Portal de Periódicos da CAPES

Gray platelet syndrome: Novel mutations of the NBEAL2 gene

2016; Wiley; Volume: 92; Issue: 2 Linguagem: Inglês

10.1002/ajh.24610

1096-8652

Roberta Bottega, Elena Nicchia, Caterina Alfano, Ana C. Glembotsky, Annalisa Pastore, Debora Bertaggia Calderara, Bettina Bisig, Michel A. Duchosal, Guillermo Arbesú, Lorenzo Alberio, Paula G. Heller, Anna Savoia,

Cell Adhesion Molecules Research

Gray platelet syndrome (GPS) is a rare inherited macrothrombocytopenia characterized by reduction of α-granules in platelets and megakaryocytes associated with mild-to-moderate bleeding and myelofibrosis 1. As reported in at least 28 unrelated families 1, GPS is caused by mutations of NBEAL2, the gene encoding for the neurobeachin-like-2 protein. NBEAL2 is a member of the family containing the BEACH (BEige And Chediak Higashi) domain, a conserved region involved in vesicular trafficking that may be critical for the α-granule development 2. Here, we report novel mutations of NBEAL2 in two affected individuals (P1 and P2), who were previously diagnosed with immune thrombocytopenia (ITP) and then suspected to have GPS because of absence of azurophilic granules on May-Grünwald-Giemsa staining. In P1, sequencing analysis identified an homozygous missense variant (c.6212G > C; p.Arg2071Pro; Supporting Information Fig. S1). Since her parents were not available for the segregation analysis, we hypothesised that the two mutant alleles were identical by descent because of homozygosity of all the polymorphic markers at the NBEAL2 locus (data not shown). Moreover, the potential hemizygous condition was excluded using statistical analyses of NBEAL2 amplicon coverage as previously reported (Supporting Information Fig. S1C) 3. In P2, we detected one maternal nonsense (c.3839C > T; p.Arg1280*) and one paternal missense (c.6477C > G; p.His2159Gln; Supporting Information Fig. S1). The three NBEAL2 variants are reported in SNPs databases but with a minor allele frequency <0.01%. Whereas the deleterious effect of nonsense mutations is usually associated with loss of function, the pathogenic role of the missense variants is not readily obvious. Several bioinformatics tools have been developed to predict the effect of amino acid substitutions. Among these, the CADD (Combined Annotation Dependent Depletion) score is relatively high for both p.Arg2071Pro and p.His2159Gln (Table 1). Moreover, the two affected residues are well conserved in orthologs of different species, suggesting that they are important for the NBEAL2 activity (Supporting Information Fig. S1G). Indeed, they affect the BEACH domain, a region that is highly homologous with that of another two human proteins, neurobeachin (NBEA) and LPS-responsive vesicle trafficking, beach and anchor protein (LRBA; Supporting Information Fig. S2A) 2. Even in NBEA and LRBA residues Arg2071 and His2159 are conserved (Supporting Information Fig. S2B). Using the crystal structure of the BEACH domain in NBEA and LRBA, we found that Arg2071 is placed in a rigid turn of a helical hairpin and is involved in a hydrogen bond with Glu2270 (Supporting Information Fig. S2C). His2159 is unusually shielded in the core of the structure and not exposed to the solvent because of a polar interaction with Thr2333. Of note, residues Glu2270 and Thr2333 are also conserved in NBEAL2, NBEA, and LRBA (Supporting Information Fig. S2B). Therefore, we can hypothesize that the Arg2071Pro and His2159Gln amino acid substitutions affect the energetics of the hydrogen bonding, which needs to be optimal to allow rapid sampling and kinetics of folding, conferring stability to the structure. Although the specific function of the BEACH domain remains unclear, it is likely to interact with the PH domain and form a large groove as a binding site for ligands. Therefore, p.Arg2071Pro and His2159Gln, as well as the others missense mutations affecting the BEACH domain, are likely to prevent interactors from binding NBEAL2 (Supporting Information Fig. S2A). Consistent with the identification of mutations in the NBEAL2 gene, the two affected individuals had thrombocytopenia with platelet anisocytosis and in their platelets α-granules were almost absent (Table 1; Supporting Information Fig. S3). Accordingly with our previous study 4, the parents of P2 had a moderate reduction of the α-granule number. Although we cannot exclude genetic heterogeneity in GPS, NBEAL2 is the only gene known