Vitamin K2 Supports Hematopoiesis through Acting on Bone Marrow Mesenchymal Stromal/Stem Cells

Aya Fujishiro, Yasuo Miura, Masaki Iwasa, Sumie Fujii, Akihiro Tamura, Atsushi Sato, Asumi Yokota, Noriko Sugino, Hideyo Hirai, Akira Ando, Tatsuo Ichinohe and Taira Maekawa


[Background] Myelodysplastic syndrome is an intractable disorder characterized by ineffective hematopoiesis. Although allogeneic hematopoietic stem cell transplantation is the only curative therapy for eligible patients, hematopoiesis-supportive pharmacotherapy is practically important for transplant-ineligible patients to overcome transfusion dependency and infections. Vitamin K2 (VK2, menatetrenone) is a drug used to aim at improvement of hematopoiesis in MDS patients (Leukemia 14: 1156, 2000). However, the exact mechanism how VK2 improves hematopoiesis remains largely unknown. It was reported that VK2 induces MDS cells to undergo apoptosis (Leukemia 13: 1399, 1999). Here, we investigated our hypothesis that VK2 exerts its hematopoiesis-supportive effects through acting on mesenchymal stem/stromal cells (BM-MSCs) in the bone marrow microenvironment.

[Methods] Normal bone marrow (BM) samples from healthy adult volunteers were purchased from AllCells (Emeryville, CA). BM-CD34+ cells were isolated from BM-mononuclear cells using anti-CD34 immunomagnetic microbeads (Miltenyi Biotec, Bergisch Gladbach, Germany). Human BM-MSCs were isolated according to our previously published methods (Stem Cells 32:2245, 2014). In co-culture experiments, BM-MSCs with or without VK2 treatment were seeded on a 24-well culture plate. BM-CD34+ cells were applied on the MSC-grown plate and co-cultured in SFEM (StemCell Technologies, Vancouver, Canada) supplemented with 100 ng/mL SCF, 100 ng/mL Flt-3 ligand, 50 ng/mL TPO and 20 ng/mL IL-3. After 10 days of co-culture, the number and surface marker expression of the expanded hematopoietic cells were examined by flow cytometric analysis.

[Results] We first tested the direct effect of VK2 on BM-CD34+ cells. BM-CD34+ cells were treated with VK2 at various concentrations ranged from 0 µM to 10 µM for 24 hours and then cultured in SFEM in combinations with cytokines. Surprisingly, viable hematopoietic cells were hardly detected in the expansion culture of BM-CD34+ cells treated with 10 µM VK2. Even with 1 µM treatment, the number of CD45+ cells was decreased, as compared to that of expansion culture of untreated BM-CD34+ cells. The apoptosis analysis showed that the percentage of AnnexinV+ PI+ cells in the expanded hematopoietic cells is increased by VK2 treatment. We next examined the effect of VK2 on the hematopoiesis-supportive capability of BM-MSCs. BM-MSCs were pretreated with VK2 at various concentrations and then co-cultured with BM-CD34+ cells. The numbers of CD34+ cells and CD45+ cells were increased in a VK2 dose-dependent manner. These results demonstrated that VK2 shows different effects on distinct stem/progenitor cells: the induction of apoptosis in BM-CD34+ cells and the enhancement of hematopoiesis-supportive capability of BM-MSCs. We then investigated whether apoptosis-related cell death of BM-CD34+ cells by VK2 treatment is ameliorated in the presence of BM-MSCs. Both BM-CD34+ cells and BM-MSCs were treated with VK2 for 24 hours, and then co-cultured. The number of CD34+ cells was not decreased significantly in contrast to its severe decrease in single culture of VK2-treated BM-CD34+ cells.

We further analyzed the effect of VK2 on BM-MSCs. Subpopulation analysis in co-culture of CD34+ cells with VK2-treated BM-MSCs showed that the expansion efficacy of CD34+CD38+ cells is higher in comparison to that of CD34+CD38- cells. In addition, the percentages of CD34-CD33+ cells and CD34-CD13+ cells were higher than those in co-cultures with untreated BM-MSCs. Therefore, VK2-treated BM-MSCs supported the expanded CD34+ cells to skew their phenotype toward myeloid lineage. The presence of a transwell in the co-culture system was unrelated to the expansion pattern of CD34+ cells, which suggested the involvement of soluble factors with respect to the underlining mechanism. We therefore compared the levels of hematopoiesis-supporting cytokine mRNA expression in VK2-treated and untreated BM-MSCs: VK2-treated BM-MSCs showed lower expression of CXCL12/SDF-1 mRNA and a trend toward higher expression of GM-CSF mRNA.

[Summary] VK2 acted on BM-MSCs to support their ability to enhance expansion and myeloid differentiation of BM-CD34+ cells probably via altered GM-CSF and CXCL12/SDF-1 expression in MSCs. These findings may help to identify the mechanisms of therapeutic effects of VK2 in patients with MDS (Figure).

Disclosures No relevant conflicts of interest to declare.

  • * Asterisk with author names denotes non-ASH members.