Main

The enteric nervous system (ENS) is recognized as a distinct third portion of the autonomic nervous system, which also includes the sympathetic and parasympathetic systems (1). The ENS is involved in peristalsis and, singularly, other spontaneous movements still persist following its isolation from all nervous inputs (2,3,4). The interstitial cells of Cajal are crucial in mediating nervous impulse onto smooth muscle cells acting as the intrinsic pacemaker of the bowel, while the ENS controls the continuous influence of the sympathetic and parasympathetic systems. The cholinergic (postganglionic) parasympathetic neurons increase peristalsis, secretions, and vasodilation, while the noradrenergic (postganglionic) sympathetic fibers project onto the submucosal and myenteric plexuses, where they play an inhibitory effect on the cholinergic neurons promoting an inhibition of peristalsis and secretions and stimulation of vasoconstriction (5). Parasympathetic fibers reach the gut via vagal nerves to celiac and superior mesenteric plexuses to over the mid-transverse colon, while the rest of the gut is supplied by fibers arising from pelvic splanchnic nerves via the sacral nerves 2–4 going through the pelvic plexus (5).

Hirschprung’s disease (HSCR; MIM# 142623), one disorder of the ENS, is a rare congenital developmental disorder of the gastrointestinal tract characterized by a failure of vagal system derived enteric neural crest (NC) cells (ENCC) (neurocristopathy) to fully migrate cranio-caudally during embryonic development and adequately colonize the entire gut, leaving an aganglionic portion of variable length (6,7,8,9). Although original studies suggested colonization of the entire length of the human gut by enteric neural precursors is not complete until the 12th week of gestation, more recent studies seem to support complete colonization by the 7th week, which corresponds more closely with data obtained from animal models as well (10). HSCR is named after Dr. Harald Hirschsprung who first described this phenotype at “The Queen Louise Hospital for Children” in Copenhagen, Danemark. Aganglionosis is defined as the absence of ganglion cells in the myenteric and submucosal plexuses of the intestinal wall with concomitant hypertrophy of parasympathetic nerve fibers (11,12) ( Figure 1 ). When suspected, HSCR is diagnosed by standard histopathological evaluation with or without auxiliary special stains or immunohistochemistry that confirms the diagnosis following biopsy of the distal rectum ( Figure 1a – c ). Expression of calretinin, a vitamin D–dependent calcium-binding protein found in ganglion cells and nerves, has been described as an adjunctive or primary diagnostic test on gut biopsy specimens in HSCR with lack of specific calretinin staining confirming the diagnosis of aganglionosis ( Figure 1d ) (8). Classifying HSCR clinically is not an easy task, because the nervous system colonization failure may be variable or discontinuous (9,13,14,15). Three phenotypes are usually recognized, including (i) total colonic aganglionosis (TCA), which involves the entire colon which is aganglionic with a potential proximal extension into varying lengths of small bowel (usually no more than 50 cm of small bowel proximal to the ileocaecal valve), (ii) total colonic and small bowel aganglionosis, which may involve very long segments of small bowel aganglionosis, and (iii) the more frequent rectal or rectal/sigmoid colonic aganglionosis (RA or RSCA). A debate over the definition and occurrence of several phenotypical entities, including the ultrashort segment and the skip-segmental aganglionosis is ongoing (13,16,17,18,19,20,21,22). The majority of treatment remains surgical, while intense efforts are exploring the use of ENS stem cells by means of transplantation (12,23). This review reports on current knowledge about syndromic forms of HSCR, genes and biochemical pathways of HSCR as well as new views regarding pathogenesis of HSCR involving gene modules, microRNAs (miRNAs), and future perspectives.

