Background
Epithelial ovarian carcinoma (EOC) is the leading cause of death from gynecologic malignancies in the United States and is the fourth most common cause of cancer death in women [
1]. Over 70% of women with EOC present with advanced stage disease and tumor dissemination throughout the peritoneal cavity [
2]. Despite the standard therapy with surgical cytoreduction and the combination of cisplatin and paclitaxel, the treatment efficacy is significantly limited by the frequent development of drug resistance [
3]. Novel complementary strategies are urgently needed to improve the outcomes of ovarian cancer.
Much data suggest that immunotherapy for EOC should be effective [
4]. Firstly, EOC cells express tumor-associated antigens against which specific immune responses have been detected [
5‐
9]. Secondly, the studies pioneered by Coukos and colleagues indicate tumor immune surveillance plays a role in clinical outcomes in EOC supported by the close correlation between survival and tumor infiltration with CD3
+ T cells in the large annotated clinical samples [
10]. Thirdly, although EOC is a devastating disease, metastases are frequently restricted to the peritoneal cavity where the tumor microenvironment is directly accessible, which prevents the need for systemic delivery of immunostimulatory treatments [
11]. Despite the abundant evidence that anti-tumor immunity could be effective, clinical success with immune-based therapies for EOC has generally been modest [
12].
T-cell immunoglobulin and mucin domain 3 (TIM-3), as a relatively newly described co-inhibitory molecule, was expressed by IFN-γ–secreting T-helper 1 (Th1) cells and subsequently on CD8
+ T cytotoxic type 1 (Tc1) cells, DCs and monocytes [
13‐
16]. The galectin-9, a soluble molecule widely expressed and upregulated by IFN-γ, was identified as TIM-3 ligand [
17,
18], which induces cell death via binding to TIM-3 expressed on Th1 cells [
19], suggesting a role for TIM-3 in negatively regulating Th1 responses. Emerging data has implicated TIM-3 a critical role in regulating tumor immune response. Early studies reported that the growth of 4 T1 mammary tumors was inhibited in TIM-3-deficient mice, and anti-TIM-3 monoclonal antibody (mAb) could suppress the growth of established subcutaneous EL4 lymphoma, suggesting TIM-3 as a potential target for cancer immunotherapy [
20]. Recent studies observed that the expression of TIM-3 and PD-1 was up-regulated on circulating tumor-specific and tumor-infiltrating CD8
+ T cells from patients and mice bearing advanced malignancies respectively, which correlated with the severely exhausted phenotype defined by failure to proliferate and produce effector cytokines, and combined blockade of both TIM-3 and PD-1 pathway reversed tumor-induced T-cell dysfunction and effectively suppressed the experimental tumor growth. This finding is further validated by the experiments demonstrating that combined anti-TIM-3/PD-1 mAbs significantly prevented established tumor growth and even cured a fraction of mice in methylcholanthrene-induced fibrosarcomas and 6 different experimental mouse tumor models, supporting the potential of blocking TIM-3 in combination with other immune-regulatory mAbs for the treatment of cancer.
CD137 (as known as CD137) belongs to the Tumor Necrosis Factor Receptor (TNFR) superfamily and is transiently upregulated on both CD4
+ and CD8
+ T cells following activation [
21]. Upon engagement, CD137 co-stimulates CD8
+ T cells promoting their proliferation, Th1-type cytokine production, and survival [
22]. Much evidence demonstrate the promising effects for anti-CD137 mAbs in the treatment of mice bearing established tumors [
23,
24]; this is not only achieved by agonist antibodies but also by dimeric RNA aptamers or tumor cells expressing a surface-attached anti-CD137 single chain antibody [
25,
26]. This preclinical evidence has led to clinical trials with 2 human mAbs directed against CD137 [
27].
Although antagonist TIM-3 or agonistic CD137 antibodies can promote the rejection of some murine tumors, however, poorly immunogenic tumors such as ID8 ovarian cancer do not respond to antibody therapy alone [
28]. We hypothesized that combined TIM-3 blockade and CD137 activation would strengthen the antitumor effect by synergistically releasing the brake for CD4
+ cells and promoting the function of CD8
+ cells. In this study, using ID8 murine ovarian cancer model, we evaluated the therapeutic effect of single or combined anti-TIM-3 and anti-CD137 mAbs and found that combined anti-TIM-3/CD137 significantly suppressed the 10 days established peritoneal ID8 tumor growth, resulting in 60% of treated mice tumor free 90 days after tumor injection. We further characterized the cellular and molecular mechanisms driving this combined antitumor effect elucidating the basic processes necessary to achieve immune-mediated tumor rejection.
