| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Departments of
*
Immunology and
Surgery, Mayo Foundation, Rochester, MN 55905
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
However, there has been limited investigation of the diversity and
characteristics of TCRs specific for individual minor HA peptides.
Until recently, only limited numbers of CTL clones specific for single
HA peptides have been selected so that characterization of these TCRs
has been impossible. The vast majority of studies of TCR diversity have
centered on T cell responses to exogenous peptides, e.g., cytochrome
c (7) and influenza nucleoprotein (8, 9). However, donor
minor HA are endogenous Ags that most likely differ from their
recipient homologues by limited, but significant, amino acid
interchanges; if recipient homologue peptides were to include specific
binding motifs for class I molecules (10), it is possible that these
recipient peptides are involved in the selection of responding T cells.
Therefore, TCRs that recognize these endogenous peptides may not be
comparable in diversity and characteristics with TCRs specific for
exogenous peptides that may not have comparably related host
counterparts. We have recently completed an analysis of TCR 
- and
ß-chains expressed by extensive panels of CTL clones that recognize
the H4 and CTT-1 immunodominant minor HA peptides (11, 12). The results
of these experiments indicate that specific V and/or Vß genes may
be overrepresented in CTL clones specific for single peptides, but
there is no clear requirement for expression of individual V or Vß
genes. The most distinctive characteristic of these TCRs is the
presence of charged residues in ß CDR3s; these panels were
characterized by net negative charges in the CDR3s with placement of
the charges at positions that may bind to specific amino acids in the
carboxy ends of class I-bound peptides or the -helices of the
presenting class I molecules (11).
The study and interpretation of diversity of TCR subunits specific for the H4 and CTT-1 peptides are hindered by the fact that the sequences of these peptides are unknown. This lack of sequence data impedes the analysis of the interactions between CDR3s and peptide:class I complexes. The only minor HA peptides identified and sequenced to date in mice are derived from two male-specific (HY) Ags recognized by CTL; one peptide is presented by Kk molecules (5) and the second is presented by Db molecules (6, 13). The HY model system should provide the opportunity to analyze the interactions between minor HA peptide:class I complexes and TCR subunits and identify characteristics of each component that facilitate CTL recognition.
Since a number of CTL clones that recognize HY in the context of Db molecules have been described previously (14, 15), we chose to investigate the diversity of CTLs involved in HY-disparate skin allograft rejection. Such in vivo analyses are complicated by the complexities of T cell populations that infiltrate graft sites. These populations of graft-infiltrating T cells would be expected to include both Ag-specific and nonspecific bystander T cells so that the chosen experimental approach must allow differentiation between the ß-chains expressed by these two T cell populations. Considerable information can be gained by amplifying, subcloning, and sequencing TCR subunit-encoding transcripts derived from sites of inflammation/rejection, but this approach is handicapped by the potential for sampling errors. Previous studies have indicated that in vivo populations of Ag-specific T cells are characterized by reduction in diversities of Vß usage as well as ß CDR3 length in comparison with ß-chains expressed by bystander and normal T cell populations (16, 17). The technique of spectratyping has been developed to identify ß-chain transcripts that exhibit these characteristics and has been used successfully to investigate the diversity of TCRs in sites of acute graft vs host disease and tumors in humans (16, 18, 19). The method involves Vß gene-specific PCR amplification, followed by electrophoresis of the products to identify overrepresented Vß genes and the lengths of products that are dependent on CDR3 lengths. In a number of cases, reamplification with Jß-specific primers restricts the products sufficiently to obtain CDR3 sequences directly from PCR products (16). Accordingly, spectratyping provides a rapid scan of all ß-chain transcripts grouped according to the utilized Vß gene and the lengths of the products. The direct sequencing of PCR products minimizes Taq polymerase errors in comparison with sequencing of cloned PCR products that requires sequencing of cloned products from replicate PCRs to confirm sequences.
