Fig 1.
CD20-TCB mediates efficient tumor cell killing through stable synapse formation.
a) Viability assay by AnnexinV (AnnV+) and Propidium Iodide (PI+) staining on WSU DLCL2 and OCI-Ly18 cells, 16 hours after co-culture with CD3/CD28 activated CD8+ T cells in the presence of the indicated CD20-TCB concentrations. n = 4 per point. Mean and +/- s.d. are shown. 2way-Anova, *p = 0.05, **p = 0.005, *** p<0.0005. b) 3D reconstruction of representative confocal imaging of synapse formation between CD8+ T cell (green) and tumor cell (blue), mediated by CD20-TCB treatment (red). c) 3D reconstruction of representative confocal imaging of Granzyme B (red) polarization toward the synapse between CD8+ T cell (green) and tumor cells (blue), mediated by CD20-TCB treatment (white). d-e) Confocal live cell imaging quantification of (d) Percentage of T cell-tumor cell contacts lasting more than 20 minutes; (e) CD8+ T cells speed (μm/min). Cells have been treated with +/- CD20-TCB (200 ng/ml), in the presence of WSU DLCL2 or OCI-Ly18 cells as target. Mean and +/- s.d. are shown, n = 3. 2way-Anova, **p<0.005; ****p<0.0001; n.s.: not significant. f-h) Confocal live cell imaging quantification of (f) Percentage of T cell-WSU DLCL2 contacts lasting more than 20 minutes; (g) contact area between T cells and WSU DLCL2, normalized by the total surface of tumor cells and (h) CD8+ T cells speed (μm/min). Cells have been treated with CD20-TCB at the indicated doses. Mean and +/- s.d. are shown, n = 3. 2way-Anova, *p<0.05; **p<0.005; ****p<0.0001; n.s.: not significant. i) Representative confocal imaging of LFA-1 (white) localization at the synapse between T cell (F-actin is shown in green) and target cell (blue). j) Viability assay by AnnV+ and PI+ staining of WSU DLCL2 cells, 16 hours after co-culture with CD3/CD28 activated CD8+ T cells in the presence of the indicated CD20-TCB concentrations and +/- LFA-1 inhibitor (10 μg/ml) or +/- FAS-L inhibitor (10 μg/ml). n = 4 per point. Mean and +/- s.d. are shown. 2way-Anova, *** p<0.0005, **** p<0.0001.
Fig 2.
A xenoreaction-free model to quantify T cell dynamics in response to therapy.
a) Workflow schematics: skinfold chambers were installed on NSG mice (day -2). 48 hours later (day 0): WSU DLCL2 (blue), unstained CT26 cells, and CD2+ T cells (pink) freshly purified from human PBMCs or from HSC-NSG mice were injected intra-dermally in the skinfold chamber together with labeled CD20-TCB (0.5 mg/kg) or with suitable vehicle. Cells were imaged 2 hours post treatment by MP-IVM. Adapted from https://smart.servier.com/. b) 3D representative rendering of MP-IVM imaging on skinfold chamber of HSC-NSG-NSG mice showing localization of therapy (white) at the contact site between WSU DLCL2 cells (blue) and T cells (pink), 2 hours post treatment. c) MP-IVM analysis of T cells tracks in the skin fold chamber of PBMC-NSG (top) vs HSC-NSG-NSG mice (bottom), +/- CD20-TCB. T cell tracks are plotted according to their displacement in the X and Y axes. Total number of tracks for each plot is: Top left: Vehicle n = 330. Top right: CD20-TCB n = 759. Bottom left: Vehicle n = 741. Bottom right: CD20-TCB n = 185. d-e) Quantification of (d) Track Speed (μm/min) and (e) Track displacement (μm) of T cells in PBMC-NSG or HSC-NSG-NSG mice, +/- CD20-TCB. Shown in yellow are mean values +/- s.d. Unpaired t-test; ****p<0.0001; n.s.: not significant.
Table 1.
Reference average values of T cell dynamic indexes in tumors treated with 0.5 mg/kg CD20-TCB intravenously or with suitable vehicle, at early (less than 24 h) or late (more than 24h) time points.
Fig 3.
