MOLECULAR AND CELLULAR BIOLOGY, July 2002, p. 4781–4791 0270-7306/02/$04.00⫹0 DOI: 10.1128/MCB.22.13.4781–4791.2002 Copyright © 2002, American Society for Microbiology. All Rights Reserved.
Vol. 22, No. 13
Kinetics of a Gamma Interferon Response: Expression and Assembly of CIITA Promoter IV and Inhibition by Methylation Ann C. Morris, Guy W. Beresford, Myesha R. Mooney, and Jeremy M. Boss* Department of Microbiology and Immunology, Emory University School of Medicine, Atlanta, Georgia 30322 Received 23 October 2001/Returned for modification 3 December 2001/Accepted 28 March 2002
Chromatin immunoprecipitation assays were employed to assess the kinetics of transcription factor assembly and histone modifications that occur during gamma interferon (IFN-␥) induction of CIITA gene expression. CIITA is the master regulator of major histocompatibility complex class II transcription. Promoter IV (PIV), the major IFN-␥ responsive promoter for CIITA expression, requires both STAT1 and IFN regulatory factor 1 (IRF-1) for induction by IFN-␥. STAT1 binding to PIV was detected first and was accompanied by a modest acetylation of histones H3 and H4 that were associated with the region. Despite these changes, which occurred within 30 min of IFN-␥ treatment, CIITA mRNA was not detected until IRF-1 protein was synthesized and bound to its site, a process that required >120 min. In contrast to these events, fetal trophoblast-like cell lines, which are refractory to CIITA induction by IFN-␥, failed to assemble the above factors or modify their chromatin, suggesting that accessibility to the promoter is blocked. Bisulfite sequencing of PIV showed strong hypermethylation of PIV, providing a link between methylation, chromatin structure, and factor binding. Together, this analysis provides a kinetic view of the activation of the CIITA gene in response to IFN-␥ and shows that regulatory factor assembly, chromatin modification, and gene expression proceed in discrete steps. The presentation of exogenously derived antigens to CD4⫹ helper T cells by major histocompatibility complex (MHC) class II molecules results in the initiation of immune responses (reviewed in reference 44). MHC class II molecules are ␣/ heterodimeric glycoproteins expressed constitutively on the surface of antigen-presenting cells such as activated macrophages, B cells, dendritic cells, and the thymic epithelia (reviewed in reference 13). MHC class II genes can be induced on many cell types by exposure to the inflammatory cytokine gamma interferon (IFN-␥) (4, 9). MHC class II genes are regulated at the transcriptional level by a series of DNA-binding factors, RFX, X2BP/CREB, and NF-Y, that each have subunits that interact directly with CIITA, a transcriptional coactivator required for expression (reviewed in references 3 and 36). CIITA in turn can interact with a variety of general transcription factors and coactivators, including TBP (27), TAFII32 (11), CBP/p300 (12, 22), PCAF (39), and pTEFb (21). Thus, it appears as though CIITA may function as a bridge between DNA bound factors, chromatin modifiers, and the RNA polymerase machinery. Indeed, the association of CIITA with the promoter bound factors in vivo has been shown to lead to the acetylation of histones H3 and H4 at the MHC class II promoter (2). CIITA has been termed the master regulator for MHC class II gene expression in that its expression correlates nearly perfectly with class II expression. Cells that do not express MHC class II genes do not express CIITA. CIITA expression can be induced by IFN-␥ in a time frame that precedes that of MHC class II genes (6, 40). Transcription of CIITA is regulated in a cell-type- and development-specific fashion by four distinct promoters, each of which directs expression of a unique first
exon (32, 49). CIITA promoter I has been shown to be primarily active in dendritic cells. The function of promoter II is not clear at this time. Promoter III is primarily responsible for directing constitutive expression of CIITA in B cells, whereas promoter IV (PIV) regulates IFN-␥-inducible expression of CIITA (32, 49). PIV contains three elements that are required for transcription in response to IFN-␥: a GAS element, which binds the factor STAT1; an E-box, which is bound by the ubiquitous factor USF-1; and an IFN regulatory factor 1 (IRF-1) binding site (31, 35). STAT1 and USF-1 have been shown to bind cooperatively to their respective sites in vitro (31). Although expression of MHC class II genes is inducible in most cell types by treatment with IFN-␥, trophoblast cells, which comprise the embryonic portion of the placenta, fail to upregulate class II upon exposure to this cytokine (34, 50). This provides one possible mechanism of maternal-fetal tolerance, whereby the maternal immune system does not react to placental tissues expressing paternally derived genes. MHC class II gene expression in placenta and in trophoblast-derived choriocarcinoma cell lines is blocked at the transcriptional level (34). The failure to induce class II genes after exposure of trophoblast-derived cells to IFN-␥ is due to inhibition of expression of CIITA (29, 30, 33). This inhibition is likely a result of cytosine methylation of PIV as shown by the reversion of CIITA and MHC class II induction after treatment of these cells by the methylation inhibitor 5-aza-2⬘-deoxycytidine (5AC) (30). Recently, it was shown that CIITA is also transcriptionally repressed in trophoblast cells cultured from placental chorionic villi (46), suggesting that the mechanism of CIITA silencing may be the same in vivo. The induction of MHC class II genes is delayed compared to other IFN-␥-inducible genes (16, 24). This is due to the necessity to synthesize CIITA. However, CIITA expression appears also to be delayed, taking up to 2 h to detect mRNA by reverse
* Corresponding author. Mailing address: Department of Microbiology and Immunology, Emory University School of Medicine, 1510 Clifton Rd., Atlanta, GA 30322. Phone: (404) 727-5973. Fax: (404) 727-1719. E-mail:
[email protected]. 4781
4782
MORRIS ET AL.
