HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 277, Issue 29, 26460-26467, July 19, 2002

This Article Abstract Full Text (PDF) All Versions of this Article: 277/29/26460    most recent M110538200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Grimm, M. Articles by Liao, J. K. Search for Related Content PubMed PubMed Citation Articles by Grimm, M. Articles by Liao, J. K. Social Bookmarking                      What's this?

Inhibition of Major Histocompatibility Complex Class II Gene Transcription by Nitric Oxide and Antioxidants^*

Michael Grimm, Martin Spiecker^§, Raffaele De Caterina^¶, Wee Soo Shin, and James K. Liao^**

From the Vascular Medicine Unit, Brigham & Women's Hospital and Harvard Medical School, Boston, Masachusetts 02115

Received for publication, November 2, 2001, and in revised form, April 22, 2002

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Interferon (IFN)- facilitates cellular immune response, in part, by inducing the expression of major histocompatibility complex class II (MHC-II) molecules. We demonstrate that IFN- induces the expression of HLA-DRA in vascular endothelial cells via mechanisms involving reactive oxygen species. IFN--induced HLA-DRA expression was inhibited by nitric oxide (NO) and antioxidants such as superoxide dismutase, catalase, pyrrolidine dithiocarbamate, and N-acetylcysteine. Nuclear run-on assays demonstrated that NO and antioxidants inhibited IFN--induced HLA-DRA gene transcription. Transient transfection studies using a fully functional HLA-DRA promoter construct ([300]DR.CAT) showed that inhibition of endogenous NO synthase activity by N-monomethyl-L-arginine or addition of exogenous hydrogen peroxide (H₂O₂) augmented basal and IFN--stimulated [300]DR.CAT activity. However, H₂O₂ and N-monomethyl-L-arginine could induce HLA-DRA expression suggesting that H₂O₂ is a necessary but not a sufficient mediator of IFN--induced HLA-DRA expression. Electrophoretic mobility shift assay and Western blotting demonstrated that NO and antioxidants had little or no effect on IFN--induced IRF-1 activation or MHC-II transactivator (CIITA) expression but did inhibit IFN--induced activation of STAT1 (p91) and Y box transcription factors, NF-Y_A and NF-Y_B. These results indicate that NO and antioxidants may attenuate vascular inflammation by antagonizing the effects of intracellular reactive oxygen species generation by IFN-, which is necessary for MHC-II gene transcription.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Structures bearing major histocompatibility complex (MHC)¹-related molecules play an important role in cellular immunity, recognition, and differentiation (1, 2). The MHC class II (MHC-II) genes, composed of HLA-DR, -DQ, -DP, and -DM, encode  and  chains of heterodimeric cell surface molecules that present processed antigens to CD4⁺ T-lymphocytes. In contrast to MHC class I (MHC-I) molecules that are expressed on virtually all cell surfaces, constitutive MHC-II expression is restricted to only a few cell types, classically B-lymphocytes, thymic epithelial cells, dendritic cells, and macrophages. However, other cell types such as type 1 astrocytes, vascular endothelial and smooth muscle cells, and fibroblasts can express MHC-II in response to interferon (IFN)-, interleukin (IL)-4, and IL-10 (3-5). Indeed, the induction of MHC-II on endothelial cells was the first reported example of "endothelial activation" (4), defined as the appearance of novel gene products on the endothelial cell surface, which allow endothelial cells to perform new functions (6). Endothelial MHC-II expression was found on the endothelium in atherosclerotic lesions obtained from patients who died of cardiovascular or neurologic diseases (7), and the induction of MHC-II on vascular wall cells, in part, mediates the cellular immune response associated with transplantation arteriosclerosis (8). Moreover, a recent clinical study found increased expression of MHC-II molecules in arterial tissue from transplanted hearts to be predictive of arteriosclerosis and graft failure (9).

Besides expressing MHC-II, cytokine-activated endothelial cells express various other cell surface adhesion molecules such as vascular and intercellular adhesion molecules, VCAM-1 and ICAM-1, which allow binding of mononuclear cells to the vessel wall. The signaling pathways leading to the induction of VCAM-1 and ICAM-1 expression involves the activation of oxidant-sensitive proinflammatory transcription factors, nuclear factor-B (NF-B), and activated protein-1 (10, 11). Indeed, antioxidants such as N-acetylcysteine (NAC) and pyrrolidine dithiocarbamate (PDTC) have been shown to inhibit NF-B activation and cytokine-induced endothelial cell activation (10). Furthermore, we have shown recently (12, 13) that another endogenous ROS, nitric oxide (NO), can attenuate cytokine-induced VCAM-1 and macrophage-colony-stimulating factor gene transcription via stabilization and induction of the NF-B inhibitor, IB. However, it is not known whether the induction of MHC-II is under similar redox control since its minimal promoter (136 bp to +31 bp), which is required for full response to IFN-, does not contain B or activated protein-1 cis-acting elements (14).

