Cancer progression and metastasis involves interactions between tumor cells and the tumor microenvironment (TME). We reported that mice deficient for cytosolic phospholipase A2 (cPLA2-KO) are protected against the development of lung tumors. The goal of this study was to examine the role of cPLA2 in the TME. Mouse lung cancer cells (CMT167 and Lewis lung carcinoma cells) injected directly into lungs of syngeneic mice formed a primary tumor, and then metastasized to other lobes of the lung and to the mediastinal lymph nodes. Identical cells injected into cPLA2-KO mice showed a dramatic decrease in the numbers of secondary metastatic tumors. This was associated with decreased macrophage staining surrounding the tumor. Wild-type mice transplanted with cPLA2-KO bone marrow had a marked survival advantage after inoculation with tumor cells compared with mice receiving wild-type (WT) bone marrow. In vitro, coculturing CMT167 cells with bone marrow–derived macrophages from WT mice increased production of interleukin 6 (IL-6) by cancer cells. This increase was blocked in cocultures using cPLA2-KO macrophages. Correspondingly, IL-6 staining was decreased in tumors grown in cPLA2-KO mice. These data suggest that stromal cPLA2 plays a critical role in tumor progression by altering tumor-macrophage interactions and cytokine production. [Cancer Res 2009;69(5):1733–8]
Lung cancer is the leading cause of cancer deaths. At diagnosis, the majority of patients display metastases, and available treatments fail to prolong survival. During the past 25 years, researchers studying epithelial cancers have focused on genetic changes in tumor cells. However, it has become apparent that cancer progression and metastasis requires complex interactions between tumor cells and the surrounding stroma (1). We have studied the role of cytosolic phospholipase A2 cPLA2 in the development of lung cancer, focusing on tumor cells themselves. Activation of cPLA2, resulting in arachidonic acid release, controls production of a family of eicosanoids, including prostaglandins. In non–small cell lung cancer (NSCLC) cells, oncogenic K-Ras leads to elevated levels of cPLA2 and prostaglandin E2 production; blocking this pathway inhibits transformed growth in vitro and in xenograft models (2). However, little is known regarding the role of cPLA2 in the tumor microenvironment (TME). Although global deletion of cPLA2 in mice inhibits chemically induced lung tumorigenesis (3), these studies do not discriminate between the actions of cPLA2 in tumor cells versus the TME. In this study, we used a model using mouse lung tumor cells injected into the lungs of syngeneic, immune-competent mice. These cells form well-defined primary tumors, which metastasize to other lobes of the lungs and into the mediastinal lymph nodes. Comparing tumor progression in wild-type (WT) versus cPLA2-KO mice allowed us to specifically assess the role of this enzyme in the TME.
Materials and Methods
Cells. CMT167 cells obtained from Dr. Alvin Malkinson (School of Pharmacy, University of Colorado Denver, Denver, Colorado), and Lewis lung carcinoma cells were transfected with pGL3-Control Vector (Promega) containing firefly luciferase constitutively driven by an SV40 promoter and a vector containing neomycin resistance. Stable transfectants were selected using medium containing G418, and individual clones were screened for luciferase activity. A clone with high luciferase activity (CMT167/luc or LLC/luc) was used for injection into animals. Bone marrow–derived cells were isolated from femurs and tibias of WT and cPLA2 knockout mice and cultured in the presence of M-CSF to promote macrophage maturation as previously described (4). After 3 to 7 d in culture, these cells have the morphology of macrophages, and were >95% F4/80 positive.
