T cell receptor repertoire of infiltrating T cells in lips of Sjögren's syndrome patients

(1)

Amendment history:

Correction

(July 1992)

T cell receptor repertoire of infiltrating T cells in

lips of Sjögren's syndrome patients.

T Sumida, … , H Tomioka, S Yoshida

J Clin Invest.

1992;

89(2)

:681-685.

https://doi.org/10.1172/JCI115635

.

Infiltrating T cells around salivary glands in the lips of Sjögren's syndrome (SjS) patients are

crucial in the pathogenesis of this disease. To analyze the nature of infiltrating T cells, their

T cell receptor repertoire was examined with quantitative polymerase chain reaction. The

repertoire of V beta transcripts in lips of SjS was not restricted; however, the V beta 2 and V

beta 13 genes were predominantly expressed on the T cells of lip specimens in six and four

of seven lips, respectively. Predominance of these genes was specific in lips because no

predominant V beta transcripts were found in lips from healthy subjects and PBLs from SjS

patients. These results indicated that the V beta 2- and V beta 13-positive T cells expanded

specifically and preferentially in SjS lips, thereby suggesting the possible role in triggering

the autoimmunity of this disease.

Research Article

Find the latest version:

(2)

T

Cell Receptor Repertoire

of

Infiltrating

T Cells

in

Lips of Sjogren's Syndrome

Patients

Takayuki Sumida, Fumiko Yonaha, Toshiro Maeda, Emiko Tanabe,* Takao Koike,HisaoTomioka, and ShoYoshida Second Department of Internal Medicine and *Department of Dermatology, School ofMedicine, Chiba University, Chiba City, Chiba 280,Japan

Abstract

Infiltrating T cells around salivary glands in the lips of Sjogren's syndrome (SjS) patientsarecrucialinthe pathogene-sis of this disease. To analyze thenature ofinfiltratingTcells, their T cellreceptorrepertoirewasexaminedwithquantitative

polymerase chain reaction. The repertoire of VB transcripts in lips of SjS was not restricted; however, the Vfi2 and

V#13

genes werepredominantly expressedontheT cells oflip speci-mensin sixand four ofsevenlips,respectively. Predominance of thesegenes wasspecificinlips becausenopredominant

Vft

transcriptswerefoundinlipsfromhealthysubjects and PBLs

from SjS patients. These results indicated that the V#2-and

VB13-positiveTcellsexpandedspecifically andpreferentially

inSjS lips, thereby suggesting the possible role intriggering the autoimmunity of this disease. (J. Clin. Invest. 1992.

89:681-685.) Key words:autoreactive T cells*Sjogren's syn-drome*Tcell receptor

V/#

gene

Introduction

Sjogren's syndrome (SjS)' isanautoimmunedisease character-izedbylymphocytic infiltration oflacrimal and salivary glands, which leadstosymptomaticdryeyesandmouth (1). Immuno-histological studies (2, 3) have clarified that the majority of the infiltratinglymphocytes around salivary glandswere

CD4-posi-tiveTcells and thatCD8-positiveTcellswereinthesalivary ductepithelium. However, there have beennoreportson the nature ofthe infiltrated T cells in the salivary glands of SjS patients.

Recently, the quantitative polymerase chain reaction (PCR)wasestablished by Choietal. (4) andmay servewellin

therapid estimation ofthe human T cell repertoire,asthereare

limitednumbers of MAbs againstatleast 20 different V(3 fami-lies in humans (5). In thispaper,toexamine the mechanism of autoimmunity of SjS,wehave analyzed the repertoire of T cell receptor(TCR)genesfrominfiltrating T cells around salivary

glands in lips of SjS patients by using quantitative PCR (6). We obtainedevidence that therepertoire of TCR VBgene on

infil-Address correspondencetoDr.TakayukiSumida, SecondDepartment of InternalMedicine,Schoolof Medicine, Chiba University, 1-8-1 Ino-hana,ChibaCity, Chiba,Japan280.

Receivedfor publication 14 February 1991 and in revisedform14 Au-gust1991.

