Dysregulated bone morphogenetic proteins (BMP) may contribute to the development and progression of renal cell carcinoma (RCC). Herein, we report that BMP-6 promotes the growth of RCC by interleukin (IL)-10–mediated M2 polarization of tumor-associated macrophages (TAM). BMP-6–mediated IL-10 expression in macrophages required Smad5 and STAT3. In human RCC specimens, the three-marker signature BMP-6/IL-10/CD68 was associated with a poor prognosis. Furthermore, patients with elevated IL-10 serum levels had worse outcome after surgery. Together, our results suggest that BMP-6/macrophage/IL-10 regulates M2 polarization of TAMs in RCC. Cancer Res; 73(12); 3604–14. ©2013 AACR.

Kidney cancer is the seventh most common cancer with an estimated 58,240 new cases and 13,040 deaths in 2010 (1). Histologically, the most common type of kidney cancer is clear cell renal cell carcinoma (RCC; 80%–85%; ref. 2). As with most solid malignancies, localized RCC are usually cured with surgery. However, approximately 30% of the patients with clinically localized RCC eventually recur after resection.

Bone morphogenetic proteins (BMP) comprise the largest family within the TGF-β superfamily. Although originally reported as factors that induce bone and cartilage formation (3), BMPs have been shown to be critical for development (4–6). More recently, we and others have reported that BMPs regulate the immune system as BMP-6 inhibits B- and T-cell proliferation (7, 8) and activates macrophages (9, 10). In the context of RCC, BMP-4, -6, and -7 are often overexpressed (11, 12), whereas BMP antagonist Sclerostin domain–containing-1 (SOSTDC1) is downregulated (13). At the same time, we observed that human RCC cells frequently have a loss of expression of BMP receptors (11), suggesting a paracrine role for BMP-6 in RCC.

Macrophages are an essential component of the host defense system and have critical roles in both innate and adaptive immune responses (14). In solid malignancies, a large number of macrophages is usually found infiltrating the tumor. These tumor-associated macrophages (TAM; refs. 15, 16) are recruited from the circulating peripheral blood monocytes by chemotactic factors and chemokines (17, 18). When tumor cells arise, macrophages are capable of mounting an antitumor response (19). However, the decision to mount an antitumor response is in part regulated by the balance of pro- and anti-inflammatory factors present within the tumor microenvironment (14, 20). Indeed, a body of evidence suggests that macrophages in tumors can promote cancer progression and metastasis by stimulating angiogenesis, tumor growth, and cellular migration and invasion (21).

One cytokine that has been suggested to play a role in regulating macrophages is interleukin (IL)-10, a type II cytokine. IL-10 is produced by monocytes, activated macrophages, and a subtype of dendritic cells (22). In cancers, additional sources of IL-10 include the alternatively activated M2 macrophages and TAMs (23, 24). Because the major stimulus for the IL-10 expression is inflammation, IL-10 plays a major role in the negative feedback loop that prevents uncontrolled inflammation (25). In solid tumors, IL-10 has been detected in a number of malignancies (26). Recently, it has been suggested that a local production of IL-10 results in a tumor microenvironment that favors cancer cell survival and metastases (23, 27). In RCC, IL-10 and IL-10 receptor expression has been correlated with increased incidence of metastasis (28). Nevertheless, the clinical significance of IL-10 and the mechanism of its induction in cancers remain largely unknown. In this study, we report that BMP-6 in murine RCC cell lines is protumorigenic and involves IL-10–mediated M2 polarization of TAMs.

Animal studies

Balb/c and IL-10KO mice (The Jackson Laboratory) were inoculated subcutaneously with BMP-6/RenCa and BMP-6/RAG cells. For BMP-6 induction, drinking water containing 2 mg/mL of doxycycline (Sigma-Aldrich) in 5% sucrose was used. For adoptive transfer of macrophages, 5 × 106 cells were injected directly into tumors. Unless indicated, 5 animals were used per group.

Cell culture

RAW 264.7, RenCa, and RAG were purchased from the American Type Culture Collection and routinely maintained in as recommended by the manufacturer. To obtain murine peritoneal macrophages (PMφ), thioglycollate (Sigma-Aldrich) was dissolved in dH2O and injected intraperitoneally into mice. Animals were sacrificed 3 days later and PMφs were isolated from the peritoneal lavage and cultured as described (29). Primary human peripheral blood monocytes (hPBMC) were purchased from STEMCELL Technologies, Inc. The STAT3 inhibitor WP1066 (EMD Chemicals) was used at 10 μg/mL. BMP-6 was obtained from R&D Systems.

BMP-6–inducible murine renal cell carcinoma cell lines

Human BMP-6 cDNA was cloned into the tetracycline-inducible vector pLenti4/Dest (Invitrogen). Simultaneously, RenCa and RAG were infected sequentially with the tetracycline repressor virus followed by pLenti4/BMP-6 virus.

Reverse transcriptase PCR

Total RNA was extracted with TRIzol (Invitrogen) and reverse transcriptase PCR (RT-PCR) was carried out using the One-step SuperScript RT-PCR kit (Invitrogen). Reverse transcription was completed as recommended. For all primers, the annealing temperature was 55°C. Primer sequences are shown in Supplementary Table S1.

Immunoblot

Cells were harvested using the Cell Lysis Buffer (Cell Signaling Technology). After centrifugation, electrophoresis was carried out with 30 μg of protein using 12% SDS-PAGE gel and analyzed using enhanced chemiluminescence (Thermo Scientific).

Antibodies

The following antibodies were used in this study: mouse IL-10 (VWR), human IL-10 (R&D Systems), β-actin (Sigma), human BMP-6 blocking antibody (R&D), human BMP-6 (Abcam), Myc (Sigma), mouse STAT3, mouse phospho STAT3 (Cell Signaling), Smad 5 (Cell Signaling), phospho Smad 5 (Cell Signaling), Smad 1/5 (Santa Cruz), phospho Smad 1/5 (Cell Signaling), mouse CD206 (Serotec), human CD206 (BD), mouse CD163 (Santa Cruz), human CD163 (Abcam), human GAPDH (Santa Cruz), and human ki67 (Abcam).

Luciferase assay

Luciferase assay was conducted using the Dual-Luciferase Reporter Assay System (Promega) as per manufacturer's protocol. Cells were transfected using Lipofectamine 2000 (Invitrogen). All experiments were repeated at least 3 times.

Transient transfection of BMP receptors

RAW 264.7 cells were plated onto 6-well plates and Lipofectamine 2000 (Invitrogen) at 1 μL/mL was used to transfect 1 μg/mL of plasmids.

RNA knockdown studies

Sequences of short hairpin RNA (shRNA) and construction of expression system have previously been published (9, 29). Knockdown of STAT3 and STAT4 was carried out using siRNA (Qiagen).

Chromatin immunoprecipitation assay

EZ-ChIP Kit (Millipore) was used. Cells were harvested, fixed, and incubated with the indicated antibodies. Then DNA was sonicated and purified with spin column and analyzed with PCR.

