Influenza virus-specific CD8+ T cell clonotypes generated and maintained in C57BL/6J mice after respiratory challenge were found previously to distribute unequally between the CD62Llow “effector” (TEM) and CD62Lhigh “central” (TCM) memory subsets. Defined by the CDR3β sequence, most of the prominent TCRs were represented in both the CD62Lhigh and CD62Llow subsets, but there was also a substantial number of diverse, but generally small, CD62Lhigh-only clonotypes. The question asked here is how secondary challenge influences both the diversity and the continuity of TCR representation in the TCM and TEM subsets generated following primary exposure. The experiments use single-cell RT-PCR to correlate clonotypic composition with CD62L phenotype for secondary influenza-specific CD8+ T cell responses directed at the prominent DbNP366 and DbPA224 epitopes. In both the acute and long-term memory phases of the recall responses to these epitopes, we found evidence of a convergence of TCR repertoire expression for the CD62Llow and CD62Lhigh populations. In fact, unlike the primary response, there were no significant differences in clonotypic diversity between the CD62Llow and CD62Lhigh subsets. This “TCR homogenization” for the CD62Lhigh and CD62Llow CD8+ populations recalled after secondary challenge indicates common origin, most likely from the high prevalence populations in the CD62Lhigh central memory set. Our study thus provides key insights into the TCR diversity spectrum for CD62Lhigh and CD62Llow T cells generated from a normal, unmanipulated T cell repertoire following secondary challenge. A better understanding of TCR selection and maintenance has implications for improved vaccine and immunotherapy protocols.

The establishment and persistence of CD8+ T cell memory provides significant protection against challenge with intracellular pathogens like the influenza A viruses (1). Recent work on human and murine T cell memory has suggested that there are distinct “central” (TCM,3 CD62Lhigh, CCR7high) and “effector” (TEM, CD62Llow, CCR7low) memory subsets that, while both can be found in peripheral blood and spleen (2, 3, 4, 5), recirculate preferentially through lymphoid (TCM) or nonlymphoid (TEM) microenvironments. Apart from this anatomical compartmentalization, these TCM and TEM memory populations are thought to differ at the functional level (2, 3, 5). The TEM are thought of as “immediate” CTL effectors, but with limited proliferative capacity, whereas the TCM are characterized by high proliferative potential but delayed effector function.

Assessing the TCR diversity characteristics of the CD62Lhigh and CD62Llow Ag-specific memory populations allows us to define the clonotypic characteristics of the TCM and TEM subsets generated in normal, unmanipulated (non-TCR transgenic) mice. To date, the extent of clonotypic diversity between and within the TEM (CD62Llow) and TCM (CD62Lhigh) subsets has received comparatively little attention. In humans, a longitudinal analysis of TCRs expressed on influenza virus-specific CD8+ memory T cells indicated that the extent of clonal diversity was greater for the CD62Lhigh TCM set (6). Mouse experiments using adoptively transferred TCRβ transgenic lymphocytes with variable TCRα-chains established that two-thirds of the memory TEM and TCM clones shared a common naive precursor (7). Paralleling the observations in humans (6), nonshared (between the CD62Lhigh and CD62Llow subsets) CD8+ T cell clonotypes were found predominantly in the CD62Lhigh subset (7).

Our recent, single-cell TCR CDR3β analysis of the primary, endogenous B6 (H2b) mouse response to the prominent influenza A virus DbNP366 and DbPA224 epitopes (8) defined the extent of clonotypic diversity and composition for the CD62Lhigh and CD62Llow subsets, from the earliest phase (day 8) through to the very long term (day 690). The spectrum of TCR repertoire diversity was greater for the CD62Lhigh set (8), as found for humans and TCR transgenics (6, 7). A substantial component of the more varied CD62Lhigh TCM pool was comprised of small clones expressing unique TCR characteristics, as defined by sequence, Jβ usage, and CDR3β length. It thus seems that, while the majority (at least two-thirds) of the TCR repertoire in “primary memory” segregates between both the TCM and TEM compartments (6, 7, 8), the CD62Lhigh population contains additional clonotypes with less “consensus” TCR characteristics (8).

The roles of the TCM and TEM CD8+ T cell subsets in secondary immune responses are not well understood. Several studies (5, 9, 10, 11) have observed differential contributions of the TCM and TEM sets to both early and late secondary immune responses. The nature of these differences appears to be infection-dependent. Studies of the CD8+ T cell response to lymphocytic choriomeningitis virus (5) and vesicular stomatitis virus (11) in mice indicated that the TCM subset makes a greater contribution than does the TEM subset after the secondary infection. Conversely, secondary responses to Plasmodium berghei malaria in mice were dominated by the TEM precursors (12). Both the TCM and TEM sets made comparable contributions to the secondary response and the secondary memory population following Listeria monocytogenes infection (11). Interestingly, secondary responses to Sendai virus in mice were dominated by the TEM set early after infection, while the TCM population was more prominent in the late infection (10). The varied nature of the recall responses for these different pathogens suggests two possible scenarios for CD8+ T cell responses to secondary challenges: 1) both the TCM and the TEM are stimulated and proliferate comparably in response to the secondary infection; and 2) the TCM subset dominates the recall response, due to the greater proliferative potential of the TCM pool (5). The further possibility that much of the recall response comes from the TEM pool is unlikely if the finding of early, but not later, prominence for Sendai virus (10) turns out to be generally true.

A pattern of differential contribution to recall responses could be explained, at least in part, by unequal proliferation rates for the various clonotypes, or by the involvement of only a restricted portion of the TCM and/or TEM subsets following secondary challenge. To better understand the role of the TCM and TEM populations in secondary responses, we studied the diversity of TCR repertoires for the tetramer+ CD62Lhigh and CD62Llow subsets isolated at the peak (day 8) of acute infection and in long-term (>day 300) secondary memory. In particular, we were interested to see whether the differences between the CD62Lhigh and CD62Llow subsets observed in the primary responses persisted following secondary challenge. The present study involves assessment of clonal diversity and composition for two secondary influenza-specific CD8+ T cell populations (DbNP366+CD8+ and DbPA224+CD8+) with distinct TCR characteristics (13, 14).

Female C56BL/6J (H-2b) mice were bred at the Department of Microbiology and Immunology, University of Melbourne (Parkville, Victoria, Australia). Naive mice at 6–8 wk of age were injected i.p. ith 1.5 × 107 PFU the PR8 (H1N1) influenza A virus. At least 6 wk after primary infection, mice were anesthetized by isoflurane inhalation and infected intranasally (i.n.) with 104 PFU of the HKx31 (H3N2) influenza A virus in 30 μl of PBS (secondary challenge). Both virus stocks were grown in the allantoic cavity of 10-day embryonated hens’ eggs and quantified as PFU on monolayers of Madin-Darby canine kidney cells. All animal work was performed in compliance with the guidelines set out by the University of Melbourne Animal Experimental Ethics Committee.

Spleens were recovered from mice at the acute (days 6 and 8) or long-term memory (≥day 300) time points of the secondary response. Spleens were disrupted and enriched for CD8+ T cells by panning on goat anti-mouse IgG and IgM Ab coated plates (Jackson ImmunoResearch Laboratories) for 1 h at 37°C. Lungs were recovered at an early time point after secondary challenge (day 6), disrupted, and depleted of RBC. Cells were washed and resuspended either in FACS buffer (1% BSA/0.2% NaN3 in PBS) for phenotypic analysis or in sort buffer (0.1% BSA in PBS) for single-cell sorting.

Lymphocytes from spleens or lungs of secondarily challenged mice were stained with either the DbNP366-PE or DbPA224-PE tetramers for 1 h at room temperature. Cells were then washed twice and stained with conjugated mAbs against surface CD8α-PerCP, CD62L-FITC, and CD44-APC for 30 min at 4°C. After two washes, lymphocytes were analyzed by flow cytometry.

CD8+ T cell-enriched lymphocytes from spleen were stained with DbNP366 or DbPA224 tetramers conjugated to streptavidin-PE (Molecular Probes) for 60 min at room temperature. Cells were washed twice in FACS buffer (10% BSA/0.02% NaAz in PBS). As a measure of TCR avidity, spleen samples were used in tetramer dissociation assay (15, 16). After cells were stained with tetramers as described above, cells were washed and incubated in the presence of anti-H2Db Ab (28-14-8, BD Biosciences) at 5 μg/ml at 37°C to prevent tetramer rebinding. At specific times, cells were removed into FACS buffer and placed on ice, stained with anti-CD8 mAb conjugated to FITC (BD Biosciences) for 30 min on ice, washed twice, and analyzed by flow cytometry. Loss of tetramer+CD8+ T cells at particular time points was calculated in comparison to tetramer staining at ti = 0 min.