1.Abbreviationsusedinthispaper:PCR,polymerasechain reaction; SjS, Sjogren's syndrome; TCR,T cellreceptor.

J. Clin.Invest.

The AmericanSocietyforClinicalInvestigation,Inc. 0021-9738/92/02/0681/05 $2.00

Volume89, February 1992, 681-685

tratingTcells from lipsin SjS patients was not restricted; how-ever, the V32 and VJ3 13 genes were predominantly expressed

inSjS lips. The possibleroleof these T cells in the progression ofSjS is discussed.

Patients. SevenSjS patientswerereferred forstudytotheChiba

Univer-sity Hospital andall metthedefinition forSjS.Thecriteriaforthe

diagnosisincludedkeratoconjunctivitis sicca, xerostomia,and

mono-nuclear cell infiltration of the salivary gland in lip biopsy samples

(grade4) (7).

PCR andquantitation of TCR V,3gene transcripts. Total RNA

(5-10Mg)wasprepared byusingRNAzolsolution (Biotecx Laboratories

Inc., East Houston, TX), fromlipspecimensofSjS patientsthat had obviousinfiltratingTcells aroundsalivaryglands byhistological stud-ies (grade 4) orfromnormalcontrollip specimens(grade0).cDNAs were synthesized in a

_20-M1

reaction mixture containing oligo(dT) primerby AMV reversetranscriptase from5Mgtotal RNA. Amplifica-tionwasperformed with Taq polymerasein50Ml ofstandardbuffer by useof0.2Mlof cDNA(correspondingto50 ngoftotalRNA) with 20

different V(3and

_Co

primerstodetecteachV13gene. The cDNA corre-spondingto0.5 ng RNAwas asuitabledose for PCR with the5'-C3and

3'-Cf primersto detect total

_Co

gene. The sequences ofsome V,3

primerswerefrompreviously publishedreports(8), and the sequence

for

Co3

primerwasthecommon onein the

_Co

1 andC#2gene segments.

Oligonucleotidesweresynthesizedby DNAsynthesizer (Applied

Bio-systemsInc.,FosterCity, CA). Thedenature stepwascarriedout at 94'C for 1.5 min, theannealingstepat60°C for1 min, and the exten-sionstepat72'Cfor 1 minfor30cycleson aDNAthermalcycler

(ZymoreactorV2;AttoCo. Ltd., Japan).One-fifth ofthe sample was loaded on a 2% agarose gel with ethidium bromide, and amplified

DNAbands(250-300 bp)bearing

VDJC,3

genes could be detected. Forquantitation ofthe TCR V(3 and

_Co

genetranscripts, PCR productswere hybridizedwith the 32P-labeled Bgl2 fragmentof the

JUR-,B2gene(9)

(CO)

and theintensities oftheautoradiographicbands

werequantitated by densitometer (Fujix BAS 2000, Fujifilm I&I Co.). The relativeamounts ofTCR VB and

_Co3

geneswerecalculated by

using standardizationcurves.Tomakestandardizationcurves, cDNAs from PBLs in healthysubjectswereserially diluted(correspondingto1

X I05 pgRNA) and subjectedtoPCR with individual

_Vf3

andCB

primers,or

_5'-Cf3

and

3'-C,3

primersasacontrolfor30 cyclesof1.5min

at94°C, 1 minat60°C,and 1 minat72°C.Byusingeach standard

curve, the relative amountof cDNAsencodingindividual

V0

genes and total CB genes was calculated as RNA (cDNA) from PBL of

healthysubjects, respectively.The VB geneexpressionwasrepresented

as

V3/C3

(%).

The cDNAs(correspondingto3 ngRNA) fromPBLsofSjS pa-tients andfrom healthy individualswereusedfor PCRtoestimate the

expressionof VB genesonPBLs, and hundredthamountsofeachDNA (correspondingto30 pgRNA)wereusedtoestimate theexpression of

Co

genes.

