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
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 arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Böhm, V.
Right arrow Articles by Reddehase, M. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Böhm, V.
Right arrow Articles by Reddehase, M. J.

 Previous Article  |  Next Article 

Journal of Virology, December 2008, p. 11637-11650, Vol. 82, No. 23
0022-538X/08/$08.00+0     doi:10.1128/JVI.01510-08
Copyright © 2008, American Society for Microbiology. All Rights Reserved.

The Immune Evasion Paradox: Immunoevasins of Murine Cytomegalovirus Enhance Priming of CD8 T Cells by Preventing Negative Feedback Regulation{triangledown}

Verena Böhm, Christian O. Simon, Jürgen Podlech, Christof K. Seckert, Dorothea Gendig, Petra Deegen, Dorothea Gillert-Marien, Niels A. W. Lemmermann, Rafaela Holtappels, and Matthias J. Reddehase*

Institute for Virology, Johannes Gutenberg University, Mainz, Germany

Received 18 July 2008/ Accepted 12 September 2008


arrow
ABSTRACT
 
Cytomegaloviruses express glycoproteins that interfere with antigen presentation to CD8 T cells. Although the molecular modes of action of these "immunoevasins" differ between cytomegalovirus species, the convergent biological outcome is an inhibition of the recognition of infected cells. In murine cytomegalovirus, m152/gp40 retains peptide-loaded major histocompatibility complex class I molecules in a cis-Golgi compartment, m06/gp48 mediates their vesicular sorting for lysosomal degradation, and m04/gp34, although not an immunoevasin in its own right, appears to assist in the concerted action of all three molecules. Using the Ld-restricted IE1 epitope YPHFMPTNL in the BALB/c mouse model as a paradigm, we provide here an explanation for the paradox that immunoevasins enhance CD8 T-cell priming although they inhibit peptide presentation in infected cells. Adaptive immune responses are initiated in the regional lymph node (RLN) draining the site of pathogen exposure. In particular for antigens that are not virion components, the magnitude of viral gene expression providing the antigens is likely a critical parameter in priming efficacy. We have therefore focused on the events in the RLN and have related priming to intranodal viral gene expression. We show that immunoevasins enhance priming by downmodulating an early CD8 T-cell-mediated "negative feedback" control of the infection in the cortical region of the RLN, thus supporting the model that immunoevasins improve antigen supply for indirect priming by uninfected antigen-presenting cells. As an important consequence, these findings predict that deletion of immunoevasin genes in a replicative vaccine virus is not a favorable option but may, rather, be counterproductive.


arrow
INTRODUCTION
 
Cytomegaloviruses (CMVs) interfere with antigen presentation in the major histocompatibility complex class I (MHC-I) pathway of antigen presentation to CD8 T cells by encoding glycoproteins that specifically interact with peptide-loaded class I molecules to prevent their bulk flow vesicular trafficking to the cell surface (for reviews, see references 1, 22, 39, 41, 49, 51, 54, 56, 66, and 71). These specialized viral glycoproteins are referred to as immunoevasins (for a review, see reference 54) or viral proteins interfering with antigen presentation (49, 71). More recently, we have proposed the neutral acronym viral regulators of antigen presentation (vRAPs) to account not only for inhibitory but also for supportive effects of viral proteins on antigen presentation and/or CD8 T-cell priming (26). According to this definition, immunoevasins, or viral proteins interfering with antigen presentation, are "negative" vRAPs. Although vRAPs of different CMV species differ in the molecular modes of action in cells of their respective host species, a phenomenon which is likely to reflect the evolutionary adaptation of these strictly host species-specific viruses, the common trait is the inhibition of antigenic peptide presentation.

In the specific case of murine CMV (mCMV), three vRAPs have been identified (for a review, see reference 54). m152/gp40, when expressed alone, is most efficient in blocking the cell surface transport of peptide-loaded class I molecules but less efficient in downmodulating total cell surface class I (26, 68). m152/gp40 transiently interacts with class I molecules and mediates their retention in an endoplasmic reticulum cis-Golgi intermediate compartment (14, 72) by a mechanism that is not well understood. m06/gp48 stably binds to class I molecules and links them to the heterotetrameric cellular adaptor proteins AP-1A and AP-3A that mediate sorting of the complexes into the late endosome/lysosome pathway for degradation (57, 58). When expressed alone, m06/gp48 is most efficient in downmodulating total cell surface class I but less efficient in preventing the presentation of newly formed MHC-peptide complexes (26, 68). Finally, m04/gp34 also binds stably to class I molecules, with the complex reaching the cell surface (32, 33). When expressed alone, m04/gp34 does not act as an immunoevasin (26, 50) and may even exert a positive vRAP function when expressed along with the negative vRAP m152/gp40 in cells infected with mutant virus mCMV-{Delta}m06 (26). Its classification as an immunoevasin is based solely on some alleviation of inhibition observed in cells infected with mutant virus mCMV-{Delta}m04 (26, 31, 50).

With regard to the in vivo role of vRAPs, their effects could possibly modulate the priming of a CD8 T-cell response, the shaping of the memory CD8 T-cell repertoire, and the infection surveillance by CD8 T cells in target organs of CMV disease. These important issues cannot be experimentally addressed for human CMV in humans. So, various mouse models of CMV infection and disease are used to derive paradigmatic predictions by comparing mCMV-wild type (WT) and vRAP gene deletion mutants of mCMV lacking the dominant vRAP m152 or all three vRAPs (for reviews, see references 15 and 27). Notably, vRAPs were found to have remarkably little impact on the control of infection in immunocompetent C57BL/6 mice (17), which may in part be explained by a dominant natural killer (NK) cell response in this particular mouse strain mediated through interaction of the viral gene product m157 with the activating NK cell receptor Ly49H (5, 10, 64; for a recent review, see reference 38). Nevertheless, expression of vRAPs in infected host tissues significantly reduced the efficacy of antiviral control exerted by adoptively transferred CD8 T cells in the immunocompromised host (34, 35, 56), including immunocompromised C57BL/6 mice (28; reviewed in reference 27). The finding that the effect of vRAPs is barely significant in most organs of immunocompetent mice is probably also due to the fact that WT virus expressing vRAPs is already quite efficiently controlled so that the increment for improvement arising from a deletion of vRAPs is small. In accordance with this interpretation, the influence of vRAPs became visible under conditions of limited CD8 T-cell control, namely, in the salivary glands (42) that represent a privileged site of persistent CMV infection (reviewed in reference 11). It has been proposed, therefore, that facilitation of virus transmission via saliva might be the evolutionary role of immunoevasins (42).

The finding that deletion of vRAPs improves the presentation of antigenic peptides and may even lead to presentation of epitopes that are otherwise not presented at all raised hope that a vRAP gene deletion mutant might qualify as a superior vaccine that excels natural infection in the priming of a protective CD8 T-cell response. It was therefore both challenging and disillusioning when Gold and colleagues (16) presented a first example to indicate that deletion of vRAP m152 does not improve CD8 T-cell priming. This was further elaborated by showing that deletion of vRAP m152 as well as of all three vRAPs does not notably alter the immunodominance hierarchies for a panel of epitopes in both the C57BL/6 (H-2b) and the BALB/c (H-2d) model (27, 45, 56), at least not in the spleen. It was not unreasonable to speculate that vRAPs play no role in priming because priming might occur through antigen uptake and presentation by uninfected dendritic cells (DCs), in which vRAPs are absent, rather than through direct antigen presentation by infected cells expressing vRAPs.

Here, we have focused on the earliest events in CD8 T-cell priming, which take place in the peripheral perifollicular region of the regional lymph node (RLN) that drains the local site of virus exposure. We show that antiviral effector CD8 T cells are rapidly induced in the RLN and have the potential to limit priming in a sort of negative feedback mode by controlling the infection early on, which is reflected by a reduced intranodal viral gene expression. Negative feedback operates in the absence of vRAPs when epitopes are presented by the infected cells but is blocked in the presence of vRAPs. This mechanism provides an explanation for the paradox reported here that immunoevasins facilitate CD8 T-cell priming.

(This research was conducted by Verena Böhm in partial fulfillment of the requirements for a doctoral degree (Dr. rer. nat.) from Philipps-University, Marburg, Germany.)


arrow
MATERIALS AND METHODS
 
Procedures of infection. Subcutaneous, intraplantar infection of 8- to 10-week-old female BALB/cJ mice (haplotype H-2d) was performed at the left hind footpad (for details, see reference 52) with 105 PFU of bacterial artificial chromosome (BAC)-derived mCMV MW97.01 (69), here referred to as mCMV-WT.BAC, or immunoevasin gene deletion mutant mCMV-{Delta}m04+m06+m152 (68), here referred to as mCMV-{Delta}vRAP. For determining the dose-response curves of CD8 T-cell priming and of intranodal viral transcription, infection doses were graded in log10 steps ranging from 105 PFU to 1 PFU. Sucrose density gradient-purified virions were used throughout (37, 52). Inactivated virions were generated by exposure to high-intensity radiation in the UVC range (Low-Heat UV sterilizer UVC-CRL 400; Umwelt und Technik GmbH, Horb, Germany) for 15 min on ice, immediately before use. Inactivation was verified by the absence of intranuclear IE1 (immediate-early 1) protein in mock-infected mouse embryo fibroblasts (MEF) in a fluorescence analysis.

For testing of viral fitness in vivo, mice were immunocompromised by hematoablative total-body gamma irradiation with a single dose of 6.5 Gy delivered by a 137Cs gamma-ray source. Mice were bred and housed under specific-pathogen-free conditions at the Central Laboratory Animal Facility of the Johannes Gutenberg University, Mainz, Germany. Animal experiments were approved according to German federal law, permission numbers 177-07/021-28 and 177-07-04/051-62.

Comparison of the replicative fitness of viruses in tissues of the immunocompromised host. The in vivo replicative potential of viruses was determined by establishing virus growth curves for host tissues in the absence of immune control (61, 70). Specifically, BALB/cJ mice were immunocompromised and infected (see above) with the viruses under investigation. At defined time points after infection, virus replication in spleen and lungs was assessed by quantification of infectious virus in the respective organ homogenates by using a virus plaque assay (PFU assay) on subconfluent second-passage MEF monolayers with the technique of centrifugal enhancement of infectivity, as described in greater detail elsewhere (52). The number of infected cells in tissue sections of livers was determined by immunohistochemistry (IHC) specific for the viral IE1-pp89/76 protein using the peroxidase-diaminobenzidine-nickel method for black staining of infected nuclei (18, 30, 52). Growth curves were established on the basis of four to five mice tested individually per time point.

