- Email: [email protected]
Colonic atresia without mesenteric vascular occlusion. The role of the fibroblast growth factor 10 signaling pathway
Journal of Pediatric Surgery (2005) 40, 390 – 396
www.elsevier.com/locate/jpedsurg
Colonic atresia without mesenteric vascular occlusion. The role of the fibroblast growth factor 10 signaling pathway Timothy J. Fairbanks, Robert C. Kanard, Pierre M. Del Moral, Fred G. Sala, Stijn P. De Langhe, Chrissy A. Lopez, Jacqueline M. Veltmaat, David Warburton, Kathryn D. Anderson, Saverio Bellusci, R. Cartland Burns* Department of Pediatric Surgery, Developmental Biology Program, Children’s Hospital Los Angeles, Los Angeles, CA 90027, USA Index words: Congenital intestinal atresia; Intestinal atresia; Colonic atresial; Fibroblast growth factor receptor 2b; Fgfr2b; Fibroblast growth factor 10; Fgf10; Gastrointestinal tract development; Mesenteric vascular accident
Abstract Background/Purpose: Colonic atresia occurs in 1:20,000 live births, offering a neonatal surgical challenge. Prenatal expression of fibroblast growth factor 10 (Fgf10), acting through fibroblast growth factor receptor 2b (Fgfr2b), is critical to the normal development of the colon. Invalidation of the Fgf10 pathway results in colonic atresia, inherited in an autosomal recessive pattern. Classically, disturbance of the mesenteric vasculature has been thought to cause many forms of intestinal atresia. The purpose of this study was to evaluate the role of vascular occlusion in the pathogenesis of colonic atresia. Methods: Wild type (Wt), Fgf10 / , and Fgfr2b / mutant mouse embryos were harvested from timed pregnant mothers. Immediately following harvest, filtered India ink was infused via intracardiac microinjection. The gastrointestinal tract was dissected, and photomicrographs of the mesenteric arterial anatomy were taken at key developmental time points. Results: Photomicrographs after India ink microinjections demonstrate normal, patent mesenteric cascades to the atretic colon at the time points corresponding to the failure of colonic development in the Fgf10 / and Fgfr2b / mutants. The mesenteric arterial anatomy of the colon demonstrates no difference between the Wt and mutant colonic atresia. Conclusions: The absence of embryonic expression of Fgf10 or its receptor Fgfr2b results in colonic atresia in mice. India ink microinjection is a direct measure of mesenteric arterial patency. Colonic atresia in the Fgf10 / and Fgfr2b / mutants occurs despite normal mesenteric vascular development. Thus the atresia is not the result of a mesenteric vascular occlusion. The patent colonic mesentery of the Fgf10 / and Fgfr2b / mutants challenges an accepted pathogenesis of intestinal atresia. Although colonic atresia can occur as a result of vascular occlusion, new evidence exists to suggest that a genetic mechanism may play a role in the pathogenesis of this disease. D 2005 Elsevier Inc. All rights reserved.
Presented at the 51st Annual Congress of the British Association of Paediatric Surgeons, Oxford, England, July 27-30, 2004. * Corresponding author. University of Virginia Health System, PO Box 800709, Charlottesville, VA 22908, USA. Tel.: +1 434 924 2476; fax: +1 434 924 2656. E-mail address: [email protected] (R.C. Burns). 0022-3468/05/4002-0018$30.00/0 D 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.jpedsurg.2004.10.023
Intestinal atresia is a significant clinical problem facing the neonatal patient population and can lead to many complicated problems including short bowel syndrome. The morbidity and mortality of intestinal insufficiency are significant and may lead to dependence on parenteral nu-
A nonvascular cause of colonic atresia trition, with infection, loss of vascular access, and hepatic failure complicating the course. Familial hereditary forms of duodenal, jejuno-ileal, colonic, and multiple atresias have been reported clinically. The associated presence of nongastrointestinal developmental defects raises the possibility of intestinal atresia being a heritable condition in some cases [1-20]. The developmental mechanisms of intestinal atresia remain incompletely understood. We have previously described the role of the Fgf10/Fgfr2b pathway on cecal, duodenal, and anorectal development [21-23]. Fgf10 expression is localized to the mesenchyme of the distal colon early in the normal development [23]. Fgfr2b is expressed ubiquitously throughout the epithelium of the entire gastrointestinal (GI) tract [24]. The function of these genes in the developing colon is incompletely understood. There are two classic theories of intestinal atresia. Tandler, in 1900, proposed a lack of recanalization of the solid cord stage of intestinal development as the cause of intestinal atresia in the duodenum [25]. Fifty-five years later, Louw and Barnard presented their observations on the origin of intestinal atresia in the midgut [26]. Observations of clinical pathological specimens led to the hypothesis that intestinal atresia was caused by an embryologic bvascular accident.Q Late term embryonic canine pup mesenteric vessels were ligated in utero and the pregnancies were allowed to continue to term. Two pups were found to have intestinal atresia after ligation of the mesenteric vessels, providing proof of their hypothesis. The intestinal atresia was characterized by proximal dilation and distal decompression. The purpose of the current study is to determine the role a bvascular accident Q (as described by Louw and Barnard) has in the development of intestinal atresia. The Fgf10 / and Fgfr2b / knockout mutant animal model is evaluated to determine the effect of deletion of embryonic GI growth factors on colonic development. The GI tracts of the Fgf10 / and Fgfr2b / mice are shown to demonstrate the phenotypic effect of homozygous deletion of either the Fgf10 or Fgfr2b gene. To assess the mesenteric blood supply to the colon, filtered India ink was given via intracardiac injection at time points corresponding to the development of colonic atresia in the Fgf10 / and Fgfr2b / mutant. The GI tract was dissected and photomicrographs were taken.
1. Materials and methods 1.1. Mutant embryos Fgf10 / and Fgfr2b / mice were generated as previously described in the C57Bl/6 murine strain [27,28]. C57Bl/6 (wild type [Wt]) littermates were used as controls. All embryos were the result of timed matings at stages E11.5, E12.5, E13.5, E14.5, and E18.5.
391 Mouse use is conducted under approved IACUC protocol #32-02.
1.2. Biological samples Embryos were removed from the uteri of timed pregnant Wt, Fgf10 +/ , and Fgfr2b +/ mice at stages E11.5 to E18.5. The GI tract was dissected from all other structures to demonstrate the intestinal phenotype. Photomicrographs of the specimen were taken as fresh samples.
1.3. Intracardiac India ink injection Higgins Black India ink (#4415) was passed through a 0.45-lm filter. Drummond Scientific Company glass micropipettes were held over a Bunsen burner and spun to a fine caliber microinjector. The uterus was removed from timed pregnant mothers and rinsed in room temperature phosphate buffered saline. The uteri and placenta were opened to reveal the Wt, Fgf10 / , and Fgfr2b / mutant embryos. Care was taken to leave the umbilical artery and vein intact connected to the placenta. The embryos were again rinsed in room temperature phosphate buffered saline. The embryos were then secured with forceps while India ink was injected into the left ventricle via the microinjector. Myocardial contraction distributed the India ink throughout the systemic circulation. India ink was infused until the cardiac contraction terminated. All India ink injections were performed by the same author (TJF) in a similar fashion in all mice at all stages. The GI tracts of the India ink–injected embryos were then dissected and photomicrographed as fresh samples. The small intestine was removed to demonstrate the mesenteric vasculature and colonic anatomy.
