International Immunology Advance Access published online on March 15, 2007
International Immunology, doi:10.1093/intimm/dxm018
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Repertoires of T cell receptors expressed by graft-infiltrating T cells evolve during long-term recall responses to single minor histocompatibility antigens
1 Department of Immunology
2 Department of Surgery, Mayo Clinic College of Medicine, 200 First Street, Southwest, Rochester, MN 55905, USA
Correspondence to: Correspondence to: P. J. Wettstein; E-mail: wettstein.peter{at}mayo.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionDisclosuresReferences |
|---|
Long-lived mammals such as humans respond over decades with CTL and helper T lymphocytes to acute and chronic infections. Maintaining these extended recall responses requires established memory populations of sufficient size and diversity to effectively respond to these infections. Studies in mice have indicated that cytotoxic T cells that respond to secondary viral infections are principally those that were activated in primary responses and maintained through memory. However, long-term recall responses in humans must involve many more responses with increased opportunities for recruitment of naive T cells and competition between memory and naive cells. We hypothesized that increased numbers of antigenic challenges prolong selection pressure on responding T cells resulting in continuously changing populations characterized by evolving TCR repertoires. This hypothesis was tested by transplanting recipients with 10 sets of H4-incompatible skin allografts that were harvested at times of rejection for spectratyping of TCR alpha and beta transcripts expressed by graft-infiltrating T cells. Amplicons with single complementarity-determining region 3 (CDR3) lengths were sequenced, and CDR3-specific primers were used for amplifications to identify graft sets in which these sequences were present. The results showed that TCR repertoires did not stabilize in these extended recall responses since the majority of identified alpha and beta CDR3s was present in varying numbers of graft sets with only a minority being present throughout. Selection for specific beta CDR3s was suggested by observed associations between the extent of the recall responses and shortened CDR3 lengths and restricted distributions of net charges in CDR3s.
Keywords: allograft rejection, antigenic challenge, CDR3, evolution
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionDisclosuresReferences |
|---|
TCRs are heteroduplexes of alpha and beta subunits that determine the specificity of T cells for peptides presented by products of class I and class II genes that map to MHCs of mammals. This specificity is directed by the utilized variable (V) and joining (J) regions in alpha and beta subunits as well as the diversity (D) regions in beta subunits (1). The process of recombination between V and J gene segments results in the formation of complementarity-determining region 3s (CDR3s) that include the carboxy and amino termini of the V and J segments, respectively, as well as variable numbers of randomly selected nucleotides inserted between the V and J segments. In the case of beta chains, variable lengths of D regions are also included in the CDR3s. CDR3s impact antigenic specificity through their lengths and amino acid sequences/motifs (2–5) which is consistent with the contact of alpha and beta CDR3s with the amino and carboxy termini of MHC-bound peptides, respectively (6). The variable numbers of nucleotides inserted between V and J segments result in CDR3s with variable lengths such that populations of T cells express arrays of alpha and beta subunits with diverse CDR3 lengths. The analysis of the diversity of beta CDR3s linked to individual beta variable (BV) genes has been achieved with spectratyping, or immunoscope, that involves amplification of beta transcripts in RT–PCR (7–9). Numerous studies have shown that normal T cell populations in mice and humans are characterized by Gaussian distributions of beta CDR3 lengths (9), and similar observations have been made for alpha CDR3s in humans (10). However, restricted diversity has been observed with T cell populations that infiltrate inflammatory sites (11), tumors (9) and allografts (12). In the most extreme cases, distributions are reduced to single lengths carried by single transcripts (11).
Experiments have demonstrated that primary responses to viral infections and antigenic protein challenges are mediated by CTLs and helper T lymphocytes (HTLs) that are characterized by diverse repertoires of TCR alpha and beta transcripts. The levels of diversity at the levels of variable region gene usage and CDR3 characteristics vary depending on the target peptides (3, 13, 14). In the case of viral infections, this diversity shrinks during the final stages of clearance of the virus, in parallel with viral load, and is maintained through memory (14, 15). Further reduction in diversity is observed in responses to secondary infections with identification of beta transcripts that are present in both primary and secondary responses (13). Similar results have been obtained with HTL and CTL responses to single peptides (4, 16). The identification of the same transcripts in T cells responding to primary and secondary challenges has led to proposals that TCR repertoires are selected within primary responses and are sustained through subsequent recall responses (13, 16). Studies in mice have been limited to primary and secondary challenges/infections, but humans have the potential to experience considerably greater numbers of challenges throughout their decades-long lifespans. It is plausible that the apparent stability in TCR repertoire observed with only two challenges in mice cannot be extrapolated to relatively long-term recall responses in humans.
