Screening of CD96 and ASXL1 in 11 patients with Opitz C or Bohring–Opitz syndromes

INTRODUCTION

Opitz C syndrome (OTCS, also known as the C trigonocephaly syndrome, MIM #211750) is a rare disorder of trigonocephaly and associated hypertelorism, multiple buccal frenula, limb defects, wide alveolar ridges, visceral anomalies, redundant skin, hypotonia, psychomotor, and mental retardation. Less than 60 cases are known worldwide [Opitz et al., 2006].

Although the molecular basis of OTCS and its inheritance are unknown, this disorder was recently associated with mutations in the CD96 gene (Cluster of Differentiation 96, also called TACTILE for “T cell activation increased late expression,” MIM *606037) [Kaname et al., 2007]. In an OTCS individual, these authors discovered a balanced chromosomal translocation, t(3;18), that disrupted the CD96 gene at the 3q13.3 breakpoint. They also identified a missense mutation (c.839C > T, p.Thr280Met) in exon six of CD96 in one BOS patient. The CD96 gene encodes the CD96 antigen, a member of the immunoglobulin superfamily, which plays a role in cell adhesion [Fuchs et al., 2004]. Cells with mutated CD96 protein display loss of adhesion and growth activities in vitro [Kaname et al., 2007]. Moreover, the expression patterns of this gene, including bone and fetal brain are consistent with organs and tissues involved in the abnormalities of the C syndrome.

Recently, the role of CD96 in OTCS was questioned by Darlow et al. [2013]. These authors found a disruption of the CD96 gene in patients bearing a t(2;3) balanced translocation and who developed renal cell carcinoma without symptoms of OTCS. To reconcile both findings, it was proposed that the breakpoint region of chromosome 18 involved in the translocation found by Kaname et al. [2007] could bear the gene responsible for OTCS in that patient.

Regarding inheritance, some reports of affected sibs support autosomal recessive segregation [Antley et al., 1981; Opitz et al., 2006]. Likewise, normal karyotypes in most patients, unaffected parents with multi-affected offspring, an equal sex ratio of affected individuals, and consanguineous mating support autosomal recessive inheritance. However, many other patients are sporadic cases, which suggests the possibility of dominant inheritance or germ-line mosaicism [Sargent et al., 1985]. OTCS is highly variable, so that many symptoms that define OTCS overlap with Bohring–Opitz syndrome (BOS, MIM #605039). While some experts have defined these pathologies as two different conditions, others think that they are different presentations of the same disorder, ranging from a milder form (OTCS) to a more severe condition (BOS) [Bohring et al., 1999]. BOS is a rare malformation syndrome with only 30 cases described so far [Hastings et al., 2011]. It is characterized by exophthalmos, hypertelorism, severe intrauterine growth retardation, poor feeding, profound mental retardation, trigonocephaly, prominent metopic suture, nevus flammeus of the forehead, upslanting palpebral fissures, hirsutism, and flexion at the elbows, and wrists. The molecular basis of BOS and its inheritance are incompletely understood. It was associated with dominant mutations in the ASXL1 gene (Additional Sex Combs-Like 1; MIM *612990). ASXL1 gene maps to chromosome position 20q11.21 and it encodes the additional sex comb-like protein 1, which is a mammalian homolog of Drosophila asx. Recently, Hoischen et al. [2011] discovered that seven out of 13 BOS patients were heterozygotes for de novo mutations of this gene. Similar mutations in ASXL1 were found in two BOS patients by Magini et al. [2012]. Although BOS is considered an autosomal dominant pathology due to de novo mutations in ASXL1, the case of two sibs diagnosed with BOS [Greenhalgh et al., 2003], suggests that this malformation syndrome could also display an autosomal recessive pattern of inheritance (although gonadal mosaicism should also be considered in this case, consistent with AD inheritance).

In this study, CD96 and ASXL1 were analyzed in 11 patients with OTCS or BOS.