CD8+ T cell-enriched lymphocyte populations were stained either with DbNP366-APC or DbPA224-APC for 60 min at room temperature, followed by two washes in sort buffer. Cells were then stained with anti-CD8-APC-Cy7, anti-CD62L-APC, and anti-Vβ8.3-FITC or anti-Vβ7-FITC Abs for DbNP366+CD8+ or DbPA224+CD8+ T cells, respectively (BD Pharmingen). After two washes, cells were resuspended in 500 μl sort buffer and transferred to polypropylene FACS tubes (BD Biosciences) for single-cell sorting. Lymphocytes were isolated using a MoFlo sorter (DakoCytomation). Single DbNP336+VB8.3+CD8+ or DbPA224+VB7+CD8+ T cells were sorted directly into a 96-well PCR plate (Eppendorf) containing 5 μl of cDNA reaction mix (13, 14). Negative controls were interspersed between the samples (1 in 10), and 80 cells were sorted per plate. After sorting, plates were incubated at 37°C for 90 min for cDNA synthesis, followed by 5 min at 95°C. The Vβ8.3 and Vβ7 transcripts were amplified and sequenced (13, 14).

Simpson’s diversity index and the Morisita-Horn similarity index were used as measures of clonotypic diversity and similarity, respectively. These indices account for both the variety of distinct clonotypes (defined by nucleotide sequences; each clonotype is given an equal value irrespective of size) and the clone size (number of copies) of each clonotype involved in the epitope-specific response within each mouse. The Simpson’s diversity indices and the Morisita-Horn similarity indices were calculated in conjunction with a randomization procedure to correct for the differences in sample sizes between the CD62Llow and CD62Lhigh subsets across all mice (17), and they were estimated as if 24 TCR sequences had been obtained for each sample. Data from sets with small sample sizes (<20 TCR sequences) were excluded from this analysis. The percentage of shared clonotypes/TCR repertoire and the diversity between the CD62Llow and CD62Lhigh subsets were compared using a Wilcoxon rank-sum test. A Mann-Whitney U test was used to compare the percentage of shared clonotypes/clones and the diversity and similarity of the CD62Llow and CD62Lhigh subsets between primary and secondary responses.

Naive B6 mice were primed i.p. with the PR8 (H1N1) influenza A virus, then challenged i.n. at least 6 wk later with the serologically distinct HKx31 (H3N2) virus that shares the NP366–374 and PA224–236 peptides. They were then sampled for FACS phenotyping and/or single-cell CDR3β repertoire analysis (using the DbNP366 and DbPA224 tetramers) from day 6 to day 510 following the HKx31 secondary exposure. The analysis focuses on the spleen, reflecting the ease of recovery for the relatively low-frequency CD8+ memory T cells from this large lymphoid organ and the presence of both CD62Llow and CD62Lhigh CD8+ T cell subsets. Earlier experiments have established that the clonotypic composition of epitope-specific CD8+ T cells present in spleen is comparable to that found in the peripheral blood, the site of infection (lungs and bronchoalveolar lavage) and nonlymphoid organs such as the liver (13, 14).

During primary infection, the DbNP366+CD8+ and DbPA224+CD8+ T cell responses are approximately equivalent in magnitude, although the DbPA224+CD8+ set peaks 1–2 days earlier and the numbers of early memory DbNP366+CD8+ T cells can be a little higher (18, 19, 20). Following secondary challenge, however, the acute phase (day 8) of the DbNP366+CD8+ T cell response is massively immunodominant, constituting up to 80% of the total virus-specific CD8+ T cell pool (18, 21). Although this hierarchy is readily observed at the acute phase of the secondary infection (day 8), the difference in magnitude between the DbNP366+CD8+ and DbPA224+CD8+-specific T cell responses is reduced in long-term secondary memory, as DbNP366+CD8+ T cells appear to decline at a higher rate than the DbPA224+CD8+ T cell set in the memory phase. By day 274, both DbNP366+CD8+ and DbPA224+CD8+ T cell memory pools appear to be of comparable magnitude (Fig. 1 A).

FIGURE 1.

CD62L and CD44 staining of CD8+DbNP366+ and CD8+DbPA224+ T cells following secondary influenza A virus challenge. A, Frequency for DbNP366+ and DbPA224+ T cells within the total CD8+ set. B, Percentage of CD62Lhigh T cells within the tetramer+ populations. C and D, Representative FACS plots (gated on the CD8+ T cells) from the acute (day 10) or long-term memory (day 274) phases after secondary i.n. challenge of PR8-primed mice with the HKx31 influenza A virus. Tetramer+CD8+ T cells were costained with (C) CD62L or (D) CD44. Percentages in the right upper quadrants represent (C) CD62Lhigh set of tetramer+CD8+ T cells and (D) proportion of tetramer+CD8+ T cells. Lymphocytes were obtained from spleens of individual mice (n = 3–5) at days 6–432 after secondary influenza virus challenge. Enriched CD8+ T cells were stained with the DbNP366-PE or DbPA224-PE tetramers, anti-CD8α-PerCP, anti-CD62L-FITC, and anti-CD44-APC. Statistically significant (∗, p < 0.01; #, p < 0.05) differences are shown.

FIGURE 1.

CD62L and CD44 staining of CD8+DbNP366+ and CD8+DbPA224+ T cells following secondary influenza A virus challenge. A, Frequency for DbNP366+ and DbPA224+ T cells within the total CD8+ set. B, Percentage of CD62Lhigh T cells within the tetramer+ populations. C and D, Representative FACS plots (gated on the CD8+ T cells) from the acute (day 10) or long-term memory (day 274) phases after secondary i.n. challenge of PR8-primed mice with the HKx31 influenza A virus. Tetramer+CD8+ T cells were costained with (C) CD62L or (D) CD44. Percentages in the right upper quadrants represent (C) CD62Lhigh set of tetramer+CD8+ T cells and (D) proportion of tetramer+CD8+ T cells. Lymphocytes were obtained from spleens of individual mice (n = 3–5) at days 6–432 after secondary influenza virus challenge. Enriched CD8+ T cells were stained with the DbNP366-PE or DbPA224-PE tetramers, anti-CD8α-PerCP, anti-CD62L-FITC, and anti-CD44-APC. Statistically significant (∗, p < 0.01; #, p < 0.05) differences are shown.

Close modal

An inverse relationship between CD62Lhigh phenotype and the size of the virus-specific CD8+ T cell pool is apparent for both secondary responses from the acute phase (days 6–8) into “established” (day 58) memory (Fig. 1,A–C). This reproduces (although at much greater magnitude) the situation following primary influenza virus exposure (8). Although all effector and memory phase DbNP366+CD8+ and DbPA224+CD8+ T cells had the “Ag-experienced” CD44high phenotype (Fig. 1,D), the smaller DbPA224+CD8+ set tended to contain more CD62Lhigh representatives than seen in the DbNP366+CD8+ set throughout this long time course. Thus, the proportion of CD62Lhigh cells in the DbPA224+CD8+- and DbNP366+CD8+-specific responses was 15.9 vs 9.3% on day 8, 32.5 vs 11.5% on day 58, and 75.2 vs 64.2% on day 432, respectively (Fig. 1 B). Furthermore, as these responses progress to the memory phase, the rate at which the CD62Lhigh proportion of the virus-specific T cell population increases is slower for both the DbNP366- and DbPA224-specific “secondary” T cells than is the case following primary infection (8).

Previous work characterized TCR repertoires of DbNP366+CD8+ and DbPA224+CD8+ T cell populations and found that these two epitope-specific CD8+ T cell populations used TCR repertoires of distinct clonotypic characteristics. Although the DbNP366+Vβ8.3+CD8+ response is restricted and characterized by a substantial proportion of TCRs that are public (present in most mice) (14, 22, 23), the DbPA224+Vβ7+CD8+ response is more diverse and private (specific to individual mice) (13, 24). Although “recurrent” (present in more than one mouse) DbPA224+Vβ7+CD8+ sequences can be found when extensive numbers of mice are tested (25), the more diverse nature of the DbPA224+Vβ7+CD8+ response makes the sampling of recurrent TCR sequences less likely.

Analysis of the CD62Llow and CD62Lhigh subsets of the DbNP366+CD8+ and DbPA224+CD8+ responses during primary infection (8) showed that many of the large, clonally dominant TCRs (predominantly public or recurrent TCRs) were present in both the CD62Llow and CD62Lhigh subsets. However, the repertoire diversity was found to be substantially higher within the CD62Lhigh population, primarily due to a broad spectrum of clonotypes unique to the CD62Lhigh subsets that were present in very low numbers (8). Thus, it was shown that while there is some commonality between the CD62Llow and CD62Lhigh subsets in a primary immune response, there are also some apparent differences. We have studied the clonotypic composition of CD62Llow and CD62Lhigh subsets for DbNP366+CD8+ and DbPA224+CD8+ T cells during secondary acute and long-term memory phases of infection. DbNP366+Vβ8.3+CD8+ and DbPA224+Vβ7+CD8+ T cell populations from the acute secondary response (day 8) or long-term secondary memory (≥day 300) were single-cell sorted as either CD62Lhigh or CD62Llow T cell sets. The clonotypic composition and the extent of TCR diversity in both subsets were assessed by cDNA expansion using RT-PCR and a CDR3β sequencing approach. The present analysis of CDR3β clonotypes for CD62Llow and CD62Lhigh T cells during acute and long-term memory phases of the secondary response consists of 1638 sequences. CDR3β analysis is summarized for all DbNP366+Vβ8.3+CD8+ samples tested from day 8 to day 510 (Table I), while only representative mice for acute (day 8) and long-term memory (day 300) time points are shown for DbPA224+Vβ7+CD8+ T cells due to the great diversity and fewer recurrent clonotypes of the underlying repertoire (Tables II and III).

