Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2017 Dec 12;8(69):114156-114172.
doi: 10.18632/oncotarget.23169. eCollection 2017 Dec 26.

HDAC inhibition potentiates immunotherapy in triple negative breast cancer

Manuela Terranova-Barberio  1 , Scott Thomas  1 , Niwa Ali  2 , Nela Pawlowska  1 , Jeenah Park  1 , Gregor Krings  3 , Michael D Rosenblum  2 , Alfredo Budillon  4 , Pamela N Munster  1
Affiliations

HDAC inhibition potentiates immunotherapy in triple negative breast cancer

Manuela Terranova-Barberio et al. Oncotarget. .

Abstract

Triple-negative breast cancer (TNBC) represents a more aggressive and difficult subtype of breast cancer where responses to chemotherapy occur, but toxicity is significant and resistance often follows. Immunotherapy has shown promising results in various types of cancer, including breast cancer. Here, we investigated a new combination strategy where histone deacetylase inhibitors (HDACi) are applied with immune checkpoint inhibitors to improve immunotherapy responses in TNBC. Testing different epigenetic modifiers, we focused on the mechanisms underlying HDACi as priming modulators of immunotherapy. Tumor cells were co-cultured with human peripheral blood mononuclear cells (PBMCs) and flow cytometric immunophenotyping was performed to define the role of epigenetic priming in promoting tumor antigen presentation and immune cell activation. We found that HDACi up-regulate PD-L1 mRNA and protein expression in a time-dependent manner in TNBC cells, but not in hormone responsive cells. Focusing on TNBC, HDACi up-regulated PD-L1 and HLA-DR on tumor cells when co-cultured with PBMCs and down-regulated CD4+ Foxp3+ Treg in vitro. HDACi significantly enhanced the in vivo response to PD-1/CTLA-4 blockade in the triple-negative 4T1 breast cancer mouse model, the only currently available experimental system with functional resemblance to human TNBC. This resulted in a significant decrease in tumor growth and increased survival, associated with increased T cell tumor infiltration and a reduction in CD4+ Foxp3+ T cells in the tumor microenvironment. Overall, our results suggest a novel role for HDAC inhibition in combination with immune checkpoint inhibitors and identify a promising therapeutic strategy, supporting its further clinical evaluation for TNBC treatment.

Keywords: HDAC inhibitor; checkpoint inhibitor; epigenetics modulators; immunotherapy; triple negative breast cancer.

PubMed Disclaimer

Conflict of interest statement

CONFLICTS OF INTEREST The authors declare no competing interest.

