Inhibitors of histone deacetylase 1 reverse the immune evasion phenotype to enhance T-cell mediated lysis of prostate and breast carcinoma cells

Abstract

The clinical promise of cancer immunotherapy relies on the premise that the immune system can recognize and eliminate tumor cells identified as non-self. However, tumors can evade host immune surveillance through multiple mechanisms, including epigenetic silencing of genes involved in antigen processing and immune recognition. Hence, there is an unmet clinical need to develop effective therapeutic strategies that can restore tumor immune recognition when combined with immunotherapy, such as immune checkpoint blockade and therapeutic cancer vaccines. We sought to examine the potential of clinically relevant exposure of prostate and breast human carcinoma cells to histone deacetylase (HDAC) inhibitors to reverse tumor immune escape to T-cell mediated lysis. Here we demonstrate that prostate (LNCAP) and breast (MDA-MB-231) carcinoma cells are more sensitive to T-cell mediated lysis in vitro after clinically relevant exposure to epigenetic therapy with either the pan-HDAC inhibitor vorinostat or the class I HDAC inhibitor entinostat. This pattern of immunogenic modulation was observed against a broad range of tumor-associated antigens, such as CEA, MUC1, PSA, and brachyury, and associated with augmented expression of multiple proteins involved in antigen processing and tumor immune recognition. Genetic and pharmacological inhibition studies identified HDAC1 as a key determinant in the reversal of carcinoma immune escape. Further, our findings suggest that the observed reversal of tumor immune evasion is driven by a response to cellular stress through activation of the unfolded protein response. This offers the rationale for combining HDAC inhibitors with immunotherapy, including therapeutic cancer vaccines.

Keywords: antigen-processing machinery; entinostat; histone deacetylases; immunogenic modulation; vorinostat.

Conflict of interest statement

CONFLICTS OF INTEREST

Authors have no other potential conflicts of interest.

Figures

Figure 1. Vorinostat decreases pan-HDAC activity and proliferation of human carcinoma cells in an exposure-dependent manner

Human prostate (LNCaP) and breast (MDA-MB-231) carcinoma cells were exposed to vorinostat (1 μM, grey circles and bars; 3 μM, black circles and bars), or vehicle (DMSO, open squares and bars). A. and B. HDAC activity determined at 96h. Results are presented as mean ± S.E.M. from replicate wells. C. and D. Cell number at the indicated time points. Insets denote viability at 96 h. Results are presented as mean ± S.D. from 6 replicate wells. Asterisks denote statistical significance relative to control cells exposed to vehicle (DMSO, P < 0.001). This experiment was repeated 2-3 times with similar results.

Figure 2. Carcinoma cells exposed to vorinostat are significantly more sensitive to CTL-mediated killing

Human prostate (LNCaP) and breast (MDA-MB-231) carcinoma cells were exposed to vorinostat (3 μM, black bars) or to vehicle (DMSO, open bars) as described in Materials and Methods, prior to being used as targets for antigen-specific CTL lysis using CEA-, brachyury-, MUC1-, or PSA-specific CD8+ T cells as effector cells (E:T = 30:1). To verify that effector T cells were HLA-restricted, CTLs were incubated with HLA-A2 negative AsPC-1 pancreatic carcinoma cells exposed to vehicle (DMSO) or vorinostat. Results are presented as mean ± S.E.M. from 3-6 replicate wells, and are representative of 1-4 independent experiments. Asterisks denote statistical significance relative to controls.

Figure 3. Vorinostat-induced immunogenic modulation of MDA-MB-231 carcinoma cells is mediated by HDAC1

MDA-MB-231 cells were exposed to siRNA control or targeting HDAC1 for 24 h prior to being exposed to vehicle (DMSO) or vorinostat (3 μM) as described in Materials and Methods. A. At the end of treatment, total cell lysates were examined by Western blotting to determine expression of HDAC1. GAPDH was used as internal control for total protein levels. Silencing ratio denotes HDAC1 protein expression relative to GAPDH, further normalized to protein levels after treatment in the presence of silencing RNA control. B. At the end of treatment, MDA-MB-231 cells were used as targets in a CTL-lysis assay where effector brachyury-specific CD8+ T cells were used at an E:T ratio of 30:1. Results are presented as mean ± S.E.M. from 4-6 replicate wells. Asterisks denote statistical significance relative to controls (*P = 0.002). Data is representative of two independent experiments.

Figure 4. HDAC inhibition activates the ER stress responsive element in LNCaP carcinoma cells in a dose-dependent manner

A. Single-cell clones of LNCaP cells stably transduced with an ER stress responsive element driving firefly luciferase expression were exposed to vorinostat or entinostat, at the designated concentrations, or DMSO controls, as described in Materials and Methods. At the end of treatment, firefly and renilla luciferase activities were determined. Data are shown as the ratio of firefly luciferase activity relative to that of control renilla luciferase within each well, further normalized to DMSO control. Results are presented as mean ± S.E.M. from 4-6 replicate wells, and are representative of two independent experiments. B. Parental LNCaP prostate carcinoma cells were exposed to vorinostat (3 μM), entinostat (500 nM) or to vehicle (DMSO) controls as described in Materials and Methods, prior to being used as targets for antigen-specific CTL lysis using PSA-specific CD8+ T cells as effector cells (E:T = 30:1). Results are presented as mean ± S.E.M. from 6 replicate wells. Asterisks denote statistical significance relative to controls (P < 0.05). C. Schematic representation of immunogenic modulation induced by HDAC inhibition in human carcinoma cells.

Figure 5. Vorinostat-induced immunogenic modulation is mediated by the unfolded protein response

MDA-MB-231 cells were exposed to siRNA control or targeting ERN1 or PERK for 24 h prior to being exposed to vehicle (DMSO) or vorinostat (3 μM) as described in Materials and Methods. A. and B. At the end of treatment, total cell lysates were examined by Western blotting to determine expression of ERN1 (A) or PERK (B). GAPDH was used as internal control for total protein levels. Silencing ratio denotes target protein expression relative to GAPDH, further normalized to protein levels after treatment in the presence of silencing RNA control. C. and D. At the end of treatment, MDA-MB-231 cells were used as targets in a CTL lysis assay using CEA-specific CD8+ T cells as effectors (E:T = 30:1). Results are presented as mean ± S.E.M. from 6 replicate wells, and are representative of 2-3 independent experiments. Asterisks denote statistical significance relative to controls (P < 0.0001).