Tumor-associated macrophage-mediated survival of myeloma cells through STAT3 activation

Nathan De Beule1, Kim De Veirman1, Ken Maes1, Elke De Bruyne1, Eline Menu1, Karine Breckpot2, Hendrik De Raeve3, Rian Van Rampelbergh3, Jo A. Van Ginderachter4,5, Rik Schots6, Els Van Valckenborgh1*, Karin Vanderkerken1*

Abstract

Overcoming drug resistance is one of the greatest challenges in the treatment of multiple myeloma (MM). The interaction of myeloma cells with the bone marrow (BM) micro-environment is a major contributing factor to drug resistance. Tumor-associated macrophages or TAMs with different polarization states are an important component of this micro-environment. Previous studies revealed a role of TAMs in MM survival and drug resistance; however, the impact of macrophage polarization (anti-tumoral “M1” versus pro-tumoral “M2”-like phenotype) in this process is not yet described. Here the presence of TAMs was confirmed in MM patients on BM sections both at diagnosis and relapse using two M2 markers, CD163 and CD206. By following different TAM subpopulations during disease progression in the syngeneic murine 5T33MM model, we demonstrated a decrease in inflammatory monocytes and increase of M2-oriented TAMs in BM. Co-culture experiments demonstrated that macrophages induce a survival benefit to myeloma cells that is maintained after treatment with several classes of anti-myeloma agents (melphalan and bortezomib) and the biggest effect was observed with M2 polarized macrophages. The pro-survival effect was associated with an activation of the STAT3 pathway in 5T33MM cells, less cleavage of caspase-3 and thus less apoptosis. AZD1480, an ATP-competitive JAK2 inhibitor, abrogated the observed TAM mediated MM cell survival and partially inhibited resistance to bortezomib. Despite only a small quantitative impact on myeloid cells in vivo, AZD1480 treatment alone and in combination with bortezomib could significantly reduce tumor load. In conclusion, M2 TAMs are present in the MM microenvironment and contribute to MM cell survival and protection from drug-induced apoptosis. As a result of TAM-induced activation of the STAT3 pathway, 5T33MM cells are sensitized to AZD1480 treatment.

Keywords: multiple myeloma, tumor-associated macrophages, drug resistance, STAT3
 
Introduction

Multiple myeloma (MM) is characterized by the accumulation of malignant plasma cells in the bone marrow (BM), associated with end-organ damage such as kidney failure, lytic bone lesions, anemia and immunodeficiency [1]. MM represents 1.8% of all new cancer cases and 2% of all cancer deaths in the USA, with a median age at diagnosis of 69 years. Despite the introduction of several new classes of drugs including proteasome inhibitors, immunomodulatory agents, histone deacetylase inhibitors and monoclonal antibodies, the 5-year survival is only around 50% for patients diagnosed between 2006 and 2012 [2]. One of the major challenges in the treatment of MM patients is relapse due to drug resistance induced by intrinsic and extrinsic mechanisms. The interaction of MM cells with the BM micro-environment containing BM stromal cells and extracellular matrix proteins contributes to their proliferation, survival, angiogenesis and drug resistance [3–7]. Studies have shown that the BM of MM patients is infiltrated by an excess of macrophages compared to healthy controls and that the degree of infiltration can serve as a prognostic histopathological marker in newly diagnosed patients [8,9]. Polarization into either M1 (classically activated) or M2 (alternatively activated) macrophages is influenced by the different signals they encounter. Chemokines like CCL2, CCL3, CCL14 and CXCL12 stimulate homing of macrophages to the BM and also promote proliferation and M2 polarization [10,11]. M1 macrophages are considered as anti-tumoral and proinflammatory, in contrast to M2 macrophages, that have an anti-inflammatory, proangiogenic and wound healing capacity. Moreover, comparison between M1 and M2 revealed increased expression of Arg1, Mrc1, Fizz1 and Stab1 by M2 macrophages and more expression of Nos2 and Tnf in M1 macrophages [12,13].
These two polarization states of macrophages represent the ends of a large spectrum and oversimplify the great variety of functional activation states present in real life [14,15]. In mice, heterogeneous monocyte/macrophage activation states were identified using the markers Ly6C and MHCII including MHCIIhi-M1 like TAM and MHCIIlo-M2 like TAM [16]. M1 markers described in humans include surface markers such as MHCII, CCR7, CD80, CD86 and pro-inflammatory cytokines like IL-12 and TNFα. M2 macrophages express CD163, CD206, CD204 and IL-10 [17–19].
Recent studies revealed that the M2 markers CD163 (hemoglobin scavenger receptor) and CD206 (mannose receptor) can be used as soluble biomarkers in MM [20,21].
Moreover, myeloma cells overexpress CD47, which protects them from phagocytosis [22]. Few studies have investigated the role of TAMs in MM development. It has been suggested that TAMs stimulate angiogenesis, since the accumulation of CD206+Tie2+ macrophages correlated with an increased production of proangiogenic cytokines and increased microvessel density (MVD) in the Vk*MYC mouse model [23,24]. Other studies showed that myeloma-macrophage interactions induced MM drug resistance and this was mediated by PSGL-1 (P-selectin glycoprotein ligand-1)/selectins and ICAM-1/CD18 interactions, leading to the activation of Src, Erk1/2 kinases and c-Myc pathways [6,8]. However, the impact of macrophage polarization on drug resistance in MM is still unknown. The aim of our study was to confirm the presence of TAMs in MM patients and their biodistribution during disease progression in the 5T33MM model. In addition, we investigated the underlying intracellular and molecular mechanisms of TAM-mediated survival and drug resistance.

