Investigating the Human Genome Insights into Human Variation and Disease Susceptibility Moyra Smith
Vice President, Publisher: Tim Moore Associate Publisher and Director of Marketing: Amy Neidlinger Acquisitions Editor: Kirk Jensen Editorial Assistant: Pamela Boland Operations Manager: Gina Kanouse Senior Marketing Manager: Julie Phifer Publicity Manager: Laura Czaja Assistant Marketing Manager: Megan Colvin Cover Designer: Chuti Prasertsith Managing Editor: Kristy Hart Senior Project Editor: Lori Lyons Copy Editor: Hansing Editorial Services Proofreader: Kathy Ruiz Senior Indexer: Cheryl Lenser Compositor: Nonie Ratcliff Manufacturing Buyer: Dan Uhrig © 2011 by Pearson Education, Inc. Publishing as FT Press Upper Saddle River, New Jersey 07458 FT Press offers excellent discounts on this book when ordered in quantity for bulk purchases or special sales. For more information, please contact U.S. Corporate and Government Sales, 1-800-382-3419,
[email protected]. For sales outside the U.S., please contact International Sales at
[email protected]. Company and product names mentioned herein are the trademarks or registered trademarks of their respective owners. All rights reserved. No part of this book may be reproduced, in any form or by any means, without permission in writing from the publisher. Printed in the United States of America First Printing June 2011 ISBN-10: 0-13216814-6 ISBN-13: 978-0-13216814-4 Pearson Education LTD. Pearson Education Australia PTY, Limited. Pearson Education Singapore, Pte. Ltd. Pearson Education Asia, Ltd. Pearson Education Canada, Ltd. Pearson Educación de Mexico, S.A. de C.V. Pearson Education—Japan Pearson Education Malaysia, Pte. Ltd. Library of Congress Cataloging-in-Publication Data Smith, Moyra. Investigating the human genome : insights into human variation and disease susceptibility / Moyra Smith. p. ; cm. Includes bibliographical references. ISBN-13: 978-0-13-216814-4 (hardback : alk. paper) ISBN-10: 0-13-216814-6 (hardback : alk. paper) 1. Human genome. 2. Disease susceptibility. I. Title. [DNLM: 1. Genome, Human. 2. Genetic Predisposition to Disease. QU 470] QH447.S625 2011 611’.01816—dc22 2011006494
This work is dedicated to the memory of three remarkable people who inspired and encouraged me long ago: My mother Florence Van Eyk Smith, my grandfather Manard James Van Eyk, and our beloved family physician, Dr. Colin Roy Cockcroft.
This page intentionally left blank
Contents Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii Chapter 1
Genome Architecture and Sequence Variation in Health and Disease . . . . . . . . . 1
Chapter 2
Genes and Transcripts: Insight into Regulation at Different Levels . . . . . . . . . 23
Chapter 3
Epigenetics: Modifications of DNA, Chromatin, and Gene Expression . . . . . . . 37
Chapter 4
Gene Environment Interactions . . . . . . . . 53
Chapter 5
Pathways, Phenotypes, and Phenocopies . . . . . . . . . . . . . . . . . . . . . . . . 67
Chapter 6
Dynamic Function, Synaptic Activity, and Plasticity . . . . . . . . . . . . . . . . . . . . . . . . 81
Chapter 7
Late Onset Neurodegenerative Diseases . . . . . . . . . . . . . . . . . . . . . . . . . . 101
Chapter 8
Genes and Genomes in Cancer: Targeted Therapies . . . . . . . . . . . . . . . . . . 131
Chapter 9
Functional Genomics: Personalized Medicine and Therapeutics . . . . . . . . . . . 153 Epilogue . . . . . . . . . . . . . . . . . . . . . . . . . . 167 References . . . . . . . . . . . . . . . . . . . . . . . . 171 About the Author . . . . . . . . . . . . . . . . . . . 203 Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205
Acknowledgments I express my gratitude to researchers and clinicians who have worked hard and long to further our knowledge of genetics and genetic diseases. Preparation of this book would not have been possible without the amazing resources of the University of California library services and the online resources of the National Library of Medicine—I want to acknowledge all who contribute to making these resources available. I want to acknowledge Kirk Jensen, who encouraged me to write this book. I thank Lori Lyons of Pearson for moving this manuscript along in the publishing process and copy editor Krista Hansing for her thorough input.
Preface In 2009, Georgina Ferry wrote, “The optimism of science is twofold: that its methods might reveal, one tiny pixel at a time, more of the wonder of the natural world and that this knowledge might be able to solve practical human problems.” Progress in the fields of genetics and genomics since a draft sequence of the human genome was published in 2001 is indeed a cause for optimism. However, this discovery has left some people disappointed because development of new therapies to treat disease has been slower than anticipated. Availability of this sequence information has fueled groundbreaking studies in genetics, genomics, and epigenetics that provide insight into human variation and the pathogenesis of both common and rare diseases. The goal of this book is to briefly review several of those groundbreaking studies and new insights. My own experiences during a 40-year career as a clinical geneticist and researcher in genetics and genomics influenced the choice of topics discussed in this book. I explore new insights into human origins, migrations, and human population diversity gained though genomic analyses. I consider insights into the etiology of common diseases such as diabetes and coronary heart disease. I also consider studies on synapses and synaptic plasticity, representing pathways to understanding mind and cognition. I discuss complexities of late-onset neurological diseases and efforts to utilize genetic and genomic methodologies to unravel the pathogenesis of these disorders. I also consider new insights into aspects of protein misfolding and clearance or deposition as aggregates that sequester other proteins. In considering gene environment interactions, I focus on aspects of DNA damage and repair and DNA
viii
INVESTIGATING THE HUMAN GENOME
instability. An appropriate movement is underway toward translational research and greater emphasize on treatment. I review examples of treating primary defects and downstream effects of genetic disorders. I review new information on regulating gene expression at the levels of transcription, translation, and post-translational modifications. Growing evidence indicates that modifications of DNA, histones, and of nonhistone proteins greatly impact gene expression and the function of gene products, and I review aspects of research in these areas, sometimes referred to as epigenetics. In a chapter related to cancer, I review new discoveries in genetics and genomics that have direct relevance to therapy. Growing evidence points to the importance of protein interactions and webs of molecular interactions that determine regulation and growth and the operation of systems, and I consider these topics. In a closing chapter I consider the relevance of genomics and systems biology to personalized medicine.
1 Genome architecture and sequence variation in health and disease Availability of information on DNA sequence in human genomes and advances in technologies to amplify and sequence DNA have led to significant progress in delineating sequence differences that lead to disease. These techniques have also led to the discovery of sequence variants that occur in healthy individuals. Studies of variation in the human genome are greatly facilitated through the availability of microarrays designed to detect single nucleotide polymorphisms (SNPs) that occur with frequencies greater than 1% to 5% in the population. Gene loci that are close to each other are often coinherited. SNP analyses can determine a series of alleles of loci in a specific region (a haplotype). Microarray technologies enable analysis of as many as one million SNPs on each array. These microarrays can also determine structural variation and copy number changes, defined as deletion or duplications greater than 1 kilobase (kb). Specific probes for regions known to frequently harbor copy number changes are also present on SNP microarrays such as the Affymetrix 6.0 array. Advances in technologies in DNA sequencing include massively parallel sequencing, often referred to as next-generation sequencing. This chapter explores aspects of structural genomic variation and sequence variation in different populations and the role of sequence differences in the etiology of common disorders such as diabetes
1
2
INVESTIGATING THE HUMAN GENOME
mellitus, obesity, and coronary heart disease. It also covers nextgeneration sequencing and examples of its application to the discovery of gene defects that lead to disease. Through the use of polymerase chain reaction techniques, samples with low concentrations of DNA can be used to derive material for DNA sequencing. This chapter discusses applications of these techniques to discover how the sequence in modern humans differs from that of Neanderthals and early modern humans. Also presented are reports of studies of DNA extracted from two teeth from a man who died in 1783. DNA analysis enabled researchers to diagnose the disease that afflicted him and analyze the specific mutation and surrounding polymorphisms that connected him to present-day patients with the same disease.
Structural variation In the human genome, segmental duplications with highly identical sequence are usually interspersed and separated by more than 1 megabase. She, et al. (2006), identified more than 400 duplication blocks within the human genome. Segmental duplications are frequently clustered in pericentric and subtelomeric regions (Marques-Bonet, et al., 2009). Evidence indicates that pericentric and subtelomeric duplications evolved independently from intrachromosomal duplications. Core duplicons of 5–30kb occur in intrachromosomal duplications. One example of a core duplicon is LCR16a, which is rich in Alu repeats. Unequal crossover between directly oriented duplicated segments may lead to dosage changes or altered structure and function of a gene. Marques-Bonet, et al., noted that most copy number polymorphisms result from this mechanism. Regions between segmental duplications may be deleted, duplicated, or inverted as a result of unequal crossover. The existence of
CHAPTER 1 • GENOME ARCHITECTURE AND SEQUENCE VARIATION
3
highly similar duplicated segments on two different chromosomes may lead to translocation events. Polymorphisms also exist within the segmental repeats, and in different individuals, these regions may be larger or smaller. Segmental duplications are particularly abundant in certain chromosome regions, such as 15q11-q13, and these regions are frequent sites of deletions and duplications. A key question is whether a specific structural variant, such as a deletion or a duplication (copy number variant) that includes unique sequence DNA, is a direct cause of phenotypic abnormality. Genomic syndromes often occur as a result of deletion or duplication of genomic regions that are flanked by segmental duplication blocks. In these syndromes, specific phenotypes result from the deletion of specific regions; for example, Williams syndrome results from the deletion of chromosome 7q11.2. Characteristic phenotypic features of this syndrome include cognitive and behavioral impairments, distinct facial features, and cardiac malformations. Girirajan and Eichler (2010), reviewed findings in a subset of genomic structural changes in which a particular genomic change results in a series of phenotypes in which specific clinical features differ in different individuals. Differences occur in the degree to which individuals with the same defect are affected—that is, there are varying degrees of penetrance. The clinical consequences of a particular dosage change in a specific region may be influenced by dosage changes or mutations elsewhere in the genome. Examples of specific regions where deletions are associated with a variety of phenotypes include 16p11.2. In some cases with deletion in this region, severe obesity occurs; other cases with the same deletion are diagnosed with autism, while in others, congenital malformations and developmental delay occur. Diverse phenotypes have been described in cases with deletion of 17q12; some cases present with hereditary neuropathy, with a tendency to pressure palsy (HNPP);
4
INVESTIGATING THE HUMAN GENOME
and in other cases, schizophrenia occurs. Other diagnoses encountered in patients with 17q12 deletion include renal cystic disease or maturity-onset diabetes of the young. Deletion in 1q21.1 may be associated with a learning disability in some cases and with congenital heart disease or schizophrenia in others. The copy number variants associated with diverse phenotypes are sometimes found with low frequency in control populations. One question that arises is whether the different phenotypic consequences result from slight differences in the position of deletion breakpoints and whether sequence differences occur in the same region on the homologous chromosome. Another genetic factor that may play a role in some cases is that the deletion of a specific locus on one chromosome unmasks a recessive mutant allele at that locus on the homologous chromosome. Other important possible explanations for the phenotypic variation are that additional genetic modifiers elsewhere in the genome modify the phenotype. Girirajan and Eichler (2010), proposed that a two-hit genomic model most likely explains the variable phenotypes in individuals with copy number variants in 16p12.1 or 22q11.2. Copy number variations and deletions, in particular, are most often considered to be of clinical relevance if they arise de novo—that is, if they are present in a child but absent in the parents. However, growing evidence indicates that parents who carry specific copy number variants may have subclinical manifestations attributable to the genomic change. CNVs and microarray are illustrated in Figure 1.1.
Human genetic sequence variation During the past decade there have been significant advances in technologies for DNA sequencing that have facilitated studies of variation in ancient and modern humans.
CHAPTER 1 • GENOME ARCHITECTURE AND SEQUENCE VARIATION
5
Figure 1.1 Results of analyzing SNP alleles and copy number variants using an Affymetrix 6.0 array and genotyping console on twin females with autism. Rows 1 and 2 at the top of the figure show the distribution of A and B alleles of specific SNPs. Note the identical patterns of alleles in the twins. Rows 3, 4, 5, and 6 show a chromosomal region with a copy number variant. Each twin has three copies of the CNV that encompasses three genes, shown at the bottom of the figure. A known population variant region is indicated, but the variant region is shorter and does not encompass a gene.
Studies in ancient fossil remains In 2010, Green, et al., published data on four billion nucleotides of DNA sequence from three different Neanderthal individuals. They noted that DNA extracted from these late Pleistocene remains had degraded to segments less than 200 nucleotides in length and that it had been chemically modified. In addition, they found substantial contamination from microorganisms. To enrich the Neanderthal DNA, samples were digested with restriction endonucleases that selectively cleave microbial DNA.
6
INVESTIGATING THE HUMAN GENOME
Green, et al., examined the DNA sequence in loci with specific alleles that are known to differ in different modern human populations. They determined that Neanderthals shared 1% to 4% of genotypes at the sequences with Europeans and Asians. At these loci, Neanderthals did not share alleles with Sub-Saharan African populations. Sequence analyses also led to the identification of genes that apparently underwent positive selection in modern humans. Specific sequences in these genes impact protein function. Green, et al., identified specific functional sequence changes that occurred in modern humans but were absent from Neanderthals, and the Neanderthal sequence matched the sequence present in chimpanzees.
Studies on an ancient Saqqaq individual In the past decade, we have seen the confluence of paleontological analyses of bone fossils and cultural artifacts with DNA analyses. Rasmussen, et al. (2010), examined DNA recovered from the hair roots of an individual from Greenland, estimated to have lived 4,000 years ago, who was of member of the Saqqaq culture. Analysis of DNA polymorphisms from the hair roots indicated that the closest match was with individuals from eastern Siberia. One advantage of analyses from hair roots is that they are less contaminated with fungi and bacteria than samples isolated from bone fossils. The high quality of the DNA isolated from hair roots of the Saqqaq individual enabled the analysis of 350,000 SNPs. Earlier studies by Rasmussen’s group generated information on the complete mitochondrial DNA gene from permafrost-preserved Saqqaq individuals. Given the length of the tracts of homozygosity they found, Rasmussen, et al., concluded that the inbreeding coefficient was high. Rasmussen studied DNA sequence at functional polymorphic sites. The combination of SNPs at the HERC2 and OCA2 (oculocutaneous
CHAPTER 1 • GENOME ARCHITECTURE AND SEQUENCE VARIATION
7
albinism gene2) indicated that the individual most likely had brown eyes and dark hair. Analyses also revealed that the Saqqaq individual was most closely related to three Northern Old World Arctic populations and was more distantly related to New World Amerinds. Researchers were not able to detect evidence of West Eurasian population admixture. Nuclear DNA SNP analyses and studies on the mitochondrial and Y chromosome haplotypes of the Saqqaq individuals matched most closely with those of North East Asian populations. The Saqqaq culture is a component of the Arctic small tool transition and is estimated to have existed between 4,750 and 2,500 years ago.
Sequence variation in different populations and regions In an analysis of 650,000 common SNPs, Li, et al. (2008), collected samples from populations in 51 geographic regions. Populations studied were drawn from Sub-Saharan Africa, North Africa, the Middle East, Europe, East, South, and Central Asia, Oceania, and the Americas. They carried out haplotype analysis to identify linked alleles at specific loci. They detected finer haplotype substructure in different regions. They noted, for example, that Palestinians, Druze, and Bedouins have haplotype contributions from the Middle East, Europe, and South and Central Asia. Li, et al., concluded that nonrandom differences between populations have accumulated at a number of different loci. However, they also concluded that within population differences accounted for most of the genetic diversity. Results of their analyses revealed that heterozygosity was greatest in Africa and was reduced as geographic distance from Addis Ababa increased. Tishkoff, et al. (2009), studied genotypes in 121 African populations, in 60 non-African populations, and in the African-American population. They studied microsatellite repeat polymorphisms and insertion deletion polymorphisms. They obtained evidence for
8
INVESTIGATING THE HUMAN GENOME
regional differences in the allele frequency of markers; however, their analyses also revealed evidence for substantial population admixture.
Homozygosity mapping Because recombination occurs between homologous chromosomes during meiosis, the presence of identical alleles over long stretches of genomic DNA on homologous chromosomes (homozygosity) was thought to occur only in consanguineous pedigrees or inbred populations. Gibson, et al. (2006), studied 262 individuals in four different populations and identified 20 different genomic regions where homozygosity extended over 1 megabase or more. They noted that the lowest number of homozygous tracts occurs in the Yoruba population and that this reflects the more ancient roots of the population; over longer time periods, segments of chromosomes have broken up. Gibson, et al., noted that, in modern populations, tracts of homozygosity often occurred in similar genomic regions, indicating regions with a lower frequency of recombination. Examples of blocks of homozygosity are illustrated in Figure 1.2.
Studies on hereditary disorders and population history Currently, 36 disorders are considered to comprise the Finnish disease heritage. Norio (2003) reviewed the history and studies of these disorders and related them to the historical origins, migrations, and settlements of the Finnish population. Thirty-two of these disorders have autosomal recessive inheritance patterns, two are autosomal dominant, and one is an X-linked disorder. Norio noted that the Finnish population has been relatively stable for many years, without evidence of continuous migration into Finland. Internal migration of families from the southeastern parts of the country around Sevo into middle and northern regions of the country occurred around 1600. The migrant families settled in small clusters. Each cluster was often located at some distance from other
CHAPTER 1 • GENOME ARCHITECTURE AND SEQUENCE VARIATION
9
clusters, and with little admixture between clusters, mating occurred within clusters. In later generations, couples who married often shared founders six or seven generations ago.
Figure 1.2 Blocks of homozygosity that are identical in twins (rows 1 and 2). Strikingly large blocks of homozygosity are present in the individual illustrated in row 3, likely due to consanguinity of parents. Rows 4 and 5 indicate positions of genes on chromosome 16.
In 1999, Peltonen, et al., reported that gene loci for 32 of the Finnish hereditary diseases were mapped to specific chromosomes and causative genes for 17 of the disorders were isolated. As expected, marked linkage disequilibrium occurred for markers in the vicinity of the disease alleles, and homogeneity of the disease-causing alleles was observed. Peltonen and colleagues studied molecular mechanisms in these genetic disorders. Molecular analyses of products encoded by genes in regions where the diseases mapped resulted in the discovery of new proteins and enzymes. Peltonen emphasized that analysis of the
10
INVESTIGATING THE HUMAN GENOME
disease genes facilitated disease diagnosis, and, importantly, healthcare was available following diagnosis. Peltonen and coworkers also reported that analyzing linkage disequilibrium is useful in identifying gene loci that contribute to the risk of complex common diseases. Kilpinen, et al. (2009), studied regions of linkage disequilibrium in a unique pedigree with multiple cases of autism. Individuals in this pedigree shared ancestors in the 17th century. Analyses revealed areas of linkage disequilibrium in three chromosomal regions at 15q11-q13, 19p13, and 1q23.
Genetic variations, single nucleotide polymorphisms (SNPs) and genome wide association studies (GWAS) The design of genome wide association studies (GWAS) is predicated on the hypothesis that common DNA sequence variants contribute to the etiology of common disease. Results indicate that even when statistically significant associations between disease and a specific SNP are determined, the overall contribution of specific SNPs to disease risk is often low.
Genome wide association studies and insight into etiology of type 2 diabetes and obesity In type 2 diabetes, the pancreatic beta cell–secreting capacity becomes inadequate to overcome the progressive peripheral resistance to insulin uptake. Factors that play roles in the development of peripheral insulin resistance include age, inactivity, and weight gain. McCarthy (2010) reviewed the discovery of genes that impact susceptibility to diabetes and obesity. He considered three waves of discovery. The first included family-based linkage studies. These studies led to the identification of genes involved in a number of Mendelian forms of early-onset diabetes, including neonatal diabetes and maturity-onset diabetes of the young (MODY). Genes that were found to
CHAPTER 1 • GENOME ARCHITECTURE AND SEQUENCE VARIATION
11
play a role in MODY included NEUROD1 (neurogenic differentiation 1); GCK (glucokinase); hepatic nuclear factor genes HNF1A, HNF1B, and HNF4A; and IPF (insulin promoter factor). Family studies also led to the discovery of a mitochondrial DNA mutation that predisposes carriers to diabetes and deafness. McCarthy noted that family studies of childhood obesity led to the discovery of rare forms of this condition due to mutations in any one of three genes: leptin, leptin receptor, and pro-opiomelanocortin. The second phase of investigation into diabetes and obesity involved searching for variants in candidate genes. These studies led to the identification of common variants of modest effect in PPARG (peroxisome proliferation activated receptor gamma) and KCNJ11 (potassium inwardly-rectifying channel, subfamily J, member 11). Resequencing of the melanocortin 4-receptor gene led to the identification of associated variants in 2% to 3% of cases of obesity. A third wave of studies involved large-scale analysis of common DNA sequence variants (SNPs). McCarthy considered this to be the most successful wave of studies. Important diabetes-associated loci identified in these studies include the transcription factor that modulates pancreatic function TCF7L2; cyclin-dependent kinases CDKAL1, CDKN2A, and CDKN2B, which regulate cyclin; and HHEX, a gene involved in beta cell development. Each copy of a susceptibility allele at one of these loci leads to a 15% to 20% increase in the risk for diabetes. McCarthy reported at least 40 known loci with alleles associated with increased risk of diabetes. Of interest is the fact that five of the loci with common variant alleles associated with increased risk of diabetes also harbor rare variants involved in familial or syndromic diabetes. These four loci are wolframin (NFS1); hepatocyte nuclear factors HNF1A and HNF1B; the melatonin receptor MTNR1B; and IRS1 insulin receptor substrate 1, which impacts insulin action.
12
INVESTIGATING THE HUMAN GENOME
Pathways involved in diabetes McCarthy reported that the loci with the strongest evidence of association with type 2 diabetes impact insulin secretion. Examples include cyclin-dependent kinases CDKAL1, CDKNL2A, and CDKN2B. At these loci, risk variants predispose to reduced pancreatic beta cell mass. The diabetes risk alleles in TCF7L2, MTNR1B, and KCNJ11 predispose to beta cell dysfunction. Risk alleles in the FTO locus (gene related to fat mass and obesity) contribute to obesity and to peripheral resistance to insulin. The PPARG and IRS1 (insulin receptor substrate 1) loci impact insulin resistance and obesity. In the monogenic forms of diabetes and in autoimmune diabetes in adults, information about the underlying causative gene can influence therapeutic decisions, such as whether insulin is required, whether dietary management may be sufficient, or whether sulfonylureas are required. In type 1 diabetes commonly associated with HLA variants (latent autoimmune diabetes) or with defects in the insulin genes INS or PTPN22 (protein tyrosine phosphatase non receptor type 22), insulin is likely necessary. Maturity-onset diabetes of the young (MODY) due to GCK (glucokinase) deficiency may respond adequately to dietary management. MODY due to HNF1A deficiency may require treatment with sulfonylureas. McCarthy reviewed the results of genome wide association studies designed to identify common variants associated with increased body mass index (BMI) and noted that at least 30 such loci have been associated. The strongest signal is associated with the FTO locus (gene related to fat mass obesity). He noted that signals of risk alleles were also detected in genes with neuronal function, such as BDNF (brain derived neurotrophic factor), SH2B1 (signaling protein), and NEGR1 (neuronal growth regulator). He indicated that obesity may be partly a disease of disordered hypothalamic function. In studies that involved analyzing fat mass distribution, risk alleles in 15 loci
CHAPTER 1 • GENOME ARCHITECTURE AND SEQUENCE VARIATION
13
were identified. Evidence indicates that risk alleles at these loci impact adipocyte development and function. McCarthy noted that clinical translation of these findings is impacted at least partly by the modest effect of the risk alleles. Homozygotes for the FTO risk allele are an average of 2 to 3 kilograms heavier than in individuals without the risk allele. However, he noted that identifying risk-altering genes contributes to our understanding of the biology of disease. Another important consideration is that most of the risk alleles lie outside the coding regions of genes, and it’s not clear how they impact the regulation of gene expression. McCarthy predicted that large-scale genome-wide resequencing efforts now underway would clarify relationships between sequence variants and clinical phenotypes.
Tracking genes involved in coronary heart disease after GWAS In 2007, Samani, et al., carried out genome wide association studies in coronary heart disease subjects. They identified several genetic loci that affect the risk of coronary artery disease (CAD), including loci at 9p21.3 and 1p13.5. In 2008, Kathiresan, et al., identified two loci associated with abnormal levels of low-density lipoprotein cholesterol (LDL cholesterol), one locus mapped to chromosome 1p13 and the other mapped to 19p13. They noted that the 1p13 locus maps near the gene SORT1 (sortilin 1). Kjolby, et al. (2010), demonstrated that sortilin protein encoded by SORT1 is an intracellular sorting receptor for apolipoprotein ApoB100. They noted that SORT1 regulates plasma low-density lipoprotein levels through hepatic export of ApoB100 containing lipoproteins. In studies on mice, they determined that sortilin 1 overexpression stimulates the hepatic release of lipoproteins and increases plasma LDL levels.
14
INVESTIGATING THE HUMAN GENOME
Musunuru, et al. (2010), carried out studies in cohorts of human subjects and in human-derived hepatocytes. They determined that a noncoding polymorphism SNP rs12740374 in 1p13 impacts a transcription factor binding site that alters hepatic expression of SORT1. The risk allele G in rs12740374 disrupts the C/EBP transcription factor binding site and is significantly associated with LDL cholesterol levels, p=1X10-170. In studies on mouse livers, Musunuru, et al., determined that sortilin 1 impacts plasma levels of low-density lipoprotein (LDL) and very low-density lipoprotein (VLDL) cholesterol. They demonstrated that knockdown of sortilin 1 expression in mice led to a 46% increase in total cholesterol compared with controls. The studies of Musunuru, et al., demonstrated the clinical relevance of non-protein-coding DNA variants identified through GWAS. They concluded that the sortilin pathway is a promising new target for therapeutic intervention in hyperlipidemia and myocardial infarction. These investigators noted that, in some individuals, aggressive treatment with statins fails to lower the levels of LDL cholesterol. Statins inhibit cholesterol synthesis through inhibiting hydroxy-3methyl-glutaryl coenzyme A reductase and reduce levels of both LDL cholesterol and total cholesterol. In GWAS analysis of blood lipids on 100,000 individuals, LinselNitschke, et al. (2010), reported evidence for the involvement of 18 genes that were previously shown to play roles in Mendelian lipid disease. The significance values for association were much higher with these variants than those obtained for other variants. Highly associated loci included LPL (lipoprotein lipase) 2X10-115, APOA1 (Apolipoprotein A1) 7X10-240, CETP (cholesterol ester transfer protein) 7X10-380, LDLR (LDL receptor) 4X10-117, APOE (apolipoproteins A) 9X10-147, APOB (Apolipoprotein B) 4X10-114, SORTL1 (sortilin1) 1X10-170, and GCKR (glucokinase regulator) 6X10-133. Therefore, evidence indicates that genes that play roles in the etiology of rare Mendelian forms of diseases such as diabetes and hyperlipidemia also play roles in the common polygenic forms of these diseases.
CHAPTER 1 • GENOME ARCHITECTURE AND SEQUENCE VARIATION
15
Disease-specific mutation in a Hunterian museum skeleton and his living relatives In 2011, Chahal, et al., reported that they had identified a specific mutation in the arylhydrocarbon receptor interacting protein (AIP) in four families from Northern Ireland in whom familial isolated pituitary adenoma occurred. The specific mutation in these families was a nucleotide substitution c.(910 CtoT); p.(R304X). A termination codon replaces amino acid 304, leading to a loss of 26 amino acids from the AIP protein. Chahal, et al., obtained permission from the directors of the Hunterian museum in London to extract DNA from two teeth of the skeleton of an Irish giant who died in 1783. Harvey Cushing examined this skeleton in 1909. He concluded on the basis of the degree of enlargement of the pituitary fossa that the man had a pituitary adenoma. Chahal, et al., discovered that the same AIP mutation was present in the Hunterian giant with pituitary adenoma and in the four families from Northern Ireland they studied. Analysis of DNA polymorphisms (microsatellite repeat polymorphisms) revealed that the giant skeleton DNA and adenoma patients in the four Irish families shared a haplotype that extended for 2,068 megabases on chromosome 11q13.2 and included the AIP gene. Taking into account polymorphisms, mutation rates, and generation length, Chahal, et al., concluded that the skeleton and the four families shared a common ancestor 57 to 66 generations ago.
Discovery of familial-inherited adenomas in different populations and the role of AIP In 2006, Vierimaa, et al., reported two clusters of families from Northern Finland who had familial pituitary adenoma that led to increased secretion of growth hormone and prolactin. Analysis of SNP polymorphisms in these families revealed a link between adenoma development and chromosome 11q12-q13.
16
INVESTIGATING THE HUMAN GENOME
Sequencing of genes in this region revealed a defect in the aryl hydrocarbon interacting protein AIP. Subsequent analyses in the Finnish population led to the identification of a Q14X mutation in 6 out of 45 patients with acromegaly. Karhu and Aaltonen (2007) noted that the function of AIP was not known. The amino-terminal region of AIP contains FKBP domains. These domains usually are involved in protein folding and trafficking. In the carboxyterminal region of AIP are three tetratricopeptide repeats. These repeats usually form scaffolds for the formation of multiprotein complexes. The carboxy-terminal region of AIP interacts with arylhydrocarbon receptor and with the HSP 90 heat shock protein. Low expression of AIP in pituitary adenomas is a marker for invasive growth hormone producing tumors. In 2009, Jennings, et al., reported studies on Polynesian kindred with three members who presented with pituitary macro-adenoma in childhood or adolescence. These patients had AIP mutation R271W. They presented with headaches, visual disturbances, and excessive height. Features of acromegaly were absent. Acromegaly features include frontal bossing and overgrowth of hands and feet. In 2010, Daly, et al., reported results of an international study on 96 patients with germline AIP mutations and pituitary adenomas. They noted that the patients were usually young and that the first symptoms occurred in children or adolescents. Males constituted 63.6% of the patients. The majority of the tumors were macro-adenomas. Excessive secretion of growth hormone occurred in 78% of tumors. In 13 of the 96 cases, prolactin secretion was excessive; 7 tumors were nonsecreting. In 2010, Chahal, et al., reported that they had identified 49 different AIP mutations in patients with familial-inherited pituitary adenomas. These included deletions, insertion, segmental duplications, nonsense, missense, and splice site mutations. In addition, whole
CHAPTER 1 • GENOME ARCHITECTURE AND SEQUENCE VARIATION
17
exon deletion or deletion of the entire AIP gene occurred in some patients. They noted that in the cohort of families they studied, approximately 30% of the individuals who carried a germline AIP mutation presented with pituitary tumors. Chahal, et al. (2010), concluded that the physiological role of the arylhydrocarbon receptor (ARH) likely includes cell proliferation and differentiation. ARH occurs in the cytoplasm as a multiprotein complex with AIP, HSP90, and co-chaperone p23. This complex binds to xenobiotics. It is transferred to the nucleus, where its binds with hypoxia inducible factor HIF1b, also known as ARNT. They noted that several proteins involved in the regulation of hypoxia-induced proteins play a role in tumor susceptibility. These include succinate dehydrogenase fumarate hydratase and Von Hippel Lindau proteins. (These proteins are discussed further in Chapter 5, “Pathways, Phenotypes, and Phenocopies.”) Evidence also indicates that AHR binds to ubiquitin ligase and plays a role in the degradation of estrogen and androgen receptors.
Next-generation sequencing Key elements in next-generation sequencing are the miniaturization of sequencing reactions, the sequencing of short fragments of DNA bound to solid matrices, and real-time photo-capture of the sequencing reactions. Different companies have developed a number of different platforms for next-generation sequencing; precise methods for capturing fragments and sequencing vary depending on the sequence platform used. In some cases, fragments are ligated with specific oligonucleotides at each end, and these are hybridized to matching oligonucleotides on the solid matrix sequencing platform. In other cases, DNA fragments are biotin labeled and then captured with streptavidin beads; the beads are subsequently captured on the sequencing platform. Detecting the sequencing reaction is enabled through use of nucleotides A G C T, each labeled with different colors
18
INVESTIGATING THE HUMAN GENOME
of fluorescent dyes, and fluorescent images are captured. The flow cells used as sequencing platforms are partitioned into several channels so that a number of samples can be simultaneously analyzed. Next-generation sequencing is referred to as massively parallel sequencing because thousands of short sequences are read at the same time and each sequence is read optimally about 30 times. Sequence data generated on the platform is submitted to a computer and is subsequently aligned to reference sequence. Whole-genome sequencing in humans generates a very large amount of data for analysis. In determining disease-causing mutations in humans, capturing specific genomic regions and capturing the human exome represent methods that reduce the complexity of the data analysis. Data analysis may be further simplified by prioritizing genomic regions or genes to be studied through filtering at the levels of bioinformatic analysis. Roach, et al. (2010), carried out studies on two siblings affected with an autosomal recessive disorder called Miller syndrome and their parents. They applied previous information on polymorphic markers and haplotype analysis to select areas of the genome for computational analysis following whole-genome sequencing. They focused their analysis on 22% of the genome where both affected offspring inherited the same genomic segments from both parents. In reviewing the application of next-generation sequencing to the discovery of rare gene defects that cause Mendelian diseases, Ng, et al. (2010b), emphasized that linkage information may narrow the genomic region that needs to be sequenced or computationally analyzed. Additional studies are usually required to definitively establish the significance of sequence alterations that are likely candidates for disease causation. Significant changes include chain termination substitution, deletions, and nonsynonymous nucleotide substitution that cause amino acid substitutions that likely alter the structure or
CHAPTER 1 • GENOME ARCHITECTURE AND SEQUENCE VARIATION
19
localization of a gene product. Follow-up studies include applying PCR (polymerase chain reaction amplification) and Sanger sequencing. Downstream follow-up includes biochemical and physiological studies.
Whole-genome sequencing and the discovery of mutation leading to Charcot-Marie-Tooth Neuropathy Charcot-Marie-Tooth neuropathies (CMT) are a group of disorders characterized by peripheral motor and sensory neuropathies with different modes of inheritance, including autosomal dominant, autosomal recessive, and X-linked inheritance. They are characterized clinically by symmetric distal polyneuropathy. Progressive muscle weakness and atrophy occur particularly in the peroneal muscles, leading to foot-drop and abnormal gait. CMT results from mutations in at least 39 different genes. Lupski, et al. (2010), reported that mutation testing is available in the United States for 15 of the 39 genes and costs $15,000. Lupski, et al., reported results of whole-genome sequencing and follow-up analysis on a family with CMT. Sequencing yielded 89 gigabytes of sequence data; the depth of coverage was 30, indicating that each base was sequenced 30 times. The sequence derived from the affected proband was compared to the human reference genome sequence, and differences between the two were documented. These differences included single base substitutions, small deletions, and insertions and copy number changes. Copy number variants were examined by array comparative hybridization and by sequence analysis. No copy number variants were identified that impacted genes known to play roles in CMT. Lupski, et al., focused attention on the 9,069 single nucleotide substitutions that led to nonsynonymous codon changes. Of these 121 were nonsense mutations. Data was examined to search for single nucleotide substitutions in the proband that impacted genes known
20
INVESTIGATING THE HUMAN GENOME
to cause neuropathic conditions. Two nucleotide substitutions were found in the SH3TC2 gene, one missense mutation that led to Y169H and one nonsense mutation R954X. Lupski, et al., noted that mutations in the SH3TC2 gene were previously described in Eastern European, Turkish, and Spanish gypsy patients and that the R954X mutation was present in some of these patients. In the family reported by Lupski, et al., the R954X mutation occurred in one parent of the proband and the Y169H mutations occurred in the other parent. The proband had three siblings affected with CMT, and all three carried both the R954X and Y169H mutations. Subclinical phenotypes revealed by neurophysiological studies occurred in heterozygotes for each of the mutations.
Earlier studies on SH3TC2 and Charcot-Marie-Tooth neuropathy Demyelinating autosomal recessive CMT was mapped to chromosome 5q23-q33 through homozygosity mapping in consanguineous families. Subsequently, sequence analysis of genes in this region revealed mutations in the SH3TC2 gene (Azzedine, et al., 2006). In ten consanguineous families, eight different mutations were found. Six of the mutations occurred in exon 11. Two cases had R954X mutations. Azzedine, et al., noted that the patients had foot deformities and that spinal abnormalities (kyphoscoliosis) also occurred. In an analysis of 23 English patients with autosomal recessive CMT, Houlden, et al. (2009), identified 5 patients with SH3TC2 mutations. Affected members in four families were homozygous for the R954X mutation, and in one family, the affected members were compound heterozygotes for the R954X mutation and E657K mutation. Houlden, et al., noted clinical heterogeneity in the families with
CHAPTER 1 • GENOME ARCHITECTURE AND SEQUENCE VARIATION
21
respect to the severity of neuropathy. Neuropathology on sural nerve biopsies revealed demyelinating fibers and an abnormal Schwann cell that formed onion bulb–like structures. The SH3TC2 protein localizes to the cellular plasma membrane and to the membrane of vesicles in the endocytic membrane trafficking pathway (Lupo, et al., 2009). Disruption in this pathway apparently leads to impaired interactions between Schwann cells and axons. Roberts, et al. (2010), demonstrated interaction between SH3TC2 protein and the membrane small GTPase Rab 11. Rab11 is known to regulate the recycling of internalized membranes in the endosomal pathway.
Exome sequencing Analysis is simplified when exome sequencing rather than wholegenome sequencing is carried out, because the exome constitutes approximately 1% of the genome, approximately 30 megabases (Mb). Ng, et al. (2010a), carried out exome sequencing on four unrelated individuals affected with Miller syndrome. Clinical features in Miller syndrome include micrognathia, cleft lip and palate, and eye and limb anomalies. To derive the sequence, Ng, et al., used array-based capture of exomes. Their study was initiated using DNA from affected siblings, which facilitated a search for changes in regions where siblings had identical polymorphisms and nucleotide substitutions. They identified a mutation in the dihydro-orotate dehydrogenase gene DHODH. They subsequently carried out studies on individuals affected by Miller syndrome in three unrelated families. Sequence analysis established that these affected individuals were compound heterozygotes for DHODH mutations. The DHODH gene product plays a role in pyrimidine metabolism.
22
INVESTIGATING THE HUMAN GENOME
Next-generation sequencing continues to shed light on DNA sequence changes and their potential roles in diseases. Bioinformatic analysis of sequence data is challenging, and continued development of resources for analysis is important. Equally important will be downstream analysis of the biochemical and physiological effects of sequence changes.
2 Genes and transcripts: insights into regulation at different levels
An insight of the decade In an editorial in Science entitled “Shining a Light on the Genome’s Dark Matter,” Elizabeth Pennisi (2010) reported that the elucidation of the role of conserved non-protein coding DNA in regulating gene expression constituted an insight of the decade. The dark matter of the genome gives rise to non-protein coding RNA that includes long and short non-protein coding RNAs. Both forms exert their effects partly by modifying epigenetic effects (chromatin modification). Short non-coding RNAs, including microRNAs, frequently exert their influence at the level of mRNA translation.
ENCODE Project The goal of the project designated Encyclopedia of DNA Elements (ENCODE) is to identify all functional elements in the DNA sequence. Results of the pilot phase of this project were published in 2007 (Birney and the ENCODE Project Consortium). Among the highlights of findings in this phase was the identification of transcripts that overlap more than one protein coding locus, the identification of
23
24
INVESTIGATING THE HUMAN GENOME
many non-protein coding transcripts, and the identification of numerous previously unidentified transcription start sites. Analyses also revealed that regulatory sequences frequently surround transcription start sites upstream and downstream. Control of transcription involves cis-acting regulatory sequences, including promoters, enhancers, silencers, and locus control regions. Transcription is impacted by the availability of transcription factors and the accessibility of DNA sequence for transcription factor binding. This accessibility is influenced by the modification of DNA and histones. Evidence indicates that gene expression and the packaging of DNA in chromatin are directly connected. ENCODE studies revealed that trimethylation of histone 3 lysine 4 (H3K4) occurred near actively expressed genes. Large numbers of intercalated transcripts were found, and the authors noted that the ENCODE project provided evidence that the genome encodes a network of transcripts, only some of which directly encode proteins. They advocated altering perceptions of transcription. Investigators involved with the ENCODE project determined that many of the regulatory elements previously defined as distal enhancers are located near alternate transcription start sites. They identified a large number of regulatory elements with sequences that indicated an escape from evolutionary constraint.
Changing definitions of a gene In 2002, the Human Gene Nomenclature committee defined a gene as “a DNA segment that contributes to phenotype/function, and in the absence of demonstrated phenotype/function, a gene may be characterized by sequence, transcription or homology.” Gingeras (2007) emphasized that transcripts should be considered as the fundamental elements in cataloging genome sequences.
CHAPTER 2 • GENES AND TRANSCRIPTS
25
RNA sequencing and transcriptomics Wang, et al. (2009), defined three the goals of transcriptomics. The first is to develop a catalog of all transcripts, including mRNAs, non-coding RNAs, and small RNAs. The second goal is to determine the transcriptional gene structure, including 5 prime and 3 prime (5' and 3')transcription start sites, mRNA splice patterns, and post-transcriptional modifications. The third goal is to quantify changes in expression levels of specific transcripts at different locations and at different time points. Techniques regularly applied to transcript analyses include microarray analysis and Sanger sequencing of cDNA generated by reverse transcriptase from mRNA. RNA high-throughput sequencing enables sequencing the total population of cellular RNA. Total RNA, or polyadenylated RNA, is often converted to cDNA. Use of polyadenylated mRNA to generate cDNA results in enrichment for sequence at the 3' ends of genes. Large RNAs need to be fragmented prior to ligation of adapters (tagging). In some cases, RNA is not converted to cDNA RNA; it is tagged and sequenced directly. Wang, et al., reported that the sequences read are typically 30 to 400 basepairs (bp) in length. They reported that small RNAs, including microRNAs, may be directly tagged and sequenced.
Available software programs, including ELAND, facilitate storage and retrieval of data Wang, et al., noted that high-throughput RNA sequencing is quantitative and variations in expression levels may therefore be readily determined. They noted that through direct sequence analysis of RNA, splicing diversity may be more readily defined and novel transcripts more readily identified.
26
INVESTIGATING THE HUMAN GENOME
Alternate splicing Through alternate splicing, genes generate more than one mRNA isoform. Hallegger, et al. (2010), reviewed information on the extent of alternate splicing revealed through large-scale genome analysis and transcriptome sequencing. Further insight into alternate splicing and relative transcript abundance can be obtained through sensitive microarray analysis, real-time PCR (RT-PCR), and analysis of RNA binding proteins. Halleger, et al., estimated that 92% to 94% of genes in the human genome are alternately spliced. They noted that misregulated splicing may lead to disease. These investigators emphasized the importance of RT-PCR and deep sequencing in analyzing co-regulated mRNAs. Transcription patterns may be analyzed using exon microarrays and microarrays that are enriched not only for exons, but also for exon junctions. Specific methods of identifying proteins that bind to DNA and impact transcription include UV cross-linking and immunoprecipitation (CLIP) followed by sequencing.
Insights into variation in gene expression through RNA sequencing Pickrell, et al. (2010), analyzed mRNA from lymphoblastoid cell lines; they converted this to cDNA and then carried out sequencing. Their studies led to the identification of extensive (3' untranslated (UTR) RNA regions not annotated in databases. They noted that 3' UTR regions in general are poorly annotated in databases. They also obtained evidence for unannotated exons and evidence for quantitative loci that impact the level of gene expression. They documented individual differences in expression of different alleles at specific loci. Pickrell, et al., also found evidence for individual variation in splice sites use and isoform generation. Montgomery, et al. (2010), carried out next-generation sequencing of transcriptomes in individuals genotyped in Hap Map studies. They were able to correlate differences in expression of specific loci
CHAPTER 2 • GENES AND TRANSCRIPTS
27
with particular single nucleotide polymorphisms (SNPs). These investigators also determined that specific quantitative trait loci influenced levels of transcribed non-protein coding RNA. When data from European population Hap Map samples (CEU) were compared with data on Hap Map samples derived from the Yoruba population, there was evidence that the two populations shared 33% of the quantitative trait loci. Montgomery, et al., reported that 86% of their sequence reads mapped to known exons documented in databases. This may mean that many exons are not documented in databases. They examined the generation of alternate isoforms and determined that 41% of isoforms arose as a result of exon skipping and that 17% arose as a result of alternate splice acceptor site usage. They also concluded that extensive individual variation exists in isoform generation.
Non-protein coding RNA Definite evidence indicates that a large percentage (up to 93%) of the genome is transcribed and that the majority of the transcripts do not encode proteins (Carninci, 2010). The question that arises is whether these non-protein coding RNA (ncRNAs) are functional or whether they represent transcriptional noise. High-throughput sequencing of RNA transcripts has revealed the presence of long and short (shorter than 200 nucleotides) species. Evidence also suggests that, in some cases, short ncRNAs are derived through cleavage of long ncRNAs. Carninci emphasized the instability of RNA and noted that a subset of short RNAs may represent degradation products. However, short RNAs are often produced from introns, and they include microRNA that constitute a distinct subclass. Carninci noted that ncRNAs sometimes overlap protein coding RNAs completely or partially. Long ncRNAs are transcribed across genes and may be longer than 2 megabases. Long ncRNAs frequently arise in imprinted regions of the genome and are antisense to the protein coding mRNA.
28
INVESTIGATING THE HUMAN GENOME
One class of long ncRNAs overlaps the 5' end of genes and upstream control regions. Carninci reported that shorter ncRNAs occur that overlap the middle exons of the gene, and some of these have the 5' CAP structure that is typical of mRNAs. He postulates that 5' CAP structures may have been added to these ncRNAs. The 5' capping process involves removing a phosphate group from the terminal nucleotide of the RNA transcript and adding guanosine triphosphate to generate an unusual 5' phosphate linkage. The capping enzyme complex catalyzes this process. Methyltransferase subsequently adds a methyl group to guanosine nitrogen residue. Gingeras (2007) reviewed evidence that interleaved transcription units that include protein coding RNAs and long and short noncoding RNAs often populate genic regions. He noted that small RNAs are derived in some cases from intergenic regions and in other cases from internal exons, and that they may cross-splice junctions. Evidence indicates that a subset of the small RNAs may bind to other DNA sequences and constitute components of regulatory circuits.
