Regulation of Class II Expression in Monocytic Cells after HIV-1 Infection¹

Macrophages play a critical role in immune regulation through Ag processing that is critical for T cell recognition of Ag (1). Macrophages process and present Ag through a multistage pathway that involves the breakdown of proteins into immunogenic peptides that complex with class II molecules, expression of costimulatory molecules, and production of accessory cytokines (2). Macrophages also have an important role in HIV-1 infection, serving not only as a reservoir for virus, but also contributing to the immune defects seen in HIV-1-infected patients (3, 4, 5, 6, 7, 8). A number of monocyte defects have been reported during different stages of HIV-1 infection, including impaired class II Ag expression, chemotactic defects, and aberrant cytokine production (9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19). Progressive impairment of monocytic function may contribute to the overall immune dysregulation that is observed in HIV-1-infected patients (20). The study of monocytic function after HIV-1 infection has been hampered by the lack of in vitro model systems. Our laboratory has generated a panel of human macrophage hybridoma cell lines that possess many normal monocytic functions and can be uniformly infected by HIV-1 (21, 22, 23, 24, 25). Using this system, we have demonstrated an impaired ability of HIV-1-infected hybridomas to present soluble Ag to MHC-matched responder T cells, which was associated with the loss of HLA-DR mRNA expression (22, 25). The block in mRNA expression could be overcome by transfection with HLA-DR-α and HLA-DR-β genes driven by a CMV promoter gene, suggesting that HIV-1 was mediating this effect by altering the regulation of mRNA production (25). Recently, this finding has been confirmed by other investigators (26).

Constitutive and induced class II expression in APC is regulated by a series of DNA-binding proteins that bind to conserved regulatory elements (W/S/Z, X-1, X-2, and Y boxes) and the class II MHC transactivator (CIITA),³ a non-DNA-binding protein transcription factor (27, 28, 29, 30, 31). In the present study, we first investigated whether HIV-1 infection interfered with DNA binding to the W/S/Z, X-1, X-2, Y box elements, or whether there was a direct effect on CIITA. We also determined whether the production of DNA-binding proteins RFX-5-W/Z/S and X-1, hX-2BP-X-2, NF-YA, and NF-YB-Y box) was suppressed after HIV-1 infection. In our previous studies, even though HLA-DR expression was restored after transfection with HLA-DR, the HIV-1-infected human macrophage hybridomas were incapable of inducing Ag-specific T cell proliferation (22, 25). In this study, we investigate whether there may be either a loss of invariant chain (Ii) production that chaperones HLA-DR into different subcellular compartments, an effect on HLA-DM-α and HLA-DM-β, nonpolymorphic MHC gene products necessary for Ag processing, or impaired sorting of HLA-DR into subcellular compartments.

Human macrophage hybridomas were obtained by fusing macrophages (obtained by allowing monocytes to mature into macrophages in Teflon bag cultures) with a hypoxanthine-guanine phosphoribosyltransferase-deficient promonocytic line (U937), as previously described (21). We have uniformly infected and characterized one clone, 43, with different strains of HIV-1 (43_HIV) (25).

Mononuclear cells were separated from buffy coats obtained from normal healthy volunteers by Ficoll-Hypaque (Pharmacia, Piscataway, NJ) density gradient centrifugation. The cells were washed three times with sterile PBS and resuspended in RPMI 1640 (Life Technologies, Grand Island, NY) supplemented with 10% FCS (Life Technologies), 2 mM l-glutamine, and 1% penicillin/streptomycin (Life Technologies), henceforth called complete medium (CM). Freshly isolated PBMCs were incubated at 37°C in CM and allowed to adhere for 45 min. The nonadherent cells were removed, and adherent cells were washed with sterile PBS, harvested with a rubber policeman, and stained with monocyte-specific anti-CD14 mAbs to assess the purity of the preparation. Ninety percent of the isolated cells expressed CD14 (21).

Monocytes or clone 43 cells were infected with HIV-1_ADA, HIV-1_87.9, HIV-1_BaL, and HIV-1_IIIB, respectively, as previously described (25). These reagents were obtained through the AIDS Research and Reference Reagent Program, Division of AIDS, National Institute of Allergy and Infectious Diseases (Bethesda, MD). Dilutions of HIV-1-containing supernatant standardized to contain reverse-transcriptase activity (80,000 cpm/ml) were incubated for 90 min, followed by three washes with PBS.

