Acerca del autor

Dr. Patrizia Ferretti. Developmental Biology Unit, Institute of Child Health, University College London.

Prof. Andrew Copp (Dean of Institute). Neural Development Unit, Institute of Child Health, University College London.

Prof. Cheryll Tickle. Professor of Anatomy & Physiology, The Wellcome Trust Building, University of Dundee.

Prof. Gudrun Moore. Institute of Child Health, University College London.

The editors are all distinguished developmental biologists with a broad range of expertise in human birth defects. Andrew Copp holds an endowed chair in Developmental Neurobiology at University College London and is Dean of the world-renowned Institute of Child Health.

De la contraportada

In the Western world, birth defects constitute the greatest single cause of infant mortality and a significant cause of infant morbidity, with a major impact on healthcare services and the affected families. Birth defects are the consequence of defective embryonic development that can be due to genetic, epigenetic or teratogenic factors.

The first edition of Embryos, Genes and Birth Defects, edited by the late Peter Thorogood, was a radical new book aimed at bridging the gap between the medical disciplines of embryology and dysmorphology, and recent advances in cellular, molecular and developmental biology. This new edition remains unique in its breadth and brings up to date our understanding of birth defects and of the strategies utilized to gain such knowledge. It features new chapters on human cytogenetics, mutagenesis and the yes and ears.

The book present key topics in developmental biology and explains how they provide the foundations of an understanding of clinical birth defects. The first six chapters introduce concepts and strategies adopted to elucidate developmental anomalies leading to birth defects. The book then focuses on specific organs and reviews the cellular and molecular mechanisms affecting their development and how disruption of these mechanisms by genetic or environmental factors may underlie certain birth defects. The chapters are concise and provide an up to date coverage of topics in a format that is easily accessible and yet at the forefront of research.

Written primarily for paediatricians, obstetricians, clinical geneticists and allied workers, this book guides the reader through the contribution of modern molecular biology to our understanding of human development. Developmental and cellular biologists will learn how errors in cellular and genetic mechanisms can lead to classical disorders, diseases and syndromes.

Fragmento. © Reproducción autorizada. Todos los derechos reservados.

Embryos, Genes and Birth Defects

John Wiley & Sons

Chapter One

Philip Stanier and Gudrun Moore

Introduction

Without even considering early fetal loss, it is reported that as many as 3.5% of all live-born babies have some kind of major abnormality, referred to as a birth defect. Actual incidences may vary according to locality, culture, ethnicity and the efficiency of recognition and reporting. If minor abnormalities such as cleft lip are included, then the incidence is nearer to 5%. In the Western world, birth defects constitute the greatest single cause of infant mortality and have a major impact on national health care budgets (http://www.modimes.org/).

In this introductory chapter some basic precepts and concepts are presented and explained. For a comprehensive introduction to embryonic development per se, the reader is referred to any one of several excellent publications that already exist (e.g. Alberts et al., 2002; Gilbert, 2003; Wolpert, 2002). What this chapter attempts to provide is the information that might be necessary for a clinician or advanced student specializing in paediatric medicine to understand and appreciate in context what follows. In that sense, an element of unorthodoxy might be discerned by some readers. However, we hope that this rationale will be justified as the reader progresses through the book.

The relationship between genotype and phenotype

The term 'genotype' is generally used to refer to the genetic make-up or constitution of an individual organism, be it virus, fruit fly or human. In contrast, we use the word 'phenotype' to cover the form and functioning of an individual, to the extent that it may encompass metabolism and behaviour (and thus we can refer to 'behavioural phenotypes'). The word 'genotype' is subtly but distinctly different from the term 'genome', which refers not to the totality of genes in an individual cell but to the array of genes in a complete haploid set of genes characteristic for that species. In this sense, a genome is a species-specific concept, whereas genotype is a concept applying to an individual of the species in question.

The complexity of the phenotype reflects largely but not entirely the complexity of the genotype. However, there is not necessarily a simple and direct relationship, since genome size and genome complexity are rather different entities. Overall genome size, in terms of DNA, is to some extent determined by the relative proportion of non-protein coding sequences contained within it. Thus, some plant, insect and amphibian species contain far more total DNA in their genomes than does Homo sapiens, even though they are phenotypically simpler and contain fewer genes (indeed, some amphibian species contain up to 9 x [10.sup.11] nucleotide bases per haploid genome, as opposed to the 2:85 x [10.sup.9] nucleotides recently sequenced in humans; International Human Genome Sequencing Consortium, 2004). Much of this increase in DNA content is thought to represent a greater than normal proportion of non-coding, repetitive sequences. If we consider genome complexity in terms of the number of genes present, then a more systematic relationship emerges. In simple organisms, such as viruses, the limited number of genes in the genome can be accurately determined. However, for more complex multicellular organisms, total gene number is an estimate based on confirmed genes and potential coding regions identified by predictive methods. Therefore, the size of these estimates has changed as our ability to visualize the DNA sequence and our understanding of genomic organization has evolved. Currently, Drosophila melanogaster, the fruit fly, is estimated to contain some 14 000 genes in its genome, whereas the genome of Homo sapiens is thought to comprise between 20 000 and 25 000. However, this latter set of figures is still subject to revision and does not take into account the considerable protein variation that can accrue from alternate usage and splicing of exons or the existence of functional non-coding RNAs.

