Human Cytomegalovirus-Encoded Immune ... - Journal of Virology

JOURNAL OF VIROLOGY, Feb. 2009, p. 1359–1367 0022-538X/09/$08.00⫹0 doi:10.1128/JVI.01324-08 Copyright © 2009, American Society for Microbiology. All Rights Reserved.

Vol. 83, No. 3

Human Cytomegalovirus-Encoded Immune Modulators Partner To Downregulate Major Histocompatibility Complex Class I Molecules䌤† Vanessa M. Noriega and Domenico Tortorella* Department of Microbiology, Mount Sinai School of Medicine, New York, New York 10029 Received 24 June 2008/Accepted 4 November 2008

Throughout the course of natural evolution with its host, the human cytomegalovirus (HCMV) has developed a variety of strategies to avoid immune recognition and clearance. The major histocompatibility complex (MHC) class I antigen presentation pathway is a major target of the virus. HCMV encodes at least six gene products that modulate the processing of endoplasmic reticulum (ER)-resident MHC class I molecules. Here, we show that two virus-encoded proteins, US2 and US3, coordinate their functions toward the common goal of attenuating class I protein surface expression. In cells stably expressing both US2 and US3, class I molecules were almost completely downregulated from the cell surface. In addition, pulse-chase analysis revealed that the proteasome-dependent turnover of class I molecules occurs more rapidly in cells expressing both US2 and US3 than either US2 or US3 alone. The ability of US3 to retain class I molecules in the ER produces a target-rich environment for US2 to mediate the destruction of class I heavy chains. In fact, expression of US3 enhanced the association between US2 and class I molecules, thus encouraging their dislocation and degradation. This immune evasion strategy ensures that viral antigens are not presented on the cell surface during the early phase of HCMV infection, a critical time of replication and viral proliferation. US10 encodes a gene product that delays class I protein complex trafficking (5). The US2 and US11 proteins exploit the cellular process known as ER quality control to target class I heavy chains for proteasome degradation (12, 25). Through inhibition of antigenic peptide presentation, HCMV can prevent immune detection and clearance. HCMV gene expression occurs in a tightly regulated cascade of immediate early, early, and late phases of replication (20). The immediate early transcription of US3 occurs between 2 and 8 h postinfection (1), while US2 expression begins at about 6 h postinfection during the early phase of replication (12). The appearance of both viral proteins coincides with the rapid destabilization of class I heavy chains in the infected cell (11). Coexpression of US2 and US3 leads to decreased surface class I protein and increased turnover of newly synthesized class I heavy chains. US3 retains class I molecules in the ER as targets for US2-mediated degradation and, furthermore, facilitates their interaction. The data presented here demonstrate a novel relationship between two immune modulators working collaboratively to promote viral subterfuge.

The human immune system bears the enormous task of coordinating protective responses against a wide variety of pathogens. While the innate immune branch offers immediate, nonspecific responses to microorganisms (10), the adaptive branch is based on the creation of unlimited variability of immune receptors and clonal expansion of pathogen-specific cells (19). However, infectious agents including bacteria and viruses successfully circumvent immune recognition through multiple routes, including secretion of compounds that diminish the host immune response (e.g., interferon antagonists), antigenic variation, and inhibition of lymphocyte activation pathways (4). Under pressure from the immune system, pathogens have evolved elaborate strategies to subvert suppressive responses. Evolving alongside its host for millions of years, the human cytomegalovirus (HCMV) has committed a large percentage of its genome toward modulation of the cellular response to infection (21). HCMV manipulates the host environment to facilitate efficient infection and, ultimately, lifelong persistence. The HCMV unique short (US) genomic region encodes at least five glycoproteins that modulate major histocompatibility complex (MHC) class I molecule surface expression, thereby hindering antigenic presentation to cytotoxic T lymphocytes (CTL). The viral US3 glycoprotein binds and retains tapasindependent class I molecules within the endoplasmic reticulum (ER), preventing their egress to the cell surface (24). The US6 gene product prevents translocation of antigenic peptide by the transporter associated with antigen presentation (TAP) (8).

MATERIALS AND METHODS Cells, antibodies, and cDNA constructs. Human U373-MG astrocytoma cells, U373-MG transfectants (see below), Gp2-293 cells, and normal human dermal fibroblasts (Cambrex) were maintained in Dulbecco’s modified Eagle’s medium as described previously (23). Rabbit polyclonal anti-US2 antibody and anti-class I heavy chain antibody were generated as described previously (23). Rabbit polyclonal anti-US3 antibody was a gift from H. Ploegh (Massachusetts Institute of Technology). US2 (pMIg), US3 (pLpCX; Clontech), ICP47 amended with a COOH-terminal hemagglutinin (HA) tag (pLgPW), adenovirus E3/19K cDNA (kind gifts from M. Bouvier [University of Illinois, Chicago, IL] and W. Wold [Univeristy of St. Louis, St. Louis, MO]), and US2/CD4/US2 (pLgPW) were introduced into U373-MG cells using retrovirus transduction (23). Flow cytometry analysis and immunoprecipitation. Cells lysis and immunoprecipitation were carried out as described previously (23). Proteasome inhibitor (carboxylbenzyl-leucyl-leucyl-leucine vinyl sulfone [ZL3VS]) was a gift from H.

