点击显示 收起
Cancer Research UK Institute for Cancer Studies, University of Birmingham, Birmingham, United Kingdom
Leukocyte immunoglobulin-like receptor1 (LIR-1) is an inhibitory receptor that negatively regulates T cell effector functions after interaction with host class I major histocompatibility complex molecules and, additionally, binds to UL18, a human cytomegalovirus (HCMV)encoded class I homologue. Here, we demonstrate that virus-specific cytotoxic T lymphocytes (CTLs) differentially express LIR-1, with high frequencies of expression on HCMV-specific CD8+ T cells and intermediate and low frequencies of expression on influenza virusspecific and Epstein-Barr virus (EBV)specific CTLs, respectively. Expression of LIR-1 was dependent on CTL-antigen specificity and was associated with a differentiated effector memory phenotype, as demonstrated by decreased expression of CD28 and increased expression of CD57. During primary HCMV and EBV infections, expression of LIR-1 on virus-specific CTLs was low and increased slowly. These results indicate that expression of LIR-1 increases during differentiation of virus-specific CD8+ effector T cells. Furthermore, they suggest that a potential immunoregulatory function of UL18 may be to preferentially target highly differentiated HCMV-specific effector memory T cells during persistent infection.
The leukocyte immunoglobulin-like receptor (LIR) family (LIRs, immunoglobulin-like transcripts , monocyte/macrophage immunoglobulin-related receptors, LILRs, and CD85) comprises a set of related immunoreceptors with inhibitory (LIR-1, -2, -3, -5, and -8), activatory (LIR-6 and -7 and ILT-7, -8, and -11), or soluble (LIR-4) functions encoded within the leukocyte receptor cluster on chromosome 19 [1, 2]. LIR-1, the most broadly expressed member, is a type I transmembrane protein that is present on monocytes, macrophages, and dendritic cells and, additionally, on B cells, T cells, and a variable proportion of NK cells. Ligation of cell-surface LIR-1 leads to transmission of inhibitory signals to the effector cell, which is mediated by association of intracellular phosphatases, such as Src homology 2 domaincontaining phosphatase 1, with intracellular tyrosine-based inhibitory motif sequences in the LIR-1 cytoplasmic tail. Coligation of LIR-1 with activatory receptors results in significant suppression of effector functions, such as Ca2+ mobilization and phosphorylation (in monocytes, macrophages, dendritic cells, and B cells) and cytotoxicity (in T cells and NK cells) [1]. Consequently, LIR-1/ligand interactions may regulate diverse immune effector cells.
The first LIR-1 ligand identified was UL18, a class I major histocompatibility complex (MHC) homologue encoded by human cytomegalovirus (HCMV) [3]. Subsequent work showed that LIR-1 and the inhibitory receptor LIR-2 each recognize a broad range of cellular class I MHC proteins, including classical (HLA-A, -B, and -C) and nonclassical (HLA-E, -F, and -G) molecules [3, 4]. LIR-1 and LIR-2 bind class I MHC proteins with similarly low affinities [4, 5]. However, despite the comparable nature of these interactions, cell-staining studies indicated that UL18-Fc protein specifically bound cell-surface LIR-1 but did not bind other LIR molecules [3]. One possible explanation for specific targeting of LIR-1 by UL18 is that, in contrast to LIR-2, LIR-1 is expressed on lymphocytes and may mediate critical regulation of HCMV-specific B cell, T cell, or NK cell responses. UL18 was originally hypothesized to inhibit NK cell lysis by engaging cell-surface inhibitory receptors on NK cells, thereby preventing lysis of HCMV-infected target cells [6]. However, initial studies indicated that expression of LIR-1 by NK cells was restricted to minor subsets varying between individuals [1]. Furthermore, expression of UL18 marginally increased NK cell killing of HCMV-infected human fibroblasts [7]. Consequently, the role that UL18 plays in immune evasion of NK cell responses is unclear. Alternatively, UL18 may regulate HCMV-specific B and T cell responses.
Results of studies of animal models and immunocompromised humans have suggested that cellular immunity plays a critical role in the control of HCMV replication and subsequent disease [810]. Cytotoxic T lymphocytes (CTLs) have been implicated as principal mediators of protection against HCMV. Initial studies of LIR-1 suggested that it is the only LIR expressed by T cells, whereas other LIRs are expressed by NK cells (LIR-4, -7, and -8) and B cells (LIR-3, -4, and -6) [1]. Furthermore, the LIR-1 inhibitory signalling pathway is functional in T cells [11, 12]. Previous studies have established predominant expression of LIR-1 by CD8+ CTLs [13] rather than by CD4+ T cells. Furthermore, in patients who have undergone lung transplantation, increased expression of LIR-1 by the NK and T cell pool has been correlated with occurrence of HCMV disease. However, the extent of expression of LIR-1 by HCMV-specific T cells, or by those specific for other viruses, is currently unclear, and an understanding of the control of expression of LIR-1 during viral infection is also lacking. To gain a greater understanding of the role that LIR-1 plays in the regulation of virus-specific T cell responses and in HCMV immune evasion, we chose to examine cell-surface expression of LIR-1 during acute and persistent viral infection.
