Figure 1.
Highly metastatic LNM35 tumors are highly angiogenic and lymphangiogenic and rich in macrophages.
(A) Comparison of tumor growth rates between N15 and LNM35 cells inoculated subcutaneously at day 0. Tumor growth rates were significantly different (**p<0.01, n = 6). (B) IHC analysis using antibodies against vascular endothelium (CD31) and lymphatic vessels (LYVE-1), a macrophage-specific antibody (F4/80), and a neutrophil-specific antibody (Gr-1). N15 and LNM35 tumors were quantitatively analyzed, scoring five areas in each tumor section for microvascular density (MVD), lymphatic vessel density (LVD), F4/80-positive cells, and Gr-1-positive cells; *p<0.05 and ** p<0.01. (C) Comparison of lymph node weights between N15 and LNM35 tumors on day 35 after subcutaneous inoculation (**p<0.01, n = 6). (D) H&E staining of lymph node samples from mice subcutaneously injected with N15 and LNM35 cells (n = 2 each); T: tumor, LN: lymph node. (E) Comparison of tumor growth rates following the subcutaneous inoculation of N15 and LNM35 cells at day 0. N15 tumors at day 38 and LNM35 tumors at day19 were of similar size and were further analyzed. (F) IHC analysis using antibodies against vascular endothelium (CD31), lymphatic vessels (LYVE-1), macrophages (F4/80), and neutrophils (Gr-1). Five N15 and LNM35 tumors were quantitatively analyzed, scoring five areas in each tumor section for MVD, LVD, F4/80-positive cells, and Gr-1-positive cells. ** p<0.01.
Figure 2.
Enhanced expression of IL-1s, CXCL8/IL-8 and VEGFs in highly metastatic tumor.
(A) The relative expression of IL-1α, VEGF-A, VEGF-C, MCP-1/CCL2, Groα/CXCL1, ENA-78/CXCL5, IL-8/CXCL8, and IL-6 was determined by ELISA of the culture medium of N15 and LNM35 cells. *p<0.05 and **p<0.01 indicate significant differences between N15 and LNM35 cells (triplicate cultures). The data were normalized to 1.0 based on N15 expression. (B) Western blots showing the levels of IL-1α and IL-1RI expression in LNM35 and N15 cells. (C) Comparison of human and mouse IL-1α and IL-1β protein and/or mRNA expression in N15 and LNM35 tumors. Protein expression was determined by ELISA, and mRNA levels by qRT-PCR. Six tumors for each cancer cell line were analyzed on day 35; **p<0.01. (D) Human CXCL8/IL-8 protein expression in N15 and LNM35 tumors was compared by ELISA in six tumors each on day 35; *p<0.05. (E) Mouse VEGF-A, VEGF-C, and VEGF-D mRNA expression in N15 and LNM35 tumors was compared by qRT-PCR in six tumors each on day 35; *p<0.05. (F) IHC analysis using the macrophage-specific antibody (F4/80) and human and mouse VEGF-A- or VEGF-C-specific antibodies. F4/80+ macrophages (green) colocalized with VEGF-A (red) (upper panel) and VEGF-C (red) (lower panel), as shown in yellow (merged). (G) Determination of M1-type and M2-type macrophages in tumors. VEGF-A, VEGF-C, IL-12, iNOS, IL-10, and arginase expression was determined by qRT-PCR of purified TAMs derived from both tumors. ** p<0.01.
Figure 3.
Effect of macrophage depletion on tumor growth, lymph node metastasis, and VEGF-A and VEGF-C expression.
(A) Anti-tumor effect of Cl2MDP-LIP on tumor growth by LNM35 xenografts. Mice were subcutaneously inoculated with LNM35 cells at day 0, and tumor growth was followed until day 35 in animals intravenously injected twice weekly with PBS-LIP or Cl2MDP-LIP. *p<0.05 between PBS-LIP and Cl2MDP-LIP-treated groups (n = 5 mice per group). (B) Inhibitory effect of Cl2MDP-LIP on lymph node metastasis. The area occupied by cancer cells in the lymph node was determined by H&E staining. Relative tumor area is expressed as the percent of the total lymph node area (n = 5 mice per group); **p<0.01 compared with the PBS-LIP-treated group. (C) Effect of Cl2MDP-LIP on angiogenesis, lymphangiogenesis, and macrophage and neutrophil infiltration by LNM35 tumors. IHC analysis was performed on day 35 using specific antibodies for vascular endothelium (CD31), lymphatic vessels (LYVE-1), infiltrated macrophages (F4/80), and infiltrated neutrophils (Gr-1). Five areas of each tumor section from five tumor samples were quantitatively analyzed; **p<0.01. (D) Effect of Cl2MDP-LIP on the expression of mouse VEGF-A, VEGF-C, and VEGF-D mRNA in LNM35 tumors, determined by qRT-PCR analysis of five tumors on day 35; *p<0.05 compared with the PBS-LIP-treated group.
