The action of TdT on mouse TCR genes accounts for ∼90% of T cell repertoire diversity. We report that in TdT−/− mice, total TCD8+ responses to influenza and vaccinia viruses are reduced by ∼30% relative to wild-type mice. We find that TCD8+ responses to three subdominant influenza virus determinants are reduced to background values in TdT−/− mice while responses to three immunodominant determinants undergo a major reshuffling. A similar reshuffling occurs in TCD8+ responses to immunodominant vaccinia virus determinants, and is clearly based on broad differences in TCR family usage and CDR3 length between wild-type and TdT−/− mice. These findings demonstrate that TdT plays a critical role in the magnitude and breadth of anti-viral TCD8+ responses toward individual determinants and suggests that germline TCR repertoire bias toward the most dominant determinants is a major factor in establishing immunodominance hierarchies.

Infection of vertebrate cells with viruses with the largest genomes generates millions of distinct viral peptides that represent potential targets for CD8+ T lymphocyte (TCD8+)3 recognition. TCD8+ responses, however, are limited to a minute fraction of viral peptidomes. Even among the chosen few peptide determinants that elicit measurable TCD8+ responses, the magnitude of TCD8+ responses varies widely, creating a hierarchical pattern that is highly consistent between individuals sharing a given set of MHC class I alleles (1). Accordingly, immunodominant determinants provoke the most robust TCD8+ responses, while subdominant determinants elicit lower numbers of TCD8+.

Immunodominance is a consistent feature of TCD8+ and TCD4+ responses to viruses and other intracellular pathogens, tumor cells, and transplanted tissues. In model mouse-virus systems, numerous factors are known to participate in shaping TCD8+ immunodominance hierarchies including the following: 1) abundance and kinetics of viral gene product expression, 2) proteolytic generation of peptides by proteasomes and other proteases, 3) specificity of TAP for transporting cytosolic peptides into the endoplasmic reticulum, 4) affinity of peptide binding to class I allomorphs, 5) ability of the TCD8+ repertoire to respond to a given peptide-class I complex based on precursor frequency and proliferative capacity, 6) immunodomination, i.e., suppression of subdominant-determinant TCD8+ by immunodominant-determinant TCD8+ at the level of individual APCs, and 7) suppression of immunodominant-determinant TCD8+ by naturally occurring CD4+CD25+ regulatory T cells (2).

In this study, we have examined the shaping of the TCR repertoire, one of the most poorly defined factors in establishing immunodominance hierarchies. The TCR repertoire is thought to encompass ∼107 and 108 distinct TCRs, respectively, in mice and humans. The diversity of TCRs is predominantly exhibited in their CDRs that establish contact with cognate peptide-MHC complexes. CDR3 regions of Abs and TCRs are unequivocally the most diverse structures known in biology.

TCR structural diversification is achieved by several mechanisms during the rearrangement of receptor genes and assembly of receptor chains in the thymus (reviewed in Refs. 3, 4). Stochastic recombination of noncontiguous germline DNA segments known as variable (V), diversity (D), and joining (J) segments is initiated by RAG-1 and RAG-2 proteins, giving rise to a multitude of V(D)J fusions. TCR diversity is further enhanced by variable removal of nucleotides from the exposed V, D, and J termini, and by addition of template-dependent or palindromic (P) as well as non-template-encoded (N) nucleotides at the V-D, D-J and V-J junctions. Finally, random pairing of TCR-α and -β chains followed by selection for functional TCRs introduces another layer of diversity to the TCR repertoire.

N-region diversity results from the transient expression during V(D)J recombination of TdT. Mice with a targeted disruption of TdT (TdT−/− mice) demonstrate the nearly complete absence of N nucleotides in their B and T cells (5, 6). TdT belongs to the pol X family of polymerases, a subgroup of an ancient nucleotidyltransferase superfamily defined by homologies within their nucleotide binding domains and active site motifs. TdT is unique among DNA polymerases in adding nucleotides to free 3′-OH ends of fragmented or nicked DNA in a template-independent fashion (7). TdT is expressed in human and mouse thymocytes at 20 wk postgestation and 3–5 days after birth, respectively (8, 9). In both cases, TdT expression temporally correlates with the onset of N-region diversity in immature thymocytes undergoing differentiation. Strikingly, TdT has been estimated to be responsible for at least 90% of diversity in αβ TCRs (10).

Mice with a targeted deletion in TdT were generated by Mathis, Benoist, and colleagues >15 years ago (6) and to date have demonstrated little in the way of disease susceptibility or alterations in their TCD4+ responses to soluble Ags. Gavin and Bevan (11) reported that TCD8+ clones from TdT−/− mice were more generally cross-reactive between peptides present in a large synthetic library. But what is the effect of the knockout on the magnitude and breadth of TCD8+ responses to individual viral determinants? In this study, we provide the initial description of the critical role that TdT plays in establishing the immunodominance hierarchy in TCD8+ anti-viral responses.

Adult sex- and age-matched mice were used in all experiments. TdT−/− mice on a BALB/c background (H-2d) and BALB/c mice expressing the Thy-1.1 congenic marker were provided by John Kearney (University of Alabama, Birmingham, AL) and Ethan Shevach (National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, MD), respectively. These mice were bred and housed together with wild-type (WT) BALB/c mice purchased from Taconic Farms in our animal care facility under specific, pathogen-free conditions.

Influenza A virus (IAVs) used in this study, namely IAV (PR8, A/Puerto Rico/8/34) IAV-SEQ12, and J-1 were grown in 10-day-old embryonated chicken eggs and used as infectious allantoic fluid. Mice received a single i.p. dose of IAV approximating 600 hemagglutinating units. In memory TCD8+ responses, mice were primed with IAV and boosted 1 mo later with IAV-SEQ12. In several experiments, we infected mice i.p. with 106 PFUs of recombinant vaccinia virus (rVV)-ES-HA518.

All peptides used in this study (listed in Table I) were procured or synthesized, purified by HPLC, and analyzed by mass spectrometry by or under the supervision of the Biologic Resource Branch, National Institute of Allergy and Infectious Diseases. In each case, substances with the predicted mass constituted >95% of the material analyzed. Stock solutions of peptides were prepared at 1 mM in DMSO and stored at −30°C.

Table I.