MATERIALS AND METHODS
Collection and isolation of peripheral blood mononuclear cells (PBMCs).
HCMV serostatus was determined from serum samples using an agglutination kit (CMVScan; BD Biosciences). HLA tissue typing was performed on DNA samples extracted from blood, using the DNeasy Tissue Kit (Qiagen). For cell staining, PBMCs were isolated from blood samples using Lymphoprep (Axis-Shield UK). Cells were then washed twice in RPMI 1640 medium and resuspended in RPMI 1640 medium supplemented with 10% fetal calf serum (FCS), 2 mmol/L L-glutamine, and 100 U/mL penicillin-streptomycin.
Generation of tetrameric HLA-peptide complexes.
Soluble HLA-peptide monomers were produced by dilution refolding, biotinylated, and tetramerized by conjugation to phycoerythrin (PE)conjugated streptavidin, as described elsewhere [14]. HLA-peptide tetramers used the epitopes [1529] outlined in table 1. HCMV peptides were derived from either pp65 (a major viral tegument protein that is highly immunodominant), immediate-early 1 protein (IE-1; an abundantly expressed transcriptional regulator), or pp50 (a viral DNA-binding protein that acts as a polymerase processivity factor). Epstein-Barr virus (EBV) peptides were derived from either latent membrane protein 2A (LMP2A; a viral membrane protein that is expressed during and is important for maintenance of viral latency), EBV nuclear antigen 3A (EBNA3A; a protein expressed during latent EBV infection that regulates transcription of LMP1 and LMP2), or BZLF1, BRLF1, BMLF1, and BMRF1 (a set of transcriptional regulators that play a key role in the switch from latent to lytic infection). Influenza virus peptides were derived from either the matrix protein (a highly immunodominant antigen that forms a protein layer surrounding the nucleocapsid) or the influenza nucleoprotein (an abundant protein that encapsidates the single-stranded RNA viral genome for the purposes of transcription and packaging).
Flow-cytometric staining.
PBMCs were stained with the appropriate HLA-peptide tetramer (for 30 min at 37°C), were washed in magnetic-activated cell sorting (MACS) buffer, and were incubated with LIR-1specific HP-F1 antibody or isotype control (for 30 min at 4°C). NK cell receptor antibodies were also used: CD158a (HP-3E4 and EB6), CD158b (GL183), CD94 (HP-3B1), NKG2A (Z199), and NKB1 (DX9). After being washed, cells were incubated with fluorescein isothiocyanate (FITC)conjugated goat antimouse IgG, and mouse serum was added to prevent secondary antibody from binding other mouse antibodies. Finally, cells were incubated with PC5-conjugated mouse anti-CD8 antibody. For analysis of expression of LIR-1 by CD4+ and CD8+ T cell subsets, cells were stained with either FITC-conjugated anti-CD3, PC5-conjugated anti-CD4, or anti-CD8 antibodies and PE-conjugated goat antimouse secondary IgG. After staining, cells were washed in MACS buffer, resuspended, and analyzed by flow-cytometric analysis. Phenotyping of LIR-1+ and LIR-1- antigen-specific CTLs was performed by 4-color flow-cytometric analysis using a Fc500 flow cytometer (Beckman-Coulter) with HP-F1, Cy5-conjugated rabbit antimouse IgG, ECD-conjugated anti-CD8 monoclonal antibody (MAb), and FITC-conjugated anti-CD28, anti-CD57, anti-CD45RA, or anti-CD45RO MAbs. Data were interpreted using WinMDI software (version 2.8).
Intracellular cytokine staining.
To assess antigen-specific production of interferon (IFN), PBMCs were isolated from HCMV-seropositive individuals and incubated separately for 5 h with HCMV peptide, irrelevant control peptide, media alone, and staphylococcus enterotoxin B. Cells were then washed and incubated with HP-F1 followed by goat antimouse PE and then with CD8- and CD3-specific antibodies conjugated to ECD and allophycyanin, respectively. Cells were then permeablized, incubated with antiIFN- antibody, and fixed using the BD Intracellular Cytokine Staining Kit (BD Biosciences). Results were assessed by 4-color flow-cytometric analysis using a BD FACSCalibur (BD Biosciences).
Statistical analysis.
Statistical analysis was performed using the Analyse-It statistics software package (Analyse-It Software; version 1.68) with Microsoft Excel. The 2-tailed Mann-Whitney U test was used to determine the significance of differences between 2 groups.
RESULTS
Enhanced expression of LIR-1 by CD4+ and CD8+ T cell subsets during persistent HCMV infection.