In this communication, we present the results of spectratype analysis of CTL-infiltrating second- and third-set HY-incompatible skin grafts. Sequencing of single-copy products revealed a net negative charge in CDR3s of chains carrying Vß8 subfamily members that were represented prominently, but the removal of Vß8+CTLs by Ab-mediated depletion resulted in not only the use of alternative Vß genes, but also the reversal of the net charges in the CDR3s.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Female and male C57BL/6ByJ (B6) mice were purchased from The Jackson Laboratory (Bar Harbor, ME). All mice were housed in the barrier facility in Mayo Clinic Division of Animal Medicine (Rochester, NY). The HY-specific and Db-restricted CTL-10 clone (15) was generously provided by Dr. Derry Roopenian, The Jackson Laboratory.
Skin grafting
Orthotopic tail skin grafts were transplanted according to the previously described technique (20). Each recipient of primary allografts to be scored for rejection time received a single autograft and two allografts from female donors. Skin grafts were scored at routine intervals for the condition of epidermal scale pattern, pigment, and hair; rejection was scored when no viable signs were observed for both allografts. Recipients of skin grafts for the purpose of priming for subsequent sets of allografts received two allografts that were scored at routine intervals to confirm rejection. Subsequent sets of skin grafts were then applied with one autograft and two allografts applied to the tail of each recipient.
In vivo Ab-mediated T cell depletion
CD4+ T cells were depleted by i.v. injection of GK1.5 mAb (21) in ascites fluid (1 mg total protein/recipient) 1 day before grafting and every 7 days thereafter (100 µg/recipient) until rejection. A comparable regimen was used for depletion of Vß8+ T cells with i.v. doses (the day before and 6 days after grafting) of 5 mg (total protein) of ascites fluid containing the F23.1 mAb (22); this amount was shown by flow cytometry to be double the amount required for >95% depletion of Vß8+ T cells in peripheral blood from 2C transgenic mice (23) (data not shown).
PCR primers
Primers were synthesized by Mayo Molecular Biology Core
Facility. All primers were purified on the basis of retention of the
5'-protecting group (dimethoxytrityl) to eliminate partially
synthesized oligonucleotides. Primers with the 5'-protecting group were
purified with oligonucleotide purification columns (Applied Biosystems,
Perkin-Elmer, Foster City, CA). Nested Cß region primers (Table I
) were designed to amplify the noncoding
cDNA strand and progress up the C region segment, with Nest I being
homologous to a region near the 3' end of the C region and Nest III
near the 5' end. CßSeq corresponds to the first 20 bp of the 5' end
of the Cß region. The CßSpectra primer is identical in sequence to
CßSeq, but is labeled with 6-carboxyfluorescein (6-FAM) (Applied
Biosystems) at the 5' end. Vß-specific primers are homologous to
specific regions of the Vß segment that distinguish different genes
and subfamilies (24) and are described in Table I.
|
cDNA was reverse transcribed from the CßNest I primer
using the Ampliwax PCR Gem protocol (Applied Biosystems). The lower
layer contained 1 µl 10x PCR buffer (Promega Corp., Madison, WI), 2
µl MgCl2 (25 mM), 2 µl of a dNTP mixture (1.25 mM
each), 10 pmol CßNest I, 7.5 U (0.25 µl) RNasin ribonuclease
inhibitor (Promega Corp.), and 2 U Moloney murine leukemia virus RT
(Life Technologies, Grand Island, NY). Ten nanograms of total RNA that
had been extracted directly from graft tissue and CTL-10 cells using
the Total RNA Extraction Kit (5'
3'; Boulder, CO) were added with
sufficient sterile water to achieve a final volume of 10 µl.
Synthesis of cDNA was performed in a thermal cycler (MJ Research,
Watertown, MA) by incubating at 37°C for 25 min and 42°C for 30
min. Five minutes before completion of the final incubation, a wax
pellet was added; the tube was incubated at 100°C for 5 min to
inactivate the Moloney murine leukemia virus RT, and the tube was
cooled to 25°C. Reaction tubes received an upper layer that consisted
of 33.25 µl sterile water, 2 µl MgCl2 (25 mM), 4 µl
10x PCR buffer (Promega Corp.), 10 pmol of Vß-specific primers in a
0.5 µl vol, and 1.25 U Taq DNA polymerase (0.25 µl;
Promega Corp.). The PCR conditions were 94°C, 1 min; 37°C, 1 min;
and 72°C, 2 min. After two cycles, the annealing temperature was
raised to 42°C. After an additional cycle, the annealing temperature
was raised to 55°C, and these conditions were maintained for 35
cycles, followed by a final extension of 8 min.