Functions of long noncoding RNAs In reviewing functions of long noncoding RNA (ncRNAs), Mercer, et al. (2009), reported that many are expressed during specific developmental processes and that specific long ncRNAs are expressed in the brain. HAR1 is a particularly interesting long ncRNA that is expressed in specific neurons in the developing neocortex, namely the Cajal Retzius neurons. Mercer, et al., reported evidence that a number of long ncRNAs impact the expression of nearby protein-encoding genes and that this regulation is mediated by chromatin modification. The long ncRNA HOTAIR is an antisense transcript derived from the homeobox locus HOXC. It silences expression of the HOXD locus by binding to it and recruiting chromatin remodeling complexes. HOTAIR acts in
CHAPTER 2 • GENES AND TRANSCRIPTS
29
trans to silence HOXD. Mercer, et al., noted that this process illustrates how a small number of chromatin remodeling complexes can influence expressions of a number of different developmental processes because ncRNAs binding to DNA provide specificity of repression. Long noncoding antisense RNA may impact chromatin status and methylation of corresponding and overlapping genes and induce epigenetic silencing. Tufarelli, et al. (2003), reported studies on an individual with thalassemia who had deletion of the alpha1 globin gene. This deletion led to juxtaposition of the alpha2 globin gene and a chromosomal segment that normally lies 18kb distant from alpha2 globin. They observed that the structurally normal alpha2 globin gene was not expressed. Analysis of the sequence in the region of this alpha2 globin gene revealed that it was now juxtaposed to the LUC7L gene that encodes an RNA binding protein and is transcribed in the opposite direction than the alpha2 globin. They established through PCR and sequence analysis that the transcript initiated from LUC7L continued into the alpha2 globin gene and was antisense to it. They demonstrated further that CpG dinucleotides in the alpha2 globin gene that was adjacent to LUC7A had become methylated. Whitehead, et al. (2009), reported that ncRNAs of lengths varying from 0.5Kb to 100Kb regulate gene expression through different mechanisms, including silencing promoters of a single gene or silencing a cluster of genes. Binding of long ncRNA may impact the interaction of the promoter with RNA polymerase or with transcription factors.
LINC RNAs Large intervening noncoding RNAs referred to as LINC RNAs arise from intergenic regions. They have evolutionarily conserved sequences and are implicated in biological functions. Guttman, et al. (2009), identified LINC RNAs. They observed that these genes are
30
INVESTIGATING THE HUMAN GENOME
transcribed by polymerase II. They exhibit trimethylation of lysine 4 in histone H3 at the promoter and trimethylation of lysine 36 in histone H3 along the length of the transcribed region. Guttman, et al. used chromatin immunoprecipitation with antibodies designed to capture these regions and then carried out sequencing. They identified 1,600 LINC RNAs in the mouse and established that the expression of these RNAs was regulated by p53 and by transcription factors SOX2, OCT, and NFkappa B. They also determined that LINC RNAs underwent splicing to produce mature transcripts. Loewer, et al. (2010), identified ten LINC RNAs that play critical roles in establishing the pluripotency of stem cells.
Alternate splicing of mRNA transcripts and multiple isoforms In-depth analysis of transcriptomes in different cells and tissues revealed that 92% to 94% of transcripts derived from human genes undergo alternate splicing (Wang, et al., 2008). Studies by these investigators revealed that patterns of splicing and polyadenylation of transcripts often coincided, indicating that common regulatory factors may be involved in splicing and polyadenylation site selection. Wang, et al., identified examples of tissue-specific splicing and mutually exclusive exon inclusion. One example involved the nuclear gene SLC25A3 that encodes mitochondrial phosphate transporter. They determined that transcripts of this gene include either exon 3a or exon 3b. In heart and skeletal muscle, transcripts with exon 3a predominate. In testes, liver, and other tissue, transcripts with exon 3b predominate. Wang, et al., noted that different events lead to generation of alternate transcripts; these include use of alternate 5' splice sites or alternate 3' splice sites, exon skipping, and alternate transcription initiation sites. In addition, alternate polyadenylation sites may be used, leading to differences in the length of 3' untranslated region. Wang, et al., emphasized the importance of analyzing the same tissue when comparing different individuals because most transcript
CHAPTER 2 • GENES AND TRANSCRIPTS
31
sequence differences they observed occurred between different tissues. They observed a high degree of tissue-specific regulation at the 3' end of genes that impacted the length of the 3' untranslated region of transcripts. They identified specific sequence elements that potentially bind FOX1 and FOX2 and CELF RNA binding proteins, and noted that these proteins impact alternate splicing and polyadenylation.
MicroRNAs and translational control Primary microRNAs are derived from the introns of genes or from intergenic regions. They are generated as stem-loop structures and are processed in the nucleus through activity of the DROSHA RNAase complex, to generate precursor microRNAs approximately 70 nucleotides in length. The DROSHA complex associated proteins include DGCR8, a protein encoded in the DiGeorge syndrome critical region on chromosome 22. Precursor microRNAs are exported from the nucleus into the cytoplasm. In the cytoplasm, precursor microRNAs undergo cleavage mediated by the Dicer complex. They are then loaded onto the RISC complex that contains the Argonaute protein. One strand remains associated with this complex and becomes the mature microRNA, approximately 21 to 25 nucleotides in length, that binds to the 3' untranslated region of its target mRNA (Bartel, 2009). Through their impact on gene expression, microRNAs impact key processes, including differentiation, proliferation, and metabolism. Thomas, et al. (2010), noted that each microRNA potentially regulates the expression of many different mRNAs, because only partial complementary over a short stretch of nucleotide sequence is required for binding. They noted that identifying miRNA targets is often problematic despite in silico analysis and the existence of databases. They emphasized that targets and specific functions of microRNAs may be identified through knockout of their expression
32
INVESTIGATING THE HUMAN GENOME
or through ultraviolet cross-linking and immune precipitation of microRNA-containing complexes. Thomas, et al., reported that distinct microRNA expression patterns occur in particular tissues and in tumors that arise from those tissues. Evidence indicates that overexpression of specific microRNAs may destabilize tumor suppressor genes. In some cases, microRNAs apparently act as oncogenes and impact critical signaling pathways. MicroRNAs play critical roles in brain development and synaptic plasticity (Forero, et al., 2010). The miRNA-132 and miRNA-219 impact expression of brain-derived neurotrophic factor (BDNF) and CREB (cyclic AMP response element binding protein). Knockout of the DGCR8 gene that plays a role in processing of microRNA and down-regulation of DICER lead to structural brain abnormalities. Forero, et al., reported that patterns of expression of microRNAs are altered in post-mortem brains from patients with psychiatric diseases including schizophrenia, bipolar disorder, and autism. They emphasized that further studies are required to determine the functional consequences of altered microRNA levels in these conditions.
MicroRNAs and the cardiovascular system In a review of microRNAs (miRNAs) and the cardiovascular system, Small and Olson (2011) noted that the human genome encodes approximately 1,000 miRNAs. Many of these are transcribed from a specific single miRNA gene; some are within longer transcripts that contain several miRNAs. Micro RNAs are sometimes derived from the introns of protein coding genes. Small and Olson reported a redundancy of function of miRNAs. A specific miRNA often inhibits the translation of a number of different mRNAs that are involved in a specific biological function. They noted examples of miRNAs that impact specific functions. miRNA29 regulates fibrotic response, miRNA1 impacts cardiac conduction, and miRNA 145 impacts actin cytoskeletal dynamics. Evidence indicates
CHAPTER 2 • GENES AND TRANSCRIPTS
33
that, in some situations, several different microRNAs may cooperatively impact a specific functional process. MicroRNA function can be investigated through the use of oligonucleotides that are antisense to the miRNA; they block its binding to an mRNA target. The concentration of specific miRNAs may be reduced by using decoys or sponges that have sequences similar to their mRNA targets Specific microRNAs play important roles in cardiac and vasculature development. Small and Olson noted that microRNAs play roles in cardiac disease and heart failure. Under these conditions, the pattern of microRNA expression closely resembles that observed in the fetal heart. An example of stress-dependent microRNA expression involves microRNAs encoded within the myosin heavy chain gene. These include miR2081, miR208b, and miR499. Scar formation in damaged heart tissue is accentuated by down regulation of the miR29 family; members of this family normally inhibit expression of collagen genes.
Post-translational protein modification The entire protein spectrum is referred to as the proteome. In humans, it is three orders of magnitude greater than the number of protein coding genes: Approximately 30,000 known protein coding genes exist, along with more than 1 million molecular species of proteins. Protein diversity results from differences in transcription, differences in splicing, and post-translational modification of proteins. Approximately 5% of the genome is dedicated to carrying out posttranslational modification (Walsh, et al., 2005). Post-translational modification may involve a specific chemical group attaching to one or more amino acids in the protein, or it may involve specific cleavage of the protein in limited proteolysis. Walsh, et al., noted that limited proteolysis controls cellular location and activity of proteins.
34
INVESTIGATING THE HUMAN GENOME
Specific amino acid residues tend to undergo specific modification: Asparagine serine, threonine, tyrosine, and histidine undergo phosphorylation; asparagine and serine may undergo glycosylation; and proline residues undergo hydroxylation. Lysine and arginine residues undergo methylation, and lysine also undergoes acetylation. Cysteine residues may be involved in disulfide bond formation. At least 500 different enzymes carry out protein phosphorylation. Adding phosphate residues through kinase activity frequently leads to altered protein conformation and initiates signal transduction in the case of membranes or to oncogene activation. The ABL oncogene is phosphorylated on 11 distinct residues. Addition of lipid molecules occurs; examples include addition of palmitic acid at sulfhydryl residues in cysteine and addition of myristic acid at N-terminal glycine residues. Monoubiquitination or polyubiquitination occurs at lysine residues through E3 ubiquitin ligase. Excessive modification of specific proteins occurs in certain disease states. Hyper-phosphorylated forms of the protein Tau occur in a number of different neurodegenerative conditions. Under specific cellular conditions, acetylation of lysine residues occurs in enzymes involved in glycolysis, gluconeogenesis, tricarboxylic acid metabolism, glycogen metabolism, and the urea cycle (Zhao, et al., 2010). These authors concluded that the acetylation of lysine plays a key role in regulating metabolic pathways.
Protein palmitoylation and synaptic activity The most frequent lipid modification of neural proteins involves palmitoylation, the addition of 16-carbon palmitic acid to form a stable thioester bond with a specific protein. Fukata and Fukata (2010) reported that palmitoylation is a reversible post-translational modification. This reversibility allows proteins to shuttle between cellular
CHAPTER 2 • GENES AND TRANSCRIPTS
35
compartments and palmitoylation; de-palmitoylation cycles occur in conjunction with specific physiological processes. Neuronal palmitoylation is considered in detail in Chapter 5, “Pathways, Phenotypes, and Phenocopies.”
This page intentionally left blank
3 Epigenetics: modifications of DNA, chromatin, and gene expression The definition of epigenetics has changed over the past decade as scientists carry out in-depth analyses of chromatin modification and gene expression. In 2007, Adrian Bird proposed that the term epigenetics refer to “the structural adaptation of chromosomal regions so as to register, signal, or perpetuate altered activity states.” Bird noted that this definition includes transient modifications and stable changes maintained across multiple cell generations. Information on the extent and nature of epigenetic modifications has increased rapidly over the past decade with the advent of techniques that enable genome-wide methylation analysis. Important advances in the analysis of chromatin and histones include techniques to isolate chromatin through the use of antibodies that react with chromatin-bound proteins.
Histone modifications Epigenetic studies over the past decade have revealed that acetylation of amino acid residues in histones takes place primarily in the chromatin of expressed genes. Enzymes with histone acetyl transferase activity carry out histone acetylation. Enzymes with histone deacetylase activity remove acetyl groups from histones, and deacetylated histone occurs more commonly in genes that show reduced 37
38
INVESTIGATING THE HUMAN GENOME
expression. Specific amino acid residues in histone may also undergo phosphorylation or methylation. Post-translational modification occurs primarily on lysine residues, but it may occur on arginine and serine residues as well. Histones may also undergo ubiquitination, ADP ribosylation, and sumoylation. Three classes of histone deacetylases exist. Within Class I histone deacetylase are at least 11 distinct enzymes. Class II histone deacetylases are quite distinct; this class includes sirtuins that act as nicotinamide adenine dinucleotide (NAD) dependent acetylases and sirtuins have many different cellular functions. Wang, et al. (2008), studied combinatorial patterns of histone modification in CD4 T lymphocytes. Their studies revealed that EP300 and CREBBP histone acetyl transferases associated with promoters and enhancer regions of genes while other forms of histone acetyl transferase associated not only with promoters, but also within the bodies of transcribed genes. Their studies provided evidence for cross-talk between the different forms of histone modification, including acetylation, deacetylation, and methylation. They proposed that histone acetyl transferase (HAT) and histone deacetylase (HDAC) binding to genes undergoes dynamic cycling.
Methylation and epigenetic modifications Methylation impacts DNA, RNA, histones, and nonhistone proteins. Evidence indicates that specific histone modifications and DNA methylation are linked. Approximately 50% of genes have CpG islands in their promoter regions or first exons. These are stretches of DNA sequence rich in cytosine and guanine residues, frequently located upstream of transcription start sites. Illingworth and Bird (2009) reported that tissue-specific methylated CpG islands occur
CHAPTER 3 • EPIGENETICS
39
not only in the promoters of genes, but also within the body of genes. Methylation of these islands inhibits gene expression. Following methylation of DNA at carbon 5 of cytosine in specific CpG dinucleotide sites, proteins bind to DNA at those sites. Bound proteins recruit histone deacetylases. DNA methyl transferases DNMT1, DNMT 2,DNMT3a, and DNMT3b catalyze methylation. DNMT3L acts by binding to other DNMT proteins and altering their DNA methylation activity. Several proteins have affinity for methylated DNA through their methyl-binding domains. These include proteins that have methyl-binding domains MBD1, MBD2, MBD3, and MECP2. Evidence suggests that MBD2 functions as a demethylase of methylated cytosine. Methylation of DNA also occurs in noncoding repetitive DNA sequences and in retroviral elements that are incorporated into genomic DNA. Within CpG islands (regions of the genome rich in cytosine guanine residues), cytosine residues may undergo methylation. CpG methylation is important in X chromosome inactivation and in imprinting. Evidence indicates that specific cells within the brain may differ with respect to the specific bases that undergo methylation. Kriauconis and Heintz (2009), reported that 5-hydroxy-methyl 2-deoxycytidine is a constituent of nuclear chromatin in brain cells. It is not found in many other cell types.
Insulators and maintenance of boundaries between active and silenced chromatin domains Specific DNA sequence elements function as insulators between silent heterochromatic genome regions and expressed regions (Giles, et al., 2010). Two forms of insulation occur. Enhancer block insulation
40
INVESTIGATING THE HUMAN GENOME
impacts the formation and stabilization of chromatin loops. Barrier insulation impacts the propagation of histone modification. Evidence suggests that the zinc finger protein CTCF (CCCTC binding factor) plays a role in barrier activity in insulation. Cuddapah, et al. (2009), reported that CTCF binds to DNA that separates chromatin domains with histone H3 lysine 27 methylation (H3K27Me3), from regions with histone 2a lysine 5 acetylation (H2AK5ac). H3K27Me3 occurs more frequently in regions where expression is repressed, and H2AK5ac occurs more frequently in regions with expressed sequences. A further finding of importance in the studies of Cuddapah, et al., is that CTCF binding sites differed between the two different cell types they studied, CD4 lymphocytes and HeLa cells. In the maturing erythroblast, most genes are shut down. The progressive loss of erythroblast gene expression is associated with the presence of condensed chromatin. Benz (2010) noted that a few hundred genes that encode proteins essential for oxygen transport and erythrocyte viability and membrane integrity are expressed until the terminal stages of erythroblast maturation. Expressed genes include globin and ankyrin, a membrane protein. Gallagher, et al. (2010), postulated that a region of the ankyrin erythroid promoter functions as a barrier insulator to prevent gene silencing in mature erythroblasts. They identified the insulator sequences in the region upstream of the ankyrin promoter. They then determined that specific mutations within this sequence occurred in individuals with hereditary spherocytosis and hemolytic anemia.
Analysis in living cells of interphase human chromosomes Development of techniques to metabolically label specific chromosomes and to photo-activate histone H3 facilitate the analysis of
CHAPTER 3 • EPIGENETICS
41
interphase nuclei in living cells (Muller, et al., 2010). These studies reveal that, following cell division, chromosomes occupy discrete territories in the interphase nucleus. In some interphase nuclei, sister chromosomes occupy similar positions—both are located at the periphery of the nucleus. Muller, et al., reported that chromosomes in interphase nuclei project chromatin loops. They noted that, in actively transcribing loci, chromatin loops sometimes remained local, often loops projected outside the chromosome territory. Phillips and Corces (2009) investigated whether the network of local and long-range intra-chromosomal loops and inter-chromosomal contacts occur stochastically or whether they signify specific biological processes. They presented evidence that CTCF (CCCTC binding factor) plays a key role in mediating chromatin contacts and in forming the chromatin interactome. Phillips and Corces proposed, therefore, that the zinc finger protein CTCF plays a key role in regulating chromatin architecture and that CTCF is a component of an epigenetic system that regulates interplay among DNA methylation, higherorder chromatin structure, and lineage-specific gene expression.
ATP-dependent chromatin remodeling At least 30 different ATP-dependent enzymes are involved in chromatin remodeling, the movement of nucleosomes in chromatin. Nucleosome mobility is required to form open regions of chromatin that allow transcription factor binding to DNA and facilitate gene expression. Nucleosome mobility is also important in the formation of dense chromatin that inhibits transcription factor binding and silences gene expression. Ho and Crabtree (2010) reviewed proteins that play a role in chromatin remodeling. One of the best-studied families of remodeling proteins is the SWI SNF family, first discovered in Drosophila. In vertebrates, this family includes Brahma-related factors (BAF)
42
INVESTIGATING THE HUMAN GENOME
encoded by the Brahma-related gene BRG1 and bromodomain proteins encoded by BRD. Other important factors involved in remodeling are proteins in the chromodomain family; these proteins are encoded by at least nine different genes. Mutations in the chromodomain 7 gene (CHD7) lead to CHARGE syndrome, associated with developmental abnormalities such as coloboma of the eye, heart defects, choanal atresia, retardation of growth, genital anomalies, and ear abnormalities. Products encoded by the genes p400 and SCRAP constitute a separate class of ATP-dependent chromatin modifiers that incorporate a variant form of histone H2AZ into chromatin by replacing H2A. Borrelli, et al. (2008), identified three distinct classes of chromatin-remodeling molecules: writers, readers, and erasers. The writers add acetyl, methyl, or phosphate groups to specific molecules in chromatin. Readers contain domains that recognize specific modifications of chromatin, and they may recruit additional molecules to chromatin. Within the category of readers are molecules involved in chromatin remodeling. The erasers remove chromatin post-translational modifications. This category includes histone deacetylases, demethylases, and phosphatases.
Chromatin modifications in response to neuronal stimulation Evidence indicates that the stimulation of neurons elicits enzymatic reactions that control gene expression. Borrelli, et al. (2008), reported that high levels of acetylated histone H3 and H4 are present in the promoter regions of actively expressed neuronal genes and that methylated lysine 27 in histone H3 (H3K27me) is associated with the repression of expression. Differences in the degree of methylation (mono, di-, or tri-methylation) correlate with differences in levels of gene expression.
CHAPTER 3 • EPIGENETICS
43
Borrelli, et al. (2008), reviewed evidence that chromatin modification plays an important role in synaptic plasticity. Synaptic plasticity refers to the repeated cycles of up-regulation and down-regulation of neuronal transmission through synapses and the capacity of transmission to lead to long-term potentiation or long-term depression. Activity in neurons results in dynamic changes in DNA methylation, and activity of histone acetyl transferase enzymes increases. Increased expression of the transcription factor CREB and recruitment of CREB-binding proteins occur. Activity in neurons leads to reduction of methylation and release of repressive complexes in the promoter region of BDNF (brain-derived neurotrophic factor). In specific neurons in the hippocampus, striatum and cerebellum histone deacetylases are transported out of the neurons. Borrelli, et al., reported that histone modifications play roles in the synaptic plasticity associated with behavioral memory.
Epigenetic changes in enriched environments Studies on the hippocampus in rodents have revealed that enriched environments increase BDNF messenger RNA expression and neurogenesis. Kuzumaki, et al. (2010), reported that exposure to an enriched environment leads to an increase in H3K9 trimethylation in specific BDNF promoters.
NAD-dependent class III histone deacetylases, sirtuins, circadian rhythm, and metabolism The CLOCK gene that regulates circadian rhythm encodes a protein with acetyl transferase activity. The CLOCK protein acetylates
44
INVESTIGATING THE HUMAN GENOME
histone and nonhistone proteins such as BMAL1 (also known as ARNTL1 arylhydrocarbon nuclear receptor). This protein forms a dimer with the CLOCK protein, and together they bind to specific canonical DNA sequences (CANNTG) upstream of promoters of specific genes regulated by CLOCK. Evidence indicates that CLOCK gene activity and chromatin remodeling play roles in linking circadian rhythm to metabolic activity (Eckel-Mahan and SassoneCorsi, 2010). The histone acetylase activity of the CLOCK protein is counterbalanced by the activity of sirtuin 1 (SIRT1) and NAD (nicotinamide adenine dinucleotide)-dependent histone deacetylase (Nakhata et al., 2009). NAD levels in the cell are determined by the synthesis and salvage of NAD, and may also be impacted by dietary intake of nicotinic acid. The synthesis of NAD is dependent on the presence of the amino acid tryptophan. In the initial synthesis reaction, tryptophan is transformed to N-formyl kynurenine. Thereafter, seven additional enzymes are required for the synthesis of nicotinic acid, which is then converted to nicotinamide adenine dinucleotide through the activity of glutamine-dependent NAD synthetase. NAD gives rise to phosphorylated forms (NADP) and to reduced forms (NADH and NADPH). These coenzymes play key roles as hydride acceptors and donors in metabolism and energy production (Belenky, et al., 2006). In acting as NAD-dependent deacetylases, situins transfer the acetyl group to the ribose carbon 1 in the ADP (adenosine di phosphate) portion of NAD. The NAD salvage pathway in cells is controlled by the enzyme nicotinamide phosphoribosyl transferase (NAMPT). The CLOCK gene product regulates the circadian expression of NAMPT. Nakahata aet al. (2008) determined that SIRT1 is also recruited to the NAMPT promoter regions and that it plays a role in the synthesis of its coenzyme NAD. SIRT associates with CLOCK in response to changes in cellular NAD levels. The coupling of CLOCK, BMAL1,
CHAPTER 3 • EPIGENETICS
45
and SIRT1 to the regions upstream of the NAMPT promoter enhances the transcription of NAMPT.
Linking environmental cues to gene expression Chromatin-modifying complexes usually occur in large multiprotein complexes. One example of a modifying complex in neurons contains CREB (CRE binding factor) and CBP (CREB binding protein). Heterozygous deletion of CBP leads to developmental defects, particularly in the central nervous system. Riccio (2010) reviewed the importance of CREB in determining gene activity in neurons and noted that mechanisms by which extracellular signals regulate chromatin modification have just begun to be characterized. Kuzumaki, et al. (2010), reported an example of age-related epigenetic changes that impact gene expression in neurons. They determined that doublecortin gene expression is reduced in the hippocampus of aged mice and is associated with a significant decrease in histone H3 lysine 4 trimethylation and a specific increase in H3K27 trimethylation. Other genes that impact chromatin architecture include ATRX, MECP2 (methyl CpG binding protein 2), and the cohesin-related proteins SMC1A, SMC3, and NIPBL. Defects in cohesins lead to Cornelia de Lange Syndrome. ATRX is defective in alphathalassemia, mental retardation syndrome. MECP2 mutations are associated with Rett syndrome in females. MECP2 was previously thought to bind to methylated DNA; however, evidence now suggests that it may bind to unmethylated DNA (Hansen, et al., 2010). The ATRX protein contains a helicase ATPase domain within its C terminal domain, and the N terminal domain contains DNMT3 and DNMT3L-like domains. ATRX deficiency leads to alphathalassemia and mental retardation. Law, et al. (2010), identified
46
INVESTIGATING THE HUMAN GENOME
ATRX binding targets that include CpG dinucleotides and also G (guanine)-rich tandem repeats. These repeats are particularly abundant at telomeres and have repeats of the sequence (CGCGGGGCGGGGG)n. Law, et al., analyzed the sequence of a region that lies 1kb upstream of the alpha globin locus on 16p13.3 in different individuals and determined that polymorphisms exist in the length of this repeat. These polymorphisms influence the degree of ATRX binding. Longer repeats bind more ATRX, leading to greater suppression of alpha globin expression. Evidence indicates that ATRX binds in other telomere regions, such as in 4q35.
Diverse mechanisms by which MECP2 impacts gene expression The gene MECP2 was identified through studies on female patients with Rett syndrome, a condition characterized by cognitive and physical regression, stereotypic movements especially involving the hands, and autistic behaviors (Amir, et al., 1999). Initial studies revealed that MECP2 protein binds to methylated DNA sequences via a methyl CpG binding domain and leads to silencing of expression. Subsequent studies revealed that silencing of expression involves not only binding of MECP2, but also recruitment of other proteins, including SIN3A and ATRX (Nan, et al., 2007) and CREB2 (Chahrour, et al., 2008). Forlani, et al. (2010), carried out bioinformatic analyses to identify MECP2 interacting proteins. They determined that the transcription factor Yin-Yang 1 (YY1) binds to MECP2 and that YY1 protein and MECP2 protein interact in vivo and in vitro. A gene on chromosome 14q32 encodes YY1 protein. Gabellini, et al. (2002), determined that YY1 protein forms part of a multiprotein complex that binds to a repetitive DNA element
CHAPTER 3 • EPIGENETICS
47
(D4Z4) on chromosome 4q35. Binding of Y1Y1 leads to altered chromatin structure and altered expression of genes located upstream of D4Z4. These genes include ANT1 (adenine nucleotide translocator 1), which is active in transferring nucleotides into mitochondria. Forlani, et al. (2010), determined that, in MECP2-deficient mice, the expression of ANT1 was reduced. This finding is of particular interest in view of evidence that altered mitochondrial function occurs in Rett syndrome. Gabellini, et al., demonstrated that deletions of D4Z4 repeats on 4q35 occur in fascio-scapulo-humeral muscular dystrophy (FSH). Genes located upstream of D4Z4 have been found to be inappropriately overexpressed, specifically in FSHD muscle. An element within D4Z4 has been shown to behave as a silencer that provides a binding site for a transcriptional repressing complex. These results suggest a model in which deletion of D4Z4 leads to the inappropriate transcriptional derepression of 4q35 genes, including ANT1 overexpression, resulting in disease (Tupler and Gabellini, 2004). The transcriptional repressor complex that is not appropriately bound when D4Z4 repeats are deleted includes MECP2, YY1, and ATRX.
Therapeutic interventions based on epigenetic changes Because deactivation of genes and reduced expression is commonly associated with histone acetylation, many investigators have proposed using histone deacetylase (HDAC) inhibitors to treat specific diseases. However, evidence now suggests that several classes of histone deacetylase exist and that these enzymes have functions other than the deacetylation of histones (for example, they deacetylate other proteins). Inhibitors of histone deacetylase developed for therapeutic intervention thus far are pan-HDAC inhibitors that have side effects. Fischer, et al. (2010), noted the necessity of developing and
48
INVESTIGATING THE HUMAN GENOME
testing HDAC inhibitors with increased specificity. Furthermore, they noted that additional information on the distribution of specific HDACs in particular cells and tissues is required. Histone acetylation in eukaryotes is dependent upon acetyl coenzyme A. Citrate derived primarily from glucose metabolism is converted to acetyl coenzyme A through activity of adenosine triphosphate (ATP) citrate lyase. Studies by Wellen, et al. (2009), revealed that the inhibition of ATP citrate lyase impairs histone acetylation that is required for cellular response to growth factors. They noted also that histone acetylation increases the affinity for other proteins and strengthens the interaction between DNA and histones. Enzymes that impact histone acetylation include histone acetyl transferase, which increases histone acetylation and histone deacetylase, which removes acetyl residues. Sirtuins also function as histone deacetylases and require nicotinamide adenine dinucleotide (NAD) as a co-factor. Sirtuins are responsive to the NAD/NADH ratio in the cell. Wellen, et al. (2009), studied the impact of short inhibitory RNA mediated silencing of ATP citrate lyase on histone acetylation. They demonstrated that this leads to decreased acetylation of all core histones. This effect could be rescued by inhibition of histone deacetylase.
Identification of mutations in chromatin modifier genes in tumors Wiegand, et al. (2010), carried out sequence analysis of whole transcriptomes in 119 tumors identified as ovarian clear cell carcinoma. They identified mutations of the ARID1A gene in 46% of cases. Jones, et al. (2010), carried out exome sequence analysis of cancer cells immune-precipitated from ovarian clear cell carcinoma tumors. They determined that 57% of tumors had mutations in ARIDA.
CHAPTER 3 • EPIGENETICS
49
The ARIDA gene locus on chromosome 1p36.11 encodes BAF250, a subunit of the SWI-SNF chromatin remodeling complex. This complex interacts with ATP and plays a role in the movement of nucleosomes and the modification of chromatin structure to modulate the accessibility of chromatin for binding transcription factors. Mutations in SNF5 that encodes a different subunit of the SWI SNF complex play a role in cancer. Mutations in epigenetic effector enzymes have been found in many different tumors. Specific effector enzymes mutated in cancer include DNA methyltransferase DNMT3A, EZH2 histone methyltransferase, and UTX histone demethylase (Wilson, et al., 2010).
Variant histones Histone H2AX is recruited to regions of DNA damage. In chromatin surrounding double-stranded DNA breaks, histone H2AX rapidly becomes phosphorylated (van Attikum and Gasser, 2005). Histone H2AZ accumulates in promoter regions at the time of transcription and plays a role in the recruitment of RNA polymerase II, the key enzyme in transcription. Hardy, et al. (2009), demonstrated that H2AZ molecules associated with promoters are acetylated, whereas in heterochromatin that is not expressed, H2AZ that accumulates is not acetylated.
Parent of origin allelic expression Improved techniques in transcriptome analyses in different tissue and enhanced capabilities to analyze polymorphisms on transcripts have led to greater insights into the existence of allele-specific gene expression and the definition of parent of origin expression. Gregg, et al. (2010), used high-throughput RNA sequencing to examine transcripts present in different areas of the mouse brain.
50
INVESTIGATING THE HUMAN GENOME
They determined that 1,300 genes showed parent of origin expression effect—that is, in a particular transcript, either the maternally derived alleles or the paternally derived alleles were expressed, Furthermore, they established that the specific parental allele expressed varied at different ages. Gregg, et al., noted that earlier studies in human and mice revealed that approximately 100 imprinted genes exist. Imprinting usually refers to the phenomenon of complete silencing of expression of the alleles derived from one parent, from early life onward. Gregg, et al., uncovered 824 genes with parent of origin expression effects in specific tissues. They demonstrated that, in the mouse brain, the cadherin 15 genes show preferential expression of the paternal allele. In humans, intellectual disability is associated with variants in cadherin 15. Gregg, et al., reported evidence for parent-specific expression of particular isoforms. In the case of the Herc3 gene, specific isoforms were expressed from the maternally derived allele, while other isoforms were expressed from the paternally derived allele. Findings of Gregg, et al., indicate that parental bias in gene expression is a widespread phenomenon. This has implications in cases where mutations in specific genes are present. The phenotypic effect of the mutation depends on the parent of origin of expression. For example, if a mutation is inherited from the mother but transcription from that gene is silenced on the maternal chromosome, the mutation will not have an impact.
Epigenetic studies in phenotypically discordant monozygotic twins Because monozygotic (MZ) twins are derived from the fusion of a single egg and a single sperm, they are assumed to be genetically
CHAPTER 3 • EPIGENETICS
51
identical. However, phenotypic differences between MZ twins are sometimes observed. These differences may result from epigenetic factors or from post-zygotic structural changes or mutations in DNA. The fertilized egg has many mitochondria. Differences in MZ twins may also arise because of differential segregation of mitochondria in the early stages of cell division. Nguyen, et al. (2011), carried out large-scale CpG island methylation profiling in a lymphoblastoid cell line derived from MZ twins discordant for autism. They also studied nonautistic siblings of the twins. Two genes showed differential methylation in discordant autistic MZ twins. These genes were RORA (retinoic acid-related orphan receptor alpha) and BCL2 (apoptosis regulator). Differential methylation was also observed between autistic individuals and their siblings. RORA expression was found to be reduced in brain samples from autistic individuals. The RORA gene plays an important role in nervous system development. The BCL2 gene is involved in cell death and apoptosis.
DNA methylation and aging Hernandez, et al. (2011), analyzed DNA methylation at more than 27,000 CpG sites throughout the genome. They specifically analyzed DNA recovered from the brain, including the frontal cortex, temporal cortex, pons, and cerebellum. They discovered a highly specific correlation between methylation status and age. They reported that age-related methylation changes particularly impact genes that play a role in transcription regulation. In most cases, increased methylation correlates with increased age.
52
INVESTIGATING THE HUMAN GENOME
Epigenetics and stem cell biology Epigenetic changes are an important consideration in stem cell biology. Pluripotent stem cells are not completely reprogrammed epigenetically by the factors that induce pluripotency. They retain epigenetic features of the cells from which they were derived. Stem cells produced by somatic nuclear transfer show more complete programming than stem cells produce through pluripotent induction with specific factors. Nevertheless, aberrant chromatin remodeling is sometimes present in cells derived using somatic nuclear transfer (Polo, et al., 2010). Abundant evidence indicates that epigenetic events, including modification of DNA chromatin and proteins, play key roles in the temporal and spatial control of gene expression. Epigenetic factors are also involved in linking gene expression with metabolism.
4 Gene-environment interactions In many genetic disorders, evidence indicates that individuals who carry the same mutation in the same gene do not necessarily have the same disease manifestations. This is true even in families that harbor the same mutation. One question that arises is to what extent modifier genes influence expression. Another important question is whether environmental factors, internal or external, interact with specific gene mutations and alter their impact. Penetrance of a specific allele that impacts the phenotype is defined as the probability that an individual who carries that allele will manifest the corresponding phenotype. Evidence also indicates that individuals with mutations in a specific gene occur more frequently in some regions of the world. The question that arises whether natural selection operates and whether heterozygotes for specific mutations have a selective advantage in particular environments.
Factors that impact penetrance and expressivity Factors that influence penetrance include age, genetic background and environmental factors. Examples of the impact of age on disease penetrance include neurodegenerative disorders, such as
53
54
INVESTIGATING THE HUMAN GENOME
Huntington’s disease, which follow Mendelian inheritance patterns yet result in disease symptoms only in middle age or later. Variable expressivity refers to the observation that, even in disorders with clear-cut Mendelian inheritance, the specific manifestation of the disease may vary among family members. Variable expressivity may be influenced by modifier genes, subsequent somatic mutations or rearrangements, or environmental factors. Variable expressivity in dominant disorder tuberous sclerosis is partly influenced by the occurrence of second hits on the homologous chromosome. For example, in a patient with a defect in the TSC2 gene on chromosome 16p13.3, demonstrated in blood cells and fibroblasts, the angiomyolipoma tumor in the kidney had the TSC2 mutation on one chromosome and a deletion on chromosome 16p13.3 that removed the remaining TSC2 gene (Green, et al., 1994). However, additional aspects of variable expressivity affect tuberous sclerosis; for example, hormonal factors may explain why women with tuberous sclerosis who are of child-bearing age are at greater risk for developing lymphangioleiomyomatosis of the lung than are males (Yu, et al., 2004). Comprehensive genomic analysis will facilitate the identification of modifier genes that play roles in variable penetrance and expressivity.
Penetrance in adult-onset hemochromatosis In Northern Europeans, the predominant cause of adult-onset hemochromatosis is homozygosity for the C282Y mutation in the protein encoded by the HFE gene (hemochromatosis gene, also known as high iron) on chromosome 6p21.3. The prevalence of C282Y homozygosity is 1 in 200 in these populations. However, not all homozygotes for this allele present with clinical features of hemochromatosis or with biochemical evidence of iron overload.
CHAPTER 4 • GENE-ENVIRONMENT INTERACTIONS
55
Rochette, et al. (2010), reviewed hemochromatosis and noted that expression of iron overload depends on interaction of the HFE genotype with modifier genes and environmental factors, including infection with hepatitis virus and alcohol consumption. In homozygotes for the C282Y mutation, the iron burden is usually apparent only after the second decade of life. In women, the manifestations often occur later because menstruation and associated iron loss provide some protection. Different symptoms occur at different stages; early symptoms include joint pain and general weakness. Later symptoms include skin pigmentation changes, evidence of liver dysfunction, and diabetes mellitus. Biochemical penetrance of the homozygous C282Y mutation is assessed by measuring serum iron, serum transferrin, transferrin iron saturation, and serum ferritin. The degree of iron saturation of transferrin is considered to be the best marker. Levels of iron saturation of transferrin in patients range between 40% and 100%. Rochette, et al., noted evidence that common polymorphisms in a number of different genes modify the risk of developing clinical manifestations of hemochromatosis. Modifier polymorphisms occur in the haptoglobin gene, in the inflammatory cytokine TNF alpha, and in BMP2 (bone morphogenic protein). These polymorphisms impact the synthesis of major iron regulator hepcidin. It is important to note that rare forms of hemochromatosis occur with other genes involved in controlling iron absorption and release.
Genetic factors and protection against disease induced by environmental agents Malaria Genetic factors play a major role in determining disease course and survival in malaria. Williams (2006) reviewed human red-cell polymorphisms that impact malaria. He emphasized that the red cell
56
INVESTIGATING THE HUMAN GENOME
plays a major role in controlling the malaria life cycle. Variants in the enzyme glucose-6-phosphate dehydrogenase (G6PD) that lead to decreased activity of this enzyme occur more frequently in areas of the world where malaria is endemic. G6PD is encoded on the X chromosome. Comprehensive population studies have revealed that heterozygotes (females) and hemizygotes (males) for G6PD deficiency are resistant to severe malaria. Mutations leading to G6PD deficiency occur in Africa, Asia (particularly India), Southern Europe, and regions of the Middle East. The most important pathological consequence of G6PD deficiency is drug- and infection-induced hemolysis. Drugs that induce hemolysis include sulfonamides (such as sulfacetamide, sulfanilamide, and sulfapyridine), nitrofuradantin, and antimalarials (including primaquine). Hemolysis may lead to weakness, body aches, and, in some cases, renal failure. G6PD mutations that involve the NADP binding site cause the most severe reactions. G6PD typing is an important consideration when testing and prescribing antimalarial drugs. Mutant alleles that give rise to abnormal forms of beta globin (such as HbS, HbC, and HbE) occur in regions of the globe where Plasmodium falciparum malaria is endemic. Analysis of haplotypes in individuals who have the HbS allele indicates that it likely arose more than once in Africa and in the Middle East (Weatherall, 2008). The exact mechanism by which HbS provides protection against malaria is not clear. Proposed mechanisms include enhanced clearance of parasitized HbS containing cells from the spleen, reduced oxygen tension in HbS cells and generation of an environment less supportive to parasite growth, and the presence of a higher titer of IgG antibodies to the Plasmodium falciparum parasite in HbS individuals. Hemoglobin C is common in West Africa. Homozygotes for HbC have a mild hemolytic anemia; both homozygotes and heterozygotes
CHAPTER 4 • GENE-ENVIRONMENT INTERACTIONS
57
have increased resistance to malaria. Evidence indicates that altered properties of HbC-containing red cells lead to reduced parasite cytoadherence and increased clearance of cells infected with Plasmodium falciparum parasites (Kwiatkowski, 2005). Hemoglobin E occurs with very high frequency in eastern India and Southeast Asia, regions where malaria is endemic. Mild anemia occurs in HbE homozygotes. Evidence suggests that HbE-containing cells are more resistant to invasion by malaria parasites (Chotivanich, et al., 2002). Beta-thalassemia associated with a deficiency of beta globin chains due to mutations also occurs more commonly in regions of the world where malaria is endemic. Alpha-thalassemia and malaria Alpha-thalassemia occurs in individuals when one of the alpha globin genes is deleted or inactivated due to mutation. It results in mild anemia in homozygotes and produces no symptoms in heterozygotes. It occurs with very high frequency in India, Southeast Asia, and Melanesia. Alpha0-thalassemia is associated with the deletion of both alpha genes and results in more severe anemia. A survey of hospital admissions revealed that, compared to individuals with normal globin genes, the risk for admission to the hospital with severe malaria decreased to 0.66 in alpha-thalassemia heterozygotes and to 0.4 in alpha-thalassemia homozygotes (Weatherall, 2008). Severe malaria is defined as the occurrence of three clinical manifestations, severe anemia, respiratory distress, and cerebral malaria. Parasite-infected erythrocytes in alpha-thalassemia are less likely to form rosettes with normal cells and are less likely to adhere to endothelial membranes. Evidence indicates that individuals who are heterozygous for sickle hemoglobin HbS or heterozygous for HbC may be more efficient transmitters of parasite gametocytes to the Anopheles mosquito gut (Gouagna, et al., 2010).
58
INVESTIGATING THE HUMAN GENOME
Red-cell membrane proteins that impact malaria infestations The malaria parasite binds to and enters the red cell. Genetic variations in red-cell membrane proteins impact the efficiency of this binding. Deletion of the red-cell membrane protein Glycophorin C reduced Plasmodium falciparum invasion of cells. This deletion gives rise to the Gerbich blood group that occurs in Papua New Guinea (Maier, et al., 2003). The red cell complement receptor CR1 plays a role in rosette formation, a phenomenon that involves the adherence of normal red cells to parasite-infected cells and clumping. Individuals with low expression of CR1 manifest decreased rosette formation and resistance to severe malaria. Mutations in CR1 leading to low expression are reportedly common in endemic malarial areas, such as Papua New Guinea. Duffy blood group and malaria The frequency of Plasmodium vivax infection and malaria is lower in individuals with the Duffy negative blood group. The Duffy antigen determines gene functions as a receptor for the Plasmodium vivax parasite. This observation led to development of a vaccine that specifically targets the P. vivax protein that binds to the Duffy receptor (Beeson and Crabb, 2007).
Gene-environment interactions in the pathogenesis of chronic obstructive pulmonary disease In a review of lung diseases characterized by abnormal fibrogenic reactions, Araya and Nishimura (2010), noted that numerous environmental factors play roles in the etiology of this disease, including exposure to fumes, smoke, metal particles, and antigens. Individual differences exist in the propensity for environmental factors to induce damage and in the dosage of these factors that is required for pathology to develop. The key pathological feature is airway wall
CHAPTER 4 • GENE-ENVIRONMENT INTERACTIONS
59
fibrosis that may occur in chronic obstructive pulmonary disease (COPD) and asthma. Parenchymal fibrosis may also occur. Pneumocytes that normally cover alveolar surface are lost, and a fibrotic response is present. Araya, et al., noted that the key factor involved is transforming growth factor beta pathway. In response to chronic injury, TGFbeta becomes a profibrogenic cytokine that leads to fibroblast accumulation, increased expression of type 1 and type III collagen in the pulmonary interstitium and in airways. TGFbeta is normally stored in a latent form in the extracellular matrix associated with the protein LAP (latency associated protein). This protein may undergo conformational change or cleavage in response to interaction with denaturing substance, reactive oxygen species, or metalloproteinase. Ito, et al. (2008), reported that a specific haplotype structure of eight SNPs within the TGFbeta gene occurred with significantly greater frequency in smokers with emphysema than in unaffected smokers. The discovery of the important role of TGFbeta activation in this disease opens the way for a number of therapeutic possibilities.
DNA damage and repair Lesions in DNA may result from hydrolytic reactions and can be induced by reactive oxygen species. Environmental agents that damage DNA include UV light, ionizing radiation, heavy metals, and chemical pollutants such as those present in tobacco. Jackson and Bartek (2009) reviewed DNA damage in biology and disease. They noted that different physiological mechanisms are used to repair different forms of DNA damage. These include mismatch repair, base excision repair, and nucleotide excision repair. The latter involves the excision of 22 to 30 nucleotides in DNA, followed by single-strand repair. In double-stranded DNA repair, DNA lesions require repair by nonhomologous end joining or homologous
60
INVESTIGATING THE HUMAN GENOME
recombination. Nonhomologous end joining can occur throughout the cell cycle but is error prone. Homologous recombination occurs in G2 and primarily in the S phase, and requires sister chromatids to repair damage. Insight into processes involved in the repair of DNA damage has come through studies in patients with conditions in which chromosome breakage and abnormal chromosome fusions arise in somatic cells and through the mapping and isolation of genes involved. These conditions include ataxia telangiectasia and Nijmegen breakage syndrome, and were reviewed by Derheimer and Kastan (2010). In ataxia telangiectasia (AT), patients develop manifestations of cerebellar ataxia and neurodegeneration. They also manifest subtle telangiectasia often visible in the eye. In addition, they manifest increased sensitivity to radiation and increased cancer predisposition. Cultured cells have shown evidence of defects in the cell cycle and of genomic instability. The AT gene locus was mapped to chromosome 11q22.23 and was found to encode a specific serine threonine protein kinase designated ataxia telangiectasia mutated (ATM). Derheimer and Kastan (2010) noted that ATM is a member of the phosphoinositide-3-kinase family (PIKK) that includes protein kinases that interact with serine and threonine and that are involved in signaling following cellular stress. Members of this PIKK family of kinases form complexes with other proteins, and these interactions are necessary for their activation and function. Proteins that form complexes with ATM include other proteins involved in the radiation response: RAD50 (DNA repair protein) MRE1 (metal regulatory transcription factor, also abbreviated as MTF1); and NBS1 (nibrin), encoded by the Nijmegen breakage syndrome locus. ATM plays an important role in controlling the cell cycle and is required to inhibit S-phase DNA synthesis following exposure to DNA damage, such as damage induced by ionizing radiation. It is
CHAPTER 4 • GENE-ENVIRONMENT INTERACTIONS
61
extremely important that entry into S phase be delayed following exposure to radiation in order for repair to take place prior to new DNA synthesis. ATM plays a role at the checkpoint before entry into S phase from G1. A key function of ATM is phosphorylation of threonine 68 in the checkpoint protein CHK2. Evidence also indicates that ATM plays a role in the repair of double-stranded DNA breaks. Derheimer and Kastan noted that intact function of ATM and NBS1 are necessary for the disruption of nucleosomes at sites of DNA damage. The histone H2AX binds to an adaptor protein and is recruited to double-stranded DNA breaks. There it may act as a docking station for other proteins involved in repairing DNA breaks. ATM recruited to these damage sites leads to the phosphorylation of histone H2AX. Derheimer and Kastan reported that chromosome end-to-end fusions, involving telomeric regions of chromosomes, often occur in ataxia telangiectasia patients. The shelterin complex caps and protects telomere ends. This complex consists of multiple proteins, including TRF1 and TRF2 (telomere repeat binding factors). Depletion of TRF2 disrupts the shelterin complex, leading to uncapped telomeres and activation of the ATM-dependent DNA damage response.