The upstream regions of all MHC-II genes contain conserved DNA elements, called W, X, and Y, which are essential for full transcriptional activation (2, 15). Contained in the Y box is a CCAAT motif that interacts with the constitutively expressed heterotrimeric transcription factor NF-Y. The DNA binding activity of NF-Y resides in the specific association of three nonidentical A, B, and C subunits (16, 17). The CCAAT motif resembles the oxyR-response element of prokaryotic cells, which is known to mediate redox-dependent transcriptional activation of bacterial genes coding for peroxide-inactivating enzymes such as catalase (katG), NADPH-dependent alkylhydroperoxidase (aphFC), and others (18, 19). Previous studies have indicated that the oxyR-response element can also function as a redox-dependent transcriptional activator in mammalian cells (20) and that NF-Y itself is regulated by cellular redox (21).

Although it is known that stimulation of endothelial cells with IFN- induces the generation of superoxide anion (O), it is not known whether the production of ROS is necessary and/or sufficient for MHC-II gene transcription (22). The purpose of this study, therefore, was to determine the role of ROS in mediating MHC-II gene transcription in vascular endothelial cells.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Reagents-- All standard laboratory and culture reagents, unless otherwise indicated, were obtained from Sigma and JRH Biosciences (Lenexa, KS), respectively. [-³²P]CTP (3000 Ci/mmol), [-³²P]ATP (3000 Ci/mmol), and [-³²P]UTP (800 Ci/mmol) were supplied by PerkinElmer Life Sciences. Human recombinant IFN- was purchased from Genzyme (Cambridge, MA). The NO donor, S-nitroso-L-glutathione (GSNO), was purchased from Calbiochem and from Cayman Chemical (Ann Arbor, MI). 2',7'-Dichlorofluorescein diacetate (DCFH-DA) was obtained from Amersham Biosciences. The antibodies to STAT1 (p91) and interferon regulatory factor-1 (IRF-1) were obtained from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). The antibodies to nuclear factor (NF)-Y_A and NF-Y_B were purchased from Rockland Inc. (Gilbertsville, PA). The oligonucleotide corresponding to the Y box motif in the HLA-DRA proximal promoter was synthesized by Genosys (The Woodlands, TX). The cDNA probe for HLA-DRA and the HLA-DRA promoter construct linked to the chloramphenicol acetyltransferase reporter gene ([300]DR.CAT) were generously provided by J. Strominger (Harvard University, Cambridge, MA) and L. Glimcher (Harvard School of Public Health, Boston). A murine monoclonal antibody directed against human HLA-DRA was obtained from A. Friedman (Dana Farber Cancer Institute, Boston).

Cell Culture-- Human saphenous vein endothelial cells were cultured and passaged as described previously (23). Cellular confluence was maintained for all treatment conditions. Cellular viability was assessed by morphology, cell number, DNA content, and trypan blue exclusion.

Cell Surface Enzyme Immunoassay-- Cytokine-stimulated endothelial cells were cultured on 96-well Falcon plates (Lincoln Park, NJ), rinsed with PBS and 2% fetal calf serum, and incubated with the indicated murine monoclonal antibody to human HLA-DRA for 2 h. After rinsing three times with PBS, cells were incubated with biotinylated secondary antibody (horse anti-mouse IgG, Vector Laboratories, Inc., Burlingame, CA, 1:1000) for 1 h before incubation with streptavidin-alkaline phosphatase (Zymed Laboratories Inc., South San Francisco, CA) for 30 min. Cells were then treated with p-nitrophenyl phosphate (1 µg/ml) for 30 min at 22 °C. Light absorbance was measured in a plate reader (Dynatech) at 410 nm, using cells without primary antibody as a blank. Integrity of the monolayers was checked before analysis. Each experiment was performed in quadruplicate.

Northern Blotting-- Equal amounts of total RNA (20 µg) were separated by 1.2% formaldehyde-agarose gel electrophoresis, transferred overnight onto nylon membranes by capillary action, and baked for 2 h at 80 °C. Radiolabeling of the full-length HLA-DRA, CIITA, or -actin cDNA probe was performed using random hexamer priming, [-³²P]CTP, and DNA polymerase I (Klenow fragment, Amersham Biosciences). The membranes were hybridized with the probes overnight at 45 °C in a solution containing 50% formamide, 5× SSC, 2.5× Denhardt's solution, 25 mM sodium phosphate buffer (pH 6.5), 0.1% SDS, and 250 µg/ml salmon sperm DNA. All Northern blots were subjected to stringent washing conditions (0.2× SSC, 0.1% SDS at 65 °C) before autoradiography with an intensifying screen for 24 h to 72 h at 80 °C.

