The JI
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     
 

The Journal of Immunology, 2008, 180, 969-979
Copyright © 2008 by The American Association of Immunologists, Inc.
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Maffei, M.
Right arrow Articles by Ciccone, E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Maffei, M.
Right arrow Articles by Ciccone, E.

Human Cytomegalovirus Regulates Surface Expression of the Viral Protein UL18 by Means of Two Motifs Present in the Cytoplasmic Tail1

Massimo Maffei2,*, Fabio Ghiotto2,*, Marzia Occhino*, Maria Bono*, Amleto De Santanna{dagger}, Lorenzo Battini{ddagger}, G. Luca Gusella{ddagger}, Franco Fais*, Silvia Bruno3,* and Ermanno Ciccone*

* Human Anatomy Section and {dagger} Histology Section, Department of Experimental Medicine, University of Genoa, Genova, Italy; and {ddagger} Division of Renal Medicine, Mount Sinai School of Medicine, New York, NY 10029


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 References
 
UL18 is a trans-membrane viral protein expressed on human cytomegalovirus (HCMV)-infected cells, and its surface expression determines the interaction of infected cells with lymphocytes expressing the CD85j (LIR-1/ILT2) receptor. We previously showed that the UL18–CD85j interaction elicits activation of T lymphocytes. However, in in vitro cell models UL18 displays mostly undetectable surface expression. Thus, we asked how surface expression of UL18 is regulated. Domain-swapping experiments and construction of specific mutants demonstrated that two motifs on its cytoplasmic tail, homologous to YXX{Phi} and KKXX consensus sequences, respectively, are responsible for impairing UL18 surface expression. However, the presence of the whole HCMV genome, granted by HCMV infection of human fibroblasts, restored surface expression of either UL18 or chimeric proteins carrying the UL18 cytoplasmic tail, starting from the third day after infection. It is of note that the two motifs responsible for cytoplasmic retention are identical in all 17 HCMV strains examined. We disclosed a control mechanism used by the HCMV to regulate the availability of UL18 on the infected-cell surface to allow interaction with its ligand on T and NK cells.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 Disclosures
 References
 
The human cytomegalovirus (HCMV)4has coexisted with eukaryotic cells for millions of years and has evolved multiple mechanisms both to evade the immune system and to avoid the complete clearance of the viral genome. Depending on the geographical area, 70–90% of the population in the world carry the HCMV virus integrated in their genomes. When primary infection occurs in immunocompetent hosts, immune control mechanisms effectively prevent overt disease and terminate virus replication (1). However, ultimate clearance of the viral genome is not achieved, and the virus remains lifelong at specific sites of its host in a nonproductive form. Under circumstances that associate with relevant impairment of host cell-mediated immunity, the viral replication cycle can be reactivated and results in relapsing disease that can be fatal. Thus, a fine-tuning and balancing of immune evasion mechanisms, on the one side, and immune activation, on the other side, made it possible for the HCMV to avoid its complete elimination after the primary infection and eventually to exist in an occult state. Viral immune evasion mechanisms that down-regulate immune recognition by cytolytic T lymphocytes (2, 3) and NK cells (4, 5, 6, 7) have been elucidated. On the other side, NK and T lymphocytes are crucial immune cells for the control of infection, both with human and murine CMV (8, 9, 10). This demonstrates the existence of viral-mediated immune activation mechanisms that have evolved to avoid death of the host and consequently of the virus. Altogether, it is clear that the HCMV has developed a vast array of immunostimulatory and immunosuppressive functions that participate in the maintenance of a delicate balance between effective immunity and immune subversion through a complex interplay between receptors on immune cells and virus-encoded genes.

One part of this array is UL18. UL18 displays a structural homology with HLA class I molecules as well as a significant amino acid identity (11). It associates with β2-microglobulin (12), and the stability of the trimeric complex depends on the presence of a loaded peptide (13). It is a late HCMV Ag in that its transcription occurs from 54 to at least 120 h postinfection (14) and is not essential for HCMV replication (15).

The receptor for UL18 is CD85j/LIR-1/ILT2 (CD85j), a molecule of the Ig superfamily (16). It is a transmembrane molecule with four cytoplasmic ITIMs that mediate transduction of inhibitory signals (17), and it is expressed by lymphoid and myelomonocytic cells (18, 19, 20). In addition to UL18, CD85j recognizes broadly MHC class I molecules (16, 21).

When CD85j is engaged by specific cross-linking Abs or by HLA class I molecules, it delivers negative signals to NK and T lymphocytes (18, 19, 22, 23, 24).

Instead, when CD85j is bound by UL18, its inhibitory activity is a matter of debate. NK cell-mediated lysis was seen to be inhibited (4, 25, 26) or increased (27), depending on the experimental setting. With respect to T lymphocytes, two studies have addressed the effect of CD85j engagement by UL18, and both show that UL18 mediates an activating signal, in one case through its interaction with CD85j (28), and in the other case by still unexplained mechanisms (29). Whatever the mechanism, the UL18–CD85j interaction certainly plays a pivotal role in the immune response to HCMV infection. CD85j displays an affinity for UL18 that is >1000-fold higher than for cellular MHC class I molecules (30, 31), and its expression is up-regulated on HCMV-specific T cells (32, 33) and on T and NK cells of lung-transplanted patients later developing HCMV disease (29, 34). Binding of CD85j to UL18 requires surface expression of the viral protein. However, we and other groups (35) have observed that UL18 accumulates in the cytoplasm and displays no detectable surface immunofluorescence in in vitro cell models, such as HCMV-infected fibroblasts, UL18-transfected tumor cell lines, or UL18-transduced fibroblasts. Only in cases of high multiplicity of infection (MOI) by HCMV or UL18-carrying adenovirus or vaccinia virus has spillover of UL18 from the cytoplasm to the cell surface been observed (28, 35, 36).

Accordingly, we focused our attention on regulatory mechanisms that control trafficking and surface expression of UL18. We searched for domains on UL18 that could be responsible for intracellular retention and verified whether their interplay with virus-encoded genes may regulate UL18 surface expression.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 Disclosures
 References
 
Cells and transfections

HeLa, HEK, COS-7, and Sp2/0 cell lines were obtained from the American Type Culture Collection and maintained in RPMI 1640 medium with GlutaMAX (Invitrogen) supplemented with penicillin, streptomycin, and 10% FCS (Invitrogen).

The human foreskin fibroblast (HFF) line Hs-27 (kindly provided by L. Lanfranconi, IFOM-IEO, Milan, Italy) was cultured in DMEM supplemented with 20% FCS and used at the 21st passage.

Transfections were performed using Lipofectamine 2000 (Invitrogen) according to the manufacturer’s instructions. To obtain stable transfectants, antibiotics were added 24 h post transfection: 500 µg/ml G418 (Invitrogen) for pIRESneo, 1 µg/ml puromycin (Sigma-Aldrich) for pIRESpuro3, and 400 µg/ml G418 for phuPSIg2.0. The Sp2/0-stable transfectant was cloned by limiting dilution, and a CD85jFc-producing clone was selected.

DNA constructs

pIRESneo/UL18. The UL18 gene was cloned in an expression vector amplifying the coding region by PCR using, as a template, the supernatant of fibroblasts infected with the laboratory strain AD169. Primers used were: forward primer UL18CPF GGAATTCACCATGATGACAATGTGGTGTCTGACG and reverse primer UL18CPR CGGGATCCTCATGACGACCGGACCTTGCG. The product was digested with EcoRI and BamHI restriction enzymes and cloned into the pIRESneo vector (BD Biosciences/Clontech).

pIRESneo/HLA-A2. HLA-A*0201 (HLA-A2) cDNA (kindly provided by R. Biassoni, G. Gaslini Institute, Genova, Italy) was amplified by PCR using the forward primer A2CPF GGAATTCACCATGGCCGTCATGGCGCCCCGAAC and the reverse primer A2CPR CGGGATCCTCACACTTTACAAGCTGTGAGAGAC. The product was purified and digested with EcoRI and BamHI restriction enzymes and cloned into the pIRESneo vector.

Swapping domain chimeric proteins. HLA-A2/UL18 chimeric proteins were constructed using a two-step PCR-based strategy.