Mice. cPLA2-KO mice (5) were backcrossed >10 generations to C57BL/6 mice in the Center for Laboratory Animal Care at the University of Colorado Denver. Wild-type littermates were used as controls in all studies. All procedures were performed under a protocol approved by the Institutional Animal Care and Use Committee at the University of Colorado Denver. For bone marrow transplant, donor mice were sacrificed, femurs and tibias were aseptically removed, and bone marrow were obtained by aspiration. Cells were suspended in sterile PBS + 1% FCS. Recipient WT and cPLA2-KO mice were γ-irradiated (900–1,200 RAD split doses) by Cesium source at age 7 to 8 wk. Twenty-four h later, halothane anesthetized, irradiated recipients were injected with donor marrow via retro-orbital injection (5 × 106 BM MNC/mouse). UBI-EGFP mice were used as WT bone marrow donors to track bone marrow by green fluorescent protein (GFP) expression; cPLA2-KO mice, without GFP-labeled bone marrow, were used for knockout bone marrow transplants.
Tumor cell injections. Mice were directly injected with the indicated tumor cells (105 per 40 μL), suspended in 10% Growth Factor Reduced Matrigel Matrix (BD Biosciences) PBS, through the rib cage into the left lobe of the lung using 30-g needles. For bioluminescence imaging, mice were injected i.p. with 300 mg/kg body weight luciferin before sedation and imaged using The IVIS Imaging System 50 Series (Caliper Life Sciences/Xenogen Corp.). At time of sacrifice, some lungs were inflated with India ink to better visualize and quantify additional large (>0.4 mm) and micro (<0.3 mm) secondary tumors. Tumor size was quantitated using digital calipers.
Immunohistochemistry/fluorescence staining. For immunohistochemistry, formalin-fixed, paraffin-embedded tissues were deparaffinized, rehydrated, and underwent antigen retrieval by heating for 20 min at 115°C in a decloaking chamber (Biocare). Sections were then exposed to specific antibodies overnight at 4°C. Antigen/antibody complexes were visualized using kits from Vector Laboratories and sections lightly counterstained with hematoxylin. Negative controls included the use of mouse or rabbit IgG. Sections were visualized using an Olympus light microscope equipped with SPOT software. For double immunofluorescence labeling, tissue sections were treated as above. After incubations with primary antibodies, antigen/antibody complexes were visualized using rhodamine (Alexa Fluor-568)–coupled or FITC (Alexa Fluor-488)–coupled secondary antibodies (Molecular Probes); sections were sequentially incubated with specific primary and secondary antibodies. Coverslips were mounted with VectaShield medium containing 4′,6-diamidino-2-phenylindole to detect all cell nuclei (Vector Laboratories), and sections were visualized using a Nikon inverted fluorescence microscope equipped with Metamorph software. Antibodies used include polyclonal anti-F4/80 (1:100; Caltag), FITC-conjugated anti-GFP (1:100; Abcam), and polyclonal anti–interleukin 6 (IL-6; 1:2,000; Abcam).
Coculture. Bone marrow–derived macrophages were grown on the bottom of Transwell inserts. CMT/167 cells were grown on Transwell filters. After 24 h in coculture, cells were separated and each cell type was placed in fresh medium. Condition medium was collected after 24 h and assayed for IL-6 production by specific ELISA.
Results and Discussion
The CMT167 cell line derived from a spontaneous alveolar lung carcinoma of a C57BL/6 female mouse (6) was stably transfected with firefly luciferase (CMT167/luc) to allow bioluminescent imaging to be performed. CMT167/luc cells suspended in Matrigel were injected through the rib cage into the left lobe of the lung of C57/BL6 mice as previously described (7, 8), and tumor formation were analyzed as a function of time by bioluminescent imaging (Fig. 1A). At 3 weeks, mice consistently developed large primary tumors at the sight of injection (Fig. 1A and B). Lungs were inflated with India ink to visualize and quantify additional large (>0.4 mm) and micro (<0.3 mm) secondary tumors. Wild-type mice exhibited multiple secondary tumors throughout the left and right lung lobes (Fig. 1C, and D, a–c). By 4 weeks postinjection, all mice showed clinical features of mediastinal lymph node metastasis (Fig. 1D , d).
