Animal models of ventral body wall closure defects A personal perspective on gastroschisis.код для вставкиСкачать
American Journal of Medical Genetics Part C (Seminars in Medical Genetics) 148C:186– 191 (2008) A R T I C L E Animal Models of Ventral Body Wall Closure Defects: A Personal Perspective on Gastroschisis TREVOR WILLIAMS* Malformations affecting the ventral body wall comprise one of the leading categories of human birth defects. Gastroschisis is a particularly important body wall closure defect as its incidence is rising worldwide. Although the occurrence of such defects is relatively common their molecular and cellular basis is very poorly understood. A robust animal model system to study the etiology of gastroschisis would be very useful, but several problems currently hamper the identification of such a model. A concerted effort is required to recognize, characterize, and classify ventral body wall defects in animal model species so that progress can be made in determining the mechanisms of ventral body wall closure during human development as well as combating the increased incidence of gastroschisis worldwide. ß 2008 Wiley-Liss, Inc. KEY WORDS: gastroschisis; omphalocele; umbilical hernia; ventral body wall; animal models How to cite this article: Williams T. 2008. Animal models of ventral body wall closure defects: A personal perspective on gastroschisis. Am J Med Genet Part C Semin Med Genet 148C:186–191. INTRODUCTION Ventral body wall closure abnormalities are common human birth defects present in about one out of every 2000 live births [Chabra and Gleason, 2005; Feldkamp et al., 2007]. Aberrant formation and closure of the ventral body wall in humans can result in several pathologies with distinct appearances and these can be understood and classified with respect to the involvement of the umbilical cord and associated umbilical ring [Brewer and Williams, 2004a; Chabra and Gleason, 2005; Feldkamp Trevor Williams is the Timpte/Brownlie Chair in Craniofacial/Molecular Biology and an Associate Professor at the University of Colorado Denver Anschutz Medical Campus. He received his undergraduate and graduate training in Pathology and Molecular Biology in England before moving to U.C. Berkeley for post-doctoral studies. His research interest include the transcription factors and signaling molecules regulating vertebrate embryonic development with particular focus on birth defects of the face, eye, limb, and body wall. *Correspondence to: Trevor Williams, Department of Craniofacial Biology, Mailstop 8120, P.O. Box 6511, Aurora, CO 80045. E-mail: firstname.lastname@example.org DOI 10.1002/ajmg.c.30179 ß 2008 Wiley-Liss, Inc. et al., 2007; Vauthay et al., 2007]. Such defects include umbilical cord hernia, omphalocele, and gastroschisis (see Figs. 1 and 2) of which the latter two have the greatest clinical impact for newborns and are the costliest to treat (http://www.marchofdimes.com/aboutus/ 680_2173.asp). There are now several animal models for omphalocele including the Tcfap2a (AP-2a) knockout mouse and a chick model using cadmium as a teratogen [Brewer and Williams, 2004b; Thompson and Bannigan, 2007]. However, animal models with gastroschisis are rare, and a definitive model of an isolated human gastroschisis is currently lacking. The absence of a suitable model system adversely impacts our understanding of the human animal models with gastroschisis are rare, and a definitive model of an isolated human gastroschisis is currently lacking. The absence of a suitable model system adversely impacts our understanding of the human pathology pathology. Below, I discuss issues of terminology that have complicated the characterization of gastroschisis in animals as well as suggestions to identify such models. An umbilical cord hernia occurs when the ventral body wall has closed normally about the umbilical cord, but a portion of the gut occupies the base of the cord along with the normal blood vessels. This pathology may be caused by incomplete retraction of the midgut into the body cavity during earlier stages of development. In omphalocele the umbilical ring, which is the transition zone between the ectoderm and mesoderm of the body wall and the amnion, is abnormal and enlarged. During normal development the umbilical ring closes so that it is the same size as the circumference of the umbilical cord by the end of the fetal period. A mature body wall covers the ventral surface surrounding the ring and the