A Short Isoform of Human Cytomegalovirus US3 Functions as a Dominant Negative Inhibitor of the Full-Length Form
http://www.100md.com
《病菌学杂志》
Department of Biological Sciences, Center for Functional Cellulomics, Seoul National University, Seoul 151-747, Korea
College of Life Science and Biotechnology, Korea University, Seoul 136-704, Korea
Department of Biomedical Sciences, 228 Irvine Hall, Ohio University College of Osteopathic Medicine, Athens, Ohio 45701
ABSTRACT
Human cytomegalovirus encodes four unique short (US) region proteins, each of which is independently sufficient for causing the down-regulation of major histocompatibility complex (MHC) class I molecules on the cell surface. This down-regulation enables infected cells to evade recognition by cytotoxic T lymphocytes (CTLs) but makes them vulnerable to lysis by natural killer (NK) cells, which lyse those cells that lack MHC class I molecules. The 22-kDa US3 glycoprotein is able to down-regulate the surface expression of MHC class I molecules by dual mechanisms: direct endoplasmic reticulum retention by physical association and/or tapasin inhibition. The alternative splicing of the US3 gene generates two additional products, including 17-kDa and 3.5-kDa truncated isoforms; however, the functional significance of these isoforms during viral infection is unknown. Here, we describe a novel mode of self-regulation of US3 function that uses the endogenously produced truncated isoform. The truncated isoform itself neither binds to MHC class I molecules nor prevents the full-length US3 from interacting with MHC class I molecules. Instead, the truncated isoform associates with tapasin and competes with full-length US3 for binding to tapasin; thus, it suppresses the action of US3 that causes the disruption of the function of tapasin. Our results indicate that the truncated isoform of the US3 locus acts as a dominant negative regulator of full-length US3 activity. These data reflect the manner in which the virus has developed temporal survival strategies during viral infection against immune surveillance involving both CTLs and NK cells.
INTRODUCTION
Human cytomegalovirus (HCMV), a member of the Betaherpesviridae, leads to lifelong persistence in the infected host, followed by causing opportunistic pathogenesis in immunocompromised individuals such as neonates, organ transplant recipients, and AIDS patients. Both innate and adaptive immune systems participate in host defense against the viral infection in which natural killer (NK) cells and cytotoxic T lymphocytes (CTLs) function as the major effector cells. NK cell response plays an important role in initial defense, and the deficiency of NK cells results in susceptibility to severe herpesvirus infections (8). NK cell activation is controlled by the interaction of inhibitory receptors of NK cells with major histocompatibility complex (MHC) class I antigens on target cells (26). The missing-self hypothesis proposes that NK cells recognize and attack target cells that lack an adequate level of cell surface-expressed self-MHC class I molecules (25). Generally, the immune response of the host against a virus, including HCMV, is mediated by CTLs, which recognize and eliminate infected cells expressing viral peptide-loaded MHC class I molecules (41).
Since the MHC class I molecule is a key factor in the modulation of the immune response against viral pathogens, it is not surprising that class I molecules are preferred targets of viruses to avoid immune surveillance and clearance (27, 32, 40). HCMV has evolved strategies for down-regulating MHC class I molecules on the surface of an infected cell, by which HCMV keeps a balance with the CTL-mediated immune response (28). It has been well established that the four unique short (US) region products, namely, US2, US3, US6, and US11, use specialized mechanisms for the down-regulation of MHC class I molecules on the cell surface. The US2 and US11 gene products induce the rapid export of MHC class I heavy chains out of the endoplasmic reticulum (ER) into the cytoplasm, where they are degraded by the proteasome (45, 46). The US6 glycoprotein interferes with the activity of the transporter associated with antigen presentation (TAP) complex and peptide loading of MHC class I molecules (2, 18). The ER-resident US3 glycoprotein interferes with the intracellular transport and maturation of MHC class I molecules during the immediate-early phase of HCMV infection. US3 binds physically to peptide-loaded MHC class I heterodimers and arrests them in the ER (1, 19). Furthermore, as a dual mechanism for US3-mediated retention of MHC class I molecules, we have recently provided evidence that US3 binds to tapasin and inhibits tapasin-mediated peptide optimization, thereby displaying allelic specificity to tapasin-dependent MHC class I molecules (31). Because the down-regulation of MHC class I surface expression by HCMV US proteins results in susceptibility to attack by NK cells (12), HCMV has also developed immune evasins to shield itself against NK cell antiviral function, for example, UL16 (9), UL18 (4), and UL40 (39) proteins.
The 22-kDa US3 protein, a type I transmembrane protein, consists of a signal sequence of 15 amino acids, a luminal domain of 146 amino acids, and 20 membrane-spanning residues followed by a short, 5-amino-acid cytoplasmic tail (1). The structural and functional determinants of US3 have been revealed by deletion and chimeric mutant analyses. The luminal domain is sufficient for ER retention of US3 itself, and the Ser58, Glu63, and Lys64 sequences of the luminal domain play an essential role in the retention activity. On the other hand, the ability to associate with MHC class I molecules requires the transmembrane as well as the luminal domain of US3 (20, 21). Interestingly, the US3 gene encodes two additional variants from alternatively spliced transcripts. These include a singly spliced 17-kDa truncate (SS isoform), which lacks a transmembrane domain and shares the N-terminal 134 amino acids containing a glycosylation site with the full-length US3 gene product, except for the C-terminal 15 amino acids, and a doubly spliced 3.5-kDa fragment (DS isoform) (24, 38, 44). These isoforms are successively and coincidentally generated during the immediate-early phase of HCMV infections (7, 24, 38). However, evidence regarding the function of these isoforms is unavailable. In this report, we show that the truncated isoform of HCMV US3 alone neither sequesters MHC class I molecules in the ER nor inhibits the function of tapasin. However, coexpression of the truncated isoform and full-length US3 suppresses the inhibitory action of full-length US3 on tapasin function. In other words, in the presence of the truncated isoform, full-length US3 can no longer interfere with tapasin-mediated peptide optimization. Considered together, our results reveal that the truncated form acts as an endogenous dominant negative regulator of full-length US3 activity, indicating a novel mechanism for the self-regulation of US3 activity.
MATERIALS AND METHODS
Cells and transfections. HeLa and 293T cells were cultured as an adherent monolayer in Dulbecco's modified Eagle's medium with 10% fetal bovine serum (HyClone, Logan, UT), and the cells were transfected by the calcium phosphate precipitation method or with Lipofectamine 2000 (Invitrogen). MHC class I-negative K562 human leukemia cells and previously established K562 stable cells expressing HLA-B4402 (K562/B44) and HLA-B2705 (K562/B27) alleles (31) were grown in RPMI 1640 medium supplemented with 10% fetal bovine serum and 0.5 mg/ml of G418. K562 cells were transfected by the electroporation method using a GenePulser Xcell apparatus (Bio-Rad Laboratories).
Antibodies and plasmids. The monoclonal W6/32 antibody (Ab) recognizes only the complex of the MHC class I heavy chain (HC) and 2-microglobulin (2m), while the polyclonal K455 Ab recognizes HC and 2m in both assembled and nonassembled forms (3). Polyclonal antisera against the N-terminal portion (residues 16 to 35) of US3 (1) and against human TAP1 (14) have been described previously. Anti-tapasin and anti-green fluorescent protein (GFP) antibodies were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). The wild-type and GFP-tagged full-length US3 (US3-GFP) cDNAs have been described previously (20). The plasmid encoding soluble tapasin was generously provided by Frank Momburg (German Cancer Research Center, Heidelberg, Germany).
Flow cytometry. The expression of MHC class I glycoproteins on the membrane was determined by flow cytometry (FACSCalibur; Becton Dickinson, Mountain View, CA) and analyzed by CellQuest software after indirect immunofluorescence staining by using a saturating amount of monoclonal mouse anti-class I Ab (W6/32) and fluorescein isothiocyanate (FITC)-conjugated goat anti-mouse secondary Ab (Jackson ImmunoResearch Laboratories, Inc., West Grove, PA).
Metabolic labeling and immunoprecipitation. For pulse-chase radiolabeling, the cells were starved in methionine/cysteine-free medium for 30 min, pulsed with 0.1 mCi/ml of [35S]methionine/cysteine (NEM, Boston, MA) for 10 min or 30 min, chased in normal medium for the indicated times, lysed using 1% Nonidet P-40 (NP-40; Sigma-Aldrich, St. Louis, MO) or 1% digitonin (Calbiochem) in phosphate-buffered saline (PBS) supplemented with a protease inhibitor cocktail, and centrifuged at 12,000 x g for 20 min. The supernatants were incubated with protein G-Sepharose beads (Amersham Pharmacia Biotech, Piscataway, NJ). Following the removal of beads showing nonspecific binding to proteins, the lysates were immunoprecipitated with appropriate antibodies coupled to protein G-Sepharose. The protein binding beads were washed four times with 0.1% NP-40 or 0.1% digitonin, and the proteins were separated by sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis (PAGE) and then analyzed by using a phosphorimaging system (BAS-2500; Fuji Film Company). For the endoglycosidase H (endo H) test, immunobeads were digested with 3 mU of endo H (Roche, Indianapolis, IN) at 37°C overnight in a solution containing 50 mM sodium acetate (pH 5.6), 0.3% SDS, and 150 mM -mercaptoethanol.
