Sunitinib facilitates the activation and recruitment of therapeutic anti-tumor immunity in concert with specific vaccination

Mice

Female 6–8 week old C57BL/6 (B6; H-2^b) mice were purchased from the Jackson laboratory (Bar Harbor, ME) and maintained in micro-isolator cages. Animals were handled under aseptic conditions, per an Institutional Animal Care and Use Committee (IACUC)-approved protocol.

Cell lines and Media

The MO5 (B16.OVA; H-2^b) melanoma has been described previously (24). This cell line was free of mycoplasma contamination and was maintained in complete media [CM: RPMI 1640 supplemented with 10% heat-inactivated fetal bovine serum, 100 U/ml penicillin, 100 μg/ml streptomycin, and 10 mM L-glutamine (all reagents from Life Technologies, Inc., Grand Island, NY)] in a humidified incubator at 5% CO₂ and 37°C.

Viral vector

The Ad.mIL12 recombinant adenoviral vector (25) encoding the p35 and p40 subunits of murine IL-12, and the control (empty) Ad.ψ5 virus were produced and provided by the University of Pittsburgh Cancer Institute Vector Core Facility (a Shared Resource).

Peptides

Peptides OVA_257–264 (SIINFEKL) and OVA_323–339 (ISQAVHAAHAEINEAGR) were synthesized by 9-fluorenylmethoxycarbonyl (Fmoc) chemistry by the University of Pittsburgh Cancer Institute's Peptide Synthesis Facility (a Shared Resource). Peptides were >96% pure based on high performance liquid chromatography profile and mass spectrometric analysis performed by the University of Pittsburgh Cancer Institute's Protein Sequencing Facility (a Shared Resource).

DC.IL12 Vaccines

DC were generated from BM precursors isolated from the tibias/femurs of C57BL/6 mice and then infected on day 5 of culture with recombinant adenovirus encoding IL-12p70 (Ad.mIL-12) as previously described (25). On day 7, IL-12 gene modified DC (DC.IL12) were loaded with a mixture of 10 μM of each of the OVA_257–264 and OVA_323–339 synthetic peptides (for 4h at 37°C and 5% CO₂), which serve as CD8⁺ and CD4⁺ T cell epitopes, respectively, in C57BL/6 mice (26, 27).

Animal Experiments

C57BL/6 mice received s.c. injections with 2 × 10⁵ MO5 (B16.OVA) tumor cells in the right flank on day 0. On day 7, the animals were randomized into cohorts of 5 mice each exhibiting average tumor sizes of approximately 30–50 mm³. Tumor-bearing mice were either left untreated or treated with s. c. injection of 1×10⁶ DC.IL12/OVA in 50 μl PBS in the left flank on day 7, days 7 and 14, or days 14 and 21 post-tumor inoculation as indicated in text. SUT (SUTENT™, Pfizer, New York, NY) was dissolved in Labrasol (Gattefossé Canada Inc., Toronto, Canada) and administered daily (at doses of 0.1 – 1.0 mg in 50 μl of Labrasol), via oral gavage for 7–14 consecutive days, beginning on day 7, 14 or 21 post-tumor inoculation, as indicated in text. In some experiments, 200 μg of blocking antibodies against murine CXCR3 (CXCR3-173: hamster IgG; non-T cell depleting; the kind gift of Dr. Robert D. Schreiber, Washington University School of Medicine; ref. 28) or murine VCAM-1 (R & D Systems, Minneapolis, MN; rat IgG) or species/isotype-matched control antibodies (Santa Cruz Biotech., San Diego, CA) were injected i.p. at the time of initiating treatment and every 2 days thereafter through day 15 (i.e. days 7, 9, 11, 13 and 15 post-tumor inoculation) to determine impact on therapeutic outcome. In all cases, individual tumor sizes were then assessed every 3–4 days and recorded in mm³ as determined by the formula: tumor volume = 0.5 × a² × b, in which a is the smallest and b, the largest (orthogonal) superficial diameter. All measurements were taken using vernier calipers, with data reported as mean tumor volume ± SD.