whose biallelic mutations are associated with severe α-granule deficiency. Considering the expression level of glycoproteins (GP) on the platelet membrane, GPI/IX/V and GPIIb/IIIa were expressed at the normal levels whereas GPVI was slightly reduced at least in P1 (Table 1). Indeed, after stimulation with convulxin, a direct agonist of collagen receptor GPVI, the platelets of P1 were defective in P-selectin exposure, dense granule secretion, and GPIIb/IIIa activation. Considering that defects of platelet aggregation with collagen have previously been reported in GPS 1, a systematic evaluation of the GPVI expression would be of interest to ascertain whether its reduction is a symptomatic feature in GPS. Similarly, the PAR1 mediated platelet response could be another common defect in the disease, as a recent study showed low expression level of the PAR1 and PAR4 thrombin receptors 5. Therefore, PAR1 was analysed in P2, confirming the data from the literature. As a consequence, after stimulation with thrombin or thrombin receptor activating peptides (TRAP-6), we observed the same defective platelet activation as that observed with convulxin (Table 1). Consistent with the identification of the NBEAL2 mutations, P1 had slight/moderate bone marrow fibrosis but no enlarged spleen 6. Megacaryocytes were small in size with hypolobulated nuclei and extensive emperipolesis. Of note, the patient had 2.7–6.0 ×109/L white blood cells with underlying lymphocytopenia of 0.8–1.4 ×109/L but without a history of recurrent or severe infections (Table 1). Being not reported in GPS, this mild lymphocytopenia could be independent of α-granules deficiency. On the contrary, P2 had increased serum B12, splenomegaly, and emperipolesis but not myelofibrosis (Table 1). Considering that P1 is older than P2, there is significant variable expressivity of the disease between the two patients. This is also consistent with the degree of their bleeding diathesis. Whereas P1 had negative bleeding history and underwent two elective uneventful caesarean sections with prophylactic treatments, P2 suffered from severe epistasis that required platelet transfusions. Although the α-granule deficiency is of similar extent in the two patients, we cannot exclude that the homozygous p.Arg2071Pro mutation identified in P1 allows NBEAL2 to retain some residual activity. In conclusion, we report novel biallelic NBEAL2 mutations associated with variable expressivity of GPS. Moreover, we support the hypothesis that defective expression of GPVI and the thrombin receptors on the GPS platelets are additional signs of the disease, whose recognition is fundamental to prevent misdiagnosis of ITP and undue therapy. All authors agreed with the content of the manuscript and approved the final version. Study design: R. Bottega and A. Savoia Collected clinical data: M.A. Duchosal, G. Arbesu, L. Alberio, and P.G. Heller Laboratory analysis: R. Bottega, E. Nicchia, A.C. Glembotsky, D. Bertaggia Calderara, and B. Bisig Data analysis and manuscript writing: R. Bottega, C. Alfano, A. Pastore, L. Alberio, P.G. Heller, and A. Savoia The authors thank Mrs C. Chapuis-Bernasconi (Institute of Pathology, CHUV University Hospital and University of Lausanne, Lausanne, Switzerland) for her excellent work on platelet electronic microscopy. This work was supported by grants of the Italian Ministry of Health (RF-2010-2309222), Telethon Foundation (Grant GGP13082), IRCCS Burlo Garofolo (Ricerca Corrente), and Cariplo Foundation (2012-0529). Roberta Bottega1, Elena Nicchia2, Caterina Alfano3, Ana C. Glembotsky4,Annalisa Pastore3, Debora Bertaggia-Calderara5, Bettina Bisig6,Michel A. Duchosal5, Guillermo Arbesú7, Lorenzo Alberio5, Paula G. Heller4,Anna Savoia1,2 1Department of Medical Sciences, University of Trieste, Italy 2Institute for Maternal and Child Health – IRCCS Burlo Garofolo, Trieste, Italy 3Maurice Wohl Institute, King's College London, UK4Instituto de Investigaciones Médicas Alfredo Lanari, Universidad de Buenos Aires, CONICET, Buenos Aires, Argentina 5Service and Central Laboratory of Hematology, University Hospital of Lausanne (CHUV), Switzerland 6Institute of Pathology, CHUV University Hospital and University of Lausanne, Switzerland 7Hospital Pediátrico Dr. Humberto Notti, Mendoza, Argentina Additional Supporting Information may be found in the online version of this article. Supporting Information Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.