T Cell Receptor Expression and Differential Proliferative Responses by T Cells Specific to a Myasthenogenic Peptide

CELLULAR IMMUNOLOGY ARTICLE NO.

180, 20–28 (1997)

CI971171

T Cell Receptor Expression and Differential Proliferative Responses by T Cells Specific to a Myasthenogenic Peptide1 Susan L. Kirshner, Ari Waisman, Einat Zisman, Avi Ben-Nun, and Edna Mozes Department of Immunology, Weizmann Institute of Science, Rehovot, Israel 76100 Received February 17, 1997; accepted June 27, 1997

an autoimmune disease whose symptoms are mediated by Ab that bind to the nicotinic acetylcholine receptor (AChR) of the neuromuscular junction (4). Several lines of experimental evidence indicate that helper T cells are required for the induction and maintenance of the autoantibody response in MG: (i) AChR-reactive T cells occur with higher frequency in MG patients and in experimental autoimmune myasthenia gravis-afflicted rodents than in normal subjects (5). (ii) CD40/0, CD80/0, and CD40/0 CD80/0 C57B1/6 mice are resistant to the induction of experimental autoimmune MG (6). (iii) MHC class II knock-out mice do not develop experimental autoimmune MG (7). (iv) T cells that provide help to anti-AChR Ab-producing B cells have been isolated from humans and mice (8, 9). (v) Decreased T cell autoreactivity has been correlated with clinical improvement in an MG patient experimentally treated with anti-CD4 Ab (10). Our laboratory has previously reported that peptides representing sequences 195–212 and 259–271 (p195– 212 and p259–271) of the human AChR could induce PBL from a high percentage of MG patients to proliferate. Proliferative responses to the peptides were correlated with the expression of HLA DR5 and DR3, respectively. In addition, Ab recognizing these peptides were detected in sera from the MG patients (11, 12). Moreover, studies with Torpedo-derived AChR-immunized mice demonstrated that p195–212 and p259–271 were immunodominant T cell epitopes in the SJL and BALB/ c mouse strains, respectively, whereas they were cryptic epitopes in the C3H.SW mouse strain (13). To further study of the role of autoreactive T cells in MG-related autoimmune manifestations, T cell lines specific to p195–212 and to p259–271 were established in SJL (TCSJL195-212) and BALB/c (TCBALB/c259271) mice. We found that MG-related autoimmune manifestations, namely anti-AChR Ab titers and compound muscle action potential decrements, could be induced in naive mice following their inoculation with activated syngeneic Ag-specific T cell lines (14). Thus, peptides representing human T cell epitopes also induced MG-related pathogenic responses in some mouse strains.

Myasthenia gravis (MG) is a T-cell-regulated autoimmune disease in which a pathological autoantibody response is mounted against the nicotinic acetylcholine receptor of the neuromuscular junction. Our laboratory previously identified a T cell epitope, p195–212, derived from the human acetylcholine receptor a subunit, which triggered PBL to proliferate from about 70% of MG patients tested. p195–212 was also found to be an immunodominant T cell epitope in SJL mice and a cryptic epitope in C3H.SW mice. Inoculation of naive SJL mice with cells from a p195–212-specific syngeneic T cell line caused MG-related autoimmune manifestations in those mice. In these studies we analyzed TCR a and b chain sequences used by T cell lines and clones from both high- and low-responder mouse strains in response to p195–212. T cell lines generated from either strain expressed single TCR Vb gene segments (Vb17 in SJL mice and Vb8 in C3H.SW mice). By deleting Vb17-expressing T cells in p195–212-immunized SJL mice we established a T cell line that expressed the Vb6 gene product, suggesting that SJL mice are not limited to using a single Vb gene segment in response to p195–212. In addition, we found that N- and/ or C-terminal-truncated peptides of p195–212, presented under the same culture conditions to different clones bearing the same TCR ab chain, could elicit very different proliferative responses from the clones. Thus, even within a constrained system, factors other than TCR sequence contribute to the differential stimulation of T cell responses. q 1997 Academic Press

INTRODUCTION T cells have long been implicated in the etiopathology of autoimmune diseases in their capacity as direct mediators of tissue destruction and/or as regulators of immune responses (1–2). Myasthenia gravis (MG)2 is 1 This research was supported in part by the Basic Research Foundation administered by the Israel Academy of Sciences and Humanities. 2 Abbreviations used: myasthenia gravis, MG; acetylcholine receptor, AChR.

20

0008-8749/97 $25.00 Copyright q 1997 by Academic Press All rights of reproduction in any form reserved.

