Histone deacetylase (HDAC) inhibition modulates dendritic cell (DC) functions and regulates experimental graft-vs-host disease and other immune-mediated diseases. The mechanisms by which HDAC inhibition modulates immune responses remain largely unknown. STAT-3 is a transcription factor shown to negatively regulate DC functions. In this study we report that HDAC inhibition acetylates and activates STAT-3, which regulates DCs by promoting the transcription of IDO. These findings demonstrate a novel functional role for posttranslational modification of STAT-3 through acetylation and provide mechanistic insights into HDAC inhibition-mediated immunoregulation by induction of IDO.
Histone deacetylase (HDAC)3 inhibitors potently modulate experimental graft-vs-host disease, allograft rejection, and autoimmune diseases (1, 2, 3, 4, 5, 6, 7, 8, 9, 10), partly through the regulation of dendritic cells (DCs) (1, 11, 12, 13). The molecular mechanisms underpinning their immunosuppressive effects on DCs are not well understood.
STAT-3 negatively regulates DC functions and its activation requires posttranslational phosphorylation (14, 15, 16, 17). However, whether other types of posttranslational modifications are relevant for its function in modulating DC responses is not known (18, 19, 20). Emerging data suggest that STATs, including STAT-3, can also be acetylated (21, 22, 23, 24), but the functional relevance of acetylation in DC responses remains undefined.
We recently demonstrated that HDAC inhibitors partly modulate DC functions through induction of IDO (1). Seminal studies by Puccetti and colleagues demonstrate that STAT-3 activation induces IDO (25). Because HDAC inhibitors can target nonhistone proteins such as STAT-3 (24), we hypothesized that STAT-3 acetylation might be critical for induction of IDO and modulation of DCs. Indeed, HDAC inhibition acetylated and activated STAT-3, which was critical for the induction of IDO and the regulation of DCs.
Materials and Methods
C57BL/6 (B6) mice were purchased from The Jackson Laboratory and were cared for under the regulations of the University Laboratory Animal Medicine guidelines.
DC isolation and PC3 cell lines
To obtain DCs, bone marrow cells from B6 mice were cultured with murine recombinant GM-CSF (10 ng/ml; BD Pharmingen) and IL-4 (10 ng/ml; Peprotech) for 7–8 days and harvested as described previously (1). DCs were harvested and positively selected by the autoMACS Pro Separator (Miltenyi Biotec) for CD11c+ cells. Stable PC3 cell lines expressing pcDNA3 empty vector (STAT3-null), wild-type STAT3, and Stat3K685R were obtained as described previously (24).
Reagents, treatment, and Abs
Suberoylanilide hydroxamic acid (SAHA) and ITF2357, both hydroxamic acid-containing molecules, were obtained and used as described previously (1). DCs were treated for 14–16 h with SAHA, ITF2357, LPS and JSI-124 (cucurbitacin I). STAT-3, phosphor-STAT-3 (Tyr705), acetylated lysine (Cell Signaling), polyclonal IDO (Alexis), acetyl-histone H4 Ab (Upstate), and p300 and β-actin Abs (Upstate) were used.
ELISA, RNA isolation, RT-PCR, immunoprecipitation, immunoblotting, and chromatin immunoprecipitation assay (ChIP)
PCR of mouse IDO and immunoprecipitation, immunoblotting, and ChIP were performed as described before (1, 24, 261, 2). ChIP was performed as described previously, with some modifications (3). Briefly, cells were cross-linked with 1% formaldehyde for 30 min and quenched with glycine. They were then harvested, washed twice with cold PBS, resuspended in lysis buffer, sonicated, and cooled in an ice bath by using a Fisher Scientific 550 Sonic Dismembrator followed by centrifugation at 13,000 × g + concentration was adjusted to 200 mM by adding NaCl followed by incubation at 55°C for 3 h and 65°C for 6 h to reverse the formaldehyde cross-linking. DNA fragments were precipitated. For PCR, 2 μl from 30 μl of DNA extraction was used. PCR primers correspond to sequences within IDO promoter gamma-activated sequence (GAS) regions as follows: forward, 5′-CTCCTTTTATGGGTGATTGTTTCC-3′; reverse, 5′-GAGAACTCCTAAGTTTATGTCCAC-3′ (1).
