Over 50 patients have to date been reported with Xq28 duplications varying in size from 0.2 Mb to larger scale duplications extending beyond Xq27. Most have included the MECP2 and L1CAM genes, although none of them have been reported to present with Hirschsprung disease [19–21]. It has been suggested that the clinical features associated to neonatal encephalopathy in males carrying the Xq28 duplication might be solely due to the MECP2 gene, as patients with duplications involving this region alone have been reported with the same classical phenotype [19, 21]. Indeed, an OMIM number is used to specifically design the disorder caused by duplication or triplication of MECP2 (OMIM#300260; Lubs X-Linked mental retardation syndrome). Although spasticity, severe learning disability, axial hypotonia and frequent chest infections have been common findings within this group of patients, gastrointestinal symptoms have been much rarer. Only worth of note, in the report by Clayton-Smith et al., eight families with several male affected patients carrying Xq28 duplications had presented with intestinal pseudo-obstruction or bladder distension, but these features were attributed exclusively to the duplication of the FLNA gene . Several arguments supported such assumption, since these authors had also found 2 other families carrying intragenic duplications of FLNA and presenting exclusively the enteric phenotype. Moreover, Gargiulo et al. had already reported point FLNA mutations in X-linked families with pseudo-obstruction . The pathogenic mechanism proposed by the authors to explain how both duplications and truncating mutations of the FLNA gene could lead to intestinal pseudo-obstruction, was the hypothesis that the amount of FLNA would be critical for neuronal migration, and either its increased or decreased expression could interfere with the normal migration process . In our particular case we could propose exactly the same pathogenic mechanism for the connection between the L1CAM gene and Hirschsprung disease in our patient carrying the duplication at Xq28. The L1 cell adhesion molecule is a membrane glycoprotein belonging to a large class of immunoglobulin superfamily cell adhesion molecules (CAMs) that mediate cell-to-cell adhesion at the cell surface. The L1CAM protein is found primarily in the nervous system and is important in neuronal adhesion, migration, neurite outgrowth, and myelination . Mutations in the L1CAM gene cause neurological abnormalities of variable severity, including congenital hydrocephalus, agenesis of the corpus callosum, spastic paraplegia, bilaterally adducted thumbs, aphasia, and mental retardation (see OMIM). Interestingly, L1CAM is the only gene included in the duplicated region which has been verified to be a ENS-expressed gene and to play a key role in its development during embryogenesis , and some cases of patients with either X-linked hydrocephalus or acrocallosal syndrome and HSCR have been reported to present L1CAM mutations (visit HGMD http://www.hgmd.cf.ac.uk/ac/all.php). In contrast, none of the remaining genes encompassed by the rearrangement, such as IRAK1 or IDH3G have been previously related to HSCR. All the L1CAM mutational events reported so far to be responsible for the manifestation of the corresponding clinical pictures, are point sequence changes [4, 6, 8, 10, 11] or small deletions [5, 9], whose proposed pathogenic mechanisms lead to the suspicion that decreased L1CAM may be a modifying factor in the development of HSCR . It would be licit to speculate that also an increase in the amount of L1CAM, due to a duplication in the dosage of this gene, might be involved someway in the pathogenesis of Hirschsprung in this particular patient. In other words it would be plausible that the amount of L1CAM would be critical during ENS development, and either an increase or decrease of this protein could interfere with the processes of neuronal adhesion and migration.
The association between encephalopathy and HSCR should not be considered surprising as brain development is largely controlled by the same neural growth factors acting in the ENS . Because the frequency of Lubs X-linked mental retardation syndrome is estimated to be extremely low (only around 50 cases being reported to date), and the frequency of HSCR is approximately 1:5000, then the anticipated incidence of both together would be almost inexistent by chance alone. So it is not probable that the association of both clinical phenotypes has occurred accidentally and by different molecular events. This statement is supported by the fact that L1CAM abnormality has been previously proposed to contribute to HSCR onset [8, 9, 25]. Parisi et al. hypothesized that in those cases in which X-linked hydrocephalus presents together with HSCR, either RET or another HSCR gene contributes to aganglionosis under the influence of a defective L1CAM gene, and L1CAM may act as an X linked modifier gene for the development of HSCR . We therefore propose a similar role for L1CAM in the context of the association of Lubs syndrome and HSCR. Given the complex nature of HSCR, it would be conceivable that the duplication at Xq28 region including the L1CAM gene, together with another still unidentified molecular events, could lead to the manifestation of the disease in our patient. The existence of those additional molecular events would also explain why no other patients carrying MECP2-L1CAM duplications reported to date present with HSCR [19–21]. In addition, we have also to argue that children with congenital severe neonatal encephalopathy cannot communicate symptoms such as constipation, and originally, they have a constipation tendency; so there may be several cases of HSCR among those patients that goes undiagnosed because of a premature death, just as it occurred with the cousin of our index patient.