* Corresponding author. Mailing address: One Gustave L. Levy Place, Box 1124, New York, NY 10029. Phone: (212) 241-5447. Fax: (212) 241-7336. E-mail: [email protected]. † Supplemental material for this article may be found at http://jvi .asm.org/. 䌤 Published ahead of print on 12 November 2008. 1359

1360

NORIEGA AND TORTORELLA

J. VIROL.

FIG. 1. HCMV US2/US3 expression alters cell surface class I molecules. (A) Total cell lysates from U373-MG (U373), U373-MGUS2 (US2), U373-MGUS3 (US3), and U373-MGUS2/US3 (US2/US3) cells were subjected to immunoblot analysis using anti-US2 (lanes 1 to 4), anti-US3 (lanes 5 to 8), and anti-GAPDH (lanes 9 to 12) antibodies. US2, US3, GAPDH, and molecular size standards are indicated. (B) U373-MG, U373-MGUS2, U373-MGUS3, and U373-MGUS2/US3 cells were analyzed by flow cytometry using W6/32 monoclonal antibody followed by an anti-mouse immunoglobulin G (IgG)-Alexa 647. Surface class I protein plots are represented by normalized cell number versus fluorescence signal.

Ploegh (Massachusetts Institute of Technology). Samples were resolved using sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and subjected to immunoblot analysis. Quantitative flow cytometry analysis of surface MHC class I molecules was assessed as described previously (23) using a Cytomics FC 500 flow cytometer. Plots of surface class I proteins are represented by normalized cell number versus fluorescence signal. Pulse-chase analysis. Pulse-chase analysis was performed as described previously (25). In brief, cells metabolically pulsed with [35S]Met for 15 min and chased up to 30 or 40 min were lysed and incubated with the appropriate antibody. Precipitates were resolved using SDS-PAGE, and the polyacrylamide gel was exposed to autoradiography film.

RESULTS U373-MG astrocytomas support expression of HCMV US2 and US3. The immediate-early and early phase of HCMV infection generates viral antigens that elicit a potent CTL response to the lower matrix protein pp65 and major immediateearly protein IE1 (14, 27). Accordingly, the immune modulators US2 and US3 are some of the first gene products to be

visualized in the infected cell, with US3 appearing between 2 and 6 h after infection (1). US2 expression is detected during the early phase of HCMV replication, at approximately 4 h postinfection (12). Hence, there is a time when both proteins are present within the ER. To define the cooperation that may exist between US2 and US3 toward class I molecule downregulation, U373-MG cells that stably express US2, US3, or both US2 and US3 (U373-MGUS2, U373-MGUS3, and U373MGUS2/US3, respectively) were generated. U373-MG cells have been extensively utilized to characterize US2 and US3 function (12, 13, 26). Protein expression was confirmed when total cell lysates of U373-MG, U373-MGUS2, U373-MGUS3, and U373MGUS2/US3 cells were subjected to immunoblot analysis (Fig. 1A, lanes 1 to 8). Anti-GAPDH (glyceraldehyde-3-phosphate dehydrogenase) immunoblot analysis confirmed equal protein loading (Fig. 1A, lanes 9 to 12). Notably, US2 levels in US2/ US3-expressing cells were lower than in US2-expressing cells,

VOL. 83, 2009

HCMV US2 AND US3 COLLABORATE TO DEGRADE MHC CLASS I

1361

FIG. 2. US2/US3 cells demonstrate decreased levels of class I molecules. (A) W6/32 precipitates (lanes 1 to 4) and cell lysates (lanes 5 to 8) from U373-MG (U373), U373-MGUS2 (US2), U373-MGUS3 (US3), and U373-MGUS2/US3 (US2/US3) cells were subjected to an anti-class I heavy chain immunoblot. The asterisks indicate nonspecific polypeptides. (B) Class I molecules recovered from U373-MG, U373-MGUS2, U373-MGUS3, U373-MGUS11 (US11), and U373-MGUS2/US3 cells using W6/32 were left untreated or treated with EndoH and then subjected to an anti-class I heavy chain immunoblot. The arrows indicate a nonspecific polypeptide in EndoH-treated samples. Class I heavy chains (HCs), glycosylated [HC(⫹)CHO] and deglycoslyated [HC(⫺)CHO] class I molecules, and molecular standards are indicated.