AID

CI 1171

/

6c27$$$$$1

09-11-97 08:16:38

cia

TCR USAGE IN RESPONSE TO A MYASTHENOGENIC PEPTIDE

To further elucidate the role of p195–212-specific T cells in the etiopathology of MG-related autoimmune responses we analyzed TCR sequences from p195–212specific T cell lines (n Å 3) and clones (n Å 11) originating from the high-responder strain SJL and T cell lines (n Å 2) originating from the low-responder strain C3H.SW. These studies revealed that SJL mice primarily utilized the TCR Vb17 family when responding to p195–212. Similarly, lines originating from the lowresponder strain were dominated by a single Vb family, Vb8. Nevertheless, SJL mice in which TCR Vb17 expression was suppressed could still respond to p195– 212 by recruiting cells that expressed the TCR Vb6 gene family. Although all the clones from one of the SJL-derived T cell lines had identical TCR sequences, some of the clones differed in their ability to proliferate to a panel of N- and/or C-terminal, truncated sequences of the native peptide. Therefore, we suggest that factors other than TCR sequence contribute to the differential stimulation of T cell responses. MATERIALS AND METHODS Antigens. The synthetic peptides of the human AChR a-subunit p195–212 (DTPYLDITYHFVMQRLPL), N- and/or C-terminal truncations of p195–212 (14), and p259–271 (VIVELIPSTSSAV) were synthesized and characterized as previously described (15). Establishment of antigen-specific T cell lines and clones. T cell lines specific to p195–212 were established from popliteal LN cells of immunized SJL or C3H.SW mice according to previously published methods (16). LN cells from five mice were pooled to establish each line. Briefly, cells (30 1 106/flask) were cultured in enriched RPMI 1640 medium containing 1% normal syngeneic mouse serum and 50 mg/ml of p195– 212. After 4 days of incubation, cells were washed and resuspended in 5 ml enriched medium supplemented with 10% FBS and 10% Con A supernatant. Cells were exposed to the stimulating peptide (50 mg/ml) presented on irradiated syngeneic spleen cells every 14 days. Three T cell lines were established for these studies, two from SJL mice (TCSJL195-212 and TCSJL195212.2) and one from C3H.SW mice (TCSW195-212.2). A previously reported T cell line of C3H.SW origin, TCSW195-212, was also included in these studies (16). The T cells were cloned by the limiting dilution technique (17). The line was cloned on two separate occasions, once on the occasion of the 4th boost and once on the occasion of the 12th boost. Typically, cells of the line were diluted to 1 cell/ml, and 200 ml was alliquoted into wells of 96-well plates. Ten clones were obtained during the first cloning and 15 during the second cloning. TCR genes from 11 clones were sequenced.

21

to specific TCR Vb families (50 ml supernatants): RR47 [Vb6 (18)], F23.1 [Vb8.1 / 8.2 / 8.3 (19)], RR3-15 [Vb11 (20)], KJ23a [Vb17a (21)], or the pan ab TCR Ab H57-597 (Jackson Immunoresearch, West Grove, PA). Thereafter, the cells were washed twice with a cold solution of 5% FBS in PBS–0.05% sodium azide and incubated with fluorophore-conjugated Ab specific to the species and isotype of the first antibody. All incubations were for 30 min on ice in the dark. Fluorescence of at least 5000 cells/sample was assessed by a FACScan analyzer and Lysis software (Becton–Dickinson, Mountain View, CA). Proliferative responses of T cell lines and clones. Seven or 10 days after antigenic stimulation, cells of the T cell lines or clones were tested for their specific proliferative responses. Cells (104 per well) were cultured with 0.5 1 106 irradiated (3000 rad) syngeneic spleen cells in the presence of different concentrations of the peptides. Cultures were established in 200 ml enriched medium containing 10% FBS in flat-bottom microtiter plates. At the end of a 48-hr incubation period, 0.5 mCi of [3H]thymidine (5 Ci/mmol; Nuclear Research Center, Negev, Israel) was added. Sixteen hours later cells were harvested and radioactivity was counted (16). Cellular RNA isolation. Total RNA from T cells (lines and clones) was isolated using Tri reagent (Molecular Research Center, OH), according to the manufacturer’s instructions. Briefly, 1–2 1 106 cells were separated on a Ficol/gradient, washed, and homogenized in 1 ml of reagent. The cells were always taken either from frozen stock or 2 weeks after the antigenic boost, a time when APC were no longer present (15). After homogenization, RNA was extracted with phenol/ chloroform and precipitated with ethanol. cDNA preparation. For preparation of cDNA from total RNA poly(dT)15 primer (Boehringer-Mannheim, Germany) was used. First-strand cDNA was synthesized using reverse transcriptase (USB, Cleveland, OH) and half of the RNA obtained from the cellular RNA isolation (see above). Briefly, 0.5 to 1 ml primer was added to 16.5 ml of the RNA preparation, incubated at 557C for 0.5 hr, and then allowed to cool slowly to 417C. Thereafter, 2 ml 101 RTC buffer (1 M Tris, pH Å 8.15, at 417C, 1 M MgCl2r6H2O, 2 M KCl2 , 1 M DTT), 10 U RNasin (Promega, Madison, WI), and 10 U reverse transcriptase were added and the sample was incubated 1–2 hr at 417C. Thereafter, 180 ml H2O was added to each sample and the samples were frozen until further analysis.

Indirect immunofluorescent staining for cell surface proteins. T cells (106/sample) were incubated with Ab

Polymerase chain reaction (PCR). The cDNA was subjected to PCR amplification, as described (21). Briefly, forward primers located in the constant region and backward primers located in the V region were

AID

cia

CI 1171

/

6c27$$$$$2

09-11-97 08:16:38

22

KIRSHNER ET AL.

