, J. (1996b). Maxillary alveolar cleft repair in dogs using recombinant human bone morphogenetic protein-2 and a polymer carrier. Plast. Reconstr. Surg. 98, 247–259. Medtronic Sofamor Danek, I. (2002). InFUSETM Bone Graft/LT-CAGETM Lumbar Tapered Fusion Device. http://www.fda.gov/cdrh/pdf/P000058.html. Miki, T., Masaka, K., Imai, Y., and Enomoto, S. (2000). Experience with freeze-dried PGLA/ HA/rhBMP-2 as a bone graft substitute. J. Craniomaxillofac. Surg. 28, 294–299. Mishina, Y., Suzuki, A., Gilbert, D. J., Copeland, N. G., Jenkins, N. A., Ueno, N., and Behringer, R. R. (1995). Genomic organization and chromosomal location of the mouse type I BMP-2/4 receptor. Biochem. Biophys. Res. Commun. 206, 310–317. Mizuno, H., Zuk, P. A., Zhu, M., Lorenz, H. P., Benhaim, P., and Hedrick, M. H. (2002). Myogenic diVerentiation by human processed lipoaspirate cells. Plast. Reconstr. Surg. 109, 199–209; discussion 210–211. Moghadam, H. G., Sandor, G. K., Holmes, H. H., and Clokie, C. M. (2004). Histomorphometric evaluation of bone regeneration using allogeneic and alloplastic bone substitutes. J. Oral Maxillofac. Surg. 62, 202–213. Mulliken, J. B., and Glowacki, J. (1980). Induced osteogenesis for repair and construction in the craniofacial region. Plast. Reconstr. Surg. 65, 553–560. Mundy, G. R., Boyce, B., Hughes, D., Wright, K., Bonewald, L., Dallas, S., Harris, S., GhoshChoudhury, N., Chen, D., Dunstan, C., et al. (1995). The eVects of cytokines and growth factors on osteoblastic cells. Bone 17, 71S–75S. Muschler, G. F., Hyodo, A., Manning, T., Kambic, H., and Easley, K. (1994). Evaluation of human bone morphogenetic protein 2 in a canine spinal fusion model. Clin. Orthop. 1, 229–240. Nehrer, S., Breinan, H. A., Ramappa, A., ShortkroV, S., Young, G., Minas, T., Sledge, C. B., Yannas, I. V., and Spector, M. (1997). Canine chondrocytes seeded in type I and type II collagen implants investigated in vitro. J. Biomed. Mater. Res. 38, 95–104.

280

Cowan et al.

Nishimatsu, S., and Thomsen, G. H. (1998). Ventral mesoderm induction and patterning by bone morphogenetic protein heterodimers in Xenopus embryos. Mech. Dev. 74, 75–88. Niswander, L., and Martin, G. R. (1993). FGF-4 and BMP-2 have opposite eVects on limb growth. Nature 361, 68–71. Noth, U., Tuli, R., Seghatoleslami, R., Howard, M., Shah, A., Hall, D. J., Hickok, N. J., and Tuan, R. S. (2003). Activation of p38 and Smads mediates BMP-2 eVects on human trabecular bone-derived osteoblasts. Exp. Cell. Res. 291, 201–211. Nuttelman, C. R., Mortisen, D. J., Henry, S. M., and Anseth, K. S. (2001). Attachment of fibronectin to poly(vinyl alcohol) hydrogels promotes NIH3T3 cell adhesion, proliferation, and migration. J. Biomed. Mater. Res. 57, 217–223. Ogawa, R., Mizuno, H., Watanabe, A., Migita, M., Shimada, T., and Hyakusoku, H. (2004). Osteogenic and chondrogenic diVerentiation by adipose-derived stem cells harvested from GFP transgenic mice. Biochem. Biophys. Res. Commun. 313, 871–877. Onishi, T., Ishidou, Y., Nagamine, T., Yone, K., Imamura, T., Kato, M., Sampath, T. K., ten Dijke, P., and Sakou, T. (1998). Distinct and overlapping patterns of localization of bone morphogenetic protein (BMP) family members and a BMP type II receptor during fracture healing in rats. Bone 22, 605–612. Ono, I., Yamashita, T., Jin, H. Y., Ito, Y., Hamada, H., Akasaka, Y., Nakasu, M., Ogawa, T., and Jimbow, K. (2004). Combination of porous hydroxyapatite and cationic liposomes as a vector for BMP-2 gene therapy. Biomaterials. 25, 4709–4718. Orban, J. M., Marra, K. G., and Hollinger, J. O. (2002). Composition options for tissueengineered bone. Tissue Eng. 8, 529–539. Owen, M. (1980). The origin of bone cells in the postnatal organism. Arthritis Rheum. 23, 1073–1080. Parikh, S. N. (2002). Bone graft substitutes: Past, present, future. J. Postgrad. Med. 48, 142–148. Peng, H., Wright, V., Usas, A., Gearhart, B., Shen, H. C., Cummins, J., and Huard, J. (2002). Synergistic enhancement of bone formation and healing by stem cell-expressed VEGF and bone morphogenetic protein-4. J. Clin. Invest. 110, 751–759. Peng, Y., Kang, Q., Cheng, H., Li, X., Sun, M. H., Jiang, W., Luu, H. H., Park, J. Y., Haydon, R. C., and He, T. C. (2003). Transcriptional characterization of bone morphogenetic proteins (BMPs)-mediated osteogenic signaling. J. Cell. Biochem. 90, 1149–1165. Petite, H., Viateau, V., Bensaid, W., Meunier, A., de Pollak, C., Bourguignon, M, Oudina, K., Sedel, L., and Guillemin, G. (2000). Tissue-engineered bone regeneration. Nat. Biotechnol. 18, 959–963. Pittenger, M. F., Mackay, A. M., Beck, S. C., Jaiswal, R. K., Douglas, R., Mosca, J. D., Moorman, M. A., Simonetti, D. W., Craig, S., and Marshak, D. R. (1999). Multilineage potential of adult human mesenchymal stem cells. Science 284, 143–147. Prockop, D. J., Gregory, C. A., and Spees, J. L. (2003). One strategy for cell and gene therapy: Harnessing the power of adult stem cells to repair tissues. Proc. Natl. Acad. Sci. USA 100, 11917–11923. Puleo, D. A. (1997). Dependence of mesenchymal cell responses on duration of exposure to bone morphogenetic protein-2 in vitro. J. Cell. Physiol. 173, 93–101. Purpura, K. A., Aubin, J. E., and Zandstra, P. W. (2004). Sustained in vitro expansion of bone progenitors is cell density dependent. Stem Cells 22, 39–50. Quarto, R., Thomas, D., and Liang, C. T. (1995). Bone progenitor cell deficits and the ageassociated decline in bone repair capacity. Calcif. Tissue Int. 56, 123–129. Ramoshebi, L. N., Matsaba, T. N, Teare, J., Renton, L., Patton, J., and Ripamonti, U. (2002). Tissue engineering: TGF-beta superfamily members and delivery systems in bone regeneration. Expert. Rev. Mol. Med. 2002, 1–11.

8. Evolving Concepts in Bone Tissue Engineering

281

Rauch, F., Lauzier, D., Croteau, S., Travers, R., Glorieux, F. H., and Hamdy, R. (2000). Temporal and spatial expression of bone morphogenetic protein-2, -4, and -7 during distraction osteogenesis in rabbits. Bone 26, 611–617. Raval, P., Hsu, H. H., Schneider, D. J., Sarras, M. P., Jr., Masuhara, K., Bonewald, L. F., and Anderson, H. C. (1996). Expression of bone morphogenetic proteins by osteoinductive and non-osteoinductive human osteosarcoma cells. J. Dent. Res. 75, 1518–1523. Reddi, A. H. (1981). Cell biology and biochemistry of endochondral bone development. Coll. Relat. Res. 1, 209–226. Reddi, A. H. (1994). Bone and cartilage diVerentiation. Curr. Opin. Genet. Dev. 4, 737–744. Reddi, A. H. (2001). Bone morphogenetic proteins: From basic science to clinical applications. J. Bone Joint Surg. Am. 83-A, S1–S6. Rengachary, S. S. (2002). Bone morphogenetic proteins: Basic concepts. Neurosurg. Focus 13, 1–6. Rickard, D. J., Sullivan, T. A., Shenker, B. J., Leboy, P. S., and Kazhdan, I. (1994). Induction of rapid osteoblast diVerentiation in rat bone marrow stromal cell cultures by dexamethasone and BMP-2. Dev. Biol. 161, 218–228. Rose, F. R., and OreVo, R. O. (2002). Bone tissue engineering: Hope vs hype. Biochem. Biophys. Res. Commun. 292, 1–7. Rosen, V., Nove, J., Song, J. J., Thies, R. S., Cox, K., and Wozney, J. M. (1994). Responsiveness of clonal limb bud cell lines to bone morphogenetic protein 2 reveals a sequential relationship between cartilage and bone cell phenotypes. J. Bone Miner. Res. 9, 1759–1768. Rosso, F., Giordano, A., Barbarisi, M., and Barbarisi, A. (2004). From cell-ECM interactions to tissue engineering. J. Cell. Physiol. 199, 174–180. Rowley, J. A., and Mooney, D. J. (2002). Alginate type and RGD density control myoblast phenotype. J. Biomed. Mater. Res. 60, 217–223. Ruhe, P. Q., Hedberg, E. L., Padron, N. T., Spauwen, P. H., Jansen, J. A., and Mikos, A. G. (2003). rhBMP-2 release from injectable poly(DL-lactic-co-glycolic acid)/calcium-phosphate cement composites. J. Bone Joint Surg. Am. 85-A, 75–81. SaVord, K. M., Hicok, K. C., SaVord, S. D., Halvorsen, Y. D., Wilkison, W. O., Gimble, J. M., and Rice, H. E. (2002). Neurogenic diVerentiation of murine and human adipose-derived stromal cells. Biochem. Biophys. Res. Commun. 294, 371–379. Sampath, T. K., Coughlin, J. E., Whetstone, R. M., Banach, D., Corbett, C., Ridge, R. J., Ozkaynak, E., Oppermann, H., and Rueger, D. C. (1990). Bovine osteogenic protein is composed of dimers of OP-1 and BMP-2A, two members of the transforming growth factorbeta superfamily. J. Biol. Chem. 265, 13198–13205. Sampath, T. K., Maliakal, J. C., Hauschka, P. V., Jones, W. K., Sasak, H., Tucker, R. F., White, K. H., Coughlin, J. E., Tucker, M. M., Pang, R. H., et al. (1992). Recombinant human osteogenic protein-1 (hOP-1) induces new bone formation in vivo with a specific activity comparable with natural bovine osteogenic protein and stimulates osteoblast proliferation and diVerentiation in vitro. J. Biol. Chem. 267, 20352–20362. Sandhu, H. S., Kanim, L. E., Kabo, J. M., Toth, J. M., Zeegen, E. N., Liu, D., Delamarter, R. B., and Dawson, E. G. (1996). EVective doses of recombinant human bone morphogenetic protein-2 in experimental spinal fusion. Spine 21, 2115–2122. Sandhu, H. S., Kanim, L. E., Toth, J. M., Kabo, J. M., Liu, D., Delamarter, R. B., and Dawson, E. G. (1997). Experimental spinal fusion with recombinant human bone morphogenetic protein-2 without decortication of osseous elements. Spine 22, 1171–1180. Sato, M., Ochi, T., Nakase, T., Hirota, S., Kitamura, Y., Nomura, S., and Yasui, N. (1999). Mechanical tension-stress induces expression of bone morphogenetic protein (BMP)-2 and BMP-4, but not BMP-6, BMP-7, and GDF-5 mRNA, during distraction osteogenesis. J. Bone Miner. Res. 14, 1084–1095.

282

Cowan et al.

Schimandle, J. H., Boden, S. D., and Hutton, W. C. (1995). Experimental spinal fusion with recombinant human bone morphogenetic protein-2. Spine 20, 1326–1337. Schmoekel, H., Schense, J. C., Weber, F. E., Gratz, K. W., Gnagi, D., Muller, R., and Hubbell, J. A. (2004). Bone healing in the rat and dog with nonglycosylated BMP-2 demonstrating low solubility in fibrin matrices. J. Orthop. Res. 22, 376–381. Schlegel, K. A., Donath, K., Rupprecht, S., Falk, S., Zimmermann, R., Felszeghy, E., and Wiltfang, J. (2004). De novo bone formation using bovine collagen and platelet-rich plasma. Biomaterials 25, 5387–5393. Schmid, B., Furthauer, M., Connors, S. A., Trout, J., Thisse, B., Thisse, C., and Mullins, M. C. (2000). Equivalent genetic roles for bmp7/snailhouse and bmp2b/swirl in dorsoventral pattern formation. Development 127, 957–967. Schneider, G. B., English, A., Abraham, M., Zaharias, R., Stanford, C., and Keller, J. (2004). The eVect of hydrogel charge density on cell attachment. Biomaterials 25, 3023–3028. Schowing, J. (1968). [Demonstration of the inductive role of the brain in osteogenesis of the embryonic skull of the chicken]. J. Embryol. Exp. Morphol. 19, 83–94. Schulze-Tanzil, G., de Souza, P., Villegas Castrejon, H., John, T., Merker, H. J., Scheid, A., and Shakibaei, M. (2002). RediVerentiation of dediVerentiated human chondrocytes in highdensity cultures. Cell Tissue Res. 308, 371–379. Schwartz, M. A. (1997). Integrins, oncogenes, and anchorage independence. J. Cell Biol. 139, 575–578. Schwartz, Z., Somers, A., Mellonig, J. T., Carnes, D. L., Jr., Wozney, J. M., Dean, D. D., Cochran, D. L., and Boyan, B. D. (1998). Addition of human recombinant bone morphogenetic protein-2 to inactive commercial human demineralized freeze-dried bone allograft makes an eVective composite bone inductive implant material. J. Periodontol. 69, 1337–1345. Shang, Q., Wang, Z., Liu, W., Shi, Y., Cui, L., and Cao, Y. (2001). Tissue-engineered bone repair of sheep cranial defects with autologous bone marrow stromal cells. J. Craniofac. Surg. 12, 586–593; discussion 594–595. Shea, C. M., Edgar, C. M., Einhorn, T. A., and Gerstenfeld, L. C. (2003). BMP treatment of C3H10T1/2 mesenchymal stem cells induces both chondrogenesis and osteogenesis. J. Cell Biochem. 90, 1112–1127. Shen, H. C., Peng, H., Usas, A., Gearhart, B., Cummins, J., Fu, F. H., and Huard, J. (2004). Ex vivo gene therapy-induced endochondral bone formation: Comparison of muscle-derived stem cells and diVerent subpopulations of primary muscle-derived cells. Bone 34, 982–992. Shi, Y., and Massague, J. (2003). Mechanisms of TGF-beta signaling from cell membrane to the nucleus. Cell 113, 685–700. Shukunami, C., Ohta, Y., Sakuda, M., and Hiraki, Y. (1998). Sequential progression of the diVerentiation program by bone morphogenetic protein-2 in chondrogenic cell line ATDC5. Exp. Cell Res. 241, 1–11. Slonaker, M., and Goswami, T. (2004). Wear mechanisms in ceramic hip implants. J. Surg. Orthop. Adv. 13, 94–105. Solloway, M. J., Dudley, A. T., BikoV, E. K., Lyons, K. M., Hogan, B. L., and Robertson, E. J. (1998). Mice lacking Bmp6 function. Dev. Genet. 22, 321–339. Spector, J. A., Luchs, J. S., Mehrara, B. J., Greenwald, J. A., Smith, L. P., and Longaker, M. T. (2001). Expression of bone morphogenetic proteins during membranous bone healing. Plast. Reconstr. Surg. 107, 124–134. Stevenson, S., Cunningham, N., Toth, J., Davy, D., and Reddi, A. H. (1994). The eVect of osteogenin (a bone morphogenetic protein) on the formation of bone in orthotopic segmental defects in rats. J. Bone Joint Surg. Am. 76, 1676–1687. Storm, E. E., and Kingsley, D. M. (1999). GDF5 coordinates bone and joint formation during digit development. Dev. Biol. 209, 11–27.

8. Evolving Concepts in Bone Tissue Engineering

283

Suzuki, A., Kaneko, E., Maeda, J., and Ueno, N. (1997). Mesoderm induction by BMP-4 and -7 heterodimers. Biochem. Biophys. Res. Commun. 232, 153–156. Suzuki, Y., Tanihara, M., Suzuki, K., Saitou, A., Sufan, W., and Nishimura, Y. (2000). Alginate hydrogel linked with synthetic oligopeptide derived from BMP-2 allows ectopic osteoinduction in vivo. J. Biomed. Mater. Res. 50, 405–409. Sweeney, T. M., Opperman, L. A., Persing, J. A., and Ogle, R. C. (1995). Repair of critical size rat calvarial defects using extracellular matrix protein gels. J. Neurosurg. 83, 710–715. ten Dijke, P., Fu, J., Schaap, P., and Roelen, B. A. (2003). Signal transduction of bone morphogenetic proteins in osteoblast diVerentiation. J. Bone Joint Surg. Am. 85-A, 34–38. Thies, R. S., Bauduy, M., Ashton, B. A., Kurtzberg, L., Wozney, J. M., and Rosen, V. (1992). Recombinant human bone morphogenetic protein-2 induces osteoblastic diVerentiation in W-20-17 stromal cells. Endocrinology 130, 1318–1324. Tholpady, S. S., Katz, A. J., and Ogle, R. C. (2003). Mesenchymal stem cells from rat visceral fat exhibit multipotential diVerentiation in vitro. Anat. Rec. 272A, 398–402. Ting, K., Vastardis, H., Mulliken, J. B., Soo, C., Tieu, A., Do, H., Kwong, E., Bertolami, C. N., Kawamoto, H., Kuroda, S., and Longaker, M. T. (1999). Human NELL-1 expressed in unilateral coronal synostosis. J. Bone Miner. Res. 14, 80–89. Toma, C. D., SchaVer, J. L., Meazzini, M. C., Zurakowski, D., Nah, H. D., and Gerstenfeld, L. C. (1997). Developmental restriction of embryonic calvarial cell populations as characterized by their in vitro potential for chondrogenic diVerentiation. J. Bone Miner. Res. 12, 2024–2039. Toquet, J., Rohanizadeh, R., Guicheux, J., Couillaud, S., Passuti, N., Daculsi, G., and Heymann, D. (1999). Osteogenic potential in vitro of human bone marrow cells cultured on macroporous biphasic calcium phosphate ceramic. J. Biomed. Mater. Res. 44, 98–108. Toriumi, D. M., Kotler, H. S., Luxenberg, D. P., Holtrop, M. E., and Wang, E. A. (1991). Mandibular reconstruction with a recombinant bone-inducing factor. Functional, histologic, and biomechanical evaluation. Arch. Otolaryngol. Head Neck Surg. 117, 1101–1112. Tou, L., Quibria, N., and Alexander, J. M. (2003). Transcriptional regulation of the human Runx2/Cbfa1 gene promoter by bone morphogenetic protein-7. Mol. Cell. Endocrinol. 205, 121–129. Toung, J. S., GriYn, A., Ogle, R. C., and Lindsey, W. H. (1998). Repair of nasal defects using collagen gels containing insulin-like growth factor 1. Laryngoscope. 108, 1654–1658. Uludag, H., D’Augusta, D., Golden, J., Li, J., Timony, G., Riedel, R., and Wozney, J. M. (2000). Implantation of recombinant human bone morphogenetic proteins with biomaterial carriers: A correlation between protein pharmacokinetics and osteoinduction in the rat ectopic model. J. Biomed. Mater. Res. 50, 227–238. Uludag, H., D’Augusta, D., Palmer, R., Timony, G., and Wozney, J. (1999a). Characterization of rhBMP-2 pharmacokinetics implanted with biomaterial carriers in the rat ectopic model. J. Biomed. Mater. Res. 46, 193–202. Uludag, H., Friess, W., Williams, D., Porter, T., Timony, G., D’Augusta, D., Blake, C., Palmer, R., Biron, B., and Wozney, J. (1999b). rhBMP-collagen sponges as osteoinductive devices: EVects of in vitro sponge characteristics and protein pI on in vivo rhBMP pharmacokinetics. Ann. NY Acad. Sci. 875, 369–378. Uludag, H., Gao, T., Porter, T. J., Friess, W., and Wozney, J. M. (2001). Delivery systems for BMPs: Factors contributing to protein retention at an application site. J. Bone Joint. Surg. Am. 83-A, S128–S135. Urist, M. R. (1965). Bone: Formation by autoinduction. Science 150, 893–899. Valentin-Opran, A., Wozney, J., Csimma, C., Lilly, L., and Riedel, G. E. (2002). Clinical evaluation of recombinant human bone morphogenetic protein-2. Clin. Orthop. 1, 110–120. Wakitani, S., Saito, T., and Caplan, A. I. (1995). Myogenic cells derived from rat bone marrow mesenchymal stem cells exposed to 5-azacytidine. Muscle Nerve 18, 1417–1426.

284

Cowan et al.