Immunoprecipitation and immunoblot

NE-PER (Thermo Scientific) was used to separate cytoplasmic and nuclear proteins. After preclearing with protein A sepharose (GE healthcare), indicated primary antibody was added to the lysate and incubated at 4°C. Antibody was pulled down with protein A sepharose and washed with ice-cold lysis buffer and proteins were analyzed by immunoblot analysis.

ELISA

IL-10 and BMP-6 levels in the culture media and serum were measured using an ELISA Kit (R&D Systems). The indicated values represent the average of 4 separate measurements.

Clodronate liposome

Clodronate- and PBS-loaded liposomes were kindly provided by Clodronate Liposomes.org. Mice were injected with 0.2 mL of either clodronate liposomes (final dose of 25 mg/kg) or PBS liposomes every 4 days intraperitoneally or intratumorally.

Human RCC tissues, serum immunohistochemistry, and confocal microscopy

Seventy-four formalin-fixed paraffin-embedded human RCC tissues were obtained from Chungbuk National University (Cheongju, Korea), whereas human clear cell RCC tissue array containing 50 cases were purchased from Imgenex. Serum samples were available in 56 of the patients from Chungbuk National University. The sections were processed using a routine laboratory procedure and counterstained with hematoxylin. The results were interpreted by 2 independent investigators who were blinded to each other's reading.

Statistical analysis

Student t test and Pearson correlation were conducted. Clinical data were evaluated using Kaplan–Meier and multivariate Cox analysis. P < 0.05 was considered statistically significant.

BMP-6 is protumorigenic in RCC cells

Because published studies have suggested the dysregulation of BMPs in RCC (11, 13), tetracycline-inducible human BMP-6 expression system based on lentivirus was established in a murine RCC cell line, RenCa. Human BMP-6 cDNA was chosen to track the expression of exogenous BMP-6. Previously, the use of human BMP-6 in mouse has been validated (9, 29). Because RenCa cells were originally derived from Balb/c mice, syngeneic tumors can be established in immunocompetent animals. A representative result showing the inducibility of BMP-6 mRNA and protein by tetracycline is shown in Fig. 1A and B. One clone with a relatively high inducible level of BMP-6 was designated as RenCa/BMP-6. When BMP-6 expression was induced in RenCa/BMP-6 in tissue culture (Tet+), no significant effect on cellular proliferation was observed (Fig. 1C). However, in Balb/c mice, the induction of BMP-6 (doxy+) showed a significantly faster tumor growth without a noticeable change in histopathology (Fig. 1D). This protumorigenic effect of BMP-6 only in vivo implicates a paracrine mechanism. Along this line, it has recently been reported that BMPs regulate immune effector cells including macrophages (9, 10). To investigate whether macrophages mediate the protumorigenic effect of BMP-6 in RCC, clodronate liposome was used to selectively remove macrophages (30). The efficacy of clodronate liposome in removing macrophages was confirmed in RenCa/BMP-6 syngeneic tumor model (Supplementary Fig. S1A and S1B). As predicted, protumorigenic effect of BMP-6 was completely blocked with clodronate liposome (Fig. 1E). In these tumors, clodronate liposome had no effect on BMP-6 induction by doxycycline (Supplementary Fig. 1C).

Figure 1.

A, representative screening using semiquantitative qRT-PCR for human BMP-6 expression in RenCa is shown. Clone #2 was selected for subsequent experiments. B, ELISA for BMP-6 levels. Up to 120 pg/mL of BMP-6 was detected when cells were treated with tetracycline. C, MTT assay showed that the induction of BMP-6 expression (Tet+) had no significant effect on the rate of cellular proliferation in vitro. D, when RenCa/BMP-6 was inoculated subcutaneously into WT mice and BMP-6 expression was induced (doxycycline; doxy+), the tumor growth rate increased significantly. Hematoxylin staining showed no significant differences between doxy(-) and doxy(+) tumors. Induction of BMP-6 was confirmed by a RT-PCR and immunoblot analysis. E, when BMP-6 expression was induced in mice that were administered clodronate (CL/Doxy+), the increased tumor growth rate observed in the control group (PBS/Doxy+) was completely blocked. Hematoxylin staining showed no significant change in histopathology when macrophages were depleted.

Figure 1.

A, representative screening using semiquantitative qRT-PCR for human BMP-6 expression in RenCa is shown. Clone #2 was selected for subsequent experiments. B, ELISA for BMP-6 levels. Up to 120 pg/mL of BMP-6 was detected when cells were treated with tetracycline. C, MTT assay showed that the induction of BMP-6 expression (Tet+) had no significant effect on the rate of cellular proliferation in vitro. D, when RenCa/BMP-6 was inoculated subcutaneously into WT mice and BMP-6 expression was induced (doxycycline; doxy+), the tumor growth rate increased significantly. Hematoxylin staining showed no significant differences between doxy(-) and doxy(+) tumors. Induction of BMP-6 was confirmed by a RT-PCR and immunoblot analysis. E, when BMP-6 expression was induced in mice that were administered clodronate (CL/Doxy+), the increased tumor growth rate observed in the control group (PBS/Doxy+) was completely blocked. Hematoxylin staining showed no significant change in histopathology when macrophages were depleted.

Close modal

Next, the identical tetracycline-inducible BMP-6 system was developed in another murine kidney cancer cell line, RAG. The best clone with inducible BMP-6 expression was designated RAG/BMP-6. When RAG/BMP-6 was subcutaneously inoculated into Balb/c mice, induction of BMP-6 (doxy+) significantly enhanced tumor growth (Supplementary Fig. S2). Furthermore, clodronate liposome injection reversed the oncogenic effect of BMP-6. These results collectively suggest that the protumorigenic activity of BMP-6 requires macrophages.

Macrophages mediate the protumorigenic effect of BMP-6 via IL-10

When the murine macrophage cell line RAW 264.7 was treated with BMP-6, multiplex PCR revealed that BMP-6 upregulated the expression of a number of cytokines including IL-10. When RAW 264.7, murine PMφs, THP-1 (human monocyte cell line), and hPBMC were treated with BMP-6, IL-10 mRNA and protein were induced in a concentration- and time-dependent manner (Fig. 2A and B); quantitative RT-PCR (qRT-PCR) in RAW 264.7 is shown in Supplementary Fig. S3A. IL-10 mRNA induction by BMP-6 was detected as early as 1 hour and peaked at 24 hours in PMφs. ELISA revealed that the concentration of IL-10 in the conditioned media exceeded 120 pg/mL when RAW 264.7 was treated with 100 ng/mL BMP-6 (Supplementary Fig. S3B). This level of secreted IL-10 was comparable with that when macrophages were stimulated with lipopolysaccharide (LPS).

Figure 2.

A, BMP-6 induced IL-10 mRNA (top) and protein (bottom) in a concentration-dependent manner in RAW 264.7, murine PMφs, THP-1 human monocyte cell line, and hPBMC. LPS was used as a positive control. B, BMP-6 (100 ng/mL) induced IL-10 mRNA (top) and protein (bottom) expression in a time-dependent manner in RAW 264.7, PMϕ, THP-1, and hPBMC. C, when RenCa/BMP-6 was inoculated subcutaneously into IL-10KO mice, the induction of BMP-6 (doxy+) led to a dramatic regression of tumors and by 4 weeks, the tumors were no longer palpable. D, RenCa/BMP-6 tumors were harvested before complete tumor regression. Hematoxylin staining showed significant necrosis and neutrophil infiltration in tumors harvested from IL-10KO mice overexpressing BMP-6 (Doxy+).

