A Novel HDAC1-Selective Inhibitor Attenuates Autoimmune Arthritis by Inhibiting Inflammatory Cytokine Production

Abstract

Rheumatoid arthritis (RA) is systemic autoimmune arthritis that causes joint inflammation and destruction. Accumulating evidence has shown that inhibitors of class I histone deacetylases (HDACs) (i.e., HDAC1, 2, 3, and 8) are potential therapeutic candidates as targeted synthetic disease-modifying antirheumatic drugs (tsDMARDs). Nevertheless, the inhibition of class I HDACs has severe adverse effects because of their broad spectrum. We evaluated the therapeutic effect of a novel selective HDAC1 inhibitor TTA03-107 for collagen-induced arthritis (CIA) and collagen antibody-induced arthritis (CAIA) models in mice. We also examined the effect of TTA03-107 in bone marrow-derived macrophages (BMDMs) and T helper 17 (Th17) cells in vitro. Here, we delineate that TTA03-107 reduced the severity of autoimmune arthritis without obvious adverse effects in CIA and CAIA models. Moreover, TTA03-107 suppressed the production of inflammatory cytokines, such as interleukin (IL)-1β, tumor necrosis factor (TNF)-α, and IL-17A, in serum and joint tissue. In vitro treatment of BMDMs with TTA03-107 dampened the M1 differentiation and inflammatory cytokine production. TTA03-107 also suppressed the differentiation of Th17 cells. These results demonstrate that TTA03-107 can attenuate the development of arthritis in experimental RA models by inhibiting the differentiation and activation of macrophages and Th17 cells. Therefore, TTA03-107 is a potential tsDMARD candidate.

INTRODUCTION

Rheumatoid arthritis (RA) is a systemic, chronic autoimmune arthritis characterized by synovitis and the progressive destruction of cartilage and bone, leading to irreversible joint destruction. Macrophages and osteoclasts have been demonstrated as major cellular components in the pathogenesis of RA. Indeed, the infiltration of macrophages into synovial tissues is a hallmark of RA in the early phase. Activated macrophages in synovial tissues produce pro-inflammatory chemokines and cytokines, such as interleukin (IL)-1β, IL-6, and tumor necrosis factor (TNF)-α, which recruit other immune cells and promote joint inflammation. In addition, enhanced differentiation and activation of osteoclasts are observed in RA, resulting in the destruction of cartilage and bone.^1,2) Both macrophages and osteoclasts in joint tissue originate from circulating monocytes. Thus, current antirheumatic therapy mainly targets the activation and differentiation of macrophages and osteoclasts to reduce disease severity.

Histone deacetylases (HDACs) play a critical role in the regulation of transcription by controlling reversible histone lysine acylation. On the basis of the similarity of nucleic acid sequence and function, HDACs are classified into four classes, with a total of 18 members in humans and mice. The therapeutic effects of pan-HDAC inhibitors (HDACi) have been examined in rodent models of autoimmune diseases as well as in tissue samples from patients. Their anti-inflammatory effects were reported to be prominent in both cases.^3–5) However, their adverse effects, such as fatigue, anorexia, diarrhea, vomiting, weight loss, and alterations in serum biochemical markers, have greatly limited their clinical use in RA.⁶⁾ Against this background, it has been proposed that HDAC isozyme-selective inhibitors may have fewer adverse effects than pan-HDAC inhibitors for the treatment of chronic disease. The current emerging trend in drug development is to identify HDAC isozyme-selective inhibitors with both immunomodulatory activity and improved safety.

The class I HDAC family consists of HDAC1, HDAC2, HDAC3, and HDAC8, the first two of which are ubiquitously expressed. HDAC1 and HDAC2 form homo- and heterodimers, thereby working together as catalytic components of the co-repressor complex at some gene loci.⁷⁾ Interestingly, recent observations in RA patients have identified elevated transcript levels of HDAC1 in synovial fibroblasts and tissues.^8,9) Similarly, lipopolysaccharide (LPS) stimulation upregulates Hdac2 expression in macrophages; this is essential for the subsequent production of pro-inflammatory cytokines, such as IL-12p70 and TNF-α.¹⁰⁾ Strikingly, T-cell-specific deletion of Hdac1 completely prevents the development of collagen-induced arthritis (CIA), suggesting that HDAC1 activity in T cells plays an indispensable role in the pathogenesis of RA.⁹⁾ Furthermore, the injection of HDAC1-targeting small interfering RNA (siRNA) into CIA mice attenuates arthritis, probably by increasing IL-10 production.⁸⁾ These observations shed light on the high therapeutic potential of chemical inhibitors against class I HDAC isozymes, especially HDAC1 and HDAC2, in RA. Indeed, studies in the past two decades have shown that synthesized class I HDAC inhibitors targeting mostly HDAC1 and HDAC2 have anti-rheumatoid effects in CIA and collagen antibody-induced arthritis (CAIA) mouse models of autoimmune arthritis. The inhibition of HDAC1 and HDAC2 suppresses osteoclast differentiation and activity and the production of inflammatory mediators from monocytes and macrophages.^4,11–13) In addition to the chemical compounds, a commensal bacteria-derived natural class I HDAC inhibitor, butyrate, attenuates autoimmune arthritis in osteoclast-dependent and -independent manners.^14–18) Nonetheless, until recently, no HDAC1- and HDAC2-selective inhibitors have been available.