Whereas gene mapping refers to identification of the chromosomal location of an individual gene, genome mapping is a programme of research designed to identify the chromosomal location of all genes in the genome of a particular species. Although it is the international Human Genome Project that has received wide media attention, it should be noted that genome mapping projects for other species are also under way or recently completed. These include a number of model organisms, such as the mouse, fruitfly, toad and nematode worm, as well as those of economically important food species, such as cow, pig and chicken (http://www.ncbi.nlm.nih.gov/Genomes/ index.html). The mapping of individual genes, or of candidate gene loci, means that chromosomal 'maps' of congenital abnormality can be drawn up (see Chapters 2, 3 and 4; also OMIM: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=OMIM), whereby the location of genes, in which mutation produces a particular dysmorphology or inherited metabolic disease, can be displayed (Figure 1.1).

At this point we should ask ourselves what kind of information is encoded within the genes. Are the genes really the 'blueprint' to which they are often analogized? A blueprint implies some kind of descriptive specification. Is that indeed how the genome is organized? In fact, the information content of genes is one-dimensionally complex, since it is specified by the nature of the linear sequence of nucleotide bases along the DNA molecule. In dramatic contrast, the phenotype is three-dimensionally complex (and four-dimensionally complex if we include dynamic phenomena, such as metabolism and homeostasis, rather than just morphology); yet the linear nucleotide sequence itself conveys no sense of what the phenotype might look like. To appreciate just how phenotypic complexity might be generated we have to move away from the rather dated analogy of a descriptive specification and think of the genome and its implementation as a generative programme. The more appropriate and meaningful analogy of origami has been proposed to illustrate the characteristics of a generative programme (Wolpert, 1991). Here, the instructions for creating a topologically complex shape from a sheet of paper contain within them no description of the final outcome. The complexity is generated progressively by implementing those instructions, which may in themselves be very simple, even though the outcome is complex. In this way, the genome, or at least the developmentally significant parts of it, can be seen as assembly rules for building an embryo.

In one sense, genes 'simply' encode proteins. Transcription of a gene produces a message that is translated from the four-letter alphabet (nucleotides) of the nucleic acids to the approximately 20-letter alphabet (amino acids) of the proteins, by virtue of the genetic code. The primary structure of a protein, i.e. the linear sequence of amino acids, together with any post-translational modifications, determines its secondary and tertiary structure. Proteins endow cells with properties such as characteristic metabolisms, behaviour, polarity, adhesiveness and receptivity to signals (Figure 1.2) and it is this functional level that marks the implementation of those assembly rules. Within the increasingly multicellular embryo, cell interactions and inductions are initiated, cell lineages are established, and morphogenesis, growth and histogenesis proceed. Thus, interactions of proteins, cells and tissues during development generate progressively higher-order complexity (Figure 1.2), from the one-dimensional complexity of the genotype and primary protein structure to the three-dimensionally complex phenotype. Embryonic development is therefore a typical generative programme. From a limited range of fundamental cell properties, an almost infinite range of complex phenotypes can be built, simply by deploying these cell properties in varying ways. The diverse range of phenotypic from across extant and extinct species bears witness to the morphogenetic power of these basic cell properties over an evolutionary time-scale.

Thus, it is the morphogenetic potential of cell properties and the mechanisms of embryonic development that causally link genotype and phenotype. And from this brief and perhaps simplistic rationalization, one can see that during development there will be significant, higher-order events taking place in the absence of direct genetic control but which are themselves the inevitable consequences of genetic specification (Figure 1.2, from the level of 'cell properties' upwards). Thus, the phenotypic expression of an individual's genotype is influenced by a variety of non-genetic factors that might involve variables such as diet, infection and ageing. These factors can have direct effects on gene expression or may influence more subtle control mechanisms, such as DNA or histone methylation. This class of phenomena is sometimes described as epigenetic and, clearly, much morphological complexity is generated within this so-called epigenetic domain (Alberch, 1982; McLachlan, 1986; Gottesman and Hanson, 2005).

Developmental biologists are interested in defining assembly rules and elucidating their operation at tissue, cellular and molecular/genetic levels. To understand dysmorphogenesis it is necessary to clarify what happens when certain assembly rules are either mis-specified or wrongly interpreted and a birth defect results. Clearly, understanding a particular birth defect involves much more than simply identifying a mutated gene or an environmental teratogen. It requires knowledge of the consequences of these on the mechanisms operating within the embryo, an understanding of how the generative programme has been perturbed and how that produces an abnormal phenotype. Furthermore, just as an understanding of normal development can help clarify abnormal development, so analysis of abnormal development can sometimes throw light on hitherto unknown aspects of normal mechanisms.