Further amplification was accomplished using nested PCR reactions. One microliter of the product from the first PCR was transferred to a fresh tube, and a nested PCR reaction was run with the substitution of CßNest II and subsequently CßNest III in a total of 50 µl (with appropriate scaling of components). The PCR conditions were 94°C, 1 min; 60°C, 1 min; and 72°C, 2 min for 35 cycles. PCR products were gel purified (25) before sequencing using the Wizard PCR Purification Kit following the manufacturer’s protocol (Promega Corp.). Sequencing was performed on an ABI200 automated sequencer at Mayo Clinic Molecular Biology Core Facility using CßSeq as the primer. Approximately 250 bp were sequenced for each PCR product.
Spectratyping
After nested PCR amplification using CßNest II was completed, 6 µl of the product was transferred to a fresh reaction tube containing 2.15 µl sterile water, 0.6 µl MgCl2 (25 mM), 1 µl 10x PCR buffer (Promega Corp.), 10 pmol Cß Spectra primer in a 0.05 µl vol, and 0.5 U Taq DNA polymerase (0.1 µl; Promega Corp.). Single-stranded primer extension was accomplished by 20 cycles of PCR under the following conditions: 94°C, 1 min; 60°C, 1 min; and 72°C, 2 min. Spectratype analysis was performed by mixing 0.75 µl labeled product, 1.25 µl formamide, 0.5 µl gel-loading buffer (50 mg/ml blue dextran, 25 mM EDTA), and 0.5 µl Genescan-500 TAMRA internal lane size standards (Perkin-Elmer). This solution was heat denatured for 3 min at 95°C, and 1.8 µl was loaded into each lane of a 4.75% polyacrylamide (acrylamide/bis ratio of 19:1), 6 M urea denaturing gel. The gel had a well-to-read length of 36 cm, used 0.22-mm spacers, and was run in an Applied Biosystems model 377 WTR DNA-sequencing machine for 2 h at 3 kV. Product length determination was performed using the Prism Genescan version 2.0.1 software package (Applied Biosystems), and data analysis was performed using the Genotyper 1.1 software package (Applied Biosystems).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Infiltration of allografts by recipient T cells expectedly involves the immigration of alloantigen-specific T cells accompanied by nonspecific T cells that are attracted by the inflammatory response. It has been reported that the majority of T cells present at inflammatory sites comprise the latter population that would expectedly be characterized by the diverse Vß usage and CDR3 length observed in normal T cell populations. However, alloantigen-specific T cells should also be present and that population would be expected to exhibit a relatively reduced level of diversity of Vß usage and CDR3 length as a result of clonal expansion (16, 17). Spectratyping was used to amplify ß-chain transcripts with Vß gene-specific primers and separate PCR products by electrophoresis to identify overrepresented Vß genes and those transcripts with reduced CDR3 length diversity. Total RNA was extracted from HY-disparate skin allografts that were judged to be in the process of rejection. cDNA was reverse transcribed with a Cß primer, and Vß gene-specific amplification was accomplished with nested Cß primers paired with primers specific for individual Vß genes. The amplified products were separated by gel electrophoresis, and the sizes of products were determined by comparison with molecular standards.
We have hypothesized that multiple, sequential sets of allografts incrementally reduce the diversity of Vß usage and CDR3 length exhibited by HY-specific CTLs. The decreased time of rejection observed with subsequent allografts should provide an advantage to specific memory T cells such that their frequency should increase in the infiltrating population. If this were the case, spectratyping should reveal ever-decreasing complexity in CDR3 length and Vß usage and focus attention on important Vß segments, which can be characterized for both common Vß gene usage between multiple individuals and CDR3 length restriction.