Methods
Mice
Female C57BL (6–8 wk old) were purchased from the Animal Experimental Center of the China Medical University. Animal use was approved by our institution (China Medical University).
Cell lines
ID8, a clone of the MOSEC ovarian carcinoma of C57BL/6 origin was a gift from Dr. George Coukos (University of Pennsylvania, Philadelphia, USA). Murine B16 melanoma cells, TC-1 lung carcinoma cells and T cell lymphoma EL4 cells were purchased from ATCC (Manassas, VA). Tumor cells were cultured in the complete DMEM medium supplemented with 10% FBS (Thermo Scientific, Rockford, IL), 100 U/mL penicillin and 100 μg/mL streptomycin before cell suspensions were prepared and transplanted to mice. The EL4 cells and splenocytes were maintained in a complete medium of RPMI-1640 supplemented with 10% FBS, 25 mM HEPES, 2 mM glutamine, 100 U/mL penicillin and 100 μg/mL streptomycin.
Antibodies
Therapeutic anti-CD137 (Clone lob12.3; Catalog#BE0169), anti-TIM-3 (Clone RMT3-23; Catalog#BE0115), anti-CD4 (Clone GK1.5; Catalog#:BE0003-1), anti-CD8 (Clone 2.43; Catalog#:BE0061), anti-NK1.1 (Clone PK136; Catalog#:BE0036), anti-CD19 (Clone 1D3; Catalog#:BE0150) and control (Clone 2A3; Catalog#:BE0089) were purchased from BioXcell (West Lebanon, NH). Antibodies used for flow cytometry were purchased from Tianjing Sungene (Tianjing, China) and eBioscience (San Diego, CA).
Tumor challenge and treatment experiments
In the experiments with ID8 ovarian tumor (Additional file
1: Figure S1), mice (5 or 10 mice/group) were injected intraperitoneally (i.p.) with 1 × 10
6 ID8 cells in 0.1 mL of PBS. At day 3, 7 and 11 (3 days established tumor model) or days 10, 14 and 18 (10 days established tumor model) post-tumor injection, each mouse received the i.p. injection of 250 μg of control, anti-TIM-3, anti-CD137 or combined anti-TIM-3/CD137 mAb in 250 μL of PBS as shown in the figure legends. The mice were weighted twice weekly and checked daily for the clinical sign of swollen bellies indicative of ascites information and for the evidence of toxicity such as respiratory distress, mobility, weight loss, diarrhea, hunched posture, and failure to eat while histopathology was conducted on major organs (i.e., liver, kidney, intestines, lungs, and colon). Following institutional guidelines, mice were killed when they developed ascites and had a weight increase > 30%. The survival of each mouse was recorded and overall survival was calculated.
For assessing the development of immune memory, pooled (2 independent experiment) 9 long-term surviving mice (90 days after first tumor injection) from combined anti-TIM-3/CD137 therapy group or age-matched naïve mice (which served as control) were challenged i.p. or subcutaneously (s.c.) with 1 × 106 ID8 cells or 1 × 106 syngeneic but antigenically different TC1 cells. Three perpendicular diameters of s.c. tumors were measured every second day using a caliper and tumor volumes were calculated according to the formula: 1/2 × (length) × (width)2. Mice were sacrificed when they seemed moribund or their tumors reached 10 mm in diameter.
For depletion of immune cells, mice were injected i.p. with 500 μg of mAbs to CD8, CD4, NK1.1, or CD19, 1 day before and two days after tumor challenge, followed by injection of 250 μg every 5 days throughout the experiment. The efficacy of cell depletion was verified by staining peripheral blood leukocytes for specific subsets after depletion (data not shown).
