Article Figures & Data
Figures
- Figure 1. Luc‐CTL assay design used for identification of immune‐checkpoint molecules
Assay principle: RNAi was performed with luciferase‐expressing cells that were challenged with or without CTLs and bi‐specific antibody. Before readout, cell supernatant was removed and the remaining intact cells were lysed to measure the residual‐cell‐associated luciferase. To identify immune‐checkpoint regulators, the difference between normalized luciferase measurements for conditions with CTLs and without CTLs was calculated. siRNAs enhancing CTL cytotoxicity would only reduce normalized luciferase levels under conditions with CTLs; hence, the difference between luciferase measurements will be > 0.
Luc‐CTL assay performed at different T cell to MCF7 cell ratio with PBMC‐derived CD8+ T cells and anti‐EpCAM x CD3 bi‐specific antibody (○). Anti‐CD19 × CD3 bi‐specific Ab (■) was used as a specificity control since CD19 is a B‐lymphocyte‐specific antigen and therefore this bsAb fails to cross‐link tumor to T cells. Lower luciferase intensity indicates higher lysis. Error bars denote ± SEM.
Luc‐CTL assay was performed with MCF7 cells transfected with control or PD‐L1‐specific siRNAs and cocultured with or without CTLs and bsAb. For each condition, the luciferase activity of PD‐L1‐siRNA‐treated cells was normalized to that of the control siRNA; n = 8. Error bars denote ± SEM.
Comparison between the Luc‐CTL assay (■) and the classical chromium (Cr)‐release assay (○) with MCF7 as target cells and survivin‐specific T cells as effector cells at varying effector to target (E:T) ratios. Error bars denote ± SEM.
- Figure 2. Layout and analysis of the RNAi screen used to identify immune‐modulatory tumor genes
-
A. Workflow and experimental layout of the RNAi screen which was performed thrice, each time in duplicates, along with an additional CTG‐based viability screen to filter out lethal genes from the final hit list. Hits were analyzed after data normalization using the cellHTS2 package.
-
B. Correlation between replicates for screen 2 for both viability and toxicity sets, as determined by the Pearson rank correlation test.
-
C–E Graphical summary of gene function related to modification of T cell‐mediated tumor lysis and cell viability for screens 1, 2, and 3, respectively. Positive score = reduced cancer cell viability; negative score = increased viability. X‐axis: influence on cell viability without addition of T cells. Y‐axis: influence on cell viability with addition of T cells. Appropriate immune‐suppressive (PD‐L1, CEACAM‐6, GAL‐3) and lethality (UBC, PLK‐1) controls, along with few positive (GRM4, GRK5) and negative (CCR9, IL8) candidate immune‐modulatory hits are highlighted herein.
-
- Figure 3. Heat map representation of potential positive and negative immune modulators identified from the RNAi screens
Differential scores were used to identify positive immune modulators (yellow) the knockdown of which enhance CTL‐mediated cell killing and negative immune modulators (blue) the knockdown of which reduce CTL‐mediated cell killing. Differential scores prior to filtering are shown for all genes tested in the 3 different screens (see Materials and Methods). Selected representative clusters of high‐confidence hits are displayed herein.
- Figure 4. CCR9 knockdown sensitizes tumor cells to immune attack
-
A. MCF7 cells were transfected with the described siRNA sequences and harvested after 72 h for mRNA and protein estimation using RT–PCR (upper) and immunoblot (lower) analysis, respectively. GAPDH and beta‐actin were used as controls for RNA and protein normalization, respectively.
-
B. Luc‐CTL cytotoxicity assay with PBMC‐derived CTLs and bi‐specific Ab as effector population and MCF7 as target cells, which were transfected with individual (s1–s4) or pooled CCR9 siRNA sequences. PD‐L1 and non‐specific control siRNAs were used as positive and negative controls, respectively, for CTL‐mediated cytotoxicity.
-
C, D Cr‐release assay showing % specific lysis of MCF7 cells by survivin‐specific T cells at different ratios upon CCR9 knockdown (C) or overexpression (D). MCF7 cells were transfected with either CCR9 siRNA s1 (∆), pooled siRNA sequences (○), positive control PD‐L1 (□), and non‐specific control siRNA (■) (C) or with pCMV6‐AC‐His control vector (■) and pCMV6‐AC‐His‐CCR9 expression construct (○) (D) 72 h prior to the assay.
-
E. Cr‐release assay showing % specific lysis of MDA‐MB‐231 breast tumor cell line by survivin‐specific T cells at different ratios upon CCR9 knockdown (○) in comparison to the control knockdown (■).
-
F, G Cr‐release assay showing lysis of patient‐derived melanoma cells (M579‐A2) by tumor‐infiltrating lymphocytes (TIL 412) (F) or lysis of PANC‐1 pancreatic adenocarcinoma cells by patient‐derived pancreatic TIL 53 (G) at different E:T ratios upon CCR9 (○) or control (■) knockdown.