DNA damage response and chromatin: Fanconi anemia genes As noted previously, double-stranded DNA breaks elicit a signaling cascade in which the chromatin surrounding the DNA is modified. Evidence indicates that the ubiquitination of proteins plays a key role in this response and that a number of different ubiquitin ligases are involved. Insight into the role of ubiquitin ligases in doublestranded DNA break response came in part through studies on specific genetic disorders. One of these disorders is Fanconi anemia.
62
INVESTIGATING THE HUMAN GENOME
Gene-mapping studies and in vitro complementation studies and subsequent DNA sequencing at mapped loci led to the discovery that mutations in any of 13 different genes can lead to Fanconi anemia. At 12 of these loci, the disease is inherited as an autosomal recessive condition; and at one locus, it is inherited as an X-linked recessive condition (Kee and D’Andrea, 2010). The Fanconi disease locus FANCD1 that maps to human chromosome 13 is identical to BRCA2, the breast ovarian cancer gene. A functional characteristic of Fanconi cell lines shared by BRCA2 negative cells is that exposure of these cells to DNA cross-linking agents such as diepoxybutane (DEB) results in an increase in chromosome damage and the formation of tri-radial chromosomes. Diagnosis in suspected cases of Fanconi anemia may be made by using the DEB-induced chromosome damage test. Evidence now suggests that proteins encoded by the Fanconi anemia loci interact with a number of different proteins involved in the DNA repair pathway. Kee and D’Andrea (2010) reported that somatic mutation and epigenetic silencing of a number of different Fanconi anemia genes occurs in various cancers and that defects in DNA repair and genomic DNA instability occur in cancer. The genes FANCF and FANCA are mutated in some forms of myeloid leukemia. Genes in the FANCA pathway are mutated in some forms of early-onset pancreatic cancer. Andrea reported that the Fanconi anemia genes work in concert to control the monoubiquitinated state of FANCD2 and FANCI. Ubiquitin is added to the internal lysine of the proteins, and the presence of DNA damage activates monoubiquitination. Ubiquitinated FANC1 and FANCD2 move to the chromatin surrounding the DNA break and then interact with other proteins at that site, including the Fanconi anemia core complex, composed of eight other FA proteins. A key step in the formation of this complex is phosphorylation of FA proteins by ataxia telangiectasia kinase ATM.
CHAPTER 4 • GENE-ENVIRONMENT INTERACTIONS
63
Following the repair of DNA damage, ubiquitin is removed from FANC1 and FANCD2 through the action of specific proteases.
DNA damage response and genomic integrity Genomic instability is a consequence of the functional impairment of DNA damage-response mechanisms. These mechanisms include the ATM pathway (ataxia telangiectasia mutated) and the ATR response (ataxia telangiectasia and Rad3 related). Evidence indicates that the ATM-ATR response is impaired in Seckel syndrome. This syndrome is characterized by impaired growth leading to dwarfism; microcephaly; and unusual facial features that include a sloping forehead, a high nasal bridge, a beaked nose, and retrognathia. Alderton, et al. (2004), reported that cells from Seckel syndrome patients displayed impaired phosphorylation of ATR protein, impaired cell-cycle checkpoint arrest, and elevated numbers of micronuclei. These abnormalities were induced upon exposure to ultraviolet light and agents that led to stalling of DNA replication. They also reported that Seckel syndrome patient cells had an increased number of centrosomes. In 2008, Griffith, et al., reported that mutations in the pericentrin protein lead to Seckel syndrome. Pericentrin is expressed in the centrosome. It plays an essential role in centrosome function and in cellcycle progression. The centrosome is the microtubule building and organizing center of the cell. During cell division, when the nuclear membrane disintegrates, centrosomal-related microtubules contribute to the mitotic spindle, and they associate with chromosomes. In 2011, Kalay, et al., reported the result of SNP genotyping studies that they carried out in four Seckel syndrome patients in three different Turkish families. They found evidence for linkage of Seckel
64
INVESTIGATING THE HUMAN GENOME
syndrome to chromosome 15q21.1-q21.2 in these families. Their studies also revealed homozygosity for a 3.4 megabase region in this location, and they identified a founder haplotype. They concluded that the gene CENP152 represents the most likely candidate gene in this region for Seckel syndrome. This gene encodes centrosomal protein 152. DNA sequencing revealed homozygosity for a splice-site mutation in intron 4 of CENP152 in all affected individuals in these families. In subsequent studies on Seckel-affected individuals in families from other parts of Europe, Kalay, et al., identified homozygosity for seven other CENP152 mutations, including nonsense mutations and splice-site mutations. CENP152 protein is located within the centrosome and colocalizes with pericentrin. Kalay, et al., reported that metaphase karyotyping on lymphocytes from patients with CENP152 mutation revealed aneuploidies in 15 out of 109 metaphase spreads. Prematurely separated sister chromatids were also observed. In addition, many cells contained multiple nuclei, and fragmented chromosomes were also present. They determined that the impaired CENP152 protein in Seckel patients leads to increased phosphorylation of histone H2AX, indicating replication stress.
Ubiquitin ligases and double-stranded DNA breakage response One ubiquitin ligase involved in the response to double-stranded DNA breakage is ring finger protein RNF168, involved in the cascade of reactions that lead to the ubiquitination of histone H2AC and phosphorylated histones. De-ubiquitinating enzymes are negative regulators of this cascade. Bekker-Jensen, et al., 2010, reported that HERC2 forms a complex with RNF168. Following interaction with HERC2, RNF8 assembles with ubiquitin-conjugating enzymes, and
CHAPTER 4 • GENE-ENVIRONMENT INTERACTIONS
65
this promotes the formation of lysine 63-linked ubiquitin chains. The HECT domain in HERC2 is essential for its function. Of particular interest is the fact that ionizing radiation induces the phosphorylation of HERC2, and this phosphorylation is required for the interaction with RNF168. HERC2-depleted cells manifest increased sensitivity to ionizing radiation. Herc2 locus disruption in mice leads to developmental delay, neuromuscular problems, and early lethality.
Signatures of environmentally induced DNA damage The risk of lung cancer is 20 times greater in smokers than in nonsmokers; more than 60 different mutagens are present in tobacco smoke. Pleasance, et al. (2010), used massively parallel DNA sequencing to investigate somatic mutations in a small-cell lung cancer line. Their sequence analyses revealed more than 22,910 somatic mutations in the tumor cells. They reviewed tobacco carcinogen-related mutagenesis and noted that three processes are involved. These include mutageninduced chemical modification of purine nucleotides, failure of repair, and incorrect nucleotide incorporation during replication. They found that guanine-to-thymidine (G-to-T) transversions were highest in the tumor cells. These transversions have been linked to exposure to aromatic hydrocarbons and acrolein. Pleasance, et al., also found increased guanine-to-cytosine (G-toC) transversions, particularly at unmethylated CpG sites. They noted that previous reports have indicated that G-to-C transversions are associated with exposure to polycyclic aromatic hydrocarbons that contain a cyclopentane ring.
66
INVESTIGATING THE HUMAN GENOME
Their studies revealed that transcription-coupled repair and expression-coupled repair are more common in tumors. Somatic acquired unbalanced genomic rearrangements were also identified in the tumor cells. A number of these involved chromosomes 1p32-p36 and 4q25-q28. Rearrangements of chromodomain helicase CHD7 were also common in small-cell cancer cell lines. In a melanoma cell line, Pleasance, et al. (2010), found evidence of increased frequency of cytosine-to-thymidine (C-to-T) transversions, particularly at pyrimidine dinucleotides CC to TT. These transversions are signatures of ultraviolet exposure.
5 Pathways, phenotypes, and phenocopies The challenge of the 21st century for all scientific and engineering disciplines is understanding and managing complexity. —C. Aufray, D. Charron, L. Hood, 2010 Identifying a specific gene responsible for a genetic disease is cause for celebration because it opens the way for diagnostic testing and in-depth investigation of the molecular pathogenesis of disease. Through such studies, therapies may be developed. From time to time, patients who present with symptoms and signs typical of a specific genetic disease fail to reveal changes in the genes known to cause that disease. These patients may have a different disease that is referred to as a phenocopy of the known disease because the phenotypes are highly similar. Research is then required to identify the gene responsible for the phenocopy. In very rare cases, environmental factors may cause a disease with manifestations that closely resemble those of a genetic disease. In some cases, a specific genetic disorder and its phenocopies may be due to mutations in genes that are active in the same physiological pathway. Similar disease manifestations may arise from mutations in any one of the proteins that occur in a multiprotein complex. However, sometimes the different genes that lead to a specific syndrome when mutated have not yet been mechanistically linked.
67
68
INVESTIGATING THE HUMAN GENOME
Phenocopies in Mendelian disease Huntington’s disease (HD) is characterized by movement disorder, cognitive decline, psychiatric symptoms, and a history indicative of autosomal dominant inheritance. Wild and Tabrizi (2007) emphasized that phenotypic variation occurs in patients with HD; 90% of patients develop movement disorders, and cognitive impairment is common. However, psychiatric symptoms may be absent in some patients. The movement disorder is variable and may include pure dystonia, ataxia, akinetic syndromes, and rigid syndromes. HD is due to trinucleotide repeat expansion in the gene that encodes the protein huntingtin, and this leads to expansion of the number of glutamate amino acids in the protein. Wild and Tabrizi (2007) reported that, in 1% of patients thought to have Huntington’s disease, the CAG repeat expansion in the huntingtin encoding gene is not detected. They carried out molecular studies on patients with the HD phenotype who did not have HD CAG repeat expansion. They determined that some of these patients have familial prion disease due to mutations or insertions in the PRNP gene. This disease is sometimes designated as HDL1. Another group of patients have a triplet repeat expansion in the gene that encodes the protein junctophilin 3 (JPH3). This disorder, designated HDL2, is more common in patients of African or Middle Eastern descent. It is not clear how mutant forms huntingtin and junctophilin lead to similar disease manifestations.
Mutations in a specific gene and occurrence of different disease phenotypes at different ages Individuals with repeat expansion in the promoter region of the Fragile X gene FMR1 that are classified as pre-mutations (55–200 repeats) do not usually manifest symptoms in childhood or in early adulthood. Through studies on multigeneration families, Hagerman
CHAPTER 5 • PATHWAYS, PHENOTYPES, AND PHENOCOPIES
69
and Hagerman (2004) determined that FMR1 premutation carriers are at risk for a number of symptoms in adult life and in the later decades. Older males with the premutation in the FMR1 promoter region are at risk for a well- studied disorder designated Fragile X tremor ataxia syndrome associated (FRXTAS/FXTAS). Individuals with this syndrome manifest tremors and balance problems; they may also have executive function and memory deficits. In these individuals, the concentrations of FMR1-encoded protein are normal, but the levels of FMR1 mRNA are increased. Abnormal neuronal inclusions occur that are positive for FMR1 mRNA. Ross-Inta, et al. (2010), reported that specific mitochondrial functional changes occur in premutation carriers, leading to uncoupling of ATP synthesis and electron transport activity in the mitochondrial respiratory complex. They reported that a 50% reduction in function of specific mitochondrial complexes occurred in individuals with an FMR1 CGG repeat number of 58 +/– 2. Studies revealed that changes in mitochondrial function occurred earlier than the accumulation of neuronal inclusions.
mTOR: A key pathway involved in Mendelian disorders, cancer, and diabetes Chapter 6, “Dynamic Function, Synaptic Activity, and Plasticity,” describes mTOR functions relevant to the synapse. Studies on the regulation and functions of the mTOR pathway have revealed the relevance of this pathway in a number of diseases (Zoncu, et al., 2011). mTOR is a serine threonine kinase that acts as the catalytic component of two complexes, mTORC1 and mTORC2. mTOR is a target of the fungal-derived drug rapamycin; binding of this substance to mTOR is dependent upon the protein FKBP12, which interacts with
70
INVESTIGATING THE HUMAN GENOME
Rapamycin and with the kinase domain of mTOR. The mTOR pathway is illustrated in an abbreviated form in Figure 5.1. Growth factors Nutrients absent
RhebGDP
PIP3
RhebGTP
Tuberin harmatin active
PTEN T
PIP2
H
Growth factors Nutrients present
*P
mTORinactive AKT
AKT P
RhebGTP
PIP2 PPP PIP3
mTOR
T H
Tuberin phosphorylated Complex inactive S6 kinase active Translation activated EIF4BP binding to EIF4B lifted Translation initiation begins
Figure 5.1
Increase in protein synthesis cell proliferation
TSC1, TSC2, and mTOR pathway
The tuberous sclerosis genes TSC1 and TSC2 control the activity of mTOR. Rheb GTP stimulates mTOR activity. TSC1 and TSC2 proteins act as Rheb GTPases; they undergo phosphorylation (for example, in response to kinase AKT or AMPK). Phosphorylation may be positive, to stimulate activity, or negative, depending on which serine residues (S) are phosphorylated. Duvel, et al. (2010), examined gene expression altered by mTORC1 induction. They determined that the specific metabolic pathways involved include the pentose phosphate pathway, glycolysis,
CHAPTER 5 • PATHWAYS, PHENOTYPES, AND PHENOCOPIES
71
fatty acid biosynthesis, and cholesterol biosynthesis. They determined that these metabolic pathways are more active in TSC1-TSC2 negative cells Activation of these pathways can be reversed by rapamycin treatment. Duvel, et al., noted that cells that lack TSC1-TSC2 proliferate in the absence of growth factors. In the absence of the TSC1-TSC2 complex, Rheb GTPase activity is deficient, leading to decreased conversion of RhebGTP to RhebGDP. This leads to increased quantities of RhebGTP and stimulation of the mTORC1 pathway. These investigators determined that signaling through mTORC1 activates enzymes throughout the glycolysis pathway, from glucose 6phosphate to the generation of pyruvate and lactate. Expression of enzymes involved in the pentose phosphate pathway is increased by mTORC1 expression, including enzymes involved in the metabolism of glucose-6-phosphate to ribulose-5-phosphate and aldolase. Analysis of TSC2 negative cells also revealed increased activity of enzymes involved in the synthesis of sterols, isoprenoids, and fatty acids. Duvel, et al., then sought to identify transcription factors downstream of mTORC1 that activate the expression of genes that encode enzymes involved in metabolism. They determined that the alpha subunit of transcription factor HIF1 was upregulated in TSC2-deficient cells. HIF1 consists of a constitutively expressed subunit HIF1B that complexes with the HIF1 alpha subunit to become active. Duvel, et al., also determined that increased mTORC1 activity that leads to increased S6K1 activity leads to increased production of the DNA-binding element SREBP (sterol regulatory element binding protein), and this stimulates lipid biosynthesis. They noted that full-length SREBP is an inactive transmembrane protein in the endoplasmic reticulum. The active SREBP is derived by proteolytic cleavage of the full-length element. SREBP DNA-binding element is also involved in the induction of G6PD expression and in oxidation in the pentose phosphate shunt.
72
INVESTIGATING THE HUMAN GENOME
mTOR and bioenergetic regulation in cancer Many tumors are characterized by active glycolysis and apoptosis resistance. Khatri, et al. (2010), demonstrated that mTORC1dependent glycolysis increases when levels of the FOX3A transcription factor decrease. Their studies further revealed that FOX3A knockdown leads to decreased expression of the TSC1 gene and increased expression of mTORC1.
Regulation of mTOR expression by stress Expression levels of mTOR change in response to signals from growth factors, nutrient levels, cellular energy levels, and stress conditions. Specific stressors that impact mTOR expression include osmotic shock, heat shock, and hypoxia. Inoki, et al. (2005), identified two stress-induced molecules that inhibit mTOR signaling. They also noted that, under low energy conditions, adenosine monophosphate kinase (AMPK) is activated, which phosphorylates TSC2 and leads to decreased mTOR expression. They determined that AKT, on the other hand, phosphorylates TSC2 at different sites and inactivates TSC2 Rheb GTPase activity, leading to increased mTOR expression. Inoki, et al., demonstrated that proteins encoded by the REDD1 and REDD2 genes (DNA damage inducible transcripts) inhibit mammalian mTOR activity and act upstream of TSC2 and RHEB. They further demonstrated that expression of these genes is upregulated by hypoxia and glucocorticoid treatment and by DNA damage, such as damage induced by alkylating agents such as methyl methane sulfonate. REDD1 protein production increases under conditions of hypoxia and in response to DNA damage. REDD1 proteins stimulate TSC1 and TSC2 activity, thereby inhibiting mTOR activity.
CHAPTER 5 • PATHWAYS, PHENOTYPES, AND PHENOCOPIES
73
It is important to note that activation of TSC1/TSC2 and inactivation of the mTOR pathway play critical roles in down-regulating cell proliferation in situations where DNA is damaged. Increased cellular production of GSK3B kinase leads to the phosphorylation of REDD1. This facilitates the binding of the ubiquitin ligase complex to the REDD1 protein and its degradation in the proteosome, lifting mTOR repression (Katiyar, et al., 2009).
Integrating environmental signals and growth The mTOR pathway plays a key role in integrating environmental cues with the regulation of protein synthesis and growth (Ma and Blenis, 2009). Low nutrient conditions inhibit mTOR signaling. The RAG proteins are small GTPase that act as amino acid-sensing molecules. Ma and Blenis noted that, under conditions of amino acid starvation, mTOR is randomly distributed in the cells. However, in the presence of amino acids, it is located in proximity to Rheb. RAG proteins respond to amino acid concentration and cellular uptake, and thus act independently of TSC1 and TSC2. Side effects of treatment with rapamycin and its analogs include metabolic effects hyperglycemia and hyperlipidemia, but these are treatable. Wacheck (2010) proposed that rapalogs might be useful in combination with other targeted cancer therapies.
Pathways in seven different syndromes in which renal cancer occurs Germline mutations in seven different genes predispose to renal cancer. Linehan, et al. (2010), reviewed these syndromes and noted that the genes included act in pathways involved in determining response to metabolic stress or nutrient stimulation. The seven genes
74
INVESTIGATING THE HUMAN GENOME
that may harbor mutations predisposing to renal cancer include Von Hippel Lindau gene (VHL), MET oncogene, folliculin (FLCN), tuberous sclerosis 1 and 2 (TSC1, TSC2), fumarate hydratase (FH), and succinate dehydratase (SDH).
Von Hippel Lindau gene In Von Hippel Lindau syndrome, germline mutations are present in the VHL gene. Somatic mutations occur, and evidence indicates that loss of heterozygosity in VHL is present in a high percentage of cases of primary renal carcinoma. The VHL gene product in combination with protein elongins B and C and cullin binds to hypoxia inducible factor (HIF), and HIF is then targeted for degradation in the ubiquitin proteosome system. The process occurs under conditions of normal oxygen concentration. The specific binding of HIF to the VHL complex requires hydroxylation of a proline residue in HIF by proline hydroxylase. This critical hydroxylation requires the presence of oxygen (O2), 2-oxoglutarate ascorbate, and iron (Fe). The critical reaction between the VHL complex and HIF does not take place under conditions of low oxygen concentration, and under these conditions, accumulation of HIF occurs. Linehan, et al., noted that genes upregulated by HIF include vascular endothelial growth factor (VEGF), platelet-derived growth factor (PDGF), and the mTOR pathway (mammalian target of rapamycin). HIF contains different subunits that bind to the promoter regions of a number of genes, including VEGF and PDGF. Linehan, et al., proposed that drugs that target HIF transcription may be valuable in treatment of renal cancer. The drug topotecan represses HIF alpha transcription.
CHAPTER 5 • PATHWAYS, PHENOTYPES, AND PHENOCOPIES
75
Turcotte, et al. (2008), reported studies on specific agents that induce toxicity in VHL negative cells (VHL-/-) independent of HIF levels. The compound STF 62247 leads to autophagy of VHL negative cells.
MET proto-oncogene and renal cancer Linehan, et al., reported that MET proto-oncogene activating mutations occur in hereditary papillary renal carcinoma. Activating MET mutations also occur in some cases of sporadic papillary carcinoma. These mutations activate the phospho-inositol-3-kinase (PI3K) signaling pathway and are associated with increased expression of nutrient transporters on the cell surface. The PI3K pathway also impacts the AKT and mTORC2 pathways. Linehan, et al., reported that early evidence indicates that Foretinib, an inhibitor of MET and VEGF receptors, is useful in treating renal papillary cancers.
Tuberous sclerosis and renal tumors Renal angiomyolipomas may occur in patients with tuberous sclerosis. Lack of TSC1 or TSC2 function leads to activation of the mTOR pathway. Linehan, et al., noted that HIF likely is also increased in these tumors due to an increase in mTOR. Sirolimus, a rapamycin-related compound, is used to treat these tumors. It forms complexes with a protein in the mTORC1 complex, thereby inhibiting function of this complex. Clear-cell renal carcinoma and other forms of renal cancer also occur in tuberous sclerosis.
76
INVESTIGATING THE HUMAN GENOME
Folliculin gene mutations and associated syndrome Germline mutations in the folliculin (FLCN) gene on chromosome 17p11.2 occur in Birt Hogge Dube (BHD) syndrome. This syndrome is associated with the development of skin tumors fibrofolliculomas, pulmonary cysts, and renal cancer, most frequently chromophobe tumors. Loss of heterozygosity in 17p11.2 is often observed in the tumors. Schmidt, et al. (2001), first mapped this syndrome to 17p11.2. Nickerson, et al. (2002), identified folliculin as the mutated gene. Overlap in symptoms occurs between tuberous sclerosis and Birt Hogge Dube syndrome; in both disorders, benign skin tumors, pulmonary cysts, renal cysts, and renal cancer occur. Linehan, et al. (2010), reported that, in a mouse model where folliculin is inactivated, polycystic kidneys and kidney tumors develop; in these tumors, mTORC1 and mTORC2 activity is altered. Rapamycin treatment reduced the size of the kidney tumors and lengthened survival time. Folliculin binds to folliculin interacting proteins FNIP1 and FNIP2, and also interacts with 5'AMP activated kinase (AMPK). Hasumi, et al. (2009), proposed that reduced activity of FLCN and the FLCN-FNIP complex leads to increased activity of AMPK, mTORC1, and mTORC2, and to increased downstream activity of S6 Kinase and HIF1alpha. Evidence also indicates that the TGFbeta signaling pathway plays an important role in the pathogenesis of renal cancer in BHD syndrome (Hong, et al., 2010).
Tricarboxylic acid metabolism and renal cancer syndromes Deficiency of the enzyme fumarate hydratase occurs in a syndrome associated with skin tumors (piloleiomyomas), uterine
CHAPTER 5 • PATHWAYS, PHENOTYPES, AND PHENOCOPIES
77
leiomyomas, and renal cancer. Linehan, et al. (2010), noted that the kidney tumors in these patients are histologically diverse. Patients with this syndrome have germline mutations in fumarate hydratase, leading to increased tissue levels of fumarate. Increased fumarate inhibits the hydroxylation of hypoxia inducible factor (HIF) by proline hydroxylase and its subsequent degradation. Increased levels of HIF lead to increased expression of a number of genes involved in cell proliferation and angiogenesis, and result in highly aggressive tumors. Linnehan, et al., noted that inhibitors that target VEGF and glucose transporter GLUT1 could potentially be effective in treatment of these tumors. Xie, et al. (2009), studied tumor growth in a mouse model of this fumarate hydratase-related disorder. They determined that inhibition of LDHA results in increased apoptosis of tumor cells.
Succinate dehydrogenase mutations endocrine tumors and renal tumors Succinate dehydrogenase germline mutations occur in familial paraganglioma and pheochromocytomas. Linehan, et al., noted that increased succinate also inhibits proline hydroxylation of hypoxia inducible factor (HIF), leading to decreased degradation of HIF. They proposed that targeting glucose transport or vascular promoting genes such as vascular endothelial growth factor (VEGF) may constitute therapies for these tumors. Linehan, et al., proposed that renal cancer is a metabolic disease because genes involved in energy, oxygen, and iron sensing are involved.
78
INVESTIGATING THE HUMAN GENOME
Congenital malformations and ribosome biogenesis pathway Ribosomes consist of 80 core proteins, plus approximately 70 small nucleolar RNAs and more than 150 associated proteins (Narla and Ebert, 2010). Growing evidence suggests that a number of different syndromes characterized by congenital malformations that involve bone, cartilage, hair, and sometimes bone marrow are due to mutations in genes in the ribosome biogenesis pathway. Syndromes due to ribosome biogenesis defects include Blackfan Diamond anemia and Treacher Collins syndrome. Characteristic craniofacial abnormalities occur in these syndromes and include micrognathia, deformation of the external ears, and absence of the lower eyelashes. In Treacher Collins syndrome, malar and maxillary hypoplasia are striking features. Evidence now indicates that Treacher Collins syndrome is genetically heterogeneous. The first identified gene defect in this disorder involved the gene TCOF1, which encodes a protein referred to as treacle. It is a glycoprotein that encodes a gene responsible for the transcription of ribosomal RNA. It is involved in the transcription of ribosomal RNA through its interaction with an upstream binding factor (Dauwerse, et al., 2010). These investigators reported that, in at least 10% of patients with Treacher Collins syndrome, no TCOF1 mutations were found. Dauwerse, et al., carried out genome-wide copy number analysis in a Treacher Collins patient who had no TCOF1 mutation. They discovered that this patient had a 156-kilobase deletion on chromosome 13q12.2. This resulted in deletion of the POLR1D gene that encodes a ribosomal polymerase subunit. They then carried out studies in other Treacher Collins patients who were negative for TCOF1 mutation and discovered ten patients with mutations in the POLR1D gene. This gene encodes a subunit that is present in RNA polymerase
CHAPTER 5 • PATHWAYS, PHENOTYPES, AND PHENOCOPIES
79
I and RNA polymerase III; these polymerases are responsible for ribosomal gene transcription and the production of ribosomal RNA.
Epistasis Epistasis is defined as the influence of alleles in one gene on the phenotype or disease risk determined by a gene at a different locus. Specific SNPs at different loci may, in combination, exert a larger effect on disease risk. Cantor, et al. (2010) reported an example of epistatic interaction. Specific alleles in the interleukin4 receptor gene interacted with specific alleles in the interleukin13 receptor gene to increase the risk of asthma. They drew attention to extensive computer resources required to detect epistasis in large genome-wide association studies.
Proteostasis network This network sustains functional proteins through processes that survey protein folding and remove unfolded or unstable proteins that are present in the endoplasmic reticulum. Chaperone proteins such as HSP70 and HSP90 in the cytosol and in the endoplasmic reticulum play key roles in protein folding and protein loading onto membranes. Chaperone activity is dependent upon ATP and the presence of coactivators, including DNAJ and HSP40 proteins. Evidence indicates that folding and unfolding are linked to the energy status of the cell. Degradation of unfolded proteins requires activity of ubiquitin ligase, ubiquitin conjugating enzymes, and guidance of ubiquitinated proteins to the proteosome for degradation. Hutt, et al. (2009), emphasized that understanding the role of the proteostasis network may pave the way to better treatment of disease.
This page intentionally left blank
6 Dynamic function, synaptic activity, and plasticity The goal of this chapter is to present a few examples of insights gained into neuronal and brain functions through molecular genetic studies, facilitated by the availability of data on gene sequence. Through functional genomic studies on synapses, we have gained insight into processes that, when disrupted, lead to cognitive impairment and behavioral abnormalities.
Synapses, dendrites, and cognitive impairment Histological studies on the brain have revealed that abnormal dendritic spine number and morphology are among the most consistent findings associated with cognitive impairment. A growing body of evidence indicates synaptic plasticity and changes in signal strength following activity, such as long-term potentiation and longterm depression of activity at synapses, are important in learning and memory (Cooke and Bliss, 2006). Activation of synapses leads to modulation of gene expression, synthesis of proteins, and modulation of synapses and their downstream connections. In genetic disorders characterized by cognitive impairment, researchers have discovered that structural changes or mutations occur in a number of genes that play roles in neuronal synaptic function. 81
82
INVESTIGATING THE HUMAN GENOME
Dendritic spines and synaptic activity Dendritic spines represent the post-synaptic compartment for excitatory glutamatergic neurotransmitters receptors that are aligned on the dendritic surface. These receptors include AMPAR (alpha amino-3hydroxy-5methyl isoxazole propionic acid) and NMDAR (N-methyl aspartic acid) receptors and metabotropic glutamate receptors mGluR. These receptors and the neuroligin proteins project from the spine surface toward the pre-synapse. Neurotransmitters on the spine are closely associated with the post-synaptic density (PSD). Figure 6.1 is a schematic illustration of the synapse. The PSD consists of several hundred proteins. Core scaffold proteins of the PSD include PSD95, SHANK3, HOMER, and GKAP (guanylate kinase associated protein) (Kim and Sheng, 2009). Long-term potentiation leads to the formation of new spines and to the enlargement of spines; long-term depression is associated with shrinkage and retraction of spines. Tada and Sheng (2006) noted that synaptic activity also influences the organelle content of spines. Upon stimulation, mitochondria migrate into spines. Actin filaments and propelin proteins impact actin filament assembly and are involved in morphological changes in dendrites. ARC protein (activity-regulated cytoplasmic associated protein) also plays an important role in determining spine morphology (Peebles, et al., 2010).
Neuroligins and synapses Neuroligins 1 and 3 (NL1 and NL3) predominate in excitatory synapses, whereas NL2 predominates in inhibitory synapses. Hussain and Sheng (2005) noted that trans-synaptic interactions occur between beta neurexin and NL1 and NL3 at excitatory synapses. At inhibitory synapses, beta neurexin aligns with NL2. Neuroligin and neurexin mutations and/or structural abnormalities have been
CHAPTER 6 • DYNAMIC FUNCTION, SYNAPTIC ACTIVITY, AND PLASTICITY
83
Synaptic Vesicles and Neurotransmitters
Mitochondrion Axon
Neurexins Neurotransmitters and Receptors
Ion Channels Neuroligins Post-Synaptic Density Dendritic Spine Actin Fibers
Mitochondrion
Figure 6.1 Schematic of the synapse and the dendritic spine, illustrating a subset of functional elements and associations with the post-synaptic density.
described in patients with autism, and neurexin deletions mutations occur in some patients with schizophrenia or autism (Sudhof, 2008).
Synaptic function and gene expression Wang, et al. (2010), reviewed the regulation of gene expression in neurons in response to stimulation. Messenger RNAs are transported into dendrites bound to ribonucleoproteins. During transport, translation of these mRNAs is repressed through a number of different
84
INVESTIGATING THE HUMAN GENOME
factors, including 4E-binding protein (4EBP), Fragile X mental retardation protein (FMRP), cytoplasmic polyadenylation element binding protein (CPEB), and cytoplasmic FMRP interacting proteins (CYFIP1 and CYFIP2). Growing evidence indicates that mRNA accumulates locally in dendrites and is translated in response to stimuli. Wang, et al., noted that, in recent studies of responses on mature neurons, attention has focused on the translation of transcripts present in dendrites.
Function of FMRP (Fragile X mental retardation protein) at synapses Cognitive impairment in Fragile X mental retardation is due to the absence of FMRP in neurons. FMRP normally occurs within large ribonucleoprotein particles that also contain CYFIP1, CYP1P2, and Fragile X-related proteins (FXR1P, FXR2P), as well as nuclear fragile X interacting protein (NUFIP1). The absence of FMRP is most commonly the result of impaired transcription due to expanded CCG repeats in the 5' promoter region of the gene and methylation of the expanded repeat. Feng, et al. (1997), demonstrated that FMRP undergoes nucleo-cytoplasmic shuttling and is associated with dendritic ribosomes. FMRP is found in cells throughout the body; highest expression occurs in neurons. It occurs in the post-synaptic region and in dendritic spines. FMRP is an mRNA-binding protein that binds to between 400 and 600 different mRNAs and with its own mRNA. Bassell and Waren (2008), noted that other important FMRP mRNA interactions involve MAP1B signal transduction mRNA and amyloid precursor protein mRNA. FMRP occurs in large polyribosomes and in small RNA granules. Deficiency of FMRP leads to increased numbers of dendritic spines and to morphological changes in dendritic spines; spines are
CHAPTER 6 • DYNAMIC FUNCTION, SYNAPTIC ACTIVITY, AND PLASTICITY
85
abnormally long and thin. A number of investigators have proposed that FMRP regulates synapse formation and that it may play a role in post-synaptic pruning in specific brain regions (Pfeiffer and Huber, 2009). FMRP suppresses the translation of a number of proteins through its interaction with CYFIP1, which binds the translation initiation factor EIF4E. Evidence indicates that FMRP may also suppress mRNA translation through its association with microRNAs and through its impact on the microRNA machinery, including Argonaut and RISC. In their review of FMRP, mRNA regulation, and synaptic function, Bassell, et al. (2008), noted that FMRP impacts approximately 4% of proteins in brain. FMRP has nuclear localization and nuclear export domains, and two hnRNP K binding domains that determine interaction with mRNA. Evidence suggests that FMRP regulates protein synthesis in dendritic spines and plays a role in synaptic plasticity. Specific mRNAs that are translated at synapses and are bound to FMRP include the alpha subunit of calcium/calmodulin dependent protein kinase II (alpha CAMKII), activity regulated cytoskeletal associated protein ARC, and PSD95.
Synaptic signals that trigger translation initiation Wang, et al. (2010), noted that metabotropic glutamate receptors are key regulators of activity-dependent translation at dendrites and are enriched at excitatory synapses. They noted further that signaling through mGluR1 and mGluR5 plays a key role in establishing synaptic circuitry during development. Signaling through mGluR1 and mGluR5 stimulates rapid relocation and translation of dendritic mRNAs. They drew attention to the fact that factors that promote release of transcription repressors, such as RNA granule protein 105,
86
INVESTIGATING THE HUMAN GENOME
(RNG105) induce the translation of mRNAs. Wildtype FMRP represses the translation of a wide variety of mRNAs; in the absence of FMRP, protein synthesis in the hippocampus increases by 20 percent.
MGluR theory of Fragile X syndrome Bear, et al. (2004), postulated that FMRP acts as a negative regulator of translation downstream of mGluR. The negative regulation by FMRP is normally lifted only upon mGluR activation. Evidence indicates that mGluR stimulation leads to downstream synthesis of a number of proteins, including ARC, and that ARC is necessary for AMPAR neurotransmitter internalization. In Fragile X syndrome (FXS), where FMRP is deficient, mGluR-dependent induction of ARC synthesis does not occur.
FMRP phosphorylation FMRP-mediated repression and depression of protein translation is controlled by phosphorylation and dephosphorylation. Rapid dephosphorylation of FMRP coincides with the translation of mRNAs. The enzyme S6 kinase plays an important role in phosphorylation of FMRP and leads to the repression of protein synthesis. Activation of mTOR leads to increased activity of S6 kinase. Excessive mTOR activation occurs in tuberous sclerosis due to a deficiency of TSC1 and TSC2. Bassell, et al., predict that FMRP is hypophosphorylated in tuberous sclerosis, leading to the translation of FMRP target mRNAs. They emphasize that two autism spectrum disorders likely impact the same post-synaptic space.
GABA neurotransmitter function in Fragile X syndrome GABAergic synaptic transmission is reduced in FXS. GABA receptors are central to inhibitory synaptic transmission. Bassell, et al.,
CHAPTER 6 • DYNAMIC FUNCTION, SYNAPTIC ACTIVITY, AND PLASTICITY
87
noted that FMRP is involved in regulation at the interface between excitatory and inhibitory signaling.
PSD95 domain and SHANK proteins Within the PSD95 domain, SHANK proteins interact with the carboxy terminal regions of mGluR and guanylate kinase associated protein (GKAP). Through GKAP, SHANK protein interacts with NMDA and AMPA receptors. Evidence indicates that dendritic spine abnormalities occur in mice when Shank1 protein is knocked out. In children with deletions of chromosome 22q13 and SHANK3, neurobehavioral abnormalities and autism occur (Bozdagi, et al., 2010).
mTOR A gene on chromosome 1p36.2 encodes the mTOR protein, a ubiquitously expressed kinase that is essential in regulating translation. Hoeffer and Klann (2009) reviewed the structure and function of mTOR, which acts downstream of receptors and constitutes a node of convergence of different signaling pathways. Signaling pathways that converge on mTOR include phosphoinositide-dependent kinase 1, phosphoinositide 3-kinase, AKT serine threonine kinase, and the TSC2 and TSC2-encoded proteins hamartin and tuberin.
Synaptic activity and mTOR Hoeffer and Klann (2009) reported evidence that mTOR couples receptor activity to translation machinery in the cell and brings about long-lasting changes in synapses that constitute the basis of higherorder brain function and memory. They noted that memory is dependent upon the regulated interaction of neuronal networks. This interaction is dependent upon electrochemical communication
88
INVESTIGATING THE HUMAN GENOME
between neurons that takes place at synapses. The strength of the synaptic connection is plastic or modifiable. Long-term changes in synaptic activity include long-term potentiation and depression. Long-term changes in synaptic activity are associated with changes in protein synthesis. mTOR serves as the focal point for the assembly of signaling complexes.
Structure of mTOR Specific domains within mTOR determine interactions with specific proteins. Six key domains occur within mTOR. The names of these domains are acronyms derived from the specific elements within the domain. Two HEAT domains are present within mTOR. HEAT domains occur within many different proteins, including the protein that is defective in Huntington’s disease, huntingtin. HEAT domains contribute a scaffolding function that facilitates the assembly of proteins. HEAT is an acronym for Huntington Elongation Factor, A subunit of PP2A (protein phosphatase 2 activator) and TOR. FAT domains are present in the middle of mTOR. FAT is an acronym for FKBP12 rapamycin associated protein (immunophilin), ATM ataxia telangiectasia protein, TRAP transformation transactivation domain associated protein. FAT domains occur in many proteins that belong to the phosphatidyl inositol kinase (PIK) family. The FRB domain is adjacent to the internal FAT domain. This protein domain has FKBP12 binding properties and also interacts with RHEB. The KIN domain determines the serine threonine kinase activity of mTOR. The NRD region within KIN domain contains the regulatory serine threonine residues that may be phosphorylated. Within TORC1, mTOR forms a complex with the protein Raptor, regulatory associated protein of mTOR, and with other proteins. In mTORC2, mTOR forms complexes with Rictor, rapamycin-insensitive companion of mTOR, and other proteins.
CHAPTER 6 • DYNAMIC FUNCTION, SYNAPTIC ACTIVITY, AND PLASTICITY
89
mTOR function Increased mTOR activity leads to increased translation through its effect on S6Kinase phosphorylation and phosphorylation of 4EBP. MTORC1 plays a key role in translation through activation of ribosomal S6 kinase. MTORC1 phosphorylation of the EIF4E binding protein (4EBP) leads to dissociation of this protein from the translation initiation protein EIF4E, and this facilitates the unwinding of the 5' untranslated region of mRNA. Hoeffer and Klann (2009) emphasized that the major activity of the mTORC1 complex is to facilitate translation. The activity of mTORC2 is less well characterized. Hoeffer and Klann (2009) noted evidence that mTOR signaling is involved not only in translation, but also in other processes, including transcription, protein degradation and autophagy, and cytoskeletal assembly. Through studies in which rapamycin was used to inhibit mTOR activity, investigators have demonstrated that mTOR activity plays a role in protein synthesis function within synapses. They include NMDA neurotransmitter-dependent long-term potentiation of synaptic activity. The upstream TOR activation pathway is known to include phosphatidyl inositol kinase PI3K, phosphoinositide dependent kinase (PDK1), and AKT protein kinase. Hoeffer and Klann noted that a number of syndromes associated with cognitive impairment and autism involve the mTOR signaling pathway. In neurofibromatosis, RASGTPase activity is impaired, leading to increased RAS activity, PI3K activity, and mTOR activity. In tuberous sclerosis, RHEB GTPase activity is decreased, leading to increased RHEB GTP and increased mTORC1 function. Disruption of the PTEN1 gene function occurs in some patients with autism. A gene on chromosome 10 encodes PTEN1, and PTEN1 protein normally plays a role in inhibiting the PI3K activation of mTOR. These investigators noted that S6 kinase generated by mTOR activity phosphorylated the protein encoded by the Fragile X mental retardation
90
INVESTIGATING THE HUMAN GENOME
locus (FMR) and that this phosphorylation regulates mRNA binding of FMR protein. Hoeffer and Klann emphasized that it will be important to identify specific proteins that are translated in response to mTOR activation of protein synthesis in synapses.
mTORC2 Studies by Huang and Manning (2009) revealed that the TSC1 TSC2 complex has stimulatory effect on the function of mTORC2, and this contrasts with the TSC1, TSC2 inhibitory effect on mTORC1. Their studies on TSC1 TSC2 deficient cells revealed that rapamycin treatment does not restore mTORC2 activity.
mTORC, rapamycin, and rapamycin analogs Evidence now indicates that mTORC1 consists of complexes of mTOR with a number of proteins, including Raptor, PRAS40, and LST8.mTORC1 phosphorylates S6K1 and 4EBP1. This leads to transcription and translation of the pro-angiogenic factor HIF1a and cell cycle progression factors. Rapamycin and its analogs rapalogs bind to the protein immunophilin, encoded by FKBP12, and this complex binds to a specific domain in mTOR that is adjacent to the mTORC1 catalytic site. Wacheck (2010) reported that rapalogs impact the phosphorylation of S6 kinase, which impacts the actin cytoskeleton. Wacheck noted that rapamycin does not inhibit mTORC2. Evidence suggests that more recently developed mTOR kinase inhibitors are more potent inhibitors of 4EBP1 phosphorylation. In addition, these inhibitors act on mTORC1 and mTORC2. Through studies that used rapamycin to inhibit mTOR activity, investigators determined that mTOR function plays a role in NMDAdependent long-term potentiation of synaptic activity.
CHAPTER 6 • DYNAMIC FUNCTION, SYNAPTIC ACTIVITY, AND PLASTICITY
91
Synaptic plasticity Synaptic plasticity underlies the processing and storage of information in neural networks. It is dependent upon new protein synthesis and growth and remodeling of excitatory synapses. Brain-derived neurotrophic factor (BDNF) plays a key role in determining synaptic plasticity. Waterhouse and Xu (2009) reviewed mechanisms through which BDNF mediates chemical and structural modifications of synapses. Early studies demonstrated the role of BDNF in neuronal proliferation and differentiation. Neurotrophins including BDNF; nerve growth factor (NGF); and neurotrophins 3, 4, and 5 are secreted proteins that activate the tyrosine kinase receptor (TRK). Evidence also indicates that pro-neurotrophins react with the sortilin receptor complex. BDNF may have rapid effects on synaptic activity through the promotion of phosphorylation of synaptic proteins. BDNF also has a more lasting effect on synaptic activity through its impact on protein synthesis. The NAD deacetylase SIRT1 plays key roles in a number of functions, including DNA repair and genomic stability. Gao, et al. (2010), demonstrated that Sirt1-deficient mice showed decreased performance in tasks that require hippocampal and cortical activity. In these mice, synaptophysin activity is decreased at presynaptic terminals and in hippocampal radiations. Synaptophysin is an integral membrane protein of synaptic vesicles. Levels of brain-derived neurotrophic factor and CREB protein (cyclic AMP response element binding protein) are also decreased. Gao, et al., determined that SIRT1 deficiency leads to decreased CREB protein levels but not to decreased CREB mRNA levels. This finding implies that a post-transcriptional mechanism is involved. They were able to demonstrate that Sirt1 depletion leads to upregulation of a specific micro RNA mIR134 and that CREB mRNA is an miR134 target. This microRNA also plays a role in dendritic spine formation. Gao, et al., demonstrated that
92
INVESTIGATING THE HUMAN GENOME
increased expression of Sirt1 in mice promotes synaptic activity and memory.
Cerebral cortex development: Impact of WDR62 mutations Next-generation sequencing facilitates the discovery of genes responsible for familial developmental defects that lead to neurocognitive deficits. Bilguvar, et al. (2010), carried out exome sequencing studies on two individuals with microcephaly who were members of consanguineous kindred. The two affected individuals share homozygous genomic segments, including one on chromosome 19. In the two affected individuals, Bilguvar, et al., identified a homozygous frame shift mutation that leads to a stop codon in the WDR62 gene on chromosome 19q13.12. The parents were heterozygous for this mutation. The index cases in this study had microcephaly and pachygyria. WD repeat protein 62 (encoded by WDR62) is a spindle pole protein. Bilguvar, et al., then carried out analyses in 30 probands with pachygyria or agyria who were products of consanguineous unions. They identified six different WDR62 mutations. Index cases that had WDR62 mutations presented with mental retardation and were found to have microcephaly. MRI studies revealed a range of different cortical malformations, including pachygyria, hypoplasia of the corpus callosum, and lissencephaly. Two patients had polymicrogyria, and one had schizencephaly. Studies in mice revealed that WDR62 expression is prominent in neural crest lineages and in ventricular and subventricular zones during cerebral cortical neurogenesis. WDR62 is also expressed in precursors of the cerebellar granular cells. Bilguvar, et al., emphasized that their study indicates that WDR62 plays a role in a spectrum of cortical abnormalities previously thought to represent distinct entities. It is interesting to note
CHAPTER 6 • DYNAMIC FUNCTION, SYNAPTIC ACTIVITY, AND PLASTICITY
93
that evidence now indicates that defects in three different spindle pole proteins, ASPM, STIL, and WDR62, each independently lead to microcephaly (Nicholas, et al., 2010).