CD20-TCB induces early and lasting T cell engagement and activity.
a) Time course quantification of Speed (μm/min), Track displacement (μm) and Arrest coefficient (%) of HSC-NSG-derived T cells co-injected with WSU DLCL2 tumor cells in NSG mice, treated with 0.5 mg/kg CD20-TCB i.v. or suitable vehicle and imaged starting from 30 minutes after treatment for 2 hours. Statistics: Kruskal-Wallis test, **** Adj. P value < 0.0001. Shown are mean values +/- SEM b) Time course tracking of T cells after intravenous injection of 0.5 mg/kg CD20-TCB. Highlighted are the 3 different possible behaviors of T cells: i) This T cell is moving fast at the beginning, with a straight trajectory. When it encounters a tumor cell in the presence of therapy, it suddenly stops, and starts interacting (light blue arrow). ii) This cell is interacting with the tumor since the beginning, its track is revolving around the same coordinates (yellow arrow). iii) This cell does not interact with the environment (light blue arrow). WSU DLCL2 cells are represented in blue, T cell in pink. Tracks are shown for each T cell in the field. c) Shown are T cell tracks, plotted according to their displacement in the X and Y axes. Comparison of tracks of T cells treated intravenously with 0.5 mg/kg CD20-TCB or with suitable vehicle. Each animal has been imaged over time for 1h at 24, 48 or 72h after treatment. 24h: Vehicle n = 403, CD20-TCB n = 348. 48h: Vehicle n = 195, CD20-TCB n = 49. 72h: Vehicle n = 353, CD20TCB n = 127. d-g) Quantification of (d) Speed (μm/min); (e) Track displacement (μm); (f) Arrest coefficient (%). (g) Contact duration (minutes) of T cells with tumor cells and percentage of contacts lasting more than 20 minutes (green boxes); T cells were derived from HSC-NSG spleens and co-injected with WSU DLCL2 tumor cells, treated with 0.5 mg/kg CD20-TCB i.v. or suitable vehicle and imaged 24, 48 or 72 hours post treatment. Statistics: Kruskal-Wallis test, **** Adj. P value < 0.0001, ** Adj. P value < 0.005. Shown are scattered plots and means +/- s.d.
Fig 4.
CD20-TCB induces resident T cell proliferation and recruitment of peripheral blood T cells.
a-c) Top: Representative histological staining of WSU DLCL2 tumors 24h post second treatment (0.5 mg/kg CD20-TCB or suitable vehicle i.v.). Bottom: Quantification of total number of cells/mm2 from histological images of vehicle vs CD20-TCB treatment. Whole slide scans quantification of 4 μm FFPE sections with the software (a) Definiens; (b-c) Halo. Statistical analysis: Unpaired 2-tailed t-test with Welch’s correction. *p<0.05, **p<0.005 (a) Red: CD3 staining, brown: CD31 staining. Quantification: Number of CD3+ cells b) red: Ki67, yellow: CD3, blue: DAPI. Quantification: Number of CD3+ Ki67+ cells c) Red: CXCR3, yellow: CD3, Blue: DAPI. Quantification of CD3+ CXCR3+ T cells. d) Percentage of proliferating CD8+ T cells, as assessed by CFSE dilution, freshly purified from PBMCs. Proliferation has been evaluated at 24h, 48h and 72h post CD20-TCB treatment, at the indicated doses, in the presence of WSU DLCL2 cells as target. n = 3 per group, mean and s.d. are shown. One-way Anova, *p<0.05, **p<0.005, ****p<0.0001. e) Workflow schematics: Skinfold chamber were installed on NSG mice. 48h later, WSU DLCL2 (Blue), unstained CT26 cells, and CD2+ T cells freshly purified from HSC-NSG spleens (pink) were injected intra-dermally in the skinfold chamber, together with 0.25 mg/kg of CD20-TCB or with suitable vehicle. Concomitantly, freshly purified CD2+ T cells from HSC-NSG spleens (orange) were injected i.v. to allow visualization of peripheral blood T cells. Cells were imaged 72h post treatment by MP-IVM. f) Representative MP-IVM imaging of the tumors. Blue: WSU DLCL2 cells; Pink: Resident T cells; Orange: Recruited T cells. Images were acquired 72h post intradermal treatment with 0.25 mg/kg CD20-TCB or suitable vehicle. Adapted from https://smart.servier.com/ g) Quantification of peripheral T cells (number/mm2) 72h post treatment. Mean +/- s.d. are shown. Unpaired 2-tailed t-test with Welch’s correction. **p<0.005. h) In the context of the skinfold chamber model, increasing number of T cells (Resident) were co-injected with the tumor and 0.25 mg/kg of CD20-TCB intradermally, while 2.5*106 T cells were injected intravenously (Peripheral). 72h post treatment, peripheral blood T cells were counted for each tumor from 5 representative fields. 4 tumors per group were analyzed. Shown is the count of peripheral T cells/mm2, Mean +/- s.d. per group. Statistical analysis: One-way Anova. **** p<0.0001. i) 3 hours in vitro chemotaxis assay of T cells toward preconditioned medium derived from WSU DLCL2 co-culture with CD3/CD28 pre-activated T cells. Pre-activated CD8 T cells have been plated with WSU DLCL2 cells at decreasing T cells: Tumor cells ratios, in the presence of 200 ng/ml of CD20-TCB. 24h later the supernatant has been collected and transferred to the bottom chamber of a 24-Transwell plate. In the top chamber 100.000 pre activated T cells, labeled with CFSE, have been seeded and let to migrate for 3 hours. Migration has been evaluated by counting total amount of CFSE positive migrated cells in the bottom chamber, by flow cytometry at constant volume and acquisition speed. Mean fold change and +/- s.d. are shown. n = 5, from two independent experiments 2-way Anova; **p<0.005.
