Human Molecular Genetics

Human Molecular Genetics Advance Access originally published online on August 4, 2004
Human Molecular Genetics 2004 13(19):2289-2301; doi:10.1093/hmg/ddh243

This Article Abstract FREE Full Text (PDF) An erratum has been published All Versions of this Article: 13/19/2289    most recent ddh243v1 Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to citation manager Search for citing articles in: ISI Web of Science (24) Request Permissions Google Scholar Articles by Cantrell, V. A. Articles by Southard-Smith, E. M. Search for Related Content PubMed PubMed Citation Articles by Cantrell, V. A. Articles by Southard-Smith, E. M. Social Bookmarking          What's this?

Human Molecular Genetics, Vol. 13, No. 19 © Oxford University Press 2004; all rights reserved

Interactions between Sox10 and EdnrB modulate penetrance and severity of aganglionosis in the Sox10^Dom mouse model of Hirschsprung disease

V. Ashley Cantrell^1,, Sarah E. Owens^1,, Ronald L. Chandler^1,, David C. Airey², Kevin M. Bradley¹, Jeffrey R. Smith¹ and E. Michelle Southard-Smith^1,*

¹Division of Genetic Medicine, Department of Medicine, Vanderbilt University School of Medicine, 529 Light Hall, 2215 Garland Avenue, Nashville, TN, 37232-0275, USA and ²Department of Pharmacology, Vanderbilt University School of Medicine, 8148-A Medical Research Building III, 465 21st Avenue South, Nashville, TN 37232-8548, USA

Received May 16, 2004; Accepted July 20, 2004

	ABSTRACT

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS REFERENCES

 
Cumulative evidence suggests that Hirschsprung disease (HSCR) is the consequence of multiple gene interactions that modulate the ability of enteric neural crest (NC) cells to populate the developing gut. One of the essential genes for this process is the NC transcription factor Sox10. Sox10^Dom mice on a mixed genetic background show variation in penetrance and expressivity of enteric aganglionosis that are analogous to the variable aganglionosis seen in human HSCR families. The phenotype of Sox10^Dom mice in congenic lines indicates this variation arises from modifiers in the genetic background. To determine whether known HSCR susceptibility loci are acting as modifiers of Sox10, we tested for association between genes in the endothelin signaling pathway (EdnrB, Edn3, Ece1) and severity of aganglionosis in an extended pedigree of B6C3FeLe.Sox10^Dom mice. Single locus association analysis in this pedigree identifies interaction between EdnrB and Sox10. Additional analysis of F2 intercross progeny confirms a highly significant effect of EdnrB alleles on the Sox10^Dom/+ phenotype. The presence of C57BL/6J alleles at EdnrB is associated with increased penetrance and more severe aganglionosis in Sox10^Dom mutants. Crosses between EdnrB and Sox10 mutants corroborate this gene interaction with double mutant progeny exhibiting significantly more severe aganglionosis. The background strain of the EdnrB mutant further influences the phenotype of Sox10/EdnrB double mutant progeny implying the action of additional modifiers on this phenotype. Our data demonstrates that Sox10–EdnrB interactions can influence development of the enteric nervous system in mouse models and suggests that this interaction could contribute to the epistatic network producing variation between patients with aganglionosis.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS REFERENCES

 
Hirschsprung disease (HSCR) is defined by the absence or reduction of enteric ganglia in a variable portion of the distal gastrointestinal tract (OMIM no. 142623). The congenital aganglionosis, which is a hallmark of HSCR, arises when neural crest (NC) cells fail to completely colonize the developing gastrointestinal tract. Mutations in eight different genes have been associated with disease in HSCR patients (1–10). Alterations in the RET receptor tyrosine kinase are responsible in approximately half of familial cases and a lower percentage of sporadic patients (11–13). Smaller proportions of patients are attributable to mutations in the RET ligands, (GDNF and NTN), components of the endothelin signaling pathway (EDNRB, EDN3, ECE1) and the transcription factors (SOX10 and ZFHX1B). However, significant numbers of patients cannot be ascribed to mutations within the known HSCR susceptibility loci. Moreover, incomplete penetrance and inter-familial variation are commonly observed for mutations in HSCR genes (14). Recent studies suggest that the incomplete penetrance and variable severity of HSCR are due to complex interactions between known HSCR genes or undiscovered susceptibility loci (15–17). Variation in penetrance and severity of aganglionosis among family members carrying equivalent mutations in RET, EDNRB and SOX10 provides further evidence that additional factors in the genetic background modulate the severity of this disorder (9,12,18,19). This article describes investigation of the genetic factors that influence the penetrance and severity of aganglionosis in the Sox10^Dom mouse model of HSCR.

Mouse models have added greatly to our understanding of HSCR genetics and embryologic events that construct the enteric nervous system (ENS). Induced and spontaneous mutants have led to the initial identification of essential genes that are later found to be altered in human HSCR patients (20–22). Studies in mouse HSCR models have contributed to our understanding of individual gene function and the effects of mutations on ENS development (23–27). Gene interactions indicated by human studies can be tested further in mouse models (16,28). Two-locus crosses between EdnrB and Ret mouse mutants have substantiated genetic interactions between these loci that were first suggested by human studies (16,17), and have led to the investigation of the mechanism of this interaction (29). Moreover, de novo identification of the genetic factors or ‘modifiers’ that influence the variable penetrance and inheritance patterns of complex diseases like HSCR are facilitated in mouse models, where genetic background and input alleles can be controlled in genome-wide and candidate gene approaches (30).

Sox10^Dom mice exhibit reduction of enteric ganglia precursors and melanoblasts (22,23,31,32) that presents as megacolon and spotting of the ventrum, head and feet. The phenotype of Sox10^Dom mice models the aganglionic megacolon and hypopigmentation of Hirschsprung–Waardenburg syndrome (WS4) patients. WS4 patients with SOX10 alterations typically present with syndromic HSCR, defined as aganglionosis in association with other deficits (hypopigmentation, deafness, heterochromia iridis and peripheral neuropathy) (9,33–37). This study focuses on the variation in aganglionosis between individual Sox10^Dom mice and uses this variation to identify modifiers that influence the aganglionosis aspect of the phenotype.

Sox10^Dom mice carry a single base insertion in the Sox10 locus that prematurely truncates this NC transcription factor downstream of the DNA binding domain, producing a dominant negative form of the protein (22,32,38). This mutation originally arose on the C57BL/6J haplotype in an F1 hybrid background (B6C3FeLe.B6-a F1) and has been maintained by crosses of mutants with wild-type B6C3FeLe.B6-a F1 mice (Fig. 1A) (39). Initial characterization of enteric ganglia in these mutants suggested variation in aganglionosis phenotype (23,39) analogous to the variation in penetrance and expressivity of aganglionosis seen in human HSCR families. We have characterized aganglionosis in congenic lines of Sox10^Dom mutants on C57BL/6J and C3HeB/FeJ strains (B6.Sox10^Dom and C3Fe.Sox10^Dom, respectively) to establish the effect of genetic background on the extent of aganglionosis.

View larger version (37K): [in this window] [in a new window]   Figure 1. Effect of genetic background on variation of aganglionosis in Sox10Dom mice. (A) Propagation of the B6C3FeLe.Sox10Dom pedigree by crosses to wild-type (WT) B6C3FeLe-a F1 mice and potential haplotypes are illustrated. Sketched at upper left is a B6C3FeLe.Sox10Dom mouse (Dom) with belly spot crossed to a WT B6C3FeLe-a F1 at upper right. Several examples of potential haplotypes carried by the B6C3FeLe.Sox10Dom mouse are illustrated in brackets. The recombinant chromosomes in the B6C3FeLe.Sox10Dom mouse can be transmitted intact to its progeny, replaced by transmission of a nonrecombinant chromosome or subjected to additional recombination as shown in the second row of haplotype diagrams. For simplicity only intact parental chromosomes transmitted by the WT F1 are shown and marked by an apostrophe. At each generation in the pedigree Sox10Dom progeny are crossed to another WT B6C3FeLe-a F1. To conserve space, not all possible haplotypes are shown. (B) Staining of gastrointestinal tract from postnatal day 7–10 (P7–P10) Sox10Dom/+ mouse by AChE whole-mount histochemistry reveals areas of abnormal enteric ganglia. Normal ganglia appear as brown fibrillar staining in small intestine (top). Reduced density of staining marks hypoganglionosis (middle), whereas absence of AChE staining marks aganglionic regions (bottom). Extent of aganglionic regions (% aganglionosis) are quantified by measuring length of aganglionic segment/total length of gut, gastric sphincter to anus. (C) Plotted variation in severity of aganglionosis in B6C3FeLe.Sox10Dom pups. (D) Distribution of aganglionosis in Sox10Dom/+ mutants on B6 and C3Fe backgrounds illustrates the effect of genetic strain on severity of megacolon. Duo, duodenum; Ile, ileum; Ce, cecum; An, anus.

