Heterozygotes in the bernard-soulier syndrome do not necessarily have giant platelets or thrombocytopenia.

Abstract

Although platelet size heterogeneity has been reported in heterozygotes with Bernard–Soulier syndrome (BSS), giant platelets and thrombocytopenia have been assumed to be characteristics only of individuals who are homozygous for the disease and not those who are carriers (Nurden & George, 2000). Doubt was cast on this theory following the publication of a report showing that autosomal dominant macrothrombocytopenia in Italy is most frequently associated with heterozygous BSS (Savoia et al, 2001). This was particularly apparent in patients with an Ala156→Val substitution in the glycoprotein (GP)Ibα subunit, a mutation known as the Bolzano variant. A logical conclusion was that this mutation induced adverse autosomal dominant effects during megakaryocyte development by an unknown mechanism. Another missense mutation has since been shown to be associated with the presence of giant platelets in a heterozygous form of BSS in several Japanese subjects (Kurokawa et al, 2001). This is a Tyr88→Cys mutation in the GPIbβ subunit, a substitution that severely affected GPIb–IX expression in homozygous individuals. In GPIbα knockout mice, ultrastructural modifications can be seen early in megakaryocyte development, and there is aberration of both the partitioning and even development of membrane systems as maturation proceeds (Poujol et al, 2002). The human transgene restored megakaryocyte morphology, thereby proving unambiguously that the absence of the GPIb–IX complex is associated with giant platelet production. We have recently located a novel GPIbβ mutation in a French BSS patient (Hillman et al, 2002). In this patient, Pro29 in one of the GPIbβ alleles was substituted with a Leu (GPIbβ:P29L), whereas in the second allele a microdeletion at the 22q11 locus of the GPIbbβ gene led to clinical diagnosis of the DiGeorge syndrome. The pedigree was determined and showed that the GPIbβ:P29L mutation was inherited from the father, who had no history of bleeding. Closer examination showed that the father had a whole blood platelet count of 168 × 109/l and a mean platelet volume of 9·0 fl (control range 7·0–10). This compared with a typical platelet count for the patient of 40 × 109/l and a mean platelet volume of 13·8 fl. Figure 1A shows dot-plot patterns of side scatter versus forward scatter for platelets gated using a monoclonal antibody (AP-2) directed against the αIIbβ3 integrin. The size heterogeneity and the presence of enlarged platelets in the patient's sample are clearly shown. In contrast, giant platelets were virtually absent from the father (only 3·2% of the platelets were in the lower right-hand quadrant). Western blotting analysis of both GPIbα (not shown) and GPIbβ expression (Fig 1B) clearly showed that the father expressed intermediate amounts of these subunits, a finding that was also confirmed by flow cytometry. Thus, the presence of a heterozygous GPIbβ mutation had no significant influence on either platelet size or platelet number in his case. Therefore, not all GPIbβ mutations are associated with familial thrombocytopenias of the heterozygous form. Samples of platelet-rich plasma from a control donor, the BSS patient and her father were prepared by differential low-speed centrifugation to avoid loss of giant platelets (Hillman et al, 2002). (A) Samples incubated with AP-2 followed by fluorescein isothiocyanate (FITC)-labelled anti-mouse IgG and their size parameters analysed by forward scatter and side scatter in a Becton Dickinson FACScan (Le Pont de Claix, France). (B) Washed platelets were solubilized with sodium dodecyl sulphate (SDS) and disulphide-reduced proteins (50 µg) separated by SDS polyacrylamide gel electrophoresis prior to being probed with a monoclonal antibody (MBC.257.4, kindly provided by Professor Dermot Kenny) to GPIbβ as previously described (Hillman et al, 2002). Molecular markers (MM) of 33·5, 25 and 16·5 Kb are shown. How can such apparently contradictory findings be explained? Giant platelet syndromes can have mutations in the myosin heavy chain in addition to genes encoding the GPIb–IX complex (Heath et al, 2001). For normal platelet production to occur, cytoskeleton-driven membrane reorganizations may be a vital part of megakaryocyte maturation. It is necessary for the genes involved in these processes to be identified and their roles studied. One possibility is that, as a result of genetic variability, an intermediate level of GPIb–IX production will be insufficient to ensure membrane reorganization in some patients, particularly if abnormally synthesized GPIbaα, GPIbbβ or GPIX gene products can interfere with the normal functioning of the unaffected allele product.