The chimeric protein with the {alpha}1 and {alpha}2 domains of UL18 and the {alpha}3 domain, stalk, transmembrane region, and cytoplasmic tail of HLA-A2 (HLA-A2/UL18{alpha}1{alpha}2) was constructed amplifying UL18 with the primers UL18CPF and {alpha}2UL18/{alpha}3HLA-A2R GGGCGTCCGTGCGGACGGGGGGTTGAAAC, and amplifying HLA-A2 with the primers {alpha}2UL18/{alpha}3HLA-A2F GTTTCAACCCCCCGTCCGCACGGACGCCC and A2CPR. PCR products were controlled by 1.5% agarose gel electrophoresis and used to perform a second PCR step. One microliter of both products was mixed and amplified using primers UL18CPF and A2CPR. The product thus obtained was cloned using the TOPO TA Cloning Kit (Invitrogen), and positive clones were analyzed by sequencing. A clone with the expected sequence was digested with EcoRI and BamHI restriction enzymes and subcloned into pIRESpuro3 vector (BD Biosciences/Clontech). The following chimeric proteins were constructed similarly: one chimeric protein composed of the HLA-A2 protein carrying the {alpha}3 domain of UL18 (HLA-A2/UL18{alpha}3) using the primers A2CPF, {alpha}2HLA-A2/{alpha}3UL18R GATGGGGATGGTGGGCTGGGAAGACGGCTCAGGTGAGGTAACGCTGATGG, {alpha}2HLA-A2/{alpha}3UL18F TACCTGGAGAACGGGAAGGAGACGCTGCAGCACCCAGTGGTAAAGGGCGGTG, and A2CPR; and one chimeric protein composed of the {alpha}1, {alpha}2, and {alpha}3 domains of HLA-A2 and the stalk, transmembrane region, and cytoplasmic tail of UL18 (HLA-A2/UL18CY) using the primers A2CPF, {alpha}2HLA-A2/{alpha}3UL18R TACCACTGGGTGGGTCTGCAGCGTCTCCTT, {alpha}3HLA-A2/TMUL18F CTCACCCTGAGATGGGACGACAGTTCCTCG, and UL18CPR. The clones were subcloned into pIRESneo or pIRESpuro3 expression vectors.

HLA-A2/UL18{alpha}3 C279S. To obtain this construct a mutation of the C at position 279 to S was introduced with the QuickChange II site-directed mutagenesis kit (Stratagene) according to the manufacturer’s instructions and using the following primers: forward primer MutC279SF GTTACGTAGCCATCTTTAGCAATCAAAACTACACC and reverse primer MutC279SR GGTGTAGTTTTGATTGCTAAAGATGGCTACGTAAC.

All of the above constructs were tagged at the C terminus with an enhanced GFP (EGFP) by subcloning into pEGFP-N1 vector.

Mutants of the HLA-A2/UL18CY construct were produced by PCR site-directed mutagenesis using as a template the HLA-A2/UL18CY construct. For all mutants the same forward primer A2CPF was used, coupled with a specific reverse primer. The HLA-A2/UL18CY Y362A mutant was obtained using the reverse primer MutY362ACPR TAGGATCCTCATGACGACCGGACCTTGCGAGCGCGCCACGC; the HLA-A2/UL18CY K364A mutant was produced using the reverse primer MutK364ACPR TAGGATCCTCATGACGACCGGACCGCGCGATAGCGCCACGC; and the HLA-A2/UL18CY Y362A/K364A mutant was constructed using the reverse primer MutY362A/K364ACPR TAGGATCCTCATGACGACCGGACCGCGCGAGCGCGCCACGC. Products were cloned with the TOPO TA cloning kit. Clones were sequenced and then digested with EcoRI and BamHI restriction enzymes and cloned into pIRESpuro3 vector.

The UL18 mutant UL18 Y362A/K364A was tagged at the C terminus with an EGFP by subcloning into pEGFP-N1 vector.

CD85jFc. The CD85jFc recombinant protein was obtained by cloning the CD85j coding region from aa 1 to aa 419 into the phuPSIg2.0 vector that contains a genomic Fc portion (hinge, intron, CH2, intron, and CH3) of human IgG1 Ig.

HiFi Platinum Taq (Invitrogen) was used for all PCR. DNA ligation kit 2.1 (Takara Shuzo) was used in all ligation reactions. For all constructs, positive clones were analyzed by sequencing to exclude mutations and to verify the frame maintenance. Big Dye Terminator 1.1 (Applied Biosystems) was used for all sequences that were analyzed using an ABI PRISM310 genetic analyzer (Applied Biosystems).

Lentiviral constructs

All cDNA of tagged, chimeric, and mutant contructs were subcloned, starting from the constructs described above, into the VVEW/BB: UL18/EGFP and HLA-A2/UL18CY were digested and subcloned NheI-NotI; and HLA-A2/UL18C Y362A, HLA-A2/UL18C K364A, HLA-A2/UL18C Y362A, and K364A were digested and subcloned NheI-BamHI.

Lentiviral vectors

To construct the self-inactivating lentivector VVEW/BB, the BiP/blasticidin cassette was inserted downstream of the EF-1 promoter in the VVEW vector (37). The blasticidin gene was obtained by digesting the pEF6/V5-HisA plasmid (Invitrogen) with NcoI and PmlI, while the BiP sequence was excised from the VVPW/BE vector (37) by digestion with NotI and NcoI. A triple ligation was performed to insert these fragments into the backbone VVEW vector that had been digested with XhoI, blunt-ended by Klenow polymerase, and then digested with NotI. Lentivirus production and in vitro transduction were performed as described (38).

HCMV viruses

The HCMV strain AD169 was purchased from American Type Culture Collection. The HCMV strain RV798 (US2-US11 deleted) was kindly provided by T. R. Jones (Wyeth) and A. E. Campbell (Eastern Virginia Medical School, Norfolk, VA) (42). Infections were performed as described (28) at 1 MOI.

Antibodies

The anti-UL18mAb M71 (Amgen) was used at 4 µg/ml for immunofluorescence and at 2.5 µg/ml for Western blot. Supernatants used were 10C7 and BB7.2 (American Type Culture Collection); 10C7 was also used as ascites for immunoprecipitations (kindly provided by P. Bjorkman, Caltech, Pasadena, CA). GM130 (BD Biosciences) and calreticulin (Affinity BioReagents) were used at 1:100 dilution. Anti-β-actin was obtained from Santa Cruz Biotechnology. Secondary fluorochrome-conjugated Abs were purchased from Molecular Probes or from Southern Biotechnology Associates. Secondary HRP-conjugated Abs for biochemical analysis were from Southern Biotechnology Associates or Kirkegaard & Perry Laboratories.

Surface immunofluorescence

Depending on the type of fluorescence measurement, that is, whether by flow cytometry or by fluorescence microscopy, cells were either detached with trypsin from the plasticware bottom and stained as cell suspensions or directly grown on glass coverslips and stained in adherence, respectively. In both cases, exposure to the first Ab, to the secondary fluorochrome-conjugated Abs, and washes were all performed at 4°C. A short fixation with 1% paraformaldehyde (PFA) was performed after the last washing, and surface fluorescence was measured either on single cell suspensions by a FACSCalibur flow cytometer (BD Biosciences) or on coverslips mounted with Prolong antifading mounting medium (Molecular Probes) either on a Leica DM-IRE2 fluorescence microscope or on a Leica SP2-AOBS confocal microscope (Leica Microsystems).

Immunofluorescence-based endocytosis assay

Cells growing on glass coverslips were stained with the primary Ab for 30 min at 4°C, extensively washed with ice-cold PBS, and then incubated at 37°C in prewarmed complete tissue culture medium for 30 min before fixation with 2% PFA. The latter procedure allows internalization of the surface Ag-Ab complex, if the Ag bears internalization sequences (39). After fixation, coverslips were washed with PBS, incubated with 50 mM NH4Cl in PBS for 15 min, washed again, permeabilized with 0.05% Triton X-100 for 5 min, and stained with the secondary fluorochrome-conjugated Ab. Mounted coverslips were imaged on a Leica DM-IRE2 fluorescence microscope or on a Leica SP2-AOBS confocal microscope.

Intracellular immunofluorescence

Cells grown on coverslips were fixed either with 100% ice-cold methanol for 3 min (M71 mAb) or with 3% PFA (8 min at room temperature) followed by permeabilization with 0.01% Triton X-100 (2 min at room temperature) (CD85jFc and all other Abs). After washing, cells were stained with the primary Ab, washed with PBS plus 1% BSA, and incubated for 40 min at room temperature with the fluorochrome-conjugated secondary Ab. For multicolor staining, incubation with other primary and respective secondary Abs followed sequentially. In the case of surface and intracellular immunofluorescence on the same sample, after incubation with primary and secondary Abs on live cells at 4°C for surface staining, cells underwent PFA-Triton X-100 treatment, except for M71 mAb, where surface-stained cells were subjected to a short pulse (30 s) with 1.5% PFA before methanol fixation.

Surface biotinylation, immunoprecipitation, endoglycosidase H (EndoH), and peptide N-glycosidase F (PNGaseF) digestions

Biotinylation of cell surface proteins was performed as described (28). Immunoprecipitations were conducted according to standard procedures (28). Protein A-agarose was used for BB7.2 immunoprecipitation (IP) and protein G-agarose for 10C7 IP and anti-β-actin. Samples were analyzed by SDS electrophoresis in 12% polyacrylamide gels under reducing conditions. Western blots were developed using either anti-H chain HCA2 mAb (provided by H. Ploegh (Harvard University, Boston, MA) and P. Giacomini (Regina Elena Cancer Institute, Rome)) followed by goat anti-mouse HRP-conjugated antiserum (Southern Biotechnology Associates) for BB7.2 IP, HRP-conjugated streptavidin (Santa Cruz Biotechnology) for 10C7 and anti-β-actin IP. For EndoH digestions of BB7.2 immunoprecipitates, 5 µl of 5x EndoH digestion buffer (50 mM sodium acetate (pH 5.2) in water solution) with 10 mU of enzyme was added to the beads. For PNGaseF digestions of BB7.2 immunoprecipitates, 5 µl of 5x PNGaseF buffer (100 mM sodium phosphate, 9% NaCl, 5% Nonidet P-40, pH 7.4) with 1 unit of enzyme was added to the beads. Digested proteins were incubated for 17 h at 37°C before analysis by SDS electrophoresis. Immune complexes were detected with a SuperSignal West Pico chemiluminescent substrate (Pierce Biotechnology).