Oligonucleotides were synthesized for the W/Z/S, X-1, X-2, and Y box-binding studies by Integrated DNA Technologies (Coralville, IA) and labeled with ³²P dCTP, dATP, dGTP, and dTTP (ICN Pharmaceuticals, Irvine, CA) using DNA polymerase I (Sigma, St. Louis, MO) (32). Nuclear protein was extracted from 43 and 43_HIV, and monocytes were infected with HIV-1_BaL that retained (HLA-DR⁺) and lost (HLA-DR⁻) HLA-DR expression, according to the protocol of Dingham (33). Ten million cells were resuspended in 100 mM pepstatin A (Sigma), 10 mg/ml leupeptin (Sigma), 100 mM PMSF (Sigma), and 10 mg/ml aprotinin (Sigma). The cells were lysed with 20 mM HEPES (Sigma), pH 7.9, 1.5 mM MgCl₂ (Sigma), 10 mM KCl (Sigma), and 1 mM DDT (Sigma), and then centrifuged at 10,000 rpm for 30 s. The nuclei containing supernatant were added to 20 mM HEPES (Sigma), pH 7.9, 1.5 mM MgCl₂ (Sigma), 0.2 mM EDTA (Sigma), 20% glycerol (Sigma), 0.42 mM KCl (Sigma), and 1 mM DTT (Sigma); incubated on ice for 30 min; and then centrifuged for 5 min at 15,000 rpm. The nuclei were then added to the DNA oligonucleotides in Murphy’s binding buffer (BD PharMingen, San Diego, CA) for 2 h at room temperature. After binding, the mixture was electrophoresed on a 10% polyacrylamide gel, which was exposed to x-ray films (Kodak, Renseassler, NY) overnight and developed. In some experiments, the specificity of the binding was determined by adding 10 μg unlabeled oligonucleotide probes.

The HIV-1_BaL-infected monocytes, 43, and 43_HIV cells were stained by indirect methods, as previously described, with various mAbs (see below) or with isotype-matched controls, followed by affinity-purified F(ab′)₂ FITC-conjugated goat anti-mouse Ig Ab (Tago Scientific, Burlingame, CA), and analyzed by flow cytometry gating on live cells (21). The anti-class I Abs (W6/32) were obtained from the American Type Culture Collection (ATCC; Manassas, VA); anti-DR Abs, anti-class II-associated Ii peptide (CLIP), anti-CD80, and anti-CD86 Abs were purchased from BD Biosciences (Mountain View, CA); and the human polyclonal anti-HIV-1 Abs were a kind gift of A. Pinter (New York University, New York, NY). For intracytoplasmic staining, cells were fixed and permeabilized with 70% ethanol for 30 min at 4°C. The cells were then washed three times with PBS, and polyclonal anti-HIV-1 Abs or noninfected control Abs were added for 30 min at 4°C, followed by affinity-purified FITC-conjugated goat anti-human Ig Ab (Tago Scientific), and analyzed as described above (25).

Sequences for oligonucleotide primers used for PCR amplification and the size of the predicted PCR products are as follows: CIITA predicted size PCR product, 5′-CAACTCCCTGAAGGATGTGGA-3′, 350 bp, and 5′-ACGTCCATCACCCGGAGGGAC-3′; RFX-5, 5′-CGAGAATTCAGCTGTATCTCTACCTTC-3′, 586 bp, and 5′-GTCGAATTCAGGGAAGATCTCTCTGATG-3′; hX-2BP, 5′-ACCCCTAAAGTTCTGCTTCTGTCG-3′, 738 bp, and 5′-CATTAATGGCTTCCAGCTTGGCTG-3′; NF-YA,5′-GCAATAGTTCCACAGAGCAGATCG-3′, 929 bp, and 5′-CTAGGGATTTCTGCAGACTACATCGG-3′; NF-YB, 5′-ACTCGGATGATCTGTGTTCATGGCT-3′, 490 bp, and 5′-AGCCAGCTGGTAACTGGTTAGTGA-3′; Ii, 5′-TCCCAAGCCTGTGAGCAAGATG-3′, 410 bp, and 5′-CCAGTTCCAGTGACTCTTTCG-3′; HLA-DM-α, 5′-ACTTTTCCCAGAACACTCGG-3′, 341 bp, and 5′-CTGGAAGCTGAGTCCATC-3′; HLA-DM-β, 5′-ACAGCCACCTCAACCAAAAAGA-3′, 321 bp, and 5′-GGGGTTAAGGCTAAATGGGA-3′; and actin, 5′-TGACGGGGTCACCCACACTGTGCCCATCTA-3′, 661 bp, and 5′-CTAGAAGCATTGCGGTGGACGATGGAGGG-3′.