transcription-PCR (RT-PCR). This delay could be due to the time required to synthesize and express IRF-1, an IFN-␥-inducible, STAT1-dependent gene product. However, the kinetics of factor binding and assembly, as well as changes in chromatin structure within PIV, in response to IFN-␥ are not known. To better understand the nature of IFN-␥-mediated immune responses and to understand the role of methylation of the CIITA promoter, we have employed chromatin immunoprecipitation (ChIP) assays to investigate the kinetics of factor assembly and chromatin modification of PIV of the CIITA gene in both CIITA-inducible cells and trophoblast-like cells. In cells that can induce MHC class II genes in response to IFN-␥, factor assembly and chromatin modification occurred in an ordered fashion. In contrast, trophoblast-like cells were completely refractory to factor assembly and chromatin modification. Bisulfite sequencing of PIV in these cells showed substantial methylation of the CpG dinucleotides, providing a direct link between methylation, chromatin structure, and factor assembly and accessibility. MATERIALS AND METHODS Cells and cell culture. A431 (CRL-1555; American Type Culture Collection [ATCC], Manassas, Va.) is a human vulvar epidermoid cancer cell line. JAR (HTB-144; ATCC) and JEG-3 (HTB-36; ATCC) are human choriocarcinomaderived cell lines. A431 and JEG-3 cells were grown in Dulbecco modified Eagle medium (Mediatech, Inc., Washington, D.C.) supplemented with 10% fetal bovine serum (HyClone, Inc., Logan, Utah), 2 mM L-glutamine, and 100 U of penicillin-streptomycin (Life Technologies, Inc., Grand Island, N.Y.)/ml. JAR cells were grown in RPMI (Life Technologies, Inc.) supplemented as described above. For some experiments, IFN-␥ (Biogen, Cambridge, Mass.) was added to cells at a final concentration of 200 to 500 U/ml. DNA-mediated transfection assays and flow cytometry. Transient transfection of 40 g of pUC18 and pCOIRF-1 DNAs into A431 and JAR cells was performed by electroporation as described previously (37). pCOIRF-1 expresses an active form of IRF-1 (14). For analysis of CIITA mRNA induction, cells were cultured for 40 h after transfection, at which point half the cultures received IFN-␥ for 2 h prior to RNA isolation and quantitation. CIITA mRNA levels were quantitated by real-time PCR as described below. For analysis of HLA-DR surface expression, cells were cultured for 16 h after transfection. They were then split in half and cultured for an additional 56 h in the presence or absence of IFN-␥. The cells were then stained with anti-HLA-DR phycoerythrin-conjugated antibodies (Becton Dickinson, Inc.) and analyzed on a FACSCaliber apparatus. ChIP. ChIP assays were performed essentially as described previously (2). Antibodies to acetylated H3 and H4 were purchased from Upstate Biotechnology (Lake Placid, N.Y.). Antibodies to USF, IRF-1, and STAT1␣ were from Santa Cruz Biotechnology, Inc. (Santa Cruz, Calif.). Approximately 4 ⫻ 107 cells were used for each chromatin preparation. One-tenth of the formaldehyde crosslinked chromatin sample was used per immunoprecipitation. Immune complexes were collected with protein A-Sepharose beads (60 l/precipitation). The beads were washed two times each with low-salt wash buffer (0.1% sodium dodecyl sulfate [SDS], 1% Triton X-100, 2 mM EDTA, 20 mM Tris-HCl [pH 8.1], 150 mM NaCl), high-salt wash buffer (0.1% SDS, 1% Triton X-100, 2 mM EDTA, 20 mM Tris-HCl [pH 8.1], 500 mM NaCl), LiCl wash buffer (0.25 M LiCl, 1% NP-40, 1% deoxycholate, 1 mM EDTA, 10 mM Tris-HCl [pH 8.1]), and 1⫻ Tris-EDTA. Immunoprecipitated products were quantitated by real-time PCR. Real-time PCRs were performed on an iCycler with an optical assembly unit (Bio-Rad Laboratories, Hercules, Calif.). Quantitation was accomplished by measuring the incorporation of the fluorescent dye SYBR-green into the PCR product at the end of each cycle or the hydrolysis of a TaqMan probe. All PCRs were performed in duplicate, and the results were averaged. One-tenth of the DNA purified from the immunoprecipitation was quantitated with primers for CIITA PIV. For SYBR-green measurements of CIITA PIV, the following primer pairs were used: CIITA 127 (forward primer), 5⬘-GACTCTCCCCGAA GTGGGGCTG; and CIITA 318 (reverse primer), 5⬘-TGGTCATCCTACCTC CCCGCCT (Fig. 2). For TaqMan real-time PCR of the PIV region of CIITA, TM-1 (5⬘-GCCACTGTGAGGAACCGACT) and TM-2 (5⬘-TGGAGCAACCA AGCACCTACT) primers with 5⬘-CAGGGACCTCTTGGATGCCCCA as the probe labeled at the 5⬘ end with the fluorescent molecule 6-FAM and containing
MOL. CELL. BIOL. the quenching moiety TAMRA incorporated at position 9 were used (Fig. 2). Similarly, for glyceraldehyde-3-phosphate dehydrogenase (GAPDH), the PCR primers were 5⬘-AATGAATGGGCAGCCGTTA (forward primer) and 5⬘-TA GCCTCGCTCCACCTGACT (reverse primer). The GAPDH TaqMan probe was 5⬘FAM-CCTGCCGGTGACTAACCCTGCGCTCCT with TAMRA at position 9. DNA samples were heated to 94°C for 1 to 3 min and then cooled to 85°C, at which point Taq polymerase was added. PCR cycling was performed for 40 cycles at 94°C for 15 s and at 55 to 65°C for 30 to 60 s, depending on the primer set, with an extension step at 72°C for 30 s. Each primer set produced a single, unique DNA product. Cycle number threshold values (point at which sample fluorescence is 10 times greater than the background level) from the above reactions were compared directly to a standard curve generated by using 0, 0.8, 4, 20, 100, and 500 ng of purified, sonicated, genomic DNA as a template for the PCR. The mean correlation coefficients for CIITA PIV and GAPDH standard curves were 0.989 with a standard deviation of 0.006 and 0.984 with a standard deviation of 0.021, respectively. All PCR assays were performed in duplicate, and the average value obtained for each sample was normalized to the amount of CIITA PIV chromatin DNA added into the immunoprecipitation reaction. The results from three or more independent chromatin preparations were averaged and are presented as the fold stimulation (of acetylation or factor binding, depending on the antibody) versus untreated cells to account for variations in immunoprecipitation efficiency between antibody lots. Bisulfite sequencing. Bisulfite sequencing of genomic DNA was performed essentially as described by Clark et al. (7). First, 5 g of EcoRI (10 U/g)digested genomic DNA was denatured by the addition of NaOH to a final concentration of 0.3 M, and then incubation was performed at 37°C for 15 min. During this time, 10 mM hydroquinone and 2 M sodium metabisulphite (pH 5.0) were prepared. Then, 1,040 l of sodium bisulfite and 60 l of hydroquinone were added to the denatured DNA, and the solution was overlaid with mineral oil and incubated overnight at 55°C. Removal of free bisulfite was accomplished by purification of the DNA by using the Wizard DNA Clean-Up columns (Promega Corporation, Madison, Wis.) according to the manufacturer’s instructions. NaOH was added to a final concentration of 0.3 M, and the sample was incubated at 37°C for 15 min. The sample was neutralized by addition of 10 M ammonium acetate to 3 M. The DNA was then precipitated, washed with 70% ethanol, dried, and resuspended in 50 l of distilled H2O. Bisulfite treated DNA (1 to 3 l) was used as a template in a 50-l PCR containing 200 M deoxynucleoside triphosphates, 200 nM primers, and 2.5 U of Taq polymerase. PCR (35 cycles) was carried out under the following conditions: 94°C for 30 s, 54 to 55°C for 30 s, and 72°C for 1 min. Product (1 l) from the first PCR was then used in a second PCR with a nested 