Evaluation of tumor-infiltrating immune cells (TIIC) in peritoneal lavages by flow cytometry
Mice which had been transplanted i.p. with ID8 cells were euthanized 7 days after they had been injected with the 2 mAb combination (or control) as in the therapy experiments. To obtain peritoneal immune cells, 3 ml PBS was injected into the peritoneal cavity of mice with ID8 tumors immediately after euthanasia, their belly was massaged and the fluid was removed, filtered through a 70 μM cell strainer (BD Biosciences), washed and immune cells were isolated by using a mouse lymphocyte isolation buffer (Cedarlane, Burlington, Ontario) following the manufacturer’s instruction.
For the staining of immune cells, above prepared immune cells were washed with FACS staining buffer and incubated with mouse Fc receptor binding inhibitor (eBioscience) for 10 minutes before staining with mAbs (Tianjing Sungene) against mouse CD45 (clone 30-F11), CD3 (clone 145-2C11), CD4 (clone GK1.5), CD8 (clone 53–6.7), CD19 (clone eBio1D3), CD11b (clone M1/70) and Gr-1 (clone RB6-8C5) for 30 minutes. For intracellular staining of FoxP3 (clone FJK-16 s; eBioscience), cells were fixed, permeabilized and stained following the instruction of Cytofix/Cytoperm kit (BD Bioscience). Flow cytometry was performed using FACSCalibur (BD Biosciences) and the data were analyzed using FlowJo software (Tree Star). All flow cytometry experiments were performed at least 3 times.
Quantitative RT-PCR
Total cellular RNA was extracted using RNeasy Mini Kits (Qiagen, Hilden, GA) and reverse transcribed into cDNA using SuperScript III Reverse Transcriptase (Invitrogen). Expression for genes of interest was analyzed in cells of peritoneal lavage on day 7 after the third injection of mAb. The primers for all genes tested, including internal control GAPDH, were synthesized by Takara Inc., Dalian, China. Primer sequences were listed in Additional file
2: Table S1. Quantitative real-time PCR was performed via ABI PRISM 7500 Real-Time PCR Systerm (Applied Biosystems) with 1× SYBR Green Universal PCR Mastermix (Takara). Transcript levels were calculated according to the 2–ΔΔCt method, normalized to the expression of GAPDH, and expressed as fold change compared with control.
Evaluation of antigen-specific CTL immune response
Isolated splenocytes from treated mice were cultured in the presence of 10 μg/mL H-2Db-restricted mesothelin-derived peptides (amino acid 406–414) or control HPV-E7-derived peptide (amino acid 49–57; all from GenScript, Nanjing, CA) for 3 days. IFN-γ in the supernatants was determined by Mouse IFN-γ Quantikine ELISA Kit (R&D systems, Minneapolis, MN).
For CTL assays, effector cells were obtained by coculturing 5 × 106 splenocytes with 5 × 105 UV-irradiated ID8 cells for 4 days. Peptide-pulsed EL4 target cells were generated by adding 10 μg/ml of peptide and incubating for 4 hours. CTL activity was measured using the CytoTox96 Non-Radioactive Cytotoxicity Assay kit (Promega, Madison, WI) following the manufacturer’s instructions. In brief, target cells were incubated with varying numbers of effector cells for about 4 hours, and supernatants were then analyzed for lactate dehydrogenase release. The results are expressed as percent specific lysis, calculated as (Experimental release-Spontaneous release/Total release-Spontaneous release) × 100. In some experiments, effector cells were incubated with anti-CD4 or CD8 antibody (10 μg/mL) for 2 hours before CTL assay.
Antibody evaluation by flow cytometry
We detected the presence of mesothelin-specific antibodies using the method described previously [
29]. Blood was obtained from 2 mAb treated long-term surviving mice (90 days after the tumor inoculation). The presence of mesothelin-specific antibodies was determined by staining the mouse ID8 ovarian cancer cells using serum from treated mice in a 1:200 dilution, followed by Phycoerythrin-conjugated anti-mouse IgG antibody (eBioscience) staining. Staining with sera from naïve mice was used as negative control. Analysis of cell staining was performed described above.
ELISA
Mice injected i.p. with 1 × 106 ID8 cells 10 day earlier were injected thrice at 4 days interval with 250 μg of control or anti-TIM-3/CD137 mAb. Seven days after the last mAb injection, pooled peritoneal lavage cells (1 × 106/well) harvested from treated mice were stimulated in vitro with 50 ng/ml PMA and 1 μg/ml ionomycin for 6 hours prior to the analysis of IL-10 and IFN-γ production in the supernatants by ELISA according to the manual (R&D systems). The results were analyzed after normalization according to the T cell numbers.