These primer sets have been previously published (27) and were synthesized by Integrated DNA Technologies. RNA was extracted from 43 and 43_HIV, HLA-DR⁺, and HLA-DR⁻ HIV-1_BaL-infected monocytes using acid guanidium thiocynate/phenol/chloroform, as described previously (RNAzol, Linnai, Dallas, TX) (34). Known quantities of RNA were mixed with 1 μg total cellular RNA and reverse transcribed at 37°C for 60 min in 20 μl buffer containing 10 mM Tris (Sigma) (pH 8.3); 50 mM KCl (Sigma); 5 mM MgCl₂ (Sigma); 1 mM each of dATP, dCTP, dGTP, and dTTP (Sigma); 20 U RNase inhibitor (Promega, Milwaukee, WI); 0.1 μg oligo(dT)₁₅ (Boehringer Mannheim, Indianapolis, IN); and 50 U murine leukemia virus reverse transcriptase (Bethesda Research Laboratories, Bethesda, MD). The PCR for RFX-5, hX-2BP, NF-YA, NF-YB, HLA-DM-α, HLA-DM-β, Ii, and actin was performed for 40 cycles at 94°C for 1 min, at 50°C for 1 min, and at 68°C for 3 min, with a 20-s elongation step per cycle (27). The amplification cycles were performed in the same tube for all of the X and Y box-binding proteins and for HLA-DM-α, HLA-DM-β, and actin. Reactions were stopped by heat inactivation for 10 min at 95°C, annealed for 2.5 min, and extended at 65°C. Negative controls were performed omitting RNA from the cDNA synthesis and specific amplification. PCR products were separated in a 2% NuSieve agarose (FMC, Rockland, ME) or a 5% polyacrylamide gel (23).

The 43_HIV cells were transiently transfected with a bacterial plasmid containing NF-YA obtained through the ATCC using a DEAE-dextran method (35). The bacteria containing the NF-YA plasmid was cultured, ethanol precipitated, extracted, and then centrifuged and resuspended in TBS. The resuspended DNA was added to 10 mg/ml DEAE-dextran and incubated with the 43_HIV and 63_HIV cells for 4 h at 37°C. After aspirating the DEAE-dextran, the cells were shocked by adding 5 ml 10% DMSO (Sigma) in PBS for 1 min at room temperature, washed with sterile PBS, and resuspended in CM for 48 h at 37°C. In some experiments, the 43_HIV cells were treated with 10 mg/ml DEAE-dextran alone without the NF-YA DNA, while in other experiments the 43_HIV cells were treated with DEAE-dextran and jellyfish green fluorescent protein (Promega) to assess the efficiency of transfection.

Uninfected, HIV-1-infected (43_HIV), or HIV-1-infected NF-YA-transfected 43 and 63 cells (43_HIV + T) were used as accessory cells in TT-induced T cell proliferation assays. In these experiments, T cells were obtained from an HLA-matched donor (HLA-DR2⁺) and were monocyte depleted using a nylon wool column. The PBMC were incubated for 45 min and eluted. T cells obtained in this manner failed to proliferate to TT in the absence of accessory cells (23). Monocyte-depleted PBMC (10⁵) were cultured with varying concentrations of 43 and 43_HIV, as well as 43_HIV transfected with NF-YA (10³–10⁵) and TT (0.4–40 μg/ml). The cells were maintained in 0.2 ml CM in triplicate round-bottom plates (Linbro, Oxnard, CA) at 37°C in a 5% CO₂ incubator for 5 days. Eighteen hours before harvesting, 1 μCi [³H]thymidine (ICN Pharmaceuticals) was added to each well. The cells were harvested onto glass fiber filters, and incorporated radiolabel was measured by scintillation counting (23).

FITC-labeled annexin V, a phospholipid-binding protein of the annexin family, was used to measure apoptosis using a commercially available kit (Beckman Coulter, Hialeah, FL). After incubating 43 and 43_HIV with T cells, the cell samples were washed with ice-cold PBS, followed by centrifugation at 500 × g at 4°C. The cells were stained simultaneously with annexin V FITC/PE-labeled anti-CD3 mAb and incubated at room temperature for 10 min in the dark. The cells were then analyzed by flow cytometry to measure costaining of the CD3⁺ and the annexin V⁺ population, gating on the live cells (36).

The 43 and 43_HIV cells were lysed using buffer containing PMSF (100 mM), aprotonin (10 mg/ml), leupeptin (10 mg/ml), iodoacetamide (1.8 mg/ml), and 0.1% Triton X. The lysates were resolved on a 10% SDS-PAGE gel, transferred onto a nitrocellulose membrane, blocked with 5% milk in PBS at room temperature for 60 min, then incubated with the PIN.1 mAb (37) at 4°C overnight. A secondary horseradish-labeled goat anti-mouse Ig (Tago Scientific) was then added at 25°C for 2 h, and the blot was developed by chemiluminescence using a commercially available kit (DuPont Pharmaceuticals, Wilmington, DE).

Vaccinia virus-containing vectors for HIV-1 envelope, gag, pol, and nef proteins were obtained from the AIDS Research and Reference Laboratory. Vaccinia virus was added to cultures of either uninfected 43 or 63 cells at a virus to cell ratio of 10⁸:1 that has been described previously (38). Successful infection was determined by intracytoplasmic immunofluorescence, as described above.

