By continuing to browse this site you agree to us using cookies as described in About Cookies
Notice: Wiley Online Library will be unavailable on Saturday 7th Oct from 03.00 EDT / 08:00 BST / 12:30 IST / 15.00 SGT to 08.00 EDT / 13.00 BST / 17:30 IST / 20.00 SGT and Sunday 8th Oct from 03.00 EDT / 08:00 BST / 12:30 IST / 15.00 SGT to 06.00 EDT / 11.00 BST / 15:30 IST / 18.00 SGT for essential maintenance. Apologies for the inconvenience.
The amount of arginine available at inflammatory loci is a limiting factor for the growth of several cells of the immune system. IL-4-induced activation of macrophages produced arginase-1, which converts arginine into ornithine, a precursor of polyamines and proline. Trichostatin A (TSA), a pan-inhibitor of histone deacetylases (HDACs), inhibited IL-4-induced arginase-1 expression. TSA showed promoter-specific effects on the IL-4-responsive genes. While TSA inhibited the expression of arginase-1, fizz1, and mrc1, other genes, such as ym,1 mgl1, and mgl2, were not affected. The inhibition of arginase-1 occurred at the transcriptional level with the inhibition of polymerase II binding to the promoter. IL-4 induced STAT6 phosphorylation and binding to DNA. These activities were not affected by TSA treatment. However, TSA inhibited C/EBPβ DNA binding. This inhibitor induced acetylation on lysine residues 215–216, which are critical for DNA binding. Finally, using macrophages from STAT6 KO mice we showed that STAT6 is required for the DNA binding of C/EBPβ. These results demonstrate that the acetylation/deacetylation balance strongly influences the expression of arginase-1, a gene of alternative activation of macrophages. These findings also provide a molecular mechanism to explain the control of gene expression through deacetylase activity.
In processes, such as wound healing, angiogenesis, and host defense against parasites, the phenotype of macrophages is related to alternative activation [1, 2], which includes arginase-1 expression. This enzyme converts arginine to ornithine, a precursor of polyamines and proline. This conversion is critical to induce the growth and proliferation of several cell types in damaged tissues, the production of the extracellular matrix, and tissue remodeling . An excess of this type of activation contributes to fibrosis in some pathological conditions . In fact, the catabolism of arginine by macrophages has emerged as a critical mechanism for the regulation of the immune response in several parasitic diseases [5-7]. Moreover, high arginase-1 expression has been associated with a variety of conditions, such as cancer, asthma, psoriasis, cardiovascular disease, and also pregnancy . The consumption of arginine by macrophages in the inflammatory loci may impair the growth of other immune cells and thus result in potent immune suppressive activity.
Recently, considerable research effort has been devoted in macrophages to the regulation of genes in their natural setting, namely the chromatin substrate . The recruitment of histone acetyl-transferases (HATs) and histone deacetylases (HDACs) to the transcriptional machinery is a key element in the dynamic regulation of genes. HAT activity promotes the acetylation of histone proteins (particularly H3 and H4), a process that leads to relaxed chromatin structure, thus binding core transcription machinery to DNA and resulting in the initiation of the transcription of several genes. In contrast, HDACs cause the deacetylation of histones, chromosomal condensation, and gene repression . However, despite the common involvement of acetylation/deacetylation in general transcription, deacetylation is associated with activation of the transcription of particular genes, as demonstrated by the observation that histone hyperacetylation inhibits the expression of some genes . Thus, a proper balance between HAT and HDAC activity is required to control chromatin accessibility to specific or general transcription factors . Moreover, in addition to histones, these enzymes can also modify an extended group of proteins, including transcription factors, mitochondrial proteins, RNA-splicing factors, structural proteins, and chaperones. Acetylation affects the affinity of these molecules to bind to other proteins or DNA and can change their half-life, subcellular localization, and even their enzymatic activity. Thus acetylation provides a myriad of potential mechanisms to modulate gene expression. Interestingly, most nonhistone proteins targeted by acetylation are relevant in the regulation of immune functions, cell proliferation, and tumorigenesis [12, 13]. For these reasons, several specific HDAC inhibitors have emerged as anticancer drugs and more recently, due to their suppressive effect on the expression of pro-inflammatory mediators, as anti-inflammatory agents . These compounds are effective in a variety of Th1-dependent inflammatory diseases, such as ulcerative colitis and rheumatoid arthritis, thereby suggesting that cellular HDAC activity exacerbates the inflammatory process . The in vitro gene-specific regulation caused by HDAC inhibitors in macrophages has been described in the context of LPS response (classical activation). By means of genome-wide microarray analysis, it has been shown that trichostatin A (TSA), a new potential HDAC inhibitor for human diseases, modifies the macrophage transcriptome, inhibiting about 32% of all LPS-induced genes and counterregulating a significant percentage of LPS-repressed genes [16, 17]. Recently, it was shown that histone deacetylase 3 is an epigenomic brake of alternative activation of macrophages as it blocks the IL-4-dependent induction of arginase-1 . Also, in a macrophagic cell line, it has been reported that TSA leads to a concentration-dependent suppression of the arginase-1 expression induced by the cAMP analog Br-cAMP .