We have developed a mouse model to investigate the maintenance or evolution of TCR diversity in responses to multiple sets of skin allografts that carry single minor histocompatibility antigen (MiHA) peptides presented to CTLs by class I molecules. We have previously used spectratyping to identify over-represented beta transcripts expressed by CTLs that infiltrate first- and second-set skin allografts in mice (17, 18). Others have also used spectratyping to identify beta transcripts that persist in memory CTL responses following primary infections with lymphocytic choriomeningitis virus (LCMV) (19). Our initial studies have now been extended through 10 sets of H4-incompatible skin grafts with spectratyping analysis performed on each set of rejecting allografts. Spectratyping identified alpha and beta transcripts that were over-represented within rejecting allografts, and sequencing of amplified products yielded CDR3 sequences for designing CDR3-specific primers to track the presence of these sequences in sequential graft sets on individual recipients. The results of these experiments showed that the arrays of identifiable alpha and beta CDR3s that were expressed by graft-infiltrating T cells varied throughout the responses to multiple sets of MiHA-incompatible skin grafts.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionDisclosuresReferences |
|---|
Mice
C57Bl/10SnJ (B10) and B10.129-H4b (21M) mice were purchased from the Jackson Laboratory, Bar Harbor, ME, USA. All mice were housed in the barrier facility in the Mayo Clinic Division of Animal Medicine.
Skin grafting
Transplantation of orthotopic tail skin grafts (
2 x 5 mm in size) was performed according to the previously described technique (20). All skin grafting was performed with donors and recipients that were anesthetized with sodium pentobarbital. Each recipient of primary allografts to be scored for times of rejection received a single autograft and two allografts. Primary grafts were scored at routine intervals for the condition of epidermal scale pattern, pigment and hair, and rejection was scored when no viable signs were observed for both allografts. Second sets of two skin allografts were transplanted 14 days after rejection of the primary allografts. When the rejection process was observed on the basis of edema and ulceration, the allografts were harvested and replaced by syngeneic grafts to promote wound healing. Ten cycles of grafting and harvesting were performed with each recipient with 14-day intervals between allograft harvests and subsequent transplantation of allografts. Harvested grafts were immediately transferred to lysis buffer [Perfect RNA kit (Brinkman, Westbury, NY, USA)] for storage at –80°C.
Alpha variable- and BV-specific amplification and sequencing
Primers were synthesized by the Mayo Clinic Molecular Biology Core Facility. Series of nested alpha constant (AC) and beta constant (BC) region primers (Table 1) have been previously described (21). BV gene-specific primers have been described (Table 1) (22), and alpha variable (AV) subfamily-specific primers (Table 1) are the murine homologs of human AV-specific primers that have been successfully used in spectratyping studies (10). Total RNA was extracted with a Perfect RNA kit (Brinkman) from (1) harvested grafts and (2) CD8+ T cells that were enriched from normal splenocytes by anti-CD8-coated magnetic beads (Miltenyi Biotek, Auburn, CA, USA). RT–PCRs were performed in 96-well plates using either the AC Nest I or the BC Nest I reverse primer and the respective panel of AV- or BV-specific forward primers in a modified Ampliwax PCR Gem protocol (Applied Biosystems, Foster City, CA, USA). The cDNA synthesis step was performed in bulk (scaled for the required numbers of V gene-specific PCRs) in 10 mM Tris–HCl (pH 8.3), 50 mM KCl, 5 mM MgCl2 and 0.25 mM each dNTP (Promega, Madison, WI, USA), and the following components per final V gene-specific PCR: 10 pmol of reverse constant region primer, 7.5 U of RNasin ribonuclease inhibitor (Promega), 2 U of Moloney murine leukemia virus reverse transcriptase (Life Technologies, Grand Island, NY, USA) and 10 ng of total RNA. Reactions were incubated at 37°C for 25 min followed by 42°C for 30 min. Ten microliter volumes of cDNA were aliquoted to wells containing wax pellets for incubation at 95°C for 5 min to inactivate the reverse transcriptase. After cooling, 40 µl upper layers were added that included 10 mM Tris–HCl, 1.25 mM MgCl2 and 50 mM KCl, 10 pmol of an AV- or BV-specific primer and 1.25 U of Taq polymerase (Promega). The PCR conditions were 35 cycles of 94°C for 1 min, 55°C for 1 min and 72°C for 2 min; a final extension was performed at 72°C for 8 min. Fluorochrome-labeled amplicons were generated in nested PCRs of 30 cycles using 1 µl of the RT–PCR products as templates, 10 pmol WellRED-labeled (Beckman Coulter, Fullerton, CA, USA) AC Nest II or BC Nest II primer, the respective AV- or BV-specific primers, 1.25 U of Taq DNA polymerase and the same cycling conditions as used for the PCR portion of the RT–PCR.
