Assessment of new HDAC inhibitors for immunotherapy of malignant pleural mesothelioma
Malignant pleural mesothelioma (MPM) is a very rare and highly aggressive cancer of the pleura associated in most cases with asbestos exposure. To date, no really efficient treatments are available for this pathology. Recently, it has been shown that epigenetic drugs, particularly DNA methylation or histone acetylation modulating agents, could be very efficient in terms of cytotoxicity for several types of cancer cells. We previously showed that a hypomethylating agent (decitabine) and a histone deacetylase inhibitor (HDACi) (valproic acid (VPA)) combination was immunogenic and led to the induction of an anti-tumor immune response in a mice model of mesothelioma. However, VPA is not very specific, is active at millimolar concentrations and is responsible for side effects in clinic. To improve this approach, we studied four newly synthetized HDACi, two hydroxamates (ODH and NODH) and two benzamides (ODB and NODB), in comparison with VPA and SAHA. We evaluated their toxicity on immune cells and their immunogenicity on MPM cells in combination with decitabine.
All the tested HDACi were toxic for immune cells at high concentrations. Combination with decitabine increased toxicity of HDACi only towards T-cell clone. A decrease in the proportion of regulatory T cells and natural killer cells was observed in particular with VPA and ODH. In MPM cells, all HDACi combinations induced NY-ESO-1 cancer testis antigen (CTA) expression and the recognition of the treated cells by a NY-ESO-1 specific T-CD8 clone. However, for MAGE-A1, MAGE-A3 and XAGE-1b mRNA expression, the results obtained depended on the HDACi used and on the CTA studied. Depending on the MPM cell line studied, molecules alone increased moderately PD-L1 expression. When combined, a higher stimulation of this immune check point inhibitor expression was observed. Decitabine-induced anti-viral response seemed to be inhibited in the presence of HDACi.
This work shows that the combination of decitabine and HDACi could be of interest for MPM immunotherapy. However, this combination induced PD-L1 expression which suggests that an association with anti-PD-L1 therapy should be performed to induce an efficient anti-tumor immune response.
KeywordsMalignant pleural mesothelioma HDAC inhibitors Decitabine (5-aza-2′-deoxycytidine) Epigenetic drugs Immunotherapy Tumor antigen CD8+ T-cell clone PD-L1
Area under the curve
Bovine serum albumin
- CD4, CD8...
Cluster of differentiation 4, 8...
Cancer testis antigen
- DTC Platform
‘Développement et Transfert Clinique’ platform
Fluorescence Activated Cell Sorting
Fluorescence Iso Thio Cyanate
Forkhead box P3
Histone deacetylase inhibitor
Human Leukocyte Antigens
The half maximal inhibitory concentration
Immunoglobulin (antibody) G1
- MAGE-A1, MAGE-A3
Melanoma antigen-encoding genes family A
Moloney Murine Leukemia Virus
Malignant pleural mesothelioma
Messenger Ribonucleic Acid
Natural killer cells
5-(6-Dimethylamino-2-methyl-3-oxo-2,3-dihydrobenzofuran-2-yl)-4methyl-penta-2,4-dienoic acid benzamide
5-(6-Dimethylamino-2-methyl-3-oxo-2,3-dihydro benzofuran-2-yl)-4methyl-penta-2,4-dienoic acid hydroxamide
New York Oesophageal Squamous Cell Carcinoma 1
4-Methyl-5-(2-methyl-3-oxo-2,3-dihydro-benzofuran-2-yl)-penta-2,4-dienoic acid benzamide
4-Methyl-5-(2-methyl-3-oxo-2,3-dihydro-benzofuran-2-yl)-penta-2,4-dienoic acid hydroxamide
Programmed cell death 1
Programmed death-ligand 1
Peridinin-chlorophyll-protein Complex Conjugate
Retinoic acid-inducible gene I
relative median fluorescent intensity
Ribosomal Protein Lateral Stalk Subunit P0
- RPMI 1640 medium
Roswell Park Memorial Institute 1640 medium
Real Time-Polymerase Chain Reaction
Standard Error of the Mean
Suberoylanilide hydroxamic acid
Regulatory T cells
Tumor suppressor genes
X antigen family member 1b
Malignant pleural mesothelioma (MPM) is a deadly disease that develops in the pleura. It was confirmed by Wagner’s studies that the vast majority of mesothelioma cases are actually attributed to asbestos exposure . MPM was a very rare tumor before the industrial era; however, its incidence is increasing worldwide and is expected to peak in the year 2020. This cancer has an unusual molecular pathology with the loss of tumor suppressor genes being the predominant pattern of lesions . Due to its resistance to all conventional therapies, the need for developing novel therapeutic regimens is urgent in order to cure this rare and treatment-resistant cancer.
