Mycobacterium tuberculosis infection induces IL-10 gene expression by disturbing histone deacetylase 6 and histonedeacetylase 11 equilibrium in macrophages

A B S T R A C T
Mycobacterium tuberculosis (MTB) infection is a significant contributor to dysregulated T cell-mediated immune response. Here we aimed to evaluate the mechanism of MTB infection in promoting interleukin-10 (IL-10) up- regulation. The IL-10 levels in MTB infected THP-1 cells were evaluated with enzyme-linked immunosorbent assay (ELISA) and quantitative real-time PCR (qRT-PCR). In challenged THP-1 cells, the HDAC6 and HDAC11 mRNA and protein levels were monitored at varied duration after MTB infection. Further, chromatin im- munoprecipitation (ChIP) analysis was used to investigate the interaction between IL-10 expression and HDAC6 or HDAC11. HDAC6 and HDAC11 were overexpressed or silenced to study their effects on IL-10 regulation. IL-10 was upregulated after MTB challenge in a time- and dose-dependent manner. Similarly, HDAC6 and HDAC11 were also upregulated by MTB challenge. Overexpression or silencing of HDAC6 and HDAC11 changed IL-10 expression correspondingly. MTB infection disturbs the HDAC6/HDAC11 levels to induce IL-10 expression. Strategies to counteract the dysregulation of HDAC6/HDAC11 would potentially alleviate the immunological disordered following MTB infection.

1.Introduction
Tuberculosis (TB) resulted from Mycobacterium tuberculosis (MTB) infection constitutes a leading cause of mortality and morbidity worldwide. It is estimated that about 1.5 million deaths are caused by MTB infection annually [1]. Cytokine induction is a putative con- sequence of MTB infection and serves as mechanism for controlling and clearing this intracellular pathogen [2]. However, upon MTB infection, interleukin-10 (IL-10) is markedly elevated in pleural fluid, bronch- oalveolar lavage fluid, serum, sputum, etc. [3]. IL-10 is an anti-in- flammatory and immunosuppressive molecule that plays an important role in protection against autoimmunity [4]. Macrophages are primary sources of IL-10 [5]. IL-10 reduces antigen presenting and attenuates phagosome maturation, thereby impairing bacterial clearance and in- ducing long-term MTB infection in the lung [6]. In addition, over- expression of IL-10 also leads to immunological disorders that further compromise quality of life of patients [7].Understanding the precise mechanism of IL-10 induction is paramount to promoting pathogen clearance and alleviating im- munological disorders. Previous studies have suggested the indis- pensable role of immune cells in regulating IL-10 production [8]. Beside macrophages, activated effector T cells serve as another source of IL-10 and predominantly account for IL-10-induced susceptibility [9]. In line with this, in coincident diabetes mellitus, T cells respond sensitively to MTB infection through an IL-10 and TGF-β-dependent manner [10], implicating a close alliance between T cells and IL-10. Histone deace- tylase 6 (HDAC6) and histone deacetylase 11 (HDAC11) are two im- portant proteins that transcriptionally modulate IL-10 in antigen pre- senting cells (APCs) [11]. HDAC6 is a transcription activator of IL-10 expression while HDAC11 is a transcription suppressor of IL-10 [11]. It has been previously demonstrated MTB infection acts on members of HDAC family to modulate cytokine production in macrophages [12]. However, no studies have investigated how HDAC6 and HDAC11 are affected by MTB infection and how this correlates with IL-10 levels.Herein, we aim to elucidate the effect of MTB infection on HDAC6 and HDAC11 expression and how this leads to elevated IL-10 expression. Strategies to normalize HDAC6 and HDAC11 levels could potentially improve defense against TB in patients with a high IL-10 expression.