5⬘ primer. The primers used for the PCRs were specific for bisulfite converted DNA. The primers used to amplify the nontemplate strand were BSp4-1 (forward primer; 5⬘-GGTTGGATTGAGTTG GAGAGAAATAGAGAT), BSp4-2 (nested forward primer; 5⬘-TGGGGATAA GTTTTTTGTAATTTAGGA), and BSp4-3 (reverse primer; 5⬘-CTACTAATA ACCTCTCCCTCCAGCCAA). The primers used to amplify the template strand were P4BS-7 (forward primer; 5⬘-ATTTATYGTTGTTTTTYGGGTTTT), P4BS-8 (nested forward primer; 5⬘-AAAAAACAAAAACCCACCCAAAAA), and P4BS-9 (reverse primer; 5⬘-TAAAAACAAACTCCCTACAACTCAAA). Amplified DNA was ligated into the pCR2.1 vector (Invitrogen, Inc., Carlsbad, Calif.) and transformed into TOP10 competent cells (Invitrogen) as recommended by the manufacturer. Clones were sequenced by automated DNA sequencing at the Emory University DNA Sequencing Core facility. Clones with unconverted non-CpG cytosines were discarded from the analysis. Real-time RT-PCR. RNA was prepared by using TriReagent (Molecular Research Center, Inc., Cincinnati, Ohio), and RT was carried out by using the GeneAmp RNA PCR kit (Perkin-Elmer, Inc.) according to the manufacturer’s instructions. A total of 1 g of RNA was used per sample. Each reaction contained a parallel control with no reverse transcriptase. One-tenth of a reverse transcriptase reaction mixture was used as a template in a real-time PCR. Real-time RT-PCR was performed as described for the ChIP PCR. For each sample, a parallel PCR with primers for GAPDH transcripts was also performed as an input control. To compare samples, the results for CIITA RT-PCR assays were normalized to results obtained for the corresponding GAPDH RT-PCR assays, providing a relative quantitation value. The sequences of the primers used for real-time RT-PCR were as follows: GAPDH5⬘, 5⬘-CCATGGGGAAGG TGAAGGTCGGAGTC; GAPDH3⬘ no. 2, 5⬘-GGTGGTGCAGGAGGCATTG CTGATG; CIITA-3, 5⬘-CTGAAGGATGTGGAAGACCTGGGAAAGC; and CIITA-6, 5⬘-GTCCCCGATCTTGTTCTCACTC. Immunoprecipitations and immunoblot analysis. Whole-cell lysates were prepared and immunoprecipitations were performed as previously described (10). STAT1 immunoprecipitations were performed with STAT1␣ (Santa Cruz Biotechnology) antibodies, which were prebound to paramagnetic beads. After 3 h
VOL. 22, 2002
KINETICS OF CIITA INDUCTION BY IFN-␥
4783
FIG. 1. CIITA mRNA is detected after 90 min of IFN-␥ induction. CIITA and GAPDH mRNAs were quantitated by real-time RT-PCR on mRNA isolated from A431 cells after treatment with IFN-␥ for the indicated times. The relative quantitation value represents the fold stimulation of CIITA mRNA determined after normalization to GAPDH mRNA. of incubation with lysates, the beads were washed in a solution containing 150 mM NaCl, 50 mM Tris (pH 8.0), and 1% NP-40 before they were boiled in SDS-polyacrylamide gel electrophoresis (PAGE) loading buffer and separated by SDS-PAGE. Western blotting was performed on either the immunoprecipitated products or on lysates prepared directly from cells. After SDS-PAGE, blots were transferred to polyvinylidene difluoride membrane as described previously (43). For IRF-1 and STAT1 detection, membranes were incubated for 1 h at room temperature with antibodies specific to these factors. For phosphotyrosine detection, membranes were incubated overnight at 4°C. Membranes were then rinsed twice with phosphate-buffered saline containing 0.1% Tween 20 (PBST) and incubated for 1 h with the appropriate horseradish peroxidase-conjugated secondary antibody (Sigma) diluted 1:3,000. Membranes were washed with PBST and incubated with enhanced chemiluminescence substrate (Amersham Life Sciences, Inc., Arlington Heights, Ill.) for 1 min and placed on film. EMSA. DNA protein interactions were analyzed by electrophoretic mobility shift assays (EMSAs) according to protocols previously described (17, 18) with binding conditions described for USF-1 (31). A DNA probe containing the GAS element and the E-box of CIITA PIV was synthesized (5⬘-GGCCAGGCAGTT GGGATGCCACTTCTGATAAAGCACGTGGTGGCCACAG). Cold competitor DNA was identical to the probe. The methylated DNA competitor was synthesized with both the coding and the noncoding underlined cytosine methylated. Antibody supershift assays used anti-USF-1 antibodies (Santa Cruz Biotechnology), which were incubated with the DNA-binding reaction mixture for 20 min prior to electrophoresis.
RESULTS Kinetics of CIITA gene expression. The induction of MHC class II genes in response to IFN-␥ proceeds with slow kinetics, typically requiring 4 to 6 h before mRNA can be detected (40; data not shown). This delay is due to the requirement to synthesize CIITA. However, there is also a delay in the expression of CIITA mRNA in response to IFN-␥ (40). To understand the nature of this delay and to begin to elucidate the mechanism of the IFN-␥-mediated regulation of MHC class II genes in vivo, a high-resolution kinetic study of the induction of CIITA mRNA by IFN-␥ was conducted. RNA isolated at 5, 15, 30, 60, 90, 120, 180, and 240 min after IFN-␥ treatment of A431 cells, a human epidermoid carcinoma cell line that is highly inducible for MHC class II genes by IFN-␥, was subjected to quantitative, real-time RT-PCR with primers specific to CIITA and GAPDH cDNA. The results showed that CIITA mRNA is essentially undetectable until ca. 90 to 120 min, when
the level just begins to rise above the background (Fig. 1). At 3 and 4 h, CIITA mRNA levels are substantially higher and significant. CIITA mRNA levels continue to increase, peaking at ca. 24 h after IFN-␥ treatment (2; data not shown). Thus, while CIITA is inducible by IFN-␥, this induction is also delayed by ⬎120 min. Several mechanisms could be responsible for the delay, including (i) the failure of the cell to process the IFN-␥ signal efficiently, (ii) the inability to assemble the transcription factor complex, (iii) the necessity to synthesize an additional gene product, or (iv) some combination of these possibilities. STAT1 binds to the CIITA promoter within 5 min of IFN-␥ stimulation. ChIP assays provide an in vivo link between a transcription factor and a DNA sequence and are ideally suited to determining the kinetics of factor binding in vivo. To address the mechanism of CIITA induction by IFN-␥ and to examine the order and kinetics of transcription factor assembly, ChIP assays were performed on PIV, the major IFN-␥responsive promoter of the CIITA gene. Antibodies to STAT1, USF-1, and IRF-1 were chosen for this analysis for two reasons. First, previous reports described the binding and assembly of these factors on PIV in vitro (31). Second, genetics studies with STAT1 and IRF-1 knockout mice showed that IFN-␥ induction of CIITA was dependent on these factors (19, 28). The ChIP assays showed that all three proteins could be found at PIV after IFN-␥ treatment; however, the fold induction and the time course of assembly were different for each factor. STAT1 binding to PIV DNA was detected within 5 min of treatment and increased to its maximum binding in 30 min (Fig. 2). This level, which decreased slightly at 1 and 4 h of treatment, is substantially above background levels. In the absence of IFN-␥ treatment, USF-1 binding was detected at ca. 10 times the no antibody control (data not shown), suggesting that USF-1 interacts with its site in the absence of stimulation. After IFN-␥ treatment, USF-1 occupancy levels increased slightly over the length of the time course (Fig. 2). In contrast, IRF-1 levels showed the most dramatic increase over time, with an induction of ⬎20-fold at 4 h. The delay in binding
4784
MORRIS ET AL.