To colocalize HLA-DR in different subcellular compartments, we used mouse anti-cathepsin D (early endosomes) Ab (Accurate Antibodies, Westbury, NY), rabbit polyclonal anti-M6PR (late endosomes; provided by R. Dunn, University of Florida, Gainesville, FL), and mouse anti-HLA-DR (BD Biosciences). The 43 and 43_HIV cells and 43 cells infected with vaccinia virus containing different HIV-1 proteins were fixed with methanol/acetic acid (3/1) for 5 min at 4°C. Fixed cells were then blocked with PBS containing 5% goat serum and 0.2% Triton X-100 (Sigma) for 45 min before incubating with the primary Ab for 1 h. The cells were rinsed with PBS three times before incubating with Texas Red-conjugated Abs directed against either rabbit or mouse Ig for 30 min. Following this incubation, the cells were again washed three times with PBS and incubated with an FITC-labeled murine anti-HLA-DR Ab. The cells were rinsed with PBS, prepared as a cytospin, and mounted with Immu-mount (Shandon, Pittsburgh, PA) before being viewed by a Leica fluorovert laser-scanning confocal microscope (Leica, Deerfield, IL) at a step position of 1 μm on the x-y-axis or x-z-axis. To localize with 3-(2,4-dinitroanilino)-3′-N-methyldipropylamine) (DAMP), the HIV-1-infected and uninfected cells were fixed with 4% paraformaldehyde in PBS for 30 min. The cells were then washed twice with 50 mM NH₄Cl (Sigma) for 5 min each before permeabilization with 0.02% Triton X-100 (Sigma) in PBS for 30 min. The cells were then incubated with rabbit polyclonal rabbit anti-DAMP Abs (Molecular Probes, Eugene, OR) for 60 min at room temperature before mounting with Immu-mount (Shandon) and viewing by confocal microscopy (25).

Our previous studies demonstrated a loss of HLA-DR expression in the HIV-1-infected human macrophage hybridomas that could be overcome by transfecting HLA-DR-α and HLA-DR-β genes driven by a nonphysiologic CMV promoter (25). These findings suggested that HIV-1 infection had an effect on the regulation of HLA-DR mRNA production. Class II and HLA-DR mRNA production is regulated by different DNA-binding proteins that interact with a compact, conserved, multicomponent motif that contains four subelements, the W/Z/S box, the X-1 and X-2 boxes, and the Y box (39, 40, 41, 42, 43). We first investigated whether HIV-1 infection had any effect on the binding of these regulatory proteins by synthesizing oligonucleotides corresponding to the W/Z/S, X-1, X-2, and Y binding sites to use in gel-shift assays comparing HIV-1-infected with uninfected human macrophage hybridomas. We used one human macrophage hybridoma cell line (clone 43) for these experiments that lost HLA-DR expression after HIV-1 infection (22, 25). Nuclear protein extracts from 43 and 43_HIV were electrophoresed on a 5% polyacrylamide gel with ³²P-labeled or unlabeled oligoprobes (to assess the specificity of the DNA binding). In the 43 cells, W/Z/S, X-1, X-2, and Y box DNA binding was present and could be displaced by the unlabeled W/Z/S, X-1, X-2, and Y probes (Fig. 1,A). However, in the 43_HIV cells, there was no binding to either the W/Z/S, X-1, X-2, or Y boxes, respectively (Fig. 1,A). Because of the conflicting reports regarding the loss of HLA-DR expression in primary monocytes after HIV-1 infection (18, 19, 22, 25), we wanted to validate our findings in the 43_HIV cells using primary HIV-1_BaL-infected monocytes. In freshly isolated monocytes from normal blood donors, we detected decreased surface expression of HLA-DR in 3 of 10 preparations 8 days after infection with HIV-1_BaL (Table I). Similar results were obtained when we infected the primary monocytes with HIV-1_ADA, HIV-1_87.9, or primary isolates from HIV-1-infected donors (data not shown). We separately pooled the nuclear extracts from the HLA-DR⁺ and HLA-DR⁻ HIV-1_BaL-infected monocytes and performed W/Z/S, X-1, X-2, and Y box-binding studies. Consistent with the findings in the 43_HIV cells, there was reduced or absent W/Z/S, X-1, X-2, and Y box DNA binding in the HLA-DR⁻ HIV-1_BaL-infected monocytes compared with the HLA-DR⁺ HIV-1_BaL-infected monocytes (Fig. 1,A). CIITA is a non-DNA-binding protein and serves as a coactivation factor for class II expression (41). Lack of CIITA expression has been reported in class II-deficient patients with the bare lymphocyte syndrome (BLS) (40, 41). Competition between Tat and CIITA for binding to the P-TEBFb transcription factor has also been reported in HIV-1-infected THP-1 cells that lost class II expression (26). Although DNA binding to W/Z/S, X-1, X-2, and Y boxes can be demonstrated in CIITA-deficient patients (44), it is still possible that CIITA production may be affected by HIV-1 infection in our system and may contribute to the loss of HLA-DR expression. CIITA production was assessed by PCR in 43 and 43_HIV cells and in the HLA-DR⁺ and HLA-DR⁻ HIV-1_BaL-infected monocytes. A 350-bp band corresponding to CIITA was detected in both the HIV-1-infected and uninfected cells (Fig. 1 B).