|
Fluorochrome-labeled amplicons were mixed with DNA Standard 400 (Beckman Coulter) in 100% formamide for analysis by capillary electrophoresis (CE) using a CEQ 8000 Genetic Analysis System (Beckman Coulter). Spectratype peak distributions were inspected to select those products with single peaks for amplification of unlabeled products with the respective, unlabeled primers to generate template for nucleotide sequencing. These PCR reactions were performed in 25 µl volumes that included 10 mM Tris–HCl (pH 8.3), 50 mM KCl, 1.5 mM MgCl2, 0.1 mM each dNTP, 4 pmol of each primer, 0.625 U of Taq polymerase (Promega) and 1 µl of original RT–PCR product. Cycling conditions were as described above. The PCR products were sequenced with a third nested AC or BC region primer. The sequences of the CDR3s were used to design CDR3-specific forward primers with annealing temperatures near 55°C for use in PCRs with the WellRED-labeled AC or BC Nest II primers and template amplified in the original RT–PCRs. Amplicons were analyzed by CE to determine the presence or absence of specific CDR3s in the products of the original RT–PCRs.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionDisclosuresReferences |
|---|
A model of long-term recall was established in mice using multiple sets of H4-incompatible skin allografts with harvests of rejecting grafts from individual recipients (Fig. 1). The H4 target antigen was chosen because it includes a single H4 peptide (SGIVYIHL) (23), and CD4+ Th are not required for rejection of H4-incompatible skin grafts (24). We did not include depletion of CD4+ T cells with mAb due to the potential to generate anti-rat IgG antibodies during this long-term grafting regimen. Five B10 males were grafted with primary allografts from 21M donors; primary allografts were rejected with a median of
28 days as predicted by previous results (25). Recipients received second-set grafts approximately two weeks after rejection of primary grafts; second-set grafts were harvested at the time of rejection, and syngeneic grafts were placed in the resulting graft beds to promote healing. Third-set grafts were transplanted 14 days later and these grafts were harvested at rejection, and grafting was continued through 10 sets at 14-day intervals between harvesting and grafting. Survival times (times of harvest) of 3rd to 10th graft sets remained constant at 4–5 days.
|
Beta transcripts
RT–PCRs were performed with RNAs extracted from all sets of rejecting H4 allografts on all recipients; 24 BV gene-specific primers were paired with a constant region primer. RT–PCR products were labeled with WellRED by re-amplification with the BV primers and a labeled, nested BC primer. Labeled products were separated by CE, and the resulting peak distributions were inspected to identify products with single or predominant peaks with single CDR3 lengths. Attempts were made to comprehensively select products from multiple graft sets from all recipients for re-amplification for nucleotide sequencing using a third, nested BC primer. A total of 72 sequences were obtained and the amino acid sequences of the CDR3s and joining regions are presented in Table 2; five of these sequences were obtained from amplicons from two different graft sets on individual recipients. Based on the range of these transcripts, the expression of BV genes by infiltrating T cells appeared to be diverse, but 30/72 transcripts utilized either BV9, BV12 or BV14. There was also no apparent over-representation of BV genes within individual recipients. However, there did appear to be over-representation of specific beta joining (BJ) genes; the proximal BJ1.1 and BJ1.2 genes in the first BJ cluster were included in 16/29 transcripts containing BJ genes from that cluster. Forty-three transcripts carried joining segments encoded by genes in the second BJ cluster and 39 of these transcripts included the BJ2.1, 2.3, 2.5 and 2.7 genes.
|
The diversity of beta transcripts and the limited number of duplicate sequences suggested that oligoclonal T cell responses were not established and maintained throughout these extended recall responses. We hypothesized that the populations of graft-infiltrating T cells were in flux throughout the grafting series due to the variable appearance and disappearance of T cells expressing different beta transcripts. Therefore, the duration of the presence of the identified CDR3 sequences in sequential sets of rejecting allografts was investigated. CDR3-specific reverse primers from 53 beta sequences were synthesized and paired with the WellRED-labeled BC primer in PCRs. Templates for these reactions were the products from the original RT–PCRs that used RNA extracted from each of the nine sets of rejecting allografts on the respective, single recipients. These labeled PCR products were separated by CE to identify those graft sets in which the CDR3s were present. CE was used rather than gel electrophoresis due to its greater sensitivity and size discrimination, and CDR3s were considered absent when the signal dropped below 5% of the maximal signal obtained with the individual CDR3-specific primer.
PCR amplifications with beta CDR3-specific primers revealed considerable variability in both the times at which CDR3s were first present and the duration of their presence in sequential sets of allografts (Fig. 2). Representative results of these analyses are presented in Fig. 3; in this figure, the #1 CDR3 sequence exemplifies the observation that some CDR3s were not detectable (>5% of the maximal signal) in every graft set between the first and last graft sets in which the CDR3 sequence was detected. We suggest that this lack of detection was due to transient reductions in levels of available template relative to other transcripts carrying the same BV gene that were amplified in the initial RT–PCRs. CDR3s were placed into four general categories based on their times of appearance and duration: early, middle, late and long term. CDR3s in the early and long-term groups were present in the second-set grafts but were differentiated by the disappearance of early sequences prior to the eighth sets while long-term sequences persisted at least through the ninth sets. The long-term CDR3s were a minority (10/53) of the total number of tested sequences indicating that the majority of sequences was only present for parts of the recall responses. The largest group of sequences was the middle group in which CDR3s appeared at varying times after the second sets of grafts and disappeared before the 10th sets. The late category included CDR3s that appeared in the fourth sets at the earliest and persisted through the 10th sets of grafts. These results suggest that the repertoires of graft-infiltrating T cells did not become either fixed or restricted but rather continued to change throughout these extended recall responses.