Synthetic peptides used in this studya

VirusDeterminantDesignationSequenceRestricting MHCReference
IAV NP147–155 NP147 TYQRTRALV Kd Sherman et al., 1992 
IAV PB2289–297 PB2289 IGGIRMVDI Dd Chen et al., 2002 
IAV HA518–526 HA518 IYSTVASSL Kd Sweetser et al., 1989 
IAV NP39–47 NP39 FYIQMCTEL Kd Deng et al., 1997 
IAV NP218–226 NP218 TAYERMCNIL Kd Deng et al., 1997 
IAV HA462–470 HA462 LYEKVKSQL Kd Deng et al., 1997 
VV F2L23–31 F2L23 SPYAAGYDL Ld Tscharke et al., 2006 
VV A52R75–83 A52R75 KYGRLFNEI Kd Tscharke et al., 2006 
VV E3L140–148 E3L140 VGPSNSPTF Dd Tscharke et al., 2006 
LCMV NP118–126 NP118 RPQASGVYM Ld Whitton et al., 1989 
VirusDeterminantDesignationSequenceRestricting MHCReference
IAV NP147–155 NP147 TYQRTRALV Kd Sherman et al., 1992 
IAV PB2289–297 PB2289 IGGIRMVDI Dd Chen et al., 2002 
IAV HA518–526 HA518 IYSTVASSL Kd Sweetser et al., 1989 
IAV NP39–47 NP39 FYIQMCTEL Kd Deng et al., 1997 
IAV NP218–226 NP218 TAYERMCNIL Kd Deng et al., 1997 
IAV HA462–470 HA462 LYEKVKSQL Kd Deng et al., 1997 
VV F2L23–31 F2L23 SPYAAGYDL Ld Tscharke et al., 2006 
VV A52R75–83 A52R75 KYGRLFNEI Kd Tscharke et al., 2006 
VV E3L140–148 E3L140 VGPSNSPTF Dd Tscharke et al., 2006 
LCMV NP118–126 NP118 RPQASGVYM Ld Whitton et al., 1989 
a

Predicted TCR contact residues for Kd-restricted determinants (Mitaksov and Fremont, 2006) are shown in small caps.

Anti-CD16/CD32 (clone 2.4G2, rat IgG2b, Fc Block), CyChrome-, or Alexa Fluor 647-conjugated anti-mouse CD8α (clone 53–6.7, rat IgG2a), PE-conjugated anti-mouse Thy-1.2 (clone 53–2.1, rat IgG2a), and FITC-conjugated anti-mouse IFN-γ (clone XMG1.2, rat IgG1) mAbs were all from BD Biosciences.

Erythrocyte-depleted splenocytes were prepared and peritoneal exudate cells (PECs) were collected via peritoneal lavage using sterile PBS. ICS was performed as described (12).

CTL induction was assessed by conventional 51Cr release assay. Erythrocyte-depleted splenocytes were prepared on day 7 postinfection and used as effector cells at indicated ratios against 51Cr-labeled P815 target cells that were presensitized with 100 nM IAV-derived peptides. Target cells were seeded at 104 cells/well together with effector splenocytes in 96-well U-bottom plates. The plates were spun at 400 × g for 5 min at the end of a 6-h incubation period at 37°C. A 100 μl aliquot of supernatant was then harvested from each well, and the 51Cr content of the samples was determined by gamma counting. Specific lysis of the target cells was determined using following formula: percent specific lysis = [(ER-SR)/(TR-SR)] × 100, where ER (experimental release) is obtained from wells containing both effector and target cells, whereas SR (spontaneous release) and TR (total release) are determined from wells receiving only target cells plus medium or target cells plus a 1/7 dilution of 3.5% (w/v) cetrimide, respectively. In vivo killing assay was performed as described (13) with minor modifications. Syngeneic erythrocyte-depleted splenocytes were split into three populations and were pulsed with either lymphocytic choriomeningitis virus (LCMV)-NP118 at 1 μM and stained with 0.025 μM CFSE (CFSElow) while CFSEint (0.2 μM CFSE) and CFSEhigh (1.6 μM CFSE) populations were preincubated with HA518 (1 μM) and NP147 (1 μM), respectively. After extensive washing, equal numbers of cells from each population were mixed, and each mouse (naive or previously infected with IAV) received a total of 3 × 107 target cells in 500 μl of PBS i.v. Two, 4, or 6 h later, spleens were harvested and analyzed by flow cytometry. Up to 1 × 104 CFSEhigh events were collected for each mouse for analysis and the percent specific killing was calculated as follows: 100− [(% cognate peptide-pulsed in primed mouse/% control peptide-pulsed in primed mouse)/(% cognate peptide-pulsed in naive mouse/% control peptide pulsed in naive mouse)] × 100.

TCD8+ from pooled splenocytes of naive TdT−/− mice were magnetically separated using an AutoMACS (Miltenyi Biotec). Anti-CD11c beads were simultaneously used during TCD8+ negative selection to eliminate splenic dendritic cells. Cell preparations thus obtained were further enriched for TCD8+ using a FACSAria cell sorter (BD Biosciences) to achieve ∼99.9% purity. Between 5–10 × 106 TdT−/− TCD8+ were then injected i.v. into BALB/c mice expressing the Thy-1.1 congenic marker 1 day before i.p. infection with IAV. IAV-specific responses of recipient (Thy-1.1+ WT) and donor (Thy-1.2+ TdT−/−) TCD8+ were evaluated by ICS 7 days postinfection after live gating on CD8+ events and using Thy-1.2 staining to distinguish donor cells from recipient cells.

BALB/c and TdT−/− mice were immunized i.p. with 1 × 106 PFU of VV strain WR. Spleens and PECs were harvested 6 days following injection. CD8+ T cells were prepared from pooled spleens and PECs using a CD8+ isolation kit and autoMACs separator (Miltenyi Biotec, 130–090-859). The flow-through fraction containing CD8-enriched cells was collected, and resulted in >95% CD8+ expression in all populations. A fraction of cells was set aside for further CD8+ purification by flow sorting. The rest were stimulated in vitro for 3 h at 37°C with A52R peptide pulsed P815 cells. Stimulated CD8+ T cells were incubated for 20 min on ice to block the release of IFN-γ. Cells were resuspended in 100 μl of cytokine catch reagent consisting of an anti-CD45 Ab conjugated to an anti-IFN-γ Ab (Miltenyi Biotec) and incubated on ice for 5 min. Ten milliliters of warm DMEM was added and cells were incubated for another 45 min at 37°C under slow continuous rotation. This step is critical to reduce the amount of “trans” capture of IFN-γ. Cells were washed with cold buffer, incubated with Fc Block, and labeled with anti-FITC CD8+ Ab (BD Pharmingen) and anti-APC IFN γ+ Microbeads (Miltenyi Biotec) for 15 min. CD8+, IFN-γ+ cells were sorted from IFN-γ cells using a FACSVantage DIVA (BD Biosciences). Post-sort analysis revealed that for WT and TdT−/− populations, respectively, 98 and 45% of cells were CD8+, IFN-γ+ (data not shown). This discrepancy is not related to differences in the singlet population sorted, which were highly similar. We believe instead that it is due to increased losses post-sorting in surface bound anti-IFN-γ Ab by the TdT−/− population, which remained 99% CD8+. We note that any putative differences in the purity of TdT−/− vs WT cells would minimize, not accentuate, the differences we observed in the oligoclonal populations detected by CDR3 length analysis. Cells were snap frozen in TRIzol (Invitrogen) and shipped to TcLand SA (Nantes, France) for Vβ typing and CDR3 length analysis as previously described (21). In brief, mRNA was reverse transcribed to cDNA that was then PCR amplified in 19 separate reactions using a Cβ primer and primers for each of 19 Vβ genes examined. Each amplification product was elongated using a dye-labeled Cβ primer and its CDR3 length was then determined by sizing via acrylamide-urea gel electrophoresis. Immunoscope software was used to display the CDR3 lengths of the amplified gene products. Real time PCR was used to determine the relative levels of each 19 Vβ genes transcripts. For each reaction, hydroxyphosphoribosyltransferase transcripts were used as an internal control to correct for variability in RNA preparations and the conversion efficiency of the reverse transcription reaction.

