"Re-educating" tumor-associated macrophages by targeting NF-kappaB

Abstract

The nuclear factor kappaB (NF-kappaB) signaling pathway is important in cancer-related inflammation and malignant progression. Here, we describe a new role for NF-kappaB in cancer in maintaining the immunosuppressive phenotype of tumor-associated macrophages (TAMs). We show that macrophages are polarized via interleukin (IL)-1R and MyD88 to an immunosuppressive "alternative" phenotype that requires IkappaB kinase beta-mediated NF-kappaB activation. When NF-kappaB signaling is inhibited specifically in TAMs, they become cytotoxic to tumor cells and switch to a "classically" activated phenotype; IL-12(high), major histocompatibility complex II(high), but IL-10(low) and arginase-1(low). Targeting NF-kappaB signaling in TAMs also promotes regression of advanced tumors in vivo by induction of macrophage tumoricidal activity and activation of antitumor activity through IL-12-dependent NK cell recruitment. We provide a rationale for manipulating the phenotype of the abundant macrophage population already located within the tumor microenvironment; the potential to "re-educate" the tumor-promoting macrophage population may prove an effective and novel therapeutic approach for cancer that complements existing therapies.

Figure 1.

IKKβ inhibition increases macrophage tumoricidal activity. (A) WT, MyD88^−/−, TLR-2^−/−, and TLR-4^−/− or mock, IKKβ^DN or IKKβ^f/f, and IKKβ^Δ BMDMs were co-cultured with ID8-Luc in a modified Boyden chamber without direct cell–cell contact for 72 h. In addition, CD11b⁺-selected TAMs from the ascites of ID8 tumor–bearing mice were mock (TAM_mock) or IKKβ^DN infected (TAM_DN). Invasion of ID8-Luc cells was assessed by luciferase activity in the lower part of the chamber. Macrophages deleted in MyD88, IL-1R, or IKKβ significantly reduce ID8-Luc invasion (P < 0.01; t test with Welch's correction). Data are represented as mean ± SEM of n = 6. Representative data are shown from at least three independent experiments. (B) ID8 cells were co-cultured with IKKβ-targeted BMDMs or TAMs and respective control cells. Fluorescence caspase 3/7 activity was assessed after 0, 3, and 6 h. Co-culture with IKKβ^DN and IKKβ^Δ BMDMs or TAMs expressing IKKβ^DN (TAM_DN) significantly increases caspase 3/7 activity in ID8 cells (P < 0.01; t test with Welch's correction). Data are represented as mean ± SD of n = 6. Representative data are shown from at least three independent experiments. (C) [¹¹¹In]oxine release assay after a 24-h co-culture of ¹¹¹In-labeled ID8 cells with mock, IKKβ^DN, TAM_DN or IKKβ^f/f, and IKKβ^Δ macrophages. ¹¹¹In-release in cell-free culture supernatants was measured after 24 h with a scintillation counter. IKKβ-targeted macrophages promote increased cytotoxicity. Data are represented as mean of n = 3. Representative data are shown from at least three independent experiments.

Figure 2.

IKKβ maintains the alternative phenotype of tumor-polarized macrophages. WT, mock, IKKβ^DN or IKKβ^f/f, and IKKβ^Δ BMDMs were co-cultured with ID8 cells for 24 h. Supernatant was collected after 24 h and analyzed for IL-10, IL-12p70, TNF-α, and NO production by ELISA. (A) Co-cultured WT macrophages express an immunosuppressive IL-10^high IL-12p70^low TNF-α^high profile. However, supernatant from co-cultured IKKβ-deleted macrophages show a significant decrease in IL-10 and TNF-α (P < 0.01; t test with Welch's correction) but increase in IL-12p70 (P < 0.01; t test with Welch's correction). Data are represented as mean ± SD of n = 3. Representative data are shown from at least three independent experiments. (B) In parallel, total RNA was isolated for real-time PCR analysis of IL-12p40 and arginase-1 in co-cultured macrophages. Targeting IKKβ in macrophages switches their phenotype to an IL-12^high arginase-1^low NOS2^high profile, consistent with an M1 phenotype. In_addition, IKKβ-targeted macrophages secrete significantly higher levels of NO, measured as nitrite by Griess assay (P < 0.01; Mann-Whitney test). Data are represented as mean ± SD of n = 3. Representative data are shown from at least four independent experiments. (C) Macrophages were co-cultured with ID8 cells, and protein extracts were prepared at the indicated time points for biochemical analysis. Expression of NF-κB p65, serine 536 phosphorylation of p65, IKKβ, Stat1, tyrosine 701 phosphorylation of Stat1 (Tyr701), and NOS2 was measured by immunoblot analysis of cell lysates using β-actin as a loading control. Representative data are shown from at least three independent experiments. (D) [¹¹¹In]oxine release assay after co-culture of ID8 ovarian cancer cells with IKKβ^DN or IKKβ^Δ BMDMs. IKKβ-targeted macrophages promote increased cytotoxicity that was reversed by the addition of the selective NOS2 inhibitor 1400W. Data are represented as mean of n = 3. Representative data are shown from at least three independent experiments.

