Introduction
Joubert syndrome (JBTS; Mendelian Inheritance in Man, MIM #213300) is a rare congenital disorder characterized by a distinctive brain malformation, developmental delay with hypotonia, ocular motor apraxia, and breathing abnormalities [
1]. The diagnosis is based on a characteristic feature of brain imaging, the “molar tooth sign” (MTS) reflecting agenesis of the cerebellar vermis, and brainstem alterations. The MTS has also been described in many syndromes included in the designation “Joubert syndrome and related disorders” (JSRDs) with involvement of several other organs including the eye, kidney, and liver [
2‐
4]. JBTS is a genetically heterogeneous disorder. Thus far, 23 genes have been identified of which 22 are autosomal recessive and one is X-linked recessive.
Most genes causing JBTS encode proteins involved in primary cilia functions, thus JBTS is included in a group of disorders called ciliopathies. Mutations in several genes implicated in JBTS have been identified in other ciliopathies such as MORM syndrome (mental retardation, truncal obesity, retinal dystrophy, and micropenis; MIM #610156) [
5] and nephronophthisis (NPHP; MIM #256100), Meckel syndrome (MIM #249000) [
6,
7], and oral-facial-digital syndrome type 1 (OFD1; MIM #311200) or type 6 (MIM #277170), thus suggesting that various clinically distinct ciliopathies may be allelic. The considerable phenotypic overlap and wide variability of these ciliopathies can be explained by their common molecular and cellular etiology [
8].
We describe here a consanguineous Moroccan family with three affected siblings (two boys and one girl) showing typical signs of JBTS with hypotonia, mental retardation, bilateral nystagmus, ataxia, retinitis pigmentosa and a MTS on brain magnetic resonance imaging (MRI). To identify the underlying genetic defect in this family, we performed a genome-wide homozygosity mapping using the high-resolution Affymetrix single nucleotide polymorphism (SNP) 6.0 array followed by direct sequencing.
Discussion
The prevalence of JBTS ranges between 1:80,000 and 1:100,000 [
9‐
11]. JBTS is genetically heterogeneous with 23 genes identified thus far. We used a whole-genome approach to identify the homozygous loci in this family, followed by a targeted Sanger sequencing of
AHI1.
The homozygous p.Thr304AsnfsX6 mutation found in this family was reported in one affected patient from a consanguineous family originating from Spain [
12] and in two patients originating from the Netherlands [
11] (exon 7 for NM_017651.3). The spanish patient (family MTI-107 in the article) had compound mutations respectively in exon 8 (exon 7 in NM_017651.4) and in exon 14. The p.Thr304AsnfsX6 mutation predicts a prematurely truncated protein. Contrary to what was reported by Kroes
et al. [
11], this mutation is therefore not specific for the Dutch population.
Most
AHI1 mutations reported by Valente
et al. [
12] and Kroes
et al. [
11] are truncating, except the two V443D and R723Q missense mutations. Most mutations cluster within exons 7 to 16. The occurrence of the p.Thr304AsnfsX6 mutation in two unrelated families (Moroccan and Spanish) may be due to a founder effect, as reported for many other mutations in patients with JBTS. The c.218 G>T mutation, resulting in p.Arg73Leu in
transmembrane protein 216 (
TMEM216) has been reported with a founder effect in the Ashkenazi Jewish population [
12]. As an Arab population (Moorish or Moriscos) lived in Spain for hundreds years and many converted to Christianity, the presence of the same mutation in Arabic Moroccan and Spanish families with JBTS suggests that they might share the same founder; however, the two other patients from the Netherlands with the same (p.Thr304AsnfsX6) mutation reported by Kroes
et al. [
11] have a Dutch origin, as well as their parents, as far as several generations back (personal communication by Kroes). To investigate the genetic history of this mutation in the Moroccan, Spanish, and Dutch families with JBTS, the analysis of linkage disequilibrium (LD) between flanking polymorphic markers and AHI locus should shed light on the origin of this mutation and estimate the age of the mutation, that is, whether it was carried by a common ancestor or arose due to a mutational hotspot.