Infection of BALB/c mice with the A/Puerto Rico/8/34 strain of IAV induces a well defined TCD8+ immunodominance hierarchy (14, 15) that can be measured by ICS for IFN-γ. Following i.p. infection, responses to individual determinants can be quantitated both locally in PECs and systemically in spleens. The typical immunodominance hierarchy is seen in Fig. 1 where NP147 is clearly the immunodominant determinant among the seven defined IAV determinants (Table I) tested for their recognition by splenic or peritoneal TCD8+. Overall anti-IAV responses can be measured by the activation of TCD8+ by IAV-infected P815 cells. This clearly represents only a fraction of the total response because the numbers of responding cells are less than the sum of the responses to the defined individual determinants (and other major determinants are recognized by BALB/c mice, our unpublished observations). Still, this provides a measure of the magnitude of the anti-IAV TCD8+ response.

FIGURE 1.

Distinct IAV-specific TCD8+ hierarchy in TdT−/− mice. WT and TdT-deficient mice were injected i.p. with IAV. Seven days later, splenocytes and PECs were examined ex vivo for IFN-γ accumulation following restimulation with indicated peptides (left panels) or IAV-infected P815 cells (right panels) used at 4 × 105 cells/well. NP118 is an H-2d-binding immunodominant peptide of LCMV that was used as a negative control. Values are subtracted from the background obtained from wells receiving no peptides and are expressed as mean ± SEM of four individual mice/group for splenic responses. For peritoneal responses, PECs were pooled and analyzed simultaneously. Data are representative of >10 independent experiments yielding similar results.

FIGURE 1.

Distinct IAV-specific TCD8+ hierarchy in TdT−/− mice. WT and TdT-deficient mice were injected i.p. with IAV. Seven days later, splenocytes and PECs were examined ex vivo for IFN-γ accumulation following restimulation with indicated peptides (left panels) or IAV-infected P815 cells (right panels) used at 4 × 105 cells/well. NP118 is an H-2d-binding immunodominant peptide of LCMV that was used as a negative control. Values are subtracted from the background obtained from wells receiving no peptides and are expressed as mean ± SEM of four individual mice/group for splenic responses. For peritoneal responses, PECs were pooled and analyzed simultaneously. Data are representative of >10 independent experiments yielding similar results.

Close modal

In the same experiment, we measured TCD8+ responses in IAV-infected TdT−/− mice. We consistently observed that although the cellularity of TdT−/− spleens was reduced by ∼20% relative to WT spleens, this was usually compensated by a 10–20% increase in the fraction of TCD8+, so that the percentage of responding TCD8+ in TdT−/− vs WT mice generally corresponds well to the absolute numbers of responding cells. Overall IAV-specific splenic and peritoneal TCD8+ responses were reduced by ∼30% relative to WT mice. Remarkably, the immunodominance hierarchy was greatly altered in TdT−/− mice: there was a reduction in TCD8+ specific for the immunodominant determinant, NP147, with a concomitant increase in HA518-specific TCD8+, which now assumed immunodominant status. The rank promotion of HA518 was highly consistent and observed in 10 independent experiments, each typically consisting of three to four mice/group. Further, we also consistently observed a narrowing of the response, with decreases to background levels in responses to NP39, NP218, and HA462.

IFN-γ secretion is but one of multiple TCD8+ functions. To broaden these findings, we measured the in vitro and in vivo cytotoxic functions of IAV-specific TCD8+ in WT and TdT−/− mice. This confirmed that splenic TCD8+ from WT and TdT−/− mice demonstrated a similar reversal in cytotoxic activity against target cells pulsed with NP147 or HA518 (Fig. 2,a). To extend this finding in vivo, we injected mice infected 7 days earlier with a mixture of three target cell populations (BALB/c splenocytes pulsed with synthetic lymphocytic choriomeningitis virus (LCMV) NP118 [negative control], NP147, or HA518) that were distinguished by levels of labeling CFSE (13). Although most NP147-pulsed target cells were cleared from WT spleens 4 h after i.v. injection, only a small fraction of HA518-pulsed target cells were cleared (Fig. 2 b). By contrast, the opposite pattern was observed in TdT−/− mice. We observed a similar hierarchical switch when spleens were harvested 2 or 6 h after target cell injection (data not shown). Similar numbers and ratio of target cells were recovered from naive TdT−/− vs WT mice, indicating that the results are unlikely to be attributed to differences in target cell homing patterns in WT vs TdT−/− mice (data not shown).

FIGURE 2.

Altered immunodominance hierarchy of anti-IAV TCD8+ is also manifest at the level of CTL effector function both ex vivo and in vivo. a, Three mice/group were infected i.p. with IAV. Seven days later, WT and TdT-deficient splenocytes from each group were pooled and used at indicated E:T ratios against P815 cells sensitized with either NP147 or HA518 in a 6-h 51Cr release assay as indicated. Data are from one experiment; another independent experiment gave similar results. Background lysis obtained from wells containing nonsensitized P815 or P815 cells pulsed with NP118 was always between 1 and 4% at the highest E:T ratio used. Spontaneous release of P815 target cells was always <10%. Each data point represents the mean of triplicate samples. b, Target splenocytes were pulsed with NP118 (control) peptide, HA518, or NP147, and stained with CFSE at 0.025, 0.2, and 1.6 μM, respectively. These cells were washed thoroughly, mixed in equal numbers, and injected into tail veins of WT or TdT−/− mice. Mice were either uninfected or infected with PR8 7 days previously. Two hours after injection, splenocytes from each mouse were harvested and analyzed by flow cytometry using differential CFSE fluorescence to distinguish target cells. The percent specific killing of each target cell population was calculated as described in the Material and Methods section and is shown in the figure. Similar results were obtained in two additional experiments.

FIGURE 2.

Altered immunodominance hierarchy of anti-IAV TCD8+ is also manifest at the level of CTL effector function both ex vivo and in vivo. a, Three mice/group were infected i.p. with IAV. Seven days later, WT and TdT-deficient splenocytes from each group were pooled and used at indicated E:T ratios against P815 cells sensitized with either NP147 or HA518 in a 6-h 51Cr release assay as indicated. Data are from one experiment; another independent experiment gave similar results. Background lysis obtained from wells containing nonsensitized P815 or P815 cells pulsed with NP118 was always between 1 and 4% at the highest E:T ratio used. Spontaneous release of P815 target cells was always <10%. Each data point represents the mean of triplicate samples. b, Target splenocytes were pulsed with NP118 (control) peptide, HA518, or NP147, and stained with CFSE at 0.025, 0.2, and 1.6 μM, respectively. These cells were washed thoroughly, mixed in equal numbers, and injected into tail veins of WT or TdT−/− mice. Mice were either uninfected or infected with PR8 7 days previously. Two hours after injection, splenocytes from each mouse were harvested and analyzed by flow cytometry using differential CFSE fluorescence to distinguish target cells. The percent specific killing of each target cell population was calculated as described in the Material and Methods section and is shown in the figure. Similar results were obtained in two additional experiments.

Close modal

These data collectively demonstrate that TdT−/− mice exhibit a marked change in the breadth and composition of the TCD8+ immunodominance hierarchy elicited by IAV infection as measured either by IFN-γ secretion or lytic activities.