Biotinylation of cell surface protein. The cells were washed gently twice in ice-cold PBS with 1 mM MgCl2 and 0.1 mM CaCl2 followed by incubation in sulfo-N-hydroxysuccinimidyl-biotin working solution (0.5 mg/ml; Pierce) with gentle shaking for 30 min at 4°C. The biotin-labeled cells were lysed in Triton X-100 buffer (1% Triton X-100, 20 mM Tris [pH 7.4], 150 mM NaCl, 1 mM phenylmethylsulfonyl fluoride, and 2 μg/ml leupeptin) and precipitated using W6/32 monoclonal antibody (MAb). The immunopellets were separated by SDS-PAGE and transferred to Immobilon-P membranes (Millipore, Bedford, MA). The membranes were blocked with a solution containing PBS, 5% bovine serum albumin, and 0.1% Tween 20 and incubated with horseradish peroxidase-conjugated streptavidin (Pierce) for 1 h at room temperature. After extensive washing with PBS-Tween (PBS and 0.1% Tween 20), W6/32-reactive biotin class I molecules were developed using SuperSignal West Pico chemiluminescent substrate (Pierce).
Thermostability assay. The biotin-labeled or radiolabeled cells were lysed in Triton X-100 buffer and centrifuged. Postnuclear supernatants were precleared with protein G-Sepharose beads for 30 min at 4°C and then divided into equal aliquots. The samples were kept in a heat block incubator at 4°C, 37°C, and 50°C for 60 min. Thereafter, the heat-treated samples were subjected to immunoprecipitation using the conformation-dependent W6/32 MAb, and the resulting immunopellets were separated by SDS-PAGE and visualized by immunoblot or autoradiography.
RESULTS
The individual products from alternatively spliced US3 transcripts do not affect the expression of MHC class I molecules on the cell surface. The full-length US3 protein binds and retains MHC class I heterodimers in the ER (1, 19). Alternatively, US3 targets tapasin function and thereby causes ER retention of tapasin-dependent class I alleles (31). To test whether the isoforms of the US3 protein could down-regulate MHC class I molecules in an allele-specific manner, similar to the full-length US3, we investigated the effect of individual US3 isoforms (Fig. 1A) on the surface expression levels of MHC class I molecules in K562/B44 and K562/B27 cells stably expressing either the HLA-B4402 or HLA-B2705 allele that represents tapasin-dependent and tapasin-independent class I molecules, respectively. The cDNAs encoding each of the US3 forms were transfected into the cells, and the surface expression levels of MHC class I alleles were determined by flow cytometry using anti-MHC class I complex MAb W6/32. The full-length US3 reduced the expression of surface class I molecules in K562/B44 cells (Fig. 1B), whereas it did not affect the surface expression levels of HLA-B2705 allele (Fig. 1C), which was consistent with a previous report (31). The isoforms of US3 (SS and DS) had no effect on the surface expression of MHC class I molecules in K562/B44 and K562/B27 cells (Fig. 1B and C, bottom panels). Metabolic labeling and immunoprecipitation experiments with polyclonal anti-US3 sera displayed expression of significant levels of ectopic US3 and SS in both stable cell lines (Fig. 1D). These results indicate that at least the isoforms of US3 alone do not have the ability to down-modulate the surface expression of MHC class I molecules, regardless of their tapasin dependence.
Coexpression of the truncated isoform and full-length US3 inhibits the activity of full-length US3. The US3 gene products are simultaneously produced during HCMV infection (7, 24). Therefore, we explored the possibility of whether the isoforms of US3 can affect the inhibitory action of the full-length US3 form on MHC class I molecules. We cotransfected full-length US3 (US3) and either the truncated (SS) or the fragmented (DS) isoform of US3 into HeLa cells and monitored the surface expression and intracellular maturation of class I molecules by flow cytometry and endo H experiments. Consistent with our observation in K562/B44 cells (Fig. 1B), both of the isoforms of US3 did not affect the surface expression of class I molecules in HeLa cells (Fig. 2A, middle panel). Interestingly, cotransfection of SS and US3 resulted in a significant increase in the surface level of class I molecules compared to that in cells transfected with US3 alone, even though a combination of DS and US3 did not affect the activity of full-length US3 in down-regulating the surface expression of MHC class I molecules (Fig. 2A, bottom panel). These results were further supported by pulse-chase experiments. The transfectants were metabolically labeled, chased, and immunoprecipitated using MAb W6/32. The equally separated samples were treated with either control buffer or endo H. In the presence of full-length US3 alone, MHC class I molecules were still sensitive to endo H digestion even after a 60-min chase (Fig. 2B, second panel), indicating that full-length US3 caused ER retention of MHC class I molecules. Almost all W6/32-reactive MHC class I molecules were resistant to endo H digestion after a 60-min chase in mock, SS, or US3-plus-SS transfectants (Fig. 2B, first, third, and fourth panels, respectively). It is noteworthy that at the 30-min chase point, the class I molecules in mock transfectant are endo H resistant, while the class I molecules in US3/SS transfectants are still endo H sensitive. Nevertheless, these results, coupled with the data from the flow cytometry experiment, suggest that the SS truncated isoform could act as a negative regulator of the full-length form.
The truncated isoform neither affects the stability and ER localization of full-length US3 nor prevents it from interacting with MHC class I molecules. To understand the mechanism underlying the counteraction of the truncated isoform on full-length US3, we investigated whether the truncated isoform influences the stability or cellular localization of full-length US3. Coexpression of the truncated isoform did not alter the stability of full-length US3 (Fig. 3A). In the presence of the truncated isoform, full-length US3 remained sensitive to endo H throughout the 90-min chase (Fig. 3B), indicating that ER localization of full-length US3, which is essential for its function, did not change. Next, we investigated whether the truncated isoform could block the binding of full-length US3 to MHC class I molecules. HeLa cells were cotransfected with cDNAs encoding US3 variant forms, the cells were radiolabeled, and digitonin lysates were immunoprecipitated with either anti-US3 or anti-class I (K455) Ab. Only the full-length US3 was coprecipitated along with MHC class I HC and 2m, and the amount of coprecipitated full-length US3 was not affected by the presence of the truncated isoform (Fig. 3C).
The truncated isoform competes with full-length US3 for binding to tapasin and restores the US3-mediated disruption of TAP/tapasin complexes. The finding that the truncated isoform does not affect the association between full-length US3 and MHC class I molecules prompted us to explore the possibility that the truncated isoform can prevent full-length US3 from subverting tapasin, another target of US3. Immunoprecipitation of HeLa transfectants with anti-US3 Ab revealed that both full-length US3 and the truncated isoform have an intrinsic property to bind tapasin (Fig. 4A), implying that the truncated isoform can induce the dissociation of full-length US3 from tapasin via competition for binding to tapasin. To further examine whether the truncated isoform inhibits the interaction between full-length US3 and tapasin proteins, we used US3-GFP due to the advantage of it distinguishing the full-length US3-associated tapasin by using anti-GFP Ab instead of anti-US3 Ab. We cotransfected US3-GFP with or without the truncated isoform into HeLa cells and analyzed the amounts of tapasin binding to US3-GFP by coimmunoprecipitation experiments using anti-GFP Ab. Surprisingly, the amount of tapasin binding to US3-GFP was extremely reduced in the presence of the truncated isoform (Fig. 4B, top panel, compare lanes 2 and 3). The reduction of US3-GFP-associated tapasin by the truncated isoform is not due to a decrease in the levels of expressed US3-GFP and tapasin, because comparable amounts of US3-GFP and tapasin were expressed in each transfectant, as shown by immunoblotting (Fig. 4B, middle and bottom panels). The relation between full-length US3 and the truncated isoform was further investigated in 293T cells that express little tapasin (13). The 293T cells were cotransfected with cDNAs as indicated and analyzed by coprecipitation and immunoblot (Fig. 4C). Despite the comparable expression levels of US3-GFP and soluble tapasin (Fig. 4C, middle and bottom panels), the amount of US3-GFP coprecipitated with soluble tapasin was remarkably decreased in the presence of the truncated SS isoform (Fig. 4C, top panel, compare lanes 3 and 4).