cord. With omphalocele, the umbilical ring is much larger than the insertion point of the cord and the mature body wall is limited to the peripherary of the defect. Typically, the umbilical cord will be in its normal position at the centre of this defect, but unlike an umbilical hernia, the gut contents are not present in the cord. Instead, the visceral contents including ARTICLE AMERICAN JOURNAL OF MEDICAL GENETICS PART C (SEMINARS IN MEDICAL GENETICS) 187 Figure 1. Graphical representation of the normal appearance of the ventral body wall in a newborn mouse along with the three body wall closure defects discussed in this review. Each panel shows a whole mount ventral view of the mouse torso (left) and a transverse section through the region of the umbilical cord (right). A, amnion; D, dorsal; G, gut; L, liver; T, tail; UC, umbilical cord; VBW, ventral body wall. the liver and gut can protrude through the enlarged ring into the amniotic sac. The extent of the omphalocele is variable and the effected area may occupy just the immediate vicinity of the cord, or it may be far more extensive and include both the abdomen and the thorax. In gastroschisis, the body wall closure defect and the cord are often juxtaposed, but can also be separated by a strip of mature body wall. Thus, as opposed to an umbilical cord hernia or an omphalocele, the cord is not necessarily part of the affected region of Figure 2. Lateral views of E18.5 mice with a normal ventral body wall (A), an omphalocele (B), or an apparent gastroschisis (C). UC, umbilical cord. The mouse in Panel C presented as a rare isolated event in a litter in the laboratory of the author. the ventral body wall in a gastroschisis. The defect is normally to the right of the umbilicus and presents with loops of bowel extending out of the fissure in the body wall with no amniotic covering. Note that in an omphalocele the visceral contents can also sometimes rupture I would therefore suggest the following system to identify a gastroschisis in any vertebrate species: loops of bowel extending through body wall with no amnionic covering; a normal umbilical cord (both in length and morphology); an umbilical ring that is not unusually large. 188 AMERICAN JOURNAL OF MEDICAL GENETICS PART C (SEMINARS IN MEDICAL GENETICS) through the amnion, so this aspect of the pathology is not a distinguishing feature of a gastroschisis. I would therefore suggest the following system to identify a gastroschisis in any vertebrate species: loops of bowel extending through body wall with no amnionic covering; a normal umbilical cord (both in length and morphology); an umbilical ring that is not unusually large. Unresolved questions with regards to human gastroschisis include its developmental origins, the mechanism responsible for the right side laterality bias, the overall incidence of skin bridges in the condition, and if the presence or absence of skin bridges provide clues to the etiology of this birth defect. Unfortunately, to date, studies in other animal species have been unable to provide insight into these issues. The classification of human ventral body wall closure defects listed above was essentially adopted in the 1950s [Chabra and Gleason, 2005; Chabra, 2007]. Before that point, all ventral body wall closure defects were generally referred to as a ‘‘gastroschisis’’. However, the switch in terminology with respect to gastroschisis in the human condition—where it now has a much more specific meaning—does not appear to have been widely applied in veterinary pathology [Szabo, 1989]. This has meant a degree of confusion when ventral body wall defects in other mammalian species are compared with the human condition. Specifically, references to gastroschisis in studies on mouse, rats, or farm animals can cover any type of ventral body wall closure defect and not just a human ‘‘gastroschisis.’’ In many such reports the term gastroschisis is often used only in a table of observed defects with no definition or illustrations of the actual body wall phenotype. Therefore, without an overt description, illustration, or photodocumentation of the pathology, great care should be taken in extending findings on ‘‘gastroschisis’’ in animal models to the precisely defined human condition. One clear recommendation for the future is that nomenclature for body wall closure defects be consistent between human and other animal species. VENTRAL WALL BODY DEFECTS The nomenclature aside, there are several other problems facing scientists and veterinarians confronted with a ventral body wall closure defect. Foremost is that the process of normal ventral body wall closure is very poorly understood in comparison to many other developmental processes [Brewer and Williams, 2004a]. Thus, compared with for example limb, brain, eye, or pancreas development, there is no clear morphogenetic framework that can be used as a guide to interpret one’s findings [Kaufman, 1992; Kaufman and Bard, 1999; Rossant and Tam, 2002]. In vertebrates, there is no consensus regarding the tissues required, the cell shape changes that might drive the process, or the biological mechanisms responsible for movement and fusion of the body wall components. Similarly, few genes have been identified that regulate this process, and their expression patterns in the tissue layers purported to govern ventral body wall closure are almost completely unknown at present [Brewer and Williams, 2004a]. This is of considerable concern because it will tend to result in body wall closure defects receiving little attention in the many new mouse models being characterized compared with associated pathologies in developmental systems that are better understood. The study of body wall closure defects, particularly gastroschisis, in non-human mammals is also prone to some technical problems. If the offspring are studied after birth, it is likely that their mothers will cannibalize those newborns possessing body wall closure defects soon after parturition. Thus, animals will need to be observed and characterized during embryogenesis and/or immediately after parturition. Although this is an important consideration, it has not necessarily hampered studies on other pathologies such as neural tube closure defects. But another problem facing those wishing to find animal models for gastroschisis is the common method used for the isolation of embryos—particularly mouse embryos ARTICLE that constitute by far the largest experimental sample. In mice, a ventral body wall defect will not be readily apparent until after embryonic day 14.5. At these later stages of development, the common procedure to isolate the embryos would be to obtain an intact yolk sac. The yolk sac would then be breached and the embryo separated from the attached placenta by pulling on the intervening umbilical cord. This methodology almost inevitably leads to damage at the attachment site of the umbilicus to the abdominal wall. If a gastroschisis is present, it is unlikely that the thin strip of mature skin between the umbilicus and the body wall defect would remain intact. This would therefore tend to obscure the difference between the various types of possible body wall closure defects in any initial analysis. So, a further recommendation from this review is that embryos with a clear body wall closure defect are processed and characterized with the umbilical attachment to the placenta still intact. Notwithstanding the issues noted above, it would be very valuable to have a robust animal model of gastroschisis that could be used to probe the etiology of the pathology and to test specific treatments for efficacy at suppressing this birth defect, for example, maternal dietary supplements. Ventral body wall closure defects have been documented to occur occasionally in offspring from many vertebrate species including mice, rats, guinea pigs, rabbits, dogs, cats, horses, cattle, sheep, and pigs [Szabo, it would be very valuable to have a robust animal model of gastroschisis that could be used to probe the etiology of the pathology and to test specific treatments for efficacy at suppressing this birth defect, for example, maternal dietary supplements. ARTICLE 1989]. The incidence is often quite low, <1%, although in some studies of kittens and piglets the frequency of the defect has approached 5%. As noted above, it is often not possible to determine whether many of these defects are equivalent to the modern definition of a human gastroschisis or represent other types of body wall defect due to the lack of documentation. Previous analyses also failed to produce convincing data for a genetic basis for gastroschisis in these animal species. Nevertheless, it is worth noting that several studies on cats, mice, and rabbits indicate that ventral body wall closure defects are more prevalent in certain breeds, indicating a potential genetic component [Collins et al., 2006; Hillebrandt et al., 2003; Szabo, 1989]. Similarly, certain