Blockade of surface-bound TGF-β on regulatory T cells abrogates suppression of effector T cell function in the tumor microenvironment.

Mice

Mouse experiments were performed in accordance with institutional guidelines under a protocol approved by the Memorial Sloan-Kettering Cancer Center Institutional Animal Care and Use Committee. All mice were maintained in a pathogen-free facility according to National Institutes of Health Animal Care guidelines. C57BL/6J mice (females, 6 to 10 weeks old) and OT-1 TCR transgenic mice ( 35 ) were purchased from The Jackson Laboratory. Pmel-1 TCR transgenic mice ( 36 ) were obtained from N. Restifo (National Institutes of Health). Foxp3-GFP knock-in mice were a gift from A. Rudensky (Memorial Sloan-Kettering Cancer Center). Foxp3-DTR (Foxp3-GDL) mice were a gift from G. Hämmerling (DKFZ). CCR2-DTR mice were generated by T. Hohl (Memorial Sloan-Kettering Cancer Center). Cell lines and tumor challenge The B16-F10 mouse melanoma line was originally obtained from I. Fidler (M.D. Anderson Cancer Center, Houston, TX). These cells were maintained in RPMI-1640 medium containing 7.5% fetal bovine serum (FBS) and L-glutamine. B16F10 cells were transfected with plasmid encoding full-length ovalbumin (OVA) protein to generate B16-OVA cells as previously described ( 37 ). YFP-B16 cells used for imaging experiments were generated as previously described ( 10 ). Tumor cells were maintained in RPMI-1640 medium containing 7.5% FBS. For B16-OVA and YFP-B16 cells, the growth medium was supplemented with G418 (0.5 mg/ml). For tumor challenge experiments, 1 × 10 5 viable B16-OVA cells in 100 μl of phosphate-buffered saline (PBS) were injected intradermally into the right flank of C57BL/6 mice. For ex vivo analysis of immune infiltrates, mice were injected subcutaneously with the indicated numbers of tumor cells reconstituted in 150 μl of growth factor–reduced Matrigel (BD Biosciences). In vitro activation of OT-1 and Pmel CD8 + T cells OT-1 CD8 + T cells express a transgene encoding a TCR that specifically recognizes the OVA peptide (Ser-Ile-Ile-Asn-Phe-Glu-Lys-Leu) in the context of mouse MHC-I H-2k b ( 35 ). Pmel-1 CD8 + T cells express a transgene encoding a TCR that specifically recognizes the Pmel-1 (gp100) peptide (Glu-Gly-Ser-Arg-Asn-Gln-Asp-Trp-Leu) in the context of mouse MHC-I H-2D b ( 36 ). Activated OT-1 or Pmel-1 T cells were generated by incubation of peptide-pulsed mouse splenocytes (5 × 10 6 cells/ml) in vitro for 5 to 7 days in the presence of IL-2. Briefly, a mouse spleen was homogenized to generate a single-cell suspension and the released cells were pelleted and resuspended in 3 ml of ACK lysis buffer (Lonza) for 1 min to lyse red blood cells. The splenocytes were washed, resuspended at 5 × 10 6 cells/ml in T cell growth medium [RPMI1640, 100 U/ml penicillin, streptomycin (100 mg/ml), 10% FBS, 2 mM L-glutamine, 50 μM 2-mercaptoethanol, and 1 mM sodium pyruvate) containing OVA peptide or gp100 peptide (0.75 μg/ml) and incubated at 37°C in a 95% air and 5% CO 2 humidified atmosphere. On days 3 and 5, 25 ml of fresh T cell growth medium containing recombinant mouse IL-2 (20 U/ml, eBioscience) was added to the cultures. On day 7, viable cells were purified by centrifugation at 400 g for 30 min at room temperature over a Histopaque gradient (density = 1.083, Sigma-Aldrich). This method yielded antigen-specific CD8 + T cells that were 90 to 95% tetramer + for their respective peptides.