 
Genes in the endothelin signaling pathway play an essential role in ENS development. Endothelin signaling is required at embryonic stages (E11 and E12), when enteric progenitors are migrating from the foregut into the hindgut (40,41). Absence of appropriate endothelin signaling in mice (20,21,42) and human patients (5,7,19) results in short-segment aganglionosis. We hypothesized that the variation in phenotype of Sox10^Dom mice could be attributed to interactions between Sox10 and known HSCR susceptibility genes, like those in the endothelin signaling pathway. To determine whether genes in the endothelin signaling pathway are interacting with Sox10 to influence aganglionic megacolon in the Sox10^Dom model, we evaluated alleles at EdnrB, Edn3 and Ece1 in an extended pedigree of B6CeFeLe.Sox10^Dom mice. We report a highly significant effect of the EdnrB locus with B6 alleles more often present in Sox10^Dom mice affected by aganglionosis and more severe aganglionosis in Sox10^Dom mice homozygous for B6 alleles at EdnrB. Alleles at Edn3 and Ece1 were not associated with penetrance or extent of a ganglionosis in Sox10^Dom HSCR model. The effect of combining mutant alleles at Sox10 and EdnrB was investigated by crossing Sox10^Dom with EdnrB^sl and B6;129-EdnrB^tm1Ywa mutants. These double mutant progeny demonstrate that complete loss of EdnrB signaling in the context of Sox10^Dom results in near total aganglionosis. These findings are consistent with the hypothesis that the Sox10^Dom phenotype is modulated by interactions with other HSCR genes and specifically demonstrate that interaction between Sox10 and EdnrB can influence the severity of aganglionosis.

	RESULTS

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS REFERENCES

 
Effect of genetic background on aganglionosis phenotype in Sox10^Dom mutants
Within an extended pedigree of B6C3FeLe.Sox10^Dom mice (Fig. 1A, n=500, 20 generations), we observed significant variability in survival of individual Sox10^Dom mutants. Some pups succumb to megacolon by post-natal day 5 (P5), whereas other mice live longer than 18 months. To establish that the differences in survival of B6C3FeLe.Sox10^Dom mice correspond to intestinal phenotype, we evaluated the extent of aganglionosis in P7–P10 pups by whole-mount acetylcholinesterase (AChE) staining (Fig. 1B). We documented aganglionosis in mice moribund with megacolon before weaning (P21) and in older animals (>15 months) to establish that survival in these mutants depends on severity of aganglionosis. Only animals with <2% aganglionosis in the distal colon survive to old age (n=9), whereas pups afflicted by >10% aganglionosis do not survive to breeding age, P42 (n=10) (data not shown). Pups that are most severely affected by megacolon die between birth and P15. The range of aganglionosis in B6C3FeLe.Sox10^Dom pups (Fig. 1C) is similar to the incomplete penetrance and variable severity of aganglionosis commonly seen in familial HSCR (15,16,43).

To investigate the effect of genetic background on severity of aganglionosis, we backcrossed Sox10^Dom mutants to C57BL/6J (B6, congenic line past the 17th generation) and C3HeB/FeJ (C3Fe, congenic line past the 15th generation) inbred strains. These congenic lines exhibit distinct differences in death of pups due to megacolon, with only 15% of B6.Sox10^Dom pups surviving to weaning in contrast to 78% survival in our C3Fe.Sox10^Dom line (Table 1). To confirm that the differences in survival of B6.Sox10^Dom and C3Fe. Sox10^Dom lines correspond to severity of intestinal phenotype, we evaluated the percent length of intestine that is aganglionic in pups from these lines. Fifteen percent of C3Fe.Sox10^Dom pups (n=39) exhibited normal innervation, whereas none of the B6.Sox10^Dom pups (n=39) appeared normal and were either affected by aganglionosis (Fig. 1D) or hypoganglionosis. In the B6.Sox10^Dom line only four of 39 pups evaluated (10.2%) exhibited <2% aganglionosis. Fifteen percent of B6.Sox10^Dom pups survive megacolon to weaning, but by breeding age (P42) only 8% of total B6.Sox10^Dom animals survive. Thus we conclude that the difference in survival of Sox10^Dom mice on the two strain backgrounds is attributable to distinct differences in the severity of aganglionosis and suggest that other genes in the genetic background of these animals significantly modulate the severity of the Sox10^Dom phenotype.

View this table: [in this window] [in a new window]   Table 1. Effect of background strain on Sox10Dom viability

 
Association testing between endothelin pathway genes and severity of Sox10^Dom aganglionosis
To facilitate mapping of loci that modify severity of aganglionosis in B6C3FeLe.Sox10^Dom animals, we documented age, cause of death and collected DNA samples from individual mice in the B6C3FeLe.Sox10^Dom pedigree over more than 10 generations. Intercrosses to B6C3FeLe F1 wild-type mice were performed to avoid selection of alleles that might introduce survival bias (Fig. 1A). Individual animals within the B6C3FeLe.Sox10^Dom pedigree were classified based either on survival or the exact extent of aganglionosis in the intestine to identify the phenotypic extremes. Sox10^Dom/+ mice that survived to breeding age and propagated the pedigree were designated as a mildly affected cohort (mild, n=142). DNA samples and information were also collected for severely affected pups (P5–P29) that exhibited externally obvious megacolon or died secondary to megacolon (severe, n=47). Additionally, we evaluated extent of aganglionosis in a group of P7–P10 B6C3FeLe.Sox10^Dom/+ pups by whole-mount AChE staining (Fig 1B, n=389) and selected those animals with no aganglionosis (mild, n=30) or >25% aganglionosis (severe, n=34). These two groups represent the extremes of the survival and aganglionosis phenotypes for B6C3FeLe.Sox10^Dom mice.

To test for any association between HSCR phenotype and genes in the endothelin signaling pathway, we identified simple tandem repeat (STR) markers closely flanking the EdnrB, Edn3 and Ece1 loci that are polymorphic between the B6 and C3FeLe strains. Markers in the Ece1 interval were selected after identification of the mouse ortholog of the human and rat ECE1 locus on mouse chromosome 4 at 136 Mb (Materials and Methods). Custom software was used to identify microsatellite repeats in genomic sequence within or flanking loci by 250 kb (STRFinder, J.R. Smith, unpublished data). All markers with repeat units greater than 10 were evaluated in these regions. No polymorphic markers were identified in the Edn3 interval (n=97 screened) or in the 3' flank of EdnrB (n=14 screened). More distant markers from public databases (MGI, http://www.informatics.jax.org/ Ensembl, http://www.ensembl.org/Mus_musculus/) were evaluated to identify the closest flanking polymorphic markers for these regions (Table 2).

View this table: [in this window] [in a new window]   Table 2. Informative markers flanking genes in the endothelin signaling pathway

 
We genotyped the extreme phenotypes, mildly and severely affected animals, from the B6C3FeLe.Sox10^Dom pedigree to define haplotypes at EdnrB, Edn3 and Ece1 (Table 3) on the basis of our flanking markers. Animals whose haplotypes were recombinant between the flanking markers were excluded from the analysis. Haplotypes at each locus were evaluated independently by chi-square (²) for association with aganglionosis phenotype. There was no difference in distribution of B6 or C3FeLe alleles between the mild and severe cohorts at Edn3 or Ece1. However, B6 haplotypes at EdnrB were present more often in pups that did not survive due to severe aganglionic megacolon (P=0.0011).

View this table: [in this window] [in a new window]   Table 3. Haplotypes at EdnrB, Edn3 and Ece1 in phenotypic extremes of the B6C3FeLe.Sox10Dom pedigree

 
Selective genotyping of individuals from the tails of the phenotype distribution, as described earlier, gives more power to detect a quantitative trait locus (QTL) than using the total distribution (44). This approach has been broadly successful in first pass analysis to detect loci that are then analyzed further in the total distribution (45–47). Our analysis of the extreme phenotypes initially indicated a significant association at EdnrB in the subset of mice stratified by survival. To estimate the effect of EdnrB in the full B6C3FeLe.Sox10^Dom pedigree population, a larger cohort of B6C3FeLe.Sox10^Dom/+ pups (N=820) were genotyped with the STR markers flanking EdnrB (Table 2). The observed/expected numbers of B6/B6, B6/C3FeLe and C3FeLe/C3FeLe genotypes were 200/204.5, 419/410 and 201/205.5, respective, in the Hardy–Weinberg equilibrium [² (1)=0.395, P = 0.530]. Association analysis was performed at both chromosomal (presence of a B6 allele) and animal (number of B6 alleles) levels.

Logistic regression models revealed that B6 alleles increased the probability of penetrance of aganglionosis or total affected length, but not hypoganglionosis. This was true at both the chromosomal level and the animal level in the analysis. Use of a gut length covariate did not alter these conclusions. For modeling dosage, for example, each B6 allele increased the odds of aganglionosis by a factor of 1.258 [P<0.05, 95% CI 1.01–1.56; overall ² (1)=4.28, P<0.05; goodness-of-fit ² (1)=0.75, P=0.390]. For total affected length, each B6 allele increased the odds by a factor of 1.403 [P<0.05, 95% CI 1.02–1.93; overall ² (1)=4.35, P<0.05; goodness-of-fit ² (1)=0.06, P=0.804].

The effect of EdnrB genotype on the measured length of hypoganglionosis and aganglionosis and the total affected length (cm) was analyzed by linear regression. This analysis did not reveal any statistically significant relationship between either the presence or the number (dosage) of B6 alleles and severity of reduced AChE staining, although suggestive trends were evident for total affected length (0.10<P<0.20). Multiple non-parametric trend tests indicated a suggestive relation between the number of B6 alleles and total affected length (0.05<P<0.10), whereas a method of median regression showed a significant relation (P<0.05) between the number of B6 alleles and percent total affected length.

