Abstract
RUNX1 (also known as CBFA2 or AML1) is a transcription factor that plays a major role in hematopoiesis. Haplodeficiency of RUNX1 has been associated with familial thrombocytopenia, impaired megakaryopoiesis, impaired platelet function and predisposition to acute myeloid leukemia. We have reported a patient with inherited thrombocytopenia and abnormal platelet function (Gabbeta et al, Blood87:1368–76, 1996). The patient platelets showed impaired phosphorylation of pleckstrin and myosin light chain, diminished GPIIb-IIIa activation and decreased platelet protein kinase C-𝛉. This was associated with a heterozygous nonsense mutation in transcription factor RUNX1 (Sun et al, Blood103: 948–54, 2004). Platelet transcript profiling showed a striking downregulation of myosin light chain 9 (MYL9) by ~77-fold relative to normal platelets (Sun et al, J. Thromb Haemost.5: 146–54, 2007). Myosin light chains (MLCs) play an important role in platelet responses to activation, in platelet biogenesis, and are involved in cellular processes such as cytokinesis, cell adhesion, cell contraction, cell migration. We have addressed the hypothesis that MYL9 is a direct transcriptional target of RUNX1. Studies were performed in human erythroleukemia (HEL) cells treated with phorbol 12-myristate 13-acetate (PMA) for 24 h to induce megakaryocytic transformation. To determine endogenous interaction of RUNX1 with MYL9 promoter, we performed chromatin immunoprecipitation (ChIP) assay using anti-RUNX1 antibody. These studies revealed RUNX1 binding to MYL9 chromatin at −742/−529 bp upstream of the ATG codon. TFSEARCH revealed four RUNX1 sites within this region. We performed electrophoretic mobility shift assay (EMSA) using probes containing each of the RUNX1 motifs and PMA-treated nuclear extracts from HEL cells. With each probe, protein binding was observed that was competed by excess unlabelled probe and inhibited by anti-RUNX1 antibody indicating RUNX1 as the protein involved. This protein binding was not competed by oligos containing mutations in the specific RUNX1 sites. No binding was noted directly to the mutant probes. To further corroborate our findings, we performed transient-ChIP analysis where wild type luciferase reporter construct −691/+4 and constructs with each of the RUNX1 sites individually mutated were transiently transfected into HEL cells. ChIP was performed using these cells and anti-RUNX1 antibody, and the expression analyzed by PCR amplification with a forward primer from MYL9 promoter sequence and reverse primer from luciferase vector sequence. Amplification was observed with immunoprecipitated wild type construct but not with any of the mutant constructs. Thus, RUNX1 interacts in vivo with MYL9 promoter, and the multiple RUNX1 sites interact with each other, as also shown for other genes. To test the functional relevance, the wild type construct −691/+4 containing all 4 RUNX1 sites or mutant constructs with each site individually deleted were cloned into firefly luciferase reporter gene vector and transfected into HEL cells. Deletion of RUNX1 site 1, 2, 3 or 4 caused ~60–90% reduction in the activity indicating that each site was functional. Lastly, siRNA mediated knock down of RUNX1 in HEL cells was associated with a decrease in both RUNX1 and MYL9 protein.
Conclusions: Our results provide the first evidence that MYL9 gene is transcriptionally regulated by RUNX1. They provide evidence for the presence of multiple RUNX1 sites in MYL9 promoter, as also observed in other genes. Moreover, these studies provide a cogent mechanism for the MYL9 transcript downregulation and the impaired MLC-phosphorylation we have previously described in association with RUNX1 haplodeficiency.
Prolactin inhibits cell loss and decreases matrix metalloproteinase expression in the involuting mouse mammary gland but fails to prevent cell loss in the mammary glands of mice expressing IGFBP-5 as a mammary transgene