2. Results 2.1. Absence of Fgf10 or its receptor Fgfr2b results in colonic atresia The E18.5 stage is near complete murine gestation. This stage represents the murine equivalent to the pathology seen clinically in the neonate. Fig. 1 demonstrates the differences in the phenotype of the Wt, Fgf10 / , and Fgfr2b / colon at near complete gestation, E18.5. The mesentery has been removed to visualize the intestinal tube. The Wt or control colon is the intestinal tissue bordered proximally by the ileocecal valve and distally by cutaneous anus. The diameter of the colon, cecum, and terminal ileum is relatively equal at this stage. The Fgf10 / and Fgfr2b / mutants demonstrate a normal small intestine and a small atretic cecum without a lumen in the same location as the cecum in the Wt. A very proximal colonic atresia is present in both the Fgf10 / and Fgfr2b / mutants. The colonic atresia phenotype has
392
T.J. Fairbanks et al.
Fig. 1 Near term colonic phenotype. Invalidation of Fgf10 and Fgfr2b in the Fgf10 / and Fgfr2b / mutants results in colonic atresia. The gastrointestinal tract of Wt mice shown as control from terminal ileum (TI), normal cecum to the cutaneous structures of the anus (A). Fgf10 / and Fgfr2b / mutants were dissected and photographed as shown. The colon (Col) of all Fgf10 / and Fgfr2b / mutants was atretic, as was the cecum (Ce). The total colonic length of the mutants was approximately 15%-20% of the Wt control. All samples are shown at E18.5.
100% penetrance with a consistent phenotype. Unlike the cecal atresia previously described, the proximal colon has an epithelial lumen [21]. Most of the complex intestinal organogenesis occurs between stages E11.5 and E13.5 in the mouse. In Fig. 2, the E14.5 stage was examined to determine whether the colonic atresia was the result of an early developmental defect or late term regression of a normally developed structure. The colon of the Wt at stage E14.5 is a well-developed structure with both mesenchymal and epithelial elements. The colon of the Fgf10 / and Fgfr2b / mutants at stage E14.5 demonstrates a proximal atresia. The colon ends in a blind pouch approximately 15%-20% the length of the Wt colon. There is also a ribbon of mesenchymal tissue extending from the distal portion of the atretic colon to the pelvis and normal site of the anus. There is no epithelial component in the ribbon of mesenchymal tissue. The presence of colonic atresia at E14.5 indicates that the pathogenic event in the Fgf10 / and Fgfr2b / mutants is an error of early
organogenesis, not a late term regression of a normally developed structure.
2.2. Fgf10 / mutants develop intestinal atresia with patent mesenteric blood flow The colonic atresia of the Fgf10 / mutant occurs between stages E11.5 and E13.5. At stage E10.5, there is no phenotypic difference in the colon of the Wt, Fgf10 / , and Fgfr2b / mutant (data not shown). At stage E11.5, the colon is slightly shorter in the mutants compared with the Wt. This is the first phenotypic evidence of the deletion of Fgf10 in the colon. By E12.5 and E13.5, the colonic atresia is clearly present. The time points of E11.5, E12.5, and E13.5 were evaluated, as they represent the earliest time points at which the pathogenesis of colonic atresia is evident in the Fgf10 / mutant. Systemic arterial perfusion of the India ink is used to visually confirm the adequacy of arterial perfusion. The India
Fig. 2 Colonic atresia is the result of an early developmental defect. The colonic phenotype was examined at stage E14.5 to determine whether the atresia is the result of an early developmental defect or a late term regression of a normally developed structure. E14.5 stage was examined to determine whether the colonic atresia was the result of an early developmental defect or late term regression of a normally developed structure. The colon of the Wt is a well-developed structure with both mesenchymal and epithelial elements. Shown from the terminal ileum (TI) and cecum extending to the cutaneous anus (A). The colon of the Fgf10 / and Fgfr2b / mutants at stage E14.5 demonstrates a proximal atresia. The colon, shown from terminal ileum (TI) and cecum, ends in a blind pouch approximately 15%-20% the length of the Wt colon. A ribbon of mesenchymal tissue extends from the distal portion of atretic colon to the pelvis. There is no epithelial component in the ribbon of mesenchymal tissue.
A nonvascular cause of colonic atresia
393
ink is noted to infiltrate the superior mesenteric artery supply to the herniated second loop of intestine. Closer inspection of the cecum, small intestine, and colon of the Wt and Fgf10 / mutants shows India ink infiltration to the local arterial cascades supplying those end organs. This pattern is maintained throughout stages E11.5, E12.5, and E13.5. There is no difference in the pattern of mesenteric blood flow of the Wt and Fgf10 / mutants as shown by India ink infiltration of the mesenteric vasculature. Fig. 3 demonstrates Fgf10 / mutant and Wt whole embryos after India ink injection at E11.5. The absence of upper and lower limbs is characteristic of the Fgf10 / mutant. Note the preferential perfusion of the upper half of the embryo compared with the lower half in both Wt and Fgf10 / mutant seen at this stage. The infiltration of India ink in the mesenteric artery supply to the cecum, small intestine, and colon is shown. There are small perfusing arteries shown in both samples to all end organ structures. The Fgf10 /
Fig. 3 India ink infiltration at E11.5. Wt and Fgf10 / mutant whole embryos after India ink injection. The black arrows denote the upper and lower limbs of the Wt, and their absence is characteristic of the Fgf10 / mutant. Note at this stage the preferential perfusion of the upper half of the embryo compared with the lower half in both specimens shown. The gastrointestinal tract has been dissected. Note the infiltration of India ink in the mesenteric artery supply to the cecum, small intestine, and colon as shown by black arrows. The Fgf10 / mutant colon has not developed the atretic phenotype although it is slightly shorter. The early cecal bud of the Fgf10 / mutant is also slightly shorter. Note the presence of small perfusing tributaries to all end organ structures in both the Wt (W) and Fgf10 / mutant (10). There is no difference in the India ink infiltration in the Fgf10 / mutant colon compared with the Wt control.
Fig. 4 India ink infiltration at stage E13.5. A Wt and Fgf10 / mutant are shown at E13.5 as whole embryos, entire gastrointestinal tracts, and the proximal colon. The Wt embryo with normally developed upper and lower limbs and the Fgf10 / mutant with absent limb. Both specimens show the physiological ventral intestinal herniation that is normal at this stage of murine development. The terminal gastrointestinal tract from stomach (S) to anus is shown in the second row. The cluster of intestinal tissue seen at the end of a mesenteric stalk in both the Wt and Fgf10 / mutant is the portion found herniated outside of the abdomen. Note the Wt (W) colon extending from the ventral herniation to the anus. The colon is overlying the mesenteric blood supply, which is intact. The gastrointestinal tract of the Fgf10 / mutant (10) does not have distal colonic tissue. The small ribbon of mesenchymal tissue connecting the colon to the pelvis is shown with the gray arrow immediately adjacent to the patent superior mesenteric blood supply infused with India ink. The bottom row demonstrates the infiltration of India ink to the proximal colon. At stage E13.5 note the infiltration of India ink in the mesenteric artery supply to the terminal ileum, cecum, and proximal colon as shown by the yellow arrows in both the Wt and Fgf10 / mutant. The gray arrow shows colonic atresia of the Fgf10 / mutant. Note how well the India ink perfuses these tissues. The India ink infiltration of the colon shows no difference between Wt and Fgf10 / mutant at stage E13.5.
