of July 16, 2018.This information is current as

Repertoire in Myasthenia GravisCDR3 Spectratyping Analysis of the TCR

Kondo, Satoshi Kotorii and Noritoshi ShibuyaTakayukiIl-Kwon Park, Yukiko Tsukada, Kuniko Kohyama,

Yoh Matsumoto, Hidenori Matsuo, Hiroshi Sakuma,

http://www.jimmunol.org/content/176/8/5100doi: 10.4049/jimmunol.176.8.5100

2006; 176:5100-5107; ;J Immunol 

Referenceshttp://www.jimmunol.org/content/176/8/5100.full#ref-list-1

, 16 of which you can access for free at: cites 37 articlesThis article

        average*  

4 weeks from acceptance to publicationFast Publication! •    

Every submission reviewed by practicing scientistsNo Triage! •    

from submission to initial decisionRapid Reviews! 30 days* •    

Submit online. ?The JIWhy

Subscriptionhttp://jimmunol.org/subscription

is online at: The Journal of ImmunologyInformation about subscribing to

Permissionshttp://www.aai.org/About/Publications/JI/copyright.htmlSubmit copyright permission requests at:

Email Alertshttp://jimmunol.org/alertsReceive free email-alerts when new articles cite this article. Sign up at:

Print ISSN: 0022-1767 Online ISSN: 1550-6606. Immunologists All rights reserved.Copyright © 2006 by The American Association of1451 Rockville Pike, Suite 650, Rockville, MD 20852The American Association of Immunologists, Inc.,

is published twice each month byThe Journal of Immunology

by guest on July 16, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

by guest on July 16, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

CDR3 Spectratyping Analysis of the TCR Repertoire inMyasthenia Gravis1

Yoh Matsumoto,2* Hidenori Matsuo,†‡ Hiroshi Sakuma,* Il-Kwon Park,* Yukiko Tsukada,*Kuniko Kohyama,* Takayuki Kondo,† Satoshi Kotorii,† and Noritoshi Shibuya†‡

Because myasthenia gravis (MG) is an autoimmune disease mediated by Abs specific for the acetylcholine receptor, helper T cellsplay a role in Ab production. In this study, we have performed large-scale cross-sectional and longitudinal TCR studies by CDR3spectratyping using PBL and thymus tissues from MG patients. We found that there was no preferential usage of any particularTCR �-chains that was identical among MG patients. However, the longitudinal study clearly demonstrated that one or more TCRV� expansions persisted frequently in MG patients. Importantly, persistent TCR expansions correlated with clinical severity andhigh anti-acetylcholine receptor Ab titer. Finally, examinations of T cells expressing CXCR5, i.e., follicular B-helper T cells,revealed that spectratype expansions in MG patients were detected mainly in the CD4� CXCR5� T cell populations, whereasCD8� T cells were the major source of clonal expansion in healthy subjects. These findings suggest that persistent clonal expan-sions of T cells in MG patients are associated with the development and maintenance of MG. Close examination of pathogenic Tcells in MG provides useful information to elucidate the pathogenesis and to estimate the disease status. The Journal of Immu-nology, 2006, 176: 5100–5107.

M yasthenia gravis (MG)3 is an autoimmune disordercharacterized by weakness and fatigability of skeletalmuscles caused by Abs against the acetylcholine re-

ceptor (AChR). The Abs bind to AChR at the postsynaptic mem-brane, leading to receptor destruction and disturbance in neuro-muscular transmission (1). Although it is clear that Abs againstAChR in most patients are final effectors, T cells play an importantrole in helping the Ab production in MG. Another characteristicfeature of MG is abnormalities in the thymus. Lymphofollicularhyperplasia of the thymic medulla occurs in 65% and thymoma isfound in 15% of MG patients (2). Furthermore, it was demon-strated that AChR, a key molecule for the development of MG, isexpressed in the thymus (3) and that thymi showing hyperplasia orwith thymoma generate more T cells, including AChR-specific Tcells (4–6). On the basis of these findings, it is generally believedthat anti-AChR Abs are generated with the help of T cells in thethymus of MG patients. Thus, characterization of the AChR-spe-cific T cells, especially their TCR usage, is essential for the elu-cidation of the pathomechanisms of MG and for the developmentof specific immunotherapy for the disease.

To date, analysis of the TCR usage by pathogenic T cells in MGpatients has revealed extremely controversial results. Cohen-Kaminsky and colleagues (7) showed that V�5.1-positive T cellsin the thymus were slightly increased in MG patients comparedwith control subjects (8). In contrast, other groups observed oli-goclonal expansion of V�4, V�6, V�12, or V�13 in MG patients(9–11). Moreover, Infante et al. (12) found a diverse TCR reper-toire with no preferential usage of particular TCRs in MG patients.Although the precise reasons for these discrepancies remain un-clear, they may reflect the difference in the materials used and themethods for the TCR analysis. In a previous study of multiplesclerosis (MS) patients by CDR3 spectratyping analysis (13), wehave demonstrated persistent or transient oligoclonal expansionsof V�5.2 in the PBL and/or cerebrospinal fluid T cells in �70% ofthe patients studied longitudinally but only in 30% of MS patientsin a cross-sectional study. These results strongly suggest that alongitudinal study is essential for TCR repertoire analyses.