2.Materials and methods
Rabbit anti-HDAC6 and anti-HDAC11 were purchased from Abcam (Cambridge, MA). Horseradish peroXidase (HRP)-conjugated secondary antibodies were from Sigma-Aldrich (St. Louis, Mo, USA). Enhanced chemiluminescence (ECL) reagents were from Thermo Fisher (Waltham, MA, USA). All other reagents were purchased from Sigma- Aldrich, unless otherwise stated.Handling of virulent MTB strain H37Rv and the avirulent strain H37Ra was carried out in a Biosafety Level III facility. Middlebrook 7H9 Collection (ATCC), were cultured in RPMI-1640 medium supplemented with 10% fetal bovine serum (FBS), 2 mM L-glutamine, 25 mM HEPES,1.5 g/L sodium bicarbonate. Cells (3 × 106/ml) were plated for dif- ferentiation into macrophages for 24 h after passaging in complete RPMI containing phorbol 12-myristate 13-acetate (PMA, Sigma- Aldrich, USA) (20 ng/ml). After removing the supernatant, the cells were infected with MTB using the multiplicity of infection (MOI) of1–10 as indicated Fig. 1, according to previously established doses [13–15]. After incubation for 1–48 h at 37 °C, cells were then washed with complete RPMI supplemented with gentamycin (10 μg/mL) to remove extracellular bacilli. We carried out most of our experimentwith MOI of 10 and incubation time of 24 h.Total RNA was extracted from macrophages using TRIzol Reagent (Gibco, Grand Island, NY, USA) according to the manufacturer’s pro- tocol. RNA of 1 μg was converted to cDNA using the GoScript™ Reverse Transcription System (Promega, Madison, WI, USA). Then, the cDNA broth (Difco) supplemented with 0.4% glycerol, 0.05% Tween 80 and was used for qPCR using the SYBR® Green SupermiX (Bio-Rad 10% OADC (Becton Dickinson), was used for bacteria culture. Mid-log culture was used for further studies after being harvested by cen- trifugation and dispersion by passing the bacterial solution through a 30-gauge needle.THP-1 cells, which were acquired from American Type Culture Laboratories, Hercules, CA, USA) on a thermal cycler (Bio-Rad, USA).

EXpression of human GAPDH was used for normalization. The primers used for qPCR are as follows: IL-10, TCACCTTCCAGTGTCTCGGA (for- ward) and TAGCTGGGATTACAGGTGCG (reverse); HDAC6, AAGTAG GCAGAACCCCCAGT (forward) and GTGCTTCAGCCTCAAGGTTC (re- verse); HDAC11, TCTTCCTCCCCAACTTCCCTT (forward) and CTCCAC ACGCTCAAACAGAA (reverse); GAPDH, GGAGCCAAAAGGGTCATCAT(forward) and CTCCACACGCTCAAACAGAA (reverse). Supernatants of cell infected macrophages were collected after centrifugation and undergo filtering. IL-10 was quantified determined using an ELISA kit acquired from R&D System (Minneapolis, MN, USA) in accordance to manufacturer’s recommendations.After MTB infection (MOI 20:1), nuclear extracts of the macro- protein A sepharose beads (Abcam, UK) with gentle shaking. After washing with buffer, DNA was eluted, purified and assessed by qRT- PCR.Cells were transfected with siRNAs for HDAC6 or HDAC11 for 36 h using Lipofectamine (Invitrogen, Carlsbad, USA) in Opti-MEM medium (Invitrogen, USA). Cells then underwent infection with MTB H37Rv at an MOI of 20:1. All siRNAs were purchased from Thermo Fisher. phages were prepared using Nuclear EXtraction Kit (Bio-Rad Laboratories, Hercules, USA). Protein concentration was determined using BCA assay kit (Thermo Fisher Scientific, Rockford, IL, USA). Proteins were then undergone SDS-PAGE using polyacrylamide gels (12%), following by electrotransfer to methanol-activated PVDF mem- branes (Millipore, Billerica, MA, USA). After primary antibody in- cubation, HRP-conjugated secondary antibody was added and the proteins were detected by chemiluminescence and analyzed using ImageJ software. GAPDH was used loading control.ChIP was performed using the ChIP kit according to manufacturer’s protocols. Briefly, macrophages were plated at the density of 3 × 106 cells/ml and infected MTB H37Rv. Uninfected macrophages were used as control. Macrophage cells were then treated with 1% PFA for 10 min at 37 °C to crosslink histones to DNA. After cell lysis, the chromatin was sheared by sonication and centrifuged at 14,000 rpm for 15 min at 4 °C. Fragmented chromatin in the supernatant were im- munoprecipitated with antibodies against HDAC6 or HDAC11 using All statistical analyses were performed with SPSS. Data were ex- pressed as mean values ± standard deviations. Statistical analysis was performed using non-parametric analysis of variance (one-way ANOVA). A p value of less than 0.05 was considered significant.