To determine the extent of expression of LIR-1 on T cells during persistent HCMV infection, the HP-F1 antibody was used to compare staining of cell-surface LIR-1 on CD4+ and CD8+ T cell subsets from 10 healthy HCMV-seropositive and 10 healthy HCMV-seronegative individuals (figure 1A and 1B). There was no significant difference in the mean (±SD) ages of the 2 groups (33.3 [±11.6] vs. 35.8 [±12.4] years; P = .97), and each group contained 6 men and 4 women. LIR-1 was preferentially expressed on the CD8+ T cell subsets from both groups, which is in agreement with previous studies [13]. Expression of LIR-1 was significantly enhanced on T cells from HCMV-seropositive individuals, since the mean frequency of expression of LIR-1 on CD3+ T cells was significantly higher for HCMV-seropositive individuals than for HCMV-seronegative individuals. These results are consistent with those of a recent study that also indicated enhanced expression of LIR-1 on CD3+ T cells from HCMV-seropositive individuals [30]. In addition, this difference was statistically significant for both the CD8+ (mean ± SD, 35.5% ± 15.6% vs. 17.2% ± 8.1%; P = .012) and CD4+ (mean ± SD, 2.8% ± 3.1% vs. 0.6% ± 1.0%; P = .007) T cell subsets.
Low frequency of expression of LIR-1 on EBV-specific CTLs.
Having demonstrated that the majority of HCMV-specific CTLs expressed LIR-1 on the cell surface, we determined the extent of expression of LIR-1 on CTL populations specific for other viruses. We examined expression of LIR-1 on EBV-specific CTLs from HCMV-seronegative and HCMV-seropositive individuals (figure 2) by examining responses specific for peptide antigens from the lytic (GLC and RAK) and the latent (CLG and FLR) phases of infection. Eight healthy HCMV-seronegative individuals (5 men and 3 women) in whom EBV-specific CTL populations could be detected were identified (figure 2). In this HCMV-seronegative cohort, EBV-specific CTLs expressed LIR-1 at significantly lower frequencies than did HCMV-specific CTLs previously assessed (mean ± SD, 13.3% ± 6.1% vs. 72.0% ± 20.5%; P < .0001) (figure 2A and 2B). Similarly, in an EBV-infected HCMV-seropositive cohort (14 coinfected individuals), the frequency of expression of LIR-1 on EBV-specific CTLs was similar to that in an EBV-infected HCMV-seronegative cohort (mean ± SD, 14.2% ± 11.5% vs. 13.3% ± 6.1%) and was significantly lower than that on HCMV-specific CTLs from these coinfected individuals (mean ± SD, 14.2% ± 11.5% vs. 71.5% ± 22.1%; P < .001) (figure 2C). These results indicate that, irrespective of HCMV serostatus, EBV-specific CTLs exhibit a lower frequency of expression of LIR-1 than do HCMV-specific CTLs. Furthermore, in both HCMV-seropositive and HCMV-seronegative cohorts, expression of LIR-1 on CTLs specific for EBV latent-cycle antigens was lower than that on CTLs specific for lytic-cycle antigens (mean ± SD, 7.6% ± 1.7% vs. 15.8% ± 5.6% [P = .0196, for the HCMV-seronegative cohort]; 6.3% ± 4.4% vs. 18.1% ± 11.9% [P = .005, for the HCMV-seropositive cohort]) (figure 2B). These results strongly suggest that, as for HCMV responses, antigen-specific effects influence expression of LIR-1 on EBV-specific CTLs.
Expression of LIR-1 on influenza virusspecific CTLs.
As a second comparison, expression of LIR-1 on CTLs specific for influenza A virus was determined. Two HLA-peptide tetramers (A2 GILGFVFTL and B8 ELRSRYWAI) were used to detect influenza virusspecific CTL responses in 6 healthy individuals. The mean ± SD frequency of expression of LIR-1 on influenza virusspecific CTLs (28.4% ± 15.1%) was significantly lower than that on HCMV-specific CTLs but was higher than that on EBV-specific CTLs (figure 2B). Four of these 6 individuals exhibited detectable EBV-specific CTLs, and, in each case, the frequency of expression of LIR-1 on influenza virusspecific CTLs was substantially greater than that on EBV-specific CTLs, whereas expression of LIR-1 on EBV-specific CTLs was similar to that previously measured (11.1% vs. 13.3%14.2%). In the 2 individuals with detectable CD8+ T cell responses to influenza virus and HCMV, expression of LIR-1 on influenza virusspecific CTLs was considerably lower (26.5% vs. 78.6% and 8.0% vs. 41.0%). Finally, in the 1 individual with detectable HCMV-, EBV-, and influenza virusspecific CTL responses, the frequencies of expression of LIR-1 were hierarchical, in the following order: HCMV (78.6%), influenza virus (26.5%), and EBV (7.7% [lytic response] and 1.6% [latent response]).
Enhanced expression of differentiation markers on LIR-1+ HCMV-specific CTLs, compared with that on LIR-1- HCMV-specific CTLs.
