| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Molecular Diagnostics and Genetics |
1 Division of Paediatric Surgery, Department of Surgery, University of Hong Kong Medical Center, Queen Mary Hospital, Hong Kong SAR, China.
2 Department of Biochemistry and 3
Genome Research Centre, The University of Hong Kong, Hong Kong SAR, China.
aAuthor for correspondence: Fax 852-2817-3155; e-mail paultam{at}hkucc.hku.hk.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResults and DiscussionReferences |
|---|
Methods: Using PCR amplification and direct sequencing, we screened for mutations and polymorphisms in the coding regions and intron/exon boundaries of the RET, GDNF, EDNRB, and EDN3 genes of 84 HSCR patients and 96 ethnically matched controls.
Results: We identified 10 novel and 2 previously described mutations in RET, and 4 and 2 novel mutations in EDNRB and in EDN3, respectively. Potential disease-causing mutations were detected in 24% of the patients. The overall mutation rate was 41% in females and 19% in males (P = 0.06). RET mutations occurred in 19% of the patients. R114H in RET was the most prevalent mutation, representing 7% of the patients or 37% of the patients with RET mutations. To date, such a high frequency of a single mutation has never been reported in unrelated HSCR patients. Mutations in EDNRB, EDN3, and GDNF were found in four, two, and none of the patients, respectively. Two patients with mutations in genes of the EDNRB pathway also harbored a mutation in RET. Three novel and three reported polymorphisms were found in EDNRB, EDN3, and GDNF.
Conclusion: This study identifies additional HSCR disease-causing mutations, some peculiar to the Chinese population, and represents the first comprehensive genetic analysis of sporadic HSCR disease in Chinese.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResults and DiscussionReferences |
|---|
4:1 for SSA-HSCR patients and 1:1 for LSA-HSCR patients (1). Approximately 20% of HSCR cases are familial. The recurrence risk for siblings of SSA-HSCR probands varies from 1.5% to 3.3%, whereas the risk for siblings of LSA-HSCR probands varies from 3% to 18% (1). HSCR is frequently associated with chromosomal abnormalities, with other neurodevelopmental disorders, such as Waardenburg syndrome type 4, and with a variety of additional isolated anomalies and syndromes (1)(2). HSCR has a complex genetic etiology, with many studies indicating the receptor tyrosine kinase gene (RET) as the major susceptibility gene for HSCR (1)(2)(3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13)(14). Mutations in the RET gene account for up to 50% of familial cases and 7%–35% of sporadic cases (5)(6)(7)(8)(9)(10)(11)(12)(13)(14). Other HSCR genes identified to date mainly code for protein members of the RET and endothelin receptor B (EDNRB) signaling pathways, which are interrelated and involved in the development of enteric ganglia from specific lineage of neural crest cells (15)(16)(17)(18)(19)(20)(21)(22)(23)(24)(25)(26)(27). Mutations in these genes account for a small proportion of HSCR patients (7%) and have reduced penetrance and various effects on the length of the aganglionosis (1)(2), suggesting that modifier genes influence the penetrance and severity of the phenotype (28)(29). Receptor–ligand relationships underlie the physiologic basis for some types of complex inheritance diseases, of which HSCR is one.
We analyzed germline mutations in genes encoding protein members of the RET and EDNRB signaling pathways to investigate how mutations in those genes correlate with the manifestation of the disease in Chinese HSCR patients. In addition to the RET gene, we investigated the gene coding for the glial cell-line-derived neurotrophic factor (RET ligand; GDNF), the EDNRB gene, and the gene coding for its ligand, endothelin 3 (EDN3). To our knowledge, this is the largest sample of HSCR patients concomitantly studied for these genes and the first comprehensive genetic analysis of HSCR in the Chinese population.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResults and DiscussionReferences |
|---|
Healthy controls (96 individuals) were unselected, unrelated, ethnic Chinese individuals. Patients and controls assented to molecular analysis. The study was approved by the Institutional Review Board of the University of Hong Kong.