394 mutant colon has not developed the atretic phenotype, although it is slightly shorter compared with the Wt. The early cecal bud of the Fgf10 / mutant is also slightly shorter. There is no difference in the India ink infiltration in the Fgf10 / mutant colon compared with the Wt control. The mesenteric arterial blood flow remains patent at stages E12.5 and E13.5. Fig. 4 demonstrates the results of intracardiac India ink injection at stage E13.5. A Wt and Fgf10 / mutant are shown as whole embryos, dissected GI tracts, and the proximal colon. The Wt embryos develop normal upper and lower limbs, whereas the Fgf10 / mutant has no limbs. Both specimens show the physiological ventral intestinal herniation that is normal at this stage of murine development. The GI tract from stomach to anus is shown in the second row. The cluster of intestinal tissue seen at the end of a mesenteric stalk in both the Wt and Fgf10 / mutant is the portion of intestine found herniated outside of the abdomen. Note the Wt colon extending from the ventral herniation to the anus. The colon is overlying the intact mesenteric blood supply. The GI tract of the Fgf10 / mutant does not have distal colonic tissue. The small ribbon of mesenchymal tissue connecting the colon to the pelvis is shown adjacent to the patent superior mesenteric blood supply infused with India ink. The bottom row demonstrates the infiltration of India ink to the proximal colon. At stage E13.5, note the infiltration of India ink in the mesenteric artery supply to the terminal ileum, cecum, and proximal colon as shown in both the Wt and Fgf10 / mutant. The proximal colonic atresia of the Fgf10 / mutant is shown in a magnified view. Note how well the India ink perfuses the tissues of the atretic colon. The India ink infiltration of the colon shows no difference between Wt and Fgf10 / mutant at stage E13.5. As the colonic phenotype of the Fgfr2b / and Fgf10 / mutants are essentially equal, all India ink data are presented in the Fgf10 / mutant. Intracardiac Fgfr2b / mutant of India ink injection at E11.5, E12.5, and E13.5 demonstrated patent mesenteric blood supply to all intestinal tissues in a manner equivalent to the Fgf10 / mutant (data not shown).
3. Discussion Currently, there are two classically accepted theories for the cause of intestinal atresia. In 1900, Tandler proposed that intestinal atresias resulted from a failure in the revacuolization of the solid cord stage of duodenal development [25]. Tandler’s theory of a solid stage of development is not relevant to this model of intestinal atresia because the Wt control colon does not pass through a solid stage in which a canalization process occurs. Louw and Barnard proposed in 1955 that intestinal atresia resulted from an intrauterine mesenteric vascular accident of normally developed intestine late in gestation, resulting in necrosis and subsequent resorption [26]. Louw and Barnard presented very convincing bobservationsQ on
T.J. Fairbanks et al. the origin of intestinal atresia. Dr Louw noted clinical observations of atresias of the midgut. Of primary interest was the anomalous vascular supply, necrosis of the proximal blind end, and postoperative atony and ileus. A vascular hypothesis was tested with intrauterine ligation of the mesenteric vasculature in 2 canine pups. The pups’ resulting intestinal atresia confirmed their clinical observations and resulted in a generally accepted etiologic pathway for the development of intestinal atresia. The fibroblast growth factors (Fgfs) are a family of signaling molecules composed of at least 22 members involved in different aspects of organogenesis [17]. Fgf10 is associated with instructive mesenchymal/epithelial interactions, which occur during branching and budding morphogenesis. The expression of Fgf10 in the developing lung mesenchyme anatomically correlates with eventual sites of epithelial bud formation [29]. Fgf10 has also been shown to be a potent chemoattractant for the distal lung epithelium [30]. Mice deficient of Fgf10 exhibit multiple organ defects including lung, limb, and mammary gland [31,32]. The Fgf10 / mutants are viable embryos, but succumb to respiratory failure at birth because of lung agenesis. The Fgfs act through tyrosine kinase transmembrane receptors [33]. The Fgf receptor (Fgfr) gene family has genetic linkage to skeletal dysplasias and autosomal dominant craniosynostosis syndromes [34]. Thus far the Fgf receptor family has not been implicated in GI tract malformation in humans. In mammals, 4 Fgf receptors have been identified. They are composed of an extracellular domain containing 2 to 3 immunoglobulin-like domains, a transmembrane domain, and an intracellular tyrosine kinase domain. Alternative splicing of the Fgfr2 gene generates 2 isoforms, termed IIIb and IIIc, which will differentially bind different FGF ligands. Fgfr2IIIb is found mainly in epithelial cells and is activated by Fgf1, Fgf3, Fgf7, Fgf10, and Fgf22. Fgfr2bIIIc is located primarily in the mesenchyme and, with the exception of Fgf1, is activated by a different set of ligands [35]. A null mutation of Fgfr2 results in peri-implantation lethality at gestational age embryonic day 4.5 to 5.5 (E4.5E5.5) [36]. A homozygous hypomorphic allele results in death around E10.5 with no limb buds and a defective placenta [37]. A Cre-mediated excision was used to generate mice lacking the Fgfr2IIIb isoform (Fgfr2b). These mice retained a normal Fgfr2IIIc receptor. The Fgfr2b / mutant embryos, such as the Fgf10 / mutants, are viable until birth at which time they have respiratory failure secondary to lung agenesis. They also exhibit severe defects in the development of the limbs, anterior pituitary gland, mammary glands, salivary glands, inner ear, teeth, skin, and skull [31,32]. Similar, but not identical, defects were observed in Fgf10 / embryos suggesting that Fgfr2b is the primary receptor for Fgf10 during embryonic development. We have previously presented data demonstrating that invalidation of the Fgf10/Fgfr2b pathway plays a causal role in the pathogenesis of cecal atresia and, now, colonic atresia
A nonvascular cause of colonic atresia in a reproducible autosomal recessive inheritable pattern. The pathology of murine cecal atresia does not result in failure of luminal patency. It is plausible that a gene controlling the development of complex structures could be mutated or misexpressed, resulting in a malformation or developmental defect of the GI tract. It is equally plausible that a genetic abnormality could affect the development of the mesenteric blood supply and subsequently cause intestinal atresia. The India ink injection in both Wt and Fgf10 / mutants reveals perfusion to the main mesenteric arteries supplying the developing colon. There is also perfusion of terminal arterioles supplying the colon. There is no difference in the pattern of mesenteric blood flow of the Wt and Fgf10 / mutants. Although the mesenteric vasculature is clearly patent during the early stages of development, the Fgf10 / mutant colon fails to develop resulting in proximal colonic atresia. These data exclude an occlusive vascular event as the causal mechanism in the pathogenesis of intestinal atresia in the Fgf10 / mutant model of colonic atresia. One limitation of this experimental model is that the India ink microinjection demonstrates only patent arterial blood flow. The India ink is trapped in the small end organ capillaries and does not demonstrate the venous system; thus the venous system is not evaluated. Although we demonstrate the patent arterial development, we have not directly assessed local vasculogenesis or angiogenesis and the role of the Fgf10 pathway in their regulation. This is an area of active investigation for our laboratory. Although the classic work done by Louw and Barnard clearly proves that intestinal atresia can result from an intrauterine bvascular accidentQ (mesenteric arterial occlusion), the current data demonstrate that not all intestinal atresias should be thought to result from a vascular occlusion. The prenatal expression of growth factors, such as Fgf10, is critical to the normal development of the GI tract. In the absence of this and other critical developmental genes, developmental defects likely occur. The data presented here suggest intestinal atresia can result from a genetic defect in development as well as vascular occlusion.