Figure 1
figure 1

Ganglion cell maturation and Hirschsprung’s disease. (a) The myenteric plexus of the distal intestinal tract of a baby of 23 wk of gestation highlighting the high nucleus to cytoplasm ratio of the premature ganglion cells (400×, hematoxylin-eosin staining, bar: 400 μm), while panel b shows the relatively more mature ganglion cells of a term newborn baby at level of the submucosa of the lower intestinal tract (400×, hematoxylin-eosin staining, bar: 630 μm). (c) The lack of ganglion cells and hypertrophy of nerve fibers of a baby born at term (50×, hematoxylin-eosin staining, bar: 50 μm), while panel d shows Hirschsprung’s disease in a newborn baby confirming the absence of ganglion cells using a monoclonal antibody against calretinin, a calcium-binding protein of 29 kDa and calcium-dependent regulator with positive staining in the perivascular cells of blood vessels (internal control). Moreover, positive calretinin staining may be recognized in the lower right corner showing characteristic dark-brown granular nerve twigs in the muscularis mucosae. No calretinin staining is identified in nerve fibers at the center of the microphotograph (200×, anti-calretinin immunohistochemical staining, avidin-biotin complex, bar: 200 μm).

PowerPoint slide

Full size image

Clinical Dysmorphology

HSCR has a population-dependent incidence of 1.5–2.8 per 10,000 births (1.5 in Caucasians and 2.8 in Asians; average 1/5,000 live births) and is a largely variable heterogeneous condition both from the clinical and genetic point of view. There can be varying lengths of aganglionosis, a male preponderance (4:1) for short segment aganglionosis, a familial incidence of 5–20%, but usually appearing as sporadic congenital anomalies (24,25). HSCR occurs as an isolated trait in 70% of the cases, with about 1/3 only occurring in a syndromal setting. Complex genetic susceptibility factors may contribute substantially to the etiology of HSCR and, except for syndromal cases that show various patterns of Mendelian inheritance, it is generally accepted that HSCR follows a multifactorial pattern of inheritance with incomplete penetrance and 5–10% of HSCR cases having additional congenital anomalies (26). These associations suggest that at least some of the susceptibility loci may probably comprise genes with pleiotropic effects (27,28). A synopsis of clinical genetic entities with HSCR as a feature are summarized in Table 1 . In this section, the most common examples of clinical syndromes associated with HSCR are depicted.

Table 1 Clinical genetic syndromes
Full size table

Chromosomal anomalies have been described in up to 12% of HSCR patients, although this rate may be higher because not all patients with HSCR undergo karyotyping (24). More subtle changes can be identified with microarray. Trisomy 21 syndrome or Down syndrome (DS) is found in about 2–10% of the HSCR cases (29). The HSCR risk ratio in DS is greater than the risk given by any single gene mutations associated with HSCR (40-fold) (30). A role for the RET gene in DS-HSCR pathogenesis has been proposed as RET hypomorphic alleles have been found more often in DS-HSCR compared to HSCR or DS alone (31). New experimental evidence shows that zebrafish overexpression of ATP5O (ATP synthase, H+ transporting, mitochondrial F1 complex, O subunit), a gene found on chromosome 21, leads to enteric hypoganglionosis potentially explaining the association between DS and HSCR (Chauhan, AJHG, Baltimore, 2015, abstract communication).

The human prototype syndrome characterized by mutations in the RET gene is multiple endocrine neoplasia type 2 (MEN2) and familial medullary thyroid carcinoma (FMTC) (32,33). FMTC is part of MEN2. MEN2 is represented by three entities: MEN2A, FMTC, and MEN2B. MEN 2B is suspected when individuals with a Marfanoid habitus have distinctive features including mucosal neuromas of the lips and tongue, enlarged lips, HSCR, and early-onset medullary thyroid cancer. About 40% of affected individuals have diffuse ganglioneuromatosis of the gastrointestinal tract (34,35,36). RET mutations in MEN2A have been shown to be activating mutations by in vitro studies, while haploinsufficiency seems to be the most likely mechanism in sporadic HSCR (37,38,39). To the best of our knowledge, this data has not yet been translated into clinical guidance for screening of individuals with HSCR in the absence of a family history suggestive of MEN or FMTC.