Generation of reporter constructs, transfection, and assessment of promoter activity
A 1500-bp DNA fragment upstream of the mouse IDO gene start codon was cloned from the mouse small intestine DNA library and the fragment was inserted into luciferase reporter plasmid pGl4.20Luc. Using the Transformer site-directed mutagenesis kit, various mutants were generated that were transfected into CD11c+ DCs using N-[1-(2,3-dioleoyloxy)propyl]-N,N,N-trimethylammonium methyl sulfate) (see supplemental data).4
All statistical analyses were performed by using the paired t test.
Results and Discussion
STAT-3 was acetylated and IDO protein was induced in murine BM DCs that were treated with both SAHA and ITF2357 (Fig. 1⇓, A and B). We reasoned that the inhibition of HDAC enzymes likely facilitated STAT-3 acetylation by the endogenous histone acetyltransferase enzyme p300 in DCs (24). Accordingly, STAT-3 was recovered from the immunoprecipitated p300 prepared from the DCs (supplemental Fig. 1A). Moreover, as shown in supplemental Fig. 1A, STAT-3 formed dimers in DCs treated with HDAC inhibitors but not in the control-treated DCs, demonstrating that acetylation is associated with dimerization and activation of STAT-3 in DCs (24). These results show that the enzymatic activity of p300 in the STAT-3-p300 complex was inhibited by HDAC enzymes in untreated DCs and that this inhibition was reversed upon treatment with HDAC inhibitors.
We next explored the functional relevance of STAT-3 acetylation in DCs. BM DCs were pretreated with either diluent control or JSI-124, a drug that specifically disrupts STAT-3 DNA complex formation (27). DCs were then conditioned with HDAC inhibitors and thereafter stimulated with LPS for an additional 6 h. SAHA and ITF2357 markedly reduced LPS-induced secretion of TNF-α from the DCs (Fig. 1⇑C). By contrast, pretreatment with the STAT-3 inhibitor JSI-124 abrogated the suppressive effect of HDAC inhibitors on LPS-induced TNF-α secretion (1). JSI-124 pretreatment also abrogated HDAC inhibition-mediated reduction of allogeneic T cell proliferation by the DCs (Fig. 1⇑D).
DCs were next pretreated with JSI-124 or vehicle and then conditioned with HDAC inhibitors alone or with LPS (a potent inducer of STAT-3 phosphorylation). As expected, LPS induced STAT-3 phosphorylation whereas treatment with HDAC inhibitors alone did not enhance phosphorylation of STAT-3 (Fig. 1⇑, E and F). Collectively these data show that HDAC inhibitors acetylate and activate STAT-3 without altering its phosphorylation status and that disruption of STAT-3 activity with JSI-124 reverses HDAC inhibition-mediated suppression of DC functions.
We had previously demonstrated that HDAC inhibition modulates DC functions, in vivo and in vitro, partly through the induction of IDO (1). Because blockade of STAT-3 with JSI-124 mitigated the suppressive effects of HDAC inhibition, we determined whether HDAC inhibition-mediated induction of IDO is dependent on STAT-3. Although we have treated the DCs with noncytotoxic doses of SAHA, it is possible that JSI-124 might be directly cytotoxic either alone or after subsequent treatment with SAHA. To rule out any confounding effects of cytotoxicity, we first evaluated the viability of DCs with annexin staining after treatment with JSI-124 alone or followed by treatment with SAHA. As shown in supplemental Fig. 1B, JSI-124 and SAHA treatment did not significantly alter DC viability at the doses that were used. Induction of IDO mRNA by HDAC inhibition was abrogated upon pretreatment with JSI-124, suggesting that STAT-3 is necessary for transcription of IDO (Fig. 2⇓A). To confirm a direct role for STAT-3, we used TESS (Transcription Element Search Software) promoter analysis software and searched the IDO promoter for the potential STAT binding consensus sequence TTCN3GAA, designated as gamma-activated sequence or GAS (28). We found two such sites upstream of the start codon designated as GAS-1 and GAS-2 (Fig. 2⇓B). A ChIP assay demonstrated that SAHA induced STAT-3 binding to the IDO gene promoter (Fig. 2⇓C). As expected, SAHA also acetylated histone 4 at the IDO promoter, but STAT-3 and acetylated histone 4 were not bound to each other (supplemental Fig. 1C).