References [1] Benawra R, Puppala BL, Mangurten HH, et al. Familial occurrence of congenital colonic atresia. J Pediatr 1981;99:435 - 6. [2] Berant M, Kahana D. Familial duodenal atresia. Arch Dis Child 1970;45:281 - 2. [3] Best LG, Wiseman NE, Chudley AE. Familial duodenal atresia: a report of two families and review. Am J Med Genet 1989;34:442 - 4. [4] DallaVecchia LK, Grosfeld JL, West KW, et al. Intestinal atresia and stenosis: a 25-year experience with 277 cases. Arch Surg 1998; 133:490 - 6. [5] Gibson MF. Familial multiple jejunal atresia with malrotation. J Pediatr Surg 1987;22:1013 - 4. [6] Gross E, Armon Y, Abu-Dalu K, et al. Familial combined duodenal and jejunal atresia. J Pediatr Surg 1996;31:1573. [7] Gungor N, Balci S, Tanyel FC, et al. Familial intestinal polyatresia syndrome. Clin Genet 1995;47:245 - 7.
395 [8] Guttman FM, Braun P, Garance PH, et al. Multiple atresias and a new syndrome of hereditary multiple atresias involving the gastrointestinal tract from stomach to rectum. J Pediatr Surg 1973;8: 633 - 40. [9] Kilani RA, Hmiel P, Garver MK, et al. Familial jejunal atresia with renal dysplasia. J Pediatr Surg 1996;31:1427 - 9. [10] Lemire EG, Evans JA, Giddins NG, et al. A familial disorder with duodenal atresia and tetralogy of Fallot. Am J Med Genet 1996; 66:39 - 44. [11] Melhem RE, Salem G, Mishalany H, et al. Pyloro-duodenal atresia. A report of three families with several similarly affected children. Pediatr Radiol 1975;3:1 - 5. [12] Mishalany HG, Najjar FB. Familial jejunal atresia: three cases in one family. J Pediatr 1968;73:753 - 5. [13] Mishalany HG, Der Kaloustian VM, Ghandour MH. Familial congenital duodenal atresia. Pediatrics 1971;47:633 - 4. [14] Mishalany HG, Der Kaloustian VM. Familial multiple-level intestinal atresias: report of two siblings. J Pediatr 1971;79:124 - 5. [15] Nixon HH, Tawes R. Etiology and treatment of small intestinal atresia: analysis of a series of 127 jejunoileal atresias and comparison with 62 duodenal atresias. Surgery 1971;69:41 - 51. [16] Nixon HH. Duodenal atresia. Br J Hosp Med 1989;41:134,138,140. [17] Ornitz DM, Itoh N. Fibroblast growth factors. Genome Biol 2001;2: [REVIEWS3005]. [18] Puri P, Fujimoto T. New observations on the pathogenesis of multiple intestinal atresias. J Pediatr Surg 1988;23:221 - 5. [19] Seashore JH, Collins FS, Markowitz RI, et al. Familial apple peel jejunal atresia: surgical, genetic, and radiographic aspects. Pediatrics 1987;80:540 - 4. [20] Thiele H, Musil A, Nagel F, et al. Familial arhinia, choanal atresia, and microphthalmia. Am J Med Genet 1996;63:310 - 3. [21] Burns RC, Fairbanks TJ, Sala F, et al. Requirement for fibroblast growth factor 10 or fibroblast growth factor receptor 2-IIIb signaling for cecal development in mouse. Dev Biol 2004;265:61 - 74. [22] Fairbanks TJ, Kanard RC, De Langhe SP, et al. Deletion of fibroblast growth factor receptor 2b (Fgfr2b) gene results in anorectal malformation in mouse. J Surg Res 2003;114:306 - 7. [23] Fairbanks TJ, De Langhe S, Sala FG, et al. Fibroblast growth factor 10 (Fgf10) invalidation results in anorectal malformation in mice. J Pediatr Surg 2004;39:360 - 5. [24] Burns RC, Fairbanks TJ, Sala F, et al. Requirement for fibroblast growth factor 10/fibroblast growth factor receptor 2-IIIb signaling for cecal development in mouse. Dev Biol 2004;265(1):61 - 74. [25] Tandler J. Zur Entwicklungsgeschichte des Menschlichen Duodenum in Fruhen Embryonalstadien. Morphol Jahrb 1900;29:187 - 216. [26] Louw JH, Barnard CN. Congenital intestinal atresia: observations on its origin. Lancet 1955;2:1065 - 7. [27] De Moerlooze L, Spencer-Dene B, Revest J, et al. An important role for the IIIb isoform of fibroblast growth factor receptor 2 (FGFR2) in mesenchymal-epithelial signalling during mouse organogenesis. Development 2000;127:483 - 92. [28] Sekine K, Ohuchi H, Fujiwara M, et al. Fgf10 is essential for limb and lung formation. Nat Genet 1999;21:138 - 41. [29] Bellusci S, Grindley J, Emoto H, et al. Fibroblast growth factor 10 (FGF10) and branching morphogenesis in the embryonic mouse lung. Development 1997;124:4867 - 78. [30] Park WY, Miranda B, Lebeche D, et al. FGF-10 is a chemotactic factor for distal epithelial buds during lung development. Dev Biol 1998;201:125 - 34. [31] De Moerlooze L, Spencer-Dene B, Revest J, et al. An important role for the IIIb isoform of fibroblast growth factor receptor 2 (FGFR2) in mesenchymal-epithelial signalling during mouse organogenesis. Development 2000;127:483 - 92. [32] Mailleux AA, Spencer-Dene B, Dillon C, et al. Role of FGF10/ FGFR2b signaling during mammary gland development in the mouse embryo. Development 2002;129:53 - 60.
396 [33] McKeehan WL, Wang F, Kan M. The heparin sulfate–fibroblast growth factor family: diversity of structure and function. Prog Nucleic Acid Res Mol Biol 1998;59:135 - 76. [34] De Moerlooze L, Dickson C. Skeletal disorders associated with fibroblast growth factor receptor mutations. Curr Opin Genet Dev 1997;7:378 - 85. [35] Ornitz DM. Regulation of chondrocyte growth and differentiation by fibroblast growth factor receptor 3. Novartis Found Symp 2001; 232:63 - 76.