Figure 2.

A, BMP-6 induced IL-10 mRNA (top) and protein (bottom) in a concentration-dependent manner in RAW 264.7, murine PMφs, THP-1 human monocyte cell line, and hPBMC. LPS was used as a positive control. B, BMP-6 (100 ng/mL) induced IL-10 mRNA (top) and protein (bottom) expression in a time-dependent manner in RAW 264.7, PMϕ, THP-1, and hPBMC. C, when RenCa/BMP-6 was inoculated subcutaneously into IL-10KO mice, the induction of BMP-6 (doxy+) led to a dramatic regression of tumors and by 4 weeks, the tumors were no longer palpable. D, RenCa/BMP-6 tumors were harvested before complete tumor regression. Hematoxylin staining showed significant necrosis and neutrophil infiltration in tumors harvested from IL-10KO mice overexpressing BMP-6 (Doxy+).

Close modal

To evaluate the role of IL-10 in vivo, IL-10 knockout (IL-10KO) mice were used. After inoculation, BMP-6 expression was induced (doxy+) when tumors became palpable. The results surprisingly showed not only a retardation of tumor growth but also a complete elimination of tumors within 4 weeks after the induction of BMP-6 (Fig. 2C). When these mice were rechallenged with RenCa/BMP-6 and RenCa, no tumors developed (data not shown). To examine the histopathologic changes of the tumors during regression in IL-10KO mice, doxycycline administration was started when the tumor size reached 5 mm in diameter. All tumors were harvested 2 weeks later. Hematoxylin staining showed a dramatic increase in neutrophil infiltration and extensive tumor necrosis in tumors expressing BMP-6 (Fig. 2D). These observations suggest that BMP-6 induces local inflammation and tumor rejection in the absence of IL-10.

BMP-6 induces IL-10 expression in macrophages via BMP-RII and ALK2/3

To determine the mechanism of IL-10 induction by BMP-6, we used the transcription inhibitor actinomycin D (ActD) and the translation inhibitor cycloheximide (CHX). IL-10 induction was blocked only by actinomycin D in both RAW 264.7 and PMφs (Fig. 3A). To expedite the dissection of the signaling pathway involved in BMP-6–induced IL-10 expression, we next constructed a luciferase reporter vector containing the IL-10 promoter (IL-10-Luc). When IL-10-Luc was transiently transfected into RAW 264.7, BMP-6 increased luciferase activity approximately 2.5-fold within 6 hours (Supplementary Fig. S3C).

Figure 3.

A, semiquantitative RT-PCR showed that actinomycin D (ActD) but not cycloheximide blocked the induction of IL-10 by BMP-6 in PMφs and RAW 264.7. CTL, control. B, each of the 3 type II BMP receptors (ActRIIA, ActRIIB, and BMPRII) and IL10-Luc reporter plasmid were cotransfected into RAW264.7. When treated with BMP-6, only the transfection of BMP-RII led to a significant increase in luciferase activity. C, each of the 3 type II BMP receptors was knocked down using the shRNA approach. When IL10-Luc was cotranfected, only the knockdown of BMP-RII blocked IL-10 induction by BMP-6. D, constitutively active type I receptors—CA-ALK2 and CA-ALK3—were cotransfected along with IL10-Luc into RAW 264.7. Increased luciferase activity was seen with the expression of either CA-ALK2 or CA-ALK3. E, when either ALK2 or ALK3 was knocked down using shRNA, IL-10 induction by BMP-6 was blocked. *, statistically significant.

Figure 3.

A, semiquantitative RT-PCR showed that actinomycin D (ActD) but not cycloheximide blocked the induction of IL-10 by BMP-6 in PMφs and RAW 264.7. CTL, control. B, each of the 3 type II BMP receptors (ActRIIA, ActRIIB, and BMPRII) and IL10-Luc reporter plasmid were cotransfected into RAW264.7. When treated with BMP-6, only the transfection of BMP-RII led to a significant increase in luciferase activity. C, each of the 3 type II BMP receptors was knocked down using the shRNA approach. When IL10-Luc was cotranfected, only the knockdown of BMP-RII blocked IL-10 induction by BMP-6. D, constitutively active type I receptors—CA-ALK2 and CA-ALK3—were cotransfected along with IL10-Luc into RAW 264.7. Increased luciferase activity was seen with the expression of either CA-ALK2 or CA-ALK3. E, when either ALK2 or ALK3 was knocked down using shRNA, IL-10 induction by BMP-6 was blocked. *, statistically significant.

Close modal

BMPs signal through a heteromeric complex of type I and II receptors. When each of the 3 known BMP type II receptors were cotransfected with IL-10-Luc into RAW 264.7, only BMP-RII overexpression increased IL-10 promoter activity upon treatment with BMP-6 (Fig. 3B). Conversely, knockdown experiments using shRNA (9) showed the suppression of luciferase activity only when BMP-RII was targeted (Fig. 3C). To identify the functional BMP type I receptor(s), constitutively active mutant constructs were used. Macrophages express only 2 of the 3 known BMP type I receptors—ALK2 and 3 (9). Overexpression of constitutively active ALK2 and 3 (CA-ALK2 and CA-ALK3, respectively) increased the luciferase activity more than 10-fold (Fig. 3D). When ALK2 and 3 were knocked down using shRNA (9), BMP-6 no longer induced luciferase activity (Fig. 3E). These results collectively suggest that BMP-RII along with ALK2 and 3 are the functional BMP-6 receptors that mediate IL-10 induction in macrophages.

IL-10 induction requires Smad5

Following the activation of receptors, the canonical BMP signaling requires receptor-mediated Smads (R-Smads; Smad1, 5, and 8) and co-Smad (Smad4). When the Smad-dependent pathway was blocked in RAW 264.7 with dominant-negative Smad4 (Smad4DN), IL-10 induction by BMP-6 was completely blocked (Fig. 4A). When each of the R-Smads was overexpressed in RAW 264.7, only Smad5 transfection increased luciferase activity upon treatment with BMP-6 (Fig. 4B). This activation of IL-10 promoter by BMP-6 was then blocked when Smad5 was knocked down using shRNA (Fig. 4C; ref. 29). Serial deletion of IL-10 promoter subsequently showed that the BMP-6 response element within IL-10 promoter is located between 588 and 800 bp 5′ to the transcription start site (Fig. 4D). Chromatin immunoprecipitation (ChIP) analysis revealed that Smad5 bound to the −588 to −800 region of IL-10 promoter (Fig. 4E).

Figure 4.