pig, rat, dog, cattle and horse breeds may have a predisposition to umbilical hernias [Szabo, 1989]. However, the low incidence of such ventral body wall closure defects in most of these animal species means that they do not currently provide a robust model system to follow the development and genetic linkage of this pathology. One exception is the inbred HLG/Ze mouse strain, in which it has been possible to map several loci that are responsible for an increase in ventral body wall closure defects following radiation exposure [Hillebrandt et al., 1998, 2003]. Also of interest is a large-scale genetic analysis of various traits in different dog breeds that could potentially be used to determine genetic linkage in breeds prone to this pathology [Szabo, 1989; Pennisi, 2007]. ANIMAL MODELS Clear evidence that gene mutations can cause ventral body wall closure defects has been obtained by forward and reverse genetic approaches in the mouse [Hubner et al., 2001; Spiegelstein et al., 2004; Brewer and Williams, 2004a; Ogi et al., 2005; Shimizu et al., 2005; Thumkeo et al., 2005; Goldman et al., 2006; Hart et al., 2006; Hirano et al., 2006; Wojcik et al., 2006; Sun et al., 2007]. Here, a growing list of genes and chromosomal loci are associated with body wall closure defects, AMERICAN JOURNAL OF MEDICAL GENETICS PART C (SEMINARS IN MEDICAL GENETICS) but again exactly how many of these mutants model human gastroschisis will require further study. For this reason, it would be very useful if there were a dedicated central repository with a trained pathologist to study and classify new mouse models. Such a center could be modeled on those available through The Jackson Laboratory, for example, the ‘‘Craniofacial Mutant Resource’’ (http://www.jax.org/cranio/index.html). Experimental models of gastroschisis and omphalocele can also be generated in chick, rodents, rabbits, and cats with various teratogens, by irradiation, or via other insults [Szabo, 1989; Baatout et al., 2002; Hillebrandt et al., 2003; Brewer and Williams, 2004a; Yan and Hales, 2006; Feldkamp et al., 2007; Thompson and Bannigan, 2007]. But here again it is often unclear if reports of gastroschisis are comparable with the modern human definition. Other issues in teratogenesis studies are a low frequency of the pathology and inconsistent phenotype in the extent of the body wall closure defect. Multiple overlapping anomalies in other developmental systems can also be problematic. Of note, a mouse model combining low dietary protein and zinc with excess exposure to carbon monoxide produced a high incidence (50%) of ventral body wall closure defects [Singh, 2003]. Several of the affected embryos had the hallmark of gastroschisis—loops of bowel extruding through a hole in the body wall separate from the umbilicus [J. Singh, personal communication]—although this was not documented photographically. These mice also had multiple additional developmental defects and so do not serve as a definitive model of human gastroschisis, which usually occurs as an isolated defect. Nevertheless, this teratogenic model may still serve as one of the best paradigms to study the development of gastroschisis during mammalian embryogenesis. Ideally, though, it would be valuable to identify a genetic model that produced a fully penetrant and consistent gastroschisis phenotype that could be followed throughout embryogenesis. Finally, given the prevalence of the mouse as both a genetic and teratogenic 189 model system to study ventral body wall closure defects it is pertinent to ask whether it represents the best possible system to model this process in human development. One concern is simply the difference in size between the newborn mouse and human. Thus, it might be imagined that the ability to form or sustain a small strip of skin between the umbilicus and a gastroschisis defect would be more difficult in the mouse than in the larger human. Under such circumstances, if the skin bridge broke the gastroschisis would ‘‘migrate’’ to the umbilicus and appear on first examination more like a small omphalocele that had subsequently ruptured through the amnion. A second problem concerns the different morphogenetic programs that occur during mouse and human development around gastrulation and shortly thereafter. At the end of gastrulation, around E8.0, the mouse embryo is a U-shaped structure with the primitive gut located on the outer convex surface, facing the antimesometrial pole of the