MATERIALS AND METHODS

Patients and Samples

Most of the patients were diagnosed as OTCS or BOS by two of us (J.M.O., G.N.), and patients P9a and b, and P10 were diagnosed elsewhere (Fig. 1). Written informed consent was obtained from all patients’ parents. We have obtained the consent to publish pictures of patients P5, P6, P7, P9a, P9b, P10a, and P10b.

Genomic DNA was obtained from the patients and parents from peripheral blood using Wizard® Genomic DNA Purification Kit (Promega, Madison, WI). Quantity and quality of the gDNA was assessed using a Nanodrop ND-1000 Spectrophotometer (NanoDrop Technologies Inc., Wilmington, DE).

RESULTS

Patients’ Phenotypes

The available clinical findings of the patients are summarized in Table III and pictures of most of them are shown in Figure 2. Patient P1 was previously reported (and pictures included) in Lalatta et al. [1990].

Table III. Clinical Features

	P1	P2	P3	P5	P6	P7	P8	P9a	P9b	P10a	P10b	P11
Age at diagnosis (y)	4	14	NA	2	2	2	4	0.5	0	1	0	6
Consanguinity	No	No	NA	NA	No	No	No	No	No	No	No	No
MANIFESTATIONS CNS
Developmental retardation	+	+	+	+	+	+	+	Very mild	No	+	+	+
Motor developmental delay								+	+
Seizures				+	+	+	+
Brain MRI				(1)		(2)	(3)	(4)	Normal			(5)
HEAD
Trigonocephaly	+	+	+	+	+	+	(6)	+	+	+	+	+
Facial anomalies	+	+	+	+		+	+	+	+	+	+	+
Myopia				+						+	NA
Strabismus	+	+	+							+	NA
Low-set and abnormally shaped ears	+			+	+		+			+	+	+
Forehead hemangioma										+	+
Cutis marmorata/livedo reticularis							+					+
NECK
Short, webbed neck			+		+					+	+
THORAX
Widely spaced nipples	+										+
Scoliosis						Mild		Severe	Severe			Severe
LIMBS
Short limbs	+
Flexion deformities	+				+	+		+	+
Brachydactyly	+									+
Single palmar crease	+							+	+
Ulnar deviation of the fingers	+				+			+
OTHER
Generalized muscular atrophy	+							+	+
Muscular hypotrophy												+
Hypotonia				+	+	+	+	+	+	+	+	+
Cryptorchidism		Bilateral	Bilateral					Left	Bilateral
Horseshoe kidneys		+
Failure to thrive				+	+					+

• NA, not available. (1) Posterior thinning corpus callosum noted on MRI; (2) Mild enlargement of the supratentorial ventricular system, mainly in the frontal horns, and asymmetric lateral ventricles. Mild elevation of the choline in the spectroscopy. (3) A brain MRI showed moderate distension of the bodies and back horns of tricorns and of parieto-occipital subarachnoid spaces; (4) MRI CNS: Dimorphic ventriculomegaly with thin corpus callosum and white matter gliosis; (5) Brain MRI showed peritrigonal white matter thinner and mild ventricular enlargement; (6) mild brachycephaly, prominence of the metopic suture without definite trigonocephaly.

DISCUSSION

Little is known of the molecular bases of the OTCS and BOS syndromes. To date, only one gene, CD96, has been associated with OTCS by Kaname et al. [2007], who reported an OTCS patient bearing a balanced chromosomal translocation, t(3;18)(q13.13;q12.1) that disrupted CD96 at 3q13. The precise breakpoint was within CD96 exon five and led to a premature termination of the protein. In that patient, CD96 mRNA expression was reduced to 50%. Moreover, in the mutation analysis of nine karyotypically normal patients diagnosed with OTCS or BOS, Kaname et al. [2007] also identified a missense mutation (c.839C > T, p.Thr280Met) in one BOS patient, not found in controls. In our study, we did not find any CD96 mutation in the 11 patients included in the present study.