A, when RAW 264.7 was transfected with the dominant-negative Smad4 (Smad4DN), the BMP-6–induced IL-10 mRNA induction was completely blocked. B, each of the 3 R-Smads (Smad1, 5, and 8) was cotransfected with IL-10-Luc into RAW 264.7. When treated with BMP-6 (100 ng/mL), only Smad5 overexpression significantly increased IL-10 promoter activity. *, statistically significant. C, each of the 3 R-Smads was knocked down with shRNA in RAW264.7. When cotransfected with IL-10-Luc reporter and treated with BMP-6 (100 ng/mL), shRNA targeting Smad5 was the most efficient blocker of BMP-6–induced IL-10 promoter activity. D, promoter deletion analysis showed that the most significant decrease in the BMP-6–induced IL-10 promoter activity was observed when the region between −800 to −588 bp 5′ upstream of the transcription start site was deleted. E, ChIP assay showed that Smad5 bound to the −800 to −588 bp region of IL-10 promoter in a ligand-dependent manner. As a control, exon 3 of IL-10 gene was amplified.

Figure 4.

A, when RAW 264.7 was transfected with the dominant-negative Smad4 (Smad4DN), the BMP-6–induced IL-10 mRNA induction was completely blocked. B, each of the 3 R-Smads (Smad1, 5, and 8) was cotransfected with IL-10-Luc into RAW 264.7. When treated with BMP-6 (100 ng/mL), only Smad5 overexpression significantly increased IL-10 promoter activity. *, statistically significant. C, each of the 3 R-Smads was knocked down with shRNA in RAW264.7. When cotransfected with IL-10-Luc reporter and treated with BMP-6 (100 ng/mL), shRNA targeting Smad5 was the most efficient blocker of BMP-6–induced IL-10 promoter activity. D, promoter deletion analysis showed that the most significant decrease in the BMP-6–induced IL-10 promoter activity was observed when the region between −800 to −588 bp 5′ upstream of the transcription start site was deleted. E, ChIP assay showed that Smad5 bound to the −800 to −588 bp region of IL-10 promoter in a ligand-dependent manner. As a control, exon 3 of IL-10 gene was amplified.

Close modal

STAT3 is required for IL-10 induction

Analysis of IL-10′s BMP-6–responsive region revealed no consensus Smad-binding element. However, one STAT-binding motif was identified. Previously, it has been reported that STAT5 interacts with Smad2 of the TGF-β signaling pathway (31). Of the known subtypes of STATs, STAT3 has been reported regulate IL-10 expression (32). Accordingly, immunoblot was carried out to determine the phosphorylation status of STAT3 after treating RAW 264.7 with BMP-6. The results showed increased phosphorylation of STAT3 (pSTAT3) within 30 minutes after BMP-6 treatment (Fig. 5A). Blocking STAT3 with the small-molecule inhibitor WP1066 (33) completed neutralized IL-10 induction by BMP-6 (Fig. 5B). Next, studies using dominant-negative STAT3 (STAT3DN) and small-interfering RNA (siSTAT3) also showed a complete abrogation of IL-10 induction by BMP-6 in RAW 264.7 (Fig. 5C and Supplementary Fig. S4A). When RAW 264.7 was transiently transfected with siSTAT3 and IL10-Luc, BMP-6 no longer increased luciferase activity (Supplementary Fig. S4B). The complementary experiment in which STAT3 and IL-10-Luc were overexpressed in RAW 264.7, BMP-6 increased the luciferase activity significantly (Fig. 5D). Surprisingly, overexpression of STAT4 repeatedly blocked the activation of IL-10 promoter by BMP-6.

Figure 5.

A, RAW 264.7 was treated with BMP-6 (100 ng/mL) and immunoblot was carried out. The results showed a dramatic increase in phosphorylated STAT3 within 15 minutes (0.25 hour) of BMP-6 treatment. B, inhibition of STAT3 with WP1066 (10 μg/mL) showed the abrogation of IL-10 induction by BMP-6 in RAW 264.7. WP1066 did not inhibit the LPS-induced IL-10 expression. C, transfection of dominant-negative STAT3 (STAT3DN) or siRNA-targeting STAT3 blocked the induction of IL-10 by BMP-6 in RAW 264.7. The knockdown of STAT4 (siSTAT4) had no significant effect on IL-10 mRNA. D, when RAW 264.7 was cotransfected with either STAT3 or STAT4 and IL-10-Luc reporter and treated with BMP-6 (100 ng/mL), the overexpression of STAT3 significantly increased IL-10 promoter activity. Interestingly, STAT4 transfection repeatedly blocked the activation of IL-10 promoter. *, statistically significant. E, ChIP assay showed that STAT3 bound to the BMP-6–responsive region (−800 to −588 bp) within the IL-10 promoter in RAW 264.7. As control, exon 3 of IL-10 gene was amplified. F, immunoprecipiation against STAT3 pulled down Smad5. Conversely, immunoprecipitation against Smad5 pulled down STAT3. G, immunoprecipitation was carried out in cytosolic and nuclear fractions from RAW 264.7. Coimmunoprecipitation of Smad5 and STAT3 was observed in the nuclear fraction. The low level of STAT3 immunoprecipitated out with Smad5 in the nucleus in the absence of BMP-6 stimulation likely reflects the basal leakage of BMP-6. H, in RenCa/BMP-6 tumors, there was a significant increase in phosphorylated levels of STAT3 and Smad5 in samples obtained from mice with BMP-6 induction (Doxy+). I, confocal immunofluorescence microscopy revealed that both pSTAT3 and pSmad1/5 colocalized to macrophages (F4/80+), especially in tumors harvested from Doxy(+) mice.

Figure 5.

A, RAW 264.7 was treated with BMP-6 (100 ng/mL) and immunoblot was carried out. The results showed a dramatic increase in phosphorylated STAT3 within 15 minutes (0.25 hour) of BMP-6 treatment. B, inhibition of STAT3 with WP1066 (10 μg/mL) showed the abrogation of IL-10 induction by BMP-6 in RAW 264.7. WP1066 did not inhibit the LPS-induced IL-10 expression. C, transfection of dominant-negative STAT3 (STAT3DN) or siRNA-targeting STAT3 blocked the induction of IL-10 by BMP-6 in RAW 264.7. The knockdown of STAT4 (siSTAT4) had no significant effect on IL-10 mRNA. D, when RAW 264.7 was cotransfected with either STAT3 or STAT4 and IL-10-Luc reporter and treated with BMP-6 (100 ng/mL), the overexpression of STAT3 significantly increased IL-10 promoter activity. Interestingly, STAT4 transfection repeatedly blocked the activation of IL-10 promoter. *, statistically significant. E, ChIP assay showed that STAT3 bound to the BMP-6–responsive region (−800 to −588 bp) within the IL-10 promoter in RAW 264.7. As control, exon 3 of IL-10 gene was amplified. F, immunoprecipiation against STAT3 pulled down Smad5. Conversely, immunoprecipitation against Smad5 pulled down STAT3. G, immunoprecipitation was carried out in cytosolic and nuclear fractions from RAW 264.7. Coimmunoprecipitation of Smad5 and STAT3 was observed in the nuclear fraction. The low level of STAT3 immunoprecipitated out with Smad5 in the nucleus in the absence of BMP-6 stimulation likely reflects the basal leakage of BMP-6. H, in RenCa/BMP-6 tumors, there was a significant increase in phosphorylated levels of STAT3 and Smad5 in samples obtained from mice with BMP-6 induction (Doxy+). I, confocal immunofluorescence microscopy revealed that both pSTAT3 and pSmad1/5 colocalized to macrophages (F4/80+), especially in tumors harvested from Doxy(+) mice.