Statistics
Results were expressed as mean ± SEM. All statistical analyses were performed using GraphPad Prism 5. Student’s t test was used to compare the statistical difference between two groups and one-way ANOVA was used to compare three or more groups. Survival rates were analyzed using the Kaplan–Meier method and evaluated with the log-rank test with Bonferroni correction. Significant differences were accepted at p < 0.05.
Discussion
The antitumor efficacy of immunotherapy remains insufficient to achieve durable clinical responses in patients with advanced EOC. In this study, we demonstrate that combined anti-TIM-3/CD137 mAbs inhibited the outgrowth of ID8 ovarian cancer cells injected 10 days previously, resulting in the long-lasting survival of 60% of mice while either mAb alone was ineffective in tumor protection. The findings provide evidence that combined TIM-3 blockade and CD137 activation may serve as a novel immunotherapeutic option for treatment of ovarian cancer.
We next sought to understand the mechanisms underlying increased tumor-rejecting effect by simultaneously removing a major brake on expansion via blockade of the negative regulator TIM-3, while at the same time actively driving proliferation and survival through activation of the co-stimulatory receptor CD137. We found that single TIM-3 blockade significantly increased the percentage of CD4
+ T cells and slightly elevated the percentage of CD8
+ T cells while it had little effects on the immunosuppressive CD4
+FoxP3
+ Treg, CD11b
+GR-1
+ MDSCs and CD19
+ B cells in peritoneal lavage. On the contrast, single CD137 activation promoted the accumulation of CD8
+ T cells with significantly elevated percentage and absolute number in peritoneal lavage. Quantitative RT-PCR data demonstrated moderately increased IFN-γ expression although significantly increased accumulation of CD8
+ T cells in peritoneal cavity from anti-CD137 mAb treated mice, indicating most of CD8
+ T cells were without function or immunologically ignorant as described previously [
30]. This was consistent with the lack of antitumor effect by single anti-CD137 mAb in ID8 model, which may be due to the concomitant expansion of Treg and MDSC by anti-CD137 mAb. In accordance with the more pronounced CD137 expression on CD8
+ versus CD4
+ T cells, marginal effect of anti-CD137 mAb on CD4
+ T cells were observed. Remarkably, combined anti-TIM-3/CD137 mAb increased the percentage and absolute number of tumor-infiltrating CD4
+ and CD8
+ T cells, while at the same time decreasing the immunosuppressive Treg and MDSC in peritoneal lavage, which gave rise to the significantly increased ratio of CD4
+ and CD8
+ T cells to immunosuppressive cells.
Importantly, we detected a systemic antigen-specific CD8
+ T-cell mediated CTL immune response to mouse mesothelin in anti-TIM-3/CD137 mAb treated mice with ID8 tumors, as evidenced by mesothelin epitope-specific IFN-γ production and cytotoxicity by CD8
+ T cells from these mice. As an endogenous non-mutated antigen, mesothelin should be naturally tolerized against. The induction of mesothelin-specific CD8
+ CTL by anti-TIM-3/CD137 mAb in the mesothelin-expressing ID8 model indicates that endogenous tolerance to mesothelin was overcome, which is consistent with previous studies showing the presence of mesothelin-specific humoral or cellular immune response in patients with cancer expressing high level of mesothelin, such as pancreatic cancer, ovarian cancer or mesothelioma [
31,
32]. We did not detect mesothelin-specific antibodies in sera harvested 90 days after tumor challenge in 2 mAb treated mice by flow cytometry, but the presence of these antibodies cannot be completely excluded in view of comparatively low sensitivity of this approach and possibly suboptimal time point for sera collection. Serially collection of sera at different time points after mAb injection and utilization of more sensitive approaches such as ELISA should be warrant in our future work.