To assess the role of cPLA2 in the TME, equal numbers of CMT167/luc cells were injected into WT or cPLA2-KO mice, and lungs were harvested after 4 weeks. Quantification of tumor metastases was limited to large secondary tumors visible under a dissecting microscope. Growth of the primary tumor was not significantly different between the two groups of mice (Fig. 2A). CMT167 cells injected into WT mice metastasized to other lung lobes and into mediastinal lymph nodes. However, CMT167 cells growing in cPLA2-KO mice showed fewer metastases and little lymph involvement; five of five WT mice, but only one of five KO mice, exhibited lung metastases and lymph involvement by macroscopic morphologic examination (Fig. 2B). To extend the generality of these findings, we performed similar experiments with Lewis lung carcinoma cells, another lung cancer cell line derived from C57BL/6 mice. These cells are more aggressive as reflected by more rapid proliferation of the primary tumor and increased numbers of secondary tumors. However, as with CMT167 cells, we observed no statistical difference in the size of primary tumors (data not shown), but a marked decrease in secondary tumor number (Fig. 2C) in cPLA2-KO mice compared with WT. CMT167 cells were used for all subsequent studies.
Although numerous cell types comprise the TME, attention has focused on tumor-associated macrophages (TAM) as mediators of tumor progression and metastasis (9). Macrophages express high levels of cPLA2, and produce proangiogenic cytokines. We showed that macrophage recruitment to the lung is impaired in cPLA2-KO mice after an inflammatory stimulus (10). TAMs surrounding tumors were assessed by F4/80 staining. Tumors grown in WT mice showed abundant accumulation of TAMs surrounding developing tumors. In contrast, fewer TAMs were detected in tumors of KO mice (Fig. 2D).
Recruitment of bone marrow–derived circulating monocytes/macrophages is associated with more aggressive, malignant tumors. To assess the role of cPLA2 in bone marrow–derived cells on tumor progression, WT mice received bone marrow transplants from either WT or cPLA2-KO mice. Animals were allowed to recover for 5 weeks, and then injected with CMT167/luc cells. Mice transplanted with cPLA2-KO bone marrow had a marked survival advantage over mice transplanted with WT bone marrow (Fig. 3A). In separate studies, bone marrow derived from either WT or cPLA2-KO mice was transplanted into a limited number of either WT or cPLA2-KO recipients to examine recruitment of bone marrow–derived macrophages to the tumor. These studies used UBI-EGFP/B6 transgenic mice, which express EGFP in all cells, as WT bone marrow donors. Lungs were harvested 4 weeks after CMT167/luc injection and analyzed histologically. Mice receiving cPLA2-KO bone marrow, independent of genotype, exhibited less inflammatory infiltrate surrounding the primary tumor compared with mice receiving WT bone marrow (Fig. 3B). Mice receiving WT bone marrow were examined for EGFP and stained for F4/80. Both groups of mice receiving WT bone marrow showed large numbers of EGFP+;F4/80+ cells surrounding the tumor (Fig. 3C) indicating recruitment of bone marrow–derived macrophages. In both groups of mice receiving cPLA2-KO bone marrow, fewer F4/80-positive cells were detected, consistent with fewer macrophages. These data suggest that the stromal protective effects of cPLA2 depletion are mediated in large part through bone marrow–derived cells, and support a model in which cPLA2 expression in bone marrow–derived macrophages is critical for recruitment of these cells to the site of the tumor.