Waardenburg syndromes (WS) are a group of autosomal dominant (AD) conditions found in 2–5% of congenital deafness cases, characterized clinically by distinctive facies with sensorineural hearing loss and pigmentary anomalies and an incidence of about 1/5,000 live births. Clinically and genetically heterogeneous, WS is classified in four subsets, three following an AD and one an autosomal recessive (AR) inheritance pattern. WS4 is underlined by homozygous mutations in three genes. Approximately 20–30% of cases are due to homozygous or heterozygous mutations within the EDN3 (endothelin-3) or the EDNRB (endothelin receptor type B) genes, whereas approximately 45–55% result from heterozygous mutations within the gene encoding the SOX10 transcription factor, suggesting further genetic heterogeneity (40,41). To the best of our knowledge, only four EDN3 mutations have been found in families with WS4, including three being homozygous for the mutation and one harboring a heterozygous missense mutation in addition to heterozygous mutations of SOX10 identified in 4 out of 15 patients (40,42,43). There is growing evidence for Alu-mediated deletions in noncoding regions around SOX10 as a recent mechanism involved in WS4 pathogenesis (44).

Haddad syndrome (HS; MIM 209880) is a genetic syndrome characterized by congenital central hypoventilation syndrome (CCHS, MIM 209880) and HSCR occurs in about 1/5 of CCHS individuals. There is no sex predilection and patients with HS have more often L-HSCR or TCA (45,46,47). Mutations in PHOX2B (paired-like homeobox 2b), a transcription factor involved in the development of noradrenergic neurons, has been extensively demonstrated in CCHS, a rare disorder with autonomic nervous system dysregulation and/or tumors of NC origin (48,49,50). In some cases of syndromic neuroblastoma (MIM 256700) that can present with CCHS or HSCR, heterozygous PHOX2B mutations have also been found (51). Two types of mutations are observed in CCHS: polyalanine repeat expansion mutations (PARM; normal range 22–33 repeats), and nonpolyalanine repeat expansion mutations (NPARM) which typically are out-of-frame deletions/duplications of variable size (1 to 38 nucleotides) (52). It has been shown that individuals harboring a heterozygous 20/27 PARM genotype are at increased risk for HSCR, and nearly all patients with NPARM have HSCR (53,54,55,56). Functional studies support PHOX2B as a rare cause of HSCR (33), and a hypomorphic HSCR RET predisposing allele can also be a risk factor for the HSCR phenotype in CCHS and of particular importance are issues related to genetic counseling including germline and somatic mosaicism, as well as late-onset disorder (36).

Goldberg-Shprintzen syndrome (GOSHS; MIM 609460) is an AR multiple congenital anomaly syndrome characterized by moderate intellectual disability, microcephaly, cleft palate, ocular colobomas, and a recognizable pattern of facial dysmorphisms (32,47,57). An abnormal neuronal migration including polymicrogyria has been observed. GOSHS is caused by a homozygous truncating mutation in the KIAA1279 gene, which encodes KIF-binding protein (KBP) on chromosome 10q21.1, a protein of poorly understood function and it has been suggested that the GOSHS phenotype may result from defects in development of both the ENS and central nervous system (CNS) (58). GOSHS patients usually have truncating homozygous KIAA1279 mutations (p.Arg90X, p.Ser200X or p.Arg202IlefsX2) leading to nonsense-mediated mRNA decay and loss of KBP function. KBP expression directly affected neurite growth in the human neuroblastoma SH-SY5Y cell line, in keeping with the central (polymicrogyria) and enteric (HSCR) neuronal developmental defects seen in GOSHS patients, providing the first evidence that an actin-microtubule cross-linking protein may be involved in neuronal development in humans.