Material and methods

Patient selection and BM trephine biopsies

Specimens were collected at the Department of Hematology and Pathology (University Hospital, Brussels, Belgium) and OLV Hospital (Aalst, Belgium). All samples available from MM patients (n=23) in the period 2005-2015, with both diagnostic and relapsed tumor bone biopsy samples, were systematically selected. From the same period, non-MM patients with normal BM samples were randomly systematically selected as controls. The study protocol was approved by the local ethics committee (B.U.N. 14320152611).

Mice and cell lines

C57BL/KaLwRij mice were purchased from Envigo (Horst, the Netherlands). Housing and maintenance was done according to the regulations approved by the Ethical Committee for Animal Experiments, Vrije Universiteit Brussel (CEP n° 14-281-2). The 5T33MM mouse model, used in these studies, was previously described [25]. 5T33MMvv cells were isolated from the BM of diseased mice by crushing hind legs and vertebrae. After red blood cell lysis, purity of >90% was confirmed on May-Grünwald Giemsa stained cytospins. The in vitro stroma-independent growing 5T33MMvt cell line was cultured in RPMI1640 medium (Lonza, Basel, Switzerland) supplemented with 100U/ml penicillin/streptomycin, 1mM sodium pyruvate, 2mM glutamine (Gibco, Eggenstein, Germany) and 10% fetal calf serum (FCS, Hyclone, UT, USA) at 37°C in 5% CO2.

Analysis of TAMs with flow cytometry

BM cells were isolated from mice by crushing hind legs and spleen and removing red blood cells. Antibodies used were CD11b-FITC (101206), LY6G-BV421 (562737), SiglecF-BV421 (562681), Ly6C-APC (560595), MHCII-PECy7 (107629), CD206-PE (141706) and their isotype controls. All antibodies were obtained from Biolegend (San Diego, CA, USA). Before staining, FcR blocking was performed (Miltenyi Biotech, Leiden, The Netherlands). Data were generated with a FACS Canto or Fortessa (Becton Dickinson (BD), Franklin Lakes, NJ, USA)) and analyzed with FACS Diva Software (BD).

Drugs

Bortezomib and quisinostat were purchased from Selleckchem (Munich, Germany). Melphalan and AZD1480 were obtained from Sigma Aldrich (Bornem, Belgium) and Active Biochemicals Company (Hong Kong, China) respectively. Bortezomib, quisinostat and AZD1480 (for in vitro use) were dissolved in dimethylsulphoxide. Melphalan was dissolved in acidified ethanol. AZD1480 for in vivo use was dissolved in sterile water supplemented with 0,5% hydroxyl-propyl-methylcellulose (Sigma) and 0.1% Tween-80 (Sigma).