MOL. CELL. BIOL.
FIG. 2. Kinetics of factor binding and assembly after IFN-␥ treatment. Real-time ChIP assays were performed on A431 cells before and after treatment with IFN-␥ with antibodies specific to STAT1, USF-1, and IRF-1. Values were obtained from a standard curve generated from genomic DNA. Samples were normalized to the amount of CIITA PIV chromatin added to each immunoprecipitation. The results are expressed as fold stimulation of binding over untreated cells and are the mean ⫾ the standard error of the mean of three to six experiments. A schematic displaying the CIITA PIV region and real-time primers and probes is shown. Primers used in SYBR-green detection were C127 and C318. The fluorescently labeled hydrolyzable probe with its fluorescent (star) and quenching (circle) moieties is shown by a dashed arrow, and the associated primer set for that probe is TM-1–TM-2. G, E, and I represent the GAS, E-box, and IRF sites, respectively.
of IRF-1 to PIV compared with STAT1 could be explained by the fact that STAT1 is latent in the cell at the time of activation, whereas IRF-1 activation requires new protein synthesis and is itself dependent on STAT1 activation. Indeed, the binding of IRF-1 parallels the accumulation of IRF-1 in the cell, as determined by Western blotting of whole-cell extracts (Fig. 3A). The timing of IRF-1 binding was coincident with the accumulation in CIITA mRNA transcripts, suggesting that this is a likely reason for the delay in CIITA expression. Thus, the kinetics of IFN-␥-mediated induction of CIITA through PIV occur in a stepwise fashion with STAT1 binding occurring rapidly after stimulation and IRF-1 binding after its synthesis. IRF-1 expression is not sufficient for CIITA expression. Previous work with STAT1 knockout animals demonstrated that STAT1 was required for CIITA expression (28). It is also known that STAT1 is required for IFN induction of IRF-1 expression (28, 42). Thus, it is formally possible that only IRF-1 expression is required for CIITA expression. Studies by the Ting group showed that a transfected IRF-1 gene could partially restore the activity of a transiently transfected PIV reporter construction (35). If this is the case, then the transfection of an IRF-1 expression plasmid into A431 cells should result in the expression of the endogenous CIITA in the absence of IFN-␥. To determine if this could be the case, A431 cells were transfected with empty vector DNA or an IRF-1 expression plasmid. At 48 h posttransfection, IRF-1 levels could be detected in IRF-1 plasmid-transfected cells at a substantially higher level than in untreated cells (Fig. 4A). This level was slightly less than the equivalent number of cells treated with IFN-␥ for 4 h. However, analysis of the transfected cells by flow cytometry for MHC class II expression demonstrated that IRF-1 transfection was not sufficient for MHC class II induction and that IFN-␥ treatment is required
(Fig. 4B). Real-time RT-PCR analysis for CIITA mRNA conducted on the IRF-1-transfected cells showed no induction of endogenous CIITA gene in the absence of IFN-␥ (data not shown). These data combined with the STAT1 knockout mouse studies (28), as well as experiments showing that mutations in the GAS element prevent CIITA induction by IFN-␥
FIG. 3. IFN-␥ mediates IRF-1 expression and STAT1 phosphorylation in A431 and JEG-1 cells. (A) Western analysis with IRF-1 antibodies was performed on whole-cell lysates from A431 and JAR cells treated with IFN-␥. Equal amounts of protein were loaded into each lane. The position of the IRF-1-specific band is indicated. (B) STAT1 is phosphorylated in response to IFN-␥ in JEG-3 cells. STAT1 was immunoprecipitated from control and IFN-␥-treated A431 and JEG-3 cells. The immunoprecipitated protein was divided and resolved by SDS-PAGE and then assayed by Western blotting for the presence of phosphotyrosine residues and the presence of STAT1.
VOL. 22, 2002
KINETICS OF CIITA INDUCTION BY IFN-␥
4785
FIG. 4. Transfected IRF-1 is not sufficient to induce MHC class II expression. (A) A431 cells transfected with pCOIRF-1 express high levels of IRF-1. Western blots performed on A431 cells, on A431 cells treated with IFN-␥ for 4 h, and on A431 transiently transfected with pCOIRF-1 were stained with anti-IRF-1 antibodies. The position of the IRF-1 band is indicated. (B) A431 cells were transfected with pCOIRF-1 or pUC18 DNA. After 12 h, cells were treated or not treated with IFN-␥ for an additional 56 h and assayed for expression of HLA-DR surface expression by flow cytometry with phycoerythrin-conjugated HLA-DR-specific antibodies. Open and closed histograms represent samples that were untreated or treated with IFN-␥, respectively.
(31), argue that STAT1, which is induced by IFN-␥, is required for expression of the endogenous CIITA gene. Histone acetylation occurs in a time course that parallels STAT1 binding to PIV. Acetylation of histones on nucleosomes close to the promoters of several genes, including MHC class II genes, has been associated with increases in mRNA levels (2, 5, 15, 23, 41, 45, 48). The kinetics described above for PIV pose the question of whether there are changes in histone acetylation and whether these changes are associated with STAT1 or IRF-1 binding since these two events occur at distinct times during the activation process. Thus, if acetylation occurred only after IRF-1 binding, this would suggest that full assembly of the promoter was required for changes in acetylation. Alternatively, if acetylation occurred prior to IRF-1 binding, this may suggest that acetylation of the promoter may play a role in the opening of the local chromatin structure to allow efficient IRF-1 binding. To distinguish between these possibilities and to characterize the changes in acetylation that may occur during the IFN-␥-mediated activation of PIV, ChIP assays with antibodies to acetylated histones H3 and H4 were performed (Fig. 5). In response to IFN-␥, a modest increase in histone H3 acetylation was observed within 15 min of IFN-␥ treatment. Acetylation declined slightly by 4 h but remained above the levels of unstimulated cells. A similar increase in acetylation of histone H4 occurred within the same time frame. Parallel analysis of the GAPDH gene showed virtually no change in acetylation of either histone H3 or H4 (Fig. 5). These results demonstrate that IFN-␥ treatment induces an increase in histone acetylation at PIV, and this occurs prior to the binding of IRF-1 and the accumulation of CIITA transcripts. CpG methylation extends along the length of PIV in trophoblast cell lines. Fetal trophoblasts are refractory to IFN-␥ induction of MHC class II genes (1, 30, 33). Previous analysis showed that the defect in the induction pathway was in the
failure to induce CIITA expression (30, 33, 46). Moreover, the repression of CIITA was linked to hypermethylation of DNA sequences in PIV. Hypermethylation of CIITA PIV could prevent transcription by inhibiting factor binding at a specific
FIG. 5. Histones H3 and H4 at PIV are acetylated in A431 cells in response to IFN-␥. Real-time ChIP assays conducted to examine the changes in histones H3 and H4 were conducted on A431 cells before and after IFN-␥ treatment. The samples are expressed as the fold induction over the control, nontreated cells as described for Fig. 2.