Wang, E. A., Israel, D. I., Kelly, S., and Luxenberg, D. P. (1993). Bone morphogenetic protein-2 causes commitment and diVerentiation in C3H10T1/2 and 3T3 cells. Growth Factors 9, 57–71. Wang, E. A., Rosen, V., D’Alessandro, J. S., Bauduy, M., Cordes, P., Harada, T., Israel, D. I., Hewick, R. M., Kerns, K. M., LaPan, P., et al. (1990). Recombinant human bone morphogenetic protein induces bone formation. Proc. Natl. Acad. Sci. USA 87, 2220–2224. Wang, J. C., Kanim, L. E., Yoo, S., Campbell, P. A., Berk, A. J., and Lieberman, J. R. (2003). EVect of regional gene therapy with bone morphogenetic protein-2-producing bone marrow cells on spinal fusion in rats. J. Bone Joint Surg. Am. 85-A, 905–911. Wang, J. C., Yu, W. D., Sandhu, H. S., Betts, F., Bhuta, S., and Delamarter, R. B. (1999). Metal debris from titanium spinal implants. Spine 24, 899–903. Warnke, P. H., Springer, I. N., Wiltfang, J., Acil, Y., Eufinger, H., Wehmoller, M., Russo, P. A., Bolte, H., Sherry, E., Behrens, E., and Terheyden, H. (2004). Growth and transplantation of a custom vascularised bone graft in a man. Lancet 364, 766–770. Warren, S. M., Fong, K. D., Chen, C. M., Loboa, E. G., Cowan, C. M., Lorenz, H. P., and Longaker, M. T. (2003). Tools and techniques for craniofacial tissue engineering. Tissue Eng. 9, 187–200. Wickham, M. Q., Erickson, G. R., Gimble, J. M., Vail, T. P., and Guilak, F. (2003). Multipotent stromal cells derived from the infrapatellar fat pad of the knee. Clin. Orthop. 1, 196–212. Wiltfang, J., Kloss, F. R., Kessler, P., Nkenke, E., Schultze-Mosgau, S., Zimmermann, R., and Schlegel, K. A. (2004). EVects of platelet-rich plasma on bone healing in combination with autogenous bone and bone substitutes in critical-size defects. An animal experiment. Clin. Oral Implants Res. 15, 187–193. Winn, S. R., Hu, Y., Sfeir, C., and Hollinger, J. O. (2000). Gene therapy approaches for modulating bone regeneration. Adv. Drug Deliv. Rev. 42, 121–138. Winn, S. R., Uludag, H., and Hollinger, J. O. (1999). Carrier systems for bone morphogenetic proteins. Clin. Orthop. 1, S95–S106. Winter, G. D., and Simpson, B. J. (1969). Heterotopic bone formed in a synthetic sponge in the skin of young pigs. Nature 223, 88–90. Woo, B. H., Fink, B. F., Page, R., Schrier, J. A., Jo, Y. W., Jiang, G., DeLuca, M., Vasconez, H. C., and DeLuca, P. P. (2001). Enhancement of bone growth by sustained delivery of recombinant human bone morphogenetic protein-2 in a polymeric matrix. Pharm. Res. 18, 1747–1753. Wozney, J. M., and Rosen, V. (1998). Bone morphogenetic protein and bone morphogenetic protein gene family in bone formation and repair. Clin. Orthop. 1, 26–37. Wozney, J. M., Rosen, V., Celeste, A. J., Mitsock, L. M., Whitters, M. J., Kriz, R. W., Hewick, R. M., and Wang, E. A. (1988). Novel regulators of bone formation: Molecular clones and activities. Science 242, 1528–1534. Wright, V., Peng, H., Usas, A., Young, B., Gearhart, B., Cummins, J., and Huard, J. (2002). BMP4-expressing muscle-derived stem cells diVerentiate into osteogenic lineage and improve bone healing in immunocompetent mice. Mol. Ther. 6, 169–178. Yamaguchi, A., Ishizuya, T., Kintou, N., Wada, Y., Katagiri, T., Wozney, J. M., Rosen, V., and Yoshiki, S. (1996). EVects of BMP-2, BMP-4, and BMP-6 on osteoblastic diVerentiation of bone marrow-derived stromal cell lines, ST2 and MC3T3-G2/PA6. Biochem. Biophys. Res. Commun. 220, 366–371. Yamamoto, M., Kato, K., and Ikada, Y. (1997). Ultrastructure of the interface between cultured osteoblasts and surface-modified polymer substrates. J. Biomed. Mater. Res. 37, 29–36. Yamamoto, M., Takahashi, Y., and Tabata, Y. (2003). Controlled release by biodegradable hydrogels enhances the ectopic bone formation of bone morphogenetic protein. Biomaterials 24, 4375–4383.

8. Evolving Concepts in Bone Tissue Engineering

285

Yaoita, H., Orimo, H., Shirai, Y., and Shimada, T. (2000). Expression of bone morphogenetic proteins and rat distal-less homolog genes following rat femoral fracture. J. Bone Miner. Metab. 18, 63–70. Yasko, A. W., Lane, J. M., Fellinger, E. J., Rosen, V., Wozney, J. M., and Wang, E. A. (1992). The healing of segmental bone defects, induced by recombinant human bone morphogenetic protein (rhBMP-2). A radiographic, histological, and biomechanical study in rats. J. Bone Joint Surg. Am. 74, 659–670. Yi, S. E., Daluiski, A., Pederson, R., Rosen, V., and Lyons, K. M. (2000). The type I BMP receptor BMPRIB is required for chondrogenesis in the mouse limb. Development 127, 621–630. Yoshikawa, H., Rettig, W. J., Takaoka, K., Alderman, E., Rup, B., Rosen, V., Wozney, J. M., Lane, J. M., Huvos, A. G., and Garin-Chesa, P. (1994). Expression of bone morphogenetic proteins in human osteosarcoma. Immunohistochemical detection with monoclonal antibody. Cancer 73, 85–91. Zanchetta, P., Lagarde, N., and Guezennec, J. (2003). Systemic eVects on bone healing of a new hyaluronic acid-like bacterial exopolysaccharide. Calcif. Tissue Int. 73, 232–236. Zegzula, H. D., Buck, D. C., Brekke, J., Wozney, J. M., and Hollinger, J. O. (1997). Bone formation with use of rhBMP-2 (recombinant human bone morphogenetic protein-2). J. Bone Joint Surg. Am. 79, 1778–1790. Zhang, H., and Bradley, A. (1996). Mice deficient for BMP2 are nonviable and have defects in amnion/chorion and cardiac development. Development 122, 2977–2986. Zhang, X., Carpenter, D., Bokui, N., Soo, C., Miao, S., Truong, T., Wu, B., Chen, I., Vastardis, H., Tanizawa, K., Kuroda, S., and Ting, K. (2003). Overexpression of Nell-1, a craniosynostosis-associated gene, induces apoptosis in osteoblasts during craniofacial development. J. Bone Miner. Res. 18, 2126–2134. Zhang, X., Kuroda, S., Carpenter, D., Nishimura, I., Soo, C., Moats, R., Iida, K., Wisner, E., Hu, F. Y., Miao, S., Beanes, S., Dang, C., Vastardis, H., Longaker, M., Tanizawa, K., Kanayama, N., Saito, N., and Ting, K. (2002). Craniosynostosis in transgenic mice overexpressing Nell-1. J. Clin. Invest. 110, 861–870. Zhu, G., Mallery, S. R., and Schwendeman, S. P. (2000). Stabilization of proteins encapsulated in injectable poly (lactide-co-glycolide). Nat. Biotechnol. 18, 52–57. Zhu, J. and Emerson, S. G. (2004). A new bone to pick: Osteoblasts and the haematopoietic stem-cell niche. Bioessays 26, 595–599. Zuk, P. A., Zhu, M., Ashjian, P., De Ugarte, D. A., Huang, J. I., Mizuno, H., Alfonso, Z. C., Fraser, J. K., Benhaim, P., and Hedrick, M. H. (2002). Human adipose tissue is a source of multipotent stem cells. Mol. Biol. Cell 13, 4279–4295. Zuk, P. A., Zhu, M., Mizuno, H., Huang, J., Futrell, J. W., Katz, A. J., Benhaim, P., Lorenz, H. P., and Hedrick, M. H. (2001). Multilineage cells from human adipose tissue: implications for cell-based therapies. Tissue Eng. 7, 211–228.

9

____________________________________________________________________________

Cranial Suture Biology Kelly A Lenton, Randall P. Nacamuli, Derrick C. Wan, Jill A. Helms, and Michael T. Longaker Children’s Surgical Research Program Division of Plastic and Reconstructive Surgery Department of Surgery Stanford University School of Medicine Stanford, California 94305-5148

I. Introduction II. Skull Vault Development III. Craniosynostosis A. Clinical Genetics of Craniosynostosis IV. The Murine Model of Suture Morphogenesis A. The Posterior Frontal Suture: An Example of Physiological Suture Fusion B. Tissue Interactions V. Molecular and Cellular Mechanisms Governing Suture Morphogenesis A. Growth Factors B. Transcription Factors C. Apoptosis VI. Conclusions and Perspectives Acknowledgments References

I. Introduction The term craniosynostosis was first used in 1830 by Otto to describe the premature fusion of cranial sutures (Otto, 1830). Since this first identification of craniosynostosis as a distinct clinical entity, several theories have been proposed to explain both the pathogenesis of premature suture fusion and the resultant aberrations in calvarial growth that result in a dysmorphic skull. The first of these theories, proposed by Virchow in the 1850s, presumed that craniosynostosis was linked with cretinism or inflammation of the meninges (Virchow, 1851). Virchow’s Law states that calvarial growth in a child with craniosynostosis occurs in a plane parallel to that of the fused suture, with minimal growth in the perpendicular plane. Thus, a child with fusion of the longitudinally oriented sagittal suture develops scaphocephaly, or a boat-shaped skull, with pronounced growth in the anteroposterior axis Current Topics in Developmental Biology, Vol. 66 Copyright 2005, Elsevier Inc. All rights reserved. 0070-2153/05 $35.00

287

288

Lenton et al.

and minimal growth laterally. In the 1920s a second theory suggested that congenital abnormalities in the suture mesenchyme were the cause of cranial suture fusion (Park and Powers, 1920). More recently, Moss noted that simply removing an aVected cranial suture does not restore normal calvarial development, and that patients with craniosynostosis often have skull base abnormalities (Moss, 1959). Moss hypothesized that the primary pathology in craniosynostosis was aberrant morphogenesis and development of the basicranium with subsequent impairment in the growth of the brain. He postulated that these aberrations result in altered mechanical forces that are transmitted to the cranial sutures via the dura mater, resulting in premature suture fusion. This hypothesis led to a radical change in the surgical treatment of patients with craniosynostosis, since it implied that merely removing a synostosed suture was insuYcient to correct the underlying pathology, and that instead complex craniofacial procedures were needed to eVect cranial expansion. Tessier, the father of modern craniofacial surgery, pioneered these procedures in the 1960s (Tessier, 1967, 1970). As early as 1912, Crouzon observed that there was a familial pattern of inheritance to the craniosynostotic syndrome that now bears his name (Crouzon, 1912). This was the first indication that there was a genetic component to craniosynostosis. Recent advances in clinical genetics have resulted in the identification of genetic mutations in the major craniosynostostic syndromes. Despite these insights into the rudimentary disturbances leading to craniosynostosis, the processes by which mutations in these genes trigger premature suture fusion remain largely unknown. Rodents are proving to be extremely valuable in unraveling the cellular and molecular mechanisms of cranial suture morphogenesis and pathology. Prior to the recent generation of genetically engineered mice that mimic human craniosynostoses, pioneering work by Opperman and colleagues demonstrated that in an environment free from mechanical forces and humoral eVects, tissue interactions between specific components of the cranial suture complex were able to sustain suture patency (Opperman et al., 1993, 1994, 1995). It was at this time that an observation made in the 1950s by Moss was rediscovered—that the posterior frontal suture in mice and rats (analogous to the human metopic or interfrontal suture) undergoes predictable physiological fusion during the first few weeks of life (Moss, 1958). This is in contrast to murine coronal and sagittal sutures, which normally remain open throughout the life of the animal. Armed with this observation, researchers now had a model of physiological suture fusion and patency that was easily accessible and facilitated analysis of the molecular and cellular events occurring in sutures before, during, and after the fusion process.

9. Cranial Suture Biology

289

II. Skull Vault Development The membranous skull vault is composed of paired frontal, parietal, and squamosal bones, and the anterior portion of the occipital bone. The bones are of mixed embryonic origin arising from either neural crest or paraxial mesoderm (Couly et al., 1993; Jiang et al., 2002; Noden, 1978, 1988). Development of the skull vault initiates with the migration of mesenchymal cells to positions between the brain and surface ectoderm. There the cells condense, forming mesenchymal blastemas, which diVerentiate along an osteogenic pathway that generates new bone without the formation of a cartilaginous intermediate. As intramembranous ossification proceeds, the flat bones of the skull vault expand from the region closest to the skull base toward the apex of the cranium (Johnson et al., 2000; Rice et al., 2000). The cranial sutures form as the margins of the developing bones approximate, and the mesenchymal tissue separating the bone fronts is recruited into the osteogenic fronts (Opperman, 2000). It remains unknown whether the formation and spatial arrangement of the sutures occurs as a response to the approximation of the bone fronts, if they are prepatterned in a manner similar to the joints in the developing limb, or if signals from the surrounding epithelia determine the position of the sutures. Once formed, the cranial sutures constitute the major growth centers in the skull vault, and further growth proceeds by apposition at the osteogenic fronts (Wilkie, 1997; Wilkie et al., 2001). Mesenchymal cells diVerentiate into osteoprogenitor cells and then into osteoblasts that express collagen 1, bone sialoprotein, and osteocalcin, and synthesize bone matrix at the osteogenic fronts (Ornitz and Marie, 2002). The developing calvarial bones are suspended within the meningeal and outer periosteal membranes. As these membranes grow, the bones are displaced and drawn apart, generating tensile forces within the sutures, which have been postulated to be the primary stimulus for the formation of new bone at the osteogenic fronts (Moss, 1959). In addition to circumferential growth, the calvarial bones are subject to continual remodeling in order to fit the developing curvature of the expanding skull vault (Enlow, 2000). The cranial sutures include the metopic or interfrontal suture (between the frontal bones), the sagittal suture (between the parietal bones), the coronal suture (between the frontal and parietal bones), and the lambdoid sutures (between the parietal and interparietal bones) (Fig. 1). The sutures can be thought of as a complex consisting of four principal components: (1) the osteogenic fronts of the approximating bone plates; (2) the suture mesenchyme spanning the osteogenic fronts; (3) the overlying pericranium or cranial periosteum; and (4) the underlying dura mater, a tough, fibrous membrane that constitutes the outer meningeal layer that envelops the brain and forms the inner lining of cranial bones and sutures. The osteogenic fronts of the transversely oriented coronal and lambdoid sutures

290

Lenton et al.

Figure 1 Schematic illustrating the skeletal components, sutures, and tissue origins of the mouse skull vault. The calvarium is composed of paired parietal and frontal bones and the interparietal bone. The sutures are situated between the bones. The interfrontal suture is between the frontal bones, the sagittal suture is between the parietal bones, the coronal suture is between the frontal and parietal bones, and the lambdoid sutures are between the parietal and interparietal bones. Tissues derived from neural crest (blue) include the dura mater (not shown), frontal bones, and midline sutures. Tissues derived from paraxial mesoderm (red) include the parietal bones and coronal sutures. The interparietal bone is a composite with origins in both neural crest and paraxial mesoderm. The embryonic origin of the lambdoid sutures is unknown.

overlap, whereas the osteogenic fronts of the anterior-to-posterior–oriented interfrontal and sagittal sutures abut end to end (Fig. 2). The derivation of the neurocranium from paraxial mesoderm and cranial neural crest is well established; however, the embryonic origins of the individual elements that make up the skull vault were not clearly delineated until recently (Couly et al., 1993; Jiang et al., 2002; Noden, 1978, 1988). Characterization of a novel double transgenic mouse has defined the cellular origins of the cranial skeleton in mammals (Chai et al., 2000; Jiang et al., 2002). This mouse is doubly heterozygous for the Wnt1–Cre transgene, which drives expression of the Cre recombinase in all cells of neural crest origin, and the R26R allele, which selectively encodes -galactosidase in cells that produce Cre recombinase (Soriano, 1999). Thus, all cells of neural crest origin indelibly express -galactosidase and can be tracked using X-gal staining. Analysis of neural crest migration pathways using X-gal staining

9. Cranial Suture Biology

291

Figure 2 Hematoxylin and Eosin stained histological sections of postnatal day 6 mouse sutures. The sutures shown are the posterior frontal suture (A), the sagittal suture (B), and the coronal suture (C). The frontal bone consists of two bone tables, an inner bone table (IBT) and an outer bone table (OBT) with an intervening marrow space (A). The principle components of the sutures include the osteogenic fronts (OF) of the approximating bone plates (outlined in yellow), the suture mesenchyme (SM), the overlying pericranium (PC), and the dura mater (DM). The osteogenic fronts of the transversely oriented coronal suture overlap (C), whereas the osteogenic fronts of the anterior to posterior oriented posterior frontal and sagittal sutures abut end-to-end (A and B). SS, sagittal sinus; PB, parietal bone; FB, frontal bone.

together with mesodermal tracing by DiI labeling in this model have revealed the mixed tissue origins of the composite elements of the skull vault. Although the dura mater and frontal bones originate from cranial neural

292

Lenton et al.

crest, the parietal bones are derived from mesoderm. Interestingly, diVerent combinations of neural crest– and mesodermal-derived tissues comprise the cranial sutures. The major growth centers of the skull vault, the coronal and sagittal sutures, form at interfaces between mesodermal and neural crest– derived tissues. In contrast, the fusing posterior frontal suture is formed entirely from neural crest cells, with both the osteogenic fronts of the flanking frontal bones and the intervening suture mesenchyme originating from the neural crest (Jiang et al., 2002) (Fig. 3). In addition to acting as connective tissue, the sutures perform several vital functions in early life. At birth, the flexible joints provided by the sutures permit overlap of the cranial bones during passage through the birth canal. From fetal life through infancy and early childhood, the sutures constitute the major calvarial growth centers, facilitating expansion of the skull vault in concert with the rapidly growing brain. Sutures are also major signaling centers that regulate the balance between the proliferation of osteogenic precursors and their diVerentiation to bone (Iseki et al., 1997, 1999; Kim et al., 1998; Rice et al., 2000; Zhou et al., 2000). With the exception of the metopic suture, which undergoes fusion during the first year of human life, the cranial sutures normally remain patent well into adulthood (Weinzweig et al., 2003).

III. Craniosynostosis Normal calvarial development is dependent upon coordinated growth between the brain and overlying cranial bone plates. Perturbations in the complex communications between the brain, dura mater, suture mesenchyme, and osteogenic fronts during development can manifest in premature, pathological fusion of cranial sutures (craniosynostosis). Craniosynostosis is a common developmental disorder with an incidence of approximately 1:2500 live births (Hunter and Rudd, 1976, 1977; Lajeunie et al., 1995, 1996). Premature suture fusion disables normal cranial vault enlargement, resulting in numerous morphological and physiological abnormalities, including dysmorphic cranial shape, hydrocephalus, elevated intracranial pressure, deafness, blindness, midface hypoplasia, and airway compromise (Posnick, 2000). Premature suture fusion can occur either as an isolated condition, or as part of several syndromes with a constellation of abnormalities, often including limb anomalies (Wilkie et al., 2001). Children aZicted with craniosynostosis require physiologically challenging, complex surgery to remodel the cranial vault and prevent or correct the functional and morphological deficits. The surgically remodeled skull vault is itself prone to refuse, once again restricting growth of the brain and necessitating multiple reconstructive surgical procedures. Despite its prevalence, the etiopathogenesis of craniosynostosis remains poorly understood. An improved understanding

9. Cranial Suture Biology

293

Figure 3 Schematic illustration of the tissue origins and development of the mouse posterior frontal (PF), sagittal (SAG), and coronal (COR) sutures. (A) Development of the skull vault begins when mesenchymal cells of either neural crest (nc) or mesodermal (meso) origins (purple) interpose themselves between the surface ectoderm and the dorsal neuroectoderm (both epithelia are indicated with yellow). (B) Whether it is derived from neural crest or mesoderm, this mesenchyme will eventually diVerentiate into bone. The presumptive dura mater, which is derived from neural crest mesenchyme (blue), is positioned dorsal to the neuroectoderm and ventral to the osteogenic mesenchyme. Depending on its position in the head, the osteogenic mesenchyme is derived from neural crest or mesoderm and diVerentiates as indicated in 1, 2, and 3. (1) In the frontal bone the undiVerentiated suture mesenchyme, the diVerentiating osteogenic mesenchyme, and the underlying dura mater are derived from cranial neural crest (blue). In the posterior frontal suture itself the frontal bone splits into ectocranial and endocranial bone tables with an intervening marrow (diploic) space (orange). The suture mesenchyme separating the endocranial bone plates undergoes chondrogenesis in the first few weeks of life, and bony fusion proceeds by endochondral ossification. The ectocranial bone tables are separated by connective tissue of unknown origin (white) and this tissue remains undiVerentiated throughout the life of the animal. (2) In the region of the sagittal suture,

294

Lenton et al.

of the genetic, molecular, and tissue interactions involved in physiological and pathological suture fusion may result in the development of improved biologically based treatments for craniosynostosis.

A. Clinical Genetics of Craniosynostosis The vast majority of craniosynostoses are nonsyndromic; however, more than 100 craniosynostotic syndromes have been identified (Wilkie, 1997). During the last decade the major syndromes have been linked to dominant mutations in genes encoding fibroblast growth factor receptors (FGFR1, FGFR2, and FGFR3), and the transcription factors TWIST and MSX2 (Bellus et al., 1996; El Ghouzzi et al., 1997; Howard et al., 1997; Jabs et al., 1994, 1993; Oldridge et al., 1995; Reardon and Winter, 1994). These mutations account for
25% of all cases of craniosynostosis (Twigg et al., 2004). Identification of the genetic basis of premature suture fusion has led to the emergence of molecular diagnoses for patients with syndromic craniosynostoses. The use of genetic criteria as a diagnostic tool is advantageous in terms of early prenatal detection, improved genetic counseling, and more informed patient management (Ferreira et al., 1999). Importantly, a number of studies have demonstrated that molecular diagnosis is superior to clinical diagnosis as a predictor of surgical outcome (Cassileth et al., 2001; von Gernet et al., 2000). Genetic screening is a useful adjunct to clinical diagnosis because of the phenotypic heterogeneity associated with craniosynostoses. For example, a mild case of Crouzon syndrome without limb anomalies may be phenotypically indistinguishable from sporadic coronal synostosis in the early stages of the pathology. Genetic studies also contribute to our understanding of the etiopathogenesis of craniosynostosis. The association of MSX2 gain-of-function mutations with premature suture fusion and MSX2 loss-of-function mutations with calvarial ossification defects implies sensitivity to MSX2 gene dosage (Wilkie et al., 2000). Genetic studies linking specific gene mutations with phenotypes expressed in clinically distinct syndromes have provided clues to the molecular interactions that might occur during suture morphogenesis.

the mesenchyme is derived from neural crest (blue) while the diVerentiating osteogenic fronts are mesoderm in origin (red). The dura is derived solely from neural crest. (3) The coronal suture is comprised of mesodermal suture mesenchyme (light red), diVerentiating osteogenic mesenchyme from both mesoderm and neural crest, and neural crest-derived dura. In all cases, the mesodermal parietal bone is positioned superficially to and overlaps the neural crest-derived frontal bone.