Macrophage generation and polarization

BM cells (400 000 cells/ml/24-well) were cultured in DMEM medium (Lonza) supplemented with M-CSF (0,05µg/ml), penicillin/streptomycin, non-essential amino acids (NEAA), sodium pyruvate, glutamine (Gibco) and 10% fetal bovine serum (Hyclone) at 37°C in 5% CO2 and refreshed after 3 days. After 6 days of culture with M-CSF, macrophages were polarized towards M1 (0.05 µg/ml IFNy), M2 (0.04 µg/ml IL-4) or M5T33 (conditioned medium (CM) of 5T33MMvt cell line). CM was obtained by culturing 5T33MMvt cells for 48 hours in RPMI1640 medium (with 10% FCS and supplements) at a density of 106 cells per l. For experiments, CM was diluted 1:1 with fresh RPMI1640 medium. M-CSF was purchased from Immunotools (Friesoythe, Germany). IFNy and IL-4 were purchased from eBioscience (San Diego, USA). All cytokines were dissolved in PBS (Lonza).

In vivo treatment with AZD1480

C57Bl/KaLwRij mice were intravenously injected with 5T33MMvv cells (0.5×106 per mouse) and divided into 2 groups (n=5 per group). Two days after inoculation, mice were treated with vehicle or AZD1480 (30 mg/kg, twice a day, gavage) for 7 days. All mice were sacrificed at day 9 after inoculation. A complete blood count was done with an automated cell counter Vet abc plus (Scil, Altorf, France). TAM subtypes were evaluated by flow cytometry.

In vivo treatment with AZD1480 and bortezomib

C57Bl/KaLwRij mice were intravenously injected with 5T33MMvv cells (0.5×106 per mouse) and divided into 4 groups (n=7 per group). Two days after inoculation, mice were treated with vehicle or AZD1480 (30 mg/kg, twice a day, gavage) for the first week, bortezomib (0.6 mg/kg, twice a week, subcutaneous) for 3 weeks, or the combination of AZD1480 and bortezomib. All mice were sacrificed when first mice showed signs of morbidity. Plasmacytosis was evaluated on BM cytospins stained with May-Grünwald-Giemsa. TAM subtypes were evaluated by flow cytometry.

Statistics

Statistical analysis was done using GraphPad Prism 6.0 software. All data represent the mean ± standard deviation (SD). The Mann-Whitney U test was used to analyze the data. p<0.05 (*), p<0.01 (**), p<0.001 (***), p<0.0001 (****) were considered statistically significant.
Details of immunohistochemistry of BM trephine biopsies and quantification, quantitative real-time PCR, western blot, viability, apoptosis and cell proliferation assays are provided in the Supplementary material and methods.
 
Results

Presence of TAMs in MM patients

Macrophage numbers were compared between healthy controls, MM patients at diagnosis and MM patients at relapse by immunohistochemistry staining. Patient characteristics are shown in Table 1. CD68 is a general macrophage marker, while CD163 and CD206 represent M2 markers. The CD206+ cell count was higher in MM patients at diagnosis and relapse compared to healthy controls (p<0.05), while no differences in CD68 and CD163 could be observed (Figure 1A). The median CD206+/CD68+ ratio was also higher in diagnostic and relapsed MM patients than the control group (Figure 1B). There were no significant differences observed in TAM infiltration between MM samples at relapse compared to diagnosis. Interestingly, there was a positive correlation between a high CD68 count at diagnosis and a high CD68 count at relapse (r=0.447, p<0.05). Representative photographs of immunohistochemistry stainings on BM sections of control subjects and MM patients are shown in Figure 1C. Overall, these results indicate a higher presence of a CD206+ macrophage subset in MM patients at diagnosis and relapse.

Presence of TAMs in the 5T33MM mouse model

We next evaluated the presence of different TAMs populations in the 5T33MM mouse model during disease progression. Mice were sacrificed at day 9 (early stage of disease) or day 21 (end stage of disease) after inoculation of 5T33MMvv cells and compared with naive C57BL/KaLwRij mice. The different BM TAMs were analyzed by flow cytometry based on the strategy described by Laoui et al [16]. A general monocytic/macrophage population was selected after positive gating for cells of the myeloid lineage (CD11b+ cells) and gating out neutrophils (Ly6G+ cells) and eosinophils (SiglecF+ cells). Further discrimination of TAM subpopulations was done based on Ly6C and MHCII expression (Figure 2A). Ly6Clo cells were also F4/80hi which confirms their macrophage phenotype (data not shown). A decrease in inflammatory monocytes and increase of both immature M2 TAMs and M2 TAMs was observed during tumor progression (Figure 2B). This shift in monocytic/macrophage subpopulations towards cells with an M2 phenotype is in accordance with the observations made on patient samples.