Previous studies using HLA-peptide tetramers have defined a characteristic pattern of cell-surface differentiation markers by HCMV-specific CD8+ T cells from persistently infected individuals, comprising an absence of expression of secondary lymphoid homing receptors CD62L and CCR7, low levels of costimulatory molecules CD27 and CD28, increased expression of the carbohydrate antigen CD57 (compared with naive cells), and incomplete reversion of antigen-experienced CD45RO+ cells to the CD45RA+ phenotype [18, 3133]. This phenotype is thought to be characteristic of differentiated effector memory CTLs that secrete cytokines, express Fas ligand, contain perforin and granzyme B, and exhibit high cytolytic activity without prestimulation [18, 33, 34]. To assess whether cell-surface expression of LIR-1 on HCMV-specific CTLs was correlated with the degree of cellular differentiation, we determined the expression of CD57, CD28, CD45RO, and CD45RA on LIR-1+ versus LIR-1- HCMV-specific CTLs from 6 healthy HCMV-seropositive individuals. A substantially higher proportion of LIR-1+ CTLs were CD28- and CD57+, compared with LIR-1- CTLs, indicating that expression of LIR-1 correlated with acquisition of a more highly differentiated effector memory phenotype (figure 3A). However, no significant differences in expression of CD45RA or CD45RO were observed between LIR-1+ and LIR-1- subsets of HCMV-specific CTLs (figure 3A). To formally test whether LIR-1+ HCMV-specific CTLs were capable of effector function, we assayed production of IFN-, using intracellular cytokine staining (figure 3B). In 3 of 4 HCMV-specific CTL responses, the overwhelming majority of cells that produced IFN- (90%, 87%, and 68%) were LIR-1+, as assessed by HP-F1 staining, indicating that expression of LIR-1 is consistent with significant effector function. To provide a comparison with these analyses, expression of CD27, CD28, CD57, CD45RA, and CD45RO on the HCMV-specific CTL population from the individual exhibiting the lowest frequency of expression of LIR-1 (6.2%) was determined. Unlike CTLs from the majority of HCMV-seropositive individuals, a high percentage of HCMV-specific CTLs from this individual expressed CD28 (88.2%) and did not express CD57 (93.5%). CD27 was also expressed at a high frequency (94.6%) by these CTLs. These results confirm that, conversely, low-level expression of LIR-1 correlates with incomplete differentiation of HCMV-specific CTLs.
Expression of LIR-1 during primary viral infection.
In addition to expression of LIR-1 during persistent viral infection, we also studied expression of LIR-1 during primary symptomatic EBV and HCMV infections. In 3 individuals with infectious mononucleosis due to primary EBV infection, the magnitude of CTL responses to latent-cycle epitopes increased over time, whereas the magnitude of remaining CTL responses decreased (figure 4A), which is consistent with our previous results [26]. Expression of LIR-1 on EBV-specific CTLs was initially low (<5%) and increased substantially (to >30%) among CTL populations specific for lytic-cycle epitopes. For EBV latent-cycle CTL responses, the frequency of expression of LIR-1 increased but remained 10% at the final time point.
PBMCs from a single patient hospitalized with severe mononucleosis syndrome due to primary HCMV infection were also available. Expression of LIR-1 was detected on only 2% of HCMV-specific CTLs during mononucleosis but increased 7-fold, to 14% of CTLs, after 320 days of follow-up (figure 4B). These data demonstrate that the frequency of expression of LIR-1 on HCMV-specific CTLs during primary viral infection is initially low and increases over time.
Expression of LIR-1 during HCMV reactivation.
Expression of LIR-1 on HCMV-specific CTLs from 3 patients who had received allogeneic stem cell transplantation (SCT), were being treated with immunosuppressive drugs, and had experienced episodic HCMV reactivation was also studied (figure 5). The frequency of expression of LIR-1 on these HCMV-specific CTLs was determined over a range of time points. The proportion of HCMV-specific CTLs that expressed LIR-1 was variable but generally lower than the mean expression of LIR-1 on HCMV-specific CTLs during persistent infection, and it increased at successive time points. Over the same time course, the magnitude of HCMV-specific CTL responses typically decreased. The rate of change of expression of LIR-1 varied between responses, the most dramatic being an increase from 6.3% to 79.0% over the course of 242 days.
Expression of other class I MHCbinding NK cell immunoreceptors on HCMV-specific CTLs.
Having established detectable cell-surface expression of LIR-1 on the majority of HCMV-specific CTLs during persistent infection, we determined whether other known class I MHCbinding inhibitory receptors (killer immunoglobulin-like receptors and CD94/NKG2A/B/E) [35] were expressed by HCMV-specific CTLs from 8 persistently infected healthy HCMV-seropositive individuals (figure 6). In each individual, HCMV-specific CTL populations expressed KIRs (CD158a, CD158b, and NKB1) at frequencies 1%. In only 3 individuals was CD94/NKG2A expressed at frequencies >10% (greatest expression, 23.1% and 18.2% for CD94 and NKG2A, respectively). In contrast, HP-F1 stained LIR-1 on the majority of these HCMV-specific CTLs. Similar results were obtained for CTLs specific for both EBV and influenza virus (data not shown). CTLs from the individual with primary HCMV infection had marginally greater frequencies of expression of KIR, as well as of CD94, on 12.1% of cells and of NKG2A on 1.9% of cells (data not shown).