|
|
The variable appearance and disappearance of CDR3s during the recall responses enabled us to investigate potential shifts in CDR3 lengths and motifs over time. In regard to CDR3 lengths, there was a perceptible change in median length between that of the early group (median = 10 aa) and those of the late (median = 8 aa) and long-term groups (median = 9.0 aa) (Fig. 4). These differences were significant at P < 0.025 and P
0.025 (Wilcoxon Rank Sum test), respectively. These results suggest that the presence of CDR3s in rejecting allografts at extended times was associated with shorter CDR3 lengths regardless of whether these sequences first appeared in late graft sets or were present in the second-set grafts and persisted throughout the entire recall responses.
|
We have previously reported that beta CDR3s from cloned CTLs and CTLs that infiltrate second-set H4-incompatible grafts were characterized by net negative charges (17, 21, 26), and this characteristic was also observed in CDR3s that were identified in the present experiments. As cited above, specific BJ segments appeared to be over-represented in CDR3s. This over-representation may be accounted for by the net negative charges exhibited by the six BJ segments that are included in 78% of the sequenced CDR3s. The identification of CDR3s at different times in the recall responses made it possible to investigate shifts in charge characteristics. There appeared to be a consolidation to a net charge of –1 within the CDR3 regions over time (Fig. 5). The presence of CDR3s with a net –1 charge increased from 25 to 50% and then to 80% through the early, middle and late categories, respectively. The long-term CDR3s that were observed throughout the entire recall responses appeared to retain a basal, broad representation of CDR3 net charges while favoring the maintenance of those CDR3s that carried a single net –1 charge. Overall, 65% of the CDR3s present at the rejection of 10th set grafts (late + long term) carried –1 charges suggesting a preference for this CDR3 characteristic.
|
Alpha transcripts
Spectratyping in mice has been limited to beta transcripts in part due to the size of the AV gene family that includes 82 functional genes in a series of subfamilies with variable numbers of subfamily members (27). We synthesized a series of 22 subfamily-specific primers that were homologous to those utilized in spectratyping studies in humans (Table 1) (10). This panel of AV primers was tested in RT–PCRs with RNA extracted from normal B6 CD8+ T cells; forward AV primers were paired with the AC I constant region primer (Table 1) for RT–PCRs. Amplified products were re-amplified with WellRED-labeled AC II primer to generate labeled amplicons for separation by CE. The spectratypes obtained with four AV primers are presented in Fig. 6; the alpha CDR3 lengths in these products and those obtained with the other AV primers exhibited Gaussian distributions as observed with beta CDR3s expressed by normal T cell populations.
|
This panel of AV-specific primers was paired with the AC I constant region primer to reverse transcribe and amplify RNA extracted from all sets of H4-incompatible grafts harvested at all time points from all grafted recipients. Amplified products were re-amplified in a nested PCR with the respective AV-specific primers, and the resulting products were analyzed by CE. Products with single peaks were re-amplified for sequencing and the obtained sequences are presented in Table 3. The reduced number of alpha sequences relative to beta sequences was linked to a lower number of amplicons with single CDR3 lengths presumably due to the specificity of AV primers for subfamilies of AV genes rather than individual V genes. The alpha sequences included AV genes from eight subfamilies that include variable numbers of genes ranging from 1 (AV19) to 12 (AV4). There was no apparent preference for specific alpha joining (AJ) segments since the distribution of utilized AJ segments was relatively diverse with a total of 15 AJ segments included in the 22 sequences. The analysis of the duration of alpha CDR3s in sequential skin allografts revealed a distribution similar to that observed with beta CDR3s in that a stable representation of alpha CDR3s was not reached during the 2–10 sets of H4-incompatible grafts (Fig. 7). With two exceptions, all CDR3s were found in at least two graft sets, and the majority of CDR3s could be classified as early, middle and late. A minority of long-term alpha CDR3s was identified throughout the complete recall responses. These results coupled with those obtained with beta CDR3s support the hypothesis that these long-term recall responses are mediated by ever-changing populations of T cells that infiltrate sequential skin allografts.
|
|
The relatively low numbers of identified alpha transcripts made it difficult to identify clear trends in CDR3 length and motif. There was no significant reduction in lengths of alpha CDR3s between early and late graft sets. The alpha CDR3s were virtually all positively charged with the vast majority of positive charges carried by the utilized AJ segments. This is not unexpected given the distribution of positively charged residues in AJ segments (27). What is most interesting is not which AJ segments were utilized but rather which were not included in these transcripts. AJ segments that carried negatively charged amino acids in their amino termini were poorly represented, and if they were, the nucleotide stretches that encoded these amino acids were deleted during rearrangement of AV and AJ genes. A case in point is sequence #21 that includes AJ30; an Asp residue was clipped off and replaced with Arg encoded by the randomly inserted nucleotides. The selection of AJ segments, rearrangements and nucleotide insertions led to alpha CDR3s that as a group carried 29 positive charges and only three negative charges.