In this study, we investigated the therapeutic efficacy of a novel HDAC1 isozyme-specific inhibitor, TTA03-107, in the CIA and CAIA models. We found that HDAC1 inhibition attenuated the severity of arthritis in both of these models. HDAC1 inhibition by TTA03-107 decreased IL-17A production in vivo and inhibited RORγt⁺ T helper 17 (Th17) cell differentiation in vitro while barely affecting osteoclast differentiation. TTA03-107 also prevented macrophage differentiation and inflammatory cytokine production. Together, our data suggest that TTA03-107 is a potential candidate for targeted synthetic disease-modifying antirheumatic drugs (tsDMARDs) targeting macrophages.

MATERIALS AND METHODS

Mice

C57BL/6J Jcl mice and DBA/1JJmSlc mice were purchased from CLEA Japan (Tokyo, Japan) and Sankyo Labo Service (Tokyo, Japan), respectively. They were fed CE-2 pellet chow (CLEA Japan). All mice were housed and bred at Keio University Faculty of Pharmacy under protocols approved by Keio University Institutional Animal Care and Use Committee.

HDAC Inhibitors

TTA03-107 and sodium valproate (VPA; FUJIFILM Wako Pure Chemical Corporation, Osaka, Japan) were dissolved in dimethyl sulfoxide (DMSO) as stock solution. TTA03-107 has high kinetic selectivity to HDAC1 over HDAC2 despite their 83% amino acid homology (Supplementary Fig. S1).

CIA Model

Six- to seven-week-old male DBA/1JJmsSlc mice were injected intracutaneously with 50 µL (2 mg/mL) of immunization-grade bovine type II collagen (CII; Chondrex, Woodinville, WA, U.S.A.) emulsified in 50 µL (1 mg/mL) of complete Freund’s adjuvant (Chondrex) at the base of the tail on day 1. On day 21, the mice were injected with bovine CII emulsified in incomplete Freund’s adjuvant (Chondrex) as a booster. On day 21, they were randomly divided into three groups of 12 mice each, namely, control DMSO, VPA, and TTA03-107 groups. TTA03-107 and VPA were suspended in MediDrip Sucralose (Clear H₂O; Westbrook, ME, U.S.A.) at a concentration of 0.03 or 8 mg/mL, respectively. Vehicle DMSO was diluted in MediDrip Sucralose at 0.003%. The reagents were orally administered as drinking water starting on the day of booster immunization. All mice were dissected on day 43.

Purification of Anti-collagen Immunoglobulin G (IgG)

To purify collagen recognizing IgG, Amicon Ultra-15 Centrifugal Filter Units (100 kDa, Millipore, Burlington, MA, U.S.A.) were used for serum collected from CIA model mice, in accordance with the manufacturer’s instructions. Briefly, after pre-rinsing the filter with MilliQ water, serum was added up to a volume of 15 mL and centrifuged at 5000 × g for 30 min. After centrifugation, the concentrated sample was collected and the protein concentration was assayed to adjust the concentration to 50 mg/mL. The samples were stored at −80 °C until later use.

CAIA Model

Six- to seven-week-old male DBA/1JJmsSlc mice were randomly divided into four groups consisting of 10 mice each, namely, control DMSO, VPA (reference group), prednisolone (Tokyo Chemical Industry) (reference group), and TTA03-107 groups, on day 0. To produce collagen antibody-induced arthritis, 5 mg of IgG purified from CIA mice was administered intraperitoneally on day 0, followed by 50 µg of LPS (E. coli O111; Wako, Osaka, Japan) on day 3. TTA03-107, VPA, and prednisolone were suspended in MediDrip Sucralose and orally administered as drinking water at concentrations of 0.03, 8, and 0.015 mg/mL, respectively, starting on day 1. All mice were dissected on day 22.

Clinical Evaluation of Arthritis

Clinical parameters were measured using an articular index score. The articular index scoring was performed using a scale of 0–4 (Supplementary Fig. S2): 0, normal; 1, red and swollen toe joint; 2, swollen toe joint and ankle; 3, swelling of the paw; and 4, swelling of all of the feet, including the ankle joint. The sum of the scores of the four limbs was expressed as the clinical score, with a maximum of 16 points.

Histological Analysis

For histological analysis, paws were skinned and fixed in Mildform 10N (FUJIFILM), decalcified in ethylenediaminetetraacetic acid (EDTA) buffer for 15 d, and then embedded in paraffin. The tissue cross-sections were stained with hematoxylin–eosin (H&E; SKK Organization Science Research Institute, Yokohama, Japan). The histological changes were examined under a microscope by blinded observers. Synovial inflammation, bone erosion, cartilage damage, and leukocyte infiltration were scored using a three-parameter system in which individual scores were summed: joint inflammatory cell infiltration, synovial hyperplasia, and cartilage and bone destruction (0 = normal, 1 = mild, 2 = moderate, 3 = marked, 4 = severe). The histological score was calculated by summing the scores.