9. Cranial Suture Biology

295

1. FGFR Mutations In mice and humans, the FGF family of ligands consists of at least 22 structurally related proteins (Ornitz and Itoh, 2001). The FGFs are grouped by amino acid sequence homology into six subfamilies that share common biochemical and functional properties and expression profiles (Ornitz and Itoh, 2001). FGF signaling pathways are involved in the regulation of a variety of developmental processes, including endochondral and intramembranous bone formation (Ornitz and Marie, 2002). FGF signal transduction is mediated via a family of four single pass transmembrane receptors with tyrosine kinase activity (Jaye et al., 1992; Johnson and Williams, 1993). Additional isoforms of FGFR1, FGFR2, and FGFR3 are generated by diVerential splicing of two adjacent exons that alternatively encode the C-terminal half of the third immunoglobulin (Ig)like loop of the extracellular ligand-binding domain (Johnson and Williams, 1993). The variant isoforms exhibit unique ligand-binding specificities and tissue-specific localization, with expression of the ‘‘IIIb’’ and ‘‘IIIc’’ splice forms generally restricted to epithelial and mesenchymal tissues, respectively (Chellaiah et al., 1994; Johnson and Williams, 1993; Miki et al., 1992; Naski and Ornitz, 1998; Orr-Urtreger et al., 1993). FGF signal transduction occurs via ligand-induced receptor homodimerization and heterodimerization activating autophosphorylation and transphosphorylation of intracellular tyrosine residues. This ligand–receptor complex is stabilized not only by heparan-like glycosaminoglycans, but also through receptor–dimer interactions between the IgII and IgIII linker regions of each molecule (Powers et al., 2000). Cytoplasmic signal transduction occurs through at least two separate pathways with recruitment of target proteins to the activated receptor complex. Phosphorylated tyrosine residues have been shown to interact with the Src-homology 2 (SH2) domains of both phospholipase C-gamma (PLC ) and Crk proteins. Activated PLC cleaves phosphatidyl-inositol-4,5-bisphosphate to generate inositol trisphosphate (IP3) and diacylglycerol (DAG). IP3, in turn, releases calcium from the endoplasmic reticulum to interact with DAG and thereby activate protein kinase C (PKC) (Powers et al., 2000). Like PLC , Crk also contains a SH2 domain that associates with phosphorylated tyrosine residues, ultimately increasing activity of both downstream extracellular regulated protein kinase (ERK)-1/2 and Jun kinase. FGF receptors have also been shown to interact with and activate SNT-1/FRS2, a 90-kDa protein that links FGFRs with the Ras/MAP kinase pathway (Wang, 1996). Heterozygous FGFR mutations are the most common genetic abnormality associated with craniosynostotic syndromes. The mutations are mostly missense, or small in-frame insertions or deletions that aVect the ligandbinding domain (Wilkie, 1997). Other mutations involve the tyrosine kinase

296

Lenton et al.

domain of FGFR2 and are presumed to impact on downstream signaling pathways (Kan et al., 2002). Most FGFR mutations are dominant and are thought to result in a gain of function through increased ligand binding aYnity, decreased specificity of receptor–ligand interactions, or increased receptor dimerization and stabilization. FGFR2 mutations cause most craniosynostotic syndromes, and two activating mutations in FGFR2—Ser252Trp and Pro253Arg—account for almost all cases of Apert syndrome. The analogous Pro252Arg mutation in FGFR1 has been noted in some cases of PfeiVer syndrome (Robin et al., 1994; Wilkie et al., 1995b). Two substitutions in the linker region between IgII and IgIII in FGFR3—Pro250Arg and Arg248Cys—have been associated with Muenke syndrome and type I thanatophoric dysplasia, respectively (Bellus et al., 1996; Muenke et al., 1997, 1994; Tavormina et al., 1995). A number of mechanisms leading to gain of function in mutated FGFRs have been described. For example, mutations in the extracellular IgII and IgIII domains enhance aYnity for FGF ligands and promote receptor dimerization (Anderson et al., 1998; Wilkie et al., 1995a). Other mechanisms include altered ligand specificity and intracellular kinase activity, and autonomous receptor function (Powers et al., 2000). Recent x-ray crystallographic analysis with Apert syndrome mutant FGFR2–FGF2 complexes has demonstrated additional intermolecular contacts, providing a structural explanation for receptor gain of function (Ibrahimi et al., 2001). Similar analysis has also been performed on the PfeiVer syndrome FGFR1 mutation, revealing three additional hydrogen bonds between the receptor and its ligand (Ibrahimi et al., 2004). By stabilizing FGF–FGFR interactions, mutations in the IgII–IgIII linker region promote receptor dimerization with subsequent upregulation of the intracellular signaling pathway. Ligand-binding analysis of these mutations has yielded additional information regarding changes in receptor specificity. The Pro252Arg mutations in FGFR1 and FGFR3 have been shown to increase ligand binding and exhibit enhanced FGF9 binding aYnity relative to wild-type receptors (Ibrahimi et al., 2004). Because wildtype FGFR1 and FGFR3 both demonstrate minimal aYnity for FGF9, it has been suggested that uncharacteristic receptor interaction with this ligand may contribute to the pathogenesis of craniosynostosis (Ibrahimi et al., 2004). Therefore, mutations in the IgII–IgIII linker regions of FGFR1, FGFR2, and FGFR3 not only increase receptor aYnity for FGF, but also decrease ligand specificity. Several cysteine substitutions involving the IgIII domain of FGFR2 have been identified in Crouzon syndrome (Cys342Ser, Cys342Arg, Csy342Tyr, C324Phe, and Cys342Trp) (Cohen, 1997b). Mutations resulting in loss of a cysteine residue have been suggested to eliminate the disulfide bond in the IgIII domain, resulting in altered conformation and ligand aYnity (Park et al., 1995). Splicing site mutations in exon 9 of FGFR2

9. Cranial Suture Biology

297

have also been noted to result in PfeiVer syndrome (Cohen, 1997b). This mutation may disrupt normal splicing mechanisms, altering gene expression and thus receptor function. The Cys342Arg FGFR2 mutation, already implicated in Crouzon syndrome, has also been associated with PfeiVer syndrome (Oldridge et al., 1995; Park et al., 1995). This phenotypic variability suggests the probable involvement of several other gene products interacting with the FGF signaling pathway to determine ultimate syndrome presentation. 2. TWIST Mutations TWIST is a transcription factor with basic helix-loop-helix and DNA binding domains originally identified as a key inducer of mesoderm formation in Drosophila, and it has since been found to be important for the specification and patterning of mesoderm across phyla (Castanon and Baylies, 2002; Nusslein-Volhard et al., 1984). There are two TWIST proteins in mammals, TWIST-1 and TWIST-2 (formerly Dermo-1) (Li et al., 1995; Wolf et al., 1991). Heterozygous TWIST-1 mutations have been identified in 80% of patients with Saethre–Chotzen syndrome (Chun et al., 2002; Johnson et al., 1998). Saethre–Chotzen syndrome is an autosomal dominant disorder with high penetrance and variable expressivity characterized by craniosynostosis involving one or both coronal sutures, midface hypoplasia, facial asymmetry, low frontal hairline, ptosis, and limb defects, including two-thirds syndactyly of the hands and bifid great toes (El Ghouzzi et al., 1997; Johnson et al., 1998; Reardon and Winter, 1994). The majority of TWIST-1 mutations are missense and nonsense mutations that cluster in the functional domains and are predicted to cause loss of function (El Ghouzzi et al., 1997; Johnson et al., 1998). A significant proportion of TWIST-1 mutations associated with Saethre–Chotzen syndrome are large deletions (Johnson et al., 1998). Developmental delay distinguishes patients with TWIST-1 deletions from those with intragenic mutations (Chun et al., 2002; Johnson et al., 1998). In contrast to the MSX2 and FGFR gain-of-function mutations associated with craniosynostosis, the frequency of TWIST-1 deletions and nonsense mutations associated with Saethre–Chotzen syndrome suggests that the genetic mechanism in this syndrome is haploinsuYciency (Gripp et al., 2000). Genetic screening of patients with Saethre–Chotzen syndrome has also linked FGFR mutations with this syndrome (Chun et al., 2002; Paznekas et al., 1998). The observation that a single syndrome can be caused by mutations in either TWIST-1, FGFR2, or FGFR3, and that phenotypic overlap occurs between Saethre–Chotzen syndrome and syndromes associated with FGFR mutations such as Crouzon and PfeiVer syndromes, suggests that TWIST-1 and FGFRs are active in the same signaling networks during suture morphogenesis (Jabs, 2001; Paznekas et al., 1998).

298

Lenton et al.

3. MSX2 Mutations MSX2 (muscle segment homeobox 2) is a highly conserved homeobox gene that is involved in the regulation of inductive tissue interactions during embryogenesis (Jabs et al., 1993). The role of MSX2 in craniofacial morphogenesis is highlighted by its association with Boston-type craniosynostosis, a fully penetrant, autosomal dominant disorder with variable expression in terms of suture involvement and cranial morphology (Muller et al., 1993; Warman et al., 1993). Boston-type craniosynostosis results from an MSX2 cysteine to adenosine transversion that leads to an amino acid substitution of a histidine for a proline at position 148 (Pro148His) within the homeodomain, a highly conserved DNA-binding motif (Jabs et al., 1993). The Pro148His mutation occurs at position 7 in the homeodomain and results in enhanced DNA-binding aYnity and augmentation of the normal function of Msx2 (Ma et al., 1996). Conversely, deletions and missense mutations within the MSX2 homeodomain result in haploinsuYciency with reduced binding to the target DNA binding site and resultant ossification deficiencies that manifest as parietal foramina (Wilkie et al., 2000). The association of the MSX2 gain-of-function mutation with premature suture fusion, and MSX2 haploinsuYciency with calvarial ossification defects, indicates a critical role for MSX2 in cranial morphogenesis that is dosage dependent (Wilkie et al., 2000).

IV. The Murine Model of Suture Morphogenesis In comparison to clinical genetics studies, analyses of postoperative specimens of synostosed sutures have been less informative. Studies based on the analysis of clinical samples reflect only a ‘‘snapshot’’ in time and are hampered by genetic diversity, a paucity of synostotic and control tissues, and usually an inability to obtain samples of dura mater. Most importantly, clinical samples do not allow for analysis before, during, and after suture fusion. Because of these impediments, researchers have turned to animal models of craniosynostosis, including primates, canines, ovines, lagomorphs, and murine models (Butow, 1990; Christensen and Clark, 1970; Levine et al., 1998; Mooney et al., 1996; Stelnicki et al., 1998). Craniofacial development, molecular specification, and structure are highly conserved in humans and mice, making the mouse an excellent model in which to study human suture development and pathology (Richtsmeier et al., 2000; Schneider et al., 1999; Waterston et al., 2002; Wilkie and Morriss-Kay, 2001). Rats, as well as mice, are suitable models of suture fusion, since the posterior frontal suture undergoes physiological fusion in a predictable manner in early postnatal life, facilitating analysis of the cellular

9. Cranial Suture Biology

299

and molecular events occurring at critical time points before and during suture fusion. Furthermore, all sutures, aside from the posterior frontal suture, remain patent in rats and mice; thus, the processes involved in normal physiological suture fusion can be compared and contrasted with those involved in the maintenance of suture patency. Mouse and rat models have enabled both in vitro and in vivo experiments to elucidate the mechanisms governing both cranial suture development and fate (fusion patency) vs.

A. The Posterior Frontal Suture: An Example of Physiological Suture Fusion The murine posterior frontal suture is equivalent to the human metopic suture, and in both humans and rodents this suture fuses in early postnatal life. In rats and mice all other sutures remain patent throughout the life of the animal, whereas in humans the other sutures fuse at various stages during adulthood (Cohen, 1997a). In the region of the posterior frontal suture, the frontal bone consists of two tables of bone with an intervening marrow space. The murine posterior frontal suture normally undergoes fusion in the first several weeks of postnatal life. Fusion is initiated at the anterior aspect of the inner table and proceeds in a posterior direction. The outermost portion of the ectocranial bone table usually remains unfused. The formation of cartilage in the posterior frontal suture has previously been noted in rats, and in the metopic suture in humans (Manzanares et al., 1988; Moss, 1958). We have recently revisited and analyzed the mouse posterior frontal suture in detail and have found that physiological posterior frontal suture fusion involves a cartilage intermediate. To explore the possibility that posterior frontal suture fusion actually occurs by endochondral rather than intramembranous ossification, we analyzed the expression of genes that mark the various stages of chondrogenesis and osteogenesis during endochondral bone formation. The spatial and temporal expression patterns of genes, including collagen 1, collagen 2, osteopontin, and osteocalcin, were consistent with endochondral bone formation. Furthermore, PECAM staining of endothelial cells demonstrated that although vasculature was excluded from cartilaginous tissues, transition to bone was associated with new vessels (Lenton, 2004). Based on these observations, we hypothesize that physiological fusion of the posterior frontal suture involves a switch from intramembranous to endochondral ossification of neural crest–derived mesenchyme. A number of laboratories have focused their investigations on the mouse coronal and sagittal sutures, which are the most commonly aVected sutures in craniosynostosis. However, as described earlier, in rats and mice these two

300

Lenton et al.

sutures remain patent throughout life. Our laboratory has focused on investigations of physiological murine posterior frontal suture fusion in order to gain insight into the mechanistic diVerences between physiological suture fusion and maintenance of cranial suture patency. In the following sections we discuss the cellular and molecular interactions that are involved in posterior frontal suture fusion in the murine model in relation to results obtained studying coronal and sagittal sutures.

B. Tissue Interactions Studies of the interactions between the pericranium, suture mesenchyme, underlying dura mater, and flanking osteogenic fronts have demonstrated that the dura mater plays a critical role in directing suture fate in the murine model. 1. Pericranium The calvarial periosteum, also referred to as the pericranium, is a source of osteogenic growth factors and is required for bone formation during growth and repair, suggesting a potential role for calvarial periosteum in suture morphogenesis (Centrella and Canalis, 1985). The results of calvariectomies on neonatal rabbits indicated that the growth of new sutures and reestablishment of normal skeletal architecture were dependent on the maintenance of periosteal continuity at the injured site (Mabbutt and Kokich, 1979; Mabbutt et al., 1979). Removal of periosteum overlying coronal sutures did not aVect suture patency in rats or young rabbits, although it was associated with fusion in a proportion of treated sagittal sutures (Moss, 1960; Nappen and Kokich, 1983). In experiments designed to study the eVects of periosteal tissue on coronal suture patency, fetal and neonatal rat coronal sutures were transplanted to surgically created defects in the parietal bones of adult rat hosts (Opperman et al., 1994). Grafted fetal sutures continued to develop normally, and neonatal suture patency was maintained regardless of the presence or absence of donor or host periosteum, demonstrating that periosteum does not contribute to the development or maintenance of the coronal suture. 2. Dura Mater The osteogenic capacity of neonatal dura mater has been observed clinically and demonstrated in several animal models. Skull vault reossification following calvariectomy in infants and young animals is dependent on the maintenance of intact dura mater (Babler et al., 1982; Mabbutt and Kokich,

9. Cranial Suture Biology

301

1979; Mabbutt et al., 1979; Mossaz and Kokich, 1981). Trephine defects created in adult animals fail to heal, whereas intact neonatal dura mater is associated with reossification at the injured site (Aalami et al., 2004; Hobar et al., 1993). Transplantation of neonatal rat dura mater into epithelial– mesenchymal pockets in rat skin resulted in the generation of new bone without the involvement of other central nervous system components (Yu et al., 1997). The ability of neonatal dura mater to induce bone formation, and the intimate association between the dura mater and cranial sutures, prompted investigations to determine a potential role for dura mater in suture morphogenesis and fate. These studies utilized tissue transplantation and organ culture techniques specifically designed to eliminate potential biomechanical influences from the dura mater and cranial base that have been postulated to influence cranial suture fate. Transplantation of presumptive embryonic (E19) and diVerentiated postnatal (P1) coronal sutures with or without dura mater to the midparietal bones of syngeneic adult rats identified a role for dura mater in maintaining patency in postnatal coronal sutures (Opperman et al., 1993). Neither E19 nor P1 explants were able to resist osseous obliteration in the absence of their native, suture-associated dura mater, whereas explants of both sutures transplanted with their dura mater intact remained patent. Similar results were obtained when E19 rat coronal sutures were placed in an in vitro organ culture system, with intact sutures remaining patent and those lacking their dura mater fusing (Opperman et al., 1995). These experiments demonstrated that factors and/or physical cues from the dura mater underlying coronal sutures play a key role in the maintenance of late embryonic and early postnatal coronal suture patency. Subsequent studies by Opperman et al. (1996) were then performed to further characterize the nature of the dura mater–derived signals. In a novel set of experiments, conditioned media from dura mater cultures was added to coronal sutures placed in organ culture without the suture-associated dura mater. Not only was dura mater–conditioned media capable of preventing the expected osseous obliteration of the suture, but when conditioned media was divided into heparan-binding and non-binding fractions, only the heparan-binding associated media was capable of maintaining coronal suture patency. Thus, these experiments suggested that heparan-binding growth factors, such as FGFs and transforming growth factor- (TGF- ) superfamily members, were responsible for the observed modulations in coronal suture fate. Dura mater–mediated maintenance of suture patency has also been demonstrated in the sagittal suture (Kim et al., 1998). Embryonic (E16) mouse sagittal suture explants cultured without dura mater fused within 3 days, whereas control explants with intact dura mater remained patent. In contrast, no diVerences were detected in the patency of sagittal suture explants

302

Lenton et al.

from P1 mice cultured with or without the associated dura mater. Expression analysis of genes known to act in conserved signaling pathways throughout development revealed changing patterns of expression in the osteogenic fronts of the sagittal suture at the end of embryonic development, suggesting that suture morphogenesis is controlled by diVerent molecular mechanisms during embryonic and postnatal stages. It has been hypothesized that in the mouse sagittal suture, dura mater may be involved in the maintenance of patency prior to birth, whereas signals from the osteogenic fronts are primarily responsible for maintenance of suture patency during neonatal development (Kim et al., 1998). The potential for dura mater–derived signaling to regulate the process of physiological fusion of the posterior frontal suture has also been demonstrated. Separating the posterior frontal dura mater from the overlying suture either by temporary disruption of the dura mater–suture interface, or by interposition of an impermeable silicone membrane, significantly delayed fusion of the posterior frontal suture, implicating dura mater–suture interactions in the progression of physiological posterior frontal suture fusion (Roth et al., 1996). Evidence that soluble factors derived from the dura mater were associated with both the maintenance of coronal and sagittal suture patency and the process of programmed posterior frontal suture fusion led to the hypothesis that the dura mater is a regionally specialized tissue that guides the development and fate of the overlying sutures. This hypothesis was tested by Levine et al. (1998) by manipulating the orientation of sutures with respect to the underlying dura mater. A strip craniectomy incorporating the posterior  frontal and sagittal suture was rotated 180 such that the posterior frontal suture was placed in contact with dura mater normally associated with the sagittal suture, and the sagittal suture was in contact with dura mater normally associated with the posterior frontal suture. This arrangement resulted in fusion of the normally patent sagittal suture and continued patency of the posterior frontal suture in vivo. The results of the previous experiments by several investigative groups clearly demonstrate that signaling interactions occur between the dura mater underlying the coronal, sagittal, and posterior frontal sutures and components of the suture itself (osteoblasts and/or mesenchymal cells). These studies also suggest that regional variation of the dura mater is responsible, at least in part, for the very diVerent fates encountered by these sutures in vivo. 3. Dura Mater–Osteoblast Interactions The hypothesis that paracrine signaling occurs between calvarial osteoblasts and the underlying dura mater has been investigated in vivo in a co-culture model (Spector et al., 2002). In this model, calvarial osteoblasts were

9. Cranial Suture Biology

303

cultured in combination with non-suture-associated dura mater from neonatal rats. This culture system was designed to mimic the in vivo anatomical arrangement of these cell populations while preventing direct cell-to-cell contact in order to establish the role of soluble paracrine factors in duramater–osteoblast interactions. Compared with osteoblasts cultured alone, osteoblasts co-cultured with non-suture-associated dura mater demonstrated increased rates of proliferation and expressed increased levels of collagen 1, osteopontin, osteocalcin, and alkaline phosphatase mRNA. Co-cultured osteoblasts also produced significantly greater numbers of mineralized bone nodules, an in vitro surrogate for bone formation. Interestingly, transcripts of the heparan-binding osteogenic growth factors Tgf 1 and Fgf2 were expressed at higher levels in cultured cells isolated from dura mater compared with osteoblasts, suggesting that the dura mater may be the primary source of these ligands in vivo. These data confirmed that immature, non-suture-associated dura mater can influence the biological activity of calvarial osteoblasts. Phenotypic diVerences have been observed in primary cultures of rat cells harvested from posterior frontal or sagittal suture–associated dura mater (Mehrara et al., 1999a). First passage dural cells microdissected and harvested from under the sagittal sutures of P6 rat pups demonstrated decreased cellular contact inhibition and increased proliferation compared with dural cells harvested from under the posterior frontal suture. In contrast, the amount of alkaline phosphatase activity and collagen 1 protein expression was more than double in dural cells isolated from under the posterior frontal suture compared with dural cells harvested from under the sagittal suture. These observations led to the exploitation of the in vitro co-culture assay to determine if paracrine signaling derived from regional dura mater was capable of inducing diVerent biological eVects in calvarial osteoblasts (Warren et al., 2003b). In these experiments, primary calvarial osteoblasts were co-cultured with dura mater cells derived from the dura mater underlying either the sagittal or posterior frontal suture. Expression analysis of alkaline phosphatase, bone sialoprotein, osteopontin, and osteocalcin mRNA demonstrated increased levels of these transcripts in osteoblasts co-cultured with posterior frontal-associated dural cells compared with osteoblasts co-cultured with sagittal-associated dural cells. These data suggest that regional dura mater isolated from the posterior frontal suture enhances the expression of osteogenic genes to a greater extent than sagittal suture-derived dural cells. The observation that suture-derived dural cells have an osteoinductive role in vitro may have implications for suture biology in vivo.