Macrophages induce MM cell survival

Although it has been described that macrophages can induce MM cell survival and drug resistance [6–8], it has never been investigated whether the polarization state of macrophages is important for this effect. Therefore, an in vitro macrophage culture system was optimized by efficiently differentiating mouse BM cells into macrophages using M-CSF and polarizing them into M1 with IFNγ, M2 with IL-4 and TAM-like macrophages with CM of the 5T33MMvt cell line (M5T33). The polarization state of the macrophages was checked with qRT-PCR: M1 genes Tnfa and Nos2 versus M2 genes Stab1, Fizz1, Arg1 and Mrc1 (coding for CD206 protein). Similarly to M2 macrophages, M5T33 macrophages had a lower Tnfa and Nos2 expression and a higher Stab1, Fizz1 and Mrc1 expression compared to M1 macrophages (Figure 3AE). Arg1 expression in M5T33 macrophages was similar in M1 macrophages (Figure 3F).
We evaluated the effect of anti-myeloma agents on the viability of the differently polarized macrophages. Sensitivity to bortezomib (proteasome inhibitor), melphalan (alkylating agent) and quisinostat (second generation histone deacetylase inhibitor) was measured after 24 hours (Figure 4A). Bortezomib did not significantly reduce viability. Melphalan reduced viability of M2 and M5T33 macrophages at the highest concentration. Interestingly, all three polarization states of macrophages were sensitive to quisinostat treatment (5 nM and 10 nM).
Next, we investigated the effect of the polarization state on survival and drug resistance in 5T33MMvt and 5T33MMvv cells. As determined by AnnexinV/7AAD staining, survival of 5T33MM cells was increased after co-culture with differently polarized macrophages (Figure 4B-C). There is a trend that the largest effects were observed with M2-oriented macrophages (M2 and M5T33). This survival benefit for myeloma cells was maintained when bortezomib and melphalan were added and was confirmed by reduced cleavage of caspase-3 (Figure 4B-E). To exclude that the increased amount of living cells was due to phagocytosis of apoptotic cells by macrophages, cytochalasin D was added to the co-cultures. Even when phagocytosis was inhibited by cytochalasin D, co-culturing 5T33MMvt/vv cells with macrophages resulted in an increased survival of MM cells (Supplementary Figure 1A). Furthermore, TAMs also increased the survival of 5T33MMvt cells in hypoxic (1% O2) conditions with or without treatment with bortezomib or melphalan compared to control (Supplementary Figure 1B).

TAM-mediated MM cell survival can be abrogated by STAT3 inhibition

To investigate the underlying mechanism of TAM-mediated MM survival, co-culture experiments were performed in the presence or absence of a transwell. The experiments were only performed with M5T33 macrophages, as the polarization by 5T33vt CM is more representative of the in vivo situation and provides the best protection against apoptosis. The increase in 5T33MMvt/vv cell survival was mediated by direct cell contact (Figure 5A-B). By measuring bromodeoxyuridine incorporation, the effects of macrophages on MM cell proliferation were investigated (Supplementary Figure 1C). No significant changes were observed, indicating that TAMs do not induce MM cell proliferation but rather MM cell survival. Therefore we investigated survival signaling pathways in MM cells co-cultured with TAMs. Western blot analysis revealed increased phosphorylation of STAT3 in 5T33MMvt/vv cells and also increased Akt phosphorylation in 5T33MMvv cells in co-culture with TAMs, while no changes in (p)ERK or MCL-1 levels were observed (Figure 5C). Furthermore, STAT3 activation was already observed 2 hours after initiation of direct cell contact (Supplementary Figure 1D). To investigate the involvement of pSTAT3 in TAM induced survival of MM cells the ATP-competitive JAK2 inhibitor AZD1480 was used. 5T33MMvt/vv cells alone had a slight but not significantly decreased survival when treated with AZD1480.
Interestingly, addition of AZD1480 to the co-culture resulted in an abrogation of the M5T33-induced survival benefit (Figure 5D-E). Furthermore, AZD1480 partially abrogated resistance to bortezomib-induced apoptosis (Figure 5F). We also evaluated the effect of STAT3 inhibition on M5T33 macrophages alone, and observed that AZD1480 decreased the viability of M5T33 macrophages (Figure 5G). Inhibition of STAT3 phosphorylation in 5T33MMvv cells and M5T33 macrophages was confirmed (Supplementary Figure 1E-F).

