A novel TLR-9 agonist C792 inhibits plasmacytoid dendritic cell-induced myeloma cell growth and enhance cytotoxicity of bortezomib

Abstract

Our prior study in multiple myeloma (MM) patients showed increased numbers of plasmacytoid dendritic cells (pDCs) in the bone marrow (BM), which both contribute to immune dysfunction as well as promote tumor cell growth, survival and drug resistance. Here we show that a novel Toll-like receptor (TLR-9) agonist C792 restores the ability of MM patient-pDCs to stimulate T-cell proliferation. Coculture of pDCs with MM cells induces MM cell growth; and importantly, C792 inhibits pDC-induced MM cell growth and triggers apoptosis. In contrast, treatment of either MM cells or pDCs alone with C792 does not affect the viability of either cell type. In agreement with our in vitro data, C792 inhibits pDC-induced MM cell growth in vivo in a murine xenograft model of human MM. Mechanistic studies show that C792 triggers maturation of pDCs, enhances interferon-α and interferon-λ secretion and activates TLR-9/MyD88 signaling axis. Finally, C792 enhances the anti-MM activity of bortezomib, lenalidomide, SAHA or melphalan. Collectively, our preclinical studies provide the basis for clinical trials of C792, either alone or in combination, to both improve immune function and overcome drug resistance in MM.

Figure 1. C792 restores the ability of MM patient pDCs to trigger both allogeneic and autologus T cell proliferation

(A) Normal CD4 T cells (1 × 10⁵) were cultured with normal healthy donor (npDC) or MM patient BM pDCs at 1:10 (pDC/T cell) in the presence or absence of C792 for 5 days, and then analyzed for T cell proliferation with WST assay (BM-pDCs from 5 MM patients were utilized; mean ± SD; p < 0.005 for all patients). (B) MM patient BM pDCs (1 × 10⁴) were cultured with autologous T cells (1× 10⁵) at 1:10 (pDC/T cell) ratio in the presence or absence of C792 (0.1 μg/ml) for 5 days, and then analyzed for T cell proliferation using WST assay (mean ± SD; p < 0.01 for all patients).

Figure 2. C792 inhibits pDC-induced MM cell growth

(A) MM.1S cells (5 × 10⁴/200 μl) and MM BM-pDCs (1 × 10⁴/200 μl) were cultured either alone or together at 1:5 (pDC/MM) ratio in the presence or absence of C792 for 72h, and then analyzed for growth using WST proliferation assays. Non-CpG-ODN 1040 served as a negative control (mean ± SD; p< 0.05; n=6). (B) Freshly isolated pDCs from six normal healthy donors were cultured in DCP-MM medium, in the presence of C792 (5–10 ug/ml) for 72h, and analyzed for viability by MTT assays (mean ± SD, p< 0.05). (C) MM.1S and IL-6-dependent ANBL-6 cells were cultured with indicated concentrations of C792 for 72h, and analyzed for viability by MTT assays (mean ± SD; p< 0.05, n=3). Non-CpG-ODN 1040 was also utilized as a control. (D) Patient MM cells were cultured with or without pDCs at 1:5 (pDC/MM) ratio for 72h, in the presence or absence of C792, and cell growth was analyzed using WST assay (mean ± SD of triplicate cultures; p< 0.05 for all samples). (E) Patient MM cells were cultured with or without autologous BM pDCs at 1:5 (pDC/MM.1S) ratio in the presence of indicated concentrations of C792 for 72h, and cell growth was analyzed using WST assay (mean ± SD; p <0.04, n=5). (F) Patient BM pDCs were cultured with or without MM.1S cells at 1:5 (pDC/MM) ratio in the presence or absence of C792 for 72h, and cell growth was analyzed by WST assay (mean ± SD; p < 0.01, n = 5). Data is fold change in MM cell growth normalized to growth in the absence of pDCs (mean ± SD; p< 0.05, n=3).

Figure 3. C792 inhibits pDC-induced MM cell growth in vivo in a murine xenograft model of human MM

(A) Average and standard deviation of tumor volume (mm³) in mice (n = 4/group) versus time (days) when tumor was measured. CB-17 SCID-mice were randomized into three groups (four mice each group): the first group was subcutaneously injected with MM.1S cells alone (1.25 × 10⁶ cells in 100 μl of serum free RPMI-1640 medium); the second group received resting pDC (0.25 × 10⁶ cells) plus MM.1S cells (1.25 × 10⁶ cells); and the third group was injected with ex-vivo C792-activated pDCs plus MM.1S cells. Tumor growth was measured by calculating tumor volume using the formula: $\text{Volume}={(\text{width})}^2\times \text{length}/2$. Error bars indicate standard deviation (SD). (B) Kaplan-Meier survival plot shows significantly increased survival of mice receiving C792-treated pDCs plus MM.1S cells versus mice injected with untreated pDCs plus MM.1S cells (P = 0.0002, Log-rank (Mantel-Cox) Test): median survival was 24 days in MM.1S-injected mice; 18 days in mice injected with MM.1S plus untreated-pDCs; and 49 days in mice receiving MM.1S plus C792-treated pDCs (CI 95%). Tumor-bearing mice were sacrificed with a tumor volume > 2 cm².