dna sequence analyses
DNA was extracted from peripheral blood by use of a QIAamp Blood Kit (Qiagen). Unaffected parents were also screened when available. Using PCR and direct sequencing, we screened all exons of the RET, EDNRB, EDN3, and GDNF genes, including intron/exon boundaries, for mutations and polymorphisms. The primers and PCR conditions for amplification of the RET and EDN3 genes have been described previously (6)(12). For amplification of RET exon 21 and the EDNRB and GDNF genes, we generated new pairs of primers (Table 1
). PCR products were sequenced using an ABI PRISM® Big DyeTM Terminator, Ver. 2.0, Cycle sequencing assay (Applied Biosystems) and an ABI 3100 automated sequencer (Applied Biosystems). For those samples in which a DNA sequence variation was observed, PCR amplification from genomic DNA and sequencing using both forward and reverse primers were repeated.
|
|
Results and Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResults and DiscussionReferences |
|---|
. Of 16 patients with associated anomalies, only 4 had mutations in the genes investigated. The distribution of the patients according to gender, mutation status, and length of aganglionosis is depicted in Table 3
.
|
|
On the basis of previous reports (29), we defined a sequence alteration as a mutation if it failed to occur in 96 ethnically matched controls. Otherwise it was termed a polymorphism.
sequence alterations in ret and gdnf
In the RET gene we identified 10 novel as well as 2 previously described mutations (Fig. 1
). The new mutations consisted of two point deletions, two small deletions, and six nucleotide substitutions [5'UT -37G>C, c360 C>T (T120T), c434 T>G (V145G), c2081 G>A (R694Q), c2862 G>A (G954G), and c2881 T>C (F961L)]. Seven of 12 (58%) of the RET mutations were de novo, which reaffirmed the pathogenic role of RET in sporadic HSCR.
|
We predict that the four novel deletions found in this study are all probably disease-causing mutations. Both point deletions, c1449delC (Y483X) in the cadherin domain (CD) and c1908delG (V636fsX1) in the cysteine domain, lead to stop codons. In these cases, the mRNA transcripts are likely to be degraded by nonsense-mediated mRNA decay surveillance mechanisms (30) and, if translated, would lead to an altered or truncated RET protein. The 10-nucleotide deletion c1685_1690 + 4del abolishes the splice site between exon and intron 8, probably leading to an abnormally spliced mRNA. The 36-nucleotide deletion IVS11 + 15 (15_36del), starting at nucleotide +15 of intron 11 and spanning to nucleotide +36, probably affects intronic cryptic regulatory sites.
We found two missense mutations in the extracellular domain of the RET protein, the novel V145G and R114H. Generally, HSCR missense mutations located in the extracellular domain would affect the folding of the RET protein, impairing its maturation and preventing it from reaching the cell surface (31). V145 is a conserved residue of the prototypical strand of the cadherin fold in CD1 (Fig. 1
) and part of a hydrophobic core crucial to the structure of the protein (32). R114H was the most prevalent mutation in our series (observed in six patients). R114H was described recently (33) in one Japanese patient with isolated CCHS, a condition that has an incidence of 1.9% in HSCR (34). In this study, none of the HSCR patients with R114H presented with CCHS features. The only HSCR patient with CCHS (patient 97) had a G>C change at nucleotide -37 of the 5'-untranslated region (5'UT -37G>C). Because R114H has been found only in Asian patients, it would seem that R114H is a founder mutation contributing to HSCR in Asia.
The other novel missense mutations that we found (R694Q and F961L) are located in the intracellular domain. R694Q is in a transition zone between the transmembrane domain and the tyrosine kinase (TK) domain, whereas F961L is in the TK domain. HSCR mutations in the TK domain usually cause impairment of the kinase activity (31). Even mutations affecting the nonconserved amino acids of the TK domain may partially impair the kinase activity (35).
We observed the R982C change in three patients (patients 9, 55, and 92) but not in the 96 controls. R982C has been debated widely (36)(37); it was initially described as a mutation but later was observed in control chromosomes (11)(38). Interestingly, patient 55, who inherited R982C from her mother, also had two de novo silent mutations, T120T and G945G, both novel. The fact that both mutations are de novo, both are in the same patient, and neither was found in the controls could indicate that they contribute to the disease, either independently, combined, or having a joint effect with R982C. Other silent mutations are known to alter mRNA processing, and in the RET gene in particular, the silent mutation I647I, interferes with normal transcription, leading to decreased protein concentrations (39).