Western Blotting-- Nuclear extracts were diluted 1:2 with Laemmli Sample buffer (Bio-Rad), boiled for 5 min, and centrifuged for 2 min at 14,000 × g. Protein concentration was determined with the Micro BCA Protein Assay (Pierce). Samples (15 µg of protein) were separated by SDS-PAGE (10% running, 4% stacking). The separated proteins were electrophoretically transferred to nitrocellulose membranes with 0.45-µm pore size from Osmonics (Westborough, MA) using the Mini Trans-Blot Cell (Bio-Rad). The blots were blocked for 1 h at room temperature in PBS buffer (containing 0.1% Tween 20 and 5% nonfat milk), before incubation with the primary antibody (anti-NF-Y_A or anti-NF-Y_B, Rockland Inc., Gilbertsville, PA) overnight at 4 °C. After washing the membranes four times in PBS with Tween 20 buffer, a peroxidase-conjugated secondary antibody (anti-rabbit IgG, Rockland Inc., 1:4000) was added for 45 min. Immunodetection was accomplished using the Renaissance chemiluminescence kit (PerkinElmer Life Sciences).

Fluorescent Measurement of Intracellular Oxidation-- HSVEC of less than three passages were cultured in 35-mm dishes (Corning Glass) coated with 0.1% gelatin. Phenol-free M199 medium + 15 mM Hepes (pH 7.4) was used. Before seeding the endothelial cells, a sterile coverslip was placed on the bottom of each dish. To subtract background fluorescent activity from intracellular fluorescence, these coverslips were eliminated before measurement in order to have a cell-free control field. Intracellular generation of reactive oxygen species (ROS) was quantified using DCFH-DA (Amersham Biosciences). This esterified form is cell membrane-permeable and undergoes deacetylation by intracellular esterases. Upon oxidation, DCFH is converted to dichlorofluoroscein (DCF), a fluorescent compound. Confluent endothelial cell monolayers were incubated 30 min with 30 µM DCFH-DA before stimulation with the indicated substances. Fluorescence was monitored under 5% CO₂ at 37 °C using an inverted microscope (Zeiss Axiovert 405 M, Oberkochen, Germany). A mercury lamp with a 490-nm filter was used as a light source for excitation. Excitation time (3 s) was constant for all conditions. Emission wavelength was set to 525 nm. Images were acquired using a CCD camera (Photometrics CH 250, Tucson, AZ) with a 512 × 512 pixel format. Analysis was performed with ISEE software version 3.6 (Inovision, Durham, NC) from six different representative fields for each condition in each experiment. Background fluorescence activity (cell-free area) was subtracted.

Nuclear Run-on Assay-- Confluent endothelial cells (~10⁸ cells) were stimulated with IFN- (1000 units/ml) alone or in combination with GSNO (0.2 mM) for 24 h. Cells were subsequently washed twice with PBS, trypsinized, and centrifuged at 300 × g for 5 min at 4 °C. The cellular pellet was gently resuspended in a buffer containing 10 mM Tris-HCl (pH 7.4), 10 mM NaCl, 3 mM MgCl₂, and 0.5% Nonidet P-40, allowed to swell on ice for 15 min, and lysed by a Dounce homogenizer (60-70 strokes) with intermittent inspection of nuclei. The lysate was recentrifuged at 300 × g, and the resulting nuclear pellet was resuspended in 100 ml of buffer containing 20 mM Tris-HCl (pH 8.1), 75 mM NaCl, 0.5 mM EDTA, 1 mM dithiothreitol, and 50% glycerol. In vitro transcription using the nuclear pellet (100 µl) was performed in a shaking water bath at 30 °C for 30 min in a buffer containing 10 mM Tris-HCl (pH 8.0), 5 mM MgCl₂, 300 mM KCl, 50 mM EDTA, 1 mM dithiothreitol, 0.5 units of RNasin (Promega, Madison, WI), 0.5 mM CTP, ATP, GTP, and 250 µCi of [-³²P]UTP as described previously (24). Equal amounts (1 µg) of purified, denatured full-length HLA-DRA, human -tubulin (ATCC #37855), and linearized pGEM-3z cDNA were vacuum-transferred onto nylon membranes using a slot blot apparatus (Schleicher & Schuell). The membranes were baked and prehybridized as described for Northern blots. The precipitated radiolabeled transcripts (~8 × 10⁷ cpm) were resuspended in 2 ml of hybridization buffer containing 50% formamide, 5× SSC, 2.5× Denhardt's solution, 25 mM sodium phosphate buffer (pH 6.5), 0.1% SDS, and 250 mg/ml salmon sperm DNA. Hybridization of radiolabeled transcripts to the nylon membranes was carried out at 45 °C for 48 h. The membranes were then washed with 1× SSC, 0.1% SDS for 1 h at 65 °C before autoradiography for 72 h at 80 °C.

Electrophoretic Mobility Shift Assay-- Nuclear extracts were prepared as described (12). Oligonucleotides corresponding to the HLA-DRA Y box (5'-ATTTTTCTGATTGGCCAAAGAGTA-3') were radiolabeled with [-³²P]ATP and T₄ polynucleotide kinase (New England Biolabs) and purified by Sephadex G-50 columns (Amersham Biosciences AB). Nuclear extracts (10 µg) were added to ³²P-labeled oligonucleotides (~20,000 cpm, 0.2 ng) in a buffer containing 4 µg of poly(dA·dT) (Amersham Biosciences), 10 µg of bovine serum albumin, 10 mM Tris-HCl (pH 7.5), 25 mM NaCl, 50 mM MgCl₂, 1 mM dithiothreitol, 1 mM EDTA, and 5% glycerol (total volume of 20 µl). DNA-protein complexes were resolved on 4% non-denaturing polyacrylamide gel electrophoresed at 12 V/cm for 3 h in low ionic strength buffer (0.5× TBE) at 4 °C. For supershift assays, the indicated antibody (15 µg/ml) was added to the nuclear extracts for 10 min before the addition of radiolabeled probe. To determine the specificity of shifted bands, excess unlabeled oligonucleotide (20 ng) was added directly to the nuclear extracts for 10 min before addition of corresponding radiolabeled probe.

