M1-like macrophages change tumor blood vessels and microenvironment in murine melanoma

Abstract

Tumor-associated macrophages (TAMs) play a significant role in at least two key processes underlying neoplastic progression: angiogenesis and immune surveillance. TAMs phenotypic changes play important role in tumor vessel abnormalization/ normalization. M2-like TAMs stimulate immunosuppression and formation of defective tumor blood vessels leading to tumor progression. In contrast M1-like TAMs trigger immune response and normalize irregular tumor vascular network which should sensitize cancer cells to chemo- and radiotherapy and lead to tumor growth regression. Here, we demonstrated that combination of endoglin-based DNA vaccine with interleukin 12 repolarizes TAMs from tumor growth-promoting M2-like phenotype to tumor growth-inhibiting M1-like phenotype. Combined therapy enhances tumor infiltration by CD4+, CD8+ lymphocytes and NK cells. Depletion of TAMs as well as CD8+ lymphocytes and NK cells, but not CD4+ lymphocytes, reduces the effect of combined therapy. Furthermore, combined therapy improves tumor vessel maturation, perfusion and reduces hypoxia. It caused that suboptimal doses of doxorubicin reduced the growth of tumors in mice treated with combined therapy. To summarize, combination of antiangiogenic drug and immunostimulatory agent repolarizes TAMs phenotype from M2-like (pro-tumor) into M1-like (anti-tumor) which affects the structure of tumor blood vessels, improves the effect of chemotherapy and leads to tumor growth regression.

Fig 1. Inhibition of B16-F10 tumor growth in response to combined therapy involving endoglin-based DNA vaccine and IL-12.

Mice (n = 10) were inoculated with B16-F10 cells (1x10⁵/animal) and then, on days 1, 8 and 15, SL7207/mCD105 bacteria were administered orally (1x10⁸ cfu/animal, in 100 μL PBS¯). Additionally, on days 9, 11 and 13 following inoculation with cancer cells, pBCMGSNeo/mIL-12 plasmid was injected directly into tumors (20μg DNA per 100 μL PBS¯). “Control” mice received PBS¯ only. Combined therapy was highly effective in inhibiting tumor growth compared to control. *P <0.001, the Cochran’s C test; **P <0.001, the Mann-Whitney U test. (A, B). Photographs were taken on 20^th day of therapy (C). Mice were sacrificed on 20^th day and tumor material was collected for H&E staining. Less necrotic areas (red arrows) and increased tumor infiltration by immune cells (black arrows) was observed in tumor sections from mice treated with combined therapy. Magnification 20× (D).

Fig 2. Effect of combined therapy (ENG vaccine + IL-12) on TAMs accumulation.

One day after inoculating mice with B16-F10 melanoma cells, oral administration of endoglin-based DNA vaccine was initiated. ENG vaccine was administered three times, 1 week apart. Additionally, on days 9, 11 and 13 following inoculation with cancer cells, IL-12 was injected directly into tumors (see Materials and methods). On 20^th day of therapy, mice were sacrificed and tumors were excised for staining with antibodies against CD206 and F4/80. TAM macrophages (F4/80⁺) were located mainly on the peripheral zone in tumors of control mice and in central area of tumors of treated mice. Whereas M2-like macrophages (F4/80⁺/ CD206⁺) both in control and treated mice were present in necrotic areas of central zone and on the periphery of tumors. The area of TAM and M2-like TAM macrophages were determined from six tumors per group, in each tumor 8–12 visual field were analyzed (lens magnification: 20×). *P<0.0001, the Mann-Whitney U test.

Fig 3. Effect of combined therapy (ENG vaccine + IL-12) on the myeloid cell phenotypes.

One day after challenge with B16-F10 cells, oral administration of the ENG vaccine was initiated. ENG vaccine was administered three times, 1 week apart. Additionally, on days 9, 11 and 13 following inoculation with cancer cells, IL-12 was injected directly into tumors (Materials and methods). On the 20^th day of the experiment, tumors (n = 10) were excised. Obtained single-cell suspensions were then used to quantitate TAMs, TANs and DCs levels. Combined therapy increased macrophages (TAMs, defined as DAPI¯/ CD11b⁺/ CD11c¯/ Ly6-G¯), neutrophils (TANs, defined as DAPI‾/ CD11b⁺/ CD11c‾/ Ly-6G⁺) and dendritic cells (DCs, defined as DAPI¯/ CD11b⁺/ CD11c⁺/ Ly6-G¯) levels compared with control tumors. Furthermore, the ratio of M1-like (anti-tumor; defined as DAPI¯/ CD11b⁺/ F4/80⁺/ CD206¯) TAMs to M2-like (pro-tumor, defined as DAPI¯/ CD11b⁺/ F4/80⁺/ CD206⁺) TAMs was more than three times increased. *P<0.001, the Mann-Whitney U test; **P<0.02, the Student’s t-test; ***P<0.007, the Student’s t-test; ****P<0.007, the Cochran’s C test.