Another genetic condition somewhat similar to GOSHSD is Mowat-Wilson syndrome (MIM 235730), an AD disorder with a wide spectrum of multiple congenital anomalies, caused by a de novo mutation in the ZFHX1B (zinc finger homeobox 1B) on chromosome 2q22 (59,60,61,62). Individuals with MOWS generally present with global developmental delay, epilepsy, and congenital anomalies including brain, eye, and heart defects with some patients having HSCR. Heterozygous de novo deletions encompassing the ZFHX1B gene or truncating mutations within the gene have been found in over 100 MOWS cases (24). The ZFHX1B gene encodes Smad-interacting protein-1 (SMADIP1 or SIP1), a transcriptional repressor involved in the TGFβ signaling pathway that is widely expressed during embryonic development. Studies of mutant mice that have lost ZFHX1B demonstrate defects of melanocyte and ENS development and loss of vagal NCCs (63). ZFHX1B appears to be a susceptibility gene for syndromic rather than isolated HSCR.

In addressing genetic counseling, sporadic HSCR should be considered to be a multigenic and sex-modified trait with a variable pattern of inheritance and an overall 4% recurrence risk in siblings of the proband (relative risk = 200) (24) ( Table 2 ). Genetic studies revealed that there is an elevated risk to relatives in terms of heritability in comparison to the general population. There is a higher incidence of HSCR in some individuals of Chinese descent versus other populations and accepted figures include an incidence of 1.0, 1.5, 2.2, 2.8 per 10,000 live births in Hispanics, Caucasian-Americans, African-Americans and Asians, respectively (24,64,65). Furthermore, this recurrence risk is dependent on the sex of both the affected individuals and the relative ( Table 3 ). The sex bias exhibited in HSCR depends on the length of the aganglionic segment, and is manifested in the observation that both incidence and penetrance is 2–4-fold higher in males. Another hallmark of this complex disorder is that both penetrance of known mutations and the compatible models of inheritance vary with the presence or absence of associated syndromic features and the highest recurrence risk should be for a male sibling of a female proband with L-HSCR (66). Recurrence risk in genetic conditions with HSCR following Mendelian patterns of inheritance as the few examples mentioned above follows the respective pattern (AD 50%, AR 25%). Taking into account various intricacies of HSCR etiology, this can make genetic counseling complicated. Even when a specific mutation has been identified in a family, predicting whether S- or L-HSCR will develop, or whether HSCR is found in the context of syndromic or nonsyndromic conditions, can often be challenging.

Table 2 Epidemiology and sex-dependent recurrence risk in Hirschsprung disease
Full size table
Table 3 miRNA studies to date exploring involvement of miRNA in HSCR
Full size table

Genes and Pathways Involved in Sporadic HSCR

The complex genetic etiology, which entangles HSCR, is intriguing and linked with mutations in the genes that encode mostly signaling molecules crucial for the proper development of the ENS. The most important genes and pathways that have been investigated include: Glial cell-derived pathway genes (RET, GDNF, NTN, SOX10, PHOX2b), Endothelin pathway genes (EDN3, EDNRB, SOX10), and TGFβ signaling pathway genes (ZFHX1B) (12,48,67,68). These genes seem to perform distinctly with RET and EDNRB using the receptor tyrosine kinase (RTK) and G-protein-coupled receptor (GPCR) signal-transduction pathways (69). The incidence and severity of intestinal aganglionosis is influenced by potentially multiple interactions between known HSCR associated genes. The mechanism behind these interactions is not yet fully known, but Ret and Ednrb might interact by activating common downstream signaling molecules. Other than genetic interactions, it may be important to emphasize that the incomplete penetrance and variation among families affected with HSCR suggest the involvement of modifier gens.

