Tumor Microenvironment Modulates Immunological Outcomes of Myeloid Cells with mTORC1 Disruption

Abstract

The role of mTOR signaling pathway in different myeloid cell subsets is poorly understood in the context of tumor. In this study, myeloid cell-specific Raptor KO mice were used to determine the roles of mTOR complex 1 (mTORC1) in regulating macrophage function from Lewis lung carcinoma (LLC) subcutaneous (s.c.) tumors and lung tumor metastasis. We found no difference of tumor growth between conditional Raptor KO (cKO) and control mice in the s.c. tumor models although depletion of mTORC1 decreased the immunosuppressive function of TAM. Despite the decreased immunosuppressive activity of TAM, M1-like TAM differentiation was impaired in the s.c. tumor microenvironment (TME) of mTORC1 cKO mice due to downregulated CD115 expression on macrophages. In addition, TNF-α production by mTORC1-deficient myeloid cells was also decreased in the s.c. LLC tumors. On the contrary, disruption of mTORC1 in myeloid cells promoted lung cancer metastasis. Accordingly, immunosuppressive interstitial macrophages/metastasis-associated macrophages (IM/MAM, CD11b⁺F4/80^high) were accumulated in the lungs of Raptor KO mice in LLC lung metastasis model, leading to decreased Th1 responses. Taken together, our results demonstrate that differential TME dictates the immunological outcomes of myeloid cells with mTORC1 disruption leading to different tumor growth phenotypes.

Introduction

It has now become clear that the inflammatory milieu of the tumor microenvironment (TME) plays important roles in regulating cancer progression, metastasis and therapies (1, 2). Tumor-associated macrophages (TAM) are one of the most abundant inflammatory cells in the TME. The roles of TAM in tumor progression, angiogenesis, metastasis and immunosuppression have been well established (3). TAM exhibit predominantly M2-like pro-tumor and immunosuppressive phenotype, particularly in the late stages of cancer. Therefore, immunosuppressive TAM are an important target for cancer treatment (4, 5). However, recent studies have demonstrated that TAM function is more complex due to macrophage heterogeneity (6, 7). It is well known that TAM are mainly differentiated from bone marrow-derived monocytes. However, tissue resident macrophages also contribute to the pool of TAM in tumor-bearing tissues such as lung (8). In addition, the local environmental factors also have a role in regulating TAM function (9, 10).

The mechanistic target of rapamycin complex 1 (mTORC1) is a highly conserved serine–threonine kinase belonging to the phosphatidylinositol kinase-related protein kinases family. mTORC1, which is characterized by the adaptor protein Raptor, phosphorylates and activates S6K and 4E-BP1. The mTOR pathway plays a central role in cellular homeostasis and has been implicated in a number of cellular events including cell growth, survival, and metabolism (11, 12). A growing body of evidence identifies activation of mTOR signaling as a common occurrence in human cancers. Furthermore, oncogenic mTOR signaling recruits myeloid-derived suppressor cells (MDSC) to promote tumor initiation (13). These findings have made mTOR an attractive target for the development of targeted therapies. Several mTORC1 inhibitors have demonstrated strong effects on tumor cell growth and have been approved for treatment in some types of cancer. However, the overall therapeutic efficacy of these mTORC1 inhibitors in cancer is limited (14–16). One of the potential reasons could be due to an immune regulatory function of mTORC1 inhibitor on host cells. In addition, the relative contributions of different TME to the anti-cancer efficacy of mTORC1 inhibitors have not been fully characterized. There are controversies in literature regarding the role of mTOR signaling in regulating the activation of different myeloid cell subsets in response to different environmental factors, particularly in the context of tumor (17–20).

In the present study, we examined the effect of disruption of mTORC1 signaling in myeloid cells on subcutaneous (s.c.) tumor development and lung cancer metastasis. We demonstrated that depletion of mTORC1 signaling in myeloid cells did not delay s.c. tumor progression although polarized M2 macrophages and TAM from s.c tumors displayed decreased expression of Arginase 1 (Arg1) and diminished immunosuppressive activity. The decreased Th1 T cell response in the s.c. TME was also observed in tumor-bearing Raptor cKO mice. This effect was associated with decreased M1-like TAM differentiation and reduced pro-inflammatory cytokine TNF-α production in myeloid cells from mTORC1-deficient TME. Further lung cancer metastasis study showed that disruption of mTORC1 in myeloid cells promoted lung cancer metastasis. The increased accumulation of interstitial macrophages/metastasis-associated macrophages (IM/MAM, CD11b⁺F4/80^high) with enhanced expression of Arg1 was observed in the LLC-bearing lungs of Raptor KO mice. These findings reveal complex roles of mTORC1 signaling in myeloid cells on regulating anti-tumor immunity in different environments. Our data suggest that differential TMEs may dictate the immunological outcomes of myeloid cells with mTORC1 disruption.