Flowcytometry. PBLs from healthy subjectswereisolatedby Ficoll-Paqueseparation.Theywereincubatedwith MAbsagainst VB8 (16G8) or Vl12geneproducts(S511) (T Cell Sciences Inc.,Cambridge, MA)

(3)

byFACScan@(BectonDickinsonImmunocytometrySystems,Moun- a N M V l

C W 0

O - N M 0 W

O)° I- IN 2

tain View, CA). >

bp -492 -369 ---246 -123

Results

Toexamine the efficiency ofthe V(3 primers,total RNAwas

preparedfrom healthy humanPBLs andcDNAs were synthe- b

sizedasdescribed in Methods. ThecDNAsencoding VDJCB

wereamplified with individualsetsofVj andCB primers.20

different VB-specific oligonucleotides representing human bp

TCR V3 families were usedfor the

5'-primers,

and a C,3 se- - W 246 quence was used for the 3'-primer (Table I). Each

V,-Cf,

--123

primeryieldedaband of 250-300 bpasvisualized with

ethi-dium bromide by PCR, and amplified DNAs were further con- Figure 1.Detection of PCR-amplified products encoding TCR VBon

firmed by hybridization with a radiolabeled CB probe (9) as PBLs. (a) Total RNA (20

_.g)

wasprepared from PBLs ofhealthy

showninFig.1. Theseresults show that the PCR using each VB individuals andwasused forsynthesisof cDNAs. ThecDNAs

corre-primer and the

C#

primer could

amplify

DNAs bearing individ- spondingto 200 ng RNAwereused for PCR toamplifyDNAs

en-ualVDJCBgenes.The PCR products encoding total C,3 genes codingof TCRVDJCBor C( genes as described in Methods. Onefifth

yieldedaband of 388 bp, using 5'-C(B and 3'-Ci3 primers (Fig. 1). ofthe amplified productswas loaded on a 2% agarose gel with ethid-For

quantification

of the

V#.

transcripts

onthe

infiltrating

T iumbromide,andamplifiedDNAbands

(VDJCGI:

250-300bp,

celi

antc

lipsVofSjSpatientsBthesnt

the

folowing

tnexrintswere1

arrow;

CS:

388

bp)

could be

detected.

(b)

Amplified

DNAswere

hy-cells inthe

lips

of

SjS patients,

the

following experiments

were _{bridized with the}

_32P-labeled

_CB

_{probe. The signals correlated with}

carried out: total RNA was prepared from seven lip biopsy thebands stained by ethidiumbromide.

specimens of SjS as well as from normal control lips, cDNAs were synthesized, and PCR was carried out. The amplified DNAproducts were hybridized with the 32P-labeled C( probe

and theintensity of the radioactive bandswasmeasured with a PCR and the radioactivity of PCR products (Fig. 2). To obtain densitometer. The cDNAs encoding 20 differentVj3genes were thestandardization curves, cDNAs from normal PBLs, were

quantitated by individual standardization curves that repre- serially diluted and subjected to PCR and Southern blot

analy-sented therelationship between theamount of cDNAused for sis. When the radioactivity of each PCR product was plottedin