The association of full-length US3 with tapasin causes the dissociation of tapasin from TAP complexes (31). Therefore, we investigated whether the truncated isoform could counteract the activity of full-length US3 that causes the dissociation of tapasin from TAP. The association of tapasin with TAP1 was blocked in HeLa cells transfected with full-length US3 alone (Fig. 4D, compare lanes 3 and 4). However, it was recovered in the cotransfectants with full-length US3 and the truncated isoform (Fig. 4D, compare lanes 4 and 6), suggesting that the truncated isoform counteracted the activity of full-length US3 in weakening the TAP/tapasin association. Based on the above-described results, we can conclude that the truncated isoform competes with full-length US3 for binding to tapasin, suggesting that the truncated isoform is a negative regulator of full-length US3.
The truncated isoform hinders full-length US3 in the inhibition of tapasin-dependent peptide optimization of MHC class I molecules. Full-length US3 inhibits peptide optimization of MHC class I peptide cargo, which is mainly mediated by tapasin (31). To test whether the truncated isoform could counteract the activity of full-length US3 in the inhibition of peptide optimization, we analyzed the thermostability of MHC class I complexes, which is greatly linked to the binding affinity of their peptide cargo (11, 29, 31, 35, 47). The transfected cells were either radiolabeled (Fig. 5A) or biotinylated (Fig. 5B) to differentiate between the newly synthesized and surface-expressed MHC class I heavy chains, treated with heat, and precipitated using conformation-dependent MAb W6/32. Quantification of W6/32-reactive class I heavy chains showed that the differences in the thermostability of MHC class I molecules were most prominent at 37°C. The thermostability of heavy chains was significantly reduced in the presence of full-length US3 alone at 37°C (Fig. 5A and B, second panels), whereas the truncated isoform alone failed to decrease the thermostability of heavy chains at 37°C (Fig. 5A and B, third panels). Interestingly, in the case of coexpression of full-length US3 and the truncated isoform, both surface and newly synthesized class I heavy chains were thermostable at 37°C in comparison with the thermostability of MHC class I molecules in either mock- or SS-transfected cells (Fig. 5A and B, fourth panels). Considered together, these results strongly suggest that the truncated isoform prevents full-length US3 from inhibiting the tapasin function of optimizing peptide cargo.
DISCUSSION
MHC class I molecules mediate the adaptive immune response by presenting non-self antigens to CTLs, followed by the elimination of target cells. The 22-kDa US3 glycoprotein is encoded within the HCMV US region genes that are related with immune escape against CTLs. This viral protein makes use of a dual mechanism to block the antigen presentation by MHC class I molecules. First, US3 directly attacks MHC class I molecules; that is, US3 physically binds to MHC class I molecules and sequesters them in the ER (1, 19). The second mechanism is a rather indirect action in which US3 targets tapasin, a component of peptide-loading complexes, and binds to it; this results in the inhibition of tapasin-dependent peptide loading and optimization (31). The dual mechanism may enable the virus to evade CTLs more efficiently, thereby contributing to the establishment of a persistent and latent infection. Since the US3 protein is expressed at the immediate-early phase of infection with no connection to de novo protein synthesis (44) and since NK cells rapidly respond to viral infections in the innate immune system, the US3-mediated down-regulation of the expression of MHC class I molecules on the cell surface renders the infected cells susceptible to NK lysis. This immediate-early protein of HCMV therefore requires tight regulation, both positive and negative, although the window of time when the virus encounters NK cells or CTLs is not known.
In this study, we demonstrate that the truncated isoform generated by an alternative splicing of the HCMV US3 gene transcript negatively regulates the action of full-length US3 in down-regulating the expression of MHC class I molecules via competing with full-length US3 for binding to a target molecule, tapasin. The relative abundance of US3 transcripts varies during HCMV infections; the full-length US3 transcript is most abundant early during infection, followed by singly spliced and doubly spliced transcripts (7, 24). Each of the encoded proteins might thus contribute to the viral life cycle in the infected host. Considering the sequential expression of US3 transcripts and our current observations, it is likely that full-length US3 and the truncated isoform have evolved to evade immune surveillance by CTLs and NK cells, respectively, during early infection. This mode of regulation, in which the function of a full-length form is controlled by a spliced truncate, has been reported in not only viral infection but also various biological contexts, for example, ICP0/ICP0R (36), HIF-1/HIF-1736 (15), and DNOS1/DNOS4 (37). The adenovirus E3/19K protein is a structural and functional homolog of full-length US3. It is an ER-resident type I transmembrane glycoprotein and inhibits MHC class I molecules from reaching the cell surface by two mechanisms targeting both tapasin and MHC class I molecules (5, 31). Similar to the action of the truncated US3 isoform on the full-length form of US3, the E3/19K molecule lacking the C-terminal ER retention motif forms a heterodimer with wild-type E3/19K and competes with E3/19K for Kd binding, thereby suppressing Kd retention by E3/19K (10).
The ability of US3 to associate with MHC class I molecules requires the transmembrane domain in addition to the luminal domain of US3 (21). Since the truncated isoform lacks the transmembrane domain that is responsible for interacting with MHC class I molecules (21), it is unlikely that the truncated isoform prevents full-length US3 from binding to the MHC class I molecules. This idea was supported by the observation that MHC class I molecules were coprecipitated only with full-length US3 but not with the truncated isoform and that the association of MHC class I molecules with full-length US3 was not reduced even in the presence of the truncated isoform (Fig. 3C).
The precise function of tapasin in MHC class I assembly is uncertain. It plays a critical role in bridging MHC class I molecules to TAP (30, 42), tethering them in the ER (16 34), and in turn enhancing peptide loading and optimization (17, 33, 47). Our study showed that not only the full-length US3 but also the truncated isoform are able to bind to tapasin (Fig. 4A). A notable observation was that the total amounts of tapasin associated with either US3 or SS proteins were not additive (Fig. 4A, lanes 3 to 5) and that in the presence of SS, the level of association between US3 and tapasin was remarkably reduced (Fig. 4B). Interestingly, the transport kinetics of class I molecules are similar in both US3/SS and US3-alone transfectants at the 30-min chase point but show clear differences between them at the 60-min chase (Fig. 2B). This might suggest that the inhibitory effect of the truncated isoform is sequentially turned on after the action of full-length US3. Since the truncated isoform does not inhibit the association of full-length US3 with class I molecules and since the soluble truncated isoform is more free to move and diffuse from tapasin than the membrane-bound full-length US3, the truncated isoform seems to lag behind in its action. Based on these results, we can conclude that the truncated isoform competes with full-length US3 for binding to tapasin and thereby nullifies the activity of US3 in inhibiting tapasin function.
The current study also provides clues for the functional determinants of US3 and tapasin. Our finding that the truncated US3 isoform binds to full-length tapasin (Fig. 4A and B) and that full-length US3 binds to soluble tapasin (Fig. 4C) indicates that the luminal domain of each protein mediates the interaction. Importantly, the truncated US3 isoform is able to bind to tapasin, but unlike full-length US3, it does not affect the transport of MHC class I molecules (Fig. 2), demonstrating that the mode of interaction between the truncated US3 isoform and tapasin is unproductive and that the transmembrane domain of the US3 protein is required for inhibiting tapasin function. The separated usages of the US3 domain are similar to the case of US11, another US region gene product, the luminal domain of which is sufficient for MHC class I binding, but the transmembrane domain is crucial for MHC class I heavy-chain dislocation (22, 23).
Transcription of the US3 gene is controlled by a complex network of viral regulators (6); however, the regulatory mechanism for spliced variants remains unclear. Full-length US3 may exhibit contradictory effects on the survival of the virus. While inactivation of full-length US3 contributes to avoiding NK lysis, it makes virus-infected cells vulnerable to lysis by CTLs. Thus, biosynthesis or activities of the truncated US3 isoform should be precisely controlled, for instance, by posttranslational modifications such as ubiquitination of ICP0R during productive infection (43). Considered together, our results suggest that in HCMV, the truncated variant of US3 acts as a negative regulator of full-length US3 activity during viral infection. This mode of regulation that we observed in HCMV could also be broadly involved in the regulation of activities of immunoevasive proteins in other viruses.
ACKNOWLEDGMENTS
We are grateful to Frank Momburg for providing tapasin cDNA.
This work was supported by a grant from the Korea Research Foundation (C00479) and by grant R01-2005-000-10235-0 from the KOSEF and a grant from the Research Center for Functional Cellulomics of the KOSEF.
REFERENCES
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.
Ahn, K., A. Gruhler, B. Galocha, T. R. Jones, E. J. Wiertz, H. L. Ploegh, P. A. Peterson, Y. Yang, and K. Fruh. 1997. The ER-luminal domain of the HCMV glycoprotein US6 inhibits peptide translocation by TAP. Immunity 6:613-621.
Andersson, M., S. Paabo, T. Nilsson, and P. A. Peterson. 1985. Impaired intracellular transport of class I MHC antigens as a possible means for adenoviruses to evade immune surveillance. Cell 43:215-222.
Beck, S., and B. G. Barrell. 1988. Human cytomegalovirus encodes a glycoprotein homologous to MHC class-I antigens. Nature 331:269-272.