Table I.

Frequency of TCRβ public/recurrent and unique amino acid sequences in the CD62Llow and CD62Lhigh sets of DbNP366+Vβ8.3+CD8+ T cells during secondary response to influenza virus infection

CDR3β RegionaaFrequency (%)a
Day 8 (M1)Day 8 (M2)Day 8 (M3)Day 300 (M4)Day 360 (M5)Day 480 (M6)Day 510 (M7)
LowHighLowHighLowHighLowHighLowHighLowHighLowHigh
Public/recurrent                 
 SGGSNTGQL 2S2 92 85 55 51 66 61 9.2 10 3.1 96 47 47 50 
 SGGANTGQL 2S2 1.6 1.5 9.9 32 1.6  4.2 5.3 14 14    11 
 SGGGNTGQL 2S2      3.1      42 23 
 RGGANTGQL 2S2   2.8 11      1.5  2.1   
 RGGGNTGQL 2S2         4.1 1.5     
 KGGANTGQL 2S2 1.6  28 4.1 1.6 4.7 49 53       
 KGGGNTGQL 2S2     14 6.3         
 KGGSNTGQL 2S2  1.5 2.8     1.3    2.1   
 KGGGGTGQL 2S2       5.6 6.6       
 SARTANTEV 1S1 4.7 12  2.7 1.6 4.7         
 SDAANTEV 1S1     16 17         
 SDSANTEV 1S1              4.5 
Unique                 
 GGGANTGQL 2S2   1.4            
 GGGSNTGQL 2S2      1.6         
 SSDHRNTEV 1S1         63 80     
 SDARQTEV 1S1         8.2      
 SDATTTEV 1S1            38   
 SDVGTVSNERL 1S2 11            2.1   
 SRRDRGGNTL 1S3 10             11 6.8 
 DGNRGGNTL 1S3              2.3 
 RKGVNSDY 1S4 11            2.1   
 SDWGGYAEQ 2S1      1.5         
 SDARGNQDTQ 2S5 10            2.1   
 SDVYEQ 2S6       34 25       
 SDGGEQ 2S6            2.1   
 SERQGGEQ 2S6            2.1   
 SEGGPYEQ 2S6              2.3 
Total clones   64 65 71 73 64 64 71 76 49 65 24 47 19 44 
CDR3β RegionaaFrequency (%)a
Day 8 (M1)Day 8 (M2)Day 8 (M3)Day 300 (M4)Day 360 (M5)Day 480 (M6)Day 510 (M7)
LowHighLowHighLowHighLowHighLowHighLowHighLowHigh
Public/recurrent                 
 SGGSNTGQL 2S2 92 85 55 51 66 61 9.2 10 3.1 96 47 47 50 
 SGGANTGQL 2S2 1.6 1.5 9.9 32 1.6  4.2 5.3 14 14    11 
 SGGGNTGQL 2S2      3.1      42 23 
 RGGANTGQL 2S2   2.8 11      1.5  2.1   
 RGGGNTGQL 2S2         4.1 1.5     
 KGGANTGQL 2S2 1.6  28 4.1 1.6 4.7 49 53       
 KGGGNTGQL 2S2     14 6.3         
 KGGSNTGQL 2S2  1.5 2.8     1.3    2.1   
 KGGGGTGQL 2S2       5.6 6.6       
 SARTANTEV 1S1 4.7 12  2.7 1.6 4.7         
 SDAANTEV 1S1     16 17         
 SDSANTEV 1S1              4.5 
Unique                 
 GGGANTGQL 2S2   1.4            
 GGGSNTGQL 2S2      1.6         
 SSDHRNTEV 1S1         63 80     
 SDARQTEV 1S1         8.2      
 SDATTTEV 1S1            38   
 SDVGTVSNERL 1S2 11            2.1   
 SRRDRGGNTL 1S3 10             11 6.8 
 DGNRGGNTL 1S3              2.3 
 RKGVNSDY 1S4 11            2.1   
 SDWGGYAEQ 2S1      1.5         
 SDARGNQDTQ 2S5 10            2.1   
 SDVYEQ 2S6       34 25       
 SDGGEQ 2S6            2.1   
 SERQGGEQ 2S6            2.1   
 SEGGPYEQ 2S6              2.3 
Total clones   64 65 71 73 64 64 71 76 49 65 24 47 19 44 
a

M indicates individual mouse; days indicated are after secondary challenge.

Table II.

Frequency of TCRβ amino acie sequences in the CD62Llow and CD62Lhigh sets of DbPA224+Vβ7+CD8+ T cells at the peak (day 8) of acute secondary responsea

CDR3β RegionaaCD62Llow Frequency (%)CD62Lhigh Frequency (%)
SPDRGEV 1S1 6.3 1.6 
SLGGEV 1S1 4.8 0.0 
SAGTEV 1S1 3.2 1.6 
SSGAEV 1S1 1.6 3.1 
SLGDRL 1S4 17.5 7.8 
SSDRGRL 1S4 4.8 4.7 
SFGQAP 1S5 9.5 17.2 
SAGEAP 1S5 6.3 4.7 
SLGEAP 1S5 3.2 4.7 
SEGRAP 1S5 3.2 0.0 
SPGQGDYAEQ 2S1 10 1.6 4.7 
SWGAEQ 2S1 1.6 1.6 
SSPDTQ 2S5 9.5 4.7 
SLGGEQ 2S6 7.9 10.9 
SSGGEQ 2S6 1.6 7.8 
SEYEQ 2S6 1.6 4.7 
ARGGYEQ 2S6 1.6 1.6 
SSYEQ 2S6 1.6 1.6 
SLGTGEV 1S1 1.6  
TGGSDY 1S2 1.6  
SLSWDRGREV 1S4 10 1.6  
SSGETL 2S3 1.6  
SWGDTQ 2S5 1.6  
SQGIEQ 2S6 1.6  
STGGEQ 2S6 1.6  
SWGDEQ 2S6 1.6  
SWDRGQV 1S1  3.1 
STSTEV 1S1  1.6 
SLSYRGPNSDY 1S2 11  1.6 
SFGAEQ 2S1  1.6 
SSPAEQ 2S1  1.6 
SSDWGSQNTL 2S3 10  1.6 
SLTKDTQ 2S5  1.6 
SWDRGEQ 2S6  1.6 
SWGDGEQ 2S6  1.6 
SWGGEQ 2S6  1.6 
Total clones   63 64 
CDR3β RegionaaCD62Llow Frequency (%)CD62Lhigh Frequency (%)
SPDRGEV 1S1 6.3 1.6 
SLGGEV 1S1 4.8 0.0 
SAGTEV 1S1 3.2 1.6 
SSGAEV 1S1 1.6 3.1 
SLGDRL 1S4 17.5 7.8 
SSDRGRL 1S4 4.8 4.7 
SFGQAP 1S5 9.5 17.2 
SAGEAP 1S5 6.3 4.7 
SLGEAP 1S5 3.2 4.7 
SEGRAP 1S5 3.2 0.0 
SPGQGDYAEQ 2S1 10 1.6 4.7 
SWGAEQ 2S1 1.6 1.6 
SSPDTQ 2S5 9.5 4.7 
SLGGEQ 2S6 7.9 10.9 
SSGGEQ 2S6 1.6 7.8 
SEYEQ 2S6 1.6 4.7 
ARGGYEQ 2S6 1.6 1.6 
SSYEQ 2S6 1.6 1.6 
SLGTGEV 1S1 1.6  
TGGSDY 1S2 1.6  
SLSWDRGREV 1S4 10 1.6  
SSGETL 2S3 1.6  
SWGDTQ 2S5 1.6  
SQGIEQ 2S6 1.6  
STGGEQ 2S6 1.6  
SWGDEQ 2S6 1.6  
SWDRGQV 1S1  3.1 
STSTEV 1S1  1.6 
SLSYRGPNSDY 1S2 11  1.6 
SFGAEQ 2S1  1.6 
SSPAEQ 2S1  1.6 
SSDWGSQNTL 2S3 10  1.6 
SLTKDTQ 2S5  1.6 
SWDRGEQ 2S6  1.6 
SWGDGEQ 2S6  1.6 
SWGGEQ 2S6  1.6 
Total clones   63 64 
a

A representative mouse is shown.

Table III.