The M1064T mutation in exon 20 was found in a male affected with SSA. DNA from the father was not available, and analysis of maternal DNA was negative for the mutation (see Table 2
). The M1064T mutation was first described in a HSCR patient with SSA who had inherited it from an unaffected parent. Because this mutation is specific to the isoform RET51, it was suggested that this isoform was essential for normal enteric development (7). Functional analyses of M1064 failed to demonstrate significant changes in the activity of RET (31), and experiments in transgenic mice indicated that RET51 is not required for the development of the ENS (40). However, based on the demonstration of the formation of homo- and heterodimers between RET9 and RET51, it has recently been suggested that there may be a critical role for the RET51:RET9 ratio in addition to that of the absolute amounts of the two isoforms (41).
A detailed analysis of the RET single-nucleotide polymorphisms (SNPs) is reported elsewhere (42). In brief, using case–control statistics and the Transmission Disequilibrium Test, we found that the variant alleles of both A45A (c135G>A in exon 2) and L769L (c2307T>G in exon 14) were overrepresented in the patient population. Conversely, the variant allele of A432A (c1296G>A in exon 7), was significantly underrepresented. The frequencies of these associated alleles were significantly higher than those reported in other populations, including those of the controls, which may explain the higher incidence of HSCR in Asians. As for RET haplotypes comprising the disease-associated SNPs, we found that 66% of the Chinese HSCR patients were represented by one main haplotype: allele A of A45A (c135G>A), allele G of A432A (c1296G>A), and allele G of L769L (c2307T>G; A-G-G, P = 0.000002). This association was independent of the RET mutational status of the patients. Thus, this haplotype is either functional itself, conferring susceptibility to HSCR and/or acting as a modifier and contributing to the variable expression of the phenotype, or represents linkage disequilibrium with a susceptibility locus located in noncoding regions of RET. This would explain the paucity of RET mutations and their low penetrance and variability in the presentation of the HSCR phenotype.
The only sequence alteration found in the GDNF gene was the transition c278G>A in exon 2, which causes a nonsynonymous change in the protein (R93Q). This change was observed in one control and one SSA patient, who was also affected with parathyroid adenoma. No other mutations were found in this patient. Interestingly, a nucleotide substitution, also affecting codon 93 (R93W), was identified as a GDNF familial mutation leading to HSCR in conjunction with a RET mutation (23). To date, GDNF mutations have been found in five patients and are a rare cause of HSCR. Four of those patients had additional contributory factors, such as mutations in RET (23)(24)(25)(26).
sequence alterations in ednrb and edn3
We found four novel mutations [p.P383_L386delinsP, D241D (c723T>C), N426N (c1278T>C), and IVS4 -14T>C] in the EDNRB gene. The p.P383_L386delinsP mutation encodes a 9-bp deletion (nucleotides c1149–1157) in exon 6, affecting the codons for proline-383, isoleucine-384, alanine-385, and lysine-386 and substituting them for a proline. The result would be a shorter protein lacking highly conserved residues of the seventh transmembrane domain of the receptor. A missense mutation affecting the same codon (P383L) is known to be responsible for the HSCR phenotype (17). Functional analysis of P383L demonstrated that ligand-binding sites were reduced and that signal transduction was impaired (43). Thus, p.P383_L386delinsP could indeed lead to HSCR. Interestingly, N426N was found in a patient who also had a mutation in RET. Although D241D (exon 3), N426N (exon 7), and IVS4 -14T>C (intron 4) were not found in the controls, their contribution to disease should be taken with caution.
Three polymorphisms [5'UT -26G>A, V185M (c553G>A), and L277L (c831G>A)] in the 5'-untranslated regions of exons 2 and 4, respectively, were also detected. The nonsynonymous change V185M is new and was found in two mutation-free SSA patients (maternal inheritance) and in one control. 5'UT -26G>A, initially reported as a mutation (14)(17), has recently been reported as a SNP (rs2070591) after a comprehensive study of SNPs in a Japanese population (44). In our study, 5'UT -26G>A was found in one control and two SSA patients, one of whom (patient 61) also harbored IVS4 -14T>C (both inherited from the father). L277L [(c831G>A; dbSNP (rs5351)] was found in both the heterozygous and homozygous states.
We found two novel mutations in the EDN3 gene: a missense mutation (E48D) in exon 2 resulting from a G>C transversion; and a de novo C>A transversion at nucleotide -19 from the start codon (5'UT -19C>A), which could be affecting as yet unexplored regulatory regions. E48D was found in patient 83, who also harbored a deletion in RET and was affected with TCA.