Transfection and CAT Assay-- The human HLA-DRA proximal promoter containing the chloramphenicol acetyltransferase (CAT) reporter gene ([300].DR.CAT) was described previously by Hehlgans and Strominger (25). Bovine aortic endothelial cells were transfected with each reporter plasmid (50 µg) using the calcium phosphate precipitation method. As an internal control for transfection efficiency, pRSV.GAL plasmid (10 µg) was co-transfected in all experiments. Preliminary results using -galactosidase staining indicate that cellular transfection efficiency was ~15%. Cells (60-70% confluent) were stimulated 24 h after transfection with IFN- (1000 units/ml) in the presence and absence of GSNO (0.2 mM), and cellular extracts were prepared 24 h later using lysis buffer (100 µg/ml leupeptin, 50 µg/ml aprotinin, 0.1 ml of phenylmethylsulfonyl fluoride, 5 mM EDTA, 5 mM EGTA, 100 mM NaCl, 5 mM Tris-HCl (pH 7.4)) and one freeze-thaw cycle. The cellular extracts were centrifuged at 12,000 × g for 10 min, and the supernatant was subjected to CAT and -galactosidase assay as described previously (26). The relative CAT activity was calculated as the ratio of CAT to -galactosidase activity. Each experiment was performed three times in duplicate, and all experiments included both positive (highly expressed pSV40.CAT) and negative (promoterless pCAT) controls.

Data Analysis-- Band intensities from Northern blots, nuclear run-on assays, and EMSA blots were analyzed densitometrically by the NIH Image program (National Institutes of Health, Bethesda). All values are expressed as mean ± S.E. compared with controls and among separate experiments. Paired and unpaired Student's t tests were employed to determine the significance of changes in absorbance values and densitometric measurements. p values of less than 0.05 were considered significant.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Inhibition of MHC-II Molecules by Nitric Oxide and Antioxidants-- To investigate the immunomodulating role of NO and antioxidants, we treated human vascular endothelial cells with the NO donor, GSNO, and a variety of antioxidants on the cell surface expression of major histocompatibility complex class II (MHC-II) antigen, HLA-DR. No constitutive MHC-II expression was detected on the surface of unstimulated endothelial cells by enzyme immunoassay. INF- induced the expression of MHC-II in a concentration-dependent manner. Tumor necrosis factor- (1000 units/ml) had no effect on MHC-II expression but stimulated basal MHC-I expression by 2.6-fold (Fig. 1A). GSNO (10-500 µM) inhibited IFN--induced MHC-II expression in a concentration-dependent manner. More than 60% inhibition of MHC-II expression was achieved with 500 µM GSNO. The basal MHC-I expression was not affected by the NO donor even at highest concentrations, indicating that the observed effects on MHC-II expression were not due to altered cellular viability. Inhibition of endothelial NO production by the NO synthase inhibitor, N-monomethyl-L-arginine (L-NMA, 1 mM), did not augment IFN--induced MHC-II cell surface expression and did not per se induce MHC-II expression. Stimulation of endothelial cells with IFN- in the presence of the membrane-permeant antioxidant NAC (30 mM) or PDTC (0.2 mM) decreased IFN--induced MHC-II expression by 60 and 93%, respectively (Fig. 1B). L-Arginine deprivation of the culture medium reduces the inhibitory effect of PDTC (data not shown), indicating a synergistic effect of PDTC and endogenous NO in inhibiting MHC-II expression. These effects were not mediated by cGMP, since the effects on MHC-II expression were not affected by treatment with the membrane-permeable cGMP analogues, dibutyryl cGMP and 8-bromo-cGMP (1 µM).

View larger version (17K): [in this window] [in a new window]   Fig. 1.   Expression of MHC in human vascular endothelial cells. A, comparative expression of MHC-I and MHC-II in response to tumor necrosis factor- or IFN- and inhibition of IFN--induced MHC-II expression by GSNO. B, role of reactive oxygen species in expression of MHC-II. *, p < 0.05 compared with IFN- alone.