DISCUSSION
In the present study, we used the LIR-1specific antibody HP-F1 [36] and HLA-peptide tetramers [14] to assess the relative frequency of expression of LIR-1 on CTLs specific for different viruses. Our results indicate that, during persistent HCMV infection, LIR-1 is preferentially expressed on HCMV-specific CTLs, compared with expression on CTLs specific for other viruses, namely influenza virus and EBV. In addition, the enhanced expression of LIR-1 on CD4+ T cells from HCMV-seropositive individuals suggests that HCMV-specific CD4+ T cells might also preferentially express LIR-1 on their surface. Despite the fact that the LIR-1specific MAbs anti-CD85 and M405 have been found to exhibit similar patterns of staining HP-F1, it is possible that some degree of cell-surface LIR-1 was present on HP-F1nonreactive cells, since LIR-1 was detected on the surface of all T cells using the M402 MAb [11]. Differential cell-surface expression of LIR-1 was not simply a reflection of the high frequency of HCMV-specific CTLs, since EBV lytic antigenspecific CTLs circulating at high frequencies (up to 5.8% of the CD8+ T cell population) retained low-level expression of LIR-1 (23.5%), compared with HCMV-specific CTLs; nor was differential expression due to differences in overall propensity for expression of LIR-1 in individuals, since similarly contrasting frequencies of expression of LIR-1 on CTLs were observed between individuals coinfected with EBV, influenza virus, or both. Together, these data strongly argue that LIR-1 is differentially expressed on CTLs specific for different viruses, with especially high expression on HCMV-specific CTLs during persistent infection.
Our data support the idea that expression of LIR-1 increases during T cell differentiation. During primary EBV and HCMV infections, expression of LIR-1 was initially low, compared with that during persistent infection, and increased substantially over time as the CTL response evolved. One potential caveat is that, since symptomatic primary HCMV infection is unusual, whether virus-specific responses in the single patient studied are characteristic of most patients undergoing primary infection is not known. Comparison of cell-surface markers on HCMV-specific CTLs during persistent infection revealed that substantially higher proportions of LIR-1+ CTLs were CD28- and CD57+, compared with the LIR-1- subset. In several studies, down-regulation of the costimulatory molecules CD28 and CD27 has been associated with differentiation to an effector cell phenotype [32, 34, 3739]. Furthermore, increased expression of CD57 has been associated with highly differentiated effector T cells [33]. Up-regulation of expression of LIR-1 on HCMV-specific CTLs is therefore indicative of a highly differentiated effector memory phenotype. These data are consistent with those of previous studies that established predominant expression of LIR-1 on memory/CTL effector populations [13]. In contrast, no substantial differences were observed in expression of the tyrosine phosphatase isoforms CD45RA and CD45RO between the LIR-1+ and LIR-1- subsets. This could reflect the fact that reversion from expression of CD45RO to expression of CD45RA by antigen-experienced HCMV-specific CTLs occurs independently of up-regulation of expression of LIR-1 during differentiation.
Up-regulation of cell-surface expression of LIR-1 on differentiated effector CTLs could have important functional consequences. Such cells typically express cytolytic molecules, such as perforin [18] and granzyme B [34], and exhibit both cytotoxicity and cytokine secretion. Consistent with this, LIR-1+ HCMV-specific CTLs were competent producers of IFN-. Up-regulation of expression of LIR-1 on such cells, which typically have lost the requirement for costimulatory signaling through CD27 and CD28, is likely to provide negative signals that significantly restrain their highly efficient effector functions. Consistent with this, LIR-1/class I MHC interactions have been shown to inhibit cytokine secretion and cytotoxicity of CD8+ T cell clones in antibody-blocking experiments [11]. Alternatively, expression of LIR-1 may also limit the proliferative potential of such cells and render them more susceptible to apoptotic cell death induced by activation [13]. Expression of LIR-1 was not associated with up-regulation of other class I MHCbinding inhibitory receptors, since <1% of HCMV-specific CTLs expressed KIRs and since CD94/NKG2A was expressed in substantial numbers (23%/18% respectively) in only a minority of individuals. Similar patterns of expression of KIR and CD94/NKG2A were detected on EBV- and influenza virusspecific CTLs (data not shown). Therefore, up-regulation of expression of LIR-1 may commonly provide a nonredundant means by which effector CTLs receive class I MHCmediated negative regulatory signals. Regulated expression of LIR-1 during T cell differentiation parallels the situation in B cell development, in which analogous changes in expression have been noted, with LIR-1 absent at the preB cell stage and expression progressively increasing as these cells mature [40]. Consequently, a key role of LIR-1 might be to restrain the activity of diverse immune cells once they have differentiated into highly potent effector cells.