RET Signaling Pathway

RET, a proto-oncogene, encoding for a RTK, is the major and most extensively studied gene implicated in HSCR pathogenesis (12,48,70,71,72,73,74). Loss of function mutations seem to be most commonly seen in patients with familial HSCR than sporadic HSCR cases and in individuals with L-HSCR rather than S-HSCR. There are more than 100 unique RET changes in families with HSCR including large deletions encompassing the RET gene, microdeletions and insertions, nonsense, missense, and splicing mutations (75). The first susceptibility locus was mapped to 10q11.2 in a group of multigenerational families segregating HSCR as an incompletely penetrant AD trait, while interstitial deletion at chromosome 10q11 in TCA with intellectual disability made this region a hot spot for biochemical studies (24). Epigenetic changes of RET have also been described (76,77). Linkage analysis has shown that in 90% of HSCR families, the colonic phenotype is linked to the RET locus (78,79,80,81); however, most familial HSCR cases that show a RET locus linkage fail to reveal coding-sequence mutations (82,83). Transmission disequilibrium testing has demonstrated that various RET polymorphisms and haplotypes at polymorphic loci are associated with HSCR (84). Functional analysis of HSCR-associated RET promoter single-nucleotide polymorphisms (SNPs) shows a reduction of RET transcription in the presence of the respectively associated alleles (85). Different SNPs of coding, noncoding regions including conserved enhancer-like sequence in intron 1, and the promoter of RET have been identified to increase HSCR susceptibility several fold when compared to control cases in different ethnic groups (86). The RET RTK comprises a signal peptide, a CYS-rich region, a transmembrane region, a conserved intracellular TK-catalytic domain, and an extracellular domain ( Figure 2 ). RET is expressed through the developing nervous system and following activation by glial cell-derived neurotrophic factor (GDNF) family ligands, RET mediates signals through a range of pathways including: RAS/ERK, p38MAPK, NF-κB, PI3/AKT, and JNK, driving cell proliferation, survival, differentiation, migration, and apoptosis, under the support of a glycosyl-phosphatidyl-inositol-linked GDNF coreceptor-α (GFRα1) (78). The development and maintenance of both central and peripheral neurons are linked to specific combinations of proteins that signal through RET, including four related glycosyl-phosphatidyl-inositol-linked coreceptors GFRA1-4 and four soluble growth factor ligands of RET: GDNF, neurturin (NTN), persephin (PSPN), and artemin (ARTN). Interestingly, Tang et al. investigated a Han Chinese population with RET mutations and noted an HSCR association with IKBKAP suggesting population specificity (87,88). RET in neural crest development was elucidated by the expression pattern of RET during mouse embryogenesis and the phenotype of Ret null mice (Ret-/-) (89,90,91). Ret-/- mice demonstrate pyloric stenosis, a dilation of the proximal bowel, and an empty urinary bladder, although early stages of RET-positive vagal NCC migration seem to be RET signaling independent.

Figure 2
figure 2

Impact of mutations on Ret gene: unknown/unavailable (black), short segment (red), and long segment (green) (see also text).

PowerPoint slide

Full size image

GDNF Signaling Pathway

GDNF acts in conjunction with RET, but is considered a rare susceptibility HSCR gene (<5%) (12,80,92). GDNF and neurturin (NTN) are two structurally related neurotrophic factors that play crucial roles in the control of survival and differentiation of neurons. GDNF family receptor alpha 1 (GFRA1) has been shown to interact with GDNF and RET. GDNF and GFRA1 interact in a heterotetrameric complex with RTK and RET (12,80). Gdnf-/- and Gfrα-/- mice demonstrate aganglionotic gut with peristaltic failure. Gdnf +/- mice have frequent obstruction of the lower intestine and, in the absence of GDNF signaling, deficits in the initial appearance of NCC can be traced in the gut anlagen. In humans, GDNF germline mutations have been found in combination with RET mutations (93,94,95,96).

Endothelin Signaling Pathway (EDNRB)