while US3 protein levels were the same in both cell lines. In summary, cell lines were generated to study the impact of US2/US3 on downregulation of class I molecules. HCMV US2 and US3 robustly downregulate surface class I molecules. Flow cytometry was utilized to examine surface class I molecules in cells expressing US2/US3 (Fig. 1B). U373MG, U373-MGUS2, U373-MGUS3, and U373-MGUS2/US3 cells were incubated with a monoclonal antibody recognizing the properly folded class I molecules (W6/32). U373-MG cells were used as a negative control for surface class I protein downregulation (Fig. 1B, solid black line), while an immunoglobulin isotype control was used to determine background fluorescence (Fig. 1B). A significant reduction (⬃80%) in surface class I molecules was observed in U373-MGUS2 cells (Fig. 1B, left panel). In comparison, U373-MGUS3 cells demonstrated similar levels of surface class I molecules as U373-MG cells (Fig. 1B, center panel), likely due to the transient retention of class I molecules and the half-life of surface class I molecules (7). Interestingly, HeLa cells (16) and normal human dermal fibroblasts (see Fig. S1 in the supplemental material) expressing US3 revealed a slight decrease in surface expression of class I molecules, highlighting the possibility that the effectiveness of US3 may slightly vary in different cell lines. Therefore, the U373-MG cells provide a good model system to examine the contribution of multiple US gene products that downregulate class I molecules. Strikingly, in U373-MGUS2/US3 cells, levels of class I molecules were considerably lower than in U373-MGUS2 or U373-MGUS3 cells (Fig. 1B, right panel). Presentation of surface class I molecules in U373-MGUS2/US3 cells was reduced by approximately 97%. These results demonstrate that US2 and US3 cooperate to downregulate surface class I molecules.

Class I molecules are efficiently degraded in cells expressing US2/US3. We next investigated the stability of class I molecules in U373-MGUS2/US3 cells. Total cell lysates and W6/32 precipitates from U373-MG, U373-MGUS2, U373-MGUS3, and U373-MGUS2/US3 cells were subjected to immunoblot analysis (Fig. 2A). The class I heavy chains from U373 cells were used as a positive control (Fig. 2A, lanes 1 and 5). Levels of class I molecules were reduced in U373-MGUS2 cells (Fig. 2A, lanes 2 and 6), while increased amounts of class I heavy chain were observed in U373-MGUS3 cells (Fig. 2A, lanes 3 and 7). The US3 molecule’s ability to retain class I molecules in the ER delays class I protein turnover (1). Strikingly, in U373-MGUS2/US3 cells, practically no class I molecules were recovered (Fig. 2A, lanes 4 and 8). These results suggest a synergistic effect imparted on degradation of class I molecules by expression of both US2 and US3. To examine the ER residency of class I molecules in cells expressing US2/US3, class I molecules were subjected to endoglycosidase H (EndoH) digestion. ER-resident forms of class I molecules are susceptible to cleavage by EndoH (25). W6/32 precipitates from U373-MG, U373-MGUS2, U373MGUS3, U373-MGUS11, and U373-MGUS2/US3 cells were treated with EndoH and subjected to immunoblot analysis (Fig. 2B). Steady-state levels of class I molecules from U373-MG cells were resistant to EndoH cleavage as these molecules have exited the ER (Fig. 2B, lanes 1 and 2). The small pool of EndoH-resistant class I molecules from U373MGUS2 cells escaped degradation and also exited the ER (Fig. 2B, lanes 3 and 4). On the other hand, class I molecules from U373-MGUS3 cells were partially retained within the ER, as observed by the faster-migrating species upon digestion (Fig. 2B, lanes 5⬘ and 6⬘). A lighter exposure demonstrates EndoH-

1362


J. VIROL.

FIG. 3. Enhanced class I heavy chain turnover in U373-MGUS2/US3 cells. (A) U373-MG (U373), U373-MGUS2 (US2), U373-MGUS3 (US3), and U373-MGUS2/US3 (US2/US3) cells were labeled with [35S]methionine for 15 min and chased for up to 30 min. Class I heavy chains, US2, and US3 proteins recovered from SDS lysates were resolved by SDS-PAGE and exposed to autoradiographic film. Class I heavy chains (HCs), US2, US3, and molecular standards are indicated. (B) Class I protein levels were quantified by densitometry by using the 0-min chase point as 100%.