Effect of NO and ROS on HLA-DRA mRNA Expression-- IFN- induction of HLA-DRA mRNA expression could be detected as early as 6 h, reaching a maximum level at 12 h (Fig. 2). We next determined the effect of co-stimulation with GSNO. In the presence of GSNO (0.2 mM) induction of IFN--stimulated HLA-DRA mRNA expression occurred later, at 12 h, and densitometric analysis of autoradiographic bands showed a 6.3-fold decreased steady state HLA-DRA mRNA level at 24 h compared with IFN- stimulation alone. In addition, the steady state expression of HLA-DRA mRNA at 24 h was blunted by NAC and PDTC and, to a lesser extent, by membrane-permeable PEG-SOD (100 units/ml) and PEG-catalase (500 units/ml), suggesting involvement of reactive oxygen species in IFN--induced modulation of HLA-DRA mRNA (Fig. 3). The decreased steady state HLA-DRA mRNA expression was in agreement with the protein expression as quantified by cell surface enzyme immunoassay. Similar Northern analyses indicate that GSNO (0.5 mM), L-nitroarginine methyl ester (10 µM), or H₂O₂ (150 µM) had no effect on the induction of CIITA by IFN-. (Fig. 4). We detected no basal CIITA mRNA expression. IFN- (1000 units/ml) induced CIITA mRNA expression between 2 and 6 h (Fig. 4A). Like GSNO, co-stimulation with either PDTC, NAC, catalase, or SOD had no effect on IFN--induced CIITA mRNA expression at 24 h (data not shown). L-NAME or H₂O₂ alone was not sufficient to induce CIITA mRNA expression after 24 h (Fig. 4B).

View larger version (54K): [in this window] [in a new window]   Fig. 2.   Northern analyses (20 µg of total RNA/lane) showing time-dependent effects of GSNO (0.2 mM) on HLA-DRA steady state mRNA expression in HSVEC stimulated with IFN- (1000 units/ml). The experiment was repeated three times with similar results.

View larger version (71K): [in this window] [in a new window]   Fig. 3.   Northern analyses (20 µg of total RNA/lane) showing the effects of PEG-SOD (100 units/ml), PEG-catalase (500 units/ml), and the antioxidants NAC (30 mM) and PDTC (0.2 mM) on IFN- (1000 units/ml)-induced HLA-DRA steady state mRNA levels at 24 h. Equal RNA loading for each experiment was verified by hybridization to -actin. Experiments were performed three times with similar results.

View larger version (47K): [in this window] [in a new window]   Fig. 4.   A, Northern analyses (20 µg total RNA/lane) showing time-dependent effects of GSNO (0. 2 mM) on CIITA steady state mRNA expression in HSVEC stimulated with IFN- (1000 units/ml). B, Northern analyses (10 µg total RNA/lane) showing effects of L-NAME (100 µM) or H2O2 (150 µM) on CIITA steady state mRNA expression (24 h) in HAEC under basal and IFN- (1000 units/ml)-stimulated conditions. RNA loading was verified by ethidium bromide-stained 28 S ribosomal RNA. Three different experiments showed similar results.

IFN- Induces ROS Generation-- Inhibition of IFN--induced expression of MHC-II molecules and HLA-DRA mRNA by a variety of antioxidants suggests that cellular responses to IFN- involve ROS. We measured DCF fluorescence in endothelial cells as a marker of oxidative stress to determine the cellular redox status under our experimental conditions (Fig. 5). The fluorescence intensities were recorded 10 min following stimulation with the indicated substances. Under basal conditions, there was a very slow increase in endothelial cell fluorescence (19.5 ± 5.6 units, Fig. 5A). Treatment with IFN- (1000 units/ml) causes a marked increase in DCF fluorescence within 5 min compared with unstimulated cells (74 ± 21 units, Fig. 5B). The difference between resting and stimulated cells persisted at least 60 min (data not shown). Co-stimulation with the antioxidant NAC (30 mM) completely reversed this effect (10.9 ± 3.1 units; Fig. 5A), whereas L-NMA did not (58.1 ± 16.8 units). L-NMA alone had no effect (24 ± 6.9 units). A very high fluorescence signal was achieved by direct stimulation of the cells with exogenous H₂O₂ (420 ± 121 units, Fig. 5A). These findings suggest that ROS are involved in cellular response to IFN-. DCFH oxidation cannot be attributed to a single reactive oxygen species, because several intermediates during the reduction of hydrogen peroxide (H₂O₂) oxidize DCFH (27-29). Therefore, formation of DCF from DCFH should be considered as an index of overall oxidative stress (30).

View larger version (66K): [in this window] [in a new window]   Fig. 5.   Production of intracellular reactive oxygen species. A, DCF fluorescence showing the effect of antioxidants and IFN- on intracellular oxidation. Fluorescence intensity was recorded at base line (control) and after 10 min of stimulation with IFN- (1000 units/ml) with and without NAC (30 mM). Cells were also stimulated with exogenous H2O2 (150 µM). B, DCF fluorescence showing the effect of PEG-SOD (100 units/ml) and PEG-catalase (500 units/ml) on IFN- (1000 units/ml)-stimulated ROS. Each experiment was performed twice in quadruplicate.