We examined the potential contribution of altered Ag presentation in TdT−/− mice to the phenomena described above. We used three approaches to address this issue. First, we generated bone marrow-derived dendritic cells (BMDCs) from WT and TdT−/− mice. IAV-infected BMDCs or P815 cells were tested for their ability to activate splenic and peritoneal primary (day 7) TCD8+ obtained from IAV-infected WT mice as determined by ICS. BMDCs from WT and TdT−/− mice were equally potent stimulators on a per cell basis under the condition activation is limited by numbers of APCs (data not shown).

Next, we examined HA518-specific TCD8+ responses following infection of mice with a rVV that expresses HA518 as an ER-targeted minigene product (rVV-ES-HA518) (Fig. 3, right). We previously provided evidence that this virus generates supernormal levels of HA518-Kd complexes on infected cells (16). If enhanced presentation of HA518 contributes to the enhanced immunodominance status of HA518 in TdT−/− mice, then saturating HA518 presentation should equalize responses between WT and TdT−/− mice. Following 7-day infection, vaccinia virus (VV)-infected TdT−/− mice maintained enhanced primary splenic and peritoneal TCD8+ responses to HA518.

FIGURE 3.

Effect of TdT on response to VV-encoded Ags. Three mice/group were injected with rVV-ES-HA518 i.p. Seven days later, individual spleens and pooled PECs were examined for the presence of TCD8+ responding to HA518 or IAV-infected P815 cells (right panels). At the same time, TCD8+ responses to three H-2d-resricted peptides of VV were evaluated in WT and TdT−/− mice (left panels). Nearly identical results were obtained in an additional experiment also consisting of three mice/group.

FIGURE 3.

Effect of TdT on response to VV-encoded Ags. Three mice/group were injected with rVV-ES-HA518 i.p. Seven days later, individual spleens and pooled PECs were examined for the presence of TCD8+ responding to HA518 or IAV-infected P815 cells (right panels). At the same time, TCD8+ responses to three H-2d-resricted peptides of VV were evaluated in WT and TdT−/− mice (left panels). Nearly identical results were obtained in an additional experiment also consisting of three mice/group.

Close modal

Finally, we adoptively transferred highly purified TCD8+ from TdT−/− mice (which express the Thy 1.2 allele) into WT BALB/c mice expressing the Thy-1.1 allele (Fig. 4 a). Using a fluorochrome-labeled mAb specific for Thy 1.2, we could thereby distinguish donor vs recipient TCD8+ responses following IAV infection. This revealed a clear dichotomy between responses of WT and TdT−/− TCD8+ in a single environment: WT cells responded more vigorously to NP147 while donor cells preferred HA518.

FIGURE 4.

Mechanism of altered immunodominance status of HA518. a, Approximately 10 million highly purified TCD8+ from pooled splenocytes of naive TdT−/− mice were injected i.v. into BALB/c mice expressing the Thy-1.1 congenic marker 1 day before i.p. infection with IAV. NP147- and HA518-specific responses of recipient (Thy-1.1+ WT) and donor (Thy-1.2+ Tk) TCD8+ were evaluated by ICS 7 days postinfection after live gating on CD8+ events and using Thy-1.2 staining to distinguish donor cells from recipient cells. The frequencies of determinant-specific TCD8+ are shown from one representative mouse. This experiment was performed twice, each involving multiple mice, and similar results were obtained for each individual mouse. b, WT and TdT−/− mice were immunized with IAV or the J-1 reassortant virus lacking IAV HA. NP147- and HA518-specific TCD8+ responses were measured by ICS 7 days later. NP147-direced responses were similarly decreased following infection with IAV or J-1 (55.6 and 50.6% reduction, respectively).

FIGURE 4.

Mechanism of altered immunodominance status of HA518. a, Approximately 10 million highly purified TCD8+ from pooled splenocytes of naive TdT−/− mice were injected i.v. into BALB/c mice expressing the Thy-1.1 congenic marker 1 day before i.p. infection with IAV. NP147- and HA518-specific responses of recipient (Thy-1.1+ WT) and donor (Thy-1.2+ Tk) TCD8+ were evaluated by ICS 7 days postinfection after live gating on CD8+ events and using Thy-1.2 staining to distinguish donor cells from recipient cells. The frequencies of determinant-specific TCD8+ are shown from one representative mouse. This experiment was performed twice, each involving multiple mice, and similar results were obtained for each individual mouse. b, WT and TdT−/− mice were immunized with IAV or the J-1 reassortant virus lacking IAV HA. NP147- and HA518-specific TCD8+ responses were measured by ICS 7 days later. NP147-direced responses were similarly decreased following infection with IAV or J-1 (55.6 and 50.6% reduction, respectively).

Close modal

These experiments clearly indicate that the reversal in the immunodominance hierarchy between NP147 and HA518 is due largely, if not exclusively, to properties of the TdT-deficient TCD8+ repertoire.

We next examined the role of immunodomination mediated by HA518-specific TCD8+ in the decreased response to NP147 in TdT−/− mice. Immunodomination is experimentally defined as augmented responses to subdominant determinants when responses to immunodominant determinants are diminished by removing or altering the determinant per se, the restricting MHC class I allomorph, or immunodominant determinant-specific TCD8+. We compared the responses of WT and TdT−/− to IAV vs a recombinant virus (J-1) that possesses seven genes from IAV with an HA gene derived from the A/Hong Kong/68 IAV strain (HK). The HK HA does not cross-react with HA518-specific TCD8+, and in fact, does not appear to induce a TCD8+ response in BALB/c mice (17).

Following infection with J-1 virus, TdT−/− mice demonstrate a decreased response to NP147 relative to the response of WT mice (Fig. 4 b) (the alteration in the magnitude of the response is most likely due to differences in viral tropism and replication rather than to alterations in presentation of the viral peptide determinants). This decrease of a nearly identical magnitude (in relative terms) to the difference observed to IAV in the same experiment supports the conclusion that enhanced immunodominance status of these clones in TdT−/− mice is not due to immunodomination mediated by HA518 TCD8+.

There is evidence that TCD8+ with increased sensitivity for cognate peptide-class I complexes exhibit a proliferative edge in vivo (18). We therefore measured the ability of NP147- and HA518-specific TCD8+ to respond to APCs incubated with decreasing concentrations of cognate synthetic peptide determinants (Fig. 5). Day 7 splenic TCD8+ obtained from WT vs TdT−/− mice specific for NP147 demonstrated highly similar sensitivity (half-maximal activation at ∼ 10−1 M). Critically, however, HA518-specific TCD8+ derived from TdT−/− mice demonstrated ∼10-fold higher sensitivity to peptide than corresponding TCD8+ from WT mice. A similar trend was observed in an independent experiment involving four mice per group that used APCs pulsed with peptide and washed before incubation with TCD8+.

FIGURE 5.

Ag sensitivity of TCD8+ from WT and TdT−/− mice. Two WT and two TdT−/− mice were infected i.p. with IAV. Seven days later, splenic TCD8+ were restimulated with indicated concentrations of NP147 or HA518 and IFN-γ production by these cells was assessed by ICS. The concentration at which a half maximal response was achieved is demonstrated. These results were repeated in an additional experiment looking at TCD8+ responses of four mice/group.

FIGURE 5.