sensitive class I molecules in U373-MG and U373-MGUS3 cells (Fig. 2B, lanes 1⬘ and 2⬘ and lanes 5⬘ and 6⬘). Immunoprecipitates from U373-MGUS11 cells revealed nonspecific proteins visible in EndoH-treated samples (Fig. 2B, lanes 7 and 8). Interestingly, in U373-MGUS2/US3 cells, not only were reduced amounts of class I molecules recovered compared to U373MGUS2 cells, but also the remaining proteins were sensitive to EndoH (Fig. 2B, lanes 9 and 10). These results suggest that US3 retains class I molecules within the ER for US2-mediated degradation. US2/US3 increase the degradation rate of class I molecules. To further define the impact of US3 upon US2-mediated downregulation, degradation kinetics of class I molecules were examined (Fig. 3A). U373-MG, U373-MGUS2, U373-MGUS3, and U373-MGUS2/US3 cells were labeled with [35S]methionine for 15 min and chased up to 30 min. Class I heavy chains from control U373-MG cells were stable over the chase period (Fig. 3A, lanes 1 to 3). As expected, the quantity of class I heavy chains from cells expressing US2 significantly decreased during the chase (Fig. 3A, lanes 4 to 6). Some class I heavy chains escaped degradation, as observed by the mature, slower-migrating proteins during the chase (Fig. 3A, lanes 5 and 6). In U373-MGUS3 cells, class I protein levels slightly increased over the chase period (Fig. 3A, lanes 10 to 12), indicative of ER retention. Unexpectedly, in U373-MGUS2/US3 cells, increased amounts of class I heavy chains were recovered from the 0-min chase point compared to U373-MGUS2 cells (Fig. 3A, compare

lanes 4 and 7). However, the levels of class I molecules dramatically diminished during the chase (Fig. 3A, compare lanes 6 and 9, and B). Accelerated degradation of class I molecules occurred in U373-MGUS2/US3 cells despite lower levels of US2 protein (Fig. 3A, lanes 16 to 21). These results demonstrate that the degradation kinetics of class I heavy chains are increased in the presence of US2 and US3. Inhibition of proteasome function in cells expressing US2 causes the accumulation of both glycosylated and deglycoslyated, dislocated forms of heavy chains (23). Does the shorter half-life of heavy chains in cells expressing US2/US3 translate to an increase in deglycoslyated intermediates? To address this question, class I molecules were recovered from U373-MG, U373-MGUS2, U373MGUS3, and U373-MGUS2/US3 cells left untreated or treated with the proteasome inhibitor ZL3VS and subjected to immunoblot analysis (Fig. 4A). Proteasome inhibition resulted in the accumulation of glycosylated and deglycoslyated species of class I molecules in U373-MGUS2 cells (Fig. 4A, lane 4). Note that the glycosylated species is comprised of a population of class I molecules with modifications of its N-linked glycan and that they have exited the ER. The small quantity of deglycoslyated class I molecules from inhibitor-treated U373-MGUS3 cells may be due to the combined effects of ER retention and a decrease in the peptide pool (Fig. 4A, lane 6). Interestingly, increased amounts of deglycoslyated intermediates and decreased amounts of glycosylated class I molecules were recovered from ZL3VS-treated U373-MGUS2/US3 cells compared to

VOL. 83, 2009


1363

FIG. 4. Increased recovery of class I protein degradation intermediates from U373-MGUS2/US3. (A) Class I molecules recovered from U373-MG (U373), U373-MGUS2 (US2), U373-MGUS3 (US3), and U373-MGUS2/US3 (US2/US3)cells, either untreated or treated with ZL3VS (2.5 ␮M for 16 h), using anti-class I heavy chain serum (HC) were subjected to an anti-class I heavy chain immunoblot. Class I heavy chains, US2, US3, and molecular standards are indicated. (B) Class I protein levels were quantified by densitometry. Ig, immunoglobulin; HC(⫹)CHO, glycosylated class I heavy chain; HC(⫺)CHO, deglycoslyated class I heavy chain.