Isolation of total RNA from the PLN. Popliteal lymph nodes (PLNs) were shock-frozen in liquid nitrogen, and total RNA of the PLN of each individual mouse was isolated by using an RNeasy Lipid Tissue Mini Kit (catalog no. 74804; Qiagen, Hilden, Germany) as recommended by the supplier. In brief, PLNs were homogenized and lysed in a Qiagen MM300 mixer mill with QIAzol lysis reagent and two steel balls (0.118 in.) at 30 Hz, twice for 3 min, interrupted by rotation of the Mixer Mill rack to ensure even homogenization. Subsequent addition of chloroform separated the sample into three phases. The aqueous upper phase was transferred after the addition of ethanol to an RNeasy Mini spin column. After a first washing step, an in-column DNase digestion was performed by using the RNase-free DNase set (catalog no. 79254; Qiagen). After three further washing steps, the RNA was eluted twice in 30 µl of RNase-free water for storage at –70°C.

Quantification of viral transcription. For the quantification of spliced IE1 (m123/ie1) transcripts in total RNA derived from infected PLNs, real-time one-step reverse transcription-PCR (RT-PCR) was performed with primers and a dual-labeled probe, as described in greater detail previously (61). Absolute quantifications were performed with graded numbers of IE1 in vitro transcripts (36) on an ABI Prism 7500 Sequence Detection System (Applied Biosystems, Foster City, CA) with PCR conditions as described before (61), except that 500 ng of total RNA was used as a template and that standard titrations ranged from 106 to 101 in vitro transcripts.

Quantification of viral genomes in viral inoculum. Viral DNA was isolated from log10-graded doses of infectious virus, ranging from 105 PFU to 10 PFU, as recommended by the supplier by using a High Pure Viral Nucleic Acid Kit (catalog no. 11858874001; Roche, Mannheim, Germany), except that the DNA was eluted in 100 µl of elution buffer. Absolute quantification was performed on an ABI Prism 7500 Sequence Detection System with a QuantiTect Sybr Green PCR kit (catalog no. 204143; Qiagen). A 2-µl aliquot of the DNA was added as template DNA to a reaction mixture that included the QuantiTect Sybr Green PCR master mix and a 1 µM concentration of each primer. Primers LCgB-forw and LCgB-rev were used for amplification of a 135-bp fragment of the viral gene M55 (gB) as described previously (63). PCR was performed with the following cycler conditions: an initial 15 min at 95°C for HotStarTaq DNA polymerase activation, followed by 50 cycles of 15 s at 94°C, 30 s at 62°C, and 45 s at 72°C. Data were obtained during the extension period. After amplification, melting curve analysis of the PCR products was performed by raising the temperature to 95°C for 15 s, cooling down rapidly to 60°C for 1 min, and then slowly raising the temperature again to 95°C, with the fluorescence being recorded continuously. Standard curves for the quantifications were established by using graded numbers of linearized plasmid pDrive_gB_PTHrP_Tdy as a template (63).

gB-specific IHC analysis of lymph node infection. PLNs were embedded in paraffin and were processed for the preparation of 2-µm tissue sections and subsequent detection of viral proteins in infected tissue cells according to standard histological procedures (52). Infected cells in PLNs of immunocompetent as well as of immunocompromised BALB/cJ mice were visualized by using a modification of the alkaline phosphatase-anti-alkaline phosphatase method to label the viral glycoprotein gB. Specific staining was achieved with custom-made (Eurogentec, Köln, Germany) polyclonal rabbit antibodies directed against the synthetic gB peptide 113-AVSTDLVRFGKSIDC-127 (amino acid sequence according to EMBL accession number M86302) that was predicted by antigenicity algorithms. Paraffinized sections of PLN tissue were digested for 15 min at 37°C in EDTA-trypsin solution (1:1; 1.25 g/liter each), followed by heat-induced epitope retrieval performed in a 10 mM trisodium citrate-dihydrate buffer at pH 6 for 5 min at 95°C. To block unspecific antibody binding sites, tissue sections were overlaid with goat serum (diluted 1:10 in Tris-buffered saline [TBS]; catalog no. 16210064; Invitrogen Life Technologies) for 20 min at ~22°C. Samples were then incubated for 18 h at ~22°C with affinity-purified polyclonal rabbit antibodies (1:500 in TBS). An alkaline phosphatase-conjugated mouse {alpha}-rabbit "bridging antibody" (catalog no. A2306; Sigma), diluted 1:25 in TBS, was added, followed by an incubation for 20 min at ~22°C. Alkaline phosphatase-rabbit anti-alkaline phosphatase soluble complex (catalog no. A9811; Sigma), diluted 1:10 in TBS, was added, followed by an incubation for 30 min at ~22°C. Red staining was performed by using a DakoCytomation Fuchsin Substrate-Chromogen Kit (catalog no. K0624; Dako, Hamburg, Germany) as recommended by the supplier. Finally, sections were stained with hematoxylin for a few seconds at ~22°C, resulting in a light-blue counterstaining.

Depletion of lymphocyte subsets in vivo. The depletion of CD8 T lymphocytes was performed by a single intraperitoneal injection of 1 mg of purified monoclonal antibody (MAb) directed against CD8 (clone YTS169.4) (12) into mice at 24 h prior to infection with the viruses under investigation. NK cells were depleted likewise with 25 µl of rabbit antiserum directed against asialo-GM1 (catalog no. 986-10001; Wako Chemicals, Osaka, Japan). The efficacy of lymphocyte subset depletion was assessed by cytofluorometric analysis of splenocytes and of PLN cells. In the case of CD8 subset depletion, cells were labeled for a two-color analysis with fluorescein isothiocyanate-conjugated MAb to mouse CD8a (clone 53-6.7; catalog no. 553031; BD Biosciences Pharmingen) and phycoerythrin-conjugated MAb to mouse T-cell receptor (TCR) beta chain (clone H57-597; catalog no. 553172; BD Biosciences Pharmingen). In the case of NK cell depletion, a three-color analysis was performed with fluorescein isothiocyanate-conjugated MAb to mouse TCR beta chain (clone H57-597; catalog no. 553171; BD Biosciences Pharmingen), phycoerythrin-conjugated MAb to mouse CD49b (clone DX5; catalog no. 553858, BD Biosciences Pharmingen), and allophycocyanin-conjugated MAb to mouse NKG2D (clone CX5; catalog no. 17-5882-82; eBioscience Inc., San Diego, CA). Measurements were performed with a FACSort instrument (Becton Dickinson) using CellQuest software.

In vivo priming of antiviral effector and memory CD8 T cells. Immunocompetent BALB/cJ mice were infected as described above with virus doses specified in the figure legends. For controlling the requirement of viral gene expression, mice were inoculated likewise with 105 PFU equivalents of UV-inactivated mCMV-WT.BAC virus (mCMV-WT.BACUV) or mCMV-{Delta}vRAPUV virus. Acutely sensitized CD8 T cells were derived from the PLNs or from spleens during the first week postinfection. Memory CD8 T cells were isolated from spleens at 5 months postinfection, when productive infection was cleared and viral latency established. CD8 T cells were purified from lymph nodes (pool of six PLNs) or spleens (pool of at least three spleens) by positive immunomagnetic cell sorting, as described previously (48).

Antigenic peptides and IE1 epitope-specific CTLL. A list of the currently known mCMV-specific H-2d class I (Kd, Dd, and Ld)-restricted antigenic peptides has been published previously (25), including the only recently identified Kd-restricted antigenic peptides M105 and m145 (29). Custom peptide synthesis to a purity of >80% was performed by Jerini Peptide Technologies (Berlin, Germany). An IE1 peptide (168-YPHFMPTNL-176)-specific, polyclonal cytolytic T-lymphocyte (CTL) line (IE1-CTLL) was generated and tested for purity, broadness of TCR Vβ-chain usage, and functional avidity as described previously (8, 48).

ELISPOT assay. A gamma interferon (IFN-{gamma})-based enzyme-linked immunospot (ELISPOT) assay was used to detect sensitization of CD8 T cells by MHC-I-presented synthetic or naturally processed peptides resulting in the secretion of IFN-{gamma} and consequent spot formation. Presentation of the IE1 epitope by MEF infected with mCMV-WT.BAC or mCMV-{Delta}vRAP was detected by using the IE1-CTLL as responder cells and the corresponding infected MEF as stimulator cells. For measuring frequencies of in vivo primed CD8 T cells, of acutely sensitized or memory CD8 T cells, optimized epitope presentation was achieved by using P815 (H-2d) mastocytoma cells as stimulator cells exogenously loaded for 1 h at ~ 22°C with synthetic peptides at a saturating concentration. For the determination of functional avidities, the loading concentrations of peptides were graded in log10 steps, as indicated in the figures, followed by washing to remove unbound peptide. The assay was performed as described in greater detail previously (reference 48 and references therein) in triplicate cultures with 105 stimulator cells per assay culture and with graded numbers of responder cells. After 18 h of cocultivation, plates were developed, and spots were counted. Spot sizes were measured with the automated ELISPOT reader Eli.Expert, version 2.1, using ELI.Analyze software, version 4.2 (A.EL.VIS GmbH, Hannover, Germany).

Mathematical calculations and statistical analyses. (i) Calculation of doubling times and half-lives. Linear regression lines in the semilogarithmic plot of log N(t) = at + log N(0), where N(t) is the experimentally determined virus titer or the number of infected cells at time t after infection, a is the slope of the regression line, and log N(0) is its ordinate intercept, were calculated by using the software Mathematica Statistics Linear Regression, version 6.0.1.0 (Wolfram Research, Inc., Champaign, IL). The doubling time of infectious virus or of infected cells in tissues was then calculated according to the formula log 2/a. The upper and lower 95% confidence limit values of slope a (determined from the ellipsoidal parameter confidence region) give the 95% confidence interval(s) (95% CI) of the doubling time. Half-lives of transcript quantities were calculated accordingly with negative values of the slope.

(ii) Calculation of CD8 T-cell frequencies from ELISPOT data. Frequencies of IFN-{gamma}-secreting, spot-forming cells and the corresponding 95% CIs were calculated based on the spot counts from triplicate cultures for each of three responder cell numbers seeded, that is, from nine values altogether, by intercept-free linear regression analysis as described previously (48), using the software Mathematica, version 6.0.1.0.