Insights into consciousness EEG recordings reflect voltage changes in the brain. Synchrony or coherence of wave patterns may occur within specific regions or may be more widely distributed. Christie and Westbrook (2006) described a specific pattern of synchronous EEG waves designated as gamma synchrony. They observed that these waves were elicited in olfactory bulb dendrites in response to small and odorant molecule induction. Hameroff (2010) reported that electroencephalographic (EEG) studies reveal that conscious neuronal activity is distinguished by specific gamma synchrony wave patterns of 30–90 Herz. He noted that this specific EEG pattern is due to transient syncytial formation and the linking of dendritic gap junctions. He applied the term “conscious pilot” as a metaphor for the mobile gamma synchronized dendritic web that characterizes conscious activity. Hameroff noted that, in addition to the well-known neuronal activity whereby neuronal dendrites receive chemical input from axons of other neurons, there is also lateral coupling of dendrites through gap junctions that constitute electrical synapses. He proposed that dendritic gamma wave synchrony moving through the brain constitutes a vehicle for consciousness. He noted that, under general anesthesia, consciousness and gamma wave synchrony are lost. Anesthesia impacts cytoplasmic and membrane proteins, including connexin 36, which is present at gap junctions. Hameroff noted that the highest levels of frequency and amplitude of coherent gamma waves across cortical regions occur in meditating Tibetan monks.
94
INVESTIGATING THE HUMAN GENOME
Epilepsy Epilepsy is associated with hyper-synchrony of neuronal firing. In reviewing the etiology of epilepsy, Steinlein and Bertrand (2010) noted that the prevalence of epilepsy is 1 in 100 or 1 in 200. They noted further that epilepsies predominantly result from environmental factors and that epilepsies due to genetic factors represent the minority. They estimated a very low prevalence rate for monogenic epilepsies. They noted further that no straightforward relationship exists between genotype and phenotype in monogenic epilepsies. One form of monogenic epilepsy is autosomal dominant nocturnal frontal lobe epilepsy (ADNFLE), caused by mutations in nicotinic acetylcholine receptor subunits. ADNFLE seizures start with vocalizations, followed by hyperkinetic activity. Sleepwalking may occur. Similar seizures are often present in family members and show autosomal dominant inheritance. Seizures usually start during adolescence but may start later. Occasionally, spontaneous remission of seizures occurs. Mutations in the CHRNA4 gene or the CHRNB2 gene, which encode alpha 4 and beta 2 subunits, respectively, of the nicotinic acetylcholine receptors, most frequently lead to ADNFLE. Mutations in CHRNA2 may also give rise to seizures with a similar clinical presentation. Mutations that lead to ADNFLE are gain-of-function mutations that lead to imbalance between excitation and inhibition. Molecular studies on cholinergic receptors revealed the complexity of ligand gated ion channels such as nicotinic acetylcholine receptor (nACHR). Following the binding of neurotransmitters to the receptor, ionic pores in the post-synaptic membrane open. Each receptor consists of five subunits, and the receptor spans the membrane. The alpha subunits have two adjacent cysteine residues; they occur in the extracellular domain and are key components of the receptor. Transmembrane segments also form the ionic pore. Steinlein and Bertrand noted that the amino acids facing the pore determine
CHAPTER 6 • DYNAMIC FUNCTION, SYNAPTIC ACTIVITY, AND PLASTICITY
95
the ionic selectivity of the pore. Seventeen different genes encode subunits that form muscle and neuronal acetylcholine receptors. Homomeric receptors occur (for example, consider the CHRNA7 receptor that consists only of alpha 7 subunits). Other ACHR receptors consist of a different number of alpha and beta receptors. Different subunit compositions confer different properties on receptors.
Antiepilepsy drugs Antiepilepsy drugs interact with ion channels and neurotransmitter receptor channels. Vigabatrin, for example, specifically impacts GABA receptors. Drugs that impact different mechanisms have also been found to be useful in epilepsy. The carbonic anhydrase inhibitor acetazolamide acts by altering neuronal pH and concentrations of potassium (K) and bicarbonate ions HCO3(Katayama, et al., 2002).
UBE3A and synaptic function: Insights gained from study of Angelman syndrome Angelman syndrome in humans is characterized by normal development during the first year of life and subsequent profound delay in cognitive development. It is due to a deficiency in UBE3A. This gene is expressed from a locus on the maternally derived copy of chromosome 15. The paternally derived locus is normally silenced through imprinting mechanisms. UBE3A protein is an E3 ubiquitin ligase. It contains a HECT domain that catalyzes the attachment of a polyubiquitin chain to target proteins. This attachment leads to degradation of the target protein in the proteosome system. Significant advances in the understanding of the pathogenesis of cognitive impairment and neurological deficits in Angelman syndrome have been made through studies on mouse models of this disorder. A number of different models have been developed in which
96
INVESTIGATING THE HUMAN GENOME
the maternal Ube3A gene is deleted Ube3Am-/p+. Yashiro, et al. (2009), used a Ube3Am-/p+ mouse model to study development of the visual cortex. They demonstrated that these mice have significant impairment of experience-driven synaptic development. Through studies in mouse models of Angelman syndrome, Greer, et al. (2010), determined that Ube3A plays a key role in regulating synapse development by facilitating the degradation of the Arc protein (activity regulated cytoplasmic associated protein). They established that, in the absence of Ube3A, there are elevated levels of Arc in neurons because Arc is not polyubiquitin tagged and is not degraded in the proteosome system. Increased levels of Arc lead to excess internalization of AMPA neurotransmitter glutamate receptors. A reduction in Arc levels leads to increased surface expression of Ampa receptors. These findings indicate that drugs that promote Ampa neurotransmitter receptor activity may reverse symptoms of Angelman syndrome. Greer, et al., reported that reduced or absent Ube3A impairs degradation of a number of substrates. These include Rho guanine nucleotide exchange molecules and ephexin5, which restrict the number of synapses that form. They note that additional proteins in neurons likely are not degraded adequately in the absence of Ube3A.
Synapses and autism Diagnostic criteria for autism include impaired social interaction, impairments in communication, and presence of repetitive stereotypic behaviors. Toro, et al. (2010), reviewed factors involved in the etiology of autism. They emphasized that many heterogeneous genes contribute to autism etiology and noted that high-penetrance and low-penetrance mutations are involved. They noted that many of the documented genetic defects in autism impact synaptic homeostasis. Evidence gathered over the past decade supports the conclusion that
CHAPTER 6 • DYNAMIC FUNCTION, SYNAPTIC ACTIVITY, AND PLASTICITY
97
“autisms” are phenotypically heterogeneous and involve different genes in common pathways. De novo copy number changes that involve genomic segments and impact gene dosage occur with higher frequency in autism (5%–10%) than in the general population (1%) (Sebat, et al., 2007). Toro, et al., noted that gene dosage changes and gene mutations in autism subjects frequently impact genes involved in synaptogenesis, formation of neuronal circuits, and synaptic function, they involve genes that regulate levels of proteins in synapses. These investigators concluded that a polygenic model for some cases of autism cannot be excluded. In a polygenic model mutations, each of low effect, in a number of different genes may lead to autism.
Synaptic activity and secondary modification of proteins Palmitoylation plays a role in protein targeting in presynaptic and post-synaptic membranes to facilitate signal transduction. Fukata and Fukata (2010) reviewed neuronal protein palmitoylation and provided several examples of this process. Palmitoylation facilitates clustering of the post-synaptic density protein (PSD95) that is required for AMPAR neurotransmitter receptor function. Other essential proteins at synaptic junctions that are modified by reversible palmitoylation include serotonin receptor and metabotropic glutamate receptor NMDAR (N-methyl D aspartate receptor) on the post-synaptic membrane. On the presynaptic membrane side K ion channels and the synaptic vesicle proteins synaptobrevin, synaptogamin, synaptosomal associated protein 25 (SNAP25) and DNAJC5 undergo palmitoylation. Axon outgrowth requires palmitoylation of NCAM (neural cell adhesion molecule), integrin, and fibroblast growth factor. Fukata and Fukata (2010) reported that, in recent years, specific enzymes involved in palmitoylation and depalmitoylation have been
98
INVESTIGATING THE HUMAN GENOME
identified. In humans and mice, at least 25 palmitoylation enzymes have been described. Enzymes in the DHHC domain zinc finger family play important roles in the palmitoylation of neuronal proteins. Different DHHC domain zinc finger enzymes differ with respect to the position in the substrate of the amino acid they modify. DHHC21 acts on substrates with a cysteine residue at the N terminus, DHHC17 palmitoylates internal cysteine residues. In current nomenclature, these genes have a Z symbol (as in ZDHHC17). Depalmitoylating enzymes include acylprotein thioesterase (APT1) and palmitoyl protein thioesterase (PPT1). A particularly interesting discovery involved a demonstration that a neuronal enriched microRNA (miRNA) targets and inhibits APT1. Activity of the miRNA leads to increased palmitoylation of RHO guanine nucleotide exchange factor, and this leads to shrinkage of dendritic spines.
Palmitoylation and neurological diseases Deletion of chromosome 22q11.2 is associated with schizophrenia. The single nucleotide polymorphic marker (SNP) within this region that shows the strongest association with schizophrenia is within the DHHC8 gene (Xu, et al., 2008). Mukai, et al. (2008), demonstrated that deficiency in DHHC8 that results from 22q11.2 microdeletion leads to deficits in dendritic spine formation and growth, as well as to reduced palmitoylation of the post-synaptic density protein PSD95. DHHC17 is encoded by a locus on 12q21.2; it palmitoylates neuronal proteins including huntingtin, SNAP 25 synaptic protein, and PSD95. Evidence indicates that palmitoylation of huntingtin is important in its function and trafficking. Yanai, et al. (2006), reported that decreased palmitoylation of huntingtin occurs as a consequence of the glutamine repeat expansion that is present in individuals with Huntington’s chorea.
CHAPTER 6 • DYNAMIC FUNCTION, SYNAPTIC ACTIVITY, AND PLASTICITY
99
Fukata and Fukata (2010) noted that deletions of DHHC15 and DHHC9 impact outcome in Fragile X mental retardation. DHHC15 maps to Xq13.3, and DHHC9 maps to Xq26.1. They emphasized the important technical advances that facilitated the analysis of the role of palmitoylation in neuronal protein function and trafficking. One advance was the development of palmitic acid analogues that may be used to metabolically label proteins in growing cells. A further important development is the use of the acyl biotinyl exchange method that facilitates the purification of palmitoylated proteins from complex solutions and tissue samples.
Neuregulin 1-ERBB4 in post-synaptic densities: Relevance in schizophrenia Genome-wide association studies have revealed evidence of an association between polymorphisms in the Neuregulin 1 (NRG1) gene and schizophrenia in Icelandic and Scottish populations (Stefansson, et al., 2004). A specific NRG1 SNP haplotype was associated with the disease in these pedigrees. Li, et al. (2004), identified a specific SNP haplotype within the NRG1 gene boundaries that was associated with schizophrenia in the Han Chinese population. The alleles in this haplotype were different than those associated with schizophrenia in the Icelandic and Scottish populations. Papiol, et al. (2011), carried out comprehensive phenotyping on 1,071 patients diagnosed with schizophrenia or schizoaffective disorder in the Gottingen study. They identified specific NRG1 haplotypes that were under-represented in the schizophrenia population, compared to the control population. They considered the under-represented haplotypes to be protective. Nicodemus, et al. (2010), carried out studies to examine epistasis and interactions between polymorphisms in genes that encode NRG1 and its binding partners, including ERBB4 and AKT1, and pathogenesis of schizophrenia. AKT1 is a serine threonine kinase critical for
100
INVESTIGATING THE HUMAN GENOME
the development of the nervous system. In schizophrenia case/control studies, they observed interactions between NRG1 polymorphisms and polymorphisms in its receptor ERBB4 (significance p = .035). They also observed significant three-way interactions among SNPs in NRG1, ERBB4, and AKT. Nicodemus, et al., concluded that genes in the NRG1, ERBB4, AKT pathway are important in the pathogenesis of schizophrenia. NRG1 on chromosome 8p12 encodes a glycoprotein that interacts with ERBB4 receptor tyrosine kinase. The NRG1 locus gives rise to many different isoforms derived through the use of alternate promoters and alternate splice sites. Evidence also supports tissue specificity of isoforms. In studies on the developing mouse brain, Fazzari, et al. (2010), demonstrated that mouse Erbb4 localizes in the post-synaptic densities of Gaba-ergic interneurons. They reported that the neuregulin 1Erbb4 signaling system is required to rewire Gaba-mediated circuits in the post-natal brain. Ting, et al. (2011), reported that ERBB4 protein stabilizes the post-synaptic density protein PSD95 in Gaba-ergic interneurons. They noted that NRG1 stimulates the formation of new synapses. It also strengthens existing synapses and acts in an ERBB4-dependent manner. Evidence for the role of the NRG1 ERBB4 signaling pathway provides further support for the hypothesis that schizophrenia is a neurodevelopmental disorder.
7 Late-onset neurodegenerative diseases
Alzheimer’s disease Alzheimer’s disease affects more than 35 million people worldwide (Grosgen, et al., 2010). Abundant evidence supports the hypothesis that amyloid beta plays a key role in this disease. This evidence derives from studies of familial Alzheimer’s disease in which mutations in specific genes, the presenilins, PSEN1 and PSEN2, lead to increased generation of the Abeta peptide from amyloid precursor protein (APP). However, mutations in the presenilin genes have not been found in late-onset Alzheimer’s disease. It is not clear whether abnormal amyloid processing and deposition constitute primary defects in lateonset Alzheimer’s disease or whether they occur secondarily in response to other insults. APP is a transmembrane glycoprotein that is sequentially cleaved by secretases (Wilquet and De Stooper, 2004). When APP is first cleaved by alpha secretase, it yields soluble alpha amyloid and a membrane-bound carboxy-terminal fragment that is subsequently cleaved by gamma secretase to give rise to P3 peptide. When APP is first cleaved by beta secretase, it gives rise to the amyloid beta domain. Subsequent cleavage of this domain by gamma secretase
101
102
INVESTIGATING THE HUMAN GENOME
leads to generation of amyloid beta peptide, the amyloidogenic peptide. The gamma secretase cleavage site is heterogeneous and gives rise to Abeta peptides of different sizes, most frequently Abeta40, with 40 amino acids, and Abeta42, with 42 amino acids. The Abeta42 form is more insoluble and prone to aggregation. Presenilins are components of the gamma secretase protein complex. Mutations in PSEN1 or PSEN2 impair the normal function of their gene products, namely cleavage of APP by gamma secretase. Mutations in the presenilin genes occur in rare forms of familial Alzheimer’s disease. The major risk factor for late-onset Alzheimer’s disease is the E4 allele of apolipoprotein E (APOE).
APOE polymorphism APOE3 constitutes the most common allele at the APOE locus. It occurs in 77% of the population in the United States; the APOE2 allele occurs in 8%. The APOE4 allele occurs in 15% of the control population in the United States. However, APOE4 is present in 40% of patients with Alzheimer’s disease (Bu, 2009). APOE4 occurs in 50% of individuals with late-onset Alzheimer’s disease. APOE4 is also a risk factor for atherosclerosis and cerebral amyloid angiopathy. The APOE2 allele is apparently protective against Alzheimer’s disease. APOE2, APOE3, and APOE4 differ at amino acids 112 and 158. APOE2 has cysteine at position 112 and 158. APOE3 has cysteine at position 112 and arginine at position 158. APOE4 has arginine at 112 and 158. These amino acid substitutions impact the three-dimensional structure of the molecules and their interactions (Mahley, et al., 2006). Evidence indicates that APOE4 is associated with increased resting calcium in neurons and with increased apoptosis.
CHAPTER 7 • LATE-ONSET NEURODEGENERATIVE DISEASES
103
APOE and amyloid precursor protein Beta amyloid (Abeta) accumulation may be due to increased synthesis or to reduced clearance. Bu (2009) reviewed amyloid clearance pathways that include receptor-mediated clearance and endopeptidase cleavage. Amyloid Abeta binds to the C terminal domain of APOE protein, the same domain that binds lipids. Binding of excess Abeta to the domain impairs lipid binding. Bu noted that although binding to APOE ultimately facilitates Abeta degradation in the endosome lysosome system, it also can lead to an accumulation of Abeta in neuronal multi-vesicular bodies. Amyloid precursor protein is taken up into cells in clathrincoated pits. It is then transferred to endosomes, where it is cleaved sequentially by beta secretase (BACE) and gamma secretase to generate toxic Abeta peptides. These may be secreted from the cells or they may aggregate in endosomes or lysosomes. Amyloid precursor protein was originally defined as a transmembrane protein that undergoes glycosylation during its passage through the Golgi apparatus. Cleavage of APP was initially thought to take place primarily when APP was membrane bound. The receptor SORTL1 shuttles APP to the Golgi and the secretory system. A growing body of data indicates that APP and beta and gamma secretases are associated with lipid rafts (Vetrivel and Thinakaran, 2010). The function of APP has not been clearly delineated. However, evidence suggests that it plays an important role in early development and impacts neuronal migration.
104
INVESTIGATING THE HUMAN GENOME
Amyloid Abeta-independent roles of apolipoprotein E4 in Alzheimer’s disease Transgenic mouse lines expressing human APOE3 and lines expressing human APOE4 have been developed. Huang (2010) reported that the transgenic mice expressing human APOE4 developed learning and memory deficits. These deficits did not occur in the lines that expressed human APOE3. Huang reported that, in the transgenic mice with human APOE4, dendritic spine density was lower than that in transgenic mice with human APOE3. In cultured dorsal root ganglion neurons and Neuro2a cell cultures, differences also occurred. APOE3 and VLDL (very low density lipoprotein) stimulated neurite extension; APOE4+VLDL inhibited neurite extension. Huang (2010) reported that neuron-specific tau phosphorylation increases in APOE4 versus APOE3 transgenic mice; tau phosphorylation in astrocytes is not different in the two groups. Of particular interest is the fact that the drug rosiglitazone rescues the dendritic loss in APO4 transgenic mice. Rosiglitazone is an insulin sensitizer and a mitochondrial activator. Evidence indicates that brain insulin receptors are down-regulated and functionally impaired by amyloid Abeta oligomers (Zhao, et al., 2008). Chang, et al. (2005), reported that APOE4 fragments containing the LDL receptor-binding region interacted with components of the mitochondrial respiratory complexes II and IV. Evidence also suggests that mitochondrial transport within neurons is negatively impacted by APOE4. Huang (2010) concluded that APOE4 has amyloid Abeta-dependent and -independent effects that contribute to the pathogenesis of Alzheimer’s disease. He proposed that stressors and injurious agents impact the synthesis of amyloid Abeta and that APOE4 gives rise to fragments that impair mitochondrial function.
CHAPTER 7 • LATE-ONSET NEURODEGENERATIVE DISEASES
105
Apolipoprotein E and brain lipid metabolism APOE4 impacts cholesterol and lipid metabolism and receptors; the alterations in lipid metabolism impair the metabolism of amyloid precursor protein and lead to amyloid beta aggregation (Bu, 2009). These findings have implications for the design of treatments for Alzheimer’s disease. In a review of APOE function, Bu (2009) noted that these proteins function as transporters of cholesterol and other lipids. In the brain, APOE binds to and transports lipoprotein synthesized by astrocytes. This lipoprotein resembles but is not identical to peripheral high-density lipoprotein. Bu noted that the N terminal domain of APOE interacts with cell surface receptors, and the C-terminal amino acids form the lipid-binding domain. Apolipoprotein E binds to the low-density lipoprotein receptor LDLR that plays a key role in cholesterol homeostasis. Apolipoprotein E also binds to LRP1 receptors. These receptors transport a number of different ligands from the cell surface to the interior of the cell. They are highly expressed in liver and brain. Astrocytes produce and secrete APOE, which is then loaded with lipids, including cholesterol, cholesterol esters, and phospholipids. Activity of the ATP binding cassette transporter protein ABCA1 plays a key role in the lipid loading of APOE. Specific transcription factors known as liver X receptors (LXR) increase APOE and ABCA1 gene expression. Lipid-loaded APOE is then transported in the cerebrospinal fluid to neurons. Lipids and cholesterol are important in the formation of synapses and their repair.
Cholesterol biosynthesis in brain In the central nervous system, cholesterol is a precursor of myelin and steroid hormones. Brain cholesterol is synthesized locally, and acetyl CoA is the precursor. The enzyme 3-hydroxy-3-methyl-glutaryl
106
INVESTIGATING THE HUMAN GENOME
coenzyme A reductase (3HMGCoA reductase) catalyzes the conversion of 3HMG CoA to mevalonic acid. Different brain regions differ in their content of cholesterol. Lipid rafts are cholesterol- and sphingolipid-rich microdomains that constitute platforms for the assembly of proteins. They regulate membrane trafficking and cell signaling, and are present in neuronal and glial cells (Korade and Kenworthy, 2008). Proteolysis of many transmembrane proteins, including amyloid precursor protein, take place in lipid rafts, and ionotropic and neurotransmitter transporters are located in them. Techniques such as FRAP (fluorescence recovery after photobleaching) and FRET (fluorescence resonance energy transfer) have facilitated analysis of lipid rafts. Evidence indicates that activities of enzymes involved in cholesterol biosynthesis are regulated by neurotrophic factors such as brain derived neurotrophic factor (BDNF). The effects of BDNF are mediated through specific receptors, including TRKB (neurotrophic tyrosine receptor kinase, also abbreviated as NTRK2) and p75 neurotrophin receptor. Nerve growth factor (NGF) may also play a role in regulating brain cholesterol synthesis. Korade and Kenworthy emphasized that growth factor stimulation of their receptors regulates membrane raft structure; in turn, membrane rafts play an essential role in signaling through receptors because receptor tyrosine kinases are located in rafts. Defective CNS cholesterol biosynthesis plays a role in specific inborn errors of metabolism. One example is Smith Laemli Opitz syndrome, in which 7dehydroxycholesterol accumulates and can be incorporated into lipid rafts, where it alters function. Niemann Pick disease type C (NPC) is associated with the accumulation of unesterified cholesterol, and it shares pathological features of Alzheimer’s disease, including generation of hyperphosphorylated Tau and abnormal accumulation of amyloid Abeta42. Cholesterol biosynthesis is down-regulated in the brain in Huntington’s disease.
CHAPTER 7 • LATE-ONSET NEURODEGENERATIVE DISEASES
107
Alzheimer’s disease: Role of brain cholesterol Hudry, et al. (2010), carried out studies on brain cholesterol and its clearance in a mouse model of Alzheimer’s disease APP23. They noted that previous studies determined that cholesterol is synthesized in the brain. However, cholesterol is not degraded in the brain and requires modification to be exported out of the brain. This modification is dependent on the conversion of cholesterol to 24S hydroxycholesterol, and the enzyme encoded by CYP46A1 catalyzes this conversion. In studies on transgenic mice prone to Alzheimer’s disease, Hudry, et al., determined that overexpression of CYP46A1 decreases amyloid plaque. Evidence for the important role of lipids in Alzheimer’s disease derives from many different studies. Ehehalt, et al. (2003), reported that amyloidogenic processing of amyloid precursor protein is dependent on the presence of lipid rafts and that processing of APP to beta amyloid takes place on these rafts. Hudry, et al. (2010), defined lipid rafts as detergent-resistant membranes that are microdomains enriched in cholesterol and in beta and gamma secretases. A number of investigators, including Xiong, et al. (2008), have reported that, as the cholesterol content of brain increases, activity of gamma and beta secretase increases. Brain cholesterol is converted to 24S hydroxycholesterol, and this interacts with LXR receptors that are transcription factors that regulate the expression of ATP binding cassette transporter ABCA1 and the expression of APOE. Leoni, et al. (2010), reported that, in elderly patients with mild cognitive impairment, cerebrospinal fluid levels of tau, ptau, APOE, and 24 S hydroxycholesterol are increased. They postulated that, during neurodegeneration, cholesterol is released and converted to 24Shydroxycholesterol. This then induces increased quantities of APOE.
108
INVESTIGATING THE HUMAN GENOME
Grosgen, et al. (2010), reported that APP amyloid precursor protein is essential for lipid homeostasis and that lipids regulate beta and gamma secretase processing of APP. They concluded that pathological alterations in lipid homeostasis or overproduction of Abeta peptide promotes neurodegeneration.
Tau phosphorylation, Alzheimer’s disease, and biomarkers Growing evidence indicates that analysis of specific biomarkers in the cerebrospinal fluid (CSF) facilitates a diagnosis of Alzheimer’s disease even in the early stages. Fagan, et al. (2009), explored the relationship between brain amyloid and CSF marker amyloid Abeta 42, as well as the relationship of neurofibrillary tangles and phosphorylated tau (ptau 181). In vivo cortical amyloid was determined using positron emission tomography with the amyloid binding agent Pittsburgh compound B (pib). Their analyses revealed that CSF Abeta 42 levels drop early in the disease process and remain low, while levels of pib in the cortex increase as the disease progresses. Their findings also suggest that CSF Abeta 42 levels drop prior to an ability to detect cortical amyloid with a PET scan and pib. Fagan, et al., reported a positive linear correlation between CSF ptau181 and the amount of cortical amyloid in cognitively normal individuals. They also reported that, in some individuals, a substantial amyloid load was present prior to the development of detectable cognitive impairment. These investigators emphasized that biomarkers will be useful in detecting early stages of Alzheimer’s disease.
Neurodegeneration in Alzheimer’s disease A number of investigators have emphasized that neurodegeneration occurs early in Alzheimer’s disease and that monomeric and
CHAPTER 7 • LATE-ONSET NEURODEGENERATIVE DISEASES
109
oligomeric forms of amyloid Abeta induce neuronal damage prior to the development of fibrillar tangles. Zempel, et al. (2010), demonstrated that Abeta oligomers cause passage of endogenous tau into dendrites and abnormal tau phosphorylation. They demonstrated further that dendrites with increased levels of Tau showed loss of mitochondria and redistribution of other cytoskeletal elements, such as neurofilaments. Furthermore, they observed that mis-sorted tau is hyperphosphorylated. Zempel, et al., noted that Tau in a normal brain is phosphorylated, on average, in 2 sites; in brain tissue from Alzheimer’s disease patients, it is often phosphorylated at 8 sites and may sometimes be phosphorylated at 20 sites. They noted further that hyperphosphorylation of tau led to its dislocation from microtubules. Through Mitotracker staining, they determined that mitochondria were reduced in regions where tau was hyperphosphorylated. They emphasized that toxic effects of amyloid Abeta are upstream of tau changes, including tau phosphorylation and mis-sorting.
Genome-wide association studies (GWAS) and Alzheimer’s disease Lambert, et al. (2009), analyzed SNP markers in 2,032 individuals from France with Alzheimer’s disease and in 5,328 controls. In a follow-up study, markers that showed evidence of an association with Alzheimer’s disease in the initial study were then analyzed in 3,978 Alzheimer’s disease individuals from Belgium, Italy, and Spain. Lambert, et al., obtained evidence for an association of Alzheimer’s disease with the clusterin encoding gene CLU (significance7.5X10–9). In addition, their study revealed an association with CRI p=3.7X10–9. CRI is a gene on chromosome 1 that encodes complement component 3b/4b receptor1. They noted that both of these genes encode products that play roles in the clearance of beta amyloid.
110
INVESTIGATING THE HUMAN GENOME
GWAS studies carried out on 19,000 Alzheimer’s patients by Seshadri, et al. (2010), have led to an identification of interacting factors that impact the pathogenesis of late-onset Alzheimer’s disease (LOAD). This study was carried out in two phases. In the first phase, 529,205 autosomal single nucleotide polymorphisms (SNPs) were typed in 11,789 subjects. Results of this stage confirmed the previously documented association of Alzheimer’s disease and APOE; in addition, five SNPs close to APOE yielded high scores, rs 2075650 within an intron of the TOMM40 gene showed significant association (p = 1.8X10–157), and rs6859 within the 3' untranslated region of PVRL2 showed significant association (p = 6.9X10–41). Two additional associations were identified in this study. Alzheimer’s disease was associated with a SNP in an intron of the clusterin gene CLU (p = 1.4X10–9) and with a SNP in the PICALM gene (p = 1.9X10–8). PICALM encodes phosphatidyl inositol binding clathrin assembly protein. In the second stage, SNP typing was carried out on a sample composed of 2,023 Alzheimer’s disease cases and 2,340 age-matched and cognitively screened controls. Harold, et al. (2009), reported that meta-analyses revealed significant evidence of associated with the clusterin gene and within PICALM. CLU SNP rs11136000 was significantly associated (p = 8.5X10–10), and SNP rs3851179 within PICALM yielded a significance score of p = 1.3X10–9. These genes are of particular interest because of their function. The CLU and PICALM genes have a relatively modest effect on Alzheimer’s disease risk, compared to APOE4. It is likely that susceptibility genes in combination impact Alzheimer’s disease predisposition and age of onset.
Clusterin and Alzheimer’s disease Clusterin encoded by a gene on 8p21-p12 is expressed in all tissues. Reports indicate that clusterin expression increases in response
CHAPTER 7 • LATE-ONSET NEURODEGENERATIVE DISEASES
111
to brain injury. Calero, et al. (2000), reported that increased expression of clusterin occurs in regions of the brain affected by Alzheimer’s disease. Clusterin acts as a chaperone for amyloid Abeta, and reports indicate that clusterin impacts the conversion of amyloid Abeta to insoluble forms (Boggs, et al., 1996).
GWAS studies mitochondrial folate metabolism and 1 C metabolism Results of genome-wide association studies carried out by Naj, et al. (2010), revealed a highly significant association of Alzheimer’s disease with APOE and also an association with a SNP within the gene MTHFD1L that encodes methylene tetrahydrofolate (NADP dependent) like 1 protein. They carried out an initial study of two independent cohorts of late Alzheimer’s disease patients, with a total of 931 cases. In a follow-up study, they analyzed data from 1,338 cases and 2,003 controls. A SNP within the APO E locus rs2075650 was associated with a p value of 1.9X10–36. The SNP rs11754661 in MTHFD1L was associated with Alzheimer’s disease, significance p = 4.7X10–8. The initial and replication data combined revealed an association between Alzheimer’s disease and rs11754661 with a significance of p = 1.90X10–10. Naj. et al. emphasized that MTHFD1L is of particular interest because it is involved in homocysteine-related metabolism. Elevated levels of homocysteine have been reported in a number of studies on Alzheimer’s disease patients. It is interesting to note that a specific MTHFD1L polymorphism rs6922269 is associated with risk of coronary artery disease (Tibbetts and Appling, 2010). These investigators reported that MTHFD1L has a mitochondrial targeting sequence and that it plays a role in the one-carbon metabolism in mitochondria.
112
INVESTIGATING THE HUMAN GENOME
In a recent longitudinal study carried out in Finland, Hooshmand, et al. (2010), suggested that both total homocysteine and total holotranscobalamin (holoTC), the active fraction of vitamin B12, may be involved in Alzheimer’s disease.
Possible role of prion protein in Alzheimer’s disease Evidence indicates that amyloid Abeta oligomers interact with cellular prion protein (Lauren, et al., 2009). These investigators reported that impairment of synaptic plasticity by amyloid depends on its interaction with cellular prion protein. Gimbel, et al. (2010), crossed Alzheimer’s susceptible mice with the APPswe/Psen Delta E9 mutations with prion protein-negative mice. They determined that the mice lacking cellular prion protein developed amyloid plaques but that specific defects in learning and memory were not present. They proposed that the cellular prion protein significantly impacts amyloid Abeta toxicity and that binding of Abeta oligomers to prion protein impacts synaptic function. They reported that alpha secretase that cleaves amyloid precursor protein also cleaves cellular prion protein. These observations indicate that increased alpha secretase activity may impact amyloid Abeta production and cell prion protein content. They noted further that prion protein-negative mice are viable and have apparently normal synaptic function. Cisse and Mucke (2009) reported that prion protein is anchored to lipid rafts and plays a role in maintaining white matter. Takahashi, et al. (2010), reported that the neuritic plaques that occur in Alzheimer’s disease consist of an amyloid Abeta core and also contain cellular prion protein PrPc. They hypothesized that the presence of PrPc and aggregated amyloid Abeta or aggregated tau protein reflects impaired cellular degradation and transport of these components.
CHAPTER 7 • LATE-ONSET NEURODEGENERATIVE DISEASES
113
Can amyloid behave as prions? Evidence indicates that APOE genotype and clusterin influence the development of Alzheimer’s disease and cerebral amyloid angiopathy through their effects on amyloid Abeta. Holtzman (2004) noted that APOE impacts the conformation and clearance of amyloid. Soluble Abeta peptide has a random coil and alpha helical structure. Pathologic changes involve conversion to a beta sheet. The question arises whether disease-associated protein aggregates, including amyloid aggregates, can behave as prions under certain circumstances. The prion hypothesis postulates that normal cellular protein transforms to assume a beta sheet conformation and that this conformation leads to aggregation and ongoing recruitment of cellular prion protein to the beta sheet conformation and aggregation. Frost and Diamond (2010), reviewed features of neurodegenerative diseases that are reminiscent of prionopathies. They noted that neurodegenerative diseases often begin with pathology in a discrete brain region and often in a particular network, and then progress in a predictable manner. Caughey, et al. (2009), reported that a key feature that distinguishes the transmissible spongiform encephalopathy (TSE) prion protein from other aggregated proteins is the existence on the TSE prion of a glycophosphatidylinositol membrane anchor. They postulate that this anchor leads to major membrane distortion in the brain.
Molecular-based treatment design in Alzheimer’s disease Neprilysin is an endopeptidase that is expressed in a number of different tissues, including the brain. Studies on mouse tissues
114
INVESTIGATING THE HUMAN GENOME
revealed that tissue-specific transcripts are generated and that transcripts in neurons, oligodendrocytes, and kidneys each differ with respect to the exons they contain (Iwata, et al., 2001). Studies on mice hemizygous for neprilysin deficiency revealed increased brain amyloid Abeta42 levels. Evidence also indicates that neprilysin levels are decreased in Alzheimer’s disease and that neprilysin levels are negatively correlated with amyloid levels. Wang, et al. (2010), proposed that measures that increase neprilysin levels might constitute a pathway to Alzheimer’s disease treatment.
Modulation of secretase levels and treatment of Alzheimer’s disease A number of different membrane proteases have alpha secretase activity, and research is ongoing to identify drugs that have the highest specificity for amyloid precursor protein. De Strooper, et al. (2010), reported that at least three different drugs that activate alpha secretase activity are in clinical trials. A beta secretase inhibitor and a gamma secretase inhibitor are also in clinical trials. De Strooper, et al., noted that current trials involve only subjects with moderate to advanced Alzheimer’s disease and that it will be important to carry out testing of these drugs earlier in the stage of the disease. Earlier diagnosis of Alzheimer’s disease may be achieved through analysis of specific biomarkers, including levels of phosphorylated forms of tau and amyloid beta42 in cerebrospinal fluid. One problem that has emerged with the use of beta secretase inhibitors is that this enzyme is also important in the processing of NOTCH (neurogenic Notch homolog). Wolfe and Selkoe (2010) emphasized that important approaches to treating Alzheimer’s disease include developing therapies based on increased cleavage of amyloid precursor protein by alpha secretase, to reduce cleavage by beta secretase. Another important therapeutic
CHAPTER 7 • LATE-ONSET NEURODEGENERATIVE DISEASES
115
approach rests on identifying agents that promote neuronal protection and repair through enhanced neurotrophin signaling. Activating the transcription factor beta retinoic acid receptor can potentially achieve both of these goals through enhanced expression of SIRT1 deacetylase (Donmez, et al., 2010). Approaches to treatment that enhance alpha secretase activity are particularly important, given the problems that have emerged with beta and gamma secretase inhibitors; the latter two secretases are required for processing a range of precursor proteins other than amyloid APP. Alpha secretase cleavage potential is a property of a number of different A type disintegrin and metalloproteinases (ADAM protease). SIRT1 (sirtuin 1) stimulates the expression of ADAM10 protease. Donmez, et al., demonstrated that activation of the retinoic acid receptor by retinoic acid leads to enhanced SIRT1 expression, enhanced NOTCH signaling, and enhanced expression of ADAM10 protease. The question arises whether retinoic acid can be safely used in therapy or whether other less toxic retinoic acid analogs will be required.
Treatment based on restoration of homeostasis of brain cholesterol Pani, et al. (2010), reviewed evidence that lipid rafts directly promote protein misfolding and that this process is directly impacted by the intracellular cholesterol concentration. They proposed that interventions that restore normal brain cholesterol homeostasis could have positive effects. In this context, it is important to note that Hudry, et al. (2010), reduced amyloid plaque in mouse models of Alzheimer’s disease
116
INVESTIGATING THE HUMAN GENOME
through gene therapy and increased expression of cholesterol 24S hydroxylase that facilitates cholesterol clearance from the brain.
Should the role of prion protein be taken into account in designing treatment for Alzheimer’s disease? Chung et al. (2010) reported that cellular prion protein PrPc is a receptor for amyloid beta oligomers. They determined that intraperitoneal infusion in Alzheimer’s disease-prone mice with an antibody that blocked binding of amyloid Abeta oligomers to PrPc was useful in preventing cognitive defects. Researchers in the Prusiner laboratory (Ghaemmaghami, et al., 2010) screened a library of 10,000 small molecules and identified 121 compounds that reduced levels of prion in neuroblastoma cells lines. They concluded that two aminothiazole compounds had anti-prion properties.
Ongoing questions regarding Alzheimer’s disease and approaches to treatment The first key question is whether defects in amyloid precursor protein synthesis, processing, and deposition as aggregates represent primary defects or whether they occur secondarily in response to other factors. Is the primary defect related to amyloid Abeta synthesis, or to deposition or clearance? Is it necessary to treat the primary defect, or can the secondary effects be treated? Another key question is whether the initiating event in Alzheimer’s disease is neuronal destruction by soluble forms of Abeta or whether neuronal damage is due to aggregates of Abeta and plaque formation. Evidence suggests that cognitive impairment is not directly related to the extent of amyloid plaque burden.
CHAPTER 7 • LATE-ONSET NEURODEGENERATIVE DISEASES
117
Does mitochondrial malfunction play a role in Alzheimer’s disease? Is it involved in the early stages, or does mitochondrial damage occur only in the later stages of the disease? To what extent should variations in the APOE genotype be taken into account in design of clinical trials?
Amyotrophic lateral sclerosis Amyotrophic lateral sclerosis (ALS), sometimes known as Lou Gehrig’s disease or motor neuron disease, affects both lower and upper motor neurons. Onset is most commonly between 50 and 60 years of age. The lifetime risk of ALS is 1 in 2,000. Death often occurs within five years of the initial diagnosis and is most often due to respiratory muscle paralysis. More than 90% of cases are sporadic; between 5% and 10% are inherited, usually in an autosomal dominant mode (Rothstein, 2009). ALS serves as a paradigm for complex late-onset diseases. A number of genes have been implicated in its pathogenesis. Mutations in these genes lead to familial ALS that may display autosomal dominant or recessive inheritance. Penetrance of these gene mutations varies even within a single family, and in some families, older members with the same mutation as affected younger members do not manifest symptoms. Some evidence suggests that environmental factors influence penetrance. Some cases with motor neuron disease typical of ALS may also manifest fronto-temporal dementia. Approaches to identifying candidate disease genes and causative mutations have included linkage studies. Genome-wide association studies have also been used to identify causative genes and risk alleles. Analysis of specific disease lesions, such as the aggregates that accumulate in neurons, led to the identification of TDP43 as a key
118
INVESTIGATING THE HUMAN GENOME
component of these lesions. Subsequently, mutations in TDP43 were identified in a subset of patients with ALS. In 1993, Rosen, et al., discovered mutations in the SOD1 gene in familial ALS. Evidence now indicates that SOD1 mutations may also play a role in approximately 1% of sporadic cases of ALS. Rothstein (2009) noted that 114 different ALS-related mutations occur and are distributed throughout the SOD1 gene. SOD1 encodes a cytosolic enzyme that converts superoxide anions to hydrogen peroxide (H2O2). ALS is apparently due not to SOD1 deficiency, but to toxic properties of the mutant enzyme. The precise mechanisms through which mutant SOD1 causes neuronal damage are not known. Evidence from mouse models of human SOD1 mutations indicates that aggregates of SOD1 occur early in the disease and that these aggregates sequester other material. Rothstein (2009) noted that mitochondrial damage is more evident with some SOD1 mutations than with others. He noted that, in SOD1 mutant mice, apoptosis is prominent in motor neurons. Evidence suggests that SOD1 aggregates damage neurons and axons. Rothstein concluded that research data provides evidence that mutant SOD and that SOD1 aggregates inhibit chaperone and proteosome activity; these aggregates also disrupt mitochondrial function and lead to oxidative damage. In a study of 162 index cases in unrelated families with ALS, Millecamps, et al. (2010), found 20 cases with SOD1 mutation; in these cases, 18 different mutations occurred. They reported that, within the group of ALS patients with SOD1 mutations, approximately half had a rapid course: Time from onset of disease to death was three years. They noted that fronto-temporal dysfunction was not present in any of the ALS patients with SOD1 mutations.
TAR DNA-binding protein 43 In 2006, Neumann, et al., and Arai, et al., determined that transactivation response (TAR) DNA-binding protein TDP43 is a key
CHAPTER 7 • LATE-ONSET NEURODEGENERATIVE DISEASES
119
component of ubiquitinated protein aggregates that accumulate in patients with sporadic ALS and in patients with fronto-temporal dementia. Symptoms of ALS and dementia occur together in some patients. Lagier-Tourenne, et al. (2010), reported 48 different mutations in the TDP43 in approximately 4% of familial ALS patients and in less than 1% of sporadic ALS patients. They noted that most patients with TDP43 mutations do not have fronto-temporal mutations and that rare TDP43 mutations may lead to FTD with or without motor neuron disease. Most of the disease-causing TDP43 mutations are in the C terminal, glycine-rich domain of the TDP43 protein. These investigators reported that post-mortem studies have revealed widespread TDP43 inclusions in the brains of patients with ALS, even in the absence of cognitive impairment. TDP inclusions show evidence of hyper-phosphorylation and ubiquitination. TAR DNA-binding protein 43 (TDP43) inclusions were found to be present in Alzheimer’s disease, dementia with Lewy bodies, and Parkinson’s disease with dementia. Schwab, et al. (2009), and LagierTourenne, et al. (2010), reported that TDP43 inclusions are present in familial British dementia, a condition associated with amyloid deposition in blood vessels in the brain (amyloid angiopathy).
FUS DNA RNA-binding protein in ALS Mutations in this protein were first reported in cases of familial ALS in 2009 by Kwiatkowski, et al., and by Vance, et al. LagierTourenne, et al. (2010), noted that (fused protein) FUS mutations also occur in rare sporadic cases of ALS. In familial ALS inheritance of FUS mutations usually lead to autosomal-dominant inheritance of ALS. In one family from the Cape Verde Islands, FUS mutations led to disease in homozygotes only, indicating a recessive pattern of inheritance. Most diseasecausing mutations are missense mutations, but a 29 base pair deletion
120
INVESTIGATING THE HUMAN GENOME
mutation and an 11 base pair insertion mutation have been described. In addition, splice-site mutations in FUS may lead to disease. A number of mutations show evidence of incomplete penetrance. LagierTourenne, et al. (2010), noted that patients with a specific FUS mutation, R521C, develop an unusual early symptom of a dropped head, due to weakness of the extensor neck muscles. FUS protein aggregates occur primarily in the nucleus. They contain ubiquitin but do not contain TDP43 (Lagier-Tourenne, et al., 2010).
Aggregates in other forms of late-onset neurodegenerative conditions Lagier-Tourenne et al. (2010) reviewed information on the presence of FUS or TDP43 aggregates in late-onset neurodegenerative condition. They noted that, in some cases of fronto-temporal dementia, FUS protein occurs in aggregates. TDP43 protein is present in inclusions in Parkinson’s disease, Alzheimer’s disease, Huntington’s disease, some forms of spino-cerebellar ataxia, and several forms of myopathy. They noted further that, in Alzheimer’s disease and Parkinson’s disease, TDP43 inclusion may be separate from or partially associated with tau or synuclein aggregates. TDP43 and FUS have similar molecular structures, and both bind mRNA and DNA. FUS plays a role in transcription and interacts with both the promoter region of genes and specific transcription factors. FUS is a multifunctional protein; it constitutes a component of heterogeneous nuclear riboprotein complexes (hnRNP). It also plays roles in gene expression in mRNA and microRNA processing and in regulation of genomic integrity. FU5 protein encoded by a gene on 16p11.2 contains a 90 amino acid RNA recognition motif and RNA binding domain that plays a role in RNA processing and alternate splicing. In addition, it contains three separate domains that have
CHAPTER 7 • LATE-ONSET NEURODEGENERATIVE DISEASES
121
homology to splicing factors. The TARDBP gene on chromosome 1p36.22 encodes the TDP43 protein. Baumer, et al. (2010), concluded that growing evidence indicates that motor neuron degeneration results from mutations in genes that encode products that play roles in ribonucleoprotein biogenesis and RNA processing. The latter includes regulation of expression, splicing, transcript stabilization, localization of RNA, and RNA translation.
Genotype-phenotype studies in ALS Millecamps, et al. (2010), carried out mutation analyses of SOD1, TARDBP, and FUS in index cases in 162 families with ALS and in 500 controls. They identified disease-causing mutation in 36 cases. Segregation of a specific mutation with ALS was documented in 11 families, including 4 with SOD1 mutations, 3 with TARDP mutations, and 4 with FUS mutations. They documented variable penetrance of disease; age of onset and initial site of symptoms varied among members of a family. Millecamps, et al., concluded that the occurrence of a specific mutation in SOD1, TARDP, or FUS genes is not the only factor that determines the ALS course. Finding a specific mutation and documenting the course of disease in one family member does not permit prediction of occurrence or rate of disease in other family members. They concluded further that, despite advances in genetics, considerable work is required to determine interacting genetic and environmental factors in ALS.