Osteoclast Differentiation Assay

Bone marrow (BM) was isolated from mice in accordance with the following protocol. In brief, erythrocyte lysis was performed on BM cells from the femur and tibia from male C57BL/6J Jcl mice (6–8 weeks old; Clea Japan), and the surviving cells were cultured in a differentiation medium. To induce osteoclast differentiation, BM cells were cultured in alpha minimum essential medium (α-MEM; Gibco, New York, NY, U.S.A.) with 10% heat-inactivated fetal bovine serum (FBS) (MP Biomedicals, Irvine, CA, U.S.A.), penicillin/streptomycin mixed solution (P/S, 100×; Nacalai Tesque, Kyoto, Japan), β-mercaptoethanol (1000×, 55 µM; Thermo Fisher Scientific, Waltham, MA, U.S.A.), and hydroxyethyl piperazine ethanesulfonic acid buffer solution (HEPES, 10 mM; Nacalai Tesque). The cells were cultured in complete α-MEM containing 20 ng/mL recombinant mouse macrophage colony-stimulating factor (rmM-CSF; BioLegend, San Diego, CA, U.S.A.) for 3 d. Cells were then seeded into 96-well plates and cultured with complete α-MEM containing 20 ng/mL M-CSF and 50 ng/mL recombinant mouse receptor activator of nuclear factor kappa-Β ligand (RANKL; BioLegend) with or without TTA03-107 at 37 °C in a 5% CO₂ incubator. The culture medium was replaced every 2 d. Cells were fixed and stained for tartrate-resistant acid phosphatase (TRAP) activity using a TRAP staining kit, in accordance with the manufacturer’s protocol (Cosmo Bio, Tokyo, Japan) 6 d later. TRAP-positive multinucleated cells containing three or more nuclei were counted as osteoclasts.

T Cell Differentiation

To establish the effect of TTA03-107 on T cells differentiation, naïve CD4⁺ T cells isolated from normal C57BL/6 Jcl mouse spleen were seeded on a high-binding 96-well plate (Corning, New York, NY, U.S.A.), which had been incubated with 5 µg/mL anti-mouse beta chain of the mouse T-cell receptor antibody (anti-TCRβ monoclonal antibody (mAb), H57-597; BioXCell, Lebanon, NH, U.S.A.) in advance. The cells were cultured in complete RPMI-1640 (Nacalai Tesque) with 10% heat-inactivated FBS, P/S (100×), 2-ME (55 µM), and HEPES (10 mM) for 2 d under helper T cell-polarizing conditions. To induce naïve CD4⁺ T cell differentiation into Th17 cells, the cells were cultured in complete media supplemented with anti-mouse CD28 antibody (anti-mCD28, 2 µg/mL, 37.51; BioXCell), anti-mouse interleukin-2 antibody (anti-mIL-2, 10 µg/mL, JES6-1A12; BioLegend), anti-mouse interleukin-4 (anti-mIL-4, 10 µg/mL, 11B11; BioXCell), anti-mouse interferon-gamma antibody (anti-mIFN-γ, 10 µg/mL, R4-6A2; BioXCell), recombinant mouse interleukin-6 (rmIL-6, 25 ng/mL; BioLegend), recombinant mouse interleukin-1β (rmIL-1β, 10 ng/mL; BioLegend), and recombinant mouse interleukin-23 (rmIL-23, 10 ng/mL; BioLegend). Then, the medium was replaced with a new one and the cells continued to be cultured under these conditions for 3 d with or without different concentrations of TTA03-107 at 37 °C in a 5% CO₂ incubator. To naïve CD4⁺ T cell differentiation induce into Treg and Tfr cells, the cells were cultured in complete media supplemented with anti-mCD28 (2 µg/mL, 37.51; BioXCell), recombinant mouse interleukin-2 (rmIL-2, 20 ng/mL; BioLegend), recombinant human transforming growth factor-β1 (rhTGF-β1, 0.1 ng/mL; BioLegend), anti-mIL-4 (10 µg/mL, 11B11; BioXCell), and anti-mIFN-γ (10 µg/mL, R4-6A2; BioXCell). Then, the medium was replaced with a new one and the cells continued to be cultured under these conditions for 2 d with or without different concentrations of TTA03-107 for Treg cells differentiation. For Treg cells differentiation, cells were washed, and 1/4 volume of each well was resuspended in complete culture media supplemented with rmIL-2 (20 ng/mL; BioLegend), rhTGF-β1 (0.1 ng/mL; BioLegend), anti-mIL-4 (10 µg/mL, 11B11; BioXCell) and anti-mIFN-γ (10 µg/mL, R4-6A2; BioXCell) and cultured for another 3 d with the compound to be tested. For Tfr cells differentiation, cells were washed and 1/4 volume of each well was resuspended in complete culture media supplemented with rmIL-2 (2 ng/mL; BioLegend), rhTGF-β1 (0.1 ng/mL; BioLegend), recombinant mouse interleukin-21 (rmIL-21, 25 ng/mL; BioLegend), rmIL-6 (25 ng/mL; BioLegend), anti-mIL-4 (10 µg/mL, 11B11; BioXCell), anti-mIFN-γ (10 µg/mL, R4-6A2; BioXCell) and anti-mouse inducible T cell costimulatory (mICOS, 5 µg/mL, 7E.17G9; BioXCell) antibodies and cultured for another 3 d with the compound to be tested.