T.J. Fairbanks et al. [36] Arman E, Haffner-Krausz R, Chen Y, et al. Targeted disruption of fibroblast growth factor (FGF) receptor 2 suggests a role for FGF signaling in pregastrulation mammalian development. Proc Natl Acad Sci U S A 1998;95:5082 - 7. [37] Xu X, Weinstein M, Li C, et al. Fibroblast growth factor receptor 2 (FGFR2)–mediated reciprocal regulation loop between FGF8 and FGF10 is essential for limb induction. Development 1998;125: 753 - 65.
www.elsevier.com/locate/jpedsurg
Colonic atresia without mesenteric vascular occlusion. The role of the fibroblast growth factor 10 signaling pathway Timothy J. Fairbanks, Robert C. Kanard, Pierre M. Del Moral, Fred G. Sala, Stijn P. De Langhe, Chrissy A. Lopez, Jacqueline M. Veltmaat, David Warburton, Kathryn D. Anderson, Saverio Bellusci, R. Cartland Burns* Department of Pediatric Surgery, Developmental Biology Program, Children’s Hospital Los Angeles, Los Angeles, CA 90027, USA Index words: Congenital intestinal atresia; Intestinal atresia; Colonic atresial; Fibroblast growth factor receptor 2b; Fgfr2b; Fibroblast growth factor 10; Fgf10; Gastrointestinal tract development; Mesenteric vascular accident
Abstract Background/Purpose: Colonic atresia occurs in 1:20,000 live births, offering a neonatal surgical challenge. Prenatal expression of fibroblast growth factor 10 (Fgf10), acting through fibroblast growth factor receptor 2b (Fgfr2b), is critical to the normal development of the colon. Invalidation of the Fgf10 pathway results in colonic atresia, inherited in an autosomal recessive pattern. Classically, disturbance of the mesenteric vasculature has been thought to cause many forms of intestinal atresia. The purpose of this study was to evaluate the role of vascular occlusion in the pathogenesis of colonic atresia. Methods: Wild type (Wt), Fgf10 / , and Fgfr2b / mutant mouse embryos were harvested from timed pregnant mothers. Immediately following harvest, filtered India ink was infused via intracardiac microinjection. The gastrointestinal tract was dissected, and photomicrographs of the mesenteric arterial anatomy were taken at key developmental time points. Results: Photomicrographs after India ink microinjections demonstrate normal, patent mesenteric cascades to the atretic colon at the time points corresponding to the failure of colonic development in the Fgf10 / and Fgfr2b / mutants. The mesenteric arterial anatomy of the colon demonstrates no difference between the Wt and mutant colonic atresia. Conclusions: The absence of embryonic expression of Fgf10 or its receptor Fgfr2b results in colonic atresia in mice. India ink microinjection is a direct measure of mesenteric arterial patency. Colonic atresia in the Fgf10 / and Fgfr2b / mutants occurs despite normal mesenteric vascular development. Thus the atresia is not the result of a mesenteric vascular occlusion. The patent colonic mesentery of the Fgf10 / and Fgfr2b / mutants challenges an accepted pathogenesis of intestinal atresia. Although colonic atresia can occur as a result of vascular occlusion, new evidence exists to suggest that a genetic mechanism may play a role in the pathogenesis of this disease. D 2005 Elsevier Inc. All rights reserved.
Presented at the 51st Annual Congress of the British Association of Paediatric Surgeons, Oxford, England, July 27-30, 2004. * Corresponding author. University of Virginia Health System, PO Box 800709, Charlottesville, VA 22908, USA. Tel.: +1 434 924 2476; fax: +1 434 924 2656. E-mail address: [email protected] (R.C. Burns). 0022-3468/05/4002-0018$30.00/0 D 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.jpedsurg.2004.10.023
Intestinal atresia is a significant clinical problem facing the neonatal patient population and can lead to many complicated problems including short bowel syndrome. The morbidity and mortality of intestinal insufficiency are significant and may lead to dependence on parenteral nu-
A nonvascular cause of colonic atresia trition, with infection, loss of vascular access, and hepatic failure complicating the course. Familial hereditary forms of duodenal, jejuno-ileal, colonic, and multiple atresias have been reported clinically. The associated presence of nongastrointestinal developmental defects raises the possibility of intestinal atresia being a heritable condition in some cases [1-20]. The developmental mechanisms of intestinal atresia remain incompletely understood. We have previously described the role of the Fgf10/Fgfr2b pathway on cecal, duodenal, and anorectal development [21-23]. Fgf10 expression is localized to the mesenchyme of the distal colon early in the normal development [23]. Fgfr2b is expressed ubiquitously throughout the epithelium of the entire gastrointestinal (GI) tract [24]. The function of these genes in the developing colon is incompletely understood. There are two classic theories of intestinal atresia. Tandler, in 1900, proposed a lack of recanalization of the solid cord stage of intestinal development as the cause of intestinal atresia in the duodenum [25]. Fifty-five years later, Louw and Barnard presented their observations on the origin of intestinal atresia in the midgut [26]. Observations of clinical pathological specimens led to the hypothesis that intestinal atresia was caused by an embryologic bvascular accident.Q Late term embryonic canine pup mesenteric vessels were ligated in utero and the pregnancies were allowed to continue to term. Two pups were found to have intestinal atresia after ligation of the mesenteric vessels, providing proof of their hypothesis. The intestinal atresia was characterized by proximal dilation and distal decompression. The purpose of the current study is to determine the role a bvascular accident Q (as described by Louw and Barnard) has in the development of intestinal atresia. The Fgf10 / and Fgfr2b / knockout mutant animal model is evaluated to determine the effect of deletion of embryonic GI growth factors on colonic development. The GI tracts of the Fgf10 / and Fgfr2b / mice are shown to demonstrate the phenotypic effect of homozygous deletion of either the Fgf10 or Fgfr2b gene. To assess the mesenteric blood supply to the colon, filtered India ink was given via intracardiac injection at time points corresponding to the development of colonic atresia in the Fgf10 / and Fgfr2b / mutant. The GI tract was dissected and photomicrographs were taken.
1. Materials and methods 1.1. Mutant embryos Fgf10 / and Fgfr2b / mice were generated as previously described in the C57Bl/6 murine strain [27,28]. C57Bl/6 (wild type [Wt]) littermates were used as controls. All embryos were the result of timed matings at stages E11.5, E12.5, E13.5, E14.5, and E18.5.
391 Mouse use is conducted under approved IACUC protocol #32-02.
1.2. Biological samples Embryos were removed from the uteri of timed pregnant Wt, Fgf10 +/ , and Fgfr2b +/ mice at stages E11.5 to E18.5. The GI tract was dissected from all other structures to demonstrate the intestinal phenotype. Photomicrographs of the specimen were taken as fresh samples.
1.3. Intracardiac India ink injection Higgins Black India ink (#4415) was passed through a 0.45-lm filter. Drummond Scientific Company glass micropipettes were held over a Bunsen burner and spun to a fine caliber microinjector. The uterus was removed from timed pregnant mothers and rinsed in room temperature phosphate buffered saline. The uteri and placenta were opened to reveal the Wt, Fgf10 / , and Fgfr2b / mutant embryos. Care was taken to leave the umbilical artery and vein intact connected to the placenta. The embryos were again rinsed in room temperature phosphate buffered saline. The embryos were then secured with forceps while India ink was injected into the left ventricle via the microinjector. Myocardial contraction distributed the India ink throughout the systemic circulation. India ink was infused until the cardiac contraction terminated. All India ink injections were performed by the same author (TJF) in a similar fashion in all mice at all stages. The GI tracts of the India ink–injected embryos were then dissected and photomicrographed as fresh samples. The small intestine was removed to demonstrate the mesenteric vasculature and colonic anatomy.