(iii) Significance analysis. The statistical significance of differences between two independent sets of data, such as numbers of viral transcripts present in the PLNs after infection with two different viruses or numbers of transcripts after infection with one virus in the presence or absence of a certain lymphocyte subset, was evaluated by using distribution-free Wilcoxon-Mann-Whitney (rank sum) statistics. An online calculator is provided on the website http://elegans.swmed.edu/~leon/stats/utest.html (Ivo Dinov, Statistics Online Computational Resources, UCLA Statistics, Los Angeles, CA). Differences are regarded as significant if the P value (two-tailed test) is <0.05.


arrow
RESULTS
 
vRAPs largely reduce but do not prevent the presentation of antigenic peptides. It is a pending question why CD8 T cells protect against infection with mCMV-WT, which expresses vRAPs, although vRAPs appear to completely inhibit the recognition of infected target cells if assessed in a cytolysis assay in vitro (26, 50). As shown previously in a direct comparison of a cytolytic assay and ELISPOT assay, measuring the sensitization of CD8 T cells on the basis of IFN-{gamma} secretion is more sensitive and a fairly reliable predictor for in vivo function (26). When tested in the ELISPOT assay with an IE1 (168-YPHFMPTNL-176)-Ld epitope-monospecific but still polyclonal IE1-CTLL (8, 48) representing a broad range of functional avidities, MEF infected with the vRAP gene deletion mutant mCMV-{Delta}vRAP stimulated almost all CTLs seeded (frequency [f], 184/200; 95% CI, 170/200 to 198/200) (Fig. 1A). By contrast, using cells of the very same IE1-CTLL as effector cells in the very same assay, only few cells (f, 14/200; 95% CI, 12/200 to 16/200) responded to stimulator cells infected with mCMV-WT.BAC expressing the three vRAPs (Fig. 1B). Notably, enhanced IE1 epitope presentation after deletion of vRAPs was also reflected by the amount of secreted IFN-{gamma} as indicated by increased spot sizes. Nonetheless, although the effect of vRAPs is impressive, the inhibition of peptide presentation was not complete since 7% (6% to 8%) of the seeded CTLs, most probably those with the highest functional avidity, still detected sufficient antigen to become activated. It should be noted that this percentage of responding cells differs between different CTLLs, even between those of the same epitope specificity, dependent upon the particular avidity distributions (D. Gillert-Marien, unpublished data). An identical message, albeit with numerically different data, was derived also from CTLLs specific for a panel of H-2d- and H-2b-restricted epitopes (26) or with ex vivo isolated mCMV-specific memory CD8 T cells (R. Holtappels, unpublished data) tested with infected MEF as stimulator cells, as well as from testing an M45 (507-VGPALGRGL-515)-Dd epitope-specific CTLL on infected bone marrow-derived DCs (26).


Figure 1
View larger version (26K):
[in this window]
[in a new window]

  		FIG. 1. Inhibition of IE1 epitope presentation by vRAPs. (A) ELISPOT analysis of the sensitization of IE1-CTLs by MEF stimulator cells infected with mCMV-{Delta}vRAP. (B) ELISPOT analysis of the sensitization of IE1-CTLs by MEF stimulator cells infected with mCMV-WT.BAC. Photographs of ELISPOT filters show IFN-{gamma} spots caused by epitope-sensitized responder cells among 200 IE1-CTLs seeded (a1 and b1). Original spot counts from triplicate cultures/filters for 50, 100, and 200 IE1-CTLs seeded and the resulting regression lines calculated by intercept-free linear regression analysis are shown (a2 and b2). Complete linear regression analysis indicates how frequencies (f) of responding cells and the corresponding 95% CIs (shown in parentheses) are determined from the calculated linear regression lines (central lines) and their 95% CIs (a3 and b3). Bars represent the estimated most probable frequencies, and error bars represent the corresponding 95% CIs. Spot size-frequency diagrams for the three cultures with 200 IE1-CTLs seeded are also shown (a4 and b4). Spot sizes are given in 50–1µm2.

Viral in vivo replicative fitness in the absence of immune control is not influenced by vRAPs. For the correct interpretation of any data concerning the in vivo role of vRAPs, it is obligatory to establish first that the deletion of vRAP genes does not show an unintended, nonimmunological phenotype with respect to viral replication in host tissues. Previous work has shown that mutant virus mCMV-{Delta}vRAP replicates like mCMV-WT.BAC in liver, spleen, and adrenal glands of immunocompromised BALB/c mice after intravenous coinfection (68). As we focus here on CD8 T-cell priming in the RLN that drains the local site of virus exposure, virus dissemination from the portal of entry to the RLN as well as to distant target organs is a parameter not covered by the previous study on intravenous infection that has bypassed putative local restraints. We have therefore tested virus dissemination to target organs and have determined the doubling times of viral infectivity in lungs and spleen as well as of the numbers of infected cells in the liver after intraplantar infection (Fig. 2). Viruses mCMV-WT.BAC and mCMV-{Delta}vRAP reached distant target organs at the same time (lungs and liver, between days 4 and 6; spleen, between days 2 and 4) and replicated there with doubling times that were identical within the 95% CI (lungs, ~14 h; spleen, ~11 h; and liver, ~24 h). In conclusion, in the immunocompromised host, vRAPs have no influence on virus dissemination to target tissues or on virus multiplication within target tissues. Thus, the mutant virus is not attenuated in terms of replicative fitness in the absence of immune control.


Figure 2
View larger version (16K):
[in this window]
[in a new window]

  		FIG. 2. Viral replicative fitness in vivo. Shown are virus growth curves in organs of immunocompromised BALB/c mice. For lungs and spleen, virus titers were determined in organ homogenates by the PFU assay. For the liver, the number of infected hepatocytes in representative tissue section areas of 10 mm2 was determined by IHC analysis specific for the intranuclear IE1 protein. Dots represent data from individual mice per time point, with median values (horizontal bars) marked. DL, detection limits of the respective assays. The doubling times ([DT] 95% CIs given in parentheses) were calculated by log-linear regression analysis. Solid regression lines and gray circles represent infection with mCMV-WT.BAC (WT.BAC); dotted regression lines and open circles represent infection with mCMV-{Delta}vRAP ({Delta}vRAP).

Infection in the draining lymph node localizes primarily to a peripheral region beneath the SCS. Adaptive immune responses are known to be initiated in the RLNs that drain the sites of antigen exposure, in the PLN in the specific case of intraplantar infection. Elegant previous work by the groups of J. R. Bennink and J. W. Yewdell has shown in vaccinia virus and vesicular stomatitis virus-based models of naïve CD8 T-cell priming that virions rapidly reach the RLN via the lymphatics, probably within minutes, and infect lymph node-resident antigen-presenting cells (APCs), including DCs, in a demarcated zone underneath the subcapsular sinus (SCS) (24, 46). Obviously, depending on cell tropism and length of the replication cycle, different viruses, in particular if they belong to different virus families, may behave differently in this respect. Since mCMV can indeed infect macrophages (reviewed in reference 19) as well as DCs (4, 13, 26, 43, 45), it was of interest to learn if mCMV, as a representative of the betaherpesvirus subfamily, also infects preferentially cells in the RLN periphery.

Therefore, immunocompetent and immunocompromised BALB/c mice were infected at the left hind footpad with mCMV-WT.BAC, and PLNs were analyzed by IHC specific for the essential virion envelope glycoprotein M55/gB, resulting in cytoplasmic staining of infected cells (Fig. 3). In both instances, the analysis was performed at the time of highest viral gene expression (Fig. 4), which was on day 2 and day 4 for immunocompetent and immunocompromised mice, respectively. Note the significant difference in PLN size and cellularity secondary to lymphocyte trapping in the PLN of the immunocompetent host and generalized leukocytopenia in the gamma-irradiated host. Whereas infected cells were scarce and thus barely detectable in the PLN in the presence of immune control (Fig. 3A), numerous infected cells were found to localize to a zone beneath the SCS, in the peri-SCS region, in immunocompromised mice (Fig. 3B, panel b1). When an area of this region was resolved to greater detail, some gB-positive cells showed morphology typical of DCs, with gB literally imaging the filipodia (Fig. 3B, panel b2). That cytoplasmic staining of gB reflected infection of the cells and not just antigen uptake from the virion inoculum is verified by the presence of an intranuclear inclusion body (Fig. 3B, panel b2), which represents the site of viral DNA packaging and nucleocapsid assembly indicative of a late stage in the viral replication cycle. We attempted to further characterize the infected cells that were morphologically classified as DCs, but double staining for the intranuclear viral regulatory protein IE1 and DC markers, such as CD11c and MIDC-8 antigen, did not reveal IE1-positive DCs (data not shown). This does not necessarily argue against the presence of infected DCs, as mCMV infection of DCs is known to downmodulate the expression of MHC-I and MHC-II molecules as well as of various costimulatory molecules and adhesion molecules, including CD11c (7). It should be noted, however, that we occasionally observed infected macrophages (data not shown), and, based on the broad cell tropism of mCMV, also stromal cells and endothelial cells are likely targets of infection in the PLN.


Figure 3
View larger version (61K):
[in this window]
[in a new window]

  		FIG. 3. Localization of infected cells in the RLN draining the site of virus exposure. PLNs of immunocompetent (A) and immunocompromised (B) BALB/c mice were analyzed by gB-specific IHC (red cytoplasmic staining) on days 2 and 4 after intraplantar infection with mCMV-WT.BAC, respectively. Total PLN overview section reveals a demarcated peripheral zone of infection beneath the SCS, i.e., the peri-SCS region. An area of the infected peri-SCS region is resolved to greater detail to reveal the gB-expressing cells, some of which show a DC-like morphological phenotype (arrow) (b2). Asterisks mark intranuclear inclusion bodies that indicate a late stage of the viral productive cycle. Marked are two prominent examples, but other gB-positive cells show inclusion bodies as well. Magnifications are indicated by size markers.


Figure 4
View larger version (14K):
[in this window]
[in a new window]

  		FIG. 4. Time course of intranodal viral gene expression depending on the immune status. (A) Viral gene expression in the PLN of immunocompromised (6.5 Gy of gamma irradiation) mice. (B) Viral gene expression in the PLN of immunocompetent mice. Spliced IE1 transcripts coding for the IE1 epitope were quantitated by real-time RT-PCR at the indicated time points after intraplantar infection with 105 PFU of mCMV-WT.BAC (WT.BAC; gray circles) or mCMV-{Delta}vRAP ({Delta}vRAP; open circles). Symbols represent data from the ipsilateral PLN of individual mice. Median values are marked by horizontal bars. In panel B, log-linear regression lines are indicated.

The absence of vRAPs accelerates the clearance of intranodal viral gene expression in the immunocompetent host. It has been reported that naïve CD8 T cells encounter antigen in the peripheral interfollicular region of the RLN for direct priming (24). Intranodal viral gene expression is therefore likely to be important for the efficacy of priming, in particular, for the priming by epitopes that are not derived from virion proteins. The IE1 epitope 168-YPHFMPTNL-176 (55; reviewed in reference 54) is derived from the regulatory nonvirion protein IE1, and its generation thus depends on viral gene expression. For quantitating infection of the PLN in the time course, we employed a real-time RT-PCR specific for spliced IE1 transcripts. Transcription is the most sensitive measure of infection, more sensitive than the PFU assay, and unlike the quantitation of viral genomes by real-time PCR, transcription is not obscured by input virion DNA. In the immunocompromised host, ~106 IE1 transcripts/PLN were found as early as 24 h after intraplantar infection, and this amount increased 100-fold to ~108 transcripts/PLN by day 4 when saturation was reached (Fig. 4A). Importantly, in accordance with the virus growth curves shown in Fig. 2, no difference was imposed by the absence or presence of vRAPs, indicating that vRAPs have no direct influence on intranodal viral gene expression and that vRAP-controlled innate and adaptive immune functions were successfully abrogated by the immunocompromising regimen.