In this study, we have performed both cross-sectional and lon-gitudinal studies using PBL and thymus tissues from MG patientsand PBL from healthy subjects. It revealed no preferential usage ofTCRs that recurred in MG patients in the cross-sectional study.However, persistent oligoclonal expansions of V�s, which differedfrom patient to patient, were detected in the longitudinal study. Im-portantly, they correlated with clinical severity and high anti-AChRAb titer. These findings suggest that activated and clonally expandedT cells detected by CDR3 spectratyping are associated with the de-velopment of MG and provide useful information to elucidate itspathogenesis and to develop tailor-made immunotherapy.

Materials and MethodsPatients and healthy subjects

Thirty-eight MG patients, 8 males and 30 females (mean age, 57.9 � 15.5years; range, 25–80 years), attending the Departments of Neurology, Na-gasaki Medical Center of Neurology, or Nagasaki University Hospital wereincluded in the present study. All subjects’ consent was obtained, and thestudy was approved by the Institute Review Board. The diagnosis ofMG was made on the basis of the clinical presentation and confirmedby laboratory examinations, such as repetitive nerve stimulation tests,

*Department of Molecular Neuropathology, Tokyo Metropolitan Institute for Neuro-science, Fuchu, Tokyo, Japan; †Division of Clinical Research and Department ofNeurology, National Hospital Organization, Nagasaki Medical Center of Neurology,and ‡Department of Clinical Neurosciences, Nagasaki University Graduate School ofBiomedical Sciences, Kawatana, Nagasaki, Japan

Received for publication October 27, 2005. Accepted for publication February2, 2006.

The costs of publication of this article were defrayed in part by the payment of pagecharges. This article must therefore be hereby marked advertisement in accordancewith 18 U.S.C. Section 1734 solely to indicate this fact.1 This study was supported in part by Grants-in-Aid from the Ministry of Education,Japan and from the Neuroimmunological Disorders Research Committee of the Min-istry of Health, Labor and Welfare of Japan.2 Address correspondence and reprint requests to Dr. Yoh Matsumoto, Department ofMolecular Neuropathology, Tokyo Metropolitan Institute for Neuroscience, Musash-idai 2-6 Fuchu, Tokyo 183-8526, Japan. E-mail address: matyoh@tmin.ac.jp3 Abbreviations used in this paper: MG, myasthenia gravis; AchR, acetylcholine re-ceptor; MS, multiple sclerosis.

The Journal of Immunology

Copyright © 2006 by The American Association of Immunologists, Inc. 0022-1767/06/$02.00

by guest on July 16, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

anti-AChR, and computerized tomography images of mediastinum. Forclinical assessment of the severity of the disease, the Myasthenia GravisFoundation of America clinical classification (from class I: purely oculardisease to class V: defined by intubation) (14) and MG-activities of dailyliving (15) were used. Muscle-specific tyrosine kinase-seropositive MGpatients were not included in this study. Forty-four healthy volunteers, 21males and 23 females (mean age, 40.1 � 17.6 years; range, 22–88 years)who had no episodes of common cold or influenza infection within theprevious month were examined as controls. Ten milliliters of heparinizedblood was drawn from patients and control subjects, and PBL was isolatedusing the density gradient method. Thymus tissues (n � 9) of MG patientswere obtained by thymectomy. Six of them showed hyperplasia and threewere uninvolved thymus adjacent to a thymoma.

Cross-sectional and longitudinal studies

The cross-sectional study was performed as described above. When thesame person was examined more than once, the results of the first exam-ination were included in the cross-sectional study. The longitudinal studywas performed on 22 MG patients and 17 healthy subjects, and the resultsobtained from patients and subjects with sufficient follow-up periods wereselected for analysis. The spectratype expansion pattern obtained in thelongitudinal study was classified into the three categories: 1) heteroge-neous, where no particular spectratype expansion was observed in multipleexaminations in which the normal pattern was included frequently; 2) per-sistent, where spectratype expansion of particular V�s were constantlyobserved during the observation period; and 3) intermediate, where a par-ticular spectratype expansion was occasionally observed.

cDNA synthesis and PCR amplification

RNA was extracted from PBL using RNAzol B (Biotecx Laboratories).cDNA was then synthesized by reverse transcription using ReverTra Ace(Toyoba) and amplified in a thermal cycler (PerkinElmer) using primerpairs for TCR. Primers for Va�1-24 were the same as those used in a pre-vious study (16). Ca� primer was labeled with Cy-5 or rhodamine or was leftunlabeled.