Materials and Methods

Mice

LysM-Cre mice and Raptor^flox/flox mice were purchased from The Jackson Laboratory (Bar Harbor, ME, USA) and interbred to generated myeloid cell Raptor knockout mice (LyzM^Cre/+Raptor^fl/fl, referred to as Raptor cKO) and control mice (LyzM^+/+Raptor^fl/fl, referred to as Control). Ovalbumin (OVA) T-cell receptor (TCR) Tg OT-II mice were purchased from The Jackson Laboratory. Rag2 deficient OVA TCR Tg OT-I mice were purchased from Taconic Biosciences (Germantown, NY). All animals were maintained under specific pathogen-free conditions and handled in accordance with the protocols approved by the Institutional Animal Care and Use Committee of the University of Louisville.

Tumor models

LLC (Lewis lung carcinoma), E0771 (murine breast cancer cell line), and LLC-GFP cells were cultured in the completed DMEM medium containing 10% FBS. To establish subcutaneous (s.c.) tumors, 1×10⁶ LLC or E0771 tumor cells were suspended in PBS and inoculated s.c. in the flank of mice. Tumor sizes were measured twice a week with a caliper. Lung cancer metastasis model was established by intravenous injection of LLC-GFP cells (0.5 ×10⁶ in 250 μl PBS) and quantitatively analyzed by measuring percentage of GFP⁺ cells in viable lung cells.

Bone marrow (BM)-derived M2 macrophage differentiation and TAM purification

BM cells were isolated by flushing femurs and tibias with complete medium and treated with ACK lysis buffer. After washing the BM cells were cultured in DMEM complete medium with M-CSF (20 ng/ml). On day 4, the culture medium and non-adherent cells were removed. The adherent cells were then cultured with fresh medium containing M-CSF. On day 7, the cells were cultured in the presence of IL-4 (20 ng/ml) and IL-13 (20 ng/ml) for 48 h to obtain M2-polarized macrophages. For TAM and MAM purification, exercised tumors and tumor-bearing lungs were minced into small pieces and digested with medium containing 0.5 mg/ml collagenase A, 0.2 mg/ml hyaluronidase (type V), and 0.02 mg/ml DNase I for 30 min on a rotating platform. The resulting cell suspensions were filtered, washed with DMEM complete medium. The TAM cells (CD11b⁺Ly6C⁻Ly6G⁻F4/80⁺) from s.c tumors and MAM (CD45⁺CD11b⁺F4/80⁺) from lungs were sorted using BD FACS Aria III. Fixable viability dye was used to exclude dead cells.

Antibodies and flow cytometry

Single-cell suspensions were blocked in the presence of anti-CD16/CD32 at 4°C for 15 min and stained on ice with the appropriate antibodies and isotype controls in PBS containing 1% FBS. The fluorochrome-labeled antibodies against mouse CD11b, Ly6G, Ly6C, F4/80, CD4, CD8, CD11c, CD115, IL-2, IFN-γ, TNF-α, Foxp3 and their corresponding isotype controls were purchased from BioLegend (San Diego, CA). Mouse VEGFR1 APC-conjugated antibody was purchased from R&D Systems (Minneapolis, MN). Fixable Viability Dye eFluor™ 780 was from Thermo Fisher Scientific. For intracellular staining, the cells were fixed and permeabilized following surface staining. The samples were acquired using FACSCalibur or FACSCanto cytometer (BD Bioscience, San Jose, CA) and analyzed using FlowJo software (Tree Star, Ashland, OR).

T cell suppressive assay

Freshly isolated splenocytes from OT-I mice were lysed of red cells and labeled with carboxyfluorescein succinimidyl ester (CFSE, 1μM) for 10 min at 37°C and washed with cold complete medium. M2 macrophages, TAM, or IM/MAM were cultured with 1×10⁵ CFSE-labeled OT-I splenocytes at different ratio in the presence of OVA (25 ug/ml) for 3 days. T cell proliferation was analyzed using flow cytometry by measurement of CFSE dilution.