Bennett, E. M., J. R. Bennink, J. W. Yewdell, and F. M. Brodsky. 1999. Cutting edge: adenovirus E19 has two mechanisms for affecting class I MHC expression. J. Immunol. 162:5049-5052.
Biegalke, B. J. 1999. Human cytomegalovirus US3 gene expression is regulated by a complex network of positive and negative regulators. Virology 261:155-164.
Biegalke, B. J. 1995. Regulation of human cytomegalovirus US3 gene transcription by a cis-repressive sequence. J. Virol. 69:5362-5367.
Biron, C. A., K. S. Byron, and J. L. Sullivan. 1989. Severe herpesvirus infections in an adolescent without natural killer cells. N. Engl. J. Med. 320:1731-1735.
Cosman, D., J. Mullberg, C. L. Sutherland, W. Chin, R. Armitage, W. Fanslow, M. Kubin, and N. J. Chalupny. 2001. ULBPs, novel MHC class I-related molecules, bind to CMV glycoprotein UL16 and stimulate NK cytotoxicity through the NKG2D receptor. Immunity 14:123-133.
Cox, J. H., J. R. Bennink, and J. W. Yewdell. 1991. Retention of adenovirus E19 glycoprotein in the endoplasmic reticulum is essential to its ability to block antigen presentation. J. Exp. Med. 174:1629-1637.
Fahnestock, M. L., I. Tamir, L. Narhi, and P. J. Bjorkman. 1992. Thermal stability comparison of purified empty and peptide-filled forms of a class I MHC molecule. Science 258:1658-1662.
Falk, C. S., M. Mach, D. J. Schendel, E. H. Weiss, I. Hilgert, and G. Hahn. 2002. NK cell activity during human cytomegalovirus infection is dominated by US2-11-mediated HLA class I down-regulation. J. Immunol. 169:3257-3266.
Ford, S., A. Antoniou, G. W. Butcher, and S. J. Powis. 2004. Competition for access to the rat major histocompatibility complex class I peptide-loading complex reveals optimization of peptide cargo in the absence of transporter associated with antigen processing (TAP) association. J. Biol. Chem. 279:16077-16082.
Fruh, K., Y. Yang, D. Arnold, J. Chambers, L. Wu, J. B. Waters, T. Spies, and P. A. Peterson. 1992. Alternative exon usage and processing of the major histocompatibility complex-encoded proteasome subunits. J. Biol. Chem. 267:22131-22140.
Gothie, E., D. E. Richard, E. Berra, G. Pages, and J. Pouyssegur. 2000. Identification of alternative spliced variants of human hypoxia-inducible factor-1alpha. J. Biol. Chem. 275:6922-6927.
Grandea, A. G., III, P. J. Lehner, P. Cresswell, and T. Spies. 1997. Regulation of MHC class I heterodimer stability and interaction with TAP by tapasin. Immunogenetics 46:477-483.
Grandea, A. G., III, and L. Van Kaer. 2001. Tapasin: an ER chaperone that controls MHC class I assembly with peptide. Trends Immunol. 22:194-199.
Hengel, H., J. O. Koopmann, T. Flohr, W. Muranyi, E. Goulmy, G. J. Hammerling, U. H. Koszinowski, and F. Momburg. 1997. A viral ER-resident glycoprotein inactivates the MHC-encoded peptide transporter. Immunity 6:623-632.
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.
Lee, S., B. Park, and K. Ahn. 2003. Determinant for endoplasmic reticulum retention in the luminal domain of the human cytomegalovirus US3 glycoprotein. J. Virol. 77:2147-2156.
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.
Lee, S. O., S. Hwang, J. Park, B. Park, B. S. Jin, S. Lee, E. Kim, S. Cho, Y. Kim, K. Cho, J. Shin, and K. Ahn. 2005. Functional dissection of HCMV US11 in mediating the degradation of MHC class I molecules. Biochem. Biophys. Res. Commun. 330:1262-1267.
Lilley, B. N., D. Tortorella, and H. L. Ploegh. 2003. Dislocation of a type I membrane protein requires interactions between membrane-spanning segments within the lipid bilayer. Mol. Biol. Cell 14:3690-3698.
Liu, W., Y. Zhao, and B. Biegalke. 2002. Analysis of human cytomegalovirus US3 gene products. Virology 301:32-42.
Ljunggren, H. G., and K. Karre. 1990. In search of the ‘missing self’: MHC molecules and NK cell recognition. Immunol. Today 11:237-244.
Lodoen, M. B., and L. L. Lanier. 2005. Viral modulation of NK cell immunity. Nat. Rev. Microbiol. 3:59-69.
Lybarger, L., X. Wang, M. Harris, and T. H. Hansen. 2005. Viral immune evasion molecules attack the ER peptide-loading complex and exploit ER-associated degradation pathways. Curr. Opin. Immunol. 17:71-78.
Mocarski, E. S., Jr. 2002. Immunomodulation by cytomegaloviruses: manipulative strategies beyond evasion. Trends Microbiol. 10:332-339.
Neefjes, J. J., G. J. Hammerling, and F. Momburg. 1993. Folding and assembly of major histocompatibility complex class I heterodimers in the endoplasmic reticulum of intact cells precedes the binding of peptide. J. Exp. Med. 178:1971-1980.
Ortmann, B., J. Copeman, P. J. Lehner, B. Sadasivan, J. A. Herberg, A. G. Grandea, S. R. Riddell, R. Tampe, T. Spies, J. Trowsdale, and P. Cresswell. 1997. A critical role for tapasin in the assembly and function of multimeric MHC class I-TAP complexes. Science 277:1306-1309.
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.
Petersen, J. L., C. R. Morris, and J. C. Solheim. 2003. Virus evasion of MHC class I molecule presentation. J. Immunol. 171:4473-4478.
Purcell, A. W., J. J. Gorman, M. Garcia-Peydro, A. Paradela, S. R. Burrows, G. H. Talbo, N. Laham, C. A. Peh, E. C. Reynolds, J. A. Lopez De Castro, and J. McCluskey. 2001. Quantitative and qualitative influences of tapasin on the class I peptide repertoire. J. Immunol. 166:1016-1027.
Schoenhals, G. J., R. M. Krishna, A. G. Grandea III, T. Spies, P. A. Peterson, Y. Yang, and K. Fruh. 1999. Retention of empty MHC class I molecules by tapasin is essential to reconstitute antigen presentation in invertebrate cells. EMBO J. 18:743-753.
Schumacher, T. N., M. T. Heemels, J. J. Neefjes, W. M. Kast, C. J. Melief, and H. L. Ploegh. 1990. Direct binding of peptide to empty MHC class I molecules on intact cells and in vitro. Cell 62:563-567.
Spatz, S. J., E. C. Nordby, and P. C. Weber. 1996. Mutational analysis of ICP0R, a transrepressor protein created by alternative splicing of the ICP0 gene of herpes simplex virus type 1. J. Virol. 70:7360-7370.
Stasiv, Y., B. Kuzin, M. Regulski, T. Tully, and G. Enikolopov. 2004. Regulation of multimers via truncated isoforms: a novel mechanism to control nitric-oxide signaling. Genes Dev. 18:1812-1823.
Tenney, D. J., L. D. Santomenna, K. B. Goudie, and A. M. Colberg-Poley. 1993. The human cytomegalovirus US3 immediate-early protein lacking the putative transmembrane domain regulates gene expression. Nucleic Acids Res. 21:2931-2937.
Tomasec, P., V. M. Braud, C. Rickards, M. B. Powell, B. P. McSharry, S. Gadola, V. Cerundolo, L. K. Borysiewicz, A. J. McMichael, and G. W. Wilkinson. 2000. Surface expression of HLA-E, an inhibitor of natural killer cells, enhanced by human cytomegalovirus gpUL40. Science 287:1031.
Tortorella, D., B. E. Gewurz, M. H. Furman, D. J. Schust, and H. L. Ploegh. 2000. Viral subversion of the immune system. Annu. Rev. Immunol. 18:861-926.
Townsend, A., and H. Bodmer. 1989. Antigen recognition by class I-restricted T lymphocytes. Annu. Rev. Immunol. 7:601-624.
Turnquist, H. R., J. L. Petersen, S. E. Vargas, M. M. McIlhaney, E. Bedows, W. E. Mayer, A. G. Grandea III, L. Van Kaer, and J. C. Solheim. 2004. The Ig-like domain of tapasin influences intermolecular interactions. J. Immunol. 172:2976-2984.
Weber, P. C., S. J. Spatz, and E. C. Nordby. 1999. Stable ubiquitination of the ICP0R protein of herpes simplex virus type 1 during productive infection. Virology 253:288-298.
Weston, K. 1988. An enhancer element in the short unique region of human cytomegalovirus regulates the production of a group of abundant immediate early transcripts. Virology 162:406-416.
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.
