A new mutation of ANO 6 in two familial cases of Scott syndrome

Scott syndrome is one of the less frequently occurring platelet function disorders. The diagnosis of this rare disorder relies on flow cytometry and molecular diagnosis. To date, only three mutations (two intronic transitions and one exonic insertion) of the gene encoding Anoctamin 6 (ANO6) have been reported (Suzuki et al, 2010; Castoldi et al, 2011). We describe here, in two affected siblings, the same anomalies of a new exonic substitution in ANO6, associated with a large deletion of this gene on the other allele. Scott Syndrome (Online Mendelian Inheritance in Man 262890) is a very rare inherited bleeding disorder, related to an impaired exposure of phosphatidylserine (PS) on the outer leaflet of the platelet membrane and a severe decrease in the production of microparticles (PMP) after platelet activation. Both these defects are responsible for a reduced catalytic platelet surface. As a consequence, vitamin K-dependent clotting factors cannot correctly interact with cofactors, which leads to reduced thrombin generation and impaired clot formation. Phosphatidylserine exposure in platelets is under the control of three types of proteins: flippases (inward directed pumps), floppases (outward directed pumps) and scramblases, which can generate bidirectional redistribution. After platelets’ activation, scramblase and floppase are triggered, while flippase is inhibited, thereby allowing for the exposure of PS on the outer leaflet of the platelets’ membrane. The exact mechanisms involved in PS exposure are complex but are on the way to be elucidated. Indeed, Ishihara et al (2016) recently demonstrated the role of calcium in sustaining phospholipd scrambling. They notably demonstrated the role of point mutations in transmembrane proteins such as the ANO6 (also termed TMEM16F) scramblase in impairing this critical role of calcium. The abnormal PS exposure and PMP production observed in Scott syndrome are the consequence of a protein defect. ANO6, a scramblase of 910 amino acids long, with 8 transmembrane segments, is a key regulator of calcium-dependent PS exposure. This protein is ineffective in Scott syndrome patients. Yet, under proapoptotic conditions not requiring calcium, PS exposure is normal in these patients (Kojima et al, 1994; Suzuki et al, 2010; Castoldi et al, 2011; van Kruchten et al, 2013; Fujii et al, 2015). Furthermore, ANO6 may form chloride (Cl) channels, although the link between Cl-channels and PS exposure is still an unresolved question (Tian et al, 2012; Harper & Poole, 2013). Changes of calcium concentrations in platelets seem to be important for PMP generation, although the fine regulation of microparticles shedding remains unclear. The molecular bases of Scott syndrome have recently been described, by the demonstration of a mutation of the ANO6 gene located on the long arm of chromosome 12 (Suzuki et al, 2010). Two other mutations in one patient have been described since (Castoldi et al, 2011). Here we report two siblings who were referred to our department for a bleeding phenotype in surgical situations. The whole haemostasis exploration was normal except for increased residual serum prothrombin (respectively 65% and 67%, normal range <5%). Scott syndrome was suspected and the analysis of PS expression on activated platelets’ surface by flow cytometry (Halliez et al, 2015) confirmed the anomaly. We previously reported that, for both patients, no more than 7% of platelets expressed PS after activation by ionophore, as shown by fluorescent annexin binding, versus more than 95% in controls (Fig 1A) (Halliez et al, 2015). This confirmed the absence of flip-flopping and thus of scramblase, consistent with increased residual serum prothrombin. Moreover, thrombin generation assay (using PRP Reagent low® [Stago, Asnières, France] and a Thrombinoscope® [Thrombinoscope BV, Maastricht, Netherlands]) was performed according to the recommendations of the manufacturer. This showed a dramatic difference between a healthy subject and one of the patients: the endogenous thrombin potential was 1470 nmol/min for the healthy subject and 465 nmol/min for the patient (Fig 1B). In order to confirm the implication of ANO6 protein, after obtaining informed consent, genetic analysis of ANO6 was performed, using the HaloPlex Target Enrichment System (Agilent Technologies, Santa Clara, CA, USA) followed by sequencing on a MiSeq Platform (Illumina, San Diego, CA, USA). ANO6 is located on the 12q12-q13.11 chromosomal region and spans 2733 pb with 21 exons. A nonsense variant c.889C>T (p.Arg297*) was identified on exon 8 and confirmed to be homozygous by Sanger sequencing (Fig 2A) on both the proband and her affected sister. In order to determine whether there was a homozygous status of the mutation c.889C>T or an heterozygous configuration with a deletion on the second allele, we performed a copy number variation (CNV) analysis by read depth approach on next generation sequencing (NGS) raw data (Tan et al, 2014). For each exon of ANO6, the mean read depth per patient was calculated and normalized to the proportion of reads per patient in the whole sequencing run (23 patients with other suspected anomalies). The CNV ratio for each exon was calculated by the mean read depth of the patient compared to the mean of the 23 other patients of the whole run. This identified the same deletion in both patients, at least from exon 1 to exon 10 (Fig 2B). The deletion was confirmed by a specifically designed quantitative polymerase chain reaction (data not shown). The consequence of this deletion is likely to be the absence of expression of ANO6, associated, through the p.Arg297* mutation, with an activation of nonsense-mediated decay, both contributing to a complete abolition of ANO6 production, thereby explaining the pathophysiology of Scott syndrome in these two siblings. Only three different deleterious sequence variations have been described in Scott syndrome, which are splicing variations. We report the first association of a deletion and a nonsense variation explaining the pathophysiology of Scott syndrome. The additional originality of these observations is that both sisters probably inherited the monoallelic ANO6 deletion from one parent and the monoallelic mutation from the other parent. Because the condition however remained clinically silent for a long time, it was not possible to collect DNA from the parents. Moreover, there were no other siblings in the family. Nonetheless, the in-depth molecular analysis of ANO6 in these sisters strongly supports this hypothesis. Pierre Boisseau, Marie-Christine Bene, Stéphane Bezieau and Marc Fouassier designed the study and contributed to manuscript writing. Pierre Boisseau, Thomas Besnard, Sinthuja Pachchek, Mathilde Giraud, Patricia talarmain, Nelly Robillard, and Marie Annick Gourlaouen performed research and analysed data. Marc Fouassier and Marie-Christine Bene contributed to clinical data analysis. None declared.