304

Lenton et al.

V. Molecular and Cellular Mechanisms Governing Suture Morphogenesis In general, guidance for laboratory investigations into the molecular and cellular biology regulating cranial suture fusion and patency has primarily been derived from clinical genetics data. Thus, the association of Saethre– Chotzen syndrome with loss-of-function mutations in TWIST, and Bostontype craniosynostosis with gain-of-function mutations in MSX2, has given researchers clues as to the possible roles of these genes in both cranial suture development and osteoblast diVerentiation (El Ghouzzi et al., 1997; Howard et al., 1997; Muller et al., 1993; Warman et al., 1993). Similarly, the wealth of genetic data linking activating mutations in FGFR1, FGFR2, and FGFR3 has spurred researchers to examine this signaling pathway in great detail, investigating the expression and function of both FGF ligands and receptors (Bellus et al., 1996; Muenke et al., 1997, 1994; Tavormina et al., 1995; Wilkie et al., 1995b). However, not all insights into cranial suture biology have been born of genetic clues. Observations that heparan-binding growth factors from the dura mater were important regulators of cranial suture development led to investigation of the role of TGF- in craniosynostosis (Opperman et al., 1996). Other avenues of investigation have taken advantage of recent advances in genomic profiling, using techniques such as diVerential-display polymerase chain reaction (PCR) and microarray analysis to suggest candidate genes with a role in physiological, and therefore potentially pathological, suture fusion (Nacamuli et al., 2004a; Ting et al., 1999; Warren et al., 2003a). In the following section we review advances in basic science research that have been used to gain insight into cranial suture development and pathology.

A. Growth Factors 1. TGF-b The TGF- superfamily includes a large number of signaling molecules with central roles in processes ranging from embryogenesis to tissue formation and wound repair (Gold et al., 1997; Kingsley, 1994). Family members include the eponymous TGF- isoforms (1, 2, and 3), bone morphogenetic proteins (BMPs), Activin, Nodal, and growth and diVerentiation factor 5 (GDF-5). These growth factors all utilize a pair of transmembrane receptors (types I and II) that dimerize in the presence of ligand and activate a serine/threonine protein kinase. Activated kinases then phosphorylate SMAD proteins, which translocate to the nucleus and form multisubunit transcriptional activators (Massague and Chen, 2000).

9. Cranial Suture Biology

305

Members of the TGF- superfamily are abundant in bone, being both expressed by osteoblasts and bound to the extracellular matrix of skeletal tissue (Centrella et al., 1994). It is well established that TGF- s are intimately associated with osteoblast biology, skeletogenesis, bone remodeling, and fracture repair (Bouletreau et al., 2000; Erlebacher et al., 1998; FilvaroV et al., 1999). Opperman et al. (2000, 1999, 1997) have demonstrated that during development, tissue interactions between the dura mater and the overlying cranial sutures are mediated (at least in part) by Tgf- isoforms. These studies demonstrated that in the context of an embryonic mouse coronal suture organ culture model, specific Tgf- isoforms are involved with the determination of fusion or patency. Specifically, coronal sutures from E19 rats cultured with their subjacent dura mater remained patent, but upon the addition of an anti-Tgf- 3 antibody would fuse. Similarly, culture of the coronal suture in the absence of the associated dura mater led to suture fusion, whereas addition of an anti-Tgf- 2 antibody was associated with patent sutures (Opperman et al., 1999). Thus, Tgf- 3 is associated with patency and Tgf- 2 with fusion of the embryonic rat coronal suture. Similarly, both dura mater and Tgf- signaling have been implicated in fusion of the posterior frontal suture. Localization of Tgf- 1 and Tgf- 2 mRNA and/or protein in the rat posterior frontal suture and associated dura mater is increased during the window of fusion when compared with the patent sagittal suture (Most et al., 1998; Roth et al., 1997a). Furthermore, immunostaining of Tgf- receptor isoforms I and II demonstrated intense immunoreactivity in the dura mater under the posterior frontal suture and in osteoblasts at the posterior frontal suture margins during posterior frontal suture fusion, relative to osteoblasts and dura mater associated with the patent sagittal suture (Mehrara et al., 1999b). Spatial examination of the expression of Tgf- ligand and receptors in these studies has shown that expression is strongest in the dura mater underlying the posterior frontal suture. Specific studies of the dura mater by mRNA analysis of freshly isolated or cultured dura mater underlying the fusing posterior frontal suture or patent sagittal suture have also confirmed that Tgf- 1 expression is markedly elevated in dura mater associated with the posterior frontal suture (Greenwald et al., 2000). In contrast, examination of the suture complexes minus their associated dura mater has demonstrated equivalent levels of Tgf- 1 mRNA in the posterior frontal and sagittal sutures (Spector et al., 2000). Similar findings have also been obtained from the sutures of mice (Gosain et al., 2004; Nacamuli et al., 2004b; Sagiroglu et al., 1999). Functional studies have corroborated the previous findings and demonstrated that Tgf- signaling very much plays a role in the regulation of posterior frontal suture fusion. Mehrara et al. (2002) and Song et al.

306

Lenton et al.

(2004a) have manipulated Tgf- signaling in vitro in a postnatal mouse organ culture model. Results from these studies demonstrated that by over-expressing a dominant-negative Tgf- type II receptor in the dura mater under the posterior frontal suture (thereby blocking signal transduction by all three Tgf- ligands), suture fusion could be abrogated (Mehrara et al., 2002). The absence of suture fusion may have been due to prevention of osteoblast diVerentiation, because osteoblasts co-cultured with dura mater cells in the presence of the dominant-negative receptor fail to downregulate expression of Msx2 (an inhibitor of osteoblast diVerentiation) or increase expression of the extracellular matrix molecule osteopontin (Song et al., 2004a). Other investigators have perturbed Tgf- signaling by delivering recombinant Tgf- 3 or antibodies to Tgf- 2 to the posterior frontal suture of rats, resulting in abnormal patency and further demonstrating that these two isoforms have complementary roles in physiological suture fusion (Moursi et al., 2003; Opperman et al., 2002). It is interesting to note that despite the widespread involvement of TGF- in appendicular and craniofacial skeletogenesis, and cranial suture biology in particular, there are no craniosynostotic syndromes associated with a mutation in the genes encoding either TGF- ligand or its cognate receptors. In fact, only one mutation in the TGF- superfamily exists, giving rise to Hunter–Thompson chondrodysplasia (Thomas et al., 1996). Despite this fact, human clinical data further implicate TGF- signaling in cranial suture pathology. TGF- 2 immunoreactivity is increased in synostotic as compared with patent sutures from patients with coronal craniosynostosis (Roth et al., 1997a). Conversely, greater TGF- 3 was noted in patent coronal sutures. Increased immunostaining for TGF- has also been shown in the aVected sutures of children with persistent plagiocephaly (Lin et al., 1997). Collectively, these data underscore the importance of TGF- signaling in normal suture patterning and development, and suggest that TGF- may be a downstream mediator of suture fusion in syndromic and non-syndromic craniosynostoses. 2. FGFs and FGFRs The significance of FGF signaling in premature suture fusion has been established by linkage of multiple FGFR gain-of-function mutations with craniosynostotic syndromes. Several FGF ligands have been investigated in relation to suture morphogenesis and pathology, including Fgf1, Fgf3, Fgf4, Fgf9, and Fgf18 (Carlton et al., 1998; Greenwald et al., 2001; Iseki et al., 1997, 1999; Kim et al., 1998; Liu et al., 2002; Mathijssen et al., 2000; Mehrara et al., 1998; Moore et al., 2002; Most et al., 1998; Ohbayashi et al., 2002; Rice et al., 2000; Sarkar et al., 2001; Spector et al., 2000). Of the FGF ligands, Fgf2 has been the most intensively studied, and an

9. Cranial Suture Biology

307

accumulating body of evidence points to a role for Fgf2 in suture morphogenesis and fate determination. Avian mesencephalic neural crest cells, destined to form the neurocranium, respond to Fgf2 in a dose-dependent fashion, with long-term exposure at high doses leading to bone formation (Sarkar et al., 2001). In a model of in vivo chick embryonic calvarial organ culture, inhibition of Fgf2 biological activity led to reduced calvarial osteogenesis through abrogation of cell proliferation and diVerentiation within the cranial tissue (Moore et al., 2002). Data from our laboratory have demonstrated a correlation between posterior frontal suture fusion and increased expression of Fgf2 mRNA and protein in the associated rat posterior frontal dura mater (Mehrara et al., 1998; Most et al., 1998; Spector et al., 2000). Furthermore, using an Fgf2 overexpression adenovirus, we have demonstrated aberrant fusion of normally patent rat coronal sutures, whereas inhibition of Fgf signal transduction in the posterior frontal suture was associated with abnormal patency in that suture (Greenwald et al., 2001). Normal suture growth and morphogenesis is dependent upon a balance between the proliferation of osteoprogenitors within the suture mesenchyme and diVerentiation to osteoblasts at the osteogenic fronts. FGF signaling appears to have a key role in maintaining this balance (Iseki et al., 1997, 1999; Kim et al., 1998; Song et al., 2004b). Iseki et al. demonstrated that in embryonic mouse coronal sutures, Fgfr2 transcripts were localized to regions of rapid cell proliferation at the osteogenic fronts and were found to be mutually exclusive with that of osteopontin, an early marker of osteoblast diVerentiation (Iseki et al., 1997, 1999; Morriss-Kay et al., 2001). Conversely, Fgf2 protein expression was most abundant in diVerentiated regions and low in the Fgfr2 expression domains. FGF2-mediated upregulation of the osteoblast diVerentiation markers osteopontin and osteonectin was concomitant with reduced rates of proliferation in mouse embryonic sutures (Iseki et al., 1999). In these studies, Fgfr1 expression was upregulated at the onset of diVerentiation in areas of highest Fgf2 concentration, whereas in regions of lower Fgf2 activity, both Fgfr2 expression and cellular proliferation rates were significantly increased (Iseki et al., 1997, 1999; Morriss-Kay et al., 2001). Consistent with observations in the coronal suture, application of exogenous FGF4 to the osteogenic fronts in the sagittal suture resulted in accelerated suture closure (Kim et al., 1998). These data suggest that Fgf modulates diVerential expression of Fgfr1 and Fgfr2 in a dose-dependent manner, and that signaling through Fgfr2 regulates cell proliferation, whereas Fgfr1 regulates osteoblast diVerentiation. Our laboratory has recently demonstrated that both the transcription and translation of Fgfr1 and Fgfr2 are altered in rat calvarial osteoblasts undergoing diVerentiation, tipping the balance towards increased levels of Fgfr1 (Song et al., 2004b).

308

Lenton et al.

The hypothesis that perturbations in Fgfr1 and Fgfr2 signaling alter the balance of osteoprogenitor proliferation and diVerentiation, leading to premature osseous obliteration of sutures, has been further investigated in a variety of mouse models. Fgfr2-null mutations cause lethality prior to skeletogenesis; however, conditional inactivation has provided a means to study the role of Fgfr2 in osteogenesis (Yu et al., 2003). Homologous introduction of cre recombinase into the Dermo-1 (Twist-2) gene locus resulted in cre expression in mesenchymal precursors of both the osteoblast and chondrocyte lineages. Inactivation of a floxed Fgfr2 allele with Dermo-1-cre produced a phenotype with dwarfism, multiple skeletal abnormalities, and decreased bone mineral density relative to age-matched controls. DiVerentiation to an osteoblast lineage was preserved; however, defective osteogenesis and decreased proliferation were clearly appreciated in the Fgfr2 conditional knockouts. Investigation of the sutures revealed sustained patency of the midline posterior frontal and sagittal sutures. Similar abrogation of Fgfr2-mediated signal transduction has been accomplished through insertion of a stop codon into exon 9 of the mouse Fgfr2 gene (Eswarakumar et al., 2002). This approach specifically eliminated the Fgfr2c splice variant while maintaining Fgfr2b transcription. Analysis of these mutants again showed skeletal abnormalities, including a shortened, rounded skull and multiple ossification defects. Surprisingly, although sagittal and posterior frontal sutures remained patent, isolated coronal suture fusion was observed in these mice. This has been suggested to be secondary to a decrease in osteoblast proliferation, with premature loss of the growth center noted only in the coronal suture. An Fgfr2 gain-of-function mouse has recently been generated by introduction of a Cys342Tyr replacement into Fgfr2c (Eswarakumar et al., 2004). The mutation is equivalent to the human FGFR2 mutation associated with Crouzon syndrome and the FGFR1 mutation associated with PfeiVer syndrome (Jabs et al., 1994; Muenke et al., 1994). Mice heterozygous for the mutation are viable and fertile with shortened face, protruding eyes, and premature fusion of coronal sutures. Histomorphometric analysis and analysis of cultured bone marrow stromal cells demonstrated a significant increase in osteoprogenitors in gain-of-function mice compared with wild-type littermates. This observation is consistent with the association of Fgfr2 signaling with osteoprogenitor proliferation. In contrast to the Fgfr2c loss-of-function mutants, the gain of function is associated with increased expression of Runx2 and osteopontin transcripts, indicating that Fgfr2c is upstream of these genes and that it regulates osteogenesis in early stages. These results complement those of Zhou et al. (2000) who demonstrated upregulation of Runx2 and osteocalcin transcripts in cells stimulated with Fgf ligand and in cells expressing activating Fgfr1 mutations as described later.

9. Cranial Suture Biology

309

Human craniosynostosis has been modeled in mice by the introduction of Fgfr mutations equivalent to those associated with specific craniosynostotic syndromes. A mouse Fgfr1 Pro250Arg mutation orthologous to the FGFR1 Pro252Arg missense mutation in PfeiVer syndrome has been introduced into mice using gene targeting (Zhou et al., 2000). Mice carrying the mutation exhibit multiple phenotypic abnormalities typical of PfeiVer syndrome, including premature sagittal and coronal suture fusion, facial asymmetry, and midface hypoplasia. Increased bone mineralization and elevated expression of osteopontin, osteocalcin, and bone sialoprotein are also associated with the mutation, suggesting accelerated osteoblast diVerentiation. RNA hybridization analysis demonstrated upregulated expression of Runx2, a transcription factor essential for osteogenesis. Taken together, these data suggest that the Fgfr1 gain-of-function mutation may precipitate premature suture fusion through promotion of osteoblast diVerentiation and bone formation. This same mutation has also been introduced into mice employing a novel bacterial artificial chromosome (Hajihosseini et al., 2004). In these mice, increasing expression levels of mutated Fgfr1 resulted in premature posterior frontal suture fusion detected as early as embryonic day (E) 18.5. Other phenotypic characteristics noted in this model include a shortened midface, curved maxilla, and fusion of zygomatic arch bones. Interestingly, no sagittal or coronal suture fusion was apparent on histological examination. Apert syndrome caused by a Ser252Trp substitution in FGFR2 has been modeled in mice engineered to express the orthologous Ser250Trp mutation (Chen et al., 2003). Mice carrying the mutation developed malformations similar to those found in Apert syndrome patients, including a brachycephalic phenotype secondary to premature coronal suture fusion. In contrast to the Fgfr1 Pro250Arg mutation, which was associated with an increase in markers of osteoblast diVerentiation, the Fgfr2 Ser250Trp mutation resulted in increased Bax expression, implicating apoptosis as a potential mediator of pathological suture fusion (Chen et al., 2003). In summary, given the abundant clinical data, and data from transgenic mouse data related to FGFs and FGFRs, the investigation of downstream gene activation in response to FGF signal transduction through specific FGF receptor isoforms will undoubtedly prove to be central to our comprehension of the molecular mechanisms underlying craniosynostosis. 3. BMPs and BMP Antagonists BMPs were originally identified by their ability to induce ectopic bone formation (Wozney et al., 1988). Subsequent studies have identified roles for BMPs in a range of developmental processes that include mesoderm induction, skeletal patterning, and limb development (Duprez et al., 1996;

310

Lenton et al.

Francis et al., 1994; Kingsley et al., 1992; Northrop et al., 1995). In situ hybridization analysis of the embryonic mouse sagittal suture demonstrated expression of both Bmp2 and Bmp4 in the osteogenic fronts, and expression of Bmp4 in the sagittal suture mesenchyme and in the dura mater (Kim et al., 1998). In postnatal sutures, Bmp4 protein was localized to the suture mesenchyme and osteogenic fronts in fusing posterior frontal and patent sagittal and coronal sutures (Warren et al., 2003a). The BMP antagonist Noggin has recently emerged as a protein closely involved with FGF2 activity and determination of suture fate (Warren et al., 2003a). Originally described in Xenopus by Smith and Harland, the 26-kDa protein Noggin was found to play a role in normal dorsal development (Smith and Harland, 1992). Noggin induces lateral mesodermal tissues to form more dorsal elements including muscle, heart, and the pronephros. Noggin was also noted to rescue dorsal patterning in ventralized embryos (Brunet et al., 1998). Interestingly, many of these eVects were found to oppose those of BMPs, which have been known to possess rival ventralizing potential. The function of Noggin as a BMP antagonist was first demonstrated in cultured murine bone marrow stromal cells (Takao et al., 1996). Noggin protein was found to bind Bmp4 with high aYnity and abolish Bmp4 activity by blocking binding to cognate cell-surface receptors (Zimmerman et al., 1996). Recent crystallographic analysis of Noggin bound to BMP7 has revealed a conformational shift in BMP7 that blocks the molecular interfaces of the binding epitopes for both type I and type II receptors, thereby inhibiting BMP signal transduction and its downstream sequelae (Groppe et al., 2003). Recent studies in our laboratory have suggested a role for the BMP antagonist Noggin in the maintenance of cranial suture patency. Analysis of gene expression has shown that although Bmp levels are relatively equivalent between fusing and patent mouse sutures, Noggin expression is primarily limited to patent sagittal and coronal sutures (Warren et al., 2003a). Furthermore, increased FGF signaling, through the application of ectopic FGF2 or transfection with gain-of-function FGFR2, suppresses Noggin expression by mouse osteoblasts in vitro. These findings suggest that Noggin inhibition of BMP activity may be critical for the maintenance of suture patency in the sagittal and coronal sutures. In contrast, FGF signaling in the posterior frontal suture may suppress Noggin expression, with the resultant unopposed BMP activity precipitating suture fusion. To further explore the role of Noggin in determining suture fate, adenoviral transfection was used to force Noggin expression in posterior frontal sutures in organ culture, resulting in abnormal, widely patent sutures after 30 days. This finding was confirmed in vivo, with mice demonstrating an increased intercanthal distance secondary to frontal bone growth and an abnormally patent posterior frontal suture noted after 50 days (Warren et al., 2003a).