Data information: All experiments were performed in triplicates and are representative of at least three independent experiments. Error bars denote ± SEM, and statistical significance was calculated using the unpaired, two‐tailed Student's t‐test.
-
- Figure 5. Tumor‐specific CCR9 impedes Th1‐type immune response
-
A, B ELISpot assay showing IFN‐γ (A) and granzyme B (B) secretion by survivin‐specific T cells, as spot numbers, upon CCR9 knockdown (black bars) in MCF7 cells compared to the control knockdown (white bars). T cells (TC) alone (grey bars) were used as control for background spot numbers.
-
C. Luminex assay showing cytokine levels in the supernatant from the coculture of survivin‐specific TC and either CCR9hi MCF7 (transfected with CCR9‐specific siRNA) or CCR9lo MCF7 (transfected with control siRNA) cells.
-
D. Phospho‐plex analysis showing the phospho‐STAT levels in survivin‐specific TC upon encountering CCR9hi or CCR9lo MCF7 cells. Log2 ratio of mean fluorescent intensity (MFI) of the respective analytes to the unstimulated TC is plotted herein.
-
E. Immunoblot analysis showing the phospho‐STAT1 levels in the CCR9hi‐treated, CCR9lo‐treated or unstimulated TC using the phospho‐specific STAT1 (pTyr701) antibody. Beta‐actin was used as the loading control.
Data information: In all the cases, experiments were performed in triplicate with at least two independent repeats. Mean ± SEM are shown herein, unless stated otherwise, with statistical significance assessed using unpaired, two‐tailed Student's t‐test.Source data are available online for this figure.
Source Data for Figure 5E [emmm201404414-sup-0002-source_data_Fig5E.pdf]
-
- Figure 6. Tumor‐specific CCR9 interacts directly with T cells inducing prominent changes in the gene expression signature
-
A. ELISA showing CCL25 levels in cell lysates from indicated tumor cell lines. CCR9 knockdown (k.d.) in MCF7 cells was achieved using specific shRNA (see Materials and Methods).
-
B. Cr‐release assay showing % specific lysis of MCF7 cells by survivin TC upon CCL25 (□) or CCR9 (○) inhibition using specific siRNAs in comparison to the control siRNA (■). Mean ± SEM are depicted herein.
-
C. MCF7 cells were transfected with control or CCR9‐specific siRNAs, and 48 h later, the supernatants (CCR9lo or CCR9hi SSN, respectively) were used to culture survivin TCs overnight. Supernatant‐treated TCs were then used as effector cells against CCR9lo or CCR9hi MCF7 tumor cells in the Cr‐release assay along with wild‐type MCF7 cells. Mean ± SEM are depicted herein.
-
D. Cr‐release assay showing % specific lysis of MCF7 cells that were pre‐treated with or without pertussis toxin (PTX), or knocked down for CCR9 using specific siRNA. Mean ± SEM are depicted herein.
-
E, F MCF7 cells transfected with control siRNA (CCR9hi) or CCR9 siRNA (CCR9lo) were cocultured with survivin TCs for 12 h. Gene microarray was performed with the total RNA extracted from purified T cells after the coculture. Volcano plot (E) illustrating fold change (FC; log2) in gene expression intensities compared with P‐value (−log2) between CCR9hi‐ and CCR9lo‐treated TCs. Horizontal bar at y = 4.32 represents a statistical significance of P = 0.05 (genes in gray below this line did not reach significance). LogFC cutoff at ± 0.5 is represented by the vertical lines. Heatmap representation of the top upregulated (LogFC > 0.5) and downregulated (LogFC < −0.85) genes (F) with P ≤ 0.05. Individual replicates per sample group are shown herein.
-
- Figure 7. In vivo inhibition of CCR9 significantly reduces tumor outgrowth in response to adoptive TIL therapy
-
A. Cr‐release assay showing TIL 209‐mediated lysis of CCR9+ M579‐A2 (transduced with control shRNA) or CCR9− M579‐A2 cells (transduced with CCR9‐specific shRNA). Curves represent mean ± SEM.
-
B. Scheme for the in vivo mouse experiment involving the s.c. injection of CCR9+ (shControl) or CCR9− (shCCR9) M579‐A2 tumor cells in the left and right flank, respectively, of the NSG mice. Following this, at d2 and d9, mice received i.v. injection of TIL 209 in PBS (n = 7) or PBS alone (control group for tumor growth; n = 3) and measured for tumor growth.
-
C, D Tumor growth curves showing mean ± SEM tumor volume of CCR9+ or CCR9− M579‐A2 tumors in TIL‐treated mice (C) or the PBS alone group (D). Statistical difference was calculated using the unpaired one‐sided Mann–Whitney U‐test.
-
Supplementary Materials
Supplementary Information [emmm201404414-sup-0001-supp_info.pdf]