In many cancers including mesothelioma, a hypermethylation of some gene promoters, including tumor suppressor genes (TSG) promoters, and overall DNA hypomethylation have been observed [3, 4]. Likewise, a decrease of acetylation of histones H3 and H4 was described [5, 6]. Therefore, epigenetics based therapies could be an interesting opportunity to improve MPM treatment. It was shown in different cancers that the use of hypomethylating agents in combination or not with histone deacetylase inhibitors (HDACi) could be immunogenic [7, 8, 9, 10, 11]. It has been shown, in the laboratory and in another team, that MPM cell death induced by the combination of a hypomethylating agent/HDAC inhibitor could be immunogenic through the induction of tumor antigen expression especially cancer testis antigens (CTA) [12, 13]. Even if epigenetic drugs are efficient especially as a treatment of hematopoietic cancers such as leukemia, they display poor clinical benefits on solid tumors. Moreover, HDACi are unspecific and toxic to healthy cells. In addition, they have side effects such as hematologic toxicity, decreasing the number of platelets and leukocytes . It is, therefore, necessary to develop new epigenetic drugs that are more specific, less toxic to healthy cells and specifically for immune cells, and that can act in low doses.
In order to identify more potent and more convenient HDACi for clinic, we tested four new HDACi we recently synthetized and characterized [15, 16]. First of all, we evaluated the effect of these HDACi in comparison with valproic acid (VPA) and suberoylanilide hydroxamic acid (SAHA), two well-known HDACi, on total lymphocytes and on CD8+ T cell clones viability, and on some important populations of immune cells (natural killer (NK) and regulatory T (Treg) cells). Secondly, we tested the effect of the combination decitabine/new HDACi, in comparison with decitabine/VPA or SAHA combinations, on cancer testis antigens (CTA) expression, NY-ESO-1, MAGE-A1, MAGE-A3 and XAGE-1b, at the mRNA level in a malignant pleural mesothelioma cell lines. Indeed, CTA expression is associated with spontaneous antitumor immune responses and their discovery has led to the development of immunotherapy strategies and to antigen-specific cancer vaccines. Following treatment with decitabine/HDACi combinations, recognition of MPM cells by a NYESO-1-specific CD8+ T-cell clone was assessed. PD-L1 mRNA expression, a molecule that is implicated in the major mechanism of immuno-suppression within the tumor microenvironment , was also measured. Recently, it was demonstrated that hypomethylating agent can induce interferon pathway through activation of endogenous retroviral elements in colorectal, breast, and ovarian cancer [10, 11]. Therefore, the effect of the combinations was studied also on the mRNA expression of MDA5 (melanoma-differentiation-associated gene 5) and RIG-1 (retinoic-acid-inducible protein I), two essential immunoreceptors implicated in interferon signaling pathway and in RNA-sensing pathway.
The tumor cell lines used in this work were Meso96, Meso34, and Meso45, three human pleural mesothelioma cell lines, established from MPM patients’ effusions obtained by thoracocentesis (the patient was informed and gave signed consent). The cell lines were characterized and cultured in our laboratory as described . Cells were cultured in Roswell Park Memorial Institute (RPMI) 1640 medium supplemented with 10% heat-inactivated fetal calf serum, 2 mM L-glutamine, 100 IU/ml penicillin, and 200 μg/ml streptomycin.