Because there was no DNA binding to the W/Z/S, X-1, X-2, and Y boxes, we investigated whether there was loss of the W/Z/S, X-1, X-2, or Y box-binding proteins in 43_HIV, and in the HLA-DR⁻ HIV_BaL-infected monocytes. Because many DNA-binding proteins have been found to associate with class II regulatory sequences (45, 46, 47, 48, 49, 50, 51, 52, 53), we only assayed for those binding proteins that have been associated with the loss of class II expression in BLS or that cooperatively interact with other binding proteins (54). To this end, we determined the presence or absence of four DNA-binding proteins: RFX-5 (W/Z/S, X-1), hX-2BP (X-2), NF-YA, and NF-YB (Y box) by PCR in the 43_HIV cells. Amplified fragments corresponding to RFX-5 (504 bp)-, hX-2BP (738 bp)-, and NF-YB (490 bp)-binding proteins could be demonstrated in the uninfected 43 and 43_HIV cells (Fig. 2). However, a 926-bp fragment, corresponding to NF-YA, was absent in the 43_HIV cells. We again wanted to validate these findings in the HLA-DR⁻ HIV_BaL-infected monocyte preparations. Fragments corresponding to RFX-5 (504 bp), hX-2BP (738 bp), and NF-YB (490 bp) DNA-binding proteins were present, but there was no detectable NF-YA (926 bp) (Fig. 2). Because the loss of NF-YA could potentially explain the loss of HLA-DR after HIV-1 infection in our system, we transfected the 43_HIV cells with NF-YA in an attempt to restore expression.

We transfected full-length cDNA for the NF-YA DNA-binding protein driven by a CMV IE promoter into 43_HIV cells using DEAE-dextran. The surface expression of HLA-DR was lost in the 43_HIV cells 2 wk after infection, but was transiently restored after transfection with NF-YA (Fig. 3). Because the NF-YA-transfected 43_HIV cells expressed HLA-DR, we investigated whether these were functional by assessing whether they could present TT to HLA-matched (HLA-DR2⁺) responder T cells. APC-depleted T cells and TT (40 μg/ml) were cocultured in the presence of varying concentrations (10³-10⁵) of either 43 and 43_HIV cells or NF-YA box-transfected 43_HIV cells. In the absence of accessory cells, there was no T cell proliferation in response to TT (Fig. 4). Uninfected 43 cells induced a TT-specific proliferative response in the HLA-DR2⁺ T cells, inducing increases in thymidine incorporation, whereas 43_HIV failed to induce any T cell proliferation. After transfection with NF-YA, the 43_HIV cells were still unable to induce T cell proliferation in response to TT (Fig. 4). One possible explanation is loss of CD80 and CD86 expression in the 43_HIV cells. However, the surface expression of CD80 and CD86 was unchanged in the 43_HIV cells compared with the uninfected 43 cells (Fig. 5). We have reported that after prolonged HIV-1 infection (35 days), 43_HIV can induce apoptosis in cocultured T cells through gp120, Fas ligand, and the production of a soluble proapoptotic factor (22, 36). Even though the 43_HIV cells were infected for only 2 wk, we performed annexin V staining to ensure that apoptosis of the cocultured T cells didn’t account for the lack of a T cell response. There was no difference in annexin V staining in the T cells cocultured with 43_HIV or 43_HIV cells transfected with NF-YA compared with T cells cultured with the uninfected cells (data not shown). These findings are in line with our previous studies, which demonstrated that HLA-DR-transfected 43_HIV cells that expressed surface HLA-DR were still incapable of inducing an Ag-specific T cell response (25). Furthermore, in colocalization experiments, we demonstrated a diminished capacity to form HLA-DR-Ag complexes (25). The HLA-DR molecules assemble in the endocytoplasmic reticulum associated with Ii, a nonpolymorphic class II gene product that is coordinately expressed with HLA-DR and targets the molecule to either the late endosome or lysosome (55). Decreased production of Ii in the 43_HIV cells may explain the inability of the cells to induce an Ag-specific T cell response in our system. We performed PCR for Ii production in 43_HIV cells and in the HLA-DR⁺ and HLA-DR⁻ HIV-1_BaL-infected monocytes to determine whether HIV-1 infection also blocked Ii mRNA production.