CDR3 spectratyping

CDR3 spectratyping was performed as described previously (17). cDNAwas amplified with V�-specific and rhodamine-labeled C� primers, andundiluted or diluted PCR products were added to an equal volume of for-mamide/dye loading buffer and heated at 94°C for 2 min. Two microliters

of the samples was applied to a 6% acrylamide sequencing gel. Gels wererun at 30 W for 3 h 30 min at 50°C. Then, the fluorescence-labeled DNAprofile on the gel was directly recorded using an FMBIO fluorescenceimage analyser (Hitachi). Spectratype expansion was evaluated by visualinspection and density analysis of the image using software attached to thefluorescence image analyzer as shown in Fig. 1, C and D, and graded intotwo categories: 1)“Oligoclonal pattern,” appearing as an increase of thedensity and thickness of one band within a normal spectratype profile (dis-tortion of the Gaussian distribution) (18); and 2) “Monoclonal” one, witha marked increase of the density and thickness of one band with faint or noadditional spectratypes (18). We tested for contaminations by the reagentsused in PCR every 10 PCR analyses by doing PCR without templates; ifpresent, all reagents used and results obtained during the period werediscarded.

Sequencing of PCR products

cDNA isolated from spectratypes of interest on the acrylamide gel wasreamplified with V� and unlabeled C� primers to remove the fluorescenceattached to the primer used for CDR3 spectratyping. This process wasessential for cloning. To avoid biased amplification of a particular TCRclone, PCR was performed for only five cycles. Then, PCR products wereligated into pT-Adv vector and cloned using the AdvanTAge PCR CloningKit (Clontech Laboratories) according to the manufacturer’s instructions.The plasmid DNA was then sequenced using Cy5-labeled C� primer andThermo Sequenase Fluorescent Labeled Primer Cycle Sequencing Kit onan ALFexpress DNA sequencer (Pharmacia Biotech). CDR3 length is de-fined as a region starting from an amino acid residue after the CASS se-quence of most V� segments and ending before the GXG box in the J�region, as described previously (19).

Isolation and TCR analysis of CXCR5-positive T cells

Because the majority of CXCR5-expressing cells were CD4� T and B cells(20), we decided to examine CD4�, CD8�, and CD4�CXCR5� T cells.The number of CD8�CXCR5� T cells was too small for the analysis.CD4�CXCR5� T cells were isolated using an AutoMACS (Miltenyi Bio-tec) following the manufacturer’s instructions. In brief, we removed ad-herent cells and some B cells from MG and control PBL by negativeselection with 1F5 (anti-CD20) plus LM2/1.6.11 (anti-Mac-1) followed byanti-mouse IgG-microbeads (Miltenyi Biotec). The cells were then posi-tively (the CD8� T cell population) or negatively (the CD4� T cell pop-ulation) selected with OKT8. Finally, CD4� T cells were further separatedinto CD4�CXCR5� and CD4�CXCR5� T cells with anti-CXCR5 mAb

FIGURE 1. CDR3 spectratypingprofiles of PBL showing the normalspectratype pattern (A) in a healthysubject and spectratype expansion(B) in a MG patient. In the normalpattern, each V� shows a Gaussiandistribution without any spectratypeexpansion (A). In contrast, this MGpatient exhibits expansion of severalspectratypes indicated by arrows (B).The profiles of V�4, V�8, and V�20spectratypes of a healthy subject (A)and V�3, V�8, and V�14 spectra-types of a MG patient (B) revealed bythe density analysis are shown in Cand D, respectively.

5101The Journal of Immunology

by guest on July 16, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

(DakoCytomation) and anti-mouse IgG-microbeads. Three populations,CD4� T, CD8� T, and CD4�CXCR5� T cells, were subjected to the TCRanalysis. We determined the purity of isolated cells by flow cytometry afterincubation of the cells with FITC-labeled anti-mouse IgG and confirmedthat each population was �90% pure.

ResultsCross-sectional study of MG patients and healthy individuals byCDR3 spectratyping

We first performed a cross-sectional study of 38 MG patients and44 healthy subjects to characterize the TCR repertoire in MG. Of38 patients, six cases of the ocular type, 15 cases of MG withthymoma, 13 cases of the early onset, and two cases of the lateonset were included. Our previous studies on neuroimmunologicaldisorders revealed that a particular type of V� TCR, i.e., V�5.2,showed clonal expansion more frequently in MS (13), whereasthere was no preferential usage of TCR in Guillain-Barre syn-drome (21). When multiple examinations were performed on thesame individuals, the results obtained in the first examination wereincluded in the cross-sectional study. Representative spectratypingprofiles are shown in Fig. 1 and the results summarized in Table I.Nearly half of controls (46.5%) showed a normal spectratype pat-tern (Fig. 1A). The density analysis revealed unskewed Gaussiandistributions (Fig. 1C). In contrast, only a quarter of the MG pa-tients showed the normal pattern ( p � 0.05) (Table I). Remainingcontrol and MG samples contained one or more expanded spec-tratypes (arrows in Fig. 1B). The distorted Gaussian distributions(V�3 and V�8 in Fig. 1D) and the monoclonal pattern (V�14 inFig. 1D) were detected in the density analysis. In healthy subjects,

V�9 spectratype expansions seemed to be more frequent comparedwith other V�s, but there was no significant difference. In MGpatients, 29 cases (77.3%) showed the expansion of one or morespectratypes (Table I). Although these were more frequent in theMG patients than in healthy subjects, they did not show preferen-tial usage of a common V�, unlike MS.

We also examined nine thymus samples from MG patients. His-tological examination demonstrated hyperplasia in six cases andnormal looking tissues taken adjacent to thymoma in three cases.Interestingly, they all showed the normal spectratype pattern with-out any spectratype expansion (Fig. 2).