used for amplification using Taq DNA polymerase (Promega). Samples were amplified for 30 cycles: 957C for 1 min, 547C for 2 min, and 727C for 3 min using a Perkin–Elmer Cetus DNA thermal cycler. The sequences of the primers for the murine T cell Ca, Cb, Va, and Vb genes have been previously published (23). Seventeen primers specific for different Vb genes and 20 primers specific for different Va genes were used as the forward primers. A single Cb or Ca primer was used as the reverse primer. Southern blots. Southern blots of PCR products were carried out as described before (22). Briefly, PCR products of T cell receptor cDNA Va or Vb genes were run on 2% agarose gel. Thereafter, the gel was treated with 0.5 N NaOH and neutralized, and DNA was transferred to Hybond-N nylon sheets (Amersham, UK). The blot was prehybridized (61 SSC, 51 Denhardt’s solution, 0.05% sodium pyrophosphate, 0.5% SDS, 100 mg/ ml herring sperm DNA, boiled) at 377C for 2 hr and then hybridized (61 SSC, 11 Denhardt’s solution, 100 mg/ml yeast tRNA, 0.05% sodium pyrophosphate) overnight at 427C with internal constant primer (Cb1 or Ca1) end-labeled with g-ATP (40 1 106 cpm). Blots were washed (twice at room temperature in 61 SSC, 0.05% sodium pyrophosphate for 10 min; twice at 427C 21 SSC, 0.05% sodium pyrophosphate for 30 min) and exposed to X-ray sensitive film. Cloning and sequencing of TcR a and b chains. To sequence the dominant TCR Vb genes used by cells of the p195–212-specific T cell lines TCSJL195-212.2, TCSJLKJ, and TCSW195-212, the PCR products that were obtained as described above were cloned into the M13 vector. Briefly, M13mp18 or mp19 DNA (Boehringer-Mannheim) was cut by either the SmaI or the SalI restriction enzyme. The PCR products were separated from the primers using Primer Remover (Advanced Genetic Technologies, MD), ligated into the DNA, and inserted into competent bacterial cells by electroporation. Positive colonies were picked using specific internal primers that were end-labeled with gATP. Positive colonies were grown, and ssDNA was prepared as described previously (22). From each line 20 colonies were derived and sequenced either by the dideoxy termination method (24) using Sequenase 2.0 kit (USB) or by an automated sequencer (Applied Biosystems) with 100 ng DNA. For the TCSJL195-212 line and T cell clones derived from that line, the PCR products were sequenced directly using an automated sequencer. Sequences were analyzed using Applied Biosystems and GCG package programs.

FIG. 1. SJL-derived p195–212-specific T cell lines and clones predominantly express the TCRVb17 gene product: fluorescence analysis of TCR Vb expression. (A) Cells of the TCSJL195-212 line. (B) Cells from two representative clones of the TCSJL195-212 lines. (C) Cells of the TCSJL195-212.2 line. Cells (106 cells/sample) were incubated for 0.5 hr on ice with 7.5 mg of one of the mAb preparations. Thereafter, the cells were washed and incubated for another 0.5 hr on ice with FITC-conjugated rabbit anti-mouse Ig (diluted 1:40). Binding was monitored by FACS.

p195–212-specific T cell lines from SJL mice predominantly use the TCR Vb17a gene product. The myasthenogenic T cell line TCSJL195-212 was previously

reported to proliferate specifically in response to p195– 212 and to be I-AS restricted (14). To further elucidate the T cell–p195–212 interaction, we determined the predominant TCRVb gene family expressed by this line using FACS analysis. To this end, cells of the TCSJL195-212 line and cells of 11 clones of that line were stained with different anti-Vb-specific mAb on two to five different occasions. Figure 1A depicts a representative FACS profile of the line, whereas Fig. 1B shows the FACS profiles of 2 representative clones. As can be seen there was a clear shift toward higher fluo-

AID

cia

RESULTS

CI 1171

/

6c27$$$$$2

09-11-97 08:16:38

23

TCR USAGE IN RESPONSE TO A MYASTHENOGENIC PEPTIDE

SJL predominantly expressed the TCR Vb17a gene product.

FIG. 2. Differential proliferative responses of cells of the TCSJL195-212 line to N- and/or C-terminal-truncated peptides of p195–212. Cells were incubated in the presence of various concentrations of each peptide (1–50 mM) along with irradiated syngeneic APC for 48 hr. [3H]Thymidine was added to the wells for a further 16-hr incubation. Thereafter, cells were harvested and thymidine incorporation was measured. The data are presented as the maximum proliferative responses of the clones to each peptide relative to their maximum proliferative responses to p195–212 (% efficacy).

rescence intensity by the cells incubated with the antiVb17 mAb KJ23a, with 81% of the cells labeled above background. There was no significant binding to the control anti-Vb8 mAb F23.1. In addition, it can be seen that cells of the line also bound the mAb L3T4 (antimurine CD4). Similarly, all of the clones of the TCSJL195-212 line were found to express the TCR Vb17 gene product. Thus, cells of the TCSJL195-212 line are CD4/ and predominantly use the TCR Vb17 gene segment. To confirm that SJL mice predominantly utilize the TCR Vb17 gene family product in response to p195– 212, a second T cell line was established according to the same protocol used to generate the TCSJL195-212 line (this second line was designated TCSJL195-212.2). The TCSJL195-212.2 line proliferated specifically in response to p195–212 and was I-AS restricted and CD4/ (data not shown). Figure 1C shows that cells of the TCSJL195-212.2 line were stained 87% above background by the KJ23a mAb but not by the F23.1 mAb. Similarly an anti-Vb6 mAb did not bind to cells of either the TCSJL195-212 or the TCSJL195-212.2 lines (data not shown). In addition, p195–212-specific T cell lines were independently generated on two other occasions and were also found predominantly to express the TCR Vb17 gene product (M. Dayan, M. PaasRozner, and E. Mozes unpublished observations). Altogether these data indicate that p195–212-specific T cell lines generated from the high-responder mouse strain