The time course of intranodal ie1 gene expression was distinctively different in immunocompetent mice. Again, high expression levels were initially reached, but under the influence of immune control the numbers of intranodal viral transcripts declined in a log-linear fashion after day 3 and after day 2 in the presence and absence of vRAPs, respectively. Importantly, deletion of vRAPs was found to accelerate this decline, as indicated by diverging regression lines (Fig. 4B). Specifically, the half-life of IE1 transcript quantity was 20.9 h (95% CI, 19.2 to 23.0 h) in the presence of vRAPs and 16.2 h (14.5 to 18.2 h) in the absence of vRAPs. Since the molecular half-life of IE1 mRNA is >24 h (61, 62), the decline in the numbers of IE1 transcripts does not appear to be caused by molecular decay of IE1 mRNA molecules but, rather, reflects elimination of infected cells by immune cells. Extrapolation of the regression lines to the time scale (abscissa) intercept predicts clearance of transcriptional activity by day 10.2 (95% CI, day 9.2 to 11.5) and day 14.2 (day 13.0 to 15.6) after infection with mCMV-{Delta}vRAP and mCMV-WT, respectively. This shows that vRAPs not only enhance transcription magnitude but also prolong the period of transcription in the RLN.

vRAP-modulated early control of intranodal viral gene expression involves NK cells and CD8 T cells. The very early onset of immune control of intranodal viral gene expression suggested a role for innate immune mechanisms rather than for CD8 T cells. NK cells represent a prime candidate since they are rapidly activated systemically during mCMV infection by IFN-{alpha}/β and interleukin-12 for cytotoxic activity and IFN-{gamma} production, respectively (47). The fact that vRAPs, specifically m152/gp40, downmodulate RAE-1 ligands of the activating NK cell receptor NKG2D (34, 40; reviewed in reference 38) might explain the improved control observed in the absence of vRAPs at early time points when a role for CD8 T cells was not to be expected. We therefore investigated if depletion of NK cells with anti-asialo-GM1 prior to infection abrogates the effect of vRAPs on intranodal ie1 gene expression in the first 3 days after infection with 105 PFU of the respective viruses (Fig. 5A). A reproducibly significant effect of vRAPs was observed in undepleted control groups for time points later than 48 h (Fig. 5, a1 and b1), indicating that immune control sets in between day 2 and day 3. Although the treatment with anti-asialo-GM1 successfully depleted TCR-β– CD49b+ NKG2D+ NK cells (Fig. 5A, a2) and although the effect of vRAPs was diminished by ~1 log10 on day 3, ie1 gene expression remained significantly lower in the absence of vRAPs (Fig. 5A, a1) (>1 log10 difference with a P of 0.01). This suggested an involvement of an additional antiviral effector cell type whose antiviral function is also influenced by the expression of vRAPs.


Figure 5
View larger version (28K):
[in this window]
[in a new window]

  		FIG. 5. Involvement of lymphocyte subsets in the early control of PLN infection. (A) Role for NK cells studied by in vivo depletion with anti-asialo-GM1 ({alpha}-ASGM1) antibody. (B) Role for CD8 T cells studied by in vivo depletion with anti-CD8 antibody. Shaded panels highlight groups with the indicated depletions. Spliced IE1 transcripts were quantitated by real-time RT-PCR at the indicated time points after intraplantar infection with 105 PFU of mCMV-WT.BAC (WT.BAC) or mCMV-{Delta}vRAP ({Delta}vRAP) (a1 and b1). Symbols represent data from the ipsilateral PLN of individual mice. Median values are marked by horizontal bars. P values (two-sided) for significance of differences are indicated for group comparisons of major interest. *, significance at P of <0.05. Three-color cytofluorometric verification of the depletion of TCR-β– CD49b+ NKG2D+ NK cells was performed at 72 h postinfection (a2). Shown are two-dimensional dot plots with 25,000 cells analyzed and 5,000 events displayed. Upper panel shows the forward scatter (FSC) versus TCR-β chain expression with the indicated gate restricting the analysis to TCR-β-negative cells. Lower panel shows NKG2D expression versus CD49b expression of cells preselected by the gates defined in the upper panel. The percentages of CD49b+ NKG2D+ cells are indicated. Two-color cytofluorometric verification of the depletion of CD8 T cells was performed at 72 h postinfection (b2). Shown are two-dimensional dot plots of CD8 expression versus TCR-β chain expression with 25,000 cells analyzed and 2,500 events displayed. The percentages of TCR-β+ CD8+ cells are indicated. {alpha}, anti.

We therefore asked in an independent second experiment if CD8 T cells are involved in the early control of intranodal viral gene expression (Fig. 5B). Notably, CD8 T-cell depletion, unlike NK cell depletion, not only reduced but abolished the effect of vRAPs at 72 h and 96 h. It must be emphasized that during the first 4 days, depletion of NK cells (Fig. 5A, a1) (P = 0.06 on day 3) or of CD8 T cells (Fig. 5B, b1) (P = 0.11 on day 4) had only slight impact on the absolute level of intranodal viral gene expression of mCMV-WT since numerical differences in the respective median values of transcript loads did not reach statistical significance. This finding is in accordance with the dual function of vRAP m152/gp40, which inhibits the recognition of infected cells by NK cells and by CD8 T cells (34). In sharp contrast, the vRAP gene deletion mutant mCMV-{Delta}vRAP clearly profited from the cell depletions in absolute terms (~2 log10 of transcript load) and with statistical significance (Fig. 5) (P = 0.01 and P = 0.03, respectively).

Altogether, these data show that vRAPs indirectly enhance intranodal viral gene expression by preventing immune recognition of infected cells and imply that priming of naïve CD8 T cells for antiviral effector function is a remarkably rapid process.

CD8 T-cell priming in the draining PLN is a rapid process. So far, the data have collectively shown that vRAPs have an impact on intranodal viral gene expression of mCMV-WT in that they increase viral fitness in the immunocompetent host by evading early antiviral effector cells. We therefore wondered how intranodal viral gene expression eventually translates into the efficiency of CD8 T-cell priming. To test this, frequencies of CD8 T cells specific for a panel of mCMV epitopes representing the immediate-early (IE1 epitope) and early (m164, m145, and M105 epitopes) phases of viral gene expression (25) were determined in the PLN at daily intervals until day 7 after infection with mCMV-WT and mCMV-{Delta}vRAP. Epitope-specific CD8 T cells became first detectable on day 3 at a low but statistically significant frequency that increased steadily until day 7 (Fig. 6) in inverse correlation with intranodal ie1 gene expression (Fig. 4B). In accordance with the data on CD8 T-cell depletion (Fig. 5B), this inverse correlation suggested that increasing immune control, primarily by the newly primed CD8 T cells, was responsible for the control of PLN infection. Strikingly, as a paradox of immune evasion, priming against mCMV-WT turned out to be more efficient than priming against mCMV-{Delta}vRAP, which is just the opposite of what one might have expected in view of the by far more efficient antigen presentation in cells infected with mCMV-{Delta}vRAP (Fig. 1). The advantage obviously taken from increased intranodal viral gene expression in the presence of vRAPs was not identical for all epitopes tested but was most pronounced for the IE1 epitope and was actually insignificant for the M105 epitope. Whether this relates to the fact that the M105 epitope is the only epitope tested that is derived from a virion protein (65) remains to be studied.


Figure 6
View larger version (11K):
[in this window]
[in a new window]

  		FIG. 6. Time course of CD8 T-cell priming in the PLN. CD8 T lymphocytes were isolated from ipsilateral PLN at the indicated time points after intraplantar infection with 105 PFU of mCMV-WT.BAC (WT.BAC) or mCMV-{Delta}vRAP ({Delta}vRAP) and were used as responder cells in an ELISPOT assay. Stimulator cells were P815 mastocytoma cells exogenously loaded with the indicated synthetic peptides at saturating concentrations of 10–7 M. Bars represent the frequencies of epitope-specific CD8 T cells, and error bars represent the corresponding 95% CI determined by intercept-free linear regression analysis, as outlined in greater detail in the legend of Fig. 1. Ø, no peptide added.

Intranodal viral gene expression in immunocompetent mice correlates with the dose of infection. It was predictable that the dose of infectious virions administered locally determines the degree of infection of the RLN, at least in the first round of viral replication, but the minimal dose of virus required for initiating detectable intranodal viral gene expression was unknown. We therefore related the amount of IE1 transcripts generated in the PLN to the dose of infection. To minimize the influence of immune control, the analysis was performed on day 1, that is, after the first round of virus replication but before the appearance of antiviral CD8 T cells (Fig. 7).


Figure 7
View larger version (13K):
[in this window]
[in a new window]

  		FIG. 7. Virion dose-dependence of intranodal viral gene expression. (A) Genome-to-infectivity relation in purified virion preparations of mCMV-WT.BAC (WT.BAC) and mCMV-{Delta}vRAP ({Delta}vRAP). Symbols represent data from independent samples. Median values are indicated by short horizontal bars. Inoculum virus doses were graded in log10 steps ranging from 10 PFU to 105 PFU, and viral genomes were quantitated by M55/gB gene-specific real-time PCR. (B) Corresponding quantification of IE1 transcripts in ipsilateral PLN by real-time RT-PCR on day 1 after intraplantar infection with the indicated doses of mCMV-WT.BAC (gray circles) or mCMV-{Delta}vRAP (open circles). Requirement for infectious virus was tested by mock infection with 105 PFU equivalents of UV light-inactivated virions (UV-105). Symbols represent data from the ipsilateral PLN of individual mice with median values being marked. DL, detection limit. Intercept-free linear regression analysis confirmed a correlation between the two variables with P values of <0.01 (null hypothesis of random distribution) for both viruses.

As an important technical control, the accuracy of the log10 dilution steps of purified virions is documented by log-linear regression analysis of viral genomes, using M55/gB-specific real-time PCR for quantitation (63) (Fig. 7A). As shown previously, the genome-to-infectivity ratio is ~500 genomes per PFU in the case of mCMV-WT.Smith (18, 37, 61), which is in fair accordance with the ordinate intercept of the mCMV-WT.BAC regression line of 237 (95% CI, 176 to 347) genomes per PFU. The corresponding analysis of mCMV-{Delta}vRAP revealed a somewhat higher genome-to-infectivity ratio, namely 1,460 (95% CI, 1,140 to 1,800) and, thus, a higher proportion of inefficient genomes. In our experience, such a variance between mCMV recombinants is not unusual and is thus not specific for mutant mCMV-{Delta}vRAP. As a consequence, priming doses were standardized on PFU instead of on viral genomes or viral particle numbers. That standardizing doses on the basis of the number of PFU indeed equalizes replicative fitness has been proven as documented above (Fig. 2 and 4A).