Bone Marrow-Derived Macrophages (BMDMs) Isolation and Culture

To induce BMDMs, BM was isolated as described previously and cultured in a differentiation medium for 3 d, then the medium was changed to a new one for 3 d with or without different concentrations of TTA03-107. The differentiation medium consisted of high-glucose Dulbecco's Modified Eagle Medium (Nacalai Tesque), 10% heat-inactivated FBS, P/S (100×), 2-ME (55 µM), and HEPES (10 mM), and a final concentration of 20 ng/mL rm M-CSF (BioLegend) was also added. BMDMs were polarized to M1 macrophages by 10-h stimulation with 100 ng/mL LPS (E. coli O111125-05181; Wako), followed by 6-h stimulation with 20 ng/mL nigericin (InvivoGen, San Diego, CA, U.S.A.) to activate the NLRP3 inflammasome. After culture under the polarized conditions, cell supernatants and cells were collected for enzyme-linked immunosorbent assay (ELISA) and flow cytometry assays, respectively.

Cytokine Measurements by ELISA

The serum and joint were collected from CIA and CAIA model mice for cytokine measurements. Frozen tissues (whole joints, including synovium, adjacent tissues, and bone) were pulverized with a mortar and pestle filled with liquid nitrogen. Approximately 80 mg of tissue was transferred to a 5 mL tube, resuspended in total 2 mL of RIPA of tissue, and frozen overnight at −80 °C. This was followed by centrifugation at 500 × g for 10 min at 4 °C. The supernatant was transferred to a 1.5 mL tube and collected for cytokine analysis. The levels of IL-1β, TNF-α, and IL-17A were measured using ELISA kits (all from BioLegend) in accordance with the manufacturer’s instructions. Joint cytokine data are expressed as pg cytokine/g tissue (data were multiplied by a dilution factor for conversion from pg/mL to pg/g). For cytokine measurements in the in vitro experiment, the cell supernatant was collected at the end of culture under the M1 polarized conditions. IL-1β, TNF-α, and IL-6 were measured using ELISA kits (all from BioLegend) in accordance with the manufacturer’s instructions.

Flow Cytometry

Th17 cells were collected after being cultured under Th17 polarized conditions. For intracellular cytokine staining, Th17 cells were stimulated with a cell activation cocktail (500×; BioLegend) with final concentrations of phorbol 12-myristate 13-acetate (PMA) of 50 ng/mL and of ionomycin of 1000 ng/mL for 4 h. A protein transport inhibitor cocktail (500×; eBioscience, San Diego, CA, U.S.A.) was added after 1 h. Staining was then performed for cell surface BV605-conjugated anti-CD4 (RM4-5; BD Biosciences), BV786-conjugated anti-CD25 (2A3; BD Biosciences), PCP-Cy5.5-conjugated anti-CD45 (30-F11; BioLegend), and APC-Cy7-conjugated anti-FVS780. After washing, cells were permeabilized for intracellular staining of PE-conjugated anti-Foxp3 (FJK-16s; eBioscience) and PE-CF594-conjugated anti-RORγt (Q21-559; BD Biosciences). Macrophages were collected from the study after being treated with the M1 polarized culture system and were stained for cell surface BUV 395-conjugated anti-CD45 (30-F11; BD Biosciences), BUV 737-conjugated anti-CD11b (M1/70; BD Biosciences), PE-conjugated anti-F4/80 (QA17A29; BioLegend), and APC-Cy7-conjugated anti-Fixable Viability Stain 780 (FVS780; BD Biosciences). After washing, cells were permeabilized for intracellular staining of APC-conjugated anti-Arg1 (A1exF5; Invitrogen, Waltham, MA, U.S.A.) and PCP-Cy5.5-conjugated anti-inducible nitric oxide synthase (iNOS) (CXNFT; Invitrogen). Flow cytometry was performed on a BD FACSCelesta or a BD LSR II and analyzed with FlowJo software (v10.8; BD Biosciences).

Statistical Analysis

Statistical analysis was performed using GraphPad Prism (version 19.0). All data were expressed as mean ± standard error of the mean (S.E.M.). Repeated-measures two-way ANOVA or one-way ANOVA was used for comparison between multiple means. The statistical significance of differences between groups was determined using an unpaired Student’s t-test or Welch’s t-test. The data were considered statistically significant when p < 0.05.

RESULTS

Antiarthritic Efficacy of TTA03-107 in Mouse CIA Model

TTA03-107 is a thermodynamically and kinetically selective HDAC1 inhibitor (HDAC2 K_i/HDAC1 K_i = 5.47; HDAC2 k_-2/HDAC1 k_-2 = 2.55) (Supplementary Fig. S1). To examine the therapeutic effect of TTA03-107 in the RA mouse model, we administered TTA03-107 in drinking water (approximately 5 mg/kg/d) to CIA-induced DBA/1J mice at the booster immunization onward. We also administered a class I HDAC inhibitor, VPA (approximately 1200 mg/kg/d), which was shown to prevent CIA development¹⁷⁾ as a positive reference. The mice administered TTA03-107 had a higher rate of body weight gain than the mice in the control group, while the VPA-treated group displayed lower body weight gain with a sharp reduction at the initial phase (Fig. 1A). Water consumption was also significantly lower in the VPA-treated group than in the other groups (Fig. 1B). We found that TTA03-107 treatment alleviated the severity of arthritis (Figs. 1C, D). In contrast, there was no change in clinical scores between the VPA-treated and control groups (Figs. 1C, D). Upon histological examination, a broken joint structure with a large amount of inflammatory cell infiltration and cartilage destruction was evident in the control group, whereas the treatment with TTA03-107 significantly ameliorated these changes with few signs of inflammation (Fig. 1E). We also evaluated the serum levels of CII-specific IgG, a hallmark of CIA, 3 weeks after the booster immunization. The administration of TTA03-107 reduced CII-specific IgG_2b but not IgG₁ and IgG_2a, nor CII-specific total IgG (Fig. 1F). Because CII-specific IgG₁ and IgG_2a play primary roles in the pathogenesis of CIA, TTA03-107 may attenuate arthritis through a mechanism other than preventing autoantibody production. Therefore, we further explored the effect of TTA03-107 on the production of inflammatory cytokines, TNF-α and IL-17A, which make critical contributions to the pathogenesis of CIA.¹⁹⁾ The serum levels of these inflammatory cytokines were moderately but significantly reduced in the TTA03-107-treated groups (Fig. 1G). These observations suggest that TTA03-107 mitigates the development of arthritis symptoms in CIA mice by suppressing inflammatory cytokine production rather than CII-specific IgG immune responses.