Frequency of TCRβ amino acid sequences in the CD62Llow and CD62Lhigh sets of DbPA224+Vβ7+CD8+ T cells at memory phase (day 300) of secondary responsea

CDR3β RegionaaCD62Llow Frequency (%)CD62Lhigh Frequency (%)
SLDRAEV 1S1 1.3 1.4 
SLGGFEV 1S1 3.9 4.1 
SWDRGLV 1S1 9.1 11.0 
SWDRGRV 1S1 13.0 9.6 
SPDRGRV 1S1 3.9 2.7 
AGPSEV 1S1 6.5 1.4 
SFGERL 1S4 3.9 4.1 
SPGQAP 1S5 1.3 2.7 
SDGQAP 1S5 9.1 6.8 
TGGAEQ 2S1 1.3 2.7 
TSGTGQL 2S2 6.5 1.4 
TNTGQL 2S2 2.6 1.4 
SYSGSAETL 2S3 1.3 1.4 
SWGGEQ 2S6 3.9 2.7 
TSGGEQ 2S6 2.6 1.4 
SSYEQ 2S6 3.9 11.0 
SLGDEQ 2S6 1.3 1.4 
SLGGQQ 2S6 1.3 1.4 
SLGGEQ 2S6 1.3  
SLDRGEV 1S1 1.3  
SLDRGVW 1S1 1.3  
SWDRGEV 1S1 2.6  
SSTVPME 2S6 1.3  
SPDRGEQ 2S6 5.2  
TEGGEQ 2S6 1.3  
ALGGYEQ 2S6 8.0  
SFGNEV 1S1 1.3  
SSLDRGVW 1S1  2.7 
SLGGRV 1S4  1.4 
STGQAP 1S5  1.4 
EGGPNYAEQ 2S1  1.4 
SFGGQL 2S2  12.3 
HSAETL 2S3  1.4 
SWGNEQ 2S6  1.4 
SWGDEQ 2S6  1.4 
SWGSEQ 2S6  1.4 
TGGAEQ 2S6  1.4 
SLAGYEQ 2S6  2.7 
SLGGYEQ 2S6  1.4 
SRGDEQ 2S6  1.4 
Total clones   77 73 
CDR3β RegionaaCD62Llow Frequency (%)CD62Lhigh Frequency (%)
SLDRAEV 1S1 1.3 1.4 
SLGGFEV 1S1 3.9 4.1 
SWDRGLV 1S1 9.1 11.0 
SWDRGRV 1S1 13.0 9.6 
SPDRGRV 1S1 3.9 2.7 
AGPSEV 1S1 6.5 1.4 
SFGERL 1S4 3.9 4.1 
SPGQAP 1S5 1.3 2.7 
SDGQAP 1S5 9.1 6.8 
TGGAEQ 2S1 1.3 2.7 
TSGTGQL 2S2 6.5 1.4 
TNTGQL 2S2 2.6 1.4 
SYSGSAETL 2S3 1.3 1.4 
SWGGEQ 2S6 3.9 2.7 
TSGGEQ 2S6 2.6 1.4 
SSYEQ 2S6 3.9 11.0 
SLGDEQ 2S6 1.3 1.4 
SLGGQQ 2S6 1.3 1.4 
SLGGEQ 2S6 1.3  
SLDRGEV 1S1 1.3  
SLDRGVW 1S1 1.3  
SWDRGEV 1S1 2.6  
SSTVPME 2S6 1.3  
SPDRGEQ 2S6 5.2  
TEGGEQ 2S6 1.3  
ALGGYEQ 2S6 8.0  
SFGNEV 1S1 1.3  
SSLDRGVW 1S1  2.7 
SLGGRV 1S4  1.4 
STGQAP 1S5  1.4 
EGGPNYAEQ 2S1  1.4 
SFGGQL 2S2  12.3 
HSAETL 2S3  1.4 
SWGNEQ 2S6  1.4 
SWGDEQ 2S6  1.4 
SWGSEQ 2S6  1.4 
TGGAEQ 2S6  1.4 
SLAGYEQ 2S6  2.7 
SLGGYEQ 2S6  1.4 
SRGDEQ 2S6  1.4 
Total clones   77 73 
a

A representative mouse is shown.

Following the secondary challenge and the consequent increase in Ag-specific T cell numbers (Fig. 1,A), the large, public and recurrent DbNP366+Vβ8.3+CD8+ clonotypes (8, 14) were usually found in both the CD62Llow and CD62Lhigh sets (M1–M3, Table I) in a manner reminiscent of the primary response (8). This relationship persisted into secondary long-term memory (M4–M7, Table I). In primary infection the CD62Lhigh subset contained many “private” TCRs (present in only one mouse), which were not found in the CD62Llow subset (8)). In secondary infection, such private TCRs were only detected in two of the seven mice after the recall in the acute secondary CD62Lhigh population. In these two long-term secondary memory mice, private clonotypes could be found mostly in the CD62Lhigh subset (M6 at day 480 with six unique clonotypes and M7 at day 510 with three unique clonotypes; Table I). Large, private clonotypes were sometimes present in other animals, but in these cases were mostly found in both CD62Llow and CD62Lhigh subsets (e.g., SSDHRNTEV, day 360, Table I). This loss of diversity in the CD62Lhigh compartment led to a greater similarity and sharing of TCR between the CD62Lhigh and CD62Llow compartments.

In primary infection, the DbPA224+Vβ7+CD8+ T cell responses were more diverse than the DbNP366+Vβ8.3+CD8+ responses, but they showed the same characteristic of sharing of large clonotypes between the CD62Lhigh and CD62Llow compartments, as well as many more private clones in the CD62Lhigh compartment than in the CD62Llow compartment. In secondary infection, most high-frequency clonotypes for diverse DbPA224+Vβ7+CD8+ T cell responses were also found in both CD62Llow and CD62Lhigh subsets (Tables II and III). However, the proportion of clonotypes shared between CD62Llow and CD62Lhigh subsets in the secondary recall response appears to be higher than the proportion of clonotypes shared between these two sets in the primary response (8). Analysis of nonshared clonotypes during the recall response reveals that they are equally distributed between CD62Llow and CD62Lhigh sets and represent predominantly low-frequency clonotypes with typical TCR characteristics. This is in contrast to the primary response, in which nonshared clonotypes preferentially were found in the CD62Lhigh population.

For statistical purposes, data were analyzed according to the proportion of clonotypes (distinct TCR sequences irrespective of size) or TCR repertoires (number of copies of particular clonotypes) that were shared between CD62Llow and CD62Lhigh subsets, (i.e., CD62Lhigh sets shared with CD62Llow populations, and CD62Llow sets shared with CD62Lhigh subsets) (Fig. 2). Assessment of both DbNP366+Vβ8.3+CD8+ and DbPA224+Vβ7+CD8+ T cells clearly showed that there was no difference in the proportion of shared TCR repertoires (Fig. 2,A–C) or clonotypes (Fig. 2,D–F) between CD62Llow and CD62Lhigh subsets during the peak of acute (day 8) or memory (day ≥ 300) phases of the secondary response. At least 70% of the total DbNP366+CD8+ and DbPA224+CD8+ T cell populations were represented in both the CD62Lhigh and CD62Llow subsets (Fig. 2,C). Looking instead at the number of different CDR3β signatures, an analysis that does not take account of clone size, the prevalence of shared (between CD62Lhigh and CD62Llow) clonotypes was still ∼50% for both epitope-specific T cell sets (Fig. 2,F). Lack of any significant difference in sharing between the CD62Lhigh and CD62Llow populations suggests that the same TCR signatures were recalled in both subsets and subsequently survived into secondary long-term memory. Interestingly, recall of the same TCR signatures within secondary CD62Llow and CD62Lhigh subsets did not alter pMHC-TCR avidity of influenza-specific CD8+ T cells. Similar results were observed for pMHC-TCR avidity of secondary DbNP366+CD8+ (Fig. 3,A) and DbPA224+CD8+ (Fig. 3 B) sets when compared with primary populations. Thus, homogenization of TCR repertoires within secondary CD62Llow and CD62Lhigh CD8+ T cell populations does not result in pMHC-TCR avidity maturation.

FIGURE 2.

Shared TCRβ signatures between CD62Lhigh and CD62Llow CD8+DbNP366+ and CD8+ DbPA224+ T cell sets. The proportion of shared (A–C) TCR repertoires and (D and E) clonotypes that are represented in both the CD62Lhigh and CD62Llow subsets are shown for CD8+Vβ8.3+DbNP336+ and CD8+ Vβ7+DbPA224+ T cells from mice sampled at acute (day 8) or long-term memory (≥day 300) phases after secondary challenge. Clonotypes define distinct TCR sequences (irrespective of their size; i.e., each amino acid sequence is counted as “1”), whereas TCR repertoires take into account a number of times that particular clonotypes appear (i.e., if a clonotype was found 10 times, it is given a value of “10”). (A and D) DbNP336+Vβ8.3+CD8+and (B and E) DbPA224+Vβ7+CD8+ T cells from individual mice are shown. C and F, The results shown are expressed as cumulated data (mean ± SD). A–C, Clone size was taken into account to show the proportion of TCR repertoires that are shared between the CD62Lhigh and CD62Llow subsets, that is, CD62Llow sets shared with CD62Lhigh populations and CD62Lhigh sets shared with CD62Llow subsets. D–F, Each CDR3β sequence is represented as an equal value irrespective of size. Lymphocytes obtained from spleens of influenza-infected mice were stained with either the DbNP366-PE or DbPA224-PE tetramer and mAbs against CD8 (APC-Cy7), CD62L (APC), and Vβ8.3 or Vβ7 (FITC). Single CD62Lhigh or CD62Llow tetramer++CD8+ T cells were sorted into 96-well plates, amplified for CDR3β region by nested PCR, and sequenced.