Figures

Figure 1
Figure 1. HDACi modulate PD-L1 expression in ER- breast cancer cell lines in a dose- and time-dependent manner
(A) PD-L1, acetyl-H3 (AcH3) and acetyl-H4 (AcH4) protein expression was determined by western blot in MDA-MB231, SKBR3, T47D and MCF-7 cells untreated or treated for 24, 48 and 72 hours with increasing doses of vorinostat. (B) PD-L1, AcH3, AcH4 protein expression was evaluated in MDA-MB231, SKBR3, T47D and MCF-7 cells untreated or treated for 48 hours with different HDACi: valproic acid (VPA), panobinostat and entinostat. (C) PD-L1 mRNA expression was evaluated by qReal-Time PCR in MDA-MB231, SKBR3, T47D and MCF-7 cells untreated or treated with various HDACi for 6, 10 and 24 hours. β-actin was used as protein loading control in western blot or housekeeping control gene to normalize qReal-Time PCR reactions.
Figure 2
Figure 2. Vorinostat induces PD-L1 expression in breast cancer cell lines directly at a transcriptional level and this translates to increased PD-L1 expression on the cell surface
(A) MDA-MB231 cells were untreated or treated with vorinostat (1.5μM) and/or actinomycin D (AD 5μg/mL) for 10 hours. The expression levels of PD-L1 and GADD45a mRNA were determined by qReal-Time PCR. (B) MDA-MB231 were untreated or treated with vorinostat (1.5μM) alone or in combination with azacitidine (Aza 2μM) or GSK126 (GSK 300nM). PD-L1 mRNA expression was evaluated by qReal-Time PCR with β-actin used as housekeeping control gene. (C) MDA-MB231 cells untreated or treated with vorinostat (1.5μM) for 48 hours, were fixed, stained for PD-L1 (red) and DAPI for nuclei (blue) and observed by microscope. Representative images show PD-L1+ cells with 20x or 40x magnification. (D) MDA-MB231 untreated or treated with increasing doses of vorinostat for 24 and 48 hours were collected for flow cytometry analysis: cells were stained for PD-L1 before or after fixation/permeabilization steps to distinguish between surface and intracellular markers staining, respectively. Flow cytometric quantification of PD-L1 MFI in live tumor cells is shown (expressed as fold change relative to the control). Statistical comparisons are relative to respective intracellular or surface control. Statistical significance is indicated by p-values as * P ≤ 0.05; ** P ≤ 0.01; *** P ≤ 0.001; **** P ≤ 0.0001 and ns: non significant. Data are presented as the mean ± SD for (A) and (B) and ± SEM for (D).
Figure 3
Figure 3. Vorinostat effect on TNBC cells and PBMCs co-cultured together
(A) MDA-MB231 cells untreated or treated with increasing doses of vorinostat were collected 24 and 48 hours after treatment and stained for flow cytometry analysis to quantify PD-L1 expression (expressed as fold change relative to the control). Statistical comparisons are relative to respective control at 24 or 48 hours. (B) MDA-MB231 cells were cultured alone or in presence of PBMCs obtained from healthy donors. Cells were untreated or treated with vorinostat (1.5μM) for 24, 48 and 72 hours and then collected and stained with a comprehensive multicolor flow cytometry panel. Representative flow cytometric plot for PD-L1 expression in MDA-MB231 with or without vorinostat treatment. (C) Flow cytometric quantification of PD-L1 expression (expressed as fold change relative to the control) in MDA-MB231 cells alone or co-cultured with PBMCs with or without vorinostat treatment. (D) Flow cytometric quantification of HLA-DR expression (expressed as fold change relative to the control) in MDA-MB231 cells treated or untreated with increasing doses of vorinostat for 24 and 48 hours. Statistical comparisons are relative to respective control. (E) Representative flow cytometric plot for HLA-DR expression in MDA-MB231 with or without vorinostat treatment. (F) Flow cytometric quantification of HLA-DR expression (expressed as fold change relative to the control) in MDA-MB231 cells alone or co-cultured with PBMCs with or without vorinostat treatment. (G) Representative flow cytometric plots for Foxp3+ CTLA-4high cells in PBMCs with or without vorinostat treatment. (H) Flow cytometric quantification (expressed as fold change relative to the control) of CD4+ Foxp3+ CTLA-4high T cells in PBMCs from healthy donors alone or in presence of MDA-MB231 after 24, 48 and 72 hours of vorinostat treatment. (I) Flow cytometric quantification of the number of CD4+ Foxp3+ CTLA-4high T cells in PBMCs with or without vorinostat treatment for 24, 48 and 72 hours. (J) Representative flow cytometric plots for CD25/CTLA-4 co-expression in live CD4+, CD127low, Foxp3+ PBMCs with or without vorinostat treatment. (K) Flow cytometric quantification of CD25high CTLA-4high Treg in PBMCs of healthy donors with or without vorinostat treatment for 24 and 48 hours. Plots in (J) and (K) are pre-gated on live CD4+ CD127low Foxp3+ cells. Data are presented as the mean ± SEM. Statistical significance is indicated by p-values as * P ≤ 0.05; ** P ≤ 0.01; *** P ≤ 0.001; **** P ≤ 0.0001.
Figure 4
Figure 4. HDACi effect on tamoxifen-resistant MCF-7 breast cancer cells
(A) MCF-7 TamR cells were untreated or treated with increasing doses of vorinostat; PD-L1 and acetyl-H3 protein expression was evaluated at 24, 48 and 72 hours after treatment by western blot analysis. (B) PD-L1 protein expression was evaluated by western blot in MCF-7 TamR cells exposed to VPA, panobinostat and entinostat for 48 hours. (C) PD-L1 mRNA expression was quantified by qReal-Time PCR in MCF-7 TamR cells after treatment with different HDACi. β-actin was used as protein loading control in western blot and housekeeping control gene to normalize qReal-Time PCR reactions. (D) MCF-7 TamR cells, untreated or treated with increasing doses of vorinostat for 24 and 48 hours, were collected for flow cytometry analysis, as described in the Material and Methods, to distinguish between surface and intracellular PD-L1 expression. Flow cytometric quantification of PD-L1 MFI for live tumor cells (expressed as fold change relative to the control) is shown. Flow cytometric quantification of PD-L1 (E) and HLA-DR (F) expression in MCF-7 TamR cells alone or co-cultured with PBMCs and evaluated at 24, 48 and 72 hours after treatment with vorinostat (1.5μM). (G) Flow cytometric quantification of CD4+ Foxp3+ CTLA-4high T cells in PBMCs of healthy donors alone or in presence of MCF-7 TamR cells after 24, 48 and 72 hours of vorinostat treatment. Statistical significance is indicated by p-values as * P ≤ 0.05; ** P ≤ 0.01; *** P ≤ 0.001; **** P ≤ 0.0001. Data are presented as the mean ± SD for (C) and ± SEM for (D-G).
Figure 5
Figure 5. Anti-tumor activity of vorinostat, anti-PD-1 and anti-CTLA-4 on established mouse breast cancer allografts
(A) 4T1 cells (1 x 106) were s.c. injected into BALB/C mice. When established tumors were palpable, mice were treated with vorinostat (100 mg/kg i.p.), anti-PD-1 (10 mg/Kg i.p.) combined with anti-CTLA-4 (10mg/Kg i.p.) or a combination of the three drugs as described in the Material and Methods. Relative tumor volume curves for 4T1 allograft; measurements are shown as mean ± SEM tumor volume (n = 8). (B) Tumor volume averages from each group at day 0 and day 25 (end of treatment) were compared and presented as percentages of vehicle. (C) Effect of vorinostat and/or anti-PD1 + anti-CTLA-4 treatments on the survival of 4T1 allograft mice. (D) Mouse PD-L1 mRNA was measured in FFPE 4T1 tumor samples by RNAscope assay. Hybridization signals were amplified and visualized with RNAscope 2.0 HD detection kit. Data are presented as the mean ± SEM. (E) Murine PD-L1 RNAscope images were captured under a bright field at 40x magnification. One representative image for each treatment is shown. Positive signals showed as brown punctuate dots were analyzed by scoring with ImageJ software. Statistical significance is indicated by p-values as * P ≤ 0.05; ** P ≤ 0.01; *** P ≤ 0.001; **** P ≤ 0.0001; Ns: non significant.
Figure 6
Figure 6. Effect of vorinostat/immunotherapy treatment on immune cell subset, proliferation and apoptosis in 4T1 allograft tumors
Paraffin-embedded tissues were generated from each tumor for IHC and immunofluorescence analysis. Slices were stained for CD4 (A), Foxp3 (B) and CD8 (C) by IHC or immunofluorescence. Mitotic counts (D) were performed on H&E-stained sections, while apoptosis was measured by TUNEL assay (E). Data are presented as the mean ± SEM. Statistical significance is indicated by p-values as * P ≤ 0.05; ** P ≤ 0.01; *** P ≤ 0.001; **** P ≤ 0.0001; Ns: non significant. Representative image for CD4 (F), Foxp3 (G), CD8 (H), and TUNEL assay (I) are shown. Images were captured with a 40x objective on a light microscope. Scale bars correspond to 20μm. Scale bars correspond to 20μm.
Figure 7
Figure 7. Hypothetical mechanism by which HDACi potentiate checkpoint inhibitors treatment in TNBC
HDACi are responsible for multiple different effects: on one side they induce anti-proliferative and pro-apoptotic effects on tumor, while they induce PD-L1 and HLA-DR expression on tumor cells, making the tumor more susceptible for tumor-antigen recognition. On the other hand, HDACi increase CD4+ and CD8+ T cell tumor infiltration and reduce the frequency of Treg in the tumor microenvironment.
See this image and copyright information in PMC