9. Cranial Suture Biology

311

Complementary studies focusing on the abrogation of Noggin in sagittal and coronal sutures have been limited by lethality in Noggin-null mutants. Homozygous deletion of Noggin results in mice with multiple skeletal abnormalities, including defects in vertebrae, ribs, and limbs (McMahon et al., 1998). The generation of Noggin conditional knockout mice will assist in further defining the role Noggin plays in suture development and help to elucidate the interplay between Noggin and BMP signaling in determining suture fate. BMP3 has been shown to be an antagonist of BMP signaling, being capable of inhibiting ventralization in Xenopus laevis embryos and preventing BMP2-induced osteoblastic diVerentiation of multipotent mesenchymal cells (Daluiski et al., 2001). In contrast to Noggin, BMP3 does not inhibit BMP signaling by binding to BMP ligands. Rather, BMP3 mediates its eVects through Activin receptors, competing for shared SMAD proteins and thereby antagonizing BMP signaling. Microarray analysis of rat cranial sutures has identified a potential role for Bmp3 in suture biology (Nacamuli et al., 2004a). Bmp3 expression is increased in patent sagittal sutures relative to the posterior frontal suture during the period of predicted suture fusion. Similar to Noggin, expression of Bmp3 is suppressed in cultured primary calvarial osteoblasts by application of FGF2. Although these findings suggest a role for Bmp3 in the regulation of cranial suture patency, Bmp3-null mutants have yielded little information regarding suture development. Aside from increased bone mineral density, no other gross abnormalities have been appreciated in these mice (Daluiski et al., 2001). It remains to be determined whether Bmp3-mediated antagonizism of BMP signaling is critical or redundant with existing Noggin pathways. 4. IGF Relatively less attention has been directed toward the role of insulin-like growth factors (IGFs) in cranial suture fusion. Originally detected in bone matrix, two isoforms (IGF1 and IGF2) have been identified that are mitogenic for osteoblasts and stimulate production of type I collagen and osteocalcin (Canalis and Lian, 1988; Canalis et al., 1988; Frolik et al., 1988). Furthermore, receptors for IGFs have been identified on osteoblasts, highlighting a possible role for IGF1 and IGF2 in bone formation and repair (McCarthy et al., 1989). Analysis of biopsy samples from patients with single-suture craniosynostosis demonstrated increased intensity of IGF1 immunoreactivity in actively fusing sutures when compared with patent controls (Roth et al., 1997b). Interestingly, immunoreactivity for IGF1 was most concentrated along the osteogenic fronts. Complementing these findings, Bradley et al. (1999) noted similarly high levels of Igf1 peptide

312

Lenton et al.

within the large osteoblasts of the suture margin during the period of predicted posterior frontal fusion in Sprague-Dawley rats. In situ hybridization analysis localized Igf1 and Igf2 transcripts to the underlying posterior frontal dura mater, with expression levels increasing 2–10 days prior to onset of fusion. Similar levels of Igf mRNA and peptide were not observed in patent coronal sutures during this time period (Bradley et al., 1999). Together, these data suggest the involvement of IGFs in the determination of suture fate, possibly by promoting osteoblast proliferation and bone deposition. 5. NELL-1 DiVerential-display PCR has been used as a tool to search for novel genes expressed in fusing sutures from patients with unilateral coronal craniosynostosis. Using this approach, Ting et al. (1999) identified upregulated expression of NELL-1 in fusing and fused coronal sutures as compared with patent coronal sutures. NELL-1 is an 810-amino acid polypeptide containing six epidermal growth factor-like repeats and a hydrophobic amino terminus that may be a signal peptide. In situ hybridization studies localized human NELL-1 expression to mesenchymal cells and osteoblasts at the osteogenic front, as well as to condensing mesenchymal cells of bone at sites of fusion. Nell-1 was also shown to be expressed in cultures of rat calvarial osteoblasts, but not in tibial osteoblasts or fibroblasts, suggesting that Nell-1 has a specific role in intramembranous bone formation. Studies in mice have demonstrated that Nell-1 over-expression leads to increased osteoblast diVerentiation, apoptosis, and suture fusion (Zhang et al., 2002, 2003). 6. EPHRIN-B1 EPHRIN-B1 is a member of the Ephrin family of transmembrane ligands for Eph receptor tyrosine kinases. The murine ortholog of EFNB1 is preferentially expressed in rat brain and is thought to play a role in the developing nervous system (Contractor et al., 2002; Cowan et al., 2000; Tang et al., 1995) Efnb1 expression has also been localized to the frontonasal neural crest in mice (Flenniken et al., 1996). Heterozygous loss-of-function mutations in EFNB1 have been identified as the cause of craniofrontal syndrome (Twigg et al., 2004; Wieland et al., 2004). Craniofrontal syndrome is an X-linked craniofacial disorder that includes coronal synostosis in females with this disease. Males with craniofrontal syndrome are not as severely aVected as females and do not exhibit coronal synostosis. The majority of EFNB1 mutations result from single-nucleotide substitutions that result in splicing, nonsense, and missense mutations, with one example

9. Cranial Suture Biology

313

of a single-nucleotide deletion resulting in a frameshift. These mutations can be inherited or can arise de novo from unaVected parents (Twigg et al., 2004). In situ hybridization analysis of murine Efnb1 at E9.5 and E10.5 during formation of the frontal–parietal boundary demonstrated expression in the neural crest-derived mesenchyme destined to form the frontal bone, and the underlying neuroepithelium, but not in mesenchyme destined to form the parietal bone (Twigg et al., 2004). The coronal suture forms at a boundary between tissues derived from neural crest and paraxial mesoderm. The authors note that the pattern of Efnb1 expression reflects the anticipated pattern of expression for a molecule involved in boundary formation at the coronal suture. Future studies will be directed at characterizing the molecular interactions that determine boundary formation and positioning of the sutures. B. Transcription Factors 1. MSX2 Studies in transgenic mice have demonstrated a role for Msx2 in the regulation of cell proliferation, and support the notion that Msx2 dosage is critical for normal craniofacial development. Mice carrying an Msx2 transgene encoding the murine equivalent of the human Pro148His mutation were viable with increased osteogenic cell proliferation in the osteogenic fronts of calvarial bones in early postnatal stages and premature fusion of the sagittal suture (Liu et al., 1999). In this model, overexpression was estimated to be 2-fold or less (Ma et al., 1996). However, integration of 13–22 copies of the mutated MSX2 transgene resulted in perinatal lethality and severe craniofacial malformations, but without craniosynostosis (Winograd et al., 1997). The diVerence in the severity of the phenotypes generated in these studies may reflect diVerences in gene dosage (Wilkie, 1997). Msx2 mutant mice exhibit persistently unossified areas of calvaria reminiscent of human parietal foramina. The frontal bone is aVected in mice, whereas human MSX2 haploinsuYciency causes parietal foramina. The defect is more exaggerated in Msx2 homozygous null mice compared with heterozygotes (Ishii et al., 2003). Both apoptosis and deficient neural crest cell migration have been ruled out as contributors to the ossification deficiencies observed in Msx2 mutants. Instead, defects in diVerentiation and proliferation of the neural crest-derived population of skeletogenic mesenchymal cells that compose the frontal bone have been found to result in ossification defects that result in calvarial foramina (Ishii et al., 2003; Satokata et al., 2000). Exaggeration of the calvarial defects found in Msx2 mutant mice have been observed in genetic interactions with Alx4, Twist-1, and Msx1 (Antonopoulou et al., 2004; Ishii et al., 2003; Satokata et al., 2000).

314

Lenton et al.

2. TWIST Twist-1 expression is localized to osteoprogenitor cells within the suture mesenchyme of mouse embryonic coronal and sagittal sutures, but is decreased postnatally (Johnson et al., 2000; Rice et al., 2000). Twist-1-null mouse embryos die at E11.5 with defects in head mesenchyme, branchial arch formation, somite organization, limb buds, and failure of neural tube closure (Chen and Behringer, 1995). However, mice heterozygous for Twist1 are viable and fertile with phenotypic defects that parallel those found in Saethre–Chotzen syndrome. These include hind leg digit duplications and abnormalities in calvarial bones, including poorly developed squamosal bones and overdevelopment of the interparietal bones and supraoccipital bones (Bourgeois et al., 1998; Carver et al., 2002; El Ghouzzi et al., 1997). The majority of Twist-1 heterozygous mice are aVected by complete or partial synostosis of the coronal sutures (Carver et al., 2002). No significant diVerences in skull shape have been detected between Twist-1 heterozygous mice and wild-type littermates; however, some Twist-1 heterozygous mice exhibit torsion of the snout (Bourgeois et al., 1998; Carver et al., 2002). Given that craniosynostosis in Saethre–Chotzen syndrome is associated with TWIST-1 haploinsuYciency, the phenotype of Twist-1 heterozygous mice suggests that the role of the Twist-1 gene in suture morphogenesis is conserved between mice and humans. The expression pattern of Twist-1 in the cranial sutures, and the phenotypes associated with Twist-1 loss of function in mouse mutants and in Saethre–Chozen syndrome, suggest that Twist-1 may be involved in the regulation of osteoblast proliferation and diVerentiation. This hypothesis is supported by data from cell culture experiments (Lee et al., 1999; Murray et al., 1992). In cultured calvarial osteoprogenitor cells, downregulation of Twist-1 expression is correlated with osteoblast diVerentiation, whereas Twist-1 overexpression maintains cells in an osteoprogenitor-like state. Furthermore, forced expression of Twist-1 has been linked to uncontrolled proliferation in several experimental systems (Gullaud et al., 2003; Maestro et al., 1999; Pajer et al., 2003). Taken together, these data suggest that Twist1 may act as a master switch in initiating bone cell diVerentiation by regulating the proliferation and diVerentiation of the osteogenic cell lineage in a dose-dependent manner. The role of Twist-1 in osteogenesis and suture morphogenesis has been further elucidated by examination of the functional relationships between it and other factors involved in osteogenesis. In Drosophila, Twist protein is required for RNA expression of dfr1/htl (an orthologue of mammalian FGFRs) (Shishido et al., 1993). This observation has prompted analysis of the relationship between Twist and Fgf signaling in mice; however, a definitive interaction has not been established. Expression analysis of Twist-1 and Fgfrs in the embryonic mouse coronal

9. Cranial Suture Biology

315

suture revealed that Twist-1 expression precedes Fgfr expression in the developing suture (Johnson et al., 2000; Morriss-Kay et al., 2001). In later stages, Twist-1 is expressed in suture mesenchyme, whereas Fgfr expression is localized to the developing bone; however, there is some overlap of Twist-1 and Fgfr2 expression in immature proliferating cells (Johnson et al., 2000; Morriss-Kay et al., 2001). In the embryonic sagittal suture, Twist-1 and Fgf2 are expressed in an overlapping pattern in the midsutural mesenchyme (Rice et al., 2000). Interestingly, in E15 mouse calvarial organ culture, FGF2-soaked beads induced Twist-1 expression in suture mesenchyme with concomitant inhibition of the osteoblast diVerentiation marker bone sialoprotein (Rice et al., 2000). Furthermore, in Twist-1þ/ mice, Fgfr2 protein expression was localized in an ectopic location in the midsutural mesenchyme (Rice et al., 2000). These data are consistent with an FGF-mediated induction of Twist-1, which in turn regulates Fgfr2 protein expression (O’Rourke and Tam, 2002). A mechanism for Twist inhibition of osteoblast diVerentiation via transient inhibition of Runx2 function has recently been established in mice (Bialek et al., 2004). Runx2 is a master regulator of osteoblast diVerentiation and the earliest marker of the osteoblast lineage (Ducy et al., 1997; Komori et al., 1997; Otto et al., 1997). However, during development Runx2 expression precedes the appearance of osteoblasts by several days, suggesting that other regulatory proteins are involved in the initiation of osteogenesis. Twist-1 and Twist-2 are transcribed in cells expressing Runx2 throughout the skeleton during early development, but osteoblast-specific gene expression in these cells does not occur until Twist expression is downregulated. These observations prompted Bialek et al. (2004) to study osteoblast diVerentiation in a series of mouse mutants with heterozygous deletions of Twist and Runx2. As described earlier, Twist-1þ/ mice develop large interparietal bones and exhibit premature fusion of the coronal sutures. A comparison of osteocalcin gene expression in the frontal and parietal bones of Twist-1þ/ and wild-type mice indicates that Twist-1 haploinsuYciency results in premature osteoblast diVerentiation, leading to coronal synostosis. In contrast, Runx2þ/ mice have delayed ossification of the cranial bones, resulting in open fontanelles and widely spaced cranial sutures; however, this phenotype is rescued in mice that also carry a heterozygous deletion of Twist-1 (Bialek et al., 2004; Otto et al., 1997). Specifically, Runx2þ/Twist-1þ/ mice have a normally shaped skull, interparietal bones close to normal size, and patent coronal sutures. Further analyses revealed that Twist-1 overexpression inhibits osteoblast diVerentiation without aVecting Runx2 expression. The antiosteogenic eVects of Twist are mediated by the Twist box, which interacts with the Runx2 DNA-binding domain and inhibits Runx2 function. Together these results demonstrate that Twist is an inhibitor of osteoblast diVerentiation (Bialek et al., 2004).

316

Lenton et al.

In Saethre–Chozen syndrome, frameshift mutations and mutations that result in premature termination before the TWIST-1 C-terminal domain have the potential to eliminate a functional TWIST box, which would be expected to result in precocious osteoblast diVerentiation and premature suture fusion. However, the mechanism of missense mutations associated with Saethre–Chozen syndrome cannot be accounted for in this model, suggesting that other mechanisms may be at play (Bialek et al., 2004).

C. Apoptosis Apoptosis has putative roles in sutural development, as well as bone formation and resorption; however, the exact function of apoptosis in normal and pathological cranial suture biology is unknown (Rice et al., 1999). Both increased and decreased rates of apoptosis have been detected in cultured osteoblasts derived from patients with syndromic craniosynostosis (Dry et al., 2001; Lemonnier et al., 2001). The results of several experimental studies have demonstrated increased apoptosis in patent sutures, and others have associated increased apoptosis with suture fusion (Agresti et al., 2003; Furtwangler et al., 1985; Lemonnier et al., 2001; Opperman et al., 2000). Recently, overexpression of the craniosynostosis-associated gene Nell-1 has been shown to lead to multiple craniofacial abnormalities (Ting et al., 1999). Apoptosis assays demonstrated Nell-1 expression associated with increased levels of apoptosis in the osteoblasts lining the osteogenic fronts of the fusing sutures (Zhang et al., 2003). We have recently utilized quantitative, in vivo imaging of programmed cell death to compare apoptosis in the posterior frontal and sagittal sutures of rats during the period of posterior frontal suture fusion. The data indicate that there are no significant diVerences in the level of apoptotic activity between the physiologically fusing posterior frontal suture and the patent sagittal suture (Fong et al., 2004). In contrast, the results of microarray analysis indicate distinct, suture-specific expression patterns of genes associated with specific apoptotic pathways. In the patent sagittal suture, expression of genes associated with death receptor-mediated apoptosis is increased relative to the fusing posterior frontal suture. Conversely, in the fusing posterior frontal suture transcription of genes associated with mitochondrial apoptosis is increased relative to the sagittal suture (Fong et al., 2004). These observations imply that diVerent pathways of programmed cell death are activated in fusing and nonfusing sutures, suggesting that specific populations of cells may be targeted for apoptosis in each suture, possibly in response to suture-specific patterns of growth factor expression. Further studies are required to elucidate the exact function of apoptosis in the cranial sutures.

9. Cranial Suture Biology

317

VI. Conclusions and Perspectives For over 100 years, craniosynostosis has been diagnosed on the basis of clinical features, with phenotypic overlap commonly noted between syndromes. Although Crouzon first realized in the early 1900s that there was a potential link between inheritance and premature suture fusion, it is only in the past decade that specific advances have been made in the identification of the genetic basis of syndromic craniosynostoses. The association of syndromic craniosynostoses with mutations in the FGFRs and the transcription factors TWIST and MSX2 has highlighted a role for these genes in normal and pathological suture morphogenesis. The linkage of specific genetic mutations with clinically defined craniosynostotic syndromes, together with genetic screening of patients, has revealed that identical mutations can be associated with multiple syndromes, thereby providing a genetic basis for at least some of the observed phenotypic overlap. Mechanistically, the association of specific phenotypes with diVerent gene mutations implies that the mutated gene products may function within the same signaling networks. Thus, an improved understanding of the genetic basis of craniosynostosis has proven to be advantageous in the clinical setting, oVering the potential for improved case management. Over the past 10 years, murine models have been utilized extensively to decipher the cellular and, more recently, molecular biology regulating physiological cranial suture patency and fusion. The murine model system has provided us with a wealth of information regarding the importance of tissuespecific interactions and growth factor expression in suture biology. Furthermore, as the genetics of the craniosynostotic syndromes are defined, investigators are able to study the expression patterns of genes associated with craniosynostosis and manipulate their expression both in vitro and in vivo in order to analyze the functional eVects of genetic mutations that lead to craniosynostosis in humans. Numerous transgenic mice carrying mutations in genes and in elements of signal transduction pathways related to craniosynostosis are helping to elucidate the molecular basis of normal suture development and pathology. Analyses of the temporal and spatial pattern of gene expression in rodents and the characterization of transgenic mice have begun to elucidate the interplay between the products of genes mutated in craniosynostosis. Several key themes have emerged from these studies, providing insights into the nature of both physiological and pathological suture fusion. The balance between osteoblast proliferation and diVerentiation appears to becritical for normal suture morphogenesis and maintenance of patency, with gain-of-function FGFR mutations shifting the balance from osteoblast proliferation to diVerentiation. The BMP inhibitor Noggin may be a key

318

Lenton et al.

downstream regulator of FGF signaling, determining when and where bone formation occurs. TWIST has a demonstrated role in osteoblast diVerentiation, its presence being a negative regulator of osteogenesis. Alterations in MSX2 gene dosage result in either ossification defects or precocious ossification and suture fusion, suggesting that it is critically involved with proliferation and diVerentiation of skeletogenic mesenchyme. What is the ultimate goal of our investigations into cranial suture biology? The main objective is to obtain a thorough understanding of normal and pathological suture morphogenesis and development. Armed with this knowledge, researchers will be prepared to devise biologically based therapeutic strategies that could be used either in utero or postnatally to prevent craniosynostosis, potentially alleviating any adverse sequelae and avoiding the morbidity of current surgical approaches.

Acknowledgments This work was funded by grants from the Oak Foundation and NIH R01-13194 to M.T.L., and from the ACS to R.P.N.

References Aalami, O. O., Nacamuli, R. P., Lenton, K. A., Cowan, C. M., Fang, T. D., et al. (2004). Applications of a mouse model of calvarial healing: DiVerences in regenerative abilities of juveniles and adults. Plast. Reconstr. Surg. 114, 713–720. Agresti, M., Schaefer, R. B., Recinos, R. F., and Gosain, A. K. (2003). Detection of apoptosis in normal postnatal mouse cranial sutures. Plastic Surgery Research Council, Las Vegas, NV. Anderson, J., Burns, H. D., Enriquez-Harris, P., Wilkie, A. O. M., and Heath, J. K. (1998). Apert syndrome mutations in fibroblast growth factor receptor 2 exhibit increased aYnity for FGF ligand. Hum. Mol. Genet. 7, 1475–1483. Antonopoulou, I., Mavrogiannis, L. A., Wilkie, A. O., and Morriss-Kay, G. M. (2004). Alx4 and Msx2 play phenotypically similar and additive roles in skull vault diVerentiation. J. Anat. 204, 487–499. Babler, W. J., Persing, J. A., Persson, K. M., Winn, H. R., Jane, J. A., et al. (1982). Skull regrowth after coronal suterectomy, periostectomy, and dural transection. J. Neurosurg. 56, 529–535. Bellus, G. A., Gaudenz, K., Zackai, E. H., Clarke, L. A., Szabo, J., et al. (1996). Identical mutations in three diVerent fibroblast growth factor receptor genes in autosomal dominant craniosynostosis syndromes. Nat. Genet. 14, 174–176. Bialek, P., Kern, B., Yang, X., Schrock, M., Sosic, D., et al. (2004). A twist code determines the onset of osteoblast diVerentiation. Dev. Cell 6, 423–435. Bouletreau, P. J., Steinbrech, D., Spector, J. A., Warren, S. M., Greenwald, J. A., et al. (2000). Gene expression of transforming growth factor-beta 3 and tissue inhibitor of metalloproteinase type 1 during membranous bone healing in rats. J. Craniofac. Surg. 11, 521–526. Bourgeois, P., Bolcato-Bellemin, A. L., Danse, J. M., Bloch-Zupan, A., Yoshiba, K., et al. (1998). The variable expressivity and incomplete penetrance of the twist-null heterozygous

9. Cranial Suture Biology

319

mouse phenotype resemble those of human Saethre-Chotzen syndrome. Hum. Mol. Genet. 7, 945–957. Bradley, J. P., Han, V. K., Roth, D. A., Levine, J. P., McCarthy, J. G., et al. (1999). Increased IGF-I and IGF-II mRNA and IGF-I peptide in fusing rat cranial sutures suggest evidence for a paracrine role of insulin-like growth factors in suture fusion. Plast. Reconstr. Surg. 104, 129–138. Brunet, L. J., McMahon, J. A., McMahon, A. P., and Harland, R. M. (1998). Noggin, cartilage morphogenesis, and joint formation in the mammalian skeleton. Science 280, 1455–1457. Butow, K. W. (1990). Craniofacial growth disturbance after skull base and associated suture synostoses in the newborn chacma baboon: A preliminary report. Cleft Palate J. 27, 241–251. Canalis, E., and Lian, J. B. (1988). EVects of bone associated growth factors on DNA, collagen and osteocalcin synthesis in cultured fetal rat calvariae. Bone 9, 243–246. Canalis, E., McCarthy, T., and Centrella, M. (1988). Isolation of growth factors from adult bovine bone. Calcif. Tissue Int. 43, 346–351. Carlton, M. B., Colledge, W. H., and Evans, M. J. (1998). Crouzon-like craniofacial dysmorphology in the mouse is caused by an insertional mutation at the Fgf3/Fgf4 locus. Dev. Dyn. 212, 242–249. Carver, E. A., Oram, K. F., and Gridley, T. (2002). Craniosynostosis in Twist heterozygous mice: A model for Saethre-Chotzen syndrome. Anat. Rec. 268, 90–92. Cassileth, L. B., Bartlett, S. P., Glat, P. M., Gripp, K. W., Muenke, M., et al. (2001). Clinical characteristics of patients with unicoronal synostosis and mutations of fibroblast growth factor receptor 3: A preliminary report. Plast. Reconstr. Surg. 108, 1849–1854. Castanon, I., and Baylies, M. K. (2002). A Twist in fate: Evolutionary comparison of Twist structure and function. Gene 287, 11–22. Centrella, M., and Canalis, E. (1985). Transforming and nontransforming growth factors are present in medium conditioned by fetal rat calvariae. Proc. Natl. Acad. Sci. USA 82, 7335–7339. Centrella, M., Horowitz, M. C., Wozney, J. M., and McCarthy, T. L. (1994). Transforming growth factor-beta gene family members and bone. Endocr. Rev. 15, 27–39. Chai, Y., Jiang, X., Ito, Y., Bringas, P., Jr., Han, J., et al. (2000). Fate of the mammalian cranial neural crest during tooth and mandibular morphogenesis. Development 127, 1671–1679. Chellaiah, A. T., McEwen, D. G., Werner, S., Xu, J., and Ornitz, D. M. (1994). Fibroblast growth factor receptor (FGFR) 3. Alternative splicing in immunoglobulin-like domain III creates a receptor highly specific for acidic FGF/FGF-1. J. Biol. Chem. 269, 11620–11627. Chen, L., Li, D., Li, C., Engel, A., and Deng, C. X. (2003). A Ser250Trp substitution in mouse fibroblast growth factor receptor 2 (Fgfr2) results in craniosynostosis. Bone 33, 169–178. Chen, Z. F., and Behringer, R. R. (1995). Twist is required in head mesenchyme for cranial neural tube morphogenesis. Genes Dev. 9, 686–699. Christensen, F. K., and Clark, D. B. (1970). The eVect of restricted suture growth on brain growth in dogs. Surg. Forum 21, 439–440. Chun, K., Teebi, A. S., Jung, J. H., Kennedy, S., Laframboise, R., et al. (2002). Genetic analysis of patients with the Saethre-Chotzen phenotype. Am. J. Med. Genet. 110, 136–143. Cohen, M. M., Jr. (1997a). Transforming growth factor beta s and fibroblast growth factors and their receptors: Role in sutural biology and craniosynostosis. J. Bone Miner. Res. 12, 322–331. Cohen, M. M., Jr. (1997b). Transforming growth factor betas and fibroblast growth factors and their receptors: Role in sutural biology and craniosynostosis. J. Bone Miner. Res. 12, 322–331. Contractor, A., Rogers, C., Maron, C., Henkemeyer, M., Swanson, G. T., et al. (2002). Transsynaptic Eph receptor-ephrin signaling in hippocampal mossy fiber LTP. Science 296, 1864–1869.