Given the important role of arginase-1 in the regulation of immune response , here we studied the effect of TSA on the expression of this enzyme when induced by IL-4. Global acetylation induced by TSA had promoter-specific effects on IL-4-responsive genes. While arginase-1, fizz 1, and mrc1 induction was inhibited, other genes such as ym1 and macrophage galactose N-acetyl-galactosamine-specific Lectin CD301a, also called mgl1, showed increased expression. In the case of mgl2, no modifications were observed. To further extend our results, we studied the molecular mechanisms involved in TSA-mediated arginase-1 inhibition. The induction of this gene by IL-4 requires a DNA response element composed by adjacent STAT6 and C/EBPβ transcription factor-binding sites. Moreover, acetylation on lysine residues 215–216 of C/EBPβ is an important regulator that modulates protein-DNA binding on the arginase-1 promoter and, therefore, modulates IL-4-dependent arginase-1 expression in macrophages.
Not all the genes induced during the alternative activation of macrophages are downregulated by TSA
In our experiments, we used bone marrow-derived macrophages, a homogeneous population of primary and quiescent cells. Treatment of these cells with several cytokines causes several modifications that allow them to develop their functional activities . To test whether deacetylase activity has any functional implication in IL-4-dependent arginase activity of macrophages, we used the pharmacological HDAC inhibitor TSA in our assays. The TSA concentration used (20 nM) did not induce cellular toxicity in our experimental conditions . We studied whether acetylation is required for all the IL-4-dependent genes induced in macrophages. For this purpose, we explored the mRNA expression of several genes induced by IL-4 treatment in a dose- and time-dependent manner. As expected, fizz1, mrc1, arginase-1, mgl1, and ym1 were strongly induced at 10 h of IL-4 treatment, a time-point at which expression was clearly detectable  (Fig. 1A and B). Although the treatment with TSA inhibited the expression of several of these genes, it was not a general phenomenon. The HDAG inhibitor significantly reduced fizz1, mrc1, and arginase-1 expression (Fig. 1A), but not that of other genes such as mgl1, mgl2 and ym1 (Fig. 1B). The latter may be due to the positive role of histone acetylation in transcription. These results confirm that deacetylation is required for the transcriptional activation of some IL-4-responsive genes, as previously reported in the context of LPS response [16, 17], and suggest that other nonhistone-related targets of HDACs are involved in the transcriptional regulation of these genes. Although all the genes induced by IL-4 are dependent on Stat-6, not all have the same transcriptional or post-transcriptional mechanisms of induction.