The endothelins are vasoactive molecules, which make a family of 21 amino acid isopeptides (EDN1, EDN2, and EDN3) with each molecule containing two intra-chain disulphide bonds encoded by a separate gene (97). Mature and active endothelins are produced by the actions of endothelin-converting enzymes (ECE). So far, there are at least four known endothelin receptors, all of which are G protein-coupled receptors whose activation result in elevation of intracellular-free calcium. The final action is the constriction or relaxation of the smooth muscles of the blood vessels, raising or lowering the blood pressure, among other functions. Endothelin receptor type B (EDNRB) is coded by EDNRB, which is located on 13q22 and is nonselective with two types of EDNRB arising from the same gene. Mutations in the EDNRB gene are associated with ABCD syndrome (an acronym for albinism, black lock, neuronal migration disorder of the gut, and sensorineural deafness) and some forms of Waardenburg syndrome (40). EDNRB is rapidly desensitized by phosphorylation by the GPCR kinase type 2 once a ligand is bound, followed by internalization via a clathrin-dependent pathway, and transfer to the lysosomal compartment. In ontogenesis, EDNRB is expressed in the neural tube before the initiation of NCC migration and continues to be expressed by ENS precursors as they begin to migrate (98). EDN3 and Endothelin Converting Enzyme 1 (ECE1) are expressed by the mesenchyme surrounding the neural tube, along the dorsal migration pathway of the melanoblasts and in the developing gut mesoderm. Enteric EDN3 expression, however, is highest in the embryonic cecum and enhances the proliferation-promoting effects of GDNF on the ENS progenitors and coordinated interaction between RET and EDRB signaling pathways controls the development of the ENS (99) ( Figure 3 ). The EDNRB knockout mouse (Ednrb-/-) and piebald-lethal mouse are almost completely lacking coat color, do not survive to adulthood, and have megacolon (91,100,101). By comparison, Edn3 (Edn3-/-, lethal-spotting) mouse is pigmented over about 1/3 of the body, and only about 15% survive to adulthood. The Ece1-deficient mouse lacks enteric neurons and choroidal/epidermal melanocytes, and presents with a phenotype very similar to the EDNRB and EDN3 knockout mice. In humans, HSCR screening for EDNRB pathway associated genes has shown mutations of EDNRB, EDN3, and ECE1 accounting for about 5% of HSCR cases. A heterozygous ECE-1 mutation has occurred in a patient with HSCR and accompanied with craniofacial and cardiac defects (102). However, both EDN3 and EDNRB homozygous and heterozygous mutations have been reported in Waardenburg syndrome type 4 and dosage changes may be modifiers of the EDNRB pathway.

Figure 3
figure 3

Schematic representation of functionally related-genes that appear to play a key role in the differentiation of NCCs providing a framework of subnetworks related to signal transduction during four stages of embryogenesis in ENS development (see text).

PowerPoint slide

Full size image

SOX10

The SOX family of transcription factors is characterized by the presence of a DNA binding high-mobility group domain, and is involved in a wide range of developmental processes (43,45,103,104,105,106). SOX10 (Sry bOX10) has a potent transcription activating domain at its C-terminus that functions through two important types of DNA response elements both binding SOX10 monomers and favoring SOX10 dimers with DNA. Transcriptional activation by SOX10 is dependent on cooperating partners that are currently being identified. SOX10 is indeed expressed in NC cells as they leave the neural tube and expression continues during their migration. Although NCC formation may be normal in the absence of SOX10 and does not seem to be required for early migration, SOX10 is crucial for their survival. An in vitro study by Kulbrodt et al. analyzed the effect of these mutations on SOX10 function and concluded that each mutation likely leads to functional inactivation the protein. The heterozygous Dom mouse with Sox10 mutation has aganglionosis and hypopigmentation, while the Sox10-/- embryo is lethal. Mutations in SOX10 that result in the loss of the trans activating domain or that disrupt protein–protein interactions with PAX3 result in Waardenburg syndrome types 2 (WS2) without HSCR (71). SOX10 mutations, likely resulting in haploinsufficiency, are present in some cases with WS4 (43), and are only rarely associated with isolated HSCR, probably not exceeding more than 5% of HSCR cases, while mutations in evolutionarily conserved regulatory elements of SOX10 have been shown to be the cause of isolated, previously unexplained HSCR.