The difference observed between the high statistical significance of EdnrB alleles in the phenotypic extremes of the pedigree and the suggestive trends found in the entire distribution of the pedigree is not surprising. Because selective genotyping in the tails of the distribution is used to gain additional power for QTL detection by comparison with the total sample (44), it is expected that smaller effect would be observed in the entire distribution.

Effects of EdnrB alleles in Sox10^Dom F1 intercross progeny
To extend our analysis of the effects of EdnrB alleles on Sox10^Dom phenotype, we investigated the effect of this locus in progeny from a standard F1 intercross. Sox10^Dom/+ F2 progeny (n=1043) were generated from B6.Sox10^DomxC3Fe F1 mice crossed to wild-type B6C3Fe F1s and genotyped with STR markers flanking EdnrB (Table 2). The observed/expected numbers of B6, B6C3 and C3 genotypes were 275/253, 478/521 and 290/268, respectively, departing from the Hardy–Weinberg equilibrium [²(1)=7.72, P=0.007]. Because of this departure, analysis was performed at the animal rather than at the chromosomal level.

We evaluated the effect of B6 alleles at EdnrB on penetrance of Sox10^Dom enteric innervation deficits by logistic-regression analysis. This analysis revealed that the number of B6 alleles increased the probability of penetrance of both aganglionosis and hypoganglionosis (Fig. 2A). For aganglionosis, each B6 allele increased the odds by a factor of 1.874 [P<0.001, 95% CI 1.56–2.25; overall ² (1)=48.02, P<0.001; goodness-of-fit ² (1)=0.08, P=0.992]. For hypoganglionosis, each B allele increased the odds by a factor of 1.781 [P<0.001, 95% CI 1.37–2.31; overall ² (1)=19.38, P<0.001; goodness-of-fit ² (1)=0.11, P=0.783]. Because almost all animals with hypoganglionosis also had aganglionosis (except three), there is essentially no difference between the prediction of the presence of hypoganglionosis versus the presence of hypoganglionosis or aganglionosis (data not shown).

View larger version (33K): [in this window] [in a new window]   Figure 2. Effect of alleles at EdnrB locus on Sox10Dom gut phenotype. (A) Qualitative effects of B6 alleles at EdnrB on penetrance of enteric deficits in Sox10Dom/+ F2 mice. Proportion of affected animals in each genotype class is plotted to illustrate the relationship between presence of aganglionosis/hypoganglionosis and the number of B6 alleles. (B) Quantitative effects of B6 alleles on severity of enteric ganglia defects in Sox10Dom/+ F2 mice. Genotype means of the natural log transformed lengths of hypoganglionosis (hypo), aganglionosis (agang) and their sum (total), showing the relation to number of B6 alleles. Mean extent of gut length affected is greater for groups of mice with B6 or B6C3 genotypes at EdnrB. Error bar=1 standard deviation.

 
We investigated the effect of B6 alleles at EdnrB on the extent of affected gut in Sox10^Dom mice by linear-regression analysis. This analysis revealed a significant statistical relation between the number of B6 alleles and the measured length of aganglionosis (N=680), the summed length of aganglionosis and hypoganglionosis (N=917), but not the length of hypoganglionosis (N=914) (Fig. 2B). Using a codominant genetic model to predict the length of aganglionosis in affected animals, the addition of each B6 allele increased aganglionosis 0.073 ln units, 0.21 standard deviations, or an 9% increase in means between each genotype class (F=14.01, P<0.001). The total variance in aganglionosis explained by the model was R²=2.1% (F=7.29, P<0.001). The total affected length was also related to the number of B6 alleles, by 0.108 ln units per allele, 0.28 standard deviations, or an 12% increase in means between each genotype class (F=42.38, P<0.001). The total variance in affected length explained by the EdnrB model was R²=4.2% (F=20.04, P<0.001). No dominance effects were detectable in any of these models.

To augment the results from linear regression, non-parametric methods were applied to test the relation between the number of B6 alleles and both the raw and natural log transformed length of affected gut. Both a non-parametric trend test and a method of median regression showed a significant relation (P<0.05) between the number of B6 alleles and the raw or transformed length of aganglionosis and total affected length, but not the length of hypoganglionosis (Fig. 3).

View larger version (16K): [in this window] [in a new window]   Figure 3. Sox10Dom phenotype variation within EdnrB genotype classes of F2 mice. Box plots of raw (cm) and natural log transformed [ln(x+1)] lengths of hypoganglionosis, aganglionosis and the summed total length are plotted against the number of B6 alleles in each genotype class (0, 1, 2). White bars indicate median values. Outlier values are indicated by dots. Positive trends in the medians are present and significant (P<0.05) for aganglionosis and the total lengths, but not for hypoganglionosis length. Residual large spread within each of the genotype classes is present.

 
Thus, the number of B6 alleles (0, 1 or 2) increases both the amount of aganglionosis in affected animals and the proportion of animals with aganglionosis. However, although the number of B6 alleles does not appear to affect the amount of hypoganglionosis by our methods of measurement, the number of B6 alleles does increase the probability of presence of hypoganglionosis.

To address potential complicating effects of gender and gut length, we examined covariates for these factors in each of the parametric models described earlier (due to missing data in covariates, the total N for these models was reduced to 1019). In logistic models, inclusion of simple covariates and two-way interaction terms resulted in no substantive changes in significance or interpretation for the effect of EdnrB alleles on aganglionosis or hypoganglionosis. In linear-regression models, the only effect that was noted from inclusion of covariates was the discovery of a significant dominance by gender interaction in the prediction of aganglionosis length or total affected gut length. This interaction is manifested by the mean of male EdnrB heterozygotes being closer to the mean of the male homozygotes carrying two B6 alleles at EdnrB, whereas the mean of female EdnrB heterozygotes is closer to the mean of female homozygotes carrying two C3Fe alleles at EdnrB. This effect is not apparent when gender is not identified (Fig. 2) because the direction of dominance is opposite in each sex, canceling its detection. Besides this aspect, the effect of the number of B6 alleles at EdnrB seems robust to differences between genders or differences in gut length between individual mice.

Phenotype of EdnrB/Sox10 double mutants
The EdnrB^s-l allele (piebald lethal) corresponds to a 1 cM deletion spanning the EdnrB locus on mouse chromosome 14 (48). Aganglionic megacolon in EdnrB^s-l mutants is recessive and produces aganglionosis confined to a short segment of the distal colon (49). To test the gene interaction indicated by our genetic analysis, we crossed EdnrB^s-l/s-l mutants with Sox10^Dom lines. Matings of EdnrB^s-l/s-lxB6.Sox10^Dom produced both genotype classes at the anticipated ratios. EdnrB^s-l/+,Sox10^Dom/+ offspring did not exhibit greater aganglionosis than Ednrb^s–l/+, Sox10^+/+ littermates (Fig. 4A, P=0.3). A few EdnrB^s-l/+;Sox10^Dom/+ pups born (4 of 33 total) were found dead before P10; however, this rate of loss was not greater than the rate of death secondary to megacolon observed for B6.Sox10^Dom/+ litters.

View larger version (69K): [in this window] [in a new window]   Figure 4. Progeny from crosses of EdnrB and Sox10 mutant lines demonstrate in vivo interaction between these genes. (A) A litter of pups from EdnrBs-l/s-lxB6.Sox10Dom/+ crosses. Mean extent of aganglionosis and total lengths of gut affected are tabulated according to genotype. Ventral surfaces of two pups display typical spotting for these genotype classes. (B) A litter of pups from EdnrBs-l/+,Sox10Dom/+xEdnrBs-l/+ crosses are shown. EdnrBs-l/s-l,Sox10Dom/+ pups from EdnrBsl/+,Sox10Dom/+xEdnrBsl/+ crosses are severely runted by P7 and exhibit extreme aganglionosis of the gastrointestinal tract. (C) Abnormalities of enteric ganglia in proximal duodenum of EdnrBs-l/s-l,Sox10Dom/+ and EdnrBs-l/s-l,Sox10+/+ mice. AChE stained guts were imaged in the 1 cm interval below the gastric sphincter (20x magnification). All panels are oriented with proximal gut at top and distal gut below. Image of EdnrBs-l/s-l,Sox10Dom/+ gut reveals few residual AChE stained neurons and transition zone to distal aganglionic region. Image of EdnrBs-l/s-l,Sox10+/+ gut shows decreased density and intensity of AChE stained fibers relative to Sox10Dom/+ and wild-type littermate stained in parallel.

 
To evaluate the effect of complete absence of EdnrB signaling on the Sox10^Dom phenotype, we established Ednrb^s-l/+, Sox10^Dom/+xEdnrB^s-l/+,Sox10^+/+ crosses. Pups of all six genotype classes were born and EdnrB^s-l/s-l,Sox10^Dom/+ pups were present in numbers nearing the anticipated frequency. However, EdnrB^s-l/s-l,Sox10^Dom/+ pups were readily distinguished from littermates due to severe runting as early as P2, and upon evaluation exhibited extreme levels of aganglionosis (Fig. 4B). Of the 11 EdnrB^s-l/s-l, Sox10^Dom/+ pups born, gastrointestinal tracts from eight were evaluated by whole-mount AChE staining. In all of the EdnrB^s-l/s-l,Sox10^Dom/+ pups, aganglionosis extended from the anus to the upper duodenum and was accompanied by abnormalities of remaining neural fibers in the duodenum (Fig. 4C).