Since these data support a model whereby cell-surface expression of LIR-1 is initially low and is up-regulated as CTLs differentiate, it is likely that differences in the extent of differentiation of virus-specific CTL populations are related to their differential expression of LIR-1. Notably, EBV-specific memory CTLs are uniformly CD27+ and heterogenous for expression of CD28 and CD45RO/RA status [41], and CTLs specific for the influenza matrix protein are predominantly CD27+, CD28+, CD45RO+/RAlow, and CCR7+ [4244]. In contrast, HCMV-specific memory CTLs are predominantly CD27-, CD28-, and CD45RA+ [18]. This is consistent with HCMV, which elicits higher levels of CD8 immunity than does EBV or influenza virus, driving virus-specific CTLs toward a more highly differentiated phenotype and greater expression of LIR-1, compared with EBV-specific CTLs and influenza virusspecific CTLs. The nature of the antigen recognized, its relative level, and the duration of antigenic challenge that CTLs receive may be critical factors determining such increased expression of LIR-1. In the present study, HCMV-specific CTLs specific for IE-1 were significantly higher in frequency than were pp65-specific CTLs (3.8% vs. 1.3%; P = .0002) and expressed LIR-1 to a significantly greater extent (88.3% vs 66.3%; P = .02). Also, EBV-specific CTLs specific for lytic antigensexpression of which results in higher antigenic load (compared with that during latent infection [45])expressed LIR-1 to a significantly greater extent during persistent infection than did those specific for latent antigens (15.9% vs. 7.6%; P = .0196) and demonstrated more-substantial increases in expression of LIR-1 during primary EBV infection. This is consistent with previous studies indicating that, whereas EBV latent antigenspecific memory CTLs universally express CD45RO and CD28, lytic antigenspecific CTLs are heterogenous for expression of CD45RO/RA and CD28, suggesting a more highly differentiated phenotype [41]. Our results also indicate that the duration of antigen exposure clearly affects expression of LIR-1, since this increased during the course of primary EBV and HCMV infections. In the single case of primary HCMV infection, expression of LIR-1 on HCMV-specific CTLs was low (14.0%) at the final time point, compared with that during persistent infection (mean, 72.0%). However, if this rate of increase in expression of LIR-1 was sustained, this frequency would be achieved in 5 years. Furthermore, it is possible that the HCMV-seropositive individual with abnormally low expression of LIR-1 on HCMV-specific CTLs (6.2%) had been recently infected with HCMV; recent infection is consistent with limited duration of antigen exposure, lower expression of LIR-1, and incomplete differentiation, compared with the majority of HCMV-specific CTLs.
The present study has also established increases in expression of LIR-1 during acute HCMV infection due to viral reactivation, a clinically important complication of allogeneic SCT that frequently causes mortality. In each of the 3 patients who had received SCT, expression of LIR-1 on HCMV-specific CTLs increased over time. Expression of LIR-1 was greater on CTLs specific for IE-1 than on those specific for pp65, which is consistent with the situation during persistent infection. This indicates that the same factors that drive high expression of LIR-1 during persistent HCMV infection operate during acute infection in the post-SCT setting.
The present study suggests that, by engaging LIR-1 on the surface of T cells, the HCMV class I MHC homologue UL18 could preferentially target HCMV-specific effector memory CTLs. Recent studies suggest that UL18 is transcribed late during HCMV infection of fibroblasts and that it does reach the cell surface [46, 47]. Furthermore, UL18 protein can be expressed during HCMV infection in immunocompromised patients after SCT [47]. Our finding that other class I MHCinhibitory receptors are not expressed on most HCMV-specific CTL populations is consistent with recent work examining expression of NKG2C on pp65-specific CTL populations [30] and may explain why UL18 has evolved to target LIR-1, as opposed to KIRs or CD94/NKG2A. Since LIR-1 is an inhibitory receptor, the expected functional outcome of UL18/LIR-1 interactions on CTLs would be inhibitory. By analogy, LIR-1/class I MHC interactions have been shown to significantly inhibit cytokine secretion and cytotoxicity [11]. However, recent data suggest that engagement of LIR-1 (on CTLs) by UL18 (on infected cells) may actually increase CTL killing, although the mechanism underlying this effect is unclear [47]. Irrespective of the mode of action of UL18, our finding that the frequency of expression of LIR-1 on virus-specific CTLs is low during primary HCMV (and EBV) infection and increases significantly as the immune response progresses suggests that the effects of UL18 may be more significant in the maintenance of viral persistence during HCMV reactivation and prolonged infection than during primary infection.
Acknowledgments
We are extremely grateful to Karen Piper, Mark Cobbold, and Batoul Pourgheysari, for helpful discussions, critical evaluation of the work, and practical advice, and to Miguel Lopez-Botet, for kindly providing the HP-F1 antibody.