3.Results
IL-10 is an important pro-inflammatory cytokine. We hypothesized that mycobacterium tuberculosis could promote inflammatory re- sponses through downregulating IL-10. To verify that dysregulated expression of IL-10 is a consequence of MTB infection, we infected THP- 1 cells with MTB, followed by evaluating IL-10 expression with both ELISA (Fig. 1A and B) and qRT-PCR (Fig. 1C and D). It was found that after infection of MTB at MOI of 10, IL-10 was dramatically increased and at 24 h, the expression of IL-10 reached the highest level (Fig. 1A and C). Further, we showed that at 24 h, IL-10 expression also increased with MOI value (Fig. 1B and D).HDAC6 and HDAC11 are putative regulators of the inflammatory responses. Studies have suggested that IL-10 is mediated by the ex- pression of both HDAC6 and HDAC11. Considering that MTB infection elevates IL-10 levels, we were intrigued to evaluate the correlation of MTB infection and HDAC6 and HDAC11 expression. As shown in Fig. 2A–D, we observed upregulation of HDAC6 levels and down-regulation of HDAC11 levels after MTB infection, consistent with rolesof HDAC6 and HDAC11 as pro- and anti-inflammatory factors, respec- tively. The intricate regulation of HDAC6 and HDAC11 levels maintains inflammatory responses in a normal level. We therefore hypothesized that MTB infection could promote inflammatory responses by dis- turbing the expression of HDAC6 and HDAC11.

To further confirm that HDAC6 and HDAC11 are direct regulators of IL-10, we performed qPCR analysis of the ChIP products of IL-10 with HDAC6 or HDAC11. As shown in Fig. 3A–B, increasing HDAC6 and decreasing HDAC11 ChIP products with IL-10 promoter were seen afterMTB infection. These data revealed that MTB infection induced the binding between HDAC6 and the promoter region of IL-10 but reduced the recruitment of HDAC11 to IL-10 promoter, corroborating the im- portance of HDAC11/HDAC6 levels on IL-10 expression.As another validation method, in infected THP-1 cells, we over- expressed or silenced HDAC6 and HDAC11. As expected, after MTB infection, cells with HDAC6 overexpression demonstrated marked in- crease in IL-10 expression, while the HDAC6 knockdown cells showed a significantly lower IL-10 expression both in protein and mRNA levels(Fig. 4A and B). We also investigated anti-inflammatory factors, in- cluding IL-12 and IFN-γ, in these cells. In contrast to changes in IL-10 levels, HDAC6 overexpression reduced IL-12 and IFN-γ but HDAC6 si- lencing increases IL-12 and IFN-γ levels (Fig. 4C and D). Consistent with that HDAC11 downregulation was seen after MTB infection, we sawthat in HDAC11-overexpressing cells, IL-10 levels were downregulated and in HDAC11-silenced cells, IL-10 level was induced (Fig. 5A and B). The levels of IL-12 and IFN-γ followed the opposite trend. Collectively,these data indicated that HDAC6 and HDAC11 expression was para-mount to maintaining IL-10 levels and normalization of these HDAC6 and HDAC11 levels could ameliorate the dysregulated inflammation induced by MTB infection (Fig. 5C and D).