AID

CI 1171

/

6c27$$$$$2

09-11-97 08:16:38

T cell clones that differ in the fine specificity of their proliferative response profiles have identical TCR Vb and Va sequences. T cell clones were generated from the TCSJL195-212 line on two separate occasions, once at the time of the 4th boost of the line and once at the time of the 12th boost of the lines. A total of 25 clones were obtained from the two cloning events. Clones 1– 8 presented below were generated during the first cloning, whereas clones 11, 13, and 14 were generated the second time that the line was cloned. Clones of the p195 – 212-specific T cell line TCSJL195-212 can be distinguished by their ability to proliferate in response to peptides representing truncated sequences of p195–212. Figure 2 depicts a summary of the proliferative response profiles of 11 clones of the TCSJL195-212 line to the panel of N-terminally truncated peptides. In this figure the maximum proliferative response of each clone to each peptide relative to the maximum response of each clone to p195–212 is plotted (% efficacy). The responses of the clones to p195–212 are presented in Table 1. As can be seen in Fig. 2 there are extreme differences in responsiveness to some of the peptides among the clones. For example, clones 1 and 3 proliferated to p200–212 with 85% efficacy and to p201–212 with over 100% efficacy. In contrast, most of the other clones responded to these two peptides with less than 20% efficacy and clones 8 and 11 did not proliferate at all in response to these two peptides. To determine whether the differences found between the clones were due to differences in their TCR, the expressed b chain gene from the 11 clones was se-

TABLE 1 Proliferative Responses of Clones of the TCSJL195-212 Line to p195-212 Clone 11 13 14 1 3 2 4 5 6 7 8

Background 1751 1544 2615 645 498 646 1115 913 931 2437 1319

{ { { { { { { { { { {

588 441 954 90 236 171 273 336 400 780 107

p195-212 17,708 16,602 42,794 34,205 43,653 83,867 35,815 36,024 74,497 33,419 52,534

{ { { { { { { { { { {

560 3997 987 1056 6609 5195 2908 4290 5538 4471 3998

Concentration (mM) 25 10 25 5 10 10 10 5 50 5 50

Note. Cells (104) of the clones were incubated in the presence of irradiated spleen cells from SJL mice (0.5 1 106) and various concentrations of peptide for 48 hr. Thereafter, [3H]thymidine (0.5 mCi of 5 Ci/mmol) was added and 16 hr later plates were harvested. Results are expressed as mean cpm of triplicate wells { SD at the concentration of optimal proliferative response.

cia

24

KIRSHNER ET AL.

TABLE 2 Identical TCR b Chain Sequences in 10 Clones Tested Vb17a A GCC

S AGC

N-Db-N S AGT

C

R GA

Q CAG

Jb2.5

G GGG

Q GGC

Q CAA

D GAC

T ACC

Q CAG

Y TAC

Note. TCR b chain sequences were determined for all the clones by automated sequencing of PCR-amplified cDNA.

quenced from PCR-amplified cDNA prepared from total RNA. Table 2 shows the CDR3 region sequences of the clones. All 11 clones expressed the same TCR b chain gene (Vb 17a-NDN-Jb 2.5) (26). The NDN region sequences could have come from either Db segment; therefore, we were unable to make an unambiguous germline assignment. To exclude the possibility that the identity of the sequences was due to contamination, the clones were sequenced from more than one preparation of cDNA and using different batches of reagents. In addition, we sequenced b chain genes from TCR Vb17-expressing T cells specific to other Ag and obtained different CDR3 region sequences. TCR Va gene use of the clones was analyzed with PCR-amplified cDNA. The a chain genes were amplified using 20 primers specific for different Va gene family products and a Ca primer as the reverse primer. Southern blot analysis showed that all 11 clones expressed message for Va 1 and 5 of the clones (clones 1, 4, 5, 11, and 13) had message for Va 4 as well. Figure 3 shows a representative blot of clone 4 that expressed both Va 1 and Va 4 message. The a chain sequences were determined and are shown in Table 3. All the clones had identical sequences for the Va 1 chain (Va 1-N-Ja NEW.03a; Va 1 subfamily 2 or 8) (25, 26). Similarly, the Va4 chain showed identity of sequence (Va 4-N-Ja 112-2; Va 4 subfamily 1) (25, 26) between the clones. A T cell can express functional message for two different a chain products (27). Therefore, in the absence of anti-Va1 and Va4 mAb we were unable to confirm whether one or both of the gene products were expressed at the cell surface. SJL mice can use other TCR Vb and Va gene products when responding to p195–212. Since the T cell response of the TCSJL195-212 line appeared extremely

FIG. 3. Southern blot analysis of TCR Va expression by clone 4 of the TCSJL195-212 line. cDNA was PCR amplified using 20 different Va-specific primers and a Ca-specific primer as the reverse primer. a chain expression was detected using end-labeled oligo primer from within the Ca region.