Williams, A. P., C. A. Peh, A. W. Purcell, J. McCluskey, and T. Elliott. 2002. Optimization of the MHC class I peptide cargo is dependent on tapasin. Immunity 16:509-520.(Jinwook Shin, Boyoun Park)
College of Life Science and Biotechnology, Korea University, Seoul 136-704, Korea
Department of Biomedical Sciences, 228 Irvine Hall, Ohio University College of Osteopathic Medicine, Athens, Ohio 45701
ABSTRACT
Human cytomegalovirus encodes four unique short (US) region proteins, each of which is independently sufficient for causing the down-regulation of major histocompatibility complex (MHC) class I molecules on the cell surface. This down-regulation enables infected cells to evade recognition by cytotoxic T lymphocytes (CTLs) but makes them vulnerable to lysis by natural killer (NK) cells, which lyse those cells that lack MHC class I molecules. The 22-kDa US3 glycoprotein is able to down-regulate the surface expression of MHC class I molecules by dual mechanisms: direct endoplasmic reticulum retention by physical association and/or tapasin inhibition. The alternative splicing of the US3 gene generates two additional products, including 17-kDa and 3.5-kDa truncated isoforms; however, the functional significance of these isoforms during viral infection is unknown. Here, we describe a novel mode of self-regulation of US3 function that uses the endogenously produced truncated isoform. The truncated isoform itself neither binds to MHC class I molecules nor prevents the full-length US3 from interacting with MHC class I molecules. Instead, the truncated isoform associates with tapasin and competes with full-length US3 for binding to tapasin; thus, it suppresses the action of US3 that causes the disruption of the function of tapasin. Our results indicate that the truncated isoform of the US3 locus acts as a dominant negative regulator of full-length US3 activity. These data reflect the manner in which the virus has developed temporal survival strategies during viral infection against immune surveillance involving both CTLs and NK cells.
INTRODUCTION
Human cytomegalovirus (HCMV), a member of the Betaherpesviridae, leads to lifelong persistence in the infected host, followed by causing opportunistic pathogenesis in immunocompromised individuals such as neonates, organ transplant recipients, and AIDS patients. Both innate and adaptive immune systems participate in host defense against the viral infection in which natural killer (NK) cells and cytotoxic T lymphocytes (CTLs) function as the major effector cells. NK cell response plays an important role in initial defense, and the deficiency of NK cells results in susceptibility to severe herpesvirus infections (8). NK cell activation is controlled by the interaction of inhibitory receptors of NK cells with major histocompatibility complex (MHC) class I antigens on target cells (26). The missing-self hypothesis proposes that NK cells recognize and attack target cells that lack an adequate level of cell surface-expressed self-MHC class I molecules (25). Generally, the immune response of the host against a virus, including HCMV, is mediated by CTLs, which recognize and eliminate infected cells expressing viral peptide-loaded MHC class I molecules (41).
Since the MHC class I molecule is a key factor in the modulation of the immune response against viral pathogens, it is not surprising that class I molecules are preferred targets of viruses to avoid immune surveillance and clearance (27, 32, 40). HCMV has evolved strategies for down-regulating MHC class I molecules on the surface of an infected cell, by which HCMV keeps a balance with the CTL-mediated immune response (28). It has been well established that the four unique short (US) region products, namely, US2, US3, US6, and US11, use specialized mechanisms for the down-regulation of MHC class I molecules on the cell surface. The US2 and US11 gene products induce the rapid export of MHC class I heavy chains out of the endoplasmic reticulum (ER) into the cytoplasm, where they are degraded by the proteasome (45, 46). The US6 glycoprotein interferes with the activity of the transporter associated with antigen presentation (TAP) complex and peptide loading of MHC class I molecules (2, 18). The ER-resident US3 glycoprotein interferes with the intracellular transport and maturation of MHC class I molecules during the immediate-early phase of HCMV infection. US3 binds physically to peptide-loaded MHC class I heterodimers and arrests them in the ER (1, 19). Furthermore, as a dual mechanism for US3-mediated retention of MHC class I molecules, we have recently provided evidence that US3 binds to tapasin and inhibits tapasin-mediated peptide optimization, thereby displaying allelic specificity to tapasin-dependent MHC class I molecules (31). Because the down-regulation of MHC class I surface expression by HCMV US proteins results in susceptibility to attack by NK cells (12), HCMV has also developed immune evasins to shield itself against NK cell antiviral function, for example, UL16 (9), UL18 (4), and UL40 (39) proteins.
The 22-kDa US3 protein, a type I transmembrane protein, consists of a signal sequence of 15 amino acids, a luminal domain of 146 amino acids, and 20 membrane-spanning residues followed by a short, 5-amino-acid cytoplasmic tail (1). The structural and functional determinants of US3 have been revealed by deletion and chimeric mutant analyses. The luminal domain is sufficient for ER retention of US3 itself, and the Ser58, Glu63, and Lys64 sequences of the luminal domain play an essential role in the retention activity. On the other hand, the ability to associate with MHC class I molecules requires the transmembrane as well as the luminal domain of US3 (20, 21). Interestingly, the US3 gene encodes two additional variants from alternatively spliced transcripts. These include a singly spliced 17-kDa truncate (SS isoform), which lacks a transmembrane domain and shares the N-terminal 134 amino acids containing a glycosylation site with the full-length US3 gene product, except for the C-terminal 15 amino acids, and a doubly spliced 3.5-kDa fragment (DS isoform) (24, 38, 44). These isoforms are successively and coincidentally generated during the immediate-early phase of HCMV infections (7, 24, 38). However, evidence regarding the function of these isoforms is unavailable. In this report, we show that the truncated isoform of HCMV US3 alone neither sequesters MHC class I molecules in the ER nor inhibits the function of tapasin. However, coexpression of the truncated isoform and full-length US3 suppresses the inhibitory action of full-length US3 on tapasin function. In other words, in the presence of the truncated isoform, full-length US3 can no longer interfere with tapasin-mediated peptide optimization. Considered together, our results reveal that the truncated form acts as an endogenous dominant negative regulator of full-length US3 activity, indicating a novel mechanism for the self-regulation of US3 activity.
MATERIALS AND METHODS
Cells and transfections. HeLa and 293T cells were cultured as an adherent monolayer in Dulbecco's modified Eagle's medium with 10% fetal bovine serum (HyClone, Logan, UT), and the cells were transfected by the calcium phosphate precipitation method or with Lipofectamine 2000 (Invitrogen). MHC class I-negative K562 human leukemia cells and previously established K562 stable cells expressing HLA-B4402 (K562/B44) and HLA-B2705 (K562/B27) alleles (31) were grown in RPMI 1640 medium supplemented with 10% fetal bovine serum and 0.5 mg/ml of G418. K562 cells were transfected by the electroporation method using a GenePulser Xcell apparatus (Bio-Rad Laboratories).
Antibodies and plasmids. The monoclonal W6/32 antibody (Ab) recognizes only the complex of the MHC class I heavy chain (HC) and 2-microglobulin (2m), while the polyclonal K455 Ab recognizes HC and 2m in both assembled and nonassembled forms (3). Polyclonal antisera against the N-terminal portion (residues 16 to 35) of US3 (1) and against human TAP1 (14) have been described previously. Anti-tapasin and anti-green fluorescent protein (GFP) antibodies were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). The wild-type and GFP-tagged full-length US3 (US3-GFP) cDNAs have been described previously (20). The plasmid encoding soluble tapasin was generously provided by Frank Momburg (German Cancer Research Center, Heidelberg, Germany).
Flow cytometry. The expression of MHC class I glycoproteins on the membrane was determined by flow cytometry (FACSCalibur; Becton Dickinson, Mountain View, CA) and analyzed by CellQuest software after indirect immunofluorescence staining by using a saturating amount of monoclonal mouse anti-class I Ab (W6/32) and fluorescein isothiocyanate (FITC)-conjugated goat anti-mouse secondary Ab (Jackson ImmunoResearch Laboratories, Inc., West Grove, PA).
Metabolic labeling and immunoprecipitation. For pulse-chase radiolabeling, the cells were starved in methionine/cysteine-free medium for 30 min, pulsed with 0.1 mCi/ml of [35S]methionine/cysteine (NEM, Boston, MA) for 10 min or 30 min, chased in normal medium for the indicated times, lysed using 1% Nonidet P-40 (NP-40; Sigma-Aldrich, St. Louis, MO) or 1% digitonin (Calbiochem) in phosphate-buffered saline (PBS) supplemented with a protease inhibitor cocktail, and centrifuged at 12,000 x g for 20 min. The supernatants were incubated with protein G-Sepharose beads (Amersham Pharmacia Biotech, Piscataway, NJ). Following the removal of beads showing nonspecific binding to proteins, the lysates were immunoprecipitated with appropriate antibodies coupled to protein G-Sepharose. The protein binding beads were washed four times with 0.1% NP-40 or 0.1% digitonin, and the proteins were separated by sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis (PAGE) and then analyzed by using a phosphorimaging system (BAS-2500; Fuji Film Company). For the endoglycosidase H (endo H) test, immunobeads were digested with 3 mU of endo H (Roche, Indianapolis, IN) at 37°C overnight in a solution containing 50 mM sodium acetate (pH 5.6), 0.3% SDS, and 150 mM -mercaptoethanol.