Ag sensitivity of TCD8+ from WT and TdT−/− mice. Two WT and two TdT−/− mice were infected i.p. with IAV. Seven days later, splenic TCD8+ were restimulated with indicated concentrations of NP147 or HA518 and IFN-γ production by these cells was assessed by ICS. The concentration at which a half maximal response was achieved is demonstrated. These results were repeated in an additional experiment looking at TCD8+ responses of four mice/group.

Close modal

These findings are consistent with the idea that the lack of TdT results in the generation of a TCR repertoire with greater sensitivity for detecting HA518-Kd complexes, most likely due to an intrinsically higher affinity of their TCR for the complexes. These data also indicate that TdT is not required to generate NP147-specific TCD8+ with high functional avidity for cognate peptide class I complexes.

To characterize immunodominance hierarchies of memory TCD8+, we needed to boost with an IAV that escapes neutralization by Abs induced by the priming virus. We could neatly do this by boosting with an IAV mutant selected by sequential neutralization with a panel of 12 HA-specific mAbs (19). This mutant has accumulated mutations in the globular domain but retains HA structure and possesses an identical sequence in and around the HA518 sequence. Heterologous prime-boosting with IAV and IAV-SEQ12 indeed resulted in greatly enhanced secondary TCD8+ responses relative to homologous prime-boosting with IAV (data not shown).

Primary infection of mice with IAV-SEQ12 recapitulated the differential hierarchy between WT and TdT−/− mice shown in Fig. 1 (data not shown), indicating the TCD8+ determinants are presented similarly by the two viruses in vivo. We next boosted IAV-primed WT and TdT−/− mice with IAV-SEQ12 (Fig. 6). As in primary responses, the dominance pattern of TCD8+ in TdT−/− mice was changed in favor of HA518 and in a very similar fashion to what is observed in primary responses. Equally notable is the severe narrowing of the TCD8+ response and the decreased overall response to IAV. Despite the prime-boost regimen, we failed to detect responses to three subdominant determinants, NP39, NP218, and HA462. Reciprocal priming with IAV-SEQ12 and boosting with IAV gave similar results (data not shown).

FIGURE 6.

TdT shuffles and narrows the immunodominance hierarchy in memory TCD8+ responses. Mice were primed with IAV and boosted a month later with IAV-SEQ12. Seven days after the boost, splenic and peritoneal TCD8+ responses of WT and TdT-deficient mice were examined against IAV-derived antigenic peptides (left panels) as well as IAV-infected P815 cells (right panels). Error bars for splenic responses represent SEM among three mice/group. The same trend was found in three similar experiments.

FIGURE 6.

TdT shuffles and narrows the immunodominance hierarchy in memory TCD8+ responses. Mice were primed with IAV and boosted a month later with IAV-SEQ12. Seven days after the boost, splenic and peritoneal TCD8+ responses of WT and TdT-deficient mice were examined against IAV-derived antigenic peptides (left panels) as well as IAV-infected P815 cells (right panels). Error bars for splenic responses represent SEM among three mice/group. The same trend was found in three similar experiments.

Close modal

These findings demonstrate that the alterations in the ID hierarchy of TdT−/− mice extends to secondary responses to IAV, and underscores the marked narrowing of the response as well as its decreased magnitude.

We generalized these findings to anti-VV responses by measuring the TCD8+ responses to three H-2d-restricted determinants (20) (Fig. 3 a, these data were obtained in the same experiment that examined responses to the HA518 minigene product). Infection of WT BALB/c mice with VV leads to a highly robust TCD8+ response to F2L23, A52R75, and E3L140 in order of hierarchical dominance. Remarkably, TdT−/− mice demonstrate alterations in their immunodominance hierarchy similar to that observed in IAV responses. Reponses to the immunodominant determinant F2L23 were decreased 2–3-fold while responses to the subdominant determinant A52R75 were increased by ∼2-fold, placing it atop the immunodominance hierarchy.

To gain molecular insight into the effect of TdT on the anti-viral TCD8+ response, we examined TCR Vβ gene usage (19 of 27 mouse Vβ genes analyzed) and CDR3 length by immunoscope analysis of mRNA (21) isolated from purified TCD8+ from naive and day 6 primary VV-infected mice (Fig. 7). CDR3 is the sole variable region of each α and β TCR gene product, and it typically forms the most intimate contacts with MHC-bound peptides in TCR-MHC structures determined by x-ray crystallography (22). Hence, CDR3 plays the most critical role in T cell specificity of cognate peptides.

FIGURE 7.

Vβ repertoire analysis of TCD8+ TCD8+ were purified from splenocytes pooled from six naive or ten TdT−/− mice immunized 6 days previously with VV via i.p. injection, or from two WT or one TdT−/− naive mice. RNA prepared from purified TCD8+ was analyzed for Vβ usage and CDR3 length using immunoscope technology. The percentage of total TCD8+ expressing the indicated Vβ-chain is given in the upper left corner of each histogram. The distribution of CDR3 lengths is shown by the peak heights plotted against CDR3 length on the x-axis (each histogram starts with CDR3 = one amino acid and extends to the longest CDR3 segment detect). Arrows depict populations demonstrating significant expansion relative to naive mice.

FIGURE 7.

Vβ repertoire analysis of TCD8+ TCD8+ were purified from splenocytes pooled from six naive or ten TdT−/− mice immunized 6 days previously with VV via i.p. injection, or from two WT or one TdT−/− naive mice. RNA prepared from purified TCD8+ was analyzed for Vβ usage and CDR3 length using immunoscope technology. The percentage of total TCD8+ expressing the indicated Vβ-chain is given in the upper left corner of each histogram. The distribution of CDR3 lengths is shown by the peak heights plotted against CDR3 length on the x-axis (each histogram starts with CDR3 = one amino acid and extends to the longest CDR3 segment detect). Arrows depict populations demonstrating significant expansion relative to naive mice.

Close modal

The validity of our immunoscope analysis is supported by the following: 1) the paucity of TCD8+ expressing Vβ3, Vβ5, and Vβ12; this is expected due to the deletion of clones bearing these TCRs due their interaction with endogenous retroviral elements in BALB/c mice (23) and 2) the distribution of Vβ-chains, which is consistent with previous reports of the naive BALB/c T cell repertoire, as is the Gaussian distribution of CDR3 segment length centered around seven amino acids (24). Surprisingly, while TdT deletion had the anticipated effect of shortening the average length of CDR3 sequences in each Vβ family by 1–2 amino acids, it had only minor effects on Vβ-chain representation in the naive repertoire, which was nearly identical to WT mice. This indicates that Vβ usage occurs independently of TdT expression, which is consistent with the conservation between Vβ usage in neonatal and adult T cell repertoires (25).

Ag-driven clonal expansions can frequently be detected as perturbations in the Gaussian CDR3 distribution as a relative increase in the fraction of TCRs with a given CDR length. These can be observed even in a total TCD8+ population if the overall response is of sufficient magnitude (24) (which is why we studied TCD8+ induced by VV instead of IAV, because the response is of much greater magnitude). On day 6 of a primary VV infection, major expansion of individual CDR3 populations are indicated in Fig. 7 by a red arrowhead. It can be seen that 11 such populations can be identified by eye in the WT anti-VV response. A disproportionate number of these responses occur in minor populations (Vβ3, 5, 12, 18, and 20). This is to be expected because these populations should have less diversity (based on their selection of avoiding super Ag deletion), which should facilitate detection of oligoclonal expansions. Expansions are also detected in Vβ1 and Vβ7 populations, however, which together constitute 20% of the total TCD8+ response. Notably, by the criterion of detectable CDR3 population expansions, the polyclonal anti-VV response of TdT−/− mice is broader than WT mice, with 17 clearly perturbed populations with the increase due to more perturbations in the predominant Vβ families (i.e., the families expressed by the highest percentage of TCD8+) (Fig. 7).