Fig. 1. Antiarthritic Efficacy of TTA03-107 in Mouse CIA Model

Body weight change (A) and drinking volume (B) among the groups (n = 12 mice/group, two-way ANOVA or one-way ANOVA followed by Dunnett’s post hoc test). (C) Arthritis scores and scores of AUC (D) for CIA mice after booster immunization (n = 12, two-way ANOVA or one-way ANOVA followed by Dunnett’s post hoc test). (E) H&E staining graphs and pathology score for histological changes of TTA03-107-treated mouse joints (n = 6, Welch’s t-test or unpaired Student’s t-test). (F) Serum levels of total IgG, anti-CII IgG, anti-CII IgG₁, anti-CII IgG_2a, and anti-CII IgG_2b (n = 12, Welch’s t-test or unpaired Student’s t-test). (G) Serum levels of TNF-α, IL-17A and IL-1β (n = 12, Welch’s t-test or unpaired Student’s t-test). * p < 0.05, ** p < 0.01, *** p < 0.001.

TTA03-107 Exerts an Antiarthritic Effect in the Later Phase

To test whether TTA03-107 is engaged in arthritis at the later phase of pathogenesis after autoantibody generation, we treated CAIA model mice with TTA03-107 via drinking water. We also administered the conventional first-line DMARD prednisolone as a positive reference. Similar to the CIA experiment, the TTA03-107-treated group did not show a change in body weight gain (Fig. 2A). By contrast, treatment with VPA or prednisolone decreased body weight, especially until day 13 after disease onset (Fig. 2A). The VPA treatment group consumed less water at the beginning of the experiment, whereas water consumption was comparable among the other three groups (Fig. 2B, Supplementary Fig. S3). TTA03-107 treatment ameliorated arthritic symptoms such as swelling of the ankle joint and paw, skin congestion, and stiffness (Figs. 2C, D). The therapeutic effect of TTA03-107 was comparable to that of prednisolone. Meanwhile, VPA did not significantly alter the severity of arthritis (Figs. 2C, D). We further evaluated inflammatory cytokines in the serum and joint tissue extract in the CAIA model and observed that TTA03-107 markedly suppressed the levels of TNF-α, IL-1β, and IL-17A in the serum (Fig. 2E). A similar tendency was also observed in the joint tissue extract (Fig. 2F). These findings indicate that TTA03-107 exerts a therapeutic effect in the later phase of arthritis development.

Fig. 2. Antiarthritic Efficacy of TTA03-107 in Mouse CAIA Model

Body weight change (A) and drinking volume (B) among the groups (n = 10 mice/group, two-way ANOVA or one-way ANOVA followed by Dunnett’s post hoc test). (C) The evolution of arthritis scores and scores of AUC (D) of CAIA mice (n = 10, two-way ANOVA or one-way ANOVA followed by Dunnett’s post hoc test). (E, F) The levels of IL-17A, TNF-α, and IL-1β in serum (E) and in knee joint tissues (F) (n = 10, Welch’s t-test or unpaired Student’s t-test). ** p < 0.01, *** p < 0.001.

TTA03-107 Treatment Suppresses Osteoclast Differentiation

TTA03-107 treatment almost completely reduced the damage of cartilage and bone in the CIA model (Fig. 1D), and therefore we hypothesized that it might reduce arthritis severity by affecting osteoclast differentiation. To test this hypothesis, we assessed the osteoclast differentiation of BM cells in the presence or absence of TTA03-107 by TRAP staining. We observed that TTA03-107 moderately suppressed osteoclast formation only at higher concentrations (Figs. 3A, B), suggesting that the direct effect of HDAC1 inhibition on osteoclast differentiation may be minimal and is unlikely to contribute to its antiarthritic activity.

Fig. 3. In Vitro Effects of TTA03-107 on Osteoclast Differentiation

Representative microphotographs (A) and quantification (B) of multinucleated TRAP-positive osteoclasts from in vitro bone marrow cell cultures stimulated with M-CSF and RANKL and treated with different concentrations of TTA03-107 (n = 5, Dunnett's test, compared to the control untreated group). * p < 0.05. All experiments were repeated at least twice.