In vivo effects of STAT3 inhibition in combination with bortezomib on tumor load

Our in vitro data indicated that AZD1480 can target TAMs. Therefore, we first evaluated the effect of in vivo STAT3 inhibition on different myeloid cell subpopulations (e.g. monocytic (Ly6C+) cells, granulocytic (Ly6G+) cells and TAM subpopulations). AZD1480 treatment decreased the monocytic cells in peripheral blood (Figure 6A-B), while it did not change lymphocyte and granulocyte numbers (Supplementary Figure 2A-B). Furthermore, no significant changes were observed on total Ly6C+ and Ly6G+ cell populations in the BM (Supplementary Figure 2C-D) and the different TAM subpopulations based on Ly6C and MHCII expression (Figure 6C). To confirm the capacity of AZD1480 to inhibit phosphorylation of STAT3 in vivo, we performed western blot analysis for pSTAT3 and total STAT3 on BM cells isolated from mice 2 hours after gavage with AZD1480. In mice treated with AZD1480, a clear reduction of STAT3 phosphorylation was observed, with no impact on total STAT3 (Figure 6D).
Since we observed in vitro that STAT3 inhibition partially inhibited TAM-mediated bortezomib resistance in MM cells, we next investigated whether the combination of AZD1480 and bortezomib in vivo could further decrease tumor burden. At the end stage of the disease (day 21), tumor load was determined by evaluating plasmacytosis (Figure 6E). In the BM, a significant reduction of tumor load was seen in all treatment groups compared to control group, with the strongest effects in the combination group. A possible correlation of a reduced tumor load with a change in distribution of the TAM subpopulations was evaluated with flow cytometry in the BM. The total Ly6C+ cell population was not affected in any of the conditions (Supplementary Figure 2E). Significant changes were seen in the combination group, with an increase of Ly6G+ cells and inflammatory monocytes and a decrease of M1 TAMs and immature M2 cells in the BM, indicating that the combination treatment prevented the increase of TAMs during tumor progression (Figure 6F + Supplementary Figure 2F).
 