320

Lenton et al.

Couly, G. F., Coltey, P. M., and Le Douarin, N. M. (1993). The triple origin of skull in higher vertebrates: A study in quail-chick chimeras. Development 117, 409–429. Cowan, C. A., Yokoyama, N., Bianchi, L. M., Henkemeyer, M., and Fritzsch, B. (2000). EphB2 guides axons at the midline and is necessary for normal vestibular function. Neuron 26, 417–430. Crouzon, O. (1912). Dysostose cranio-faciale he´ re´ ditaire. Bull. Mem. Soc. Med. Hop. Paris 33, 545–555. Daluiski, A., Engstrand, T., Bahamonde, M. E., Gamer, L. W., Agius, E., et al. (2001). Bone morphogenetic protein-3 is a negative regulator of bone density. Nat. Genet. 27, 84–88. Dry, G. M., Yasinskaya, Y. I., Williams, J. K., Ehrlich, G. D., Preston, R. A., et al. (2001). Inhibition of apoptosis: A potential mechanism for syndromic craniosynostosis. Plast. Reconstr. Surg. 107, 425–432. Ducy, P., Zhang, R., GeoVroy, V., Ridall, A. L., and Karsenty, G. (1997). Osf2/Cbfa1: A transcriptional activator of osteoblast diVerentiation. Cell 89, 747–754. Duprez, D., Bell, E. J., Richardson, M. K., Archer, C. W., Wolpert, L., et al. (1996). Overexpression of BMP-2 and BMP-4 alters the size and shape of developing skeletal elements in the chick limb. Mech. Dev. 57, 145–157. El Ghouzzi, V., Le Merrer, M., Perrin-Schmitt, F., Lajeunie, E., Benit, P., et al. (1997). Mutations of the TWIST gene in the Saethre-Chotzen syndrome. Nat. Genet. 15, 42–46. Enlow, D. H. (2000). Normal Craniofacial Growth. In ‘‘Craniosynostosis Diagnosis Evaluation and Management’’ (M. M. Cohen, Jr. and R. E. MacLean, Eds.), pp. 35–47. Oxford University Press, New York. Erlebacher, A., FilvaroV, E. H., Ye, J. Q., and Derynck, R. (1998). Osteoblastic responses to TGF-beta during bone remodeling. Mol. Biol. Cell 9, 1903–1918. Eswarakumar, V. P., Horowitz, M. C., Locklin, R., Morriss-Kay, G. M., and Lonai, P. (2004). A gain-of-function mutation of Fgfr2c demonstrates the roles of this receptor variant in osteogenesis. Proc. Natl. Acad. Sci. USA 101, 12555–12560. Eswarakumar, V. P., Monsonego-Ornan, E., Pines, M., Antonopoulou, I., Morriss-Kay, G. M., et al. (2002). The IIIc alternative of Fgfr2 is a positive regulator of bone formation. Development 129, 3783–3793. Ferreira, J. C., Carter, S. M., Bernstein, P. S., Jabs, E. W., Glickstein, J. S., et al. (1999). Second-trimester molecular prenatal diagnosis of sporadic Apert syndrome following suspicious ultrasound findings. Ultrasound Obstet. Gynecol. 14, 426–430. FilvaroV, E., Erlebacher, A., Ye, J., Gitelman, S. E., Lotz, J., et al. (1999). Inhibition of TGF-beta receptor signaling in osteoblasts leads to decreased bone remodeling and increased trabecular bone mass. Development 126, 4267–4279. Flenniken, A. M., Gale, N. W., Yancopoulos, G. D., and Wilkinson, D. G. (1996). Distinct and overlapping expression patterns of ligands for Eph-related receptor tyrosine kinases during mouse embryogenesis. Dev. Biol. 179, 382–401. Fong, K. D., Song, H. M., Nacamuli, R. P., Franc, B. L., Mari, C., et al. (2004). Apoptosis in a rodent model of cranial suture fusion: In situ imaging and gene expression analysis. Plast. Reconstr. Surg. 113, 2037–2047. Francis, P. H., Richardson, M. K., Brickell, P. M., and Tickle, C. (1994). Bone morphogenetic proteins and a signalling pathway that controls patterning in the developing chick limb. Development 120, 209–218. Frolik, C. A., Ellis, L. F., and Williams, D. C. (1988). Isolation and characterization of insulinlike growth factor-II from human bone. Biochem. Biophys. Res. Commun. 151, 1011–1018. Furtwangler, J. A., Hall, S. H., and Koskinen-MoVett, L. K. (1985). Sutural morphogenesis in the mouse calvaria: The role of apoptosis. Acta Anat. (Basel) 124, 74–80. Gold, L. I., Sung, J. J., Siebert, J. W., and Longaker, M. T. (1997). Type I (RI) and type II (RII) receptors for transforming growth factor-beta isoforms are expressed subsequent to

9. Cranial Suture Biology

321

transforming growth factor-beta ligands during excisional wound repair. Am. J. Pathol. 150, 209–222. Gosain, A. K., Recinos, R. F., Agresti, M., and Khanna, A. K. (2004). TGF-beta1, FGF-2, and receptor mRNA expression in suture mesenchyme and dura versus underlying brain in fusing and nonfusing mouse cranial sutures. Plast. Reconstr. Surg. 113, 1675–1684. Greenwald, J. A., Mehrara, B. J., Spector, J. A., Warren, S. M., Crisera, F. E., et al. (2000). Regional diVerentiation of cranial suture-associated dura mater in vivo and in vitro: Implications for suture fusion and patency. J. Bone Miner. Res. 15, 2413–2430. Greenwald, J. A., Mehrara, B. J., Spector, J. A., Warren, S. M., Fagenholz, P. J., et al. (2001). In vivo modulation of FGF biological activity alters cranial suture fate. Am. J. Pathol. 158, 441–452. Gripp, K. W., Zackai, E. H., and Stolle, C. A. (2000). Mutations in the human TWIST gene. Hum. Mutation 15, 150–155. Groppe, J., Greenwald, J., Wiater, E., Rodriguez-Leon, J., Economides, A. N., et al. (2003). Structural basis of BMP signaling inhibition by Noggin, a novel twelve-membered cystine knot protein. J. Bone Joint Surg. Am. 85-A(Suppl. 3), 52–58. Gullaud, M., Delanoue, R., and Silber, J. (2003). A Drosophila model to study the functions of TWIST orthologs in apoptosis and proliferation. Cell Death DiVer. 10, 641–651. Hajihosseini, M. K., Lalioti, M. D., Arthaud, S., Burgar, H. R., Brown, J. M., et al. (2004). Skeletal development is regulated by fibroblast growth factor receptor 1 signalling dynamics. Development 131, 325–335. Hobar, P. C., Schreiber, J. S., McCarthy, J. G., and Thomas, P. A. (1993). The role of the dura in cranial bone regeneration in the immature animal. Plast. Reconstr. Surg. 92, 405–410. Howard, T. D., Paznekas, W. A., Green, E. D., Chiang, L. C., Ma, N., et al. (1997). Mutations in TWIST, a basic helix-loop-helix transcription factor, in Saethre-Chotzen syndrome. Nat. Genet. 15, 36–41. Hunter, A. G., and Rudd, N. L. (1976). Craniosynostosis. I. Sagittal synostosis: Its genetics and associated clinical findings in 214 patients who lacked involvement of the coronal suture(s). Teratology 14, 185–193. Hunter, A. G., and Rudd, N. L. (1977). Craniosynostosis. II. Coronal synostosis: Its familial characteristics and associated clinical findings in 109 patients lacking bilateral polysyndactyly or syndactyly. Teratology 15, 301–309. Ibrahimi, O. A., Eliseenkova, A. V., Plotnikov, A. N., Yu, K., Ornitz, D. M., et al. (2001). Structural basis for fibroblast growth factor receptor 2 activation in Apert syndrome. Proc. Natl. Acad. Sci. USA 98, 7182–7187. Ibrahimi, O. A., Zhang, F., Eliseenkova, A. V., Linhardt, R. J., and Mohammadi, M. (2004). Proline to arginine mutations in FGF receptors 1 and 3 result in PfeiVer and Muenke craniosynostosis syndromes through enhancement of FGF binding aYnity. Hum. Mol. Genet. 13, 69–78. Iseki, S., Wilkie, A. O., Heath, J. K., Ishimaru, T., Eto, K., et al. (1997). Fgfr2 and osteopontin domains in the developing skull vault are mutually exclusive and can be altered by locally applied FGF2. Development 124, 3375–3384. Iseki, S., Wilkie, A. O., and Morriss-Kay, G. M. (1999). Fgfr1 and Fgfr2 have distinct diVerentiation- and proliferation-related roles in the developing mouse skull vault. Development 126, 5611–5620. Ishii, M., Merrill, A. E., Chan, Y. S., Gitelman, I., Rice, D. P., et al. (2003). Msx2 and Twist cooperatively control the development of the neural crest-derived skeletogenic mesenchyme of the murine skull vault. Development 130, 6131–6142. Jabs, E. W. (2001). A TWIST in the fate of human osteoblasts identifies signaling molecules involved in skull development. J. Clin. Invest. 107, 1075–1077.

322

Lenton et al.

Jabs, E. W., Li, X., Scott, A. F., Meyers, G., Chen, W., et al. (1994). Jackson-Weiss and Crouzon syndromes are allelic with mutations in fibroblast growth factor receptor 2. Nat. Genet. 8, 275–279. Jabs, E. W., Muller, U., Li, X., Ma, L., Luo, W., et al. (1993). A mutation in the homeodomain of the human MSX2 gene in a family aVected with autosomal dominant craniosynostosis. Cell 75, 443–450. Jaye, M., Schlessinger, J. , and Dionne, C. A. (1992). Fibroblast growth factor receptor tyrosine kinases: Molecular analysis and signal transduction. Biochim. Biophys. Acta 1135, 185–199. Jiang, X., Iseki, S., Maxson, R. E., Sucov, H. M., and Morriss-Kay, G. M. (2002). Tissue origins and interactions in the mammalian skull vault. Dev. Biol. 241, 106–116. Johnson, D., Horsley, S. W., Moloney, D. M., Oldridge, M., Twigg, S. R., et al. (1998). A comprehensive screen for TWIST mutations in patients with craniosynostosis identifies a new microdeletion syndrome of chromosome band 7p21.1. Am. J. Hum. Genet. 63, 1282–1293. Johnson, D., Iseki, S., Wilkie, A. O., and Morriss-Kay, G. M. (2000). Expression patterns of Twist and Fgfr1, -2 and -3 in the developing mouse coronal suture suggest a key role for twist in suture initiation and biogenesis. Mech. Dev. 91, 341–345. Johnson, D. E., and Williams, L. T. (1993). Structural and functional diversity in the FGF receptor multigene family. Adv. Cancer Res. 60, 1–41. Kan, S. H., Elanko, N., Johnson, D., Cornejo-Roldan, L., Cook, J., et al. (2002). Genomic screening of fibroblast growth-factor receptor 2 reveals a wide spectrum of mutations in patients with syndromic craniosynostosis. Am. J. Hum. Genet. 70, 472–486. Kim, H. J., Rice, D. P., Kettunen, P. J., and ThesleV, I. (1998). FGF-, BMP- and Shh-mediated signalling pathways in the regulation of cranial suture morphogenesis and calvarial bone development. Development 125, 1241–1251. Kingsley, D. M. (1994). The TGF-beta superfamily: New members, new receptors, and new genetic tests of function in diVerent organisms. Genes Dev. 8, 133–146. Kingsley, D. M., Bland, A. E., Grubber, J. M., Marker, P. C., Russell, L. B., et al. (1992). The mouse short ear skeletal morphogenesis locus is associated with defects in a bone morphogenetic member of the TGF beta superfamily. Cell 71, 399–410. Komori, T., Yagi, H., Nomura, S., Yamaguchi, A., Sasaki, K., et al. (1997). Targeted disruption of Cbfa1 results in a complete lack of bone formation owing to maturational arrest of osteoblasts. Cell 89, 755–764. Lajeunie, E., Le Merrer, M., Bonaiti-Pellie, C., Marchac, D., and Renier, D. (1995). Genetic study of nonsyndromic coronal craniosynostosis. Am. J. Med. Genet. 55, 500–504. Lajeunie, E., Le Merrer, M., Bonaiti-Pellie, C., Marchac, D., and Renier, D. (1996). Genetic study of scaphocephaly. Am. J. Med. Genet. 62, 282–285. Lee, M. S., Lowe, G. N., Strong, D. D., Wergedal, J. E., and Glackin, C. A. (1999). TWIST, a basic helix-loop-helix transcription factor, can regulate the human osteogenic lineage. J. Cell. Biochem. 75, 566–577. Lemonnier, J., Hay, E., Delannoy, P., Fromigue, O., Lomri, A., et al. (2001). Increased osteoblast apoptosis in apert craniosynostosis: Role of protein kinase C and interleukin-1. Am. J. Pathol. 158, 1833–1842. Lenton, K. A. (2004). Personal communication. Levine, J. P., Bradley, J. P., Roth, D. A., McCarthy, J. G., and Longaker, M. T. (1998). Studies in cranial suture biology: Regional dura mater determines overlying suture biology. Plast. Reconstr. Surg. 101, 1441–1447. Li, L., Cserjesi, P., and Olson, E. N. (1995). Dermo-1: A novel twist-related bHLH protein expressed in the developing dermis. Dev. Biol. 172, 280–292. Lin, K., Nolen, A., Gampper, T., Jane, J., Opperman, L., et al. (1997). Elevated levels of transforming growth factors beta 2 and beta 3 in lambdoid sutures from children with persistent plagiocephaly. Cleft Palate Craniofac. J. 34, 331–337.

9. Cranial Suture Biology

323

Liu, Y. H., Tang, Z., Kundu, R. K., Wu, L., Luo, W., et al. (1999). Msx2 gene dosage influences the number of proliferative osteogenic cells in growth centers of the developing murine skull: A possible mechanism for MSX2-mediated craniosynostosis in humans. Dev. Biol. 205, 260–274. Liu, Z., Xu, J., Colvin, J. S., and Ornitz, D. M. (2002). Coordination of chondrogenesis and osteogenesis by fibroblast growth factor 18. Genes Dev. 16, 859–869. Ma, L., Golden, S., Wu, L., and Maxson, R. (1996). The molecular basis of Boston-type craniosynostosis: The Pro148–>His mutation in the N-terminal arm of the MSX2 homeodomain stabilizes DNA binding without altering nucleotide sequence preferences. Hum. Mol. Genet. 5, 1915–1920. Mabbutt, L. W., and Kokich, V. G. (1979). Calvarial and sutural re-development following craniectomy in the neonatal rabbit. J. Anat. 129, 413–422. Mabbutt, L. W., Kokich, V. G., MoVett, B. C., and Loeser, J. D. (1979). Subtotal neonatal calvariectomy. A radiographic and histological evaluation of calvarial and sutural redevelopment in rabbits. J. Neurosurg. 51, 691–696. Maestro, R., Dei Tos, A. P., Hamamori, Y., Krasnokutsky, S., Sartorelli, V., et al. (1999). Twist is a potential oncogene that inhibits apoptosis. Genes Dev. 13, 2207–2217. Manzanares, M. C., Goret-Nicaise, M., and Dhem, A. (1988). Metopic sutural closure in the human skull. J. Anat. 161, 203–215. Massague, J. , and Chen, Y. G. (2000). Controlling TGF-beta signaling. Genes Dev. 14, 627–644. Mathijssen, I. M., van Leeuwen, J. P., and Vermeij-Keers, C. (2000). Simultaneous induction of apoptosis, collagen type I expression and mineralization in the developing coronal suture following FGF4 and FGF2 application. J. Craniofac. Genet. Dev. Biol. 20, 127–136. McCarthy, T. L., Centrella, M., and Canalis, E. (1989). Insulin-like growth factor (IGF) and bone. Connect. Tissue Res. 20, 277–282. McMahon, J. A., Takada, S., Zimmerman, L. B., Fan, C. M., Harland, R. M., et al. (1998). Noggin-mediated antagonism of BMP signaling is required for growth and patterning of the neural tube and somite. Genes Dev. 12, 1438–1452. Mehrara, B. J., Greenwald, J., Chin, G. S., Dudziak, M., Sagrioglu, J., et al. (1999a). Regional diVerentiation of rat cranial suture-derived dural cells is dependent on association with fusing and patent cranial sutures. Plast. Reconstr. Surg. 104, 1003–1013. Mehrara, B. J., Mackool, R. J., McCarthy, J. G., Gittes, G. K., and Longaker, M. T. (1998). Immunolocalization of basic fibroblast growth factor and fibroblast growth factor receptor-1 and receptor-2 in rat cranial sutures. Plast. Reconstr. Surg. 102, 1805–1817. Mehrara, B. J., Spector, J. A., Greenwald, J. A., Ueno, H., and Longaker, M. T. (2002). Adenovirus-mediated transmission of a dominant negative transforming growth factorbeta receptor inhibits in vitro mouse cranial suture fusion. Plast. Reconstr. Surg. 110, 506–514. Mehrara, B. J., Steinbrech, D. S., Saadeh, P. B., Gittes, G. K., and Longaker, M. T. (1999b). Expression of high-aYnity receptors for TGF-beta during rat cranial suture fusion. Ann. Plast. Surg. 42, 502–508. Miki, T., Bottaro, D. P., Fleming, T. P., Smith, C. L., Burgess, W. H., et al. (1992). Determination of ligand-binding specificity by alternative splicing: Two distinct growth factor receptors encoded by a single gene. PNAS 89, 246–250. Mooney, M. P., Aston, C. E., Siegel, M. I., Losken, H. W., Smith, T. D., et al. (1996). Craniosynostosis with autosomal dominant transmission in New Zealand white rabbits. J. Craniofac. Genet. Dev. Biol. 16, 52–63. Moore, R., Ferretti, P., Copp, A., and Thorogood, P. (2002). Blocking endogenous FGF-2 activity prevents cranial osteogenesis. Dev. Biol. 243, 99–114. Morriss-Kay, G. M., Iseki, S., and Johnson, D. (2001). Genetic control of the cell proliferationdiVerentiation balance in the developing skull vault: Roles of fibroblast growth factor receptor signalling pathways. Novartis Found. Symp. 232, 102–116.

324

Lenton et al.

Moss, M. L. (1958). Fusion of the frontal suture in the rat. Am. J. Anat. 102, 141–165. Moss, M. L. (1959). The pathogenesis of premature cranial synostosis in man. Acta Anat. (Basel) 37, 351–370. Moss, M. L. (1960). Inhibition and stimulation of sutural fusion in the rat calvaria. Anat. Rec. 136, 457–467. Mossaz, C., and Kokich, V. G. (1981). Redevelopment of the calvaria after partial craniectomy in growing rabbits: The eVect of altering dural continuity. Acta. Anat. 109, 321–331. Most, D., Levine, J. P., Chang, J., Sung, J., McCarthy, J. G., et al. (1998). Studies in cranial suture biology: Up-regulation of transforming growth factor-beta1 and basic fibroblast growth factor mRNA correlates with posterior frontal cranial suture fusion in the rat. Plast. Reconstr. Surg. 101, 1431–1440. Moursi, A. M., Winnard, P. L., Fryer, D., and Mooney, M. P. (2003). Delivery of transforming growth factor-beta2-perturbing antibody in a collagen vehicle inhibits cranial suture fusion in calvarial organ culture. Cleft Palate Craniofac. J. 40, 225–232. Muenke, M., Gripp, K. W., McDonald-McGinn, D. M., Gaudenz, K., Whitaker, L. A., et al. (1997). A unique point mutation in the fibroblast growth factor receptor 3 gene (FGFR3) defines a new craniosynostosis syndrome. Am. J. Hum. Genet. 60, 555–564. Muenke, M., Schell, U., Hehr, A., Robin, N. H., Losken, H. W., et al. (1994). A common mutation in the fibroblast growth factor receptor 1 gene in PfeiVer syndrome. Nat. Genet. 8, 269–274. Muller, U., Warman, M. L., Mulliken, J. B., and Weber, J. L. (1993). Assignment of a gene locus involved in craniosynostosis to chromosome 5qter. Hum. Mol. Genet. 2, 119–122. Murray, S. S., Glackin, C. A., Winters, K. A., Gazit, D., Kahn, A. J., et al. (1992). Expression of helix-loop-helix regulatory genes during diVerentiation of mouse osteoblastic cells. J. Bone Miner. Res. 7, 1131–1138. Nacamuli, R. P., Fong, K. D., Lenton, K. A., Song, H. M., Fang, T. D., et al. (2004a). Expression and possible mechanisms of regulation of BMP3 in rat cranial sutures. Plas. Reconst. Surg. (manuscript submitted). Nacamuli, R. P., Song, H. M., Fang, T. D., Fong, K. D., Mathy, J. A., et al. (2004b). Quantitative transcriptional analysis of fusing and non-fusing cranial suture complexes in mice. Plast. Reconstr. Surg. 114, 1818–1825. Nappen, D. L., and Kokich, V. G. (1983). Experimental craniosynostosis in growing rabbits. The role of the periosteum. J. Neurosurg. 58, 101–108. Naski, M. C., and Ornitz, D. M. (1998). FGF signaling in skeletal development. Front. Biosci. 3, 781–794. Noden, D. M. (1978). The control of avian cephalic neural crest cytodiVerentiation. I. Skeletal and connective tissues. Dev. Biol. 67, 296–312. Noden, D. M. (1988). Interactions and fates of avian craniofacial mesenchyme. Development 103, 121–140. Northrop, J., Woods, A., Seger, R., Suzuki, A., Ueno, N., et al. (1995). BMP-4 regulates the dorsal-ventral diVerences in FGF/MAPKK-mediated mesoderm induction in Xenopus. Dev. Biol. 172, 242–252. Nusslein-Volhard, C., Wieschaus, E., and Kluding, H. (1984). Mutations aVecting the pattern of the larval cuticle in Drosophila melanogaster. Zygotic loci on the second chromosome. Wilhelm Roux’s Arch. Dev. Biol. 193, 267–282. Ohbayashi, N., Shibayama, M., Kurotaki, Y., Imanishi, M., Fujimori, T., et al. (2002). FGF18 is required for normal cell proliferation and diVerentiation during osteogenesis and chondrogenesis. Genes Dev. 16, 870–879. Oldridge, M., Wilkie, A. O., Slaney, S. F., Poole, M. D., Pulleyn, L. J., et al. (1995). Mutations in the third immunoglobulin domain of the fibroblast growth factor receptor-2 gene in Crouzon syndrome. Hum. Mol. Genet. 4, 1077–1082.