Genomic DNA extraction and UL18 amplification

Genomic DNA was extracted from HCMV-seropositive donors. Briefly, 3 ml of peripheral venous blood was diluted with an equal volume of PBS, layered over 3 ml of Lymphoprep (Axis-Shield), and spun at 800 rpm for 20 min. Mononuclear cells were washed with PBS. DNA was extracted with the GeneElute mammalian genomic DNA miniprep kit (Sigma-Aldrich) according to the manufacturer’s instructions. Genomic DNA was amplified using the forward primer UL18–5'F2 CGCCATGATGACAATGTGGTG and the reverse primer UL18–3'R2 GCGTCGCGTGAGAAACATGAC. After 50 cycles of amplification the product was run on a 1% agarose gel. Positive samples were purified and sequenced.

Bioinformatic analysis

The amino acid sequence of the UL18 cytoplasmic tail was analyzed using the PSORTII program (http://www.psort.org/) (40) and the ELM program (http://www.elm.eu.org/) (41).


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 Disclosures
 References
 
Viral UL18 glycoprotein is mostly distributed in the endoplasmic reticulum (ER) and cis-Golgi

HFFs transduced with UL18-GFP and subjected to immunofluorescence staining by the anti-UL18 mAb M71 display a significant intracellular distribution (Fig. 1A), mainly residing in the ER and the cis-Golgi (Fig. 1B). UL18 accumulation in these compartments was demonstrated previously by biochemical analysis of the protein sensitivity to EndoH (35). Conversely, UL18 was undetectable by immunofluorescence on the cell surface (Fig. 1A). The same results were achieved with other in vitro UL18-expressing cell systems, such as UL18-transfected HeLa cells or HFFs infected at 1 MOI by AD169 or by the AD169-derived deletion mutant RV798 (42), either by using M71 or the recombinant protein CD85jFc for immunofluorescence (not shown).


Figure 1
View larger version (75K):
[in this window]
[in a new window]

  		FIGURE 1. UL18 is not detected by immunofluorescence on the plasma membrane. A, HFFs transduced with UL18-GFP-carrying lentiviral vectors were subjected to surface staining with the recombinant protein CD85jFc and Alexa 633-conjugated goat anti-human IgG, followed by a pulse with PFA and fixation with methanol, and then subjected to intracellular staining with M71 mAb and Alexa 546-conjugated GAM IgG. Although intracellular M71 fluorescence perfectly overlaps GFP expression pattern, no surface staining is observed. However, when surface Ag-Ab complex was allowed to be endocytosed by a 30-min incubation of surface-stained cells at 37°C (see Materials and Methods: Immunofluorescence-based endocytosis assay), a significant punctuate fluorescence appeared. B, UL18-GFP transduced HFFs were fixed with PFA, permeabilized with Triton X-100, and subjected to intracellular staining with GM130 plus Alexa 546-conjugated GAM IgG (virtual red color) and calreticulin plus Alexa 633-conjugated goat anti-rabbit IgG (virtual blue color). Two regions of interest are magnified in the second and third rows. Merging images show co-localization of UL18 with the ER and the cis-Golgi, and white pixels shown in merged region-of-interest images correspond to areas with significant co-localization (Leica CF2D software). C, UL18 is detected on the cell surface by biochemical analysis. Left, Surface biotin-labeled proteins from HeLa cells transfected with pIRES/UL18 or with pIRES alone were immunoprecipitated by 10C7 and immunoblotted by HRP-streptavidin. In addition to an unspecific 101-kDa band present in both samples, a 69-kDa band corresponding to the low glycosylated form of UL18 is derived from the surface of UL18-transfected cells. Middle, Control for cell surface specificity. Surface biotin-labeled proteins from pIRES/UL18-transfected HeLa cells were immunoprecipitated (IP) by anti-β-actin Ab and immunoblotted (IB) either by HRP-streptavidin or anti-β-actin Abs. No band is observed on the streptavidin blot, thus excluding cytoplasmic contaminants. Right, Western blot of total cell lysate revealed by M71 displays two bands corresponding to the 69-kDa, low glycosylated form and the 116-kDa, high glycosylated form.

 
We asked ourselves whether the UL18 molecules are unable to reach the cell membrane, or whether they are able to egress toward the cell surface but undergo rapid internalization. This latter mechanism would shorten significantly their residency time on the cell surface. To ascertain this, we conducted immunofluorescence-based endocytosis assays (see Materials and Methods). Briefly, CD85jFc was administered to cells at 4°C for 30 min, which allows binding to UL18 (if present). After removal of CD85jFc from the medium, cells were cultured for another 30 min at 37°C, then fixed, permeabilized, and stained with fluorochrome-associated secondary Abs. Negative fluorescence would have indicated that no UL18 molecules were present on the cell surface during exposure to CD85jFc. On the contrary, we observed an intracellular punctuated fluorescence pattern (Fig. 1A), suggesting that UL18 molecules were indeed reaching the cell membrane during incubation with CD85jFc, but then underwent internalization during the subsequent 30-min culture and became visible due to the accumulation of the UL18/CD85jFc complex in intracellular compartments, such as endosomes.

The presence of a small amount of UL18 molecules on the cell membrane is also demonstrated on UL18-transfected HeLa cells by immunoprecipitation of surface-labeled proteins 48 h after transfection (Fig. 1C). The anti-UL18 mAb 10C7 was used because M71 did not work in our hands for immunoprecipitation experiments. In addition to a nonspecific band of 101 kDa observed also in cells transfected with the vector alone, a 69-kDa band corresponding to the low glycosylated form of UL18 is detected (Fig. 1C, left), not derived from cytoplasmic contaminants (Fig. 1C, middle). Western blots of total lysates revealed with M71 (Fig. 1C, right) showed two bands corresponding to the 69-kDa, low glycosylated form and the 116-kDa, high glycosylated form of UL18.

Altogether, the data indicate that UL18 expression is mostly confined in intracellular compartments. Molecules that reach the cell membrane are few in number and/or their residency time is very short.

The cytoplasmic region of UL18 regulates its surface expression

To understand the cytoplasmic retention of UL18, domain-swapping experiments were conducted. The HLA-A*0201 (HLA-A2) molecule was chosen as partner, due to its structural homology with UL18 and because it is physiologically expressed on the cell surface. Three different hybrid molecules were constructed in which either the extracellular domains {alpha}1{alpha}2 or {alpha}3 or the stalk-transmembrane-cytoplasmic region of UL18 was substituted to the respective domains of HLA-A2 (Fig. 2). HeLa cells, which are HLA-A2 negative, as demonstrated by HLA genotyping and by the negative staining with the HLA-A2 allele-specific BB7.2 mAb (Fig. 3B), were transfected with GFP-tagged expression vectors carrying the HLA-A2 cDNA or the hybrid molecules, namely HLA-A2/UL18 stalk-transmembrane-cytoplasmic regions (herein called HLA-A2/UL18CY), HLA-A2/UL18{alpha}1{alpha}2, or HLA-A2/UL18{alpha}3. For determining the Ab specificity for each of the hybrid proteins, the following Abs or recombinant chimeric proteins were tested: the BB7.2 mAb specific for the {alpha}2 domain of HLA-A2, the M71 mAb raised against UL18 and specific for its {alpha}1{alpha}2 domain, and the recombinant protein CD85jFc. According to intracellular immunofluorescence analysis, the most appropriate Abs, in terms of specificity and sensitivity, for each of the hybrids were: BB7.2 mAb for HLA-A2/UL18CY, M71 for HLA-A2/UL18{alpha}1{alpha}2, and CD85jFc for HLA-A2/UL18{alpha}3. Note that HLA-A2/UL18{alpha}3 was first undetectable by CD85jFc, despite the presence of β2-microglobulin and the UL18 {alpha}3 domain, which is known to contain the primary site for UL18 binding to CD85jFc (30). Analysis of the amino acid sequence of the {alpha}3 domain of UL18 by means of prediction model programs revealed the presence of one cysteine (at position 279 in the wild-type UL18 molecule) available for disulfide bonds (51), possibly interfering, in the hybrid molecule, with proper epitope recognition by CD85jFc. To verify this hypothesis, we mutated the cysteine to serine in HLA-A2/UL18{alpha}3 and observed a significant immunofluorescence with CD85jFc (not shown). Additionally, HeLa transfected with unmutated HLA-A2/UL18{alpha}3 and processed for CD85jFc intracellular immunofluorescence after 20 min incubation with 20 mM 2-ME displayed a significant fluorescence signal (not shown). Therefore, experiments on HLA-A2/UL18{alpha}3 transfectants were always conducted after 2-ME treatment. When the proper detection reagents were identified for each of the hybrid molecules, surface immunofluorescence and confocal microscopy were performed on transient GFP-tagged transfectants to find out which domain of UL18 was responsible for impairing HLA-A2 surface expression. Results are displayed in Fig. 2. All hybrids retained surface expression, except for HLA-A2/UL18CY that was detected by immunofluorescence intracellularly (not shown) but not on the plasma membrane. No surface BB7.2 fluorescence was observed when HLA-A2/UL18CY transfectants were exposed to 2-ME. These results suggest that the portion of UL18 containing the stalk, the transmembrane region, and the cytoplasmic tail might bear sequences involved in the regulation of UL18 surface expression.