TableI. SequenceofHuman TCellReceptor V#Primers

Primer Sequences

5'-TCT AGA ATT CCA

5'-TCT AGA ATT CCC

5'-TCT AGA ATT CGA

5-TCT AGA ATT CCA

5'-TCT AGA ATT CAA

5'-TCT AGA ATT CAG

5'-TCT AGA ATT CAG 5'-TCT AGA ATT CAT

5'-TCT AGA ATT CTG 5'-TCT AGA ATT CAA 5'-TCT AGA ATT CAA 5'-TCT AGA ATT CAA

5'-TCT AGA ATT CAT 5'-TCT AGA ATT CAA 5'-TCT AGA ATT CAA 5'-TCT AGA ATT CCT

5'-TCT AGA ATT CGG 5'-TCT AGA ATT CGG 5'-TCT AGA ATT CGA

5'-TCT AGA ATT CCT

5'-TCT

AGA ATT CTT

5'-CCC ACA CCC AAA

5'-CAT AGA GGA TGG

AAA GGA AAC ATT

ACA TAC GAG CAA

AAA AGG AGA TAT

TAT GAG AGT GGA

AGG AAA CTT CCC

ATG ACT CAG GGC

TGT GCC AAG TCG

AGA TGA TTC AGG

AAA CAG TTC CAA

AGC AGA AAT AAT

GGG AGA TCT TTC

AGG AGA AGT CTC

GGC TAC AAT GTC GGG AGA TGT TCC

AGG AGA GAT CTC

TTA TCG ACG TGT

AGA TAT AGC TGA

AAT GCC AAA GGA

GAT GCA CAA GAA

GCA GGC AGG GGC

CTG ATG GCT CAA

AGG CCA-3' TGG CAG-3' CTl GAA GGC GTC TCC TGA TTT GTC TGG TCG TGC CCA CTT

CTC

GAT GCC ATC GCT CAA TGA CTC TGA AGA TGG TCC AGA TGA AGG TGA TGG TAT GGG AGG GTA ACG ATT GCG ATT CTC CAG ACA C-3'

682 T.Sumida,F. Yonaha, T.Maeda,E. Tanabe, T.Koike,H.Tomioka, and S. Yoshida

Vol

V#2

V#3

V14

V#5

V#6

V37

V#8

V#9

V10

Vo~l

1

V#12

V#13

V#14

V#15

V#16

V#17 V#18

Vflig

V#20

5'-Cfl

3'-CB C-3' G-3' G-3' A-3' A-3' A-3' A-3' C-3' T-3' G-3' G-3' C-3' T-3' G-3' A-3' A-3' C-3' T-3' C-3' C-3'

Thesize ofamplifiedproductsusing

5'V1

and

_3'C#

primers rangedfrom - 250 to300 bp.Primers for

V#

and

CG

genescontained additional

EcoRIandXbaI restriction enzyme sites for furthercloningexperiments.

5'-Cf3

and

3'-CO

primerswereusedtodetect total

CQ

transcriptsas

(4)

104- alog-log graph

against

theamountofcDNA (RNA) encoding

each individual V: gene, a linear relationship was obtained (Fig. 2). Alinear relationshipwas alsoobtained in the case of

E the

_Co

gene segmentamplified with

5'-C3

and

_3'-CO

primers, as

0. 103₁₀

o3-Xacontrol. We usedthis system to

quantify

the amountofthe

cDNAs

encoding

V:and

Co

genes,

respectively.