implantation site. The yolk sac is attached at the anterior end of the foregut and the posterior end of the hindgut and loops back to surround the entire embryo before integrating with the ectoplacental cone at the mesometrial pole. Thus, the yolk sac cavity is in direct contact with the gut and also surrounds the entire embryo up until the ectoplacental cone. The mouse embryo undergoes a process of ‘‘turning’’ between E8.5-9.5 resulting in an inversion of the germ layers. This process establishes the final relative positions of the tissue layers associated with the ventral body wall [Brewer and Williams, 2004a]. Any defect in turning will therefore have a profound effect on mouse body wall formation. After turning, the association of the yolk sac with the embryo proper in the midgut region is lost, but the sac maintains its attachment at the periphery of the placenta. Consequently, the yolk sac eventually forms a fluid filled bag that surrounds the embryo throughout gestation [Kaufman, 1992; Kaufman and Bard, 1999]. In contrast, the human embryo develops from a flat disc with a ventrally located yolk sac, and does not undergo a 190 AMERICAN JOURNAL OF MEDICAL GENETICS PART C (SEMINARS IN MEDICAL GENETICS) dynamic turning process [Sadler and Feldkamp, 2008]. Instead, the germ layers maintain their same relative position and four body folds are formed on the ventral surface of the embryo. These four folds grow ventrally and eventually fuse to form the umbilical ring. During this process the yolk sac regresses and eventually may only exist as a remnant by mid gestation. Despite these differences, the mouse still provides the best model to probe the genetic causes of ventral body wall closure defects given the resources available for manipulating and mapping its genome. In future, though, it would be worthwhile to consider a survey of embryogenesis in other mammalian species to identify one that better matches the process of normal human body wall formation. In conclusion, slow but steady advances are occurring in our knowledge of ventral body wall closure in both human and mouse. However, much still needs to be done to achieve a basic understanding of normal development of this structure as well as its associated pathology. Recommendations for future analysis of body wall closure in vertebrate species are listed below: 1. Clear terminology–consistent with the human nomenclature. 2. Increased awareness of the clinical relevance of the animal data. 3. A better understanding of the normal process of ventral body wall closure and its pathologies in multiple species. 4. Mouse embryo dissections to leave the umbilical cord intact and undamaged. 5. Detailed photo-documentation and/ or overt description of pathology. 6. A central repository for mouse genetic models of ventral body wall closure defects. With the rising incidence of gastroschisis in the human population, the cause of this birth defect should be garnering increased attention. Perhaps a necessary first step to achieve this goal would be for those involved with body wall closure defects in human and animal model species to come together to standardize terminology, share findings, define issues, and advocate for resources to tackle these prevalent but understudied birth defects. ACKNOWLEDGMENTS I am most grateful to Irene Choi for preparing the illustrations used in Figure 1. I thank Dr. Jarnail Singh, Dr. Cory Brayton, Dr. Susan Little, Dr. Ramona Skirpstunas, Dr. Bethany Reid, Dr. Vida Melvin, Dr. Marcia Feldkamp, and members of the Utah Gastroschisis Workshop for stimulating advice and discussion that have shaped this perspective. I am also grateful to the two anonymous reviewers for their important clarifications and suggestions. I have tried to reference the primary literature when possible, but due to space constraints I have also relied heavily on previous review articles for overviews of various topics. I apologize for any oversights in this regard. REFERENCES Baatout S, Jacquet P, Michaux A, Buset J, Vankerkom J, Derradji H, Yan J, von Suchodoletz H, de Saint-Georges L, Desaintes C, Mergeay M. 2002. Developmental abnormalities induced by X-irradiation in p53 deficient mice. In Vivo 16: 215–221. Brewer S, Williams T. 2004a. Finally, a sense of closure? Animal models of human ventral body wall defects. Bioessays 26:1307– 1321. Brewer S, Williams T. 2004b. Loss of AP-2a impacts multiple aspects of ventral body wall development and closure. Dev Biol 267: 