This analysis of total TCD8+ populations provides a rough idea of clonal diversity, as it is obviously limited by the presence of the majority of TCD8+ that do not respond to VV. To get a better idea of the effect of TdT in shaping the antiviral repertoire, and particularly the phenomenon of the promotion of subdominant determinant to immunodominant status, we analyzed TCD8+ that were sorted ex vivo on the basis of IFN-γ secretion upon activation by A52R75–83 peptide-coated APCs. In this case, nearly every Vβ population in both WT and TdT−/− mice exhibited oligoclonal expansion, demonstrating the validity of the immunoscope analysis and the effectiveness of purifying A52R75-specific TCD8+. These data are all the more remarkable because the analysis was performed on TCD8+ pooled from many individual animals (ten WT and six TdT−/− mice were used). Nearly all of the specifically expanded populations in TdT−/− polyclonal anti-VV population were also present in the A52R75-specific population, while only a few of expanded populations in WT mice were similarly matched. This is completely consistent with the immunodominance status of A52R75-specific TCD8+ in TdT−/− vs WT mice.

The most important finding in this experiment is that A52R75-specific responses of WT and TdT use a similarly wide, but clearly distinct range of Vβ-chains. Notably, the CDR3 regions of responding TCD8+ in TdT−/− mice are shorter by an average of 1.25 amino acids (5.1 vs 6.35), and are more uniform in length as reflected by a narrower SD (TdT − SD = 1.1 (n = 20) vs WT SD = 1.9 (n = 21)). These data indicate that the greater immunogenicity of A52R75 in TdT−/− mice is not due to a “jackpot” scenario of expansion of a single closely related family of clones. Rather, it reflects the activation of a wide array of clones scattered among virtually all of the Vβ families examined. Notably, most of responding TdT clones use shorter CDR3 segments than their counterparts in the corresponding WT Vβ families.

The quality, quantity, and kinetics of Ag presentation have been the focus of most mechanistic investigations into immunodominance in TCD8+ responses to intracellular pathogens. This is but half of the equation, however, as much remains to be learned about how determinant-specific TCD8+ clonal variation affects immunodominance. Variation in TCD8+ precursor frequencies and doubling times would obviously be intrinsic contributors to the immunodominance hierarchy. Clonal differences in the numbers of complexes or degree of costimulation needed for TCD8+ activation must also contribute to immunodominance, though more subtly, as they would operate in conjunction with variables in Ag presentation. Clonal differences in the abilities of TCD8+ to modulate the behavior of APCs and other T cells would also influence immunodominance.

The TdT gene is highly conserved among vertebrates. The sole defined function of TdT is to generate lymphocyte Ag receptor diversity. It might be anticipated that TdT−/− mice would demonstrate easily detected differences in their immunity to pathogens. Strikingly, TdT−/− mice do not exhibit increased susceptibility to pathogens that commonly cause morbidity and/or mortality in immunodeficient animals in conventional animal care facilities (6). TdT−/− mice survived an outbreak of Sendai virus that annihilated immunocompromised strains in the same colony. TdT−/− mice also cope well with LCMV, a natural mouse pathogen. Gilfillan et al. (26) further showed that WT and TdT−/− mice are similar in T cell proliferation and Ab responses to soluble protein immunogens and CTL responses to LCMV, VV, or vesicular stomatitis virus infections. No consistent differences were noted in TCD4+ immunodominance hierarchies to soluble Ages. Antiviral TCD8+ responses in this work were measured by 51Cr release assay using virus-infected cells, which provides a semiquantitative measure for enumerating TCD8+ responses because it is based not only on numbers of responding TCD8+ but also their capacity to lyse multiple target cells during the course of the assay. Responses to individual defined determinants were not measured, so the role of TdT in generating the immunodominance hierarchy was not explored.

In this study, we provide the initial evidence for the importance of TdT in antipathogen immunity. We show that TdT has a major influence in establishing the immunodominance hierarchy to IAV and VV, two radically different viruses in terms of complexity, structure, genome, replication strategy, and host cell range. Detailed study of the IAV response conclusively demonstrates that these differences are directly related to intrinsic differences between the TCD8+ compartment of WT and TdT−/− mice and are not due to alterations in Ag presentation.

We find that TdT plays a major role in setting the hierarchy among the most dominant determinants in both IAV and VV infection. In IAV infections, it appears that increased sensitivity of TCD8+ activation explains the enhanced status of HA518-specific clones. Assuming that this is due to increased TCR affinity for the HA518-Kd complex (which remains to be established experimentally), this would mean that TdT expression results in the loss of a fraction of high affinity clones for some determinants. The implication is that shorter CDR3 regions enable higher affinity TCRs for a subset of determinants. The key word here is subset, because the cost of better recognition appears to be diminished recognition of a larger subset of determinants that for IAV includes NP147 and all of the subdominant determinants.

Gavin and Bevan (11) reported that highest affinity TCD8+ clones from TdT−/− mice were more cross-reactive for peptides present in a large synthetic library. The authors speculated that this was based on higher affinity docking with class I α-helices and further, that such clones might provide protective immunity against multiple pathogens at the expense of increased susceptibility to autoimmune abnormalities. To date, however, it appears that the opposite is true. TdT−/− mice are reported to be protected against autoimmune diabetes in the NOD model (27) and glomerulonephritis in lupus-prone MRL-Faslpr mice (28) and (NZB × NZW) F1 mice (29). A second school of thought advocates a link between higher numbers of TCR specificities (rather than cross-reactivity) and autoimmune process (30). The delay in TdT expression in ontogeny may therefore allow peripheral tolerance mechanisms to develop before the acquisition of a highly diverse repertoire with increased self-recognition.

Turner, Doherty, and colleagues have recently explored the relationship between peptide class I complex structure and the mouse TCR repertoire (31, 32). They found a correlation between the dominant use of public TCR specificities (i.e., sequences closely related to germline sequences shared between all responding individual inbred mice) and the lack of prominent peptide side chain structures protruding from the class I groove. Based on Mitaksov and Fremont’s (33) recent x-ray crystallographic structural determination of NP147-Kd complexes, NP147 qualifies as a “structure rich” peptide (with Thr, Arg, and Arg oriented to the TCR at positions P1, P4, and P6, respectively), while HA518 is more “vanilla” (Ile, Ala, Ala) (Table I). The decreased recognition of NP147 and increased recognition of HA518 by TdT−/− TCD8+ is consistent with the Turner-Doherty model. On the other hand, TdT−/− TCD8+ also demonstrate enhanced recognition of the structure rich Kd-restricted A52R75 VV determinant. This suggests either that the correlation between peptide structure and public TCR usage is not universal, or that TdT activity is necessary to generate the core public specificities that demonstrate the proposed correlation.