Analysis of protein aggregates in ALS Deng, et al. (2010), analyzed immunoreactive inclusions in postmortem tissues from 78 cases of clinically and pathologically diagnosed ALS cases. They analyzed spinal cord sections. In ten cases
122
INVESTIGATING THE HUMAN GENOME
diagnosed with dementia, they analyzed hippocampus and frontal lobe sections. Their studies revealed FUS positive inclusions in spinal cord anterior horn neurites in 52 sporadic ALS cases and in 10 cases that had been diagnosed with ALS and dementia. FUS inclusions were also present in the spinal cord of one case with a TDP mutation and in one case with an FUS mutation. Importantly, FUS inclusions were not present in four cases with SOD1 mutations. FUS positive inclusions were not found in mice with SOD1 mutations. Deng, et al., concluded that the pathogenic pathway in SOD1 ALS is distinct from that in other forms of ALS.
Proteinopathies, induction of misfolding, and formation of protein aggregates In some cases, changes can be identified in genes that encode the proteins that are precipitated. In many cases, gene changes are not detected, but specific proteins precipitate and form aggregates. Why? Does gene expression change so that larger quantities of the specific protein are formed? Do cellular factors change and promote secondary modification? In reviewing the formation of amyloid aggregates, Holtzman (2004) noted that normally amyloid Abeta peptide has a random coil and alpha helical structure, and is soluble. Pathological changes involve conversion of that form to a beta sheet formation. Holtzman cited evidence that the APOE genotype influences the occurrence of Alzheimer’s disease and cerebral amyloid angiopathy through its effects on amyloid Abeta. These include effects on amyloid conformation and clearance.
CHAPTER 7 • LATE-ONSET NEURODEGENERATIVE DISEASES
123
Aggregates in Parkinson’s disease Alpha synuclein aggregates occur in Parkinson’s disease. Burre et al. (2010) studied the physiological functions of synuclein. They reported that alpha synuclein protein stimulates the assembly of a specific protein complex SNARE on presynaptic membranes. They noted that the function of alpha synuclein is particularly important during increased synaptic activity. They proposed that sequestration of synuclein into Lewy bodies in Parkinson’s disease or degradation of synuclein contribute to neurodegeneration through impaired presynaptic activity. Tau aggregates occur in more than 20 different neurodegenerative diseases. In most of these diseases, the tau encoding gene is apparently not mutated. However, in fronto-temporal dementia with Parkinsonism, mutations occur in the tau encoding gene MAPT.
Neurodegenerative diseases and mitochondria In reviewing the role of mitochondria in neurodegenerative diseases, Lin and Beal (2006) noted that mitochondrial mutations and increased production of reactive oxygen species (ROS) are key factors in aging, and aging is the greatest risk factor for development of neurodegenerative diseases. In aging, both large-scale deletions of mitochondrial DNA and point mutations commonly occur. Kraytsberg, et al. (2006), reported that, in Parkinson’s disease, mitochondrial DNA deletions are common and lead to functional impairment. Lin and Beal (2006) noted that mitochondria contain multiple electron carriers capable of producing ROS and that aging results in diminished defenses against ROS.
124
INVESTIGATING THE HUMAN GENOME
They listed mitochondrial enzymes capable of generating ROS. These include the tricarboxylic acid enzymes aconitase and alphaketoglutarate dehydrogenase; pyruvate dehydrogenase and electron transport complexes I, II, and II; glycerol-3-phosphate dehydrogenase; the urea cycle enzyme dihydro-orotate dehydrogenase; monoamine oxidase; and cytochrome b2 reductase. Lin and Beal noted that the generation of superoxide is more likely when redox carriers are abundantly charged with electrons. The key antioxidant defense systems include factors such as coenzyme Q, glutathione, alpha tocopherol (vitamin E derived) and enzymes including manganese dependent superoxide dismutase (MnSOD), glutathione reductase, and thioredoxins. Activity of the thioredoxins is dependent on the generation of adequate NADPH through activity of isocitrate dehydrogenase and nicotinamide transhydrogenase that impacts membrane potential. Lin and Beal noted that evidence of oxidative damage occurs early in Alzheimer’s disease, before the development of amyloid plaques. In a transgenic APP mutant mouse with hemizygous deficiency of the manganese-dependent superoxide dismutase, levels of amyloid Abeta and amyloid deposition were increased more markedly than those in an APP transgenic mouse with normal MnSOD activity. Evidence from studies in guinea pigs also indicates that oxidative stress increases the expression of beta secretase and tau phosphorylation.
Neuronal oxidative and nitrosative stress and protein modification Evidence suggests that oxidative and nitrosative stress leads to protein modification and misfolding. Hardingham and Lipton (2010) reviewed neuroprotective pathways that counteract these stresses and noted that master regulators of these pathways include NRF2
CHAPTER 7 • LATE-ONSET NEURODEGENERATIVE DISEASES
125
(nuclear factor) and PGC1alpha. They noted that an imbalance between production and neutralization of reactive oxygen species (ROS) and reactive nitrogen species (RNS) is damaging to cells, particularly neurons. Antioxidative systems include glutathione and the thiol-reducing systems thioredoxin and peroxiredoxin. Hardingham and Lipton reported that an increased expression of nerve growth factor (NGF) leads to upregulation of the glutathione redox pathway. NGF also prevents translocation of BAX (BCL2 associated apoptosis regulator) to mitochondria and subsequent ROS generation, release of cytochrome C and apoptosis. They reported that synaptic activity, caloric restriction, and exercise enhance the expression of PGC1alpha. NRF2 is a transcription factor that plays a key role in activating the expression of genes that encode antioxidative enzymes and proteins. Large quantities of ROS and RNS are produced under conditions of cerebral ischemia. Hardingham and Lipton noted that clinical trials of small-molecule antioxidants have had limited effects in reducing ROS levels. They emphasize that boosting intrinsic antioxidant defenses may offer a more fruitful approach to reducing levels of ROS. Gu, et al. (2010), postulated that nitrosative and oxidative stress contribute to protein misfolding and to mitochondrial fragmentation in patients with neurodegeneration. Specific examples of proteins and enzymes impacted by S-nitrosylation include Parkin, ubiquitin E3 ligase, the chaperone protein disulfide isomerase, and DRP1, a mitochondrial fusion and fission protein. S-nitrosylation involves the reaction of nitrous oxide (NO) with cysteine residues in target proteins and the formation of S-nitrosothiols. Nitric oxide is produced from the amino acid arginine through the activity of nitric oxide synthase (NOS). Gu, et al., noted that three types of nitric oxide synthase occur: neuronal, endothelial, and
126
INVESTIGATING THE HUMAN GENOME
inducible NOS. The latter is usually induced as part of the inflammatory response.
Prionlike activity of aggregates in neurodegenerative conditions A key question is whether the aggregates that accumulate in neurodegenerative condition are cell autonomous, indicating that they arise de novo in each cell, or whether aggregates may be transferred between cells, as is the case with prions. Evidence from in vitro studies on animal models of neurodegenerative diseases indicates that aggregates are transferred from one cell to another. In reviewing evidence for such a transfer, Goedert, et al. (2010), considered potential mechanisms underlying this transfer. They proposed that abnormal aggregates, including neurofibrillary tangles or synuclein-containing Lewy bodies, are taken up into exosomes that are then extruded from donor cells. These exosomes may then fuse to cell membranes and endosomes of acceptor cells. Aggregates may subsequently be released into the cytoplasm and polymerize to form larger inclusions. Another possible mechanism of transfer proposed is that tunneling nanotubes evolve that connect donor and acceptor cells. Rajendran (2006) reported that amyloid-derived peptides are transferred in exosomes. Goedert, et al. (2010), emphasized that, in inherited forms of neurodegenerative disease, mutant proteins aggregate or mutant proteins may be overproduced. However, in sporadic forms of neurodegenerative disease, the proteins within aggregates are not mutant, nor are they overproduced. Results of studies carried out between 2006 and 2009 led Goedert, et al., to propose that sporadic neurodegenerative disease could originate in a specific localized region and then spread to other
CHAPTER 7 • LATE-ONSET NEURODEGENERATIVE DISEASES
127
regions. Origin in a localized area opens the way for different therapeutic approaches, including immunotherapy.
Inducible proteopathies Walker, et al. (2006), used the term “inducible proteopathies” to define degenerative disease characterized by the accumulation and polymerization of specific proteins. These investigators noted that a number of proteopathies are induced in animal models through exposure to exogenous material. They emphasized that understanding the earliest events that induce accumulation and polymerization of specific protein will be important in developing treatments.
Prions Prion proteins assume many different conformations. The native protein may exist in a soluble nonprion form. It may sequester additional prion proteins and undergo conformational change. The modified prion serves as a template for the conversion of other native prion protein molecules. It replicates, therefore, by bringing about the conversion of other native prion protein molecules (Halfmann and Lindquist, 2010). In Kuru disease, the converted prion protein acts as the infectious agent. The key question, then, is what triggers prion proteins to take on abnormal conformations? Evidence from studies of prion proteins in yeast indicate that environmental stress, including changes in temperature, pH, and intracellular metabolite concentrations, impact the rate of prion protein formation. Halfmann and Lindquist noted that the activities of a large number of proteins are altered through covalent modifications, including
128
INVESTIGATING THE HUMAN GENOME
disulfide bridge formation, phosphorylation, ubiquitination, and glycosylation, and through protein-protein interactions. Two separate proteins may interact to give rise to a prion conformation.
Mitochondrial function and regulators: possibilities for therapy Electrons derived from the metabolism of glucose, amino acids, and fatty acids are transmitted through the mitochondrial electron transport complexes and ultimately drive the conversion of ADP to ATP, oxidative phosphorylation, and energy generation. During electron transport passage through these complexes, protons are generated and pass into the mitochondrial matrix and through the inner mitochondrial membrane to the intermembrane space, to give rise to a proton gradient. If protons interact with uncoupling protein UCP1, heat is generated. Impairments in oxidative phosphorylation may arise when levels of metabolites derived from nutrients exceed ADP levels or if electron passage through electron transport complexes is impaired. Impaired oxidative phosphorylation leads to the generation of reactive oxygen species (ROS), which leads to oxidative damage of lipids, proteins, and DNA (Wallace, 2005). Master regulators of mitochondrial function include PGC1alpha. Evidence also indicates that sirtuins play key roles on mitochondrial metabolism and biology. Verdin, et al. (2010), reviewed functions of the sirtuin protein SIRT1-7. SIRT1, 6, and 7 occur in the nucleus. SIRT2 is active in the cytosol, and SIRT3, 4, and 5 occur in the mitochondria. Each sirtuin protein has a core domain composed of an NAD binding site and a catalytic site. The key enzymatic activity of sirtuins is deacetylation of the coenzyme NAD (nicotinamide adenine dinucleotide). Evidence suggests that mitochondrial sirtuins function as metabolic sensors.
CHAPTER 7 • LATE-ONSET NEURODEGENERATIVE DISEASES
129
Under nutrient deprivation, NAD levels rise and lysine residues in metabolic enzymes and mitochondrial proteins are heavily acetylated. Mitochondrial situins are particularly active during cellular stress and maintain metabolism. SIRT3, 4, and 5 are located in different subcompartments in mitochondria. SIRT3 activity leads to deacetylation of these enzymes and proteins.
Potential therapeutic roles of sirtuins in neurodegenerative diseases Evidence indicates that overexpression of nuclear SIRT1 through calorie reduction or through pharmacological activation (for example, with resveratrol) leads to increased alpha secretase activity and decreased generation of amyloid Abeta peptides. In 2009, de Oliveira, et al., reported that injection of modified lentivirus containing SIRT1 genes led to protection against development of neurodegeneration in transgenic Alzheimer’s disease-prone mice. Increased SIRT1 expression leads to increased activity of PPAR gamma, PGC1alpha and NFkappa B, and FOXO transcription factors. Sirtuin 1 activity is increased by resveratrol, and SIRT2 activation is observed following quercetin administration. A concern in designing treatment is that increased sirtuin activity may lead to decreased availability of NAD, which is required for many metabolic processes.
Questions regarding late-onset neurodegenerative diseases Late-onset neurodegenerative diseases are characterized by inclusions and aggregates of protein in the brain. Interestingly, more recent studies indicate that the same protein may form aggregates in different diseases. This is particularly true of tau protein and TDP43.
130
INVESTIGATING THE HUMAN GENOME
Evidence also indicates that the clinical features in a particular patient may include dementia and Parkinsonism. In patients with ALS, dementia may be absent, but in some cases, it is present. Perhaps these diseases are more closely related to each other than neurological classifications lead us to believe. It seems likely that most cases of late-onset neurodegenerative disease are multifactorial in etiology. Different predisposing factors, genetic and environmental, come together to induce phenotypic defects. Do extrinsic factors induce aggregation, and do these factors include oxidants, dietary factors, or perhaps infectious agents? The most powerful inducer of these diseases is a change that is directly related to aging. Few reports address research on the specific age-related factor. Perhaps age-related mitochondrial factors are key.
8 Genes and genomes in cancer: targeted therapies The goal of this chapter is to consider advances in the understanding of the biology of tumors through genomic and genetic studies. These studies have increasingly led to delineation of specific signaling pathways that are impacted in tumors. Insights into these pathways have led to advances in cancer therapy. However, technologies that enable in-depth analysis of the clones within tumors reveal the extent of genomic changes that occur in tumors and the difficulty in distinguishing driver mutations that are key in tumor progression from passenger mutations that may play secondary roles in progression.
Molecular studies in tumors and therapeutic developments In a review of the history of cancer therapeutics and practices currently being adopted, Lord and Ashworth (2010) noted that treatment of cancer previously involved chemical toxins, such as nitrogen mustard. This approach to treatment was followed by the use of agents that generally impact DNA synthesis; cell division and proliferation, such as folate analogs (aminopterin); and purine analogs (6mercaptopurine). Lord and Ashworth noted that breast cancer treatment became more effective following the development of antihormone agents such 131
132
INVESTIGATING THE HUMAN GENOME
as tamoxifen, which acts as an estrogen receptor antagonist. Tamoxifen is useful in treating tumors that express estrogen receptors. These authors emphasized that the molecular characterization of tumors that started in the 1970s and 1980s has dramatically changed approaches to cancer therapy. One of the most striking examples is the identification of a specific chromosomal anomaly, the Philadelphia chromosome, in chronic myelogenous leukemia. Studies of the origin of this anomalous chromosome followed. Chromosome breakage and subsequent fusion result in the chromosome 9 ABL oncogene locus fusing to the BCR locus on chromosome 22. This fusion leads to constitutive activation of ABL kinase. Development of a small molecule inhibitor of kinase activity, Gleevec (imatinib), followed and opened the way to successful therapy. Imatinib targets the catalytic domain of activated ABL. In some cases, resistance to therapy emerges. An allosteric kinase inhibitor was identified that binds to a different site on the activated ABL protein. Lord and Ashworth noted that the two inhibitors of activated ABL are sometimes used in combination in leukemia therapy. They noted further that imatinib is sometimes effective in inhibiting other related kinases, such as KIT and platelet-derived growth factor PDGFB. The latter is activated in chronic myelomonocytic leukemia. Specific inhibitors of epidermal growth factor receptor (EGFR) include gefitinib and erlotinib. EGFR is activated in a number of cancers, including head, neck, and lung cancer. The epidermal growth factor receptor 2 (EGFR2 or HER2) is often activated in breast cancer, and these cancers may be successfully treated with Herceptin (trastuzmab). The enzyme aromatase p450 that plays a role in hormone synthesis is the target of a number of therapeutic agents used in treating breast cancer.
CHAPTER 8 • GENES AND GENOMES IN CANCER
133
Hedgehog signaling pathway in cancer Insight into the role of the hedgehog pathway in tumorigenesis was derived in part from information obtained from patients with a rare genetic disorder, basal cell nevus syndrome. This disorder is characterized by the development of skin tumors and, in some, cases brain tumors (medulloblastoma). Basal cell nevus syndrome was mapped to human chromosome 9q22.2 and was subsequently shown to be caused by mutations in the Patched 1 gene (PTCH1) (Hahn, et al., 1996). Somatic mutations in the PTCH1 gene occur in some cases of basal cell carcinoma and medulloblastoma. PTCH1 is a receptor for a secreted protein HH that is homologous to a drosophila protein hedgehog (SHH). PTCH1 binding of HH leads to repression of the gene SMO (homologous to a drosophila gene called smoothened). In the absence of adequate PTCH1, SMO cannot be repressed. Excess SMO activates expression of the GLI gene. This then activates transcription of a number of genes, leading to tumor cell proliferation. Von Hoff, et al. (2009), reported that a specific small molecule GDC-0449 targets SMO and inhibits the hedgehog signaling pathway. Rudin, et al. (2009), reported that tumors that become resistant to GDC-0449 therapy have developed mutations in the SMO. Medulloblastomas with aberrant SHH signaling often occur in very young patients and have a poor prognosis. Gibson, et al. (2010), reported that these tumors arise from granule precursor cells in the cerebellum. Medulloblastomas with activation of the WNT signaling pathway occur in older children, arise from a different cell type, and have a different location. They are often located in the dorsal brainstem. These tumors frequently have beta-catenin mutations.
134
INVESTIGATING THE HUMAN GENOME
In an analysis of 62 medulloblastomas, Kool, et al. (2008), identified five molecular subtypes. Nine cases were characterized by WNT signaling pathway abnormalities, and all of these cases had betacatenin mutations. The SHH pathway was involved in 15 cases, and all had PTCH1 mutation. Abnormalities in neuronal differentiation genes and/or photoreceptor genes occurred in the remaining cases. Determination of the molecular subtype is important in designing therapies. Evidence now indicates that the hedgehog signaling pathway is activated in other cancers, including pancreatic cancer (Jones, et al., 2008) and colorectal cancer (Saif and Chu, 2010).
Drivers and passengers A specific single gene is the primary driver for tumor development in relatively few cancers. Jones, et al. (2008), determined that pancreatic cancer, for example, results from genetic changes in a large number of genes that operate in relatively few pathways. These include signaling, DNA damage control, apoptosis, and cell adhesion pathways. They concluded that therapeutic agents could be developed that target nodal points in pathways. Therapeutic agents that act downstream to target metabolism or angiogenesis in tumors may also be effective in combating tumor growth. Detailed microarray studies to determine genomic changes such as deletion, duplication, and comprehensive sequence analyses have revealed that many genes are altered in a malignant tumor. Somatic mutations may be drivers or passengers. Driver mutations confer growth and selective advantages and play key roles in cancer. DNA sequence analyses have revealed that driver mutations are most often nonsynonymous mutations. Although tumors are often highly heterogeneous in terms of their individual mutations, they are less heterogeneous with respect to the pathways involved (Jones, et al., 2008).
CHAPTER 8 • GENES AND GENOMES IN CANCER
135
There are, however, examples of specific types of cancer where mutations involving a specific gene occur in a relatively high percentage of individuals. For example, studies carried out by a number of different individuals have revealed that 40%–60% of melanomas and 7%–8% of all cancers have activating mutations in the serine threonine protein kinase BRAF. Furthermore, 90% of the BRAF cancer mutations lead to substitution of glutamic acid for valine at position 600 in the BRAF protein (V600E) (Flaherty, et al., 2010). This mutation leads to constitutive activation of BRAF and activation of the Map kinase signaling pathway. The drug PLX4032 is a potent inhibitor of V600E form of BRAF. Flaherty, et al. (2010), carried out clinical trials specifically in melanoma patients who had this BRAF mutation. They found that PLX4032 treatment resulted in complete or partial tumor regression in the majority of melanoma patients. Side effects were mild and were proportional to the drug dosage. Unfortunately, in some cases, tumors that were initially sensitive to PLX4032 later become resistant to this treatment. Johannessen, et al. (2010), carried out studies to identify the specific genes that led melanomas to become resistant to therapy. Initially, they used cultured melanoma cells and transfected these cells with individual ORF (open reading frame) clones that each expressed a specific kinase gene. They determined that increased expression of the kinase MAP3K8, also known as COT, drives resistance to kinase inhibitor PLX4032. They then examined tumor samples from three patients who were resistant to therapy with PLX4032 and showed increased expression of MAP3K8 mRNA in tumors from two of the three patients. They then demonstrated that treatment with a small inhibitor of MAP3K8 (COT) suppresses growth of the resistant tumors. Johannessen, et al., noted that single-agent targeted therapy is often very effective initially but that it is almost invariably followed by
136
INVESTIGATING THE HUMAN GENOME
resistance. Screening methods to identify the molecular basis of such resistance opens the way for the design of combined therapies.
Synthetic lethal therapeutic strategies The concept of synthetic lethality is based on the observation that two proteins, each consistent with cell survival when individually inactivated, may lead to cell death when both are inactivated. One striking example of this is the relationship between inactivating BRCA mutations and poly-ADP ribose polymerase (PARP). PARP plays a role in repairing single-stranded DNA breaks. Inhibition of this protein is compensated for by DNA repair via a homologous recombination mechanism. The latter is usually controlled by BRCA1 and BRCA2. Deficiency of one of the latter due to mutations or deletions renders a cell incapable of compensating for the inhibition of PARP activity. Fong, et al. (2009), translated this observation into clinical trails and demonstrated that PARP inhibitors had significant antitumor activity in patients with BRCA mutations leading to breast or ovarian cancer. PARP inhibitors reportedly had few side effects.
Exploiting molecular networks in cancer therapy Lord and Ashworth (2010) emphasized that analysis of tumor biology needs to have a major role in cancer therapy. They emphasized that patients should be treated on the basis of biomarker analysis of their tumors. De Bono and Ashworth (2010) stated that clinical trials should be designed to define the best treatment for a specific patient rather than determining the best treatment for the average
CHAPTER 8 • GENES AND GENOMES IN CANCER
137
patient. Trials should also assess variability between patients, with a goal of finding the dose of therapeutic agents that leads to maximum antitumor activity with minimal toxicity.
Hypothesis testing clinical trials One example of such a clinical trial was the use of a small molecule inhibitor of BRAF in melanoma patients using PLX4032. Another example was the use of small molecule inhibitors of the KIT oncogene and PDFGR (platelet-derived growth factor receptor), where these genes were activated in gastrointestinal stromal tumors. Surgery is first-line treatment; imatinib and sunitinib may be useful as a second-line therapy (Martin-Broto, et al., 2010).
Identification of nodal points in pathways to design therapeutic interventions Evidence indicates that increased activity in the beta-catenin signaling pathway facilitates growth, survival, and invasiveness of colorectal cancer cells. Firestein, et al. (2008), carried out experiments to identify oncogenes that modulate beta-catenin expression in colon cancer. In one series of experiments, they carried out RNA interference screens. The goal of these experiments was to identify genes necessary for the proliferation of cultured cells from different colon cancer samples. They identified nine genes that, when down-regulated through RNA interference, altered colon cancer cell proliferation. They then used SNP microarrays to search for copy number changes in genomic DNA derived from 123 primary colorectal adenocarcinomas. The determined that one of the nine genes identified in the RNA inhibition studies mapped in a region that was amplified, as a
138
INVESTIGATING THE HUMAN GENOME
result of copy number changes, in primary colon cancer tumors. This gene was cyclin-dependent kinase 8 (CDK8) that maps to 13q12.13-q12.2. Chromosome 13 regions showed amplification in 58 of 123 tumors studies. They determined that the minimal region of amplification encompassed 16 genes that included CDK8; in some cases, the retinoblastoma gene RB1 was also amplified. Firestein, et al., then carried out studies on each of the amplified genes to determine which were necessary to maintain the proliferation in colon cancer cell lines. They determined that CDK8 was the only 13q gene necessary for proliferation. In subsequent experiments, they determined that colon cancer cell lines with the highest levels of CDK8 expression also showed greater dependence on beta-catenin for proliferation. Firestein et al. determined that suppressing CDK8 expression with shRNA inhibition reduced beta-catenin activity. Cyclin C is a cofactor for CDK8, and cyclin C knockdown negatively impacted the growth of colon cancer cell lines with 13q amplification. Firestein, et al., also established that CDK8 modulates the expression of a subset of other genes implicated in the growth of colon cancer, including MYC oncogene and AXIN2 and LEF1 transcription factor encoding genes. In summary, through integrated analysis involving RNA inhibition, studies in cultured cells, and copy number analyses in tumor samples, Firestein et al., established that high CDK8 levels lead to high levels of beta-catenin expression and colon cancer cell proliferation. These investigators concluded that targeting CDK8 activity might be of therapeutic value in colon cancer patients. In specific forms of colon cancer, the adenomatous polyposis cancer tumor suppressor gene (APC) is deleted. In normal cells, the APC gene product binds to beta-catenin. APC, GSK3B, and axin interact with beta-catenin and bring about its degradation. Beta-catenin inter-
CHAPTER 8 • GENES AND GENOMES IN CANCER
139
acts with transcription factors TCF and LEF1, and this complex enters the nucleus and enhances the expression of genes that encode tumor-promoting proteins MYC and cyclin.
Beta-catenin and hepatocellular carcinoma Different signaling pathways are impacted in hepatocellular carcinoma. Whittaker, et al. (2010), reported that between 8% and 13% of these tumors have mutations in genes that activate beta-catenin and that 17% of hepatocellular carcinomas have mutations in betacatenin. Whittaker, et al., noted that key factors in the pathogenesis of hepatocellular carcinoma are tissue damage leading to regeneration, and cirrhosis, and mutations in oncogenes or tumor suppressor genes. Liver tissue damage most frequently results from viral infection with Hepatitis B or C virus, and toxin exposure including alcohol or aflatoxin. In specific inborn errors of metabolism, abnormally high concentrations of tyrosine and metabolites may lead to liver damage. Liver damage may also occur as a complication of alpha-1-antitrypsin deficiency. Quercetin is a small molecule inhibitor shown to be useful in cancer treatment partly through its impact on the Wnt/beta-catenin signaling pathway.
Cancer therapy networks and oncogene addiction Identification of the increased expression of an oncogene in a tumor and the possibility that tumor growth is dependent on increased expression of that oncogene (oncogene addiction) present an avenue for therapeutic intervention. Protein kinases often are required for tumor growth and may be inhibited through the use of specific inhibitors (Lord and Ashworth, 2010).
140
INVESTIGATING THE HUMAN GENOME
Analysis of the biology of primary tumors, metastatic tumors, and circulating tumor cells Identification of optimal tumor targets for therapy depends on detailed studies of tumor biology. De Bono and Ashworth (2009) noted that because resistance to any chemotherapy is likely to develop, it is important that metastatic tumors be studied. Recent technological developments facilitate the isolation and analysis of circulating tumor DNA; in some cases, circulating tumor cells may be isolated. Strategies for the purification of tumor cells include immunomagnetic bead capture using antibodies directed against tumor surface antigens. Other tumor cell capture techniques include microfluidic-based methods in which cells circulate through microwells coated with specific antibodies.
Fusion genes in specific forms of cancer and their utility as biomarkers for diagnosis and treatment In experiments designed to identify novel transforming genes in lung tumors, Soda, et al. (2007), isolated mRNA from these tumors. They then reverse transcribed tumor mRNA and derived cDNAs that they amplified and cloned to develop a library of plasmid clones. Individual plasmid clones were then transfected into mouse 3T3 fibroblasts to identify which plasmid clones gave rise to transformed clones. One of the transforming clones derived from lung tumors was found to encode a novel 1,059 amino acid protein. The amino terminal 496 amino acid in this clone corresponded in sequence to the EML4 protein (echinoderm microtubule associated protein-like). The carboxy terminal portion of the novel protein (amino acids 497–1,059) corresponded in sequence to the intracellular domain of the human ALK gene (anaplastic lymphoma kinase). EML4 maps to
CHAPTER 8 • GENES AND GENOMES IN CANCER
141
chromosome 2p21. ALK maps to human chromosome 2p23. The two genes are normally 12 megabases (Mb) apart. The novel gene isolated from the lung tumor is therefore a fusion gene. Soda, et al., sequenced genomic DNA directly from the tumor and determined that in the tumor EML4 is disrupted 3.6 kilobases (Kb) downstream of exon 13 and is fused to an intron in ALK that is 297 basepairs upstream of exon 21. The abnormal fusion product was not present in normal cells. Importantly, Soda, et al., were able to detect the abnormal fusion product in the sputum of the tumor patient using PCR technology. A number of additional studies have confirmed that EML4–ALK fusions occur in subsets of patients with non-small cell lung cancers (NSCLC). Shaw, et al. (2009), studied 149 NSCLC tumors and determined that 19 tumors (13%) had the EML4–ALK fusion gene. Reports indicate that tumors with ALK–EML4 fusions respond to treatments with specific ALK inhibitors. EML4–ALK fusion positive tumors were present in patients who had never smoked or who had a light smoking history. On average, patients in this group were younger than other NSCLC patients. Evidence now suggests that, in a number of different cancers, ALK may fuse with different genes to trigger signaling cascades and cancer growth (Garber, 2010). Fusions that have been reported include ALK/TPM4, ALK/TPM3 (tropomyosins 3 and 4), ALK/ CLTC (clathrin heavy chain), and ALK/NPM (nucleoplasmin). The compound Crizotinib was found to be a potent inhibitor of the growth of cultured cells with ALK/EML4 fusion. Garber (2010) noted that therapy with this compound is not curative in patients with EML4–ALK fusions. Tumors initially shrink, but resistance later develops. Shaw, et al. (2009), determined that 31 of the 141 NSCLC tumors they studied carried mutations in EGFR (epidermal growth factor receptor). Very important findings were that patients with EGFR
142
INVESTIGATING THE HUMAN GENOME
mutation-bearing tumors responded to therapy with EGFR tyrosine kinase inhibitors; 70% of patients with EGFR mutations had a positive response to treatment with inhibitors such as erlotinib. On the other hand, patients with the EML4ALK fusions were resistant to these agents. The ALK kinase is inactive in the absence of ligand. In the present of bound ligand or when constitutively activated through fusion with other genes, ALK decreases apoptosis (Allouche, 2007). ALK expression is important in development, particularly in neuronal development. Allouche (2007) noted that ALK expression increases in neuronal crest-derived tumors such as neuroblastomas and glioblastomas.
Targeting oncogene activation in tumor therapies Gastrointestinal stromal tumor is a solid tumor that arises in connective tissue. Gain of function mutations in the KIT oncogene occur in approximately 80% of these tumors. In most cases, these are somatic mutations that arise in the tumor. Rare forms of familial GIS tumors occur and are associated with germline mutations in the KIT oncogene (Heinrich and Corless, 2010). KIT is an Imatinib-sensitive tyrosine kinase inhibitor. Continuous treatment with Imatinib is necessary to avoid tumor recurrence. Activated KIT expression in GIS tumors leads to increased activity in the RAS, RAF, and MEK signaling pathways. An increased MEK signal reduces the proteosomal degradation of ETV1, leading to increased levels of ETV1 (an ETS-like transcription factor). ETV1 enters the nucleus and enhances gene transcription and facilitates tumor growth. Heinrich and Corless proposed the use of ETV1 ETS transcription factor inhibitors in combination with KIT inhibitors to treat GIS tumors.
CHAPTER 8 • GENES AND GENOMES IN CANCER
143
Gene rearrangements leading to increased expression of ETS transcription factors Tomlins, et al. (2005), screened prostate tumors to identify genes that were markedly overexpressed. They determined that two transcription factors, ERG and ETV1, that belong to the ETS family of transcription factors were overexpressed. They then carried out genomic analysis in these tumors and determined that a translocation between ETV1 and TMPRSS2 (transmembrane protease serine 2) was present. The latter is an androgen-responsive gene. Promoter elements from TMPRSS2 were translocated to ERG or to ETV1 and led to overexpression of the transcription factors. Expression of the transcription factors was then directly controlled by androgens. In 2009, Tomlins, et al., reviewed reports of ETS gene fusions and prostate cancer. They noted that ETS-like transcription factors had a number of fusion partners and that the most common fusion of TMPRSS2–ERG was reported to be present in approximately 50% of localized prostate cancers. ETS fusions were detected in urine of men with prostate cancer, with a specificity rate of >90% in PSAscreened cohorts. ETS fusion analysis thus provides a biomarker for prostate cancer, and the test is noninvasive. Combinations of antiandrogens and signal transduction pathway inhibitors, including mTOR inhibitors, offer promise in treating prostate cancer (Schayowitz, et al., 2008).
Targeted mutational analysis of tumors to guide personalized cancer treatment Dias-Santagata, et al. (2010), developed a high-throughput genotyping platform that would allow prospective patient selection for personalized cancer therapy.
144
INVESTIGATING THE HUMAN GENOME
In 2010, they reported using such a platform to test for mutations in 13 different cancer genes. The purpose of the testing was to determine the presence of mutations that confer sensitivity or resistance to specific targeted therapies. McBride, et al. (2010), reported that, in hematological malignancies characterized by the presence of tumor-specific rearrangements, sensitive assays have been developed to assess residual disease burden and recurrence. These investigators noted that next-generation sequencing is efficient in revealing rearrangements in DNA in solid tumors. Solid tumors release DNA into the plasma, thus facilitating tumor DNA analysis. Following the identification of rearrangements in tumors through massively parallel paired end sequencing, McBride, et al., utilized Sanger sequencing to confirm rearrangements and designed PCR assays to subsequently track the presence of DNA with the tumorspecific rearrangements. One case they followed had osteosarcoma and rearrangements that led to the deletion of a segment on chromosome 11.
Systems Biology and Cancer Cancer is often associated with dysregulation of multiple pathways. Kreeger and Lauffenburger (2010) explored how tumor genomic transcriptome and metabolic data might be used in computational modeling to derive actionable understanding. They emphasized analysis of specific pathways that provide an organizing principle for approaches to therapy. They noted that pathways frequently interact. They also stressed the importance of systems biology modeling techniques to further the understanding of the key processes in cancer.
CHAPTER 8 • GENES AND GENOMES IN CANCER
145
A number of investigators have emphasized the importance of distinguishing driver mutations, which have a major impact on tumor characteristics, from passenger mutations.
Clonal heterogeneity in tumors As scientists design therapies based on tumor-specific genomic changes, clonal heterogeneity within tumors must be taken into account. Anderson, et al. (2011), examined cells from patients with childhood acute lymphoblastic leukemia. The cases chosen were positive for the ETV6–RUNX1 translocation detected by fluorescence in situ hybridization (FISH). Cells varied, in that a significant percentage of cells (between 15% and 80% in different cases) had deletion of the untranslocated ETV6 gene. Of these cells, a significant number also showed deletion of the PAX5 gene or deletion of both the PAX5 and CDKN2A genes. Anderson, et al., reported that distinctive additional copy number changes were present in subclones within the tumors. The simplest architectural variation involved two or three genetically distinct subclones. Most cases had more distinctly different subclones, and in many, ten subclones types were present.
Bioenergetic regulation as a target in cancer therapy The end product of glucose metabolism in the absence of oxygen, (anaerobic metabolism) is primarily lactate. In the presence of oxygen, glucose is metabolized through pyruvate to acetyl CoA, which then enters the mitochondrial citric acid cycle. Pyruvate in mitochondria undergoes oxidation through activity of pyruvate dehydrogenase.
146
INVESTIGATING THE HUMAN GENOME
The Warburg effect that is typical of cancer cells involves the generation of high levels of lactate and limited oxidation of pyruvate. McFate, et al. (2008), demonstrated that inhibition of the activity of the pyruvate dehydrogenase complex contributes to the Warburg effect in squamous carcinomas of the head and neck. They specifically demonstrated that, in these cancers, increased expression of pyruvate dehydrogenase kinase PDK1 leads to inhibitory phosphorylation of the PDH1alpha subunit encoded by the PDHE1alpha gene and decreased PDH complex activity. They used short hairpin inhibitory RNAs (shRNA1) to knock down PDK1 expression and determined that this restored pyruvate dehydrogenase complex activity and decreased tumor growth and invasiveness. McFate, et al. (2008), suggested that high PDK1 expression promoted the activation of hypoxia inducible factor 1 alpha (HIF1alpha) and that this led to tumor progression.
Chemotherapeutic approaches designed to target aberrant metabolism in tumors Tennant, et al. (2010), reviewed specific metabolic targets for cancer therapy. They emphasized that specific changes in glycolysis and glutaminolysis may serve as targets. Upstream regulators of aberrant metabolism in tumors include hypoxia inducible factor (HIF), phosphatidyl inositol kinase PI3K, AKT, mTOR, and adenosine monophosphate activated protein kinase (AMPK). They noted that, in cancer, insulin and insulin-like growth factor signaling to PI3K and AKT is often activated. Furthermore, the increased glycolysis in tumors is often associated with altered expression or localization of the glycolytic enzymes hexokinase, 6-phosphofructokinase, and fructose 2-6-biphosphatase. Tennant, et al., emphasized that the Warburg effect observed in tumors is characterized by increased activity of pyruvate dehydrogenase kinase 1 (PDK1), which leads to the inhibition of PDH. This
CHAPTER 8 • GENES AND GENOMES IN CANCER
147
then leads to decreased entry of pyruvate into the tricarboxylic acid cycle and increased production of lactate from pyruvate, due to an increased activity of lactate dehydrogenase. They noted that inhibition of PDK1 with dichloroacetate or knockdown of PDK1 or LDH with inhibitory RNAs reduce tumor growth. Tumors produce high quantities of lactate, and this must be transported out of the cell to avoid harmful lowering of the pH. Carbonic anhydrase inhibition prevents the correction of acidosis. Indisulam is an inhibitor of enzymes in the carbonic anhydrase family and is in Phase II trials for cancer therapy. Tumors require high levels of amino acids. Tennant, et al., noted that glutamine is particularly important for tumor survival and that glutaminolytic agents are being considered for cancer treatment. Asparaginase has been used for a number of years to treat childhood leukemia, particularly in cases where the tumor cells have low levels of asparagine synthetase. In those tumors, asparagine is an essential amino acid. Asparaginase is useful as treatment because it converts asparagine to aspartate and ammonia. Arginine deaminase is also being considered in cancer treatment. Arginine is an essential factor in the growth of certain tumors, including hepatocellular carcinomas and melanomas. In normal cells, arginine is not usually an essential aminoacid. Components that inhibit the conversion of citrate to acetyl CoA through the activity of ATP citrate lyase are also being investigated as antitumor agents. Tennant, et al., noted that inhibitors of the enzyme fatty acid synthase are also being investigated as anticancer agents because many tumors express high levels of these enzymes. Unusual splice forms of PKM2 (pyruvate kinase, muscle form 2) occur in some tumors. Metabolic enzymes that are mutated in tumors include succinate dehydrogenase (SH) mitochondrial complex 2, fumarate hydratase (FH), and isocitrate dehydrogenases (IDH1, IDH2).
148
INVESTIGATING THE HUMAN GENOME
Isocitrate dehydrogenases IDH1 and IDH2 Isocitrate undergoes initial dehydrogenation catalyzed by isocitrate dehydrogenase to form oxalosuccinate. This remains enzyme bound and then undergoes decarboxylation to form alpha ketoglutarate. IDH1 and IDH2 use NADP as coenzyme and are located in the cytosol and in mitochondria. IDH3 uses NAD as coenzyme and is located only in mitochondria. Evidence indicates that IDH activity plays key roles in cellular metabolism, glucose sensing, and the antioxidant response. Mutations in arginine residues in IDH1 and IDH2 occur in a number of different brain tumors and are found in more than 50% of diffuse glioblastomas, anaplastic oligodendrogliomas, oligoastrocytomas, and anaplastic astrocytomas (Reitman and Yan, 2010). IDH1 and IDH2 mutation occur in 2%–6% of cases of acute myeloid leukemia, and 96% of cases that are positive for IDH mutations do not have cytogenetic abnormalities (Patel, et al., 2011). The exact mechanisms through which IDH mutations lead to these specific malignancies are not known. Reitman and Yan (2010) reported that a key result of mutations is that IDH activity in the tumors is reduced. A consequence of this is that glucose sensing is altered partly because NADPH generation is decreased. This then signals tumor cells to increase nutrient consumption and block differentiation. They noted that reports indicate that the mutant enzymes develop a new function and act to reduce alpha ketoglutarate to D-2hydroxyglutarate and in this reaction NADPH is reduced to NADP. Evidence also suggests that cells with IDH mutations have increased stability of hypoxia inducible factor HIF alpha and that this likely results from the inhibition of prolyl hydroxylase activity. Reitman and Yan postulated that a clear understanding of the roles of mutant IDH in cancer could open the way for new target therapies.
CHAPTER 8 • GENES AND GENOMES IN CANCER
149
Genome instability, chromosome instability, and cancer Robust DNA damage surveillance and cell cycle checkpoint mechanisms are essential to avoid genomic instability and cancer. Key mechanisms involved in protection against genomic instability and cancer include nucleotide excision repair, which is especially important in repair of UV damage and involves the activity of the XP gene that is deficient in Xeroderma pigmentosum. Base excision repair is also important; impaired function of the MUTYH (mut homolog) gene leads to defects in this repair mechanism and is associated with colon cancer. Double-stranded DNA breaks are primarily repaired by homologous recombination. Products of ATM (ataxia telangiectasia mutated), NBS1 (nibrin), BRCA1, BRCA2, CHEK1, and CHEK2 (cell cycle checkpoint control) are important in this process. (See Chapter 4, “Gene–Environment Interactions,” for additional descriptions.) Double-stranded breaks activate ATM. Upon activation, ATM transitions from an inactive dimer to an active monomer that phosphorylates a number of DNA repair products, including BRCA1, Nibrin, Rad50, and MRE11 (DNA recombination and repair protein). Phosphorylation of CHEK2 leads to the activation of expression of a number of genes, including the P53 gene, that block cell cycle progression and stimulate apoptotic response. Instability at the chromosome level leads to segmental duplications, translocation, and inversions that are common in cancer. Key mechanisms in generating these changes include nonhomologous end joining following DNA breaks. Growing evidence indicates that telomeric dysfunction plays an important role in nonhomologous end joining.
150
INVESTIGATING THE HUMAN GENOME
Therapeutic agents have been developed that impact proteins and enzymes involved in DNA repair. Examples include small molecule inhibitors of ATM and checkpoint kinases CHK1 and CHK2. A competitive inhibitor of ATP binding to ATM has been found to increase the sensitivity of tumors to radiation and chemotherapy (Bolderson, et al., 2009). Inhibitors of CHK2 and CHK2 have also been shown to increase sensitivity to other antitumor treatments. Bolderson, et al., noted that these inhibitors impact normal tissue. Further molecular manipulations of these inhibitors are required to decrease their impact on normal tissues.
Telomeres and Cancer The repeat sequence DNA element at the end of telomeres (TTAGGG)n binds specific proteins that protect telomere ends. Telomere shortening or erosion commonly occurs in aging cells and in hyperproliferating cells. In aging tissue, telomere shortening frequently leads to apoptosis. In normal tissue, telomere shortening activates the senescence program that involves activation of the P53 gene and the RB (retinoblastoma) gene. In the hyperproliferation that occurs in cancer, telomeres are, in part, continually restored by increased telomerase expression. Telomerase is a ribonucleoprotein enzyme composed of an RNA subunit TERC (telomerase RNA component) and a protein subunit TERT (telomerase reverse transcriptase). TERC acts as a template for the synthesis of telomere repeats. Proteins that bind to TERC include dyskerin and a WD40 repeat protein TCAB1. TERT constitutes the catalytic component of telomerase. Clinical trials of specific inhibitors of telomerase are ongoing. Preclinical studies of the compound imetelstat indicate that it slows the proliferation of glioblastoma cells (Marian, et al., 2010).
CHAPTER 8 • GENES AND GENOMES IN CANCER
151
Genetic evolution of pancreatic cancer When first diagnosed, pancreatic cancer in most patients is already metastatic. Yachida, et al. (2010), carried out sequence analysis to evaluate the clonal relationships between primary and metastatic cancers. They concluded that the genetic heterogeneity present in metastatic tumors also exists in clones present in the primary tumor. Pancreatic cancer typically metastasizes to the peritoneum, liver, and lung. The survey of exome sequencing in the primary tumor from an individual revealed 426 somatic mutations in 388 different genes. Yachida, et al., determined that the majority of these mutations were also present in the metastases, and clones with these mutations were defined as parental clones. Mutations that were more abundant in metastases than in primary tumors were defined as progressive mutations. Yachida, et al., noted that, in parental clones, mutations occurred in genes known to be drivers of pancreatic cancer, such as KRAS, P53, and SMAD4. They noted further that homozygous mutations frequently occurred in the tumor suppressor genes SMAD4 and CDKN2A and that these mutations were often associated with chromosome instability. They could not identify specific genomic signatures of metastases. Modeling of mutations and clonal evolution led them to conclude that, on average, six to eight years elapsed between the initial tumorcausing mutation in the pancreas and metastatic seeding. Yachida, et al., concluded that developing tests for early detection of tumors or abnormal gene transcripts would therefore be valuable. Campbell, et al. (2010), undertook the characterization of genomic rearrangements in primary pancreatic tumors and metastases in 13 patients. They determined that telomere dysfunction and abnormal cell cycle control play important roles in the origin and evolution of these tumors. They noted further that there is ongoing
152
INVESTIGATING THE HUMAN GENOME
genomic instability in the metastases and that additional driver mutations may be required for metastatic seeding. These investigators reported considerable variation among patients in the degree of genomic instability and in the specific types of genomic rearrangements. A specific form of rearrangement referred to as foldback inversion accounted for one-sixth of the rearrangements. Campbell, et al., noted that breakage, fusion, and bridge processes lead to foldback inversions and that telomere loss often initiates these processes. They cited evidence that telomerase expression is low early in the process of pancreatic carcinogenesis and is higher later in the process. Their studies also revealed evidence of cell cycle defects. Their findings indicate that rearrangements that led to an increased copy number of mutant KRAS were present in early metastases and apparently drove expansion. Campbell, et al., determined that specific rearrangements were more common in metastatic lesions in a specific organ; for example, lung metastases had amplification of MYC oncogene and CCNE (cyclin e). In one patient, they determined that deletions of exon 6 in the PARK2 ubiquitin ligase encoding gene were present in lung metastases but were absent from abdominal metastases. They proposed that specific rearrangements facilitate overcoming barriers to the colonization of tumor cells in a specific organ. Key concepts of cancer therapy that have emerged from gene and genome analysis of tumors are, first, that physicians and researchers should design the best treatment for a specific patient instead of the best treatment for an average patient. Second, targeted therapy based on a single specific genetic finding is often effective initially but is frequently followed by tumor resistance. Combinations of therapies are often more useful in the long term.
