We report a maternally inherited deletion of approximately 5.8 Mb at Xq21.1, in a male subject with developmental delay, dysmorphic facial features, cleft palate, intellectual disability, short stature and hearing loss. The deleted region includes 14 OMIM genes: CYSLTR1, ZCCHC5, LPAR4, P2RY10, GPR174, ITM2A, TBX22, BRWD3, HMGN5, SH3BGRL, HMGN5, SH3BGRL, POU3F4, CYLC1 (Fig. 4 and Table 1). Among these, TBX22, BRWD3 and POU3F4 well explain some of the clinical features of this patient.
Mutations in the TBX22 gene are a well established cause of X-linked cleft palate with ankyloglossia as well as contributing to the prevalence of isolated cleft palate [5]. More rarely, other craniofacial anomalies including cleft lip and hypodontia have also been related to TBX22 variants [6]. The phenotypic spectrum of subjects bearing TBX22 mutations can vary, even within the same family, from asymptomatic females to males or females with a bifid uvula, a cleft of the soft palate, or a complete cleft of the hard and soft secondary palate, along with ankyloglossia [7].
Genetic hearing loss has an extremely varied etiology, with a plethora of genes involved in the autosomal and X-linked forms with overlapping phenotypes [8]. Among the main genetic defects, mutations within connexin 26 (GJB2) and connexin 30 (GJB6) are usually associated with severe to profound sensorineural deafness. Our patient had been screened for the presence of mutations in GJB2 before performing array-CGH, and was negative. The deafness causative gene is undoubtedly POU3F4, which causes the X-linked neurosensorial deafness DFN3. This defect is associated with either POU3F4 point mutations or small deletions in a region located at 900 Kb upstream of the gene and disturbing a regulatory element [9]. Male patients show a well characterized phenotype with both stapes fixation and progressive mixed hearing loss [10, 11]. Hearing loss was also observed in the mother of our patient as well as in about 40 % of the POU3F4 mutation carrier females. Carrier females usually exhibit a postlingual onset of the hearing impairment that progresses over time [12] with a variable expressivity attributable to variations in the degree of skewing of X inactivation.
Loss of function mutations affecting BRWD3 are associated with a phenotype including mild to moderate intellectual disability, macrocephaly, dysmorphic facial features, skeletal signs and behavioral disturbance [13, 14]. Among these alterations, a partial deletion encompassing 74 Kb and including the 30 last exons of the BRWD3 gene was reported in a male with high forehead, deep-set eyes, hypertelorism, short palpebral fissures, anteverted nares, downturned corners of the mouth, pointed chin, and skeletal anomalies. This phenotype is similar to that of the patient presented here: macrocephaly, dysmorphic facial features including prominent forehead and abnormal ears, behavioral disturbance, skeletal symptoms like pes planus, and cubitus valgus. These findings strongly suggest that the lack of BRWD3 causes intellectual disability, macrocephaly and possibly the skeletal symptoms (pes planus and cubitus valgus) observed in our patient. It has been demonstrated that BRWD3 plays a crucial role in ubiquitination, as part of the ubiquitin/proteasome system. In Drosophila, BRWD3 belongs to the CUL4-ROC1-DDB1 E3 ligase complex in which it acts as a CULLIN (CUL)4-associated factor that mediates light-dependent binding of CRY (Cryptochrome, a circadian photoreceptor) to the complex, inducing the ubiquitination of dCRY and its light-induced degradation [15]. The ubiquitin-proteasome system plays a crucial role in brain development and is a critical regulator of the synaptic plasticity and long-term memory formation [16]. Intriguingly, in humans mutations of the CUL4B, which encode a ubiquitin 3 ligase subunit cause an X-linked syndrome characterized by intellectual impairment, macrocephaly, central obesity, hypogonadism, pes cavus and tremor [17], a phenotype that largely overlaps with that observed in patients carrying BRWD3 mutations/deletions, including the present case. As both BRWD3 and CUL4B are part of the same complex it is likely that alterations of BRWD3 influence the ubiquitin-proteasome system similarly to other intellectual disability syndromes, whose prototype is represented by the Angelman Syndrome which is caused by mutations of the ubiquitin ligase-encoding UBE3A gene (MIM 105830).
The search in the Decipher and ISCA databases for similar size deletions at Xq21.1 led to the identification of six deletions in males with syndromic intellectual disability and developmental delay (DECIPHER 287069,253735, 257570,292180 and ISCA nsv533989, nsv530231, Fig. 4) and in one patient with no further phenotype details (DECIPHER 283100, Fig. 4) but none of these exactly corresponds to that of our patient. However, among those observed in female patients, there are two deletions (DECIPHER 262726 and 265126, Fig. 4) that completely cover that of our patient. The milder associated phenotype includes broad nasal tip, skeletal defects, speech delay, and cleft palate, which are also features of the present patient.
Other remarkable clinical signs of our patient are short stature and multiple pituitary hormone deficiency. The presence of mutations in PROP1, the most common genetic cause of combined pituitary hormone deficiency (CPHD) associated with short stature [18], was excluded. It is interesting that the deletion partially overlaps with regions duplicated in patients with multiple congenital anomalies, developmental delay, GH deficiency and short stature [19–21]. This suggests that Xq21.1 might contain either a protein coding gene or a regulatory element involved in pituitary hormone secretion. Within the 5.8 Mb region there is no evidence for a protein coding gene directly involved in pituitary functioning. However, ITM2A, included in the deletion, might be related to the severe short stature as it encodes an integral transmembrane protein involved in early cartilage development [22]. It has been suggested that the expression of ITM2A influences the chondrogenic differentiation potential of mesenchymal stem cells in vitro [23]. It might be hypothesized that the absence of ITM2A greatly influences the cartilage development with a possible impact on postnatal growth.