During our documentation of enteric deficits in mice from Ednrb^s-l/+,Sox10^Dom/+xEdnrB^s-l/+,Sox10^+/+ crosses, we observed abnormalities of enteric ganglia in the small intestines of EdnrB^s-l animals that have not been reported previously. Original studies of EdnrB^s-l/s-l mutants relied on histological cross-sections to count ganglia in the large intestines of these mice (49) and reported complete absence in the distal-most millimeters of the intestine. AChE whole-mount staining also identifies aganglionosis in the distal colons of EdnrB^s-l/s-l mice. However, using this method, we consistently observed abnormal patterning of AChE stained fibers in the small intestine of EdnrB^s-l/s-l mutants. The overall plexus architecture appeared normal in these mutants, but the spacing between ganglia was increased, imparting a honeycomb appearance to the enteric network (Fig. 4C). In addition, the intensity of AChE stain associated with interconnecting neuron fibers was noticeably reduced in comparison with wild-type littermates that were stained in parallel. This decrease in interneuron fiber intensity extended throughout the small intestine to the proximal ileum (data not shown). EdnrB^s-l/+ heterozygotes exhibited a similar reduction in fiber density. This aspect of the EdnrB^s-l phenotype is underappreciated and yet consistent with the recent reports of reduced numbers of enteric neural crest stem cells (NCSC) in the embryonic gut of EdnrB^sl rats (50).

The EdnrB^s-l allele is an extensive deletion that could disrupt regulatory elements and thus expression of other developmentally important genes in the interval (51). To determine that the severe aganglionosis we observed in EdnrB^s-l/s-l, Sox10^Dom/+ pups was attributable to EdnrB and not other genes in the piebald lethal deletion interval (51), we crossed B6;129-EdnrB^tm1Ywa knockout mice with B6.Sox10^Dom mutants. B6;129-EdnrB^tm1Ywa knockout mice have been specifically engineered to replace a 4.2 kb gene fragment carrying exon 3 with a neomycin selection cassette that causes premature termination and complete loss of EDNRB expression (20). We established crosses to generate EdnrB^tm1Ywa/+,Sox10^Dom/+ animals anticipating that analogous to the EdnrB^s-l allele we would need to backcross these to observe a phenotype in EdnrB^{tm1Ywa/tm1Ywa},Sox10^Dom/+ mice. Unexpectedly, all EdnrB^tm1Ywa/+,Sox10^Dom/+ pups succumbed to megacolon before or just shortly after weaning (P13–P29, n=8), unlike EdnrB^sl/+,Sox10^Dom/+ pups that were viable well past breeding age (n=4). B6;129-EdnrB^tm1YwaxSox10^Dom/+ crosses were expanded to obtain additional progeny for quantitative assessment of the gut phenotype by whole-mount AChE. EdnrB^tm1Ywa/+,Sox10^Dom/+ pups exhibited a mean affected gut length of 38% (n=21) exhibiting either hypoganglionosis or aganglionosis, in contrast to their EdnrB^+/+,Sox10^Dom/+ littermates with mean gut length affected of only 16% (n=15). The increased severity of aganglionosis in EdnrB^tm1Ywa/+,Sox10^Dom/+ double mutants confirms that the effect seen in EdnrB^slxSox10^Dom crosses is due to the EdnrB gene. However, this finding in the context of EdnrB heterozygosity is in distinct contrast to the wild-type phenotype of EdnrB^sl/+, Sox10^Dom/+ animals. It is very likely that additional modifier genes in the 129 strain are influencing interaction between EdnrB and Sox10.

	DISCUSSION

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS REFERENCES

 
We present a series of genetic studies aimed at elucidating the influence of gene interactions on severity of aganglionosis in the Sox10^Dom mouse model of HSCR. Our characterization of Sox10^Dom congenic lines is the first detailed report of genetic background effect on the extent of aganglionosis in a HSCR model. Although these congenic lines carry the identical Sox10 mutation, Dom, the differences in severity of enteric deficits between them indicates that the variation in phenotype is due to heritable factors that are amenable to dissection by genetic approaches.

We investigated potential gene interactions between Sox10 and components in the endothelin signaling pathway, EdnrB, Edn3 and Ece1, that are recognized HSCR susceptibility loci. We initially applied selective genotyping of extremes in the phenotype distribution (severe versus mild) to test for association at these candidate loci. Upon identification of a highly significant effect at the EdnrB locus in mice stratified by survival, we genotyped all animals in the distribution. Our analysis found evidence that the interaction between EdnrB and Sox10 influences the severity of aganglionosis in the B6C3FeLe.Sox10^Dom pedigree. Analysis of extreme phenotypes increases the power to detect associations with genotypes (44), but at a price of increased false positives and upwardly biased estimates of effect size, if the full distribution is not eventually genotyped at significant loci. The reduction in effect size from the pedigree extremes (OR=7.583; Table 3) to the full distribution (OR=1.258) observed in our data was therefore not surprising. Subsequently, we extended and confirmed the effect of Sox10/EdnrB interaction in the F1 intercross progeny. Both logistic- and linear-regression analyses in the F2 revealed a consistent and significant relationship between penetrance of Sox10^Dom enteric deficits detected by AChE staining and the number of B6 alleles at EdnrB. Thus, Sox10^Dom mice with EdnrB B6 alleles are more likely to exhibit defects in enteric ganglia and those defects are typically more severe. B6 alleles at EdnrB explain 4.2% of the variance in penetrance and 1.9% of the variance in severity among animals with aganglionosis. The current analysis focuses on single gene effects and does not consider EdnrB interactions with other loci. Future investigations of epistasis will focus on variance associated with EdnrB in the context of multiple gene interactions.

Segregation analyses in human HSCR shows an approximately 4 : 1 predominance of males/females (14,52). In our study, we have scored for penetrance and defined the extent of ENS defect as a quantitative trait. We did not observe any greater incidence in Sox10^Dom/+ males than in females. However, we found that the direction of dominance for measured aganglionosis length (severity) differs by gender at the EdnrB locus in Sox10^Dom mutants. This finding differs from the sex-effect reported in human HSCR in that it is an effect on severity not penetrance. The disparity between our findings on gender-effects in the Sox10^Dom model and prior human studies could arise from any of several aspects. Human studies may not have had the power or contained sufficient numbers of SOX10 patients to detect the effects we describe. Alternatively, it is possible that we have not detected any major gender-specific modifiers in our crosses because the genomic region where such a locus resides is equivalent between the B6 and C3Fe strains. It will be important to investigate the Sox10^Dom phenotype on different genetic backgrounds to determine whether there are any strains that reproduce the gender effect observed in human HSCR.

We examined the interaction between EdnrB and Sox10 further by crossing EdnrB mutant alleles with Sox10^Dom lines. The increased severity of aganglionosis in EdnrB^sl, Sox10^Dom double mutants supports interaction between these loci. The phenotype of EdnrB^tm1Ywa/+,Sox10^Dom/+ animals confirms that this effect is specifically due to the EdnrB locus. Interestingly, the increased severity of aganglionosis is only observed for EdnrB^slxSox10^Dom/+ crosses when progeny are homozygous for the EdnrB^sl allele, EdnrB^sl/sl, Sox10^Dom/+. The difference in phenotype of EdnrB^sl/+, Sox10^Dom/+ versus EdnrB^tm1Ywa/+,Sox10^Dom/+ mutants suggests additional modifiers that are distinct between the piebald lethal and 129 strain backgrounds that modulate this gene interaction. It should be noted that the EdnrB^sl allele has been maintained for many generations through EdnrB^s/sl intercrosses that may have selected for genetic background modifiers capable of suppressing megacolon. Congenic lines of EdnrB^sl on well-characterized inbred strains like B6 that would facilitate further investigations of these effects are currently not available.

The mechanism of interaction between Sox10 and EdnrB is potentially a direct function of Sox10 transcriptional control at the EdnrB regulatory elements. The mode of interaction is now of interest for further eludicating sources of variation in HSCR phenotypes. Although Sox10 and EdnrB are expressed in mature enteric ganglia, it is likely that the effect of this interaction is exerted during migration of enteric NC into the gut. Both EdnrB and Sox10 expressed in migrating enteric NC cells during gastrointestinal ontogeny (22,41,53), and Iwashita et al. (54) have demonstrated that they are both up-regulated in the NCSCs that give rise to the ENS. Zhu et al. (55) have identified essential regulatory elements that control expression of EdnrB in migrating enteric precursors. These elements contain multiple Sox consensus sites necessary for correct spatiotemporal expression and are capable of binding Sox10. Other Sox gene family members, Sox8 and Sox9, are expressed in developing NC (56,57). Sox8 has some functional redundancy with Sox10 and could potentially bind sites flanking EdnrB (58). However, no ENS deficiencies have been reported in mouse knockout models that ablate Sox8 or Sox9 (59–62). Our genetic data specifically implicate Sox10 in regulation of EdnrB. Interestingly, the regions identified as essential regulatory elements for EdnrB (55) do not vary in sequence between the B6 and C3Fe strains (J. Stein and E.M. Southard-Smith, unpublished data). Nor have we detected any sequence variants between B6 or C3Fe within coding or non-coding regions of the complete cDNA (reference sequence NM_007904). The absence of variants in these regions suggests that other gene regions (e.g. splice junctions or distant regulatory elements) must be responsible for the modifier effects we have observed.