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionDisclosuresReferences |
|---|
Humans are relatively long-lived mammals who must respond immunologically to both chronic and repeated, acute exposures to infectious agents over decades of life. Effective responses to both types of exposures require maintenance of memory T cell compartments, and little is known about the stability and levels of diversity of memory populations and their relationships to extended recall responses, despite the importance of maintaining such responses. CTLs mediating primary and secondary responses in LCMV and influenza models in mice share beta transcript repertoires(14, 15, 29), and CTLs responding to secondary infections appear to be randomly selected from memory populations that had been established following primary infections (29, 30). These results and those obtained in CTL responses to single peptides (31) have driven the hypothesis that repertoires of memory CTLs are stable and reflect the constitution and diversity of primary responses.
The experiments described here investigated the effects of increased numbers of antigenic challenges on the stability of repertoires of TCRs expressed by skin allograft-infiltrating T cells. We have operationally classified these responses as recall responses rather than continuous responses due to the characteristics of skin allograft rejection. Chronic exposure to viral pathogens is characterized by reduced but sustained viral loads with continuous T cell responses. Acute infections with higher levels of antigen stimulate rapid T cell responses resulting in either reduction or clearance of viral loads followed by shifts to memory phases that require apoptosis of the vast majority of effector CTLs and the establishment of memory populations. The accelerated rejection of 3rd to 10th set H4-incompatible grafts within 4 days results in complete destruction of the grafts including the underlying graft beds right down to the cartilage on the tails. Therefore, rejection of donor skin which is the sole, stable source of foreign antigens should be analogous to complete clearance of virus since there would no longer be a source of antigen once the graft is completely rejected. There are most certainly differences between skin allograft and acute viral infection models including the differences in stimulation of innate immunity at the different anatomic sites as well as the complete destruction of graft-infiltrating T cells as a consequence of graft rejection. The latter is particularly important since experimental harvesting of rejecting grafts should result in no more reduction in CTL populations than that encountered in the normal rejection process. Further, the short time intervals (14 days) between rejection and transplantation in these experiments is shorter than would be expected between natural, acute viral infections. However, results of previous studies in the LCMV model predict that 14 days is sufficient for reduction in numbers of effector CTLs and establishment of CTL memory when allografts are completely rejected within 4 days. LCMV loads are reduced within 9–10 days and LCMV-specific CTLs drop to memory phase frequencies within 15 days following primary infections (32, 33). Future experiments are planned to investigate the effects of increased rejection-transplantation time intervals on evolution of TCR repertoires of H4-specific CTLs.
Spectratyping dissects repertoires of alpha and beta transcripts for the identification and sequencing of individual CDR3 sequences, but it is not possible to confirm that the identified CDR3s are expressed by CTLs that mediate allograft rejection through specific recognition of target alloantigenic peptides. Allografts may be infiltrated by non-specific CTLs as well as specific and non-specific Th. We have already shown that the speed of rejection of H4-incompatible skin grafts is not affected by depletion of CD4+ Th (24). The level of non-specific infiltration of rejecting allografts is unknown but would be expected to involve random distributions of T cells in individual graft sets with associated changes in CDR3 diversity. However, the vast majority of CDR3 sequences that was analyzed for expression in multiple H4-incompatible graft sets was identified in at least two sets of grafts. These results argued that these CDR3s were expressed by T cells specifically involved in rejection since the probability of detecting identical CDR3s expressed by T cells that randomly infiltrated multiple graft sets would be expected to have been vanishingly small given the complexity of repertoires of peripheral T cell populations. Further, we recently performed comparable spectratype analyses of syngeneic male grafts and observed that the groups of identified alpha and beta CDR3s exhibited different charge characteristics than those reported here for H4-incompatible allografts (data not shown). This disparity argues against identification of CDR3s expressed by T cells that non-specifically infiltrate rejecting allografts.
The most important observation reported here is that the spectra of alpha and beta CDR3s expressed by populations of graft-infiltrating T cells did not remain stable throughout the extended recall responses. Rather, the majority of CDR3s was present in variable numbers of graft sets ranging from early to late, and changing expression of CDR3s was observed with all tested recipients. There were also CDR3s that were present throughout the complete range of allografts, suggesting that minorities of T cells were involved in rejection of grafts throughout the complete recall responses. These are the first results from a mouse model of long-term recall that involves multiple antigenic challenges in comparison with previous mouse models of primary and secondary T cell responses to viruses and protein antigens. The multiple challenges did not appear to drive the H4-specific response toward over-representation of specific BV genes as observed with CTL responses to LCMV and influenza infections that are characterized by preferential utilization of BV genes in primary CTL responses that is sustained through memory phases and secondary responses (13, 14, 29). This apparent lack of over-representation of BV genes is consistent with diverse BV gene usage in extensive panels of H4-specific CTL clones (21, 26).