Deacetylase activity is required for IL-4-dependent arginase activity
IL-4-induced arginase activity in macrophages (Fig. 2A) . Treatment with TSA significantly reduced this activity. However, of note, the treatment with TSA alone produced an increase in arginase activity. The assay for arginase activity was done with cell lysates, to which we added the substrate . Under these conditions, we were unable to differentiate the arginase activity caused by arginase-1 located in the cytosol or by arginase-2 in the mitochondria. Therefore, we studied the effects of TSA on the expression of each enzyme. Real-time PCR analysis of both mRNAs showed that TSA exerted different effects on the two arginase isoforms. Although arginase-1 mRNA-induced expression by IL-4 or 8-Br-cAMP was significantly inhibited (Fig. 2B), expression of arginase-2 was not (Fig. 2C). This observation could explain why TSA treatment increased arginase activity. Thus, taken together, these results showed that these two arginase isoforms are regulated by a specific acetylation/deacetylation mechanism in different manners in response to IL-4 and other stimuli.
TSA prevents RNA polymerase II recruitment to arginase-1 promoter
To determine whether the decrease in arginase-1 mRNA was due to an inhibition of the mRNA production or to an increase in its degradation, we measured the rate of mRNA degradation. Macrophages were treated with IL-4 for 6 h and then with DRB  at a concentration sufficient to block all further RNA synthesis, as determined by (3H)UTP incorporation . RNA was isolated from aliquots of cells at different times after the addition of DRB and actinomycin D. This approach allowed us to estimate the half-life of arginase-1 mRNA . Under these conditions, the mRNA was stable (Fig. 3A). After 6 h of DRB and actinomycin D treatment, there were no modifications for the half-life of arginase-1 mRNA. The treatment with TSA simultaneously with IL-4 did not modify the half-life of arginase-1 (Fig. 3A). As a control, we determined the half-life of the c-myc mRNA. The mRNA of this protoncogene was very unstable with a half-life of less than 1 h . When we extrapolated the amounts of c-myc RNA, we obtained a t1/2 of around 30 min (Fig. 3A). These results demonstrate that the reduced levels of mRNA were not due to a decrease in the half-life of RNA but to an inhibition at the transcriptional level.
Because HATs and HDACs play a crucial role in the formation of transcription pre-initiation complexes, we next tested the effect of TSA on the recruitment of RNA polymerase II upon IL-4 stimulation of macrophages. Using chromatin immunoprecipitation, we found that RNA polymerase II was recruited to the arginase-1 promoter after 3 or 6 h of IL-4 treatment. However, this recruitment did not occur when the cells were treated with TSA (Fig. 3B). The specificity of the reaction was checked by using unrelated antibodies or a fragment of the coding region (data not shown). This observation indicates that TSA inhibits IL-4-mediated transcription by blocking the recruitment of the basal complex machinery. Taken together, these results suggest that deacetylase activity is required to recruit RNA polymerase II and activate the transcription of IL-4-dependent arginase-1 induction in macrophages.
A detailed analysis of the time course of arginase-1 expression showed that it increased rapidly after IL-4 treatment, reaching a maximum between 6 and 10 h and then progressively diminishing, but still detectable 24 h after IL-4 activation (Fig. 3C). When TSA was administrated before or at the same time as IL-4, the expression of arginase-1 was blocked (Fig. 3C). It is known that arginase-1 expression is dependent on new protein synthesis, although the required protein remains unknown . To exclude that TSA inhibited arginase-1 expression by blocking the induction of this unknown protein, TSA was added after 3 h of IL-4 treatment. Even when transcription was initiated, the addition of TSA reduced arginase-1 induction (Fig. 3D). These data allow us to conclude that the TSA acts directly on arginase-1 transcription machinery and not through the inhibition of a protein necessary for the transcription.
Deacetylase activity is not required by STAT6 binding to the arginase-1 promoter
The transcription of arginase-1 induced by IL-4 in macrophages is regulated by a composed element, placed about 3 kb upstream of the start transcription site, which binds STAT6 and C/EBPβ  (Fig. 4A). A similar element is present in other genes induced by IL-4 and whose expression is inhibited by TSA, such as Fizz1 . Ym1, which is not downregulated by TSA, has no C/EBPβ-binding sites adjacent to STAT6 boxes . Therefore, we concentrated our efforts on determining whether TSA acts on the transcription factors that bound to the precise area in the promoter.