9. Cranial Suture Biology

325

Opperman, L. A. (2000). Cranial sutures as intramembranous bone growth sites. Dev. Dyn. 219, 472–485. Opperman, L. A., Adab, K., and Gakunga, P. T. (2000). Transforming growth factor-beta 2 and TGF-beta 3 regulate fetal rat cranial suture morphogenesis by regulating rates of cell proliferation and apoptosis. Dev. Dyn. 219, 237–247. Opperman, L. A., Chhabra, A., Cho, R. W., and Ogle, R. C. (1999). Cranial suture obliteration is induced by removal of transforming growth factor (TGF)-beta 3 activity and prevented by removal of TGF-beta 2 activity from fetal rat calvaria in vitro. J. Craniofac. Genet. Dev. Biol. 19, 164–173. Opperman, L. A., Galanis, V., Williams, A. R., and Adab, K. (2002). Transforming growth factor-beta3 (Tgf-beta3) down-regulates Tgf-beta3 receptor type I (Tbetar-I) during rescue of cranial sutures from osseous obliteration. Orthod. Craniofac. Res. 5, 5–16. Opperman, L. A., Nolen, A. A., and Ogle, R. C. (1997). TGF-beta 1, TGF-beta 2, and TGFbeta 3 exhibit distinct patterns of expression during cranial suture formation and obliteration in vivo and in vitro. J. Bone Miner. Res. 12, 301–310. Opperman, L. A., Passarelli, R. W., Morgan, E. P., Reintjes, M., and Ogle, R. C. (1995). Cranial sutures require tissue interactions with dura mater to resist osseous obliteration in vitro. J. Bone Miner. Res. 10, 1978–1987. Opperman, L. A., Passarelli, R. W., Nolen, A. A., Gampper, T. J., Lin, K. Y. K., et al. (1996). Dura mater secretes soluble heparin-binding factors required for cranial suture morphogenesis. In Vitro Cell Dev. Biol. Animal 32, 627–632. Opperman, L. A., Persing, J. A., Sheen, R., and Ogle, R. C. (1994). In the absence of periosteum, transplanted fetal and neonatal rat coronal sutures resist osseous obliteration. J. Craniofac. Surg. 5, 327–332. Opperman, L. A., Sweeney, T. M., Redmon, J., Persing, J. A., and Ogle, R. C. (1993). Tissue interactions with underlying dura mater inhibit osseous obliteration of developing cranial sutures. Dev. Dyn. 198, 312–322. Ornitz, D. M. and Itoh, N. (2001). Fibroblast growth factors. Genome Biol. 2, REVIEWS 3005.1–REVIEWS 3005.12. Ornitz, D. M., and Marie, P. J. (2002). FGF signaling pathways in endochondral and intramembranous bone development and human genetic disease. Genes Dev. 16, 1446–1465. O’Rourke, M. P., and Tam, P. P. (2002). Twist functions in mouse development. Int. J. Dev. Biol. 46, 401–413. Orr-Urtreger, A., Bedford, M. T., Burakova, T., Arman, E., Zimmer, Y., et al. (1993). Developmental localization of the splicing alternatives of Fibroblast Growth Factor Receptor-2 (FGFR2). Dev. Biol. 158, 475–486. Otto, A. W. (1830). Lehrbuch der Pathologischen Anatomie. Ru¨ cher, Berlin, Germany. Otto, F., Thornell, A. P., Crompton, T., Denzel, A., Gilmour, K. C., et al. (1997). Cbfa1, a candidate gene for cleidocranial dysplasia syndrome, is essential for osteoblast diVerentiation and bone development. Cell 89, 765–771. Pajer, P., Pecenka, V., Karafiat, V., Kralova, J., Horejsi, Z., et al. (2003). The twist gene is a common target of retroviral integration and transcriptional deregulation in experimental nephroblastoma. Oncogene 22, 665–673. Park, E. A., and Powers, G. F. (1920). Acrocephaly and scaphocephaly with symmetrically distributed malformations of the extremities. Am. J. Dis. Child 20, 235–315. Park, W. J., Meyers, G. A., Li, X., Theda, C., Day, D., et al. (1995). Novel FGFR2 mutations in Crouzon and Jackson-Weiss syndromes show allelic heterogeneity and phenotypic variability. Hum. Mol. Genet. 4, 1229–1233. Paznekas, W. A., Cunningham, M. L., Howard, T. D., Korf, B. R., Lipson, M. H., et al. (1998). Genetic heterogeneity of Saethre-Chotzen syndrome, due to TWIST and FGFR mutations. Am. J. Hum. Genet. 62, 1370–1380.

326

Lenton et al.

Posnick, J. C. (2000). Craniofacial syndromes and anomalies. In ‘‘Craniofacial and Maxillofacial Surgery in Children and Young Adults’’ (J. C. Posnick, Ed.), Vol. 1, pp. 391–527. W. B. Saunders, Philadelphia. Powers, C. J., McLeskey, S. W., and Wellstein, A. (2000). Fibroblast growth factors, their receptors and signaling. Endocr. Relat. Cancer 7, 165–197. Reardon, W., and Winter, R. M. (1994). Saethre-Chotzen syndrome. J. Med. Genet. 31, 393–396. Rice, D. P., Aberg, T., Chan, Y., Tang, Z., Kettunen, P. J., et al. (2000). Integration of FGF and TWIST in calvarial bone and suture development. Development 127, 1845–1855. Rice, D. P., Kim, H. J., and ThesleV, I. (1999). Apoptosis in murine calvarial bone and suture development. Eur. J. Oral Sci. 107, 265–275. Richtsmeier, J. T., Baxter, L. L., and Reeves, R. H. (2000). Parallels of craniofacial maldevelopment in Down syndrome and Ts65Dn mice. Dev. Dyn. 217, 137–145. Robin, N. H., Feldman, G. J., Mitchell, H. F., Lorenz, P., Wilroy, R. S., et al. (1994). Linkage of PfeiVer syndrome to chromosome 8 centromere and evidence for genetic heterogeneity. Hum. Mol. Genet. 3, 2153–2158. Roth, D. A., Bradley, J. P., Levine, J. P., McMullen, H. F., McCarthy, J. G., et al. (1996). Studies in cranial suture biology: Part II. Role of the dura in cranial suture fusion. Plast. Reconstr. Surg. 97, 693–699. Roth, D. A., Gold, L. I., Han, V. K., McCarthy, J. G., Sung, J. J., et al. (1997a). Immunolocalization of transforming growth factor beta 1, beta 2, and beta 3 and insulin-like growth factor I in premature cranial suture fusion. Plast. Reconstr. Surg. 99, 300–309. Roth, D. A., Gold, L. I., Han, V. K., McCarthy, J. G., Sung, J. J., et al. (1997b). Immunolocalization of transforming growth factor beta 1, beta 2, and beta 3 and insulin-like growth factor I in premature cranial suture fusion. Plast. Reconstr. Surg. 99, 300–309; discussion 310–306. Sagiroglu, J. S., Mehrara, B. J., Chau, D., Saadeh, P. B., Gittes, G. K., et al. (1999). Analysis of TGF-beta production by fusing and nonfusing mouse cranial sutures in vitro. Ann. Plast. Surg. 42, 496–501. Sarkar, S., Petiot, A., Copp, A., Ferretti, P., and Thorogood, P. (2001). FGF2 promotes skeletogenic diVerentiation of cranial neural crest cells. Development 128, 2143–2152. Satokata, I., Ma, L., Ohshima, H., Bei, M., Woo, I., et al. (2000). Msx2 deficiency in mice causes pleiotropic defects in bone growth and ectodermal organ formation. Nat. Genet. 24, 391–395. Schneider, R. A., Hu, D., and Helms, J. A. (1999). From head to toe: Conservation of molecular signals regulating limb and craniofacial morphogenesis. Cell Tissue Res. 296, 103–109. Shishido, E., Higashijima, S., Emori, Y., and Saigo, K. (1993). Two FGF-receptor homologues of Drosophila: One is expressed in mesodermal primordium in early embryos. Development 117, 751–761. Smith, W. C., and Harland, R. M. (1992). Expression cloning of noggin, a new dorsalizing factor localized to the Spemann organizer in Xenopus embryos. Cell 70, 829–840. Song, H. M., Fong, K. D., Nacamuli, R. P., Warren, S. M., Fang, T. D., et al. (2004a). Mechanisms of murine cranial suture patency mediated by a dominant negative transforming growth factor-beta receptor adenovirus. Plast. Reconstr. Surg. 113, 1685–1697. Song, H. M., Nacamuli, R. P., Xia, W., Bari, A. S., Shi, Y.-Y., et al. (2005). High-dose retinoic acid modulates rat calvarial osteoblast biology. J. Cell Phys. 202, 255–262. Soriano, P. (1999). Generalized lacZ expression with the ROSA26 Cre reporter strain. Nat. Genet. 21, 70–71. Spector, J. A., Greenwald, J. A., Warren, S. M., Bouletreau, P. J., Crisera, F. E., et al. (2002). Co-culture of osteoblasts with immature dural cells causes an increased rate and degree of osteoblast diVerentiation. Plast. Reconstr. Surg. 109, 631–642. Spector, J. A., Mehrara, B. J., Greenwald, J. A., Saadeh, P. B., Steinbrech, D. S., et al. (2000). A molecular analysis of the isolated rat posterior frontal and sagittal sutures: DiVerences in gene expression. Plast. Reconstr. Surg. 106, 852–861.

9. Cranial Suture Biology

327

Stelnicki, E. J., Vanderwall, K., HoVman, W. Y., Harrison, M. R., Glowacki, J., et al. (1998). A new in utero sheep model for unilateral coronal craniosynostosis. Plast. Reconstr. Surg. 101, 278–286. Takao, M., Hino, J., Takeshita, N., Konno, Y., Nishizawa, T., et al. (1996). Identification of rat bone morphogenetic protein-3b (BMP-3b), a new member of BMP-3. Biochem. Biophys. Res. Commun. 219, 656–662. Tang, X. X., Biegel, J. A., Nycum, L. M., Yoshioka, A., Brodeur, G. M., et al. (1995). cDNA cloning, molecular characterization, and chromosomal localization of NET(EPHT2), a human EPH-related receptor protein-tyrosine kinase gene preferentially expressed in brain. Genomics 29, 426–437. Tavormina, P. L., Shiang, R., Thompson, L. M., Zhu, Y. Z., Wilkin, D. J., et al. (1995). Thanatophoric dysplasia (types I and II) caused by distinct mutations in fibroblast growth factor receptor 3. Nat. Genet. 9, 321–328. Tessier, P. (1967). [Total facial osteotomy. Crouzon’s syndrome, Apert’s syndrome: Oxycephaly, scaphocephaly, turricephaly]. Ann. Chir. Plast. 12, 273–286. Tessier, P. (1970). [The treatment of facial dysmorphy peculiary to cranio-facial dysostosis (C.F.D.). Crouzon and Apert diseases. Total osteotomy and sagittal displacement of the facial mass]. Chirurgie 96, 667–674. Thomas, J. A., Manchester, D. K., Prescott, K. E., Milner, R., McGavran, L., et al. (1996). Hunter-McAlpine craniosynostosis phenotype associated with skeletal anomalies and interstitial deletion of chromosome 17q. Am. J. Med. Genet. 62, 372–375. Ting, K., Vastardis, H., Mulliken, J. B., Soo, C., Tieu, A., et al. (1999). Human NELL-1 expressed in unilateral coronal synostosis. J. Bone Miner. Res. 14, 80–89. Twigg, S. R., Kan, R., Babbs, C., Bochukova, E. G., Robertson, S. P., et al. (2004). Mutations of ephrin-B1 (EFNB1), a marker of tissue boundary formation, cause craniofrontonasal syndrome. Proc. Natl. Acad. Sci. USA 101, 8652–8657. Virchow, R. (1851). Ueber den Cretinismus, namentlich in Franken, und ueber pathologische Schaedelformen. Verh. Phys. Med. Gesellsch. Wuerzburg 2, 231–271. von Gernet, S., Golla, A., Ehrenfels, Y., SchuVenhauer, S., and Fairley, J. D. (2000). Genotype-phenotype analysis in Apert syndrome suggests opposite eVects of the two recurrent mutations on syndactyly and outcome of craniofacial surgery. Clin. Genet. 57, 137–139. Wang, J. S. (1996). Basic fibroblast growth factor for stimulation of bone formation in osteoinductive or conductive implants. Acta Orthop. Scand. Supp. 269, 1–27. Warman, M. L., Mulliken, J. B., Hayward, P. G., and Muller, U. (1993). Newly recognized autosomal dominant disorder with craniosynostosis. Am. J. Med. Genet. 46, 444–449. Warren, S. M., Brunet, L. J., Harland, R. M., Economides, A. N., and Longaker, M. T. (2003a). The BMP antagonist noggin regulates cranial suture fusion. Nature 422, 625–629. Warren, S. M., Greenwald, J. A., Nacamuli, R. P., Fong, K. D., Song, H. J., et al. (2003b). Regional dura mater diVerentially regulates osteoblast gene expression. J. Craniofac. Surg. 14, 363–370. Waterston, R. H., Lindblad-Toh, K., Birney, E., Rogers, J., Abril, J. F., et al. (2002). Initial sequencing and comparative analysis of the mouse genome. Nature 420, 520–562. Weinzweig, J., Kirschner, R. E., Farley, A., Reiss, P., Hunter, J., et al. (2003). Metopic synostosis: Defining the temporal sequence of normal suture fusion and diVerentiating it from synostosis on the basis of computed tomography images. Plast. Reconstr. Surg. 112, 1211–1218. Wieland, I., Jakubiczka, S., Muschke, P., Cohen, M., Thiele, H., et al. (2004). Mutations of the ephrin-B1 gene cause craniofrontonasal syndrome. Am. J. Hum. Genet. 74, 1209–1215. Wilkie, A. O. (1997). Craniosynostosis: Genes and mechanisms. Hum. Mol. Genet. 6, 1647–1656. Wilkie, A. O., and Morriss-Kay, G. M. (2001). Genetics of craniofacial development and malformation. Nat. Rev. Genet. 2, 458–468. Wilkie, A. O., Oldridge, M., Tang, Z., and Maxson, R. E., Jr. (2001). Craniosynostosis and related limb anomalies. Novartis Found. Symp. 232, 122–133.

328

Lenton et al.

Wilkie, A. O. M., Morriss-Kay, G. M., Jones, E. Y., and Heath, J. K. (1995a). Functions of fibroblast growth factors and their receptors. Curr. Biol. 5, 500–507. Wilkie, A. O. M., Slaney, S. F., Oldridge, M., Poole, M. D., Ashworth, J. G., et al. (1995b). Apert syndrome results from localized mutations of FGFR2 and is allelic with Crouzon syndrome. Nat. Genet. 9, 165–171. Wilkie, A. O. M., Tang, Z., Elanko, N., Walsh, S., Twigg, S. R., et al. (2000). Functional haploinsuYciency of the human homeobox gene MSX2 causes defects in skull ossification. Nat. Genet. 24, 387–390. Winograd, J., Reilly, M. P., Roe, R., Lutz, J., Laughner, E., et al. (1997). Perinatal lethality and multiple craniofacial malformations in MSX2 transgenic mice. Hum. Mol. Genet. 6, 369–379. Wolf, C., Thisse, C., Stoetzel, C., Thisse, B., Gerlinger, P., et al. (1991). The M-twist gene of Mus is expressed in subsets of mesodermal cells and is closely related to the Xenopus X-twi and the Drosophila twist genes. Dev. Biol. 143, 363–373. Wozney, J. M., Rosen, V., Celeste, A. J., Mitsock, L. M., Whitters, M. J., et al. (1988). Novel regulators of bone formation: Molecular clones and activities. Science 242, 1528–1534. Yu, J. C., McClintock, J. S., Gannon, F., Gao, X. X., Mobasser, J. P., et al. (1997). Regional diVerences of dura osteoinduction: Squamous dura induces osteogenesis, sutural dura induces chondrogenesis, and osteogenesis. Plast. Reconstr. Surg. 100, 23–31. Yu, K., Xu, J., Liu, Z., Sosic, D., Shao, J., et al. (2003). Conditional inactivation of FGF receptor 2 reveals an essential role for FGF signaling in the regulation of osteoblast function and bone growth. Development 130, 3063–3074. Zhang, X., Carpenter, D., Bokui, N., Soo, C., Miao, S., et al. (2003). Overexpression of Nell-1, a craniosynostosis-associated gene, induces apoptosis in osteoblasts during craniofacial development. J. Bone Miner. Res. 18, 2126–2134. Zhang, X., Kuroda, S., Carpenter, D., Nishimura, I., Soo, C., et al. (2002). Craniosynostosis in transgenic mice overexpressing Nell-1. J. Clin. Invest 110, 861–870. Zhou, Y. X., Xu, X., Chen, L., Li, C., Brodie, S. G., et al. (2000). A Pro250Arg substitution in mouse Fgfr1 causes increased expression of Cbfa1 and premature fusion of calvarial sutures. Hum. Mol. Genet. 9, 2001–2008. Zimmerman, L. B., De Jesu´ s-Escobar, J. M., and Harland, R. M. (1996). The Spemann organizer signal noggin binds and inactivates bone morphogenetic protein 4. Cell 86, 599–606.