Figure 4.
Effect of IL-1Ra on lymphangiogenesis and the accumulation of M2-type macrophages by Matrigel plug assay.
(A) Matrigel plugs containing N15 and LNM35 with or without anakinra (n = 4 each). (B) Tumor angiogenesis, lymphangiogenesis, and infiltrated macrophages in Matrigel plugs (n = 6 per group) were determined immunohistochemically in frozen sections using CD31 (red), LYVE-1 (red), and F4/80 (green) as specific markers for microvessels, lymphatic vessels, and infiltrated macrophages, respectively. (C) VEGF-A and VEGF-C expression in macrophages purified from Matrigel plugs was determined by qRT-PCR. *p<0.05 and ** p<0.01. (D) Expression of M1- (iNOS and IL-12) and M2-type (arginase and IL-10) specific biomarkers in macrophages purified from Matrigel plugs was determined by qRT-PCR. ** p<0.01.
Figure 5.
Effect of IL-1Ra on VEGFs expression by macrophages when co-cultured with cancer cells.
(A) Effect of anakinra on the migration of macrophages co-cultured with N15 or LNM35 cells; *p<0.05, triplicate samples. (B) Effect of anakinra on Groα/CXCL1, ENA-78/CXCL5, and IL-8/CXCL8 expression in N15 and LNM35 cells as determined by ELISA; *p<0.05 and **p<0.01, triplicate cultures. Effect of a CXCR2 antagonist (C) and a CXCR2 neutralizing antibody (D) on macrophage migration induced by highly metastatic LNM35 cells; *p<0.05 and **p<0.01, triplicate cultures. (E) Expression of mouse VEGF-A, VEGF-C, and VEGF-D in macrophages co-cultured directly or indirectly with LNM35 cells. In indirect co-cultures, mouse macrophages were placed in the lower chamber with or without cancer cells in the upper chamber. After 4 days, mouse VEGF-A, VEGF-C, and VEGF-D mRNA expression was analyzed by qRT-PCR; *p<0.05 and **p<0.01, triplicate cultures. (F) Effect of anakinra on mouse VEGF-A, VEGF-C, and VEGF-D expression in macrophages directly co-cultured with LNM35 cells. Mouse macrophages were co-cultured with or without cancer cells for 4 days in the presence or absence of 50 ng anakinra/mL, followed by the analysis of mouse VEGF-A, VEGF-C, and VEGF-D mRNA expression by qRT-PCR; **p<0.01, triplicate cultures.
Figure 6.
Effect of IL-1Ra on tumor growth and lymph node metastasis by highly metastatic cancer cells.
(A) Anti-tumor effect of anakinra on the growth of LNM35 xenografts. LNM35 cells were subcutaneously inoculated at day 0, after which tumor growth with or without a daily subcutaneous injection of anakinra was followed until day 35; *p<0.05 between non-treated and treated groups (n = 5 mice per group). (B) Inhibition of lymph node metastasis by anakinra. Three metastatic lymph nodes representative of each group at day 35 are shown. (C) Effects of anakinra on angiogenesis, lymphangiogenesis, and macrophage and neutrophil infiltration by LNM35 tumors, analyzed on day 35 by IHC staining using specific antibodies for vascular endothelium (CD31), lymphatic vessels (LYVE-1), infiltrated macrophages (F4/80), and infiltrated neutrophils (Gr-1). Five areas of each tumor section from five tumor samples were quantitatively analyzed; *p<0.05 and **p<0.01. (D) Effect of anakinra on mouse VEGF-A, VEGF-C, and VEGF-D mRNA levels in LNM35 tumors, determined by qRT-PCR analysis of five tumors at day 35; *p<0.05 and **p<0.01. (E) Effect of anakinra on human CXCL8/IL-8 protein levels in LNM35 tumors, determined by ELISA analysis of five tumors at day 35; *p<0.05. (F) Inhibitory effect of human COX2 expression in anakinra-treated tumors. COX2 expression in two representative tumors from mice treated or not with anakinra (5 mg) was analyzed by western blotting.
Figure 7.
Our hypothetical model how macrophages and cancer cells mutually promote lymph node metastasis through IL-1/IL-1R.
Lung cancer cells form lymph node metastases, promote the development of lymphatic vessels, exhibit invasive tumor growth, and constitutively express IL-1 and CXC chemokines. Highly metastatic cancer cells intrinsically produce larger amounts of IL-1 and CXC chemokines. IL-1α and, to a lesser extent, IL-1β induce the up-regulation of CXC chemokines, which in turn promotes the recruitment and activation of macrophage in the tumor microenvironment. Direct interaction of these cancer cells with these macrophages markedly enhances the production of potent angiogenic and lymphangiogenic factors, leading to further malignant progression including lymph node metastasis and lymphangiogenesis. This sequence of events is suppressed by the IL-1R antagonist anakinra.