The analysis of virus-specific CTL responses has led to the prevailing view that repertoires become stabilized following secondary responses. Repertoires of LCMV-specific CTLs appear to be determined in great part by burst sizes in primary responses since BV-specific staining of class I tetramer-positive CTLs has shown that primary and secondary CTL populations have comparable distributions of expressed BV genes (14). Similarly, spectratyping of BV-BJ amplicons revealed that T cell clonotypes responding to primary infections were detectable following secondary infections (15). Although these results point toward stabilization of repertoires within memory compartments, their significance for understanding repertoire diversity in long-term responses is reduced by the relatively limited numbers of opportunities for competition between responding CTLs.
Extension of recall responses, as described here, would be expected to increase the potential for competition between responding CTLs. This competition begins in the primary response where stochastic forces affect the initial diversity of responding CTLs that are then selected for functional avidity. Such avidity is presumably determined by TCR affinity, expression of CD8 and response to cytokines. In particular, IL-15 plays an important role in determining CD8 expression levels and driving expansion and homeostasis of high-avidity CTLs (34), and IL-15 expression by multiple cell types in skin increases during wound healing (35). The competition involved in primary responses would expectedly continue following successive challenges with the added complexity of competition between pre-existing memory and naive CTLs. Naive CTLs would include pre-existing cells that are newly trafficking through the draining lymph nodes or are newly emigrated from the thymus. Support for a contribution by newly differentiated CTLs in CTL responses to chronic infection comes from a recent publication showing that maintenance of CTLs that respond to chronic viral infections is partially dependent upon continued thymic function (36). Increased rounds of competition between CTLs should exert increased selection pressure with accompanying changes in TCR repertoires expressed by responding CTLs.
Antigen-driven selection of T cell repertoires has been shown by the study of Th responses to pigeon cytochrome C (PCC) where selection for CDR3 lengths and motifs was revealed by comparing beta and alpha transcripts expressed by PCC-specific Th involved in primary and secondary responses (2, 4). PCC-specific Th preferentially express AV11 and BV3 genes with high frequencies (37), and the alpha and beta CDR3s exhibit characteristics of antigen-driven selection since the frequencies of apparently optimal lengths and motifs increase significantly between the primary and secondary challenges (2, 4). The speed with which these characteristics increase in frequency suggests that recognition of PCC peptide requires not only constrained variable region gene usage but also constrained alpha and beta CDR3 attributes that are strongly selected for early in T cell responses. It appears that H4-specific CTLs express diverse variable region genes rather than strictly require AV and/or BV genes (21, 26) since there were no apparent shifts in usage of AV and BV genes during the extended recall responses to H4-incompatible allografts. This flexible usage of variable region genes appeared to also extend to alpha and beta CDR3s which were characterized by net positive and negative charges, respectively, but did not include motifs consisting of multiple, specific amino acids as observed for PCC-specific CDR3s and for influenza peptide-specific CTLs (4, 13). A deterministic role for antigen-driven selection in long-term recall responses to the H4 peptide was suggested by the association between numbers of graft sets and significant reductions in beta CDR3 lengths and focusing of numbers and positions of net negative charges. Therefore, antigen-driven selection may be more subtle in responses to the H4 peptide than in the responses to PCC or influenza, but significant selection of lengths and motifs may be achieved during extended recall responses.
Inspection of beta CDR3 sequences revealed that the majority of negative charges was included in the BJ segments so it was appealing to hypothesize that the selection of BJ segments was directed by preferences for negative charges. In fact, BJ segments that lacked these charges in their amino termini were poorly represented among the panel of beta CDR3 sequences. It is difficult to confirm that this apparent over-representation of specific BJ segments in graft-infiltrating T lymphocytes is significantly different from that observed in total, peripheral T lymphocyte populations. Skewing of BJ usage has been reported with mouse fetal thymocytes (38) as well as with thymocytes and peripheral T cell subsets that expressed BV17 (39). In the latter study, BJ usage changed according to the T cell sub-populations making it difficult to discern significant trends. Perhaps the most informative comparison is between BJ usage in T lymphocytes that infiltrate H4- and HY-incompatible grafts. Beta CDR3 segments derived from HY-incompatible grafts do not exhibit skewing of BJ gene usage or preferential inclusion of negative charges (18 and data not shown), as we report here for those derived from H4-incompatible grafts.