4786
MORRIS ET AL.
MOL. CELL. BIOL.
FIG. 6. CpG dinucleotides in CIITA PIV are methylated in trophoblast-derived cells and unmethylated in CIITA-inducible A431 cells. Bisulfite sequencing of the nontemplate (A) and template (B) strands of DNA from A431 cells and the trophoblast-derived cell lines JAR and JEG-3 was performed. Each CpG was assigned a number from 5⬘ to 3⬘ of PIV, as indicated in the schematics for each strand. At least 20 alleles were examined in each case. Each sequenced allele is represented by a row of boxes representing each CpG dinucleotide. Methylated CpGs are indicated by black boxes; unmethylated CpGs are indicated by white boxes.
sequence or by causing changes in chromatin that make the DNA less accessible to transcription factors. Sixteen potential targets of cytosine methylation are located within the 355-bp region of PIV (Fig. 6). Of the three factor-binding sites in CIITA PIV, only the USF-1 site contains a CpG dinucleotide. Therefore, to determine whether CpG methylation of the USF-1 site could have a consequence on regulation, the ability of USF-1 to bind to CpG methylated DNA was assessed in vitro. Nuclear extracts prepared from JEG-3, JAR, and A431 cells were incubated with a labeled PIV DNA probe, and the
protein-DNA complexes were analyzed by EMSA (Fig. 7). A single specific complex competed by unlabeled nonmethylated PIV DNA was observed with all three extracts. The complex contains USF-1 as indicated by the anti-USF-1 antibody supershifted bands (Fig. 6, lanes 5, 9, and 13). From this analysis, it appears as though the levels of USF-1 in JAR and A431 cells are similar, whereas JEG-3 cells contain less. When PIV DNA containing a methylated CpG was used as a competitor, no substantial competition was observed, indicating that USF-1 binding to methyl CpG DNA is an unfavorable event.
VOL. 22, 2002
KINETICS OF CIITA INDUCTION BY IFN-␥
4787
FIG. 7. USF-1 does not bind to a CpG methylated E-box. An EMSA was performed with a labeled PIV CIITA-specific probe containing an unmethylated E-box with nuclear extracts prepared from A431, JAR, and JEG-3 cells. Control reactions (⫺) and DNA competitions with cold DNA either identical to the probe (lanes C) or synthesized with a methylated CpG at position 4 (lanes M) were carried out. Antibody supershift experiments in which anti-IRF-1 antibodies were added to the DNA-binding reaction were also performed and are indicated by an asterisk above the appropriate lanes. The USF-1 specific and supershifted bands are indicated by arrowheads.
The above results suggest that methylation of only this CpG may be sufficient to block expression of CIITA and that this could be the mechanism by which these cell types repress CIITA and MHC class II expression. Therefore, to determine whether the USF-1 CpG was differentially methylated in CIITA-expressing cells and in trophoblast cells and to examine the methylation status of the other CpG dinucleotides in PIV, DNA sequencing of bisulfite modified genomic DNA clones was performed. Because CpG methylation can vary between individual cells within the culture, a collection of “alleles” was analyzed from each cell type studied. Genomic DNA isolated from two CIITA uninducible trophoblast cell lines (JAR and JEG-3), as well as A431 cells, was treated with sodium bisulfite, which deaminates all non-5methylcytosines and converts them to uracils. Bisulfite-treated DNA was amplified with primers specific for PIV. The PCR products from two independent assays were cloned, and ⬎20 random clones for each cell line were sequenced. Each clone represents a PCR product derived from an allele of the CIITA PIV region. The results showed that methylation was nearly absent in the CIITA-inducible cell line A431 (Fig. 6). In contrast, methylation was extensive in both the JAR and JEG-3 cell lines (Fig. 6). On the nontemplate strand (Fig. 6A), JAR cells showed prevalent methylation at CpG positions 1, 2, 3, and 7 to 12, with lower levels of CpG methylation at positions 5 and 6. In general, JEG-3 cells exhibited a lower degree of cytosine methylation, occurring mostly at positions 1, 2, 9, 11, and 12. A similar pattern of methylation was observed for the template strand (Fig. 6B) in both JAR and JEG-3 cells, with the 3⬘ portion of DNA containing a higher degree of methylation. Interestingly, methylation at CpG-4, which is located within the USF-1 binding site, was detected in only 3 of the 24 JAR alleles sequenced and in none of the JEG-3 alleles sequenced, suggesting that transcriptional repression of CIITA PIV in this cell type is unlikely to be caused directly by methylation of a transcription factor recognition sequence. The CpGs between the E-box and surrounding the IRF-1 site also showed reduced levels of methylation in both cell types. It is of note that in JEG-3 and JAR cells a small number of alleles were unmethylated, suggesting the possibility that methylation
in this region is dynamic or that the alleles were captured during replication and may represent a partial or hemimethylated state. Factor binding and histone acetylation do not occur in trophoblast cell lines in response to IFN-␥. The results presented above suggest that the methylation pattern may influence factor binding and nucleosome acetylation. This suggestion is supported by previous in vivo genomic footprinting (IVGF) analysis of PIV in the same cell lines (30), which showed an absence of factor binding in response to IFN-␥. ChIP assays are more sensitive than the IVGF owing to the fact that in order to observe a footprint in the IVGF assay, a majority of the sites within a given gene must be occupied so that the background banding pattern is obliterated. Thus, to confirm the previous result and to examine the dynamics of histone acetylation at PIV in these cells, ChIP assays were performed in JAR cells before and after IFN-␥ treatment. STAT1, USF-1, and IRF-1 binding were not detected over the time course (Fig. 8). In addition, no significant change in histone acetylation was observed for both histones H3 and H4 (Fig. 8). Furthermore, the absolute amounts of PIV DNA immunoprecipitated with each antibody, quantitated by real-time PCR, were ⬍2-fold above those observed in the no antibody control samples (data not shown). This was not due to a failure to induce the synthesis of IRF-1 protein as IRF-1 is induced in these cells after IFN-␥ treatment (Fig. 3A). The induction of IRF-1 in the trophoblast cell lines suggests that STAT1 is present and activated by IFN-␥ treatment. To show that STAT1 becomes activated, experiments were carried out to determine if STAT1 became phosphorylated after IFN-␥ treatment. Lysates prepared from control and IFN-␥treated A431 and JEG-3 cells were subjected to STAT1 immunoprecipitation and Western blot analysis of the products. The results showed that the levels of STAT1 are similar between the cell lines and that STAT1 gets phosphorylated in JEG-3 cells to a level that is similar to that in A431 cells (Fig. 3B). Taken together, these results support the model that methylation of PIV induces a repressive chromatin environment, which prevents the access of transcription factors to the DNA.
4788
MORRIS ET AL.
MOL. CELL. BIOL.
FIG. 8. Factor assembly and histone acetylation of PIV did not occur in JAR cells treated with IFN-␥. Real-time ChIP assays were performed before and after treatment with IFN-␥ as described in above. No changes in factor assembly or histone acetylation were observed. Data are presented as described in Fig. 2.