9 Functional genomics: personalized medicine and therapeutics [K]nowledge acquisition and dissemination are increasingly viewed as a collective effort in which disciplinary boundaries are dissolved and ideas and data are more freely and openly exchanged. —L. Hood, L. Rowan, D. J. Galas, J. D. Aitchison (2008) Discoveries in genomics and genetics must be systematically translated to improve patient care. Ginsburg and Willard (2009) defined genomic medicine as “the use of information of genomes and their derivatives, RNA, protein, and metabolites, to guide medical decision-making.” They defined personalized medicine as “healthcare that is informed by each person’s unique clinical, genetic, genomic, and environmental information.” This chapter presents examples of the application of genomic medicine in clinical practice. The relevance of personal genetic and genomic information in designing clinical trials and applying therapies is also discussed. The chapter closes with an example of the development of patient-specific model systems of disease through the use of induced pluripotent stem cells.
153
154
INVESTIGATING THE HUMAN GENOME
Application of genomics to clinical genetics Key examples include studies designed to establish the gene defect responsible for a specific inherited disease in individuals in population isolates. Strauss and Puffenberger (2009) carried out studies in the Amish population. In initial studies aimed at mapping disease genes, they analyzed 300–400 polymorphic microsatellite repeat markers and derived genotypes of affected individuals and their family members. Subsequently, they carried out large-scale genotyping using SNP microarrays. They used SNP data and identification of homozygous blocks to narrow genomic regions likely to harbor candidate disease genes. They compared homozygous blocks among different individuals affected with the same inherited disease. Strauss and Puffenberger (2009) emphasized that they continually shaped laboratory studies based on patient needs and that research and clinical care were inseparable. Candidate gene identification was followed by mutation analysis, and this led the way to carrier detection and early diagnosis in affected patients. Among their innovations and achievements were the capability to detect glutaric aciduria through studies on amniotic fluid and identification of maple syrup urine disease on the first day of life. These authors noted that 12% of the genetic disorders in the Old Order Amish populations are detected in the state newborn screening program. In this population, 40% of the genetic disorders cannot be diagnosed by routine clinical methods and symptoms are nonspecific. They commented that, in their practice in the Amish community, evaluation of a new patient was as likely to include using SNP analysis and gene sequencing as defining an electrolyte profile. Strauss and Puffenberger noted that genetic diagnosis followed by knowledge of pathophysiology has fueled progress in treating a number of genetic diseases. Examples presented include studies in
CHAPTER 9 • FUNCTIONAL GENOMICS
155
their patients that led to an improved understanding of the pathophysiology of maple syrup urine disease, improved care, and improved long-term outcome. They noted that key issues include achieving genotyping and sequencing results within a time frame that is clinically relevant and at a manageable cost. They reported that tissue transplantation with hematopoietic stem cells remains the most effective treatment for a number of genetic disorders. This can be facilitated by rapid donor identification by genotyping. In the Amish population, immune cell deficiencies and severe combined immune deficiency (SCID) occur as a result of at least six different gene mutations. Strauss and Puffenberger report that SCID can be rapidly improved by early umbilical stem cell transplantation. Matched donors can be identified from among siblings using SNP genotyping of HLA and MHC loci. This typing is less costly and can be more rapidly performed than traditional MHC typing. Strauss, et al. (2008), reported an example of applying their methods to disease gene characterization and follow-up. An Amish patient with SCID was found to share blocks of homozygosity on several chromosomes with unaffected siblings. One homozygosity block was found to be uniquely present in the patients. This block on chromosome 11 was 24.7 megabases in length. Strauss, et al., noted that among the 266 genes that mapped within the block were two potentially related to SCID, RAG1 and RAG2 (recombination activating genes). Through sequencing of patient DNA, they identified two nucleotide changes within RAG1; one change likely represented a rare polymorphism, and the other was considered to be pathogenic because it led to an amino acid substitution in the core of the protein. This substitution was found in heterozygous form in the parents. One unaffected sibling was found to have the same HLA MHC haplotype as the affected patient and donated bone marrow to the affected sibling.
156
INVESTIGATING THE HUMAN GENOME
Therapeutic design after gene cloning It is important to emphasize that identifying gene and mutations responsible for a specific disease does not lead directly to therapy. Therapy development usually requires analyzing the downstream effects of the mutation on cells and tissues. Identifying the specific gene that, when impacted by structural rearrangement or mutation, leads to disease opens up possibilities for developing models of the disease in lower organisms, thus facilitating analyses of the downstream effects of the disease-causing change. Availability of such model organisms or of appropriate cell cultures from patients facilitates the evaluation of therapeutic effects. Studies on Huntington’s disease (HD) serve as an interesting example. The disease gene defect, an abnormal glutamate expansion in the protein designated as huntingtin, was identified in 1983. However, currently (2011) the precise function of the gene and the precise mechanism of disease pathogenesis are still unknown. In HD, there is widespread loss of cortical cells and marked degeneration of the corpus striatum. In addition, energy metabolism is severely impacted and evidence indicates that mitochondrial function is impaired. Evidence suggests that the abnormal gene product in HD interacts with the mitochondrial electron transport complex. Gohil, et al. (2011), postulated that, in metabolic situations when there is increased reliance on mitochondrial electron transport activity for energy production, the toxicity of the mutant HD protein may be exacerbated. This then leads to increased mitochondrial function impairment, caspase activation, and apoptosis. Gohil, et al., demonstrated in models of glutamate repeat expansion that the drug meclizine subtly impacts mitochondrial function. It diminishes the production of reactive oxygen species, decreases release of cytochrome c, and suppresses caspases 3 and 7. Meclizine treatment was shown to protect against neuronal damage in model organisms. The mechanism of action may involve suppressing mitochondrial energy metabolism and increasing reliance on glucose for
CHAPTER 9 • FUNCTIONAL GENOMICS
157
energy production. Meclizine is an over-the-counter drug used to treat nausea and travel sickness. It crosses the bloodbrain barrier. The main side effect is a tendency to drowsiness. The question that arises is whether a small molecule can be designed that is similar to meclizine but without the side effects.
Exploring the mechanisms of therapeutic efficacy through functional genomics In polycystic kidney disease (PKD), fluid-filled cysts arise from the lining of renal tubules. As the cysts enlarge, they cause increasing damage to kidney tissue. The autosomal dominantly inherited forms of PKD are caused by mutations in the genes that encode polycystin 1 and polycystin 2. Patients present with flank pain, frequent urination, frequent urinary tract infections, and kidney stones. They may subsequently develop renal failure. Cysts may also occur in other organs, such as the liver and pancreas. Natoli, et al. (2010), identified a potential therapeutic target in PKD. They determined that, in kidney tissue from PKD patients, the levels of the glycosphingolipid glucosylceramide (GlcCer) and the levels of ganglioside GM3 were elevated. Levels of these compounds were also elevated in different mouse models of PKD. Treatment of affected mice with a specific inhibitor of GlcCer synthesis led to improved kidney function, decreased kidney weight relative to body weight, and decreased histological abnormalities in kidneys. Of interest is the fact the compound used inhibited glucosylceramide synthesis and the AKT mTOR pathway. It is possible that the increased levels of GlcCer are associated with the disease instead of the cause of the disease. The compound that leads to improvement may act primarily through its effect on the AKT mTOR pathway.
158
INVESTIGATING THE HUMAN GENOME
Genomic studies in pharmacologic response Lithium has been used in treating bipolar disease for more than 50 years. However, the precise molecular basis through which it induces positive effects is not known. It is also sometimes used to treat unipolar depression. In a review of lithium treatment, Cruceanu, et al. (2009), noted that 30% of patients with bipolar disorder show full response and 60% show partial response. The therapeutic range is very narrow, and frequent monitoring of lithium levels is required to avoid toxicity. Evidence indicates that genetic factors impact lithium efficacy. Cruceanu, et al., reported that the response to lithium serves as a measure to subtype bipolar disease patients. In lithium responders, there is typically an absence of other psychiatric comorbidities and also an absence of rapid cycling. They noted that lithium responders frequently have a family history of bipolar disease. Studies by Boer, et al. (2007), shed light on pathways and mechanisms impacted by lithium. They reported that lithium regulates the activity of CREB (cyclic AMP response binding protein). Boer, et al. (2008), reported that mice exposed to chronic stress had increased CREB expression. Lithium treatment of these mice led to decreased stress-induced levels of CREB. Chronic lithium treatment also downregulated the expression of glycogen synthase kinase 3 (GSK3) and its signaling cascade. Lithium has also been shown to impact the serotonin pathway through its effects on tryptophan hydroxylase, the ratelimiting enzyme in serotonin synthesis, and on serotonin transporter, SLC6A4.
Persistence of the expression of fetal hemoglobin analysis and therapeutic relevance Studies on the persistence of expression of fetal hemoglobin have provided insight into the developmental regulation of globin gene
CHAPTER 9 • FUNCTIONAL GENOMICS
159
expression. They have also revealed possible approaches to treating hemoglobinopathies that involve beta globin synthesis. Studies began with observations on patients with hereditary persistence of fetal hemoglobin expression. Family studies and analysis of segregation of this trait within families were followed by linkage analysis using polymorphic DNA markers. Following the identification of a genomic region that segregated with this trait, candidate loci were selected for sequence analysis. Functional studies of candidate loci were carried out in cell culture systems and in animal models. Subsequently, the impact of the loss of expression of specific candidate loci was investigated through use of short inhibitory RNAs (siRNAs) (Thein, et al., 2009). Fetal hemoglobin (HbF) consists of two alpha globin chains and two gamma globin chains. It is expressed primarily during fetal life. Persistent expression of HbF at levels greater than 1%–2% of total hemoglobin may indicate genetically determined hemoglobinopathies. However, in some cases, persistence of expression of HbF expression appears to be a primary genetic trait not associated with hemoglobin mutations. Family studies in an Asian-Indian kindred revealed that, in some members of the family, beta-thalassemia was mild and was associated with increased expression of HbF. Linkage analyses revealed that the locus determining this increased HbF expression mapped to chromosome 6q23. This locus was designated MYB transcriptional activator, and it segregated independently from the beta globin locus that carried the thalassemia mutation. Resequencing of the gamma globin 2 locus revealed the presence of a polymorphism in this locus that impacted expression. This locus HBG2 XMN1 was found to be a quantitative trait locus that impacted HbF levels in several different populations (Garner, et al., 2000). Studies on hemoglobin F expression on 5,184 individuals led to the identification of 179 individuals with extremely high levels of HbF, above the 95th percentile and individuals with levels below the
160
INVESTIGATING THE HUMAN GENOME
5th percentile. Genome-wide association studies with SNP markers identified a new locus that influenced HbF concentrations. This locus was in intron of the BCL11A oncogene on chromosome 2p16. These studies also detected the association of the chromosome 6q23 locus and HbF concentration (Thein, et al., 2009). Subsequent linkage studies on a Maltese family with hereditary persistence of HbF led to the identification of a linked locus on chromosome 19p13.12–p13.13 (Borg et al., 2010). These investigators determined that individuals in the family who had high levels of HbF had a consistent haplotype in 19p13.12–p13.13. Recombination events narrowed the linked interval to 663kb. Borg, et al., noted that the KLF1 gene mapped in this interval. KLF1 encodes a key erythroid transcriptional regulator. Sequence analyses revealed two sites with changes in the affected individuals. One site was a neutral substitution. The other substitution, K288X, was predicted to alter the zinc finger domain within KLF1. This domain is responsible for DNA binding of KLF1. Borg, et al., determined that this mutation was not present in 400 control individuals from the Maltese population. Individuals who carried the K288X mutation had mild hypochromic anemia. Levels of gamma globin were elevated. In addition, levels of BCL11A were down-regulated. Borg, et al., determined that, in individuals with the mutation, KLF1 concentrations were very low. Borg, et al. (2010), studied fetal erythroid precursor cells and demonstrated that KLF1 binds to the BCL11A promoter. In control individuals, KLF1 leads to decreased BCL11A expression and to increased expression of gamma globin. These investigators raised the possibility that therapeutic inhibition of KLF1 may be useful in raising HbF levels in individuals with beta globin hemoglobinopathies. Zhou, et al. (2010), carried out studies in mice. They determined that BCL11A represses gamma globin gene expression. They also demonstrated that KLF1 regulates BCL11A expression. They proposed that controlled knockdown or pharmacological inhibition of
CHAPTER 9 • FUNCTIONAL GENOMICS
161
KLF1 could provide a strategy to increase fetal hemoglobin concentrations in individuals with sickle-cell hemoglobin disease and in individuals with beta-thalassemia. Taken together, these studies led to the discovery of factors central to persistent fetal hemoglobin expression in children and adults. These discoveries have implications for therapeutic management of hemoglobinopathies such as sickle-cell disease and beta-thalassemia.
Functional genomics, data integration Hood, et al. (2008), emphasized that the exploration of phenotypes requires analyzing dynamic changes in informational molecules, including DNA, RNA, protein, and metabolites, as well as their relationships. They noted that digital genomic information must be integrated with biological and environmental information. Other factors that must be taken into account include dynamic measurement because molecules and networks change their structures and composition in the course of function. Hood, et al., noted further that systems analyses include discovery-driven and hypothesis-driven approaches. Discovery-driven approaches define the elements of the system at a specific information level, such as with information on the genome. Discovery then leads to hypothesis generation. The hypothesis-driven approach includes design and testing of models.
Integrative genomics Evidence now indicates that many gene products function as components of protein complexes. Furthermore, protein complexes play key roles in signaling metabolism and post-translational modifications. Lage, et al. (2007), emphasized that mutations in each the different genes that encode proteins that constitute members of a specific complex may individually render the complex dysfunctional
162
INVESTIGATING THE HUMAN GENOME
and that these different gene mutations may lead to comparable phenotypes. Through text mining of clinical phenotypic descriptions and integration of phenotypic descriptions with data on components of protein complexes, they created phenome-interactome networks. They utilized information on specific phenotypes, protein interaction networks, and linkage intervals to identify genes involved in the generation of specific phenotypes. One example of their studies relates to the discovery of a gene involved in epithelial ovarian cancer. Linkage analysis revealed that this condition was associated with a region on chromosome 3p25-22. They determined that the gene that encodes the Fanconi D2 protein (FANCD2) maps in this interval. This protein forms a complex with BRCA1, BRCA2, FANCD1, and nibrin proteins that play roles in DNA repair and genomic stability.
Integrative genomics and identification of a gene leading to metabolic syndrome Leigh syndrome is characterized by recurrent episodes of metabolic acidosis and coma. Pathological features include subacute neurodegeneration of brain stem and basal ganglia. Levels of cytochrome C oxidase (COX) are reduced. The French-Canadian form of Leigh syndrome was mapped through linkage studies to chromosome 2p16-21. Mootha, et al. (2003), noted that genes that contribute to COX structural subunits or assembly factors do not map this chromosome region. They used an integrative genomics approach to identify the specific gene product that leads to French-Canadian Leigh syndrome. They integrated information on the sequences of all genes in the region, information on mitochondrial expression, and mitochondrial proteomic data. The proteomic data was obtained by purifying mitochondria, extracting mitochondrial proteins, and then analyzing proteins using chromatography and mass spectroscopy. By integrating
CHAPTER 9 • FUNCTIONAL GENOMICS
163
the different forms of data, they were able to identify a single gene that, when mutated, leads to French-Canadian Leigh syndrome. The gene LRPPRC encodes leucine-rich pentatricopeptide repeat containing protein, an mRNA-binding protein that plays a role in mitochondrial mRNA transcript processing and trafficking. Integrative data set analyses are required to identify all genes encoded in the nucleus and in mitochondrial DNA to establish an inventory of genes that may be involved in mitochondrial diseases. Mitochondrially active proteins may be identified through orthology (comparison with other organisms), expression studies, histological studies, knockdown or knockout studies, protein interaction analyses, and DNA sequencing aimed at identifying genes with mitochondrial localization signals (Aiyar, et al., 2008).
Data integration: Application of systems biology to medicine Auffray, et al. (2009), emphasized that the key concept in applying systems biology to medicine is to stratify patients by molecular diagnostics and analysis of disease-perturbed networks. Analytical methods include proteomics through mass spectrometry that analysis peptides and through development in microfluidics that will enable the analysis of proteins. Metabolomic analyses include the application of chromatography, mass spectrometry, and nuclear magnetic resonance imaging of complex metabolites, including lipid and glycosylated molecules. Auffray, et al. noted that computational and mathematical applications are required to identify molecular structures and interactions, gene regulatory networks, and metabolic and signaling networks. They emphasized the importance of quality assurance and the establishment of databases and curation of data. Stated goals in the transition from medical genomics to systems medicine
164
INVESTIGATING THE HUMAN GENOME
are to establish links between genes, biological functions, and disease; the definition of signatures of pathology; the facilitation of clinical studies; and the development of therapies.
P4 medicine: Predictive, preventive, personalized, and participatory medicine Auffray, et al. (2010), considered the HLA complex as a paradigm for P4 medicine. Dausset, Benaceraff, and Snell discovered the genetically determined cell surface molecules that regulate the immune response designated as the HLA complex in humans. They were awarded the Nobel Prize for their work in 1980. Scientists have discovered that unique antigens in this system caused individuals who carried them to have an increased susceptibility to specific diseases. Different antigens within the HLA system were initially determined using antibodies and cellular assays. Prediction of disease susceptibility could then be made on the basis of antigens present in a specific individual. Subsequent cloning and sequencing of gene loci in the HLA complex led to the development of DNA-based typing of antigenic diversity. Auffray, et al., noted that scientists in the HLA community shared reagents and data and established electronic information repositories. They elucidated molecular structure and functional relationship. Therapeutic application enabled by HLA research included organ, tissue, and cell transplantation.
Relevance of personalized medicine to therapeutic design Personalized medicine includes designing therapeutic interventions that take genetic and genomic data from specific patients into account. Examples increasingly point to the importance of genetic
CHAPTER 9 • FUNCTIONAL GENOMICS
165
variation in determining patient response to specific medications. Polymorphisms within and surrounding the interleukin IL28B gene determine the efficacy of interferon alpha and ribovirin in hepatitis C treatment (Thompson, et al., 2010). Documentation of specific genetic variants that impact drug metabolism is also important. For example, evidence suggests that, in Asian individuals, metabolism of the drug Crestor that is used to treat hypercholesterolemia is different than in individuals from other populations. Specific recommendations for dosage must take population differences into account (Singer, 2005). A number of organizations, including the American Society of Oncologists, have recommended that clinical trials include collection and analysis of patient tissues (including tumors) and blood samples. This is important because specific medications may be beneficial in a restricted number of patients who have a particular genomic rearrangement or gene mutation in their tumors. Blood-based proteomic studies are recommended to determine whether there are specific profiles in patients who show a positive response to particular therapies (Singer, 2005).
Induced pluripotent stem cells to model disease and test therapies Growing evidence indicates that cells from readily accessible tissues such as the skin can be induced to pluripotent stem cells (IPS cells) and then to tissue-specific stem cells. Itzhaki, et al. (2011), cultured skin-derived cells from a patient with long Q-T syndrome that leads to cardiac arrhythmias that can be fatal. They reprogrammed these cells to myocardiocytes and demonstrated that these cells had abnormal electrophysiological properties consistent with those in the patient heart. They then used the derived myocardiocytes to test
166
INVESTIGATING THE HUMAN GENOME
medications and to identify agents that reversed the abnormal electrophysiological properties. Other conditions have been successfully modeled in IPS-derived stem cells. These include Fanconi anemia (Rodriguez-Piza, et al., 2010) and spinal muscular atrophy (Marchetto, et al., 2010).
Epilogue The study of disease begins with clinical observation of phenotype, function, and behavior, as well as changes induced by disease or birth defects. Pathology is concerned with studies at the cellular and tissue level and observations of changes that result from a disease process. Through studies in the fields of biochemistry, chemical pathology, and genetics, physicians and scientists aim to understand the dynamic mechanisms that lead to pathologies. These dynamic mechanisms include altered concentrations or structures of specific metabolites and products of gene expression, proteins, and enzymes. There are two main cornerstone concepts of medical science. The first is that understanding aspects of the pathophysiology of a disease can potentially open the way for therapeutic intervention. The second is that, delineating the underlying cause of disease offers possibilities for early diagnosis and prevention. One problem with the study of human diseases in living individuals is that tissues a disease process affects (for example, brain or heart) may not be available for analysis. Another problem is that we may be confronted with the consequences of malfunction—for example, developmental defects—at time points distant from the process that induced the malfunction or malformation. Studies on morphology and chemical pathology often fail to uncover the underlying basis of impaired cognition and aberrant behaviors. The key contributions of our ability to analyze the DNA sequence include the opportunity
167
168
INVESTIGATING THE HUMAN GENOME
to directly analyze genes that when mutated, lead to disease or malformation and that we can carry out analysis in any nucleated cells and in genes and accessible tissues. Availability of information on DNA sequence and genomic architecture has enabled greater understanding of human individual variation, population differences, and evolutionary relationships. The availability of sequence information and expeditious sequence analysis has led to progress in identifying genes responsible for rare genetic defects and for a number of common complex diseases in which genetics plays a role. Analysis of altered DNA sequence and altered genomic architecture in cancer, together with analysis of altered gene expression, has promoted insight into pathogenesis of a number of different cancers. That, in turn, has led to the development of various pharmaceutical agents for treating cancer. Availability of the genomic sequence and information on the architecture and molecular function of the genome have changed the ways we study biology, physiology, and pathology. In some cases, genome studies and studies on altered production of genome products in disease have led to a paradigm shift in therapeutic design. Continued developments in high-throughput DNA sequencing now enable more comprehensive and rapid analysis of sequence in small samples of DNA. With respect to common diseases, sequence variants found in next-generation sequencing can be used in parallel with genome-wide association study results and expression studies. Sequence variants found are more likely to have direct relevance to patient management and treatment development when the downstream effects of the specific variant on function are known. Discoveries of cures are the most highly desired outcome of research. However, we must recognize that accurate diagnosis of disease has utility in instituting appropriate medical management. Correct diagnosis of genetic diseases and knowledge about their manifestations are often empowering to patients and their families as they plan their daily lives and their future.
EPILOGUE
169
Valuable information regarding genotype and phenotype is available in knowledge bases, such as Genetests at the National Center for Biotechnology information (NCBI) and the National Library of Medicine in the United States, and Gene Cards available in Europe through the European Journal of Human Genetics. Information on genetic tests in the latter resource includes data on analytical validity (sensitivity and specificity), clinical validity (informative value in diagnosis and differential diagnosis), and clinical utility (whether testing will impact disease management, lifestyle, and prevention). Increasing evidence indicates the benefits to patients in considering personal genetics and genomics and environmental factors in their diagnoses and treatments. Family history continues to give us important insight into disease susceptibility and risk factors. Specific environmental exposures constitute risk factors for disease. Individual sensitivities to specific chemicals contained in medications must be taken into account in designing therapies. Practice of personalized medicine will benefit patients, and help researchers improve the design of clinical trials, and advance the efficiency and improved economics of the healthcare system. Sequence analysis in individuals with unusual responses to pharmaceuticals, such as increased sensitivity, different dosage requirements, or adverse reactions, have led to the discovery of genes and variants involved and to the development of appropriate tests. The Pharmacogenomics knowledge base (PharmGKB), sponsored by different institutes at the National Institutes of Health (NIH), is a growing and important resource for healthcare professionals for the Food and Drug Administration (FDA) and for the public. However, there is room for humility even in the face of accomplishment. Much remains that we do not yet understand, and growing evidence indicates great complexity and webs of interactions. At this point, modern scientists and physicians should heed the advice from scientist and sage Claude Bernard, who in 1865 wrote:
170
INVESTIGATING THE HUMAN GENOME
The first condition to be fulfilled by men of science, applying themselves to the investigation of natural phenomena, is to maintain absolute freedom of mind, based on philosophic doubt. Yet we must not be sceptical; we must believe in science, i.e., in determinism; we must believe in a complete and necessary relation between things, among the phenomena proper to living things as well as in others; but at the same time we must be thoroughly convinced that we know this relation only in a more or less approximate way and that the theories we hold are far from embodying changeless truths.
References Aiyar, R. S., J. Gagneur, and L. M. Steinmetz. (2008) “Identification of Mitochondrial Disease Genes Through Integrative Analysis of Multiple Datasets.” Methods 46: 248–255. Alderton, G. K., H. Joenje, R. Varon, A. D. Borglum, P. A. Jeggo, and M. O’Driscoll. (2004) “Seckel Syndrome Exhibits Cellular Features Demonstrating Defects in the ATR-Signalling Pathway.” Human Molecular Genetics 13: 3,127–3,138. Allouche, M. (2007) “ALK Is a Novel Dependence Receptor: Potential Implications in Development and Cancer.” Cell Cycle 6: 1,533–1,538. Amir, R. E., I. B. Van den Veyver, M. Wan, C. Q. Tran, U. Francke, and H. Y. Zoghbi. (1999) “Rett Syndrome Is Caused by Mutations in X-Linked MECP2, Encoding Methyl-CpG-Binding Protein 2.” Nature Genetics 23: 185–188. Anderson, K., C. Lutz, F. W. van Delft, C. M. Bateman, Y. Guo, S. M. Colman, H. Kempski, A. V. Moorman, I. Titley, J. Swansbury, et al. (2011) Genetic Variegation of Clonal Architecture and Propagating Cells in Leukaemia.” Nature 469: 356–361. Arai, T., M. Hasegawa, H. Akiyama, K. Ikeda, T. Nonaka, H. Mori, D. Mann, K. Tsuchiya, M. Yoshida, Y. Hashizume, et al. (2006) “TDP-43 Is a Component of Ubiquitin-Positive Tau-Negative Inclusions in Frontotemporal Lobar Degeneration and Amyotrophic Lateral Sclerosis.” Biochemical and Biophysical Research Communications 351: 602–611. Araya, J., and S. L. Nishimura. (2010) “Fibrogenic Reactions in Lung Disease.” Annual Review of Pathology 5: 77–98.
171
172
INVESTIGATING THE HUMAN GENOME
Auffray, C., D. Charron, and L. Hood. (2010) “Predictive, Preventive, Personalized, and Participatory Medicine: Back to the Future.” Genome Medicine 2: 57. Auffray, C., Z. Chen, and L. Hood. (2009) “Systems Medicine: The Future of Medical Genomics and Healthcare.” Genome Medicine 1: 2. Azzedine, H., N. Ravise, C. Verny, A. Gabreels-Festen, M. Lammens, D. Grid, J. M. Vallat, G. Durosier, J. Senderek, S. Nouioua, et al. (2006) “Spine Deformities in Charcot-Marie-Tooth 4C Caused by SH3TC2 Gene Mutations.” Neurology 67: 602–606. Azzu, V., and M. D. Brand, “The On–Off Switches of the Mitochondrial Uncoupling Proteins.” Trends in Biochemical Sciences 35: 298–307. Bartel, D. P. (2009) “MicroRNAs: Target Recognition and Regulatory Functions.” Cell 136: 215–233. Bassell, G. J., and S. T. Warren. (2008) “Fragile X Syndrome: Loss of Local mRNA Regulation Alters Synaptic Development and Function.” Neuron 60: 201–214. Baumer, D., O. Ansorge, M. Almeida, and K. Talbot. (2010) “The Role of RNA Processing in the Pathogenesis of Motor Neuron Degeneration.” Expert Reviews in Molecular Medicine 12: e21. Bear, M. F., K. M. Huber, and S. T. Warren. (2004) “The mGluR Theory of Fragile X Mental Retardation.” Trends in Neurosciences 27: 370–377. Beeson, J. G., and B. S. Crabb. (2007) “Towards a Vaccine Against Plasmodium Vivax Malaria.” PLoS Medicine 4: e350. Bekker-Jensen, S., J. Rendtlew Danielsen, K. Fugger, I. Gromova, A. Nerstedt, C. Lukas, J. Bartek, J. Lukas, and N. Mailand. (2010) “HERC2 Coordinates Ubiquitin-Dependent Assembly of DNA Repair Factors on Damaged Chromosomes.” Nature Cell Biology 12: 80–86, sup. 1–12. Belenky, P., K. L. Bogan, and C. Brenner. (2007) “NAD+ Metabolism in Health and Disease.” Trends in Biochemical Sciences 32: 12–19. Benz, E. J., Jr. (2010) “Learning About Genomics and Disease from the Anucleate Human Red Blood Cell.” Journal of Clinical Investigation 120: 4,204–4,206.
REFERENCES
173
Bernard, C. (1865) An Introduction to the Study of Experimental Medicine. Trans. by Henry Copley Greene. New York: Dover Publications, 1957. Bilguvar, K., A. K. Ozturk, A. Louvi, K. Y. Kwan, M. Choi, B. Tatli, D. Yalnizoglu, B. Tuysuz, A. O. Caglayan, S. Gokben, et al. (2010) “WholeExome Sequencing Identifies Recessive WDR62 Mutations in Severe Brain Malformations.” Nature 467: 207–210. Bird, A. (2007) “Perceptions of Epigenetics.” Nature 447: 396–398. Birney, E., J. A. Stamatoyannopoulos, A. Dutta, R. Guigo, T. R. Gingeras, E. H. Margulies, A. Weng, M. Snyder, E. T. Dermitzakis, R. E. Thurman, et al. (2007) “Identification and Analysis of Functional Elements in 1% of the Human Genome by the ENCODE Pilot Project.” Nature 447: 799–816. Boer, U., J. Eglins, D. Krause, S. Schnell, C. Schofl, and W. Knepel. (2007) “Enhancement by Lithium of cAMP-Induced CRE/CREB-Directed Gene Transcription Conferred by TORC on the CREB Basic Leucine Zipper Domain.” Biochemical Journal 408: 69–77. Boggs, L. N., K. S. Fuson, M. Baez, L. Churgay, D. McClure, G. Becker, and P. C. May. (1996) “Clusterin (Apo J) Protects Against In Vitro AmyloidBeta (1-40) Neurotoxicity.” Journal of Neurochemistry 67: 1,324–1,327. Bolderson, E., D. J. Richard, B. B. Zhou, and K. K. Khanna. (2009) “Recent Advances in Cancer Therapy Targeting Proteins Involved in DNA Double-Strand Break Repair.” Clinical Cancer Research 15: 6,314–6,320. Borg, J., P. Papadopoulos, M. Georgitsi, L. Gutierrez, G. Grech, P. Fanis, M. Phylactides, A. J. Verkerk, P. J. van der Spek, C. A. Scerri, et al. (2010) “Haploinsufficiency for the Erythroid Transcription Factor KLF1 Causes Hereditary Persistence of Fetal Hemoglobin.” Nature Genetics 42: 801–805. Borrelli, E., E. J. Nestler, C. D. Allis, and P. Sassone-Corsi. (2008) “Decoding the Epigenetic Language of Neuronal Plasticity.: Neuron 60: 961–974. Bozdagi, O., T. Sakurai, D. Papapetrou, X. Wang, D. L. Dickstein, N. Takahashi, Y. Kajiwara, M. Yang, A. M. Katz, M. L. Scattoni, et al. (2010) “Haploinsufficiency of the Autism-Associated Shank3 Gene Leads to
174
INVESTIGATING THE HUMAN GENOME
Deficits in Synaptic Function, Social Interaction, and Social Communication.” Molecular Autism 1: 15. Bu, G. (2009) “Apolipoprotein E and Its Receptors in Alzheimer’s Disease: Pathways, Pathogenesis, and Therapy.” Nature Reviews Neuroscience 10: 333–344. Burre, J., M. Sharma, T. Tsetsenis, V. Buchman, M. R. Etherton, and T. C. Sudhof. (2010) “Alpha-Synuclein Promotes SNARE-Complex Assembly In Vivo and In Vitro.” Science 329: 1,663–1,667. Calero, M., A. Rostagno, E. Matsubara, B. Zlokovic, B. Frangione, and J. Ghiso. (2000) “Apolipoprotein J (Clusterin) and Alzheimer’s Disease.” Microscopy Research and Technique 50: 305–315. Campbell, P. J., S. Yachida, L. J. Mudie, P. J. Stephens, E. D. Pleasance, L. A. Stebbings, L. A. Morsberger, C. Latimer, S. McLaren, M. L. Lin, et al. (2010) “The Patterns and Dynamics of Genomic Instability in Metastatic Pancreatic Cancer.: Nature 467: 1,109–1,113. Cantor, R. M., K. Lange, and J. S. Sinsheimer. (2010) “Prioritizing GWAS Results: A Review of Statistical Methods and Recommendations for Their Application.” American Journal of Human Genetics 86: 6–22. Carninci, P. (2010) “RNA Dust: Where Are the Genes?” DNA Research 17: 51–59. Caughey, B., G. S. Baron, B. Chesebro, and M. Jeffrey. (2009) “Getting a Grip on Prions: Oligomers, Amyloids, and Pathological Membrane Interactions.” Annual Review of Biochemistry 78: 177–204. Chahal, H. S., J. P. Chapple, L. A. Frohman, A. B. Grossman, and M. Korbonits. (2010) “Clinical, Genetic, and Molecular Characterization of Patients with Familial Isolated Pituitary Adenomas (FIPA).” Trends in Endocrinology and Metabolism 21: 419–427. Chahal, H. S., K. Stals, M. Unterlander, D. J. Balding, M. G. Thomas, A. V. Kumar, G. M. Besser, A. B. Atkinson, P. J. Morrison, T. A. Howlett, et al. (2011) “AIP Mutation in Pituitary Adenomas in the 18th Century and Today.” New England Journal of Medicine 364: 43–50. Chahrour, M., S. Y. Jung, C. Shaw, X. Zhou, S. T. Wong, J. Qin, and H. Y. Zoghbi. (2008) “MeCP2, a Key Contributor to Neurological Disease, Activates and Represses Transcription.” Science 320: 1,224–1,229.
REFERENCES
175
Chang, S., T. ran Ma, R. D. Miranda, M. E. Balestra, R. W. Mahley, and Y. Huang. (2005) “Lipid- and Receptor-Binding Regions of Apolipoprotein E4 Fragments Act in Concert to Cause Mitochondrial Dysfunction and Neurotoxicity.” Proceedings of the National Academy of Sciences USA 102: 18,694–18,699. Chotivanich, K., R. Udomsangpetch, K. Pattanapanyasat, W. Chierakul, J. Simpson, S. Looareesuwan, and N. White. (2002) “Hemoglobin E: A Balanced Polymorphism Protective Against High Parasitemias and Thus Severe P Falciparum Malaria.” Blood 100: 1,172–1,176. Christie, J. M. and G. L. Westbrook. (2006) “Lateral Excitation Within the Olfactory Bulb.” Journal of Neuroscience 26: 2,269–2,277. Chung, E., Y. Ji, Y. Sun, R. J. Kascsak, R. B. Kascsak, P. D. Mehta, S. M. Strittmatter, and T. Wisniewski. (2010) “Anti-PrPC Monoclonal Antibody Infusion As a Novel Treatment for Cognitive Deficits in an Alzheimer’s Disease Model Mouse.” BMC Neuroscience 11: 130. Cisse, M., and L. Mucke. (2009) “Alzheimer’s Disease: A Prion Protein Connection.” Nature 457: 1,090–1,091. Cooke, S. F., and T. V. Bliss. (2006) “Plasticity in the Human Central Nervous System.” Brain 129: 1,659–1,673. Costa, V., C. Angelini, I. De Feis, and A. Ciccodicola. “Uncovering the Complexity of Transcriptomes with RNA-Seq.” Journal of Biomedicine and Biotechnology 2010: 853,916. Cruceanu, C., M. Alda, and G. Turecki. (2009) “Lithium: A Key to the Genetics of Bipolar Disorder.” Genome Medicine 1: 79. Cuddapah, S., R. Jothi, D. E. Schones, T. Y. Roh, K. Cui, and K. Zhao. (2009) “Global Analysis of the Insulator Binding Protein CTCF in Chromatin Barrier Regions Reveals Demarcation of Active and Repressive Domains.” Genome Research 19: 24–32. Daly A.F., M.A. Tichomirowa, P. Petrossians, E. Heliovaara, M.L Jaffrain Rea, A. Barlier, L.A. Naves, T. Eberling, A Karhu, A Raappana et al. (2010) “Clinical Characteristics and Therapeutic Responses in Patients with Germ-Line AIP Mutations and Pituitary Adenomas: An International Collaborative Study.” Journal of Clinical Endocrinology & Metabolism 95: E373–E383.
176
INVESTIGATING THE HUMAN GENOME
Dauwerse, J. G., J. Dixon, S. Seland, C. A. Ruivenkamp, A. van Haeringen, L. H. Hoefsloot, D. J. Peters, A. C. Boers, C. Daumer-Haas, R. Maiwald, et al (2010). “Mutations in Genes Encoding Subunits of RNA Polymerases I and III Cause Treacher Collins Syndrome.” Nature Genetics 43: 20–22. de Bono, J. S., and A. Ashworth. (2010) “Translating Cancer Research into Targeted Therapeutics.” Nature 467: 543–549. de Oliveira, R. M., T. F. Pais, and T. F. Outeiro. (2010) “Sirtuins: Common Targets in Aging and in Neurodegeneration.” Current Drug Targets 11: 1,270–1,280. De Strooper, B., R. Vassar, and T. Golde. (2010) “The Secretases: Enzymes with Therapeutic Potential in Alzheimer Disease.” Nature Reviews Neurology 6: 99–107. Deng, H. X., H. Zhai, E. H. Bigio, J. Yan, F. Fecto, K. Ajroud, M. Mishra, S. Ajroud-Driss, S. Heller, R. Sufit, et al. (2010) “FUS-Immunoreactive Inclusions Are a Common Feature in Sporadic and Non-SOD1 Familial Amyotrophic Lateral Sclerosis.” Annals of Neurology 67: 739–748. Derheimer, F. A., and M. B. Kastan. (2010) “Multiple Roles of ATM in Monitoring and Maintaining DNA Integrity.” FEBS Letters 584: 3,675–3,681. Dias-Santagata, D., S. Akhavanfard, S. S. David, K. Vernovsky, G. Kuhlmann, S. L. Boisvert, H. Stubbs, U. McDermott, J. Settleman, E. L. Kwak, et al. (2010) “Rapid Targeted Mutational Analysis of Human Tumours: A Clinical Platform to Guide Personalized Cancer Medicine.” EMBO Molecular Medicine 2: 146–158. Dietrich, M. O., and T. L. Horvath. (2010) “The Role of Mitochondrial Uncoupling Proteins in Lifespan.” Pflugers Archives 459: 269–275. Donmez, G., D. Wang, D. E. Cohen, and L. Guarente. (2010) “SIRT1 Suppresses Beta-Amyloid Production by Activating the Alpha-Secretase Gene ADAM10.” Cell 142: 320–332. Duvel, K., J. L. Yecies, S. Menon, P. Raman, A. I. Lipovsky, A. L. Souza, E. Triantafellow, Q. Ma, R. Gorski, S. Cleaver, et al. (2010) “Activation of a Metabolic Gene Regulatory Network Downstream of mTOR Complex 1.” Molecular Cell 39: 171–183.
REFERENCES
177
Echtay, K. S., D. Roussel, J. St-Pierre, M. B. Jekabsons, S. Cadenas, J. A. Stuart, J. A. Harper, S. J. Roebuck, A. Morrison, S. Pickering, et al. (2002) “Superoxide Activates Mitochondrial Uncoupling Proteins.” Nature 415: 96–99. Eckel-Mahan, K., and P. Sassone-Corsi. (2010) “Aging Brains and Waning Clocks on the Process of Habituation.” Aging 2: 320–321. Ehehalt, R., P. Keller, C. Haass, C. Thiele, and K. Simons. (2003) “Amyloidogenic Processing of the Alzheimer Beta-Amyloid Precursor Protein Depends on Lipid Rafts.” Journal of Cell Biology 160: 113–123. Fagan, A. M., D. Head, A. R. Shah, D. Marcus, M. Mintun, J. C. Morris, and D. M. Holtzman. (2009) “Decreased Cerebrospinal Fluid Abeta(42) Correlates with Brain Atrophy in Cognitively Normal Elderly.” Annals of Neurology 65: 176–183. Fazzari, P., A. V. Paternain, M. Valiente, R. Pla, R. Lujan, K. Lloyd, J. Lerma, O. Marin, and B. Rico. (2010) “Control of Cortical GABA Circuitry Development by Nrg1 and ErbB4 Signalling.” Nature 464: 1,376–1,380. Feng, Y., C. A. Gutekunst, D. E. Eberhart, H. Yi, S. T. Warren, and S. M. Hersch. (1997) “Fragile X Mental Retardation Protein: Nucleocytoplasmic Shuttling and Association with Somatodendritic Ribosomes.” Journal of Neuroscience 17: 1,539–1,547. Ferry, G. (2009) “Science’s New Battle Lines.” Nature 459: 34–35. Firestein, R., A. J. Bass, S. Y. Kim, I. F. Dunn, S. J. Silver, I. Guney, E. Freed, A. H. Ligon, N. Vena, S. Ogino, et al. (2008) “CDK8 Is a Colorectal CVancer Oncogene That Regulates Beta-Catenin Activity.” Nature 455: 547–551. Fischer, A., F. Sananbenesi, A. Mungenast, and L. H. Tsai. (2010) “Targeting the Correct HDAC(s) to Treat Cognitive Disorders.” Trends in Pharmacological Sciences 31: 605–617. Flaherty, K. T., I. Puzanov, K. B. Kim, A. Ribas, G. A. McArthur, J. A. Sosman, P. J. O’Dwyer, R. J. Lee, J. F. Grippo, K. Nolop, et al. (2010) “Inhibition of Mutated, Activated BRAF in Metastatic Melanoma.” New England Journal of Medicine 363: 809–819.
178
INVESTIGATING THE HUMAN GENOME
Fong, P. C., D. S. Boss, T. A. Yap, A. Tutt, P. Wu, M. Mergui-Roelvink, P. Mortimer, H. Swaisland, A. Lau, M. J. O’Connor, et al. (2009) “Inhibition of Poly(ADP-ribose) Polymerase in Tumors from BRCA Mutation Carriers.” New England Journal of Medicine 361: 123–134. Forero, D. A., K. van der Ven, P. Callaerts, and J. Del-Favero. (2010) “miRNA Genes and the Brain: Implications for Psychiatric Disorders.” Human Mutation 31: 1,195–1,204. Forlani, G., E. Giarda, U. Ala, F. Di Cunto, M. Salani, R. Tupler, C. Kilstrup-Nielsen, and N. Landsberger. (2010) “The MeCP2/YY1 Interaction Regulates ANT1 Expression at 4q35: Novel Hints for Rett Syndrome Pathogenesis.” Human Molecular Genetics 19: 3,114–3,123. Frost, B., and M. I. Diamond. (2010) “Prion-like Mechanisms in Neurodegenerative Diseases.” Nature Reviews Neuroscience 11: 155–159. Fukata, Y., and M. Fukata. (2010) “Protein Palmitoylation in Neuronal Development and Synaptic Plasticity.” Nature Reviews Neuroscience 11: 161–175. Gabellini, D., M. R. Green, and R. Tupler. (2002) “Inappropriate Gene Activation in FSHD: A Repressor Complex Binds a Chromosomal Repeat Deleted in Dystrophic Muscle.” Cell 110: 339–348. Gallagher, P. G., L. A. Steiner, R. I. Liem, A. N. Owen, A. P. Cline, N. E. Seidel, L. J. Garrett, and D. M. Bodine. (2010) “Mutation of a Barrier Insulator in the Human Ankyrin-1 Gene Is Associated with Hereditary Spherocytosis.” Journal of Clinical Investigation 120: 4,453–4,465. Gao, J., W. Y. Wang, Y. W. Mao, J. Graff, J. S. Guan, L. Pan, G. Mak, D. Kim, S. C. Su, and L. H. Tsai. (2010) “A Novel Pathway Regulates Memory and Plasticity via SIRT1 and miR-134.” Nature 466: 1,105–1,109. Garber, K. (2010) “ALK, Lung Cancer, and Personalized Therapy: Portent of the Future?: Journal of the National Cancer Institute 102: 672–675. Garcia-Arocena, D., and P. J. Hagerman. “Advances in Understanding the Molecular Basis of FXTAS.” (2010) Human Molecular Genetics 19: R83–R89.
REFERENCES
179
Garner, C., T. Tatu, J. E. Reittie, T. Littlewood, J. Darley, S. Cervino, M. Farrall, P. Kelly, T. D. Spector, and S. L. Thein. (2000) “Genetic Influences on F Cells and Other Hematologic Variables: A Twin Heritability Study.” Blood 95: 342–346. Ghaemmaghami, S., B. C. May, A. R. Renslo, and S. B. Prusiner. (2010) “Discovery of 2-aminothiazoles As Potent Antiprion Compounds.” Journal of Virology 84: 3,408–3,412. Gibson, P., Y. Tong, G. Robinson, M. C. Thompson, D. S. Currle, C. Eden, T. A. Kranenburg, T. Hogg, H. Poppleton, J. Martin, et al. (2010) “Subtypes of Medulloblastoma Have Distinct Developmental Origins.” Nature 468: 1,095–1,099. Gibson, J., N. E. Morton, and A. Collins. (2006) “Extended Tracts of Homozygosity in Outbred Human Populations.” Human Molecular Genetics 15: 789–795. Giles, K. E., H. Gowher, R. Ghirlando, C. Jin, and G. Felsenfeld. (2010) “Chromatin Boundaries, Insulators, and Long-Range Interactions in the Nucleus.” Cold Spring Harbor Symposia on Quantitative Biology (in press). Gimbel, D. A., H. B. Nygaard, E. E. Coffey, E. C. Gunther, J. Lauren, Z. A. Gimbel, and S. M. Strittmatter. (2010) “Memory Impairment in Transgenic Alzheimer Mice Requires Cellular Prion Protein.” Journal of Neuroscience 30: 6,367–6,374. Gingeras, T. R. (2007) “Origin of Phenotypes: Genes and Transcripts.” Genome Research 17: 682–690. Ginsburg, G. S., and H. F. Willard. (2009) “Genomic and Personalized Medicine: Foundations and Applications.” Translational Research 154: 277–287. Girirajan, S., and E. E. Eichler. (2010) “Phenotypic Variability and Genetic Susceptibility to Genomic Disorders.” Human Molecular Genetics 19: R176–187. Goedert, M., F. Clavaguera, and M. Tolnay. (2010) “The Propagation of Prion-like Protein Inclusions in Neurodegenerative Diseases.” Trends in Neurosciences 33: 317–325.