Antibodies, MHC/peptide tetramer and flow cytometry analysis

Phycoerythrin (PE)-conjugated H-2K^b/SIINFEKL tetramers were purchased from Beckman Coulter (Brea, CA) and used to stain cells per the manufacturer's protocol. Single cell suspensions derived from tumors, tumor-draining lymph nodes (TDLN), vaccine-draining lymph nodes (VDLN) and splenocytes were also stained using directly-conjugated Abs (all from BD Biosciences, San Jose, CA): FITC-anti-CD8, PE-anti-CD8, FITC-anti-CD4, PE-anti-Gr1, FITC-anti-CD11b, PE-anti-CD86, PE-anti-Foxp3, PE-anti-CTLA-4 and FITC-anti-PD-1. Intracellular staining for Foxp3 performed after first staining the cell surface with FITC-antiCD4 mAb. Briefly, cells were incubated in permeabilization/fixation buffer (eBioscience, San Diego, CA) for a minimum of 30 min. After washing with permeabilization buffer, cells were blocked with the anti-FcR mAb and then stained with PE-anti-Foxp3 or isotype control mAb for 30 min at 4°C. Cells were washed in permeabilization buffer and resuspended in PBS. Flow cytometry was performed using Cell Quest software and a FACScan flow cytometer (Becton Dickinson, San Jose, CA). Data analysis was performed using FlowJo software (Tree Star Inc., Ashland, OR).

Evaluation of anti-OVA T Cell Responses

CD3⁺ T cells were isolated using specific MACS™ beads (Miltenyi Biotec) from single cell suspensions of (mechanically-disrupted) TDLN or (enzymatically-digested; ref. 25) TIL prepared 28 days after tumor inoculation. For all analyses, T cells were pooled from 3 animals per treatment cohort. After their isolation, T cells were restimulated in vitro for 5 days with irradiated (100 Gy) MO5 tumor cells at a T cell-to-tumor ratio of 10:1. Recovered T cells were then co-cultured alone or with syngenic DC pulsed with no peptide or the SIINFEKL peptide (10 μM for 4h at 37°C and 5% CO₂) for 48h. Cell-free supernatants were then harvested and assessed for levels of mIFN-γ using a specific ELISA (BD Biosciences) with a lower limit of detection of 32.5 pg/ml. Data are reported as the mean ± SD of triplicate determinations.

RT-PCR

Total RNA was extracted from single cell suspensions obtained via the enzymatic digestion of resected day 7, 10, 14, 21 and 28 tumors using Trizol (Invitrogen, Carlsbad, CA) per the manufacturer's protocol. Reverse transcription-PCR (RT-PCR) was performed using the primer pairs listed in Table S1. Cycling times and temperatures were as follows: initial denaturation at 94°C for 2 min (1 cycle), denaturation at 94°C for 30 s, annealing at 60°C for 30 s and elongation at 72°C for 1 min (35–40 cycles), final extension at 72°C for 5 min (1 cycle). PCR products were identified by image analysis software for gel documentation (LabWorks Software; UVP) following electrophoresis on 1.5% agarose gels and staining with ethidium bromide.

Western Blot

Western blots were performed on lysates generated from single cell suspensions of day 28 tumors as previously described (25). Primary blotting Abs were directed against STAT-1, STAT-3, pSTAT-1 and pSTAT-3 (all from Santa Cruz Biotechnology, San Jose, CA) and (control) β-actin (Cambridge, MA). After washed blots were incubated with horseradish peroxidase (HRP)-conjugated goat anti-rabbit antibody (Santa Cruz Biotechnology), blots were visualized by Western Lighting chemiluminescence detection kit (Perkin-Elmer, Waltham, MA) and exposed to X-Omat film (Eastman Kodak, New Haven, CT) for 30 sec to 2 min. Densitometry analysis was performed using a GDS 8000 bioimaging system and LabWorks4.6 software (UVP, LLC, Upland, CA).