Three B6 mice were primed with HY-disparate allografts and received
GK1.5 treatment to eliminate CD4+ T cells following
complete rejection of the primary allografts. CD4+
depletion was performed to restrict spectratype analysis exclusively to
CTLs. Second-set grafts were then transplanted to decrease the
variability between rejection times (26) and reduce the diversity of
TCRs expressed by HY-specific CTLs, which was evident in comparisons
between primary and secondary responses in the H4 system (manuscript
submitted). Spectratyping was performed on CTLs infiltrating second-set
allografts as well as two autografts (Fig. 1). Vß2-,Vß5.2-, and Vß8-specific
amplification was observed with almost all recipients with accompanying
evidence of CDR3 length restriction. The CDR3 length was calculated as
the number of amino acids between the terminal Vß Cys residue and the
Jß GXG motif, as proposed previously by Rock et al. (27). No products
were obtained from the autograft on mouse 2, suggesting limited, if
any, T cell infiltration at the time of graft harvest. However,
products were obtained from the autograft on mouse 1, and the profile
was similar in Vß usage to the allograft on the same recipient, but
the CDR3 length composition was almost entirely different, suggesting
that the CTL infiltrates present in these two grafts were comprised of
partially overlapping sets of CTLs. This may not be surprising since
the allografts and autograft were adjacent on the tail (
3 mm apart)
and CTL trafficking through both grafts may have occurred.
|
. Only a single
sequencing attempt was successful. The sequence derived from peak A
from mouse 2 was characterized by inclusion of Vß8.2, a negative
charge at CDR3 position 5, a positive charge at position 6, and a
neutral Jß segment (Table II and Fig. 2). The ability to sequence a single
Vß8.2 PCR product with a single-length CDR3, which was the only
product obtained from that graft, suggests that progeny of only a
single Vß8.2+CTL clone had infiltrated the graft.
|
|
The evolution of the CTL response toward HY in the same individual
and the identification of CDR3 characteristics of anti-HY TCRs were
monitored in two additional B6 mice that received primary male grafts
followed by second- and third-set male grafts. When second-set
allografts were harvested, they were replaced with female syngeneic
grafts. The evolving CTL response to HY is shown in Figure 3. Vß5.2 and Vß8 were present in
the secondary responses of both mice, similar to that seen in Figure 1.
Additional Vß genes were observed, suggesting increased common Vß
segment usage in comparison with the response described in Figure 1, but CDR3 length diversity was still generally restricted to three or
fewer components in these segments. The autograft control for mouse 5
revealed no amplification of any Vß subfamilies, suggesting there
were limiting numbers of infiltrating T cells (data not shown). The
results of spectratyping ß-chains following third-set grafts
demonstrated that Vß segment usage was further restricted to Vß4,
Vß5.2, and Vß8 genes (Fig. 3). It is also interesting to note that
CDR3 lengths seen in the third-set response were often, but not always,
present in the corresponding secondary response, suggesting that single
TCRs may not gain a selective advantage and dominate subsequent
rejection episodes in this minor HA system. However, direct sequencing
of individual Vß segments revealed a single CDR3 sequence in mouse 4
that was present in both second-set (C) and third-set (I) responses
(Fig. 2 and Table II). This result strongly suggests that the CTLs
expressing this ß-chain were involved directly in rejection of both
second- and third-set male allografts on this recipient. Discovery of
additional conserved sequences was most likely precluded by the
presence of two or more components in the PCR products.
|
), and the CDR3
sequences are included in Figure 2. Eight of the ten sequences were
obtained from Vß8 subfamily PCR products, despite sequencing attempts
on multiple non-Vß8 products. The ability to obtain single-copy
sequences directly from Vß8 products suggested that 1) the number of
different transcripts derived from these allografts was restricted
despite apparent CDR3 length diversity within the Vß8 PCR products,
and 2) there was selective expansion of CTL-expressing Vß8 genes,
presumably due to a specific response to HY rather than nonspecific
inflammation. All but one of these sequences contained negative charges
at either position 5 or 6, and all sequences contained at least one
negative charge at these or other positions. In addition, 80% of Jß
segments bore a negative charge, which is greater than the 58% usage
expected from unbiased Jß usage (28). Eighty percent of all sequences
contained two or more negative charges, which is distinctly different
from characteristics of CDR3s specific for a Db + influenza
nucleoprotein peptide (8, 9), but similar to H4- and CTT-1-specific ß
CDR3s (11, 12). Only 30% of ß CDR3 sequences carried a positive
charge, and two of the three sequences carrying positive charges also
contained two negative charges; the ratio of charged residues was 19
negative/3 positive. The other distinguishing characteristic of these
CDR3s was the inclusion of a Gly residue at position 8 in 80% of
sequences, which may either serve as a spacing component, a pocket for
a bulky peptide-derived (or class I-derived) side chain, or chance
placement of a preferentially encoded residue due to preferential
guanine insertion by terminal deoxyuucleotidyl transferase in J
regions. CTLs infiltrating HY allografts on F23.1-treated recipients
As the second- and third-set responses toward HY +
Db exhibited prominent Vß8 subfamily usage with CDR3
length restriction, we assessed the effects of F23.1 Ab-mediated
depletion of Vß8+ T cells on HY-disparate allograft
survival. Ten B6 female recipients that had rejected primary male
allografts were injected i.v. (1 day before and 6 days after grafting)
with 5 mg total protein from ascites fluid containing the F23.1 mAb;
this amount had previously been shown to deplete >95% of peripheral
blood T cells from 2C transgenic mice (23) (data not shown). The
F23.1-treated mice and 10 normal B6 females (having rejected primary
allografts) were grafted 2 days later with second-set male allografts.