TTA03-107 Treatment Reduces RORγt⁺ Th17-Cell Differentiation in Vitro

Our previous study showed that commensal bacteria-derived butyrate, as a class I HDAC inhibitor, ameliorates arthritis by inhibiting autoantibody production through enhancing the differentiation of follicular regulatory T (Tfr) cells.¹⁵⁾ However, TTA03-107 showed no effects on the production of CII specific IgG in CIA model except IgG_2b (Fig. 1F). Meanwhile, in vitro experiment demonstrated that TTA03-107 promoted Treg and Tfr cell differentiation only at very high concentrations (Supplementary Fig. S4). These data suggest that TTA03-107 affect arthritis not through Tfr cell induction.

Th17 cells serve as potent osteoclastogenic T cells in the RA model; RANKL derived from Th17 cells supports osteoclastogenesis to enhance bone destruction. Considering that pan-HDAC inhibitors inhibit Th17-cell differentiation and the T-cell-specific deletion of HDAC1 completely suppresses IL-17A production, we assumed that TTA03-107 might prevent the destruction of cartilage and bone in the CIA model by inhibiting Th17-cell differentiation. To test this assumption, we adopted an in vitro Th17 culture and analyzed the expression of RORγt, a characteristic transcription factor of Th17 cells. The frequency of RORγt⁺ Th17 cells and RORγt mean fluorescent intensity were reduced upon TTA03-107 treatment, but there was no significant dose–response relationship (Fig. 4).

Fig. 4. Effects of TTA03-107 on T Cells in Vitro under Th17 Differentiation

Frequency and MFI of RORγt cells within CD45⁺CD4⁺CD25⁻Foxp3⁻ cells (n = 3, Dunnett’s test, compared to the control untreated group). Naïve T cells were isolated from C57BL/6J Jcl mice and cultured under Th17-skewing conditions with or without different concentrations of TTA03-107. ** p < 0.01, *** p < 0.001. All experiments were repeated at least twice.

TTA03-107 Treatment Downregulates Inflammatory Cytokine Production by Macrophages

Class I HDAC inhibitors have been recognized to suppress the expression of pro-inflammatory cytokines in response to LPS stimulation in macrophages. Although the mechanism of HDAC1 involvement remained unclear, it is possible that HDAC1 inhibition also affects macrophage hyperresponsiveness. To test this, we prepared BM-derived macrophages under M1 polarizing conditions in the presence of TTA03-107 before stimulating them with LPS. The levels of IL-1β, TNF-α, and IL-6 were significantly reduced in M1 macrophages treated with TTA03-107 in a dose-dependent manner (Fig. 5A). To examine whether TTA03-107 treatment changes the M1 macrophage phenotype toward the M2 phenotype, we analyzed the expression of an M1 macrophage marker, iNOS, and an M2 macrophage marker, arginase 1 (Arg1), by flow cytometry. We found that TTA03-107 treatment downregulated iNOS expression while upregulating Arg1 expression (Figs. 5B–D, Supplementary Fig. S5). These results suggest that TTA03-107 downregulated the transcriptional status of genes involved in the differentiation and production of inflammatory cytokines of M1 macrophages, probably by promoting the conversion of M1 to M2 phenotype.

Fig. 5. Effects of TTA03-107 on Macrophages in Vitro under M1 Differentiation Culture System

(A) Effect of TTA03-107 on cytokine production under M0 culture conditions (without LPS stimulants) or M1 culture conditions (n = 3, Dunnett’s test, compared to the control untreated group). Frequency and MFI level of iNOS⁺ (B), Arg1⁺ (C) cells within CD45⁺F4/80⁺CD11b⁺ macrophage cell population under M0 or M1 conditions (n = 3, Dunnett’s test, compared to the control untreated group), (D) iNOS⁻Arg1⁺ cells (M2), and iNOS⁺Arg1⁻ cells (M1) within F4/80⁺CD11b⁺ macrophage cell population under M0 or M1 conditions (n = 3, Dunnett’s test, compared to the control untreated group). Macrophages were isolated from C57BL/6J Jcl mice and cultured under induction conditions with or without different concentrations of TTA03-107. * p < 0.05, ** p < 0.01, *** p < 0.001. All experiments were repeated at least twice.

DISCUSSION

Immunosuppressive and immunomodulatory agents for treating RA are generally called disease-modifying antirheumatic drugs (DMARDs). DMARDs are classified as either synthetic DMARDs (sDMARDs) or biologic DMARDs (bDMARDs). The most widely used DMARDs for RA therapy are conventional sDMARDs (csDMARDs), such as methotrexate, which target the entire immune system and slow down disease progression, although they take approximately a month to start working. bDMARDs and targeted sDMARDs (tsDMARDs) tend to work faster than csDMARDs.^20,21) However, bDMARDs are expensive and approximately 40% of RA patients do not respond to therapy using anti-TNF-α antibodies, a representative bDMARD. In addition, currently available tsDMARDs are limited to inhibitors of Janus kinases (JAK), which are associated with concerns about increased susceptibility to viral infection.²²⁾ Moreover, some individuals who do not respond to JAK inhibitor treatment have been reported. There is thus a need for a rational therapeutic approach with novel tsDMARDs that target non-JAK molecules for RA patients, especially bDMARD non-responders.