Fluorescence Imaging of Tumor Sections

Tumor tissue samples were prepared and sectioned as previously reported (25). All washing steps were performed using wash buffer (WB; 0.5% BSA in PBS, Sigma-Aldrich). All blocking steps were performed using 2% BSA. For analysis of T cell subsets, sections were incubated with rat-anti-mouse-CD4 (BD Biosciences) followed by Goat-anti-Rat CY3 conjugated Fab 1 Fragments (Jackson ImmunoResearch, West Grove, PA). Sections were then incubated with Alexa 488-conjugated rat-anti-mouse-Foxp3 (eBioscience). MDSC were defined using FITC-conjugated rat anti-mouse-CD11b and PE-conjugated rat-anti-mouse-Gr1 mAbs (BD Biosciences). For analysis of the CD31 versus VCAM-1 or CXCL9/Mig rat-anti-mouse-CD31 (BD Biosciences) was combined with either goat-anti-mouse-VCAM-1 or goat-anti-mouse-CXCL9 (both from R&D Systems Inc., Minneapolis, MN). Each combination was then treated with a mixture of Alexa 488-conjugated donkey-anti-rat (Invitrogen) and CY3-conjugated donkey-anti-goat CY3 (Jackson ImmunoResearch). All sections were treated with DAPI (Sigma-Aldrich) and then mounted. Images were acquired using either an Olympus 1000 Scanning Confocal Microscope or an Olympus BX 51 Fluorescent Microscope (both from Olympus America, Melville, NY). Metamorph (Molecular Devices, Sunnyvale, CA) software was used for cell quantification.

Statistical analysis

Student t-test and one-way ANOVA were used to test for overall differences among the groups (StatMate III, ATMS Co., Tokyo, Japan), with a p value < 0.05 considered significant.

Combination therapy using SUT and specific DC-based vaccination provides superior anti-tumor efficacy against s.c. MO5 melanomas

C57BL/6 mice harboring established day 7 s.c. MO5 (B16.OVA) melanomas were left untreated or they were administered SUT (1 mg/day via oral gavage for 7 consecutive days) and/or a syngenic cellular vaccine (DC.IL12 pulsed with the OVA_257–264 and OVA_323–339 peptide epitopes) on days 7 and 14 post-tumor inoculation. As shown in Fig. 1A, a seven day course of SUT alone or a single course of DC-based VAC monotherapy yielded partial protection against melanoma progression (p < 0.05 versus the untreated control for all time points after day 10). We also observed that each combination therapy provided superior protection vs. either of the single modality treatments (p < 0.05 versus single modality therapy for time points after day 17). Notably, only the combination treatment (SUT/VAC) cohort contained regressor animals, with 24% (i.e. 12/50) of mice rendered tumor-free through day 60 post-tumor inoculation (Fig. 1A).

The impact of altering the dose of co-applied SUT or the schedule of SUT or VAC delivery on the anti-tumor efficacy of the combination therapy was next determined. As shown in Fig. 1B, all doses of SUT in excess of 0.05 mg/day resulted in an improvement in the anti-tumor effectiveness of DC-based vaccination (p < 0.05 versus VAC only for all time points after day 10). The timing of SUT administration in the context of vaccination appeared important, with superior efficacy demonstrated for combinational regimens in which SUT administration was initiated at the time of primary (day 7; p < 0.05 versus VAC only after day 10) or booster (day 14; p < 0.05 versus VAC only after day 14) vaccination (Fig. 1C). In contrast, if the application of SUT was delayed until day 21 (i.e. 7 days after the booster vaccination), only a modest deflection in tumor growth was observed (p < 0.05 versus VAC only on days 24 and 28). Data provided in Fig. 1D, suggest a major anti-tumor benefit for repeat vaccination vs. single vaccine administration in the context of SUT co-therapy (p < 0.05 for SUT/VAC day 7/14 versus SUT/VAC day 7 only for all time points after day 14). Furthermore, while a delay in the initiation of vaccination until day 14 (a day after SUT discontinuation) still resulted in a small degree of added protection to treated animals (p < 0.05 vs. SUT only on days 24 and 28), optimal co-therapy benefit was associated with the temporal overlap in administration of SUT and VAC (Fig. 1D; p < 0.05 for SUT/VAC d7/14 versus SUT/VAC d14/21 after day 10).