Fig 4. Effect of ENG vaccine in combination with IL-12 on cytokine expression in TAMs.

On 20^th day of the combined therapy, mice were sacrificed and tumors were excised for TAMs FACS-sorting. TAMs (defined as DAPI¯/ CD11b⁺/ F4/80⁺) from 1–7 tumors were pooled as 4–5 samples in each group. Total RNA was extracted from sorted cells and converted to cDNA. Gene transcription in TAMs was analyzed by quantitative real-time PCR. Combined therapy up-regulated most of M1-like proinflammatory genes expression (red bars) and down-regulated most of M2-like anti-inflammatory (green bars) and proangiogenic (yellow bars) genes expression in TAMs as compared with controls. Horizontal dash: the value of 1 in controls.

Fig 5. Depletion of macrophages reduces the effect of combining ENG-based DNA vaccine with IL-12.

One day after inoculating mice (n = 7–8) with B16-F10 cells, the animals were orally vaccinated (three times at one-week intervals). Additionally, on days 9, 11 and 13 following inoculation with cancer cells, IL-12 was injected directly into tumors (Materials and methods). Moreover, liposomes (‘empty’ liposomes (control) or Clodronate liposomes (Clodr.)) were delivered intraperitoneally at a dose of 10 mg/kg and directly into tumors at a dose of 5 mg/kg, 2 times/week. Depletion of TAMs decreased the growth of control tumors. But in mice treated with combined therapy we observed enhanced tumor growth after TAMs depletion. Photographs were taken on 20^th day of the experiment. Tumors (n = 3) were collected 20 days after challenge and stained with antibodies against CD206 and F4/80. Magnification 20×. *P<0.0001, the ANOVA followed by the Tukey’s post hoc test.

Fig 6. Effect of combined therapy (ENG vaccine + IL-12) on tumor-immune cells infiltration.

On day 20^th of the combined therapy, mice (n = 9) were sacrificed and tumors were excised for flow analysis. Obtained single-cell suspensions were then used to quantitate the level of CD4⁺, CD8⁺ lymphocytes and NK cells. The percentage of CD4⁺, CD8⁺ lymphocytes and NK cells was determined in total viable cells (representative flow data for CD4⁺, CD8⁺ lymphocytes and NK cell populations). Combined therapy increased tumor-infiltrating CD4⁺ (more than three times), CD8⁺ lymphocytes (more than eight times) as well as NK cells (more than three times) levels compared with control tumors. *P<0.0001, the Cochran’s C test; **P<0.0001, the Student’s t-test.

Fig 7. Depletion of CD8⁺ lymphocytes and NK cells, but not CD4⁺ lymphocytes, reduces the effect of combining ENG-based DNA vaccine with IL-12.

One day after inoculating mice (n = 6–8) with B16-F10 cells, the animals were orally vaccinated (three times at one-week intervals). Additionally, on days 9, 11 and 13 following inoculation with cancer cells, IL-12 was injected directly into tumors (see Materials and methods). Moreover, monoclonal antibodies (anti-CD8a, anti-CD4 lymphocytes or anti-NK cells) were injected intraperitoneally at a dose of 200 μg per mouse on days -1 (1 day before the first vaccination), 1, 6, 11 and 16. Blood was collected on 10^th day after B16-F10 inoculation for flow analysis. Tumors and spleens were harvested 20 days after challenge and stained with antibodies against CD4, CD8 lymphocytes and NK cells. The percentage of CD4⁺, CD8⁺ lymphocytes and NK cells was determined in total viable lymphoid CD45⁺ cells. Representative flow data for CD4⁺, CD8⁺ lymphocytes and NK cell populations after antibody administration (A). Antibody selectively depleted cells in peripheral blood, spleens and tumors. Depletion of CD8⁺ lymphocytes and NK cells enhanced the growth of tumors in treated mice (more than 3 and 7 times, respectively). But after CD4⁺ lymphocytes depletion we observed more than 3 times decreased tumor growth in mice treated with combined therapy. Photographs were taken on 19^th day of the experiment (B). Magnification 20×. *P<0.0001 compared with control (PBS¯) group, **P<0.05 compared with NK depletion group, ***P<0.01 compared with NK depletion group; the Kruskal-Wallis followed by the post hoc multiple comparisons of rank sums test.

Fig 8. Depletion of TAMs changes accumulation of CD8⁺ T cells in tumors.