In essence, intranodal ie1 gene expression at 24 h correlated with the dose of infection, albeit with variance between individual mice within each dose group, suggesting that traveling of virions to the PLN is a critical variable (Fig. 7B). Infectivity proved to be required since even a high dose of nonreplicative UV light-inactivated virions failed to initiate any IE1 transcription in the PLN. Notably, between 10 and 100 PFU of infectious virus induced ie1 gene expression in some PLNs, whereas 103 PFU were required to initiate IE1 transcription in all PLNs tested. One may discuss a trend toward lower intranodal viral gene expression after infection in the absence of vRAPs, but this trend did not reach statistical significance.

Priming of CD8 T cells correlates with intranodal viral gene expression. We next investigated how the infection dose translates into the efficacy of CD8 T-cell priming. In accordance with the fact that IE1 is not a virion component (65), ie1 gene expression was required for priming since no IE1 epitope-specific CD8 T cells were detected in the draining PLN after high-dose intraplantar mock infection with 105 PFU equivalents of UV-inactivated virions of mCMV-WT.BACUV and mCMV-{Delta}vRAPUV (Fig. 8A). In conspicuous accordance with the dose dependence of intranodal viral gene expression (Fig. 7B), IE1-specific CD8 T cells became detectable at as little as 10 PFU, and their frequency increased with increasing doses. Thus, CD8 T-cell priming measured by ELISPOT assay is at least as sensitive as intranodal viral transcription measured by RT-PCR. Again, the inhibition of epitope presentation by vRAPs did not inhibit epitope-specific priming but enhanced it significantly.


Figure 8
View larger version (17K):
[in this window]
[in a new window]

  		FIG. 8. Impact of vRAPs on the dose efficiency of IE1 epitope-specific CD8 T-cell priming. (A) Acute response in the draining PLN. Intraplantar infection was performed with log10-graded doses of infectivity ranging from 1 PFU to 105 PFU of mCMV-WT.BAC (WT.BAC; gray bars for all panels) and mCMV-{Delta}vRAP ({Delta}vRAP; open bars for all panels). Requirement for infectious virus was tested by mock infection with 105 PFU equivalents of UV light-inactivated virions (UV-105). On day 7, CD8 T lymphocytes were isolated from pools of six PLNs per test group and were used as responder cells in an ELISPOT assay performed with P815 stimulator cells exogenously loaded with synthetic IE1 peptide at a saturating concentration of 10–7 M. Bars represent the frequencies of IE1 epitope-specific CD8 T cells, and error bars represent the corresponding 95% CI determined by intercept-free linear regression analysis as outlined in greater detail in the legend of Fig. 1. (B) Corresponding analysis of the acute response in the spleen is shown (b1). Memory response in the spleen was determined after 5 months (b2).

Expression of vRAPs enhances the IE1 epitope-specific acute and memory CD8 T-cell response in the spleen. We next addressed the questions of whether vRAP-mediated enhancement of IE1 epitope-specific priming in the RLN is transmitted to the downstream central lymphoid organ, the spleen, and whether it is maintained during the shaping of immunological memory. It is important to emphasize that the acute response in the spleen was measured in parallel to the response in the PLN of the same mice already used for the experiment shown in Fig. 8A and that the memory response in the spleen was measured for mice of the same infection groups. Over the entire virus dose range of intraplantar infection, frequencies of IE1 epitope-specific CD8 T cells present in the spleen were higher when priming was performed with mCMV-WT (Fig. 8B). This was true on day 7 for the acute phase of a primary infection (Fig. 8B, graph b1) as well as after establishment of viral latency and CD8 T-cell memory at 5 months after infection (Fig. 8B, graph b2).

Deletion of vRAPs does not notably change the avidity distribution in the IE1 epitope-specific CD8 T-cell population. As shown by Bousso and Robey in an analysis of the priming dynamics in the RLN (9), the overall avidity of interaction determines the probability that T cells will be stably captured by DCs. Furthermore, as it is known from recent work by Henrickson and colleagues (23), T-cell sensing of antigen dose sets a threshold for T-cell activation (23), so that high-avidity TCR-epitope interactions may still reach the activation threshold under conditions of limited epitope presentation. It was thus previously thought that deletion of vRAPs should lead to higher frequencies of primed CD8 T cells by recruiting more cells, in particular, those of lower avidities. As a rationale for this assumption, provided that direct priming by infected cells applies, the low epitope presentation in the presence of vRAPs after infection with mCMV-WT (Fig. 1) should suffice only for the priming of high-avidity CD8 T cells, whereas enhanced epitope presentation in cells infected with mCMV-{Delta}vRAP should allow for priming also of low-avidity CD8 T cells. Accordingly, deletion of vRAPs should alter the avidity distribution in a primed CD8 T-cell population. Specifically, it should shift the avidity distribution toward lower avidities.

For determining the avidity distributions experimentally, frequencies of CD8 T cells primed in the presence and absence of vRAPs were measured in ELISPOT assays performed with stimulator cells that were exogenously loaded with graded concentrations of synthetic IE1 peptide. In accordance with the immune evasion paradox, priming with mCMV-WT generated more IE1-specific CD8 T cells than priming with mCMV-{Delta}vRAP for all peptide concentrations tested (Fig. 9A). One has to consider that frequencies measured at any particular peptide concentration represent cumulative frequencies that include also the cells that are stimulated by all lower peptide concentrations, thus resulting in a "cumulative avidity distribution." To reveal the frequencies of cells of particular avidities, the experimentally determined cumulative avidity distributions were mathematically transformed into the avidity distributions resembling Gaussian distributions (Fig. 9B).


Figure 9
View larger version (11K):
[in this window]
[in a new window]

  		FIG. 9. Influence of vRAPs on the functional avidities of IE1 epitope-specific CD8 T cells. (A) Cumulative avidity distributions. Frequencies of IE1 epitope-specific CD8 T cells present in the spleen during the acute response were determined on day 7 after infection with 105 PFU of mCMV-WT.BAC (WT.BAC) and mCMV-{Delta}vRAP ({Delta}vRAP). CD8 T lymphocytes were isolated from a pool of 10 spleens and were used as responder cells in an ELISPOT assay performed with P815 stimulator cells exogenously loaded with synthetic IE1 peptide at the indicated, log10-graded concentrations. Frequencies (bars) and the corresponding 95% CIs (error bars) were calculated by intercept-free log-linear regression analysis as described in greater detail in the legend of Fig. 1. It is important to understand that the thus measured frequencies actually represent cumulative frequencies of cells responding to a particular peptide concentration and all lower concentrations. (B) Gaussian-like avidity distributions. To reveal the frequencies of CD8 T cells of particular avidities, the experimentally determined cumulative avidity distributions of panel A were mathematically transformed into avidity distributions by calculating the differences between each cumulative frequency bar and its right hand (lower concentration) neighboring bar. Gray bars, mCMV-WT.BAC; open bars, mCMV-{Delta}vRAP.

In essence, the avidity distributions were found to be almost congruent and thus apparently barely influenced by the presence or absence of vRAPs. This finding provides an argument against a major role for direct priming by infected cells.


arrow
DISCUSSION
 
Previous work by the group of A. B. Hill as well as our own data has revealed that deletion of vRAPs does not notably alter the immunodominance hierarchy of CD8 T-cell epitopes (16, 27, 45, 56). Initially, this finding was surprising since one expected to see an enhanced immune response against subdominant antigenic peptides when antigen presentation in infected cells was improved. Based on this observation, it was proposed that priming is accomplished primarily by APCs in which vRAPs are not operative. Since previous work using the complete panel of vRAP gene deletion mutants (68) has shown that vRAPs are operative also in DCs, at least in bone marrow-derived DCs (26), cross-presentation of viral antigens by uninfected DCs—as established for many other models (for reviews, see references 20, 21, and 67)—was a reasonable assumption also for the priming of CD8 T cells during mCMV infection (for a discussion, see references 27 and 28).

Here, we have attempted to define the mechanism behind these observations by relating CD8 T-cell priming to the infectivity that is reflected by the transcription of an epitope-encoding viral gene. Our study focused on the early events in the RLN where priming of naïve CD8 T cells takes place. This is important since the previous studies referred to the influence of vRAPs on epitope-specific CD8 T-cell frequencies during acute infection or during immunological memory in the spleen, the downstream central lymphoid organ, and were therefore possibly influenced by downstream variables that reshaped the immune response. At least, one could not simply anticipate that the situation in the spleen precisely mirrored the events in the RLN. This needed to be tested. Furthermore, previous studies, including our own studies, have not systematically investigated the influence of the infection dose and might thus have described a saturating, high-dose priming situation in which a putative difference between the presence and absence of vRAPs might have been leveled. Finally, the possibility of feedback regulation of CD8 T-cell priming by CD8 T cells' antiviral effector functions was not taken into consideration at all.

Our data show here that CD8 T-cell priming against mCMV is a fast and highly efficient process that generates epitope-specific effector cells within only 3 days and that is triggered by as few as 10 PFU. This is in accordance with a recent report on the early kinetics of the mCMV IE1 epitope-specific CD8 T-cell response in the spleen after low-dose infection by the intraperitoneal route (3). Strikingly, key features of mCMV infection and CD8 T-cell priming in the RLN closely resemble findings made with enhanced green fluorescent protein-tagged recombinant vaccinia and vesicular stomatitis viruses (24, 46). Specifically, despite the significant biological differences among the three viruses, which represent three different virus families, the distribution of infected cells was largely limited to the RLN periphery bordering the SCS. In all three examples, productive infection reached the RLN rapidly. Moreover, the decline in RLN infection occurred with half-lives of comparable orders of magnitude in both the vaccinia virus model (46) and the mCMV model (shown herein). This astounding similarity between such different viruses most likely indicates common principles in CD8 T-cell priming and control of the infection in the RLN.

In the case of mCMV infection, priming kinetics and the limiting virus dose required for infection were not essentially different in the presence or absence of vRAPs, but the overall frequencies of epitope-specific CD8 T cells were higher after priming in the presence of vRAPs, for the IE1 epitope in particular. This was the tendency also for most other epitopes tested, with the exception of the M105 epitope. Whether the fact that M105 is a virion protein is of any importance in this context remains to be investigated.