The importance of CD8
+ T-cell mediated CTL response in tumor protection was supported by in vivo antibody depletion experiments demonstrating depletion of CD8
+ T cells abrogated the antitumor effect of anti-TIM-3/CD137 mAb. The pivotal role of CD8
+ T cells in antitumor effect elicited by anti-TIM-3/CD137 mAb treatment is consistent with other combined strategies involving anti-CD137 mAb, such as combined cyclophosphamide/anti-CD137 treatment [
33], which has been shown to produce synergist antitumor effects via expansion of CD8
+ tumor-specific T cells. The depletion of CD4
+ T cells also decreased the antitumor effect of 2 mAb treatment although the effect was not as prominent as depletion of CD8
+ T cells, which may be explained by the fact that Treg depletion by CD4 antibody partially compensates a deleterious effect on effector CD4
+ T cells in view of the finding that Treg cells can be expanded by anti-CD137 antibody [
34]. Our data were concordant with a recent study showing the administration of anti-TIM-3 mAb had a tumor-suppressing effect in several transplantable and chemical carcinogen-induced fibrosarcoma tumor models via CD4
+ and CD8
+ T-cell- and IFN-γ-dependent mechanisms [
16,
35]. A recent study also shows that TIM-3 blockade enhanced the antitumor effects of vaccine-induced response against established B16 murine melanomas via NK cell-dependent mechanisms [
36], and this discrepancy may be possibly due to the different treatments and tumor microenvironments where TIM-3 may modulate distinct immune cells and the related signaling pathways that exist [
36].
Noticeably, we did not detect the expression of galectin 9 on the tumor cells in spite of the immune enhancing effect of the anti-TIM3 mAb. This finding is contrasting to the scenario of other inhibitory receptors, such as PD-1, where the presence of the ligand (PD-L1) on the tumor seems to correlate with the response to anti-PD-1 [
37]. A recent report did not detect a specific interaction between galectin 9 and TIM-3 [
38], suggesting that TIM-3 functions are independent of galectin-9, which may partially explain our finding. It warrants further explore whether TIM-3 receives an inhibitory signal from an unidentified molecule other than galectin-9 on tumor cells where blockade of this interaction by anti-TIM-3 mAb elicits an immunostimulatory effect.
Activation of the co-stimulatory receptor CD137 in the clinic has shown promise as a therapy for advanced solid tumors with manageable autoimmune adverse effects when administrated at dose levels ranging from 0.3 mg per kg to 10 mg per kg [
27]. The major side effect of systemic use of CD137 agonist antibody appears to be an as yet poorly mechanistically defined inflammatory liver toxicity by infiltration with T cells and hematologic abnormalities [
39,
40]. In our study, we did not observe any obvious toxicity such as weight or hair loss in mice receiving single or combined anti-TIM-3/CD137 mAb. Further detailed biochemical and histological analysis of liver, spleen, bone marrow and peripheral blood at different time points after antibody injection should be warranted to inform any major side effects induced by 2 mAb treatment. Recent studies provide a rationale for local delivery of anti-CD137 mAb to treat tumor. The study from A. Palazon et al. shows that hypoxia-inducing transcription factor-1α (HIF-1α) in hypoxic tumors induces the expression of CD137 on TILs [
41] and local intratumoral low-dose injection of agonist anti-CD137 mAb elicited systemic tumor-specific effector T cells capable of eradicating distant metastases. In addition, a study performed in a murine AT3 breast cancer model shows that combined radiotherapy and anti-CD137 treatment upregulated the expression of CD137 on tumor-specific CD8
+ CTLs [
42]. Therefore, it is of great interest to elucidate the role of CD137 expression on TILs in our model since local delivery of anti-CD137 antibodies might avoid the toxicity associated with systemic application without compromising the antitumor effects.
In view of recent encouraging results of immunomodulatory mAbs in clinic for treatment of multiple solid tumor [
43], our finding that TIM-3 blockade and CD137 activation synergistically induce a potent antitumor effect in a highly clinical relevant ID8 ovarian cancer model should aid the design of future trials for ovarian cancer immunotherapy.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
ZQG conceived and designed the study, performed most of the experiments and drafted the manuscript. LLW and GW carried out the flow cytometric analysis, participated in the design of the study and helped in writing the manuscript. DLC contributed in cell culture techniques and analyzed data. ZJX participated in the statistical analysis and interpretation of data. ML participated in the analysis and revised the manuscript. SLZ, head of the department, critically revised the manuscript. All authors read and approved the final manuscript.