By analyzing Vβ gene expression and CDR3 length, we found that TdT is not required to mount a diverse response to the VV A52R75 determinant, as the response includes TCD8+ expressing each of the 19 Vβ families we examined. Because we sorted A52R75-specific TCD8+ on basis of Ag-induced IFN-γ secretion, it appears that each of the Vβ families contributes functionally relevant TCD8+ that recognize this one determinant. CDR3 length analysis clearly indicates that in most of these families, the dominant clones are shorter by a single amino acid relative to the TCRs from WT mice. Thus, for some determinants, TdT-mediated insertion of amino acids in a wide variety of Vβ-chains generally results in a less fit TCR by the criteria of the magnitude of response. In contrast, our finding that overall antiviral responses are decreased and narrowed in TdT−/− mice demonstrates that the negative effect of TdT on responses to some individual determinants (like A52R75 and HA518) is outweighed by its effects on TCRs that recognize the majority of immunogenic viral determinants.

The role of TdT in shaping antiviral determinant responses was first explored by Fazilleau et al. (34) who used TdT−/− C57BL/6 mice to study TCD8+ responses to a hepatitis B virus TCD4+ determinant and an IAV TCD8+ determinant (NP366) introduced as immunogens as, respectively, a synthetic peptide or a NP-expressing plasmid. In contrast to our findings, the overall response to these determinants was extremely narrow in both WT and TdT−/− mice, being limited to a single predominant Vβ receptor (Vβ8.3 in the case of NP366). This may have resulted from the in vitro Ag-driven expansion of responding T cells from DNA vaccination or from the use of tetramers to isolate TCD8+, because NP366-specific TCD8+ induced by IAV infection have been reported to comprise at least 11 different Vβ gene products (35). Indeed, using mAbs specific for 15 individual Vβ gene products, we reported that in the BALB/c response to five different IAV determinants, multiple Vβ families (typically nine) contributed to the response to each determinant (14).

Using Db-NP366 tetramer off-rate as a measure of TCR affinity, Fazilleau et al. (34) found that TdT−/− and WT TCD8+ have a similar affinity for Db-NP366 complexes. This is consistent with our findings regarding similar sensitivity of WT and TdT−/− TCD8+ for NP147-Kd complexes. Performing detailed molecular analysis of V chain expression, Fazilleau et al. reported that TdT expression had a minor effect on the Jβ gene usage and CDR3 sequences, exerting larger effects on Jα usage and CDR sequences. We extend these findings by showing that for A52R75-specific TCD8+, TdT expression has protean effects on Vβ-chain usage itself. This suggests that in broadening the overall TCD8+ repertoire, TdT creates holes in the repertoire among a wide variety of CDR3 regions that modifies the ability of Vβ family members to recognize certain antigenic peptides. Presumably, in the case of HA518 and A52R75, the TdT-dependent holes cannot be completely compensated by recruiting other Vβ family members, thus resulting in a diminished response to these determinants in WT mice.

The positive effect of TdT expression on the overall magnitude and breadth of antiviral TCD8+ is probably our most important finding. The decreased magnitude of the TCD8+ response in TdT−/− mice may compromise their capacity to eradicate viruses (and other intracellular pathogens) under circumstances where the immune system is just able to cope with the infectious burden. Further, the narrowing of the response to fewer viral determinants should favor the generation of viral escape mutants, and concomitantly decrease the functional diversity of the response, which has been linked to TCD8+ specificity (36, 37, 38). Together, these factors likely contribute to the evolution and maintenance of the TdT in vertebrate genomes by providing positive selective pressure based on resistance to infection.

Deborah Tokarchick and Dr. James Gibbs provided outstanding technical support. We are indebted to John Kearney for providing TdT−/− mice.

The authors have no financial conflict of interest.

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

1

This work was supported by the Intramural Research Program of the National Institute of Allergy and Infectious Diseases, National Institutes of Health.

3

Abbreviations used in this paper: TCD8+, targets for CD8+ T lymphocyte; WT, wild type; IAV, influenza A virus; ICS, intracellular cytokine staining; PEC, peritoneal exudate cell; LCMV, lymphocytic choriomeningitis virus; BMDC, bone marrow-derived dendritic cell; rVV, recombinant vaccinia virus; HK, Hong Kong strain; VV, vaccinia virus.