Figure 4. C792 triggers maturation of pDCs, IFN release, and activation of TLR9-MyD88 signaling axis

(A) MM BM-pDCs (CD123⁺/BDCA-2⁺/HLA-DR⁺/CD11c⁻) were cultured in the presence or absence of C792 (1.0 μg/ml) for 12h; cells were stained with fluorophore-conjugated antibodies against CD40, CD80, CD83, CD86, or HLA-DR, followed by flow cytometry analysis. Non-CpG 1040 served as a negative control. Bar graph shows the percentage change in MFI for indicated molecules in untreated- versus C792-treated pDCs (mean ± SD, p< 0.05, n = 3). (B, left panel) pDCs were treated with indicated concentrations of C792 for 12h, followed by intracellular staining using FITC-conjugated TLR-9 antibody. Isotype-matched antibody served as control. (B, right panel) C792-triggered changes in TLR9, as shown in left panel, were quantified: Intracellular TLR-9 expression is presented as percentage change in relative fluorescence intensity in C792-treated pDCs versus the untreated-pDCs (mean ± SD; p < 0.05, n = 3). (C) MM BM-pDCs (5 × 10³ cells) were cultured in the presence or absence of indicated concentrations of C792 (Type C CpG ODN), Type-B CpG ODN, or non-CpG for 24h; supernatants from these cutures were analyzed for IFN-α using ELISA (mean ± SD, p< 0.05, n=3). (D) pDCs (1 ×10⁴), MM.1S cells (5 × 10⁴), or pDCs plus MM.1S cells were cultured in the presence or absence of indicated concentrations of C792 for 24h, and supernatant from these cultures were analyzed for IFN-α using ELISA (mean ± SD; p< 0.05, n=4). (E) MM BM-pDCs (5 × 10³) were pre-treated with 50 μM of MyD-88 inhibitor peptide for 4h, followed by addition of C792 for 24h; supernatants from these cultures were analyzed for IFN-α using ELISA (mean ± SD; p< 0.05, n=3).

Figure 5. Mechanism(s) mediating C792 activity

(A) MM.1S cells were cultured with MM BM-pDCs in the presence or absence of recombinant human-IFN-α for 48h, and cell growth was analyzed using WST assay. Data is derived after normalizing cell growth in IFN-α-treated versus untreated cultures (mean ± SD; p< 0.05, n=3). (B) pDCs from 2 MM patients were treated with C792 for 96h, and supernatants were analyzed for soluble TRAIL using ELISA (mean ± SD; p< 0.05, n=3). (C) MM cell lines were treated for 72h with supernatants derived from untreated- or C792-treated MM BM-pDC cultures, and cell viability was analyzed using MTT assay. Data is derived after normalizing MM cell viability in supernatants from untreated- versus C792-treated pDC cultures (mean ± SD; p< 0.05, n=3). (D) MM.1S and MM BM-pDCs cells were co-cultured at 1:5 (pDC/MM) ratio in the presence or absence of C792 for 12h. Cells were then stained with propidium iodide. MM (CD138-positive) cell population was selectively gated by flow cytometry for cell cycle analysis. Bar graph shows the percentage of MM cells in cell cycle phases. A significant accumulation of MM cells in G2M phase was noted in the C792-treated versus untreated co-cultures (p < 0.005, n = 3). (E) MM.1S and MM BM-pDCs cells were co-cultured at 1:5 (pDC/MM) ratio in the presence or absence of C792 for 12h; the MM (CD138-positive) population was selectively gated by flow cytometry and analyzed for apoptosis using Annexin V/PI staining assays. A significant increase in apoptotic MM cell population was noted in the C792-treated versus untreated co-cultures (25–30% increase in Annexin V⁺/PI⁻ cells) (p < 0.005, n = 3). Bar graph shows the percentage of apoptotic MM cells.

Figure 6. Combination of C792 and bortezomib, SAHA, melphalan, or lenalidomide triggers synergistic anti-MM activity

(A) Co-cultures of MM.1S plus BM-pDCs were treated with indicated concentrations of C792, bortezomib, or C792 plus bortezomib for 48h, and then analyzed for viability. Isobologram analysis shows the synergistic cytotoxic effect of C792 and bortezomib. The graph (left) is derived from the values given in the table (right). The numbers 1–6 in graph represent combinations shown in the table. Combination index (CI) <1 indicates synergy. (B) Co-cultures of MM.1S plus BM-pDCs were treated with indicated concentrations of C792, SAHA, or C792 plus SAHA for 48h, and then analyzed for viability. Isobologram analysis shows the synergistic cytotoxic effect of C792 and bortezomib. The graph (left) is derived from the values given in the table (right). The numbers 1–6 in graph represent combinations shown in the table. Combination index (CI) <1 indicates synergy. (C) Co-cultures of MM.1S plus BM-pDCs were treated with indicated concentrations of C792, melphalan, or C792 plus melphalan for 48h, and then analyzed for viability. Isobologram analysis shows the synergistic cytotoxic effect of C792 and melphalan. The graph (left) is derived from the values given in the table (right). The numbers 1–6 in graph represent combinations shown in the table. Combination index (CI) <1 indicates synergy. (D) Co-cultures of MM.1S plus BM-pDCs were treated with indicated concentrations of C792, lenalidomide, or C792 plus lenalidomide for 48h, and then analyzed for viability. Isobologram analysis shows the synergistic cytotoxic effect of C792 and lenalidomide. The graph (left) is derived from the values given in the table (right). The numbers 1–6 in graph represent combinations shown in the table. Combination index (CI) <1 indicates synergy.