FIGURE 2.

Shared TCRβ signatures between CD62Lhigh and CD62Llow CD8+DbNP366+ and CD8+ DbPA224+ T cell sets. The proportion of shared (A–C) TCR repertoires and (D and E) clonotypes that are represented in both the CD62Lhigh and CD62Llow subsets are shown for CD8+Vβ8.3+DbNP336+ and CD8+ Vβ7+DbPA224+ T cells from mice sampled at acute (day 8) or long-term memory (≥day 300) phases after secondary challenge. Clonotypes define distinct TCR sequences (irrespective of their size; i.e., each amino acid sequence is counted as “1”), whereas TCR repertoires take into account a number of times that particular clonotypes appear (i.e., if a clonotype was found 10 times, it is given a value of “10”). (A and D) DbNP336+Vβ8.3+CD8+and (B and E) DbPA224+Vβ7+CD8+ T cells from individual mice are shown. C and F, The results shown are expressed as cumulated data (mean ± SD). A–C, Clone size was taken into account to show the proportion of TCR repertoires that are shared between the CD62Lhigh and CD62Llow subsets, that is, CD62Llow sets shared with CD62Lhigh populations and CD62Lhigh sets shared with CD62Llow subsets. D–F, Each CDR3β sequence is represented as an equal value irrespective of size. Lymphocytes obtained from spleens of influenza-infected mice were stained with either the DbNP366-PE or DbPA224-PE tetramer and mAbs against CD8 (APC-Cy7), CD62L (APC), and Vβ8.3 or Vβ7 (FITC). Single CD62Lhigh or CD62Llow tetramer++CD8+ T cells were sorted into 96-well plates, amplified for CDR3β region by nested PCR, and sequenced.

Close modal
FIGURE 3.

TCR-pMHC avidity of primary and secondary DbNP336+CD8+ and DbPA224+CD8+ T cell responses. Tetramer dissociation was assessed as a measure of TCR avidity for pMHC. Splenocytes obtained from mice at the acute primary (day 10) and secondary (day 8) influenza virus infection were stained with either (A) DbNP336 or (B) DbPA224 tetramers and subsequently incubated with anti-H2Db Ab to prevent rebinding of dissociated tetramer. Loss of tetramer staining was assessed at the indicated time points. Data represent means ± SD of 4–5 mice/group.

FIGURE 3.

TCR-pMHC avidity of primary and secondary DbNP336+CD8+ and DbPA224+CD8+ T cell responses. Tetramer dissociation was assessed as a measure of TCR avidity for pMHC. Splenocytes obtained from mice at the acute primary (day 10) and secondary (day 8) influenza virus infection were stained with either (A) DbNP336 or (B) DbPA224 tetramers and subsequently incubated with anti-H2Db Ab to prevent rebinding of dissociated tetramer. Loss of tetramer staining was assessed at the indicated time points. Data represent means ± SD of 4–5 mice/group.

Close modal

Comparing the TCR repertoires of the CD62Llow and CD62Lhigh sets between the primary (8) and secondary responses, there are apparent differences between the primary and recall responses. The extent of sharing between the CD62Llow and CD62Lhigh sets is substantially greater in the secondary response than in the primary response and there are no significant differences between the diversities of the CD62Llow and CD62Lhigh sets in the secondary responses, unlike what was observed for the primary responses in which the CD62Lhigh set was significantly more diverse (Figs. 4,B and 5 B). What, then, is the mechanism underlying homogenization of TCR repertoires in CD62Llow and CD62Lhigh sets after the secondary challenge?

FIGURE 4.

Homogenization of the CD62Lhigh and CD62Llow subsets of the DbNP366+CD8+ TCR repertoire from primary to secondary responses via focusing of the CD62Lhigh subset. A, Schematic illustrating the changes in the CD62Lhigh and CD62Llow subsets between primary and secondary responses for DbNP366+CD8+ T cells characterized by a restricted TCR repertoire, and (B) a table with statistical comparisons of the various subsets. The numerical values in B represent the mean Simpson’s diversity indices and the mean Morisita-Horn similarity indices, averaged across all samples for each of the TCR subsets. The p values for significant differences are shown, with CD62Lhigh and CD62Llow subsets compared using a Wilcoxon test and primary and secondary responses compared using a Mann-Whitney U test. In the primary responses, the CD62Lhigh subsets are significantly more diverse than the CD62Llow subsets, with some similarity between the two subsets. In the secondary responses, there is a homogenization between the CD62Llow and CD62Lhigh subsets, as indicated by an increase in the similarity between the two subsets and no significant difference in the diversities of the two sets. The homogenization arises from the CD62Lhigh subsets becoming significantly less diverse and mostly losing TCRs that are not common between the CD62Llow and CD62Lhigh subsets. There is no significant change in the diversity of the CD62Llow subsets between primary and secondary responses.

FIGURE 4.

Homogenization of the CD62Lhigh and CD62Llow subsets of the DbNP366+CD8+ TCR repertoire from primary to secondary responses via focusing of the CD62Lhigh subset. A, Schematic illustrating the changes in the CD62Lhigh and CD62Llow subsets between primary and secondary responses for DbNP366+CD8+ T cells characterized by a restricted TCR repertoire, and (B) a table with statistical comparisons of the various subsets. The numerical values in B represent the mean Simpson’s diversity indices and the mean Morisita-Horn similarity indices, averaged across all samples for each of the TCR subsets. The p values for significant differences are shown, with CD62Lhigh and CD62Llow subsets compared using a Wilcoxon test and primary and secondary responses compared using a Mann-Whitney U test. In the primary responses, the CD62Lhigh subsets are significantly more diverse than the CD62Llow subsets, with some similarity between the two subsets. In the secondary responses, there is a homogenization between the CD62Llow and CD62Lhigh subsets, as indicated by an increase in the similarity between the two subsets and no significant difference in the diversities of the two sets. The homogenization arises from the CD62Lhigh subsets becoming significantly less diverse and mostly losing TCRs that are not common between the CD62Llow and CD62Lhigh subsets. There is no significant change in the diversity of the CD62Llow subsets between primary and secondary responses.

Close modal
FIGURE 5.

Homogenization of the CD62Lhigh and CD62Llow subsets of the DbPA224+CD8+ TCR repertoire from primary to secondary responses via diversification of the CD62Llow subset. A, Schematic illustrating the changes in the CD62Lhigh and CD62Llow subsets between primary and secondary responses for DbPA224+CD8+ T cells characterized by a diverse TCR repertoire, and (B) a table with statistical comparisons (as described for Fig. 4 B) of the various subsets. In the primary responses, the CD62Lhigh subsets are significantly more diverse than the CD62Llow subsets. In the secondary responses, there is a homogenization of the CD62Llow and CD62Lhigh subsets, as indicated by a significant increase in the similarity between the two subsets and no significant difference in the diversities of the two sets. The homogenization arises from the CD62Llow subsets becoming more diverse (the small but not significant p value is shown in italics) and including TCRs that are common between the CD62Llow and CD62Lhigh subsets. There is no significant change in the diversity of the CD62Lhigh subsets between primary and secondary responses.

FIGURE 5.

Homogenization of the CD62Lhigh and CD62Llow subsets of the DbPA224+CD8+ TCR repertoire from primary to secondary responses via diversification of the CD62Llow subset. A, Schematic illustrating the changes in the CD62Lhigh and CD62Llow subsets between primary and secondary responses for DbPA224+CD8+ T cells characterized by a diverse TCR repertoire, and (B) a table with statistical comparisons (as described for Fig. 4 B) of the various subsets. In the primary responses, the CD62Lhigh subsets are significantly more diverse than the CD62Llow subsets. In the secondary responses, there is a homogenization of the CD62Llow and CD62Lhigh subsets, as indicated by a significant increase in the similarity between the two subsets and no significant difference in the diversities of the two sets. The homogenization arises from the CD62Llow subsets becoming more diverse (the small but not significant p value is shown in italics) and including TCRs that are common between the CD62Llow and CD62Lhigh subsets. There is no significant change in the diversity of the CD62Lhigh subsets between primary and secondary responses.

