In silico and pharmacological evaluation of GPR65 as a cancer immunotherapy target regulating T-cell functions

Abstract

The success of cancer immunotherapies such as immune checkpoint inhibitors, CAR T-cells and immune cell engagers have provided clinicians with tools to bypass some of the limitations of cancer immunity. However, numerous tumour factors curtail the immune response against cancer and limit the efficiency of immuno-oncology (IO) therapies. Acidification of the extra-cellular tumour environment consecutive to aberrant cancer cell metabolism is a well-known promoter of oncogenic processes that also acts as an immune regulator. Yet, the suppressive mechanisms of low extra-cellular pH on anti-cancer immunity remain poorly understood. Recent reports have suggested that GPR65, a Gαs-coupled proton-sensing GPCR broadly expressed in the immune system, may act as an immune suppressant detrimental to anti-tumour immunity. So far, the immuno-regulatory properties of GPR65 in acidic milieux have mostly been documented in macrophages and myeloid cells. Our computational evaluation of GPR65's transcriptomic expression profile and potential as an IO target using public datasets prompted us to further investigate its functions in human T-cells. To this end, we identified and validated GPR65 small molecule inhibitors active in in vitro cellular assays and we showed that GPR65 inhibition promoted the killing capacity of antigen-specific human T-cells. Our results broaden the scope of GPR65 as an IO target by suggesting that its inhibition may enhance T-cell anti-tumour activity and provide useful pharmacological tools to further investigate the therapeutic potential of GPR65 inhibition.

Keywords: GPCRs; Gpr65; T-cells; TDAG8; acidosis; immunosuppression; immunotherapy; solid tumours.

Figure 1

TCGA pan-cancer survival analysis regarding GPR65 genotype. (A) Kaplan-Meier curves and logrank test between rs3742704 homozygous and non-homozygous patients in the entire TCGA database. (B) Kaplan-Meier curves and logrank test between rs3742704 homozygous and non-homozygous patients in early or late stages. (C) Cox model analysis​ of impact of being homozygous for GPR65 rs3742704 polymorphism on overall survival in various patients’ groups (HR, Hazard Ratio), focused on grouped indications and individual indications with at least 2 samples for each genotype.

Figure 2

GPR65 mRNA expression profiling in cells of the tumour environment from the single cell RNAseq public resource TISCH2.

Figure 3

Suppression of human transgenic TCR T-cells effector functions by GPR65 agonists and cAMP elevating agents in response to antigen-specific activation by tumour cells. Representative examples of real-time fluorescent A375 cell growth inhibition induced by TCR T-cell cytotoxic activity in the HLA-A2/NY-ESO1 (A) and HLA-A2/MAGE-A4 (B) antigen systems. TCR T-cells and tumour cells were co-cultured at E:T = 5:1. The graph y-values represent A375 growth through the average total red nuclei surface area per well expressed in µm² from triplicates. The co-cultures were treated with DMSO vehicle control (TCR), 5µM ZINC13684400, 5µM BTB09089, 3 µM PGE2, 50 µM forskolin (FSK) or 33 µM NECA, as indicated in the figure key. NT = non-transduced T-cells control, TCR = T-cells transduced with the NY-ESO1 or MAGE-A4 specific TCRs. (C) Compilation of the T-cell killing activity for both TCR systems in each of the above conditions with T-cells from a total of five donors (two for the NY-ESO1 system and three for MAGE-A4) at E:T = 5:1. The killing index was calculated using area under the curve (AUC) values as follows: ((AUC of untreated A375-AUC of treated A375)/AUC of untreated A375) * 100. Inhibition of IFN-γ secretion in response to antigen by TCR T-cells expressing the NY-ESO1 (D) or MAGE-A4 TCRs (E) co-cultured with A375 cells at E:T ratio = 5:1; each assay condition was performed in duplicate. Unpaired t tests were used to assess the statistical significance of the differences between the groups compared as indicated (* p value = 0.0205; ** p value = 0.0098).

Figure 4

Identification and characterization of GPR65 small molecule inhibitory compounds. (A–D) Dose-response curves showing intra-cellular cAMP levels as a function of GPR65 antagonists, concentrations measured at 3 different extra-cellular pH or with 1µM of the GPR65 agonist BTB09089 at pH 7.5. Potencies, expressed as pIC₅₀ values (in bold), and amplitudes, expressed as maximal difference in concentration (nM) of cAMP (in italic) of several antagonists were evaluated to assess sensitivity of the screening assay at different pH values and in the presence of the agonist BTB09089 (BTB) (E).

Figure 5

GPR65 inhibitors increase T-cell antigen-specific tumour cell killing. (A) NY-ESO1 TCR T cells from 2 healthy donors (generated as described in Material and Methods, DS931 and DS932) were co-cultured with red A375 cells and treated or not with selected GPR65 antagonists at 30 µM and at a 5:1 E:T cell ratio. (B) NY-ESO1 TCR T cells from DS932 were thawed and co-cultured with red A375 cells and treated or not with selected GPR65 antagonists at 6.25, 12.5 and 25 µM and at a 10:1 E:T cell ratio. (C) NY-ESO1 TCR T cells from a healthy donor (DS673) were co-cultured with red A375 cells and treated or not with selected GPR65 antagonists at different concentrations, as indicated, and at a 10:1 E:T cell ratio. Growth of tumour cells was followed over time by Incucyte through red fluorescence and depicted as area of live cells. N.B. that growth of tumour cells only and tumour cells + non transduced T cells is maintained at the same value from 48h due to biological variabilities. Raw data can be found in Supplementary Figure 6C . In total TCR-T cells from 8 donors were tested for this assay. DS931 and DS932 were used in three independent experiments and 6 additional donors were used in 2 independent experiments. NT, non transduced.

Figure 6

GPR65 inhibition is associated with a higher production of pro-inflammatory cytokines. (A) NY-ESO1 TCR T cells from 6 healthy donors (n=2 independent experiments) were co-cultured with wild type A375 cells and treated or not with selected GPR65 antagonists at different concentrations for 3 days. Frequencies of live cells among total CD45- cells were analyzed by FACS and fold change was calculated for each condition compared to the non-treated one. One-way ANOVA test was performed on frequencies of CD45-live cells (left panel). Histograms of the L/D staining in one representative donor (right panel). (B) FACS analysis was performed as described in (A) and frequencies or MFI expression of T-cell activation markers were assessed. Fold change was calculated for each treatment condition compared to the non-treated one. (C) In the second assay assessing T-cell activation, an intranuclear staining was performed following surface staining to analyze in addition IFN-γ and Ki-67 expression. (D) Supernatants from the same co-cultures were harvested at day 3 and analyzed for cytokine release using Luminex (n=6, 2 independent experiments, technical replicates=3). One-way ANOVA test was performed on cytokine concentrations.