Next, to directly test whether the binding of STAT-3 to the IDO promoter is absolutely necessary for HDAC inhibition-mediated transcription of IDO, we next performed mutagenesis studies. We cloned the IDO promoter and designed site-directed GAS deletion mutants (supplemental Fig. 1D) that were then inserted it to a pGL4 luciferase reporter vector. As shown in Fig. 3⇓A, IDO promoter-driven luciferase induction was enhanced several fold upon treatment with SAHA. Although SAHA treatment also enhanced the induction of the reporter genes containing deletions of either GAS-1 (MUT-1) or GAS-2 (MUT-2) sites, it was significantly less than the expression driven by the wild-type promoter (Fig. 3⇓A). By contrast, constructs with deleted GAS-1 and GAS-2 (MUT-3) were unable to respond to SAHA treatment. Furthermore, induction of luciferase by SAHA was completely blocked after treatment with JSI-124 in the wild-type promoter controls (Fig. 3⇓A). A similar pattern was observed in cells that were treated with ITF2357 (data not shown), demonstrating a direct requirement for IDO induction by HDAC inhibitor-induced STAT-3 activation and binding to the GAS regions of the IDO promoter.
Functional relevance of STAT-3 activation in DCs has been attributed to its posttranslational phosphorylation, and previous reports have suggested that JSI-124 regulates STAT-3 activity by inhibiting phosphorylation (27, 29). By contrast, our data show that the HDAC inhibitors acetylate STAT-3 but do not alter its phosphorylation status (Fig. 1⇑, E and F), induce IDO, and modulate the function of DCs, which was blocked by JSI-124, a small molecule that disrupts STAT-3 binding to DNA.
These observations show that HDAC inhibition increased acetylation with no alteration of phosphorylation of STAT-3, but they do not show a critical requirement for acetylation of STAT-3 in the induction of IDO. To this end we tested the effects of SAHA and ITF2357 on the induction of IDO in cell lines expressing pcDNA3 empty vector (STAT-3 null), wild-type STAT-3, and the STAT-3K685R mutant, which contains the Lys685-to-Arg substitution and therefore cannot be acetylated (K685R) (24). HDAC inhibition enhanced IDO expression in the wild-type STAT-3-transfected cells but not in the null controls (Fig. 3⇑, B and C). HDAC inhibition also failed to increase IDO expression in the cells transfected with acetylation-resistant STAT-3K685 mutant. These data thus demonstrate a critical role for STAT-3 acetylation in the induction of IDO.
Heretofore, the critical pathways responsible for the induction of IDO and the relevant posttranslational requirements for the function of STAT-3 remained unknown. In this study we demonstrate a novel role for STAT-3 acetylation in the induction of IDO with HDAC inhibition. This, to our knowledge, is also the first direct demonstration of a role for the acetylation of nonhistone proteins in direct regulation of DCs. It is however possible that acetylated STAT-3 might target other genes in addition to IDO, which might contribute either directly or indirectly to the regulation of DC function (17, 30). Furthermore, our data also do not rule out a requirement for the phosphorylation of STAT-3 in mediating its other downstream effects in DCs. Future studies will determine the specific gene targets that are dependent on the acetylation, phosphorylation, or both modifications of the STAT-3 protein in the regulation of DC functions. Although IDO is partly critical for the HDAC inhibition-mediated regulation of DCs (1), it remains to be determined whether this alone or other distinct pathways of suppression of costimulatory molecules and proinflammatory cytokines (1, 11, 12, 13) might act in concert to cause the overall DC-suppressive effects of HDAC inhibitors. Further studies will need to be performed to carefully determine the effects of these pathways in the absence of IDO secretion. Nonetheless, together our data demonstrate that posttranslational modification of STAT-3 through acetylation is responsible, in part, for the induction of IDO and the immunoregulatory effects of HDAC inhibitors on DCs.
The authors have no financial conflict of interest.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
↵1 This work was supported by National Institutes of Health Grants AI-075284 (to P.R.), HL-090775 (to P.R.), AI-15614 (to C.A.D.), and HL-68743 (to C.A.D.). P.R. is a recipient of the Alaina J. Enlow Scholar Award and the Doris Duke Clinical Scientist Development Award.
↵2 Address correspondence and reprint requests to Dr. Pavan Reddy, 6310 Comprehensive Cancer Center, 1500 East Medical Center Drive, Ann Arbor, MI 48109-0942. E-mail address:
↵3 Abbreviations used in this paper: HDAC, histone deacetylase; ChIP, chromatin immunoprecipitation; DC, dendritic cell; GAS, gamma-activated sequence; SAHA, suberoylanilide hydroxamic acid.
↵4 The online version of this article contains supplemental material.
- Received January 6, 2009.
- Accepted March 17, 2009.
- Copyright © 2009 by The American Association of Immunologists, Inc.