Discussion

The aim was to confirm the presence of TAM both in MM patients and in the 5T33MM mouse model and to investigate their role in survival and drug resistance in MM, also addressing the role of macrophage polarization in this process.
Suyani et al. [9] previously observed that the numbers of CD68+ (pan-macrophage) and CD163+ (M2) cells were both negative prognostic factors for overall survival in MM. We compared the presence of macrophages based on the expression of CD68 and two well characterized M2 markers CD163 and CD206 in the BM of healthy controls and in MM patients at diagnosis and relapse. We did not observe significantly higher CD68 and CD163 counts in the MM patients. In contrast, we were able to confirm higher CD206 counts in MM patients both at diagnosis and relapse, indicating a M2 directed polarization. There was a trend to higher CD163 and CD206 counts at relapse as compared to diagnosis, suggesting that TAMs have an increasing role throughout disease progression.
Making use of a gating strategy first reported by Laoui et al. [16], different TAM subpopulations were defined in the 5T33MM model. In accordance with patient data, M2 macrophages were also increased in end-stage diseased 5T33MM mice. It is notable that humans seem to have a dominance of M2 oriented macrophages compared to C57BL/KaLwRij mice. This can be explained by the fact that depending on the mouse strain, macrophages can be either M1-biased (as in C57BL/6 mice) or M2-biased (as in BALB/c mice) [26]. Nevertheless, we observed both in humans and mice an increase of M2 TAMs.
Furthermore, an in vitro macrophage polarization culture demonstrated that 5T33MMvt CM has the capacity to polarize macrophages to an M2 status, explaining the in vivo observation of an increasing M2 population in 5T33MM diseased mice. Addition of well-known anti-myeloma drugs to all different activation states of macrophages (M1, M2 and M5T33) had little impact on their viability, except for quisinostat, a secondgeneration HDAC inhibitor (HDACi). Recently, quisinostat’s potential anti-myeloma effects in combination with bortezomib and dexamethasone were evaluated in an open-label multicenter phase Ib study in patients with relapsed MM. An overall response rate of 88.2% was reached. [27]. Other HDACis like panobinostat, often used in combination with bortezomib and dexamethasone, have already made their entry in current treatment regimens of relapsed MM patients. Whether HDACi influence TAM number and functionality in vivo would be interesting to study in future.
Surprisingly, all types of macrophages increased MM cell survival, as demonstrated by fewer apoptotic cells and less caspase-3 cleavage; however, the most convincing results were obtained with M5T33 and M2 macrophages. Regardless of the addition of melphalan or bortezomib to the co-culture, this survival benefit was maintained. STAT3 activation was induced in MM cells in contact with TAMs. STAT3 is a crucial link between cancer and inflammation, as it is one of the most commonly activated oncogenic transcription factors in both cancer cells and micro-environmental myeloid immune cells like DCs, myeloid-derived suppressor cells (MDSCs) and TAMs, leading to pro-cancer and anti-tumor immune responses. We recently demonstrated a higher STAT3 phosphorylation in CD11b+ cells from MM-diseased mice compared to naive CD11b+ cells [28]. Moreover, cytokines/chemokines upregulated by STAT3 contribute to an autocrine activation loop [29–32]. Catlett-Falcone et al. demonstrated that constitutively activated STAT3 signaling in human U266 myeloma cells resulted in overexpression of the anti-apoptotic Bcl-xL protein and resistance to Fas-mediated apoptosis [33]. As previously described, TAM-induced MM survival is cell-cell contactmediated through PSGL-1 (P-selectin glycoprotein ligand-1)/selectins and ICAM1/CD18 interactions [6]. It has been previously shown that β1-integrins can also induce STAT3 activation. A necessity for direct cell contact to provide this protection does not exclude the possibility that changes in cytokine secretion induced by this contact by either MM cells or macrophages are involved. IL-6 and IL-10 are reported as important cytokines in micro-environmental protection of cancer cells, and are able to induce STAT3 [5,34–36]. However, blocking experiments with anti-IL-6 and anti-IL10 antibodies did not abrogate macrophage-induced MM cell survival (data not shown).
AZD1480 is a potent and competitive small-molecule inhibitor of JAK1/2 kinases and inhibits STAT3 phosphorylation. Considering its effects on both cancer cells and tumor infiltrating myeloid cells in vitro, we further investigated its potential in vivo. Maenhout et al. demonstrated that AZD1480 could delay tumor growth and decrease the percentage of MDSCs in melanoma bearing mice, much in the same way as specific knockout of STAT3 in melanoma cells [37]. However, the suppressive activity of the remaining MDSCs increased [38]. AZD1480 also exerted anti-cancer effects in several other murine cancer models, including small cell lung carcinoma, neuroblastoma and pediatric sarcoma [39,40]. Scuto et al. discovered that AZD1480 suppressed human myeloma cell growth either alone or in co-culture with BM stromal cells by blocking STAT3 and EGFR [41]. In our study, 5T33MMvt/vv cells were the most sensitive to STAT3 inhibition by AZD1480 in co-culture conditions. Interestingly, combining AZD1480 with bortezomib also resulted in a reduced survival benefit in MM cells.
The direct impact of AZD1480 on different monocyte/macrophage populations in vivo was evaluated. Except for the decrease in peripheral monocytes in blood, no major quantitative effects were seen in other white blood cells and TAM subpopulations in the BM. It is possible that the concentration of AZD1480 reached in the BM is lower than in blood, and the western blot analysis shows that STAT3 activation is not completely blocked in the BM. Furthermore, our in vitro data indicate that an efficient inhibition of STAT3 phosphorylation does not necessarily lead to a strong reduction in cell viability. In accordance with our results, no significant differences were observed in macrophage and MDSC numbers after AZD1480 treatment in an ovarian cancer immunocompetent mouse model [42]. Interestingly, in that study, AZD1480 treatment reduced pro-tumoral T cell populations in the peritoneal cavity. Despite the minor impact on myeloid cells at early stage in the 5T33MM model, AZD1480 treatment was still able to significantly reduce BM tumor load at the end-stage of disease. We hypothesize that 5T33MMvv cells become sensitive to AZD1480 due to the interaction with TAMs in the BM micro-environment. Interestingly, there is a clear correlation between the impact of the treatment on tumor load and the ratio between TAMs suggesting that targeting tumor cells and thus reducing M2 polarizing cytokines leads to a clear decrease in immature M2 TAMs. Taken together, our results, indicating that 5T33MM cells are sensitized to AZD1480 as a result of the TAM-induced activation of the STAT3 pathway, may lead to new therapeutic options in MM patients.