AID

CI 1171

/

6c27$$$$$3

09-11-97 08:16:38

restricted it was of interest to examine further the p195–212-specific T cell responses in SJL mice. To this end a T cell line was established from SJL mice in which Vb17 expression was depleted during their immunization with p915–212 by inoculation with the KJ23a mAb (this line was designated TCSJLKJ). Inoculation of SJL mice with the KJ23a mAb routinely resulted in popliteal lymph nodes with undetectable levels of Vb17 expression by FACS analysis (data not shown). Figure 4 depicts a representative dose-dependent proliferative response curve of the line. It can be seen that the TCSJLKJ line proliferated specifically in response to p195–212. In addition, the TCSJLKJ line was I-AS restricted and CD4/ (data not shown). Figure 5A shows that a TCR Vb17-specific mAb did not recognize cells of the TCSJLKJ line, although the pan b chain-specific mAb H57-597 stained all the cells of the line. Thus the TCSJLKJ line is p195–212 specific but does not utilize the TCR Vb 17 gene product. To determine the TCR Vb family used by the TCSJLKJ line, cDNA was prepared from total RNA. The b chain genes were amplified using 17 primers specific for different Vb gene family products and a Cb primer as the reverse primer. A single product of the expected size was amplified in the lane corresponding to the Vb6 gene product, as shown in Fig. 5B. The cell surface expression of the Vb6 gene product was confirmed by staining cells of the line with the antiVb6 mAb, RR4-7. Figure 5C shows a representative FACS profile of the TCSJLKJ line. All of the cells of the line were stained by the anti-Vb6 mAb RR4-7 with a clear shift to the right in fluorescence intensity. Cells of the line were not stained above background by the control, anti-Vb11 mAb RR3-15. Altogether the data presented in Fig. 5 suggest that the TCSJLKJ line predominantly expresses the TCR Vb6 gene product. Comparison of Vb genes between lines and strains. The TCR b chain genes for all three SJL-derived lines were sequenced and the results are presented in Table 4. Table 4 also presents sequence analyses of the TCR genes from two p195–212-specific T cell lines established from C3H.SW mice. Our laboratory previously reported that C3H.SW is a low-responder strain to p195–212 (13). In addition, one of the lines (TCSW195212) was previously reported as utilizing a TCR Vb8 gene family product as determined by FACS analysis

cia

25

TCR USAGE IN RESPONSE TO A MYASTHENOGENIC PEPTIDE

TABLE 3 Sequences of TCR a Chain CDR3 Regions of Clones of the TCSJL195-212 Line V a1 C TGT

A GCA

A GCT

N S AGT

E GAG

H CAT

JaNEW.03a

A GCA

L CTC

A GCA

N AAC

V a4 Y TAC

Y TAC

C TGC

K AAG

M ATG

I ATC

Ja112-2 A GCT

L CTG

S AGT

L CTA

R CGA

N AAT

N AAC

Y TAT

D GAC

Q CAG

G GGA

Note. TCR a chain sequences were determined for all the clones by automated sequencing of PCR-amplified cDNA.

(16). Germline Vb and Jb assignments are noted in the table, whereas N and P nucleotides are noted by bold lettering. The TCSJL195-212.2, TCSW195-212, and TCSW195-212.2 lines rearranged to the Db1 locus. For the TCSJL195-212 and TCSJLKJ lines the germline Db segment could not be given a definitive germline assignment. As shown in Table 4 there is no apparent CDR3 motif common to all the lines. Comparing the CDR3 regions within each strain reveals that the three lines derived from the high-responder strain are all very different from one another, whereas the two lines derived from the low-responder strain were very similar to each other. Thus, the CDR3 regions of the TCSJL195-212, TCSJL195-212.2, and TCSJLKJ lines were four, three, and one amino acids long and rearranged to the Jb2.5, Jb1.1, and Jb2.4 segments, respectively. Although the

FIG. 4. Specificity of proliferative responses of cells of the TCSJLKJ line. Cells were incubated in the presence of various concentrations of each peptide (1–50 mM) along with irradiated syngeneic APC for 48 hr. [3H]Thymidine was added to the wells for a further 16-hr incubation. Thereafter, cells were harvested and thymidine incorporation was measured. Results are expressed as mean cpm of triplicate wells {SD.

AID

CI 1171

/

6c27$$$$$3

09-11-97 08:16:38

two lines originated from the low-responder strain used two different members of the Vb8 family (Vb8.1 and Vb8.2), they both rearranged to the same Jb segment (Jb2.4) and had NDN regions that were three amino acids long with two amino acids in common (Thr-Gly). It is also notable that features found in the CDR3 regions of the two low-responder lines were also found in the high-responder lines. Thus, the TCSJL195-212

FIG. 5. The TCSJLKJ line expresses the TCR Vb6 gene segment. (A) FACS analysis of ab TCR and TCR Vb17 gene product expression. (B) Southern blot analysis of PCR products amplified from TCSJLKJ-derived cDNA. (C) Cells of the line are stained with the Vb6-specific mAb RR4-7 but not the Vb11-specific mAb RR3-15. Fluorescence analysis was performed as described in the legend to Fig. 1 with the exception that the secondary mAb was FITC-conjugated rabbit anti-rat. Southern blot analysis was performed as described in the legend to Fig. 3 with the exception that 17 b chain-specific primers and internal Cb probes were used.

cia

26

KIRSHNER ET AL.

