SEMA3A partially reverses VEGF effects through binding to neuropilin-1

- VEGFA is increased in CD34+ cells from AML patients.
- VEGFA overexpression gives AML cell advantages.
- SEMA3A is increased in BMSCs from MDS and AML patients.
- SEMA3A seems to neutralize VEGFA effects.

Cross-talk between hematopoietic stem cells (HSCs) and bone marrow stromal cells (BMSCs) is essential for HSCs regulation and leukemogenesis. Studying bone marrow of myelodysplasia patients, a pre-leukemic condition, we found mRNA overexpression of vascular endothelial growth factor A (VEGFA) in CD34+ HSCs and semaphorin 3A (SEMA3A) in BMSCs. To better understand the role of VEGFA and SEMA3A in leukemogenesis, we recruited 30 myelodysplastic syndrome (MDS) patients, 29 acute myeloid leukemia (6 secondary to MDS) patients and 12 controls. We found higher VEGFA expression in de novo AML patients (without prior MDS) group (p = 0.0073) and higher SEMA3A expression in all BMSCs patient's samples compared to control group. We then overexpressed VEGFA in an acute myelogenous leukemia cell line, KG1 cells, and in normal CD34+ cells. This overexpression increased KG1 (p = 0.045) and CD34+ cell (p = 0.042) viability and KG1 (p = 0.042) and CD34+ cell (p = 0.047) proliferation. Moreover, KG1 and CD34+ cells overexpressing VEGFA also had increased proliferation when co-cultured with human marrow stromal HS5 cells (p = 0.045 and p = 0.02, respectively). However, co-culture of these transformed cells with HS5 cells overexpressing SEMA3A reduced KG1 (p = 0.004) and CD34+ (p = 0.009) proliferation. Co-culture of KG1 transformed cells with HS27 cells overexpressing SEMA3A reduced KG1 proliferation as well (p = 0.01). To investigate whether the dominant SEMA3A effect over VEGFA could be due to competition for neuropilin1 receptor (NRP1), we performed immunoprecipitation with anti-NRP1 antibody of cell extracts of co-cultured KG1 and HS5 cells, induced or not by VEGFA and SEMA3A recombinant proteins. Results showed a preferential association of NRP1 with SEMA3A, suggesting that SEMA3A can partially reverse the effects caused by the VEGFA preventing its binding with the NRP1 receptor. Since both hematopoietic cells, leukemic and normal, showed similar behavior, we suppose that the attempt to reversion of VEGF effects by SEMA3A is a homeostatic phenomenon in the hematopoietic niche. Finally, we conclude that VEGFA overexpression confers AML cell advantages and SEMA3A may partially reverse this effect; thus, SEMA3A protein combined with VEGFA inhibitors could be beneficial for AML treatment.