References

1. Raab MS, Podar K, Breitkreutz I, et al. Multiple myeloma. Lancet.
2. National Cancer Institute. Surveillance, Epidemiology, and End Results SEER Stat Fact Sheets [Internet]. National Institutes of Health. [cited 2016 Jul 14].
3. Hazlehurst L a, Damiano JS, Buyuksal I, et al. Adhesion to fibronectin via beta1 integrins regulates p27kip1 levels and contributes to cell adhesion mediated drug resistance (CAM-DR). Oncogene. 2000;19(38):4319–27.
4. Hazlehurst LA, Enkemann SA, Beam CA, et al. Multidrug Resistance in Cancer: Role of Atp-Dependent Transporters. Cancer Res. 2001;63(22):1–11.
5. Shain KH, Yarde DN, Meads MB, et al. β1 integrin adhesion enhances IL-6mediated STAT3 signaling in myeloma cells: Implications for microenvironment influence on tumor survival and proliferation. Cancer Res. 2009;69(3):1009–15.
6. Zheng Y, Yang J, Qian J, et al. PSGL-1/selectin and ICAM-1/CD18 interactions are involved in macrophage-induced drug resistance in myeloma. Leukemia.
7. Kim J, Denu RA, Dollar BA, et al. Macrophages and mesenchymal stromal cells support survival and proliferation of multiple myeloma cells. Br J Haematol.
8. Zheng Y, Cai Z, Wang S, et al. Macrophages are an abundant component of myeloma microenvironment and protect myeloma cells from chemotherapy drug-induced apoptosis. Blood. 2009;114(17):3625–8.
9. Suyani E, Sucak GT, Akyürek N, Ş et al. Tumor-associated macrophages as a prognostic parameter in multiple myeloma. Ann Hematol. 2013;92(5):669–77.
10. Li Y, Zheng Y, Li T, et al. Chemokines CCL2, 3, 14 stimulate macrophage bone marrow homing, proliferation, and polarization in multiple myeloma. Oncotarget. 2015;6(27):24218–29.
11. Beider K, Bitner H, Leiba M, et al. Multiple myeloma cells recruit tumorsupportive macrophages through the CXCR4/CXCL12 axis and promote their polarization toward the M2 phenotype. Oncotarget. 2014;5(22):11283–96.
12. Movahedi K, Laoui D, Gysemans C, et al. Different tumor microenvironments contain functionally distinct subsets of macrophages derived from Ly6C(high) monocytes. Cancer Res. 2010;70(14):5728–39.
13. Raes G, Noël W, Beschin A, et al. FIZZ1 and Ym as tools to discriminate between differentially activated macrophages. Dev Immunol. 2002;9(3):151–9.
14. Van Overmeire E, Laoui D, Keirsse J, et al. Mechanisms driving macrophage diversity and specialization in distinct tumor microenvironments and parallelisms with other tissues. Front Immunol. 2014;26(5):127.
15. Murray PJ, Allen JE, Biswas SK, et al. Macrophage Activation and Polarization: Nomenclature and Experimental Guidelines. Immunity. 2014;41(1):14–20.
16. Laoui D, Van Overmeire E, Conza G Di, et al. Tumor hypoxia does not drive differentiation of tumor-associated macrophages but rather fine-tunes the M2like macrophage population. Cancer Res. 2014;74(1):24–30.
17. Pollard JW. Trophic macrophages in development and disease. Nat Rev Immunol. 2009;9(4):259–70.
18. Qian BZ, Pollard JW. Macrophage Diversity Enhances Tumor Progression and Metastasis. Cell. 2010;141(1):39–51.
19. Sica A, Schioppa T, Mantovani A, et al. Tumour-associated macrophages are a distinct M2 polarised population promoting tumour progression: Potential targets of anti-cancer therapy. Eur J Cancer. 2006;42(6):717–27.
20. Andersen MN, Abildgaard N, Maniecki MB, et al. Monocyte/macrophage-derived soluble CD163: A novel biomarker in multiple myeloma. Eur J Haematol. 2014;93(1):41–7.