2. Results 2.1. Absence of Fgf10 or its receptor Fgfr2b results in colonic atresia The E18.5 stage is near complete murine gestation. This stage represents the murine equivalent to the pathology seen clinically in the neonate. Fig. 1 demonstrates the differences in the phenotype of the Wt, Fgf10 / , and Fgfr2b / colon at near complete gestation, E18.5. The mesentery has been removed to visualize the intestinal tube. The Wt or control colon is the intestinal tissue bordered proximally by the ileocecal valve and distally by cutaneous anus. The diameter of the colon, cecum, and terminal ileum is relatively equal at this stage. The Fgf10 / and Fgfr2b / mutants demonstrate a normal small intestine and a small atretic cecum without a lumen in the same location as the cecum in the Wt. A very proximal colonic atresia is present in both the Fgf10 / and Fgfr2b / mutants. The colonic atresia phenotype has
392
T.J. Fairbanks et al.
Fig. 1 Near term colonic phenotype. Invalidation of Fgf10 and Fgfr2b in the Fgf10 / and Fgfr2b / mutants results in colonic atresia. The gastrointestinal tract of Wt mice shown as control from terminal ileum (TI), normal cecum to the cutaneous structures of the anus (A). Fgf10 / and Fgfr2b / mutants were dissected and photographed as shown. The colon (Col) of all Fgf10 / and Fgfr2b / mutants was atretic, as was the cecum (Ce). The total colonic length of the mutants was approximately 15%-20% of the Wt control. All samples are shown at E18.5.
100% penetrance with a consistent phenotype. Unlike the cecal atresia previously described, the proximal colon has an epithelial lumen [21]. Most of the complex intestinal organogenesis occurs between stages E11.5 and E13.5 in the mouse. In Fig. 2, the E14.5 stage was examined to determine whether the colonic atresia was the result of an early developmental defect or late term regression of a normally developed structure. The colon of the Wt at stage E14.5 is a well-developed structure with both mesenchymal and epithelial elements. The colon of the Fgf10 / and Fgfr2b / mutants at stage E14.5 demonstrates a proximal atresia. The colon ends in a blind pouch approximately 15%-20% the length of the Wt colon. There is also a ribbon of mesenchymal tissue extending from the distal portion of the atretic colon to the pelvis and normal site of the anus. There is no epithelial component in the ribbon of mesenchymal tissue. The presence of colonic atresia at E14.5 indicates that the pathogenic event in the Fgf10 / and Fgfr2b / mutants is an error of early
organogenesis, not a late term regression of a normally developed structure.
2.2. Fgf10 / mutants develop intestinal atresia with patent mesenteric blood flow The colonic atresia of the Fgf10 / mutant occurs between stages E11.5 and E13.5. At stage E10.5, there is no phenotypic difference in the colon of the Wt, Fgf10 / , and Fgfr2b / mutant (data not shown). At stage E11.5, the colon is slightly shorter in the mutants compared with the Wt. This is the first phenotypic evidence of the deletion of Fgf10 in the colon. By E12.5 and E13.5, the colonic atresia is clearly present. The time points of E11.5, E12.5, and E13.5 were evaluated, as they represent the earliest time points at which the pathogenesis of colonic atresia is evident in the Fgf10 / mutant. Systemic arterial perfusion of the India ink is used to visually confirm the adequacy of arterial perfusion. The India
Fig. 2 Colonic atresia is the result of an early developmental defect. The colonic phenotype was examined at stage E14.5 to determine whether the atresia is the result of an early developmental defect or a late term regression of a normally developed structure. E14.5 stage was examined to determine whether the colonic atresia was the result of an early developmental defect or late term regression of a normally developed structure. The colon of the Wt is a well-developed structure with both mesenchymal and epithelial elements. Shown from the terminal ileum (TI) and cecum extending to the cutaneous anus (A). The colon of the Fgf10 / and Fgfr2b / mutants at stage E14.5 demonstrates a proximal atresia. The colon, shown from terminal ileum (TI) and cecum, ends in a blind pouch approximately 15%-20% the length of the Wt colon. A ribbon of mesenchymal tissue extends from the distal portion of atretic colon to the pelvis. There is no epithelial component in the ribbon of mesenchymal tissue.
A nonvascular cause of colonic atresia
393
ink is noted to infiltrate the superior mesenteric artery supply to the herniated second loop of intestine. Closer inspection of the cecum, small intestine, and colon of the Wt and Fgf10 / mutants shows India ink infiltration to the local arterial cascades supplying those end organs. This pattern is maintained throughout stages E11.5, E12.5, and E13.5. There is no difference in the pattern of mesenteric blood flow of the Wt and Fgf10 / mutants as shown by India ink infiltration of the mesenteric vasculature. Fig. 3 demonstrates Fgf10 / mutant and Wt whole embryos after India ink injection at E11.5. The absence of upper and lower limbs is characteristic of the Fgf10 / mutant. Note the preferential perfusion of the upper half of the embryo compared with the lower half in both Wt and Fgf10 / mutant seen at this stage. The infiltration of India ink in the mesenteric artery supply to the cecum, small intestine, and colon is shown. There are small perfusing arteries shown in both samples to all end organ structures. The Fgf10 /
Fig. 3 India ink infiltration at E11.5. Wt and Fgf10 / mutant whole embryos after India ink injection. The black arrows denote the upper and lower limbs of the Wt, and their absence is characteristic of the Fgf10 / mutant. Note at this stage the preferential perfusion of the upper half of the embryo compared with the lower half in both specimens shown. The gastrointestinal tract has been dissected. Note the infiltration of India ink in the mesenteric artery supply to the cecum, small intestine, and colon as shown by black arrows. The Fgf10 / mutant colon has not developed the atretic phenotype although it is slightly shorter. The early cecal bud of the Fgf10 / mutant is also slightly shorter. Note the presence of small perfusing tributaries to all end organ structures in both the Wt (W) and Fgf10 / mutant (10). There is no difference in the India ink infiltration in the Fgf10 / mutant colon compared with the Wt control.
Fig. 4 India ink infiltration at stage E13.5. A Wt and Fgf10 / mutant are shown at E13.5 as whole embryos, entire gastrointestinal tracts, and the proximal colon. The Wt embryo with normally developed upper and lower limbs and the Fgf10 / mutant with absent limb. Both specimens show the physiological ventral intestinal herniation that is normal at this stage of murine development. The terminal gastrointestinal tract from stomach (S) to anus is shown in the second row. The cluster of intestinal tissue seen at the end of a mesenteric stalk in both the Wt and Fgf10 / mutant is the portion found herniated outside of the abdomen. Note the Wt (W) colon extending from the ventral herniation to the anus. The colon is overlying the mesenteric blood supply, which is intact. The gastrointestinal tract of the Fgf10 / mutant (10) does not have distal colonic tissue. The small ribbon of mesenchymal tissue connecting the colon to the pelvis is shown with the gray arrow immediately adjacent to the patent superior mesenteric blood supply infused with India ink. The bottom row demonstrates the infiltration of India ink to the proximal colon. At stage E13.5 note the infiltration of India ink in the mesenteric artery supply to the terminal ileum, cecum, and proximal colon as shown by the yellow arrows in both the Wt and Fgf10 / mutant. The gray arrow shows colonic atresia of the Fgf10 / mutant. Note how well the India ink perfuses these tissues. The India ink infiltration of the colon shows no difference between Wt and Fgf10 / mutant at stage E13.5.