DISCUSSION This study describes the kinetics of IFN-␥ induction of the master regulator of MHC class II gene expression, CIITA (Fig. 9). The results demonstrate a stepwise assembly of factors at CIITA’s IFN-␥ responsive promoter and suggest that cytosine methylation of PIV can block assembly and chromatin remodeling during IFN-␥ induction. An order of events can be assigned for the transcriptional activation of the CIITA gene by PIV in response to IFN-␥. After IFN-␥ stimulation, STAT1 is activated, translocates to the nucleus, and binds to the PIV GAS element. STAT1 also stimulates the transcription and expression of the IRF-1 gene. Concurrent with this binding is the binding and/or stabilization of USF-1 at the E-box. Second, acetylation of histones H3 and H4 is increased. Third, IRF-1 protein accumulates and binds to its site in PIV. Lastly, transcription of the CIITA gene initiated at PIV occurs. From this kinetic study insight into the combinatorial assembly of transcription factors and how this assembly relates to gene expression can be inferred. It has been demonstrated previously that STAT1 activation, translocation, and binding to its GAS sequence is a rapid event. The ability to detect these events within 5 min of activation in vivo was made possible by use of the ChIP assay. In the absence of other regulatory controls, it is likely that STAT1 binds to other GAS elements with at least an equal efficiency and speed. Because in vivo genomic footprints of PIV display enhanced binding to the E-box after treatment with IFN-␥, it was anticipated that USF-1 binding would increase dramatically (30, 35). This expectation was further supported by cooperative binding of
STAT1 and USF-1 to PIV DNA in vitro (31). The present data showed only a small increase in USF-1 binding after exposure to IFN-␥. This could be due to several reasons. The first is that, prior to IFN-␥ treatment, USF-1 may be binding to the E-box transiently. Thus, only the stabilization of binding by STAT1 is detected after IFN-␥ treatment. This conclusion is supported by the fact that the signal detected by the ChIP assay was substantially above those seen in the no-antibody and irrelevant-antibody controls, suggesting that the factor can be detected at the promoter prior to treatment. A second reason could be that the binding of the antibody to the USF-1 chromatin complex is inhibited as additional proteins, including STAT1 and IRF-1, bind. A third reason could be that USF-1 is one of several members of a family of proteins that can interact with this sequence and therefore, while USF-1 is clearly detected by this assay, other factors may also bind to the site in vivo as well. The role of IRF-1 in this system is intriguing for several reasons. IRF-1 is clearly required for expression, since animals containing a targeted disruption of the IRF-1 gene have defects in their ability to induce CIITA (19). Moreover, pretreatment of cells with cycloheximide results in a significant reduction in CIITA expression, suggesting that a factor must be synthesized for full expression (19; data not shown). The delay in CIITA expression correlates with the requirement to induce the expression of IRF-1. It is likely that a critical threshold of IRF-1 protein must be produced to activate expression, since the cells used in this study exhibited a low background of constitutive IRF-1 protein (Fig. 3). IRF-1 protein levels were
VOL. 22, 2002
FIG. 9. Model of assembly and chromatin remodeling of PIV of the CIITA gene after IFN-␥ induction. In untreated cells, STAT-1 and USF-1 are found in the cytoplasm and nucleus, respectively. Within 60 min of IFN-␥ treatment, STAT-1 is phosphorylated, moves to the nucleus, and binds to CIITA PIV. USF-1 binding increases over this time period, and histone H3 and H4 acetylation increases (represented by increased shading of the schematic nucleosome). During this time period, IRF-1 protein begins to be synthesized and accumulates at CIITA PIV. By 4 h, the levels of IRF-1 protein at CIITA PIV reach a threshold concentration that, in concert with the other factors, promotes transcription. When PIV is methylated, as in the case of trophoblasts, transcription factor access and nucleosome acetylation are prevented.
KINETICS OF CIITA INDUCTION BY IFN-␥
4789
substantially increased at 60 min, and this correlated with its binding to PIV. At 120 min, low but significant levels of CIITA mRNA could be detected. Thus, IRF-1 is necessary but not sufficient for CIITA gene expression, since the constitutive expression of IRF-1 in A431 cells through the use of a transient transfection assay was not sufficient to activate the CIITA gene. Since histone acetylation occurs at least an hour before transcripts are detected, it is likely that it is not sufficient to activate the CIITA gene. This is in contrast to MHC class II genes which were recently shown to be responsive to the histone deacetylase inhibitor trichostatin A (TSA) in a cell line deficient in CIITA (26). Thus, it is important to consider the role of the acetylation in activating PIV transcription. Acetylation is thought to open or relax the chromatin structure in such a way as to allow factor binding and transcriptional activation. We propose here that acetylation is functioning to allow the full assembly of factors and the stable association of these factors. This proposal is based on the timing of histone acetylation and IRF-1 binding, as well as the stability of the in vivo footprint that one sees after IFN-␥ stimulation. The overall level and change in acetylation of PIV are low compared to other genes that we have studied with similar antibodies and techniques, including the HLA-DRA gene (2) and the MCP-1 and MnSOD genes (unpublished results). If acetylation correlates with gene activity, then the changes observed for CIITA reflect the overall level of mRNA that is produced after IFN-␥ treatment, which is considered to be low, requiring the use of RT-PCR to detect. It is also possible that for poorly expressed genes the level of acetylation is not critical as long as the regulatory regions are accessible for the binding of transcription factors. PIV was found to be differentially methylated in CIITAexpressing cells and in trophoblast-derived cells (30, 46). The pattern of methylation was not due to the fact that these cells were derived from tumors since other tumor-derived cell lines did not show methylation of the CIITA promoter, suggesting that the patterns in JAR and JEG-3 represent the developmental state of the cell (47). In addition to the trophoblastderived cell lines, freshly isolated fetal trophoblasts also displayed methylation at PIV (46). However, because those experiments were done by methylation-sensitive restriction analysis, only a limited number of CpG sites could be examined. Through the use of bisulfite sequencing the methylation status of each CpG dinucleotide within PIV was determined. The pattern that emerged was complex. Although virtually no methylation was detected in the 40 alleles examined from A431 cells, the extent of methylation in the trophoblast-derived cell lines was high and varied along the length of PIV. Unexpectedly, the E-box site showed the lowest level of methylation. This suggests that the reason that PIV is repressed in the trophoblast cell lines is not due directly to a single methylcytosine, which could potentially prevent the direct binding of USF-1, but rather to the state of the chromatin packaging PIV in these cells. In contrast to the A431 cells, where in the absence of IFN-␥ USF-1 binding occurred at 10 times the background level, no binding was detected in JAR cells, which as determined by EMSA were found to contain similar levels of USF-1 protein as the A431 cells. Additionally, H3 and H4 acetylation levels were extremely low at PIV in these cells, and
4790
MORRIS ET AL.