21. Andersen MN, Andersen NF, Rødgaard-Hansen S, et al. The novel biomarker of alternative macrophage activation, soluble mannose receptor (sMR/sCD206): Implications in multiple myeloma. Leuk Res. 2015;39(9):971–5.
22. Kim D, Wang J, Willingham SB, et al. Anti-CD47 antibodies promote phagocytosis and inhibit the growth of human myeloma cells. Leuk Off J Leuk Soc Am Leuk Res Fund, UK. 2012;26(12):1–8.
23. Calcinotto A, Ponzoni M, Ria R, et al. Modifications of the mouse bone marrow microenvironment favor angiogenesis and correlate with disease progression from asymptomatic to symptomatic multiple myeloma. Oncoimmunology. 2015;4(6):e1008850.
24. Scavelli C, Nico B, Cirulli T, et al. Vasculogenic mimicry by bone marrow macrophages in patients with multiple myeloma. Oncogene. 2008;27(5):663–74.
25. Asosingh K, Radl J, Van Riet I, et al. The 5TMM series: a useful in vivo mouse model of human multiple myeloma. Hematol J. 2000;1(5):351–6.
26. Mills CD, Kincaid K, Alt JM, et al. M-1/M-2 macrophages and the Th1/Th2 paradigm. J Immunol. 2000;164(12):6166–73.
27. Moreau P, Facon T, Touzeau C, et al. Quisinostat, bortezomib, and dexamethasone combination therapy for relapsed multiple myeloma. Leuk Lymphoma. 2016;57(7):1546–59.
28. De Veirman K, Van Ginderachter JA, Lub S, et al. Multiple myeloma induces Mcl-1 expression and survival of myeloid-derived suppressor cells. Oncotarget. 2015;6(12):10532–47.
29. Matsukawa A, Kudo S, Maeda T, et al. Stat3 in Resident Macrophages as a Repressor Protein of Inflammatory Response. J Immunol. 2005;175(5):3354–9.
30. Wang T, Niu G, Kortylewski M, et al. Regulation of the innate and adaptive immune responses by Stat-3 signaling in tumor cells. Nat Med. 2004;10(1):48–
31. Yu H, Pardoll D, Jove R. STATs in cancer inflammation and immunity: a leading role for STAT3. Nat Rev Cancer. 2009 Nov;9(11):798–809.
32. Dufait I, Van Valckenborgh E, Menu E, et al. Signal transducer and activator of transcription 3 in myeloid-derived suppressor cells: an opportunity for cancer therapy. Oncotarget. 2016;7(27): 42698–715.
33. Catlett-Falcone R, Landowski TH, Oshiro MM, et al. Constitutive activation of Stat3 signaling confers resistance to apoptosis in human U266 myeloma cells. Immunity. 1999;10(1):105–15.
34. Yang C, He L, He P, et al. Increased drug resistance in breast cancer by tumorassociated macrophages through IL-10/STAT3/bcl-2 signaling pathway. Med Oncol. 2015;32(2):352.
35. Shain KH, Dalton WS. Environmental-mediated drug resistance: a target for multiple myeloma therapy. Expert Rev Hematol. 2009;2(6):649–62.
36. Chatterjee M, Stühmer T, Herrmann P, et al. Combined disruption of both the MEK/ERK and the IL-6R/STAT3 pathways is required to induce apoptosis of multiple myeloma cells in the presence of bone marrow stromal cells. Blood. 2004;104(12):3712–21.
37. Emeagi PU, Maenhout S, Dang N, et al. Downregulation of Stat3 in melanoma: reprogramming the immune microenvironment as an anticancer therapeutic strategy. Gene Ther. 2013;20(11):1085–92.
38. Maenhout SK, Du Four S, Corthals J, et al. AZD1480 delays tumor growth in a melanoma model while enhancing the suppressive activity of myeloid-derived suppressor cells. Oncotarget. 2014;5(16):6801–15.
39. Lee JH, Park KS, Alberobello AT, et al. The janus kinases inhibitor AZD1480 attenuates growth of small cell lung cancers in vitro and in vivo. Clin Cancer Res. 2013;19(24):6777–86.
40. Yan S, Li Z, Thiele CJ. Inhibition of STAT3 with orally active JAK inhibitor, AZD1480, decreases tumor growth in Neuroblastoma and Pediatric Sarcomas In vitro and In vivo. Oncotarget. 2013;4(3):433–45.