Longitudinal study of selected MG patients

We have currently performed longitudinal studies on 22 MG pa-tients and 17 healthy subjects. Among them, we selected 16 MGand 11 control cases; we were able to examine MG patients formore than 2 years and healthy subjects for 1 year (Table II andTable IV). Representative profiles obtained from MG patients areillustrated in Fig. 3, and all of the results are summarized in TableII. Longitudinal studies of oligoclonal expansions of TCR revealedthree behavior patterns. 1) In persistent type, expansions were de-tected in all samples, as illustrated in Fig. 3 and Table II. In Fig.3, V�9 and V�11 expansion persisted over 1 year and 5 mo. Nineof 16 patients showed this type of spectratype expansion during theobservation periods (patients 4, 5, 7, 8, 10, 11, 12, 13, and 15 inTable II). 2) Heterogeneous type ranged from a normal pattern toone of short-lived expansions (patients 1, 2, 3, 14, and 16 in TableII). 3) In intermediate type, a particular spectratype expansion wasintermittently observed (patients 6 and 9 in Table II).

We also examined HLA-DR haplotypes of MG patients sub-jected for the longitudinal study as shown in Table II ( fourth col-umn from the left). As far as we examined, there was no correlationbetween the HLA haplotype and TCR V� usage.

CDR3 sequences of TCR clones derived from expandedspectratypes in MG

In a series of previous studies in MS (13) and its animal models(17, 22), we demonstrated that persistently expanded spectratypes

Table I. The frequency of V� expansion in healthy subjects andpatients with MGa

V� ExpansionHealthy SubjectsIncidence (%)b

Patients with MGIncidence (%)b

Absent (normal pattern) 20 (46.5) 9 (23.7)c

1 0 32 3 13 3 34 3 25.1 1 15.2 2 16 2 37 1 28 4 49 7 (29.2) 7 (28.6)10 2 111 2 6 (20.7)12 2 113 1 314 2 6 (20.7)15 1 016 2 8 (27.6)17 0 018 1 219 0 020 1 321 2 022 4 223 2 7 (28.6)24 1 1

No. of subjects examined 44 38

a The number of spectratypes showing oligoclonal expansion is summarized inhealthy subjects and MG patients. A total of 49 expansions from 44 controls and 67expansions from MG patients are shown.

b Percentages of each V� expansion (the number of samples showing a particularspectratype expansion per total number of samples with any spectratype expansion).Samples with a normal spectratype pattern are not included in percentage calculation.

c The incidence of the normal spectratype pattern was significantly lower in MGpatients than in healthy subjects ( p � 0.05 by �2 test).

FIGURE 2. CDR3 spectratyping of the thymus tissue. All six thymustissues examined showed the normal spectratyping pattern without anyexpansion.

5102 TCR REPERTOIRE ANALYSIS IN MG

by guest on July 16, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

represent expansions of particular TCR clones. To test whetherthat also applies in MG, we did a similar analysis using PBL frompatient 7 (Table II), which showed V�11 expansion over 2 yearsin 4 of 5 examinations. The expanded spectratypes were cut from

the gel, and cDNA was extracted, cloned, and sequenced. Asclearly demonstrated in Table III, we found a TCR clone with theCASS-ATPNEQFF-GPG sequence in the CDR3 throughout the 2years. Evidently, it represents an expansion of a single T cell clone.

Table II. Clinical characteristics and spectratyping analysis of MG patientsa

a N.E., Not examined; N, the normal spectratype pattern; 1–24, the spectratype expansion of each V�.

5103The Journal of Immunology

by guest on July 16, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

Spectratype expansion found in healthy subjects was unstableand did not last for a long period

We performed similar longitudinal studies in elderly healthy sub-jects who occasionally show clonal T cell expansion (23–25) tolook for differences from those in our MG patients. As can be seenin Table IV, persistent spectratype expansions were found in sub-jects in their seventies and transient ones in their fifties and sixties.These TCR expansions are thought to occur due to exogenous Agstimulation/infections and to reduce the T cell diversity (26). In-terestingly, one volunteer (37 years old) residing in the communityhad a persistent cough when first tested (not included in Table IV).As shown in Fig. 4A, there were V�2, 3, 6, 14, 20, and 24 expan-sions at that time point. Three months later when the cough haddisappeared, the number of expanded spectratypes had reducedmarkedly (Fig. 4B). This change strongly suggests that spectratypeexpansions in middle-aged to elderly subjects can be elicited byexogenous Ag stimulation and may be short-lived.

We also tested for correlations between spectratype pattern andMG severity (Table V) or anti-AChR Ab titer. The persistent pat-tern was observed more frequently in the patients with grade III-Vthan in patients with grade I-II and controls ( p � 0.02) (Table V).With regard to anti-AChR Abs, the persistent pattern seemed to befound more frequently in patients with the titers above 10 nM

rather than below 10 nM (data not shown). Taken together, theysuggest that clonal expansions of particular V�s correlate withdisease severity and are thus involved in pathogenesis of MG.