We extracted RNA from 43 and 43_HIV cells and from the HLA-DR⁺ and HLA-DR⁻ HIV-1_BaL-infected monocytes to amplify Ii-specific base pair fragments (410 bp) (Fig. 6) by PCR. Bands (410 bp) corresponding to Ii were detected in 43 and 43_HIV as well as in the HIV_BaL HLA-DR⁺- and HLA-DR⁻-infected monocytes (Fig. 6). Other HLA gene products (HLA-DM-α and HLA-DM-β) are also coordinately expressed with Ii and are essential in Ag processing (55). HLA-DM-α and HLA-DM-β further cleave antigenic peptides in subcellular processing compartments, allowing for HLA-DR binding permitting Ag recognition by T cells (56). In the processing compartments, Ii is degraded into several peptides, including CLIP and leupeptin-induced protein (LIP). CLIP binds to the Ag-binding groove of the class II molecules and is removed first before endocytically generated antigenic peptides can bind. The removal of CLIP is catalyzed by HLA-DM. LIP is a 14-kDa NH₂-terminal fragment of the Ii that accumulates in the presence of the protease inhibitor leupeptin (37). Impaired production of HLA-DM-α and HLA-DM-β could possibly explain the failure to induce T cell proliferation in response to TT in the 43_HIV cells. We determined by PCR whether HLA-DM-α and HLA-DM-β were present in 43, 43_HIV, and in HLA-DR⁺ and HLA-DR⁻ HIV_BaL-infected monocytes. Similar to the results obtained with Ii, there was no difference in expression of either HLA-DM-α (341 bp) or HLA-DM-β (321 bp) in 43, 43_HIV, or in the HLA-DR⁺ and HLA-DR⁻ HIV-1_BaL-infected monocytes (Fig. 5). However, the Ii breakdown products CLIP and LIP were present in the uninfected 43 cells, but absent in the 43_HIV cells (Fig. 7, A and B). The PIN.1 mAb identifies not only LIP, but also the α-chain (molecular mass 30 kDa) and β-chain (molecular mass 28 kDa) of HLA-DR, which are present in both the 43 and 43_HIV cells. The presence of Ii and HLA-DM-α and HLA-DM-β in the 43_HIV cells in the absence of Ii breakdown products suggested a sorting defect.

To follow the sorting of HLA-DR into subcellular trafficking compartments, we permeabilized uninfected 43 cells along with the NF-YA-transfected 43_HIV cells and stained with FITC-labeled anti-HLA-DR Abs and Texas Red-labeled anti-early endosome (cathepsin D), anti-late endosome (M6PR), anti-lysosome (CD63), and anti-acidic compartment Abs (DAMP). In the uninfected 43 cells, HLA-DR was present (yellow staining) in the late endosomes, acidic compartments, and lysosomes (Fig. 8). However, in the NF-YA-transfected 43_HIV cells, HLA-DR did not localize into any subcellular compartment (Fig. 8). It is possible that there was competition between HIV-1 peptides and normal endosomal and lysosomal targeting that may have prevented HLA-DR sorting. To address this possibility, we introduced env, gag, pol, and nef proteins into uninfected 43 cells using vaccinia virus that contained vectors for different HIV-1 proteins, and then attempted to colocalize HLA-DR in the early and late endosomes, lysosomes, and acidic compartments.

To ensure that there were comparable amounts of the env, gag, pol, and nef proteins in the 43 cells before performing the colocalization experiments, we stained the cells intracytoplasmically with FITC-labeled human polyclonal anti-HIV-1 Ab (35). Using an empty vaccinia virus vector as a negative control, peak channel shifts corresponding to the presence of intracytoplasmic env, gag, pol, and nef proteins were detected to a comparable degree in the 43 cells after staining (Fig. 9). We then repeated the colocalization studies with FITC-labeled anti-HLA-DR Abs and the Texas Red-labeled anti-cathepsin D, anti-M6PR, anti-lysosome, and anti-DAMP Abs. HLA-DR localized (yellow staining) in the late endosomes, lysosomes, and acidic compartments in the 43 cells infected with the empty vaccinia virus (Fig. 10). HLA-DR also localized in the same compartments in the 43 and 63 cells that were infected with vaccinia virus containing the pol, gag, and nef proteins (Fig. 8). However, in the 43 cells infected with vaccinia virus that contained the env proteins, HLA-DR did not localize in any of the trafficking compartments (Fig. 8).

We have extended our previous studies investigating the effect of HIV-1 infection on class II expression in monocytic cells (22, 25). Using a human macrophage hybridoma model system, we demonstrated that HLA-DR expression was lost 2 wk after HIV-1 infection and could be restored by transfecting HLA-DR genes driven by a nonphysiological CMV promoter (25). The loss of HLA-DR expression in monocytic cell lines and primary monocytes has been variably reported by different investigators, with the percentage of monocytes infected with HIV-1 being between 30% and 70% (19, 20, 22, 25). In the present study, we observed the loss of HLA-DR in 30% of primary monocytes infected with HIV-1_BaL (Table I). These findings may relate to either varying numbers of infected cells in monocytic preparations or a different effect of HIV-1 infection on HLA-DR expression in monocyte subpopulations. Different subpopulations of monocytes have been described in the peripheral blood of healthy blood donors (57). In one report, the presence of a subtype resembling tissue macrophages expressing CD14⁺CD16⁺⁺ was found in the peripheral blood (58, 59). The CD14⁺CD16⁺⁺ monocytic population has been reported to be increased in patients with autoimmune diseases, AIDS, and AIDS dementia (60, 61). The human macrophage hybridomas were derived from peripheral blood monocytes and represent clonal expansions of different subpopulations of macrophages (21). The 43 cells express CD14⁺CD16⁺⁺ (21). It is possible that selection basis may have occurred during the generation of these cell lines, and our findings of lost HLA-DR expression may represent events occurring in a selected subpopulation of macrophages.