References
1. Borges L, Cosman D. LIRs/ILTs/MIRs, inhibitory and stimulatory Ig-superfamily receptors expressed in myeloid and lymphoid cells. Cytokine Growth Factor Rev 2000; 11:20917. First citation in article
2. Cella M, Nakajima H, Facchetti F, Hoffmann T, Colonna M. ILT receptors at the interface between lymphoid and myeloid cells. Curr Top Microbiol Immunol 2000; 251:1616. First citation in article
3. Cosman D, Fanger N, Borges L, et al. A novel immunoglobulin superfamily receptor for cellular and viral MHC class I molecules. Immunity 1997; 7:27382. First citation in article
4. Chapman TL, Heikema AP, Bjorkman PJ. The inhibitory receptor LIR-1 uses a common binding interaction to recognize class I MHC molecules and the viral MHC homolog UL18. Immunity 1999; 11:60311. First citation in article
5. Shiroishi M, Tsumoto K, Amano K, et al. Human inhibitory receptors ILT2 and ILT4 compete with CD8 for MHC class I binding and bind preferentially to HLA-G. Proc Natl Acad Sci USA 2003; 100:885661. First citation in article
6. Fahnestock ML, Johnson JL, Feldman RM, Neveu JM, Lane WS, Bjorkman PJ. The MHC class I homolog encoded by human cytomegalovirus binds endogenous peptides. Immunity 1995; 3:58390. First citation in article
7. Leong CC, Chapman TL, Bjorkman PJ, et al. 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 1998; 187:16817. First citation in article
8. Steffens HP, Kurz S, Holtappels R, Reddehase MJ. Preemptive CD8 T-cell immunotherapy of acute cytomegalovirus infection prevents lethal disease, limits the burden of latent viral genomes, and reduces the risk of virus recurrence. J Virol 1998; 72:1797804. First citation in article
9. Reusser P, Riddell SR, Meyers JD, Greenberg PD. Cytotoxic T-lymphocyte response to cytomegalovirus after human allogeneic bone marrow transplantation: pattern of recovery and correlation with cytomegalovirus infection and disease. Blood 1991; 78:137380. First citation in article
10. Walter EA, Greenberg PD, Gilbert MJ, et al. Reconstitution of cellular immunity against cytomegalovirus in recipients of allogeneic bone marrow by transfer of T-cell clones from the donor. N Engl J Med 1995; 333:103844. First citation in article
11. Saverino D, Fabbi M, Ghiotto F, et al. The CD85/LIR-1/ILT2 inhibitory receptor is expressed by all human T lymphocytes and down-regulates their functions. J Immunol 2000; 165:374255. First citation in article
12. Saverino D, Merlo A, Bruno S, Pistoia V, Grossi CE, Ciccone E. 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 2002; 168:20715. First citation in article
13. Young NT, Uhrberg M, Phillips JH, Lanier LL, Parham P. Differential expression of leukocyte receptor complex-encoded Ig-like receptors correlates with the transition from effector to memory CTL. J Immunol 2001; 166:393341. First citation in article
14. Altman JD, Moss PA, Goulder PJ, et al. Phenotypic analysis of antigen-specific T lymphocytes. Science 1996; 274:946. First citation in article
15. Longmate J, York J, La Rosa C, et al. Population coverage by HLA class-I restricted cytotoxic T-lymphocyte epitopes. Immunogenetics 2001; 52:16573. First citation in article
16. Wills MR, Carmichael AJ, Mynard K, et al. The human cytotoxic T-lymphocyte (CTL) response to cytomegalovirus is dominated by structural protein pp65: frequency, specificity, and T-cell receptor usage of pp65-specific CTL. J Virol 1996; 70:756979. First citation in article
17. Gavin MA, Gilbert MJ, Riddell SR, Greenberg PD, Bevan MJ. Alkali hydrolysis of recombinant proteins allows for the rapid identification of class I MHC-restricted CTL epitopes. J Immunol 1993; 151:397180. First citation in article
18. Khan N, Cobbold M, Keenan R, Moss PA. Comparative analysis of CD8+ T cell responses against human cytomegalovirus proteins pp65 and immediate early 1 shows similarities in precursor frequency, oligoclonality, and phenotype. J Infect Dis 2002; 185:102534. First citation in article
19. Kern F, Faulhaber N, Frommel C, et al. Analysis of CD8 T cell reactivity to cytomegalovirus using protein-spanning pools of overlapping pentadecapeptides. Eur J Immunol 2000; 30:167682. First citation in article
20. Elkington R, Walker S, Crough T, et al. Ex vivo profiling of CD8+-T-cell responses to human cytomegalovirus reveals broad and multispecific reactivities in healthy virus carriers. J Virol 2003; 77:522640. First citation in article
21. Steven NM, Annels NE, Kumar A, Leese AM, Kurilla MG, Rickinson AB. Immediate early and early lytic cycle proteins are frequent targets of the Epstein-Barr virus-induced cytotoxic T cell response. J Exp Med 1997; 185:160517. First citation in article
22. Bogedain C, Wolf H, Modrow S, Stuber G, Jilg W. Specific cytotoxic T lymphocytes recognize the immediate-early transactivator Zta of Epstein-Barr virus. J Virol 1995; 69:48729. First citation in article
23. Lee SP, Thomas WA, Murray RJ, et al. HLA A2.1-restricted cytotoxic T cells recognizing a range of Epstein-Barr virus isolates through a defined epitope in latent membrane protein LMP2. J Virol 1993; 67:742835. First citation in article