Close modal

Next, ChIP assay showed that STAT3 bound to the −588 to −800 bp region of IL-10 promoter (Fig. 5E). Because Smad5 also bound to the same region, we subsequently tested whether STAT3 interacts with Smad 5. When each of the myc-tagged R-Smads was expressed in RAW 264.7, immunoprecipitation against endogenous STAT3 pulled down Smad5, whereas immunoprecipitation against myc-Smad5 precipitated out STAT3 (Supplementary Fig. S4C). The physical interaction between endogenous Smad5 and STAT3 in macrophages, when stimulated by BMP-6, was then confirmed (Fig. 5F).

Consistent with the known functions of Smad5 and STAT3, confocal immunofluorescence microscopy showed that both STAT3 and Smad5 translocate to the nucleus of RAW 264.7 when treated with BMP-6 (Supplementary Fig. 4D). This BMP-6–induced nuclear translocation of STAT3 was still detected in RAW 264.7 even when Smad5 was knocked down (Supplementary Fig. S4E). The reverse experiment revealed that the nuclear translocation of Smad5 still occurred when STAT3 was knocked down (Supplementary Fig. S4F). Subsequently, coimmunoprecipitation of STAT3 and Smad5 was observed only in the nuclear fraction (Fig. 5G). In the absence of exogenous BMP-6, low levels of STAT3 and Smad5 were coimmunoprecipitated out in RAW 264.7, likely suggesting a low basal level of expression of BMPs.

In RenCa/BMP-6 tumors, immunoblot confirmed the presence of both phosphorylated STAT3 and Smad1/5 (Fig. 5H). Subsequently, immunofluorescence microscopy showed the colocalization of pSTAT3, pSmad1/5, and F4/80 (murine macrophages; Fig. 5I). Interestingly, the induction of BMP-6 showed a significantly higher number of macrophages within the tumors. These results confirm that both Smad5 and STAT3 are phosphorylated in macrophages upon BMP-6 induction in vivo.

BMP-6 induces M2 macrophage polarization via IL-10

IL-10 induces M2 polarization of macrophages that leads to local immunosuppression (19). When RAW 264.7 was cocultured with RenCa, there was a 10-fold increase in positive M2 marker (CD206+/CD163+) cells (3.8% vs. 40%;Fig. 6A). This increase in M2 polarization was partially reversed when RenCa/RAW 264.7 coculture was treated with BMP-6 or IL-10–neutralizing antibodies (Fig. 6A). When RAW 264.7 and hPBMC were treated directly with 100 ng/mL of BMP-6 for 3 days, there was a 4-fold increase in M2 cells (0.8% vs. 3.8% for RAW 264.7 and 3.8% vs. 16.7% for hPBMC; Fig. 6B). The induction of M2 polarization by BMP-6 was again partially inhibited by anti-IL-10 antibody.

Figure 6.

A, flow cytometry showed M2 polarization (CD206+ or CD163+) when RAW 264.7 was cocultured with RenCa. This M2 polarization was reversed partially when the cells were treated with neutralizing antibodies to either BMP-6 or IL-10. *, statistically significant. B, RAW 264.7 and hPBMC were treated with BMP-6 (100 ng/mL). Flow cytometry again showed a dramatic increase in M2 polarization of both cell types. This increase in M2 fraction induced by BMP-6 was blocked partially by IL-10–neutralizing antibodies. *, statistically significant. C, endogenous TAMs were depleted with clodronate liposomes in 40 Balb/c mice. When WT-macrophages (WT Mφ) were injected into the tumors, induction of BMP-6 (WT/doxy+) resulted in a dramatic increase in tumor growth rate when compared with the control group (WT/doxy−). In contrast, the adoptive transfer of macrophages derived from IL-10KO mice led to a significant decrease in tumor growth with or without BMP-6 induction; however, BMP-6 induction resulted in the most significant decrease in tumor growth rate. RT-PCR showed a decreased IL-10 mRNA level in the IL-10KO macrophages group (IL-10KO Mφ). The low level of IL-10 expression detected in the IL-10KOMφ/Doxy+ group reflects the repopulation of tumors by endogenous macrophages as clodronate liposome was injected only once at the beginning of the experiment. D, hematoxylin staining showed an increase in neutrophil infiltration and tumor necrosis when IL-10KO Mφs were adoptively transferred into tumors overexpressing BMP-6.

Figure 6.

A, flow cytometry showed M2 polarization (CD206+ or CD163+) when RAW 264.7 was cocultured with RenCa. This M2 polarization was reversed partially when the cells were treated with neutralizing antibodies to either BMP-6 or IL-10. *, statistically significant. B, RAW 264.7 and hPBMC were treated with BMP-6 (100 ng/mL). Flow cytometry again showed a dramatic increase in M2 polarization of both cell types. This increase in M2 fraction induced by BMP-6 was blocked partially by IL-10–neutralizing antibodies. *, statistically significant. C, endogenous TAMs were depleted with clodronate liposomes in 40 Balb/c mice. When WT-macrophages (WT Mφ) were injected into the tumors, induction of BMP-6 (WT/doxy+) resulted in a dramatic increase in tumor growth rate when compared with the control group (WT/doxy−). In contrast, the adoptive transfer of macrophages derived from IL-10KO mice led to a significant decrease in tumor growth with or without BMP-6 induction; however, BMP-6 induction resulted in the most significant decrease in tumor growth rate. RT-PCR showed a decreased IL-10 mRNA level in the IL-10KO macrophages group (IL-10KO Mφ). The low level of IL-10 expression detected in the IL-10KOMφ/Doxy+ group reflects the repopulation of tumors by endogenous macrophages as clodronate liposome was injected only once at the beginning of the experiment. D, hematoxylin staining showed an increase in neutrophil infiltration and tumor necrosis when IL-10KO Mφs were adoptively transferred into tumors overexpressing BMP-6.

Close modal

Adoptive transfer of macrophages

Although BMP-6 activates macrophages and induces the expression of type I proinflammatory cytokines including IL-6 and IL-1β (29, 34). Results of the present study suggest that the immunosuppressive effect of IL-10 seems to overcome the proinflammatory effects of IL-6 and IL-1β and mediates the overriding biologic effect of BMP-6 that results in tumor growth in RCC. Therefore, the disruption of IL-10 production by macrophages in the context of BMP-6 may tip the balance of local immune response in favor of inflammation that may lead to the activation of host immune system and tumor regression. To investigate this possibility, we again used the RenCa/BMP-6 tumor model in Balb/c mice following macrophage depletion with intraperitoneal injection of clodronate liposome (Fig. 6C). When wild-type (WT) macrophages were adoptively transferred intratumorally, the induction of BMP-6 led to a dramatic increase in tumor growth. In contrast, the injection of IL-10KO macrophages not only blocked the protumorigenic effect of BMP-6 but also decreased the rate of tumor growth significantly. The modest level of IL-10 detected in animals administered clodronate liposome is likely due to the recovery of native macrophages by the end of the experiment. Hematoxylin staining showed a dramatic increase in neutrophil infiltration and necrosis in tumors injected with IL-10KO macrophages (Fig. 6D).

