Mechanism of Platelet Factor (PF4) Deficiency with RUNX1 Mutations: RUNX1 Is a Transcriptional Regulator of PF4.

Abstract 227

RUNX1/CBFA2 (Core binding factor A2) is a major transcription factor involved in hematopoiesis. RUNX1 mutations are associated with mild thrombocytopenia, platelet dysfunction, and predisposition to acute leukemia. Several patients with mutations in RUNX1 have been shown to have alpha granule deficiency characterized by decreased platelet factor 4 (PF4) content. The mechanisms leading to PF4 deficiency remain unclear in most patients with α-granule abnormalities or the Gray platelet Syndrome (GPS). GPS is a heterogeneous disorder, and defective granule formation and targeting of proteins to the granule have been postulated. Previous studies from our group have documented a patient with mild thrombocytopenia, impaired platelet aggregation, secretion, phosphorylation of pleckstrin and myosin light chain (MLC), and GPIIb-IIIa activation, which was associated with a heterozygous mutation in transcription factor RUNX1. Platelet expression profiling of this patient showed decreased expression of several genes including chemokine PF4 and its non-allelic variant PF4V1. Platelet PF4 protein was also decreased. PF4 is mainly expressed in megakaryocytes and platelets, and it serves as a lineage-specific marker of megakaryocytic differentiation. Because PF4 is downregulated in platelets from our patient with a RUNX1 mutation, we addressed the hypothesis that PF4 and PF4V1 may be direct transcriptional targets of RUNX1. Computer based TFSEARCH analysis showed six RUNX1 consensus sites on PF4 upstream (2 kb) region and eleven RUNX1 sites on PF4V1 upstream sequence (2 kb). To assess in vivo binding of RUNX1 to PF4 upstream region chromatin immunoprecipitation (ChIP) assay was performed using HEL cell genomic DNA and RUNX1 antibody. These studies were done in HEL cells treated with phorbol 12-myristate 13-acetate (PMA) to induce megakaryocytic transformation. ChIP assay revealed in vivo RUNX1 binding in two regions encompassing RUNX1 sites at −1768 (TGTGGT) and −151 (ACCGCA) on PF4 promoter. These sites were pursued with electrophoretic mobility shift assay (EMSA) using PMA treated HEL cell nuclear extract. EMSA showed specific protein binding to DNA probes encompassing each of the above sites; this was abolished by RUNX1 antibody. The ChIP and EMSA suggest that RUNX1 binds to the PF4 promoter region. To test the functional relevance of the RUNX1 binding sites wild type PF4 upstream promoter region (−1936/−27) containing both RUNX1 sites (−1768, and −151) or containing one or both sites mutated were cloned into firefly luciferase reporter gene vector pGL4 and expressed in PMA-treated HEL cells. Mutation of the −1768 site caused ∼50% reduction in luciferase activity, mutation at −151 site caused 60% reduction in activity and mutation of both sites caused 75-80% reduction in activity. These studies suggest that each RUNX1 site contributes to the transcriptional activity of PF4 promoter. Moreover, the upstream region (−1837 to +25) of the non-allelic variant PF4V1 was cloned into pGL4 plasmid; it showed negligible luciferase activity as compared to the wild PF4 promoter containing plasmid. These studies suggest that the regulation of PF4 and its variant PF4V1 is distinctly different in HEL cells. Conclusions: Our results provide the first evidence that PF4 promoter is regulated by RUNX1, and the two RUNX1 sites at −1768 and −151 are involved in its regulation. These studies provide a cogent explanation for the α-granule PF4 deficiency in our patient and others with RUNX1/CBFA2 haplodeficiency. They extend our understanding of the potential mechanisms involved in the pathogenesis of the Gray platelet syndrome.