The advantage of mCMV-WT over mCMV-{Delta}vRAP in CD8 T-cell priming corresponded to an overall higher level and an extended period of viral gene expression, and thus of antigen supply, due to delayed control of the infection in the PLN. Whereas it was previously thought that vRAPs are not implicated at all in priming and are simply avoided by epitope cross-presentation, the most important new finding here is that vRAPs play a crucial role in that they indirectly enhance priming by protecting infected cells against an early CD8 T-cell-mediated antiviral effector function. Accordingly, in the absence of vRAPs, newly primed CD8 T cells exert a negative feedback function on the priming of further CD8 T cells and thus diminish the overall magnitude of the response. This conclusion was supported by the key experiment of this study demonstrating that CD8 T-cell depletion elevates the absolute level of intranodal viral gene expression selectively after infection with mCMV-{Delta}vRAP and, thus, abrogates the difference in intranodal viral gene expression levels between mCMV-WT and mCMV-{Delta}vRAP infection. Since priming in the RLN is usually accomplished by lymphoid-resident CD8{alpha}+ DCs (2, 6, 60), one might object that depletion with anti-CD8 antibodies has eliminated not just the CD8 T cells but also the CD8{alpha}+ DCs. This technical problem, however, is without consequence for the interpretation of the data because the final outcome is the prevention of CD8 T-cell priming. The finding that CD8 depletion did not reduce the overall level of intranodal viral gene expression after infection with mCMV-WT indicates that CD8{alpha}+ DCs are not the main infected cell type in the PLN.

Is there any role for an innate immune control of infection prior to the appearance of primed CD8 T cells? Our data have revealed that viral gene expression does not initially differ between the two viruses, suggesting that neither CD8 T cells nor NK cells significantly intervene in the first 24 h. So, there exists no evidence to propose an immediate innate defense. We have shown here that depletion of CD49+ NKG2D+ NK cells leads to an increase in intranodal viral gene expression beyond 24 h only after infection with mCMV-{Delta}vRAP, which thereby reduces the difference between the two viruses at least partially. This suggests a contribution of NK cell effector function. That NK cells selectively affect mCMV-{Delta}vRAP is compatible with the lack of downmodulation of the activating NKG2D ligand RAE-1 in the absence of vRAP m152/gp40 (34, 40).

One could also envisage, however, another role for the NK cells. As reported by Robbins and colleagues for Ly49H+ mice infected with mCMV-WT (59), NK cells activated through m157-Ly49H interaction (5, 10, 64) promote early CD8 T-cell responses to mCMV. Thus, depletion of NK cells could weaken the antiviral CD8 T-cell response. In Ly49H-negative mouse strains, such as BALB/c, this mechanism does not exist, but after infection with mCMV-{Delta}vRAP, NK cells are alternatively sensitized through interaction between RAE-1 and NKG2D. This can promote the antiviral CD8 T-cell response also in BALB/c mice and explains the enhancement of intranodal viral gene expression observed after anti-asialo-GM1 treatment and infection with mCMV-{Delta}vRAP but not mCMV-WT. In fact, only the interpretation of NK cell function as a helper function for the CD8 T cells can fully explain why NK cells had no effect after CD8 depletion (Fig. 5). Whichever mechanism applies, these findings suggest that NK cells and CD8 T cells work hand in hand in reducing CD8 T-cell priming by mCMV-{Delta}vRAP.

Although DCs are capable of sensitizing CD8 T cells for IFN-{gamma} secretion after infection with mCMV-{Delta}vRAP (26) and although mature DCs can still prime early after their infection (43), the interpretation that direct presentation played little role in our experiments is supported by the literature. Specifically, mCMV-infected DCs were reported to rather inhibit the expansion of epitope-specific CD8 T cells by selectively maintaining high-level expression of the negative costimulatory molecule PD-L1 while downregulating positive costimulatory molecules (7). Programming of DCs toward inducing T-cell anergy has frequently been discussed as a mechanism by which CMVs inhibit early adaptive antiviral responses (4, 7, 44, 53). Here, we have documented highly efficient priming in terms of both the low virus doses required for priming and the speed with which epitope-specific CD8 T cells are generated. This apparent discrepancy between our findings and the proposed inhibitory function of infected DCs is reasonably explained by a preponderance of cross-priming.

Collectively, our findings lead us to a model that proposes a role for vRAPs in preventing negative feedback regulation of CD8 T-cell priming (Fig. 10). After infection with mCMV-{Delta}vRAP (Fig. 10A), infected cells in the peri-SCS region of the RLN, which may include infected immigrant DCs and most likely also infected macrophages and stromal cells, produce and release antigens that are taken up by uninfected APCs/DCs for CD8 T-cell priming in the deeper interfollicular regions of the RLN. As shown recently by Hickman and colleagues (24), effector CD8 T cells relocate to the infected peripheral interfollicular region, a process that depends on the presentation of a cognate epitope. Since epitopes are presented in the absence of vRAPs, the infected cells can be recognized by the newly primed CD8 effector cells, which control the infection by means of their antiviral effector functions. In a negative feedback mode, resolution of infection by the newly primed CD8 T cells limits the amount of antigen available for cross-priming of other CD8 T cells. In contrast, after infection with mCMV-WT (Fig. 10B), vRAPs inhibit epitope presentation and thereby protect the infected cells against the attack by newly primed CD8 effector cells. Consequently, infection is not controlled, or is at least less controlled, and high levels of viral antigens continue to be delivered to uninfected DCs for ongoing cross-priming, thus leading to high frequencies of epitope-specific CD8 T cells.


Figure 10
View larger version (56K):
[in this window]
[in a new window]

  		FIG. 10. Explanation for the immune evasion paradox by the negative feedback model of CD8 T-cell priming. (A) Negative feedback regulation of CD8 T-cell priming in the absence of vRAPs after infection with mCMV-{Delta}vRAP ({Delta}vRAP). (B) Lack of negative feedback regulation of CD8 T-cell priming in the presence of vRAPs after infection with mCMV-WT.BAC (WT.BAC). At an early stage of the priming process in the RLN (infected cortical and uninfected paracortical regions symbolized by dark and light shading), cells infected by either virus deliver antigen(s) to uninfected APCs for cross-priming of virus-specific CD8 T cells (a1 and b1). At a later stage, cells presenting epitopes after infection with the mutant virus are attacked by newly primed cognate CD8 T cells, whereas cells infected with the WT virus are protected against the effector functions of the CD8 T cells by vRAPs inhibiting epitope presentation (a2 and b2). Consequently, viral gene expression and supply with antigens cease after infection with the mutant virus, whereas after infection with mCMV-WT.BAC, viral gene expression and antigen delivery continue for sustained priming of further CD8 T cells, thus leading to higher frequencies (a3 and b3).

In theory, the proposed mechanism of a negative feedback loop after infection with mCMV-{Delta}vRAP predicts that in the following round, the reduced frequency of CD8 T cells is associated with a recovery of intranodal viral gene expression and, thus, a refreshed antigen supply for cross-priming. As a consequence, the frequency of CD8 T cells should rise again and so forth. That such an oscillation of the feedback system was not resolvable by our experiments may be due to inertia of the system caused by the long half-life of CD8 effector cells and possibly also by the consumption and consequent shortage of infectible cells.

Interest in vRAPs of CMVs was nourished by the hope that improvement of antigen presentation by deletion of vRAP genes might yield a superior human CMV vaccine virus for the prevention of CMV disease. Our results in the mouse model dampen this optimism.


arrow
ACKNOWLEDGMENTS
 
We thank Ulrich H. Koszinowski and Markus Wagner, Max von Pettenkofer Institute, Munich, Germany, for the vRAP gene deletion mutant and the Central Laboratory Animal Facility team of the Johannes Gutenberg University, Mainz, Germany, for animal care. We are indebted to Stipan Jonjic, Rijeka, Croatia, for supplying us with antibodies.

Extramural support was provided by the Deutsche Forschungsgemeinschaft, SFB 490, individual projects E2 (C.O.S., C.K.S., D.G., and M.J.R.), E3 (P.D., D.G.-M., and R.H.), and E4 (V.B. and M.J.R.) as well as SFB 432, individual project A10 (J.P.) and Clinical Research Group KFO 183 (N.A.W.L. and M.J.R.). Special thanks go to the Gerhard and Martha Röttger Stiftung and to the Hans-Joachim and Ilse Brede Stiftung for generous donations.


arrow
FOOTNOTES
 
* Corresponding author. Mailing address: Institute for Virology, Johannes Gutenberg University, Hochhaus am Augustusplatz, 55101 Mainz, Germany. Phone: 49 6131 39 33650. Fax: 49 6131 39 35604. E-mail: Matthias.Reddehase{at}uni-mainz.de Back