Western blot analysis and quantitative real-time PCR

BM-derived macrophages were stimulated with 25% conditioned medium (CM) from LLC cells for the indicated times. These in vitro stimulated macrophages and in vitro polarized M2 macrophages were lysed in Triton X-100 lysis buffer in the presence of protease and phosphatase inhibitors. The whole-cell lysates were subjected to SDS-PAGE and electrotransferred to polyvinylidene difluoride membrane. After blocking, the membranes were probed overnight at 4°C with appropriate primary Abs and then secondary Ab. The primary Abs including phospho-p70 S6 kinase, p70 S6 kinase, phospho-4E-BP1, 4E-BP1, phospho-Akt (Ser473), Akt, anti-Arginase-1, and Raptor were from Cell Signaling Technology. The blots were developed with ECL Plus Western blotting detection reagents (GE Healthcare). For qRT-PCR, RNA was extracted with TRIzol reagent (Invitrogen) and transcribed to cDNA with a reverse transcription kit (Bio-Rad). Quantitative real-time PCR reactions were performed using SYBR Green Supermix (Bio-Rad) with the relevant primers. We normalized gene expression level to β2-microglobulin (β-MG) housekeeping gene and represented data as fold differences by the 2^−ΔΔCt method (21). The primer sequences used for real-time PCR were as follows: Arg1: F: 5’-TTTTAGGGTTACGGCCGGTG-3’ and R: 5’- CCTCGAGGCTGTCCTTTTGA-3’; Mgl2: F: 5’-ACCCACCTCCTCCTGTTCTC-3’ and R: 5’- GGGACTGGAATTCAGCCTTT-3’; iNOS: F: 5’-GGGACTGAGCTGTTAGAGACAC-3’ and R: 5’-CCAAATCCAACGTTCTCCGT-3’; VEGFA: F:5’-TTACTGCTGTACCTCCACC-3’ and R: 5’- ACAGGACGGCTTGAAGATG-3’.

Cytokine array and ELISA

TAM were sorted from LLC s.c. tumor tissues and cultured in complete medium for 24 hours. The culture supernatants were collected and cytokine/chemokine profile of culture supernatant was determined using proteome profiler mouse cytokine array kit (R&D Systems). The expression levels of cytokine/chemokine were measured by pixel density of each dot using the ImageJ software. TNF-α ELISA kit was purchased from BioLegend.

Statistical analysis

Data were analyzed using GraphPad Prism5.0 software (GraphPad Software). An unpaired Student t test was used to calculate significance. All graph bars are expressed as mean ± SEM. Significance was assumed to be reached at p < 0.05. The p values were presented as follows: *p<0.05, **p<0.01, and ***p<0.001.

Results

Depletion of Raptor in myeloid cells decreases the immunosuppressive function of polarized M2 macrophages and TAM from s.c. tumors

Previous studies have used rapamycin to examine the role of mTORC1 in regulating macrophage function (22, 23). However, high dose and persistent use of rapamycin can inhibit both mTORC1 and mTORC2 signaling. Thus, the role of mTORC1 in regulating macrophage polarization still needs to be clearly defined. To this end, we crossed Raptor flox mice with LysM-Cre mice. As shown in Fig.1A, depletion of Raptor specifically inhibited the activation of mTORC1, but not mTORC2, in macrophages induced by LLC CM. The phosphorylation of p70 S6K and 4E-BP1, makers of mTORC1 activation, was impaired in Raptor-deficient macrophages, whereas similar activation of mTORC2 target Akt S473 was not affected in Raptor-deficient macrophages. TAM are thought to more closely resemble M2-like macrophages. We thus polarized M2 macrophages from BM. Gene expression analysis revealed that Agr1 and other M2 signature gene Mgl2 were diminished whereas expression of iNOS, a M1-like macrophage marker, showed a trend of increase in Raptor-deficient M2 macrophages (Fig. 1B). Western blot analysis also showed decreased expression of Arginase 1 in Raptor-deficient M2 macrophages (Fig.1C). Consequently, Raptor-deficient M2 macrophages exhibited a decreased immunosuppressive activity on CD8 OT-I T cell proliferation compared to M2 macrophages from control mice (Fig. 1D).

FIGURE 1.