Interaction of RET-GDNF, EDNRB, and SOX10 Signaling

The complex process of how ENS development leads to adequate function of the gastrointestinal tract is far from being understood. Multiple signaling pathways, briefly summarized above, have been shown to interact in this intricate process ( Figure 3 ). EDNRB and SOX10 are components of signaling cascades that are critical to the development of the ENS together with RET-GDNF signaling. Carter et al. (66) suggest intriguing relationships between RET candidate genes (HOXB5 and PHOX2B) and RET expression. The helix loop helix (HLH) transcription factor Ascl1 retards the differentiation of myenteric neural cells in the intestine of mouse embryos toward ganglionic differentiation and induces Ret expression and neurogenesis in cell cultures of NC stem cells (66). Disruption of the adhesion molecule L1cam and the transcription factors Hoxb5 and Phox2b results in the delay or failure of migration of ENCCs to the distal intestine of murine embryos. Sox10 regulates the expression of both EDNRB and RET receptors, and it also modulates the expression of EDNRB expression in NC-derived cells as they approach the cecum—an area rich for ligands for EDN3 and GDNF. Interestingly, GDNF-induced proliferation is enhanced by EDN3, whereas GDNF-induced migration is inhibited by EDN3. Thus, a complex balance between RET and EDRB activation may allow for significant expansion of the NC stem cell pool in the cecum while affecting migration, which is dependent on cell–cell contact. Additionally, by reducing the attraction of NC cells to GDNF, EDN3 may allow the cells to migrate beyond the cecum. EDN3/EDNRB exhibits a complex interaction with RTK signaling with developmental-stage specific effects. It is intimately involved in RET signaling in the ENS lineage. Haploinsufficiency for SOX10 appears to interfere with the development and survival of ENS precursors due to its role as an important transcriptional regulator of several other genes known to be important in ENS development. SOX10 is also reported to regulate RET expression synergistically with PAX3 and SRY binds to the promoter of the RET gene at its both enhancer regions by interacting with transcription factors including PAX3 and NKX2-1 (107,108). SRY can inhibit RET expression generating haploinsufficiency of RET suppressing its function in ENS development. Figure 3 highlights functionally related genes that are associated with HSCR, providing a framework of subnetworks that are related to signal transduction during the formation and migration of ganglion cells in ENS development and genes associated with the GDNF and the EDNRB pathway interact with genes of the SOX10 signaling pathway, which play a key role in the differentiation of NCCs.

Semaphorin (SEMA) Signaling

Semaphorins are a group of proteins involved in signaling, characterized by the presence of a semaphorin domain, SEMA, at the N terminal (109). Semaphorins have an integral role in the migratory pathway of NCC during ENS development, involved in proliferation, migration, and differentiation (89). Class 3 semaphorin receptors include neuropilins 1 and 2 for binding and plexins, coreceptors for signaling. A cluster of SEMA SNPs was identified by GWAS, with allelic effects independent of RET and RET mutations may occur in patients carrying SEMA3 variants were noted (89,110,111). Not only did these mutations suggest a pathogenic effect on the disease, but the coexistence with RET mutations also substantiated the additive genetic model that has been proposed for the rarer forms of the disease (112,113). Recently, semaphorin 3C/3D signaling has been suggested to be an evolutionarily conserved regulator of the ENS development (114). Luzon-Toro et al. studying semaphorin class 3 genes through SNP analysis and by next generation sequencing technologies, associated SEMA3A (7p12.1) and SEMA3D (7q21.11) in the pathogenesis of a subset of S-HSCR (115). It seems that increased SEMA3A expression is a risk factor for HSCR through the upregulation of the gene in the aganglionic smooth muscle layer of the colon of HSCR individuals. Finally, SEMA3A polymorphisms have been discovered in different ethnic backgrounds (Caucasian, Northeastern Chinese, and Thai) (113).