Analysis of CXCR5-positive T cells provides useful informationfor identification of pathogenic T cells in MG

These longitudinal studies on total T cells may indicate persistentTCR clonal expansions in autoimmune responses in MG patients.However, some persistent clonal expansions in healthy subjectswere indistinguishable from those found in MG patients. To pursuethat further, we spectratyped CD4�CXCR5� T cells isolated fromhealthy subjects and MG patients (Table VI). Because these T cellsreportedly act as helpers for follicular B cells (20), they might playan essential role in anti-AChR Ab production and in thymic hy-perplasia, a classic sample of lymphoid neogenesis in autoimmunedisorders. As shown in Table VI, the spectratype expansions foundin three healthy subjects were mainly restricted to CD8� T cells,as reported previously (26, 27), whereas in 2 of 3 cases CD4� andCD4�CXCR5� T cells showed normal spectratype patterns. Char-acteristically, in most MG patients, spectratype expansions foundin the CD4� T cell population were also detected inCD4�CXCR5� T cells. These findings clearly demonstrate that

Table III. Amino acid and nucleotide sequences of the CDR3 of V�11 TCRs extracted from patient 7a

Sample Follow-up V� (N)D(N) J� Frequency (%)

4129 0 mob CASS A T P N E Q F F GPG 11/13 (85)CASS L S V Y A V F F GDG 1/13 (8)CACS A T H N E Q F F GPG 1/13 (8)

4148 6 moc CASS A T P N E Q F F GPG 2/3 (66)A T P N E R F F GPG 1/3 (33)

4609 1 year CASS A T P N E Q F F GPG 9/10 (90)CASS A T H N E Q F F GPG 1/10 (10)

4846 2 years CASS A T P N E Q F F GPG 8/8 (100)

a PCR products were extracted from V�11 spectratype showing expansion, and reamplified and the CDR3 were sequenced after cloning.b The first examination in the longitudinal examination.c Six months after the first examination.

FIGURE 3. Longitudinal studies of a MG patient showing the persistent pattern. Multiple examinations revealed that V�9 and V�11 expansions (A–C)indicated by arrows persisted over 1.5 years. Spectratype expansions other than V�9 and V�11 are not marked. This patient corresponds to Patient 8 inTable II.

5104 TCR REPERTOIRE ANALYSIS IN MG

by guest on July 16, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

the clonal expansions of T cells in MG patients are different fromthose in healthy subjects.

DiscussionA large body of evidence indicates that the final effectors in MGare autoantibodies directed against targets at the neuromuscularjunction (28, 29). Moreover, Ag-specific Th cells are essential forthe generation of myasthogenic autoantibodies and for the devel-opment of experimental autoimmune MG (30, 31). Hence, therehave been several attempts to characterize MG-associated T cells.However, the TCR usage of pathogenic T cells is verycontroversial.

Most studies used either FACS analysis with anti-TCR mAbs orCDR3 spectratyping. By FACS analysis using PBL or thymiccells, V�5.1- (7), V�12- (9), and V�13- (11) positive T cells werereported to be significantly increased in MG patients. In our ex-perience, however, this method is not sufficiently sensitive forTCR analysis of pathogenic T cells in autoimmune diseases. In

EAE induced in Lewis rats by immunization with myelin basicprotein, several lines of evidence indicate that encephalitogenic Tcells mainly use V�8.2. Clonal expansion of this cell type wasdetectable by CDR3 spectratyping in the lymphoid organ at thevery early stages of the disease (32). In sharp contrast, by flowcytometry, this expansion could not be detected in the lymphoidorgan throughout EAE (33). V�8.2 expansion was only detectableat the peak of EAE in the CNS, where V�8.2-positive cells ac-counted for �20% (33). This is mainly because the numbers ofcirculating encephalitogenic T cells are too small for detection byFACS analysis. So far, two CDR3 spectratyping studies have fo-cused on the TCR repertoire in MG. Navneetham et al. reportedthat V�4 and V�6 were used significantly more in MG patientsthan in controls (10). In contrast, Infante et al. (12) found no sig-nificant difference in the TCR repertoire between MG patients andcontrol subjects. The inconsistencies between these two reportsand between these reports and our results may be largely attribut-able to the number of patients and controls and to the presence or

Table IV. Longitudinal spectratyping analysis of healthy subjectsa

a N, The normal spectratype pattern; 1–24, the spectratype expansion of each V�.

5105The Journal of Immunology

by guest on July 16, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

absence of the longitudinal examination. In previous studies, thenumber of cases for the analysis was too small, and each case wasexamined only once.