Analysis in human clear cell RCC tissues

To determine the clinical relevance of the current findings, 124 human clear cell RCC tissues were examined. A typical example of positive immunohistochemistry for BMP-6, CD68, IL-10, and CD206 is shown in Fig. 7A. Confocal immunofluorescence microscopy revealed that in samples that were BMP-6–positive, CD68, IL-10, and CD206 frequently colocalized (Fig. 7B). When each marker was analyzed individually in the context of tumor stage, grade, and size, only CD68 and BMP-6 correlated with grade, whereas BMP-6 was associated with stage (Supplementary Fig. S5A). When cumulative survival rate was examined, higher tumor stage and grade were associated with poor prognosis (Supplementary Fig. S5B and S5C); this confirms the validity of our samples. Subsequent analysis of the 4 markers failed to show any association with outcome (Supplementary Fig. S5D). However, when each of the 4 markers were examined collectively or in various combinations, being positive for 3 markers (BMP-6, CD68, and IL-10) was associated with decreased cumulative survival (Fig. 7C); CD206 status did not add any further prognostic value. Next, a multivariate Cox analysis showred that the 3-marker signature (BMP-6/CD68/IL-10) was an independent predictor of cumulative survival (Supplementary Table S2). When ELISA was carried out with serum banked from 56 patients, BMP-6 levels were not high enough for detection. However, IL-10 serum level more than 1 pg/mL was associated with an increased incidence of metastasis over a 5-year period (Fig. 7D).

Figure 7.

A, examples of human clear cell RCC specimens that were positive by immunohistochemistry for BMP-6, IL-10, macrophages (CD68+), and M2 polarization (CD206) are shown. B, confocal immunofluorescence microscopy was used to colocalize BMP-6, IL-10, macrophages (CD68+), and M2 polarization (CD206+) in human RCC tissues. BMP-6 and IL-10 frequently colocalized with CD68. C, the prognostic value of BMP-6, CD68, IL-10, and CD206 was studied in 124 patients with clear cell RCC using Kaplan–Meier analysis. Having the 3-marker signature (BMP-6/CD68/IL-10) was associated with a decreased cumulative survival. D, serum levels of IL-10 were correlated with the rate of 5-year metastasis in 56 patients with clear cell RCC. In patients with IL-10 serum levels more than 1 pg/mL, there was a significantly higher rate of metastasis (P = 0.022). E, proposed model for BMP-6′s protumorigenic effect in RCC. BMP-6 produced by tumor activates TAMs and induces the type II cytokine IL-10 expression via Smad5 and STAT3 interaction. IL-10, then, induces M2 polarization of TAMs and suppresses the local antitumor immune response, thus leading to tumor proliferation and progression.

Figure 7.

A, examples of human clear cell RCC specimens that were positive by immunohistochemistry for BMP-6, IL-10, macrophages (CD68+), and M2 polarization (CD206) are shown. B, confocal immunofluorescence microscopy was used to colocalize BMP-6, IL-10, macrophages (CD68+), and M2 polarization (CD206+) in human RCC tissues. BMP-6 and IL-10 frequently colocalized with CD68. C, the prognostic value of BMP-6, CD68, IL-10, and CD206 was studied in 124 patients with clear cell RCC using Kaplan–Meier analysis. Having the 3-marker signature (BMP-6/CD68/IL-10) was associated with a decreased cumulative survival. D, serum levels of IL-10 were correlated with the rate of 5-year metastasis in 56 patients with clear cell RCC. In patients with IL-10 serum levels more than 1 pg/mL, there was a significantly higher rate of metastasis (P = 0.022). E, proposed model for BMP-6′s protumorigenic effect in RCC. BMP-6 produced by tumor activates TAMs and induces the type II cytokine IL-10 expression via Smad5 and STAT3 interaction. IL-10, then, induces M2 polarization of TAMs and suppresses the local antitumor immune response, thus leading to tumor proliferation and progression.

Close modal

Taken together, we propose that clear cell RCC–derived BMP-6 stimulates the expression of the anti-inflammatory type II cytokine IL-10 in TAMs via a physical interaction between Smad5 and STAT3. IL-10, in turn, induces M2 polarization of TAMs and suppresses the local antitumor immune response (Fig. 7E).

Because many types of solid tumors have elevated expression levels of BMPs, it has been proposed that BMPs may be oncogenic (35). Consistent with this hypothesis, we have observed that the overexpression of BMP-6 has a protumorigenic effect in 2 murine RCC cell lines. This oncogenic activity of BMP-6 though, was seen only in vivo, suggesting that BMP-6 stimulates RCC proliferation via a paracrine mechanism.

Common paracrine mechanisms of carcinogenesis include immunomodulation and angiogenesis. Results of the present study showed that BMP-6 stimulates TAMs to produce IL-10, a type II cytokine that suppresses inflammation (36). The induction of IL-10 by BMP-6 was associated with M2 polarization of macrophages. This observation is consistent with the report that IL-10 induces M2 polarization of macrophages (14). Because M2 macrophages have poor antigen-presenting capacity and suppress T-cell activity (35), the protumorigenic effect of BMP-6 is likely mediated by IL-10–induced M2 polarization of TAMs. Indeed, the removal of either macrophages or IL-10 reversed the protumorigenic effect of BMP-6. In short, these results suggest that BMP-6 induces tumor growth by altering the tumor microenvironment in favor of anti-inflammation and local immune suppression. The possibility remains though that BMP-6 may also induce angiogenesis via a yet-to-be identified mechanism.

In the context of IL-10KO mice, BMP-6 induction resulted in a dramatic tumor regression. Because these mice developed no tumors when rechallenged, BMP-6 induces antitumor immune response in the absence of IL-10. Mechanistically, we have observed that BMP-6 also induces the expression of the proinflammatory type I cytokines IL-6 and IL-1β in macrophages (29, 34). Accordingly, in the absence of IL-10 acting as a counterweight, the balance of the effect of BMP-6 in vivo may be skewed toward inflammation and antitumor response. Consistent with this concept, the adoptive transfer of macrophages from IL-10KO mice retarded BMP-6–induced tumor growth. Therefore, blocking IL-10 may be an effective therapy in a subgroup of patients with RCC with dysregulated BMP-6/IL-10.

Canonical BMP signaling requires a heterotetrameric complex of type I and II receptors. To date, 3 each of type I and II receptors have been reported (37). Although these receptors are promiscuous, there is an optimal receptor combination for each BMP as it has been shown that there are coreceptors that are relatively specific for BMP-2 and -4 (38). In the context of BMP-6/IL-10, BMP-RII, along with ALK2 and 3, efficiently transduced BMP-6 signal. Whether there is a BMP-6–specific coreceptor remains to be studied.