As mentioned above, the loss of HLA-DR expression in the human macrophage hybridomas after HIV-1 infection could be overcome by transfecting HLA-DR-α and HLA-DR-β genes driven by a nonphysiologic CMV promoter, suggesting an effect on regulatory DNA-binding proteins (25). HLA-DR and other class II genes are regulated by a series of proteins (X and Y box proteins) that bind to conserved elements of DNA (W/S/Z, X-1, X-2, and Y boxes) and CIITA, a non-DNA-binding protein that is an important transcription factor (60). Although RFX-5-, hX-2BP-, NF-YA-, and NF-YB-binding proteins are ubiquitously expressed, CIITA expression is restricted to class II-positive cells (63, 64). The identification of key class II regulatory factors has come from studies of B cell lines derived from patients with BLS, an immunodeficiency characterized by absent HLA class II expression (63, 64).

Four different types of defects have been described in BLS patients based on these studies. Class II deficiency is classified into four categories, A, B, C, and D, depending on the presence or absence of different X box DNA-binding proteins and CIITA (52, 62, 65, 66). Type A has no X box-binding proteins or CIITA activity; types B, C, and D have CIITA activity, but different defects in the X box-binding proteins. The 43_HIV cells and HIV-1_BaL-infected monocytes that lost HLA-DR expression had no W/Z/S, X-1, X-2, and Y box-binding activity, but had mRNA for CIITA (Fig. 1), resembling types B, C, and D class II deficiency. However, mRNA for RFX-5 and hX-2BP was present in these cells. The only defect noted was a loss of the NF-YA DNA-binding protein (Fig. 2) that was clearly important for HLA-DR in the human macrophage hybridomas because surface expression could be restored after transfection (Fig. 3).

The Y box proteins are necessary for class II expression and are comprised of two protein chains, NF-YA and NF-YB, which are highly homologous to the HAP2 and HAP3 transcription factors of Saccharomyces cerevisiae (67, 68, 69, 70, 71). The loss of the NF-YA DNA-binding protein as it relates to HLA-DR expression appears to be unique to our system. Others have reported that absent NF-YA increased HLA-DR expression and increased the susceptibility to rheumatoid arthritis (72, 73). Interestingly, the NF-YA protein binds to the long terminal repeat of HIV-1 and human T cell leukemia virus-1, can activate transcription of viral gene products, and is related to the Rous sarcoma virus enhancer factor I (74, 75, 76).

The regulation of class II expression in APC, including HLA-DR, HLA-DQ, HLA-DP, HLA-DM-α, HLA-DM-β, and Ii, probably involves an intricate regulatory unit consisting of multiple protein complexes (54). The high synergy in DNA-protein complex formation between RFX-5 and hX-2BP suggests that these proteins interact first to form the class II transcription complex (77, 78). The NF-YA and NF-YB proteins stabilize the interaction between RFX-5 and hX-2BP, which then recruits and binds CIITA, which activates transcription of class II gene products through its acidic activation domain (54). The data in Fig. 1,a demonstrate that there was no binding of nuclear extracts from the 43_HIV cells and the HIV-1_BaL-infected monocytes that lost HLA-DR expression to X box, Y box, and the W/S/Z box. The loss of binding of the X box proteins in the gel-shift assay (Fig. 1 a) may be due to the lack of stability of the RFX-hX2-BP complex due to the absence of NF-YA. Because RFX is also important for W/Z/S binding, the same might hold true for the absence of binding to the W/Z/S box.

One of the interesting features of our system is that there is selective loss of mRNA for HLA-DR, while mRNA for HLA-DM-α and HLA-DM-β and Ii are transcribed. Although HLA-DR, HLA-DM-α, HLA-DM-β, and Ii are reported to be transcribed coordinately by some investigators (79), more recent studies have shown that HLA-DR and Ii genes are not always coordinately transcribed (80, 81, 82). The HLA-DR and Ii promoters differ in spacing between the X and Y boxes (83, 84), and mutational analysis of HLA-DR and Ii promoters has demonstrated differences in the contribution of W/Z/S box binding (52). Subtle differences in promoter occupancy have also been noted between HLA-DR and Ii (81). Because differences in the stability of the class II transcription complex have been observed (85, 86, 87), it is conceivable that the lack of NF-YA-binding protein in vivo may alter the stability of this complex, accounting for the selective loss of HLA-DR transcription.