One day after inoculating mice with B16-F10 cells, the animals were orally vaccinated (three times at one-week intervals). Additionally, on days 9, 11 and 13 following inoculation with cancer cells, IL-12 was injected directly into tumors. Moreover, liposomes (‘empty’ liposomes (control) or Clodronate liposomes (Clodr.)) were delivered intraperitoneally (see Materials and methods). Depletion of TAMs decreased recruitment of CD8⁺ T cells to control mice tumors, but in tumors of mice treated with combined therapy we observed over 75% decrease in CD8⁺ T cells infiltration. Tumors (n = 5) were collected 20 days after challenge and stained with antibody against CD8a. Magnification 20×. *P<0.005 compared with control (PBS¯) group, **P<0.00001 compared with control (PBS¯), Clodr. and ENG+IL-12 + Clodr. groups ***P<0.00001 compared with ENG+IL-12 group; the Kruskal-Wallis followed by the post hoc multiple comparisons of rank sums test.

Fig 9. Effect of combined therapy (ENG vaccine + IL-12) on tumor blood vessels.

On 20^th day of the combined therapy, mice were sacrificed and tumors were excised for immunohistochemical staining. “Normalized” vessels were identified using several tests [35]: (A) α-SMA and CD31 staining was used to identify pericytes-covered tumor vessels (α-SMA⁺CD31⁺ vessels, percentage of CD31⁺ vessels; n = 6; 10 visual fields per tumor section; magnification 20×; *P <0.001, the Cochran’s C test). (B) Staining with pimonidazole (PIMO) was conducted to visualize hypoxic regions in tumors (PIMO⁺ area (% of tumor area); n = 5; magnification 4×; *P <0.001, the Student’s test). (C) Caspase-3 staining was used to identify cancer cells that undergo apoptosis during vasculature “normalization” (apoptotic index: caspase-3⁺/ total cells; n = 6; 10 visual fields per tumor section; magnification 40×; *P <0.001, the Mann-Whitney U test). (D) Lectin perfusion test was used to assess vessel permeability in tumors (lectin⁺CD31⁺ vessels (% of CD31⁺ vessels); n = 6; magnification 20×; *P <0.001, the U Manna-Whitneya test). The structure of tumor vessels in mice treated with combined therapy resembles a regular one: with open lumens, the walls are thick, with a fine layer of pericytes (αSMA) adjacent to their surface (A). Smaller areas of hypoxia (B) and lower number of cells undergoing apoptosis were also found in tumors of treated mice (C). Increased number of perfused blood vessels (lectin⁺CD31⁺) were observed in tumor sections from mice treated with combined therapy (D). This indicated that the B16-F10 tumor vasculature in ENG vaccine with IL-12- treated mice is mature and functional.

Fig 10. Inhibition of B16-F10 tumor growth in response to combination therapy involving endoglin-based DNA vaccine, IL-12 and chemotherapy.

One day after inoculating mice (n = 6–7) with B16-F10 cells, the animals were orally vaccinated (three times at one-week intervals). Additionally, on days 9, 11 and 13 following inoculation with cancer cells, IL-12 was injected directly into tumors (Materials and methods). Moreover, doxorubicin (Doxo) was delivered intraperitoneally at a dose of 2.5 mg/kg, 3 times/week. “Control” mice received PBS¯ only. The suboptimal doses of the cytotoxic agent doxorubicin inhibited the growth of tumors in mice treated with combined therapy, but only slightly inhibited the tumor growth in controls. * P<0.01, the ANOVA followed by the Tukey’s post hoc test.

Fig 11. Repolarization of tumor-associated macrophages by endoglin-based DNA vaccine and IL-12.

Progression of tumor strongly depends on the tumor microenvironment [–7]. Cells that form tumor milieu, especially tumor-associated macrophages (TAMs), play a significant role in at least two key processes underlying neoplastic progression: angiogenesis and immune surveillance [5,9,10]. The structure of tumor blood vessels is defective and they are functionally abnormal [,–19]. Slowed-down circulation of blood leads to underoxygenation. Hypoxia stimulates formation of novel microvessels and invasiveness of tumor cells [6,20]. TAMs phenotypic changes play an important role in tumor vessel abnormalization/ normalization. M2-like macrophages stimulate immunosuppression and formation of defective tumor blood vessels and tumor progression. In contrast M1-like macrophages can activate immune response and cause normalization of irregular tumor vascular network which should sensitize cancer cells to chemo- and radiotherapy and lead to tumor growth regression [,–38]. Combination of antiangiogenic drug and immunostimulatory agent like the endoglin-based DNA vaccine with IL-12 repolarizes TAMs phenotype from M2-like (tumor growth-promoting) into M1-like (tumor growth-inhibiting) which affects the structure of tumor blood vessels and tumor regression.