Though the effect of B6 alleles at EdnrB are clearly significant, they account for a small portion of the variability in phenotype. It is likely that other modifiers in addition to EdnrB interact with Sox10 to influence aganglionosis in this HSCR model. The residual variation in EdnrB genotype classes of Sox10^Dom intercross pups (Fig. 3) suggests that additional factors contribute to variation in aganglionic megacolon. Additionally, the difference in phenotype between the EdnrB^sl/+,Sox10^Dom/+ and EdnrB^tm1Ywa/+,Sox10^Dom/+ mutants implies the presence of other loci that modulate interactions between Sox10 and EdnrB. Studies in the HSCR families and EdnrB mouse models have demonstrated that EdnrB–Ret interactions modulate enteric phenotype (16,17). The Ret locus is an obvious candidate for future studies of additional Sox10 modifiers.

We have employed a somewhat novel approach of investigating gene interactions within a B6C3FeLe.Sox10^Dom pedigree. Typical mapping of complex traits relies on congenic lines in standard intercross schemes to initially map chromosomal locations of modifier genes, followed by subcongenic strategies to refine and isolate the map position of the locus (63–65). These strategies are successful, but very lengthy. Maintenance of Sox10^Dom mice on a B6C3FeLe-a F1 background offers an immediate and relatively rapid opportunity to identify modifiers of aganglionic megacolon because input alleles are restricted to the B6 and C3FeLe-a strains, and progeny from this line offer potential for greater recombination along a chromosome than standard F1 intercrosses. Using the described EdnrB STR markers (Table 2), we observed recombinant chromosomes within the pedigree that were passed between as many as four generations, but limitations in available polymorphic markers between the B6 and C3FeLe-a strains prevented further analysis within these intervals. The rate of recombination between the EdnrB markers in the pedigree (2.9%) was approximately equivalent to that in the F1 intercross (2.3%), as would be expected for closely linked markers. Mapping projects that need to interrogate large genomic intervals, on the order of 15–20 cM, like those identified in initial QTL genome scans would likely benefit most from the type of pedigree mapping used here. In this study, exploitation of a pedigree resource facilitated rapid association testing of candidate genes to evaluate the hypothesis that known HSCR susceptibility loci are interacting with Sox10 to modulate the severity of enteric deficits.

Our analysis focuses on the effects of Sox10/EdnrB interaction in the enteric NC; however, these genes are also expressed in the melanocyte lineage (32,41,66). Interaction between Sox10 and EdnrB could alter development of pigment cells and produce increased hypopigmentation. However, we do not observe differences in spotting between double heterozygous mutant pups, either EdnrB^s-l/+,Sox10^Dom/+ or EdnrB^{tm1Ywa/tm1Ywa},Sox10^Dom/+ mice and Sox10^Dom/+ littermates that would support interaction of these genes in melanocytes (data not shown). This is not particularly surprising as there is a reciprocal relationship between strain background and enteric phenotype versus pigmentation for Sox10^Dom/+ mutants. Although we have documented more severe enteric phenotype in B6.Sox10^Dom mice, Sox10^Dom heterozygotes on the B6 background exhibit less spotting (ventral belly spot and headspot) than C3Fe.Sox10^Dom mice (36). This observation suggests that the modifiers that interact with Sox10 to influence the melanocyte lineage are distinct from those that impact the ENS. Studies to investigate the effects of Sox10/EdnrB interaction on pigment cell phenotypes are in progress (Mollaaghababba et al., manuscript in preparation).

We did not detect any modifier effects of the Ece1 or Edn3 loci on Sox10^Dom enteric phenotype. We readily identified three polymorphic STR markers among the 10 screened at Ece1 (Table 2). Our data suggest that the B6 and C3Fe strains carry distinct haplotypes at Ece1 and that this locus does not contribute to variation of aganglionosis in the Sox10^Dom model. However, in screening markers at Edn3 we were unable to identify any polymorphic markers among 97 markers that were screened within 500 kb on either side of the locus. This suggests that the B6 and C3Fe strains carry equivalent haplotypes through this interval. Though we were able to use more distant markers to investigate potential interactions of Edn3 with Sox10, it is likely that identity by descent between B6 and C3Fe at Edn3 precludes detection of any effects in our crosses. Haplotypes at Edn3 carried by other inbred strains might have an effect on Sox10^Dom phenotype that cannot be excluded by our analysis.

Previous investigations of EdnrB HSCR models have focused primarily on the essential role of this gene for normal enteric ganglia development in the colon (29,40,50,67). The abnormal patterning of the ENS we observed in proximal small intestines of EdnrB^sl mice suggests that in addition to its role in the primary migration of enteric NC, this signaling pathway is also needed for normal patterning of enteric ganglia in the small intestine. Our observation is based on whole-mount AChE staining, which facilitates visualization of subtle alterations in enteric patterns that would not be apparent in histological sections. Whole-mount analyses also proved more informative than histological sections in a study reporting deficiencies of enteric ganglia in Gdnf haploinsufficient mice (27). Further investigation is essential to determine the developmental mechanisms that give rise to this aspect of the EdnrB^sl phenotype and establish its relevance to gastrointestinal motility disorders.

To summarize, our data demonstrates the effect of genetic background on the complexity of aganglionosis penetrance and severity in HSCR. Our analysis attributes some of this variation in the Sox10^Dom model to Sox10–EdnrB interactions and illustrates the relevance of investigating HSCR complexity in multiple mouse mutants to identify epistatic effects that may not be detected in analysis of a single-mutant model.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS REFERENCES

 
Animals
All animal protocols were approved by the Institutional Animal Care and Use Committee at Vanderbilt University. The B6C3FeLe.Sox10^Dom pedigree was maintained by crosses to B6C3Fe-a/a F1 mice. C3Fe.Sox10^Dom and B6.Sox10^Dom congenic lines were established and maintained by backcrosses to C3HeB/FeJ and C57BL/6J stocks, respectively. F1 intercross progeny were generated by crosses of (B6.Sox10^DomxC3Fe)-F1 mice with wild-type B6C3Fe-F1 mice. EdnrB^s/sl (stock no. 000308) and B6;129-EdnrB^tm1Ywa (stock no. 003295) mutants were obtained from Jackson Laboratories. EdnrB^s/sl mutants were outcrossed with B6 to expand the line and specifically identify EdnrB^sl/+ individuals. EdnrB^sl/+xB6.Sox10^Dom/+ crosses were performed to evaluate compound heterozygous phenotype. EdnrB^sl/+xSox10^Dom/+ were crossed to generate EdnrB^sl/+,Sox10^Dom/+ that were then crossed with EdnrB^sl/+ to obtain EdnrB^sl/sl,Sox10^Dom/+ pups for analysis of interactions between EdnrB and Sox10. Postnatal pups (P7–P10) were euthanized by approved procedures for harvest of gastrointestinal tracts and analysis of ENS deficits. Whole-mount AChE enzyme histochemistry was performed essentially as described by Enomoto et al. (25). Gut images were captured on a Leica MZ12.5 stereomicroscope with a DAGE-MTI, DC330 digital camera.

Biocomputational identification of Ece1 locus
We aligned human ECE1 exons to the rat ECE1 mRNA (accession no. D29683) to define exon boundaries, and then ran WU-BLAST (version 2.0) of individual rat exons against the mouse genome assembly (Ensembl version 18.30.1). We identified a genomic region on mouse chromosome 4 (136.0 Mb) that aligns to all 19 of the rat Ece1 exons (Contig AL807764). Annotation of this region predicts a transcript (XM_131743) that is 92% identical to the rat ECE1 transcript and 94% similar to the translation of the rat ECE1 protein when aligned by tBLASTx. Homology to the human ECE1 transcript is 88% at the nucleotide level, with 93% amino acid similarity by tBLASTx. Alignment of mouse- ordered EST hits identifies the same exons (data not shown). Although there are eight members of the endothelin converting enzyme gene family members scattered throughout the mouse genome, the chromosome 4 locus (136.0 Mb) bordered by D4Mit54 and D4Mit158 exhibits the highest degree of nucleotide and amino acid identity to both the rat and human ECE1 genes. Additionally, oligonucleotide primers that were used to genotype Ece1 knockout mice flank exon 15 upstream of the zinc-binding domain originally targeted by homologous recombination (21).

Genotyping
DNA was extracted from tail biopsies using standard methods. Animals were genotyped for Sox10^Dom using primers that flank the site of this single base insertion as described (22). Genotyping of EdnrB^sl/+ alleles was performed by PCR with primers that flank exon 6 (EdnrB CDS6 F 5'-GTTGTCCCGAGTCTTTATTTTG-3'; EdnrB CDS6 R 5'-CTCCAACCCCCTCTATCTTC-3') and only amplify from wild-type alleles or with D14Mit166 that discriminates between B6, C3Fe and EdnrB^sl/sl alleles. Genotypes for MIT markers were obtained by standard methods. Animals with recombinant haplotypes through the interval of the locus under evaluation were excluded from analysis.