Biotinylation of cell surface protein. The cells were washed gently twice in ice-cold PBS with 1 mM MgCl2 and 0.1 mM CaCl2 followed by incubation in sulfo-N-hydroxysuccinimidyl-biotin working solution (0.5 mg/ml; Pierce) with gentle shaking for 30 min at 4°C. The biotin-labeled cells were lysed in Triton X-100 buffer (1% Triton X-100, 20 mM Tris [pH 7.4], 150 mM NaCl, 1 mM phenylmethylsulfonyl fluoride, and 2 μg/ml leupeptin) and precipitated using W6/32 monoclonal antibody (MAb). The immunopellets were separated by SDS-PAGE and transferred to Immobilon-P membranes (Millipore, Bedford, MA). The membranes were blocked with a solution containing PBS, 5% bovine serum albumin, and 0.1% Tween 20 and incubated with horseradish peroxidase-conjugated streptavidin (Pierce) for 1 h at room temperature. After extensive washing with PBS-Tween (PBS and 0.1% Tween 20), W6/32-reactive biotin class I molecules were developed using SuperSignal West Pico chemiluminescent substrate (Pierce).
Thermostability assay. The biotin-labeled or radiolabeled cells were lysed in Triton X-100 buffer and centrifuged. Postnuclear supernatants were precleared with protein G-Sepharose beads for 30 min at 4°C and then divided into equal aliquots. The samples were kept in a heat block incubator at 4°C, 37°C, and 50°C for 60 min. Thereafter, the heat-treated samples were subjected to immunoprecipitation using the conformation-dependent W6/32 MAb, and the resulting immunopellets were separated by SDS-PAGE and visualized by immunoblot or autoradiography.
RESULTS
The individual products from alternatively spliced US3 transcripts do not affect the expression of MHC class I molecules on the cell surface. The full-length US3 protein binds and retains MHC class I heterodimers in the ER (1, 19). Alternatively, US3 targets tapasin function and thereby causes ER retention of tapasin-dependent class I alleles (31). To test whether the isoforms of the US3 protein could down-regulate MHC class I molecules in an allele-specific manner, similar to the full-length US3, we investigated the effect of individual US3 isoforms (Fig. 1A) on the surface expression levels of MHC class I molecules in K562/B44 and K562/B27 cells stably expressing either the HLA-B4402 or HLA-B2705 allele that represents tapasin-dependent and tapasin-independent class I molecules, respectively. The cDNAs encoding each of the US3 forms were transfected into the cells, and the surface expression levels of MHC class I alleles were determined by flow cytometry using anti-MHC class I complex MAb W6/32. The full-length US3 reduced the expression of surface class I molecules in K562/B44 cells (Fig. 1B), whereas it did not affect the surface expression levels of HLA-B2705 allele (Fig. 1C), which was consistent with a previous report (31). The isoforms of US3 (SS and DS) had no effect on the surface expression of MHC class I molecules in K562/B44 and K562/B27 cells (Fig. 1B and C, bottom panels). Metabolic labeling and immunoprecipitation experiments with polyclonal anti-US3 sera displayed expression of significant levels of ectopic US3 and SS in both stable cell lines (Fig. 1D). These results indicate that at least the isoforms of US3 alone do not have the ability to down-modulate the surface expression of MHC class I molecules, regardless of their tapasin dependence.
Coexpression of the truncated isoform and full-length US3 inhibits the activity of full-length US3. The US3 gene products are simultaneously produced during HCMV infection (7, 24). Therefore, we explored the possibility of whether the isoforms of US3 can affect the inhibitory action of the full-length US3 form on MHC class I molecules. We cotransfected full-length US3 (US3) and either the truncated (SS) or the fragmented (DS) isoform of US3 into HeLa cells and monitored the surface expression and intracellular maturation of class I molecules by flow cytometry and endo H experiments. Consistent with our observation in K562/B44 cells (Fig. 1B), both of the isoforms of US3 did not affect the surface expression of class I molecules in HeLa cells (Fig. 2A, middle panel). Interestingly, cotransfection of SS and US3 resulted in a significant increase in the surface level of class I molecules compared to that in cells transfected with US3 alone, even though a combination of DS and US3 did not affect the activity of full-length US3 in down-regulating the surface expression of MHC class I molecules (Fig. 2A, bottom panel). These results were further supported by pulse-chase experiments. The transfectants were metabolically labeled, chased, and immunoprecipitated using MAb W6/32. The equally separated samples were treated with either control buffer or endo H. In the presence of full-length US3 alone, MHC class I molecules were still sensitive to endo H digestion even after a 60-min chase (Fig. 2B, second panel), indicating that full-length US3 caused ER retention of MHC class I molecules. Almost all W6/32-reactive MHC class I molecules were resistant to endo H digestion after a 60-min chase in mock, SS, or US3-plus-SS transfectants (Fig. 2B, first, third, and fourth panels, respectively). It is noteworthy that at the 30-min chase point, the class I molecules in mock transfectant are endo H resistant, while the class I molecules in US3/SS transfectants are still endo H sensitive. Nevertheless, these results, coupled with the data from the flow cytometry experiment, suggest that the SS truncated isoform could act as a negative regulator of the full-length form.
The truncated isoform neither affects the stability and ER localization of full-length US3 nor prevents it from interacting with MHC class I molecules. To understand the mechanism underlying the counteraction of the truncated isoform on full-length US3, we investigated whether the truncated isoform influences the stability or cellular localization of full-length US3. Coexpression of the truncated isoform did not alter the stability of full-length US3 (Fig. 3A). In the presence of the truncated isoform, full-length US3 remained sensitive to endo H throughout the 90-min chase (Fig. 3B), indicating that ER localization of full-length US3, which is essential for its function, did not change. Next, we investigated whether the truncated isoform could block the binding of full-length US3 to MHC class I molecules. HeLa cells were cotransfected with cDNAs encoding US3 variant forms, the cells were radiolabeled, and digitonin lysates were immunoprecipitated with either anti-US3 or anti-class I (K455) Ab. Only the full-length US3 was coprecipitated along with MHC class I HC and 2m, and the amount of coprecipitated full-length US3 was not affected by the presence of the truncated isoform (Fig. 3C).
The truncated isoform competes with full-length US3 for binding to tapasin and restores the US3-mediated disruption of TAP/tapasin complexes. The finding that the truncated isoform does not affect the association between full-length US3 and MHC class I molecules prompted us to explore the possibility that the truncated isoform can prevent full-length US3 from subverting tapasin, another target of US3. Immunoprecipitation of HeLa transfectants with anti-US3 Ab revealed that both full-length US3 and the truncated isoform have an intrinsic property to bind tapasin (Fig. 4A), implying that the truncated isoform can induce the dissociation of full-length US3 from tapasin via competition for binding to tapasin. To further examine whether the truncated isoform inhibits the interaction between full-length US3 and tapasin proteins, we used US3-GFP due to the advantage of it distinguishing the full-length US3-associated tapasin by using anti-GFP Ab instead of anti-US3 Ab. We cotransfected US3-GFP with or without the truncated isoform into HeLa cells and analyzed the amounts of tapasin binding to US3-GFP by coimmunoprecipitation experiments using anti-GFP Ab. Surprisingly, the amount of tapasin binding to US3-GFP was extremely reduced in the presence of the truncated isoform (Fig. 4B, top panel, compare lanes 2 and 3). The reduction of US3-GFP-associated tapasin by the truncated isoform is not due to a decrease in the levels of expressed US3-GFP and tapasin, because comparable amounts of US3-GFP and tapasin were expressed in each transfectant, as shown by immunoblotting (Fig. 4B, middle and bottom panels). The relation between full-length US3 and the truncated isoform was further investigated in 293T cells that express little tapasin (13). The 293T cells were cotransfected with cDNAs as indicated and analyzed by coprecipitation and immunoblot (Fig. 4C). Despite the comparable expression levels of US3-GFP and soluble tapasin (Fig. 4C, middle and bottom panels), the amount of US3-GFP coprecipitated with soluble tapasin was remarkably decreased in the presence of the truncated SS isoform (Fig. 4C, top panel, compare lanes 3 and 4).
The association of full-length US3 with tapasin causes the dissociation of tapasin from TAP complexes (31). Therefore, we investigated whether the truncated isoform could counteract the activity of full-length US3 that causes the dissociation of tapasin from TAP. The association of tapasin with TAP1 was blocked in HeLa cells transfected with full-length US3 alone (Fig. 4D, compare lanes 3 and 4). However, it was recovered in the cotransfectants with full-length US3 and the truncated isoform (Fig. 4D, compare lanes 4 and 6), suggesting that the truncated isoform counteracted the activity of full-length US3 in weakening the TAP/tapasin association. Based on the above-described results, we can conclude that the truncated isoform competes with full-length US3 for binding to tapasin, suggesting that the truncated isoform is a negative regulator of full-length US3.