394 mutant colon has not developed the atretic phenotype, although it is slightly shorter compared with the Wt. The early cecal bud of the Fgf10 / mutant is also slightly shorter. There is no difference in the India ink infiltration in the Fgf10 / mutant colon compared with the Wt control. The mesenteric arterial blood flow remains patent at stages E12.5 and E13.5. Fig. 4 demonstrates the results of intracardiac India ink injection at stage E13.5. A Wt and Fgf10 / mutant are shown as whole embryos, dissected GI tracts, and the proximal colon. The Wt embryos develop normal upper and lower limbs, whereas the Fgf10 / mutant has no limbs. Both specimens show the physiological ventral intestinal herniation that is normal at this stage of murine development. The GI tract from stomach to anus is shown in the second row. The cluster of intestinal tissue seen at the end of a mesenteric stalk in both the Wt and Fgf10 / mutant is the portion of intestine found herniated outside of the abdomen. Note the Wt colon extending from the ventral herniation to the anus. The colon is overlying the intact mesenteric blood supply. The GI tract of the Fgf10 / mutant does not have distal colonic tissue. The small ribbon of mesenchymal tissue connecting the colon to the pelvis is shown adjacent to the patent superior mesenteric blood supply infused with India ink. The bottom row demonstrates the infiltration of India ink to the proximal colon. At stage E13.5, note the infiltration of India ink in the mesenteric artery supply to the terminal ileum, cecum, and proximal colon as shown in both the Wt and Fgf10 / mutant. The proximal colonic atresia of the Fgf10 / mutant is shown in a magnified view. Note how well the India ink perfuses the tissues of the atretic colon. The India ink infiltration of the colon shows no difference between Wt and Fgf10 / mutant at stage E13.5. As the colonic phenotype of the Fgfr2b / and Fgf10 / mutants are essentially equal, all India ink data are presented in the Fgf10 / mutant. Intracardiac Fgfr2b / mutant of India ink injection at E11.5, E12.5, and E13.5 demonstrated patent mesenteric blood supply to all intestinal tissues in a manner equivalent to the Fgf10 / mutant (data not shown).
3. Discussion Currently, there are two classically accepted theories for the cause of intestinal atresia. In 1900, Tandler proposed that intestinal atresias resulted from a failure in the revacuolization of the solid cord stage of duodenal development [25]. Tandler’s theory of a solid stage of development is not relevant to this model of intestinal atresia because the Wt control colon does not pass through a solid stage in which a canalization process occurs. Louw and Barnard proposed in 1955 that intestinal atresia resulted from an intrauterine mesenteric vascular accident of normally developed intestine late in gestation, resulting in necrosis and subsequent resorption [26]. Louw and Barnard presented very convincing bobservationsQ on
T.J. Fairbanks et al. the origin of intestinal atresia. Dr Louw noted clinical observations of atresias of the midgut. Of primary interest was the anomalous vascular supply, necrosis of the proximal blind end, and postoperative atony and ileus. A vascular hypothesis was tested with intrauterine ligation of the mesenteric vasculature in 2 canine pups. The pups’ resulting intestinal atresia confirmed their clinical observations and resulted in a generally accepted etiologic pathway for the development of intestinal atresia. The fibroblast growth factors (Fgfs) are a family of signaling molecules composed of at least 22 members involved in different aspects of organogenesis [17]. Fgf10 is associated with instructive mesenchymal/epithelial interactions, which occur during branching and budding morphogenesis. The expression of Fgf10 in the developing lung mesenchyme anatomically correlates with eventual sites of epithelial bud formation [29]. Fgf10 has also been shown to be a potent chemoattractant for the distal lung epithelium [30]. Mice deficient of Fgf10 exhibit multiple organ defects including lung, limb, and mammary gland [31,32]. The Fgf10 / mutants are viable embryos, but succumb to respiratory failure at birth because of lung agenesis. The Fgfs act through tyrosine kinase transmembrane receptors [33]. The Fgf receptor (Fgfr) gene family has genetic linkage to skeletal dysplasias and autosomal dominant craniosynostosis syndromes [34]. Thus far the Fgf receptor family has not been implicated in GI tract malformation in humans. In mammals, 4 Fgf receptors have been identified. They are composed of an extracellular domain containing 2 to 3 immunoglobulin-like domains, a transmembrane domain, and an intracellular tyrosine kinase domain. Alternative splicing of the Fgfr2 gene generates 2 isoforms, termed IIIb and IIIc, which will differentially bind different FGF ligands. Fgfr2IIIb is found mainly in epithelial cells and is activated by Fgf1, Fgf3, Fgf7, Fgf10, and Fgf22. Fgfr2bIIIc is located primarily in the mesenchyme and, with the exception of Fgf1, is activated by a different set of ligands [35]. A null mutation of Fgfr2 results in peri-implantation lethality at gestational age embryonic day 4.5 to 5.5 (E4.5E5.5) [36]. A homozygous hypomorphic allele results in death around E10.5 with no limb buds and a defective placenta [37]. A Cre-mediated excision was used to generate mice lacking the Fgfr2IIIb isoform (Fgfr2b). These mice retained a normal Fgfr2IIIc receptor. The Fgfr2b / mutant embryos, such as the Fgf10 / mutants, are viable until birth at which time they have respiratory failure secondary to lung agenesis. They also exhibit severe defects in the development of the limbs, anterior pituitary gland, mammary glands, salivary glands, inner ear, teeth, skin, and skull [31,32]. Similar, but not identical, defects were observed in Fgf10 / embryos suggesting that Fgfr2b is the primary receptor for Fgf10 during embryonic development. We have previously presented data demonstrating that invalidation of the Fgf10/Fgfr2b pathway plays a causal role in the pathogenesis of cecal atresia and, now, colonic atresia
A nonvascular cause of colonic atresia in a reproducible autosomal recessive inheritable pattern. The pathology of murine cecal atresia does not result in failure of luminal patency. It is plausible that a gene controlling the development of complex structures could be mutated or misexpressed, resulting in a malformation or developmental defect of the GI tract. It is equally plausible that a genetic abnormality could affect the development of the mesenteric blood supply and subsequently cause intestinal atresia. The India ink injection in both Wt and Fgf10 / mutants reveals perfusion to the main mesenteric arteries supplying the developing colon. There is also perfusion of terminal arterioles supplying the colon. There is no difference in the pattern of mesenteric blood flow of the Wt and Fgf10 / mutants. Although the mesenteric vasculature is clearly patent during the early stages of development, the Fgf10 / mutant colon fails to develop resulting in proximal colonic atresia. These data exclude an occlusive vascular event as the causal mechanism in the pathogenesis of intestinal atresia in the Fgf10 / mutant model of colonic atresia. One limitation of this experimental model is that the India ink microinjection demonstrates only patent arterial blood flow. The India ink is trapped in the small end organ capillaries and does not demonstrate the venous system; thus the venous system is not evaluated. Although we demonstrate the patent arterial development, we have not directly assessed local vasculogenesis or angiogenesis and the role of the Fgf10 pathway in their regulation. This is an area of active investigation for our laboratory. Although the classic work done by Louw and Barnard clearly proves that intestinal atresia can result from an intrauterine bvascular accidentQ (mesenteric arterial occlusion), the current data demonstrate that not all intestinal atresias should be thought to result from a vascular occlusion. The prenatal expression of growth factors, such as Fgf10, is critical to the normal development of the GI tract. In the absence of this and other critical developmental genes, developmental defects likely occur. The data presented here suggest intestinal atresia can result from a genetic defect in development as well as vascular occlusion.