no significant increase in acetylation was detected in response to IFN-␥, indicating the repressed state of this promoter. The link between methylation, histone hypoacetylation, and gene silencing was discovered through the identification of methyl-CpG binding proteins such as MeCP2 (25). It has been proposed that binding of MeCP2 to methyl-CpGs results in the recruitment of factors or complexes with histone deacetylase activity. This interaction has been demonstrated in Xenopus for MeCP2 and the transcriptional corepressor Sin3A (20). This hypothesis is also supported by work in Neurospora strains, in which reactivation of a methylated copy of the hygromycin resistance gene was achieved by exposure to TSA, a histone deacetylase inhibitor (38). To silence CIITA gene induction, the present study suggests that both methylation and hypoacetylation mechanisms are at play. However, methylation appears to be dominant to hypoacetylation because previous reports showed that CIITA gene expression could be restored after treatment of trophoblast cells with the cytosine methylation inhibitor 5AC (30). Interestingly, treatment of trophoblasts with TSA was ineffective in restoring CIITA expression, and TSA treatment did not enhance the effect of 5AC (data not shown). These results are similar to those seen with the methylated fragile X mental retardation gene, FMR1 (8). FMR1 can be reactivated with 5AC but not with TSA, which results in acetylation of histones around the FMR1 promoter. Thus, although methylation and histone hypoacetylation are linked, the dominance of one epigenetic tag over the other may vary depending on the gene and the modifications required for its expression or repression.
MOL. CELL. BIOL.
11. 12.
13.
14. 15. 16.
17. 18.
19. 20. 21. 22.
ACKNOWLEDGMENTS We acknowledge U. Nagarajan for help with flow cytometry, as well as members of the laboratory and M. Brown, P. Wade, and D. Reines for discussions about the results. This work was supported by NIH grants AI34000 and HD34440. M. Mooney is supported by NIH fellowship GM20675. A.C.M. and G.W.B. contributed equally to this study. REFERENCES 1. Athanassakis-Vassiliadis, I., D. Thanos, and J. Papamatheakis. 1989. Induction of class II major histocompatibility complex antigens in murine placenta by 5⬘-azacytidine and interferon-gamma involves different cell populations. Eur. J. Immunol. 19:2341–2348. 2. Beresford, G. W., and J. M. Boss. 2001. CIITA coordinates multiple histone acetylation modifications at the HLA-DRA promoter. Nat. Immunol. 2:652– 657. 3. Boss, J. M. 1999. A common set of factors control the expression of the MHC class II, invariant chain, and HLA-DM genes. Microbes Infect. 1:847– 853. 4. Boss, J. M., and J. L. Strominger. 1986. Regulation of a transfected human class II major histocompatibility complex gene in human fibroblasts. Proc. Natl. Acad. Sci. USA 83:9139–9143. 5. Chakravarti, D., V. Ogryzko, H. Y. Kao, A. Nash, H. Chen, Y. Nakatani, and R. M. Evans. 1999. A viral mechanism for inhibition of p300 and PCAF acetyltransferase activity. Cell 96:393–403. 6. Chang, C.-H., J. D. Fontes, B. M. Peterlin, and R. A. Flavell. 1994. Class II transactivator (CIITA) is sufficient for the inducible expression of major histocompatibility complex class II genes. J. Exp. Med. 180:1367–1374. 7. Clark, S. J., J. Harrison, C. L. Paul, and M. Frommer. 1994. High sensitivity mapping of methylated cytosines. Nucleic Acids Res. 22:2990–2997. 8. Coffee, B., F. Zhang, S. T. Warren, and D. Reines. 1999. Acetylated histones are associated with FMR1 in normal but not fragile X-syndrome cells. Nat. Genet. 22:98–101. 9. Collins, T., A. J. Korman, C. T. Wake, J. M. Boss, D. J. Kappes, W. Fiers, K. A. Ault, M. A. Gimbrone, Jr., J. L. Strominger, and J. S. Pober. 1984. Immune interferon activates multiple class II major histocompatibility complex genes and the associated invariant chain gene in human endothelial cells and dermal fibroblasts. Proc. Natl. Acad. Sci. USA 81:4917–4921. 10. DeSandro, A., U. M. Nagarajan, and J. M. Boss. 2000. Associations and
23. 24. 25.
26.
27.
28.
29. 30. 31.
32.
33.
interactions among the bare lymphocyte syndrome factors. Mol. Cell. Biol. 20:6587–6599. Fontes, J. D., B. Jiang, and B. M. Peterlin. 1997. The class II trans-activator CIITA interacts with the TBP-associated factor TAFII32. Nucleic Acids Res. 25:2522–2528. Fontes, J. D., S. Kanazawa, D. Jean, and B. M. Peterlin. 1999. Interactions between the class II transactivator and CREB binding protein increase transcription of major histocompatibility complex class II genes. Mol. Cell. Biol. 19:941–947. Glimcher, L. H., and C. J. Kara. 1992. Sequences and factors: a guide to MHC class II transcription, p. 13–50. In W. E. Paul, C. Garrison-Fathman, and H. Metzger (ed.), Annual reviews in immunology. Annual Reviews, Inc., Palo Alto, Calif. Gourley, T. S., and C. H. Chang. 2001. Cutting edge: the class II transactivator prevents activation-induced cell death by inhibiting Fas ligand gene expression. J. Immunol. 166:2917–2921. Grant, P. A., D. E. Sterner, L. J. Duggan, J. L. Workman, and S. L. Berger. 1998. The SAGA unfolds: convergence of transcription regulators in chromatin-modifying complexes. Trends Cell Biol. 8:193–197. Hamilton, T. A., N. Bredon, Y. Ohmori, and C. S. Tannenbaum. 1989. IFN-␥ and IFN- independently stimulate the expression of lipopolysaccharideinducible genes in murine peritoneal macrophages. J. Immunol. 142:2325– 2331. Hasegawa, S. L., and J. M. Boss. 1991. Two distinct B cell factors bind the HLA-DRA X box region and recognize different subsets of MHC class II promoters. Nucleic Acids Res. 19:6269–6276. Hasegawa, S. L., J. L. Riley, J. H. Sloan, and J. M. Boss. 1993. Protease treatment of nuclear extracts distinguishes between class II major histocompatibility complex X1 box DNA-binding proteins in wild type and class II deficient B cells. J. Immunol. 150:1781–1793. Hobart, M., V. Ramassar, N. Goes, J. Urmson, and P. F. Halloran. 1997. IFN regulatory factor-1 plays a central role in the regulation of the expression of class I and II MHC genes in vivo. J. Immunol. 158:4260–4269. Jones, P. L., G. J. Veenstra, P. A. Wade, D. Vermaak, S. U. Kass, N. Landsberger, and A. P. Wolffe. 1998. Methylated DNA and MeCP2 recruit histone deacetylase to repress transcription. Nat. Genet. 19:187–191. Kanazawa, S., T. Okamoto, and B. M. Peterlin. 