Herein, we addressed the therapeutic potential of a novel HDAC1-selective inhibitor, TTA03-107, for mouse models of RA. TTA03-107 ameliorated disease severity and the degree of inflammation in the joints of both CIA and CAIA models without affecting their body weight. In comparison, a class I HDAC inhibitor, VPA, reduced the body weight and had little if any effect on the development of arthritis in both models. Given that other class I HDAC inhibitors have a strong DMARD effect, the lower therapeutic effect of VPA in the current study is an anomaly,^11–13) which may be due to the different administration routes used. A previous study demonstrated that the intraperitoneal administration of VPA significantly improved the incidence and disease severity of CIA.¹⁷⁾ Meanwhile, we administered VPA orally, which may be inappropriate with regard to achieving efficacy due to the associated low bioavailability. Moreover, the VPA treatment group consumed less water presumably due to its high concentration. Therefore, the dosage of VPA was calculated based on the reduced volume of water consumption, and thus the reduction of water consumption is unlikely to affect the results.

In line with previous findings from mice with T-cell-specific deletion of HDAC1,⁹⁾ the current study shows that HDAC1 inhibition by TTA03-107 treatment significantly decreased IL-17A levels in both serum and joint tissue in the CIA model. The reduction of IL-17A is probably attributable to T-cell-intrinsic HDAC1 inhibition to arrest Th17 differentiation and also to the downregulation of IL-6 and IL-1β by macrophages in the joint tissues. Indeed, our in vitro BMDM experiment confirmed that TTA03-107 reduces the production of TNF-α, IL-1β, and IL-6. Moreover, TTA03-107 upregulated the M2 macrophage marker Arg1 while downregulating the M1 marker iNOS. These phenomena imply that HDAC1 inhibition is related to the M1 to M2 macrophage transition, although there is a need for further work aimed at developing a deeper understanding of the molecular mechanisms involved.

CII-specific IgG_2b production was slightly abated by TTA03-107 treatment in the CIA model, although the other IgG subclasses remained unchanged. We previously reported that a class I HDAC inhibitor, butyrate, suppresses germinal center responses by inducing follicular regulatory T-cell differentiation in the CIA model.^15,18) Therefore, the abatement of CII-specific IgG_2b by TTA03-007 may also be attributable to the attenuation of germinal center responses to CII. Additionally, class I HDAC inhibition by butyrate also regulates osteoclast differentiation and bone homeostasis by inducing metabolic reprogramming of osteoclasts, reducing joint inflammation and bone destruction in mice in the CIA and CAIA models.^14,16) Meanwhile, TTA03-107 showed only a minor effect on osteoclast differentiation even at a high concentration. Thus, it is plausible that osteoclast regulation by class I HDAC inhibition is mediated by its effect on HDAC2, HDAC3, or HDAC8, but not HDAC1. Taken together, our observations illustrate that TTA03-107 alleviates the development of inflammatory arthritis by inhibiting inflammatory cytokine production from inflammatory macrophages and subsequent Th17 responses in the autoimmune arthritis models.

CONCLUSION

The present study demonstrates that HDAC1 could be a target for future therapeutic approaches in RA. Because few tsDMARDs are currently available, the HDAC1-selective inhibitor TTA03-107 is a potential candidate tsDMARD. Future studies will further dissect the molecular mechanisms involved and investigate the clinical efficacy and safety, which could lead to the development of valuable therapeutic applications.

Acknowledgments

We would like to thank Shunsuke Kimura at Keio University Faculty of Pharmacy for valuable discussions. We would also like to acknowledge the Animal Facility at Keio University Faculty of Pharmacy for help with the breeding and maintenance of our mouse strains, and the Instrument Management Division for help with maintaining instruments.

This work was supported by AMED-Crest (21gm1310009h0002 to KH), the Japan Society for the Promotion of Science (20H00509, and 20H05876 to KH; JP20K07552 and JP17K15734 to DT), Keio University Special Grant-in-Aid for Innovative Collaborative Research Projects (KH), Keio Gijuku Fukuzawa Memorial Fund for the Advancement of Education and Research (DT), the SECOM Science and Technology Foundation (KH), The Uehara Memorial Foundation (KH), and The Asahi Grass Foundation (KH). The funders had no role in study design, data collection, data analysis, interpretation, writing of the report, nor the decision to publish.

Author Contributions

WZ, DT, and KH conceived the project. WZ performed most experiments, data collection, and statistical analyses. WZ, DT and KH wrote the manuscript. KH supervised the study. TT, YI, and TS synthesized and provided compounds.

Conflict of Interest

The authors declare no conflict of interest.

Supplementary Materials

This article contains supplementary materials.