Lymphocytes were obtained from the clinical transfer platform (DTC, CIC-biothérapies Nantes) of Nantes Hospital (Nantes, France). CD8 T-cell clones used for this study, HLA-A*0201/NY-ESO-1(157–165)-specific CD8 T-cell clone was described previously .
Decitabine (5-azaCdR) and VPA were purchased from Sigma-Aldrich (Saint-Quentin Fallavier, France). SAHA was from Interchim (Montluçon, France). HDACi ODB, NODB, ODH, and NODH were synthesized by Dr. Philippe Bertrand as described previously . Structure of the HDACi is provided as Additional file 1: Table S1.
For dose-response experiments, lymphocytes were seeded at 4 × 105 cells/well in RPMI 1640 8% HS in flat-bottomed 96-well plates, and CD8 T-cell clones were seeded at 50 × 103 cells/well in 180 μl RPMI 1640 8% HS containing 150 IU IL-2. The treatments were carried out 24 h after seeding by adding decitabine at 500 nM for 72 h; then, HDACi treatments were performed for additional 48 h.
For flow cytometry experiments, lymphocytes were seeded at 4 × 105cells/well in a flat-bottomed 96-well plate in 180 μl RPMI 1640 8% HS. Cells were treated with different HDACi for 48 h.
For tumor cell culture, 105 cells were seeded per well of 6-well plate in 2 ml of culture medium and then incubated at 37 °C for 24 h. Cells were then treated with a combination of 0.5 μM decitabine for 72 h followed by HDACi, VPA 5 mM, SAHA 2.5 μM, ODB 7.5 μM, NODB 2.5 μM, ODH 2.5 μM, and NODH 25 nM for 48 h.
Cell viability test
After treatments, cell viability was measured using CellTiter-Glo kit (Promega) according to the manufacturer’s protocol.
Flow cytometry analysis
The day of analysis, cells were stained in a conical-bottomed 96-well plate at 105 cells/well. The plate was centrifuged at 800×g for 1 min, and the pellets were washed using 200 μl/well of PBS containing BSA 0.1% (wash buffer). Antibodies (see Additional file 2: Table S2) were used at a dilution of 1/30 in the wash buffer, and then, the plate was incubated at 4 °C for 30 min. Later, two washings were performed using the wash buffer prior to flow cytometry analysis. All flow cytometry data were acquired with FACScalibur (BD biosciences) using the CellQuest software (BD Biosciences) and analyzed by FlowJo software.
Expression levels of the gene of interest were analyzed using real-time PCR. Reverse transcription was performed with the M-MLV Reverse Transcriptase (Invitrogen) using aliquots of total RNA extracted from MPM cells NucleoSpin® RNA kits. All real-time PCR reactions were performed using the Mx3005P QPCR Systems (Stratagen Products, Agilent Technologies), and the amplifications were done using the SYBR Green PCR Master Mix SAB bioscience (Qiagen) mixed with Oligonucleotides QuantiTect Primer (Qiagen). The thermal cycling conditions were composed of 1 cycle at 95 °C for 10 min, 40 cycles at 95 °C for 30 s and 60 °C for 1 min, and 1 cycle at 95 °C for 1 min, 60 °C for 30 s, and 95 °C for 30 s. The experiments were carried out in duplicate for each data point. All the qPCR data were analyzed by MxPro software.
Measurement of NY-ESO-1-specific CD8+ T-cells activation
MPM cells were treated or not with decitabine 72 h/HDACi 48 h prior to be seeded at 105 cells/well and then co-cultured with NY-ESO-1-specific CD8+ T-cells  at 5 × 104 cells/well in complete RPMI 1640 medium containing 10 mg mL−1 of brefeldin A (Sigma-Aldrich) for 6 h at 37 °C, then washed. Cells were stained with APC-conjugated mouse anti-human CD8 at 1/30 for 30 min at 4 °C in wash buffer, and PE-conjugated mouse anti-human IFN-γ monoclonal antibodies at 1/50 for 30 min at room temperature in permeabilization buffer. CD8 and IFN-γ expression were analyzed using flow cytometry.