Statistical analyses
Determination of statistical significance for allele frequencies in mildly versus severely affected mice of the B6C3FeLe.Sox10^Dom pedigree were assayed by ² with the Yates Correction for continuity. Determination of statistical significance in comparison to extent of aganglionosis were assayed by Student's t-test, assuming unequal variances. Significance was attributed for values P<0.05.

The effect of EdnrB genotype on the presence (probability) of hypoganglionosis or aganglionosis was analyzed separately in the B6C3FeLe.Sox10^Dom pedigree (N=820) and the F1 intercross (N=1043) by logistic regression. The presence of any amount of aganglionosis, hypoganglionosis or either was recoded to 1, and the absence of penetrance was recoded to 0. The Stata (http://www.stata.com) commands logistic and logit were used, along with supplemental commands in the spost package for categorical dependent variable model assessment (http://www.indiana.edu/jslsoc/spost.htm).

The effect of EdnrB genotype on the measured lengths of hypoganglionosis, aganglionosis, and the total affected length (cm) in affected animals was separately analyzed in the B6C3FeLe.Sox10^Dom pedigree and the F1 intercross by linear regression. The Stata commands qtlsnp (www.biostat-resources.com/stata) and regress were used to evaluate quantitative genetic models fitting both additive and dominance effects (the codominant model). The measured traits were natural log transformed to reduce distribution skew and meet variance assumptions required by linear regression.

Non-parametric methods on both raw and transformed quantitative measures were used to augment the linear regression results. The Stata command nptrend was used for a non-parametric trend test and the command qreg was used for a median-regression test.

The relation between genotype at EdnrB and disease penetrance and expressivity was further scrutinized for possible complicating effects of gender and gut length by estimating models with these factors as covariates. Gut length was available in both the pedigree and the F1 intercross, but gender (by genotype) was available only in the F1 intercross. Gut length was also used to create a percent affected length score. Because most results did not change in significance or interpretation, we here report only results from the simpler models, except when explicitly stated as otherwise.

	ACKNOWLEDGEMENTS

 
We thank Dr Al George for critical reading and discussion of the manuscript, Dr Alejandro Schaffer for insightful discussions and Harlan Simantel for contribution of the mouse sketch included in Figure 1a. This work was supported by a Howard Hughes Medical Research Scholar Award, a Foundation for Digestive Health and Nutrition Research Scholar Award and NIH NIDDK grant (DK60047) to E.M.S-S.

	FOOTNOTES

 
^* To whom correspondence should be addressed. Tel: +1 6159362172; Fax: +1 6159362661; Email: michelle.southard-smith{at}vanderbilt.edu

The authors wish it to be known that, in their opinion, the first three authors should be regarded as joint First Authors.

	REFERENCES

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS REFERENCES

 

1. Amiel, J., Attie, T., Jan, D., Pelet, A., Edery, P., Bidaud, C., Lacombe, D., Tam, P., Simeoni, J., Flori, E. et al. (1996) Heterozygous endothelin receptor B (EDNRB) mutations in isolated Hirschsprung disease. Hum. Mol. Genet., 5, 355–357.[Abstract/Free Full Text]

2. Amiel, J., Espinosa-Parrilla, Y., Steffann, J., Gosset, P., Pelet, A., Prieur, M., Boute, O., Choiset, A., Lacombe, D., Philip, N. et al. (2001) Large-scale deletions and SMADIP1 truncating mutations in syndromic Hirschsprung disease with involvement of midline structures. Am. J. Hum. Genet., 69, 1370–1377.[CrossRef][Web of Science][Medline]

3. Cacheux, V., Dastot-Le Moal, F., Kaariainen, H., Bondurand, N., Rintala, R., Boissier, B., Wilson, M., Mowat, D. and Goossens, M. (2001) Loss-of-function mutations in SIP1 Smad interacting protein 1 result in a syndromic Hirschsprung disease. Hum. Mol. Genet., 10, 1503–1510.[Abstract/Free Full Text]

4. Doray, B., Salomon, R., Amiel, J., Pelet, A., Touraine, R., Billaud, M., Attie, T., Bachy, B., Munnich, A. and Lyonnet, S. (1998) Mutation of the RET ligand, neurturin, supports multigenic inheritance in Hirschsprung disease. Hum. Mol. Genet., 7, 1449–1452.[Abstract/Free Full Text]

5. Edery, P., Attie, T., Amiel, J., Pelet, A., Eng, C., Hofstra, R.M., Martelli, H., Bidaud, C., Munnich, A. and Lyonnet, S. (1996) Mutation of the endothelin-3 gene in the Waardenburg–Hirschsprung disease (Shah–Waardenburg syndrome). Nat. Genet., 12, 442–444.[CrossRef][Web of Science][Medline]

6. Edery, P., Lyonnet, S., Mulligan, L.M., Pelet, A., Dow, E., Abel, L., Holder, S., Nihoul-Fekete, C., Ponder, B.A. and Munnich, A. (1994) Mutations of the RET proto-oncogene in Hirschsprung's disease. Nature, 367, 378–380.[CrossRef][Medline]

7. Hofstra, R.M., Valdenaire, O., Arch, E., Osinga, J., Kroes, H., Loffler, B.M., Hamosh, A., Meijers, C. and Buys, C.H. (1999) A loss-of-function mutation in the endothelin-converting enzyme 1 (ECE-1) associated with Hirschsprung disease, cardiac defects, and autonomic dysfunction. Am. J. Hum. Genet., 64, 304–308.[CrossRef][Web of Science][Medline]

8. Lyonnet, S., Edery, P., Mulligan, L.M., Pelet, A., Dow, E., Abel, L., Holder, S., Nihoul-Fekete, C., Ponder, B.A. and Munnich, A. (1994) Mutations of RET proto-oncogene in Hirschsprung disease. C. R. Acad. Sci. III, 317, 358–362.[Medline]

9. Pingault, V., Bondurand, N., Kuhlbrodt, K., Goerich, D.E., Prehu, M.O., Puliti, A., Herbarth, B., Hermans-Borgmeyer, I., Legius, E., Matthijs, G. et al. (1998) SOX10 mutations in patients with Waardenburg–Hirschsprung disease. Nat. Genet., 18, 171–173.[CrossRef][Web of Science][Medline]

10. Wakamatsu, N., Yamada, Y., Yamada, K., Ono, T., Nomura, N., Taniguchi, H., Kitoh, H., Mutoh, N., Yamanaka, T., Mushiake, K. et al. (2001) Mutations in SIP1, encoding Smad interacting protein-1, cause a form of Hirschsprung disease. Nat. Genet., 27, 369–370.[CrossRef][Web of Science][Medline]

11. Angrist, M., Bolk, S., Thiel, B., Puffenberger, E.G., Hofstra, R.M., Buys, C.H., Cass, D.T. and Chakravarti, A. (1995) Mutation analysis of the RET receptor tyrosine kinase in Hirschsprung disease. Hum. Mol. Genet., 4, 821–830.[Abstract/Free Full Text]

12. Attie, T., Pelet, A., Edery, P., Eng, C., Mulligan, L.M., Amiel, J., Boutrand, L., Beldjord, C., Nihoul-Fekete, C., Munnich, A. et al. (1995) Diversity of RET proto-oncogene mutations in familial and sporadic Hirschsprung disease. Hum. Mol. Genet., 4, 1381–1386.[Abstract/Free Full Text]

13. Seri, M., Yin, L., Barone, V., Bolino, A., Celli, I., Bocciardi, R., Pasini, B., Ceccherini, I., Lerone, M., Kristoffersson, U. et al. (1997) Frequency of RET mutations in long- and short-segment Hirschsprung disease. Hum. Mutat., 9, 243–249.[CrossRef][Web of Science][Medline]

14. Badner, J.A., Sieber, W.K., Garver, K.L. and Chakravarti, A. (1990) A genetic study of Hirschsprung disease. Am. J. Hum. Genet., 46, 568–580.[Web of Science][Medline]

15. Bolk, S., Pelet, A., Hofstra, R.M., Angrist, M., Salomon, R., Croaker, D., Buys, C.H., Lyonnet, S. and Chakravarti, A. (2000) A human model for multigenic inheritance: phenotypic expression in Hirschsprung disease requires both the RET gene and a new 9q31 locus. Proc. Natl Acad. Sci. USA, 97, 268–273.[Abstract/Free Full Text]

16. Carrasquillo, M.M., McCallion, A.S., Puffenberger, E.G., Kashuk, C.S., Nouri, N. and Chakravarti, A. (2002) Genome-wide association study and mouse model identify interaction between RET and EDNRB pathways in Hirschsprung disease. Nat. Genet., 32, 237–244.[CrossRef][Web of Science][Medline]

17. McCallion, A.S., Stames, E., Conlon, R.A. and Chakravarti, A. (2003) Phenotype variation in two-locus mouse models of Hirschsprung disease: tissue-specific interaction between Ret and Ednrb. Proc. Natl Acad. Sci. USA, 100, 1826–1831.[Abstract/Free Full Text]