Gene Ontology Studies and miRNAs

A recent international study cohort of 162 S-HSCR trios genotyped and analyzed using the Transmission Disequilibrium Test (TDT) by PLINK software confirmed a strong association of gene ontology (GO) modules related to signal transduction and regulation of ENS formation as well as other processes related to the disease (116). Their results revealed a clear association of GO terms connected to Ras signaling, a pathway known to play a key role in ENS formation. This network of 53 genes provides a current hypothesis in the context of genomic studies for the genetic complexity of HSCR (48,117). Some of the most interesting genes and networks affected that appear to be significantly associated with HSCR based on GO modules include: CRK, related to regulation of small GTPase mediated signal transduction and Ras protein signal transduction; GRB2, related to Ras protein signal transduction; ITGB1, related to cell–cell adhesion, cell migration, cell projections and neuron development; PLCG1, related to cell migration; RPS27AP16, related to synaptic transmission, cell projection organization and neuron development; SH3GL3 and TP53 which in the context of HSCR are related to CNS development. Overall, GO biological processes significantly associated with HSCR strongly support that HSCR is caused in different cell populations by specific genes belonging to the same (or related) GO modules; and these gene modules carry out biological functions that are essential to both neurogenesis and signaling.

MicroRNAs (miRNAs or miRs) are small, noncoding RNA molecules that are about 19–25 nucleotides long that regulate cell differentiation, proliferation, migration, and apoptosis (118,119,120,121,122,123,124,125,126,127,128,129). MiRs regulate target gene functions by triggering mRNA degradation or translational repression through complementary binding to the 3’-untranslated regions of target mRNA. However, their role in HSCR is yet to be clearly defined; so, a better understanding of miRNAs during ENCCs development is necessary. In the last couple of years, several miRNA targets have been proposed to play a role into the pathogenesis of HSCR including: SLIT2/ROBO1, MeCP2, NID1, SDPR, CD47/CUL3, SOX9, PTEN, and DIEXF ( Table 3 ). New methodologies and techniques are currently discovering new horizons in genetics and molecular biology and this will affect our understanding of this very complex disease.

Future Perspectives

New avenues for therapies for ENS disorders provided from recent advances in molecular and stem cell biology have led to the development of a unique ENS stem cell field (6,12,23,130) and genomic-wide association studies (GWAS) are an incredible resource with potential identification of candidate genes, which may need to be further investigated in the nearest future (6). The gene hunting has been galvanized by new molecular biology technologies including next-generation sequencing (NGS), conditional and cell-lineage-specific animal models, and stem cell biology providing an astonishing resource for investigating the early human development, etiology and progression of HSCR and defining the interaction of signaling pathways in detail. The platform of patient specific induced pluripotent stem cells will exquisitely dissect disease heterogeneity widening the perspectives introduced by whole-genome sequencing (6). The better risk prediction and development of stratified surgical approaches for some subsets of patients will be key for tailoring a personalized medical approach and bowel transplantation for short-gut syndrome may harbor novel strategies (131). In the future, stem cell-based therapeutic approaches may become available to children with TCA. HSCR remains a complex congenital anomaly and is still mysterious over a hundred years since its first report.

Statement of Financial Support

C.M.S. has received research funding from the Women and Children’s Health Research Institute, the Saudi Cultural Bureau, and the Canada Foundation for Women’s Health (CFWH) (Award 2009) and is Visiting Professor at the Wuhan University of Science and Technology (WUST), Wuhan, Hubei, P.R. China. O.C. is supported by WCHRI Innovation and seed grants, University Hospital Foundation grant, and Gilbert K Winter Fund, as well as DART Neuroscience. D.E. holds the Muriel & Ada Hole Kids with Cancer Society Chair in Pediatric Oncology and is supported by a WCHRI innovation grant (University of Alberta).

Disclosure

C.M.S. revised the original manuscript, re-drafted the gene signaling pathways and the histology, and re-shaped the work in consideration of several perspectives that may be useful for research and diagnosis of the 21st century. O.C. was responsible for the conceptual outline of the manuscript, review of clinical dysmorphology, genes, and signaling pathways as well as structuring figures and tables. H.M. was responsible of the RET network and signaling schema, and D.E. was responsible for the gene signaling pathways and critical reviews of all versions of the manuscript.