Interestingly, we identified the G>A transition leading to the A17T nonsynonymous change in 7% of the patients (none syndromic) and in 5% of the controls. Remarkably, one of the controls was homozygous for A17T. A17T was previously identified as a mutation in a nonsyndromic sporadic HSCR patient (22). To our knowledge, this is the first time that A17T has been detected in a healthy individual. A new polymorphism, (T insertion at nucleotide -35 of exon 5; IVS4 -35insT), was found in 9 patients and 11 controls.
Mutations in EDNRB and EDN3 were scarce and were mainly inherited from unaffected parents. It is worthwhile noting that at least two EDNRB and EDN3 mutations previously described in isolated HSCR (14)(17)(22) have recently been observed in healthy individuals of different ethnic origins [ (44) and this study]. Therefore, either those "mutations" do not contribute to HSCR or if they do, they require the contribution of modifiers, which could be population- or/and individual-specific.
gender bias in the incidence of the disease
For the whole cohort, the sex ratio observed was 3.94 (67 males and 17 females). The ratio was increased in SSA-HSCR (5.09) but decreased in LSA-HSCR patients (1.83). These values are consistent with a decrease in gender bias with an increase in the length of aganglionosis (1). Gender bias indicates the presence of additional gender-specific susceptibility/modifying locus (loci) (21). If the main HSCR genes harbor more severe mutations (more frequent in LSA/TCA), the effect of gender-specific modifiers on the phenotype would be diminished, explaining the diminished gender bias in sporadic LSA-HSCR.
female patients have a higher mutation rate than male counterparts
Remarkably, mutations were observed in 41% (7 of 17) of the female patients and only in 19% (13 of 67) of the male patients (Table 3
). Although the number of female patients is too small for significant statistical analysis, to our knowledge, this observation has never been reported.
genotype–phenotype correlation
Lack of genotype–phenotype correlation is a feature common to all mutations in HSCR genes. In our series, this is exemplified by R114H of the RET gene. Of six patients with R114H, five were affected with SSA and one with LSA. Analysis of parental DNA showed that this mutation could be inherited from either the unaffected mother or father. Remarkably, R114H is highly recurrent in our series, representing 7% (6 of 84) of the patients or 30% (6 of 20) of the patients with mutations in any of the genes, or 37% (6 of 16) of the patients with mutations in RET. To date, no such frequency has been reported for any of the RET mutations.
Of 17 LSA/TCA patients, 4 (24%) had potentially HSCR-causing mutations in RET (including patient 83, who also had E48D in EDN3). No mutations were found in the other genes.
Of 67 SSA patients, 16 (24%) had mutations either in RET (18%; 12 of 67) or in the other genes (Table 3
). The percentage of LSA/TCA patients harboring mutations was quite similar to that of SSA patients with mutations, probably because EDNRB mutations are more frequently associated with SSA (1). RET mutations alone were slightly more frequent in LSA/TCA patients (see Table 3
). More striking differences in mutation rates between LSA- and SSA-HSCR patients have been documented in other series of individuals (9).
The most important issue, however, is that 76% of the patients had no mutation(s) in the genes investigated. It is worth mentioning patient 38, who also had aganglionosis in the small bowel. For this patient, the possibility of a hemizygous deletion of RET cannot be ruled out because analysis of RET polymorphisms showed a homozygous state for each site. For the rest of the patients with no mutations in RET, heterozygosity was observed for several SNPs distributed throughout the gene.
The overall paucity of mutations in HSCR patients indicates that mutations exist in other as yet undiscovered genes and/or in noncoding regions of the HSCR genes identified. In addition, the fact that mutations leading to disease are inherited from unaffected parents indicates that the HSCR phenotype results from interaction of several genes and/or modifiers, as has been demonstrated with genes of the RET and EDNRB pathways (20)(21)(28)(29). In our study, only two patients had mutations in both RET and genes of the EDNRB pathway. Interestingly, in those two patients (patients 25 and 83), RET mutations were de novo, and the mutations in the genes of the EDNRB pathway were inherited from unaffected parents. Whether E48D in EDN3 and N426N in EDNRB, contribute to the phenotype in combination with the mutations in RET is not known, but this certainly adds force to the interaction between pathways (18)(19)(20)(21). Mutations in GDNF and EDNRB have been found previously in the context of mutations in the RET locus (23)(38)(39). Even specific EDNRB mutations are known to require specific RET polymorphisms for transmission (20)(28).