There was no significant difference between the median survival times
(MST) of F23.1-treated (MST = 8 days; range = 6–11 days) and
normal (MST = 8.5 days; range = 7–11 days) recipients (Table III), indicating that Vß8+
T cells were not required for rejection of second-set male
allografts.
|
). Five of the eight recipients showed
no evidence of Vß8 transcripts, indicating complete Vß8-specific
depletion. Vß8-specific amplification was observed in three
recipients of F23.1 treatment, indicating that Vß8 depletion was not
complete in these recipients (Fig. 4). In two of these failures (mice 8
and 10), only single-length Vß8.1 products were obtained in
comparison with the above-described recipients of second- and third-set
allografts, in which at least two Vß8 subfamily members were observed
in four of five recipients. A restricted Vß gene usage profile was
observed in the five Vß8-depleted mice, with only two to four Vß
subfamilies utilized in any single individual. Products from Vß2- and
Vß5-specific amplification were identified, which was characteristic
of the response of untreated individuals. More interestingly, all
depleted mice also utilized Vß7, which was a prominent component
in only one of the previously described, untreated recipients. The
Vß7-specific products had very restricted CDR3 lengths (one or two
lengths), and single Vß7 sequences were obtained from four of five
individuals (included in Table II). RNA from autografts on two
recipients were spectratyped and no products were obtained (Fig. 4).
|
. Contrary to CDR3 sequences derived
from allografts on normal female recipients, there appeared to be two
groups of CDR3s distinguished by the charges of amino acids at
positions 4–6. The first group included Vß8- and Vß2-specific
products that all included negatively charged groups at these
positions, as described in Figure 2 for CDR3s derived from allografts
on untreated recipients. Furthermore, two of three of the Vß8-linked
CDR3s, derived from recipients with incomplete Vß8 depletion,
included a Gly residue at position 8. The charge distribution of this
group of sequences is contrasted by that of the apparent, second group
that included Vß7 genes, in which the majority of sequences included
positively charged residues at positions 4–6. The majority of the
remaining sequences in this group included polar residues at these
positions. It would appear that treatment with anti-Vß8 Ab
increased the frequency with which CDR3s in the second group were
observed. However, complete elimination of Vß8+CTLs did
not result in a complete shift away from CDR3s with negatively charged
residues at positions 4–6, since Vß2-specific products retained
these charges.
|
- and ß-chains of the first described clone
have been sequenced (14); this clone expresses Vß8.2, and the CDR3
sequence is CASGDNSAETLYF, which is similar in negative charge
distribution to Vß8-joined CDR3s expressed by CTLs that infiltrate
male allografts. We have now identified the expressed Vß gene and
sequenced the ß CDR3 of the CTL-10 clone that was used to identify an
HY peptide presented by Db molecules (6). Total RNA from
CTL-10 cells was amplified with Vß-specific primers and Cß primers.