inhibitor-treated U373-MGUS2 cells (Fig. 4A, compare lanes 4 and 8, and B). The results suggest that class I molecules from U373-MGUS2/US3 cells are rapidly extracted from the ER. Retention of class I molecules in the ER does not mimic US2/US3-mediated class I protein degradation. Does the retention of class I molecules by US3 specifically contribute to the robust US2-mediated degradation of class I molecules? To address this question, we examined the US2-mediated class I protein degradation in the presence of different viral proteins that retain class I molecules within the ER (herpes simplex virus ICP47 and adenovirus E3/19K). To that end herpes simplex virus type 1 ICP47 protein was stably introduced into U373-MG and US2-expressing cells (see Fig. S2A in the supplemental material). ICP47 binds to the cytosolic face of TAP, preventing peptide translocation and indirectly impeding the egress of class I molecules (9). Flow cytometry experiments were performed to examine the impact of ICP47 expression on surface class I molecules (Fig. 5A). U373-MG cells expressing HA-tagged ICP47 (U373-MGICP47-HA) demonstrated a reduction (⬃67%) in surface class I molecules (Fig. 5A, top panel). The class I molecules are likely retained in the ER as a properly folded dimer (see Fig. S2A, lane 2, in the supplemental material). Interestingly, in U373-MG cells expressing US2 plus ICP47-HA (U373-MGUS2/ICP47-HA), surface class I molecules were also reduced by approximately 67% (Fig. 5A, bottom panel). The comparable levels of reduction in class I molecules in U373-MGICP47-HA and U373-MGUS2/ICP47-HA cells suggested that ICP47 imparted a downregulation that was distinct from US2. Hence, ICP47 probably targets a different pool of class I molecules (with or without peptide) than US2, suggesting that these two viral proteins function in an additive manner. To determine if class I molecules from U373MGUS2/ICP47-HA cells were dislocated with increased kinetics, levels of class I molecules from U373-MG, U373MGICP47-HA, U373-MGUS2, and U373-MGUS2/ICP47-HA cells, either untreated or treated with ZL3VS, were subjected to immunoblot analysis (Fig. 5B). In order to recover

all forms of class I molecules (both glycosylated and deglycoslyated polypeptides), class I heavy chains were immunoprecipitated from SDS-denatured lysates using a polyclonal anti-heavy chain serum. Decreased levels of class I proteins were recovered from U373-MGICP47-HA cells (versus U373-MG cells) both in the presence and absence of proteasome function (Fig. 5B, lanes 1 to 4). The absence of increased glycosylated class I molecules from inhibitortreated U373-MGICP47-HA cells, as well as a lack of dislocated intermediates (Fig. 5B, lane 4), suggested an ICP47mediated proteasome-independent turnover of class I heavy chains. A similar reduction in class I protein levels was observed in U373-MGUS2/ICP47-HA cells compared to U373MGUS2 cells (Fig. 5B, lanes 5 to 8). As expected, ICP47 expression did not induce increased amounts of deglycoslyated class I molecules (Fig. 5B, lanes 6 versus 8). A similar reduction was observed for properly folded class I molecules (W6/32 reactive) from U373-MGICP47-HA and U373-MGUS2/ICP47-HA cells (see Fig. S2B in the supplemental material). The data suggest that US2 and US3 specifically collaborate to enhance class I heavy chain degradation. To further address the specificity of US2/US3 collaboration, adenovirus-2 E3/19K glycoprotein was stably introduced into U373-MG (U373-MGE3/19K) and U373-MGUS2 (U373MGUS2-E3/19K) cells. The E3/19K gene product binds to MHC class I proteins within the ER and attenuates their egress to the cell surface (2, 15). As was observed in cells expressing US3, flow cytometry revealed that there was no reduction in surface class I proteins in U373-MGE3/19K cells (Fig. 5C, top panel), a finding confirmed by immunoblot analysis (Fig. 5D, lanes 1 to 4). Interestingly, there was no appreciable reduction of surface class I protein levels in U373-MGUS2-E3/19K cells (Fig. 5C, bottom panel). Consistent with the ICP47 results, E3/19K expression did not produce an increase in dislocated, deglycoslyated class I protein intermediates in U373-MGUS2-E3/19K cells (Fig. 5D, lanes 5 to 8), providing further evidence of the particularity of US2/US3 cooperation. In addition, experiments performed using brefeldin A, a drug that impedes the movement of secretory proteins from the ER (22) and most

1364


J. VIROL.

FIG. 5. Nonspecific class I protein retention does not increase US2-mediated degradation. (A) U373-MG (U373), U373-MGICP47-HA, U373MGUS2 and U373-MGUS2/ICP47-HA cells were subjected to flow cytometry using W6/32 antibody. The percentage of surface class I molecules from two independent experiments was calculated using the mean surface class I protein signal from ICP47-HA-expressing cell lines versus parental cells. (B) Class I molecules from U373-MG, U373-MGICP47-HA, U373-MGUS2 (US2), and U373-MGUS2/ICP47-HA (US2ICP47-HA) cells, either untreated or treated with the proteasome inhibitor ZL3VS (2.5 ␮M for 16 h), were recovered using anti-class I heavy chain serum and subjected to an anti-class I heavy chain immunoblot. (C) U373-MG, U373-MGE3/19K, U373-MGUS2, and U373-MGUS2-E3/19K (US2E3/19K) cells were subjected to flow cytometry using W6/32 antibody. The percentage of surface class I molecules from two independent experiments was calculated using the mean surface class I protein signal from E3/19K-expressing cell lines versus parental cells. (D) Total cell lysates from U373-MG, U373-MGE3/19K, U373-MGUS2, and U373-MGUS2-E3/19K cells, either untreated or treated with ZL3VS (2.5 ␮M for 16 h), were subjected to immunoblot analysis using anti-class I heavy chain (lanes 1 to 8) and anti-GAPDH (lanes 9 to 16) antibodies. Class I heavy chains, GAPDH, and molecular standards are indicated. IgG, immunoglobulin G; HC(⫹)CHO, glycosylated class I heavy chain; HC(⫺)CHO, deglycosylated class I heavy chain.