Parisi
et al. reported that 11 % of individuals with JBTS had
AHI1 mutations [
10], Kroes
et al. revealed that 16 % of patients with JBTS from the Netherlands had mutations in
AHI1 [
11], whereas Valente
et al. identified mutations in this gene in 7.3 % of individuals in a cohort of patients with JSRDs [
12]. Among this large cohort of 137 families with JSRDs and a demonstrated MTS, Valente
et al. [
12] identified 15 deleterious
AHI1 mutations in 10 families with pure JS or JS plus retinal and/or additional central nervous system abnormalities. They reported that no correlate was evident between the type of mutation (truncating, missense, or splicing) or the exon involved and the phenotypes observed [
12]. They concluded that
AHI1 mutations are a frequent cause of the disease in patients with specific forms of JSRDs but are not responsible for JSRDs with liver and kidney abnormalities. Valente
et al. [
12] found that
AHI1 mutations are a frequent cause of JS with retinal involvement or other central nervous system abnormalities, or both, which is consistent with our report.
Almost 80 % of patients with
AHI1 mutations have retinal dystrophy [
11,
13] and early onset congenital blindness [
12]. The three patients presented here and the two patients reported by Froes
et al. [
11] carrying the p.Thr304AsnfsX6 mutation showed retinal and oculomotor abnormalities (that is, retinitis pigmentosa or Leber congenital amaurosis, nystagmus). The ocular abnormalities were not observed in the Spanish patient (MTI-107, 10-years old) reported by Valente
et al. [
12].
Renal disease consistent with NPHP has also been described in patients with JBTS with
AHI1 mutations [
10,
11]. Renal ultrasound in our three patients did not reveal any abnormalities; renal disease was also absent in the Spanish and Dutch patients with the same p.Thr304AsnfsX6 mutation. However, breathing abnormalities were absent in our affected Moroccan siblings as compared to the Spanish and Dutch patients. Other neurological malformations are found to be variable among these patients with the same mutation.
JSRDs is inherited predominantly in an autosomal recessive manner. X-linked inheritance has been reported in OFD1-related JSRDs. Digenic inheritance has also been suggested. Lee
et al. reported a digenism in JBTS, resulting from a heterozygous
CEP41 mutation in combination with either a
coiled-coil and C2 domain containing 2A (
CC2D2A) or a Kinesin family member 7 (
KIF7) mutation [
14]. Such digenic inheritance has also been suggested in patients with
AHI1 mutations and
NPHP1 deletions [
11]. This observation suggests that
NPHP1 deletion leads to a large range of phenotypes that can be modified by
AHI1 mutations. The CNV analysis undertaken in our patients did not show any deletion of the
NPHP1 gene.
Our study demonstrates that homozygosity mapping is a powerful method to rapidly detect the disease-causing gene especially in clinically and genetically heterogeneous disorders such as JBTS.
Methods
Sampling and DNA extraction
The Moroccan family was recruited in the Genetics Department of the National Institute of Health, Rabat, Morocco. DNA was extracted from whole blood using the standard protocol for the three affected patients and their parents. Written informed consent for genetic analysis was obtained from the parents for themselves and for their children included in this study, which conforms to the Helsinki Declaration and local legislation.
Linkage analysis and homozygosity mapping
Affected individuals were selected for genome-wide SNP analysis using the high-resolution Affymetrix SNP 6.0 array containing 906,600 polymorphic SNPs and more than 946,000 unique probes for CNV detection. The genotyping of patients and parents was performed according to the manufacturer’s protocol. The regions of homozygosity (ROHs) were detected by the Affymetrix® Chromosome Analysis Suite (ChAS) software.
Mutation analysis
Exons and intron–exon boundaries of AHI1 were sequenced in one proband using 3130 Automated Sequencer (Applied Biosystems, Foster City, CA, USA). We used a touchdown protocol for the polymerase chain reaction (PCR) procedures to amplify all exons in the same conditions. Primer sequences and PCR conditions are available on request.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
SCE, AS, MM and YK made substantial contributions to the clinical investigation of the family, blood collection and DNA extraction and article redaction. MC made a contribution to array experiment. NE and TAB did the sequencing analysis. LB and TAB were involved in the interpretation of data and in article redaction and discussion. TAB, LB and AS gave approval of the final version to be published. All authors read and approved the final manuscript.