Combination SUT/VAC therapy elicits superior antigen-specific Type-1 T cell responses and losses in cells bearing a regulatory phenotype in the TME and TDLN in vivo

MO5-bearing mice were treated as outlined in Fig. 1A, with TIL, vaccine-draining lymph node (VDLN), tumor-draining lymph node (TDLN) and spleen cells harvested on day 28 post-tumor inoculation. MHC/peptide tetramer-based analyses of day 28 CD8⁺ T cells monitored by flow cytometry (Fig. 2A, 2B) revealed that treatment with SUT alone resulted in modest increases in the frequency and absolute numbers of OVA-specific CD8⁺ T cells in spleen, TDLN and tumor (p < 0.05 versus untreated). Treatment with VAC alone improved the prevalence of anti-OVA CD8⁺ T cells systemically (p < 0.05 versus untreated in all tissues analyzed). Combination SUT/VAC treatment yielded the highest frequencies of tetramer⁺ CD8⁺ T cells in all tissue compartments analyzed (p < 0.05 versus untreated or single modality therapy).

We next evaluated the ability of day 28 spleen, VDLN, TDLN and TIL CD8⁺ T cells to secrete IFN-γ in response to in vitro stimulation with the H-2K^b-presented OVA_257–264 peptide (Fig. 2C). Despite the potential for cross-priming CD8⁺ T cells reactive with OVA in MO5 (B16.OVA) tumor-bearing mice, anti-OVA Tc1 responses were negligible in all tissues analyzed from untreated animals. SUT or VAC monotherapy improved anti-OVA Tc1 responses systemically (p < 0.05 versus untreated controls in all tissue compartments evaluated), with superior antigen-specific responses observed in animals treated with combination SUT/VAC therapy (p < 0.05 versus SUT only, VAC only or untreated controls in all tissue compartments analyzed).

Previous reports have described the ability of SUT to reduce numbers of MDSC and Treg in the peripheral blood and TME of tumor-bearing hosts (12–14). Hence, we next asked whether single or combination modality treatment of MO5-bearing animals would alter the presence of cells bearing an MDSC-associated phenotype (i..e CD11b⁺Gr1⁺), a Treg-associated phenotype (i.e. CD4⁺Foxp3⁺), a mature DC-associated phenotype (i.e. CD11c⁺CD86⁺), as well as, total CD8⁺ and CD4⁺ T cells in the TME and TDLN through day 28 post-tumor inoculation. As shown in Fig. 3, while treatment of mice with SUT (or VAC) alone resulted in the partial, sustained reduction in the number of cells bearing MDSC (Fig. 3A, 3Ba–Bf) and Treg (Fig. 3A, 3Bg–Bk) phenotypes in the TME and TDLN, with a somewhat greater and more durable loss of these cell populations occurring in mice treated with the combination therapy. CD11b⁺Gr1⁺ cells in the TME were heterogeneous in their nuclear morphology, with some cells exhibiting multi- (i.e. PMN-like; Fig. 3Ba) and other cells displaying single- (i.e. monocyte-like; Fig. 3Bb) lobed nuclei. Combination SUT/VAC therapy was also superior in its capacity to acutely (within 3–7 days after initiating treatment) increase and sustain (through day 28; p < 0.05 versus single modality therapy after day 7) CD8⁺ T cell, CD4⁺ T cell and mature DC levels within the TME and TDLN (Fig. 3A).

Combination SUT/VAC therapy results in a switch from a non-Type-1- to a Type-1-biased immune profile in the TME

Given the suggestion of durable anti-tumor effects associated with day 7–14 treatment with SUT +/− VAC, we performed a more comprehensive analysis of immune parameters in the day 28 TME. Using RT-PCR and western blotting, we determined that SUT or VAC monotherapy, but even more so, SUT/VAC combination therapy resulted in a TME receptive towards Type-1 T cell recruitment (i.e. increased transcript expression for CXCR3 and the CXCR3 ligands CXCL9/Mig, CXCL10/IP-10, CXCL11/I-TAC; Fig. S1A) and functionality (i.e. increased transcripts for IFN-γ, IL-12p40 and T-bet and increased pSTAT1 protein expression; Fig. S1B, S1C, S1E). The TME post-treatment with SUT/VAC also could be characterized as being more refractory towards non-Type-1/regulatory cell recruitment (i.e. reduced transcript expression for known chemo-attractants for MDSC and Treg such as CXCL12/SDF-1α, CCL2/MCP-1, CXCL5/ENA-78, S100A8 and S100A9; Fig. S1A; refs. 29–36) and function (i.e. decreased expression of transcripts for IL-10, GATA-3 and Foxp3, as well as, the MDSC-associated gene products arginase-1, IDO-1 and iNOS/NOS2 (Fig. S1B and S1D) and reduced pSTAT3 protein expression; Fig. S1C, S1E). In contrast, expression of ROR-γt (a Th17-associated gene product) transcript levels appeared unaltered in the TME as a consequence of therapy (Fig. S1B). Additonal experiments support the TKI dose-dependency (for SUT administered at > 0.1 mg/day × 7 days; Fig. S2) of these molecular alterations in immune parameters. A confirmatory longitudinal evaluation of transcript changes in the TME suggests that single, and even more so, combination modality treatment results in the acute and sustained inhibition of many non-Type-1 immune indices (Fig. S3; p < 0.05 for SUT/VAC versus single modality therapy on day 28 for CXCR4, CXCL12, IL-10, S100A8, S100A9, Foxp3, Arg-1, IDO-1 and iNOS). A somewhat delayed kinetic (maximal by day 14–21 post-therapy initiation) was observed for improving many of the Type-1 immune indices in the TME (Fig. 4; p < 0.05 for SUT/VAC versus single modality therapy on day 28 for CXCR3, CXCL9, CXCL10, CXCL11, IFN-γ, Tbet and IL-12p40).