Depletion of Raptor in myeloid cells decreases the immunosuppressive function of M2 macrophages and TAM. (A) Bone marrow-derived macrophages from control and Raptor cKO mice were stimulated with 25% conditioned medium (CM) from LLC cells for the indicated time points. The expression of markers for mTORC1 and mTORC2 were measured by Western blot. Data are representative of two independent experiments. (B) Bone marrow-derived macrophages from control and Raptor KO mice were polarized into M2 macrophages in the presence of IL-4 and IL-13. The gene expression of Arg1, Mgl2, and iNOS in M2 macrophages was analyzed by quantitative RT-PCR. Each dot represents a single mouse. (C) The expression of arginase 1 in M2 macrophages was measured by Western blot. Data represent two independent experiments. (D) Immunosuppression assay was performed by co-culture of M2 macrophages and OT-I splenocytes in the presence of OVA (25 μg/ml) for 3 days. The T cell proliferation was measured by CFSE dilution. (E) The expression of arginase 1 in TAM was measured by Western blot. (F and G) Immunosuppressive function of TAM was determined by co-culture of TAM and OT-I splenocytes in the presence of OVA. The T cell proliferation was measured by CFSE dilution (F). The co-cultured cells were restimulated with PMA and Ionomycin in the presence of brefeldin A for 4–6 hours. The IFN-γ production by CD8⁺ T cells was determined by intracellular cytokine staining followed by flow cytometry analysis (G). Data in F and G represent at least three independent experiments.

We next examined the role of mTORC1 on TAM immunosuppressive function. TAM (CD11b⁺Ly6C⁻Ly6G⁻F4/80⁺) were sorted from LLC s.c. tumors from control and Raptor cKO tumor-bearing mice. Similar to BM polarized M2 macrophages, TAM isolated from LLC s.c. tumors of Raptor cKO mice displayed a significantly decreased expression of Arg1 (Fig. 1E) and reduced immunosuppressive activity on effector T cells (Fig. 1F and G). Taken together, these data suggest that depletion of mTORC1 signaling in myeloid cells impairs M2 macrophage polarization and promotes TAM towards M1-like phenotype with decreased immunosuppressive function in s.c. tumors.

Depletion of Raptor in myeloid cells has no impact on the growth of s.c. tumors

The decreased immunosuppression of TAM is expected to activate anti-tumor T cell responses in vivo and thereby delay the tumor progression. To test this hypothesis and understand the importance of myeloid cell mTORC1 signaling in vivo, LLC lung cancer and E0771 mammary tumor cells were inoculated s.c. into control and Raptor cKO mice, and tumor growth was followed. However, we didn’t observe the protective effect of depletion of Raptor in myeloid cells on the growth of s.c. LLC and E0771 tumors (Fig. 2A).

FIGURE 2.

Depletion of Raptor in myeloid cells has no effect on tumor progression rate in the subcutaneous LLC and E0771 models. (A) The control and Raptor KO mice were subcutaneously injected with 1 × 10⁶ LLC (n = 7–9 mice per group) or E0771 tumor cells (n = 4 mice per group). Tumor diameters were measured twice a week. Data represent two to five independent experiments. (B, C and D) The single-cell suspensions from LLC tumors were stimulated with PMA and Ionomycin in the presence of brefeldin A for 4–6 hours. The IFN-γ production by CD4⁺ T cells (B), IL-2 production by CD4⁺ T cells (C), and IFN-γ production by CD8⁺ T cells (D) were determined by intracellular cytokine staining followed by flow cytometry analysis. Each dot represents a single mouse. Data represent two to five independent experiments.

Next, we examined T cell function in tumors by measuring CD4⁺IFN-γ⁺, CD4⁺IL-2⁺, CD8⁺IFN-γ⁺ cells and regulatory T cells (Treg, CD4⁺Foxp3⁺). Although there were no differences in the percentages of CD4, CD8 T cells and CD4⁺Foxp3⁺ cells in the tumors from control and Raptor-deficient mice (data not shown), we consistently detected significant decrease of CD4⁺IFN-γ⁺, CD4⁺IL-2⁺ and CD8⁺IFN-γ⁺ cells in tumors of Raptor KO mice (Fig. 2B, C and D). These data suggest that the mTORC1 deficiency in myeloid cells impairs antitumor T cell responses despite the fact that mTORC1-deficient TAM from s.c. tumors show reduced T cell immunosuppressive function on a per cell basis.

Depletion of Raptor in myeloid cells decreases the differentiation of inflammatory monocytes into M1-like TAM in the s.c. TME

To understand the underlying potential mechanisms of the above unexpected findings, we examined whether the deficiency of mTORC1 in myeloid cells leads to alterations of myeloid cell composition during s.c. tumor progression. As shown in Fig.3A, the total myeloid cells (CD11b⁺) were decreased in LLC s.c. tumors from Raptor cKO mice. However, the percentages of MDSC, including granulocytic (G-MDSC) and monocytic (M-MDSC) were similar in tumors from control and Raptor cKO mice (Fig. 3B). Notably, the percentage of F4/80⁺ TAM cells in tumor tissues was significantly decreased in Raptor cKO mice (Fig. 3C).