Intracellularly, both Smad5 and STAT3 are necessary for IL-10 induction by BMP-6. ChIP showed that both Smad5 and STAT3 bind to the same region within the IL-10 promoter. Because the IL-10 promoter contains STAT3 but not a consensus Smad-binding element, it is likely that STAT3 interacts directly with the DNA, whereas Smad5 is a cofactor for the transcriptional activation of IL-10 by BMP-6. Further studies are necessary to verify this concept.

In patients with clear cell RCC, BMP-6, CD68, IL-10, and CD206 had no prognostic value individually. This observation is consistent with the report that BMP-6 in clear cell RCC tissues does not predict outcome (12). Notwithstanding, patients who had the 3-marker signature composed of BMP-6, CD68, and IL-10 were more likely to die from RCC. Interestingly though, CD206 (M2 macrophages) did not add any further prognostic value. This is likely due to colinearity as the presence of IL-10 and macrophages (CD68) were consistently associated with CD206. These observations suggest that the effect of BMP-6 in RCC is dictated by the tumor microenvironment and that IL-10 may be a surrogate marker of the protumorigenic effect of BMP-6.

Results of the present study implicate IL-10 as a new biomarker and a potential therapeutic target in RCC. Specifically, elevated serum levels of IL-10 were associated with increased risk of metastatic RCC. On the therapeutic front, an adoptive transfer of IL-10KO macrophages resulted in a dramatic tumor regression. Therefore, serum IL-10 levels may eventually be used to identify patients who are best candidates for targeting BMP-6/IL-10 loop. We are planning to test this concept in patients with clear cell RCC using AS101, a small-molecule inhibitor of IL-10 (39).

It should be pointed out that elevated tissue levels of IL-10 mRNA have been correlated with poor prognosis in patients with RCC (40). Herein though, IL-10 alone did not predict outcome. The underlying reason for this difference is unclear. Possible explanations include the different methodologies used to detect IL-10—immunohistochemistry versus RT-PCR. Further studies are needed to clarify this discrepancy.

In conclusion, BMP-6 is protumorigenic in RCC. This oncogenic effect of BMP-6 is mediated by macrophage-derived IL-10, which leads to M2 polarization of TAMs. At the molecular level, BMP-6 induces IL-10 expression in macrophages via a direct interaction between STAT3 and Smad5 in the nucleus. In patients with clear cell RCC, the 3-marker signature BMP-6/CD68/IL-10 predicted decreased cumulative survival after surgery, whereas elevated serum IL-10 levels were associated with increased risk of metastasis. Taken together, these results show a novel mechanism of M2 polarization of TAMs in RCC that involves the BMP-6/IL-10 axis and provides the proof-of-principle for developing a personalized immunotherapy based on serum IL-10 level.

No potential conflicts of interest were disclosed.

Conception and design: J.-H. Lee, G.T. Lee, I.Y. Kim

Development of methodology: J.-H. Lee, G.T. Lee, I.Y. Kim

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): G.T. Lee, Y.-S. Ha, I.Y. Kim

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): J.-H. Lee, G.T. Lee, Y.-S. Ha, I.Y. Kim

Writing, review, and/or revision of the manuscript: J.-H. Lee, G.T. Lee, S.J. Kwon, I.Y. Kim

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): J.-H. Lee, S.H. Woo, S.J. Kwon, I.Y. Kim