The relative .> ZD ,/; amountof

V,3

geneexpressionwasrepresentedas VA/C: (%),

0

o / ,/;t

~~~~~~~~~~~~~because

the infiltrating T cells in the individual lip biopsy

sam--0_1-2 _ples_were_{different in number.}

cs/

,,

Table II showsthat the

repertoire

of

TCR Vf3

genein

lips

of

XI

SjS patientswas not

restricted;

however, the

Vf2

gene

(8.6-16.8%)waspredominantlydetectedin six (SjS8, 9, 10, 11, 12,

10 14) of seven SjS lips compared with three lip specimens from

10 102 103 104

10o

healthy subjects (2.3-4.8%). Additionally, the

V13

gene was

RNA (pg)

preferentially expressed (13.0-29.2%)

in four lips ofSjS pa-tients(SjS 8, 10, 12, 13) in comparison with normal lips (8.1-Figure2. Standardizationcurves forquantitationof amplified prod-

96.nt(som

of the

seven

SSlpsamlso

examind

the ote

ucts.The cDNAsfrom normal PBLwereserially diluted

(corre-spondingto 1 X 105pg RNA)andsubjected to PCR with primers

V:

genes, such as Vf3, 4, 6, 7, 14, or 18, in addition toVfB2and

specificfor 20different individual V# and

CO3

genes for 30 cycles of

V:ol

3genes,werepreferentiallyexpressed.Inlipsfromhealthy

1.5minat

940C,

1 minat60'C,and1 min at

720C.

Amplified DNAs subjectswithout severeinfiltration ofT

cells,

totalRNAs en-werehybridizedwiththe

32P-labeled

Bgl2fragment of the

JUR-32

coding TCR Vie and

_CO

gene were detectable, but their amount geneand theintensities ofautoradiographicbands werequantitated was 10 times less than in SjS lips.

bydensitometer. Theradioactivity(cpm)in theV:or

Co

bands and _The_TCR

_V3

_expression

_on_PBLs_of_two

_{SjS patients}

_(SjS

therelativeamountof cDNAs(RNA) encoding

V,3

orCdgenes used 10_and

₁₂₎

_was_also_estimated

_by

_the_same_method_as_{in Table} for PCRwereplottedon alog-loggraph. Here, the standard curves,

ILasshowninF

ig.

3y

the

same Vt2 as 1n 8able

includingthe

Co3

(o),

V,32 (.), V36 (E), Vf7

(A), and

_V#

13

(x),

are II. Asshown in Fig. 3, the T cells expressingVf2(SjS 10:8.6%,

presented. SjS 12: 10.7%) and VA13genes(SjS 10: 20.9%,

SjS

12:

29.2%)

TableII. TCR VfiExpression in LipSpecimensof

SjS

Patients

V/Sfamilies (%Vfl/C/S)

Case

no. 1 2 3 4 5 6 7 8 9 10

SjS8 1.8 10.1 3.1 1.9 1.0 5.7 1.4 3.7 4.6 2.7

Sjs9 2.0 16.8 2.3 7.6 9.7 3.8 7.5 4.3 5.6 1.8

Sjs10 3.3 8.6 3.5 2.7 5.2 7.3 5.1 7.4 1.4 0.3

Sjs11 0.1 9.0 15.6 2.5 8.4 3.1 18.9 5.8 1.9 1.7

SjS12 2.0 10.0 6.4 4.2 4.3 7.8 3.7 2.8 2.5 3.2

SjS13 1.0 4.0 9.3 2.9 2.7 22.1 16.0 7.3 0.3 0.1

SjS14 2.7 10.7 12.2 10.2 1.6 3.6 5.6 2.6 7.3 3.5

Control 1 1.2 4.8 8.8 4.6 3.9 7.2 5.6 4.9 2.6 1.2

Control 2 1.0 3.2 9.3 8.7 2.8 1.4 12.8 9.2 6.2 1.0

Control 3 0.5 2.3 5.2 3.2 5.5 6.2 13.4 7.4 6.0 2.2

Vflfamilies (%V/S/Co) Case

no. 11 12 13 14 15 16 17 18 19 20

SjS8 1.0 5.6 15.2 8.4 6.8 1.8 5.5 17.4 1.4 1.2

Sjs9 0.4 1.2 8.1 16.3 4.3 1.2 2.6 3.6 0.7 0.1

Sjs 10 0.1 4.2 20.9 7.6 3.6 1.7 6.7 3.6 2.5 3.1

Sjs11 0.5 0.8 9.8 10.1 2.0 1.2 6.0 1.3 0.7 0.2

SjS12 2.2 2.6 29.2 3.9 2.6 2.5 2.3 2.0 2.5 3.1

SjS13 1.4 2.3 13.0 4.5 1.0 0.3 6.3 4.5 0.1 0.1

SjS 14 3.3 4.1 1.8 1.0 2.6 4.1 2.7 5.3 2.7 2.2

Control 1 2.1 2.9 9.1 7.2 8.2 2.9 7.8 6.8 4.3 3.8

Control 2 0.4 5.4 9.6 9.0 4.1 1.3 4.9 3.8 2.2 4.0

Control 3 2.0 4.6 8.1 9.6 5.6 4.0 4.4 5.8 3.1 1.0

ThecDNAs weresynthesized by using 5ggtotalRNAfrom lip samples ofSjS patients (grade 4)orhealthysubjects (grade 0). The