TABLE 4 Comparison of the TCR Vb Sequences Expressed by p195-212-Specific Lines Vb TCSJL195-212 Vb17a TCSJL195-212.2 Vb17a TCSJLKJ Vb6 TCSW195-212 Vb8.1 TCSW195-212/2 Vb8.2

N-Db-N

A GCC

S AGC

S AGT

C

R GA

A GCC

S AGC

S AGT

C

A GCC

S AGC

S ATG

I ATA

A GCC

S AGC

S AGT

R AGG

T ACA

A GCC

S AGC

S AGT

G GGG

T ACA

P CC

Jb Jb2.5

Q CAG

G GGG

G GGC

Q CAA

D GAC

T ACC

Q CAG

Y TAC

G GGG

T ACA

G

V TT

A GCA

N AAC

T ACA

E GAA

Q CAA

N AAC

T ACC

L TTG

Y TAC

Jb2.4

G GGT

Q CAA

N AAC

T ACC

L TTG

Y TAC

Jb2.4

G GGT

Q CAA

N AAC

T ACC

L TTG

Y TAC

Jb2.4

A GCT

V GTC

F TTC

Jb1.1

Note. TCR b chain sequences were determined for the lines by sequencing PCR-amplified cDNA that had been cloned into an M13 vector. The lines used the Db 1 gene segment with the exceptions of the TCSJL195-212 and TCSJLKJ lines for which germline Db assignments could not be made. Non-germline-encoded nucleotides are indicated by bold lettering. The deduced amino acid sequence is indicated.

and TCSW195-212 lines both contained Arg-X-Gly in their NDN region. Since the Arg was not germline encoded for either of these lines it may be the result of Ag-driven selection. Similarly, the NDN regions of the low-responder lines are three amino acids long as is the NDN region of the TCSJL195-212.2 line. Finally the TCSJLKJ line and the two C3H.SW-derived lines all rearranged to the Jb2.4 gene segment. Thus the CDR3 regions of the low-responder line are a synthesis of traits found in the CDR3 regions of the high-responder lines.

Engagement of the TCR by the Ag–MHC complex is the first step toward specifically activating T cells. In this paper we analyzed TCR sequences of p195–212specific T cell lines originated from mouse strains that are high (SJL) and low (C3H.SW) responders to the peptide. These studies revealed that p195–212-specific T cell lines originated from both responder mouse strains were restricted in their usage of Vb gene segments. Nevertheless, the TCR b chain CDR3 regions of cells from lines originated from the high-responder strain were divergent. We also found that TCR sequence could not explain differences in the proliferative responses of T cell clones derived from one of the highresonder lines to peptides representing truncated sequences of the native peptide. These studies suggest that T cell responses may be modulated even within highly constrained TCR–Ag–MHC interactions. T cells expressing the TCRVb17 gene product were most commonly observed in lines established from SJL mice immunized with the peptide p195–212. The utilization of the TCRVb17 chain was seen in four different

SJL-derived T cell lines generated over the course of a number of years by three different members of this laboratory (Fig. 1 and M. Dayan, M. Paas-Rozner, and E. Mozes, unpublished observations). Similarly, lines established from the low-responder strain C3H.SW most commonly used a member of the TCR Vb8 gene family. From these data it may be suggested that the low-responder status of C3H.SW mice is not due to an inability to generate TCR that recognize p195–212, i.e., a hole in the TCR repertoire. It should be noted that SJL mice have a genomic deletion within the Vb locus that eliminates the Vb8 gene family, whereas C3H.SW mice delete TCRVb17expressing cells. This may in part explain the utilization of different Vb gene segments in the lines derived from these two strains. Moreover, p195–212-specific T cell lines could be established from SJL mice in which TCRVb17 expression had been suppressed, confirming that a variety of TCRVb genes can participate in a p195–212-specific T cell response. Restricted TCR Vb expression has been reported for mouse and rat models of autoimmunity, both for diseases in which T cells have an effector role and for diseases in which they have a helper role. Thus, in experimental autoimmune encephalomyelitis the TCR Vb8 gene family is used in the myelin basic protein response of PL/J and B10.PL mice (H-2u), whereas the TCR Vb17 gene product is used in the response of SJL mice (H-2S) (28, 29). Similarly, in experimental autoimmune MG the TCR usage of C57BL/6 mice is dominated by the Vb6 gene product (30). With humans some studies revealed greater TCR heteroclonality when freshly isolated T cells were analyzed than when T cells were expanded in vitro in an Ag-specific manner (31, 32).