{triangledown} Published ahead of print on 24 September 2008. Back


arrow
REFERENCES
 
    1
  1. Alcami, A., and U. H. Koszinowski. 2000. Viral mechanisms of immune evasion. Immunol. Today 9:447-455.
    2. 2
  2. Allan, R. S., C. M. Smith, G. T. Belz, A. L. van Lint, L. M. Wakim, W. R. Heath, and F. R. Carbone. 2003. Epidermal viral immunity induced by CD8{alpha}+ dendritic cells but not by Langerhans cells. Science 301:1925-1928.[Abstract/Free Full Text]
    3. 3
  3. Andrews, D. M., C. E. Andoniou, P. Fleming, M. J. Smyth, and M. A. Degli-Esposti. 2008. The early kinetics of cytomegalovirus-specific CD8+ T-cell responses are not affected by antigen load or the absence of perforin or gamma interferon. J. Virol. 82:4931-4937.[Abstract/Free Full Text]
    4. 4
  4. Andrews, D. M., C. E. Andoniou, F. Granucci, P. Ricciardi-Castagnoli, and M. A. Degli-Esposti. 2001. Infection of dendritic cells by murine cytomegalovirus induces functional paralysis. Nat. Immunol. 2:1077-1084.[CrossRef][Medline]
    5. 5
  5. Arase, H., E. S. Mocarski, A. E. Campbell, A. B. Hill, and L. L. Lanier. 2002. Direct recognition of cytomegalovirus by activating and inhibitory NK cell receptors. Science 296:1323-1326.[Abstract/Free Full Text]
    6. 6
  6. Belz, G. T., C. M. Smith, D. Eichner, K. Shortman, G. Karupiah, F. R. Carbone, and W. R. Heath. 2004. Cutting edge: conventional CD8 alpha+ dendritic cells are generally involved in priming CTL immunity to viruses. J. Immunol. 172:1996-2000.[Abstract/Free Full Text]
    7. 7
  7. Benedict, C. A., A. Loewendorf, Z. Garcia, B. R. Blazar, and E. M. Janssen. 2008. Dendritic cell programming by cytomegalovirus stunts naive T cell responses via the PD-L1/PD-1 pathway. J. Immunol. 180:4836-4847.[Abstract/Free Full Text]
    8. 8
  8. Böhm, V., J. Podlech, D. Thomas, P. Deegen, M.-F. Pahl-Seibert, N. A. W. Lemmermann, N. K. A. Grzimek, S. A. Oehrlein-Karpi, M. J. Reddehase, and R. Holtappels. 2008. Epitope-specific in vivo protection against cytomegalovirus disease by CD8 T cells in the murine model of preemptive immunotherapy. Med. Microbiol. Immunol. 197:135-144.[CrossRef][Medline]
    9. 9
  9. Bousso, P., and E. Robey. 2003. Dynamics of CD8+ T cell priming by dendritic cells in intact lymph nodes. Nat. Immunol. 4:579-585.[CrossRef][Medline]
    10. 10
  10. Bubic, I., M. Wagner, A. Krmpotic, T. Saulig, S. Kim, W. M. Yokoyama, S. Jonjic, and U. H. Koszinowski. 2004. Gain of virulence caused by loss of a gene in murine cytomegalovirus. J. Virol. 78:7536-7544.[Abstract/Free Full Text]
    11. 11
  11. Campbell, A. E., V. J. Cavanaugh, and J. S. Slater. 2008. The salivary glands as a privileged site of cytomegalovirus immune evasion and persistence. Med. Microbiol. Immunol. 197:205-213.[CrossRef][Medline]
    12. 12
  12. Cobbold, S. P., A. Jayasuriya, A. Nash, T. D. Prospero, and H. Waldmann. 1984. Therapy with monoclonal antibodies by elimination of T-cell subsets in vivo. Nature 312:548-551.[CrossRef][Medline]
    13. 13
  13. Dalod, M., T. Hamilton, R. Salomon, T. P. Salazar-Mather, S. C. Henry, J. D. Hamilton, and C. A. Biron. 2003. Dendritic cell responses to early murine cytomegalovirus infection: subset functional specialization and differential regulation by interferon alpha/beta. J. Exp. Med. 197:885-898.[Abstract/Free Full Text]
    14. 14
  14. del Val, M., H. Hengel, H. Haecker, U. Hartlaub, T. Ruppert, P. Lucin, and U. H. Koszinowski. 1992. Cytomegalovirus prevents antigen presentation by blocking the transport of peptide-loaded major histocompatibility complex class I molecules into the medial-Golgi compartment. J. Exp. Med. 176:729-738.[Abstract/Free Full Text]
    15. 15
  15. Doom, C. M., and A. B. Hill. 2008. MHC class I immune evasion in MCMV infection. Med. Microbiol. Immunol. 197:191-2004.[CrossRef][Medline]
    16. 16
  16. Gold, M. C., M. W. Munks, M. Wagner, U. H. Koszinowski, A. B. Hill, and S. P. Fling. 2002. The murine cytomegalovirus immunomodulatory gene m152 prevents recognition of infected cells by M45-specific CTL, but does not alter the immunodominance of the M45-specific CD8 T cell response in vivo. J. Immunol. 169:359-365.[Abstract/Free Full Text]
    17. 17
  17. Gold, M. C., M. W. Munks, M. Wagner, C. W. McMahon, A. Kelly, D. G. Kavanagh, M. K. Slifka, U. H. Koszinowski, D. H. Raulet, and A. B. Hill. 2004. Murine cytomegalovirus interference with antigen presentation has little effect on the size or the effector memory phenotype of the CD8 T cell response. J. Immunol. 172:6944-6953.[Abstract/Free Full Text]
    18. 18
  18. Grzimek, N. K. A., J. Podlech, H.-P. Steffens, R. Holtappels, S. Schmalz, and M. J. Reddehase. 1999. In vivo replication of recombinant murine cytomegalovirus driven by the paralogous major immediate-early promoter-enhancer of human cytomegalovirus. J. Virol. 73:5043-5055.[Abstract/Free Full Text]
    19. 19
  19. Hanson, L. K., and A. E. Campbell. 2006. Determinants of macrophage tropism, p. 419-443. In M. J. Reddehase (ed.), Cytomegaloviruses: molecular biology and immunology. Caister Academic Press, Wymondham, Norfolk, United Kingdom.
    20. 20
  20. Heath, W. R., and F. R. Carbone. 2001. Cross-presentation in viral immunity and self-tolerance. Nat. Rev. Immunol. 1:126-134.[CrossRef][Medline]
    21. 21
  21. Heath, W. R., and F. R. Carbone. 2001. Cross-presentation, dendritic cells, tolerance and immunity. Annu. Rev. Immunol. 19:47-64.[CrossRef][Medline]
    22. 22
  22. Hengel, H., W. Brune, and U. H. Koszinowski. 1998. Immune evasion by cytomegalovirus—survival strategies of a highly adapted opportunist. Trends Microbiol. 6:190-197.[CrossRef][Medline]
    23. 23
  23. Henrickson, S. E., T. R. Mempel, I. B. Mazo, B. Liu, M. N. Artymov, H. Zheng, A. Peixoto, M. P. Flynn, B. Senman, T. Junt, H. C. Wong, A. K. Chakraborty, and U. H. von Andrian. 2008. T cell sensing of antigen dose governs interactive behavior with dendritic cells and sets a threshold for T cell activation. Nat. Immunol. 9:282-291.[CrossRef][Medline]
    24. 24
  24. Hickman, H. D., K. Takeda, C. N. Skon, F. R. Murray, S. E. Hensley, J. Loomis, G. N. Barber, J. R. Bennink, and J. W. Yewdell. 2008. Direct priming of antiviral CD8+ T cells in the peripheral interfollicular region of lymph nodes. Nat. Immunol. 9:155-165.[CrossRef][Medline]
    25. 25
  25. Holtappels, R., V. Böhm, J. Podlech, and M. J. Reddehase. 2008. CD8 T-cell-based immunotherapy of cytomegalovirus infection: "proof of concept" provided by the murine model. Med. Microbiol. Immunol. 197:125-134.[CrossRef][Medline]
    26. 26
  26. Holtappels, R., D. Gillert-Marien, D. Thomas, J. Podlech, P. Deegen, S. Herter, S. A. Oehrlein-Karpi, D. Strand, M. Wagner, and M. J. Reddehase. 2006. Cytomegalovirus encodes a positive regulator of antigen presentation. J. Virol. 80:7613-7624.[Abstract/Free Full Text]
    27. 27
  27. Holtappels, R., M. W. Munks, J. Podlech, and M. J. Reddehase. 2006. CD8 T-cell-based immunotherapy of cytomegalovirus disease in the mouse model of the immunocompromised bone marrow transplantation recipient, p. 383-418. In M. J. Reddehase (ed.), Cytomegaloviruses: molecular biology and immunology. Caister Academic Press, Wymondham, Norfolk, United Kingdom.
    28. 28
  28. Holtappels, R., J. Podlech, M.-F. Pahl-Seibert, M. Juelch, D. Thomas, C. O. Simon, M. Wagner, and M. J. Reddehase. 2004. Cytomegalovirus misleads its host by priming of CD8 T cells specific for an epitope not presented in infected tissues. J. Exp. Med. 199:131-136.[Abstract/Free Full Text]
    29. 29
  29. Holtappels, R., C. O. Simon, M. W. Munks, D. Thomas, P. Deegen, B. Kühnapfel, T. Däubner, S. F. Emde, J. Podlech, N. K. Grzimek, S. A. Oehrlein-Karpi, A. B. Hill, and M. J. Reddehase. 2008. Subdominant CD8 T-cell epitopes account for protection against cytomegalovirus independent of immunodomination. J. Virol. 82:5781-5796.[Abstract/Free Full Text]
    30. 30
  30. Holtappels, R., D. Thomas, J. Podlech, and M. J. Reddehase. 2002. Two antigenic peptides from genes m123 and m164 of murine cytomegalovirus quantitatively dominate CD8 T-cell memory in the H-2d haplotype. J. Virol. 76:151-164.[Abstract/Free Full Text]
    31. 31
  31. Kavanagh, D. G., M. C. Gold, M. Wagner, U. H. Koszinowski, and A. B. Hill. 2001. The multiple immune-evasion genes of murine cytomegalovirus are not redundant: m4 and m152 inhibit antigen presentation in a complementary and cooperative fashion. J. Exp. Med. 194:967-977.[Abstract/Free Full Text]
    32. 32
  32. Kavanagh, D. G., U. H. Koszinowski, and A. B. Hill. 2001. The murine cytomegalovirus immune evasion protein m4/gp34 forms biochemically distinct complexes with class I MHC at the cell surface and in a pre-Golgi compartment. J. Immunol. 167:3894-3902.[Abstract/Free Full Text]
    33. 33
  33. Kleijnen, M. F., J. B. Huppa, P. Lucin, S. Mukherjee, H. Farrell, A. E. Campbell, U. H. Koszinowski, A. B. Hill, and H. Ploegh. 1997. A mouse cytomegalovirus glycoprotein, gp34, forms a complex with folded class I MHC molecules in the ER which is not retained but is transported to the cell surface. EMBO J. 16:685-694.[CrossRef][Medline]
    34. 34
  34. Krmpotic, A., D. H. Busch, I. Bubic, F. Gebhardt, H. Hengel, M. Hasan, A. A. Scalzo, U. H. Koszinowski, and S. Jonjic. 2002. MCMV glycoprotein gp40 confers virus resistance to CD8+ T lymphocytes and NK cells in vivo. Nat. Immunol. 3:529-535.[CrossRef][Medline]
    35. 35
  35. Krmpotic, A., M. Messerle, I. Crnkovic-Mertens, B. Polic, S. Jonjic, and U. H. Koszinowski. 1999. The immunoevasive function encoded by the mouse cytomegalovirus gene m152 protects the virus against T cell control in vivo. J. Exp. Med. 190:1285-1295.[Abstract/Free Full Text]
    36. 36
  36. Kurz, S. K., M. Rapp, H.-P. Steffens, N. K. A. Grzimek, S. Schmalz, and M. J. Reddehase. 1999. Focal transcriptional activity of murine cytomegalovirus during latency in the lungs. J. Virol. 73:482-494.[Abstract/Free Full Text]
    37. 37