2000. Tat competes with CIITA for the binding to P-TEFb and blocks the expression of MHC class II genes in HIV infection. Immunity 12:61–70. Kretsovali, A., T. Agalioti, C. Spilianakis, E. Tzortzakaki, and M. Merika. 1998. Involvement of CREB binding protein in expression of major histocompatibility complex class II genes via interaction with the class II transactivator. Mol. Cell. Biol. 18:6777–6783. Kuo, M.-H., J. Zhou, P. Jambeck, M. E. A. Churchill, and C. D. Allis. 1998. Histone acetyltransferase activity of yeast Gcn5p is required for the activation of target genes in vivo. Genes Dev. 12:627–639. Lee, J. Y., and K. E. Sullivan. 2001. Gamma interferon and lipopolysaccharide interact at the level of transcription to induce tumor necrosis factor alpha expression. Infect. Immun. 69:2847–2852. Lewis, J. D., R. R Meehan, W. J. Henzel, I. Maurer-Fogy, P. Jeppesen, F. Klein, and A. Bird. 1992. Purification, sequence, and cellular localization of a novel chromosonal protein that binds to methylated DNA. Cell 69:905– 914. Magner, W. J., A. L. Kazim, C. Stewart, M. A. Romano, G. Catalano, C. Grande, F. Santaniello, and T. B. Tomasi. 2001. Activation of MHC class I, II, and CD40 gene expression by histone deacetylase inhibitors. J. Immunol. 165:7017–7024. Mahanta, S. K., T. Scholl, F.-C. Yang, and J. L. Strominger. 1997. Transactivation by CIITA, the type II bare lymphocyte syndrome-associated factor, requires participation of multiple regions of the TATA box binding protein. Proc. Natl. Acad. Sci. USA 94:6324–6329. Meraz, M. A., J. M. White, K. C. F. Sheehan, E. A. Bach, S. J. Rodig, A. S. Dighe, D. H. Kaplan, J. K. Riley, A. C. Greenlund, D. Campbell, K. CarverMoore, R. N. DuBois, R. Clark, M. Aguet, and R. D. Schreiber. 1996. Targeted disruption of the Stat1 gene in mice reveals unexpected physiologic specificity in the JAK-STAT signaling pathway. Cell 84:431. Morris, A. C., J. L. Riley, W. H. Fleming, and J. M. Boss. 1998. MHC class II gene silencing in trophoblast cells is caused by inhibition of CIITA expression. Am. J. Reprod. Immunol. 40:385–394. Morris, A. C., W. E. Spangler, and J. M. Boss. 2000. Methylation of class II trans-activator promoter IV: a novel mechansim of MHC class II gene control. J. Immunol. 164:4143–4149. Muhlethaler-Mottet, A., W. Di Berardino, L. A. Otten, and B. Mach. 1998. Activation of the MHC class II transactivator CIITA by interferon gamma requires cooperative interaction between Stat1 and USF-1. Immunity 8:157– 166. Muhlethaler-Mottet, A., L. A. Otten, V. Steimle, and B. Mach. 1997. Expression of MHC class II molecules in different cellular and functional compartments is controlled by differential usage of multiple promoters of the transactivator CIITA. EMBO J. 16:2851–2860. Murphy, S. P., and T. B. Tomasi. 1998. Absence of MHC class II antigen
VOL. 22, 2002
34.
35.
36. 37. 38. 39. 40. 41. 42.
expression in trophoblast cells results from a lack of class II transactivator (CIITA) gene expression. Mol. Reprod. Dev. 51:1–12. Peyman, J. A., P. J. Nelson, and G. L. Hammond. 1992. HLA-DR genes are silenced in human trophoblasts and stimulation of signal transduction pathways does not circumvent IFN-␥ unresponsiveness. Transplant. Proc. 24: 470–471. Piskurich, J. F., M. W. Linhoff, Y. Wang, and J. P.-T. Ting. 1999. Two distinct gamma interferon-inducible promoters of the major histocompatibility complex class II transactivator gene are differentially regulated by STAT1, interferon regulatory factor 1, and transforming growth factor . Mol. Cell. Biol. 19:431–440. Reith, W., and B. Mach. 2001. The bare lymphocyte syndrome and the regulation of MHC expression. Annu. Rev. Immunol. 19:331–373. Riley, J. L., and J. M. Boss. 1993. Class II MHC transcriptional mutants are defective in higher order complex formation. J. Immunol. 151:6942–6953. Selker, E. U. 1998. Trichostatin A causes selective loss of DNA methylation in Neurospora. Proc. Natl. Acad. Sci. USA 95:9430–9435. Spilianakis, C., J. Papamatheakis, and A. Kretsovali. 2000. Acetylation by PCAF enhances CIITA nuclear accumulation and transactivation of major histocompatibility complex class II genes. Mol. Cell. Biol. 20:8489–8498. Steimle, V., C.-A. Siegrist, A. Mottet, B. Lisowska-Grospierre, and B. Mach. 1994. Regulation of MHC class II expression by interferon-gamma mediated by the transactivator gene CIITA. Science 265:106–108. Sterner, D. E., and S. L. Berger. 2000. Acetylation of histones and transcription-related factors. Microbiol. Mol. Biol. Rev. 64:435–459. Taniguchi, T., K. Ogasawara, A. Takaoka, and N. Tanaka. 2001. IRF family of transcription factors as regulators of host defense. Annu. Rev. Immunol. 19:623–655.
KINETICS OF CIITA INDUCTION BY IFN-␥
4791
43. Towbin, H., T. Staehelin, and J. Gordon. 1979. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc. Natl. Acad. Sci. USA 76:4350–4354. 44. Unanue, E. R., C. V. Harding, I. F. Luescher, and R. W. Roof. 1989. Antigenbinding function of class II MHC molecules. Cold Spring Harbor Symp. Quant. Biol. 54:383–392. 45. Utley, R. T., K. Ikeda, P. A. Grant, J. Cote, D. J. Steger, A. Eberharter, S. John, and J. L. Workman. 1998. Transcriptional activators direct histone acetyltransferase complexes to nucleosomes. Nature 394:498–502. 46. van den Elsen, P. J., N. van der Stoep, H. E. Vietor, L. Wilson, M. van Zutphen, and S. J. P. Gobin. 2000. Lack of CIITA expression is central to the absence of antigen presentation functions of trophoblast cells and is caused by methylation of the IFN-␥-inducible promoter (PIV) of CIITA. Hum. Immunol. 61:850–862. 47. van der Stoep, N., P. Biesta, E. Quinten, and P. J. van den Elsen. 2002. Lack of IFN-␥-mediated induction of the class II transactivator (CIITA) through promoter methylation is predominantly found in developmental tumor cell lines. Int. J. Cancer 97:501–507. 48. Van Lint, C., S. Emiliani, M. Ott, and E. Verdin. 1996. Transcriptional activation and chromatin remodeling of the HIV-1 promoter in response to histone acetylation. EMBO J. 15:1112–1120. 49. Wadburger, J. M., T. Suter, A. Fontana, H. Acha-Orbea, and W. Reith. 2001. Selective abrogation of major histocompatibility complex class II expression on extrahematopoietic cells in mice lacking promoter IV of the class II transactivator gene. J. Exp. Med. 19:393–406. 50. Zuckermann, F. A., and J. R. Head. 1986. Expression of MHC antigens on murine trophoblast and their modulation by interferon. J. Immunol. 137: 846–853.
