Previous genome-wide linkage/association studies have identified several modifier genes in different genomic regions, but RET is the only gene known to play a major role in all forms of HSCR susceptibility and a quantitative study of its allelic spectrum should provide clues on the role of both rare and common sequence variants in the complex inheritance of HSCR, particularly its relationships to factors that are correlated with risk, such as gender.
After our RET CDS mutational screening, we have found that our whole series of Spanish HSCR patients presents a mutational frequency of 11.11% in sporadic cases and 18.75% in familial ones, which is concordant with previous studies that report values of up to 50% in familial forms and 7-20% in sporadic cases .
Genotyping analysis of RET common SNPs shows similar results to those previously described, confirming thereby the prominent role of either the intronic enhancer mutation (rs2435357, ) or the RET "risk haplotype" [4, 8] in the pathogenesis of sporadic HSCR.
As previously commented, one of the major features of HSCR, especially in the short-segment forms, is the sex-dependent penetrance and male predominance of 4:1 . Nevertheless the reason of this sex difference is still unclear. It has been postulated that sex differences could arise from mutations on the × chromosome, but genome-wide mapping studies have failed to identify an X-linked gene with a relevant impact per se in HSCR. Previous studies have indeed demonstrated differences in the transmission frequency of the "enhancer mutation" (rs2435357) depending on the offspring gender or parent gender [3, 5]. For this reason, we sought to perform an additional case-control study to analyze the distribution of the RET variants based in gender. We found a different allelic distribution between HSCR cases and controls in both, males and females, for rs2435357 and rs2505532. However and noteworthy, significant differences in HSCR versus controls in the male subset were considerably increased with respect the female subset, which is concordant with the sex-biased transmission of the rs2435357 variant published by Emison et al. Nevertheless, no significant differences were found when comparing males vs females, in agreement with the similarity in transmission frequency by gender after applying TDT reported by Emison et al., 2010 . The same phenomenon was observed for the haplotype analysis, which as a whole is consistent with the greater incidence in males than in females. However we cannot discard the possibility that the difference in the sample size of the corresponding subsets might be interfering to some extent in the results observed.
On the other hand, the association of rs2565206 to HSCR has been found to be sex-dependent and restricted to males. Probably, for this reason this SNP presented association to HSCR only when we extended our cohort, since we increased considerably the proportion of male HSCR cases. It is relevant to note that this specific variant has also been previously proposed to have a role in the pathogenesis of sporadic medullary thyroid cancer, since an over-representation of the T allele was found in different studies reported [8, 15, 16]. Although the most accepted hypothesis to explain these associations is the linkage with a still unidentified functional locus within or nearby RET, it has been proposed as well that this variant might slightly modulate per se the expression of the RET proto-oncogene, given that a new binding motif is created for NAFT transcription factor in the presence of the T allele . The restriction to association of the G allele to the male subset of HSCR patients could be therefore considered as another hallmark of the complex nature of Hirschsprung disease.
After the evaluation of RVs and CVs in our HSCR series, we have classified our cohort of patients according to RET mutational status in 4 subsets (CDS mutation only, enhancer mutation only, both or any of those events) obtaining similar frequencies in these subsets than Emison et al . Nevertheless, taking a further step in this study we performed a complete segregation analysis in patients with both RV and CV. 91.66% of cases with both mutational events resulted to present the enhancer variant in trans with regard to the CDS mutation. It supports that the CVs may act as modifier alleles of that bearing the RV, as speculated by Emison et al . In this way, we could be talking about a synergy effect between two alleles, which is in agreement with the additive/multiplicative model proposed for HSCR.
On the other hand, in our cohort, we have detected a 13% of cases with no RET RVs or CVs. It has been previously proposed that mutations in other HSCR genes could be responsible for the HSCR phenotype in those patients, but the analysis of other HSCR genes and chromosomal alterations can only explain 31% of these cases. Alternatively, RET deletions have been also proposed as a possible underlying molecular mechanism , although our MLPA analysis revealed the absence of this kind of mutational event affecting RET or other HSCR genes [17, 18]. Nevertheless, we cannot discard the presence of deletions in non-coding regulatory regions of RET or other HSCR genes which may lead to a loss-of-function of the protein.