FIGURE 3.

Depletion of Raptor in myeloid cells decreases the differentiation of monocytes into TAM. (A) The percentages of myeloid cell population (CD11b⁺) in tumor tissues from LLC-bearing mice. Each dot represents a single mouse. (B) The percentages of G-MDSC and M-MDSC within CD11b⁺ tumor myeloid cells. (C) The percentages of F4/80⁺ population in tumor tissues from LLC-bearing mice. (D) The subset of inflammatory monocytes (Ly6C^highMHC class II^low) and TAM (Ly6C^lowMHC class II^low) within tumor myeloid cells. The cells were gated on live CD11b⁺ Ly6G⁻ cell population from tumor tissues. The summarized percentage of monocyte versus TAM is shown. Data in A, B, C, and D represent at least three independent experiments. (E) The expression of M-CSFR (CD115) on TAM and bone marrow-derived macrophages (BMDM) was examined by Flow cytometry analysis. Data represent two independent experiments.

Previous studies have shown that inflammatory Ly6C^high monocytes are more efficiently recruited to tumors and function as TAM precursors (24, 25). We next assessed whether mTORC1 signaling affected the differentiation of TAM based on the differential expression of MHC class II and Ly6C (25). More Ly6C^high cells and less TAM (Ly6C⁻ MHC class II⁻) in CD11b⁺Ly6G⁻ population were found in the tumors from Raptor KO mice (Fig. 3D). These data suggest that mTORC1 signaling in myeloid cells is important for the monocytes differentiation into TAM in the s.c tumors. The colony-stimulating factor 1 (CSF1)/CSF1 receptor (CSF1R) axis plays essential roles in TAM differentiation and mTORC1 signaling has been shown to be critical in M-CSF-mediated myelopoiesis (26). Thus, we hypothesized that decreased TAM frequency in mTORC1 cKO s.c tumors may be the result of downregulation of CSF1R in Raptor-deficient macrophages. Indeed, we observed the decreased M-CSFR (CD115) expression on TAM from Raptor cKO tumor-bearing mice. In addition, the expression of CSF1R was also significantly decreased on bone marrow-derived macrophages (BMDM) from Raptor cKO mice (Fig. 3E). Collectively, these results suggest that mTORC1 deficiency in myeloid cells results in defective CSF1/CSF1R signaling pathway which likely contributes to the decreased TAM differentiation and infiltration in the s.c. tumors.

Expression of TNF-α is decreased in Raptor-deficient myeloid cells

Cytokines in the TME can modulate anti-tumor immune response dependent on the balance of pro- and anti-inflammatory cytokines and their relative concentrations. To further dissect the mechanisms underlying impaired anti-tumor immunity in the TME of Raptor-deficient tumors, we examined cytokine/chemokine profile of myeloid cells from LLC s.c. tumors. Cytokine array analysis revealed a decrease in the expression of several cytokines and chemokines, such as IL-1RA, KC, CCL4, and TNF-α, in the culture supernatants of TAM from Raptor cKO mice (Fig. 4A). The decreased expression of TNF-α in culture supernatants of Raptor-deficient TAM was also confirmed by ELISA (Fig. 4B). Flow cytometry analysis revealed that myeloid cells represent a major source of TNF-α in the TME of s.c. LLC tumors and CD11b⁺ cells from Raptor KO mice produced significantly lower levels of TNF-α compared to these from control mice (Fig. 4C). Similar result was also observed in Raptor-deficient TAM (Fig. 4D). To examine the contribution of TNF-α to Th1 T cell response, we performed an in vitro assay by culturing OVA-specific TCR transgenic OT-1 T cells in the presence of antigen OVA with or without exogenous TNF-α. TNF-α indeed promoted Th1 cytokine IFN-γ production by CD8 T cells (Fig. 4E). This result demonstrates an important role for myeloid cell mTORC1 signaling in TNF-α production which is critical for Th1 T cell response in the TME.

FIGURE 4.

Depletion of Raptor in myeloid cells results in decreased TNF-α expression by tumor myeloid cells. (A) The cytokine/chemokine profile of the culture supernatants of TAM was determined by mouse cytokine/chemokine array. The culture supernatants were from freshly sorted and then in vitro cultured TAM (1 ×10⁶ cells in 1 ml medium for 24 hours). The summarized data represent relative expression levels of detected cytokine and chemokine by measuring pixel density. (B) The expression levels of TNF-α in the culture supernatants were measured by ELISA. (C and D) The single-cell suspensions from tumor tissues were stimulated with PMA and Ionomycin in the presence of brefeldin A for 4–6 hours. The TNF-α expression was determined by intracellular cytokine staining followed by Flow cytometry. The percentages of TNF-α-producing myeloid cells in the total tumor cells (C) and TAM population (D) were shown. Each dot represents a single mouse. Data in B, C, and D represent at least three experiments. (E) OT-I splenocytes were cultured in the presence of OVA (25 μg/ml) with or without exogenous mouse TNF-α (5 ng/ml) for 3 days. The IFN-γ producing T cells were determined by intracellular cytokine staining followed by Flow cytometry. The cells were gated on CD8⁺ T cells. Data are representative of two independent experiments.

