A single universal primer for the T-Cell receptor (TCR) variable genes enables enzymatic amplification and direct sequencing of TCRβ cDNA of various T-cell clones

A Single Universal Primer for the T-Cell Receptor (TCR) Variable Genes Enables Enzymatic Amplification and Direct Sequencing of TCR/ cDNA of Various T-Cell Clones Fumiya Obata, Misao Tsunoda, Koichi Ito, Ichiro Ito, Takehisa Kaneko, Graham Pawelec, and Noboru Kashiwagi

ABSTRACT: We designed a primer for the PCR directed against a highly conserved sequence of the TCR V/~ gene. TheV/3-universal primer, in combination with a constant region-specific primer, enabled us to amplify TCR/~ cDNA of allo-HLA class-II-reactive T-cell clones by PCR without prior knowledge of their V/~ sequences. The amplified TCR cDNA was purified by agarose" gel electrophoresis and subjected to direct sequencing. In nine of ten T-cell clones analyzed, direct TCR sequencing gave readable sequence ladders, including two-thirds of VB, junctional, and J/3 regions. One T-cell clone gave an ABBREVIATIONS CDR3 third-complementarity-determining region D diversity J joining

unreadable mixed-profile sequence ladder, indicating that this clone expressed more than one major TCR/3 transcript. Even in this case, however, it was possible to determine two different TCR/~ sequences separately using sequence primers specific to one of the 13 J/~ segments deduced from the mixed ladder. Thus, direct sequencing utilizing the single V~-universal primer enabled a simple, rapid, and reliable sequence determination of TCR3 cDNA of all T-cell clones analyzed. Human Immunology 36, 163-167 (1993)

PCR TCR V

polymerase chain reaction T-cell receptor variable

INTRODUCTION For specific antigen recognition, T cells utilize T-cell receptor (TCR) molecules expressed on their surface. Enormous diversity of TCR molecules is generated by D N A rearrangement between the variable (V), diversity (D) (only for T C R 3 and TCRS), and joining (J) genes

From the Laboratory ofImmunology (F.O,, M. T., K.I., 1.I., T.K., N.K.), Kitasato University SchoolofMedicine, Sagamihara,Japan; and the Section for Transplantation Immunology (G.P.), University of Tiibingen Hospital Tiibingen, Germany. Address reprint requests to Dr. F. Obata, Department of Immunology, Kitasato University School of Medicine, 1-15-1 Kitasato, Sagamihara 228, Japan. ReceivedAugust 17. 1992; acceptedNovember 23, 1992. Human Immunology 36, 163-167 (1993) © American Society for Histocompatibility and Immunogenetics, 1993

as well as by junctional flexibility including N-segment insertion (see Davis [1] for a review). To clarify the TCR structure utilized for recognition of a given antigen, sequence analysis of rearranged TCR D N A or c D N A to their transcripts is necessary. Although the polymerase chain reaction (PCR) has greatly facilitated the molecular cloning of various genes starting from a small amount of material and subsequent sequence determination [2], the procedure of cloning with plasmid vectors and bacteria is still laborious and time-consuming. More seriously, bacterial cloning of the PCR-amplified products involves a risk that erroneous PCR products might be selected for sequence analysis 163 0198-8859/93/$6.00

164

F. Obata et al.

V6 "" CDR2 ""

-CDRI-

D6-J6 CDR3

C6

VBUN--* C

WY

C

C ",FJB1.5 FJB2. 3

VBUN : FPRI5 : FJBI. 5:

C

"*-FPR2T

*-FPRI5

5'-GGGG(AGCT)(AGCT) (AGCT)(AGC)T(AGT)T(ATC)(CT)TGGTA-3' 5'-AAGCCACAGTCTGCTCTACC-3' FPR27 : 5'-GGAGATCTCTGCTTCTGATG-3' 5'-ATGGAGAGTCGAGTCCCA-3' FJB2.3: 5'-ACTGTCAGCCGGGTGCCT-3'

FIGURE 1 Locations and sequences of oligonucleotide primers used in this study. W, Y. and C indicate the conserved tryptophan, tyrosine, and cysteines, respectively.

[2, 3]. In this regard, direct sequencing of whole preparations of PCR products without bacterial cloning is simple and rapid, and is also free from interference by erroneous PCR products, which generally account for only a minor proportion of the total PCR products [4]. In the case of TCR, however, the lack of prior knowledge of upstream sequences, i.e., V-region sequences, makes direct sequencing difficult. Although modified forms of PCR called anchored PCR [5] and inverted PCR [6], and a PCR method utilizing primers directed against a conserved sequence around the 5' end of V genes [7], have been reported for amplification and cloning of TCR cDNA containing unknown V genes, direct sequencing by these methods has not been described. In direct sequencing of TCR cDNA amplified by utilizing the known V-region sequences, a large number of PCR primers must be used to cover all the V genes [8]. In this article, we report direct sequencing of PCRamplified TCR/3 cDNA derived from several antiallogeneic HLA T-cell clones. For this purpose, we designed a PCR primer directed against a nucleotide sequence that is highly conserved among all known V3 genes. Using this forward PCR primer and TCR/3-constant region (C3)-specific reverse PCR primer, we successfully amplified TCR3 cDNA of T-cell clones without prior knowledge of their V/~ genes. Direct sequencing of the TCR cDNA thus amplified provided us with sequence information on the V/3, junctional (D/3 and N), and J/3 regions. MATERIALS A N D M E T H O D S

T-cell clones. An anti-DPw2 clone, 4-1A4, anti-DPw5 clones, 2-1C5, 2-1E6, and 2-3C1, and an anti-DP clone with an unknown specificity, 4-1BA, were established

from a single priming culture by the conventional cloning procedure (T. Kaneko et al., manuscript in preparation). An anti-DRw52 clone, 341-62, an anti-DQw3 clone, 341-26, an anti-DPw3 clone, 341-70, and antisupertypic class II clones, 341-1 and 341-7, were also established from another single priming culture ([9] and G. Pawelec, manuscript in preparation). Some of these clones were also submitted to the l l t h International Histocompatibility Workshop in 1991 as 11 wT37 (34126), 1 lwT38 (341-62), 11wT39 (341-1), 1 lwT40 (3417), and 11wT41 (341-70) [9, 10].

Oligonucleotides. The location and sequences ofoligonucleotide primers used in this study are shown in Fig. 1. The CB-specific primers FPR 15 and FPR27, and the J/3specific primers FJB1.5 and FJB2.3, were synthesized according to the published sequence of the TCR C3 region and those of the J/J1.5 and J/32.3 segments, respectively [11]. VB-universal primer VBUN was designed on the basis of known V3 sequences obtained from EMBL (release 29.0) and Genbank (release 70.0) [12]. cDNA synthesis, mRNA was isolated from 1-3 × 106 cells ofT-cell clones by using Dynabeads-oligo(dT) (Dynal, Oslo) by the procedure recommended by the supplier. Double-stranded TCR/3 cDNA was synthesized according to the method of Gubler and Hoffman [13] using a C~-specific primer, FPR15, for synthesis of the first strand. The cDNA was purified by extraction with phenol-chloroform-isoamyl alcohol and precipitated with ethanol.

PCR. One-tenth of the cDNA preparation was used for amplification by PCR using 25/~l of mixture containing 10 mM Tris-HCl (pH 8.8), 50 mM KCI, 2.5 mM MgCl> 100 p~g/ml gelatin, 0.5 mM each dATP, dGTP, dCTP, and dTTP, 10 tzM VBUN, 0.5/~M FPR27, and 2.5 units of AmpliTaq (Perkin Elmer Cetus, Norwalk, CT) [2]. Thirty cycles of PCR were carried out by denaturation

Direct Sequencing of TCR3

for 1 minutes at 97°C, annealing for 1 minute at 55°C, and extension for 3 minutes at 72°C. The PCR product was applied to 1.5 % SeaPlaque agarose gel (FMC, Rockland, ME) and a DNA fragment of about 320 bp was excised from the gel, purified using a Magic PCR Prep DNA Purification System (Promega, Madison, WI), and precipitated with ethanol.

Sequencing. One-fifth of the gel-purified TCR/~ cDNA preparation was subjected to 30 cycles of the sequence reaction (cycle sequencing) according to the method of Murray [14]. FPR27 was 32P-labeled and used as a sequence primer. For TCR/3 cDNA ofT-cell clone 341-62, FJB 1.5 and FJB2.3 were also used as sequence primers.

165

M

1

2

1,419

517 396

214

75 65

RESULTS For the PCR amplification ofTCR/3 cDNA without prior knowledge of its VB sequence, we chose a nucleotide sequence of 5 bases (TGGTA) as a target for the forward PCR primer. This sequence, coding for tryptophan and partly for tyrosine, is highly conserved among all V/3 genes known so far. Therefore, we designed an 18-baselong V/3-universal PCR primer designated VBUN containing the above 5 bases at the 3' end (Fig. 1). The other portion of VBUN contained two thymines, various degrees of base degeneracy designed on the basis of published V/3 sequences, and a stretch of four guanines at the 5' end for stabilization of annealing. The universality of VBUN was confirmed by PCR amplification of cloned cDNA plasmids containing various known VB genes (data not shown). TCR/3 cDNA was synthesized from mRNA isolated from 10 T-cell clones with reactivities against allogeneic HLA class II antigens and amplified by PCR with VBUN and a C/~-specific primer, FPR27. Agarose gel electrophoresis of the PCR products gave DNA fragments consisting of two-thirds of the V/3 gene, the junctional region (D/~ and N), J/3 segments, and the beginning of C/3 (Figs. 1 and 2). The exact sizes of the DNA fragments varied slightly around 320 bp depending on the length of both the junctional and V/3 regions. The amplified TCRB cDNA was recovered from the agarose gel and subjected to cycle sequencing using FPR27 as a sequence primer. In nine of ten T-cell clones analyzed, direct sequencing of their TCR/3 cDNA gave readable sequence ladders V/3 through C/3, indicating that these clones expressed single major TCR/3 transcripts (Fig. 3, lane 1 for clone 341-1). All the TCR/3 cDNA of these clones had translatable in-frame sequences from the V/3 region to the beginning of C/3, demonstrating that they were derived from productive transcripts. Each of nine TCR3 consisted of one of the known V/3 and J/3 sequences,

FIGURE 2 Agarose gel electrophoresis of TCR/~ cDNA amplified with the V/3-universalprimer. The PCR products of TCR/3 cDNA of 2-1E6 (lane 1) and 2-3C1 (lane2) are shown. The molecular weight marker (lane M) is a HinfI-digested Puc 18 plasmid.

except that two amino acid substitutions from the known V/320.1 sequence, ascribable to either an allelic variation or a new member of this V/3 subfamily, were found in V/3 of 2-1C5 (Fig. 4). The junctional sequences of these TCR/3, also known as the third complementarity determining region (CDR3), were also determined conclusively by direct sequencing. A pair ofT-cell clones, 3411 and 341-7, were found to have TCR/3 sequences completely identical to each other, including the CDR3. Because this pair of T-cell clones showed similar antiHLA specificities (G. Pawelec, unpublished data), they might have been sister clones that had originated in the single priming culture, although sequence analysis of TCR~ would be prerequisite for confirmation. On the other hand, direct sequencing ofTCR/~ cDNA of one T-cell clone 341-62 gave a mixed-profile sequence ladder whose upstream of C/3 was unreadable, suggesting that this clone expressed more than one major TCR/3 transcript (Fig. 3, lane 2). Even in this case, however, two candidate J/3 segments constituting the mixed ladder, i.e., J/31.5 and J~2.3, were identified. By using sequence primers specific to each of the two J~3 segments, we were able to determine the two TCR 3 sequences separately (Fig. 3, lanes 3 and 4). One TCRB sequence of T-cell clone 341-62 with J/32.3 was productive, whereas the other with JF31.5 had an in-frame stop condon in the junctional region, in keeping with the rule of allelic exclusion (Fig. 4).

166

F. Obata et al.

1

2

3

4

G A~C

GATC

GATC

GATC

Vs

!

m

N

b

VB

J8

m

N D o o

Cs Z

J8

,,MI*

V

FIGURE 3 Direct sequencing of TCR3 cDNA amplified with the V/3-universal primer. Sequence ladders around the junctional region are shown. Lane I, T-cell clone 341-1 with FPR27 sequence primer; and lane 2, T-cell clone 341-62 with FPR27 sequence primer. Arrows indicate the positions where J/31.5 and JB2.3 segments formed the mixed ladder: lane 3, T-cell clone 341-62 with FJB1.5 sequence primer; and lane 4, T-cell clone 341-62 with FJB2.3 sequence primer.

DISCUSSION Because of erroneous amplification in PCR, sequence analysis of PCR products after bacterial cloning often gives incorrect sequence information [2, 3]. Particularly in analysis of TCR, this could cause serious trouble, because single amino acid substitutions in the CDR3 are known to affect the specificities of T-cell clones greatly [15]. Erroneous sequences in the V/3 region would mistakenly be ascribed to a new V gene or a new allele. Even for single T-cell clones, therefore, several bacterial

clones have to be analyzed and compared with one another for conclusive sequence determination. From this viewpoint, the method of direct sequencing described here, which utilizes the V/3-universal PCR primer, is a reliable, simple, and rapid approach for sequence determination of TCR c D N A of T-cell clones. It should be stressed that our method requires only one PCR mixture with a single combination of primers for each T-cell clone. This is in marked contrast to the PCR amplification using more than 20 different primers specific to each of the V/3 genes [8], by which up to 200 different PCR mixtures would have to be prepared for analysis of 10 T-cell clones. The method is especially convenient for analysis of T-cell clones expressing a single major TCR transcript, as was the case in most of the T-cell clones analyzed in this study. Although the D N A fragment obtained by PCR-amplification with the V3-universal primer lacks the Nterminal one-third V/3 sequence of the mature polypeptide, the residual two-thirds of the V/3 gene analyzed in this study would give enough sequence information to identify VB genes at the level of their subfamily members in most cases (62 of 71 known V 3 members), providing that the unanalyzable N-terminal sequences are identical to the known V/3 sequences. The exceptional nine cases are V~31.1/1.2, V/34.1/4.2/4.3, V~5.3/5.7, VF36.6/6.7, V/38.1/8.2/8.3, V39.1/9.2, V/310.1/10.2, V311.1/11.2, and V/318.1/18.2, for which N-terminal region sequences are necessary for conclusive identification. In one of 10 clones analyzed (341-62), PCR amplification with the V3-universal primer and direct sequencing with the C~3-primer gave an unreadable sequence ladder, because of expression of more than one major TCR~ transcript, one of which was nonproductive. In a study of a larger number of T-cell clones reactive with allogeneic HLA-class II antigens, in which anchored PCR and conventional bacterial cloning were adopted, eight (about 14%) of 56 T-cell clones analyzed were found to express more than one major TCR/3 transcript, including both productive and nonproductive transcripts [10]. As far as cloned T cells are concerned, direct sequencing would become possible by using sequence primers specific to one of the 13 J3 segments, providing that the T-cell clones utilize two different J 3 segments as in the case ofT-cell clone 341-62. For a heterogeneous or uncloned T-cell population, on the other hand, bacterial cloning is prerequisite for sequence analysis of the TCR/3 consisting of various V/3 genes. In this case, the V3-universal primer may serve as a convenient PCR primer for amplifying a heterogeneous TCR c D N A preparation before it is subjected to bacterial cloning. Finally, it would also be possible to apply our method for direct sequencing of TCRc~, TCRS, and TCRy, not

Direct Sequencing of TCR/3

T ceil

clone

167

VB

--D6+N

JB - -

C6-

VB - DB - J 6

17.1 - 1.1 NEQFFGPGTRLTVLEDLKN 2-1E6 [WY]RQDPGOGLRLIYYSOIVND FOKGDIkEGYSVSREKKESFPLTVTSAQKNPTAFYLCkSSFGTGV ? SYEOYFGPGTRLTVTEDLKN 5 . 3 / 5 . 7 4-1k4 [WY]QQVLGQGPQFIFQYYEKEE RGRGNFPDRFSARQFPNYSSELNVNALLLGDSALYLCASSLAPGS QYFGPGTRLLVL EDLKN 16.1 4-1BA [WY]RRVMGKEIKFLLHFVKESKQDESGMPNNRFLAERTGGTYSTLKVQPAELEDSGVYFCASSOVLGLREN 2-3C1 [WY]KQDSKKFLKIMFSYNNKELIINE TVPNRFSPKSPDKAHLNLHINSLELGDSAVYFCASS OVLGOGKR TDTQYFGPGTRLTVL EDLKN 9 . 1 / 9 . 2 - 1. 2 0 . 1 a - 1. EKLFFGSGTQLSVLEDLNK 2-1C5 [WY]RQAAGRGLQLLFYSVGIGQIS SEVPQNLSASRPQDRQFILSSKKLLLSDSGFYLCAWSGEOGIR 2.1 - 2. 341-26 [WY]RQFPKQSLNLMATSNEGSKATYEQGVEKDKFLINHASLTLSTLTVTSAHPEDSSFYICSARDPGLAGL EQYFGPGTRLTVT EDLKN 17.1 - 2 . DTQYFGPGTRLTVLEDLKN 341-62 ~1) [¥Y]RQDPGQGLRLIYYSQIVND FQKGDIAEGYSVSREKKESFPLTVTSAQKNPTAFYLCASSSTSGGP 13.4 - 1. 341-62 !12) [WY]RQDLGLGLRLIHYSNTAG TTGKGEVPDGYSVSRANTDDFPLTLASAVPSQTSVYFCASSDLAGGSTAPAFW* KG SNQPOHFGDGTRLSILEDLNK 22.1 - 1. 341- 1 [¥Y]RQILGQKVEFLVSFYNNEISEKSEIFDDQFSVERPDGSNFTLKIRSTKLEDSAMYFCAS 22.1 - 1. KG SNQPQHFGDGTRLSlLEDLNK 341- 7 [WY]RQILGQKVEFLVSFYNNE ISEKSEIFDDQFSVERPDGSNFTLKIRSTKLEDSAMYFCAS 13.1 - 1 . ETQYFGPGTRLLVLEDLKN 341-70 [WY]RQDPGNGLRLIHYSVGEG TTAKGEVPDGYNVSRLKKONFLLGLESAAPSQTSVYFCAS NRQA ($: stop codon,

FIGURE 4 Amino acid sequence ofTCR/3 cDNA ofT-cell clones directed against allogeneic HLA class II antigens. W and Y residues in parentheses are part of the target sequence of the VB-universal primer. Two amino acid substitutions from the known V/320.1 sequence [16] are underlined.

only those of human but also those of other species such as mouse and rat, because the target sequence of the V/3-universal primer (TGGTA) is also conserved, although not perfectly, in Vex, VS, and V3' of various species.

ACKNOWLEDGMENTS This study was supported in part by grant 02670718 from the Japanese Ministry of Education and by the Deutsche Forschungsgemeinschaft (SFB 120 A1; Pa 361-1/1).

REFERENCES 1. Davis MM: T cell receptor gene diversity and selection. Annu Rev Biochem 59:475, 1990. 2. Saiki RK, Gelfand DH, Stoffel S, Scharf SJ, Higuchi R, Horn GT, Mullis KB, Erlich HA: Primer-directed enzymatic amplification of DNA with a thermostable polymerase. Science 239:487, 1988. 3. Keohavong P, Thilly WG: Fidelity of DNA polymerases in DNA amplification. Proc Natl Acad Sci USA 86: 9253, 1989. 4. Wong C, Dowling CE, Saiki RK, Higuchi RG, Erlich HA, Kazazian HH Jr: Characterization of/3-thalassaemia mutations using direct genomic sequencing of amplified single copy DNA. Nature 330:384, 1987. 5. Loh EY, Elliott JF, Cwirla S, Lanier LL, Davis MM: Polymerase chain reaction with single-sided specificity: Analysis o f T cell receptor 8 chain. Science 243:217, 1989. 6. Uematsu Y: A novel and rapid method cloning method

?: Di31.1

-

2.1

- 2.7 -2.5 - 2.3 - 1.4 - 2.7 -2.3 ~(1.5) - 1,5 - 1.5 -2.5

or 2.1)

for the T-cell receptor variable region sequences. Immunogenetics 34:174, 1991. 7. Broeren CPM, Verjans GMGM, Van Eden W, Kusters JG, Lenstra JA, Logtenberg T: Conserved nucleotide sequences at the 5' end of T cell receptor variable genes facilitate polymerase chain reaction amplification. Eur J Immunol 21:569, 1991. 8. Bragado R, Lauzurica P, L6pez D, L6pez de Castro JA: T cell receptor V/3 gene usage in a human alloreactive response: shared structural features among HLA-B27-specific T cell clones. J Exp Med 171:1189, 1990. 9. Rehbein A, Schlotz E, V6hringer D, Schaudt K, Pawelec G: Cellular alloreactivity to the DRB 1" 1303 allele-associated Dw"HAG" specificity. In Tsuji K, Aizawa M, SasazukiT (eds): HLA 1991, vol 1. Oxford, Oxford University Press, 1993. 10. Obata F, Kashiwagi N: Joint report of the T-cell component: sequence analyses of the T-cell receptor utilized for recognition of HLA-class II molecules. In Tsuji K, Aizawa M, Sasazuki T (eds): HLA 1991, vol 1. Oxford, Oxford University Press, 1993. 11. Toyonaga B, Yoshikai Y, Vadasz V, Chin B, Mak TW: Organization and sequences of the diversity, joining, and constant region genes of the human T-cell receptor 3 chain. Proc Natl Acad Sci USA 82:8624, 1985. 12. Devereux J, Haeberli P, Smithies O: A comprehensive set of sequence analysis programs for the VAX. Nucleic Acids Res 12:387, 1984. 13. Gubler U, Hoffman BJ: A simple and very efficient method for generating cDNA libraries. Gene 25:263, 1983. 14. Murray V: Improved double-stranded DNA sequencing using the linear polymerase chain reaction. Nucleic Acids Res 17:8889, 1989. 15. Engel I, Hedrick SM: Site-directed mutations in the VDJ junctional region ofa T cell receptor/3 chain cause changes in antigenic peptide recognition. Cell 54:473, 1988. 16. Leiden JM, Strominger JL: Generation of diversity of the 13chain of the human T-lymphocyte receptor for antigen. Proc Natl Acad Sci USA 83:4456, 1986.