Anti-myeloma activity of MELK inhibitor OTS167: effects on drug-resistant myeloma cells and putative myeloma stem cell replenishment of malignant plasma cells

Currently considered incurable, multiple myeloma (MM) is char-acterized by proliferation of malignant plasma cells (PC) predomi-nantly in the bone marrow, which overproduce monoclonal immunoglobulin proteins, and a perturbed tumor microenviron- ment, which promotes PC survival, inhibits osteoblast activity, increases osteoclast activity

Currently considered incurable, multiple myeloma (MM) is characterized by proliferation of malignant plasma cells (PC) predominantly in the bone marrow, which overproduce monoclonal immunoglobulin proteins, and a perturbed tumor microenvironment, which promotes PC survival, inhibits osteoblast activity, increases osteoclast activity, and leads to hallmark osteolytic bone disease, all of which contribute to the clinical manifestations of the disease. Disrupting these sequelae by incorporation of proteasome inhibitors and immunomodulatory drugs into treatment regimens has improved overall survival of MM patients; however, the majority of patients will become refractory to the most effective currently available therapeutic options and relapse. 1 It is hypothesized that a drug-resistant population of myeloma stem cells promotes disease relapse, but the characteristics and identity of this cell type(s) remain uncertain. 2 Expression of maternal embryonic leucine zipper kinase (MELK) is increased in a number of cancers and is associated with poorer prognosis. MELK activity modulates many cellular and biological processes, including proliferation, apoptosis, hematopoiesis and oncogenesis, and is believed to have a critical role in cancer stem cell maintenance. 3 We assessed the expression of MELK mRNA in malignant PC derived from MM patients and human myeloma cell lines (HMCL) and effects of the MELK inhibitor OTS167 on myeloma cells, including drug-resistant subclones. The effects of OTS167 were also tested in an in vitro cell culture model that recapitulates the bone marrow microenvironment and a malignant PC outgrowth model using peripheral blood mononuclear cells (PBMC) from patients with frank MM.
MELK gene expression analysis was performed on publically available data sets GSE5900, 4 GSE2658 (refs 5,6) and GSE6477 (ref. 7) and demonstrated significantly increased MELK mRNA expression in newly diagnosed MM PC (n = 628) compared with either normal PC (nPC; n = 37; P = 0.0189), monoclonal gammopathy of undetermined significance (MGUS) CD138+ PC (n = 65; Po0.0001), or PC from smoldering MM patients (n = 36; P = 0.0003; Figure 1a). There were no significant differences in MELK expression between nPC and either MGUS or sMM PC. Protein and mRNA expression of MELK were investigated in a panel of 26 patients from whom CD138+ MM PC were derived as well as 11 HMCL. Overall, MELK levels were variable with limited concordance between mRNA and protein (Supplementary Figures 1A and B).
Next, we tested the anti-myeloma effects of a potent, smallmolecule inhibitor of MELK kinase activity, OTS167. 8 Figure 2B). We also treated fresh bone marrow aspirates from two MM patients, one newly diagnosed and one with progressive disease, with OTS167 and analyzed apoptosis markers in either total marrow cells, CD138+ myeloma cells only, or other cells of the marrow not marked by CD138 expression (Figure 1d and Supplementary Figure 3). Although analysis of total marrow cells showed pre-existing levels of early apoptotic cells, the number of late apoptotic cells increased with MELK inhibition in a dosedependent manner. Importantly, assessment of CD138+ PC showed 100% killing in response to OTS167, whereas the remaining CD138-negative cell population was markedly less affected by MELK inhibition.
Next, we treated MM1S and U266 cells with increasing doses of OTS167 for 24 h and analyzed expression of key myeloma survival proteins ( Figure 1e). Expression of MELK and one of its known targets, FOXM1, 13 as well as FOXM1 phosphorylation increased with the lowest doses and returned to lower levels at the higher doses, potentially as a result of enhanced cell death. Serine 473-phosphorylated Akt, which marks active Akt and was only detectable in MM1S cells, decreased in a dose-dependent manner and correlated with increased cleaved total Akt, which is associated with myeloma cell death. Myeloma survival factors c-Myc and IRF4 also decreased in a dose-dependent manner in both cell lines. 12 p53 levels decreased in U266 cells but were only modestly changed in MM1S cells, and the p53 target, p21, was increased in p53 wild-type cells and fluctuated in p53 mutant cells, which agreed with previous findings. 14 Altogether, these data reveal potent apoptotic effects of OTS167 in malignant MM PC, support the potential for OTS167 to overcome drug resistance, and define novel targets of MELK whose inhibition is known to promote myeloma cell death.
The bone marrow microenvironment potently influences the growth of MM cells and contributes to chemoresistance. This occurs, in part, via bone marrow stromal cell (BMSC)-dependent production of myeloma cell survival factors as well as direct interaction with MM PC. 1 To test the potential of OTS167 to overcome growth-supportive effects conferred by BMSC, we co-cultured a subset of myeloma cell lines with the HS-5 BMSC line     Letter to the Editor co-cultures demonstrates clear induction of apoptosis-associated PARP cleavage only in the presence of myeloma cells and not with HS-5 alone (Figure 2b). We also detected robust NF-κB activation, as determined by analysis of phospho-IκBα levels, in the HS-5 cells alone and in co-culture, and to a lesser extent in MM1S cells. NF-κB activation was impaired with increasing doses of OTS167. These results coupled with our flow cytometric analysis of enumerated total bone marrow samples underscore a potent anti-myeloma effect of MELK inhibition and establish OTS167 as a potential weapon for the treatment of chemoresistant myeloma.
Despite significant efforts, there is still no consensus for the identity and location of the myeloma stem cell population. It was demonstrated over 25 years ago that malignant MM precursor cells exist in the peripheral blood and can be induced to proliferate and differentiate into clonal MM PC with interleukin-3 (IL-3) and IL-6 treatment. 15 Given the increasing evidence suggesting a role for MELK in cancer stem cells, 3 we sought to determine the effects of OTS167 on the outgrowth and differentiation of presumptive myeloma stem cells into MM PC. Therefore we treated fresh MM patient PBMC cultures with IL-3/IL-6 alone or in combination with OTS167 for 6 days and then separated the resultant cells into CD138+ and CD138 − populations using immunomagnetic positive selection (Figure 2c). IL-3/IL-6 treatment significantly increased the number of CD138+ cells and co-treatment with OTS167 impaired IL-3/IL-6dependent growth of CD138+ cells. Interestingly, MELK inhibition alone reduced the number of CD138+ cells by fivefold. In the CD138 − populations, no treatment significantly affected growth of the cells. This experiment, while indirect, demonstrates that MELK inhibition may kill a circulating MM progenitor cell population and/or prevent myeloma stem cells from re-establishing the malignant MM PC population.
In summary, this study establishes elevated MELK expression in symptomatic MM and implicates its potential in promoting myeloma cell growth and drug resistance. Importantly, MELK inhibition induced potent and rapid apoptosis of MM PC and impaired outgrowth of malignant PC derived from presumptive myeloma stem cells in the peripheral blood. These data, therefore, support clinical exploration into the potential of OTS167 as a new treatment option in MM with the ability to target active disease as well as insidious cells that may be responsible for disease relapse.