  37. Kurz, S. K., H.-P. Steffens, A. Mayer, J. R. Harris, and M. J. Reddehase. 1997. Latency versus persistence or intermittent recurrences: evidence for a latent state of murine cytomegalovirus in the lungs. J. Virol. 71:2980-2987.[Abstract]
    38. 38
  38. Lenac, T., J. Arapovic, L. Traven, A. Krmpotic, and S. Jonjic. 2008. Murine cytomegalovirus regulation of NKG2D ligands. Med. Microbiol. Immunol. 197:159-166.[CrossRef][Medline]
    39. 39
  39. Lilley, B. N., and H. L. Ploegh. 2005. Viral modulation of antigen presentation: manipulation of cellular targets in the ER and beyond. Immunol. Rev. 207:126-144.[CrossRef][Medline]
    40. 40
  40. Lodoen, M., K. Ogasawara, J. A. Hamerman, H. Arase, J. P. Houchins, E. S. Mocarski, and L. L. Lanier. 2003. NKG2D-mediated natural killer cell protection against cytomegalovirus is impaired by viral gp40 modulation of retinoic acid early inducible 1 gene molecules. J. Exp. Med. 197:1245-1253.[Abstract/Free Full Text]
    41. 41
  41. Loureiro, J., and H. L. Ploegh. 2006. Antigen presentation and the ubiquitin-proteasome system in host-pathogen interactions. Adv. Immunol. 92:225-305.[CrossRef][Medline]
    42. 42
  42. Lu, X., A. K. Pinto, A. M. Kelly, K. S. Cho, and A. B. Hill. 2006. Murine cytomegalovirus interference with antigen presentation contributes to the inability of CD8 T cells to control virus in the salivary gland. J. Virol. 80:4200-4202.[Abstract/Free Full Text]
    43. 43
  43. Mathys, S., T. Schroeder, J. Ellwart, U. H. Koszinowski, M. Messerle, and U. Just. 2003. Dendritic cells under influence of mouse cytomegalovirus have a physiologic dual role: to initiate and to restrict T cell activation. J. Infect. Dis. 187:988-999.[CrossRef][Medline]
    44. 44
  44. Moutaftsi, M., A. M. Mehl, L. K. Borysiewicz, and Z. Tabi. 2002. Human cytomegalovirus inhibits maturation and impairs function of monocyte-derived dendritic cells. Blood 99:2913-2921.[Abstract/Free Full Text]
    45. 45
  45. Munks, M. W., A. K. Pinto, C. M. Doom, and A. B. Hill. 2007. Viral interference with antigen presentation does not alter acute or chronic CD8 T cell immunodominance in murine cytomegalovirus infection. J. Immunol. 178:7235-7241.[Abstract/Free Full Text]
    46. 46
  46. Norbury, C. C., D. Malide, J. S. Gibbs, J. R. Bennink, and J. W. Yewdell. 2002. Visualizing priming of virus-specific CD8+ T cells by infected dendritic cells in vivo. Nat. Immunol. 3:265-271.[CrossRef][Medline]
    47. 47
  47. Orange, J. S., and C. A. Biron. 1996. Characterization of early IL-12, IFN-{alpha}/β and TNF effects on antiviral state and NK cell responses during murine cytomegalovirus infection. J. Immunol. 156:4746-4756.[Abstract]
    48. 48
  48. Pahl-Seibert, M.-F., M. Juelch, J. Podlech, D. Thomas, P. Deegen, M. J. Reddehase, and R. Holtappels. 2005. Highly protective in vivo function of cytomegalovirus IE1 epitope-specific memory CD8 T cells purified by T-cell receptor-based cell sorting. J. Virol. 79:5400-5413.[Abstract/Free Full Text]
    49. 49
  49. Pinto, A. K., and A. B. Hill. 2005. Viral interference with antigen presentation to CD8 T cells: lessons from cytomegalovirus. Viral Immunol. 18:434-444.[CrossRef][Medline]
    50. 50
  50. Pinto, A. K., M. W. Munks, U. H. Koszinowski, and A. B. Hill. 2006. Coordinated function of murine cytomegalovirus genes completely inhibits CTL lysis. J. Immunol. 177:3225-3234.[Abstract/Free Full Text]
    51. 51
  51. Ploegh, H. 1998. Viral strategies of immune evasion. Science 280:248-253.[Abstract/Free Full Text]
    52. 52
  52. Podlech, J., R. Holtappels, N. K. A. Grzimek, and M. J. Reddehase. 2002. Animal models: murine cytomegalovirus, p. 493-525. In S. H. E. Kaufmann and D. Kabelitz (ed.), Methods in microbiology, 2nd ed., vol. 32. Academic Press, San Diego, CA.[CrossRef]
    53. 53
  53. Raftery, M. J., M. Schwab, S. M. Eibert, Y. Samstag, H. Walczak, and G. Schönrich. 2001. Targeting the function of mature dendritic cells by human cytomegalovirus: a multilayered viral defense strategy. Immunity 15:997-1009.[CrossRef][Medline]
    54. 54
  54. Reddehase, M. J. 2002. Antigens and immunoevasins: opponents in cytomegalovirus immune surveillance. Nat. Rev. Immunol. 2:831-844.[CrossRef][Medline]
    55. 55
  55. Reddehase, M. J., J. B. Rothbard, and U. H. Koszinowski. 1989. A pentapeptide as minimal antigenic determinant for MHC class I-restricted T lymphocytes. Nature 337:651-653.[CrossRef][Medline]
    56. 56
  56. Reddehase, M. J., C. O. Simon, J. Podlech, and R. Holtappels. 2004. Stalemating a clever opportunist: lessons from murine cytomegalovirus. Hum. Immunol. 65:446-455.[CrossRef][Medline]
    57. 57
  57. Reusch, U., O. Bernhard, U. H. Koszinowski, and P. Schu. 2002. AP-1A and AP-3A lysosomal sorting functions. Traffic 3:752-761.[CrossRef][Medline]
    58. 58
  58. Reusch, U., W. Muranyi, P. Lucin, H. G. Burgert, H. Hengel, and U. H. Koszinowski. 1999. A cytomegalovirus glycoprotein re-routes MHC class I complexes to lysosomes for degradation. EMBO J. 18:1081-1091.[CrossRef][Medline]
    59. 59
  59. Robbins, S. H., G. Bessou, A. Cornillon, N. Zucchini, B. Rupp, Z. Ruzsics, T. Sacher, E. Tomasello, E. Vivier, U. H. Koszinowski, and Marc Dalod. 2007. Natural killer cells promote early CD8 T cell responses against cytomegalovirus. PLoS Pathogens 3:1152-1164.
    60. 60
  60. Schnorrer, P., G. M. Behrens, N. S. Wilson, J. L. Pooley, C. M. Smith, D. EI-Sukkari, G. Davey, F. Kupresanin, M. Li, E. Maraskovsky, G. T. Belz, F. R. Carbone, K. Shortman, W. R. Heath, and J. A. Villadangos. 2006. The dominant role of CD8+ dendritic cells in cross-presentation is not dictated by antigen capture. Proc. Natl. Acad. Sci. USA 103:10729-10734.[Abstract/Free Full Text]
    61. 61
  61. Simon, C. O., R. Holtappels, H.-M. Tervo, V. Böhm, T. Däubner, S. A. Oehrlein-Karpi, B. Kühnapfel, A. Renzaho, D. Strand, J. Podlech, M. J. Reddehase, and N. K. A. Grzimek. 2006. CD8 T cells control cytomegalovirus latency by epitope-specific sensing of transcriptional reactivation. J. Virol. 80:10436-10456.[Abstract/Free Full Text]
    62. 62
  62. Simon, C. O., B. Kühnapfel, M. J. Reddehase, and N. K. Grzimek. 2007. Murine cytomegalovirus major immediate-early enhancer region operating as a genetic switch in bidirectional gene pair transcription. J. Virol. 81:7805-7810.[Abstract/Free Full Text]
    63. 63
  63. Simon, C. O., C. K. Seckert, D. Dreis, M. J. Reddehase, and N. K. Grzimek. 2005. Role for tumor necrosis factor alpha in murine cytomegalovirus transcriptional reactivation in latently infected lungs. J. Virol. 79:326-340.[Abstract/Free Full Text]
    64. 64
  64. Smith, H. R., J. W. Heusel, I. K. Metha, S. Kim, B. G. Dorner, O. V. Naidenko, K. Iizuka, H. Furukawa, D. L. Beckman, J. T. Pingel., A. A. Scalzo, D. H. Fremont, and W. M. Yokoyama. 2002. Recognition of a virus-encoded ligand by a natural killer cell activation receptor. Proc. Natl. Acad. Sci. USA 99:8826-8831.[Abstract/Free Full Text]
    65. 65
  65. Streblow, D. N., S. M. Varnum, R. D. Smith, and J. A. Nelson. 2006. A proteomics analysis of human cytomegalovirus particles, p. 91-110. In M. J. Reddehase (ed.), Cytomegaloviruses: molecular biology and immunology. Caister Academic Press, Wymondham, Norfolk, United Kingdom.
    66. 66
  66. Tortorella, D., B. E. Gewurz, M. H. Furman, D. J. Schust, and H. L. Ploegh. 2000. Viral subversion of the immune system. Annu. Rev. Immunol. 18:861-926.[CrossRef][Medline]
    67. 67
  67. Villadangos, J. A., W. R. Heath, and F. R. Carbone. 2007. Outside looking in: the inner workings of the cross-presentation pathways within dendritic cells. Trends Immunol. 28:45-47.[CrossRef][Medline]
    68. 68
  68. Wagner, M., A. Gutermann, J. Podlech, M. J. Reddehase, and U. H. Koszinowski. 2002. MHC class I allele-specific cooperative and competitive interactions between immune evasion proteins of cytomegalovirus. J. Exp. Med. 196:805-816.[Abstract/Free Full Text]
    69. 69
  69. Wagner, M., S. Jonjic, U. H. Koszinowski, and M. Messerle. 1999. Systematic excision of vector sequences from the BAC-cloned herpesvirus genome during virus reconstitution. J. Virol. 73:7056-7060.[Abstract/Free Full Text]
    70. 70
  70. Wilhelmi, V., C. O. Simon, J. Podlech, V. Böhm, T. Däubner, S. Emde, D. Strand, A. Renzaho, N. A. W. Lemmermann, C. K. Seckert, M. J. Reddehase, and N. K. A. Grzimek. 2008. Transactivation of cellular genes involved in nucleotide metabolism by the regulatory IE1 protein of murine cytomegalovirus is not critical for viral replicative fitness in quiescent cells and host tissues. J. Virol. 82:9900-9916.[Abstract/Free Full Text]
    71. 71
  71. Yewdell, J. W., and A. B. Hill. 2002. Viral interference with antigen presentation. Nat. Immunol. 3:1019-1025.[CrossRef][Medline]
    72. 72
  72. Ziegler, H., R. Thäle, P. Lucin, W. Muranyi, T. Flohr, H. Hengel, H. Farrell, W. Rawlinson, and U. H. Koszinowski. 1997. A mouse cytomegalovirus glycoprotein retains MHC class I complexes in the ERGIC/cis-Golgi compartments. Immunity 6:57-66.[CrossRef][Medline]

Journal of Virology, December 2008, p. 11637-11650, Vol. 82, No. 23
0022-538X/08/$08.00+0     doi:10.1128/JVI.01510-08
Copyright © 2008, American Society for Microbiology. All Rights Reserved.



This article has been cited by other articles:

  • Kielczewska, A., Pyzik, M., Sun, T., Krmpotic, A., Lodoen, M. B., Munks, M. W., Babic, M., Hill, A. B., Koszinowski, U. H., Jonjic, S., Lanier, L. L., Vidal, S. M. (2009). Ly49P recognition of cytomegalovirus-infected cells expressing H2-Dk and CMV-encoded m04 correlates with the NK cell antiviral response. JEM 206: 515-523 [Abstract] [Full Text]  

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
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 arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Böhm, V.
Right arrow Articles by Reddehase, M. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Böhm, V.
Right arrow Articles by Reddehase, M. J.

Copyright © 2009 by the American Society for Microbiology.   For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»