The truncated isoform hinders full-length US3 in the inhibition of tapasin-dependent peptide optimization of MHC class I molecules. Full-length US3 inhibits peptide optimization of MHC class I peptide cargo, which is mainly mediated by tapasin (31). To test whether the truncated isoform could counteract the activity of full-length US3 in the inhibition of peptide optimization, we analyzed the thermostability of MHC class I complexes, which is greatly linked to the binding affinity of their peptide cargo (11, 29, 31, 35, 47). The transfected cells were either radiolabeled (Fig. 5A) or biotinylated (Fig. 5B) to differentiate between the newly synthesized and surface-expressed MHC class I heavy chains, treated with heat, and precipitated using conformation-dependent MAb W6/32. Quantification of W6/32-reactive class I heavy chains showed that the differences in the thermostability of MHC class I molecules were most prominent at 37°C. The thermostability of heavy chains was significantly reduced in the presence of full-length US3 alone at 37°C (Fig. 5A and B, second panels), whereas the truncated isoform alone failed to decrease the thermostability of heavy chains at 37°C (Fig. 5A and B, third panels). Interestingly, in the case of coexpression of full-length US3 and the truncated isoform, both surface and newly synthesized class I heavy chains were thermostable at 37°C in comparison with the thermostability of MHC class I molecules in either mock- or SS-transfected cells (Fig. 5A and B, fourth panels). Considered together, these results strongly suggest that the truncated isoform prevents full-length US3 from inhibiting the tapasin function of optimizing peptide cargo.
DISCUSSION
MHC class I molecules mediate the adaptive immune response by presenting non-self antigens to CTLs, followed by the elimination of target cells. The 22-kDa US3 glycoprotein is encoded within the HCMV US region genes that are related with immune escape against CTLs. This viral protein makes use of a dual mechanism to block the antigen presentation by MHC class I molecules. First, US3 directly attacks MHC class I molecules; that is, US3 physically binds to MHC class I molecules and sequesters them in the ER (1, 19). The second mechanism is a rather indirect action in which US3 targets tapasin, a component of peptide-loading complexes, and binds to it; this results in the inhibition of tapasin-dependent peptide loading and optimization (31). The dual mechanism may enable the virus to evade CTLs more efficiently, thereby contributing to the establishment of a persistent and latent infection. Since the US3 protein is expressed at the immediate-early phase of infection with no connection to de novo protein synthesis (44) and since NK cells rapidly respond to viral infections in the innate immune system, the US3-mediated down-regulation of the expression of MHC class I molecules on the cell surface renders the infected cells susceptible to NK lysis. This immediate-early protein of HCMV therefore requires tight regulation, both positive and negative, although the window of time when the virus encounters NK cells or CTLs is not known.
In this study, we demonstrate that the truncated isoform generated by an alternative splicing of the HCMV US3 gene transcript negatively regulates the action of full-length US3 in down-regulating the expression of MHC class I molecules via competing with full-length US3 for binding to a target molecule, tapasin. The relative abundance of US3 transcripts varies during HCMV infections; the full-length US3 transcript is most abundant early during infection, followed by singly spliced and doubly spliced transcripts (7, 24). Each of the encoded proteins might thus contribute to the viral life cycle in the infected host. Considering the sequential expression of US3 transcripts and our current observations, it is likely that full-length US3 and the truncated isoform have evolved to evade immune surveillance by CTLs and NK cells, respectively, during early infection. This mode of regulation, in which the function of a full-length form is controlled by a spliced truncate, has been reported in not only viral infection but also various biological contexts, for example, ICP0/ICP0R (36), HIF-1/HIF-1736 (15), and DNOS1/DNOS4 (37). The adenovirus E3/19K protein is a structural and functional homolog of full-length US3. It is an ER-resident type I transmembrane glycoprotein and inhibits MHC class I molecules from reaching the cell surface by two mechanisms targeting both tapasin and MHC class I molecules (5, 31). Similar to the action of the truncated US3 isoform on the full-length form of US3, the E3/19K molecule lacking the C-terminal ER retention motif forms a heterodimer with wild-type E3/19K and competes with E3/19K for Kd binding, thereby suppressing Kd retention by E3/19K (10).
The ability of US3 to associate with MHC class I molecules requires the transmembrane domain in addition to the luminal domain of US3 (21). Since the truncated isoform lacks the transmembrane domain that is responsible for interacting with MHC class I molecules (21), it is unlikely that the truncated isoform prevents full-length US3 from binding to the MHC class I molecules. This idea was supported by the observation that MHC class I molecules were coprecipitated only with full-length US3 but not with the truncated isoform and that the association of MHC class I molecules with full-length US3 was not reduced even in the presence of the truncated isoform (Fig. 3C).
The precise function of tapasin in MHC class I assembly is uncertain. It plays a critical role in bridging MHC class I molecules to TAP (30, 42), tethering them in the ER (16 34), and in turn enhancing peptide loading and optimization (17, 33, 47). Our study showed that not only the full-length US3 but also the truncated isoform are able to bind to tapasin (Fig. 4A). A notable observation was that the total amounts of tapasin associated with either US3 or SS proteins were not additive (Fig. 4A, lanes 3 to 5) and that in the presence of SS, the level of association between US3 and tapasin was remarkably reduced (Fig. 4B). Interestingly, the transport kinetics of class I molecules are similar in both US3/SS and US3-alone transfectants at the 30-min chase point but show clear differences between them at the 60-min chase (Fig. 2B). This might suggest that the inhibitory effect of the truncated isoform is sequentially turned on after the action of full-length US3. Since the truncated isoform does not inhibit the association of full-length US3 with class I molecules and since the soluble truncated isoform is more free to move and diffuse from tapasin than the membrane-bound full-length US3, the truncated isoform seems to lag behind in its action. Based on these results, we can conclude that the truncated isoform competes with full-length US3 for binding to tapasin and thereby nullifies the activity of US3 in inhibiting tapasin function.
The current study also provides clues for the functional determinants of US3 and tapasin. Our finding that the truncated US3 isoform binds to full-length tapasin (Fig. 4A and B) and that full-length US3 binds to soluble tapasin (Fig. 4C) indicates that the luminal domain of each protein mediates the interaction. Importantly, the truncated US3 isoform is able to bind to tapasin, but unlike full-length US3, it does not affect the transport of MHC class I molecules (Fig. 2), demonstrating that the mode of interaction between the truncated US3 isoform and tapasin is unproductive and that the transmembrane domain of the US3 protein is required for inhibiting tapasin function. The separated usages of the US3 domain are similar to the case of US11, another US region gene product, the luminal domain of which is sufficient for MHC class I binding, but the transmembrane domain is crucial for MHC class I heavy-chain dislocation (22, 23).
Transcription of the US3 gene is controlled by a complex network of viral regulators (6); however, the regulatory mechanism for spliced variants remains unclear. Full-length US3 may exhibit contradictory effects on the survival of the virus. While inactivation of full-length US3 contributes to avoiding NK lysis, it makes virus-infected cells vulnerable to lysis by CTLs. Thus, biosynthesis or activities of the truncated US3 isoform should be precisely controlled, for instance, by posttranslational modifications such as ubiquitination of ICP0R during productive infection (43). Considered together, our results suggest that in HCMV, the truncated variant of US3 acts as a negative regulator of full-length US3 activity during viral infection. This mode of regulation that we observed in HCMV could also be broadly involved in the regulation of activities of immunoevasive proteins in other viruses.
ACKNOWLEDGMENTS
We are grateful to Frank Momburg for providing tapasin cDNA.
This work was supported by a grant from the Korea Research Foundation (C00479) and by grant R01-2005-000-10235-0 from the KOSEF and a grant from the Research Center for Functional Cellulomics of the KOSEF.
REFERENCES
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.
Ahn, K., A. Gruhler, B. Galocha, T. R. Jones, E. J. Wiertz, H. L. Ploegh, P. A. Peterson, Y. Yang, and K. Fruh. 1997. The ER-luminal domain of the HCMV glycoprotein US6 inhibits peptide translocation by TAP. Immunity 6:613-621.
Andersson, M., S. Paabo, T. Nilsson, and P. A. Peterson. 1985. Impaired intracellular transport of class I MHC antigens as a possible means for adenoviruses to evade immune surveillance. Cell 43:215-222.
Beck, S., and B. G. Barrell. 1988. Human cytomegalovirus encodes a glycoprotein homologous to MHC class-I antigens. Nature 331:269-272.
Bennett, E. M., J. R. Bennink, J. W. Yewdell, and F. M. Brodsky. 1999. Cutting edge: adenovirus E19 has two mechanisms for affecting class I MHC expression. J. Immunol. 162:5049-5052.