STAT6 plays a critical role in the arginase-1 expression induced by IL-4 in macrophages since this gene is not expressed in macrophages from STAT6 KO mice . For this reason, we explored the potential role of deacetylase activity on the STAT6 transduction pathway. In this regard, it is known that IL-4 stimulation promotes STAT6 phosphorylation, thereby allowing its dimerization and nuclear translocation [30, 33, 34]. TSA treatment did not modify the IL-4-induced phosphorylation of STAT6, as shown by western blot with antibodies against tyrosine-phosphorylated STAT6 (Fig. 4B). Moreover, the in vitro DNA-binding capacity of STAT6 was tested by EMSA assay (Fig. 4C). As a probe, we used a sequence of the arginase-1 promoter, which includes the STAT6-binding element placed 2.86 Kb upstream of the transcription start site . While the nuclear extracts from untreated macrophages did not bind to this probe, a shift was observed when we used those from IL-4-treated macrophages. This band was specific because treatment with 100-fold excess of cold oligonucleotide eliminated the binding but not the competition with an oligonucleotide that contains the mutated STAT6 box. In support of the results on the STAT6 phosphorylation, the in vitro DNA binding of STAT6 was not modified by TSA treatment (Fig. 4B).
To confirm the results, we performed DNA-binding assays in vivo using the chromatin immunoprecipitation technique. The treatment of macrophages with IL-4 induced an increase in the binding of STAT6 (Fig. 4D). However, this activity was not affected by TSA treatment. Therefore, these results indicate that TSA does not affect the IL-4-STAT6 pathway involved in arginase-1 expression in macrophages.
Deacetylase activity is required for C/EBPβ binding to an enhancer element of arginase-1
So far, we can exclude that TSA affects STAT6. Therefore, we next tested the effect of TSA on the binding of C/EBPβ to the promoter of arginase-1. For this purpose, chromatin immuno-precipitation assays were performed. IL-4 treatment induced a strong binding of C/EBPβ to the arginase-1 promoter while the addition of TSA abolished this binding (Fig. 5A). The attachment of C/EBPβ to DNA was abrogated in macrophages from STAT6 KO mice (Fig. 5B), suggesting that the cooperation between STAT6 and C/EBPβ binding to the arginase-1 promoter is essential for C/EBPβ binding to this element.
C/EBPβ is regulated at multiple levels in several cell types [35-37]. Therefore, we first explored the effects of TSA in the mRNA and protein expression of C/EBPβ in IL-4-activated macrophages. IL-4 treatment did not modify C/EBPβ (Fig. 5C) or protein levels (Fig. 5D). This result is consistent with the previous findings that discarded the involvement of early genes on the TSA inhibitory effect. To date, three well-known C/EBPβ isoforms have been described. These share the 145 C-terminal amino acids that contain the basic DNA-binding domain and the leucine zipper dimerization helix. Of these, LAP (liver activator protein) is a transcriptional activator whereas the shortest form LIP (liver inhibitor protein) acts as a transcriptional repressor . It has been proposed that the LAP/LIP ratio determines the final outcome of C/EBPβ activity . However, we did not observe changes in the percentage of expression of any of the isoforms between nonstimulated (starvation conditions) or IL-4-activated cells. Nor were changes detected after addition of TSA (Fig. 5 D).
Because C/EBPβ-induced expression is required for arginase-1 expression in macrophages in response to activation with cAMP , we evaluated the role of TSA on the expression of this transcription factor upon 8-Br-cAMP treatment. After 3 or 6 h of this treatment, C/EBPβ as well as protein levels were increased, thereby confirming previous observations  (Fig. 5C and E). The addition of TSA did not have any effect on the 8-Br-cAMP-induced protein expression, as determined by LAP (Fig. 5E) or LIP (data not shown). This observation reinforces the notion that TSA does not regulate C/EBPβ activity through a modulation of LAP/LIP expression.