24. Burrows SR, Misko IS, Sculley TB, Schmidt C, Moss DJ. An Epstein-Barr virus-specific cytotoxic T-cell epitope present on A- and B-type transformants. J Virol 1990; 64:39746. First citation in article
25. Burrows SR, Gardner J, Khanna R, et al. Five new cytotoxic T cell epitopes identified within Epstein-Barr virus nuclear antigen 3. J Gen Virol 1994; 75:248993. First citation in article
26. Hislop AD, Annels NE, Gudgeon NH, Leese AM, Rickinson AB. Epitope-specific evolution of human CD8+ T cell responses from primary to persistent phases of Epstein-Barr virus infection. J Exp Med 2002; 195:893905. First citation in article
27. Saulquin X, Ibisch C, Peyrat MA, et al. A global appraisal of immunodominant CD8 T cell responses to Epstein-Barr virus and cytomegalovirus by bulk screening. Eur J Immunol 2000; 30:25319. First citation in article
28. Townsend AR, Rothbard J, Gotch FM, Bahadur G, Wraith D, McMichael AJ. The epitopes of influenza nucleoprotein recognized by cytotoxic T lymphocytes can be defined with short synthetic peptides. Cell 1986; 44:95968. First citation in article
29. Suhrbier A, Schmidt C, Fernan A. Prediction of an HLA B8-restricted influenza epitope by motif. Immunology 1993; 79:1713. First citation in article
30. Guma M, Angulo A, Vilches C, Gomez-Lozano N, Malats N, Lopez-Botet M. Imprint of human cytomegalovirus infection on the NK cell receptor repertoire. Blood 2004; 104:366471. First citation in article
31. Wills MR, Carmichael AJ, Weekes MP, et al. Human virus-specific CD8+ CTL clones revert from CD45ROhigh to CD45RAhigh in vivo: CD45RAhighCD8+ T cells comprise both naive and memory cells. J Immunol 1999; 162:70807. First citation in article
32. Gillespie GM, Wills MR, Appay V, et al. Functional heterogeneity and high frequencies of cytomegalovirus-specific CD8+ T lymphocytes in healthy seropositive donors. J Virol 2000; 74:814050. First citation in article
33. d'Angeac AD, Monier S, Pilling D, Travaglio-Encinoza A, Reme T, Salmon M. CD57+ T lymphocytes are derived from CD57- precursors by differentiation occurring in late immune responses. Eur J Immunol 1994; 24:150311. First citation in article
34. Hamann D, Baars PA, Rep MH, et al. Phenotypic and functional separation of memory and effector human CD8+ T cells. J Exp Med 1997; 186:140718. First citation in article
35. Natarajan K, Dimasi N, Wang J, Mariuzza RA, Margulies DH. Structure and function of natural killer cell receptors: multiple molecular solutions to self, nonself discrimination. Annu Rev Immunol 2002; 20:85385. First citation in article
36. Colonna M, Navarro F, Bellon T, et al. A common inhibitory receptor for major histocompatibility complex class I molecules on human lymphoid and myelomonocytic cells. J Exp Med 1997; 186:180918. First citation in article
37. Kern F, Khatamzas E, Surel I, et al. Distribution of human CMV-specific memory T cells among the CD8pos. subsets defined by CD57, CD27, and CD45 isoforms. Eur J Immunol 1999; 29:290815. First citation in article
38. Weekes MP, Carmichael AJ, Wills MR, Mynard K, Sissons JG. Human CD28-CD8+ T cells contain greatly expanded functional virus-specific memory CTL clones. J Immunol 1999; 162:756977. First citation in article
39. Vingerhoets JH, Vanham GL, Kestens LL, et al. Increased cytolytic T lymphocyte activity and decreased B7 responsiveness are associated with CD28 down-regulation on CD8+ T cells from HIV-infected subjects. Clin Exp Immunol 1995; 100:42533. First citation in article
40. Banham AH, Colonna M, Cella M, et al. Identification of the CD85 antigen as ILT2, an inhibitory MHC class I receptor of the immunoglobulin superfamily. J Leukoc Biol 1999; 65:8415. First citation in article
41. Hislop AD, Gudgeon NH, Callan MF, et al. EBV-specific CD8+ T cell memory: relationships between epitope specificity, cell phenotype, and immediate effector function. J Immunol 2001; 167:201929. First citation in article
42. He XS, Mahmood K, Maecker HT, et al. Analysis of the frequencies and of the memory T cell phenotypes of human CD8+ T cells specific for influenza A viruses. J Infect Dis 2003; 187:107584. First citation in article
43. Pittet MJ, Valmori D, Dunbar PR, et al. High frequencies of naive Melan-A/MART-1-specific CD8+ T cells in a large proportion of human histocompatibility leukocyte antigen (HLA)-A2 individuals. J Exp Med 1999; 190:70515. First citation in article
44. Ravkov EV, Myrick CM, Altman JD. Immediate early effector functions of virus-specific CD8+CCR7+ memory cells in humans defined by HLA and CC chemokine ligand 19 tetramers. J Immunol 2003; 170:24618. First citation in article
45. Rickinson AB, Kieff E. Epstein-Barr virus. In: Fields BN, Knipe DM, Howley PM, et al., eds. Fields virology. 4th ed. Philadelphia: Lippincott Williams & Wilkins, 2001:2575627. First citation in article
46. Park B, Oh H, Lee S, et al. 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 2002; 168:34649. First citation in article
47. Saverino D, Ghiotto F, Merlo A, et al. 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 2004; 172:562937. First citation in article