Biegalke, B. J. 1999. Human cytomegalovirus US3 gene expression is regulated by a complex network of positive and negative regulators. Virology 261:155-164.
Biegalke, B. J. 1995. Regulation of human cytomegalovirus US3 gene transcription by a cis-repressive sequence. J. Virol. 69:5362-5367.
Biron, C. A., K. S. Byron, and J. L. Sullivan. 1989. Severe herpesvirus infections in an adolescent without natural killer cells. N. Engl. J. Med. 320:1731-1735.
Cosman, D., J. Mullberg, C. L. Sutherland, W. Chin, R. Armitage, W. Fanslow, M. Kubin, and N. J. Chalupny. 2001. ULBPs, novel MHC class I-related molecules, bind to CMV glycoprotein UL16 and stimulate NK cytotoxicity through the NKG2D receptor. Immunity 14:123-133.
Cox, J. H., J. R. Bennink, and J. W. Yewdell. 1991. Retention of adenovirus E19 glycoprotein in the endoplasmic reticulum is essential to its ability to block antigen presentation. J. Exp. Med. 174:1629-1637.
Fahnestock, M. L., I. Tamir, L. Narhi, and P. J. Bjorkman. 1992. Thermal stability comparison of purified empty and peptide-filled forms of a class I MHC molecule. Science 258:1658-1662.
Falk, C. S., M. Mach, D. J. Schendel, E. H. Weiss, I. Hilgert, and G. Hahn. 2002. NK cell activity during human cytomegalovirus infection is dominated by US2-11-mediated HLA class I down-regulation. J. Immunol. 169:3257-3266.
Ford, S., A. Antoniou, G. W. Butcher, and S. J. Powis. 2004. Competition for access to the rat major histocompatibility complex class I peptide-loading complex reveals optimization of peptide cargo in the absence of transporter associated with antigen processing (TAP) association. J. Biol. Chem. 279:16077-16082.
Fruh, K., Y. Yang, D. Arnold, J. Chambers, L. Wu, J. B. Waters, T. Spies, and P. A. Peterson. 1992. Alternative exon usage and processing of the major histocompatibility complex-encoded proteasome subunits. J. Biol. Chem. 267:22131-22140.
Gothie, E., D. E. Richard, E. Berra, G. Pages, and J. Pouyssegur. 2000. Identification of alternative spliced variants of human hypoxia-inducible factor-1alpha. J. Biol. Chem. 275:6922-6927.
Grandea, A. G., III, P. J. Lehner, P. Cresswell, and T. Spies. 1997. Regulation of MHC class I heterodimer stability and interaction with TAP by tapasin. Immunogenetics 46:477-483.
Grandea, A. G., III, and L. Van Kaer. 2001. Tapasin: an ER chaperone that controls MHC class I assembly with peptide. Trends Immunol. 22:194-199.
Hengel, H., J. O. Koopmann, T. Flohr, W. Muranyi, E. Goulmy, G. J. Hammerling, U. H. Koszinowski, and F. Momburg. 1997. A viral ER-resident glycoprotein inactivates the MHC-encoded peptide transporter. Immunity 6:623-632.
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.
Lee, S., B. Park, and K. Ahn. 2003. Determinant for endoplasmic reticulum retention in the luminal domain of the human cytomegalovirus US3 glycoprotein. J. Virol. 77:2147-2156.
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.
Lee, S. O., S. Hwang, J. Park, B. Park, B. S. Jin, S. Lee, E. Kim, S. Cho, Y. Kim, K. Cho, J. Shin, and K. Ahn. 2005. Functional dissection of HCMV US11 in mediating the degradation of MHC class I molecules. Biochem. Biophys. Res. Commun. 330:1262-1267.
Lilley, B. N., D. Tortorella, and H. L. Ploegh. 2003. Dislocation of a type I membrane protein requires interactions between membrane-spanning segments within the lipid bilayer. Mol. Biol. Cell 14:3690-3698.
Liu, W., Y. Zhao, and B. Biegalke. 2002. Analysis of human cytomegalovirus US3 gene products. Virology 301:32-42.
Ljunggren, H. G., and K. Karre. 1990. In search of the ‘missing self’: MHC molecules and NK cell recognition. Immunol. Today 11:237-244.
Lodoen, M. B., and L. L. Lanier. 2005. Viral modulation of NK cell immunity. Nat. Rev. Microbiol. 3:59-69.
Lybarger, L., X. Wang, M. Harris, and T. H. Hansen. 2005. Viral immune evasion molecules attack the ER peptide-loading complex and exploit ER-associated degradation pathways. Curr. Opin. Immunol. 17:71-78.
Mocarski, E. S., Jr. 2002. Immunomodulation by cytomegaloviruses: manipulative strategies beyond evasion. Trends Microbiol. 10:332-339.
Neefjes, J. J., G. J. Hammerling, and F. Momburg. 1993. Folding and assembly of major histocompatibility complex class I heterodimers in the endoplasmic reticulum of intact cells precedes the binding of peptide. J. Exp. Med. 178:1971-1980.
Ortmann, B., J. Copeman, P. J. Lehner, B. Sadasivan, J. A. Herberg, A. G. Grandea, S. R. Riddell, R. Tampe, T. Spies, J. Trowsdale, and P. Cresswell. 1997. A critical role for tapasin in the assembly and function of multimeric MHC class I-TAP complexes. Science 277:1306-1309.
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.
Petersen, J. L., C. R. Morris, and J. C. Solheim. 2003. Virus evasion of MHC class I molecule presentation. J. Immunol. 171:4473-4478.
Purcell, A. W., J. J. Gorman, M. Garcia-Peydro, A. Paradela, S. R. Burrows, G. H. Talbo, N. Laham, C. A. Peh, E. C. Reynolds, J. A. Lopez De Castro, and J. McCluskey. 2001. Quantitative and qualitative influences of tapasin on the class I peptide repertoire. J. Immunol. 166:1016-1027.
Schoenhals, G. J., R. M. Krishna, A. G. Grandea III, T. Spies, P. A. Peterson, Y. Yang, and K. Fruh. 1999. Retention of empty MHC class I molecules by tapasin is essential to reconstitute antigen presentation in invertebrate cells. EMBO J. 18:743-753.
Schumacher, T. N., M. T. Heemels, J. J. Neefjes, W. M. Kast, C. J. Melief, and H. L. Ploegh. 1990. Direct binding of peptide to empty MHC class I molecules on intact cells and in vitro. Cell 62:563-567.
Spatz, S. J., E. C. Nordby, and P. C. Weber. 1996. Mutational analysis of ICP0R, a transrepressor protein created by alternative splicing of the ICP0 gene of herpes simplex virus type 1. J. Virol. 70:7360-7370.
Stasiv, Y., B. Kuzin, M. Regulski, T. Tully, and G. Enikolopov. 2004. Regulation of multimers via truncated isoforms: a novel mechanism to control nitric-oxide signaling. Genes Dev. 18:1812-1823.
Tenney, D. J., L. D. Santomenna, K. B. Goudie, and A. M. Colberg-Poley. 1993. The human cytomegalovirus US3 immediate-early protein lacking the putative transmembrane domain regulates gene expression. Nucleic Acids Res. 21:2931-2937.
Tomasec, P., V. M. Braud, C. Rickards, M. B. Powell, B. P. McSharry, S. Gadola, V. Cerundolo, L. K. Borysiewicz, A. J. McMichael, and G. W. Wilkinson. 2000. Surface expression of HLA-E, an inhibitor of natural killer cells, enhanced by human cytomegalovirus gpUL40. Science 287:1031.
Tortorella, D., B. E. Gewurz, M. H. Furman, D. J. Schust, and H. L. Ploegh. 2000. Viral subversion of the immune system. Annu. Rev. Immunol. 18:861-926.
Townsend, A., and H. Bodmer. 1989. Antigen recognition by class I-restricted T lymphocytes. Annu. Rev. Immunol. 7:601-624.
Turnquist, H. R., J. L. Petersen, S. E. Vargas, M. M. McIlhaney, E. Bedows, W. E. Mayer, A. G. Grandea III, L. Van Kaer, and J. C. Solheim. 2004. The Ig-like domain of tapasin influences intermolecular interactions. J. Immunol. 172:2976-2984.
Weber, P. C., S. J. Spatz, and E. C. Nordby. 1999. Stable ubiquitination of the ICP0R protein of herpes simplex virus type 1 during productive infection. Virology 253:288-298.
Weston, K. 1988. An enhancer element in the short unique region of human cytomegalovirus regulates the production of a group of abundant immediate early transcripts. Virology 162:406-416.
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.
Williams, A. P., C. A. Peh, A. W. Purcell, J. McCluskey, and T. Elliott. 2002. Optimization of the MHC class I peptide cargo is dependent on tapasin. Immunity 16:509-520.(Jinwook Shin, Boyoun Park)