Data presented are means ± S.E.M. The unpaired t test and one-way ANOVA test followed by Holm-Sidak’s multiple comparisons test were used to measure the statistical differences. Statistical analyses were performed using GraphPad Prism 6 (GraphPad Software Inc., San Diego, CA, USA). A P value of 0.05 or less was considered as significant.
Effect of histone deacetylase inhibitors (HDACi) on lymphocyte viability
HDACi IC50 on immune cells
+ Decitabine 500 nM
+ Decitabine 500 nM
6.75 ± 0.06 mM
3.05 ± 0.11 mM
3.80 ± 0.06 mM
0.02 ± 0.10 mM
13.66 ± 0.05 μM
9.9 ± 0.08 μM
1.29 ± 0.05 μM
0.08 ± 0.06 μM
28.81 ± 0.05 μM
0.17 ± 0.12 μM
46.01 ± 0.06 μM
26.22 ± 0.05 μM
17.99 ± 0.05 μM
0.37 ± 0.12 μM
9.62 ± 0.05 μM
5.28 ± 0.09 μM
1.86 ± 0.05 μM
0.07 ± 0.07 μM
0.43 ± 0.06 μM
0.19 ± 0.11 μM
0.03 ± 0.05 μM
0.002 ± 0.06 μM
Histone deacetylase inhibitors effect on immune cells subpopulations
The aim of our strategy is to induce an anti-tumor immune response to cure malignant pleural mesothelioma by using combinations of epigenetic drugs including HDACi. Therefore, it was necessary to study the effect of these HDACi on sub-populations of lymphocytes implicated in this immune response. Doses of compounds were chosen according to blood concentrations of VPA and SAHA measured during clinical trials [20, 21, 22, 23] and to induce approximately 50% of decrease of cell viability for ODB, NODB, ODH, and NODH on total lymphocytes (Fig. 1). Toxicity of the compounds was tested on natural killer cells (NK) (CD56+, CD16+), on regulatory T-cells (Treg) (CD4+, CD25+, CD127 low and Foxp3+), on CD4 and CD8 naïve T-cells (CD4+, CCR7+ and CD45RA+, and CD8+, CCR7+ and CD45RA+, respectively), and on CD4 and CD8 memory T cells (CD4+ and CD45RO+, and CD8+ and CD45RO+, respectively).
Effect of HDACi in combination with decitabine on cancer testis antigens expression in mesothelioma cells
Effect of decitabine alone and/or in combination with HDACi on RIG-1 and MDA5 mRNA expression in mesothelioma cells
Malignant pleural mesothelioma (MPM) is a locally invasive and rapidly fatal malignancy often diagnosed 30–40 years after exposure to asbestos. To date, there is no effective therapeutic strategy against this disease; therefore, the development of new treatment is required. The use of epigenetic drugs represents a promising approach to treat people suffering from MPM. The hypomethylating drug decitabine (5-azaCdR) and HDAC inhibitors have many positive effects either by promoting growth arrest, apoptosis or differentiation of tumor cells [27, 28]. In a previous work on a mice model of mesothelioma, we have shown that the use of these drugs alone did not have a strong anti-tumoral effect, while the use of the combination of decitabine/HDAC inhibitors, VPA or SAHA, was efficient . The use of these FDA-approved drugs had an impact on the expression of CTA in mesothelioma cells and induced an anti-tumor immune response in vivo. However, these compounds have many toxic properties, notably hematologic toxicity, which requires the development of new HDAC inhibitors that are less toxic to healthy cells, act at low doses and preserve immune cells.