References

    1. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2017. CA Cancer J Clin. 2017;67:7–30. https://doi.org/10.3322/caac.21387. - DOI - PubMed
    1. Gu G, Dustin D, Fuqua SA. Targeted therapy for breast cancer and molecular mechanisms of resistance to treatment. Curr Opin Pharmacol. 2016;31:97–103. https://doi.org/10.1016/j.coph.2016.11.005. - DOI - PubMed
    1. Gradishar WJ, Anderson BO, Balassanian R, Blair SL, Burstein HJ, Cyr A, Elias AD, Farrar WB, Forero A, Giordano SH, Goetz M, Goldstein LJ, Hudis CA, et al. NCCN guidelines insights breast cancer, version 1.2016. J Natl Compr Canc Netw. 2015;13:1475–85. - PubMed
    1. Marks PA. Histone deacetylase inhibitors: a chemical genetics approach to understanding cellular functions. Biochim Biophys Acta. 2010;1799:717–25. https://doi.org/10.1016/j.bbagrm.2010.05.008. - DOI - PMC - PubMed
    1. Minucci S, Pelicci PG. Histone deacetylase inhibitors and the promise of epigenetic (and more) treatments for cancer. Nat Rev Cancer. 2006;6:38–51. https://doi.org/10.1038/nrc1779. - DOI - PubMed