The results reported here are the first obtained with a murine model of extended recall responses that involve numbers of challenges that are comparable in number to those in long-lived humans. Since maintenance of recall responses to pathogens is essential for survival, it is important to evaluate factors that affect the evolution of TCR repertoires. These factors would expectedly include the lengths of memory phases between challenges, thymic output and immunogenicity of foreign peptides. Since humans maintain recall responses with advancing age, it will be important to understand the effects of aging on TCR repertoire diversity. These and additional factors can be effectively investigated with this model to gain a greater understanding of the selective forces that shape evolving T cell responses.
|
Disclosures |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionDisclosuresReferences |
|---|
The authors have no financial conflict of interest.
|
Acknowledgements |
|---|
The authors thank Ms DeAnn Frederixon for preparation of the manuscript and graphics. This work was supported by a grant (AI16052) from the National Institutes of Health (P.J.W.).
|
Abbreviations |
|---|
| AC, alpha constant |
| AJ, alpha joining |
| AV, alpha variable |
| BC, beta constant |
| BJ, beta joining |
| BV, beta variable |
| CDR3, complementarity-determining region 3 |
| CE, capillary electrophoresis |
| HTL, helper T lymphocyte |
| LCMV, lymphocytic choriomeningitis virus |
| MiHA, minor histocompatibility antigen |
| PCC, pigeon cytochrome C |
|
Notes |
|---|
Transmitting editor: S. Swain
Received 6 October 2006, accepted 24 January 2007.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionDisclosuresReferences |
|---|
- Chien Y-H, Gascoigne RJ, Kavaler J, Lee NE, Davis MM. (1984) Somatic recombination in a murine T-cell receptor gene. Nature 309:322.[CrossRef][Medline]
- McHeyzer-Williams MG and Davis MM. (1995) Antigen-specific development of primary and memory T cells in vivo. Science 268:106.
[Abstract/Free Full Text] - Kedzierska K, La Gruta NL, Davenport MP, Turner SJ, Doherty PC. (2005) Contribution of T cell receptor affinity to overall avidity for virus-specific CD8+ T cell responses. Proc. Natl Acad. Sci. USA 102:11432.
[Abstract/Free Full Text] - McHeyzer-Williams LJ, Panus JF, Mikszta JA, McHeyzer-Williams MG. (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.
[Abstract/Free Full Text] - Zhong W and Reinherz EL. (2004) In vivo selection of a TCR Vß repertoire directed against an immunodominant influenza virus CTL epitope. Int. Immunol. 16:1549.
[Abstract/Free Full Text] - Garcia KC, Degano M, Pease LR, et al. (1998) Structural basis of plasticity in T cell receptor recognition of a self peptide-MHC antigen. Science 279:1166.
[Abstract/Free Full Text] - Pannetier C, Cochet M, Darche S, Casrouge A, Zoller M, Kourilsky P. (1993) The sizes of the CDR3 hypervariable regions of the murine T-cell receptor ß chains vary as a function of the recombined germ-line segments. Proc. Natl Acad. Sci. USA 90:4319.
[Abstract/Free Full Text] - DePalma R and Gorsky J. (1995) Restricted and conserved T-cell repertoires involved in allorecognition of class II major histocompatibility complex. Proc. Natl Acad. Sci. USA 92:8836.
[Abstract/Free Full Text] - Caignard A, Dietrich P, Morand V, et al. (1994) Evidence for T-cell clonal expansion in a patient with squamous cell carcinoma of the head and neck. Cancer Res. 54:1292.
[Abstract/Free Full Text] - Han M, Harrison L, Kehn P, Stevenson K, Currier J, Robinson MA. (1999) Invariant or highly conserved TCR
are expressed on double-negative (CD3+CD4–CD8–) and CD8+ T cells. J. Immunol. 163:301.[Abstract/Free Full Text] - Musette P, Bequet D, Delarbre C, Gachelin G, Kourilsky P, Dormont D. (1996) Expansion of a recurrent V beta 5.3+ T-cell population in newly diagnosed and untreated HLA-DR2 multiple sclerosis patients. Proc. Natl Acad. Sci. USA 93:12461.
[Abstract/Free Full Text] - Guillet M, Brouard S, Gagne K, et al. (2002) Different qualitative and quantitative regulation of Vß TCR transcripts during early acute allograft rejection and tolerance induction. J. Immunol. 168:5088.
[Abstract/Free Full Text] - Kedzierska K, Turner SJ, Doherty PC. (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.
[Abstract/Free Full Text] - Blattman JN, Sourdive DJD, Murali-Krishna K, Ahmed R, Altman JD. (2000) Evolution of the T cell repertoire during primary, memory, and recall responses to viral infection. J. Immunol. 165:6081.
[Abstract/Free Full Text] - Lin MY and Welsh RM. (1998) Stability and diversity of T cell receptor repertoire usage during lymphocytic choriomeningitis virus infection of mice. J. Exp. Med. 188:1993.
[Abstract/Free Full Text] - Walker PR, Wilson A, Bucher P, Maryanski JL. (1996) Memory TCR repertoires analyzed long-term reflect those selected during the primary response. Int. Immunol. 8:1131.
[Abstract/Free Full Text] - Johnston SL, Borson ND, Wettstein PJ. (1997) Spectratyping of TCRs expressed by CTL-infiltrating minor histocompatibility antigen-disparate allografts. J. Immunol. 159:5233.[Abstract]
- Johnston SL and Wettstein PJ. (1998) Spectratyping of TCR expressed by CTL infiltrating male antigen (HY)-disparate allografts. J. Immunol. 160:3374.