In this study, we aimed at characterizing the toxicity of new HDACi, in comparison with VPA and SAHA, on immune cells. Then, we studied their immunogenicity, in combination with decitabine, on MPM cells by measuring the mRNA expression of CTA, PD-L1, MDA5, and RIG-1. Our results showed that toxicity on immune cells and immunogenicity on MPM cells of the tested molecules are different. This study demonstrates that a careful evaluation of HDACi is required to define the best combination strategy for immunotherapy application.
All the tested molecules were toxic for immune cells. However, whereas high doses of HDACi led to the death of 100% of CD8 T-cells clones on total lymphocytes, maximal toxicity depended on the molecule used. When compared to blood concentrations reach in clinic for SAHA, approximately 1 μM [22, 23], these results are coherent with the hematological toxicity observed in patients. Concerning VPA, blood concentrations observed during treatments are around 1 mM [20, 21]. At this concentration, VPA presents weak hematological toxicity according to our observation. Extrapolated to our new compounds, NODH seems to be the more promising. Indeed, this HDACi is 100 times more active to induce histone H3 acetylation in cells than SAHA . Whereas SAHA is toxic for both lymphocytes and CD8 T-cells at clinical doses measured in blood from patients (approximately 1 to 4 μM) [30, 31], NODH at 10 to 40 nM, 100 times lower concentrations than SAHA, was poorly toxic.
Additional analyses demonstrated that Treg and NK cell populations were particularly affected by HDACi. All the tested compounds reduced significantly the proportion of Treg. However, VPA and ODH compounds were the most active drugs on these cells. Epigenetic mechanisms could be responsible for these effects. There is no evidence of a regulation of Treg and NK cell markers through modulation of histone acetylation. FoxP3, a strong marker of Treg, is well-known to be regulated by methylation , and recent study suggests that CD16 expression could also be regulated by methylation . Decitabine alone reduced moderately Treg and NK proportion (Additional file 11: Figure S8). However, here, it is difficult to determine whether the effect was more a toxicity or an epigenetic regulation of cell markers. A mechanistic study is necessary to identify mechanisms involved.
In clinic, the effect of HDACi on Treg cell population is not clear. Some studies mentioned a decrease of Treg cell population following HDACi treatment [34, 35] whereas others reported an increase of Treg number and function . In vitro and preclinical models suggest that reduction of Treg cells by HDACi depends on the specificity and on the dose of the molecules used . According to the implication of Treg cells in the inhibition of anti-tumor immune response, this effect of HDACi could be beneficial for immunotherapy strategy. NK cells are implicated in innate immunity. HDACi were already described as toxic for these particular cells [38, 39, 40], as observed in our study. This action of HDACi could be deleterious for induction of anti-tumor response in vivo regarding the implication of NK cells in tumor immune-surveillance [41, 42] and metastasis control . However, on the contrary of Treg cells, the impact of HDACi on NK cell proportion is more moderate. Indeed, Treg cells were reduced from 65 to 95% and NK cells were reduced from 40 to 85%.
Several studies have demonstrated the immunogenicity of the combination decitabine/HDACi on cancer cells, notably on mesothelioma cells [12, 13]. In this study, we showed that all HDACi tested increased decitabine-induced CTA expression which was associated with an increase activation of a NY-ESO-1-specific CD8 T cell in vitro. This result was not associated with an increase of HLA molecules at mesothelioma cell surface (Additional file 7: Figure S4) which demonstrated that the recognition by T cell clone was driven by the increase of NY-ESO-1 expression. We previously demonstrated that HDAC1 is in part responsible for NY-ESO-1 repression . Thus, an action of all tested HDACi on decitabine-induced NY-ESO-1 expression was expected according to their inhibitory activity on HDAC1 (Additional file 12: Figure S9). For the other CTA tested, there is no available information on the HDAC implicated in their regulation. Additionally, we observed that decitabine increased RIG-1 and MDA5 expression in Meso34 and Meso45 whereas in Meso96 the modifications observed were negligible. This observation is novel in the field of MPM. RIG-1 and MDA5 are two receptors implicated in interferon signaling pathway and viral infection response. The increase of RIG-1 and MDA5 expression by decitabine was described as the consequence of the induction of endogenous virus replication and then of the activation of the interferon signaling pathway [10, 11]. In Meso34 and Meso45, combination with HDACi reduced induction of RIG-1 and MDA5 expression by decitabine. Anti-replicative effect of HDACi on viruses was already described  which could explain the results obtained. Thus, immunogenicity of decitabine could be reduced. However, induction of endogenous viruses by decitabine could also be responsible for an increase of the mutation rate in MPM cells and then to an improve adaptability to their environment. Regarding our previous results obtained in vivo with the combination decitabine/HDACi , it seems that inhibition of interferon signaling pathway induced by decitabine is not determinant for the induction of an anti-tumor response. Moreover, recent data from our laboratory has demonstrated that approximately 70% of MPM have a non-functional interferon pathway (Achard et al., Oncotarget, 2015).