Collagen-fibrin gel killing assay

The collagen-fibrin gel-based killing assay has been previously described in depth ( 11 ). We adapted this assay to examine the killing of ex vivo B16 tumors. Briefly, C57BL/6 mice (6- to 8-weeks old) were tumor challenged with 1 × 10 5 viable B16-OVA cells intradermally on the right flank. Tumors were excised on day 10 or 11 and dissected into smaller pieces. The tumors were then incubated for 5 min in collagenase (250 μg/ml) in PBS containing Ca 2+ and Mg 2+ before being homogenized through 70-μm mesh cell strainers to generate single-cell suspensions. The number of viable tumor cells and immune infiltrates were assessed using a hemocytometer and trypan blue exclusion. The fraction of immune infiltrates within the dissociated tumors was confirmed by flow cytometry with an anti-CD45 antibody.1 × 10 4 viable tumor cells (together with all infiltrating cells) were co-embedded with or without 5 × 10 5 in vitro–activated CD8 + T cells into collagen-fibrin gels (0.1 ml volume). As a control for each experiment, 1 × 10 4 viable B16-OVA cells cultured in vitro were also co-embedded with or without 5 × 10 5 in vitro–activated CD8 + T cells in collagen-fibrin gels. Duplicate gels were lysed daily with collagenase and trypsin for up to 3 days. The viable tumor cells from dissolved gels were diluted and plated in 6-well plates for colony formation. Seven days later, plates were fixed with 3.7% formaldehyde and stained with 2% methylene blue. Colonies were manually counted to assess the number of cells. For experiments in which CD8 + T cells were analyzed by flow cytometry, collagen-fibrin gels were lysed with collagenase only, which was followed by mechanical pipetting to fully dissolve the gels and recover single-cell suspensions of T cells. Depleting T regs in vivo In experiments in which diphtheria toxin (DT) was used to deplete immune cell subsets in vivo, Foxp3-DTR and CCR2-DTR mice were injected i.p. with 45 ng of DT in 0.2 ml of PBS for the depletion of either Foxp3 + cells or CCR2 + cells, respectively. For all experiments, DT was administered on day 8 or 9 after tumor inoculation and tumors were excised 48 hours later. Calculating the value for k k was calculated according to the following equation: bt = b 0 e −kpt+gt where bt = the concentration of B16 cells at time t ; b0 = the concentration of B16 cells at time 0; k = the killing rate constant (or killing efficiency) for CD8 + T cells; p = the concentration of CD8 + T cells; and g = the growth rate constant for B16 cells ( 11 ). Experimentally determined values are used to calculate k .

Purification of T regs

In some experiments, T regs were purified with MACS beads. B16-OVA tumors were excised on days 10 or 11 after tumor challenge and were dissociated as described earlier. T regs were purified from dissociated tumors in vitro by magnetic bead separation with the CD4 + CD25 + Regulatory T Cell Isolation Kit (Miltenyi). The purity of these cells was confirmed by flow cytometric analysis with fluorophore-conjugated antibodies against CD4, CD25, and Foxp3. In other experiments, T regs were purified by FACS sorting. Foxp3-GFP mice were challenged with 1 × 10 5 B16-OVA tumor cells. On day 10 or 11, the B16-OVA tumors were excised and dissociated as described earlier. T regs were sorted based on viable CD4 + GFP + cells on a Cytomation MoFlo or BD FACS Aria cell sorter in the MSKCC Flow Cytometry Core Facility. Flow cytometric analysis of cell surface antigens and intracellular proteins Cell suspensions were incubated in Fc-block (anti-CD16 and anti-CD32 antibodies, BD Biosciences) for 20 min on ice in FACS buffer (PBS containing 0.5% BSA and 2mM EDTA) before being stained for cell surface markers. Samples were incubated with fluorophore-conjugated antibodies against CD4, CD8, CD25, PD-1, and TGF-β (clone 1D11) for 20 to 30 min and then were washed three times with FACS buffer. The Foxp3 Staining Kit (eBioscience) was used for the intracellular staining of Foxp3 and Granzyme B. Dead cells were excluded from the analysis with the Fixable Viability Dye eFluor 506 (eBioscience). Samples were acquired on 12-color LSRII cytometer and data were analyzed with FlowJo software (Tree Star).