Figure 2
View larger version (64K):
[in this window]
[in a new window]

  		FIGURE 2. The intracellular region of UL18 is responsible for cytoplasmic retention of the viral protein. Hybrid HLA-A2/UL18 molecules were constructed by substituting regions of HLA-A2 with the respective regions of UL18, as schematically drawn. The hybrids were: HLA-A2 with the stalk, transmembrane region, and cytoplasmic tail of UL18 (HLA-A2/UL18CY); and HLA-A2 with extracellular {alpha}1 and {alpha}2 domains of UL18 (HLA-A2/UL18{alpha}1{alpha}2) and with an extracellular {alpha}3 domain of UL18 (HLA-A2/UL18{alpha}3). Wild-type HLA-A2 and hybrids were GFP-tagged and transiently transfected into HeLa cells, which do not express constitutively the HLA-A*0201 allele. Their plasma membrane expressions, as measured by immunofluorescence and confocal microscopy, are shown. Depending on the hybrid protein expressed by the cells, different Abs had to be used, as shown. Results indicate that the stalk-transmembrane-cytoplasmic region of UL18 impairs surface expression of HLA-A2.

 

Figure 3
View larger version (58K):
[in this window]
[in a new window]

  		FIGURE 3. Two sequences on the cytoplasmic tail of UL18 are responsible for its rapid internalization/endocytosis and ER retention/retrieval, respectively. A, Schematic representation of the hybrid molecule HLA-A2/UL18CY that bears extracellular {alpha}1, {alpha}2, and {alpha}3 domains of HLA-A2 and stalk, transmembrane region, and cytoplasmic tail of UL18. On the amino acid sequence of the UL18 cytoplasmic tail, the two sequences under investigation are highlighted. Sequence YRKV from aa 362 to aa 365 is homologous to sequence YXX{Phi} that represents a tyrosine-based motif for rapid internalization and endocytosis. Sequence KVRS from aa 364 to aa 367 is homologous to the KKXX consensus signal for endoplasmic reticulum retention retrieval signal. Mutations that abrogate one or both consensus sequences are indicated. B, Mutation of the two identified motifs abrogates internalization and leads to surface expression of the hybrid protein HLA-A2/UL18CY. HeLa cells were either not transfected or stably transfected with HLA-A2, with HLA-A2/UL18CY, or with HLA-A2/UL18CY mutated at the positions indicated in panel A. Cells were analyzed for surface immunofluorescence performed at 4°C with BB7.2 mAb plus FITC-conjugated goat GAM IgG, either by flow cytometry (left columns) or fluorescence microscopy (middle columns). Surface BB7.2-stained cells were also analyzed after 30 min incubation at 37°C (right columns) to determine internalization of the hybrid proteins complexed to BB7.2 (see Materials and Methods: Immunofluorescence-based endocytosis assay). HLA-A2 transfectants display surface expression of the protein, but not cells transfected with HLA-A2/UL18CY or the Y362A mutant. The K364A mutation reconstitutes surface expression, which is further augmented for the double mutant Y362A,K364A. Internalization experiments suggest that, although almost undetectable by surface staining, a low amount of HLA-A2/UL18CY must have reached the cell surface (right column). C, HLA-A2, HLA-A2/UL18CY, and mutants were subjected to biochemical analyses. BB7.2-immunoprecipitated proteins, undigested or digested with EndoH or PNGaseF, were analyzed by SDS electrophoresis in 12% polyacrylamide gels, under reducing conditions, followed by immunoblotting with the anti-HLA class I H chain HCA2 mAb.

 
Two amino acid sequences on the cytoplasmic tail of UL18 may be responsible for its cytoplasmic retention and re-internalization

The amino acid sequence of UL18 from the AD169 strain has been analyzed using the PSORTII and ELM programs that allow prediction of protein localization signals. Two consensus sequences were identified in the cytoplasmic tail of UL18 that may account for its intracellular retention (Fig. 3A). Sequence YRKV from aa 362 to aa 365 is homologous to the consensus sequence YXX{Phi} that represents a tyrosine-based motif for rapid internalization and endocytosis. It mediates the internalization of proteins from the cell surface and targeting to intracellular compartments, such as endosomes, TGN, or lysosomes, via clathrin-coated pits associated with the AP-2 and AP-1 adaptor complexes (43, 44). Sequence KVRS from aa 364 to aa 367 is homologous to the KKXX consensus sequence responsible for retrieval of type I transmembrane proteins from postendoplasmic reticulum compartments back to the ER. Retrieval is mediated by proteins of the coatomer protein complex I structure (45) and occurs from as far as Golgi compartments, as indicated by Golgi-specific carbohydrate modifications of the retrieved proteins (46).

Mutation of the two motifs identified on the cytoplasmic tail of UL18 leads to surface expression of the hybrid protein HLA-A2/UL18CY

To evaluate the function of the above-described amino acid motifs, mutations that abrogate one or both motifs within the hybrid molecule HLA-A2/UL18CY were produced. Sequence YRKV was mutated to ARKV by replacing Y with A (Y362A), and sequence KVRS was mutated by replacing K with A (K364A) (Fig. 3A). Stable HeLa transfectants for all constructs were generated and analyzed. In these experiments the constructs were not GFP-tagged.

As shown in Fig. 3B (4°C column), transfection with HLA-A2 yields a strong surface expression of the protein. In contrast, transfection with HLA-A2/UL18CY leads to undetectable surface expression of the hybrid. However, a low amount of protein must have reached the surface membrane, as immunofluorescence-based endocytosis conducted by incubating Ab-fed cells for 30 min at 37°C with fresh medium allows the detection of a fluorescence punctuate pattern (Fig. 3B, 37°C, 30 min column).

The Y362A mutant, which bears a mutated YRKV sequence, is ineffective for clustering and internalization of Ab-bound hybrid protein, because no fluorescent punctuated pattern was observed after 30 min incubation at 37°C. The lack of internalization, however, is not sufficient to yield a detectable surface expression (Fig. 3B, 4°C column). Note that positive expression of the hybrid is demonstrated by immunoprecipitation with BB7.2 mAb (Fig. 3C) and by intracellular immunofluorescence (Fig. 4A). Lack of re-internalization is specific for the Y362A mutant, as it is not observed in the K364A mutant (Fig. 3B, 37°C, 30 min column) that retains the unmutated YRKV sequence. Interestingly, mutation of the KVRS sequence allows recovery of hybrid surface expression. Surface expression is further augmented by the double mutant Y362A/K364A, and internalization after 30 min is difficult to evaluate, possibly also because of the mutation of the YRKV sequence. Recovery of HLA-A2/UL18CY surface expression by K364A and by Y362A/K364A mutations, as observed on stable HeLa transfectants, was confirmed in transiently transfected HeLa cells, in transiently transfected COS cells, and in lentiviral-transduced HFFs (not shown).


Figure 4
View larger version (66K):
[in this window]
[in a new window]

  		FIGURE 4. A, Co-localization with ER and cis-Golgi. HFFs transduced with lentiviral vectors carrying either HLA-A2/UL18CY or its mutants were fixed with PFA, permeabilized with Triton X-100, and subjected to intracellular staining with BB7.2 mAb followed by Alexa 546-conjugated GAM IgG2b, anti-GM130 followed by Alexa 488-conjugated GAM IgG1, and calreticulin followed by Alexa 633-conjugated goat anti-rabbit IgG (virtual blue color). Merged images of hybrid expression with ER (red plus blue) or cis-Golgi (red plus green) display white pixels corresponding to areas with significant co-localization (Leica CF2D software). Scale bars correspond to 10 µm. B, Mutation of the two identified motifs abrogates internalization and leads to surface expression of UL18. COS-7 cells, transfected either with UL18-GFP or with UL18-GFP(Y362A,K364A), were subjected to surface immunofluorescence with M71 mAb followed by Alexa 546-conjugated GAM IgG, counterstained with 4',6-diamidino-2-phenylindole (DAPI), and analyzed by confocal microscopy. Scale bars correspond to 10 µm.