Before leaving this topic, it should be noted that in Figure 1.2 there is feedback indicated from proteins to genes (see reverse arrow). This reflects the fact that the role of some proteins is to bind to DNA, typically in a highly sequence-specific manner. Genes that encode such proteins are referred to as 'regulatory genes' and the proteins themselves known as 'transcription factors', since they control (either upregulate or downregulate) transcriptional activity of the gene to which they have bound. In essence, genes work in hierarchies, with regulatory genes controlling the expression of 'downstream' genes and with elements of 'cross-talk' between regulatory genes themselves. The definition of such genetic cascades and signalling pathways is a very topical issue in contemporary developmental biology and this is reflected by the prominence given to it by many of the contributors to this volume. Such genes are, of course, pivotally important in the normal life of the cell, in its synthetic and metabolic activity, homeostasis and proliferation, but during embryonic development they have multiple and crucial roles in determining cell fate. Although many of the genes identified to date as being involved in birth defects encode enzymes or structural proteins, it is not surprising that numerous families of regulatory genes have been established as playing important roles in dysmorphogenesis (see later).

Having discussed some aspects of the genotype-phenotype relationship, it is now appropriate to point out that it can be simplistic to always interpret dysmorphogenesis on the basis of a 'one gene:one (dysmorphic) phenotype' model. It is clear that, in some cases, a diversity of phenotypes can emerge from mutations in a single gene, each disease or dysmorphic phenotype reflecting a different mutation within that gene. Thus, different mutations in the receptor tyrosine kinase gene, RET, can result in familial medullary thyroid carcinoma, multiple endocrine neoplasia types 2A and 2B (all of which accords with its original recognition as an oncogene) and in Hirschsprung's disease, a developmental anomaly of the gut (reviewed by Mani et al., 2001; and see Chapter 11). This last disorder appears to be the consequence of a failure of RET-expressing neural crest cells to migrate normally and establish a parasympathetic innervation to the gut. The thyroid cancer-associated syndromes all result from mutations causing specific amino acid substitutions that apparently alter the functionality of the receptor tyrosine kinase encoded by RET (i.e. gain-of-function mutations that may lead to hyperplasia of the RET-expressing tissues). In contrast, the Hirschsprung mutations comprise deletion, insertion, frameshift, nonsense and missense mutations that lead to a loss of function. The phenotype can be explained as due to haploinsufficiency, whereby a threshold sensitivity to absence of 50% of the gene product (due to a mutated allele) is sufficient to perturb the development of the cells normally expressing that particular gene. In this case, it is the neural crest progenitors of the gut parasympathetic neurones that are affected, leaving other RET-expressing cell populations in the embryo apparently unscathed, due to tissue-specific differences in the threshold sensitivity (Mani et al., 2001). Interestingly, RET mutations that affect one of four extracytoplasmic cysteine residues have been found in Hirschsprung's patients, as well as patients with MEN2A and familial medullary thyroid carcinoma. These findings have raised the idea that a single mutation has opposing effects, depending on the tissue in which RET is expressed, and results in uncontrolled proliferation in endocrine cell types and apoptosis in enteric neurons (reviewed in Mani et al., 2001). Furthermore, mutations in RET are found only in about half of the familial cases of Hirschsprung's disease and then frequently with variable penetrance. This suggests a higher level of complexity, involving the interaction of other genes or non-coding variants, often referred to as modifiers. Co-inheritance of mutations in distinct loci but with additive effect gives rise to a multi- or polygenic inheritance model. In this case, each of the individual mutations alone may be considered risk factors, as they are insufficient to cause the phenotype alone but do so when inherited together.

The causality of birth defects is not necessarily genetic in origin and various aetiological categories can be recognized:

Chromosomal anomalies (e.g. trisomies, translocations)

Polygenic disorders Single gene mutations

Environmental/teratogenic factors Multifactorial aetiology Unknown aetiology

Each of these six categories presents its own set of problems in determining how a particular birth defect is generated (see Chapters 2, 3, 4 and 6). It might be argued that events occurring within the epigenetic domain referred to earlier can be extended to environmental influences on development. The embryo does not occupy a completely protected and privileged environment and, in some respects, is as open to effects from its environment as the neonate, juvenile or adult. Indeed, the recognition that the intrauterine experience of the fetus is strongly influenced by maternal nutritional or hormonal status is pivotal in determining later susceptibility to a number of adult diseases, such as diabetes and coronary heart disease (reviewed by Barker, 1995).

Clearly, the phenotype, be it adult or embryonic, is always the product of the combined effects of genetic and environmental influences (Sykes, 1993), but the relative contributions of each can differ for each aspect of the phenotype (Figure 1.3). Thus, Down's syndrome, as a trisomy disorder, reflects a condition that is 100% genetic, whereas a neural tube defect such as spina bifida (see Chapter 8) may have a strong environmental component in its aetiology, coupled with a possible genetic predisposition in some cases (reviewed by Marsh, 1994).

(Continues...)

Excerpted from Embryos, Genes and Birth Defects Copyright © 2006 by John Wiley & Sons, Ltd. Excerpted by permission.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.