Intravital imaging

YFP-B16 tumors were injected in the left flank of either wild-type (WT) or Foxp3-GFP mice upstream of the inguinal lymph node (LN). The mice were imaged at multiple time points to find the time of maximal infiltration and compensate for variability associated with each set of tumor injections, priming response, and 3D tumor structures. Seven days after the transfer of fluorescently labeled CD8 + T cells, the mice were anesthetized with 1.5% isoflurane given concurrently with 1L per min O 2 . Each mouse was then placed on a heated platform maintained at 37°C. Surgery was performed to open up a skin flap, extending from the fore limbs to the hind limbs, up to the ventral midline, exposing the tumor and inguinal LN while maintaining vasculature integrity. The tumor and tumor-draining lymph node (TDLN) were then isolated under nylon washer mounted coverslips with PBS and visualized with a heated (37°C) water dipping 40× objective lens (Nikon). The temperature of the isolated tissues was checked with a thermal probe to ensure it was maintained at 37°C. Time-lapse images are acquired with a Z-depth on the average of 100 to 150 μm with 3 μm between steps, starting at ± 10 μm from the top edge of the tumor cells. Mosaic images were taken with 50-μm overlaps between adjacent regions. The video capture rate of over 20 fps enabled 6:1 frame averaging with a sample area that included up to 9 adjacent 270 μm × 270 μm × 100 μm volumes to produce a mosaic image every 80 to 120 s. Time-lapse images varied in length from 60 to 240 min with mosaic images taken for as long as possible.

Image analysis

Images were analyzed with Volocity 4.0.2 software (Improvision) and custom-developed Matlab code. Mosaic images were compiled together with Matlab before being imported into Volocity. T cell tracking was performed on individual quadrants in Volocity. Images were corrected for contrast with 3 × 3 × 3 pixel noise filtering to remove background signal where necessary. Tracks were calculated with Volocity automatic object acquisition and tracking modules and were verified for algorithmic errors. Image drift was removed from the calculated trajectory and velocity measurements by calculating the average movement for 3 tumor landmarks per image during the time lapse and adjusting the cell tracking measurement accordingly. Intra-tumor T cell positions were calculated by producing a high digital threshold map of the tumor images and then comparing Volocity-calculated cell centroid positions with the tumor map to determine cell location with respect to tumor or “not tumor” using Matlab.

Statistical comparisons of Pmel1 vs

OT1 were performed with Graphpad Prism 5 software with a student’s t test.

T reg cell proximity and contact score generation

During the verification of trajectory measurements for Pmel1 T cells in Foxp3-GFP mice, each cell was manually assessed in XY and Z for interactions with T regs . Cells received a score of 1 for each contact with or proximity to (within 10 μm) each T reg , with additional interactions per time point being additive. Scores were normalized by dividing the sum of the interactions by the number of time points for which an individual cell was tracked. Scores produced are a weighted time average.

Statistical analysis

Unless otherwise indicated, all experiments were performed at least three times with duplicate samples. Data are reported as means ± SEM for the number of experiments indicated. For statistical analyses, a Kruskal-Wallis test (nonparametric equivalent of analysis of variance) was applied when there were more than two groups. If statistically significant, pairwise comparisons with Wilcoxon test and Bonferroni correction for multiple comparisons was applied.

Supplementary Material Budhu et al, 2017 suppl Fig. S1.

Analysis of immune cell infiltrates in tumors from DT-treated

Foxp3-DTR mice. Fig. S2. In vivo depletion of CCR2 + cells in CCR2-DTR mice has no effect on the suppression of CD8 + T cell–mediated killing by the tumor microenvironment. Fig. S3. Depleting T regs ex vivo with anti-CD25 microbeads has no effect on the immunosuppression of CD8 + T cells. Fig. S4. T regs do not affect the mobility of CD8 + T cells in the tumor. Fig. S5. Suppression of CD8 + T cells by T regs is contact- or proximity-dependent. Fig. S6. Expression of TGF-β and CD51 (α V integrin) in immune cell subsets from the tumors and spleens of B16 tumor–bearing mice. Fig. S7. Effect of DT on the expression of PD-1 and granzyme B on the surface of endogenous CD8 + T cells. Movie S1 Movie S1. CFP-Pmel T cells are found in regions highly infiltrated by T regs . Movie S2 Movie S2. CFP-Pmel T cells are found within proximity to or make contact with T regs .