References [1] Benawra R, Puppala BL, Mangurten HH, et al. Familial occurrence of congenital colonic atresia. J Pediatr 1981;99:435 - 6. [2] Berant M, Kahana D. Familial duodenal atresia. Arch Dis Child 1970;45:281 - 2. [3] Best LG, Wiseman NE, Chudley AE. Familial duodenal atresia: a report of two families and review. Am J Med Genet 1989;34:442 - 4. [4] DallaVecchia LK, Grosfeld JL, West KW, et al. Intestinal atresia and stenosis: a 25-year experience with 277 cases. Arch Surg 1998; 133:490 - 6. [5] Gibson MF. Familial multiple jejunal atresia with malrotation. J Pediatr Surg 1987;22:1013 - 4. [6] Gross E, Armon Y, Abu-Dalu K, et al. Familial combined duodenal and jejunal atresia. J Pediatr Surg 1996;31:1573. [7] Gungor N, Balci S, Tanyel FC, et al. Familial intestinal polyatresia syndrome. Clin Genet 1995;47:245 - 7.
395 [8] Guttman FM, Braun P, Garance PH, et al. Multiple atresias and a new syndrome of hereditary multiple atresias involving the gastrointestinal tract from stomach to rectum. J Pediatr Surg 1973;8: 633 - 40. [9] Kilani RA, Hmiel P, Garver MK, et al. Familial jejunal atresia with renal dysplasia. J Pediatr Surg 1996;31:1427 - 9. [10] Lemire EG, Evans JA, Giddins NG, et al. A familial disorder with duodenal atresia and tetralogy of Fallot. Am J Med Genet 1996; 66:39 - 44. [11] Melhem RE, Salem G, Mishalany H, et al. Pyloro-duodenal atresia. A report of three families with several similarly affected children. Pediatr Radiol 1975;3:1 - 5. [12] Mishalany HG, Najjar FB. Familial jejunal atresia: three cases in one family. J Pediatr 1968;73:753 - 5. [13] Mishalany HG, Der Kaloustian VM, Ghandour MH. Familial congenital duodenal atresia. Pediatrics 1971;47:633 - 4. [14] Mishalany HG, Der Kaloustian VM. Familial multiple-level intestinal atresias: report of two siblings. J Pediatr 1971;79:124 - 5. [15] Nixon HH, Tawes R. Etiology and treatment of small intestinal atresia: analysis of a series of 127 jejunoileal atresias and comparison with 62 duodenal atresias. Surgery 1971;69:41 - 51. [16] Nixon HH. Duodenal atresia. Br J Hosp Med 1989;41:134,138,140. [17] Ornitz DM, Itoh N. Fibroblast growth factors. Genome Biol 2001;2: [REVIEWS3005]. [18] Puri P, Fujimoto T. New observations on the pathogenesis of multiple intestinal atresias. J Pediatr Surg 1988;23:221 - 5. [19] Seashore JH, Collins FS, Markowitz RI, et al. Familial apple peel jejunal atresia: surgical, genetic, and radiographic aspects. Pediatrics 1987;80:540 - 4. [20] Thiele H, Musil A, Nagel F, et al. Familial arhinia, choanal atresia, and microphthalmia. Am J Med Genet 1996;63:310 - 3. [21] Burns RC, Fairbanks TJ, Sala F, et al. Requirement for fibroblast growth factor 10 or fibroblast growth factor receptor 2-IIIb signaling for cecal development in mouse. Dev Biol 2004;265:61 - 74. [22] Fairbanks TJ, Kanard RC, De Langhe SP, et al. Deletion of fibroblast growth factor receptor 2b (Fgfr2b) gene results in anorectal malformation in mouse. J Surg Res 2003;114:306 - 7. [23] Fairbanks TJ, De Langhe S, Sala FG, et al. Fibroblast growth factor 10 (Fgf10) invalidation results in anorectal malformation in mice. J Pediatr Surg 2004;39:360 - 5. [24] Burns RC, Fairbanks TJ, Sala F, et al. Requirement for fibroblast growth factor 10/fibroblast growth factor receptor 2-IIIb signaling for cecal development in mouse. Dev Biol 2004;265(1):61 - 74. [25] Tandler J. Zur Entwicklungsgeschichte des Menschlichen Duodenum in Fruhen Embryonalstadien. Morphol Jahrb 1900;29:187 - 216. [26] Louw JH, Barnard CN. Congenital intestinal atresia: observations on its origin. Lancet 1955;2:1065 - 7. [27] De Moerlooze L, Spencer-Dene B, Revest J, et al. An important role for the IIIb isoform of fibroblast growth factor receptor 2 (FGFR2) in mesenchymal-epithelial signalling during mouse organogenesis. Development 2000;127:483 - 92. [28] Sekine K, Ohuchi H, Fujiwara M, et al. Fgf10 is essential for limb and lung formation. Nat Genet 1999;21:138 - 41. [29] Bellusci S, Grindley J, Emoto H, et al. Fibroblast growth factor 10 (FGF10) and branching morphogenesis in the embryonic mouse lung. Development 1997;124:4867 - 78. [30] Park WY, Miranda B, Lebeche D, et al. FGF-10 is a chemotactic factor for distal epithelial buds during lung development. Dev Biol 1998;201:125 - 34. [31] De Moerlooze L, Spencer-Dene B, Revest J, et al. An important role for the IIIb isoform of fibroblast growth factor receptor 2 (FGFR2) in mesenchymal-epithelial signalling during mouse organogenesis. Development 2000;127:483 - 92. [32] Mailleux AA, Spencer-Dene B, Dillon C, et al. Role of FGF10/ FGFR2b signaling during mammary gland development in the mouse embryo. Development 2002;129:53 - 60.
396 [33] McKeehan WL, Wang F, Kan M. The heparin sulfate–fibroblast growth factor family: diversity of structure and function. Prog Nucleic Acid Res Mol Biol 1998;59:135 - 76. [34] De Moerlooze L, Dickson C. Skeletal disorders associated with fibroblast growth factor receptor mutations. Curr Opin Genet Dev 1997;7:378 - 85. [35] Ornitz DM. Regulation of chondrocyte growth and differentiation by fibroblast growth factor receptor 3. Novartis Found Symp 2001; 232:63 - 76.
T.J. Fairbanks et al. [36] Arman E, Haffner-Krausz R, Chen Y, et al. Targeted disruption of fibroblast growth factor (FGF) receptor 2 suggests a role for FGF signaling in pregastrulation mammalian development. Proc Natl Acad Sci U S A 1998;95:5082 - 7. [37] Xu X, Weinstein M, Li C, et al. Fibroblast growth factor receptor 2 (FGFR2)–mediated reciprocal regulation loop between FGF8 and FGF10 is essential for limb induction. Development 1998;125: 753 - 65.