Nitric Oxide Inhibits the Transcription of the Human HLA-DRA Gene-- To confirm that NO decreases IFN--induced steady state HLA-DRA mRNA levels by transcriptional repression, we performed nuclear run-on experiments using human EC stimulated with IFN- (1000 units/ml) for 12 or 24 h in the presence or absence of NO (GSNO, 0.2 mM, Fig. 6). In unstimulated EC there was no basal HLA-DRA transcriptional activity. IFN- (1000 units/ml)-induced HLA-DRA gene transcription was detectable after 12 h and maximal after 24 h. Preliminary studies using different amounts of radiolabeled RNA transcripts demonstrate that under our experimental conditions, hybridization was linear and not saturable. Co-treatment with GSNO demonstrated a transcriptional effect of NO on HLA-DRA expression by repressing HLA-DRA transcription. Specificity was established by lack of hybridization to the insertless vector, pGEM. Transcription of the -tubulin gene served as an internal control.

View larger version (33K): [in this window] [in a new window]   Fig. 6.   A, nuclear run-on assay showing the effects of IFN- (1000 units/ml) and ·NO (GSNO, 0. 2 mM) on HLA-DRA gene transcription at 12 and 24 h. The density of each HLA-DRA band was standardized to the density of its corresponding -tubulin gene transcription band. The specificity of each band was determined by the lack of hybridization to the nonspecific linearized pGEM cDNA vector. B, Western blot showing the effects of IFN- (100 and 1000 units/ml) and IFN- (1000 units/ml) + PEG-SOD (100 units/ml) on eNOS protein expression. Representative of two separate experiments.

The CCAAT Motif Is a Redox-sensitive Cis-acting Element in the MHC Class II Promoter-- The human HLA-DRA proximal promoter linked to the chloramphenicol acetyltransferase (CAT) reporter gene, [300].DR.CAT, was used in transient transfection studies. These studies were performed with bovine rather than human endothelial cells due to higher transfection efficiency with bovine cells using the calcium phosphate precipitation method (12 versus <2%). The promoterless pCAT produced essentially no relative CAT activity (50 ± 40). The highly expressed pSV40.CAT containing the SV40 early promoter exhibited a high level of relative CAT activity (1400 ± 260) which is a 7.46 ± 260-fold induction compared with [300].DR.CAT (Basal, Fig. 7). Treatment of the cells with IFN- (1000 units/ml) for 12 h caused a 4.1 ± 0.3-fold increase in relative CAT activity compared with basal activity. Co-stimulation with GSNO significantly inhibited the IFN- effect (only 1.4 ± 0.16-fold induction). Interestingly, we found that H₂O₂ can directly induce the promoter activity 5.7 ± 0.6-fold, whereas inactivation of H₂O₂ by catalase completely abolished IFN- induced promoter activity (0.7 ± 0.26-fold), suggesting that H₂O₂ is a necessary mediator of IFN-. As shown above, H₂O₂ by itself is not sufficient to induce MHC-II expression (Fig. 1B), although disrupting the cellular redox homeostasis (Fig. 5A). SOD (100 units/ml) also inhibited IFN--induced promoter activity (3.0 ± 0.4-fold induction). Inhibition of endogenous NO by L-NMA (1 mM) increased both basal and IFN--stimulated HLA-DRA minimal promoter activity (4.1 ± 0.5- and 9.5 ± 1.2-fold, respectively).

View larger version (17K): [in this window] [in a new window]   Fig. 7.   Transfection studies using promoterless construct (vector), a highly expressed constitutive promoter (pSV40), and the HLA-DRA promoter construct [300]DR.CAT. * represents a significant change from basal [300]DR.CAT activity (p < 0.05). ** represents a significant change from IFN- (1000 units/ml) stimulation (p < 0.05). Experiments were performed four times with similar results.

To localize the site of inhibition on the HLA-DRA promoter by NO, we analyzed the DNA binding activity of NF-Y under cytokine stimulation and under co-stimulation with nitric oxide. This transcription factor was of particular interest because it recognizes the CCAAT box, a redox-sensitive motif within the HLA-DRA promoter, with high affinity and specificity (31, 32). The CCAAT box is necessary for HLA-DRA gene transcription. To determine whether exogenous NO inhibits NF-Y DNA binding activity, we performed EMSA using radiolabeled oligonucleotide corresponding to the Y box in the HLA-DRA promoter (Fig. 8A). We found a significant decrease in NF-Y binding activity when cells were treated with GSNO (0.5 mM) in addition to IFN- (Fig. 8A, lane 3). Stimulation with IFN- alone (Fig. 8, lanes 2, 4, and 5) seems to increase DNA binding, but this effect was not significant when compared with unstimulated cells (lane 1). The shifted complexes were specific for NF-Y since they were supershifted in the presence of an antibody to the A subunit of NF-Y (Fig. 8, lane 5) and disappeared with excess unlabeled oligonucleotide (lane 4). To investigate further the role of NF-Y in regulating the effect of NO in MHC-II gene transcription, we tested whether the altered DNA binding activity may be explained by nuclear NF-Y expression. Although IFN- alone did not alter the expression of the A and B subunits of NF-Y, we found that co-treatment with GSNO reduced the amount of both NF-YA and NF-YB (Fig. 8B).