180
INVESTIGATING THE HUMAN GENOME
Gohil, V. M., N. Offner, J. A. Walker, S. A. Sheth, E. Fossale, J. F. Gusella, M. E. MacDonald, C. Neri, and V. K. Mootha. (2011) “Meclizine Is Neuroprotective in Models of Huntington’s Disease.” Human Molecular Genetics 20: 294–300. Gouagna, L. C., G. Bancone, F. Yao, B. Yameogo, K. R. Dabire, C. Costantini, J. Simpore, J. B. Ouedraogo, and D. Modiano. (2010) “Genetic Variation in Human HBB Is Associated with Plasmodium Falciparum Transmission.” Nature Genetics 42: 328–331. Green, R. E., J. Krause, A. W. Briggs, T. Maricic, U. Stenzel, M. Kircher, N. Patterson, H. Li, W. Zhai, M. H. Fritz, et al. (2010) “A Draft Sequence of the Neandertal Genome.” Science 328: 710–722. Green, A. J., M. Smith, and J. R. Yates. (1994) “Loss of Heterozygosity on Chromosome 16p13.3 in Hamartomas from Tuberous Sclerosis Patients.” Nature Genetics 6: 193–196. Greer, P. L., R. Hanayama, B. L. Bloodgood, A. R. Mardinly, D. M. Lipton, S. W. Flavell, T. K. Kim, E. C. Griffith, Z. Waldon, R. Maehr, et al. (2010) “The Angelman Syndrome Protein Ube3A Regulates Synapse Development by Ubiquitinating Arc.” Cell 140: 704–716. Gregg, C., J. Zhang, B. Weissbourd, S. Luo, G. P. Schroth, D. Haig, and C. Dulac. (2010) “High-Resolution Analysis of Parent-of-Origin Allelic Expression in the Mouse Brain.” Science 329: 643–648. Griffith, E., S. Walker, C. A. Martin, P. Vagnarelli, T. Stiff, B. Vernay, N. Al Sanna, A. Saggar, B. Hamel, W. C. Earnshaw, et al. (2008) “Mutations in Pericentrin Cause Seckel Syndrome with Defective ATR-Dependent DNA Damage Signaling.” Nature Genetics 40: 232–236. Grosgen, S., M. O. Grimm, P. Friess, and T. Hartmann. (2010) “Role of Amyloid Beta in Lipid Homeostasis.” Biochimica et Biophysica Acta 1,801: 966–974. Gu, Z., T. Nakamura, and S. A. Lipton. (2010) “Redox Reactions Induced by Nitrosative Stress Mediate Protein Misfolding and Mitochondrial Dysfunction in Neurodegenerative Diseases.” Molecular Neurobiology 41: 55–72. Guttman, M., I. Amit, M. Garber, C. French, M. F. Lin, D. Feldser, M. Huarte, O. Zuk, B. W. Carey, J. P. Cassady, et al. (2009) “Chromatin
REFERENCES
181
Signature Reveals over a Thousand Highly Conserved Large Non-coding RNAs in Mammals.” Nature 458: 223–227. Hagerman, P. J., and R. J. Hagerman. (2004) “The Fragile-X Premutation: A Maturing Perspective.” American Journal of Human Genetics 74: 805–816. Hahn, H., C. Wicking, P. G. Zaphiropoulous, M. R. Gailani, S. Shanley, A. Chidambaram, I. Vorechovsky, E. Holmberg, A. B. Unden, S. Gillies, et al. (1996) “Mutations of the Human Homolog of Drosophila Patched in the Nevoid Basal Cell Carcinoma Syndrome.” Cell 85: 841–851. Halfmann, R., and S. Lindquist. (2010) “Epigenetics in the Extreme: Prions and the Inheritance of Environmentally Acquired Traits.” Science 330: 629–632. Halfmann, R., S. Alberti, and S. Lindquist. (2010) “Prions, Protein Homeostasis, and Phenotypic Diversity.” Trends in Cell Biology 20: 125–133. Hallegger, M., M. Llorian, and C. W. Smith. (2010) “Alternative Splicing: Global Insights.” FEBS Journal 277: 856–866. Hameroff, S. (2010) “The ‘Conscious Pilot’: Dendritic Synchrony Moves Through the Brain to Mediate Consciousness.” Journal of Biological Physics 36: 71–93. Hansen, J. C., R. P. Ghosh, and C. L. Woodcock. (2010) “Binding of the Rett Syndrome Protein, MeCP2, to Methylated and Unmethylated DNA and Chromatin.” IUBMB Life 62: 732–738. Hardingham, G. E., and S. A. Lipton. (2010) “Regulation of Neuronal Oxidative and Nitrosative Stress by Endogenous Protective Pathways and by Disease Processes.” Antioxidants and Redox Signaling (in press). Hardy, S., P. E. Jacques, N. Gevry, A. Forest, M. E. Fortin, L. Laflamme, L. Gaudreau, and F. Robert. (2009) “The Euchromatic and Heterochromatic Landscapes Are Shaped by Antagonizing Effects of Transcription on H2A.Z Deposition.” PLoS Genetics 5: e1,000,687. Harold, D., R. Abraham, P. Hollingworth, R. Sims, A. Gerrish, M. L. Hamshere, J. S. Pahwa, V. Moskvina, K. Dowzell, A. Williams, et al. (2009) “Genome-wide Association Study Identifies Variants at CLU and PICALM Associated with Alzheimer’s Disease.” Nature Genetics 41: 1,088–1,093.
182
INVESTIGATING THE HUMAN GENOME
Hasumi, Y., M. Baba, R. Ajima, H. Hasumi, V. A. Valera, M. E. Klein, D. C. Haines, M. J. Merino, S. B. Hong, T. P. Yamaguchi, et al. (2009) “Homozygous Loss of BHD Causes Early Embryonic Lethality and Kidney Tumor Development with Activation of mTORC1 and mTORC2.” Proceedings of the National Academy of Sciences USA 106: 18,722–18,727. Heinrich, M. C., and C. L. Corless. (2010) “Cancer: Oncogenes in Context.” Nature 467: 796–797. Hernandez, D. G., M. A. Nalls, J. R. Gibbs, S. Arepalli, M. van der Brug, S. Chong, M. Moore, D. L. Longo, M. R. Cookson, B. J. Traynor, et al. (2011) “Distinct DNA Methylation Changes Highly Correlated with Chronological Age in the Human Brain.” Human Molecular Genetics 20(6):1164-7 Ho, L., and G. R. Crabtree (2010) “Chromatin Remodelling During Development.” Nature 463: 474–484. Hoeffer, C. A., and E. Klann. “mTOR Signaling: At the Crossroads of Plasticity, Memory, and Disease.” Trends in Neurosciences 33: 67–75. Holtzman, D. M. (2004) “In Vivo Effects of ApoE and Clusterin on Amyloid-beta Metabolism and Neuropathology.” Journal of Molecular Neuroscience 23: 247–254. Hong, S. B., H. Oh, V. A. Valera, J. Stull, D. T. Ngo, M. Baba, M. J. Merino, W. M. Linehan, and L. S. Schmidt. (2010) “Tumor Suppressor FLCN Inhibits Tumorigenesis of a FLCN-Null Renal Cancer Cell Line and Regulates Expression of Key Molecules in TGF-Beta Signaling.” Molecular Cancer 9: 160. Hood, L., L. Rowen, D. J. Galas, and J. D. Aitchison. (2008) “Systems Biology at the Institute for Systems Biology.” Briefings in Functional Genomics Proteomic 7: 239–248. Hooshmand, B., A. Solomon, I. Kareholt, J. Leiviska, M. Rusanen, S. Ahtiluoto, B. Winblad, T. Laatikainen, H. Soininen, and M. Kivipelto. (2010) “Homocysteine and Holotranscobalamin and the Risk of Alzheimer Disease: A Longitudinal Study.” Neurology 75: 1,408–1,414. Houlden, H., M. Laura, L. Ginsberg, H. Jungbluth, S. A. Robb, J. Blake, S. Robinson, R. H. King, and M. M. Reilly. (2009) “The Phenotype of Charcot-Marie-Tooth Disease Type 4C Due to SH3TC2 Mutations and
REFERENCES
183
Possible Predisposition to an Inflammatory Neuropathy.” Neuromuscular Disorders 19: 264–269. Huang, Y. (2010) “Abeta-Independent Roles of Apolipoprotein E4 in the Pathogenesis of Alzheimer’s Disease.” Trends in Molecular Medicine 16: 287–294. Huang, J., and B. D. Manning. (2009) “A Complex Interplay Between Akt, TSC2, and the Two mTOR Complexes.” Biochemical Societies Transactions 37: 217–222. Hudry, E., D. Van Dam, W. Kulik, P. P. De Deyn, F. S. Stet, O. Ahouansou, A. Benraiss, A. Delacourte, P. Bougneres, P. Aubourg, et al. (2010) “AdenoAssociated Virus Gene Therapy with Cholesterol 24-hydroxylase Reduces the Amyloid Pathology Before or After the Onset of Amyloid Plaques in Mouse Models of Alzheimer’s Disease.” Molecular Therapy 18: 44–53. Hussain, N. K., and M. Sheng. (2005) “Neuroscience. Making Synapses: A Balancing Act.” Science 307: 1,207–1,208. Hutt, D. M., E. T. Powers, and W. E. Balch. (2009) “The Proteostasis Boundary in Misfolding Diseases of Membrane Traffic.” FEBS Letters 583: 2,639–2,646. Illingworth, R., A. Kerr, D. Desousa, H. Jorgensen, P. Ellis, J. Stalker, D. Jackson, C. Clee, R. Plumb, J. Rogers, et al. (2008) “A Novel CpG Island Set Identifies Tissue-Specific Methylation at Developmental Gene Loci.” PLoS Biology 6: e22. Illingworth, R. S., and A. P. Bird. (2009) “CpG Islands—‘a Rough Guide.’ FEBS Letters 583: 1,713–1,720. Inoki, K., M. N. Corradetti, and K. L. Guan. (2005) “Dysregulation of the TSC-mTOR Pathway in Human Disease.” Nature Genetics 37: 19–24. Ito, M., M. Hanaoka, Y. Droma, O. Hatayama, E. Sato, Y. Katsuyama, K. Fujimoto, and M. Ota. (2008) “The Association of Transforming Growth Factor Beta 1 Gene Polymorphisms with the Emphysema Phenotype of COPD in Japanese.” Internal Medicine 47: 1,387–1,394. Itzhaki, I., L. Maizels, I. Huber, L. Zwi-Dantsis, O. Caspi, A. Winterstern, O. Feldman, A. Gepstein, G. Arbel, H. Hammerman, et al. (2011) “Modelling the Long QT Syndrome with Induced Pluripotent Stem Cells.” Nature (in press).
184
INVESTIGATING THE HUMAN GENOME
Iwata, N., S. Tsubuki, Y. Takaki, K. Shirotani, B. Lu, N. P. Gerard, C. Gerard, E. Hama, H. J. Lee, and T. C. Saido. (2001) “Metabolic Regulation of Brain Abeta by Neprilysin.” Science 292: 1,550–1,552. Jackson, S. P., and J. Bartek. (2009) “The DNA-Damage Response in Human Biology and Disease.” Nature 461: 1,071–1,078. Jennings, J., M. Georgitsi, I. Holdaway, A. Daly, M. Tichomirowa, L. Aaltonen, A. Karhu, and F. Cameron. (2009) “Aggressive Pituitary Adenomas in Young Patients in a Large Polynesian Kindred with Germline R271W Mutation in the AIP Gene.” European Journal of Endocrinology 161: 799–804. Johannessen, C. M., J. S. Boehm, S. Y. Kim, S. R. Thomas, L. Wardwell, L. A. Johnson, C. M. Emery, N. Stransky, A. P. Cogdill, J. Barretina, et al. (2010) “COT Drives Resistance to RAF Inhibition Through MAP Kinase Pathway Reactivation.” Nature 468: 968–972. Jones, S., T. L. Wang, M. Shih Ie, T. L. Mao, K. Nakayama, R. Roden, R. Glas, D. Slamon, L. A. Diaz Jr., B. Vogelstein, et al. (2010) “Frequent Mutations of Chromatin Remodeling Gene ARID1A in Ovarian Clear Cell Carcinoma.” Science 330: 228–231. Jones, S., X. Zhang, D. W. Parsons, J. C. Lin, R. J. Leary, P. Angenendt, P. Mankoo, H. Carter, H. Kamiyama, A. Jimeno, et al. (2008) “Core Signaling Pathways in Human Pancreatic Cancers Revealed by Global Genomic Analyses.” Science 321: 1,801–1,806. Kahle, P. J., J. Waak, and T. Gasser. (2009) “DJ-1 and Prevention of Oxidative Stress in Parkinson’s Disease and Other Age-Related Disorders.” Free Radical Biology & Medicine 47: 1,354–1,361. Kalay, E., G. Yigit, Y. Aslan, K. E. Brown, E. Pohl, L. S. Bicknell, H. Kayserili, Y. Li, B. Tuysuz, G. Nurnberg, et al. (2010) “CEP152 Is a Genome Maintenance Protein Disrupted in Seckel Syndrome.” Nature Genetics 43: 23–26. Karhu, A., and L. A. Aaltonen. (2007) “Susceptibility to Pituitary Neoplasia Related to MEN-1, CDKN1B, and AIP Mutations: An Update.” Human Molecular Genetics 16 spec. no. 1: R73–R79. Kasuki Jomori de Pinho, L., L. Vieira Neto, L. E. Armondi Wildemberg, E. L. Gasparetto, J. Marcondes, B. de Almeida Nunes, C. M. Takiya, and M.
REFERENCES
185
R. Gadelha. (2010) “Low Aryl Hydrocarbon Receptor-Interacting Protein Expression Is a Better Marker of Invasiveness in Somatotropinomas Than Ki-67 and p53.” Neuroendocrinology (in press). Katayama, F., H. Miura, and S. Takanashi. (2002) “Long-Term Effectiveness and Side Effects of Acetazolamide As an Adjunct to Other Anticonvulsants in the Treatment of Refractory Epilepsies.” Brain and Development 24: 150–154. Kathiresan, S., O. Melander, D. Anevski, C. Guiducci, N. P. Burtt, C. Roos, J. N. Hirschhorn, G. Berglund, B. Hedblad, L. Groop, et al. (2008) “Polymorphisms Associated with Cholesterol and Risk of Cardiovascular Events.” New England Journal of Medicine 358: 1,240–1,249. Katiyar, S., E. Liu, C. A. Knutzen, E. S. Lang, C. R. Lombardo, S. Sankar, J. I. Toth, M. D. Petroski, Z. Ronai, and G. G. Chiang. (2009) “REDD1, an Inhibitor of mTOR Signalling, Is Regulated by the CUL4A-DDB1 Ubiquitin Ligase.” EMBO Reports 10: 866–872. Kee, Y., and A. D. D’Andrea. (2010) “Expanded Roles of the Fanconi Anemia Pathway in Preserving Genomic Stability.” Genes & Development 24: 1,680–1,694. Khatri, S., H. Yepiskoposyan, C. A. Gallo, P. Tandon, and D. R. Plas. (2010) “FOXO3a Regulates Glycolysis Via Transcriptional Control of Tumor Suppressor TSC1.” Journal of Biological Chemistry 285: 15,960–15,965. Kilpinen, H., T. Ylisaukko-oja, K. Rehnstrom, E. Gaal, J. A. Turunen, E. Kempas, L. von Wendt, T. Varilo, and L. Peltonen. (2009) “Linkage and Linkage Disequilibrium Scan for Autism Loci in an Extended Pedigree from Finland.” Human Molecular Genetics 18: 2,912–2,921. Kim, E., and M. Sheng. (2009) “The Postsynaptic Density.” Current Biology 19: R723–R724. Kjolby, M., O. M. Andersen, T. Breiderhoff, A. W. Fjorback, K. M. Pedersen, P. Madsen, P. Jansen, J. Heeren, T. E. Willnow, and A. Nykjaer. (2010) “Sort1, Encoded by the Cardiovascular Risk Locus 1p13.3, Is a Regulator of Hepatic Lipoprotein Export.” Cell Metabolism 12: 213–223. Kool, M., J. Koster, J. Bunt, N. E. Hasselt, A. Lakeman, P. van Sluis, D. Troost, N. S. Meeteren, N. H. Caron, J. Cloos, et al. (2008) “Integrated Genomics Identifies Five Medulloblastoma Subtypes with Distinct Genetic
186
INVESTIGATING THE HUMAN GENOME
Profiles, Pathway Signatures, and Clinicopathological Features.” PLoS ONE 3: e3,088. Korade, Z., and A. K. Kenworthy. (2008) “Lipid Rafts, Cholesterol, and the Brain.” Neuropharmacology 55: 1,265–1,273. Kraytsberg, Y., E. Kudryavtseva, A. C. McKee, C. Geula, N. W. Kowall, and K. Khrapko. (2006) “Mitochondrial DNA Deletions Are Abundant and Cause Functional Impairment in Aged Human Substantia Nigra Neurons.” Nature Genetics 38: 518–520. Kreeger, P. K., Y. Wang, K. M. Haigis, and D. A. Lauffenburger. (2010) “Integration of Multiple Signaling Pathway Activities Resolves K-RAS/NRAS Mutation Paradox in Colon Epithelial Cell Response to Inflammatory Cytokine Stimulation.” Integrative Biology (Cambridge) Integr Biol (Camb) 2: 202–208. Kreeger, P. K., and D. A. Lauffenburger. (2010) “Cancer Systems Biology: A Network Modeling Perspective.” Carcinogenesis 31: 2–8. Kriaucionis, S., and N. Heintz. (2009) “The Nuclear DNA Base 5hydroxymethylcytosine Is Present in Purkinje Neurons and the Brain.” Science 324: 929–930. Kuzumaki, N., D. Ikegami, R. Tamura, T. Sasaki, K. Niikura, M. Narita, K. Miyashita, S. Imai, H. Takeshima, T. Ando, et al. (2010) “Hippocampal Epigenetic Modification at the Doublecortin Gene Is Involved in the Impairment of Neurogenesis with Aging.” Synapse 64: 611–616. Kuzumaki, N., D. Ikegami, R. Tamura, N. Hareyama, S. Imai, M. Narita, K. Torigoe, K. Niikura, H. Takeshima, T. Ando, et al. (2011) “Hippocampal Epigenetic Modification at the Brain-Derived Neurotrophic Factor Gene Induced by an Enriched Environment.” Hippocampus 21(2):127-32 Kwiatkowski, D. P. (2005) “How Malaria Has Affected the Human Genome and What Human Genetics Can Teach Us About Malaria.” American Journal of Human Genetics 77: 171–192. Kwiatkowski, T. J., Jr., D. A. Bosco, A. L. Leclerc, E. Tamrazian, C. R. Vanderburg, C. Russ, A. Davis, J. Gilchrist, E. J. Kasarskis, T. Munsat, et al. (2009) “Mutations in the FUS/TLS Gene on Chromosome 16 Cause Familial Amyotrophic Lateral Sclerosis.” Science 323: 1,205–1,208.
REFERENCES
187
Lage, K., E. O. Karlberg, Z. M. Storling, P. I. Olason, A. G. Pedersen, O. Rigina, A. M. Hinsby, Z. Tumer, F. Pociot, N. Tommerup, et al. (2007) “A Human Phenome-Interactome Network of Protein Complexes Implicated in Genetic Disorders.” Nature Biotechnology 25: 309–316. Lagier-Tourenne, C., M. Polymenidou, and D. W. Cleveland. (2010) “TDP-43 and FUS/TLS: Emerging Roles in RNA Processing and Neurodegeneration.” Human Molecular Genetics 19: R46–R64. Lagier-Tourenne, C., and D. W. Cleveland. (2009) “Rethinking ALS: The FUS About TDP-43.” Cell 136: 1,001–1,004. Lambert, J. C., S. Schraen-Maschke, F. Richard, N. Fievet, O. Rouaud, C. Berr, J. F. Dartigues, C. Tzourio, A. Alperovitch, L. Buee, et al. (2009) “Association of Plasma Amyloid Beta with Risk of Dementia: The Prospective Three-City Study.” Neurology 73: 847–853. Lauren, J., D. A. Gimbel, H. B. Nygaard, J. W. Gilbert, and S. M. Strittmatter. (2009) “Cellular Prion Protein Mediates Impairment of Synaptic Plasticity by Amyloid-beta Oligomers.” Nature 457: 1,128–1,132. Law, M. J., K. M. Lower, H. P. Voon, J. R. Hughes, D. Garrick, V. Viprakasit, M. Mitson, M. De Gobbi, M. Marra, A. Morris, et al. (2010) “ATR-X Syndrome Protein Targets Tandem Repeats and Influences AlleleSpecific Expression in a Size-Dependent Manner.” Cell 143: 367–378. Leoni, V., A. Solomon, and M. Kivipelto. (2010) “Links Between ApoE, Brain Cholesterol Metabolism, Tau, and Amyloid Beta-peptide in Patients with Cognitive Impairment.” Biochemical Society Transactions 38: 1,021–1,025. Li, T., H. Stefansson, E. Gudfinnsson, G. Cai, X. Liu, R. M. Murray, V. Steinthorsdottir, D. Januel, V. G. Gudnadottir, H. Petursson, et al. (2004) “Identification of a Novel Neuregulin 1 At-risk Haplotype in Han Schizophrenia Chinese Patients, but No Association with the Icelandic/Scottish Risk Haplotype.” Molecular Psychiatry 9: 698–704. Li, J. Z., D. M. Absher, H. Tang, A. M. Southwick, A. M. Casto, S. Ramachandran, H. M. Cann, G. S. Barsh, M. Feldman, L. L. CavalliSforza, et al. (2008) “Worldwide Human Relationships Inferred from Genome-wide Patterns of Variation.” Science 319: 1,100–1,104.
188
INVESTIGATING THE HUMAN GENOME
Lin, M. T., and M. F. Beal. (2006) “Mitochondrial Dysfunction and Oxidative Stress in Neurodegenerative Diseases.” Nature 443: 787–795. Linehan, W. M., R. Srinivasan, and L. S. Schmidt. (2010) “The Genetic Basis of Kidney Cancer: A Metabolic Disease.” Nature Reviews Urology 7: 277–285. Linsel-Nitschke, P., N. J. Samani, and H. Schunkert. (2010) “Sorting Out Cholesterol and Coronary Artery Disease.” New England Journal of Medicine 363: 2,462–2,463. Loewer, S., M. N. Cabili, M. Guttman, Y. H. Loh, K. Thomas, I. H. Park, M. Garber, M. Curran, T. Onder, S. Agarwal, et al. (2010) “Large Intergenic Non-coding RNA-RoR Modulates Reprogramming of Human Induced Pluripotent Stem Cells.” Nature Genetics 42: 1,113–1,117. Lord, C. J., and A. Ashworth. (2010) “Biology-Driven Cancer Drug Development: Back to the Future.” BMC Biology 8: 38. Lupo, V., M. I. Galindo, D. Martinez-Rubio, T. Sevilla, J. J. Vilchez, F. Palau, and C. Espinos. (2009) “Missense Mutations in the SH3TC2 Protein Causing Charcot-Marie-Tooth Disease Type 4C Affect Its Localization in the Plasma Membrane and Endocytic Pathway.” Human Molecular Genetics 18: 4,603–4,614. Lupski, J. R., J. G. Reid, C. Gonzaga-Jauregui, D. Rio Deiros, D. C. Chen, L. Nazareth, M. Bainbridge, H. Dinh, C. Jing, D. A. Wheeler, et al. (2010) “Whole-Genome Sequencing in a Patient with Charcot-Marie-Tooth Neuropathy.” New England Journal of Medicine 362: 1,181–1,191. Ma, X. M., and J. Blenis. (2009) “Molecular Mechanisms of mTORMediated Translational Control.” Nature Reviews Molecular Cell Biology 10: 307–318. Mahley, R. W., K. H. Weisgraber, and Y. Huang. (2006) “Apolipoprotein E4: A Causative Factor and Therapeutic Target in Neuropathology, Including Alzheimer’s Disease.” Proceedings of the National Academy of Sciences USA 103: 5,644–5,651. Maier, A. G., M. T. Duraisingh, J. C. Reeder, S. S. Patel, J. W. Kazura, P. A. Zimmerman, and A. F. Cowman. (2003) “Plasmodium Falciparum Erythrocyte Invasion Through Glycophorin C and Selection for Gerbich Negativity in Human Populations.” Nature Medicine 9: 87–92.
REFERENCES
189
Marchetto, M. C., B. Winner, and F. H. Gage. (2010) “Pluripotent Stem Cells in Neurodegenerative and Neurodevelopmental Diseases.” Human Molecular Genetics 19: R71–R76. Marian, C. O., S. K. Cho, B. M. McEllin, E. A. Maher, K. J. Hatanpaa, C. J. Madden, B. E. Mickey, W. E. Wright, J. W. Shay, and R. M. Bachoo. (2010) “The Telomerase Antagonist, Imetelstat, Efficiently Targets Glioblastoma Tumor-Initiating Cells Leading to Decreased Proliferation and Tumor Growth.” Clinical Cancer Research 16: 154–163. Marques-Bonet, T., and E. E. Eichler. (2009) “The Evolution of Human Segmental Duplications and the Core Duplicon Hypothesis.” Cold Spring Harbor Symposia on Quantitative Biology 74: 355–362. Martin-Broto, J., L. Rubio, R. Alemany, and J. A. Lopez-Guerrero. (2010) “Clinical Implications of KIT and PDGFRA Genotyping in GIST.” Clinical and Translational Oncology 12: 670–676. McBride, D. J., A. K. Orpana, C. Sotiriou, H. Joensuu, P. J. Stephens, L. J. Mudie, E. Hamalainen, L. A. Stebbings, L. C. Andersson, A. M. Flanagan, et al. (2010) “Use of Cancer-Specific Genomic Rearrangements to Quantify Disease Burden in Plasma from Patients with Solid Tumors.” Genes, Chromosomes and Cancer 49: 1,062–1,069. McCarthy, M. I. (2010) “Genomics, Type 2 Diabetes, and Obesity.” New England Journal of Medicine 363: 2,339–2,350. McFate, T., A. Mohyeldin, H. Lu, J. Thakar, J. Henriques, N. D. Halim, H. Wu, M. J. Schell, T. M. Tsang, O. Teahan, et al. (2008) “Pyruvate Dehydrogenase Complex Activity Controls Metabolic and Malignant Phenotype in Cancer Cells.” Journal of Biological Chemistry 283: 22,700–22,708. Mercer, T. R., M. E. Dinger, and J. S. Mattick. (2009) “Long Non-coding RNAs: Insights into Functions.” Nature Reviews Genetics 10: 155–159. Millecamps, S., F. Salachas, C. Cazeneuve, P. Gordon, B. Bricka, A. Camuzat, L. Guillot-Noel, O. Russaouen, G. Bruneteau, P. F. Pradat, et al. (2010) “SOD1, ANG, VAPB, TARDBP, and FUS Mutations in Familial Amyotrophic Lateral Sclerosis: Genotype–Phenotype Correlations.” Journal of Medical Genetics 47: 554–560.
190
INVESTIGATING THE HUMAN GENOME
Montgomery, S. B., M. Sammeth, M. Gutierrez-Arcelus, R. P. Lach, C. Ingle, J. Nisbett, R. Guigo, and E. T. Dermitzakis. (2010) “Transcriptome Genetics Using Second Generation Sequencing in a Caucasian Population.” Nature 464: 773–777. Mootha, V. K., P. Lepage, K. Miller, J. Bunkenborg, M. Reich, M. Hjerrild, T. Delmonte, A. Villeneuve, R. Sladek, F. Xu, et al. (2003) “Identification of a gene causing human cytochrome c oxidase deficiency by Integrative Genomics.” Proceedings of the National Academy of Sciences USA 100: 605–610. Mukai, J., A. Dhilla, L. J. Drew, K. L. Stark, L. Cao, A. B. MacDermott, M. Karayiorgou, and J. A. Gogos. (2008) “Palmitoylation-Dependent Neurodevelopmental Deficits in a Mouse Model of 22q11 Microdeletion.” Nature Neuroscience 11: 1,302–1,310. Muller, I., S. Boyle, R. H. Singer, W. A. Bickmore, and J. R. Chubb. (2010) “Stable Morphology, but Dynamic Internal Reorganisation, of Interphase Human Chromosomes in Living Cells.” PLoS One 5: e11,560. Musunuru, K., A. Strong, M. Frank-Kamenetsky, N. E. Lee, T. Ahfeldt, K. V. Sachs, X. Li, H. Li, N. Kuperwasser, V. M. Ruda, et al. (2010) “From Noncoding Variant to Phenotype Via SORT1 at the 1p13 Cholesterol Locus.” Nature 466: 714–719. Naj, A. C., G. W. Beecham, E. R. Martin, P. J. Gallins, E. H. Powell, I. Konidari, P. L. Whitehead, G. Cai, V. Haroutunian, W. K. Scott, et al. (2010) “Dementia Revealed: Novel Chromosome 6 Locus for Late-Onset Alzheimer Disease Provides Genetic Evidence for Folate-Pathway Abnormalities.” PLoS Genetics 6: (9). pii: e1001130. Nakahata, Y., M. Kaluzova, B. Grimaldi, S. Sahar, J. Hirayama, D. Chen, L. P. Guarente, and P. Sassone-Corsi. (2008) “The NAD+–dependent Deacetylase SIRT1 Modulates CLOCK-Mediated Chromatin Remodeling and Circadian Control.” Cell 134: 329–340. Nakahata, Y., S. Sahar, G. Astarita, M. Kaluzova, and P. Sassone-Corsi. (2009) “Circadian Control of the NAD+ Salvage Pathway by CLOCKSIRT1.” Science 324: 654–657. Nan, X., J. Hou, A. Maclean, J. Nasir, M. J. Lafuente, X. Shu, S. Kriaucionis, and A. Bird. (2007) “Interaction Between Chromatin Proteins
REFERENCES
191
MECP2 and ATRX Is Disrupted by Mutations That Cause Inherited Mental Retardation.” Proceedings of the National Academy of Sciences USA 104: 2,709–2,714. Narla, A., and B. L. Ebert. (2010) “Ribosomopathies: Human Disorders of Ribosome Dysfunction.” Blood 115: 3,196–3,205. Natoli, T. A., L. A. Smith, K. A. Rogers, B. Wang, S. Komarnitsky, Y. Budman, A. Belenky, N. O. Bukanov, W. R. Dackowski, H. Husson, et al. (2010) “Inhibition of Glucosylceramide Accumulation Results in Effective Blockade of Polycystic Kidney Disease in Mouse Models.” Nature Medicine 16: 788–792. Neumann, M., D. M. Sampathu, L. K. Kwong, A. C. Truax, M. C. Micsenyi, T. T. Chou, J. Bruce, T. Schuck, M. Grossman, C. M. Clark, et al. (2006) “Ubiquitinated TDP-43 in Frontotemporal Lobar Degeneration and Amyotrophic Lateral Sclerosis.” Science 314: 130–133. Ng, S. B., K. J. Buckingham, C. Lee, A. W. Bigham, H. K. Tabor, K. M. Dent, C. D. Huff, P. T. Shannon, E. W. Jabs, D. A. Nickerson, et al. (2010b) “Exome Sequencing Identifies the Cause of a Mendelian Disorder.” Nature Genetics 42: 30–35. Ng, S. B., D. A. Nickerson, M. J. Bamshad, and J. Shendure. (2010a) “Massively Parallel Sequencing and Rare Disease.” Human Molecular Genetics 19: R1,19–R1,124. Nguyen, A., T. A. Rauch, G. P. Pfeifer, and V. W. Hu. (2011) “Global Methylation Profiling of Lymphoblastoid Cell Lines Reveals Epigenetic Contributions to Autism Spectrum Disorders and a Novel Autism Candidate Gene, RORA, Whose Protein Product Is Reduced in Autistic Brain.” FASEB Journal 24: 3,036–3,051. Nicholas, A. K., M. Khurshid, J. Desir, O. P. Carvalho, J. J. Cox, G. Thornton, R. Kausar, M. Ansar, W. Ahmad, A. Verloes, et al. (2010) “WDR62 Is Associated with the Spindle Pole and Is Mutated in Human Microcephaly.” Nature Genetics 42: 1,010–1,014. Nickerson, M. L., M. B. Warren, J. R. Toro, V. Matrosova, G. Glenn, M. L. Turner, P. Duray, M. Merino, P. Choyke, C. P. Pavlovich, et al. (2002) “Mutations in a Novel Gene Lead to Kidney Tumors, Lung Wall Defects, and Benign Tumors of the Hair Follicle in Patients with the Birt-HoggDube Syndrome.” Cancer Cell 2: 157–164.
192
INVESTIGATING THE HUMAN GENOME
Nicodemus, K. K., A. J. Law, E. Radulescu, A. Luna, B. Kolachana, R. Vakkalanka, D. Rujescu, I. Giegling, R. E. Straub, K. McGee, et al. (2010) “Biological Validation of Increased Schizophrenia Risk with NRG1, ERBB4, and AKT1 Epistasis Via Functional Neuroimaging in Healthy Controls.” Archives of General Psychiatry 67: 991–1,001. Norio, R. (2003) “Finnish Disease Heritage II: Population Prehistory and Genetic Roots of Finns.” Human Genetics 112: 457–469. Pani, A., A. Mandas, and S. Dessi. (2010) “Cholesterol, Alzheimer’s Disease, Prion Disorders: A Ménage à Trois?” Current Drug Targets 11: 1,018–1,031. Pantaleo, M. A., G. Nicoletti, C. Nanni, C. Gnocchi, L. Landuzzi, C. Quarta, S. Boschi, M. Nannini, M. Di Battista, P. Castellucci, et al. (2010) “Preclinical Evaluation of KIT/PDGFRA and mTOR Inhibitors in Gastrointestinal Stromal Tumors Using Small Animal FDG PET.” Journal of Experimental and Clinical Cancer Research 29: 173. Papiol, S., M. Begemann, A. Rosenberger, H. Friedrichs, K. Ribbe, S. Grube, M. H. Schwab, H. Jahn, S. Gunkel, F. Benseler, et al. (2011) “A Phenotype-Based Genetic Association Study Reveals the Contribution of Neuregulin1 Gene Variants to Age of Onset and Positive Symptom Severity in Schizophrenia.” American Journal of Medical Genetics Part B: Neuropsychiatric Genetics (in press). Patel, K. P., F. Ravandi, D. Ma, A. Paladugu, B. A. Barkoh, L. J. Medeiros, and R. Luthra. (2011) “Acute Myeloid Leukemia with IDH1 or IDH2 Mutation: Frequency and Clinicopathologic Features.” American Journal of Clinical Pathology 135: 35–45. Peebles, C. L., J. Yoo, M. T. Thwin, J. J. Palop, J. L. Noebels, and S. Finkbeiner. (2010) “Arc Regulates Spine Morphology and Maintains Network Stability In Vivo.” Proceedings of the National Academy of Sciences USA 107: 18,173–18,178. Peltonen, L., A. Jalanko, and T. Varilo. (1999) “Molecular Genetics of the Finnish Disease Heritage.” Human Molecular Genetics 8: 1,913–1,923. Pennisi, E. (2010) “Shining a Light on the Genome’s ‘Dark Matter.’” Science 330: 1,614.
REFERENCES
193
Pfeiffer, B. E., and K. M. Huber. (2009) “The State of Synapses in Fragile X Syndrome.” Neuroscientist 15: 549–567. Phillips, J. E., and V. G. Corces. (2009) “CTCF: Master Weaver of the Genome.” Cell 137: 1,194–1,211. Pickrell, J. K., J. C. Marioni, A. A. Pai, J. F. Degner, B. E. Engelhardt, E. Nkadori, J. B. Veyrieras, M. Stephens, Y. Gilad, and J. K. Pritchard. (2010) “Understanding Mechanisms Underlying Human Gene Expression Variation with RNA Sequencing.” Nature 464: 768–772. Pleasance, E. D., P. J. Stephens, S. O’Meara, D. J. McBride, A. Meynert, D. Jones, M. L. Lin, D. Beare, K. W. Lau, C. Greenman, et al. (2010) “A Small-Cell Lung Cancer Genome with Complex Signatures of Tobacco Exposure.” Nature 463: 184–190. Pleasance, E. D., R. K. Cheetham, P. J. Stephens, D. J. McBride, S. J. Humphray, C. D. Greenman, I. Varela, M. L. Lin, G. R. Ordonez, G. R. Bignell, et al. (2010) “A Comprehensive Catalogue of Somatic Mutations from a Human Cancer Genome.” Nature 463: 191–196. Polo, J. M., S. Liu, M. E. Figueroa, W. Kulalert, S. Eminli, K. Y. Tan, E. Apostolou, M. Stadtfeld, Y. Li, T. Shioda, et al. (2010) “Cell Type of Origin Influences the Molecular and Functional Properties of Mouse Induced Pluripotent Stem Cells.” Nature Biotechnology 28: 848–855. Rajendran, L., M. Honsho, T. R. Zahn, P. Keller, K. D. Geiger, P. Verkade, and K. Simons. (2006) “Alzheimer’s Disease Beta-amyloid Peptides Are Released in Association with Exosomes.” Proceedings of the National Academy of Sciences USA 103: 11,172–11,177. Rasmussen, M., Y. Li, S. Lindgreen, J. S. Pedersen, A. Albrechtsen, I. Moltke, M. Metspalu, E. Metspalu, T. Kivisild, R. Gupta, et al. (2010) “Ancient Human Genome Sequence of an Extinct Palaeo-Eskimo.” Nature 463: 757–762. Reitman, Z. J., and H. Yan. (2010) “Isocitrate Dehydrogenase 1 and 2 Mutations in Cancer: Alterations at a Crossroads of Cellular Metabolism.” Journal of the National Cancer Institute 102: 932–941. Riccio, A. (2010) “Dynamic Epigenetic Regulation in Neurons: Enzymes, Stimuli, and Signaling Pathways.” Nature Neuroscience 13: 1,330–1,337.
194
INVESTIGATING THE HUMAN GENOME
Roach, J. C., G. Glusman, A. F. Smit, C. D. Huff, R. Hubley, P. T. Shannon, L. Rowen, K. P. Pant, N. Goodman, M. Bamshad, et al. (2010) “Analysis of Genetic Inheritance in a Family Quartet by Whole-Genome Sequencing.” Science 328: 636–639. Roberts, R. C., A. A. Peden, F. Buss, N. A. Bright, M. Latouche, M. M. Reilly, J. Kendrick-Jones, and J. P. Luzio. (2010) “Mistargeting of SH3TC2 Away from the Recycling Endosome Causes Charcot-Marie-Tooth Disease Type 4C.” Human Molecular Genetics 19: 1,009–1,018. Rocchi, M., E. D. Green, N. Archidiacano, et al. (2006) “A Preliminary Comparative Analysis of Primate Segmental Duplications Shows Elevated Substitution Rates and a Great-Ape Expansion of Intrachromosomal Duplications.” Genome Research 16: 576–583. Rochette, J., G. Le Gac, K. Lassoued, C. Ferec, and K. J. Robson. (2010) “Factors Influencing Disease Phenotype and Penetrance in HFE Haemochromatosis.” Human Genetics 128: 233–248. Rodriguez-Piza, I., Y. Richaud-Patin, R. Vassena, F. Gonzalez, M. J. Barrero, A. Veiga, A. Raya, and J. C. Belmonte. (2010) “Reprogramming of Human Fibroblasts to Induced Pluripotent Stem Cells Under Xeno-free Conditions.” Stem Cells 28: 36–44. Rosen, D. R., T. Siddique, D. Patterson, D. A. Figlewicz, P. Sapp, A. Hentati, D. Donaldson, J. Goto, J. P. O’Regan, H. X. Deng, et al. (1993) “Mutations in Cu/Zn Superoxide Dismutase Gene Are Associated with Familial Amyotrophic Lateral Sclerosis.” Nature 362: 59–62. Ross-Inta, C., A. Omanska-Klusek, S. Wong, C. Barrow, D. GarciaArocena, C. Iwahashi, E. Berry-Kravis, R. J. Hagerman, P. J. Hagerman, and C. Giulivi. (2010) “Evidence of Mitochondrial Dysfunction in Fragile X–Associated Tremor/Ataxia Syndrome.” Biochemical Journal 429: 545–552. Rothfuss, O., H. Fischer, T. Hasegawa, M. Maisel, P. Leitner, F. Miesel, M. Sharma, A. Bornemann, D. Berg, T. Gasser, et al. (2009) “Parkin Protects Mitochondrial Genome Integrity and Supports Mitochondrial DNA Repair.” Human Molecular Genetics 18: 3,832–3,850. Rothstein, J. D. (2009) “Current Hypotheses for the Underlying Biology of Amyotrophic Lateral Sclerosis.” Annals of Neurology 65, Suppl. 1: S3–S9.
REFERENCES
195
Rudin, C. M., C. L. Hann, J. Laterra, R. L. Yauch, C. A. Callahan, L. Fu, T. Holcomb, J. Stinson, S. E. Gould, B. Coleman, et al. (2009) “Treatment of Medulloblastoma with Hedgehog Pathway Inhibitor GDC-0449.” New Englan Journal of Medicine 361: 1,173–1,178. Saif, M. W., and E. Chu. (2010) “Biology of Colorectal Cancer.” Cancer Journal 16: 196–201. Samani, N. J., J. Erdmann, A. S. Hall, C. Hengstenberg, M. Mangino, B. Mayer, R. J. Dixon, T. Meitinger, P. Braund, H. E. Wichmann, et al. (2007) “Genomewide Association Analysis of Coronary Artery Disease.” New England Journal of Medicine, 357: 443–453. Schayowitz, A., G. Sabnis, V. C. Njar, and A. M. Brodie. (2008) “Synergistic Effect of a Novel Antiandrogen, VN/124-1, and Signal Transduction Inhibitors in Prostate Cancer Progression to Hormone Independence In Vitro.” Molecular Cancer Therapeutics 7: 121–132. Schmidt, L. S., M. B. Warren, M. L. Nickerson, G. Weirich, V. Matrosova, J. R. Toro, M. L. Turner, P. Duray, M. Merino, S. Hewitt, et al. (2001) “Birt-Hogg-Dube Syndrome, a Genodermatosis Associated with Spontaneous Pneumothorax and Kidney Neoplasia, Maps to Chromosome 17p11.2.” American Journal of Human Genetics 69: 876–882. Schwab, C., T. Arai, M. Hasegawa, H. Akiyama, S. Yu, and P. L. McGeer. (2009) “TDP-43 Pathology in Familial British Dementia.” Acta Neuropathologica 118: 303–311. Sebat, J., B. Lakshmi, D. Malhotra, J. Troge, C. Lese-Martin, T. Walsh, B. Yamrom, S. Yoon, A. Krasnitz, J. Kendall, et al. (2007) “Strong Association of De Novo Copy Number Mutations with Autism.” Science 316: 445–449. Seshadri, S., A. L. Fitzpatrick, M. A. Ikram, A. L. DeStefano, V. Gudnason, M. Boada, J. C. Bis, A. V. Smith, M. M. Carassquillo, J. C. Lambert, et al. (2010) “Genome-wide Analysis of Genetic Loci Associated with Alzheimer Disease.” Journal of the American Medical Association 303: 1,832–1,840. Shaw, A. T., B. Y. Yeap, M. Mino-Kenudson, S. R. Digumarthy, D. B. Costa, R. S. Heist, B. Solomon, H. Stubbs, S. Admane, U. McDermott, et al. (2009) “Clinical Features and Outcome of Patients with Non-Small-Cell Lung Cancer Who Harbor EML4-ALK.” Journal of Clinical Oncology 27: 4,247–4,253.
