Abstract

Iwashita T, Kruger GM, Pardal R, Kiel MJ, Morrison SJ (Howard Hughes Medical Institute and Departments of Internal Medicine and Cell and Developmental Biology, University of Michigan, Ann Arbor, Michigan). Hirschsprung disease is linked to defects in neural crest stem cell function. Science 2003;301:972–976. Several gene mutations have been identified in Hirschsprung’s disease, a condition characterized by distal gut aganglionosis. The mechanisms by which these genetic defects produce aganglionosis are unknown. Interest has focused on aberrant stem cell function in eliciting disease in other conditions. In the present investigation, Iwashita and colleagues quantified expression of genes associated with Hirschsprung’s in gut neural crest stem cells from rat fetus and characterized stem cell responses to gene products that are defective in Hirschsprung’s disease. Gene expression profiles from isolated gut neural crest stem cells were compared with whole fetus RNA from 14.5-day rat fetuses using oligonucleotide arrays consisting of 26,379 probe sets. RNA was amplified by in vitro transcription and hybridized to the array sets. Genes corresponding to 475 probe sets were expressed ≥3-fold higher in stem cells, whereas genes relating to 970 probe sets were selectively expressed in whole fetus RNA. Of 10 genes with the greatest increased expression in the stem cells, 4 (Ret, Sox10, Gfra-1, and EDNRB) are defective in Hirschsprung’s disease. To confirm that these 4 up-regulated genes were stem cell-specific, expressions were compared in restricted neural progenitors and more differentiated cells from fetal days 14.5 and 19.5 and post-natal day 4. In these cells, expression of Ret, Sox10, Gfra-1, and EDNRB were reduced, whereas other up-regulated genes including dopamine-β-hydroxylase involved in mature neuronal neurotransmitter generation were unchanged or increased. Observed gene expressions in neural crest stem cells were complemented by investigations testing stem cell function. Glial cell line-derived neurotrophic factor (GDNF) acting on Ret receptors promotes migration of cultured neural crest cells. Upon immunostaining with Ret antibodies, most gut neural crest stem cells expressed Ret protein, whereas other cell types did not. Guts from 13.5-day to 14.5-day fetuses were cultured in collagen gels. GNDF in the gels promoted neural crest cell migration, whereas gels without GDNF exhibited little response. Migrated cells were extracted from the gels and cultured at clonal density. Approximately 2.5% of migrating cells formed neural crest stem cell colonies, which were enriched >13-fold in gels with GDNF. Finally in Ret−/− mice, neural crest stem cells were reduced 4-fold in the esophagus and 20-fold in the stomach and intestine. This investigation confirmed that genes associated with Hirschsprung’s disease are up-regulated in gut neural crest stem cells. Products of these genes are necessary for normal caudad migration of neural crest stem cells. The authors concluded that Hirschsprung’s may be caused by defects in neural crest stem cell function. Hirschsprung’s disease presents with refractory fecal retention secondary to absent hindgut intramural ganglia and affects 1 of 5000 newborns. Observations suggest a significant genetic component to its development with incompletely penetrant dominant inheritance in familial cases (Nature 1994;367:378–380). Four percent of siblings of affected individuals are born with the disorder and males are 4-fold more likely to have the disease (Clin Genet 1998;54:39–44). Early pedigree analyses of familial cases mapped a chromosomal abnormality to proximal 10q (Hum Mol Genet 1993;2:1803–1808). Over the past decade, several gene mutations have been associated with Hirschsprung’s disease. Defects in the Ret (REarranged during Transfection) gene mapped to 10q11.2 were the first characterized (Nature 1994;367:377–378). The protein product of Ret is a transmembrane tyrosine kinase that transduces signals for growth and differentiation (Nature 1994;367:380–383). Immunohistochemical staining for Ret protein is reduced in aganglionic segments in Hirschsprung’s disease ( J Pediatr Surg 1995;30:433–436). Mutations involving the tyrosine kinase domain and extracellular regions that regulate Ret protein maturation and intracellular transport are observed in disease and decrease Ret protein amount or promote loss of function (Nature 1994;367:378–380, EMBO J 1996;15:2717–2725, J Clin Invest 1998;101:1415–1423). Evidence supports pathogenic roles for Ret defects in Hirschsprung’s disease. Enteric neurons and superior cervical ganglia are absent in mice homozygous for Ret mutations (Nature 1994;367:380–383, Development 1996;122:349–358). Introduction of missense mutations associated with Hirschsprung’s disease into wild-type Ret causes loss of function in mouse fibroblast and rat pheochromocytoma lines (Nat Genet 1995;10:35–40, EMBO J 1996;15:2717–2725). Defects in the glial cell line-derived neurotrophic factor (GDNF) gene also are described in Hirschsprung’s disease. GDNF is the ligand for Ret, is expressed in the developing nervous system, and is a neuronal survival factor (Nature 1996;382:70–73, Nature 1996;382:73–76). Whereas Ret immunoreactivity is found in neural cell bodies, GDNF-like immunoreactivity is prominent in gut neural fibers (Gastroenterology 1997;112:1381–1385). GDNF-deficient mice develop pyloric stenosis and duodenal dilation and show deficits in enteric neurons, in dorsal root ganglia, and sympathetic and nodose neurons (Nature 1996;382:70–73, Nature 1996;382:76–79). GDNF mutations seem to be less crucial than Ret defects to produce Hirschsprung’s disease. GDNF variants were observed in 2 of 36 cases in one series and GDNF levels were similar in aganglionic segments in another study (Hum Mol Genet 1996;5:2023–2026, Gastroenterology 1997;112:1381–1385). Furthermore, GDNF mutations do not prevent Ret activation (Eur J Hum Genet 2002;10:183–187). Other gene mutations are associated with Hirschsprung’s disease. Gfra-1, a coreceptor for GDNF, is expressed by neural crest-and noncrest-derived cells (Dev