We now summarize intriguing findings from the large-scalecross-sectional and longitudinal analyses of PBL and thymus takenfrom MG patients and healthy subjects by CDR3 spectratyping.First, TCR spectratype expansions were more frequently found inMG patients than healthy subjects ( p � 0.05), though withoutpreferential usage of any particular TCR �-chains. Second, thelongitudinal study clearly demonstrated persistent TCR V� expan-sions more frequently in the patients with severe MG. Althoughpersistent spectratype expansions varied from patient to patient,they correlated with severe clinical status and high anti-AChR Ab

titers. Finally, examination of CXCR5� T cells may help to focuson pathogenic Th cells in MG. In MG patients, spectratype expan-sions were found in CD4� and CD4�CXCR5� T cells, whereasspectratype expansion seen in healthy subjects was observed inCD8� T cells. We have tried to exclude the possibility that pre-ceding infection occurred in MG patients may induce spectratypeexpansion unrelated to the MG pathogenesis as follows. For thispurpose, we have checked all of the patients subjected for thelongitudinal study whether they have clinical signs and laboratorydata suggesting infection. All of the patients did not have fever andincreased white blood cells at the time of spectratype examina-tions. Only patient 12 at the second examination exhibited slightlyincreased C-reactive protein. However, it is unlikely that thiswould modulate the results of CDR3 spectratyping, because theresults at the second examination were almost the same as those ob-tained in other examinations. Taken together, the present studystrongly suggests that T cells showing persistent clonal expansion areinvolved in the development and maintenance of MG.

It should be noted that all nine thymus tissues, including sixhyperplasmic thymi, exhibited normal spectratype patterns. Al-though the thymus has previously been implicated as a possiblesite of the disease origin (�75% of patients have thymic abnor-malities) (1), its role in MG is still a mystery (29). Our findingsfrom CDR3 spectratyping suggest either that the thymus includesno clonally expanded pathogenic T cells or that their frequency istoo low to detect even using this method. Importantly, �30% of

Table V. Classification of the spectratyping pattern in the follow-upstudy of MG patients and healthy subjects

Classification

MG Patients

Healthy SubjectsaTotal Grade III–Va

Persistent 9 (56%) 7 (78%) 2 (18.2%)Intermediate 2 (13%) 1 (11%) 2 (18.2%)Heterogeneous 5 (31%) 1 (11%) 7 (63.6%)Total 16 9 11

a The numbers of Grade III–V MG patients showing the persistent pattern andhealthy subjects showing the heterogeneous pattern are significantly higher than thatof their counterparts ( p � 0.02 by �2 test).

Table VI. CDR3 spectratyping analysis of CD4�, CD8�, and CD4�CXCR5� T cells isolated from healthysubjects and MG patients

CD4� T Cells CD8� T Cells CD4�CXCR5� T Cells

Healthy subjects1 Na 1�, 6�, 16� Na

2 8�, 24� 8�, 24� 8�3 Na 11�, 16� Na

MG patients1 2�, 5.1�, 9�, 11�, 12� 2�, 11�2 6�, 11�, 16� 11�, 13�, 16�, 20�, 22�, 23� 11�, 16�3 1�, 3�, 5.2�, 9� 6�, 13�, 14�, 20�, 21� 3�, 5.2�, 6�, 9�4 11� 3�, 6�, 8�, 11�, 15� 11�

a N, Normal spectratype pattern.

FIGURE 4. CDR3 spectratypingprofile of one of control subject dur-ing persistent cough (A) and after itsubsided (B). A volunteer had persis-tent cough when he was first exam-ined by CDR3 spectratyping. Asshown in A, there were V�2, -3, -6,-14, -20, and -24 expansions at thattime point. Three months later whenthe symptoms had disappeared, ex-panded spectratypes were markedlyreduced (B).

5106 TCR REPERTOIRE ANALYSIS IN MG

by guest on July 16, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

MG patients do not show clinical improvement after thymectomy(34). Collectively, it is possible that the thymus plays a role inexacerbation of the disease in some, but not all, MG patients. Inthis regard, recent progress in elucidating the mechanism of lym-phoid neogenesis provides useful information for the understand-ing of the relationship between thymic abnormalities and MG. Asreviewed by Weyand et al. (35), ectopic germinal centers areformed in several organs in the presence of autoantigens,CXCL13-producing tissue cells, follicular dendritic cells, and Band T cells. Although it remains to be elucidated whether all ofthese factors are present in the thymus of MG patients, it is pos-sible because AChRs are found in the thymic myoid cells (36).

In a series of reports, Berrih-Aknin and colleagues (7, 37) haveinsisted that V�5.1-positive T cells are pathogenic in MG, and thatcirculating anti-V�5.1 Abs may regulate the excess of pathogenicV�5.1-positive T cells (8). This conclusion was drawn on the basisthat by FACS analysis, V�5.1-positive T cells were slightly butsignificantly increased in number in the thymus of MG patients.However, this finding was not confirmed by other groups usingflow cytometry and CDR3 spectratyping. Similarly, we have notobserved significant V�5.1 clonal expansion in this study. Takingall these findings into consideration, it can be said that the con-clusion by Berrih-Aknin et al. (7, 37) remains unestablished.

In summary, we were able to identify disease-associated andclonally expanded T cells in patients with MG by CDR3 spectra-typing. Close examination of pathogenic T cells in MG providesuseful information to elucidate the pathogenesis and to estimatethe disease status.

AcknowledgmentsWe are grateful to Dr. Masakatsu Motomura (First Department of InternalMedicine, Nagasaki University, School of Medicine, Nagasaki, Japan) forproviding information about MG patients.

DisclosuresThe authors have no financial conflict of interest.

References1. Drachman, D. B. 1994. Myasthenia gravis. N. Engl. J. Med. 330: 1797–1810.2. Engel, A. G. 1992. Myasthenia gravis and myasthenic syndromes. In Myopathies,

Vol. 62. L. P. Rowland and S. Dimauro, eds. Elsevier Science Publishers, Am-sterdam, pp. 391–455.