196
INVESTIGATING THE HUMAN GENOME
She, X., G. Liu, M. Ventura, S. Zhao, D. Misceo, R. Roberto, M.F.Cardone, M. Rocchi M; NISC Comparative Sequencing Program, E.D. Green, N. Archidiacano E.E. Eichler (2006). “A preliminary comparative analysis of primate segmental duplications shows elevated substitution rates and a great-ape expansion of intrachromosomal duplications.” Genome Research. 16(5): 576-83. Singer, E. (2005) “Personalized Medicine Prompts Push to Redesign Clinical Trials.” Nature Medicine 11: 462. Small, E. M., and E. N. Olson. (2010) “Pervasive Roles of MicroRNAs in Cardiovascular Biology.” Nature 469: 336–342. Soda, M., Y. L. Choi, M. Enomoto, S. Takada, Y. Yamashita, S. Ishikawa, S. Fujiwara, H. Watanabe, K. Kurashina, H. Hatanaka, et al. (2007) “Identification of the Transforming EML4-ALK Fusion Gene in NonSmall-Cell Lung Cancer.” Nature 448: 561–566. Stefansson H, V. Steinthorsdottir, T.E. Thorgeirsson, J.R. Gulcher, K. Stefansson (2004). “Neuregulin 1 and schizophrenia.” Ann Med. 36(1): 62-71. Steinlein, O. K., and D. Bertrand. (2010) “Nicotinic Receptor Channelopathies and Epilepsy.” Pflugers Archiv 460: 495–503. Strauss, K. A., E. G. Puffenberger, N. Bunin, N. L. Rider, M. C. Morton, J. T. Eastman III, and D. H. Morton. (2008) “Clinical Application of DNA Microarrays: Molecular Diagnosis and HLA Matching of an Amish Child with Severe Combined Immune Deficiency.” Clinical Immunology 128: 31–38. Strauss, K. A., and E. G. Puffenberger. (2009) “Genetics, Medicine, and the Plain People.” Annual Review of Genomics and Human Genetics 10: 513–536. Sudhof, T. C. (2008) “Neuroligins and Neurexins Link Synaptic Function to Cognitive Disease.” Nature 455: 903–911. Tada, T., and M. Sheng. (2006) “Molecular Mechanisms of Dendritic Spine Morphogenesis.” Current Opinion in Neurobiology 16: 95–101. Takahashi, R. H., M. Tobiume, Y. Sato, T. Sata, G. K. Gouras, and H. Takahashi. (2010) “Accumulation of Cellular Prion Protein Within
REFERENCES
197
Dystrophic Neurites of Amyloid Plaques in the Alzheimer’s Disease Brain.” Neuropathology (in press). Tennant, D. A., R. V. Duran, and E. Gottlieb. (2010) “Targeting Metabolic Transformation for Cancer Therapy.” Nature Reviews Cancer 10: 267–277. Thein, S. L., S. Menzel, M. Lathrop, and C. Garner. (2009) “Control of Fetal Hemoglobin: New Insights Emerging from Genomics and Clinical Implications.” Human Molecular Genetics 18: R216–R223. Thomas, M., J. Lieberman, and A. Lal. (2010) “Desperately Seeking microRNA Targets.” Nature Structural & Molecular Biology 17: 1,169–1,174. Thompson, A. J., A. J. Muir, M. S. Sulkowski, D. Ge, J. Fellay, K. V. Shianna, T. Urban, N. H. Afdhal, I. M. Jacobson, R. Esteban, et al. (2010) “Interleukin-28B Polymorphism Improves Viral Kinetics and Is the Strongest Pretreatment Predictor of Sustained Virologic Response in Genotype 1 Hepatitis C Virus.” Gastroenterology 139: 120–129 e18. Tibbetts, A. S., and D. R. Appling. (2010) “Compartmentalization of Mammalian Folate-Mediated One-Carbon Metabolism.” Annual Review of Nutrition 30: 57–81. Ting, A. K., Y. Chen, L. Wen, D. M. Yin, C. Shen, Y. Tao, X. Liu, W. C. Xiong, and L. Mei. (2011) “Neuregulin 1 Promotes Excitatory Synapse Development and Function in GABAergic Interneurons.” Journal of Neuroscience 31: 15–25. Tishkoff, S. A., F. A. Reed, F. R. Friedlaender, C. Ehret, A. Ranciaro, A. Froment, J. B. Hirbo, A. A. Awomoyi, J. M. Bodo, O. Doumbo, et al. (2009) “The Genetic Structure and History of Africans and African Americans.” Science 324: 1,035–1,044. Tomlins, S. A., D. R. Rhodes, S. Perner, S. M. Dhanasekaran, R. Mehra, X. W. Sun, S. Varambally, X. Cao, J. Tchinda, R. Kuefer, et al. (2005) “Recurrent Fusion of TMPRSS2 and ETS Transcription Factor Genes in Prostate Cancer.” Science 310: 644–648. Tomlins, S. A., A. Bjartell, A. M. Chinnaiyan, G. Jenster, R. K. Nam, M. A. Rubin, and J. A. Schalken. (2009) “ETS Gene Fusions in Prostate Cancer: From Discovery to Daily Clinical Practice.” European Urology 56: 275–286.
198
INVESTIGATING THE HUMAN GENOME
Toro, R., M. Konyukh, R. Delorme, C. Leblond, P. Chaste, F. Fauchereau, M. Coleman, M. Leboyer, C. Gillberg, nd T. Bourgeron. (2010) “Key Role for Gene Dosage and Synaptic Homeostasis in Autism Spectrum Disorders.” Trends in Genetics 26: 363–372. Tufarelli, C., J. A. Stanley, D. Garrick, J. A. Sharpe, H. Ayyub, W. G. Wood, and D. R. Higgs. (2003) “Transcription of Antisense RNA Leading to Gene Silencing and Methylation As a Novel Cause of Human Genetic Disease.” Nature Genetics 34: 157–165. Tupler, R., and D. Gabellini. (2004) “Molecular Basis of Facioscapulohumeral Muscular Dystrophy.” Cellular and Molecular Life Sciences 61: 557–566. Turcotte, S., D. A. Chan, P. D. Sutphin, M. P. Hay, W. A. Denny, and A. J. Giaccia. (2008) “A Molecule Targeting VHL-Deficient Renal Cell Carcinoma That Induces Autophagy.” Cancer Cell 14: 90–102. van Attikum, H., and S. M. Gasser. (2005) “The Histone Code at DNA Breaks: A Guide to Repair?” Nature Reviews Molecular Cell Biology 6: 757–765. Vance, C., B. Rogelj, T. Hortobagyi, D. J. De Vos, A. L. Nishimura, J. Sreedharan, X. Hu, B. Smith, D. Ruddy, P. Wright, et al. (2009) “Mutations in FUS, an RNA Processing Protein, Cause Familial Amyotrophic Lateral Sclerosis Type 6.” Science 323: 1,208–1,211. Verdin, E., M. D. Hirschey, L. W. Finley, and M. C. Haigis. (2010) “M.C. Sirtuin Regulation of Mitochondria: Energy Production, Apoptosis, and Signaling.” Trends in Biochemical Sciences 35: 669–675. Vetrivel, K. S., and G. Thinakaran. (2010) “Membrane Rafts in Alzheimer’s Disease Beta-amyloid Production.” Biochimica et Biophysica Acta 1,801: 860–867. Vierimaa, O., M. Georgitsi, R. Lehtonen, P. Vahteristo, A. Kokko, A. Raitila, K. Tuppurainen, T. M. Ebeling, P. I. Salmela, R. Paschke, et al. (2006) “Pituitary Adenoma Predisposition Caused by Germline Mutations in the AIP Gene.” Science 312: 1,228–1,230. Von Hoff, D. D., P. M. LoRusso, C. M. Rudin, J. C. Reddy, R. L. Yauch, R. Tibes, G. J. Weiss, M. J. Borad, C. L. Hann, J. R. Brahmer, et al. (2009)
REFERENCES
199
“Inhibition of the Hedgehog Pathway in Advanced Basal-Cell Carcinoma.” New England Journal of Medicine 361: 1,164–1,172. Wacheck, V. (2010) “mTOR Pathway Inhibitors in Cancer Therapy: Moving Past Rapamycin.” Pharmacogenomics 11: 1,189–1,191. Walker, L. C., H. Levine III, M. P. Mattson, and M. Jucker. (2006) “Inducible Proteopathies.” Trends in Neurosciences 29: 438–443. Wallace, D. C. (2005) “A Mitochondrial Paradigm of Metabolic and Degenerative Diseases, Aging, and Cancer: A Dawn for Evolutionary Medicine.” Annual Review of Genetics 39: 359–407. Walsh, C. T., S. Garneau-Tsodikova, and G. J. Gatto, Jr. (2005) “Protein Posttranslational Modifications: The Chemistry of Proteome Diversifications.” Angewandte Chemie International Edition 44: 7,342–7,372. Wang, S., R. Wang, L. Chen, D. A. Bennett, D. W. Dickson, and D. S. Wang. (2010) “Expression and Functional Profiling of Neprilysin, InsulinDegrading Enzyme, and Endothelin-Converting Enzyme in Prospectively Studied Elderly and Alzheimer’s Brain.” Journal of Neurochemistry 115: 47–57. Wang, D. O., K. C. Martin, and R. S. Zukin. (2010) “Spatially Restricting Gene Expression by Local Translation at Synapses.” Trends in Neurosciences 33: 173–182. Wang, E. T., R. Sandberg, S. Luo, I. Khrebtukova, L. Zhang, C. Mayr, S. F. Kingsmore, G. P. Schroth, and C. B. Burge. (2008) “Alternative Isoform Regulation in Human Tissue Transcriptomes.” Nature 456: 470–476. Wang, Z., C. Zang, J. A. Rosenfeld, D. E. Schones, A. Barski, S. Cuddapah, K. Cui, T. Y. Roh, W. Peng, M. Q. Zhang, et al. (2008) “Combinatorial Patterns of Histone Acetylations and Methylations in the Human Genome.” Nature Genetics 40: 897–903. Wang, Z., M. Gerstein, and M. Snyder. (2009) “RNA-Seq: A Revolutionary Tool for Transcriptomics.” Nature Reviews Genetics 10: 57–63. Waterhouse, E. G., and B. Xu. (2009) “New Insights into the Role of BrainDerived Neurotrophic Factor in Synaptic Plasticity.” Molecular and Cellular Neuroscience 42: 81–89.
200
INVESTIGATING THE HUMAN GENOME
Weatherall, D. J. (2008). “Genetic Variation and Susceptibility to Infection: The Red Cell and Malaria.” British Journal of Haematology 141 276–286. Wellen, K. E., G. Hatzivassiliou, U. M. Sachdeva, T. V. Bui, J. R. Cross, and C. B. Thompson. (2009) “ATP-Citrate Lyase Links Cellular Metabolism to Histone Acetylation.” Science 324: 1,076–1,080. Whitehead, J., G. K. Pandey, and C. Kanduri. (2009) “Regulation of the Mammalian Epigenome by Long Noncoding RNAs.” Biochimica et Biophysica Acta 1,790: 936–947. Whittaker, S., R. Marais, and A. X. Zhu. (2010) “The Role of Signaling Pathways in the Development and Treatment of Hepatocellular Carcinoma.” Oncogene 29: 4,989–5,005. Wiegand, K. C., S. P. Shah, O. M. Al-Agha, Y. Zhao, K. Tse, T. Zeng, J. Senz, M. K. McConechy, M. S. Anglesio, S. E. Kalloger, et al. (2010) “ARID1A Mutations in Endometriosis-Associated Ovarian Carcinomas.” New England Journal of Medicine 363: 1,532–1,543. Wild, E. J., and S. J. Tabrizi. (2007) “Huntington’s Disease Phenocopy Syndromes.” Current Opinion in Neurology 20: 681–687. Williams, T. N. (2006) “Human Red Blood Cell Polymorphisms and Malaria.” Current Opinion in Microbiology 9: 388–394. Williams, T. N. (2006) “Red Blood Cell Defects and Malaria.” Molecular and Biochemical Parasitology 149: 121–127. Wilquet, V., and B. De Strooper. (2004) “Amyloid-beta Precursor Protein Processing in Neurodegeneration.” Current Opinion in Neurobiology 14: 582–588. Wilson, B. G., X. Wang, X. Shen, E. S. McKenna, M. E. Lemieux, Y. J. Cho, E. C. Koellhoffer, S. L. Pomeroy, S. H. Orkin, and C. W. Roberts. (2010) “Epigenetic Antagonism Between Polycomb and SWI/SNF Complexes During Oncogenic Transformation.” Cancer Cell 18: 316–328. Wolfe, M. S., and D. J. Selkoe. (2010) “Giving Alzheimer’s the Old One–Two.” Cell 142: 194–196. Xie, H., V. A. Valera, M. J. Merino, A. M. Amato, S. Signoretti, W. M. Linehan, V. P. Sukhatme, and P. Seth. (2009) “LDH-A Inhibition, a Therapeutic Strategy for Treatment of Hereditary Leiomyomatosis and Renal Cell Cancer.” Molecular Cancer Therapeutics 8: 626–635.
REFERENCES
201
Xiong, H., D. Callaghan, A. Jones, D. G. Walker, L. F. Lue, T. G. Beach, L. I. Sue, J. Woulfe, H. Xu, D. B. Stanimirovic, et al. (2008) “Cholesterol Retention in Alzheimer’s Brain Is Responsible for High Beta- and GammaSecretase Activities and Abeta Production.” Neurobiology of Disease 29: 422–437. Xu, B., J. L. Roos, S. Levy, E. J. van Rensburg, J. A. Gogos, and M. Karayiorgou. (2008) “Strong Association of De Novo Copy Number Mutations with Sporadic Schizophrenia.” Nature Genetics 40: 880–885. Yachida, S., S. Jones, I. Bozic, T. Antal, R. Leary, B. Fu, M. Kamiyama, R. H. Hruban, J. R. Eshleman, M. A. Nowak, et al. (2010) “Distant Metastasis Occurs Late During the Genetic Evolution of Pancreatic Cancer.” Nature 467: 1,114–1,117. Yanai, A., K. Huang, R. Kang, R. R. Singaraja, P. Arstikaitis, L. Gan, P. C. Orban, A. Mullard, C. M. Cowan, L. A. Raymond, et al. (2006) “Palmitoylation of huntingtin by HIP14 Is Essential for Its Trafficking and Function.” Nature Neuroscience 9: 824–831. Yashiro, K., T. T. Riday, K. H. Condon, A. C. Roberts, D. R. Bernardo, R. Prakash, R. J. Weinberg, M. D. Ehlers, and B. D. Philpot. (2009) “Ube3a Is Required for Experience-Dependent Maturation of the Neocortex.” Nature Neuroscience 12: 777–783. Yauch, R. L., G. J. Dijkgraaf, B. Alicke, T. Januario, C. P. Ahn, T. Holcomb, K. Pujara, J. Stinson, C. A. Callahan, T. Tang, et al. (2009) “Smoothened Mutation Confers Resistance to a Hedgehog Pathway Inhibitor in Medulloblastoma.” Science 326: 572–574. Yu, J., A. Astrinidis, S. Howard, and E. P. Henske. (2004) “Estradiol and Tamoxifen Stimulate LAM-Associated Angiomyolipoma Cell Growth and Activate Both Genomic and Nongenomic Signaling Pathways.” American Journal of Physiology: Lung Cellular and Molecular Physiology 286: L694–L700. Zempel, H., E. Thies, E. Mandelkow, and E. M. Mandelkow. (2010) “Abeta Oligomers Cause Localized Ca(2+) Elevation, Missorting of Endogenous Tau into Dendrites, Tau Phosphorylation, and Destruction of Microtubules and Spines.” Journal of Neuroscience 30: 11,938–11,950.
202
INVESTIGATING THE HUMAN GENOME
Zhao, S., W. Xu, W. Jiang, W. Yu, Y. Lin, T. Zhang, J. Yao, L. Zhou, Y. Zeng, H. Li, et al. (2010) “Regulation of Cellular Metabolism by Protein Lysine Acetylation.” Science 327: 1,000–1,004. Zhao, W. Q., F. G. De Felice, S. Fernandez, H. Chen, M. P. Lambert, M. J. Quon, G. A. Krafft, and W. L. Klein. (2008) “Amyloid Beta Oligomers Induce Impairment of Neuronal Insulin Receptors.” FASEB Journal 22: 246–260. Zhou, D., K. Liu, C. W. Sun, K. M. Pawlik, and T. M. Townes. (2010) “KLF1 Regulates BCL11A Expression and Gamma- to Beta-globin Gene Switching.” Nature Genetics 42: 742–744. Zoncu, R., A. Efeyan, and D. M. Sabatini. (2011) “mTOR: From Growth Signal Integration to Cancer, Diabetes, and Aging.” Nature Reviews Molecular Cell Biology 12: 21–35.
About the Author As a young child, I spent many months as a patient in the Children’s Hospital in Johannesburg, South Africa. A dedicated physician eventually found a solution to my health problem. This experience inspired me to pursue a career in medicine, and I graduated as a physician from the University of Pretoria, South Africa. The fields of genetics and biochemistry held great fascination for me, and I pursued these as a graduate student at University College London. I have had the privilege of participating in the clinical genetic services at the Children’s Hospital in Glasgow, Scotland, at the Mount Sinai Hospital in New York, and in the Pediatrics department at the University of California, Irvine. I also have been fortunate enough to be able to carry out research in gene mapping, gene isolation, and gene and enzyme analyses, and more recently in analysis of genetic factors in the etiology of autism. My forty-year career in clinical and research genetics happily coincided with eras of great expansion in our capacity to understand genome architecture and function and growing insights into the origin of disease. Although much has been accomplished in understanding the genetic basis of human disease, much remains to be done, particularly as we strive to find cures. Nevertheless, optimism seems appropriate.
203
This page intentionally left blank
INDEX Numbers 1p13 region, 13 1p13.5 region, 13 1p32-p36 chromosome, 66 1q21.1 region, 4 2p16-21 chromosome, 162 3HMGCoA reductase, 106 3p25-22 chromosome, 162 4q25-q28 chromosome, 66 5q23-q33 chromosome, 20 9p21.3 region, 13 11q12-q13 chromosome, 15 11q22.23 chromosome, 60 15q21.1-q21.2 chromosome, 64 16p11.2 region, 3 16p13.3 chromosome, 46, 54 17p11.2 chromosome, 76 17q12 region, 3 19p13 region, 13 19p13.12–p13.13 chromosome, 160 22q11.2 chromosome, 98 22q13 chromosome, 87
A Abeta accumulation in Alzheimer’s disease, 103 ABL oncogene, phosphorylation, 34 acetylation, 34, 37 ADNFLE (autosomal dominant nocturnal frontal lobe epilepsy), 94 ADP ribosylation in histone modifications, 38 adult-onset hemochromatosis, penetrance in, 54-55
aggregates analysis in ALS, 121 formation of, 122 in late-onset neurodegenerative diseases, cellular transfer, 126-127 in Parkinson’s disease, 123 aging DNA methylation and, 51 effect on penetrance, 53 neurodegenerative diseases and, 123 AIP (arylhydrocarbon receptor interacting protein), 15-17 AKT gene, 72 AKT1 gene, 99 ALK–EML4 fusion genes, 140, 142 alpha secretase activity in Alzheimer’s disease treatment, 114-115 alpha-thalassemia, 45, 57 alpha2 globin gene, long ncRNA, 29 ALS (amyotrophic lateral sclerosis), 117 FUS in, 119-120 genotype-phenotype studies, 121 protein aggregate analysis, 121 SOD1 in, 118 TDP43 in, 118-119 alternate splicing, 26, 30-31 Alzheimer’s disease alpha and beta secretase activity, 114-115 amyloid aggregates as prions, 113 APOE APOE4 versus APOE3, 104 APP and, 103 brain lipids and, 105 polymorphism, 102
205
206 APP and, 101-102 biomarkers in detection of, 108 brain cholesterol homeostasis and treatment, 115 brain cholesterol, role of, 107-108 cholesterol biosynthesis in brain, 105-106 clusterin and, 110 GWAS and, 109-112 molecular-based treatment, 113 neurodegeneration in, 108-109 prion protein, role of, 112, 116 questions regarding treatment, 116-117 Amish population studies in genomic medicine, 154-155 AMPAR receptors, 82 AMPK gene, 72, 76 amyloid aggregates as prions, 113 amyotrophic lateral sclerosis. See ALS Angelman syndrome, UBE3A gene and, 95-96 ANT1 protein, MECP2 interaction, 47 antiepilepsy drugs, 95 APC gene, 138 APOA1 gene, 14 APOB gene, 14 APOE gene in Alzheimer’s disease, 102 brain lipids and, 105 in coronary heart disease, 14 APOE3 protein versus APOE4 protein in Alzheimer’s disease, 104 APP (amyloid precursor protein), Alzheimer’s disease and, 101-103 arginine deaminase in cancer treatment, 147 ARH (arylhydrocarbon receptor), 17 ARID1A gene, 48 ARNT protein, 17 arylhydrocarbon receptor interacting protein (AIP), 15, 17 asparaginase in cancer treatment, 147 ataxia telangiectasia (AT), DNA damage and repair, 60 ATM gene, 60-61, 149 ATM-ATR response, 63 ATP-dependent chromatin remodeling, 41-42 ATRX gene, 45
INDEX autism gene mutations in, 96-97 mTOR signaling pathway and, 89 in phenotypically discordant monozygotic twins, 51
B barrier insulation, 40 basal cell nevus syndrome, 133 base excision repair, 149 BCL2 gene, 51 BCL11A oncogene, 160-161 BDNF (brain-derived neurotrophic factor), 32 in obesity, 12 in synaptic plasticity, 91 Bernard, Claude, 169 beta-catenin signaling pathway hepatocellular carcinoma, 139 identifying oncogenes in, 137-139 BHD (Birt Hogge Dube) syndrome, 76 bioenergetic regulation in cancer treatment, 145-148 biology of tumor cells, analysis of, 140-144 biomarkers in Alzheimer’s disease detection, 108 fusion genes as, 140-142 bipolar disease, 158 Bird, Adrian, 37 Birt Hogge Dube syndrome, 76 Blackfan Diamond anemia, 78 BMAL1 protein, circadian rhythms and, 44 BRAF gene, 135 brain. See also cognitive impairment cholesterol biosynthesis in, 105-106 cholesterol in, role in Alzheimer’s disease, 107-108 lipid metabolism in, 105 brain cholesterol homeostasis in Alzheimer’s disease treatment, 115 brain development, microRNA and, 32 BRCA2 gene, 62 BRD (bromodomain) proteins, 42 BRG1 (Brahma) gene, 42
INDEX
C C282Y gene, 54 cancer. See also renal cancer; tumors bioenergetic regulation in treatment, 145-148 biology analysis of tumor cells, 140-144 driver mutations, treatments for, 134-136 epithelial ovarian cancer, 162 fusion genes as biomarkers, 140-142 genome instability and, 149-152 hedgehog signaling pathway, 133-134 metabolic targets in treatment, 146-147 molecular networks, role in treatment, 136-139 molecular studies and treatments, 131-134 mTOR pathway and, 72 personalized treatments for, 143-144 prostate cancer, ETS transcription factors in, 143 systems biology modeling, 144-145 telomeres and, 150 cardiovascular system, microRNA and, 32-33 CBP (CREB binding protein), 45 CDK8 gene, 138 CDKAL1 gene, 11-12 CDKN2A gene, 11-12 CDKN2B gene, 11-12 cDNA, generating from polyadenylated RNA, 25 centrosomes, defined, 63 CENP152 gene, 64 CETP gene, 14 chaperone proteins, 79 Charcot-Marie-Tooth neuropathies (CMT), 19-21 CHARGE syndrome, 42 CHD proteins, chromatin remodeling, 42 cholesterol biosynthesis in brain, 105-106 cholesterol homeostasis, 105 cholesterol in brain, role in Alzheimer’s disease, 107-108 cholesterol levels, GWAS on, 13-14
207 CHRNA4 gene, 94 CHRNB2 gene, 94 chromatin ATP-dependent chromatin remodeling, 41-42 in Fanconi anemia, 61-65 insulators and, 39-40 interphase human chromosome analysis, 40-41 neuronal stimulation and, 42-43 chromatin modifier genes, 48-49 chromosome instability, cancer and, 149-152 chronic myelogenous leukemia, Philadelphia chromosome in, 132 chronic obstructive pulmonary disease (COPD), environmental factors in, 58-59 circadian rhythm, metabolism and, 43-45 circulating tumor cells, biology analysis, 140-144 clinical trials of molecular-based cancer treatments, 137 CLOCK gene, 43-45 clonal heterogeneity, 145 clusterin, Alzheimer’s disease and, 110 CMT (Charcot-Marie-Tooth neuropathies), 19-21 cognitive impairment synaptic activity and, 81 dendritic spines, 82 FMRP gene, 84-87 gene expression, 83-84 neuroligins, 82 WDR62 gene mutations and, 92-93 cohesins, defects in, 45 congenital malformations, ribosome biogenesis pathway and, 78-79 consciousness, EEG studies on, 93 COPD (chronic obstructive pulmonary disease), environmental factors in, 58-59 copy number variants, 3-5, 19 core duplicons, 2 Cornelia de Lange Syndrome, 45 coronary heart disease, GWAS on, 13-14 CpG islands, methylation, 38 CR1 gene, 58
208 CREB (cyclic AMP response element binding protein), 32, 45, 158 Crestor, 165 CTCF protein, 40-41 Cushing, Harvey, 15 CYFIP1 gene, 85 CYP46A1 gene, 107 cytosine-to-thymidine (C-to-T) transversions, 66
D D4Z4 DNA element, gene expression, 47 data integration in functional genomics, 161 systems biology and medicine, 163 deletions, phenotypic abnormalities and, 3-5 dendritic spines, synaptic activity and, 82 dephosphorylation of FMRP gene, 86 DGCR8 gene, 31 DHHC8 gene, 98 DHHC9 gene, 99 DHHC15 gene, 99 DHHC17 gene, 98 DHHC21 gene, 98 DHODH gene, 21 diabetes. See type 2 diabetes discovery-driven approaches in system analyses, 161 disease-specific mutations. See genome wide association studies disulfide bond formation, 34 DNA damage and repair cancer and, 149-150 environmental factors in, 59-61 in Fanconi anemia, 61-65 in Seckel syndrome, 63-64 signatures of environmentally induced damage, 65-66 ubiquitin ligases and, 64-65 DNA methyl transferases, methylation, 39 DNA methylation, aging and, 51 donor identification by genotyping, 155 double-stranded DNA breakage, ubiquitin ligases and, 64-65
INDEX driver mutations in cancer, treatments for, 134-136 DROSHA RNAase complex, 31 drug metabolism in personalized medicine, 165 Duffy blood group, malaria and, 58 duplications, segmental, 2-5
E E657K gene, 20 EEG studies on consciousness, 93 EGFR (epidermal growth factor receptor), 132 EML4–ALK fusion genes, 140-142 ENCODE project, transcription, 23-24 alternate splicing and, 26, 30-31 gene expression variation, 26-27 genes, definition of, 24 LINC RNA, 29 long ncRNA, 28-29 ncRNA, 27-28 RNA sequencing and, 25 Encyclopedia of DNA Elements. See ENCODE project enhancer block insulation, 39 enriched environments, epigenetics in, 43 environmental factors in gene expression, 45-46, 53 DNA damage and repair, 59-61 in Fanconi anemia, 61-65 penetrance, impact on, 53-59 signatures of, 65-66 environmental signals, integrating with growth, 73 epidermal growth factor receptor (EGFR), 132 epigenetics ATP-dependent chromatin remodeling and, 41-42 chromatin modifier genes in tumors, 48-49 circadian rhythm and, 43-45 defined, 37 DNA methylation and aging, 51 in enriched environments, 43 environmental cues and gene expression, 45-46 histone modifications, 37-38
INDEX insulators and, 39-40 interphase human chromosome analysis in, 40-41 MECP2 gene expression, 46-47 methylation and, 38-39 neuronal stimulation and, 42-43 parent of origin allelic expression, 49-50 in phenotypically discordant monozygotic twins, 50-51 stem cell biology and, 52 therapeutic interventions based on, 47-48 variant histones, 49 epilepsy, 94-95 epistasis, 79 epithelial ovarian cancer, 162 erasers in chromatin remodeling, 42 ERBB4 protein, 99-100 erlotinib, 132 erythroblasts, gene expression in, 40 ETS transcription factors, 143 ETV1 gene, 142 ETV6–RUNX1 translocation, 145 excitatory synapses, 82, 85 exome sequencing, 21-22 expression of fetal hemoglobin, persistence of, 158, 161
F familial pituitary adenoma, GWAS on, 15-17 FANCA gene, 62 FANCD1 gene, 62 FANCD2 gene, 62 FANCF gene, 62 FANCI gene, 62 Fanconi anemia, 61-65, 166 fascio-scapulo-humeral muscular dystrophy (FSHD), 47 FAT domains, 88 fetal hemoglobin, persistence of expression, 158, 161 Finnish disease heritage, 8, 10 FKBP domains, 16 FLCN gene mutations, 76 FMR1 gene, 69 FMRP gene, 84-87 foldback inversion, 152 folliculin gene mutations, 76
209 FOX3A gene, 72 Fragile X mental retardation protein. See FMRP gene Fragile X syndrome GABA neurotransmitter function in, 86 mGluR gene, 86 mTOR signaling pathway and, 89 palmitoylation and, 99 phenocopies in, 68-69 FRB domain, 88 French-Canadian Leigh syndrome, 162-163 FSHD (fascio-scapulo-humeral muscular dystrophy), 47 FTO gene, 12-13 fumarate hydratase, germline mutations of, 76 functional genomics, data integration in, 161. See also genomic medicine FUS positive inclusions, 121 FUS protein, 119-121 fusion genes as biomarkers, 140-142
G G6PD enzyme, malaria survival, 56 GABA neurotransmitter function, 86 gamma synchrony, 93 gastrointestinal stromal tumors, treatment of, 142 GCK gene, 11-12 GCKR gene, 14 GDC-0449 molecule, 133 gefitinib, 132 Gene Cards, 169 gene expression environmental factors, 53 DNA damage and repair, 59-61 in Fanconi anemia, 61-65 penetrance, impact on, 53-59 signatures of, 65-66 linking environmental cues to, 45-46 MECP2 impact on, 46-47 parent of origin allelic expression, 49-50 in schizophrenia, 99-100 synaptic activity and, 83-84 variation in microRNA and, 31-33 RNA sequencing and, 26-27
210 gene mutations in autism, 96-97 genes, definition of, 24 Genetests, 169 genetics, benefits of studying, 167-169 genome instability, cancer and, 149-152 genome wide association studies. See GWAS (genome wide association studies) genomic integrity, DNA damage response and, 63-64 genomic medicine. See also functional genomics Amish population studies in, 154-155 defined, 153 integration with systems biology, 163 pharmacologic response to, 158 in PKD (polycystic kidney disease), 157 therapy development, 156-157 genotype-phenotype studies in ALS, 121 genotyping, donor identification by, 155 germline mutations, 73 in Birt Hogge Dube syndrome, 76 of fumarate hydratase, 76 succinate dehydrogenase mutations, 77 in Von Hippel Lindau syndrome, 74-75 GKAP gene, 87 Gleevec, 132 glucose metabolism, 145 glutamine in cancer treatment, 147 glycolysis pathway, 71 glycosylation, 34 growth, integrating with environmental signals, 73 GSK3 gene, 158 GSK3B gene, 73 guanine-to-cytosine (G-to-C) transversions, 65 guanine-to-thymidine (G-to-T) transversions, 65 GWAS (genome wide association studies), 10 Alzheimer’s disease, 109-112 coronary heart disease, 13-14 familial pituitary adenoma, 15-17 type 2 diabetes and obesity, 10-13
INDEX
H H2AK5ac histone, 40 H2AX histone, 49, 61 H2AZ histone, 49 H3K27me histone, 42 H3K27Me3 histone, 40 hair root analysis, advantages of, 6 HAR1 gene, 28 HAT (histone acetyl transferase), 38 HbC allele, 56 HbE allele, 57 HbF (fetal hemoglobin), persistence of expression, 158, 161 HbS allele, 56 HDAC (histone deacetylase), 38 HEAT domains, 88 hedgehog signaling pathway in cancer, 133-134 hematopoietic stem cell transplants, donor identification for, 155 hemochromatosis, penetrance in, 54-55 hemoglobinopathies, 159-161 hepatocellular carcinoma, 139 HERC2 gene, 6, 64-65 Herceptin, 132 hereditary disorders, population history and, 8-10 Heterozygosity in regional populations, 7 HHEX gene, 11 HIF (hypoxia inducible factor), 74 HIF1 gene, 71 HIF1b protein, 17 high-throughput RNA sequencing, 25 histone acetyl transferase (HAT), 38 histone acetylation, 48 histone deacetylase (HDAC), 38, 47-48 histone modifications, 37-39 histones, variants, 49 HLA complex, P4 medicine and, 164 HLA gene variants in diabetes, 12 HNF1A gene, 11-12 HNF1B gene, 11-12 HNF4A gene, 11 homologous recombination, 149 homozygosity mapping, 8-9 in Saqqaq culture, 6
INDEX HOTAIR long ncRNA, 28 HOXC gene, 28 HOXD gene, 28 HSP70 chaperone protein, 79 HSP90 chaperone protein, 79 Hunterian skeleton, familial pituitary adenoma in, 15 huntingtin gene, 68 Huntington’s chorea, palmitoylation and, 98 Huntington’s disease genetic studies on, 156 phenocopies in, 68 hydroxylation, 34 hyperlipidemia, treatment for, 14 hypothesis-driven approaches in system analyses, 161 hypoxia inducible factor (HIF), 74
I IDH1 mutation, 148 IDH2 mutation, 148 IL28B gene, 165 imatinib, 132, 142 imetelstat, 150 indisulam, 147 induced pluripotent stem cells (IPS cells), 165 inducible proteopathies, 127 inhibitors in cancer treatment, 147 on mTOR protein, 90 inhibitory synapses, 82 INS gene, 12 insulators, 39-40 integrative genomics explained, 161 in French-Canadian Leigh syndrome research, 162-163 interleukin4 receptor gene, 79 interleukin13 receptor gene, 79 interphase human chromosome analysis, 40-41 intrachromosomal duplications, 2 IPF gene, 11 IPS cells (induced pluripotent stem cells), 165 iron levels in hemochromatosis, 54-55 IRS1 gene, 11-12 isocitrate dehydrogenases IDH1 and IDH2, 148
211
J–K JPH3 gene, 68 K288X gene, 160 KCNJ11 gene, 11-12 KIN domain, 88 KIT oncogene activation, 142 KLF1 gene, 160-161 Kuru disease, 127
L lactate in cancer cells, 145 large intervening noncoding RNA (LINC RNA), 29 late-onset neurodegenerative diseases aggregates, transfer methods of, 126-127 ALS. See ALS (amyotrophic lateral sclerosis) Alzheimer’s disease. See Alzheimer’s disease FUS and TDP43 in, 120-121 inducible proteopathies, 127 mitochondria and, 123-124 mitochondrial function regulation, 128-129 Parkinson’s disease, aggregates in, 123 prion protein conformations, 127-128 protein aggregates, formation of, 122 protein modification in, 124-126 questions regarding, 129-130 SIRT1 gene, role in treatment, 129 LDL cholesterol levels, GWAS on, 13-14 LDLR gene, 14 Leigh syndrome, 162-163 LINC RNA, 29 linkage disequilibrium, 9-10 lipid metabolism, APOE and, 105 lipid rafts, 106, 115 lithium treatment, 158 liver damage, 139 living cells, interphase human chromosome analysis in, 40-41 long ncRNA, 28-29 long Q-T syndrome, 165 Lou Gehrig’s disease. See ALS LPL gene, 14 LRPPRC gene, 163 LUC7L gene, 29
212
M malaria, genetic factors in survival, 55-58 MAP3K8 kinase, 135 maple syrup urine disease, 154 massively parallel sequencing. See next-generation sequencing maturity-onset diabetes of the young (MODY), 10-12 meclizine, 156 MECP2 gene, 45-47 medicine integration with systems biology, 163 P4 medicine, 164-165 medulloblastomas, 133 melanocortin 4-receptor gene, 11 melanomas, treatment of, 135 mental retardation syndrome, 45 MET proto-oncogene, 75 metabolic targets in cancer treatment, 146-147 metabolism, circadian rhythm and, 43-45 metastatic tumors, biology analysis, 140-144 methyl-binding domains, 39 methylation, 34, 38-39 aging and, 51 in histone modifications, 38 mGluR gene, 86-87 mGluR receptors, 82 mGluR1 gene, 85 mGluR5 gene, 85 microcephaly, WDR62 gene mutations and, 92-93 microRNA cardiovascular system and, 32-33 gene expression and, 31-33 Miller syndrome exome sequencing of, 21 next-generation sequencing and, 18 miR29 gene, 33 miR208b gene, 33 miR499 gene, 33 miR2081 gene, 33 miRNA1 gene, 32 miRNA29 gene, 32 miRNA-132 gene, 32 miRNA-145 gene, 32 miRNA-219 gene, 32
INDEX mitochondria, neurodegenerative diseases and, 123-124 mitochondrial diseases, integrative genomics and, 163 mitochondrial function regulation in neurodegenerative disease treatment, 128-129 MODY (maturity-onset diabetes of the young), 10-12 molecular networks, role in cancer treatment, 136-139 molecular studies in cancer treatments, 131-134 molecular-based treatment of Alzheimer’s disease, 113 monogenic epilepsy, 94 monozygotic twins, phenotypical discordance in, 50-51 motor neuron disease. See ALS mRNA FMRP binding, 84 repressing translation of, 83 MTHFD1L gene, 111-112 MTNR1B gene, 11-12 mTOR protein function of, 89-90 inhibitors on, 90 regulating expression by stress, 72-73 role in diseases, 69-72 signaling pathways on, 87 structure of, 88 synaptic activity and, 87 TSC1 TSC2 complex and, 90 mTORC1 gene, 89-90 mTORC2 gene, 88-90 mutations. See variations myocardial infarction, treatment for, 14
N NAD-dependent histone deacetylases in epigenetics, 43-45 NAMPT enzyme, 44 ncRNA, 27-29 Neanderthal fossils, sequence variation in, 5-6 NEGR1 gene, 12 neprilysin in Alzheimer’s disease, 114 NEUROD1 gene, 11
INDEX neurodegeneration in Alzheimer’s disease, 108-109 neurodegenerative disease. See lateonset neurodegenerative diseases neurofibromatosis, mTOR signaling pathway and, 89 neuroligins, synaptic activity and, 82 neurological diseases, palmitoylation and, 98-99 neuronal stimulation, 42-43 next-generation sequencing, 17-19. See also exome sequencing; whole-genome sequencing NFS1 gene, 11 Niemann Pick disease type C (NPC), 106 NIPBL gene, 45 nitrosative stress, protein modification and, 124-126 NMDAR receptors, 82 non-protein coding RNA (ncRNA), 27-28 LINC RNA, 29 long ncRNA, 28-29 nonhomologous end joining, 149 NRG1 gene in schizophrenia, 99-100 nucleosome mobility, 41 nucleotide excision repair, 59, 149
O–P obesity, GWAS on, 10-13 OCA2 gene, 6 oncogenes activation, targeting, 142 addiction, 139 identifying, 137-139 oxidative stress, protein modification and, 124-126 P4 medicine, 164-165 p400 gene, 42 palmitoylation, 34 explained, 97-98 in late-onset neurodegenerative diseases, 124-126 neurological diseases and, 98-99 post-translational protein modification, 33-35 pancreatic cancer, genome instability and, 151-152
213 parent of origin allelic expression, 49-50 Parkinson’s disease, aggregates in, 123 PARP inhibitors, 136 pathways in cancer driver mutations, 134-135 hedgehog signaling pathway in cancer, 133-134 identifying oncogenes in, 137-139 mTOR pathway regulating expression by stress, 72-73 role in diseases, 69-72 in renal cancer, 73 Birt Hogge Dube syndrome, 76 MET proto-oncogene, 75 succinate dehydrogenase mutations, 77 tricarboxylic acid metabolism, 76 tuberous sclerosis, 75 Von Hippel Lindau syndrome, 74-75 ribosome biogensis pathway, congenital malformations and, 78-79 PDGF gene, 74 PDK1 gene, 146 PDK1 inhibitors, 147 penetrance defined, 53 factors affecting, 53-59 in COPD, 58-59 in hemochromatosis, 54-55 in malaria, 55, 57-58 Pennisi, Elizabeth, 23 pentose phosphate pathway, 71 pericentric duplications, 2 pericentrin in centrosomes, 63 persistence of expression of fetal hemoglobin, 158, 161 personalized cancer treatments, 143-144 personalized medicine, 153, 164-165 pharmacologic response to genomic medicine, 158 PharmGKB (Pharmacogenomics knowledge base), 169 phenocopies defined, 67 in Fragile X syndrome, 68-69 in Huntington’s disease, 68
214 phenome-interactome networks, 162 phenotypic abnormalities, structural variation and, 3-5 phenotypical discordance in monozygotic twins, 50-51 Philadelphia chromosome, 132 phosphorylation, 34 of FMRP gene, 86 in histone modifications, 38 PI3K pathway, 75 PICALM gene, 110 PKD (polycystic kidney disease), 157 pluripotent stem cells, 52, 165 PLX4032 treatment, 135 POLR1D gene, 78 polyadenylated RNA, conversion to cDNA, 25 polyadenylation, alternate splicing and, 30-31 polycystic kidney disease (PKD), 157 population history, hereditary disorders and, 8-10 populations, sequence variation in regional populations, 7-8 post-synaptic density (PSD), 82 post-translational protein modification, 33-35 PPARG gene, 11-12 presenilins in Alzheimer’s disease, 101 primary tumors, biology analysis, 140-144 prion protein amyloid aggregates as, 113 conformations, 127-128 role in Alzheimer’s disease, 112, 116 PRNP gene, 68 prostate cancer, ETS transcription factors in, 143 protein aggregates. See aggregates protein modification. See palmitoylation proteome, defined, 33 proteostasis network, 79 PSD (post-synaptic density), 82 PSD95 domain, 87 PSEN1 gene, 101 PSEN2 gene, 101 PTCH1 gene, 133 PTPN22 gene, 12 pyruvate in cancer cells, 145
INDEX
Q–R Q14X gene, 16 quercetin, 139 R271W gene, 16 R954X gene, 20 Rab11 membrane, 21 rapamycin, 90 Raptor protein, 88 readers in chromatin remodeling, 42 red-cell membrane proteins, malaria and, 58 REDD1 gene, 72 REDD2 gene, 72 regional populations, sequence variation in, 7-8 renal cancer, pathways in, 73 Birt Hogge Dube syndrome, 76 MET proto-oncogene, 75 succinate dehydrogenase mutations, 77 tricarboxylic acid metabolism, 76 tuberous sclerosis, 75 Von Hippel Lindau syndrome, 74-75 resistance to cancer treatments, 135 Rett syndrome, 45-46 RhebGTP gene, 71 ribosome biogenesis pathway, congenital malformations and, 78-79 Rictor protein, 88 RNA sequencing, 25-27 RNF168 gene, 64-65 RNG105 gene, 86 RORA gene, 51 ROS (reactive oxygen species), 123 rosiglitazone, 104
S Saqqaq individuals, sequence variation in, 6-7 schizophrenia gene expression in, 99-100 palmitoylation and, 98 SCID (severe combined immune deficiency), 155 SCRAP gene, 42 Seckel syndrome, DNA damage response and, 63-64 secretase levels in Alzheimer’s disease treatment, 114-115
INDEX segmental duplications, 2-5 sequence variation, 4 exome sequencing, 21-22 GWAS. See GWAS (genome wide association studies) in hereditary disorders and population history, 8-10 homozygosity mapping, 8-9 in Neanderthal fossils, 5-6 next-generation sequencing, 17-19 in regional populations, 7-8 in Saqqaq individuals, 6-7 whole-genome sequencing, 18-20 serotonin pathways, lithium treatment and, 158 severe combined immune deficiency (SCID), 155 SH2B1 gene, 12 SH3TC2 gene, 20-21 SHANK proteins, PSD95 domains and, 87 “Shining a Light on the Genome’s Dark Matter” (Pennisi), 23 signaling pathways on mTOR protein, 87 signatures of environmentally induced DNA damage, 65-66 single nucleotide polymorphisms (SNPs), 1, 10 SIRT1 deficiency, 91 SIRT1 gene circadian rhythms, 43, 45 role in neurodegenerative disease treatment, 129 SIRT1-7 proteins, 128 SLC25A3 gene, 30 SMC1A gene, 45 SMC3 gene, 45 Smith Laemli Opitz syndrome, 106 SMO gene, 133 SNPs (single nucleotide polymorphisms), 1, 10 SOD1 gene, 118 somatic nuclear transfer stem cells, 52 SORT1 gene, 13 SORTL1 gene, 14 spinal muscular atrophy, 166 spines. See dendritic spines SREBP gene, 71 statins in coronary heart disease treatment, 14
215 stem cell biology, epigenetics and, 52 stress, regulating mTOR pathway expression, 72-73 structural variation, 2-5 subtelomeric duplications, 2 succinate dehydrogenase mutations, 77 sumoylation in histone modifications, 38 SWI SNF protein family, 41 synaptic activity in autism, 96-97 cognitive impairment and, 81 dendritic spines, 82 FMRP gene, 84-87 gene expression, 83-84 neuroligins, 82 mTOR function in, 89 mTOR protein and, 87 palmitoylation and, 97-98 UBE3A gene and Angelman syndrome, 95-96 synaptic plasticity, 43, 91-92 synaptophysin protein, 91 synthetic lethality, 136 synucleain aggregates in Parkinson’s disease, 123 system biology, integration with medicine, 163 systems biology modeling in cancer research, 144-145
T tamoxifen, 132 targeted mutational analysis of tumors, 143-144 tau phosphorylation, Alzheimer’s disease and, 108-109 TCF7L2 gene, 11-12 TCOF1 gene, 78 TDP43 gene, 117-121 telomerase, 150 telomere shortening, 150 telomeres, 46, 61, 150 TGFbeta gene, 59 therapeutic interventions based on epigenetics, 47-48 therapy development in genomic medicine, 156-157 TMPRSS2–ERG fusion genes, 143
216 tobacco smoke, DNA damage signatures from, 65-66 TORC1 gene, 88 transcription, ENCODE project, 23-24 alternate splicing and, 26, 30-31 gene expression variation, 26-27 genes, defintion of, 24 LINC RNA, 29 long ncRNA, 28-29 ncRNA, 27-28 RNA sequencing and, 25 transcriptomics, RNA sequencing and, 25 transfer methods for aggregates in late-onset neurodegenerative diseases, 126-127 translation, mTORC1 in, 89 translation initiation, triggering, 85 transversions in environmentally induced DNA damage, 65 trastuzmab, 132 Treacher Collins syndrome, 78-79 treacle, 78 treatments based on epigenetics, 47-48 TRF1 gene, 61 TRF2 gene, 61 tricarboxylic acid metabolism, 76 TSC1 gene, 70 TSC1 TSC2 complex, 90 TSC2 gene, 54, 70 tuberous sclerosis FMRP phosphorylation and, 86 mTOR pathway and, 70, 89 renal cancer and, 75 variable expressivity in, 54
INDEX tumors. See also cancer chromatin modifier genes in, 48-49 DNA damage signatures in, 65-66 microRNA and, 32 mTOR pathway and, 72 twins, phenotypical discordance in monozygotic twins, 50-51 type 2 diabetes, GWAS on, 10-13
U–V UBE3A gene, 95-96 ubiquitin ligases, double-stranded DNA breakage and, 64-65 ubiquitination in histone modifications, 38 variable expressivity, 54 variant histones, 49 variation. See sequence variation; structural variation VEGF gene, 74 Von Hippel Lindau syndrome, 74-75
W–Z Warburg effect, 146 WDR62 gene, 92-93 whole-genome sequencing, 18-20 Williams syndrome, 3 writers in chromatin remodeling, 42 Y169H gene, 20 YY1 protein, 46-47