Index A 2 Antiplasmin, 175 Aorta-gonad-mesonephros (AGM), 9–10, 24 Acceleration cell death 2 (ACD2), 140–141 Activin, 14, 304 and inhibin signaling, 84–86 Acute myeloid leukemia (AML), 20 Alzheimer’s disease (AD), 162, 170–171 Animal caps, xenopus, 14 Anterior visceral endoderm (AVE), 86 Anthocyanins, 151–153 synthesis of, 151

B Baselines membranes (BMs), 37–39, 41, 43, 45–46, 48, 51–55, 56–57 mediated interaction, a scheme of, from ES cells to endoderm and epiblast, 51 proteins, distribution of in the early mouse embryo, 40 Basic fibroblast growth factor (bFGF), 14 Blast-colony forming cell (BL-CFC), 11, 24 generation, 13 Blastocyst development, 39, 48 Blastomers, 38 Blood-brain barrier (BBB), 175 Bone morphogenetic proteins (BMPs), 3–4, 14–15, 17, 52, 67, 70–71, 75–84, 86–88, 91–92, 94, 96, 100, 105–107, 240–247, 304 antagonists, 309–311 Bone tissue engineering, evolving concepts in, 239–285 approaches, 247–248 critical size defects (CSDs), non-healing defects, 247, 257, 258–263 locations, 247 current bone reconstruction, 248 allogeneic, 248 autogenous, 248

prosthetic materials, 248 lessons learned from nature, 248–251 endochondral ossification, 248–250 representation of BMP-2, BMP-4 and BMP-7 during, 250 intramembranous ossification, 248–250 representation of BMP-2, BMP-4 and BMP-7 during, 250 bone morphogenetic proteins, 241–247 BMP-2 induced osteogenesis, 244–247 ADAS cells, 244–245, 252 genes upregulated 3- to 6-fold alkaline phosphatase, 244 Cbfa/Runx2, 244 collagen I, 244 junB, 244 osteocalcin, 244 osteopontin, 244 plasminogen activator inhibitor (PAI-1), 244 TBF binding protein 2 (TGF BP2), 244 latent, 244 transcription factor GIF, 244 BMP family, 241–243 belongs to the transforming growth factor-beta (TBF- ) superfamily BMPR1A, 241 BMPR1B, 241 fifteen members (BMP-1 through BMP-15), 241 multifunctional regulators of morphogenesis during embryonic development, 241 BMP heterodimers, increased potency of, 246–247 chinese hamster ovary (CHO) cells, 246 extracellular matrix (ECM), 245, 252, 254, 267 329

330 Bone tissue engineering, evolving concepts in (cont.) mitogen-activated protein kinase (MARK) pathways, 243 polylactic acid (PLA) scaffolds, rhBMP-2-coated, 245 signaling, mother against decapentaplegic (SMAD) proteins, 243 clinical trials, translating BMP-2 and BMP-7 research to, 268–270 four studies alveolar ridge preservation, 268 fibular defect, 268 maxillary sinus floor augmentation, 268 spinal fusion, 268 INFUSE Bone Graft, 269–270 LT-CAGE Lumbar Tapered Fusion Device, 269–270 Osigraft, 269–270 conclusions and future directions, 270–271 current research in the fields of, 242 introduction, 240–241 osteogenesis, 252 osteogenic BMPs, 243–244 adipose-derived adult stromal (ADAS), cells, 243, 255–257 BMP-2, 4, 6, 7, 9, 243 bone marrow stromal (BMS), 243, 255–257, 269 cranial neural crest (CNC) in craniofacial morphogenesis, 243–244 muscular-derived stem cells (MDSCs), 243 adult-derived, 255 osteoprogenitor cells, 255–257 adipocytic lineage, 256 bone marrow (BMS cells), 255 chondrocytic lineage, 256 osteoblastic cells, 255–256 lineage, 256 stem cells adult-derived, 255 embryonic, 255 parathyroid hormone (PTH), 252 pharmacokinetics (PK), 264, 271 protein-peptide binding and immobilization, 270 scaffold (aka ‘‘delivery vehicle’’ and ‘‘carrier’’ ), 252, 271

Index biodegradable, 254 definition of, 253 implantable, 253–255 biocompatibility, 253 bioresponsiveness, 253 osteoconductive, 253 osteoinductive, 253 polyhydroxyethylmethacrylate (poly-HEMA), synthetic sponge, 253 smart biomaterials, 253 structural stability, 253 interactions, BMP, 264–268, 270 absorbable collagen sponges (ACS), 265, 269 isoelectric points (pIs), 266 pharmacokinetics (PK) analyses, 264 rhBMP-2, 264–269 release rates of, 265 ACS, 265 hydroxyapatite, 265 rabbit demineralized bone matrix (rDBM), 265 superphysiological doses of, 270 using BMP-2 and BMP-4 versus BMP-7, 257–264 critical size defects (CDS), bone tissue engineering experiments using, 257, 258–263 fibroblast growth factors (FGFs), 257 human growth factor (hGF), 257 platelet-derived growth factor (PDGF), 257 TGF , 257 vascular endothelial growth factor (VEGF), 257 Brain tumors, 162 Branchial arches, first, (BA1), 93, 95 Buffalo rat liver cells (BRLCs), 2 C cAMP binding program (CBP), 19 response element-bind protein (CREB), 19 Cavitation, 51 Central nervous system (CNS), 161–165, 167, 170, 172, 175, 180 Cerebral ischemia, 162 Chemically defined medium (CBM), 15

Index Chlorophyll, 135–141 catabolic enzyme RCC reductase, 141 catabolism, 137–139 pathway, 140 photosensitive genotypes, 137 stay-greens, 137, 139 chloroplasts, 139, 229–231 degradation, 136–139 pathway, 138 regulator of protein metabolism in senescing cells, 139–141 Chromatin immunoprecipitative (ChIP), 19 Cleft palate, 100–106 Colony-forming cells (CFU-GM), 16 mixed (CFU-Mix), 165 Connective tissue growth factor (CTGF), 71 Cranial neural crest cells (CNCs), 90, 92–94, 243–244, 290–292 Cranial suture biology, 287–328 clinical genetics of, 294–298 FGFR1-3, 294, 297, 304 FGFR mutations, 295–297 muscle segment homeobox 2 (MSX2), 294, 297–298, 304, 306, 313, 317–318 Boston-type craniosynostosis, 298 definition of, 298 mutations, 298 TWIST, 294, 304, 314–316, 317–318 -1, 297, 314–316 -2, 297 definition of, 297 mutations, 295–297 Saethre-Chotzen syndrome, 314, 316 definition of, 297 conclusions and perspectives, 317–318 craniosynostosis (premature pathological fusion of cranial sutures), 287–288, 292–298, 317 definition of, 287 etiopathogenesis of, 292–293 TGF- , role of in, 304 craniosynostotic syndrome familiar pattern of inheritance to, 288, 317 hematoxylin and eosin stained histological sections of postnatal day 6 mouse sutures, 291 introduction, 287–289 molecular and cellular mechanisms governing suture morphogenesis, 304–316

331 apoptosis, 316 Nell-1, 316 growth factors, 304–313 BMPs and BMP antagonists, 309–311 Noggin, 310–311, 317 EPHRIN-B1, 312–313 EFNB1, 312–313 FGFs and FGFRs, 306–309 Apert syndrome, 309 avian mesencephalic neural crest cells, 307 Crouzon syndrome, 308 osteogenesis, 308 Pfeiffer syndrome, 308–309 Runx2, 308 insulin-like growth factors (IGF), 311–312 NELL-1, 312, 316 TGF- , 304–306 Activin, 304 bone morphogenetic proteins (BMPs), 304 eponymous isoforms TGF- 1, 2, & 3, 304 growth and differentiation factor 5 (GDF-5), 304 Hunter-Thompson chondrodysplasia, 306 Nodal, 304 posterior frontal suture, 305–306 transcription factors, 313–316 MSX2, 313 TWIST, 314–316 -1, 314–316 -2, 315 FGFRs, 314–315 Saethre-Chozen syndrome, 314, 316 murine model of suture morphogenesis, 298–303, 317 endochondral bone formation, 299 posterior frontal suture: an example of physiological suture fusion, 299–300 equivalent to human metopic suture, 299 tissue interactions, 300–303 dura mater, 300–302 osteoblast interactions, 302–303 pericranium (aka calvarial periosteum), 303 neural crest (NC), 290–292 cranial, 290–292

332 Cranial suture biology (cont.) skeletal components, sutures, and tissue origins of the mouse skull vault, 290 skull vault development, 289–292 fusing posterior frontal suture, schematic illustration of the tissue origins & development of, 293 membranous skull vault, 289 mesenchymal blastemas, 289 cells, 289 tissue, 289 sutures coronal, 289 lambdoid, 289 metopic or interfrontal, 289 sagittal, 289 D Degenerative disc disease (DDD), 269 Dementia, progressive, 173 Desmodus rotundus (DSPa 1) (saliva of vampire bat), 167–168 Differentiation-inhibiting activity (DIA), 3 Diseases demyelination, 162 neurodegenerative, 162 E E-cadherin, 42 Ectoderm differentiation, 51 primitive, 51 Embryo, mouse, 40 Embryogenesis, early, growth factors, growth factor receptors, and the basement membrane, 38–46 basement membranes and their network-forming elements, 45–46 early mammalian development, 38–40 model for, 40–41 fibroblast growth factor (FGF) signaling during early embryogenesis, 41–43 growth factors, relative localization of, growth factor receptor, and the basement membranes, 43–44 Embryoid bodies (EBs), 11, 14, 22, 48, 50, 52–54

Index cultures, 37 differentiation, mutations affecting, 53 model for early mammalian development, 40–41 Embryonal carcinoma (EC), 3 Embryonic stem cells (ES), 1–7, 22–24, 41, 48, 50–54 and/or primitive ectoderm, 51 differentiation, 43 EMT (palatal fusion), 103–104 Endoderm differentiation, 51 Endothelial cells, 1, 12–13, 15 Enhanced green fluorescent protein (EGFP), 19 -like domain, 168, 271 Epiblast polarization, 51 Epidermal growth factor domain (EGF), 163, 168 Epilepsy, myoclonus, 173 Epithelial-to-mesenchymal transdifferentiation (EMT), 92 Erk kinase activation, 173 pathway, 4 stimulation, 6 Expressed sequence tags (ESTs), 5 Extracellular matrix (ECM), 37–38, 44–45, 56–57, 162, 164–165, 166–167, 169–170, 174–175, 245, 254 cartilage, 267 components, 170 degradation, 167, 169, 174–175 protein (DSD-1-PG/phosphacan), 164–165 F Fibroblast growth factor (FGFs), 37, 41–45, 56, 105, 107, 306–309, 318 signaling and the function and assembly of basement membranes, 37–64 current questions, 55–57 introduction, 37–38 Fibroblast growth factor receptor (FGFR), 39, 41–45, 52, 56, 294, 304, 306–309, 314–315, 317 mutations, 295–297 signaling during early embryogenesis, 41–43 Follistatin-related protein (FSRP), 71 Frontonasal mass (FNM), 93

Index G GATA, 18–20, 43, 49–51, 53, 56 Growth and differentiation factors (GDF), 69, 70, 76–84, 304 Guanosine triphosphate (GTP)ases, 75 binding protein (GTP), 92 H Hedgehog dessert (Dhh), 17 Indian (Ihh), 17 signaling, 16 sonic (Shh), 17, 106 Helix-loup-helix (HLH), 4 basic (bHLH), 4, 17 domain, 18 Hematopoiesis definitive, 8 primitive, 8 Hemogenic endothelium, 10 Hematopoietic and endothelial cells, 1–36 conclusions and future directions, 24–25 embryonal carcinoma (EC) cell lines, 3 embryonic bodies (EBs), 2, 23–24 cell transplantation, 23 embryonic stem cells (ES), 1–7, 22–24, 255 colonies, 2 pluripotency, 6 schematic diagrams of in vitro differentiation of, 2 self-renewal, 5 signaling pathways regulating, 2–4 transcriptional control of, 4–7 transplantation, 22 lineage commitment, transcriptional control of, 17–22 GATA-2, 18–20 endodermal, (GATA-4, GATA-5, GATA-6), 19, 43, 50–51, 53, 56 hematopoietic (GATA-1, GATA-2, GATA-3), 18–20 In vivo potential of embryonic stem cell-derived hematopoietic progenitors, 22–24 Lmo2, 20 Runx1, 20–22; See also, acute myeloid leukemia Cgfa, 2

333 Scl, 17–18, 20–21 progenitors, 7–24 blast colony-forming cells from in vitro differentiated embryonic stem cells, the identification of, 11–12 development, an overview of, 7–10 Flk-1-expressing mesoderm to hematopoietic and endothelial cells, 12–13, 15, 21, 24 and Scl-expressing endothelial cells, 13, 15 hemangioblast, 10 hematopoietic inductive signals, 14–17 Hematopoietic stem cells (HSCs), 8–10, 22–23 Heparan sulfates, (HS), 44, 57 proteoglycans (HSPGs), 44 High proliferative potential colony-forming cells (HPP-CFCs), 8 House fly mating behavior, the genetic architecture of, 189–213 amplified fragment length polymorphism (ALFLP)-linkage mapping, 207 animal behavior, prevalence of nonadditive genetic effects in, 205–206 summary of, 207 basic evolutionary theory, 208 courtship in, 195–198 BUZZ, 196, 203 complex repertoires, qualitative comparisons of, 197 HOLD, 196, 200 LIFT, 196, 200 LUNGE, 196, 200 stalking phase, 195–196 traits, intercorrelations among the five, 198 WING OUT, 196, 200, 202–203 Drosophila melanogaster, 190, 205 future directions, 205–207 microsatellite markers, 207 Musca domestica L (common house fly), 189, 204–208 genome, sophisticated molecular techniques for, 190, 208 quantitative genetic interactions evidence of in, 198–204 bottleneck populations, assays of additive genetic variances in, 199–200, 207–208

334 House fly mating behavior, the genetic architecture of (cont.) line cross analyses, 200–201, 202–203 repeatability assays, 203–204, 207–208 ANOVA (analysis of variance), 203 evolutionary dynamics of, 192–195 dominance, 192–194, 208 epistasis, 192–194, 202, 205, 208 genotype-by-environment interactions and learning, 192–195, 199–200, 202, 204–205, 208 phenotypic values, 192–193, 202, 207 pleiotropy, 192, 195, 205, 207–208 quantitative trait loci (QTL) mapping, 207 sexual selection D. heteroneura–D silvestris system, species recognition pattern, 191 D. montana–D. littoralis system, sexual selection–species recognition continuum, 191, 205 Drosophila Hawaiian, 191 experiments, 207 intersexual, 191 process, cooperative or antagonistic (run away and chase away), 191 summary, 207–208 Human inter/leukin DA responsive cytokine, (HILDA), 3 I Id genes, 4 Inner cell mass (INC), 5 Insulin-like growth factors, 311–312 Integrin-like kinase (ILK), 57 Intercellular matrix (ICM), 38–39, 42, 54 primitive endoderm interface, 52 Internal ribosomal entry site-enhanced green fluorescent protein (IRES-EGFP), 19 J Janus-associated tyrosine kinases (Jak), 3, 6 L Laminin(s) and basement membrane-mediated signaling, 47–55

Index endoderm and ectoderm differentiation follow different pathways, 49–52 from stem cells to pregastrulation embryo: a cascade of cellular and molecular interactions, 52–54 genetic analysis of, 47–48 globular (LG), 45 isotypes, 46–48 induced epithelial polarization, 56 laminin-1 and epiblast differentiation, 48–49 mutations affecting embryogenesis or embryoid body differentiation, 47 receptors and anchorage sites, 54–55 Latency-associated peptide (LAP), 70 latent binding proteins (LTPBs), 70 Leaf senescence, nutrient recycling, and stress defenses, 135–160 chlorophyll, the role of in protein recycling, 136–141 degradation biochemistry of in senescing leaves, 136–137 genes and genetic variation for, 137–139 pathway, 138 regulator of protein metabolism in senescing cells, 139–141 conclusions, 155 dishonesty, does it pay, 152–153 in relation to programmed death of green plant cells, 140 insect preference for green leaves, 153–154 introduction, 136 non-green pigments in, 141–149 anthocyanins and other flavonoids, 144–149 generalized biosynthetic pathway of, 145 carotenoids, 141–144 revelation of autumn colors, 141, 143 pigments and stress defenses in, 149–154 autumn color as a costly signal, 15–151 color changes in senescence as signals, 149–150 possible functions of leaf color, 151–152 visual and olfactory signals, 154 senescence, definition of, 140

335

Index Leukemia inhibitory factor (LIF), 3–7 LIM kinase (LIMK1), 75 Lipoprotein receptor-related protein (LRP), 173 Long-term potentiation (LTP), 171, 177–179 late phase of (L-LTP), 177, 179 M Matrix metalloproteinases (MMPs), 162, 167, 170 Maxillary prominences (MXP), 93 Meckel’s cartilage, 95–96, 100 Medial edge epithelium (MEE), 101–102, 104, 106 alternative fates of, 103 Microglia, 165–168 Mitogen-activated protein kinases (MAPKs), 4, 74, 243, 252 Multiple sclerosis, 162 N N-methyl-D-aspartate (NMDA), 165, 168, 173, 177 Nanog (Enk), 5–6, 43 Nasal processes lateral (LNP), 93–94 medial (MNP), 93–94 Neural crest (NC), 90–92, 101, 290–292 BMP signaling delamination of, 91–92 induction of, 91 cranial (CNC), 90, 92–94, 244 the major player in head development, 90–91 -derived mesenchyme, 299 segregation, paraxial, 88–90 Neural crest cells (NCCs), 84, 88, 90–92, 95–96, 99, 103 avian mesencephalic, 307 migration, a schematic representation of, 89 tissue-specific deletion of Alk2 in, 97 Neuromuscular junction (NMJ), 47 Neuropathology, 162 Neuroserpin, 172 Nonfluorescent chlorophyll catabolites (NCCs), 136–137 formation, enzymic pathway of, 136

O Oct3/4, 6–7, 43 P Palatal development, 100–106 fusion (palatogenesis), 100–101 model of signaling interactions during, 105 shelves; See also medial edge epithelium (MEE) growth, elevation and fusion, 102 prefusion, 104–106 Para-aortic-splanchnopleure (PAS), 17 /aorta-gonad-mesonephros (AGM) region, 8–9, 24 Paraxial mesoderm, 88–90 Parkinson’s disease, 162 Plant photoreceptors and associated signaling, 215–238 Arabidopsis thaliana, 216–231 chloroplast relocation: plant movements at the subcellular level, 229–231 accumulation response, 230–231 anticlinal walls, 229 avoidance response, 229–231 fern cells, 230 mesophyll cell, 229 periclinal walls, 230 photo-stimulated, 231 phototropin- and phytochrome-dependent, 231 relocation, 230–231 mechano, 231 unusual positioning 1 mutant, isolation of, 231 conclusion, 231 cryptochromes (cry1 & cry2), 216, 219–222 and phytochromes: modulatory receptors in light-induced plant movement responses, 219–222 carboxyl-terminal extension (designated the CCT), 220 de-etiolated (light-grown) developmental states, 219 DNA, 220 etiolated (dark green) developmental states, 219 flavin adenine dinucleotide (FAD), 220

336 Plant photoreceptors and associated signaling (cont.) timing of floral induction 219 introduction, 215–216 phototropins (PHOT1 & 2), primary receptors in light-induced plant movement responses, 216–219, 223–226, 229, 230 active state of, 217–218 domain organization of, 218 flavin mononucleotide (FMN) molecule, 217 light receptor nomenclature, 217 members of LOV(1 &2) domain family, 217–219, 224 phototropism: plant movements of entire organs, 222–228 auxin response elements (AuxREs), 227–228 auxin-responsive transcription factor (ARF) proteins, 227–228 AUX/IAAs, 227–228 carboxyl-terminal dimerization motif (CTD), 227 DNA-binding domain (DBD), 227 GH3s, 227 middle region (MR), 227 NPH4/ARF7, 227–228 Small auxin-upregulated RNAs (SAURs), 227 Cholodny-Went theory, 225–226 indole-3–acetic acid (IAA), 225–227 negative root, 223 NPH3, 224–225 NRL family, 224 PIN1 & 3, 226 positive hypocotyl, 223 RPT2, 224–225 phytochromes (phyA-D), 216, 221 and cryptochromes: modulatory receptors in light-induced plant movement responses, 219–222 light-absorbing form active far red (Pfr), 221–222 inactive red (Pr), 221–222 stomatal opening: plant movements at the cellular level, 228–229 stomata, definition of, 228

Index Plasminogen activators (PAs), 162–168 inhibitor (PAIs), 164, 173, 177 activators and plasmin, 171–175 gene, 174 interactions and integrin, 165 tissue type (tPA), 162–163, 168 urokinase-type (uPA), 163, 168–170, 174, 179–180 Prechordal plate mesenchyme (aka head organizer), 87–88 Proteases, extracellular, 161–188 conclusions, 180 introduction, 162 plasmin(ogen), 170–171 precursor of serine protease plasmin, 170 plasminogen activators (PAs), 162–169 domain structure of, 163 inhibitors of, 171–175 serpins (serine protease inhibitor), 171 tissue-type (tPA), 163–168 effects on rodent behavior, 175–176, 177–179 pro-survival and anti-survival properties of, 166 urokinase-type (uPA), 163, 168–170 plasminogen and uPA, effects on rodent behavior, 179–180 Proteases, serine, 162 Protein kinase C (PKC), 74–75 R ROCK, 54 kinase, 50–51 mediated cytoskeletal rearrangement of the epiblast S Senescence, definition of, 140 Serpins (serine protease inhibitor), 171 Signal transducer and activator of transcription (Stat), 3 stat3, 3–5 F, 5 cDNA, 5 Smad proteins, 74–75, 87, 95–96, 173, 304

Index Srchomology 2 (SH2), 3 Subtilisin-like pro-protein convertases (SPC), 70 T Tgf- (transforming growth factor beta) signaling and facial development, 65–133, 173, 304–306 clinical research applications, 107–108 craniofacial fracture healing, 107 prevention of heterotopic bone formation, 107–108 teeth and periodontal regeneration, 108 conclusions, 108 eponymous isoforms (1, 2, & 3), 304 facial prominences and formation of the face, 92–94 development based on fusion of several regions of tissue, 92–94 initial phases of facial formation, 93 head organizers and early anterior development, 86–90 anterior visceral endoderm acts synergistically with derivatives of the gastrula organizer, 86–87 future: prechordal plate mesenchyme with BMP downregulation/nodal upregulation, 87–88 paraxial mesoderm, neural crest segregation, & possible involvement of BMP-4 signaling, 88–90 introduction, 66–67 mandibular development, 94–100 early development of, 100 identity of the first branchial arch, 95 intramembranous ossification of, 99–100 lower jaw, BMP signaling is critical in the rostral part of, 96–99 Alk2 receptor in neural crest cells, tissue-specific deletion of, 97 Meckel’s cartilage, TBF /Smad signaling regulates growth of, 95–96 neural crest in early craniofacial development, 90–92 cranial NC, the major player in head development, 90–91 delamination of NCC and BMP signaling, 91–92

337 induction of NC and the BMP signaling, 91 migrate to multiple sites of the developing embryos, 90 palatal development and cleft palate, 100–106 apoptosis, 101–103 palatal shelf growth, elevation and fusion, 102 prefusion, epithelial-mesenchymal interactions, interactive signaling pathways, and morphogenesis of , 104–106 palatogenesis; See also medial edge epithelium alternative fates of, 103 in mice as a model for human development and disease, 100–101 role of TGF superfamily signaling in, 103–104 role of in craniosynostosis, 304 superfamily signaling, 67–75 bone morphogenetic proteins and related growth factors, 67–69 convergence by type I receptors and Smad proteins, 73–74 ligands, 68 and receptors, craniofacial phenotypes in mutants of, 76, 77–83, 84–86 activin and inhibin signaling, 84–86 BMP and GDF signaling, 76–84 Tgf- signaling, 84 antagonists and ligand heterodimers increase the signaling complexity, 70–71 family, structure of, 70 interactions of with type I and type II receptors or transmission, 72 receptors do not make our understanding of the system easier, 71–73 unconventional, alternative pathways, 74–75

338

Index

U

Y

Ultraviolet (UV), 215

Yolk sac, 8–9, 11 avian, 9, 14 blood progenitors and endothelial cells, 17 embryo, 21 extraembryonic, 14, 19, 21 mice, 8, 12–15 vasculature, abnormalities in, 16

V Vascular endothelial growth factor (VEGF), 11, 14–16, 24, 257 VENT cells, 96