 
Hybrid proteins were also analyzed as BB7.2 immunoprecipitates revealed by anti-HLA class I H chain mAbs (Fig. 3C). In all cases two bands were observed: of 44 and 46 kDa for HLA-A2 and of 43 and 41 kDa for the hybrid molecules. The two bands represent different glycosylation levels, as demonstrated by digestion with PNGaseF that cleaves all sugars. Upon digestion, HLA-A2 and hybrid molecules display one band only, of 40 and 39 kDa, respectively (Fig. 3C). Note that upon digestion with EndoH, which is unable to remove sugar residues once glycoproteins have been processed by the trans-Golgi, the lower 41-kDa band of the hybrid molecules, but not the lower 43-kDa band of HLA-A2, dropped to 39 kDa (Fig. 3C). This indicates that at least a fraction of HLA-A2/UL18CY and mutants, but not of HLA-A2, is retained in the ER and cis-Golgi compartments.

The hybrid protein co-localizes predominantly with the cis-Golgi and partially with the ER

Confocal microscopy evaluations were conducted to analyze changes in the intracellular distribution patterns of the Y362A and the K364A mutants compared with HLA-A2/UL18CY. After lentiviral transduction to induce expression of the hybrid proteins, HFFs were processed for immunofluorescence detection of the ER (by calreticulin Abs) and the cis-Golgi (by GM130 mAb) to be analyzed for co-localization with BB7.2-stained hybrid proteins. As displayed in Fig. 4A, the unmutated HLA-A2/UL18CY hybrid shows a major co-localization with the cis-Golgi and, to a lesser extent, with the ER. The same pattern is observed on the Y362A mutant, whereas the K364A mutant displays an intracellular distribution entirely confined to the cis-Golgi (Fig. 4A). Most HLA-A2/UL18CY(Y362A/K364A) hybrid molecules are revealed on the cell membrane, and their intracellular expression is hard to detect. Note that despite permeabilization procedures required for BB7.2 intracellular staining, which can affect protein distribution on the plasma membrane, surface immunofluorescence detection of Y362A/K364A is still high. Altogether the data are consistent with the expected role of the two motifs on the cytoplasmic tail of UL18, that is, rapid internalization and endoplasmic reticulum retrieval.

Mutation of the two motifs on the cytoplasmic tail of UL18 abrogates internalization and leads to surface expression of UL18

Based on the above results, a double mutant of UL18 bearing the two Y362A/K364A mutations was constructed. Surface expression of this mutant was investigated on HeLa, COS-7, and HEK tumor cell lines, as well as in HFFs. COS-7 cells transiently transfected with UL18-GFP or with UL18(Y362A/K364A)-GFP and stained for surface immunofluorescence with M71 are shown in Fig. 4B. A significant recovery of surface detectability of UL18 was observed when the two motifs on its cytoplasmic tail were mutated. Identical results were achieved in the other cell systems, both with M71 and CD85jFc detection reagents (not shown).

The HCMV genome is involved in the regulation of HLA-A2/UL18CY and UL18 surface expression in human-transduced fibroblasts

We wondered whether the two motifs found on the cytoplasmic tail of UL18 represent a mechanism used in vivo by the HCMV for regulating the egress of UL18 toward the cell surface and its consequent availability for trans-interactions with the CD85j receptor.

An in vitro cell model comprehensive of either HLA-A2/UL18CY or UL18 genes, together with the whole HCMV genome, was created. To this end, HFFs were first transduced with VVEW/BB lentiviral vectors carrying HLA-A2/UL18CY or UL18-GFP (100% infection efficiency) and subsequently infected with the HCMV. This model of HCMV infection should grant the presence of HCMV genes possibly regulating expression and trafficking of UL18 and/or HLA-A2/UL18CY. Either AD169 or the AD169-derived deletion mutant RV798 was used on UL18-GFP-expressing HFFs. In fact, cell surface expression of UL18 is not affected by US2, US3, US6, and US11 HCMV gene products (47). Conversely, experiments on HLA-A2/UL18CY-HFFs were conducted with the virus RV798 only, because it does not contain US2, US3, US6, and US11 genes that down-regulate HLA-A*0201 surface expression (42, 47). To maximize duration of the experiments, HFFs were infected with low viral titers (1 MOI). Under these conditions, in vitro cellular viability was preserved for at least 7 days, and no endogenous UL18 was detected at any time of infection. Successful infection was monitored by expression of the early DNA-binding viral protein UL44 (not shown).

Analysis of surface HLA-A2/UL18CY or UL18-GFP temporal expression was performed by staining HCMV-infected fibroblasts for surface BB7.2 or M71 immunofluorescence, respectively, and analyzing samples by confocal microscopy. Time-course observations from 0 to 6 days postinfection (dpi) are shown in Fig. 5, A and B. As expected based on previous data, expression of HLA-A2/UL18CY or UL18-GFP on the cell membrane of HCMV uninfected fibroblasts was negative (0 dpi). However, as HCMV infection progressed, surface-stained cells appeared and increased over time. The first fluorescent cells were observed at 3 dpi; at day 4 the percentage of surface-positive cells increased significantly, having increased further during the subsequent days with a much lower rate, until cells died from infection-induced toxicity in culture. Time-course of the percentage of surface HLA-A2/UL18CY- and UL18-GFP-positive cells, as calculated on samples of one representative experiment, is displayed in Fig. 5C. Very similar results were achieved when UL18-GFP-expressing fibroblasts were infected with RV798 (not shown). To exclude that surface immunofluorescence was due to undesired binding of Abs to infection-induced Fc receptors (48), wild-type HFFs or HFFs transduced either with empty lentiviral vectors without transgene insert or with GFP only were HCMV infected and processed for surface immunofluorescence with BB7.2 and M71 mAbs. No fluorescence was detected in any case and at any time (lowest row of Fig. 5, A and B, and not shown).


Figure 5
View larger version (53K):
[in this window]
[in a new window]

  		FIGURE 5. Infection by HCMV induces expression of HLA-A2/UL18CY and UL18 on the cell membrane of HFFs. A and B, HFFs transduced with lentiviral vectors carrying either HLA-A2/UL18CY (A) or UL18-GFP (B) were infected with laboratory viral strains of HCMV: HLA-A2/UL18CY-HFFs with RV798 (A), UL18-GFP-HFFs with AD169 (B). Surface expression of the hybrid protein (A) or UL18-GFP (B) was measured on a laser scanning confocal microscope after surface immunofluorescence with BB7.2 plus Alexa 488 (green)-conjugated GAM IgG (A) or with M71 plus Alexa 546 (red)-conjugated GAM IgG (B). Autofluorescence (A) or GFP fluorescence (B) are displayed for the identification of all cells. The cells were optically sectioned in the z-axis (40 slides/imaged field), and the integral of eight sections (from 16 to 24) is shown. Time-course images are displayed from day 0 postinfection (0 dpi) to 6 dpi. The lowest row shows negative control samples at 6 dpi: A, wild-type fibroblasts not expressing HLA-A2/UL18CY that were infected with RV798; and B, fibroblasts expressing GFP only that were infected with AD169. C, Time-course analysis of the percentage of HFFs displaying positive surface expression of HLA-A2/UL18CY ({blacktriangleup}) or UL18-GFP ({blacksquare}) molecules. Data from one representative experiment are used: for each time point the mean value over five randomly selected imaged fields is reported, with each field containing ~40–60 cells. Error bars define the SD. D, The ratio between surface M71 plus GAM Alexa 546 red fluorescence and intracellular GFP fluorescence was evaluated on 30 UL18-GFP-expressing HFF cells randomly selected for surface M71 positivity. Data from one representative experiment were used, and each dot corresponds to one cell. Fluorescence values were calculated by Leica software. The broadly scattered pattern of the ratio distribution demonstrates that the amount of GFP-UL18 molecules on the cell surface has no relationship to its intracellular amount, as illustrated in four examples displayed in E.

 
We wondered whether UL18 surface expression of infected cells is simply due to an increased translation with a resulting spillover of UL18 from the cytoplasm to the cell surface or if it is due to a regulated mechanism controlled by the HCMV genome. To this end we evaluated whether the two UL18 fluorescence intensities were related, namely intracellular versus surface fluorescence. By visual microscopic inspection of UL18-GFP samples, a heterogeneous pattern of GFP/surface M71 relationship was evident, as cells very bright for GFP fluorescence occurred to be negative for surface M71, or cells with weak GFP displayed strong M71 surface staining. The ratio between surface M71 plus goat anti-mouse (GAM) Alexa 546 red fluorescence and intracellular GFP fluorescence was quantitated with Leica software applied to confocal microscope measurements conducted under identical instrumental settings. Thirty cells per sample were randomly selected among cells that displayed positive surface M71 fluorescence. The broadly scattered ratio distribution in Fig. 5D demonstrates that the surface amount of UL18-GFP molecules was independent from the intracellular amount, and rules out that surface UL18 expression was induced merely by a HCMV-induced increase in its translation. As an example, imaged fields showing cells with extremely different surface versus cytoplasmic UL18 expression ratios (e.g., cells with low GFP fluorescence and high red fluorescence, or vice versa) are reported in Fig. 5E.