VOL. 83, 2009


1365

FIG. 5—Continued.

closely mimics US3 retention, did not augment class I protein degradation in U373-MGUS2 cells (see Fig. S2 in the supplemental material). Collectively, the data suggest that US2 and US3 proteins specifically enhance class I destruction. US3 prolongs the association of a class I molecule/US2 complex. Conceivably, US3 may augment US2-mediated class I protein degradation by enhancing the association between class I molecules and US2. Class I molecules were

virtually completely downregulated in US2-expressing cells (Fig. 2A, lanes 2 and 6), making it difficult to recover substantial amounts of heavy chain molecules or to visualize an association between class I molecules and US2. Therefore, to determine whether US3 induces a stable US2/class I protein complex, we examined the interaction between class I molecules and a chimeric US2 mutant (US2/CD4/US2) that fails to mediate class I protein degradation (23). Ex-

1366


J. VIROL.

FIG. 6. US3 protein enhances the association of US2 and class I molecules. (A) W6/32 precipitates from U373-MG (U373), U373MGUS2/CD4/US2 (US2/CD4), U373-MGUS3 (US3), and U373-MGUS3-US2/CD4/US2 (US3-US2/CD4) cells were subjected to sequential immunoblot analysis using anti-class I heavy chain and anti-US2 serum (lanes 1 to 4). (B) U373-MG, U373-MGUS2/CD4/US2, U373-MGUS3, and U373-MGUS3-US2/CD4/US2 cells were metabolically labeled with [35S]methionine for 15 min and chased up to 40 min. Samples were lysed in 0.5% NP-40, and class I molecules were recovered using W6/32 (lanes 1 to 12). Class I heavy chains, US2, GAPDH, a nonspecific polypeptide (ⴱ), and molecular standards are indicated.

pression of US3 and US2/CD4/US2 was confirmed by immunoblot analysis (data not shown). Precipitates of class I molecule from U373-MG, U373-MG cells expressing US2/ CD4/US2 (U373-MGUS2/CD4/US2), U373-MGUS3, and U373MGUS3-US2/CD4/US2 cells were subjected to immunoblot analysis (Fig. 6A, lanes 1 to 4). Only small amounts of US2/CD4/US2 were able to associate with class I molecules (Fig. 6A, lane 2), while in the presence of US3, the chimeric US2 mutant robustly bound to class I molecules (Fig. 6A, lane 4). The results suggest that US3 enhances a US2/class I protein interaction. We next examined the ability of US2/CD4/US2 to associate with newly synthesized class I molecules in the presence of US3 (Fig. 6B). Class I heavy chains were recovered from U373-MG, U373-MGUS2/CD4/US2, U373-MGUS3, and U373MGUS3-US2/CD4/US2 cells pulsed with [35S]methionine for 15 min and chased up to 40 min. The US2/CD4/US2 mutant coprecipitated with class I molecules at the 0-min chase point, but this association quickly declined (Fig. 6B, lanes 4 to 6). In contrast, US3 prolonged the association of US2/ CD4/US2 with class I molecules during the chase period (Fig. 6B, lanes 7 to 9). As expected, two class I polypeptides were resolved in US3-expressing cells due to the delayed modifications of the N-linked glycan. These results demonstrate that the presence of US3 facilitates the association of

US2 and class I heavy chains, thereby enhancing US2-mediated class I degradation. DISCUSSION Immune evasion is a hallmark of HCMV. HCMV-infected cells exhibit reduced surface class I molecules due to several independent loci within the US region of the genome (11). The US2 and US11 glycoproteins catalyze the extraction of the class I heavy chain component from the ER for proteasome degradation (12, 25). Other loci, such as US3 and US6, target accessory proteins of class I antigen presentation (8, 24). The temporal expression of US proteins suggests that these gene products may function together to downregulate class I molecules. We show that the virus-encoded immune modulators US2 and US3 coordinate their functions to limit surface expression of class I molecules. The emulation of US2 and US3 coexpression in vitro demonstrated dramatically diminished surface levels of class I molecules (Fig. 1) due to the accelerated turnover of newly synthesized class I protein (Fig. 3). Our data suggest a paradigm where US2 and US3 coordinate their activities during early time points of infection to prevent the presentation of viral antigens. During normal class I protein processing, the heavy chain and ␤2-microglobulin assemble within the ER. The class I