View larger version (56K): [in this window] [in a new window]   Fig. 8.   HSVECs were untreated or stimulated with IFN- (1000 units/ml) alone or in combination with GSNO (0.5 mM) for 24 h, and then nuclear extracts were prepared. A, electrophoretic mobility shift assay was performed with 10 µg of nuclear extracts from untreated cells (lane 1), IFN--stimulated cells (lanes 2, 3, and 5), or IFN-/GSNO-stimulated cells (lane 3). 100-Fold excess of unlabeled NF-Y oligonucleotide was used for competition analysis (lane 4). Supershift analysis was performed in the presence of 2 µg of anti-NF-Y (A subunit-specific) antibody (lane 5). B, immunoblots of nuclear extracts (15 µg of protein) were performed with antibodies directed against the differentially spliced 35- and 40-kDa forms of the A subunit (1:1000) and the 25-kDa form of the B subunit of human NF-Y (1:500). Data shown are representative of three experiments.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

We have shown that the induction of HLA-DRA gene transcription by IFN- is dependent upon the generation of ROS in vascular endothelial cells, in part by activating Y box trans-activating proteins. These findings are consistent with studies showing that IFN- stimulates the production of ROS in vascular endothelial cells (22) and that the Y box motif is an oxidant-sensitive cis-acting element (20). The ability of antioxidants such as NAC, PDTC, PEG-conjugated catalase, and SOD to inhibit IFN--induced HLA-DRA protein and mRNA expression suggests that ROS are required for IFN--induced HLA-DRA gene transcription. Indeed, we found that H₂O₂ can directly stimulate the HLA-DRA minimal promoter. However, neither H₂O₂ nor L-NMA, alone or in combination, could induce the expression of HLA-DRA, indicating that ROS production is necessary but, by itself, not sufficient for HLA-DRA gene transcription.

Most interesting, NO was able to inhibit IFN--induced HLA-DRA gene transcription. Inhibition of endogenous NO by L-NMA increased both basal and IFN--stimulated HLA-DRA minimal promoter activity. Endothelium-derived NO is an important endogenous modulator of vascular tone, platelet aggregation, and vascular smooth muscle proliferation (33-36). For example, supplementation with L-arginine, the precursor of NO, to the diets of cholesterol-fed rabbits leads to inhibition of endothelium-leukocyte interaction and attenuation of atherosclerotic lesions (37). Inhibition in endogenous NO with N-monomethyl-L-arginine results in enhanced leukocyte adhesiveness to the vessel wall (38). Indeed, we have shown recently (39) that exogenous NO can regulate leukocyte homeostasis in the vessel wall by inhibiting the release of soluble cytokines such as IL-6 and IL-8 and attenuating the expression of various cell surface adhesion molecules. The mechanism underlying the effects of NO on cytokine-induced VCAM-1 expression is via the inhibition of NF-B. This is similar to the effects of antioxidants, which also inhibit cytokine-induced NF-B activation and VCAM-1 expression (40, 41). Thus, NO appears to perform similar functions as antioxidants with respect to inhibiting cytokine-induced endothelial cell activation.

Our observations that NO inhibits IFN--induced HLA-DRA promoter activity and inhibition of NO synthase increased HLA-DRA promoter activity suggest a direct effect of IFN- on eNOS. This is supported by previous studies (42, 43) showing that exposure of endothelial cells to cytokines, among them IFN-, reduced eNOS expression. However, we did not observe changes in eNOS expression upon stimulation with IFN-. The finding that L-NMA was able to increase further IFN--induced activity of the HLA-DRA promoter construct suggests that IFN- does not significantly alter eNOS function but inhibits physiological functions of NO, probably by generating superoxide which rapidly reacts with NO (44).

The induction of MHC class II gene transcription by IFN- occurs relatively slowly and is mediated by the MHC class II transactivator (CIITA), a non-DNA-binding protein, and depends on the presence of the transcription factors STAT1 (45) and IRF-1 (46). STAT1 induces transcription of CIITA both by direct binding to CIITA promoter and by inducing transcription of IRF-1. CIITA is both essential and sufficient for MHC-II expression (47). In addition, stereo-specific alignment of the X and Y box motif is required for the MHC-II promoter response to IFN- (48-50). Although CIITA does not directly bind to enhancer elements in the MHC-II genes, its activation is necessary for the coordinate binding of W, X, and Y box trans-activating proteins (51).

In our study, NO or antioxidants inhibited expression and binding of Y box-binding proteins. Furthermore, electrophoretic mobility shift assays showed that co-stimulation of endothelial cells with NO or antioxidants also resulted in inhibition of STAT1 induction (data not shown). However, we observed no effect of NO on IRF-1 activity or CIITA induction in IFN--stimulated cells. This was in contrast to data published previously by Kielar et al. (52) who found that NO inhibited IFN--induced increases in CIITA gene transcription in murine macrophages. A possible explanation for this discrepancy may due to differences in cell type and species. For example, macrophages constitutively express MHC-II, whereas endothelial cells do not express MHC-II molecules under basal conditions. A possible explanation for the inhibitory effect of NO on the IFN- pathway, although without affecting CIITA expression, could be the finding in activated macrophages that nitration of tyrosine residues in STAT1 by NO inhibits IFN--induced phosphorylation of STAT1 (53). It is known that STAT1 can be specifically activated by oxidative stress (54), and we demonstrated ROS upon stimulation with IFN-, and therefore it is not surprising that antioxidants affect STAT1 activity. But again, these effects were only weak compared with the effect on Y box transcription factors.