Depletion of myeloid cell mTORC1 signaling promotes lung cancer metastasis

Previous studies have shown that in addition to cytokines and growth factors, the modulating effects of mTORC1 signaling in macrophages are also regulated by the environmental cues (12, 27). The pulmonary microenvironment represents a unique milieu as evidenced by diverse macrophage populations as well as recruited monocytes including patrolling monocytes and inflammatory monocytes (28-30). Therefore, we sought to determine the roles of myeloid cell mTORC1 signaling in regulating lung cancer metastasis. Experimental lung cancer metastasis was established by injection of LLC-GFP cells into the tail veins of age and sex-matched Raptor control and cKO mice. LLC metastasis was determined by measuring percentage of GFP positive cells within the lungs. Remarkably, LLC lung metastasis was significantly higher in Raptor cKO mice compared to that in control mice (Fig. 5A). Accordingly, Raptor cKO mice had decreased survival compared to control mice (Fig. 5B). T cell analyses revealed the reduced percentages of CD4 and CD8 T cells in Raptor cKO tumor-bearing lungs (Fig. 5C). In addition, IFN-γ and IL-2 production by T cells was also significantly decreased (Fig. 5D, E and F). These findings demonstrate that mTORC1 signaling in myeloid cells is important for the generation and maintenance of effective anti-tumor immunity in mouse metastatic lung cancer.

FIGURE 5.

Depletion of Raptor in myeloid cells promotes lung metastasis. (A) The lung metastasis was determined by measuring percentage of LLC-GFP positive cells in viable lung cells using Flow cytometry analysis. The data are pooled from three experiments (n = 13–16 mice per group). (B) The survival of control and Raptor KO mice (n = 4–5 mice per group) after intravenous injection of 5 × 10⁵ LLC-GFP cells. (C) The percentage of CD4⁺ T cells and CD8⁺ T cells among CD45⁺ leukocytes in LLC tumor-bearing lungs. (D, E and F) The single-cell suspensions from lung tissues of LLC metastasis mice were stimulated with PMA and Ionomycin in the presence of brefeldin A for 4–6 hours. The IFN-γ production by CD4⁺ T cells (D), IL-2 production by CD4⁺ T cells (E), and IFN-γ production by CD8⁺ T cells (F) were determined by intracellular cytokine staining followed by flow cytometry analysis. Each dot represents a single mouse. Data in C, D, E, and F represent two to four independent experiments.

Depletion of myeloid cell mTORC1 signaling enhances the accumulation of lung CD11b⁺F4/80^high cells in metastatic tumor-bearing mice

Next we investigated whether the lack of mTORC1 signaling in myeloid cells alters the pulmonary myeloid cells, leading to a more permissive environment that promotes LLC metastasis. Lung patrolling monocytes are important in controlling lung cancer metastasis (28). Lung macrophages are highly heterogeneous populations including alveolar macrophages (AM) and interstitial macrophages (IM). Previously, it has been shown an essential role of mTORC1 for AM homeostasis by regulating proliferative renewal (31), and contribution of AM to the premetastatic niche in a model of breast cancer (30). Thus, we assessed the percentages of monocytes and macrophages in naïve and tumor-bearing lungs of control and Raptor cKO mice. Although tumor development in the lungs significantly resulted in the increase in the population of patrolling monocytes, the percentages of inflammatory monocytes (Ly6C^highCX3CR1⁺) and patrolling monocytes (Ly6C^lowCX3CR1⁺) were similar between Raptor control and cKO mice (Fig. 6A). We also observed the decrease of AM (CD11c⁺F4/80⁺) within lung CD45⁺CD11b⁻ cells of naïve Raptor cKO mice but LLC metastasis didn’t change the infiltration of AM (Fig. 6B). In addition, there was no change of IM (CD11b⁺F4/80⁺) within lung CD45⁺ cells in naïve control and KO mice. However, LLC metastasis resulted in significantly expansion of IM. Importantly, this cell population was significantly increased in tumor-bearing lungs from Raptor cKO mice (Fig. 6C). As IM express similar surface molecules with metastasis-associated macrophages (MAM, CD11b⁺F4/80^high) which promote extravasation and survival of metastatic cancer cells (29, 32), we then compared the phenotypic changes of IM/MAM in tumor-bearing lungs from control and Raptor cKO mice. The intensity of F4/80 expression was significantly increased in IM/MAM, which was correlated with LLC lung cancer burden (Fig. 6D). Next, we measured the expression of Arg1 and found an increase of Arg1 expression by IM/MAM from Raptor cKO mice (Fig. 6E and F). Accordingly, IM/MAM from cKO mice demonstrated an enhanced immunosuppressive function for T cell proliferation (Fig. 6G). Previous studies have indicated that VEGFR1 (FLT1) signaling in MAM promotes lung metastasis (29, 33). Therefore, we further examined whether mTORC1 modulates VEGFR signaling in IM/MAM. Although there was no difference of VEGF expression by IM/MAM (Fig. 6H), the VEGFR1 expression was significantly increased in Raptor-deficient IM/MAM (Fig. 6I). Together, our data indicate that disruption of mTORC1 in the lung myeloid cells may promote lung cancer metastasis via regulating the accumulation of immunosuppressive IM/MAM with increased VEGF signaling.