4.Discussion
IL-10 is a key modulator of immune responses by adopting an in- hibitory and anti-inflammatory role [16]. It is critical in generating antigen presenting cells (APCs) with tolerogenic characteristics for preventing self-tissue damage [17]. However, the upregulated IL-10 expression upon MTB infection is linked to the capacity of MTB in evading immune responses, which results in long-term infections in the lung [18]. It has been a long-lasting endeavor to clarify the mechanism of IL-10 upregulation and reactivate immune defense against MTB. In the present study, we corroborated that IL-10 upregulation is a con- sequence of MTB infection and the upregulation is in a dose and time- dependent manner. Our study also confirms that macrophages are an important cellular source of IL-10.Moreover, we show that the IL-10 upregulation appears to be cor- related with the disturbance of the expression of HDAC6 and HDAC11. Previous studies also indicated other transcriptional factors, such as STAT3, Sp, AP-1, NFkB, C/EBPβ, and GATA3 in the regulation of IL-10 [19]. Upon MTB infection, elevated IL-10 level was accompanied by HDAC6 upregulation and HDAC11 downregulation. The direct inter- action of IL-10 with HDAC6 and HDAC11 was also demonstrated by ChIP analysis. This affirms the important role of transcriptional reg- ulation in IL-10 expression [20]. Further, we overexpressed or silenced HDAC6 and HDAC11 expression in THF-1 cells. Attenuation of HDAC6 expression and enhancement of HDAC11 expression dampens the in- crease in IL-10. This provides us the opportunity to normalize HDAC6 and HDAC11 levels to enhance immune response.

The potential of HDAC6 and HDAC11 regulation in IL-10 expression can be also used for treating other diseases associated with abnormal immune responses [21,22]. Indeed, HDAC inhibitors have been utilized in cancer, neurological diseases and immune disorders that are char- acterized by dysregulated IL-10 expression [23]. Particularly in cancer, the effects of HDAC inhibitors go beyond inducing cell apoptosis, cell- cycle arrest, etc., by also exerting regulation on non-tumor cells, such as regulatory T cells [24]. While most currently available HDAC inhibitors are natural compounds that target HDAC family in a broad-spectrum manner [24], our result suggests that inhibiting HDACs non-selectively may not achieve the optimal regulatory effect on IL-10 expression since inhibition of HDAC6 and HDAC11 exerts opposing effects. There is still a lack of isoform-selective HDAC inhibitors [25]. Here we adopted a gene-mediated method to inhibit or promote expression of specific HDAC isoforms, therefore precisely controlling the outcome of IL-10 regulation. Specific HDAC6 suppression and HDAC11 induction can hence be afforded.With significant advances of the clinical use of gene therapeutics [26,27], we can envision that gene therapy for MTB infection and other immunological disorders could dramatically improve the efficacy of TB therapy and amelioration of immune diseases. Our preliminary study potentiates gene-therapy strategies in the in vitro regulation of IL-10 in macrophages. Further in vivo preclinical and clinical study could be useful in demonstrating the clinical potential of this strategy. Further, HDAC6 and HDAC11 are also important regulatory molecules for IFN signaling, regulatory cell function, and immune synapse formation, further comprehensive studies could aid in the elucidation of complex network associated with HDAC6 and HDAC11 expression.

5.Conclusions
In sum, here we corroborated that IL-10 is upregulated in macro- phages after MTB infection. Concomitantly, HDAC6 is overexpressed and HDAC11 is suppressed. Silencing of HDAC6 and upregulating HDAC11 are effective strategies in attenuating IL-10 expression. Our data confirms that significant role of IL-10 in MTB to evade immune responses and opens a new avenue of gene therapy for counteracting the dysregulation of HDAC6 and HDAC11 to restore normal IL-10 ACY-775 levels.