This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. The protocol was reviewed and approved by the Institutional Animal Care and Use Committee at the University of South Florida (#A4100-01). Mice were humanely euthanized by CO₂ inhalation according to the American Veterinary Medical Association Guidelines. Mice were observed daily and humanely euthanized if a solitary subcutaneous tumor exceeded 1.5cm in diameter or mice showed signs referable to metastatic cancer. All efforts were made to minimize suffering. Female C57BL.6 mice (6–8 weeks old) were purchased from Harlan Laboratories (Indianapolis, IN). Mice were housed at the Animal Research Facility of the H. Lee Moffitt Cancer Center and Research Institute.

B16 and MC38 murine colon cancer cell lines were maintained by serial in vitro passages in Complete medium (CM). CM consisted of RPMI 1640 supplemented with 10% heat-inactivated FCS, 0.1 mM nonessential amino acids, 1 mM sodium pyruvate, 2 mM fresh L-glutamine, 100 mg/ml streptomycin, 100 U/mL penicillin, 50 mg/mL gentamycin, 0.5 mg/mL fungizone (all from Life Technologies, Rockville, MD), and 0.05 mM 2-ME (Sigma-Aldrich, St. Louis, MO).

The monoclonal antibodies anti-LAG3 (clone C9B7W), anti-BTLA (clone 6A6), anti-CTLA-4 (clone 9H10), anti-PD-1 (clone RMP1-14), and anti PDL-1 (clone 10F.9G2) were all purchased from BioXcell (West Lebanon, NH). Normal ratIgG or hamster IgG was used as isotype controls.

A total of 1x10⁵ MC-38 colon cancer tumor cells were injected subcutaneously (s.c.) in C57BL/6 mice. Three days later, mice were injected with 300 ug of monoclonal antibodies (isotype, anti-PD-1, anti-PD-L1, anti-Lag-3, anti-CTLA-4 or anti-BTLA) intraperitoneally (i.p.). Mice continued to receive this treatment every 3–4 days until the tumor reached a size of 400mm². Tumor size was measured and recorded every two days. In another set of experiments, mice were euthanatized at day 21 after tumor injection. Tumors and splenocytes were harvested for in vitro assays.

Tumor cell suspensions were prepared from solid tumors by enzymatic digestion in HBSS (Life Technologies) containing 1 mg/ml collagenase, 0.1 mg/ml DNAse I, and 2.5 U/ml of hyaluronidase (all from Sigma-Aldrich) with constant stirring for 2 hours at room temperature. The resulting suspension was passed through a 70-um cell strainer, washed once with HBSS and resuspended in PBS + 3% BSA to a concentration of 1 x 10⁶ cells/ml for flow cytometric analysis. Cells were labeled with anti-CD90 microbeads according to the manufacturers’ instructions (Miltenyi Biotec) and purified using an autoMACS. After autoMACS purification, TIL were cultured for 5 days in the presence of IL-2 (3000 IU/ml). On day 5, TIL were collected for in vitro assays.

A ⁵¹Cr release assay was done as described previously [3]. MC-38 cells were used as targets and B16 tumor cells were used as a control. TIL purified from tumors were used as effector cells. T cell purity was measured by flow cytometry and cells were 95% positive for CD90 (data not shown). Briefly, target MC-38 or B16 tumor cells were labeled with 100 μCi of ⁵¹Cr (Amersham Corp.) in 0.2 mL of medium at 37°C in a 5% CO₂ atmosphere for one and a half hours. The labeled tumor cells were washed three times and added to the effector cells in triplicate wells of 96-well round-bottomed microplates at 50:1 and 25:1 effector to target ratios. After 5 hours, the percentage of specific ⁵¹Cr release was determined by the following equation: (experimental cpm − spontaneous cpm)/ (total cpm incorporated − spontaneous cpm) × 100. All determinations were done in triplicate, and the SE of all assays was calculated and was typically 5% of the mean or less.

Briefly, TIL were plated at 1x10⁴ and co-cultured with 1x 10³ irradiated MC-38 or B16 tumor cells and incubated for 48 hours at 37°C. The number of IFN-γproducing cells in response to stimulation was evaluated in an ELISPOT assay. The number of spots was counted in triplicates and calculated using an automatic ELISPOT counter.