Little is known about the signaling events governing the fate of the enteric neurons during development. Each of the genes coding for protein components of the RET and EDNRB signaling pathways is to be considered a HSCR candidate gene, contributing to HSCR on its own or in combination. In patients without mutations, even polymorphisms in two or more loci of the RET and EDNRB pathway genes could produce a simultaneous decrease in protein dosage (39)(41). We have recently demonstrated that PHOX2B (45), a gene encoding a transcriptional activator necessary for the expression of RET, is also associated with HSCR. Although the molecular mechanism underlying this association remains unclear, our finding indicated the contribution of PHOX2B in the enteric neuronal development. Recently, mutations in PHOX2B have been found in patients with isolated CCHS as well as in patients with CCHS and HSCR (46).
In oligogenic diseases such as HSCR, with a complex mode of inheritance, functional analyses of mutations have their limitations. A better understanding of the molecular mechanisms of oligogenicity through the study of the interactions between a discrete number of loci could help to make genotype-based phenotypic predictions.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
1 Nonstandard abbreviations: HSCR, Hirschsprung disease; SSA, short-segment aganglionosis; LSA, long-segment aganglionosis; ENS, enteric nervous system; RET, receptor tyrosine kinase; EDNRB, endothelin receptor B; GDNF, glial cell-line-derived neurotrophic factor; EDN3, endothelin 3; TCA, total colonic aganglionosis; CCHS, congenital central hypoventilation syndrome; CD, cadherin domain; TK, tyrosine kinase; and SNP, single-nucleotide polymorphism. 
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResults and DiscussionReferences |
|---|
receptor complex, in Hirschsprung disease. Hum Mol Genet 1996;5:2023-2026.
r S, Ibáñez CF. Molecular modeling of the extracellular domain of the RET receptor tyrosine kinase reveals multiple cadherin-like domains and a calcium-binding site. J Biol Chem 2001;276:35808-35817.The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  J Amiel, E Sproat-Emison, M Garcia-Barcelo, F Lantieri, G Burzynski, S Borrego, A Pelet, S Arnold, X Miao, P Griseri, et al. Hirschsprung disease, associated syndromes and genetics: a review J. Med. Genet., January 1, 2008; 45(1): 1 - 14. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Asai, T. Fukuda, Z. Wu, A. Enomoto, V. Pachnis, M. Takahashi, and F. Costantini Targeted mutation of serine 697 in the Ret tyrosine kinase causes migration defect of enteric neural crest cells Development, November 15, 2006; 133(22): 4507 - 4516. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Soufir, R. Meziani, J.-J. Lacapere, G. Bertrand, F. Fumeron, A. Bourillon, B. Gerard, V. Descamps, B. Crickx, L. Ollivaud, et al. Association Between Endothelin Receptor B Nonsynonymous Variants and Melanoma Risk J Natl Cancer Inst, September 7, 2005; 97(17): 1297 - 1301. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. C.H. Lui, T. Y.Y. Leon, M.-M. Garcia-Barcelo, R. W. Ganster, B. L.S. Chen, J. M. Hutson, and P. K.H. Tam Novel RET Mutation Produces a Truncated RET Receptor Lacking the Intracellular Signaling Domain in a 3-Generation Family with Hirschsprung Disease Clin. Chem., August 1, 2005; 51(8): 1552 - 1554. [Full Text] [PDF]
|
|
|
|
 |
|
 M. Garcia-Barcelo, R. W. Ganster, V. C.H. Lui, T. Y.Y. Leon, M.-T. So, A. M.F. Lau, M. Fu, M.-H. Sham, J. Knight, M. S. Zannini, et al. TTF-1 and RET promoter SNPs: regulation of RET transcription in Hirschsprung's disease Hum. Mol. Genet., January 15, 2005; 14(2): 191 - 204. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. A. Cantrell, S. E. Owens, R. L. Chandler, D. C. Airey, K. M. Bradley, J. R. Smith, and E. M. Southard-Smith Interactions between Sox10 and EdnrB modulate penetrance and severity of aganglionosis in the Sox10Dom mouse model of Hirschsprung disease Hum. Mol. Genet., October 1, 2004; 13(19): 2289 - 2301. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