Close modal

Assessment of DbNP366+CD8+ T cell responses (characterized by restricted and public TCR repertoires) showed that both TCR diversity and the extent of sharing were comparable between the primary and secondary CD62Llow subsets, while the TCR repertoires of the secondary CD62Lhigh populations were significantly less diverse than those of the primary CD62Lhigh sets (Fig. 4,B). Thus, the homogenization of TCR repertoires arises from the CD62Lhigh subsets becoming significantly less diverse than in primary infection and mostly losing TCRs that are not common between the CD62Llow and CD62Lhigh subsets (Fig. 4 A).

Dissection of CD62Llow and CD62Lhigh subsets for DbPA224+CD8+ T cell responses (characterized by diverse and private TCR repertoire) showed that, like the DbNP366+CD8+ T cell responses, during primary infection the CD62Lhigh subsets were significantly more diverse than the CD62Llow subsets (Fig. 5,B). Following the secondary challenge, the CD62Llow and CD62Lhigh subsets homogenize, with the similarity between the two subsets increasing significantly (Fig. 5,B). However, in the case of DbPA224+CD8+ T cells, the homogenization arises from the CD62Llow subsets becoming more diverse and including TCRs that are common between the CD62Llow and CD62Lhigh subsets (Fig. 5 A). Although the CD62Lhigh and CD62Llow subsets in the DbPA224+CD8+ T cell recall response are significantly more similar than in the primary response, the similarity between the CD62Lhigh and CD62Llow subsets in both primary and secondary responses is less for the DbPA224+CD8+ secondary responses than for the primary DbNP366+CD8+ T cell responses. This is probably due, in part, to the more diverse nature of the response to the DbPA224 epitope. However, it is evident for the DbPA224+CD8+ T cell response that there is a subset of small TCR clonotypes that are exclusive to each of the CD62Lhigh and CD62Llow subsets.

Our findings of TCR repertoire homogenization within secondary CD62Lhigh and CD62LlowCD8+ T cell populations suggest that these two subsets arise from the same precursor population, most likely the CD62Lhigh central memory compartment. This is particularly evident for CD62LlowDbPA224+CD8+ sets. Because increased clonotypic diversity was found in the secondary CD62LlowDbPA224+CD8+ subset compared with the primary CD62LlowDbPA224+CD8+ population, secondary CD62LlowDbPA224+CD8+ T cells could not be recalled from the primary memory CD62LlowDbPA224+CD8+ set per se.

There is a possibility that some of the secondary CD62Llow TCR signatures could originate from clonotypes either trafficking from other anatomical compartments or naive precursors primed after the secondary challenge (26). To address the possibility that clonal diversity within the secondary CD62Llow set could originate from other anatomical compartments, we analyzed clonotypic composition within lungs at day 6, an early time point at which influenza-specific CD8+ T cells can be detected. At this time point, most (∼99%) DbNP366+CD8+ and DbPA224+CD8+ T cells in the lung are of CD62Llow phenotype (Fig. 6) and thus could potentially be a source of TCR signatures detected in spleen CD62Llow and CD62Lhigh sets at the peak (day 8) of secondary response (Tables I–III). TCR analysis of lungs and spleen CD62Llow and spleen CD62Lhigh compartments for DbPA224+CD8+ T cells on day 6 showed comparable levels of diversity for all three compartments (Table IV, results for a representative mouse; n = 3), in agreement with our previous data showing comparable clonotypic distribution between different anatomical compartments (13, 14). Thus, it is rather unlikely that TCR diversity within the secondary CD62Llow spleen population originates from lungs.

FIGURE 6.

Comparison of CD62Lhigh expression within lung and spleen at early (day 6) effector phase of secondary influenza infection. Lymphocytes were obtained from spleens and lungs of individual mice (n = 5) at day 6 after secondary influenza virus challenge. Enriched CD8+ T cells were stained with the DbNP366-PE or DbPA224-PE tetramers, anti-CD8α-PerCP, anti-CD62L-FITC, and anti-CD44-APC. Proportion of CD62Lhigh within DbNP366+CD44highCD8+ or DbPA224+CD44highCD8+ T cells on day 6 phase of secondary influenza A virus challenge is shown.

FIGURE 6.

Comparison of CD62Lhigh expression within lung and spleen at early (day 6) effector phase of secondary influenza infection. Lymphocytes were obtained from spleens and lungs of individual mice (n = 5) at day 6 after secondary influenza virus challenge. Enriched CD8+ T cells were stained with the DbNP366-PE or DbPA224-PE tetramers, anti-CD8α-PerCP, anti-CD62L-FITC, and anti-CD44-APC. Proportion of CD62Lhigh within DbNP366+CD44highCD8+ or DbPA224+CD44highCD8+ T cells on day 6 phase of secondary influenza A virus challenge is shown.

Close modal
Table IV.

Comparison of DbPA224+Vβ7+CD8+ TCR amino acid sequences between lungs, CD62Llow spleen, and CD62Lhigh spleen compartments at the early effector phase (day 6) of the secondary responsea

CDR3β RegionaaLungsSpleen CD62LlowSpleen CD62Lhigh
SPDRGEV 1S1 1.6 3.1 2.9 
SFGGGV 1S1 3.2 1.5 4.4 
SLGKEV 1S1 12.9 12.3 30.9 
SLGGEV 1S1 1.6 1.5 1.5 
SSGERL 1S4 11.3 26.2 5.9 
SVGQAP 1S5 1.6 3.1 1.5 
HSYAEQ 2S1 14.5 10.8 2.9 
SYGGEQ 2S6 3.2 1.5 1.5 
TTGGEQ 2S6 9.7 10.8 14.7 
GGTGGYEQ 2S6 1.6 1.5 1.5 
SPDRGEQ 2S6 3.2 3.1 7.4 
SLGGEQ 2S6 1.6 4.6 1.5 
SDGTEV 1S1 12.9 7.7  
SLSGYEQ 2S6 11.3 4.6  
SLDRGEV 1S1 4.8  5.9 
SDGNCAEV 1S1 1.6   
SSPETL 2S3 1.6   
SFGGEL 2S6 1.6   
SSPTKQ 2S5  1.5 1.5 
SLGDEQ 2S6  1.5 4.4 
SFHTEV 1S1  1.5  
SIDRGEV 1S1  1.5  
SPGEAP 1S5  1.5  
SSGGGV 1S1   1.5 
SLSYRDANSDY 1S2 10   1.5 
SRGERL 1S4   2.9 
SGGTNNQAP 1S5 10   1.5 
TGGEAP 1S5   2.9 
TGGAEQ 2S1   1.5 
Total clones   62 68 65 
CDR3β RegionaaLungsSpleen CD62LlowSpleen CD62Lhigh
SPDRGEV 1S1 1.6 3.1 2.9 
SFGGGV 1S1 3.2 1.5 4.4 
SLGKEV 1S1 12.9 12.3 30.9 
SLGGEV 1S1 1.6 1.5 1.5 
SSGERL 1S4 11.3 26.2 5.9 
SVGQAP 1S5 1.6 3.1 1.5 
HSYAEQ 2S1 14.5 10.8 2.9 
SYGGEQ 2S6 3.2 1.5 1.5 
TTGGEQ 2S6 9.7 10.8 14.7 
GGTGGYEQ 2S6 1.6 1.5 1.5 
SPDRGEQ 2S6 3.2 3.1 7.4 
SLGGEQ 2S6 1.6 4.6 1.5 
SDGTEV 1S1 12.9 7.7  
SLSGYEQ 2S6 11.3 4.6  
SLDRGEV 1S1 4.8  5.9 
SDGNCAEV 1S1 1.6   
SSPETL 2S3 1.6   
SFGGEL 2S6 1.6   
SSPTKQ 2S5  1.5 1.5 
SLGDEQ 2S6  1.5 4.4 
SFHTEV 1S1  1.5  
SIDRGEV 1S1  1.5  
SPGEAP 1S5  1.5  
SSGGGV 1S1   1.5 
SLSYRDANSDY 1S2 10   1.5 
SRGERL 1S4   2.9 
SGGTNNQAP 1S5 10   1.5 
TGGEAP 1S5   2.9 
TGGAEQ 2S1   1.5 
Total clones   62 68 65 
a

A representative mouse is shown. Results are expressed as frequency (%) of particular clonotypes.

Furthermore, clonal diversity within CD62Llow and CD62Lhigh spleen sets was comparable for days 6 (Table IV) and 8 (Tables II and III) after secondary challenge. Because negligible numbers of primary influenza-specific CD8+ T cells are found on day 6 (19), the majority of clonotypes detected on day 6 after secondary challenge are derived from primary memory. Similar diversity within day 6 and day 8 populations suggests that clonotypes within CD62Llow and CD62Lhigh recall responses originate predominantly from primary memory CD8+ T cells rather than naive precursors during the secondary challenge.