cDNAscorrespondingto50or0.5 ngRNAwereusedfor PCR with V: andC:primersor5'-Cfland

3'-CO

primersasdescribed in Methods. The

amplifiedDNAs wereseparatedon a 2%agarosegelhybridizedwith the32P-labeledBgl2 fragmentof theJUR-f32gene. The intensities ofthe autoradiographicbandswerequantitatedbydensitometer, and thecDNAsencodingV#andCBgene segmentswerecalculatedbyusingthe re-spectivestandardizingcurves(Fig. 2).Therelativeamountof TCRVOtranscriptswasdescribedasthe

VO/C3

ratio (%).Thepercentage ofT

(5)

30-20

10*

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 V,8 families

in lips from SjS 10 and SjS 12 patients increasedinnumber compared with findingsinPBLsfromthesamepatients (Vj32: 2.6 and 4.7%; V 133: 9.9 and 9.1%, respectively). These results indicate that the T cellsexpressing V32 and Vfi1 3genes

ex-panded specificallyinthelips of SjS patients.

To confirm the results obtained bythequantitative PCR,

we examined the actualpercentageofTcellsexpressing V18 and

_V#

12 molecules on PBLs from the same three healthy subjects by usingflowcytometryandanti-V,3 MAbs.Thedata

show that the percentagesof V38(4.9±0.50%,mean±SD)-and

V,312 (2.1±0.68%)-positive Tcells calculated by quantitative PCRweresimilartothose estimatedbyflowcytometry (V38: 4.1±0.24,

_V#132:

2.4±0.81%).

Discussion

The TCR V(3 repertoire of infiltrating T cells in lips ofSjS patientswasexaminedwith thequantitativePCRmethod. We

obtainedevidence that therepertoireofVie transcriptswasnot

restricted; however, the Vf32 (8.6-16.8%) and V 1l3 (13.0-29.5%)geneswerepredominantly expressed onTcellsinSjS lips. Predominance of theVj32and Vj 13geneswasspecificin lips because lesser percentages of these two genes were

ex-pressedonPBLs(V#2: 2.6-4.7%,V1313:9.1-9.2%)inthesame

subjects. Thus,TcellsbearingVf2orVf 13genes maybe

re-sponsiblefortriggeringtheautoimmunityofthisdisease.

Studies(10-13)of human autoimmune diseases have pro-videdsomeevidence foroligoclonalitybytherestriction frag-mentlength polymorphismmethod inpatientswith myasthe-niagravis, multiple sclerosis,and rheumatoid arthritis.Indeed, Oksenbergetal.(14), usingPCRtechnologywith brainbiopsy specimens, reported that TCR Va10gene waspreferentially expressed, suggesting the limited heterogeneity of TCR Va transcriptsinbrains ofmultiplesclerosispatients.Sottini etal. (15) demonstrated that synovial fluid T cells expressed V#7 transcripts andnonrestricted Vaelements. The studyinthis paperis the firstreportonthe TCRV,3 repertoireofinfiltrating Tcells inlips of SjS patients.

Choietal.(4) reportedthatTcellsexpressing certainVf3 genes could bestimulated by some bacterialtoxins, such as

toxic-shocksyndrome toxin and staphylococcal enterotoxin,as

a"superantigen." It isinterestingto notethat Vf32-positive T cellscanbestimulatedby toxic-shock syndrome toxin1and T

Figure 3.Expressionof

_Vft

genes in lips and on PBLs of

SjSpatients. Toestimatethe VB geneexpression,the cDNAs(correspondingto3 ngRNA) fromPBLsoftwo

SjS patients (a)

SjS

12(b)SjS 10 andcDNAs

(corre-spondingto50 ng RNA)from lips ofthe same subjects were used for PCR, respectively, as described in Table

II.

For the

Ci3

gene, hundredth amounts of each cDNA were subjected to PCR. The relative amount of TCR

Vj3

transcripts in lip

(u)

and on PBL(o)werecalculated with the same method as in TableII.The percentageofTcells in PCR was the meanoftwoindividual experiments.