Spleens or TIL were harvested under sterile conditions. Single-cell suspensions were prepared, and red blood cells were removed using ACK lysis buffer. For analysis of immune cell populations, one million cells (splenocytes or tumor digest suspension) were incubated for 20 minutes on ice in staining medium with relevant antibodies for surface expression analysis according to the manufacturer’s instructions (all from BD Biosciences). Samples were analyzed using an LSRII (BD Biosciences) and the data was analyzed using FlowJo software (Tree Star).

TIL were purified from either NrIgG or anti- PD-L1 antibody treated mice as described above. After autoMACS purification, TIL were cultured for 5 days in the presence of IL-2 (3000IU/ml) and TIL were used for in vivo studies. On day 4 following tumor injections, 5x10⁶ TIL per mouse were i.v. transferred. Beginning on day 4 and continuing every 12 hours for three days, mice also received 2.5e5 IU of IL-2 i.p. Following this treatment, tumor size was measured and recorded every 2 days.

A Mann-Whitney test (unpaired) or a Student’s t-test was used to compare between two treatment groups. All statistical evaluations of data were performed using Graph Pad Prism software. Statistical significance was achieved at p<0.05.

To examine the phenotype of infiltrating lymphocytes in tumors of B16 and MC-38-bearing mice, CD4+ and CD8+ T cells were analyzed by flow cytometry by gating on viable cells. As shown in Fig 1A and 1B, TIL isolated from MC-38 and B16 tumors had increased CD8+ T cell infiltration compared to CD4+ T cells. We next tested the tumor specificity and function of TIL isolated from MC-38 and B16 tumor bearing mice. TIL isolated from MC-38 (Fig 1C) and B16 (Fig 1D) when co-cultured in the presence of specific tumor cells had significant levels of IFN-γ production compared to irrelevant tumor cells. This data shows that TIL isolated from MC-38 and B16 tumors contain tumor-specific CD8+ T cells.

Inhibitory receptors such as PD-1, CTLA-4, Lag-3 and BTLA and expressed on T cells and ligands such as PD-L1 have been shown to contribute to immune mediated suppression. We examined the expression of inhibitory immune checkpoint receptors on CD8+ T cells isolated from B16 and MC-38 tumors and measured expression of PD-1, CTLA-4, BTLA, and LAG-3. We also measured 41BB expression on TIL to determine activation status (Fig 2A and 2B). Reports have shown that PD-L1 expression on tumor cells mediates negative signaling through PD-1 interaction on T cells. Previous study from our lab has shown that B16 tumor cells express PD-L1 on their surface [23]. We examined PD-L1 expression on MC-38 tumor cells. As shown in Fig 2C, MC-38 expresses PD-L1.

We next examined the feasibility of expanding TIL in vitro from B16 and MC-38 tumors. C57/BL6 mice were injected with B16 or MC-38 and tumors were collected on day 21. Tumor cell suspensions were prepared from tumors by enzymatic digestion as described in methods section. Cells were labeled with anti-CD90 microbeads and purified using an autoMACS. After autoMACS purification, TIL were cultured for 5 days in the presence of IL-2 (3000 IU/ml). T cell purity was checked by flow cytometry and > 95% cells were positive for CD3. On day 5, TIL were collected, counted and used for in vitro assays TIL from MC-38 tumor expanded two-fold after 5 days of culture in media containing IL-2 (data not shown). However, we were not able to expand B16 TIL in culture for 5 days. Hence we utilized MC-38 tumor model for our TIL studies. An ELISPOT assay was performed to determine the number of IFN-γ producing TIL in response to stimulation by the MC-38 or irrelevant tumor cells. We compared freshly isolated TIL and TIL that were cultured for 5 days in media containing 3000 IU/ml IL-2. As shown in Fig 2D, both fresh and cultured TIL secreted IFN-γ in response to MC-38 cells. Significantly lower secretion of IFN-γ was measured in response to irrelevant B16 cells. Higher production of IFN-γ was measured in TIL cultured in the presence of IL-2 compared to fresh TIL (p<0.05). Next, we evaluated the ability of expanded TIL to mediate specific cell lysis in vitro. TIL isolated from MC-38 tumor bearing mice were cultured in vitro for 5 days in the presence of IL-2. On day 5, TIL were co-cultured with ⁵¹Cr-labeled MC-38 or B16 cells at effector to target ratios of 50:1 and 25:1. TIL exhibited cytotoxicity against MC-38 tumor cells at a 50:1 and 25:1 ratio and were tumor-specific (Fig 2E, p<0.001 compared to killing of irrelevant B16 cells). This data shows that cultured TIL mediates specific cytotoxicity against MC-38 cells.