TSA abolished the DNA binding of C/EBPβ without modifying the amounts of the protein. This finding prompted us to look for post-transcriptional modifications of this transcription factor in response to TSA treatment that allow us to explain the results observed. C/EBPβ has many lysine residues that are potential substrates of acetylation. Using an antibody against C/EBPβ acetyl-lysine residues 215 and 216, we observed that TSA treatment of macrophages induced C/EBPβ acetylation (Fig. 5F). The acetylation of 215 and 216 residues, placed in the DNA-binding domain of the protein, have been implicated in the regulation of the DNA interaction with C/EBPβ . In conclusion, these data suggest that C/EBPβ acetylation on lysine residues 215–216, which inhibits DNA binding, is at the basis of arginase-1 inhibition.
Chromatin remodeling is an essential mechanism that regulates gene transcription. Acetylation and deacetylation play pivotal roles in modifying not only histone but also the activity of several transcription factors [10, 11, 42]. Here, we provide evidence that deacetylase activity is required for IL-4- and cAMP-dependent arginase-1 induction. The requirement of deacetylase activity is related to arginase-1, as well as other IL-4-induced genes such as fizz1 or mrc1, but is not a common feature since TSA does not inhibit the expression of ym1, mgl1, or mgl2. We have shown that IL-4 and cAMP induce arginase-1 expression in macrophages at transcriptional level and that TSA does not exert its function through an increase in mRNA degradation. These results were confirmed by the observation that TSA abolishes the IL-4- or cAMP-dependent binding of the polymerase II to the arginase-1 promoter. Also, these results suggest that the acetylation of transcription factors or co-activators is involved in the inhibition of transcription.
The transcription of arginase-1 induced by IL-4 or cAMP in macrophages is regulated by a composed element, placed about 3 kb upstream of the transcription start site, which binds STAT6 and C/EBPβ [30, 40]. These elements are also present in other genes whose expression was abolished by TSA. A 3 bp mutation in the CCAAT box of the arginase-1 enhancer element abolishes the response to IL-4, thereby confirming the involvement of C/EBPβ in arginase-1 induction . Using EMSA assays, it was found that C/EBPβ binds to the arginase-1 enhancer . Using chromatin immunoprecipitation assays, we found that a small amount of C/EBPβ was associated with the promoter prior to the IL-4 stimulus. However, the recruitment of C/EBPβ increased 5–6-fold after IL-4 exposure.
STAT6 recruits many HATs, which act as co-activators. Initially, CBP/p300, like SRC1, bind directly to STAT6 [43, 44] and later p/CIP join the complex  to form an enhanceosome that contacts the general transcriptional machinery at the start site. However, the inhibition of HATs by TSA does not prevent the binding of STAT6 to the arginase-1 promoter. In addition to the modification of histone proteins, acetylation has been shown to affect the activities of transcription factors . Here, we show that the induced binding of C/EBPβ to the arginase-1 promoter by IL-4 or cAMP is lost under the effect of TSA. This inhibition of DNA binding correlates with the C/EBPβ acetylation and could be explained by the suppressive effect of TSA on HDAC. We cannot exclude that STAT6 is required for the recruitment of HDACs. Recently, it has been shown in macrophages that STAT6 is a facilitator of the nuclear receptor PPARγ, which promotes DNA binding and consequently increases the number of regulated genes as well as the magnitude of the response . However, in the absence of PPARγ, no reduction of IL-4-dependent induction of arginase-1 was observed.
The regulation of C/EBPβ activity by the acetylation/deacetylation mechanism has been described previously. In concordance with our results, it has been demonstrated that the acetylation of C/EBPβ lysine residues 215–216 represses the transcription of the inhibitor of DNA binding 1 (Id-1) gene by diminishing its DNA-binding activity . It will be interesting to mutate lysine residues 215–216 in C/EBPβ and examine whether this affects alternative activation. However, technically, this experiment is very difficult to perform in nontransformed macrophages due to the inefficient transfection capacity (maximum 5–10% of the cells).
In contrast to the acetylation of C/EBPβ in lysine residues 215–216, acetylation of lysine 39 seems to be required to allow C/EBPβ to act as a transactivator in several genes, these related mainly to adipogenesis and growth hormone response [48, 49]. Thus, acetylation on different lysine residues in the same transcription factor can result in opposite effects on the transcription of specific target genes. We propose that acetylation of C/EBPβ on lysine residues 215–216 determines the outcome of arginase-1 expression during IL-4 stimulation. In this scenario, by early hyperacetylation of these residues, TSA impedes the recruitment of C/EBPβ and RNA poll II to the arginase-1 promoter and limits arginase-1 expression without modifying STAT6 activity.