REFERENCES

• 1) Smolen JS, Aletaha D, Barton A, Burmester GR, Emery P, Firestein GS, Kavanaugh A, McInnes IB, Solomon DH, Strand V, Yamamoto K. Rheumatoid arthritis. Nat. Rev. Dis. Primers, 4, 18002 (2018).
• 2) McInnes IB, Schett G. The pathogenesis of rheumatoid arthritis. N. Engl. J. Med., 365, 2205–2219 (2011).
• 3) Nasu Y, Nishida K, Miyazawa S, Komiyama T, Kadota Y, Abe N, Yoshida A, Hirohata S, Ohtsuka A, Ozaki T. Trichostatin A, a histone deacetylase inhibitor, suppresses synovial inflammation and subsequent cartilage destruction in a collagen antibody-induced arthritis mouse model. Osteoarthritis Cartilage, 16, 723–732 (2008).
• 4) Grabiec AM, Korchynskyi O, Tak PP, Reedquist KA. Histone deacetylase inhibitors suppress rheumatoid arthritis fibroblast-like synoviocyte and macrophage IL-6 production by accelerating mRNA decay. Ann. Rheum. Dis., 71, 424–431 (2012).
• 5) Lin HS, Hu CY, Chan HY, Liew YY, Huang HP, Lepescheux L, Bastianelli E, Baron R, Rawadi G, Clément-Lacroix P. Anti-rheumatic activities of histone deacetylase (HDAC) inhibitors in vivo in collagen-induced arthritis in rodents. Br. J. Pharmacol., 150, 862–872 (2007).
• 6) Subramanian S, Bates SE, Wright JJ, Espinoza-Delgado I, Piekarz RL. Clinical toxicities of histone deacetylase inhibitors. Pharmaceuticals, 3, 2751–2767 (2010).
• 7) Hassig CA, Tong JK, Fleischer TC, Owa T, Grable PG, Ayer DE, Schreiber SL. A role for histone deacetylase activity in HDAC1-mediated transcriptional repression. Proc. Natl. Acad. Sci. U.S.A., 95, 3519–3524 (1998).
• 8) Hawtree S, Muthana M, Wilkinson JM, Akil M, Wilson AG. Histone deacetylase 1 regulates tissue destruction in rheumatoid arthritis. Hum. Mol. Genet., 24, 5367–5377 (2015).
• 9) Göschl L, Preglej T, Boucheron N, et al. Histone deacetylase 1 (HDAC1): a key player of T cell-mediated arthritis. J. Autoimmun., 108, 102379 (2020).
• 10) Wu C, Li A, Hu J, Kang J. Histone deacetylase 2 is essential for LPS-induced inflammatory responses in macrophages. Immunol. Cell Biol., 97, 72–84 (2019).
• 11) Cantley MD, Fairlie DP, Bartold PM, Marino V, Gupta PK, Haynes DR. Inhibiting histone deacetylase 1 suppresses both inflammation and bone loss in arthritis. Rheumatology, 54, 1713–1723 (2015).
• 12) Nishida K, Komiyama T, Miyazawa SI, Shen ZN, Furumatsu T, Doi H, Yoshida A, Yamana J, Yamamura M, Ninomiya Y, Inoue H, Asahara H. Histone deacetylase inhibitor suppression of autoantibody-mediated arthritis in mice via regulation of p16INK4a and p21 WAF1/Cip1 expression. Arthritis Rheum., 50, 3365–3376 (2004).
• 13) Wu C, Li A, Leng Y, Li Y, Kang J. Histone deacetylase inhibition by sodium valproate regulates polarization of macrophage subsets. DNA Cell Biol., 31, 592–599 (2012).
• 14) Kim DS, Kwon JE, Lee SH, Kim EK, Ryu JG, Jung KA, Choi JW, Park MJ, Moon YM, Park SH, Cho ML, Kwok SK. Attenuation of rheumatoid inflammation by sodium butyrate through reciprocal targeting of HDAC2 in osteoclasts and HDAC8 in T cells. Front. Immunol., 9, 1525 (2018).
• 15) Takahashi D, Hoshina N, Kabumoto Y, et al. Microbiota-derived butyrate limits the autoimmune response by promoting the differentiation of follicular regulatory T cells. EBioMedicine, 58, 102913 (2020).
• 16) Lucas S, Omata Y, Hofmann J, Böttcher M, Iljazovic A, Sarter K, Albrecht O, Schulz O, Krishnacoumar B, Krönke G, Herrmann M, Mougiakakos D, Strowig T, Schett G, Zaiss MM. Short-chain fatty acids regulate systemic bone mass and protect from pathological bone loss. Nat. Commun., 9, 55 (2018).
• 17) Saouaf SJ, Li B, Zhang G, Shen Y, Furuuchi N, Hancock WW, Greene MI. Deacetylase inhibition increases regulatory T cell function and decreases incidence and severity of collagen-induced arthritis. Exp. Mol. Pathol., 87, 99–104 (2009).
• 18) He J, Chu Y, Li J, et al. Intestinal butyrate-metabolizing species contribute to autoantibody production and bone erosion in rheumatoid arthritis. Sci. Adv., 8, eabm1511 (2022).
• 19) Komatsu N, Okamoto K, Sawa S, Nakashima T, Oh-Hora M, Kodama T, Tanaka S, Bluestone JA, Takayanagi H. Pathogenic conversion of Foxp3⁺ T cells into TH17 cells in autoimmune arthritis. Nat. Med., 20, 62–68 (2014).
• 20) Holdsworth EA, Donaghy B, Fox KM, Desai P, Collier DH, Furst DE. Biologic and targeted synthetic DMARD utilization in the United States: Adelphi Real World Disease Specific Programme for rheumatoid arthritis. Rheumatol. Ther., 8, 1637–1649 (2021).
• 21) Huang J, Fu X, Chen X, Li Z, Huang Y, Liang C. Promising therapeutic targets for treatment of rheumatoid arthritis. Front. Immunol., 12, 686155 (2021).
• 22) Harrington R, al Nokhatha SA, Conway R. Jak inhibitors in rheumatoid arthritis: an evidence-based review on the emerging clinical data. J. Inflamm. Res., 13, 519–531 (2020).