3. Wheatley, L. M., D. Urso, K. Tumas, J. Maltzman, E. Loh, and A. I. Levinson.1992. Molecular evidence for the expression of nicotinic acetylcholine receptor�-chain in the mouse thymus. J. Immunol. 148: 3105–3109.

4. Nagvekar, N., A. M. Moody, P. Moss, I. Roxanis, J. Curnow, D. Beeson,N. Pantic, J. Newsom-Davis, A. Vincent, and N. Willcox. 1998. A pathogeneticrole for the thymoma in myasthenia gravis: autosensitization of IL-4- producingT cell clones recognizing extracellular acetylcholine receptor epitopes presentedby minority class II isotypes. J. Clin. Invest. 101: 2268–2277.

5. Buckley, C., D. Douek, J. Newsom-Davis, A. Vincent, and N. Willcox. 2001.Mature, long-lived CD4� and CD8� T cells are generated by the thymoma inmyasthenia gravis. Ann. Neurol. 50: 64–72.

6. Sempowski, G., J. Thomasch, M. Gooding, L. Hale, L. Edwards, E. Ciafaloni,D. Sanders, J. Massey, D. Douek, R. Koup, and B. Haynes. 2001. Effect ofthymectomy on human peripheral blood T cell pools in myasthenia gravis. J. Im-munol. 166: 2808–2817.

7. Truffault, F., S. Cohen-Kaminsky, I. Khalil, P. Levasseur, and S. Berrih-Aknin.1997. Altered intrathymic T-cell repertoire in human myasthenia gravis. Ann.Neurol. 41: 731–741.

8. Aissaoui, A., I. Klingel-Schmitt, J. Couderc, D. Chateau, F. Romagne, F. Jambou,A. Vincent, P. Levasseur, B. Eymard, M. C. Maillot, et al. 1999. Prevention ofautoimmune attack by targeting specific T-cell receptors in a severe combinedimmunodeficiency mouse model of myasthenia gravis. Ann. Neurol. 46:559–567.

9. Grunewald, J., R. Ahlberg, A. K. Lefvert, H. DerSimonian, H. Wigzell, andC. H. Janson. 1991. Abnormal T-cell expansion and V-gene usage in myastheniagravis patients. Scand. J. Immunol. 34: 161–168.

10. Navneetham, D., A. S. Penn, J. F. Howard, Jr., and B. M. Conti-Fine. 1998.TCR-V� usage in the thymus and blood of myasthenia gravis patients.J. Autoimm. 11: 621–633.

11. Xu, B. Y., R. Giscombe, A. Soderlund, M. Troye-Blomberg, R. Pirskanen, andA. K. Lefvert. 1998. Abnormal T cell receptor V gene usage in myasthenia

gravis: prevalence and characterization of expanded T cell populations. Clin. Exp.Immunol. 113: 456–464.

12. Infante, A. J., J. Billargeon, E. Kraig, L. Lott, C. Jackson, G. J. Haemmerling,R. Raji, and C. David. 2003. Evidence of a diverse T cell receptor repertoire foracetylcholine receptor, the autoantigen of myasthenia gravis. J. Autoimmun. 21:167–174.

13. Matsumoto, Y., W. K. Yoon, Y. Jee, K. Fujihara, T. Misu, S. Sato, I. Nakashima,and Y. Itoyama. 2003. Complementarity-determining region 3 spectratypinganalysis of the T cell repertoire in multiple sclerosis. J. Immunol. 170:4846–4853.

14. Jaretzki, A. I., R. J. Barohn, R. M. Ernstoff, H. J. Kaminski, J. C. Keesey,A. S. Penn, and D. B. Sanders. 2000. Myasthenia gravis: recommendations forclinical research standards. Neurology 55: 16–23.

15. Wolfe, G. I., L. Herbelin, S. P. Nations, B. Foster, W. W. Bryan, andR. J. Barohn. 1999. Myasthenia gravis activities of daily living profile. Neurology52: 1487–1489.

16. Choi, Y., B. Kotzin, L. Herron, J. Callahan, P. Marrack, and J. Kappler. 1989.Interaction of Staphylococcus aureus toxin “superantigen” with human T cells.Proc. Natl. Acad. Sci. USA 86: 9841–8945.

17. Kim, G., N. Tanuma, T. Kojima, K. Kohyama, Y. Suzuki, Y. Kawazoe, andY. Matsumoto. 1998. CDR3 size spectratyping and sequencing of spectratype-derived T cell receptor of spinal cord T cells in autoimmune encephalomyelitis.J. Immunol. 160: 509–513.

18. Maini, M. K., L. R. Wedderburn, F. C. Hall, A. Wack, G. Casorati, andP. C. L. Beverley. 1998. A comparison of two techniques for the moleculartracking of specific T-cell responses: CD4� human T-cell clones persist in astable hierarchy but at a lower frequency than in the CD8� population. Immu-nology 94: 529–535.

19. Rock, E. P., P. R. Sibbald, M. M. Davis, and Y.-H. Chien. 1994. CDR3 length inantigen-specific immune receptors. J. Exp. Med. 179: 323–328.