The induction of arginase by IL-4 requires SHIP degradation . However, a complete understanding of arginase-1 regulation may be relevant because it may provide new targets to control the expression of this gene. Arginase-1 is receiving increasing attention because arginine consumption at inflammatory loci could be a limiting factor of the immune response as a result of the requirement of this essential amino acid for the growth of many cells involved in the immune system [5-7]. Thus, a better understanding of the molecular/signaling mechanism(s) regulating arginase-1 may provide an attractive opportunity to manipulate macrophage activation in diseases in which these cells contribute to the pathology. In line with this hypothesis, and on the bases of the results presented here, deacetylase inhibitors (e.g. TSA) prevent the induction of several genes involved in pathologies such as allergy, asthma, and fibrosis .
Materials and methods
Recombinant murine IL-4 was purchased from R&D systems, 8-Br-cAMP, Actynomicin D and 5, 6-dichlorobenzimidazole riboside (DRB) were from Sigma, and TSA was from Tocris Bioscience. The Abs used were as follows: anti-phospho-STAT6 (Cell Signaling); anti-β-actin (Sigma-Aldrich); anti-acetyl C/EBPβ (Lys215, Lys216) (Millipore); anti-RNA Pol II (N20); anti-C/EBPβ (C-19); anti-histone H1 (N-19); and anti-STAT6 (M20) (Santa Cruz Biotechnology). Peroxidase-conjugated anti-rabbit (Jackson ImmunoResearch Laboratories) or anti-mouse (Sigma-Aldrich) was used as a secondary Ab. All other chemicals were of the highest purity grade available and were purchased from Sigma-Aldrich. Deionized water was further purified with Millipore Milli-Q System A10.
Bone marrow-derived macrophages were isolated from 6-week-old male BALB/c mice (Charles River Laboratories, Wilmington, MA, USA), as described . After 7 days of culture, a homogeneous population of adherent macrophages was obtained (>99% Mac-1+). To synchronize the cells, at 80% confluence, they were deprived of M-CSF for 16–18 h before being subjected to the treatments. Animal use was approved by the Animal Research Committee of the Government of Catalonia (number 2523).
RNA extraction and real-time RT-PCR
RNA was extracted with Tri Reagent (Sigma), following the manufacturer's instructions. One microgram of RNA was retro-transcribed using Moloney murine leukemia virus reverse transcriptase RNase H Minus (Promega) and real-time PCR was performed as described . Data were expressed relative to the expression in each sample of β-actin. The primer sequences are described in Table 1.
Table 1. Primer sequences by real-time PCR
Arginase activity was measured as described . In brief, cells were lysed and arginine hydrolysis was conducted by incubating the lysate with L-arginine at 37°C for 15–120 min. The urea concentration was measured and one unit of enzyme activity was defined as the amount of enzyme that catalyzes the formation of 1 μmol of urea per minute.
Cells were lysed and nuclear extracts obtained as described . EMSA assay was used as described . The probes were synthesized by Sigma and correspond to a STAT6-binding element 2.86 Kb upstream of the arginase-1 transcription start site  and the mutant, as indicated in Figure 4C.
Chromatin immunoprecipitation (ChIP) assay
This assay was performed as described previously . The primers used correspond to the arginase-1 promoter transcription start site: 5′GGCCGTAACCCTAAAAGACA3′ 5′CGGAGCCAGTTGTTGGATA3′ or to the enhancer element placed at −2.88 Kb: 5′GGCACAACTCACGTACAGACA3′, 5′TGAGGCATTGTTCAGACTTCC3′.
The nonparametric Wilcoxon test for paired differences was used in all calculations .
This work was supported by grants BFU2007–63712/BMC and BFU2011–23662 from the Ministerio de Ciencia y Tecnología to AC. We thank Tanya Yates for editing the manuscript.
Conflict of interest
The authors declare no financial or commercial conflict of interest.