We were not confident in quantitating the "surface UL18/intracellular UL18" ratio by flow cytometric measurements that would provide statistically more significant data, because we observed a strong increase of cellular autofluorescence upon HCMV infection. Flow cytometry, being unable to discriminate the source of fluorescence, would have summed fluorescence from specific Ab binding and autofluorescence; instead, the autofluorescence signal could be reliably subtracted by means of emission wavelength scans on the spectral confocal microscope.

Altogether, the above data suggest that during HCMV infection the activity of the two UL18 cytoplasmic motifs is regulated by a finely tuned interplay between HCMV viral proteins, host proteins, and UL18 itself to provide the virus with a mechanism that controls UL18 surface expression.

The amino acid sequence of the UL18 cytoplasmic tail is highly conserved

Once we had shown that two sequences present in the cytoplasmic tail regulate the surface expression of the UL18 molecule encoded by the AD169 HCMV strain, we asked whether they are conserved in other laboratory and wild-type HCMV strains. To this end, stalk, transmembrane, and cytoplasmic portions of the UL18 gene from genomic DNA of 10 HCMV-positive donors were sequenced. Other sequences (i.e., AD169, Merlin, Toledo, and Towne strains and FIX-BAC, PH-BAC, and TR-BAC isolates) were obtained from GenBank database. ClustalW alignment of the amino acid sequences showed that the 65 carboxyl-terminal amino acids are highly conserved and that the two motifs are identical in all cases (Fig. 6).


Figure 6
View larger version (61K):
[in this window]
[in a new window]

  		FIGURE 6. The amino acid sequence of the UL18 cytoplasmic tail is highly conserved in wild-type HCMV strains. Amino acid comparison of the stalk, transmembrane region, and cytoplasmic tail of UL18 from 10 wild-type HCMV strains obtained from the genomic DNA sequences of HCMV-positive donors (accession numbers AJ972608–AJ972617). Other sequences (i.e., AD169, Merlin, Toledo, and Towne strains and FIX-BAC, PH-BAC, and TR-BAC isolates) were obtained from GenBank database (accession numbers NC_001347, AY446894, AY486470, AY315197, AC146907, AC146904, AC146906). *, Amino acid identity in all samples;., a conservative change. Blank spaces refer to nonconservative changes. Note that the two sequences YRKV from aa 362 to aa 365 and KVRS from aa 364 to aa 367 are identical in all samples.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
Disclosures
 References
 
The HCMV has developed a vast array of immunostimulatory and immunosuppressive functions that participate in the maintenance of its survival within the immunocompetent host, through a complex interplay of receptors on immune cell, particularly T and NK cells, and virus-encoded genes. One node of this network is represented by the interaction between the viral glycoprotein UL18 on HCMV-infected cells and CD85j expressed by T and NK cells, as suggested by the high affinity of UL18 for CD85j (Kd in the nanomolar range (30, 49)), and by the fact that no other ligands for UL18 have been identified. Also, UL18 is not necessary for infection (15), but the gene was found in all HCMV genomes analyzed so far, namely in four laboratory strains and in 26 samples from clinical isolates or seropositive donors (49, 50).

However, the effects of UL18 on NK and T cells are still matter of controversy. On one hand, the CD85j–UL18 interaction was proposed as a viral escaping mechanism from NK-cell-mediated lysis (4, 25), and on the other hand as a mechanism that leads to the killing of HCMV-infected cells by NK cells (27) and T lymphocytes in a non-MHC-restricted fashion (28). Nevertheless, in either instance, we expect the CD85j–UL18 interaction to operate in a controlled fashion in a specific time frame. If the interaction triggers an immunosubversive pathway, inhibition of NK cells should become operative at a time coincidental with an increased risk of NK lysis due to the virus-induced HLA class I down-regulation. On the other hand, if the CD85j–UL18 interaction exerts an activation of T lymphocytes, only a regulated pattern of surface expression would prevent the virus from being completely erased from the host and letting it undergo latency following primary infection.

Accordingly, we hypothesized the CD85j–UL18 interaction to be tightly regulated by a control mechanism of UL18 surface expression during HCMV infection. It is known that transcription of this gene occurs from 54 h postinfection onward (14), but not much information is available on its expression on the plasma membrane of infected cells.

We observed in in vitro experiments that fibroblasts or tumor cell lines respectively transduced or transfected with the UL18 gene display most molecules accumulated in the cytoplasm and almost undetectable levels on the plasma membrane. However, when culturing these cells in the presence of the whole HCMV genome, which had been introduced by in vitro HCMV infection at 1 MOI, a time-associated pattern of UL18 egress toward the cell surface could be observed. These data suggest that a network of HCMV gene activity operates on UL18 to control its surface expression and, consequently, the UL18–CD85j interaction.

Molecular dissection studies conducted for identification of the UL18 domains involved in this regulatory mechanism disclosed a relevant role for the cytoplasmic tail of UL18. Two motifs on the cytoplasmic tail of UL18 were identified that are homologous to consensus sequences known to mediate intracytoplasmic retention, namely one tyrosine-based motif for rapid internalization and endocytosis, and one consensus sequence responsible for retrieval of type I transmembrane proteins from Golgi compartments back to the ER. The two motifs were able to completely abolish surface expression of a well-known plasma membrane protein such as HLA-A2, as observed in fibroblasts transduced with a hybrid HLA-A2 protein that bears the cytoplasmic tail of UL18 (HLA-A2/UL18CY). Most HLA-A2/UL18CY molecules were retained in the ER and cis-Golgi. However, surface expression was restored when cells transduced with HLA-A2/UL18CY were infected with the HCMV, displaying a time-regulated pattern of plasma membrane egress very similar to that of UL18 in HCMV-infected UL18-transduced fibroblasts. These data further demonstrate that the HCMV uses the cytoplasmic tail of UL18 for regulating its surface expression. Importantly, the two motifs are completely identical in UL18 genes from AD169, Merlin, Toledo, and Towne strains and FIX-BAC, PH-BAC, and TR-BAC isolates, as well as in the 10 HCMV-positive donors analyzed.

One of the starting points of this study is that UL18 surface expression on in vitro cell systems is mostly hard to detect. However, a few examples in the literature apparently are not in accordance with this observation, as human fibroblasts subjected to HCMV infection (MOI 10) or to infection with UL18-carrying adenovirus (MOI 100) or vaccinia virus (MOI 10) displayed UL18 expression on their plasma membrane (28, 35, 36). We ascribe the spillover of UL18 from the cytoplasm to the cell surface observed in these experiments to the high viral titer used for the infections. However, we cannot exclude a contribution of portions of viral genomes possibly present in the DNA of tumor cell lines or immortalized fibroblasts, in a mechanism like the one we described for HCMV, especially if the viral genes involved in the regulation of the activity of KVRS and YRKV motifs are conserved among the different viruses.

For the HCMV-infection experiments on UL18-transduced fibroblasts (Fig. 5) we had to decrease the MOI down to 1 to exclude the contribution of HCMV-encoded UL18 molecules to the surface expression of UL18. Besides ameliorating cellular viability during the infection, this experimental setting provided a "clean" cell model for studying the effects of the interaction between factors encoded by the HCMV genome and transduced UL18 molecules (Fig. 5B, bottom row). To account for the negative surface expression of endogenous UL18 that is undetectable even at the 4th and 5th day after in vitro HCMV infection, we argue that a minimal amount of UL18 translated molecules might be necessary to allow immunofluorescence detection on the cell membrane. The viral load administered to the fibroblasts by an MOI of 1 might be too low to provide enough UL18 transcripts and translated proteins to be detected on the cell membrane by immunofluorescence and confocal microscopy, even when the "virally encoded" mechanism is maximally effective in our in vitro experimental system. Conversely, the UL18-GFP-fibroblasts, also when displaying very faint GFP (Fig. 5E), might produce an amount of UL18 above threshold, so that the mechanism of regulation of UL18 egress has the possibility to overcome the immunofluorescence detection barrier. Altogether, the data suggest that HCMV-encoded "factors" require a threshold amount of UL18 molecules to allow their surface detection by immunofluorescence and confocal microscopy, but the mechanism of regulation of UL18 surface egress does not simply depend on the amount of cytoplasmic UL18.

In conclusion, we disclose in this study a regulatory mechanism for the control of surface UL18 expression. Following our previous observation that the UL18–CD85j interaction triggers non-MHC-restricted TCR-independent T cell activation and subsequent lysis of UL18-expressing cells (28), we hypothesize that during the latent phase of infection, a lack of UL18 molecules on the infected cells prevents their clearance. Conversely, during the productive phase of infection, when most HCMV genes are active, the virus may allow the release of UL18 surface expression by suppressing the retention function of the YRKV and KVRS motifs. This would make UL18 available for trans-interaction with CD85j on T lymphocytes and mediate the lysis of infected cells, thus preserving the host from death and, consequently, the virus from its elimination.