Previous work by our group and others provided support for the critical roles of VLA-4 (which binds to VCAM-1 expressed by activated endothelial cells) and CXCR3 (expressed by Type-1 CD8⁺ T cells) in the ability of protective anti-tumor T cells to effectively traffic into the TME (37–39). Consistent with the enhanced licensing of the day 28 TME for preferential recruitment of Type-1 T cells after SUT/VAC treatment, we observed remarkably enhanced expression of VCAM-1 (Fig. 5Aa–Ad; 5B, p < 0.05 for SUT/VAC versus all other cohorts, p < 0.05 for SUT or VAC versus untreated) and CXCL9 (a CXCR3 ligand; Fig. 5Ae–Am; 5C, p < 0.05 for SUT/VAC versus all other cohorts, p < 0.05 for SUT or VAC versus untreated) by tumor-associated CD31⁺ vascular endothelial cells in our fluorescence microscopy analyses. In the treated TME, we also observed CXCL9 to be expressed in situ by a population of “dendritic” cells (Fig. 5Ai) that are loosely-associated with the abluminal surface of tumor blood vessels (Fig. 5Ae–Am). Some, but not all, of these CD31^negCXCL9/Mig⁺ cells appear to represent NG2⁺SMA⁺ vascular pericytes (Fig. S4).

The anti-tumor efficacy of combination SUT/VAC therapy is negated upon administration of blocking antibodies against CXCR3 or VCAM-1

To address the biologic relevance of recruiting CXCR3⁺ (VLA-4⁺) anti-OVA Tc1 cells into the therapeutic TME, we injected MO5-bearing mice i.p. with blocking (non-depleting) antibodies against CXCR3 or VCAM-1, or the corresponding species/isotype control antibodies beginning on the day of initiating SUT/VAC treatment (i.e. 7 post-tumor inoculation). Antibodies were then re-injected every 2 days thereafter through day 15 (i.e. days 9, 11, 13 and 15) to sustain specific blockade. As shown in Fig. 6A, blockade of either CXCR3 or VCAM-1 completely abrogated protection associated with SUT/VAC treatment (p < 0.05 versus SUT/VAC and not significant versus untreated), while injection of control Abs had no discernable impact on therapy outcome. Flow cytometry-based OVA tetramer analyses performed on day 17 post-tumor inoculation revealed that CXCR3 or VCAM-1 blockade did not affect the ability of SUT/VAC combination therapy to expand the frequency of OVA-specific CD8⁺ T cells in the spleens or TDLN of treated animals (Fig. 6B). However, provision of the anti-CXCR3 or anti-VCAM-1 Abs, but not control Abs, prevented therapy-induced augmentation in the number of tetramer⁺CD8⁺ T cells in the day 17 TME (Fig. 6B). Interestingly, within 8 days of discontinuing the administration of anti-CXCR3 or anti-VCAM-1 Abs, these cohorts of animals began to display evidence for the recovery of SUT/VAC anti-tumor therapy benefit (Fig. 6A, day 23; *p < 0.05 versus untreated animals), which was associated with the renewed infiltration of tumor lesions by anti-OVA CD8⁺ T cells (Fig. 6C).