VOL. 83, 2009


heterodimer then engages the peptide loading complex, composed of the accessory proteins tapasin and TAP. Tapasin facilitates loading of the antigenic peptide onto the class I molecule, and the trimeric class I complex is competent to egress toward the cell surface. In an HCMV-infected cell, the US3 glycoprotein retains the class I molecule through a transient interaction for several hours until US2 expression (1). The class I molecule is then targeted by US2 for extraction from the ER membrane and subsequent degradation by the proteasome. Early US2 and US3 coordination creates an environment where few or no class I molecules traffic out of the ER, thereby preventing the presentation of HCMV-derived antigenic peptides. Although the functional significance of US2/US3 collaboration on the immune response remains to be seen, our data correlate with the notion that the virus employs extreme measures to inhibit early presentation of pp65 and IE1-derived peptides, which dominate the CTL response to HCMV (14, 27). Association experiments performed with US2 chimeras further define the mechanism by which these two viral proteins collaborate (Fig. 6). US3 facilitates a tighter binding between class I molecules and US2, thereby guaranteeing that this immune modulator encounters and disposes of its target. The presence of US3 allows for a tighter and prolonged association of class I molecules with US2. Alternately, US3 binding to class I molecules may create a more accessible binding site for US2, thereby causing US2 to efficiently eliminate class I molecules. Both paradigms suggest that US3 retention of class I molecules generates a larger pool of potential substrates for US2-mediated class I protein degradation. Of most interest is the idea that US3 may augment the allelic specificity of US2. The decreased levels of surface class I molecules in U373-MGUS2/US3 cells implies that in the presence of US3, various alleles of class I protein could become substrates for US2-mediated degradation. This is in agreement with previous findings describing allele-specific differences for several HCMV US gene products (6, 18, 24). By coordinating their activities, US2 and US3 would allow the virus more effective coverage of a broad range of class I molecules. Of course, this could leave the infected cell susceptible to attack by NK (natural killer) cells (3). This is circumvented by the brief time of US3 transcription and by expression of inhibitors of NK cell activation (i.e., UL16 and UL40) (17). By engaging several immune evasion genes at multiple times during its replication cycle, the virus cleverly masks itself from lymphocyte attack while maintaining the appearance of normalcy at the cell surface.

3. 4. 5. 6.

7. 8. 9. 10. 11.

12. 13.

14.

15. 16.

17. 18.

19. 20. 21. 22. 23. 24.

ACKNOWLEDGMENTS We thank L. Tortorella for critical evaluation and suggestions. This study was supported by NIH grants AI060905 and U19 AI62623. D.T. is partially supported by the Irma T. Hirschl Trust. V.M.N. is a predoctoral trainee supported by NIH grant AI060905-SI. REFERENCES 1. Ahn, K., A. Angulo, P. Ghazal, P. A. Peterson, Y. Yang, and K. Fruh. 1996. Human cytomegalovirus inhibits antigen presentation by a sequential multistep process. Proc. Natl. Acad. Sci. USA 93:10990–10995. 2. Bennett, E. M., J. R. Bennink, J. W. Yewdell, and F. M. Brodsky. 1999.

25. 26.

27.