Production of IL-6 by both tumor and stromal cells contributes to tumor progression as well as angiogenesis (11). To determine if cPLA2 plays a role in IL-6 production in vivo, sections from tumors grown in WT or cPLA2-KO mice were stained for IL-6. Tumors grown in cPLA2-KO mice had markedly lower levels of IL-6 both within and surrounding the tumor, compared with tumors grown in WT mice (Fig. 4A). To examine IL-6 production in vitro, bone marrow–derived cells isolated from WT or cPLA2-KO mice were cultured in the presence of M-CSF to promote macrophage maturation as previously described (4). These cells have the morphology of macrophages, and are >95% F4/80 positive. Macrophages derived from WT mice produced twice the levels of IL-6 compared with macrophages derived from cPLA2-KO mice (Fig. 4B). Because interactions between cancer cells and macrophages effect cytokine production (12), IL-6 production was also assessed in cocultures of bone marrow–derived macrophages with CMT167 cells using Transwells, which allows diffusible mediators to act on each cell type. Coculture of macrophages with CMT167 cells increased IL-6 production in both WT and cPLA2-KO macrophages, with WT macrophages continuing to produce twice the levels compared with cPLA2-KO macrophages (Fig. 4B). CMT cells grown alone failed to produce detectable levels of IL-6. However, coculture with macrophages resulted in significant IL-6 production. Importantly, IL-6 production by CMT167 cocultured with cPLA2-KO macrophages was only 40% of the levels seen with WT macrophage-CMT cell coculture (Fig. 4C).
Although cPLA2 expression by NSCLC has been shown to be important for transformed growth (2), this is to our knowledge the first report indicating the importance of this enzyme in the TME. Our data indicate that two independent mouse lung cancer cell lines show a marked impairment in formation of secondary tumors when grown in cPLA2-deficient mice. Although the TME comprises many types of cells, we have focused on the role of bone marrow–derived cells, specifically macrophages. Specific deletion of cPLA2 in bone marrow–derived cells inhibits tumor progression and metastasis and promotes cancer survival. This protection is associated with decreased numbers of TAMs surrounding the tumors in cPLA2-KO mice. In human lung cancer, increased numbers of macrophages surrounding the tumor has been associated with an unfavorable prognosis (13). Our findings are consistent with these observations.
Macrophages play a complex role in cancer progression. Although initially mediating cytotoxic effects on tumors, TAMs have been implicated in promoting tumor progression and metastasis. Production by TAMs of proangiogenic cytokines in cooperation with tumor cells stimulates tumor angiogenesis (14). TAMs secrete factors with immunomodulatory activity, inhibiting T-cell function and other immune antitumorigenic effects. IL-6, a critical cytokine for tumor progression, promotes cancer progression through effects on tumor cells or stroma (15). Our data indicate that tumors grown in cPLA2-KO mice are exposed to lower levels of IL-6, associated with reduced tumor progression and metastasis. Consistent with this, in vitro data indicate that cPLA2 contributes to IL-6 production by macrophages, and is critical for the synergistic induction of IL-6 seen in cocultures of cancer cells and macrophages.
The mechanism whereby cPLA2 regulates IL-6 production remains to be established. cPLA2 is critical for PGE2 production, the major prostaglandin produced by macrophages and NSCLC cells. PGE2 can induce IL-6 production in cholangiocarcinoma cells (16), and modulates cytokine production by macrophages leading to promotion of M2-type macrophages (17). PGE2 can act as an immune suppressor by inhibiting proliferation of T cells, effecting immunoglobulin production by B cells, and regulating dendritic cell maturation (18). Further studies will be required to define the role of PGE2 in this model. Because cPLA2 is critical for production of other eicosanoids, including leukotrienes, other mediators may contribute to the effects of stromal cPLA2. In addition, our studies also do not rule out a role for cPLA2 in other stromal cells, including fibroblasts and vascular cells. Recent studies have implicated production of PGE2 by myeloid-derived suppressor cells as a mediator of breast cancer progression (19). Finally it should be noted that numerous studies have examined nonsteroidal anti-inflammatory drugs as therapeutic treatments for lung cancer. It seems that these agents may not only effect tumor growth but may also directly effect metastasis through targeting prostaglandin production in macrophages and other stromal cells (20).
Disclosure of Potential Conflicts of Interest
No potential conflicts of interest were disclosed.
Grant support: NIH (CA103618, CA108610, and CA58187) as well as a University of Colorado Cancer Center Seed Grant (M.C.M. Weiser-Evans).
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.