AID

cia

DISCUSSION

CI 1171

/

6c27$$$$$3

09-11-97 08:16:38

TCR USAGE IN RESPONSE TO A MYASTHENOGENIC PEPTIDE

27

In addition, active multiple sclerosis lesions showed greater heteroclonality than chronic lesions, as well as TCR repertoire differences between lesions within the same person (33). These data may be explained by differences in experimental design. Thus, Ag-specific expansion of T cell lines and clones in vitro may select for specific TCR rearrangements. In contrast, active inflamatory sites may recruit bystander T cells that are not disease specific. In addition, the repertoire of disease-specific T cells may increase during the disease process as a result of epitope spreading (34). Our T cell lines were generated against and educated with a short peptide (18 amino acids) 10 days after immunization of naive mice. Thus, it is not surprising that the T cell response was of limited heterogeneity. Although each of the lines we sequenced was dominated by a single TCR b chain sequence, the CDR3 region sequences differed between the lines. Thus, SJL mice can generate more than one TCR that recognizes the peptide p195–212. These data also suggest that in vitro selective pressures may have contributed to the domination of each line by a single TCR sequence. TCRVb17 rearrangements in both thymic and peripheral T cells of naive mice were extensively analyzed by Candeis et al. (35). Comparing our Vb17 sequences to those data revealed that the TCSJL195212 line contained the most commonly observed features of Vb17 rearrangements, whereas the TCSJL195-212.2 line included less common features. Thus, Candeias et al. (35) observed that approximately 80% of both peripheral and thymic Vb17/ TCR in naive SJL mice joined with a member of the Jb2 cluster, as did cells of the TCSJL195-212 line. In contrast, the TCSJL195-212.2 line combined with Jb1.1. Jb1.1 rearrangement was reported to be enhanced in the 20% of both thymic and peripheral Vb17/ CD4/ T cells that rearranged to a member of the Jb1 cluster. Hence, although the TCSJL195212.2 line combined with the less commonly used Jb cluster, it used the most commonly found member of that cluster. A nonrandom pattern of amino acid residue usage in the NDN region was also reported by Candeis et al. (37). For example, Arg residues were reported as being present at a relatively high frequency within the CDR3 region (7% of amino acids). This is partially because the Db2 minigene encodes an Arg residue in one of its reading frames and partially because Arg is encoded by CGN. The latter is significant since there is a bias toward the introduction of C nucleotides as 5* N region insertions. The NDN regions of the TCSW195-212 line, originated from the low-responder strain, and TCSJL195-212 line, originated from the high-responder strain, encode Arg residues at the 5* NDN junction. Since neither Arg is completely germline the presence of this motif may be the result of Ag-driven selection. Nevertheless, immunization of SJL mice with the peptide

p195 – 212 resulted in TCR b chain CDR3 regions that had features commonly found in Vb17 rearrangements in naive mice. It might be suggested that the SJL mouse strain is a high responder to the peptide p195 – 212 because it is predisposed to respond to that peptide. Three lines of evidence based on these studies and studies previously reported from our laboratory support this hypothesis. First, the SJL and C3H.SW mice were originally defined as high- and low-responder strains to p195 – 212 based on the ability of LN cells from mice that had been immunized with the entire AChR to proliferate in response to p195 – 212 (13). Thus, LN cells from SJL mice proliferated better to p195 – 212 than to the immunizing Ag, whereas LN cells from AChR-immunized C3H.SW mice had almost no proliferative response to p195 – 212. Second, our laboratory previously demonstrated that the T cell lines generated from the high-responder mouse strain, SJL, were able to proliferate in response to p195 – 212 when the Ag was presented by syngeneic APC in which Ag processing had been inhibited (36). In contrast, the low-responder mouse strain-derived T cell lines were unable to proliferate when Ag was presented under those same conditions. These data suggest that p195 – 212 is not a cryptic epitope in SJL mice. Finally, in this paper we present data that p195 – 212-specific T cells generated from SJL mice utilize common rather than unique TCR elements, strengthening the hypothesis that SJL mice are predisposed to respond to the peptide p195 – 212. Although clones of the TCSJL195-212 T cell line expressed the same TCR a and b chain sequences, differences in their ability to proliferate in response to truncated p195–212 peptides were observed. Similarly, some of the clones differed in their ability to proliferate in response to p195–212 and their cytokine secretion profiles (S.L.K. and E.M., unpublished observations). Although some of the clones expressed message for two different a chain genes, the differences in proliferative responses between the clones cannot fully be explained by the use of different TCR ab combinations. In addition to the TCR–Ag–MHC interaction, T cells require at least one other signal to become activated. Moreover, multiple receptor–ligand interactions are involved in modulating the T cell response. Preliminary data from this lab suggest that 1 week after stimulation some of the clones differ in the amount of various molecules expressed at the cell surface (e.g., LFA-1, CD25, TCR). Thus the surface expression of LFA-1 ranged in mean fluorescence intensity from 71 to 233 log units, of ILRa from 17 to 45 log units, and of TCR from 9 to 21 log units (S.L.K. and E.M., unpublished observations). This suggests that even under conditions in which TCR primary sequence and Ag presentation are held constant, T cell signaling is coordinated with other regulatory pathways. Supporting this, Green et al. (37) have

AID

cia

CI 1171

/

6c27$$$$$4

09-11-97 08:16:38

28

KIRSHNER ET AL.

reported that homotypic LFA1/ICAM3 interactions led to decreased proliferative responses from activated T cells. The specific activation of T cells is essential for Ag-driven immune responses. Consideration of these findings may be useful when designing targeted immunomodulatory therapy for the treatment of diseases characterized by pathological Ag-specific T cell activation. REFERENCES 1. Ben-Nun, A., and Lando, Z., J. Immunol. 130, 1205, 1983. 2. Newsom-Davis, J., Harcourt, G., Sommer, N., Beeson, D., Wilcox, N., and Rothbard, J. B., J. Autoimmun. 2, 101, 1989. 3. Fricke, H., Mendlovic, S., Blank, M., Shoenfeld, Y., Ben, B. M., and Mozes, E., Immunology 73, 421, 1991. 4. Lindstrom, J., Shelton, D., and Fujii, Y., Adv. Immunol. 42, 233, 1988. 5. Conti, T. B., McLane, K. E., Raftery, M. A., Grando, S. A., and Protti, M. P., Crit. Rev. Biochem. Mol. Biol. 29, 69, 1994. 6. Zhang, G. X., Xiao, B. G., Bakhiet, M., van der Meide, P., Wigzell, H., Link, H., and Olsson, T., J. Exp. Med. 184, 349, 1996. 7. Kaul, R., Shenoy, M., Goluszko, E., and Christadoss, P., J. Immunol. 152, 3152, 1994. 8. Asthana, D., Fujii, Y., Huston, G. E., and Lindstrom, J., Clin. Immunol. Immunopathol. 67, 240, 1993. 9. Fujii, Y., and Lindstrom, J., J. Immunol. 141, 3361, 1988. 10. Ahlberg, R., Yi, Q., Pirskanen, R., Matell, G., Swerup, C., Rieber, E. P., Riethmuller, G., Holm, G., and Lefvert, A. K., Neurology 44, 1732, 1994. 11. Brocke, S., Brautbar, C., Steinman, L., Abramsky, O., Rothbard, J., Neumann, D., Fuchs, S., and Mozes, E., J. Clin. Invest. 82, 1894, 1988. 12. Zisman, E., Katz-Levy, Y., Dayan, M., Kirshner, S. L., PaasRozner, M., Karni, A., Abramski, O., Brautbar, C., Fridkin, M., Sela, M., and Mozes, E., Proc. Natl. Acad. Sci. USA 93, 4492, 1996. 13. Brocke, S., Dayan, M., Rothbard, J., Fuchs, S., and Mozes, E., Immunology 69, 495, 1990. 14. Kirshner, S. L., Katz-Levy, Y., Wirguin, I., Argov, Z., and Mozes, E., Cell. Immunol. 157, 11, 1994. 15. Katz-Levy, Y., Kirshner, S. L., Sela, M., and Mozes, E., Proc. Natl. Acad. Sci. USA 90, 7000, 1993. 16. Brocke, S., Dayan, M., Steinman, L., Rothbard, J., and Mozes, E., Int. Immunol. 2, 735, 1990.

AID

CI 1171

/

6c27$$$$$5

09-11-97 08:16:38

17. Axelrod, O., and Mozes, E., Immunobiology 172, 99, 1986. 18. Kanagawa, O., Utsunomiya, Y., Bill, J., Palmer, E., Moore, M. W., and Carbone, F. R., J. Immunol. 147, 1307, 1991. 19. Staerz, U. D., Rammensee, H., Benedetto, J. D., and Bevan, M. J., J. Immunol. 134, 3994, 1985. 20. Bill, J., Kanagawa, O., Woodland, D. L., and Palmer, E., J. Exp. Med. 169, 1405, 1989. 21. Kappler, J. W., Wade, T., White, J., Kushnir, E., Blackman, M., Bill, J., Roehm, N., and Marrack, P., Cell 49, 263, 1987. 22. Sambrook, J., Fritsch, E. F., and Maniatis, T., ‘‘Molecular Cloning: A Laboratory Manual,’’ Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, 1989. 23. Zisman, E., Sela, M., Ben-Nun, A., and Mozes, E., Eur. J. Immunol. 10, 2497, 1994. 24. Sanger, F., Nicklen, S., and Coulson, A. R., Proc. Natl. Acad. Sci. USA 74, 5463, 1977. 25. Arden, B., Clark, S. P., Kabelitz, D., and Mak, T. W., Immunogenetics 42, 501, 1995. 26. Kabat, E. A., Wu, T. T., Perry, H. M., Gottesman, K. S., and Foeller, K., ‘‘Sequences of Proteins of Immunological Interest,’’ United States Department of Health and Human Services (PHS, NIH), Bethesda, MD, 1995. 27. Malissen, M., Trucy, J., Jouvin, M. E., Cazenave, P. A., Scollay, R., and Malissen, B., Immunol. Today 13, 315, 1992. 28. Acha-Orbea, H., Mitchell, D. J., Timmermann, L., Wraith, D. C., Tausch, G. S., Waldor, M. K., Zamvil, S. S., McDevitt, H. O., and Steinman, L., Cell 54, 263, 1988. 29. Bourdette, D. N., Vandenbark, A. A., Hashim, G., Whitham, R. H., and Offner, H., J. Neuroimmunol. 22, 255, 1989. 30. Infante, A. J., Levcovitz, H., Gordon, V., Wall, K. A., Thompson, P. A., and Krolick, K. A., J. Immunol. 148, 3385, 1992. 31. Van Laar, J. M., Miltenburg, A. M., Verdonk, M. J., Daha, M. R., De Vries, R. R., Van den Elsen, P. J., and Breedveld, F. C., Clin. Exp. Immunol. 83, 352, 1991. 32. Ben-Nun, A., Liblau, R. S., Cohen, L., Lehmann, D., Tournier, L. E., Rosenzweig, A., Zhang, J. W., Raus, J. C., and Bach, M. A., Proc. Natl. Acad. Sci. USA 88, 2466, 1991. 33. Wucherpfennig, K. W., Newcombe, J., Hong, L., Keddy, C., Cuzner, M. L., and Hafler, D. A., J. Exp. Med. 175, 993, 1992. 34. Lehmann, P. V., Forsthuber, T., Miller, A., and Sercarz, E. E., Nature 358, 155, 1992. 35. Candeias, S., Waltzinger, C., Benoist, C., and Mathis, D., J. Exp. Med. 174, 989, 1991. 36. Zisman, E., and Mozes, E., Int. Immunol. 6, 683, 1994. 37. Green, J. M., and Thompson, C. B., Cell. Immunol. 171, 126, 1996.

cia