PCR product was amplified with only the Vß7 primer, and the sequence
of the CDR3 was obtained (Fig. 6). The
CDR3 joined to Vß7 in this CTL clone included an amino acid (Ser) at
position 5 capable of forming hydrogen bonds, which suggests it is
similar to CDR3s derived from male allografts on F23.1-treated
recipients.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Spectratype analysis highlighted the prominence of ß-chains utilizing
Vß8 gene segments in CTLs infiltrating second- and third-set male
allografts. Although Vß8 subfamily members are the most frequently
expressed Vß genes in normal mice of C57BL strains (9), the
prominence of Vß8 PCR products and the ability to obtain multiple,
single-copy sequences from these products indicated specific
overrepresentation of Vß8 gene usage in these rejecting allografts.
In fact, PCR products derived from a second-set allograft on one female
B6 recipient were confined to a single Vß8.2 ß-chain, suggesting
monoclonality in this individual response. Furthermore, identical
ß-chains including Vß8.1 were recovered from PCR products derived
from both second- and third-set allografts on an individual recipient.
The relatively low probability of sequencing identical ß-chains from
two different allografts supports the contention that CTLs expressing
this ß-chain were involved in HY-specific, allograft rejection. The
proposed prominence of Vß8 expression in the in vivo response to HY
is supported by the fact that one of the two reported HY-specific CTL
clones (14) expresses Vß8.2, and a second unpublished clone expresses
Vß8.3 (O. Lantz, J. Ridge, and P. Matzinger, personal observation). A
disadvantage of spectratyping is, unfortunately, the focus on ß-chain
transcripts to the exclusion of analysis of V gene usage due to the
relatively large number of V genes. Even if V analysis were
feasible, it would still be impossible to pair - and ß-chain
expression in individual CTLs in vivo, as is possible with analysis of
individual CTLs that are cloned in vitro.
The restriction to dominant Vß8 gene usage following second- and third-set male allograft placement was more pronounced than we have observed previously for H4-incompatible allografts (manuscript submitted). Vß8 usage was prominent in H4-incompatible grafts, but it did not dominate spectratypes of CTLs infiltrating second- and third-set grafts in which a number of Vß genes were observed that were not present in HY-incompatible grafts. Therefore, we attempted to suppress rejection of HY-incompatible allografts with anti-Vß8 Ab. This approach has been successful at retarding the rejection of allografts expressing H2Ld that has been shown to be preferentially recognized by Vß8+CTLs in vitro (29). Anti-Vß8 treatment had no effect on the speed of male allograft rejection, so it was not surprising that new Vß genes predominated in allografts rejected by these recipients. It appears that Vß8 usage effectively excluded Vß7 in most recipients, in that Vß7 was only seen in one of five normal female recipients that expressed Vß8 transcripts. However, when Vß8+ T cells were effectively depleted, a Vß7 product with limited length heterogeneity was found in each of five recipients, suggesting that the expression of Vß7 by male allograft-infiltrating CTLs was dependent on elimination of Vß8+ T cells.
A surprising result was the change in net charge and charged residue placement in ß CDR3s following anti-Vß8 depletion. ß CDR3s of CTL-infiltrating male allografts in normal female recipients were similar to those of H4-specific CTLs in exhibiting a net negative charge (11). This characteristic was common to H4-specific CTLs that were cloned in vitro or harvested with rejecting allografts (11, 12, manuscript submitted). Furthermore, ß CDR3s derived from male allografts were characterized by strong representation of negatively charged residues at positions 4–6, which differs from H4-specific CDR3s. This specific negative charge placement also occurs in CDR3s from the two Vß8-expressing CTL clones with sequences CASGDNSAETLYF (14) and CASSDLVEVFF (O. Lantz, J. Ridge, and P. Matzinger, personal observation), which links this characteristic to in vitro derived, HY-specific CTL clones. It is apparent that the majority of these CDR3s linked to Vß8 genes included the Asp residue at position 5 due to a single codon shift in the site of recombination, since all three germline Vß8 genes include this codon following the CASS element (30). However, additional, negatively charged amino acids at this and adjacent CDR3 positions appear to have been the result of N nucleotide additions. This net charge and placement of charged residues does not appear to simply be a feature of Db-restricted CTLs, since two previously reported panels of influenza nucleoprotein-specific CTLs expressed ß CDR3s with limited representation of any charged residues (8, 9).