   Acknowledgments
 
This study is dedicated to the memory of Carlo Enrico Grossi.


   Disclosures
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
References
 
The authors have no financial conflicts of interest.


   Footnotes
 
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1 This work was supported by grants from Compagnia di San Paolo, Ministero per l’Istruzione, l’Università e la Ricerca Scientifica, and Progetto Finalizzato Ministero della Salute (to E.C.). Back

2 M.M. and F.G. contributed equally to this work. Back

3 Address correspondence and reprint requests to Dr. Silvia Bruno, Department of Experimental Medicine, Human Anatomy Section, University of Genoa, Via De Toni 14, 16132 Genova, Italy. E-mail address: silvia.bruno{at}unige.it Back

4 Abbreviations used in this paper used: HCMV, human cytomegalovirus; EGFP, enhanced GFP; EndoH, endoglycosidase H; ER, endoplasmic reticulum; GAM, goat anti-mouse; HFF, human foreskin fibroblast; MOI, multiplicity of infection; PNGaseF, peptide N-glycosidase F; PFA, paraformaldehyde; IP, immunoprecipitation. Back

Received for publication July 24, 2007. Accepted for publication October 30, 2007.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 References
 

  1. Ho, M.. 1990. Epidemiology of cytomegalovirus infections. Rev. Infect. Dis. 12: (Suppl. 7):S701-S710.
  2. Fruh, K., A. Gruhler, R. M. Krishna, G. J. Schoenhals. 1999. A comparison of viral immune escape strategies targeting the MHC class I assembly pathway. Immunol. Rev. 168: 157-166. [Medline]
  3. Xu, X. N., G. R. Screaton, A. J. McMichael. 2001. Virus infections: escape, resistance, and counterattack. Immunity 15: 867-870. [Medline]
  4. Farrell, H. E., H. Vally, D. M. Lynch, P. Fleming, G. R. Shellam, A. A. Scalzo, N. J. Davis-Poynter. 1997. Inhibition of natural killer cells by a cytomegalovirus MHC class I homologue in vivo. Nature 386: 510-514. [Medline]
  5. Tomasec, P., V. M. Braud, C. Rickards, M. B. Powell, B. P. McSharry, S. Gadola, V. Cerundolo, L. K. Borysiewicz, A. J. McMichael, G. W. Wilkinson. 2000. Surface expression of HLA-E, an inhibitor of natural killer cells, enhanced by human cytomegalovirus gpUL40. Science 287: 1031-1033. [Abstract/Free Full Text]
  6. Tomasec, P., E. C. Wang, A. J. Davison, B. Vojtesek, M. Armstrong, C. Griffin, B. P. McSharry, R. J. Morris, S. Llewellyn-Lacey, C. Rickards, et al 2005. Downregulation of natural killer cell-activating ligand CD155 by human cytomegalovirus UL141. Nat. Immunol. 6: 181-188. [Medline]
  7. Stern-Ginossar, N., N. Elefant, A. Zimmermann, D. G. Wolf, N. Saleh, M. Biton, E. Horwitz, Z. Prokocimer, M. Prichard, G. Hahn, et al 2007. Host immune system gene targeting by a viral miRNA. Science 317: 376-381. [Abstract/Free Full Text]
  8. Polic, B., H. Hengel, A. Krmpotic, J. Trgovcich, I. Pavic, P. Luccaronin, S. Jonjic, U. H. Koszinowski. 1998. Hierarchical and redundant lymphocyte subset control precludes cytomegalovirus replication during latent infection. J. Exp. Med. 188: 1047-1054. [Abstract/Free Full Text]
  9. Arase, H., E. S. Mocarski, A. E. Campbell, A. B. Hill, L. L. Lanier. 2002. Direct recognition of cytomegalovirus by activating and inhibitory NK cell receptors. Science 296: 1323-1326. [Abstract/Free Full Text]
  10. Gutierrez, A., I. Munoz, C. Solano, I. Benet, C. Gimeno, I. Marugan, M. D. Gea, J. Garcia-Conde, D. Navarro. 2003. Reconstitution of lymphocyte populations and cytomegalovirus viremia or disease after allogeneic peripheral blood stem cell transplantation. J. Med. Virol. 70: 399-403. [Medline]
  11. Beck, S., B. G. Barrell. 1988. Human cytomegalovirus encodes a glycoprotein homologous to MHC class-I antigens. Nature 331: 269-272. [Medline]
  12. Browne, H., G. Smith, S. Beck, T. Minson. 1990. A complex between the MHC class I homologue encoded by human cytomegalovirus and β2 microglobulin. Nature 347: 770-772. [Medline]
  13. Fahnestock, M. L., J. L. Johnson, R. M. Feldman, J. M. Neveu, W. S. Lane, P. J. Bjorkman. 1995. The MHC class I homolog encoded by human cytomegalovirus binds endogenous peptides. Immunity 3: 583-590. [Medline]
  14. Hassan-Walker, A. F., A. V. Cope, P. D. Griffiths, V. C. Emery. 1998. Transcription of the human cytomegalovirus natural killer decoy gene, UL18, in vitro and in vivo. J. Gen. Virol. 79: 2113-2116. [Abstract]
  15. Browne, H., M. Churcher, T. Minson. 1992. Construction and characterization of a human cytomegalovirus mutant with the UL18 (class I homolog) gene deleted. J. Virol. 66: 6784-6787. [Abstract/Free Full Text]
  16. Cosman, D., N. Fanger, L. Borges, M. Kubin, W. Chin, L. Peterson, M. L. Hsu. 1997. A novel immunoglobulin superfamily receptor for cellular and viral MHC class I molecules. Immunity 7: 273-282. [Medline]
  17. Bellon, T., F. Kitzig, J. Sayos, M. Lopez-Botet. 2002. Mutational analysis of immunoreceptor tyrosine-based inhibition motifs of the Ig-like transcript 2 (CD85j) leukocyte receptor. J. Immunol. 168: 3351-3359. [Abstract/Free Full Text]
  18. Colonna, M., F. Navarro, T. Bellon, M. Llano, P. Garcia, J. Samaridis, L. Angman, M. Cella, M. Lopez-Botet. 1997. A common inhibitory receptor for major histocompatibility complex class I molecules on human lymphoid and myelomonocytic cells. J. Exp. Med. 186: 1809-1818. [Abstract/Free Full Text]
  19. Saverino, D., M. Fabbi, F. Ghiotto, A. Merlo, S. Bruno, D. Zarcone, C. Tenca, M. Tiso, G. Santoro, G. Anastasi, et al 2000. The CD85/LIR-1/ILT2 inhibitory receptor is expressed by all human T lymphocytes and down-regulates their functions. J. Immunol. 165: 3742-3755. [Abstract/Free Full Text]
  20. Young, N. T., M. Uhrberg, J. H. Phillips, L. L. Lanier, P. Parham. 2001. Differential expression of leukocyte receptor complex-encoded Ig-like receptors correlates with the transition from effector to memory CTL. J. Immunol. 166: 3933-3941. [Abstract/Free Full Text]
  21. Fanger, N. A., D. Cosman, L. Peterson, S. C. Braddy, C. R. Maliszewski, L. Borges. 1998. The MHC class I binding proteins LIR-1 and LIR-2 inhibit Fc receptor-mediated signaling in monocytes. Eur. J. Immunol. 28: 3423-3434. [Medline]
  22. Dietrich, J., M. Cella, M. Colonna. 2001. Ig-like transcript 2 (ILT2)/leukocyte Ig-like receptor 1 (LIR1) inhibits TCR signaling and actin cytoskeleton reorganization. J. Immunol. 166: 2514-2521. [Abstract/Free Full Text]
  23. Merlo, A., D. Saverino, C. Tenca, C. E. Grossi, S. Bruno, E. Ciccone. 2001. CD85/LIR-1/ILT2 and CD152 (cytotoxic T lymphocyte antigen 4) inhibitory molecules down-regulate the cytolytic activity of human CD4+ T-cell clones specific for Mycobacterium tuberculosis. Infect. Immun. 69: 6022-6029. [Abstract/Free Full Text]
  24. Saverino, D., A. Merlo, S. Bruno, V. Pistoia, C. E. Grossi, E. Ciccone. 2002. Dual effect of CD85/leukocyte Ig-like receptor-1/Ig-like transcript 2 and CD152 (CTLA-4) on cytokine production by antigen-stimulated human T cells. J. Immunol. 168: 207-215. [Abstract/Free Full Text]
  25. Reyburn, H. T., O. Mandelboim, M. Vales-Gomez, D. M. Davis, L. Pazmany, J. L. Strominger. 1997. The class I MHC homologue of human cytomegalovirus inhibits attack by natural killer cells. Nature 386: 514-517. [Medline]
  26. Kim, J. S., S. E. Choi, I. H. Yun, J. Y. Kim, C. Ahn, S. J. Kim, J. Ha, E. S. Hwang, C. Y. Cha, S. Miyagawa, C. G. Park. 2004. Human cytomegalovirus UL18 alleviated human NK-mediated swine endothelial cell lysis. Biochem. Biophys. Res. Commun. 315: 144-150. [Medline]