399–417. Chabra S. 2007. Gastroschisis: Brief early history. J Perinat Med 35:455. Chabra S, Gleason CA. 2005. Gastroschisis: Embryology, pathogenesis, epidemiology. NeoReviews 6:e493–e499. Collins MD, Eckhoff C, Weiss R, Resnick E, Nau H, Scott WJ Jr. 2006. Differential teratogenesis of all-trans-retinoic acid administered on gestational day 9.5 to SWV and C57BL/6N mice: Emphasis on limb dysmorphology. Birth Defects Res A Clin Mol Teratol 76:96–106. Feldkamp ML, Carey JC, Sadler TW. 2007. Development of gastroschisis: Review of hypotheses, a novel hypothesis, and implications for research. Am J Med Genet Part A 143A:639–652. Goldman DC, Hackenmiller R, Nakayama T, Sopory S, Wong C, Kulessa H, Christian JL. 2006. Mutation of an upstream cleavage site in the BMP4 prodomain leads to tissuespecific loss of activity. Development 133: 1933–1942. ARTICLE Hart AW, Morgan JE, Schneider J, West K, McKie L, Bhattacharya S, Jackson IJ, Cross SH. 2006. Cardiac malformations and midline skeletal defects in mice lacking filamin A. Hum Mol Genet 15:2457–2467. Hillebrandt S, Streffer C, Montagutelli X, Balling R. 1998. A locus for radiation-induced gastroschisis on mouse Chromosome 7. Mamm Genome 9:995–997. Hillebrandt S, Matern S, Lammert F. 2003. Mouse models for genetic dissection of polygenic gastrointestinal diseases. Eur J Clin Invest 33:155–160. Hirano M, Kiyonari H, Inoue A, Furushima K, Murata T, Suda Y, Aizawa S. 2006. A new serine/threonine protein kinase, Omphk1, essential to ventral body wall formation. Dev Dyn 235:2229–2237. Hubner CA, Stein V, Hermans-Borgmeyer I, Meyer T, Ballanyi K, Jentsch TJ. 2001. Disruption of KCC2 reveals an essential role of K-Cl cotransport already in early synaptic inhibition. Neuron 30:515–524. Kaufman MH. 1992. The atlas of mouse development. San Diego: Academic Press Limited. Kaufman MH, Bard JBL. 1999. The anatomical basis of mouse development. San Diego: Academic Press. Ogi H, Suzuki K, Ogino Y, Kamimura M, Miyado M, Ying X, Zhang Z, Shinohara M, Chen Y, Yamada G. 2005. Ventral abdominal wall dysmorphogenesis of Msx1/ Msx2 double-mutant mice. Anat Rec A Discov Mol Cell Evol Biol 284:424–430. Pennisi E. 2007. Genetics. The geneticist’s best friend. Science 317:1668–1671. Rossant J, Tam PPL. 2002. Mouse development: Patterning, morphogenesis, and organogenesis. San Diego, CA: Academic. xviii, 712 p. Sadler TW, Feldkamp ML. 2008. The embryology of body wall closure: Relevence to gastroschisis and other ventral body wall defects. Am J Med Genet Part C Semin Med Genet (in press). Shimizu Y, Thumkeo D, Keel J, Ishizaki T, Oshima H, Oshima M, Noda Y, Matsumura F, Taketo MM, Narumiya S. 2005. ROCK-I regulates closure of the eyelids and ventral body wall by inducing assembly of actomyosin bundles. J Cell Biol 168:941–953. Singh J. 2003. Gastroschisis is caused by the combination of carbon monoxide and protein-zinc deficiencies in mice. Birth Defects Res B Dev Reprod Toxicol 68: 355–362. Spiegelstein O, Mitchell LE, Merriweather MY, Wicker NJ, Zhang Q, Lammer EJ, Finnell RH. 2004. Embryonic development of folate binding protein-1 (Folbp1) knockout mice: Effects of the chemical form, dose, and timing of maternal folate supplementation. Dev Dyn 231:221–231. Sun J, Liu YH, Chen H, Nguyen MP, Mishina Y, Upperman JS, Ford HR, Shi W. 2007. Deficient Alk3-mediated BMP signaling causes prenatal omphalocele-like defect. Biochem Biophys Res Commun 360:238– 243. Szabo KT. 1989. Congenital malformations in laboratory and farm animals. San Diego: Academic Press. 313 p. Thompson JM, Bannigan JG. 2007. Omphalocele induction in the chick embryo by ARTICLE administration of cadmium. J Pediatr Surg 42: 1703–1709. Thumkeo D, Shimizu Y, Sakamoto S, Yamada S, Narumiya S. 2005. ROCK-I and ROCK-II cooperatively regulate closure of eyelid and ventral body wall in mouse embryo. Genes Cells 10:825–834. AMERICAN JOURNAL OF MEDICAL GENETICS PART C (SEMINARS IN MEDICAL GENETICS) Vauthay L, Mazzitelli N, Rittler M. 2007. Patterns of severe abdominal wall defects: Insights into pathogenesis, delineation, and nomenclature. Birth Defects Res 79:211–220. Wojcik SM, Katsurabayashi S, Guillemin I, Friauf E, Rosenmund C, Brose N, Rhee JS. 2006. A shared vesicular carrier allows synaptic 191 corelease of GABA and glycine. Neuron 50:575–587. Yan J, Hales BF. 2006. Depletion of glutathione induces 4-hydroxynonenal protein adducts and hydroxyurea teratogenicity in the organogenesis stage mouse embryo. J Pharmacol Exp Ther 319:613–621.