Memory CD8+ T cells provide significant protection against infections with intracellular pathogens, including influenza viruses (1), and thus are important targets for vaccine development. Although an enormous amount of research has been done using primary and memory CD8+ T cells, few studies have addressed the secondary effector or secondary memory CD8+ T cell populations, important in prime-and-boost vaccine regimens. Recent evidence suggests different phenotypic and functional characteristics of memory CD8+ T cells following primary or secondary infection (27, 28). Compared with primary memory cells, secondary CD8+ T cell populations are slower at the acquisition of the TCM phenotype, characterized by CD62Lhigh expression and production of IL-2, but express more TEM qualities such as CD62Llow expression, increased granzyme B expression, and far superior cytotoxic properties (27, 28). Given the different phenotypic and functional characteristics of primary and secondary memory CD8+ T cells, it is important to determine whether there are also differences between primary and secondary CD8+ TCR repertoires.

We assessed secondary CD62Lhigh and CD62Llow CD8+ T cells using endogenous C57BL/6J mice after respiratory challenge. An inverse relationship between CD62Lhigh phenotype and pool size was apparent for both the influenza virus-specific DbNP366+CD8+ and DbPA224+CD8+ T cells during secondary responses, with the smaller DbPA224+CD8+ set tending to contain more CD62Lhigh representatives than did the DbNP366+CD8+ set (Fig. 1 B). This epitope-specific difference could be a reflection of signal strength that, in turn, is a function of the spectrum of Ag-presenting cell availability (29) and/or Ag dose (30). Interestingly, the rate at which the proportion of CD62Lhigh virus-specific CD8+ T cell population increased during the memory phase was slower for both the DbNP366- and DbPA224-specific “secondary” T cells than was the case following primary infection (8). This pattern of delayed reemergence of CD62Lhigh prominence following secondary challenge is in accord with findings from experiments with vesicular stomatitis virus and lymphocytic choriomeningitis virus (27, 28). The mechanism for this is not clear, although the difference between naive and recall responses could be related to the extent of clonal expansion (28) and the duration of Ag stimulation (31). However, delayed reemergence of the CD62Lhigh phenotype during secondary challenge could be related to higher functionality of the cells (27, 28).

Superior functionality of secondary memory CD8+ T cells over primary memory CD8+ T cell sets (27, 28) might be also caused by selection of specific, possibly “better-fit” clonotypes within either CD62Llow or CD62Lhigh T cell subsets during the recall response. Our TCR repertoire analysis of secondary CD62Lhigh and CD62Llow populations revealed different results to those previously observed for primary infection (8). In the primary response, clonotypes distributed unequally between CD62Lhigh and CD62Llow sets. Although most clonotypes were represented in both the CD62Lhigh and CD62Llow sets, a small number of clonotypes were found exclusively in the CD62Lhigh population. Conversely, in the secondary responses the TCR repertoire composition and diversity were more comparable between the CD62Lhigh and CD62Llow pools. The results are true for two different epitope-specific populations, DbNP366+CD8+ and DbPA224+CD8+ T cells, despite their distinct recall capacities (21), functional quality (production of multiple cytokines) (15), pMHC-TCR avidity (16), and TCR repertoire characteristics (13, 14, 24). Although the same phenomenon of homogenization of TCR repertoires between CD62Llow and CD62Lhigh sets occurs for both influenza-specific epitopes upon secondary challenge, it is mediated by slightly different mechanisms: decreasing clonotypic diversity of CD62Lhigh subset in the DbNP366+CD8+ responses, and increasing clonotypic diversity of the CD62Llow subset in the DbPA224+CD8+ responses. The reasons for this difference are not clear; however, one possibility is that this might be due, at least in part, to the fact that DbPA224+CD8+ T cell responses (in contrast to DbNP366+CD8+ T cells) consist of a diverse array of high avidity clonotypes (16, 23), which in turn may promote easier selection of “optimal” clonotypes following secondary challenge. Homogenization of TCR repertoires within CD62Lhigh and CD62Llow sets does not lead to pMHC-avidity maturation for either DbNP366+CD8+ or DbPA224+CD8+ T cells.

The homogenization of the CD62Lhigh and CD62Llow subsets in both the DbNP366+CD8+ and the DbPA224+CD8+ T cell recall responses suggests that the CD62Lhigh and CD62Llow subsets primarily originate from the same precursors, most likely from primary memory CD62Lhigh clonotypes rather than from naive precursors or other anatomical compartments. Experimental evidence suggests that the CD62Lhigh compartment has greater proliferative potential and may thus be the dominant source of cell expansion feeding into both the CD62Lhigh and CD62Llow pools during the recall response (5, 11). Consistent with this we observed that during the secondary DbPA224+CD8+ T cell responses, the CD62Lhigh and CD62Llow compartments were more similar, suggesting that they originated from the same common (CD62Lhigh primary memory) pool of cells. Additionally, the CD62Llow compartment of the secondary DbPA224+CD8+ T cell response was more diverse than in primary infection, acquiring diversity similar to the CD62Lhigh compartment in the primary response. However, in the secondary DbNP366+CD8+ T cell responses the picture was less clear. The CD62Llow and CD62Lhigh subsets became significantly more similar than in primary infection (“homogenization”). However, this was accompanied by a decrease in the diversity of both subsets (which was statistically significant for the CD62Lhigh subset), with a focusing toward the dominant and public/recurrent TCR clonotypes. This is consistent with “clonal selection” (32) leading to a more focused secondary response. It is interesting to speculate why this phenomenon may be observed in the DbNP366+CD8+ T cell but not in the DbPA224+CD8+ T cell response. The most likely explanation is that, unlike the primary response and primary memory, where the DbNP366+CD8+- and DbPA224+CD8+-specific T cell responses are codominant, in the secondary response the DbNP366+CD8+-specific response is >10× the magnitude of the DbPA224+CD8+-specific response. Thus, the additional cell divisions required to expand the DbNP366+CD8+-specific response may have allowed for further clonal selection of the DbNP366+CD8+-specific repertoire.

Previous analysis of total epitope-specific T cells (not subdivided into CD62Lhigh and CD62Llow sets) in other experimental systems indicated narrowing and homogenization of TCR clonotypes during memory differentiation and subsequent secondary challenge (32, 33, 34), suggesting preferential selection of particular clonotypes upon recall. Conversely, we (13, 14, 24) and other longitudinal studies (35, 36, 37, 38) found clonal stability from the acute response through to long-term memory and recall. TCRβ repertoires for both influenza-specific epitopes displayed equivalent levels of clonotypic diversity at every time point during the primary or secondary infection (13, 14). The reasons for such discrepancies between studies showing either TCR repertoire narrowing or clonal stability after secondary challenge are unknown but could include selection of high TCR affinity/avidity clonotypes in some experimental systems and not others, or from the magnitude of clonal expansion in different systems, leading to differences in the effects of relatively minor growth advantages of some clonotypes. The novelty of our present study lies in the finding that TCR repertoires of subdivided CD62Lhigh and CD62Llow T cell pools homogenized upon secondary challenge without narrowing of the total TCR pool.

Taken together, in contrast to the primary response, where the CD62Lhigh set is significantly more diverse than the CD62Llow set and contains a number of small clonotypes not found in the CD62Llow set, the CD62Lhigh and CD62Llow sets generated after secondary challenge show no selective bias toward greater numbers of CD62Lhigh clonotypes. The homogenization of the repertoires suggests that the same T cell precursors feed into the CD62Lhigh and CD62Llow populations that respond to secondary Ag exposure, and it seems most likely that these precursors derived from the TCM subset. Thus, our study provides key insights into TCR composition of CD62Lhigh and CD62Llow T cells generated after secondary challenge. A better understanding of TCR selection and maintenance has implications for improved vaccine and immunotherapy protocols.

We thank Nicole La Gruta, John Stambas, Lukasz Kedzierski, Justine Mintern, and Carole Guillonneau for review of the manuscript, Ken Field for help with the FACS sorting, and Dina Stockwell for technical assistance.

The authors have no financial conflicts 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 an Australian National Health and Medical Research Council Burnet Fellowship and the National Institutes of Health Grant AI70251 awarded to P.C.D. and a National Health and Medical Research Council Project Grant 454312 awarded to K.K. S.J.T. is a Pfizer Senior Research Fellow, M.P.D. is a Sylvia and Charles Viertel Senior Medical Research Fellow, and K.K. is a National Health and Medical Research Council R. D. Wright Research Fellow.

3

Abbreviations used in this paper: TCM, central memory T cell; TEM, effector memory T cell; H, viral hemagglutinin molecule; N, viral neuraminidase; i.n., intranasally; PR8, A/PR/8/34 H1N1 influenza virus; HKx31, A/HKx31 H3N2 influenza virus.