The relationship between CD96 and Opitz C syndrome was questioned recently as this gene was found disrupted in patients with renal cell carcinoma without any resemblance to Opitz C patients [Darlow et al., 2013]. Alternatively, it was proposed that in Kaname's patient, the breakpoint region on chromosome 18 (18q12.1) may be the one bearing the OTCS causative mutation. This region in now known to comprise the desmocollin and desmoglein gene clusters, responsible for tight junction formation in epithelia. Additionally, it includes the ASXL3 gene (a paralog of ASXL1), which was recently associated with a syndrome with phenotypical overlap with BOS [Bainbridge et al., 2013]. It should be very interesting to sequence this gene in a sample from the patient with the t(3;18) translocation described by Kaname et al. [2007].

In our study, we scrutinized the exon sequences within 3.22 Mb of chromosome region 18q12.1 of the six patients subjected to WES. In patient P1 (OTCS) we found two mutations in the DSC2 gene (p.Glu46Lys and p.Glu102Lys). Both mutations are present in the general population albeit with a low frequency (MAFs <0.001) and both are classified as benign in Polyphen2 (with scores of 0.006 and 0.016, respectively). The p.Glu102Lys mutation was reported previously in a heterozygous state in a patient with arrhythmogenic right ventricular cardiomyopathy (ARVC) [Anastasakis et al., 2012]. The clinical manifestations of patient P1 were previously reported by Lalatta et al. [1990] (as patient 3; C.L.), and at age 4 years, this patient did not present any kind of cardiomyopathy. However, ARVC is usually manifest in late adulthood. Interestingly, Andreasen et al. [2013] have pointed out that a large number of the mutations reported as associated with a cardiomyopathy are not, in fact, the monogenic cause of the disease. Many of these changes (14–18%), including the p.Glu102Lys mutation, are found in public databases such as the NHLBI Exome Sequencing Project (ESP). According to the observed frequency of these mutations in ESP, the expected frequency of ARVC should be 1:5, a prevalence far higher than that observed in the general population, which is estimated as 1:5000. Also, when compared to the ones not present in the public databases, the mutations found in ESP differ significantly in the number of nonsense mutations (lower) and in the PolyPhen2 prediction (often classified as “benign”). All things considered, we understand that the DSC2 mutations found in patient P1 are not responsible of her OTCS phenotype. No other putative pathogenic mutation was found in any of the remaining exons of the genes in the 18q12.1 region in the remaining patients included in the exome project. This includes ASXL3, a very interesting candidate. However, the exome analysis does not allow for the inspection of regulatory sequences located outside the exons.

In our study we have assessed the ASXL1 gene, both in OTCS and BOS patients, and we have found a loss-of-function mutation in one patient with a clear diagnosis of BOS (P5). This mutation (c.2100dupT, p.Pro701Serfs*16) consists of an insertion of a T at position 2100 of the cDNA, leading to a change in the amino acid at residue 701 and a premature truncation of the protein. This change was found in patient P5 but was absent in his parents, being a de novo change. This mutation is located in exon 13, as all but one (p.Arg404X) of the ASXL1 mutations previously associated with BOS [Hoischen et al., 2011; Magini et al., 2012]. And, as all of them, it leads to a premature end of the protein. All these facts suggest that this mutation is the cause of the BOS phenotype in the P5 patient.

In summary, we have not found any mutation in the CD96 gene or in the 18q12.1 region that could explain the OTCS or BOS phenotypes of our patients. In contrast, we have found a de novo mutation in the ASXL1 gene in the BOS patient, adding a novel mutation to those described for BOS and found no ASXL1 mutation in OTCS patients. Phenotypically, it is hard to define salient features allowing a clear distinction between these two conditions, except perhaps for the severity, which tends to be more marked in the BOS. The genes might provide a better distinction. For example, ASXL1 mutations were found only in BOS. But half of the BOS patients remain undiagnosed at the molecular level. However, it may happen that as soon as there is a new gene for OTCS, a patient with mutation in this gene will be considered BOS by some clinicians. Further analysis of the exome sequencing, in a hypothesis-free approach, currently underway, should help shed light on the molecular bases of OCTS.