FIGURE 6.

Depletion of Raptor in myeloid cells results in the expansion of interstitial macrophages/metastasis-associated macrophages (IM/MAM). (A) The percentages of inflammatory monocytes and patrolling monocytes within CD11b⁺Ly6G⁻ population of naïve and tumor-bearing lungs of control and Raptor KO mice. (B) The percentages of alveolar macrophages (AM) within live CD45⁺CD11b⁻ population of naïve and tumor-bearing lungs. (C) The percentages of IM/MAM within live CD45⁺ population of naïve and tumor-bearing lungs. Data represent two to five independent experiments (n=3–13 mice per group). (D) The correlation of LLC-GFP positive cells and the mean fluorescence intensity (MFI) of F4/80 expression by IM/MAM. Each dot represents one mouse. Data are representative of two independent experiments. (E) The expression levels of Arg1 by IM/MAM were determined by qRT-PCR. Each dot represents a single mouse. (F) The expression of arginase 1 in IM/MAM was measured by Western blot. (G) Immunosuppression assay was performed by co-culture of IM/MAM and OT-I splenocytes in the presence of OVA (25 μg/ml) for 3 days. The T cell proliferation was measured by CFSE dilution. Data represent two independent experiments. (H) The expression levels of VEGFA by IM/MAM were determined by qRT-PCR. Each dot represents one mouse. (I) Flow cytometry analysis of VEGFR1 expression by IM/MAM. Data represent two independent experiments.

Discussion

Tumor progression and metastasis involves critical interactions between the tumor and the microenvironment. Compelling evidence has emerged that host immune response has a major impact on the cancer targeted therapy efficacy (1, 34–36), and targeted therapies can also have effects on host immune responses in addition to their effects on tumor cells (37). Therefore, we need to gain a deeper understanding of these complex effects with the goal to develop more effective therapies against cancer. The present study was designed to characterize the impact of mTORC1 signaling on the tumor-associated myeloid cells which play critical roles in tumor progression and metastasis.

We show that depletion of mTORC1 signaling in tumor-associated myeloid cells exerted no benefits for delaying tumor progression in murine s.c. tumor models. This phenotype is associated with decreased effector T cell responses in myeloid cell mTORC1-deficient TME although TAM immunosuppressive function is significantly diminished. There are two reasons that may explain seemingly contradictory results. First, although ex vivo studies showed decreased immunosuppressive function of Raptor-deficient TAM, TAM trafficking and differentiation defects were observed in LLC s.c. tumors from Raptor cKO mice. CSF1 and CCL2 are the master factors for TAM trafficking and programming in the TME. Blocking M-CSF signaling results in a reduction of mature TAMs due to impaired recruitment, extravasation, proliferation, and maturation of their Ly6C^high monocytic precursors (25). A recent study has also demonstrated a critical role of mTORC1 signaling for M-CSF-mediated myelopoiesis (38). Our data reveal that mTORC1 in myeloid cells is required for the differentiation of Ly6^high monocytes into Ly6^low MHC-II^low TAM in the LLC s.c. tumors. Second, TNF-α secretion by tumor-associated myeloid cells in the s.c. TME was significantly decreased in Raptor cKO mice. TNF-α is a multifunctional cytokine playing a key role in apoptosis and inflammation (39). It has been shown that TNF-α can counterbalance the M2 macrophages by suppressing IL-13 production from activated eosinophils (40). The importance of TNF-α in CD8⁺ T cell- and NK cell-mediated killing of tumor cells has been also highlighted (41). Our in vitro experiments further demonstrate that TNF-α significantly enhances CD8 T cell IFN-γ production. Taken together, we show that although mTORC1-deficient TAM from s.c. tumors demonstrate the decreased immunosuppressive function on a per cell basis, the absolute number of TAM is low due to defective TAM differentiation mediated by depletion of mTORC1 signaling in myeloid cells. In addition, mTORC1-deficient TAM secrete low level of TNF-α. Thus, the net outcome of mTORC1 disruption in myeloid cells yields no drastic s.c. tumor growth phenotype.