1
Doherty, P. C., S. J. Turner, R. G. Webby, P. G. Thomas.
2006
. Influenza and the challenge for immunology.
Nat. Immunol.
7
:
449
-455.
2
Sallusto, F., D. Lenig, R. Forster, M. Lipp, A. Lanzavecchia.
1999
. Two subsets of memory T lymphocytes with distinct homing potentials and effector functions.
Nature
401
:
708
-712.
3
Masopust, D., V. Vezys, A. L. Marzo, L. Lefrançois.
2001
. Preferential localization of effector memory cells in nonlymphoid tissue.
Science
291
:
2413
-2417.
4
Reinhardt, R. L., A. Khoruts, R. Merica, T. Zell, M. K. Jenkins.
2001
. Visualizing the generation of memory CD4 T cells in the whole body.
Nature
410
:
101
-105.
5
Wherry, E. J., V. Teichgraber, T. C. Becker, D. Masopust, S. M. Kaech, R. Antia, U. H. Von Andrian, R. Ahmed.
2003
. Lineage relationship and protective immunity of memory CD8 T cell subsets.
Nat. Immunol.
4
:
225
-234.
6
Baron, V., C. Bouneaud, A. Cumano, A. Lim, T. P. Arstila, P. Kourilsky, L. Ferradini, C. Pannetier.
2003
. The repertoires of circulating human CD8+ central and effector memory T cell subsets are largely distinct.
Immunity
18
:
193
-204.
7
Bouneaud, C., Z. Garcia, P. Kourilsky, C. Pannetier.
2005
. Lineage relationships, homeostasis, and recall capacities of central- and effector-memory CD8 T cells in vivo.
J. Exp. Med.
201
:
579
-590.
8
Kedzierska, K., V. Venturi, K. Field, M. P. Davenport, S. J. Turner, P. C. Doherty.
2006
. Early establishment of diverse TCR profiles for influenza-specific CD62Lhi CD8+ memory T cells.
Proc. Natl. Acad. Sci. USA
103
:
9184
-9189.
9
Zaph, C., J. Uzonna, S. M. Beverley, P. Scott.
2004
. Central memory T cells mediate long-term immunity to Leishmania major in the absence of persistent parasites.
Nat. Med.
10
:
1104
-1110.
10
Roberts, A. D., K. H. Ely, D. L. Woodland.
2005
. Differential contributions of central and effector memory T cells to recall responses.
J. Exp. Med.
202
:
123
-133.
11
Klonowski, K. D., A. L. Marzo, K. J. Williams, S. J. Lee, Q. M. Pham, L. Lefrançois.
2006
. CD8 T cell recall responses are regulated by the tissue tropism of the memory cell and pathogen.
J. Immunol.
177
:
6738
-6746.
12
Berenzon, D., R. J. Schwenk, L. Letellier, M. Guebre-Xabier, J. Williams, U. Krzych.
2003
. Protracted protection to Plasmodium berghei malaria is linked to functionally and phenotypically heterogeneous liver memory CD8+ T cells.
J. Immunol.
171
:
2024
-2034.
13
Turner, S. J., G. Diaz, R. Cross, P. C. Doherty.
2003
. Analysis of clonotype distribution and persistence for an influenza virus-specific CD8+ T cell response.
Immunity
18
:
549
-559.
14
Kedzierska, K., S. J. Turner, P. C. Doherty.
2004
. Conserved T cell receptor usage in primary and recall responses to an immunodominant influenza virus nucleoprotein epitope.
Proc. Natl. Acad. Sci. USA
101
:
4942
-4947.
15
La Gruta, N. L., S. J. Turner, P. C. Doherty.
2004
. Hierarchies in cytokine expression profiles for acute and resolving influenza virus-specific CD8+ T cell responses: correlation of cytokine profile and TCR avidity.
J. Immunol.
172
:
5553
-5560.
16
La Gruta, N. L., P. C. Doherty, S. J. Turner.
2006
. A correlation between function and selected measures of T cell avidity in influenza virus-specific CD8+ T cell responses.
Eur. J. Immunol.
36
:
2951
-2959.
17
Venturi, V., K. Kedzierska, S. J. Turner, P. C. Doherty, M. P. Davenport.
2007
. Methods for comparing the diversity of samples of the T cell receptor repertoire.
J. Immunol. Methods
321
:
182
-195.
18
Belz, G. T., P. C. Doherty.
2001
. Virus-specific and bystander CD8+ T-cell proliferation in the acute and persistent phases of a γ-herpesvirus infection.
J. Virol.
75
:
4435
-4438.
19
Kedzierska, K., N. L. La Gruta, S. J. Turner, P. C. Doherty.
2006
. Establishment and recall of CD8+ T cell memory in a model of localized transient infection.
Immunol. Rev.
211
:
133
-145.
20
Stambas, J., P. C. Doherty, S. J. Turner.
2007
. An in vivo cytotoxicity threshold for influenza A virus-specific effector and memory CD8+ T cells.
J. Immunol.
178
:
1285
-1292.
21
Flynn, K. J., G. T. Belz, J. D. Altman, R. Ahmed, D. L. Woodland, P. C. Doherty.
1998
. Virus-specific CD8+ T cells in primary and secondary influenza pneumonia.
Immunity
8
:
683
-691.
22
Zhong, W., E. L. Reinherz.
2004
. In vivo selection of a TCR Vβ repertoire directed against an immunodominant influenza virus CTL epitope.
Int. Immunol.
16
:
1549
-1559.
23
Kedzierska, K., N. L. La Gruta, M. P. Davenport, S. J. Turner, P. C. Doherty.
2005
. Contribution of T cell receptor affinity to overall avidity for virus-specific CD8+ T cell responses.
Proc. Natl. Acad. Sci. USA
102
:
11432
-11437.
24
Turner, S. J., K. Kedzierska, H. Komodromou, N. L. La Gruta, M. A. Dunstone, A. I. Webb, R. Webby, H. Walden, W. Xie, 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.
25
Venturi, V., K. Kedzierska, D. A. Price, P. C. Doherty, D. C. Douek, S. J. Turner, M. P. Davenport.
2006
. Sharing of T cell receptors in antigen-specific responses is driven by convergent recombination.
Proc. Natl. Acad. Sci. USA
103
:
18691
-18696.
26
Turner, S. J., R. Cross, W. Xie, P. C. Doherty.
2001
. Concurrent naive and memory CD8+ T cell responses to an influenza A virus.
J. Immunol.
167
:
2753
-2758.
27
Masopust, D., S. J. Ha, V. Vezys, R. Ahmed.
2006
. Stimulation history dictates memory CD8 T cell phenotype: implications for prime-boost vaccination.
J. Immunol.
177
:
831
-839.
28
Jabbari, A., J. T. Harty.
2006
. Secondary memory CD8+ T cells are more protective but slower to acquire a central-memory phenotype.
J. Exp. Med.
203
:
919
-932.
29
Crowe, S. R., S. J. Turner, S. C. Miller, A. D. Roberts, R. A. Rappolo, P. C. Doherty, K. H. Ely, D. L. Woodland.
2003
. Differential antigen presentation regulates the changing patterns of CD8+ T cell immunodominance in primary and secondary influenza virus infections.
J. Exp. Med.
198
:
399
-410.
30
La Gruta, N. L., K. Kedzierska, K. Pang, R. Webby, M. P. Davenport, W. Chen, S. J. Turner, P. C. Doherty.
2006
. A virus specific CD8+ T cell immunodominance hierarchy determined by antigen dose and percursor frequencies.
Proc. Natl. Acad. Sci. USA
103
:
994
-999.
31
Wherry, E. J., J. N. Blattman, K. Murali-Krishna, R. van der Most, R. Ahmed.
2003
. Viral persistence alters CD8 T-cell immunodominance and tissue distribution and results in distinct stages of functional impairment.
J. Virol.
77
:
4911
-4927.
32
McHeyzer-Williams, M. G., M. M. Davis.
1995
. Antigen-specific development of primary and memory T cells in vivo.
Science
268
:
106
-111.
33
Busch, D. H., I. Pilip, E. G. Pamer.
1998
. Evolution of a complex T cell receptor repertoire during primary and recall bacterial infection.
J. Exp. Med.
188
:
61
-70.
34
McHeyzer-Williams, L. J., J. F. Panus, J. A. Mikszta, M. G. McHeyzer-Williams.
1999
. Evolution of antigen-specific T cell receptors in vivo: preimmune and antigen-driven selection of preferred complementarity-determining region 3 (CDR3) motifs.
J. Exp. Med.
189
:
1823
-1838.
35
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.
36
Maryanski, J. L., C. V. Jongeneel, P. Bucher, J. L. Casanova, P. R. Walker.
1996
. Single-cell PCR analysis of TCR repertoires selected by antigen in vivo: a high magnitude CD8 response is comprised of very few clones.
Immunity
4
:
47
-55.
37
Lin, M. Y., R. M. Welsh.
1998
. Stability and diversity of T cell receptor repertoire usage during lymphocytic choriomeningitis virus infection of mice.
J. Exp. Med.
188
:
1993
-2005.
38
Blattman, J. N., D. J. Sourdive, K. Murali-Krishna, R. Ahmed, J. D. Altman.
2000
. Evolution of the T cell repertoire during primary, memory, and recall responses to viral infection.
J. Immunol.
165
:
6081
-6090.