Study supervision: J.-H. Lee, S.J. Kwon, W.-J. Kim, I.Y. Kim

This work was supported in part by the Tanzman Foundation, John Shein, and Malcolm Wernik.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1.
Jemal
A
,
Siegel
R
,
Xu
J
,
Ward
E
. 
Cancer statistics, 2010
.
CA Cancer J Clin
2010
;
60
:
277
300
.
2.
Baldewijns
MM
,
van Vlodrop
IJ
,
Schouten
LJ
,
Soetekouw
PM
,
de Bruine
AP
,
van Engeland
M
. 
Genetics and epigenetics of renal cell cancer
.
Biochim Biophys Acta
2008
;
1785
:
133
55
.
3.
Nishimura
R
,
Hata
K
,
Matsubara
T
,
Wakabayashi
M
,
Yoneda
T
. 
Regulation of bone and cartilage development by network between BMP signalling and transcription factors
.
J Biochem
2012
;
151
:
247
54
.
4.
Zhao
GQ
. 
Consequences of knocking out BMP signaling in the mouse
.
Genesis
2003
;
35
:
43
56
.
5.
Sieber
C
,
Kopf
J
,
Hiepen
C
,
Knaus
P
. 
Recent advances in BMP receptor signaling
.
Cytokine Growth Factor Rev
2009
;
20
:
343
55
.
6.
Perron
JC
,
Dodd
J
. 
ActRIIA and BMPRII Type II BMP receptor subunits selectively required for Smad4-independent BMP7-evoked chemotaxis
.
PLoS ONE
2009
;
4
:
e8198
.
7.
Sivertsen
EA
,
Huse
K
,
Hystad
ME
,
Kersten
C
,
Smeland
EB
,
Myklebust
JH
. 
Inhibitory effects and target genes of bone morphogenetic protein 6 in Jurkat TAg cells
.
Eur J Immunol
2007
;
37
:
2937
48
.
8.
Kersten
C
,
Sivertsen
EA
,
Hystad
ME
,
Forfang
L
,
Smeland
EB
,
Myklebust
JH
. 
BMP-6 inhibits growth of mature human B cells; induction of Smad phosphorylation and upregulation of Id1
.
BMC Immunol
2005
;
6
:
9
.
9.
Hong
JH
,
Lee
GT
,
Lee
JH
,
Kwon
SJ
,
Park
SH
,
Kim
SJ
, et al
Effect of bone morphogenetic protein-6 on macrophages
.
Immunology
2009
;
128
:
e442
50
.
10.
Kwon
SJ
,
Lee
GT
,
Lee
JH
,
Kim
WJ
,
Kim
IY
. 
Bone morphogenetic protein-6 induces the expression of inducible nitric oxide synthase in macrophages
.
Immunology
2009
;
128
:
e758
65
.
11.
Kim
IY
,
Lee
DH
,
Lee
DK
,
Kim
BC
,
Kim
HT
,
Leach
FS
, et al
Decreased expression of bone morphogenetic protein (BMP) receptor type II correlates with insensitivity to BMP-6 in human renal cell carcinoma cells
.
Clin Cancer Res
2003
;
9
:
6046
51
.
12.
Kwak
C
,
Park
YH
,
Kim
IY
,
Moon
KC
,
Ku
JH
. 
Expression of bone morphogenetic proteins, the subfamily of the transforming growth factor-beta superfamily, in renal cell carcinoma
.
J Urol
2007
;
178
:
1062
7
.
13.
Blish
KR
,
Wang
W
,
Willingham
MC
,
Du
W
,
Birse
CE
,
Krishnan
SR
, et al
A human bone morphogenetic protein antagonist is down-regulated in renal cancer
.
Mol Biol Cell
2008
;
19
:
457
64
.
14.
Martinez
FO
,
Sica
A
,
Mantovani
A
,
Locati
M
. 
Macrophage activation and polarization
.
Front Biosci
2008
;
13
:
453
61
.
15.
Fukuda
K
,
Kobayashi
A
,
Watabe
K
. 
The role of tumor-associated macrophage in tumor progression
.
Front Biosci
2012
;
4
:
787
98
.
16.
Ruffell
B
,
Affara
NI
,
Coussens
LM
. 
Differential macrophage programming in the tumor microenvironment
.
Trends Immunol
2012
;
33
:
119
26
.
17.
Coussens
LM
,
Werb
Z
. 
Inflammation and cancer
.
Nature
2002
;
420
:
860
7
.
18.
Loberg
RD
,
Ying
C
,
Craig
M
,
Yan
L
,
Snyder
LA
,
Pienta
KJ
. 
CCL2 as an important mediator of prostate cancer growth in vivo through the regulation of macrophage infiltration
.
Neoplasia
2007
;
9
:
556
62
.
19.
Mantovani
A
,
Sozzani
S
,
Locati
M
,
Allavena
P
,
Sica
A
. 
Macrophage polarization: tumor-associated macrophages as a paradigm for polarized M2 mononuclear phagocytes
.
Trends Immunol
2002
;
23
:
549
55
.
20.
Hao
NB
,
Lu
MH
,
Fan
YH
,
Cao
YL
,
Zhang
ZR
,
Yang
SM
. 
Macrophages in tumor microenvironments and the progression of tumors
.
Clin Dev Immunol
2012
;
2012
:
948098
.
21.
Condeelis
J
,
Pollard
JW
. 
Macrophages: obligate partners for tumor cell migration, invasion, and metastasis
.
Cell
2006
;
124
:
263
6
.
22.
Bernert
H
,
Sekikawa
K
,
Radcliffe
RA
,
Iraqi
F
,
You
M
,
Malkinson
AM
. 
Tnfa and Il-10 deficiencies have contrasting effects on lung tumor susceptibility: gender-dependent modulation of IL-10 haploinsufficiency
.
Mol Carcinog
2003
;
38
:
117
23
.
23.
Zeni
E
,
Mazzetti
L
,
Miotto
D
,
Lo Cascio
N
,
Maestrelli
P
,
Querzoli
P
, et al
Macrophage expression of interleukin-10 is a prognostic factor in nonsmall cell lung cancer
.
Eur Respir J
2007
;
30
:
627
32
.
24.
Lv
M
,
Xiaoping
X
,
Cai
H
,
Li
D
,
Wang
J
,
Fu
X
, et al
Cytokines as prognstic tool in breast carcinoma
.
Front Biosci
2011
;
16
:
2515
26
.
25.
Sato
T
,
Terai
M
,
Tamura
Y
,
Alexeev
V
,
Mastrangelo
MJ
,
Selvan
SR
. 
Interleukin 10 in the tumor microenvironment: a target for anticancer immunotherapy
.
Immunol Res
2011
;
51
:
170
82
.
26.
Fortis
C
,
Foppoli
M
,
Gianotti
L
,
Galli
L
,
Citterio
G
,
Consogno
G
, et al
Increased interleukin-10 serum levels in patients with solid tumours
.
Cancer Lett
1996
;
104
:
1
5
.
27.
Mosser
DM
,
Zhang
X
. 
Interleukin-10: new perspectives on an old cytokine
.
Immunol Rev
2008
;
226
:
205
18
.
28.
Daurkin
I
,
Eruslanov
E
,
Stoffs
T
,
Perrin
GQ
,
Algood
C
,
Gilbert
SM
, et al
Tumor-associated macrophages mediate immunosuppression in the renal cancer microenvironment by activating the 15-lipoxygenase-2 pathway
.
Cancer Res
2011
;
71
:
6400
9
.
29.
Lee
GT
,
Kwon
SJ
,
Lee
JH
,
Jeon
SS
,
Jang
KT
,
Choi
HY
, et al
Induction of interleukin-6 expression by bone morphogenetic protein-6 in macrophages requires both SMAD and p38 signaling pathways
.
J Biol Chem
2010
;
285
:
39401
8
.
30.
Weisser
SB
,
van Rooijen
N
,
Sly
LM
. 
Depletion and reconstitution of macrophages in mice
.
J Vis Exp
2012; 66
:
4105
.
31.
Cocolakis
E
,
Dai
M
,
Drevet
L
,
Ho
J
,
Haines
E
,
Ali
S
, et al
Smad signaling antagonizes STAT5-mediated gene transcription and mammary epithelial cell differentiation
.
J Biol Chem
2008
;
283
:
1293
307
.
32.
Ziegler-Heitbrock
L
,
Lotzerich
M
,
Schaefer
A
,
Werner
T
,
Frankenberger
M
,
Benkhart
E
. 
IFN-alpha induces the human IL-10 gene by recruiting both IFN regulatory factor 1 and Stat3
.
J Immunol
2003
;
171
:
285
90
.
33.
Iwamaru
A
,
Szymanski
S
,
Iwado
E
,
Aoki
H
,
Yokoyama
T
,
Fokt
I
, et al
A novel inhibitor of the STAT3 pathway induces apoptosis in malignant glioma cells both in vitro and in vivo
.
Oncogene
2007
;
26
:
2435
44
.
34.
Lee
GT
,
Jung
YS
,
Lee
JH
,
Kim
WJ
,
Kim
IY
. 
Bone morphogenetic protein 6-induced interleukin-1beta expression in macrophages requires PU.1/Smad1 interaction
.
Mol Immunol
2011
;
48
:
1540
7
.
35.
Akdis
CA
,
Blaser
K
. 
Mechanisms of interleukin-10-mediated immune suppression
.
Immunology
2001
;
103
:
131
6
.
36.
Moore
KW
,
de Waal Malefyt
R
,
Coffman
RL
,
O'Garra
A
. 
Interleukin-10 and the interleukin-10 receptor
.
Annu Rev Immunol
2001
;
19
:
683
765
.
37.
Miyazono
K
,
Kamiya
Y
,
Morikawa
M
. 
Bone morphogenetic protein receptors and signal transduction
.
J Biochem
2010
;
147
:
35
51
.
38.
Xia
Y
,
Yu
PB
,
Sidis
Y
,
Beppu
H
,
Bloch
KD
,
Schneyer
AL
, et al
Repulsive guidance molecule RGMa alters utilization of bone morphogenetic protein (BMP) type II receptors by BMP2 and BMP4
.
J Biol Chem
2007
;
282
:
18129
40
.
39.
Sredni
B
. 
Immunomodulating tellurium compounds as anti-cancer agents
.
Semin Cancer Biol
2012
;
22
:
60
9
.
40.
Abe
H
,
Yamanishi
T
,
Mashidori
T
,
Arai
K
,
Kamai
T
. 
Significant association of interleukin 10 receptor mRNA levels with renal cell carcinoma metastasis
.
Biomed Res
2008
;
29
:
19
25
.