18. Fitze, G., Cramer, J., Serra, A., Schreiber, M., Roesner, D. and Schackert, H.K. (2003) Within-gene interaction between c.135 G/A genotypes and RET proto-oncogene germline mutations in HSCR families. Eur. J. Pediatr. Surg., 13, 152–157.[CrossRef][Web of Science][Medline]

19. Puffenberger, E.G., Hosoda, K., Washington, S.S., Nakao, K., deWit, D., Yanagisawa, M. and Chakravart, A. (1994) A missense mutation of the endothelin-B receptor gene in multigenic Hirschsprung's disease. Cell, 79, 1257–1266.[CrossRef][Web of Science][Medline]

20. Hosoda, K., Hammer, R.E., Richardson, J.A., Baynash, A.G., Cheung, J.C., Giaid, A. and Yanagisawa, M. (1994) Targeted and natural (piebald-lethal) mutations of endothelin-B receptor gene produce megacolon associated with spotted coat color in mice. Cell, 79, 1267–1276.[CrossRef][Web of Science][Medline]

21. Yanagisawa, H., Yanagisawa, M., Kapur, R.P., Richardson, J.A., Williams, S.C., Clouthier, D.E., de Wit, D., Emoto, N. and Hammer, R.E. (1998) Dual genetic pathways of endothelin-mediated intercellular signaling revealed by targeted disruption of endothelin converting enzyme-1 gene. Development, 125, 825–836.[Abstract]

22. Southard-Smith, E.M., Kos, L. and Pavan, W.J. (1998) Sox10 mutation disrupts neural crest development in Dom Hirschsprung mouse model. Nat. Genet., 18, 60–64.[CrossRef][Web of Science][Medline]

23. Kapur, R.P., Livingston, R., Doggett, B., Sweetser, D.A., Siebert, J.R. and Palmiter, R.D. (1996) Abnormal microenvironmental signals underlie intestinal aganglionosis in Dominant megacolon mutant mice. Dev. Biol., 174, 360–369.[CrossRef][Web of Science][Medline]

24. Kapur, R.P. (1999) Early death of neural crest cells is responsible for total enteric aganglionosis in Sox10(Dom)/Sox10(Dom) mouse embryos. Pediatr. Dev. Pathol., 2, 559–569.[CrossRef][Web of Science][Medline]

25. Enomoto, H., Araki, T., Jackman, A., Heuckeroth, R.O., Snider, W.D., Johnson, E.M., Jr and Milbrandt, J. (1998) GFR alpha1-deficient mice have deficits in the enteric nervous system and kidneys. Neuron, 21, 317–324.[CrossRef][Web of Science][Medline]

26. Heuckeroth, R.O., Enomoto, H., Grider, J.R., Golden, J.P., Hanke, J.A., Jackman, A., Molliver, D.C., Bardgett, M.E., Snider, W.D., Johnson, E.M., Jr et al. (1999) Gene targeting reveals a critical role for neurturin in the development and maintenance of enteric, sensory, and parasympathetic neurons. Neuron, 22, 253–263.[CrossRef][Web of Science][Medline]

27. Shen, L., Pichel, J.G., Mayeli, T., Sariola, H., Lu, B. and Westphal, H. (2002) Gdnf haploinsufficiency causes Hirschsprung-like intestinal obstruction and early-onset lethality in mice. Am. J. Hum. Genet., 70, 435–447.[CrossRef][Web of Science][Medline]

28. Garcia-Barcelo, M., Sham, M.H., Lee, W.S., Lui, V.C., Chen, B.L., Wong, K.K., Wong, J.S. and Tam, P.K. (2004) Highly recurrent RET mutations and novel mutations in genes of the receptor tyrosine kinase and endothelin receptor B pathways in Chinese patients with sporadic Hirschsprung disease. Clin. Chem., 50, 93–100.[Abstract/Free Full Text]

29. Barlow, A., de Graaff, E. and Pachnis, V. (2003) Enteric nervous system progenitors are coordinately controlled by the G protein-coupled receptor EDNRB and the receptor tyrosine kinase RET. Neuron, 40, 905–916.[CrossRef][Web of Science][Medline]

30. Nadeau, J.H. (2003) Modifier genes and protective alleles in humans and mice. Curr. Opin. Genet. Dev., 13, 290–295.[CrossRef][Web of Science][Medline]

31. Herbarth, B., Pingault, V., Bondurand, N., Kuhlbrodt, K., Hermans-Borgmeyer, I., Puliti, A., Lemort, N., Goossens, M. and Wegner, M. (1998) Mutation of the Sry-related Sox10 gene in Dominant megacolon, a mouse model for human Hirschsprung disease. Proc. Natl Acad. Sci. USA, 95, 5161–5165.[Abstract/Free Full Text]

32. Potterf, S.B., Mollaaghababa, R., Hou, L., Southard-Smith, E.M., Hornyak, T.J., Arnheiter, H. and Pavan, W.J. (2001) Analysis of SOX10 function in neural crest-derived melanocyte development: SOX10-dependent transcriptional control of dopachrome tautomerase. Dev. Biol., 237, 245–257.[CrossRef][Web of Science][Medline]

33. Inoue, K., Shilo, K., Boerkoel, C.F., Crowe, C., Sawady, J., Lupski, J.R. and Agamanolis, D.P. (2002) Congenital hypomyelinating neuropathy, central dysmyelination, and Waardenburg–Hirschsprung disease: phenotypes linked by SOX10 mutation. Ann. Neurol., 52, 836–842.[CrossRef][Web of Science][Medline]

34. Inoue, K., Tanabe, Y. and Lupski, J.R. (1999) Myelin deficiencies in both the central and the peripheral nervous systems associated with a SOX10 mutation. Ann. Neurol., 46, 313–318.[CrossRef][Web of Science][Medline]

35. Bondurand, N., Kuhlbrodt, K., Pingault, V., Enderich, J., Sajus, M., Tommerup, N., Warburg, M., Hennekam, R.C., Read, A.P., Wegner, M. et al. (1999) A molecular analysis of the yemenite deaf-blind hypopigmentation syndrome: SOX10 dysfunction causes different neurocristopathies. Hum. Mol. Genet., 8, 1785–1789.[Abstract/Free Full Text]

36. Southard-Smith, E.M., Angrist, M., Ellison, J.S., Agarwala, R., Baxevanis, A.D., Chakravarti, A. and Pavan, W.J. (1999) The Sox10(Dom) mouse: modeling the genetic variation of Waardenburg–Shah (WS4) syndrome. Genome Res., 9, 215–225.[Abstract/Free Full Text]

37. Touraine, R.L., Attie-Bitach, T., Manceau, E., Korsch, E., Sarda, P., Pingault, V., Encha-Razavi, F., Pelet, A., Auge, J., Nivelon-Chevallier, A. et al. (2000) Neurological phenotype in Waardenburg syndrome type 4 correlates with novel SOX10 truncating mutations and expression in developing brain. Am. J. Hum. Genet., 66, 1496–1503.[CrossRef][Web of Science][Medline]

38. Inoue, K., Khajavi, M., Ohyama, T., Hirabayashi, S., Wilson, J., Reggin, J.D., Mancias, P., Butler, I.J., Wilkinson, M.F., Wegner, M. et al. (2004) Molecular mechanism for distinct neurological phenotypes conveyed by allelic truncating mutations. Nat. Genet., 36, 361–369.[CrossRef][Web of Science][Medline]

39. Lane, P.W. and Liu, H.M. (1984) Association of megacolon with a new dominant spotting gene (Dom) in the mouse. J. Hered., 75, 435–439.[Abstract/Free Full Text]

40. Shin, M.K., Levorse, J.M., Ingram, R.S. and Tilghman, S.M. (1999) The temporal requirement for endothelin receptor-B signalling during neural crest development. Nature, 402, 496–501.[CrossRef][Medline]

41. Lee, H.O., Levorse, J.M. and Shin, M.K. (2003) The endothelin receptor-B is required for the migration of neural crest-derived melanocyte and enteric neuron precursors. Dev. Biol., 259, 162–175.[CrossRef][Web of Science][Medline]

42. Baynash, A.G., Hosoda, K., Giaid, A., Richardson, J.A., Emoto, N., Hammer, R.E. and Yanagisawa, M. (1994) Interaction of endothelin-3 with endothelin-B receptor is essential for development of epidermal melanocytes and enteric neurons. Cell, 79, 1277–1285.[CrossRef][Web of Science][Medline]

43. Torfs, C.P. (1998) An epidemiological study of Hirschsprung disease in a multiracial population. Third International Meeting: Hirschsprung Disease and Related Neurocristopathies. Evian, France.