Based on the expression levels of inhibitory immune checkpoints, we examined whether blockade of these inhibitory immune checkpoints had an effect on TIL infiltration and tumor growth. As shown in Fig 3A, antibody treatment with anti-CTLA-4 (p<0.02), anti-BTLA (p<0.01), anti- PD-L1 (p<0.01), anti- Lag-3 (p<0.001) and anti- PD-1 antibodies (p<0.02) led to a modest but a significant delay in tumor growth in MC-38 bearing mice (p values compared to mice that received isotype control antibody). We next investigated whether blockade of individual co-inhibitory molecules had an effect on TIL infiltration that could contribute to the observed delay in MC-38 tumor growth. Tumors were collected on day 21 and CD8+ T cells within the TIL population were measured by flow cytometry. Treatment with anti-BTLA or anti-PD-L1, anti-Lag3 and anti-PD-1 led to a significant increase in CD8+ T cell infiltration compared to mice treated with isotype control antibody (Fig 3B). Tumor bearing mice that received anti-PD-L1 antibody demonstrated the most significant increase in T cell infiltration (p<0.001, compared isotype control). There was no significant increase in T cell infiltration in anti-CTLA-4 treated mice. Together, this data suggests that PD-L1 blockade is an effective strategy to increase CD8+ T cell infiltration in MC-38 tumors.

Since PD-L1 blockade led to a significant increase in T cell infiltration, we next examined whether PD-L1 blockade enhanced tumor-specific immune responses in MC-38 tumor bearing mice. Administration of anti-PD-L1 antibody enhanced the percentage of CD8+ T cells in the spleen (15.3%) and tumor (28.5%) compared to 5.6% of CD8+ T cells in the spleen and 17.6% of CD8+ T cells in tumor of mice treated with NrIgG (Fig 4A, n = 8). Similarly, anti-PD-L1 antibody treated mice had increased levels of CD4+ T cells in the spleen compared to NrIgG treated mice. No significant difference was measured in CD4+ T cell infiltration into tumors between mice treated with anti-PD-L1 antibody or NrIgG. Immunohistochemical staining revealed an increase in CD3+ T cell infiltration in the tumors of PD-L1 treated mice (Fig 4B, n = 8).

We next measured IFN-γ production in response to restimulation with tumor cells as a measure of T cell function. TIL isolated from tumors of NrIgG or anti-PD-L1 antibody treated mice were restimulated with MC-38 or irrelevant B16 cells for 48 hours. As shown in Fig 4C, TIL from anti-PD-L1 antibody treated tumor bearing mice demonstrated increased IFN-γ production in response to MC-38 cells (p<0.0002) compared to the TIL from NrIgG treated mice. This data suggests that PD-L1 blockade enhances TIL function in MC-38 tumor-bearing mice.

We next examined whether PD-L1 blockade improves the cytotoxic function of TIL. TIL isolated from tumors of anti-PD-L1 antibody or NrIgG treated mice were cultured in the presence of 3000IU/ml of IL-2. On day 5, TIL were analyzed for their cytotoxic function using a ⁵¹Cr release assay. Purified TIL from anti-PD-L1 antibody treated mice co-cultured with ⁵¹Cr-labeled MC-38 tumor cells had higher cytotoxicity (27%) at 25:1 and 10:1 (22%) ratios compared to the TIL from tumors of NrIgG treated mice (18% at 25:1 and 8%. at 10:1 ratio) (Fig 4D, p<0.01). No killing of irrelevant B16 cells was measured in either group. This data supports PD-L1 blockade to improve the cytotoxic function of TIL.

Next, we investigated PD-L1 blockade for the generation of TIL for adoptive cell therapy. Mice were injected s.c. with MC-38 tumor cells on day 0 followed by total body irradiation (TBI) with 600 rad on day 3. Cultured TIL from anti-PD-L1 or NrIgG antibody treated mice were adoptively transferred on day 4. As shown in Fig 5A, mice that received TIL derived from either NrIgG or anti-PD-L1 treated mice demonstrated a significant delay in tumor growth compared to the control mice that did not receive any TIL therapy (p<0.01). However, adoptive transfer of TIL from anti-PD-L1 antibody treated mice led to a significant delay in tumor growth and enhanced survival compared to mice that received TIL from NrIgG treated mice (Fig 5B, p<0.002). This data demonstrates that PD-L1 blockade improves TIL function that may enhance anti-tumor immune responses after adoptive transfer.