The loss of class II mRNA expression in monocytic cells after HIV-1 infection, associated with defects in transcription factors, has also been reported by other investigators. Kanazawa et al. (26) have demonstrated that the Tat protein competed with CIITA for binding to P-TEFb, an activation factor that is required for class II transcription, and blocked the expression of class II genes in THP-1 cells. Tat inhibition of the binding of CIITA to P-TEFb could be occurring to impair class II mRNA transcription. However, because there was selective loss of HLA-DR in the 43_HIV cells and in the HIV-1_BaL-infected monocytes, it is uncertain how much of a role this may be playing in our system. It is possible that HIV-1 infection may have multiple effects on the regulation of class II expression in different subpopulations of monocytic cells.

We have previously reported that there was defective APC function in HIV-1-infected human macrophage hybridomas that expressed HLA-DR after transfection with full-length HLA-DR cDNA driven by a CMV promoter (25). Furthermore, in these cells there was no Ag-class II complex formation, implying dysregulation of intracellular sorting (25). Similarly, in this study, the 43_HIV cells were incapable of inducing a T cell proliferative response after HLA-DR expression was restored following transfection with NF-YA (Fig. 4), even though there was no change in CD80 and CD86 expression (Fig. 5). Despite the fact we detected mRNA for Ii in the 43_HIV cells (Fig. 6), there were no Ii breakdown products (CLIP and LIP) (Fig. 7) and HLA-DR failed to localize in either the late endosomes, lysosomes, or acidic compartments (Fig. 8). Infection of the 43 cells with vaccinia virus expressing different HIV-1 proteins demonstrated that only the env protein prevented localization of HLA-DR in either the late endosomes or lysosomes (Fig. 10). In human cells, two forms of Ii target HLA-DR molecules to endosomal processing compartments, a 33-kDa isoform and a 35-kDa isoform (87). The 33-kDa Ii isotype constitutes 80% of the Ii pool and traffics to the Ag-processing compartments via the cell surface, while the 35-kDa Ii targets endosomal compartments by a strictly intracellular route (85). Similar to the Ii 33-kDa isoform, the env protein is also synthesized in the endoplasmic reticulum, transported to the cell surface, and then endocytosed, where it enters into the processing compartments (88, 89, 90). The envelope protein contains motifs for tyrosine-based sorting signals that are present in many endocytic receptors.

It is not entirely clear how the HIV-1 env protein could inhibit Ii targeting in our system. The leucine motif present in the cytoplasmic domain of both the 33-kDa and 35-kDa isoforms targets the Ag-processing compartments (91). Proteins like the 33-kDa isoform of Ii and HIV-1 env protein that are internalized from the cell surface before delivery into the late endosomes or lysosomes cluster into regions on the cell membrane that are underlain with the coat protein clathrin (92, 93, 94). After internalization, an adapter complex is formed (95). One well-characterized adapter complex, AP-2, mediates the association of clathrin with the plasma membrane (95). The AP-2 complex consists of four protein subunits 2∼100-kDa large chains (α-adapter and either β-2 or β-1 adaptor), a 50-kDa medium chain, μ2 and a 17-kDa small ς2 protein (96). Early experiments indicate that the cytosolic domains of internalized proteins could be bound with low affinity by the AP-2 adapter complex (97, 98). Evidence from several laboratories has established that the AP-2 adapter complex recognizes both tyrosine-based and di-leucine-based sorting signals (99, 100, 101). Ohno et al. (103) have demonstrated that the HIV-1 env protein can bind to members of the AP-2 adapter complex, and overexpression of the env protein saturates intracellular sorting into different subcellular compartments (96). Competition between env protein and Ii may inhibit the localization of HLA-DR to the late endosomes, lysosomes, and acidic compartments in the 43_HIV cells and in the 43 cells infected with vaccinia virus expressing the env protein.

In conclusion, we have demonstrated that the loss of HLA-DR expression after HIV-1 infection is associated with the loss of the NF-YA Y box-binding protein in 43_HIV cells and primary HIV-1_BaL-infected monocytes that lost HLA-DR expression. Despite the fact that the block in HLA-DR production could be overcome by transfection with NF-YA driven by a CMV promoter, the cells were still incapable of inducing Ag-specific T cell proliferation. Even though there was Ii mRNA in the HIV-1-infected human macrophage hybridomas, HLA-DR failed to localize in the late endosomes, lysosomes, or acidic compartments. The HIV-1 env protein appears to interfere with the trafficking of HLA-DR into late endosomes, lysosomes, and acidic compartments in the HIV-1-infected human macrophage hybridomas. HIV-1 infection has multiple effects on Ag processing in monocytic cells, which may contribute to the immune defects seen in HIV-1-infected patients.