cellsbearingV3 13geneexpand by thestimulationof staphylo-coccalenterotoxin C2. This suggested the possibility that

pre-vious bacterial infection in the lips ofSjS patients ledto stimula-tionandexpansion of T cellsexpressingthesetwogenes.Some

of these T cells might recognize self-antigens and playacrucial

role in thepathogenesisof autoimmunity.

Theself-antigens in SjS havenot yetbeen identified, butan

association between retrovirus infection and SjS has been

sus-pected (16-18). Antibodies against HIV-l proteinwerehighly

detectedintheserumofSjS patients (16), and transgenic mice

expressing the humanTcellleukemia virus(HTLV- 1)taxgene

developedanexocrinopathy resembling SjS (17). Recently,a

human intracisternal A-type retroviral particlewasdetected in

lymphoblastoidcellsexposedtohomogenates of salivary tissue fromSjS patients (18). These studiessuggestthe possibility that the retroviral particle by itself, or virus-induced products,

might be candidates for self-antigens in theprogressionof SjS. Theanalysisoftheantigenrecognizedby limitedautoreactive

Tcells in thelips of SjS patientsmayprovetoclarify the mecha-nism of autoimmunity in SjS.

Severalinvestigations (19-21) oftheTCRgeneexpression in animal models forTcell-mediatedautoimmune diseases, suchasexperimental allergicencephalomyelitisand

collagen-inducedarthritis, have alsodemonstratedthat the TCRgene

usageinvolvedinautoimmunitywasrestricted. In thecaseof

experimentalallergicencephalomyelitis,the administration of

aMAbspecificforthepredominant V,38 molecules resultedin

nearly complete preventionofthedisease(19, 20).Inaddition,

the vaccinationofasynthetic peptide encoding for the hyper-variableregion of V38 regionsprotected against encephalitis (22). Therefore, our results may indicate a new therapeutic

approachtoautoimmunediseases suchasSjS bymeansof the

selective elimination ofpredominant T cells in involved

tis-sues by anti-TCR antibodies or the vaccination ofpeptides

forTCR.

Acknowledgments

JUR-(32genewaskindlyprovided by Prof. Tak. W. Mak.

This workwassupportedinpartby The Mochida Memorial

Foun-dation for Medical and Pharmaceutical Research.Takayuki Sumida

wassupported as apostdoctoralfellowby the JapanHealthSciences

Foundation.

684 T.Sumida,F. Yonaha, T.Maeda, E. Tanabe,T.Koike,H.Tomioka, and S. Yoshida

a

30-Qx

20-0

(6)

References

1.Bloch, K. J., W. W.Buchanan,M. J.Wohl, and J. J. Bunim. 1965. Sjogren's syndrome. Aclinical, pathologicalandserological studyofsixty-twocases. Medi-cine.44:187-231.

2.Fox, R. I., T.C. Adamson III, S. Fong, C. Young, and F. V. Howell. 1983. Characterizationofthephenotype andfunction oflymphocytes infiltratingthe salivary gland in patients withprimary Sjogren'ssyndrome. Diagn. Immunol.

1:233-239.

3. Adamson, T. C., III, R. I. Fox, D. M.Frisman,and F. V.Howell. 1983.

Immunohistologicanalysis oflymphoid infiltratesin primarySjogren's syn-dromeusingmonoclonalantibodies. J. Immunol. 130:203-208.

4.Choi, Y., B. Kotzin, L. Herron, J. Callahan, P.Marrack,andJ. Kappler. 1989.InteractionofStaphylococcusaureustoxin"superantigens"with human T cells. Proc. Natl.Acad. Sci. USA. 86:8941-8945.

5.Toyonaga, B., and T. W. Mak.1987. Genesofthe t-cellantigenreceptor in normal and malignant T cells. Annu. Rev.Immunol. 5:585-620.