44. Van Gestel, S., Houwing-Duistermaat, J.J., Adolfsson, R., van Duijn, C.M. and Van Broeckhoven, C. (2000) Power of selective genotyping in genetic association analyses of quantitative traits. Behav. Genet., 30, 141–146.[CrossRef][Web of Science][Medline]

45. Kirkpatrick, B.W., Mengelt, A., Schulman, N. and Martin, I.C. (1998) Identification of quantitative trait loci for prolificacy and growth in mice. Mamm. Genome., 9, 97–102.[CrossRef][Web of Science][Medline]

46. Pavan, W.J., Mac, S., Cheng, M. and Tilghman, S.M. (1995) Quantitative trait loci that modify the severity of spotting in piebald mice. Genome Res., 5, 29–41.[Abstract/Free Full Text]

47. Jones, H.C., Depelteau, J.S., Carter, B.J., Lopman, B.A. and Morel, L. (2001) Genome-wide linkage analysis of inherited hydrocephalus in the H-Tx rat. Mamm. Genome, 12, 22–26.[CrossRef][Web of Science][Medline]

48. Roix, J.J., Hagge-Greenberg, A., Bissonnette, D.M., Rodick, S., Russell, L.B. and O'Brien, T.P. (2001) Molecular and functional mapping of the piebald deletion complex on mouse chromosome 14. Genetics, 157, 803–815.[Abstract/Free Full Text]

49. Lane, P.W. (1966) Association of megacolon with two recessive spotting genes in the mouse. J. Hered., 57, 29–31.[Free Full Text]

50. Kruger, G.M., Mosher, J.T., Tsai, Y.H., Yeager, K.J., Iwashita, T., Gariepy, C.E. and Morrison, S.J. (2003) Temporally distinct requirements for endothelin receptor B in the generation and migration of gut neural crest stem cells. Neuron, 40, 917–929.[CrossRef][Web of Science][Medline]

51. Peterson, K.A., King, B.L., Hagge-Greenberg, A., Roix, J.J., Bult, C.J. and O'Brien, T.P. (2002) Functional and comparative genomic analysis of the piebald deletion region of mouse chromosome 14. Genomics, 80, 172–184.[CrossRef][Web of Science][Medline]

52. Lipson, A.H., Harvey, J. and Oley, C.A. (1990) Three-generation transmission of Hirschsprung's disease. Clin. Genet., 37, 235.[Web of Science][Medline]

53. Young, H.M., Bergner, A.J. and Muller, T. (2003) Acquisition of neuronal and glial markers by neural crest-derived cells in the mouse intestine. J. Comp. Neurol., 456, 1–11.[CrossRef][Web of Science][Medline]

54. Iwashita, T., Kruger, G.M., Pardal, R., Kiel, M.J. and Morrison, S.J. (2003) Hirschsprung disease is linked to defects in neural crest stem cell function. Science, 301, 972–976.[Abstract/Free Full Text]

55. Zhu, L., Lee, H.-O., Jordan, C.S., Cantrell, V.A., Southard-Smith, E.M. and Shin, M.K. (2004) Spatiotemporal regulation of Endothelin receptor-B by SOX10 in neural crest-derived enteric neuron precursors. Nat. Genet., 36, 732–737.[CrossRef][Web of Science][Medline]

56. Bell, K.M., Western, P.S. and Sinclair, A.H. (2000) SOX8 expression during chick embryogenesis. Mech. Dev., 94, 257–260.[CrossRef][Web of Science][Medline]

57. Cheung, M. and Briscoe, J. (2003) Neural crest development is regulated by the transcription factor Sox9. Development, 130, 5681–5693.[Abstract/Free Full Text]

58. Stolt, C.C., Lommes, P., Friedrich, R.P. and Wegner, M. (2004) Transcription factors Sox8 and Sox10 perform non-equivalent roles during oligodendrocyte development despite functional redundancy. Development, 131, 2349–2358.[Abstract/Free Full Text]

59. Bi, W., Huang, W., Whitworth, D.J., Deng, J.M., Zhang, Z., Behringer, R.R. and de Crombrugghe, B. (2001) Haploinsufficiency of Sox9 results in defective cartilage primordia and premature skeletal mineralization. Proc. Natl Acad. Sci. USA, 98, 6698–6703.[Abstract/Free Full Text]

60. Kist, R., Schrewe, H., Balling, R. and Scherer, G. (2002) Conditional inactivation of Sox9: a mouse model for campomelic dysplasia. Genesis, 32, 121–123.[CrossRef][Web of Science][Medline]

61. Akiyama, H., Chaboissier, M.C., Martin, J.F., Schedl, A. and de Crombrugghe, B. (2002) The transcription factor Sox9 has essential roles in successive steps of the chondrocyte differentiation pathway and is required for expression of Sox5 and Sox6. Genes Dev., 16, 2813–2828.[Abstract/Free Full Text]

62. Sock, E., Schmidt, K., Hermanns-Borgmeyer, I., Bosl, M.R. and Wegner, M. (2001) Idiopathic weight reduction in mice deficient in the high-mobility-group transcription factor Sox8. Mol. Cell. Biol., 21, 6951–6959.[Abstract/Free Full Text]

63. Bennett, B., Beeson, M., Gordon, L., Carosone-Link, P. and Johnson, T.E. (2002) Genetic dissection of quantitative trait loci specifying sedative/hypnotic sensitivity to ethanol: mapping with interval-specific congenic recombinant lines. Alcohol Clin. Exp. Res., 26, 1615–1624.[CrossRef][Web of Science][Medline]

64. MacPhee, M., Chepenik, K.P., Liddell, R.A., Nelson, K.K., Siracusa, L.D. and Buchberg, A.M. (1995) The secretory phospholipase A2 gene is a candidate for the Mom1 locus, a major modifier of ApcMin-induced intestinal neoplasia. Cell, 81, 957–966.[CrossRef][Web of Science][Medline]

65. Kuida, S. and Beier, D.R. (2000) Genetic localization of interacting modifiers affecting severity in a murine model of polycystic kidney disease. Genome Res., 10, 49–54.[Abstract/Free Full Text]

66. Opdecamp, K., Kos, L., Arnheiter, H. and Pavan, W.J. (1998) Endothelin signalling in the development of neural crest-derived melanocytes. Biochem. Cell. Biol., 76, 1093–1099.[CrossRef][Web of Science][Medline]

67. von Boyen, G.B., Krammer, H.J., Suss, A., Dembowski, C., Ehrenreich, H. and Wedel, T. (2002) Abnormalities of the enteric nervous system in heterozygous endothelin B receptor deficient (spotting lethal) rats resembling intestinal neuronal dysplasia. Gut, 51, 414–419.[Abstract/Free Full Text]

CiteULike    Connotea    Del.icio.us    What's this?

This article has been cited by other articles:

				M. Fu, Y. Sato, A. Lyons-Warren, B. Zhang, M. A. Kane, J. L. Napoli, and R. O. Heuckeroth Vitamin A facilitates enteric nervous system precursor migration by reducing Pten accumulation Development, February 15, 2010; 137(4): 631 - 640. [Abstract] [Full Text] [PDF]

				M.-M. Garcia-Barcelo, C. S.-m. Tang, E. S.-w. Ngan, V. C.-h. Lui, Y. Chen, M.-t. So, T. Y.-y. Leon, X.-p. Miao, C. K.-y. Shum, F.-q. Liu, et al. Genome-wide association study identifies NRG1 as a susceptibility locus for Hirschsprung's disease PNAS, February 24, 2009; 106(8): 2694 - 2699. [Abstract] [Full Text] [PDF]

				N. Bondurand, D. Natarajan, A. Barlow, N. Thapar, and V. Pachnis Maintenance of mammalian enteric nervous system progenitors by SOX10 and endothelin 3 signalling Development, May 15, 2006; 133(10): 2075 - 2086. [Abstract] [Full Text] [PDF]

				S. E. Owens, K. W. Broman, T. Wiltshire, J. B. Elmore, K. M. Bradley, J. R. Smith, and E. M. Southard-Smith Genome-wide linkage identifies novel modifier loci of aganglionosis in the Sox10Dom model of Hirschsprung disease Hum. Mol. Genet., June 1, 2005; 14(11): 1549 - 1558. [Abstract] [Full Text] [PDF]

				I. K. Quigley, J. L. Manuel, R. A. Roberts, R. J. Nuckels, E. R. Herrington, E. L. MacDonald, and D. M. Parichy Evolutionary diversification of pigment pattern in Danio fishes: differential fms dependence and stripe loss in D. albolineatus Development, January 1, 2005; 132(1): 89 - 104. [Abstract] [Full Text] [PDF]

This Article Abstract FREE Full Text (PDF) An erratum has been published All Versions of this Article: 13/19/2289    most recent ddh243v1 Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to citation manager Search for citing articles in: ISI Web of Science (24) Request Permissions Google Scholar Articles by Cantrell, V. A. Articles by Southard-Smith, E. M. Search for Related Content PubMed PubMed Citation Articles by Cantrell, V. A. Articles by Southard-Smith, E. M. Social Bookmarking          What's this?

Online ISSN 1460-2083 - Print ISSN 0964-6906

Copyright © 2010 Oxford University Press

Oxford Journals Oxford University Press

• Site Map
• Privacy Policy
• Frequently Asked Questions

Other Oxford University Press sites: Oxford University Press Oxford Journals China Oxford Journals Japan American National Biography Booksellers' Information Service Children's Fiction and Poetry Children's Reference Corporate & Special Sales Dictionaries Dictionary of National Biography Digital Reference English Language Teaching Higher Education Textbooks Humanities International Education Unit Law Medicine Music Online Products Oxford English Dictionary Reference Rights and Permissions Science School Books Social Sciences Very Short Introductions World's Classics