The charge distribution in ß CDR3s could be altered by pretreatment of female recipients with anti-Vß8 Ab. This treatment reduced the frequency of Vß8-specific products; those that remained still exhibited the charge distribution that was characteristic of ß CDR3s from untreated recipients. However, there was observed to be a considerable increase in ß CDR3s with either positively charged or polar amino acids, the latter of which could be expected to be involved in hydrogen ion pairing. This characteristic is shared with the CTL-10 clone that expresses Vß7 and the CASSSGNTLYF CDR3 sequence. Therefore, Vß8 depletion results in not only the usage of other Vß genes, but also alters the net charges of the expressed ß CDR3s. It should be stressed that this alteration is not complete, in that ß CDR3s with negatively charged residues in positions 4 to 6 can still be found with non-Vß8 genes, such as Vß2, even after complete Vß8 depletion.
The basis for the shift in net charge of ß CDR3s is not clear at present, but may be related to the peptide(s) recognized by HY-specific CTLs. An HY peptide presented by Db was predicted by identified mimotopes (13) and, more recently, has been identified and sequenced (6). This peptide, WMHHNDLII, was found through its recognition by the CTL-10 clone. As discussed above, the ß CDR3 of this clone is similar to those regions derived from male allografts on Vß8-depleted recipients, in which positively charged or polar amino acids are prevalent in positions 5–6. We have speculated previously that the negatively charged residues in H4-specific ß CDR3s are involved in salt bridges with a positively charged residue in the carboxy end of the H4 peptide (11) that we have recently supported through H4 mimotopes that include a positively charged amino acid at position 7 (data not shown).
It is possible that the same interaction or hydrogen bonding facilitates the binding of specific TCRs to the WMHHNDLII peptide. However, CDR3s joined to Vß8 genes exhibit a strong net negative charge that would not appear compatible with such binding to the carboxy end of this peptide. The recent crystal structures of mouse and human TCRs (31, 32) suggest that ß CDR3s are not in positions that allow binding of ß CDR3s to the His residues of this peptide. It would appear that either the CDR3s of HY-specific Vß8 chains are not required for binding, or another peptide may be recognized by Vß8+CTLs. The former possibility appears unlikely given the marked, net negative charge and concentration of negative charges at positions 5–6 of these ß CDR3s that are not observed in other Db-restricted TCR (8, 9). Support for the latter possibility comes from the observation that CTLs expressing the published Vß8+ HY-specific TCR (14) do not recognize the WMHHNDLII peptide (E. Simpson, personal observation). If this were true, it would imply that anti-Vß8 depletion resulted not only in a change in Vß usage, but also shifted the response to a second male-specific peptide that is recognized by CTLs that preferentially use the Vß7 gene and express ß CDR3s with characteristics, i.e., positively charged residues, favoring binding to that second peptide. Experiments are in progress to determine whether 1) HY-specific CTL clones expressing Vß8 and Vß7 actually recognize different peptides that can be eluted from Db molecules and be separated chromatographically, and 2) a change in recognition of these peptides occurs following Vß8 depletion in vivo. The results of these experiments will expectedly increase our understanding of the mechanisms underlying preferential T cell responses to immunodominant Ags.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Peter J. Wettstein, Department of Surgery, Mayo Clinic, Rochester, MN 55905.
3 Abbreviations used in this paper: HA, histocompatibility antigen; CDR3, complementarity-determining region 3; MST, median survival time; RT, reverse transcriptase.
Received for publication August 12, 1997. Accepted for publication December 8, 1997.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
ß T-cell receptors ’see’ peptide/MHC complexes. Immunol. Today 14:597.[Medline]
ß T cell receptor structure at 2.5A and its orientation in the TCR-MHC complex. Science 274:209.This article has been cited by other articles:
|
|
 |
|
  P. J. Wettstein, M. Strausbauch, and N. Borson Repertoires of T cell receptors expressed by graft-infiltrating T cells evolve during long-term recall responses to single minor histocompatibility antigens Int. Immunol., April 1, 2007; 19(4): 523 - 534. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Baron, I. McMorrow, D. H Sachs, and C. LeGuern Persistence of Dominant T Cell Clones in Accepted Solid Organ Transplants J. Immunol., October 15, 2001; 167(8): 4154 - 4160. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