6.Saiki, R. K., S.Scharf,F.Faloona,K. B.Mullis,G. T.Horn,H.A.Erlich, and N.Arnheim. 1985.Enzymatic amplificationof,B-globingenomicsequences and restrictionsiteanalysis fordiagnosis of sicklecellanemia.Science (Wash.

DC).230:1350-1354.

7.Chisholm,D.M.,and D. K. Mason. 1968.Labialsalivary gland biopsyin Sjogren's disease. J. Clin. Pathol.(Lond.).21:656-660.

8.Wucherpfennig,K.W.,K.Ota,N.Endo,J.G.Seidman,A.Rosenzweig,

H.L.Weiner, and D. A.Hafler.1990. Shared humanTcell receptor VB usageto immunodominant regions of myelin basic protein. Science (Wash. DC). 248: 1016-1019.

9.Yoshikai, Y., D.Anatoniou,S. P. Clark, Y.Yanagi,R.Sangster, P. V. Elsen, C. Terhorst, and T. W. Mak. 1984.Sequenceandexpression of transcripts

of the human T-cell receptorj#-chaingenes. Nature(Lond.) 312:521-524.

10.Oksenberg,J.G.,M.Sherritt,A.B.Begovich,H. A.Erlich,C. C.Bernard,

L. L.Cavalli-Sfurze, and L.Steinman. 1989. T-cell receptor Va and Ca alleles

associatedwithmultiplesclerosis andmyasthemiagravis.Proc.Natl.Acad.Sci. USA. 86:988-992.

11.Beall,S. S., P.Concannon,P.Charmley, H. F. McFarland, R.A.Gatti,

L. E.Hood, D. E.McFarlin,and W. E.Biddison. 1989. Thegermlinerepertoireof Tcell receptorf3-chaingenes inpatientswithchronic progressive multiple

sclero-sis.J.Neuroimmunol.21:59-66.

12.Hafler,D.A.,A.D.Duby,S. J.Lee,D.Benjamin,J.G.Seidman,and H. L. Weiner. 1988.OligoclonalTlymphocytesin thecerebrospinal fluid of

patientswith_multiplesclerosis. J.Exp.Med. 167:1313-1322.

13.Stamenkovic, I.,M.Stegagno,K. A.Wright,S.M.Krane,E. P.Amento,

R. B.Colvin, R.J.Duquesnoy,andJ. T. Kurnick. 1988. Clonal dominance amongT-lymphocyteinfiltrates in arthritis. Proc.NatL.Acad. Sci. USA.

85:1179-1183.

14.Oksenberg,J.R.,S.Stuart,A.B.Begovich,R.B.Bell,H. A.Erlich,L.

Steinmann,and C. C. A. Bernard. 1990. Limitedheterogeneityofrearranged

T-cell receptor Vatranscriptsin brains ofmultiplesclerosispatients. Nature

(Lond.).345:344-346.

15.Sottini, A.,L.Imberti,R.Gorla,R.Cattaneo,and D.Primi. 1991. Re-strictedexpressionof T cell receptorV(3butnotVa genes in rheumatoid arthritis. Eur.J.Immunol.21:451-466.

16.Talal,N., M. J. Dauphinee, H. Dang, S. S. Alexander, D. J. Hart, and R. F.

Garry.1990. Detection ofserumantibodiestoretroviralproteinsinpatientswith

primary Sjogren'ssyndrome (autoimmune exocrinopathy). ArthritisRheum.

33:774-781.

17.Green,J.E.,S. H.Hinrichs,J.Vogel,andG. Jay. 1989.Exocrinopathy resemblingSjdgren'ssyndrome in HTLV- I tax transgenic mice. Nature(Lond.).

341:72-74.

18.Garry,R.F.,C. D. Fermin, D. J. Hart, S.-S. Alexander, L. A. Donehower, and H. Luo-Zhang. 1990. Detection of a human intracisternal A-type retroviral

particle antigenicallyrelatedtoHIV.Science (Wash. DC). 250:1127-1129. 19.Archa-Orbea, H., D. J. Mitchell, L. Timmermann, D. C. Wraith, G.S.

Tansch, M. K. Waldor, S. S. Zamvil, H. D. McDevitt, and L. Steinman. 1988. Limited heterogeneity of T cell receptors from lymphocytes mediating autoim-muneencephalomyelitisallowsspecificimmuneintervention.Cell.54:263-273.

20.Urban,J.L., V. Kumar, D. H. Kono, C. Gomez, S. J. Horvath, J. Clayton, D.G.Ando,E. E.Sercarz,and L. Hood. 1988. Restricted use ofT cell receptorV genes in murine autoimmuneencephalomyelitis raises possibilities forantibody therapy.Cell. 54:577-592.

21. Wraith, D. C., H. D. McDevitt, L. Steinman, and H. Acha-Orbea. 1989. T cell recognitionasthe target for immune intervention in autoimmune disease. Cell. 57:709-715.