Multiple myeloma (MM) expands in a manner dependent on the bone marrow microenvironment, and develops devastating bone destruction. MM is still an incurable disease, and preferentially arises in the elderly. Therefore, novel well-tolerated therapeutic alternatives are wanted especially for elderly patients with MM. Valproic acid (VPA), a well-tolerated and safe anti-epileptic agent with extensive clinical experience, has recently been shown to be a class I- and IIa-specific HDAC inhibitor, and induce cytotoxic effects on various types of tumor cells. In the present study, we evaluated the impact of VPA on MM cell growth and survival as well as MM-induced bone marrow microenvironment. VPA reduced viable cell numbers to less than 50 % from the baseline at day 2 in all MM cell lines (5/5) as well as primary CD138-positive MM cells (4/4) tested, and a portion of B cell (2/5) and T cell (1/3) lines, but not in AML cell lines (0/4) at 100 microg/ml, a therapeutic concentration for epilepcy, which raises a possibility for VPA as a therapeutic agent against MM. Interestingly, CD138-negative non-MM bone marrow cells remained intact and CFU-GM numbers were not affected by VPA, suggesting tumor-specific actions of VPA. VPA induced death receptor- but not mitochondrial pathway-mediated apoptosis with down-regulation of cellular FLICE-inhibitory protein (c-FLIP) and cleavage of caspase 8 in RPMI8226 MM cells. Furthermore, VPA down-regulated cyclin D1, and up-regulated the cyclin-dependent kinase inhibitor p21(Cip1) with accumulation of MM cells at G0/G1 phase, suggesting the involvement of cell cycle arrest in anti-proliferation actions of VPA. Notably, VPA at therapeutically relevant concentrations potentiated the induction of apoptosis by dexamethasone which triggers the release of Smac from mitochondria. However, VPA did not enhance the cytotoxic effects of cell cycle-specific agents including doxorubicine and melphalan. The VPA-induced tumor cell dormancy may reduce the susceptibility of MM cells to such cell cycle-specific agents. In parallel with MM progression, angiogenesis as well as osteoclastogenesis are enhanced in the bone marrow. We previously demonstrated that MM cell-osteoclast (OC) interactions enhance the growth and survival of MM cells as well as angiogenesis. Therefore, we next investigated the effects of VPA on MM cell-OC interactions and angiogenesis. Although VPA showed no significant effects on osteoclastogenesis induced by MM cells, VPA suppressed the growth and survival of RPMI8226 and U266 MM cells in the presence of OCs generated from monocytes to the levels similar to those without OCs. Furthermore, VPA potently inhibited in vitro vascular tubule formation enhanced by conditioned media from co-cultures of MM cells and OCs. Such anti-angiogenic effects of VPA was further potentiated in concert with thalidomide. Collectively, the present study suggests that VPA exerts multi-factorial anti-MM actions and may serve as a novel well-tolerated therapeutic alternative against MM.