Fig 5.
CD20-TCB-induced T cell recruitment is dependent on IFNγ and CXCL10.
a) IFNγ protein quantification by multiplex analysis of supernatant derived from co-culture of WSU DLCL2 cells with CD8+ T cells freshly purified from PBMCs and stimulated with CD20-TCB (200 ng/ml) at the indicated time points. n = 3 per group. Two-way Anova. *p< 0.05, **p< 0.05, ***p < 0.001, ****p < 0.0001, n.s.: not significant. b) Flow cytometry analysis of CD8+ INFγ+ T cells, at 24, 48 and 72 hours post in vitro CD20-TCB (200 ng/ml) treatment. c) Quantification of released cytokines (pg/mL) upon IFNγ stimulation of WSU DLCL2 tumor cells for 48 hours. Shown is Mean +/- s.d. of 3 replicates. d) Quantification of released cytokines (pg/mL) upon CD20-TCB treatment (200 ng/ml) of WSU DLCL2 co-cultured with CD3/CD28 pre-activated CD8+ T cells. Shown is Mean +/- s.d. n = 4 per group. Two-way Anova *p< 0.05, n.s.: not significant. e) CXCL10 protein quantification by multiplex analysis of supernant derived from co-culture of WSU DLCL2 cells with CD8+ T cells freshly purified from PBMCs and stimulated with CD20-TCB (200 ng/ml) at the indicated time points. n = 3 per group. Two-way Anova. ***p < 0.001, ****p < 0.0001, n.s. not significant. f) Representative images from MP-IVM imaging in the skinfold chamber of HSC-NSG-NSG mice. WSU DLCL2 cells (blue) pre-treated or not with IFNγ were injected intra-dermally together with CD2+ T cells derived from the spleen of HSC-NSG (pink) and 0.25 mg/kg CD20-TCB or suitable vehicle. CD2+ T cells derived from the spleen of HSC-NSG (orange) where concomitantly injected intravenously. Top row: Tumor cells (blue) and peripheral T cells (orange). Bottom row: Tumor cells (blue), resident T cells (pink) and peripheral T cells (orange). Where indicated, antibodies against CXCL10 or IFNγ were injected intravenously. g) Quantification of peripheral blood T cells (count/mm2) 72 hours post treatment. Shown are individual counts/mm2 and mean +/- s.d. Statistical analysis: One-way Anova. ***p < 0.001, ****p < 0.0001.
Fig 6.
Inhibition of CXCL10 or IFNγ impairs CD20-TCB-dependent T cell infiltration.
a) Subcutaneous WSU DLCL2-derived tumors between 200 and 800 mm3 were intravenously treated with 0.5 mg/kg CD20-TCB in the presence of αCXCL10 or αIFNγ antibodies (10 mg/kg), or with suitable vehicle, once per week. n = 9. Shown are mean values +/- s.d. for tumor volume over time. Unpaired t-test, *p<0.05. b) Representative histology of tumors at termination. Blue: Nuclear counterstain, brown: CD3. c) Quantification of infiltration at termination (CD3+ T cells per tumor area). Statistics: One-way Anova, **p< 0.01, ***p<0.005. d) Correlation between infiltration and tumor growth.