Biol 1998;204:385–406). Ten Gfra-1 polymorphisms were detected in 269 Hirschsprung’s cases, although similar variants occurred in normals ( J Med Genet 1999;36:217–220). Glial and neuronal Gfra-1 expression is reduced in aganglionic regions in some individuals (Lab Invest 2002;82:703–712). Gfra-1-deficient mice have absent enteric neurons but normal peripheral ganglia, in contrast to Ret- and GDNF-deficient animals (Neuron 1998;21:53–62, Neuron 1998;21:317–324). Abnormalities in endothelin-3 (EDN-3), endothelin receptor B (EDNRB), endothelin-converting enzyme (ECE1), and Phox2b transcription factor genes, as well as the gene encoding the Sry-related transcription factor Sox10, have been characterized in Hirschsprung’s disease (Hum Mol Genet 1998;7:1449–1452, J Ped Surg 2000;35:1017–1025, Gut 2003;52:563–567). Chromosomal susceptibility loci mapped to 3p21, 16q23, and 19q21 may represent disease penetrance modifiers (Nat Genet 2002;31:89–93, Nat Genet 2002;32:237–244). Embryonic enteric neuronal development in the intestine and colon requires antegrade colonization by advancing vagal neural crest cells. Hirschsprung’s is postulated to result from premature termination of neural crest cell advancement. However, other data raise the possibility that neural crest cells reach the anus but fail to differentiate or survive (Semin Pediatr Surg 1998;7:140–147). Migration of neural crest cells occurs in response to aborally propagating sources of GDNF in Ret-dependent fashion, demonstrating a chemoattractant role for GDNF (Dev Biol 2001;229:503–516, Development 2002;129:5151–5160). Contained in the advancing neural crest cells are neural crest stem cells, which are multipotent and self-renewing precursors of neural progenitors, glia, and more differentiated cells in the developing enteric nervous system. In vitro, neural crest cells form clonal neurospheres that contain stem cells as well as neuronal and glial cells in various stages of differentiation (Proc Natl Acad Sci U S A 2002;99:14506–14511). Neural stem cells share properties with stem cells in other regions. Hematopoietic stem cell enriched genes are expressed in neural stem cells (Proc Nat Acad Sci 2001;98:7934–7939). A recent investigation characterized 216 genes that are relatively enriched in mouse embryonic, hematopoietic, and neural stem cells, and some of which may produce specific transcription factors and signaling molecules (Science 2002;298:597–600, Blood 2002;99:488–498). A common molecular signature has been reported for different stem cell lines (Science 2002;298:601–604). However, stem cells in different regions exhibit region-specific properties. Transplanting gut neural crest stem cells into developing peripheral nerves promotes neural formation, whereas transplanting peripheral nerve stem cells promotes gliogenesis (Neuron 2002;35:643–656). Of genes associated with Hirschsprung’s disease, Sox10 is purported to function in maintaining stem cells as well as neural and glial progenitors (Neuron 2003;38:17–31). The investigation by Iwashita et al. is the first study to quantify expression of genes associated with Hirschsprung’s disease in neural crest stem cells and to measure stem cell responses to products of these genes. Information gleaned from this research is the most comprehensive delineation to date of possible cellular and molecular pathways that underlie development of Hirschsprung’s disease. Importantly, expressions of 4 genes defective in Hirschsprung’s were shown to be enriched in neural crest stem cells, supporting the hypothesis that defective production of these gene products interferes with stem cell function. The authors confirmed the chemoattractant properties of GDNF and showed that neural crest stem cells are among the cells attracted to GDNF. Finally, they quantified reductions in neural crest stem cells in Ret-deficient mice. These findings confirm that genetic defects associated with Hirschsprung’s can be linked to defective stem cell migration. This investigation takes the next step from correlating these gene defects to providing possible mechanisms by which these defects produce disease. There is little to criticize about the science of this study. Uncertainty arises when extrapolating findings of largely in vitro studies in animal models to a human disease. Although the authors provide a compelling mechanism linking the genetic abnormalities of Hirschsprung’s disease with defects in stem cell function, stem cell activities in fetuses with Hirschsprung’s have not been measured. Such experiments would be extremely difficult to perform. Importantly, this investigation may explain only a minority of cases of Hirschsprung’s disease. Ret, GDNF, EDNRB, EDN3, and Sox10 mutations are correlated with many long-segment Hirschsprung’s cases, but with few cases of short-segment disease (Nat Genet 2002;31:89–93). Ret mutations are more common in familial Hirschsprung’s (50%) than in sporadic cases (as low as 3% in some series) (Nat Genet 1996;14:345–347, J Med Genet 1999;36:771–774). However, the present study identified 6 genes up-regulated in neural crest stem cells, in addition to the 4 associated with Hirschsprung’s disease. Defects in these genes might be responsible for cases not yet correlated to a specific gene abnormality. Alternatively, other cases may stem from abnormalities in genes not selectively expressed in stem cells or in pathways unrelated to stem cell function. One group has postulated that Hirschsprung’s disease results from a hostile extracellular matrix that precludes normal neuronal development (Semin Pediatr Surg 1998;7:140–147). It is uncertain if the findings of this study can be extended to other heritable motility disorders. Sox10 mutations are found in one form of intestinal pseudo-obstruction without aganglionosis (Hum Genet 2002;111:198–206). However, Ret, GDNF, EDNRB, and EDN3 mutations were not observed in intestinal neuronal dysplasia (Gut 2001;48:671–675). Further investigations are warranted to explore genetic factors and stem cell abnormalities in these other conditions.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call