Figure 3.

Adoptive transfer of IKKβ-targeted BMDMs inhibits tumor growth in vivo. (A) Representative bioluminescence imaging in vivo 14 d after adoptive transfer of BMDMs (n = 6; three independent experiments). ID8-Luc were injected i.p. into syngeneic mice and tumors were allowed to develop. After 35 d, WT, mock, and IKKβ^DN BMDMs were adoptively transferred and peritoneal cells were collected after an additional 7 and 14 d. Bioluminescence is presented as a pseudocolor scale: red, the highest photon flux; blue, the lowest photon flux. (B) Quantification of bioluminescence from primary tumors (n = 6 each) obtained on days 0, 7, and 14. Adoptively transferred IKKβ^DN BMDMs significantly reduced ID8 growth on day 14 (P < 0.05; t test with Welch's correction). (C) NO secretion into the peritoneal cavity. IKKβ^DN BMDMs increased NO release 4 h after adoptive transfer (P < 0.05; t test with Welch's correction). Data are represented as mean ± SD of n = 6. Representative data are shown from at least two independent experiments. (D) Cytokine profile in tumor ascites after 14 d. Ascites from mice treated with IKKβ-targeted BMDMs contained significantly lower amounts of IL-10 and TNF-α but higher amounts of IL-12 (P < 0.01; Mann-Whitney test). Total RNA was isolated from CD11b⁺-selected ascitic macrophages for real-time PCR analysis of IL-12p40 and arginase-1 expression. Data are represented as fold induction of mRNA expression compared with WT macrophages (n = 6). (E) MHC class II expression in the ascitic CD11b⁺ myeloid cell population measured by FACS (percentage of positive cells indicated). Representative histograms are shown from n = 6.

Figure 4.

Adoptive transfer of IKKβ-targeted TAMs inhibits tumor growth in vivo. (A) Representative bioluminescence imaging in vivo 14 d after adoptive transfer of TAMs isolated from established ID8 tumors (n = 6; three independent experiments). ID8-Luc were injected i.p. into syngeneic mice and tumors were allowed to develop. After 35 d, mock- and IKKβ^DN-infected TAMs (TAM_mock, TAM_DN, respectively) were adoptively transferred and peritoneal cells were collected after an additional 7 and 14 d. Bioluminescence is presented as a pseudocolor scale: red, the highest photon flux; blue, the lowest photon flux. (B) Quantification of bioluminescence from primary tumors (n = 6 each) obtained on days 0, 7, and 14. Adoptive transfer of TAM_DN significantly reduced ID8 growth on day 14 (P < 0.05; t test with Welch's correction). (C) NO secretion into the peritoneal cavity. Transfer of TAM_DN significantly increased NO production 4 h after adoptive transfer (P < 0.05; t test with Welch's correction). Data are represented as mean ± SD of n = 6. Representative data are shown from at least two independent experiments. (D) Cytokine profile in tumor ascites after 14 d. Ascitic fluid after transfer of TAM_DN contained significantly lower amounts of IL-10 and TNF-α but higher amounts of IL-12 (P < 0.01; Mann-Whitney test). Total RNA was isolated from CD11b⁺-selected TAMs for real-time PCR analysis of IL-12p40 and arginase-1 expression. Data are represented as fold induction of mRNA expression compared with TAM_mock-treated mice (n = 6). (E) MHC class II expression in the ascitic CD11b⁺ TAMs measured by FACS (percentage of positive cells indicated). Representative histograms are shown from n = 6.

Figure 5.

Increased IL-12–dependent NK cell recruitment. (A) Neutralizing IL-12p40 but not IL-23p19 rescued the antitumor effect of adoptively transferred IKKβ^DN BMDMs (P < 0.01; Mann-Whitney test). Data are represented as mean ± SEM of n = 6. Representative data are shown from two independent experiments. (B) Representative bioluminescence image of the IL-12p40–neutralizing experiment (n = 6). (C) Adoptive transfer of IKKβ^DN BMDMs increases NK cell recruitment in ID8 tumors (P < 0.01; Mann-Whitney test). Data are represented as mean ± SD of n = 6. Representative data are shown from two independent experiments. (D) Ex vivo NK cell cytotoxicity assay. Splenic DX5⁺-enriched NK cells were stimulated with ascites from tumor-bearing mice, which had either no treatment or were treated with adoptive transfer of mock or IKKβ^DN BMDMs. Ascites from mice treated by adoptive transfer of IKKβ^DN BMDMs led to a significant increase in NK cell–mediated tumor cell cytotoxicity (P < 0.01; t test with Welch's correction). The addition of a neutralizing antibody against IL-12p40 but not IL-23p19 or the respective control antibody rescued the increased tumoricidal effect. Data are represented as mean of n = 3. Representative data are shown from three independent experiments.