1
Yewdell, J. W., J. R. Bennink.
1999
. Immunodominance in major histocompatibility complex class I-restricted T lymphocyte responses.
Annu. Rev. Immunol.
17
:
51
-88.
2
Haeryfar, S. M., R. J. DiPaolo, D. C. Tscharke, J. R. Bennink, J. W. Yewdell.
2005
. Regulatory T cells suppress CD8+ T cell responses induced by direct priming and cross-priming and moderate immunodominance disparities 2.
J. Immunol.
174
:
3344
-3351.
3
Davis, M. M., P. J. Bjorkman.
1988
. T-cell antigen receptor genes and T-cell recognition.
Nature
334
:
395
-402.
4
Nikolich-Zugich, J., M. K. Slifka, I. Messaoudi.
2004
. The many important facets of T-cell repertoire diversity.
Nat. Rev. Immunol.
4
:
123
-132.
5
Komori, T., A. Okada, V. Stewart, F. W. Alt.
1993
. Lack of N regions in antigen receptor variable region genes of TdT-deficient lymphocytes.
Science
261
:
1171
-1175.
6
Gilfillan, S., A. Dierich, M. Lemeur, C. Benoist, D. Mathis.
1993
. Mice lacking TdT: mature animals with an immature lymphocyte repertoire.
Science
261
:
1175
-1178.
7
Landreth, K. S., K. McCoy, J. Clagett, F. J. Bollum, C. Rosse.
1981
. Deficiency in cells expressing terminal transferase in autoimmune (motheaten) mice.
Nature
290
:
409
-411.
8
Bonati, A., P. Zanelli, M. Savi, T. M. Neri.
1994
. TCR-β chain gene rearrangement and expression in human T-cell development and in leukemia.
Leukemia
8
:
918
-923.
9
Bogue, M., S. Gilfillan, C. Benoist, D. Mathis.
1992
. Regulation of N-region diversity in antigen receptors through thymocyte differentiation and thymus ontogeny.
Proc. Natl. Acad. Sci. USA
89
:
11011
-11015.
10
Cabaniols, J. P., N. Fazilleau, A. Casrouge, P. Kourilsky, J. M. Kanellopoulos.
2001
. Most α/β T cell receptor diversity is due to terminal deoxynucleotidyl transferase.
J. Exp. Med.
194
:
1385
-1390.
11
Gavin, M. A., M. J. Bevan.
1995
. Increased peptide promiscuity provides a rationale for the lack of N regions in the neonatal T cell repertoire.
Immunity
3
:
793
-800.
12
Chen, W., J. W. Yewdell, R. L. Levine, J. R. Bennink.
1999
. Modification of cysteine residues in vitro and in vivo affects the immunogenicity and antigenicity of major histocompatibility complex class I-restricted viral determinants.
J. Exp. Med.
189
:
1757
-1764.
13
Coles, R. M., S. N. Mueller, W. R. Heath, F. R. Carbone, A. G. Brooks.
2002
. Progression of armed CTL from draining lymph node to spleen shortly after localized infection with herpes simplex virus 1 3.
J. Immunol.
168
:
834
-838.
14
Chen, W., L. C. Anton, J. R. Bennink, J. W. Yewdell.
2000
. Dissecting the multifactorial causes of immunodominance in class I- restricted T cell responses to viruses.
Immunity
12
:
83
-93.
15
Chen, W., J. R. Bennink, P. A. Morton, J. W. Yewdell.
2002
. Mice deficient in perforin, CD4+ T cells, or CD28-mediated signaling maintain the typical immunodominance hierarchies of CD8+ T-cell responses to influenza virus.
J. Virol.
76
:
10332
-10337.
16
Deng, Y., J. W. Yewdell, L. C. Eisenlohr, J. R. Bennink.
1997
. MHC affinity, peptide liberation, T cell repertoire, and immunodominance all contribute to the paucity of MHC class I-restricted peptides recognized by antiviral CTL.
J. Immunol.
158
:
1507
-1515.
17
Bennink, J. R., J. W. Yewdell, G. L. Smith, B. Moss.
1986
. Recognition of cloned influenza virus hemagglutinin gene products by cytotoxic T lymphocytes.
J. Virol.
57
:
786
-791.
18
Kedl, R. M., J. W. Kappler, P. Marrack.
2003
. Epitope dominance, competition, and T cell affinity maturation.
Curr. Opin. Immunol.
15
:
120
-127.
19
Mozdzanowska, K., J. Feng, M. Eid, G. Kragol, M. Cudic, L. Otvos, Jr, W. Gerhard.
2003
. Induction of influenza type A virus-specific resistance by immunization of mice with a synthetic multiple antigenic peptide vaccine that contains ectodomains of matrix protein 2.
Vaccine
21
:
2616
-2626.
20
Tscharke, D. C., W. P. Woo, I. G. Sakala, J. Sidney, A. Sette, D. J. Moss, J. R. Bennink, G. Karupiah, J. W. Yewdell.
2006
. Poxvirus CD8+ T-cell determinants and cross-reactivity in BALB/c mice.
J. Virol.
80
:
6318
-6323.
21
Sebille, F., K. Gagne, M. Guillet, N. Degauque, A. Pallier, S. Brouard, B. Vanhove, M. A. Delsuc, J. P. Soulillou.
2001
. Direct recognition of foreign MHC determinants by naive T cells mobilizes specific Vβ families without skewing of the complementarity-determining region 3 length distribution.
J. Immunol.
167
:
3082
-3088.
22
Garcia, K. C., L. Teyton, I. A. Wilson.
1999
. Structural basis of T cell recognition.
Annu. Rev. Immunol.
17
:
369
-397.
23
Hodes, R. J., R. Abe.
1992
. T cell recognition of Mls-like superantigens: analysis of TCR requirements, superantigenic ligands, and signal transduction.
Semin. Immunol.
4
:
319
-327.
24
Sourdive, D. J., K. Murali-Krishna, J. D. Altman, A. J. Zajac, J. K. Whitmire, C. Pannetier, P. Kourilsky, B. Evavold, A. Sette, R. Ahmed.
1998
. Conserved T cell receptor repertoire in primary and memory CD8 T cell responses to an acute viral infection.
J. Exp. Med.
188
:
71
-82.
25
Gonzalez-Quintial, R., R. Baccala, R. S. Balderas, A. N. Theofilopoulos.
1995
. V β gene repertoire in the aging mouse: a developmental perspective.
Int. Rev. Immunol.
12
:
27
-40.
26
Gilfillan, S., M. Bachmann, S. Trembleau, L. Adorini, U. Kalinke, R. Zinkernagel, C. Benoist, D. Mathis.
1995
. Efficient immune responses in mice lacking N-region diversity.
Eur. J. Immunol.
25
:
3115
-3122.
27
Gilfillan, S., C. Benoist, D. Mathis.
1995
. Mice lacking terminal deoxynucleotidyl transferase: adult mice with a fetal antigen receptor repertoire.
Immunol. Rev.
148
:
201
-219.
28
Robey, I. F., M. Peterson, M. S. Horwitz, D. H. Kono, T. Stratmann, A. N. Theofilopoulos, N. Sarvetnick, L. Teyton, A. J. Feeney.
2004
. Terminal deoxynucleotidyltransferase deficiency decreases autoimmune disease in diabetes-prone nonobese diabetic mice and lupus-prone MRL-Fas(lpr) mice.
J. Immunol.
172
:
4624
-4629.
29
Conde, C., S. Weller, S. Gilfillan, L. Marcellin, T. Martin, J. L. Pasquali.
1998
. Terminal deoxynucleotidyl transferase deficiency reduces the incidence of autoimmune nephritis in (New Zealand Black x New Zealand White)F1 mice.
J. Immunol.
161
:
7023
-7030.
30
Benedict, C. L., S. Gilfillan, T. H. Thai, J. F. Kearney.
2000
. Terminal deoxynucleotidyl transferase and repertoire development.
Immunol. Rev.
175
:
150
-157.
31
Turner, S. J., K. Kedzierska, H. Komodromou, N. L. La Gruta, M. A. Dunstone, A. I. Webb, R. Webby, H. Walden, W. Xie, J. McCluskey, et al
2005
. Lack of prominent peptide-major histocompatibility complex features limits repertoire diversity in virus-specific CD8+ T cell populations.
Nat. Immunol.
6
:
382
-389.
32
Turner, S. J., K. Kedzierska, N. L. La Gruta, R. Webby, P. C. Doherty.
2004
. Characterization of CD8+ T cell repertoire diversity and persistence in the influenza A virus model of localized, transient infection.
Semin. Immunol.
16
:
179
-184.
33
Mitaksov, V., D. H. Fremont.
2006
. Structural definition of the H-2Kd peptide-binding motif.
J. Biol. Chem.
281
:
10618
-10625.
34
Fazilleau, N., J. P. Cabaniols, F. Lemaitre, I. Motta, P. Kourilsky, J. M. Kanellopoulos.
2005
. Vα and Vβ public repertoires are highly conserved in terminal deoxynucleotidyl transferase-deficient mice.
J. Immunol.
174
:
345
-355.
35
Deckhut, A. M., W. Allan, A. McMickle, M. Eichelberger, M. A. Blackman, P. C. Doherty, D. L. Woodland.
1993
. Prominent usage of Vá8.3 T cells in the H-2D b -restricted response to an influenza A virus nucleoprotein epitope.
J. Immunol.
151
:
2658
-2666.
36
Kiepiela, P., A. J. Leslie, I. Honeyborne, D. Ramduth, C. Thobakgale, S. Chetty, P. Rathnavalu, C. Moore, K. J. Pfafferott, L. Hilton, et al
2004
. Dominant influence of HLA-B in mediating the potential co-evolution of HIV and HLA.
Nature
432
:
769
-775.
37
Crowe, S. R., S. C. Miller, R. M. Shenyo, D. L. Woodland.
2005
. Vaccination with an acidic polymerase epitope of influenza virus elicits a potent antiviral T cell response but delayed clearance of an influenza virus challenge.
J. Immunol.
174
:
696
-701.
38
Yant, L. J., T. C. Friedrich, R. C. Johnson, G. E. May, N. J. Maness, A. M. Enz, J. D. Lifson, D. H. O'Connor, M. Carrington, D. I. Watkins.
2006
. The high-frequency major histocompatibility complex class I allele Mamu-B*17 is associated with control of simian immunodeficiency virus SIVmac239 replication.
J. Virol.
80
:
5074
-5077.