The lung microenvironment comprises airway, alveolar and interstitial macrophages, as well as inflammatory and patrolling monocytes. Previous studies have shown that patrolling monocytes can control tumor metastasis to the lungs (28), whereas AM contribute to the promotion of lung metastasis by suppressing Th1 response (30, 42). However, we didn’t observe the causal relationship of AM in experimental LLC lung metastasis model. The percentage of AM is significantly reduced in Raptor-deficient lung myeloid cells with no expansion in tumor-bearing lungs. In contrast to AM, IM are expanded in LLC-bearing lungs and correlated with tumor burden in the lungs. These results are in agreement with previous data showing promotion of metastasis by CD11b⁺F4/80⁺ MAM (29, 32). Although both IM in naïve mice and IM/MAM from tumor metastasis mice express F4/80, the intensity of F4/80 expression is significantly higher in tumor-bearing IM/MAM and is associated with lung tumor burden. Such association with IM/MAM accumulation in tumor-bearing lungs has not been previously reported.

Emerging evidence indicates that TME regulates the therapeutic outcomes in cancer. For example, the sensitivity of murine lung cancer cells (CMT167 and LLC) to PD-1/PD-L1 antibody blockade is different between s.c. tumors and orthotopic model (43). Transfection of MIP-1α is more effective in inhibiting tumor growth in a metastasis model than in the s.c. tumor model (44). The tissue environment also determines which cellular effector pathways are responsible for antibody-dependent tumor immunotherapy (45). In this study, we demonstrate the differential regulation of macrophages by mTORC1 in different TMEs. Depletion of mTORC1 impairs M2 macrophage polarization with reduced expression of Arg1 and diminished immunosuppressive activity of TAM from s.c. tumors. In contrast, disruption of mTORC1 in lung macrophages results in the expansion of IM/MAM with enhanced expression of Arg1 and immunosuppression. The differential effects for Arg1 expression and immunosuppression in macrophages with mTORC1 disruption may be explained by different cellular ontogeny of tumor-associated macrophages (46). The majority of macrophages in s.c. tumors are derived from BM (47), whereas expanded IM/MAM in the tumor-bearing lungs are tissue-resident macrophages. Thus, our data provide additional evidence suggesting that regulation of tumor-associated macrophage polarization by mTORC1 signaling is affected by different TMEs.

In summary, our study has demonstrated that TME differentially modulates the role of mTORC1 signaling in regulating macrophage differentiation and function. mTORC1 signaling plays complex roles in regulating immunosuppressive function of TAM, M-CSF-mediated TAM differentiation and inflammatory cytokine TNF-α production by myeloid cells in subcutaneous primary tumor. mTORC1 also controls the expansion and function of IM/MAM in the lung cancer metastasis model. Depletion of mTORC1 in the lung myeloid cells promotes lung metastasis. Our data provide more evidence suggesting that the regulation of tumor-associated macrophages by mTORC1 may be location dependent. Although it needs to be determined whether these animal models recapitulate microenvironment-specific tumor growth in human patients, findings form this study suggest that it needs to be cautious when mTOR-based targeted therapy is utilized in cancer treatment.

Abbreviations used in this article:

AM

alveolar macrophages

Arg1

Arginase 1

BMDM

bone marrow-derived macrophages

iNOS

inducible nitric oxide synthase

IM

interstitial macrophages

MAM

metastasis-associated macrophages

MDSC

myeloid-derived suppressor cells

Mgl2

macrophage galactose N-acetyl-galactosamine specific lectin 2

mTORC1

mechanistic target of rapamycin complex 1

OVA

Ovalbumin

TME

tumor microenvironment

TAMs

tumor-associated macrophages

VEGFA

vascular endothelial growth factor A

VEGFR1

vascular endothelial growth factor receptor 1