20. Moser, B., P. Schaerli, and P. Loetscher. 2002. CXCR5� T cells: follicular hom-ing takes center stage in T-helper-cell responses. Trends Immunol. 23: 250–254.

21. Koga, M., N. Yuki, Y. Tsukada, K. Hirata, and Y. Matsumoto. 2003. CDR3spectratyping analysis of the T cell receptor repertoire in Guillain-Barre andFisher syndromes. J. Neuroimmunol. 141: 112–117.

22. Matsumoto, Y., Y. Jee, and M. Sugisaki. 2000. Successful TCR-based immuno-therapy for autoimmune myocarditis with DNA vaccines after rapid identificationof pathogenic TCR. J. Immunol. 164: 2248–2254.

23. Schwab, R., P. Szabo, J. S. Manavalan, M. E. Weksler, D. N. Posnett,C. Pannetier, P. Kourilsky, and J. Even. 1997. Expanded CD4� and CD8� T cellclones in elderly humans. J. Immunol. 158: 4493–4499.

24. Monteiro, J., R. Hingorani, I. Choi, J. Silver, R. Pergolizzi, and P. K. Gregersen.1995. Oligoclonality in the human CD8� T cell repertoire in normal subjects andmonozygotic twins: implications for studies of infection and autoimmune dis-eases. Mol. Med. 1: 614–624.

25. Khan, N., N. Shariff, M. Cobbold, R. Bruton, J. A. Ainsworth, A. J. Sinclair,L. Nayak, and P. A. H. Moss. 2002. Cytomegalovirus seropositivity drives theCD8 T cell repertoire toward greater clonality in healthy elderly individuals.J. Immunol. 169: 1984–1992.

26. LeMaoult, J., I. Messaoudi, J. S. Manavalan, H. Potvin, D. Nikolich-Zugich,R. Dyall, P. Szabo, M. E. Weksler, and J. Nikolich-Zugich. 2000. Age-relateddysregulation in CD8 T cell homeostasis: kinetics of a diversity loss. J. Immunol.165: 2367–2373.

27. Messaoudi, I., J. Lemaoult, J. A. Guevara-Patino, B. M. Metzner, andJ. Nikolich-Zugich. 2004. Age-related CD8 T cell clonal expansions constrictCD8 T cell repertoire and have the potential to impair immune defense. J. Exp.Med. 200: 1347–1358.

28. Lindstrom, J., D. Sehlton, and Y. Fujii. 1988. Myasthenia gravis. Adv. Immunol.42: 233–284.

29. Vincent, A. 2002. Unravelling the pathogenesis of myasthenia gravis. Nat. Rev.Immunol. 2: 797–804.

30. Fujii, Y., and J. Lindstrom. 1988. Regulation of antibody production by helper Tcell clones in experimental autoimmune myasthenia gravis. J. Immunol. 141:3361–3369.

31. Zhang, G., B. Xiao, M. Barkhiet, P. van der Meide, H. Wigzell, H. Link, andT. Olsson. 1996. Both CD4� and CD8� T cells are essential to induce experi-mental autoimmune myasthenia gravis. J. Exp. Med. 184: 349–356.

32. Sakuma, H., K. Kohyama, Y. Jee, and Y. Matsumoto. 2004. Tracking of V�8.2-positive encephalitogenic T cells by CDR3 spectratyping and subsequent South-ern blot hybridization in Lewis rats after neuroantigen sensitization. J. Immunol.173: 4516–4522.

33. Tsuchida, M., Y. Matsumoto, H. Hirahara, H. Hanawa, K. Tomiyama, andT. Abo. 1993. Preferential distribution of V�8.2-positive cells in the central ner-vous system of rats with myelin basic protein-induced autoimmune encephalo-myelitis. Eur. J. Immunol. 23: 2399–2406.

34. Tellez-Zenteno, J. F., J. M. Remes-Troche, G. Garcia-Ramos, B. Estanol, andJ. Garduno-Espinoza. 2001. Prognostic factors of thymectomy in patients withmyasthenia gravis: a cohort of 132 patients. Eur. Neurol. 46: 171–177.

35. Weyand, C. M., P. J. Kurtin, and J. J. Goronzy. 2001. Ectopic lymphoid orga-nogenesis: a fast track for autoimmunity. Am. J. Pathol. 159: 787–793.

36. Schluep, M., N. Willcox, A. Vincent, G. K. Dhoot, and J. Newsom-Davis. 1987.Acetylcholine receptors in human thymic myoid cells in situ: and immunohisto-chemical study. Ann. Neurol. 22: 212–222.

37. Jambou, F., W. Zhang, M. Menestrier, I. Klingel-Schmitt, O. Michel,S. Caillat-Zucman, A. Aissaoui, L. Landemarre, S. Berrih-Aknin, andS. Cohen-Kaminsky. 2003. Circulating regulatory anti-T cell receptor antibodiesin patients with myasthenia gravis. J. Clin. Invest. 112: 265–274.

5107The Journal of Immunology

by guest on July 16, 2018http://w

ww

.jimm

unol.org/D

ownloaded from