CONFLICT OF INTEREST
YN is a stock holder and a scientific advisor of OncoTherapy Science, Inc. JP is a scientific advisor of OncoTherapy Science, Inc. YM and SC are employees of OncoTherapy Science, Inc. AJJ is a consultant, member of the advisory board and receives honoraria from Bristol-Myers Squibb, Celgene, Janssen Pharmaceuticals, Karyopharm Therapeutics, Millennium/Takeda Pharmaceuticals, Onyx/Amgen Pharmaceuticals, Sanofi-Aventis and SkylineDx. The remaining authors declare no conflict of interest. were obtained by Ficoll gradient separation and 2.0-2.5 × 10 6 PBMC/well in sixwell plates were cultured as indicated with recombinant human interleukin-3 (IL-3) and IL-6 (both at 5 ng/ml), purchased from R&D Systems (Minneapolis, MN, USA) (catalog number 203-IL and 206-IL, respectively), or OTS167 (10 nM) alone or in combination for 6 days. Cultures were collected and subjected to immunomagnetic positive selection to separate CD138+ plasma cells from CD138 − cells. Trypan blue-negative cells were counted and the fold change in each population in response to each treatment is depicted. Data represent six independent determinations from six separate blood samples from MM patients. Statistical significance of differences (set at Po0.05) was determined using one-way analysis of variance with Tukey's multiple comparison test in GraphPad Prism 6 (GraphPad Software). P-values are indicated. Effects on CD138 − populations are not significant (NS).
Letter to the Editor