1367

Cutting edge: adenovirus E19 has two mechanisms for affecting class I MHC expression. J. Immunol. 162:5049–5052. Carayannopoulos, L. N., and W. M. Yokoyama. 2004. Recognition of infected cells by natural killer cells. Curr. Opin. Immunol. 16:26–33. Finlay, B. B., and G. McFadden. 2006. Anti-immunology: evasion of the host immune system by bacterial and viral pathogens. Cell 124:767–782. Furman, M. H., N. Dey, D. Tortorella, and H. L. Ploegh. 2002. The human cytomegalovirus US10 gene product delays trafficking of major histocompatibility complex class I molecules. J. Virol. 76:11753–11756. Gewurz, B. E., R. Gaudet, D. Tortorella, E. W. Wang, H. L. Ploegh, and D. C. Wiley. 2001. Antigen presentation subverted: structure of the human cytomegalovirus protein US2 bound to the class I molecule HLA-A2. Proc. Natl. Acad. Sci. USA 98:6794–6799. Gruhler, A., P. A. Peterson, and K. Fruh. 2000. Human cytomegalovirus immediate early glycoprotein US3 retains MHC class I molecules by transient association. Traffic 1:318–325. Hewitt, E. W., S. S. Gupta, and P. J. Lehner. 2001. The human cytomegalovirus gene product US6 inhibits ATP binding by TAP. EMBO J. 20:387–396. Hill, A., P. Jugovic, I. York, G. Russ, J. Bennink, J. Yewdell, H. Ploegh, and D. Johnson. 1995. Herpes simplex virus turns off the TAP to evade host immunity. Nature 375:411–415. Janeway, C. 2005. Immunobiology: the immune system in health and disease, 6th ed. Garland Science, New York, NY. Jones, T. R., L. K. Hanson, L. Sun, J. S. Slater, R. M. Stenberg, and A. E. Campbell. 1995. Multiple independent loci within the human cytomegalovirus unique short region down-regulate expression of major histocompatibility complex class I heavy chains. J. Virol. 69:4830–4841. Jones, T. R., and L. Sun. 1997. Human cytomegalovirus US2 destabilizes major histocompatibility complex class I heavy chains. J. Virol. 71:2970– 2979. Jones, T. R., E. J. Wiertz, L. Sun, K. N. Fish, J. A. Nelson, and H. L. Ploegh. 1996. Human cytomegalovirus US3 impairs transport and maturation of major histocompatibility complex class I heavy chains. Proc. Natl. Acad. Sci. USA 93:11327–11333. Kern, F., I. P. Surel, N. Faulhaber, C. Frommel, J. Schneider-Mergener, C. Schonemann, P. Reinke, and H. D. Volk. 1999. Target structures of the CD8⫹-T-cell response to human cytomegalovirus: the 72-kilodalton major immediate-early protein revisited. J. Virol. 73:8179–8184. Korner, H., and H. G. Burgert. 1994. Down-regulation of HLA antigens by the adenovirus type 2 E3/19K protein in a T-lymphoma cell line. J. Virol. 68:1442–1448. Lee, S., J. Yoon, B. Park, Y. Jun, M. Jin, H. C. Sung, I. H. Kim, S. Kang, E. J. Choi, B. Y. Ahn, and K. Ahn. 2000. Structural and functional dissection of human cytomegalovirus US3 in binding major histocompatibility complex class I molecules. J. Virol. 74:11262–11269. Lopez-Botet, M., A. Angulo, and M. Guma. 2004. Natural killer cell receptors for major histocompatibility complex class I and related molecules in cytomegalovirus infection. Tissue Antigens 63:195–203. Machold, R. P., E. J. Wiertz, T. R. Jones, and H. L. Ploegh. 1997. The HCMV gene products US11 and US2 differ in their ability to attack allelic forms of murine major histocompatibility complex (MHC) class I heavy chains. J. Exp. Med. 185:363–366. Medzhitov, R. 2007. Recognition of microorganisms and activation of the immune response. Nature 449:819–826. Mocarski, E. S., Jr. 1996. Cytomegaloviruses and their replication, p. 2447– 2492. In B. N. Fields, D. M. Knipe, and P. M. Howley (ed.), Fields virology, 3rd ed. Lippincott-Raven, Philadelphia, PA. Mocarski, E. S., Jr. 2002. Immunomodulation by cytomegaloviruses: manipulative strategies beyond evasion. Trends Microbiol. 10:332–339. Nebenfuhr, A., C. Ritzenthaler, and D. G. Robinson. 2002. Brefeldin A: deciphering an enigmatic inhibitor of secretion. Plant Physiol. 130:1102– 1108. Noriega, V. M., and D. Tortorella. 2008. A bipartite trigger for dislocation directs the proteasomal degradation of an endoplasmic reticulum membrane glycoprotein. J. Biol. Chem. 283:4031–4043. Park, B., Y. Kim, J. Shin, S. Lee, K. Cho, K. Fruh, and K. Ahn. 2004. Human cytomegalovirus inhibits tapasin-dependent peptide loading and optimization of the MHC class I peptide cargo for immune evasion. Immunity 20:71–85. Wiertz, E. J., T. R. Jones, L. Sun, M. Bogyo, H. J. Geuze, and H. L. Ploegh. 1996. The human cytomegalovirus US11 gene product dislocates MHC class I heavy chains from the endoplasmic reticulum to the cytosol. Cell 84:769–779. Wiertz, E. J., D. Tortorella, M. Bogyo, J. Yu, W. Mothes, T. R. Jones, T. A. Rapoport, and H. L. Ploegh. 1996. Sec61-mediated transfer of a membrane protein from the endoplasmic reticulum to the proteasome for destruction. Nature 384:432–438. Wills, M. R., A. J. Carmichael, K. Mynard, X. Jin, M. P. Weekes, B. Plachter, and J. G. Sissons. 1996. The human cytotoxic T-lymphocyte (CTL) response to cytomegalovirus is dominated by structural protein pp65: frequency, specificity, and T-cell receptor usage of pp65-specific CTL. J. Virol. 70:7569– 7579.