Recently, immune check point inhibitors were identified as responsible for inhibition of anti-tumor immune response and then appear as promising therapeutic targets . The PD-1/PD-L1 axis is of particular interest regarding the impressive results observed in clinic with anti-PD-1 blocking antibody. Therefore, the impact of treatments on PD-L1 expression in cancer cells needs to be monitored to avoid a fail of the strategy evaluated. Indeed, several studies demonstrated that the expression of PD-L1 can be increased in cancer cells following HDACi or hypomethylating agent treatment . In MPM cells, we observed that PD-L1 was induced by the combination decitabine/HADCi. This observation suggests that epigenetic regulation of CTA expression and PD-L1 could be associated.
In view of our results, it seems that there is no really more appropriate molecule when regarding immunogenicity. Our study suggests that the combination decitabine/HDACi should be associated with anti-PD-L1 strategy given that all combinations induced PD-L1 gene expression. However, NODH could be interesting regarding its pharmacological properties. Indeed, this HDACi is active at nanomolar concentrations and at concentrations which induce gene expression, NODH is moderately toxic for lymphocytes when combined with decitabine (25% of cell death). Unfortunately, toxicity on T-cell clone should be considered in the protocol of treatment to avoid hampering an ongoing anti-tumor immune response.
From our work, it seems that more careful studies are needed on the intrinsic properties of HDACi on both immune and cancer cells. Indeed, data obtained will be useful to define the most appropriate combination of molecules for immunotherapy strategies. Likewise, a better comprehension of molecular players participating in the regulation of genes implicated in immunogenicity is required to design optimized HDACi. To date, anti-PD-1/PD-L1 strategy needs to be considered in combination with decitabine and HDACi for MPM treatment in order to overcome a possible induction of PD-L1 expression.
The authors thank the Cytocell core facility for the flow cytometry experiments and the clinical transfer platform (CIC-biothérapies Nantes) for monocyte purification.
This work was supported by INSERM, CNRS, Région Pays de la Loire, the “Institut de recherche en santé respiratoire des Pays de la Loire”, the ‘Ligue Contre le Cancer (committees of Morbihan, Sarthe, Vendée et Loire-Atlantique), ARSMESO44, Nantes University Hospital, and INSERM (France)/ CNRST (Morocco) agreement 2015/2016.
Availability of data and materials
The datasets used and analyzed in this study are available from the corresponding author on request.
MG, EM, and CB contributed to the conception and design of the study. DB and SD performed treatment of cells, mRNA extractions and analyses; DB and VD conducted immunological experiments; PB synthetized HDACi. DB, PB, and CB wrote the manuscript. All authors read and approved the final manuscript.
Ethics approval and consent to participate
Mesothelioma cell line use in this work belongs to a bio-collection of mesothelioma samples from patients (DC-2011-1399) recognized by the French ministry of research. All recruited patients had received no prior anticancer therapy and gave signed, informed consent.
Consent for publication
The authors declare that they have no competing interests.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
- 3.Pan Y, Liu G, Zhou F, Su B, Li Y. DNA methylation profiles in cancer diagnosis and therapeutics. Clin Exp Med. 2018;18:1–14.Google Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.