  27. Leong, C. C., T. L. Chapman, P. J. Bjorkman, D. Formankova, E. S. Mocarski, J. H. Phillips, L. L. Lanier. 1998. Modulation of natural killer cell cytotoxicity in human cytomegalovirus infection: the role of endogenous class I major histocompatibility complex and a viral class I homolog. J. Exp. Med. 187: 1681-1687. [Abstract/Free Full Text]
  28. Saverino, D., F. Ghiotto, A. Merlo, S. Bruno, L. Battini, M. Occhino, M. Maffei, C. Tenca, S. Pileri, L. Baldi, et al 2004. Specific recognition of the viral protein UL18 by CD85j/LIR-1/ILT2 on CD8+ T cells mediates the non-MHC-restricted lysis of human cytomegalovirus-infected cells. J. Immunol. 172: 5629-5637. [Abstract/Free Full Text]
  29. Wagner, C. S., G. C. Riise, T. Bergstrom, K. Karre, E. Carbone, L. Berg. 2007. Increased expression of leukocyte Ig-like receptor-1 and activating role of UL18 in the response to cytomegalovirus infection. J. Immunol. 178: 3536-3543. [Abstract/Free Full Text]
  30. Chapman, T. L., A. P. Heikeman, P. J. Bjorkman. 1999. The inhibitory receptor LIR-1 uses a common binding interaction to recognize class I MHC molecules and the viral homolog UL18. Immunity 11: 603-613. [Medline]
  31. Chapman, T. L., A. P. Heikema, A. P. West, Jr, P. J. Bjorkman. 2000. Crystal structure and ligand binding properties of the D1D2 region of the inhibitory receptor LIR-1 (ILT2). Immunity 13: 727-736. [Medline]
  32. Antrobus, R. D., N. Khan, A. D. Hislop, D. Montamatt-Sicotte, L. I. Garner, A. B. Rickinson, P. A. Moss, B. E. Willcox. 2005. Virus-specific cytotoxic T lymphocytes differentially express cell-surface leukocyte immunoglobulin-like receptor-1, an inhibitory receptor for class I major histocompatibility complex molecules. J. Infect. Dis. 191: 1842-1853. [Medline]
  33. Northfield, J., M. Lucas, H. Jones, N. T. Young, P. Klenerman. 2005. Does memory improve with age? CD85j (ILT-2/LIR-1) expression on CD8 T cells correlates with ‘memory inflation’ in human cytomegalovirus infection. Immunol. Cell Biol. 83: 182-188. [Medline]
  34. Berg, L., G. C. Riise, D. Cosman, T. Bergstrom, S. Olofsson, K. Karre, E. Carbone. 2003. LIR-1 expression on lymphocytes, and cytomegalovirus disease in lung-transplant recipients. Lancet 361: 1099-1101. [Medline]
  35. Griffin, C., E. C. Wang, B. P. McSharry, C. Rickards, H. Browne, G. W. Wilkinson, P. Tomasec. 2005. Characterization of a highly glycosylated form of the human cytomegalovirus HLA class I homologue gpUL18. J. Gen. Virol. 86: 2999-3008. [Abstract/Free Full Text]
  36. Prod’homme, V., C. Griffin, R. J. Aicheler, E. C. Wang, B. P. McSharry, C. R. Rickards, R. J. Stanton, L. K. Borysiewicz, M. Lopez-Botet, G. W. Wilkinson, P. Tomasec. 2007. The human cytomegalovirus MHC class I homolog UL18 inhibits LIR-1+ but activates LIR-1– NK cells. J. Immunol. 178: 4473-4481. [Abstract/Free Full Text]
  37. Fedorova, E., L. Battini, A. Prakash-Cheng, D. Marras, G. L. Gusella. 2006. Lentiviral gene delivery to CNS by spinal intrathecal administration to neonatal mice. J. Gene Med. 8: 414-424. [Medline]
  38. Naldini, L., U. Blomer, F. H. Gage, D. Trono, I. M. Verma. 1996. Efficient transfer, integration, and sustained long-term expression of the transgene in adult rat brains injected with a lentiviral vector. Proc. Natl. Acad. Sci. USA 93: 11382-11388. [Abstract/Free Full Text]
  39. Ishido, S., C. Wang, B. S. Lee, G. B. Cohen, J. U. Jung. 2000. Downregulation of major histocompatibility complex class I molecules by Kaposi’s sarcoma-associated herpesvirus K3 and K5 proteins. J. Virol. 74: 5300-5309. [Abstract/Free Full Text]
  40. Horton, P., K. Nakai. 1997. Better prediction of protein cellular localization sites with the k nearest neighbors classifier. Proc. Int. Conf. Intell. Syst. Mol. Biol. 5: 147-152. [Medline]
  41. Puntervoll, P., R. Linding, C. Gemund, S. Chabanis-Davidson, M. Mattingsdal, S. Cameron, D. M. Martin, G. Ausiello, B. Brannetti, A. Costantini, et al 2003. ELM server: a new resource for investigating short functional sites in modular eukaryotic proteins. Nucleic Acids Res. 31: 3625-3630. [Abstract/Free Full Text]
  42. Jones, T. R., L. K. Hanson, L. Sun, J. S. Slater, R. M. Stenberg, 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. [Abstract]
  43. Ohno, H., J. Stewart, M. C. Fournier, H. Bosshart, I. Rhee, S. Miyatake, T. Saito, A. Gallusser, T. Kirchhausen, J. S. Bonifacino. 1995. Interaction of tyrosine-based sorting signals with clathrin-associated proteins. Science 269: 1872-1875. [Abstract/Free Full Text]
  44. Marks, M. S., L. Woodruff, H. Ohno, J. S. Bonifacino. 1996. Protein targeting by tyrosine- and di-leucine-based signals: evidence for distinct saturable components. J. Cell Biol. 135: 341-354. [Abstract/Free Full Text]
  45. Cosson, P., Y. Lefkir, C. Demolliere, F. Letourneur. 1998. New COP1-binding motifs involved in ER retrieval. EMBO J. 17: 6863-6870. [Medline]
  46. Gaynor, E. C., S. te Heesen, T. R. Graham, M. Aebi, S. D. Emr. 1994. Signal-mediated retrieval of a membrane protein from the Golgi to the ER in yeast. J. Cell Biol. 127: 653-665. [Abstract/Free Full Text]
  47. Park, B., H. Oh, S. Lee, Y. Song, J. Shin, Y. C. Sung, S. Y. Hwang, K. Ahn. 2002. The MHC class I homolog of human cytomegalovirus is resistant to down-regulation mediated by the unique short region protein (US)2, US3, US6, and US11 gene products. J. Immunol. 168: 3464-3469. [Abstract/Free Full Text]
  48. Antonsson, A., P. J. Hugo Johansson. 2001. Binding of human and animal immunoglobulins to the IgG Fc receptor induced by cytomegalovirus. J. Gen. Virol. 82: 1137-1145. [Abstract/Free Full Text]
  49. Vales-Gomez, M., M. Shiroishi, K. Maenaka, H. T. Reyburn. 2005. Genetic variability of the major histocompatibility complex class I homologue encoded by human cytomegalovirus leads to differential binding to the inhibitory receptor ILT2. J. Virol. 79: 2251-2260. [Abstract/Free Full Text]
  50. Cerboni, C., A. Achour, A. Warnmark, M. Mousavi-Jazi, T. Sandalova, M. L. Hsu, D. Cosman, K. Karre, E. Carbone. 2006. Spontaneous mutations in the human CMV HLA class I homologue UL18 affect its binding to the inhibitory receptor LIR-1/ILT2/CD85j. Eur. J. Immunol. 36: 732-741. [Medline]
  51. Occhino, M., F. Ghiotto, S. Soro, M. Mortarino, S. Bosi, M. Maffei, S. Bruno, N. Nardini, M. Figini, A. Tramontano, E. Ciccone. 2008. Dissecting the structural determinants of the interaction between the human cytomegalovirus UL18 protein and the CD85j immune receptor. J. Immunol. 180: 957-968. [Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 J. Virol.Home page
A. Petrucelli, M. Rak, L. Grainger, and F. Goodrum
Characterization of a Novel Golgi Apparatus-Localized Latency Determinant Encoded by Human Cytomegalovirus
J. Virol., June 1, 2009; 83(11): 5615 - 5629.
[Abstract] [Full Text] [PDF]

Home page
 J. Immunol.Home page
M. Occhino, F. Ghiotto, S. Soro, M. Mortarino, S. Bosi, M. Maffei, S. Bruno, M. Nardini, M. Figini, A. Tramontano, et al.
Dissecting the Structural Determinants of the Interaction between the Human Cytomegalovirus UL18 Protein and the CD85j Immune Receptor
J. Immunol., January 15, 2008; 180(2): 957 - 968.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Maffei, M.
Right arrow Articles by Ciccone, E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Maffei, M.
Right arrow Articles by Ciccone, E.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS