Jocyndra A. Wright, Leroy

ABSTRACT. Polymorphism in the germline repertoire of T-cell receptor (TCR) variable a and j3 fVa and VP) genes could alter the relative abilities of individuals in a population to respond to particular an the number of germline Va and VP gene segmenrs has been reported in wild mice and in different inbred mouse strains. A previous study of the human Vg gene germline repertoire failed to reveal a similar degree of polymorphism in the numbers of Vg gene segments. We have now carried out a survey of 10 diierent Va gene segment subfamilies containing approximately 23 Va gene segments in a panel of 120 unrelated individuals by hybridization and failed to find any evidence for Va repertoire polymorphism. To determine if significant germhne poly-

does occur in h

277-283

(1991)

ABBREVIATIONS VU

z&C

a-chain variable region g-chain variable region major histocompatibility

PCR complex

t-E

T-cell receptor

INTRODUCTION T cells recognize foreign antigens presented in conjunction with self major h&compatibility complex (MHC) molecules through their TCR molecules. On the majority of peripheral T cells, the human TCR is a heterodimer of disulfide-linked a and fl chains 111. The genes encoding the a and /3 chains of the TCR are c ted during T-ceil development through the ordered rearrangement of discreet germline V, D, and J gene segments [for review see refs. 2 and inatorial possibilities deriving from this r process and the random pairing of Q and fl chains is augmented by somatic diversification

mechanisms,

such as the addi-

HumanImmuno~ 32.277-283 (1991) Q fhericm Socimyfor Hirmcompuibili~andImmunogenetics, 1991

277 0~~~59~911~3.50

J. A. Wright et al.

278

involving as much as 50% of the rotaI repertoire, have been reported (4, 6, 7). More limited sequence analysis has provided some examples of sequence polymorphism in the coding regions of V/3 genes, including one example of a null allele of the murine VP17 gene [S, 81. A survey of the germline VP gene segment repertoire in humans indicated that, unlike the mouse, the number of V/3 gene segments is relatively constant in unrelated individuals in human populations 191. Pulsed field gel analysis has demonstrated that insertions and deletions in the human VP region do occur, however, no V/3 gene segments that map to these deleted areas have as yet been reported [lo}. There is also evidence of allotypic polymorphism of human V/3 genes based on the reactivity of a monoclonaI antibody, OT145, which detects a polymorphic V region determinant on peripheral T cells [l I]. In addition, nucleotide sequence analysis of the human V/31 gene has detected two alleles which differ at a single amino acid position [12]. We report here the results of our study of the human Va locus for evidence of germline V gene polymorphism. A suvey of variation in the number of Va gene segments by hybridization with 10 different Va subfamily-specific probes provided no evidence for deletion or duplication of any of the 23 Va gene segments that these probes detected in human populations. We carried out nucleotide sequence analysis of a selected Va gene segment, Va21, derived from a single member Va

lines 02-001,07-001, and HeIaas well as from placenta (DJ), by standard methods. Southern blots were prepared on nylon membranes (Genatran-Plasco Industries) by the method of Reed and Mann [14]. All probes were prepared by the random priming method [ 151, and hybridized as described by Gatti et al. (161. Hybridization washes were carried out at a final stringency of 2X standard saline citrate (SSC)/O.l% sodium dodecyl sulfate (SDS) at 60°C to ensure detection of all members of Va gene segment subfamilies. The Vu gene segment probes utilized in this study were isolated from the TCRa cDNA clones described by Klein et al. [17).

subfamily in order to assess the level of coding region polymorphism that would go undetected by hybridization analysis. We identified three different alleles in eight sequences derived from seven unrelated individuals. One of these alleles is likely to be a null allele for

Ndeotide seqnence analysis. Nucleotide sequences of Va2 1 genomic clones were determined by the dideoxy sequencing method utilizing sequenase (United States Biochemicals, Cleveland, OH). All nucleotide se-

some cases

expression of the Va21 gene product because of the presence of a frameshiftmg mutation which would lead to the premature termination of the corresponding rearranged Va2 1 gene. This sequence heterogeneity found in a single, randomly selected Va gene segment suggests that germline V gene polymorphism may make important contributions to the diversity of the TCR repertoire at the population level by mechanisms other than gene duplication or deletion. MATERIALS

AND METHODS

Southern blots. High-molecular-weight DNA from the parents of 40 families of primarily French or Utah IMormon origin was supplied by CEPH (Centre d’Etude du Polymorphisme Humain) [ 131. DNA from the parents of additional Utah Mormon pedigrees was provided by Dr. Raymond White. Additional DNAs from unrelated individuals were provided by Drs. Richard Spielman and Richard Gatti. DNA was prepared from the cell

Genomic libraries. Ten micrograms of genomic DNA was digested to completion with BamHI and size fractionated on a 0.7% agarose gel. DNA in the size range of 2.0 to 4.3 kilobases (kb) was isolated from low gelling temperature agarose gels by phenol/chloroform extraction and cloned into the Xhol site of lambda-zap using the partial fill-in technique described by the manufac-

turer (Stratagene). Each of the libraries was screened for the presence of Va21 sequences with a radiolabeled EcoRI-PstI fragment from the cDNA clone AF211 1171. Plaque purified clones were subcloned into plasmids by the in vivo circularization procedure described by the manufacturer (Strategene). Additional subclones were prepared in M13mp18 and mp19 for nucleotide sequence analysis.

quences were determined completely on both strands. Additional nucleotide sequencing was performed with specific oligonucleotide primers to confirm all polymorphic differences observed. RESULTS Variation in Va gene segment number. To determine whether significant variation in the germline repertoire of Vu genes in human populations results from deletion or duplication of Va gene segments, we surveyed the germline Vu gene repertoire in a panel of unrelated individuals. High-molecular-weight DNA from a panel of 120 individuals was digested with the restriction enzymes TaqI and Mspl, blotted, and hybridized with probes derived from Va gene segment subfamilies Val , Vu2. Vu8. Va12, and V&17-22. The results of this analysis are indicated in Table 1. We observed variation in the numbers of bands detected in different individ-

uals for four of the Vu probes: Val, Va19, Va20, and Va21. Band sire differences for the Val probe in Taql-

Human T-Cell Receptor Va Gene Polymorphism

TABLE

1

279

Hybridization patterns for Va probes in human genomic DNA from multiple individuals Taql’

--

near the Va2l

MSPl” -

P&C!

Bands

Val.3 Va2.2

7’ 3

Vu8.2 vu12.1

2 1

Va17.1 ValS.1 va17.1

1 1 16

vu20.1

Va21.1 Va22.1

Individuals

Bands

individuals

79 82

9 3

86 83

80 114

1 1

87 47

80 54 PO

1 1 1

86 60 85

2

85

2b

71

2lb

a9 79

2 2

E

‘The sets of individuals tested with each of these enzymes overlaps IOvarying extents with different prober.

al. { 17).

in order to co~fitm t

We dete~ined

t

‘Minimumnumber of bands is indicated for thee polymorphic probe/enzyme combinations.

digested DNA corresponded to a previously reported restriction fragment length polymorphism ( ) 1181. Va21, The three remaining probes Vtx19. Vtt20, h only one of the two enzymes sting that the differences observed with those probes might also reflect the existence of other novel RPLPs. Additional Southern blots utilizing three other restriction enzymes also failed to reproduce the differences in band numbers detected with TaqI and/or MspI and these probes, indicating that these differences most likely derive from REW, rather

rhan from differences in the number of germline Va genes. (A detailed analysis of the segregation and linkage relationships of these Vcz RFLPs and others is in preparation.) Therefore, by hybridization analysis we were unable (0 find any convincing evidence for variation in the number of germline human VII gene seg-

polymorphism do not result in d~fe~nc~s in the ptotein sequence. One is located in the 5’ untranslated region of the gene ( ition 111 in Fig. 1). At this posi-

ments.

Analyses of germline TCR V gene repertoires based on hybridization could easily overlook a substantial amount of V gene polymorphism. Simple nucleotide substitutions either in coding or regulatory sequences of particular V gene segments could have major effects on the abilities of those genes to contribute to the functional TCR repertoire. In order to gain some insight into the extent of coding region polymorphism that might exist among human TCR Va gene segments. we determined the nucleotide sequences of eight copies of the Vcu21

gene segment derived from seven unrelated individuals. Size fractionated genomic libraries were constructed

Clone V2 lCOS, derived from the library, yielded a unique single nucleotide deletio quences at position 489. This difference directly results in the substitution of a termination codon for the highly conserved tyrosine residue norm~ly found at that tion in ail human TCR V/3 and most Va genes 131. difference also results in the loss of a RsaI restri site in the cloned DNA. In order to confirm that this difference was not the result of a cloning or sequencing t, we tested for the prese in genomic DNA from

blotting. Figure 2 shows that the predicted product of

J. A. Wright et al.

280

10 AF211 WZ1a Ve21b

20

30

&O

60

SO

70

80

90

100

GTCTGTGGAA~LACAT~4ATA~L¢~A~TT~tAGT£A~TT;~.AGCTTTCi ~TTArL~CTCCA~r~LTTGTTCATATGTAA~TrL~OG

va21e 110

120

130

1/,0

150

160

170

. L'L G A i V

AF211

Va21a V~21b Va21c

180

190

200

L X'L W ~ ' Q P D ~

A~r'~Ar.~.lAGATAAGACAATCT~AT~TT~AcAGGAr.GGATGG~AT~CT~cTG~GGGCATCAGTGCTGATTcTGTGGCTT~CT 9tgegttgt9 c c 21o •

220 .

230

2~o

;'so

260

270

280

299

300

.

AF211 Ve21a Ve21b Ve21c

catgggaggtt tgaatatagtaagaaaagct acaaggaacactacaaggctgsgagataattggagaagtt t tgt t t tgt t t t c c g p t c t tgRtggtt t

310 AF211 Ve21a VaZlb V.21c

3'0

350

360

370

390

400

420

430

440

450

460

470

480

499

500

S P S ' L S Y (~ E G R ' I $ I ' L H C D Y T N'S M F ' D Y F I" U Y K ' K Y P * TTCACCATCCCTGAGCGTCCAGGAAGGAAGAAT TTCTATTCTGAACTGTGACTATACTAACAGCATGTTTGATTATTTCCTATGGTACAANU~TACCCT X 520

A t . ~ T F L'i

530

S I'S

540

550

560

S I ~ O K .'E

570

580

0 a'R F T G F L . ' ~

590

600

S M'~.

L i

GCTGAAGGTCCTACATTCCTGATATCTATAAGTTCCATTAAGGATAAAAATGAAGATGGAAGAT TCACTGTTTTCTTAMCA.qA~TGCCAAGCACCTCT C 610

620

630

640

650

660

670

•

680

690

~0

.

?mr

AF211 Ve21a

380

¢¢tgt ggggacattaaaaataaaactgaaaatct gt t at tc tc t t tccaaaca9__

510 AF211 v,~?.la Vm?.lb Va21C

330

V i S 0 Q'K N D'O Q Q V K O N" GGGT~CAGTCAACAGAAGAATGAT~CJL~TTAAGCAAAA

&lO AF211 Va21a Va21b Ve21c

320

9mr

L O i V P S O P G D S A V Y F C A A S A CTCTCGACATTGTGCCCTCCCAGCCTGGAGACTCTGCAGTGTACTTCTGTGCAGCAAGCGCACAGTGCTCTCCAGGCACCTGCAGCCCGTACTCJUL4~T

VdK?.Ib ve21c 710 AF211 Va21a VaZlb Ve21c

720

GCTTTGGGGACTCAGACTGGGAGACACATAG

FIGURE 1 Nucleotide sequences of Vc~21 alleles. The complete corrected nucleofid~ sequence derived from the V~,21 cDNA clone AF211 is shown. Additional intronic (lower case) and flanking genomic sequences are shown as derived from allele Va21a. For all other alleles, homology to these reference sequences is indicated by horizontal lines. Polymorphic sites are indicated with the appropriate substituted nucleotide or an "X" to indicate the deletion of a nucleotide. Recombination sig.nal sequences at the 3' end of the

the joining of the 800-base pair and 154-base pair fragments, which are normally separated by a RsaI site in genomic D N A derived from controls DJ and 142301, is specifically detected in D N A from HeLa cells when cleaved with Rsal and probed with a V,~21 probe. The HeLa cells used in this experiment (subline $3) are beterozygous for this polymorphism. Subsequent screening by RsaI Southern blot for this allele in 60 unrelated

Human T-Cell Receptor Va Gene Polymorphism

differences play a fund bility [S-221.

-8OObp

-154bp

FIGURE 2 Rsal RFLP corresoondine to the Va21 sequence derived from HeLa cells. ‘Five mTcrog& aliquots of DNA derived from 142301, DJ, and HeLa were digested with RsaI, separated on a 1.5% agarosegel, transferred to a nylon membrane, and probed with the AF211 probe.

A search for de

grounds failed to reveal any other examples (data not shown). Therefore, this polymorphism is either a rare variant in human populations or has arisen as a result of in vitro mutation in the HeLa cell line. DISCUSSION Variation in the repertoire of expressed TCR genes can arise from at lease two main sources: germline differences, both in the number of V gene segments and allelic sequence polymorphism. as well as differences introduce4 by clonal elimination of cells expressing the products of selected TCR V gene segments during thymic maturation 1191. Therefore, the ability of an individual to respond appropriately or inappropriately to a specific antigen challenge may be strongly influenced by the existence of these types of variation. Indeed, such variation has the potential to create specific “holes in the repertoire” in different individuals which could lead to ineffective or aberrant immune response. In murine systems, such “holes” can profoundly influence the ability of a given strain to respond to a particular antigen. For

example. studies of experimentally induced autoim-

find a second e

role in disease suscepti-

282

ternatively, it may represent a somatic mutation arising in vitro. The presence of three alleles among the eight nucleotide sequences that we analyzed raises the possibility that heterogeneity among V gene segment sequences also contributes to the diversity of the TCR repertoire. This is consistent with a similar nucleotide sequence analysis of the V/31 gene in which two alleles were identified among seven sequences determined [~12]. In summary, these results suggest that although frequent deletion or duplication of Vcz gene segments do not occur in humans, a comprehensive study o f T C R V gene nucleotide sequences will be necessary to fully characterize the human TCR germline repertoire. ACKNOWLEDGMENTS The authors wish to thank Dr. R. White, Dr. R. Spielman, and CEPH for providing the genomic DNAs used in these studies. We also thank Dr. B. Popko for the HeLa cell cosmid library, Dr. G. Nepom for helpful comments on the manuscript, Dr. P. Charmley for reviewing the nucleotide sequencing data, and Anita Stuart for secretarial assistance. This study was supported in part by T-Cell Sciences, Inc., and the Seaver Foundation. REFERENCES 1. Meuer S, Acuto O, Hussey R, Hodgdon J, Fitzgerald K, Schlossman S, Reinherz E: Evidence for the T3-associated 90KD heterodimer as the T-cell antigen receptor. Nature 303:808, 193. 2. Kronenberg M, Siu G, Hood L, Shastri N: The molecular g~nctics uf die T-cell andgen receptor and T-cell recognition. Annu Rev Immunol 4:529, 1986. 3. Wilson RK, Lai E, Concannon P, Barth RK, Hood LE: Structure, organization, and polymorphism of murine and human T-cell receptor cx and B chain gene families. Immunol Rev 101:149, 1988. 4. Jouvin-Marche E, Trede NS, Bandeira A, Tomas A, Loh DY, Cazenave P-A: Different large deletions of T cell receptor V~ genes in natural populations of mice. Eur J Immunol 19:1921, 1989. 5. Smith LR, Plaza A, Singer PA, Theofilopoulos AN: Coding sequence polymorphisms among V~ T cell receptor genes. J Immunol 144:3234, 1990. 6. Behike M, Chou H, Huppi K, Lob D: Murine T cell receptor mutants with deletions of B-chain variable region genes. Proc Natl Acad Sci USA 83:767, 1986. 7. Klotz J, Barth RK, Kiser GL, Hood I,E, Kronenberg M: Restriction fragment length polymorphisms of the mouse T-cell receptor gene families. Immunogenetics 29:191, 1989. 8. Wade T, Bill J, Marrack PC, Palmer E, Kappler JW: Mo* lecular basis for the nonexpression of VB17 in some strains of mice. J Immunol 141:2165, 1988.

J.A. Wright et al.

9. Concannon P, Gatti RA, Hood LE: Human T-cell receptor V beta gene polymorphism. J Exp Med 165:1130, 1987. I0. Seboun E, Robinson MA, Kindt TJ, Hauser SL: Insertion/deletion-related ¢olymorphisms in the human T cell receptor J3 gene complex. J Exp Med 170:1263, 1989. 11. Li Y, Szabo P, Robinson MA, Dong B, Posnett DN: Allelic variations in the human T cell receptor VB6.7 gene products. J Exp Med 171:221, 1990. 12. Robinson MA: Allelic sequence variations in the hypervariable region of a T-cell receptor B chain: Correlation with restriction fragment length polymorphism in human femilies and populations. Proc Nail Acad Sci USA 86:9422, 1989. 13. Dausset J: Le centre d'etude du polymorphisme humain. Presse Med 15:1801, 1986. 14. Reed KC, Mann DA: Rapid transfer of DNA from agarose gels to nylon membranes. Nucleic Acids Res 13:7207, 1985. 15. Feinberg AP, Vogelstein B: A technique for radiolabelling DNA restriction endonuclease fragments to high specific activity. Anal Biochem 137:266, 1984. 16. Gatti RA, Concannon P, Salser W: Multiple use of Southern blots. BioTechniques 1:148, 1984. 17. Klein M, Concannon P, Everett M, Kim L, Hunkapiller T, Hood L: Diversity and structure of human T-cell receptor o~-chainvariable region genes. Proc Natl Acad Sci USA 84:6884, 1987. 18. So A, John S, Bailey C, Owen MJ: A new polymorphic marker of the T-cell antigen receptor a chain. Immunogenetics 25:141, 1987. 19. Kappler J, Roehm N, Marrack P: T cell tolerance by clonal elimination in the thymus. Cell 49:273, 1987. 20. Banerjee S, Haqqi TM, Luthra HS, StuartJM, David CS: Possible role of V~ T cell receptor ger~esin .~usceptibility to coUagen-induced arthritis in mice. J Exp Med 167:832, 1988. 21. Acha-Orbea H, Mitchell DJ, Timmerman L, Wraith DC, Tausch G8, Waldor MK, Zamvil SS, McDevitt HO, Steinman L: Limited heterogeneity of T cell receptors from T lymphocytes mediating autoimmune encephalomyelitis allows specific immune intervention. Cell 54:263, 1988. 22. Urban JL, Kumar V, Kono DH, Gomez C, Horvath SJ, Clayton J, Ando DG, Sercarz EE, Hood L: Restricted use of T cell receptor V genes in murine autoimmune encephalomyelitis raises possibilities for antibody therapy. Cell 54:577, 1988. 23. Demaine A, Welsh KI, Hawe BS, Farid NR: Polymorphism of the T cell receptor B-chain in Graves disease. J Clin Endocrinol Metab 65:643, 1987. 24. Hoover ML, Angelini G, Ball E, Stasmy P, Marks J, Rosenstock J, Raskin P, Ferrara GB, Tosi R, Capra JD: HLA-DO and T-cell receptor genes in insulin-dependent

Human T-Cell Receptor Va Gene Polymorphlsm

25.

26.

27.

28.

diabetes meilitus. Cold Spring Harbor Syrup Quant Biol 51:803, 1986. Millward BA, Welsh KI, Leslie RDG, Pyke DA, Demaine AG: T cell receptor beta chain polymorphlsms are associated with insulin-dependent diabetes. Clin Exp Immunol 70:152, 1987. Weetman AP, So AK, Roe C, Walport MJ, Fororfi L: T-cell receptor a chain V region polymorphism linked to primary autoimmune hypothyroidism but not Graves' disease. Hum lmmunol 20:167, 1987. Demaine ACT,Vaughan RW, Taulm DH, Welsh KI: Association of membranous nephropathy with T-cell receptor constant ~ chain and immunoglobulin heavy switch region lmlymorphisms. Immunogenetics 27:19, 1988. Gao XJ, Ball EJ, Dombrausky K, Olsen NJ, Pincus T, Khan MA, Wolfe F, Stasmy P: Class il human leukocyte

283

antigen genes and T cell receptor polymorphisms in patients with rheumatoid arthritis. AmJ Med 85:14, 1988. 29. Ito M, Tanimoto M, Kamura H, Yoneda M, Morishima Y, Takatsuki Y, Itatsu T, 8aito H: Association of HI.ADR pheno~pes and T-lymphocyte-receptor beta-chainregion RFLP with IDDM in japanese. D~betes 37:1633, 1988. 30. Beall SS, Concannon P, Charmley P, McFarland HI:, Gatti RA, Hood LE, McFarlin DE, Biddison WE: The germline repertoire ofT-cell receptor beta chain genes in patients with multiple sclerosis..J Neuroimmunol 21:59, 1989. 31. Seboun E, Robinson MA, DoolittleTH, CiullaTA, Kindt TJ, Hauser SL: A susceptibility locus for multiple sclerosis is linked to the T cell receptor 3 chain complex. Cell 57:1095, 1989.