Inhibition of endogenous NO synthesis by L-NMA did not affect IFN--induced MHC-II expression in our study, whereas we observed a 4-fold induction of the HLA-DRA promoter by NO, even in the absence of IFN-. Moreover, L-NMA further increased IFN--induced promoter activity. Taken together, these findings support our observation that elements in the MHC-II promoter are required for the NO effects on MHC-II expression. Within the HLA-DRA promoter, the Y box motif is of considerable interest for its ability to bind oxidant-sensitive transcription factors (21). The Y box contains an inverted CCAAT motif and plays a role in eukaryotic redox signaling (20). A recent study demonstrated that the prokaryotic oxyR-response element, which is found in the promoters of many bacterial genes coding for peroxidase-inactivating enzymes such as catalase (katG), is highly homologous to the eukaryotic Y box cis-acting element (20). The oxyR-response element can function as a redox-dependent transcriptional enhancer in murine cells by interacting with a member of the Y box family of DNA- and RNA-binding proteins, YB-1 (20). Thus, although the MHC-II genes lack oxidant-sensitive B and activated protein-1 sites, which are present in many pro-inflammatory genes including MHC-I, they could still be responsive to ROS production through their Y box cis-acting elements.

Recently, the ROS-mediated regulation of the Escherichia coli OxyR transcription factor could be demonstrated by crystal structure analysis (55). Reversible intramolecular disulfide bond-mediated changes in the protein structure appear to be responsible for its activity. The OxyR protein is activated in response to H₂O₂ and induces transcription of genes that protect the bacterium against oxidative stress (56). We cannot rule out the possibility that functional activities such as protein-protein interactions of the Y box proteins or their transcriptional activation may be affected by co-stimulation of IFN--stimulated cell, thus inhibiting MHC-II expression. Indeed, our findings that H₂O₂ by itself induced HLA-DRA promoter activity and antioxidants such as catalase completely inhibited IFN- induced promoter activity support the postulated function of NF-Y as an oxidant-sensitive transcription factor. In conclusion, we find that the Y box motif is an oxidant-sensitive cis-acting element in the HLA-DRA promoter that mediates anti-inflammatory responses of NO in vascular endothelial cells. The immunomodulating effect of NO on vascular wall cells is consistent with its complex role in regulating inflammatory responses in the vascular wall.

	ACKNOWLEDGEMENTS

We thank J. Strominger and L. Glimcher for HLA-DR cDNA and promoter CAT constructs and A. Friedman (Dana Farber Cancer Institute, Boston) for murine monoclonal antibody to human HLA-DR. We are also grateful to K. L. Wright (H. Lee Moffitt Cancer Center, University of South Florida, Tampa, FL) for technical assistance.

	FOOTNOTES

^* This work was supported by National Institutes of Health Grants HL-52233 and HL-48743 and an American Heart Association Bugher Foundation award.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Recipient of a Feodor Lynen fellowship (Alexander von Humboldt Foundation).

^§ Present address: Dept. of Cardiology, St. Josef-Hospital, Ruhr-University Bochum, Germany.

^¶ Present address: Consiglio Nazionale delle Ricerche Institute of Clinical Physiology, University of Pisa, Italy.

Present address: Dept. of Medicine, University of Tokyo, Tokyo, Japan.

^** To whom correspondence should be addressed: Vascular Medicine Unit, Brigham & Women's Hospital, 221 Longwood Ave., LMRC-322, Boston, MA 02115. Tel.: 617-732-6538; Fax: 617-264-6336; E-mail: jliao@rics.bwh.harvard.edu.

Published, JBC Papers in Press, May 10, 2002, DOI 10.1074/jbc.M110538200

	ABBREVIATIONS

The abbreviations used are: MHC, major histocompatibility complex; IFN-, interferon-; ROS, reactive oxygen species; NO, nitric oxide; SOD, superoxide dismutase; PDTC, pyrrolidine dithiocarbamate; NAC, N-acetylcysteine; L-NMA, N-monomethyl-L-arginine; NF-Y, nuclear factor Y; DCF, dichlorofluoroscein; GSNO, S-nitroso-L-glutathione; EMSA, electrophoretic mobility shift assay; DCFH-DA, 2',7'-dichlorofluorescein diacetate; eNOS, endothelial nitric-oxide synthase; PEG, polyethylene glycol; IL, interleukin; CAT, chloramphenicol acetyltransferase; PBS, phosphate-buffered saline; NF-B, nuclear factor-B; IRF-1, interferon regulatory factor-1; VCAM-1, vascular intercellular adhesion molecule 1; ICAM-1, intercellular adhesion molecule 1; HSVEC, human saphenous vein endothelial cells.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES