Discussion
Pathogenic mutations in
MKS1 and
B9D1 have been reported only in MS fetuses.
MKS1 is mutated in about 7-14% of MS patients, with increased frequency in northern European countries due to a founder mutation [
3‐
7]. Several studies have highlighted that mutations in
MKS1 are associated with a particularly severe MS phenotype, with high occurrence of polydactyly, bone dysplasia, encephalocele and other central nervous system anomalies [
4‐
6]. To date,
B9D1 was found mutated only in one MS fetus with cystic dysplastic kidneys, encephalocele, shortened limbs and ambiguous genitalia [
8]. Conversely, the four JS patients described here all had a relatively mild presentation characterized by a pure neurological phenotype, with the exception of retinal dystrophy in patient COR340. The degree of intellectual impairment was variable, and patient COR413 even presented with normal intellectual abilities, a rare occurrence in JS [
9]. None of the patients showed involvement of the organs that are typically affected in MS, namely the kidneys, liver and skeleton, although a future renal disease can be safely excluded only in the adult patient (COR340).
This wide phenotypic variability associated with mutations in the same genes remains an intriguing open question. Genotype-phenotype correlates have been proposed for some genes (such as
RPGRIP1L,
TMEM67,
CCD2D2A and
TCTN3), with biallelic null mutations causative of lethal phenotypes, and at least one hypomorphic missense mutation found in JS [
10‐
13]. This could also hold true for
MKS1 and
B9D1. In fact, most MS fetuses are known to carry two null mutations in these genes [
4,
8]; conversely, three of our JS patients have at least one mutation not resulting in protein truncation, and the fourth is homozygous for a splice-site mutation involving the penultimate exon of
MKS1, whose pathogenetic impact on the protein remains to be determined (See Additional file
1) (Figure
1). Interestingly, a previous study identified two hypomorphic mutations in the
MKS1 gene (a missense change and a single aminoacid deletion) in a 2-year-old Turkish patient with Bardet-Biedl syndrome, another non-lethal ciliopathy partly overlapping with JS, supporting this hypothesis [
14]. Yet, these genotype-phenotype correlates are unlikely to fully explain the extreme phenotypic variability of these allelic ciliopathies, and other mechanisms, such as the presence of modifier variants in other genes, need to be explored.
MKS1, B9D1 and B9D2 proteins are known to interact physically [
15], and are main components of the “B9” or “Tectonic” complex residing at the ciliary transition zone, that includes many other proteins mutated in JS and/or MS [
16]. In our large JS cohort,
MKS1 and
B9D1 mutations each account for less than 1% cases. We failed to identify mutations in
B9D2, but we cannot exclude that this gene may also be rarely mutated in JS.
In conclusion, we expand the genetic basis of JS to include MKS1 and B9D1, delineate genotype-phenotype correlates, and further outline JS and MS as the two ends of a common spectrum. These findings have implications for genetic testing and counselling of JS patients and their families.
Acknowledgements
This work was partly supported by grants from Italian Ministry of Health (Ricerca Corrente 2013, Ricerca Finalizzata Malattie Rare 2008), Telethon Foundation Italy (Grant GGP13145), European Research Council (ERC Starting Grant 260888), National Institute of Health (grant R01NS048453) Tunisian Minister of Health and Tunisian Minister of Scientific Research.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
Patients’ recruitment, data collection, analysis of clinical and imaging data: IK, MTD, MC, LS, FM, BH, MAC, FD, AM, AZ, NGK, EMV; molecular genetic studies: MR, AM, RR; bioinformatics analysis: TM, SC; study conception and design, manuscript drafting: MR, EMV. All authors revised the manuscript critically and approved the final version.