[Abstract/Free Full Text] - Selin LK, Lin MY, Kraemer KA, et al. (1999) Attrition of T cell memory: selective loss of LCMV epitope-specific memory CD8 T cells following infections with heterologous viruses. Immunity 11:733.[CrossRef][Web of Science][Medline]
- Bailey DW and Usama B. (1960) A rapid method of grafting skin on tails of mice. Transplant. Bull. 7:424.[Medline]
- Johnston SL and Wettstein PJ. (1997) T cell receptor diversity in CTLs specific for the CTT-1 and CTT-2 minor histocompatibility antigens. J. Immunol. 159:2606.[Abstract]
- Rodriguez M, Prayoonwiwat N, Zhou P, David C. (1993) Expression of T cell receptor Vß transcripts in central nervous system of mice susceptible and resistant to Theiler's induced demyelination. J. Neuroimmunol. 47:95.[CrossRef][Web of Science][Medline]
- Luedtke B, Pooler LM, Choi EY, et al. (2003) A single nucleotide polymorphism in the Emp3 gene defines the H4 minor histocompatibility antigen. Immunogenetics 55:284.[CrossRef][Web of Science][Medline]
- Wettstein PJ and Korngold R. (1991) T cell subsets required for in vivo and in vitro responses to single and multiple minor histocompatibility antigens. Transplantation 54:296.
- Wettstein PJ and Haughton G. (1977) The genetic control of the immune response to murine histocompatibility antigens. I. The response to H-4 alloantigens. Immunogenetics 4:65.
- Johnston SL, Graham D, Wettstein PJ. (1997) Diversity of alpha and beta subunits of T-cell receptors specific for the H4 minor histocompatibility antigen. Immunogenetics 46:17.[CrossRef][Web of Science][Medline]
- Lefranc MP, Giudicelli V, Kaas Q, et al. (2005) IMGT, the international ImMunoGeneTics information system. Nucleic Acids Res. 33:D593.
[Abstract/Free Full Text] - Wang R and Taniguchi M. (1995) Limited T cell antigen receptor repertoire in tumor-infiltrating lymphocyte and inhibition of experimental lung metastasis of murine melanoma by anti-TCR antibody. J. Immunol. 154:1797.[Abstract]
- Turner SJ, Diaz G, Cross R, Doherty PC. (2003) Analysis of clonotype distribution and persistence for an influenza virus-specific CD8+ T cell response. Immunity 18:548.
- Sourdive DJD, Murali-Krishna K, Altman JD, et al. (1998) Conserved T cell receptor repertoire in primary and memory CD8 T cell responses to an acute viral infection. J. Exp. Med. 188:71.
[Abstract/Free Full Text] - Maryanski JL, Jongeneel CV, Bucher P, Casanova JL, Walker PR. (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.[Medline]
- Moskophidis D, Lechner F, Pircher H, Zinkernagel RM. (1993) Virus persistence in acutely infected immunocompetent mice by exhaustion of antiviral cytotoxic effector T cells. Nature 362:758.[CrossRef][Medline]
- Murali-Krishna K, Altman JD, Suresh M, et al. (1998) Counting antigen-specific CD8 T cells: a re-evaluation of bystander activation during viral infection. Immunity 8:177.[CrossRef][Web of Science][Medline]
- Oh S, Perera LP, Burke DS, Waldmann TA, Berzofsky JA. (2004) IL-15/IL-15R-mediated avidity maturation of memory CD8+ T cells. Proc. Natl Acad. Sci. USA 101:15154.
[Abstract/Free Full Text] - Castagnoli C, Trombotto C, Ariotti S, et al. (1999) Expression and role of IL-15 in post-burn hypertrophic scars. J. Invest. Dermatol. 113:238.[CrossRef][Web of Science][Medline]
- Vezys V, Masopust D, Kemball CC, et al. (2006) Continuous recruitment of naive T cells contributes to heterogeneity of antiviral CD8 T cells during persistent infection. J. Exp. Med. 203:2263.
[Abstract/Free Full Text] - Hedrick SM, Engel I, McElligott DL, et al. (1988) Selection of amino acid sequences in the beta chain of the T cell antigen receptor. Science 239:1541.
[Abstract/Free Full Text] - Feeney AJ. (1991) Junctional sequences of fetal T cell receptor ß chains have few N regions. J. Exp. Med. 174:115.
[Abstract/Free Full Text] - Candeias S, Waltzinger C, Benoist C, Mathis D. (1991) The VB17+ T cell repertoire: skewed JB usage after thymic selection; dissimilar CDR3s in CD4+ versus CD8+ cells. J. Exp. Med. 174:989.
[Abstract/Free Full Text]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  P. Wettstein, M. Strausbauch, T. Therneau, and N. Borson The application of real-time PCR to the analysis of T cell repertoires Nucleic Acids Res., December 1, 2008; 36(21): e140 - e140. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||