Common T-cell Receptor Vp Usage in Oligoclonal T Lymphocytes Derived from Cerebrospinal Fluid and Blood of Patients with Multiple Sclerosis Soon Jin Lee, PhD, Kai W. Wucherpfennig, MD, PhD, Staley A. Brod, MD, Deborah Benjamin, BS, Howard L. Weiner, MD, and David A. Hafler, MD
T-cell populations were investigated in the blood and cerebrospinal fluid of patients with multiple sclerosis and other neurological diseases. Individual T cells were directly cloned from the cerebrospinal fluid and blood before in viuo expansion, and their clonotypes were compared by Southern blot analysis of the rearrangement patterns of their T-cell receptor p chain and y chain genes. This allowed the determination of whether two T cell clones shared the same Tcell receptor and thus arose from identical, clonally expanded (oligoclonal) progenitor T cells. As an extension of previous studies, oligoclonal T-cell clones were identified in both cerebrospinal fluid and blood populations in 5 of 9 patients with inflammatory demyelinating disease among a total of 486 blood and cerebrospinal fluid T-cell clones. In contrast, no clonally expanded T-cell populations were found among a total of 424 clones derived from either blood of 4 normal control subjects or blood and cerebrospinal fluid of 8 patients with other neurological diseases. Analysis of Tcell receptor V, genes among 4 oligoclonal T-cell populations derived from 3 patients with multiple sclerosis demonstrated common usage of the Vp12 gene segment. These data suggest that oligoclonal T cells share similar specificities and that clonal expansion may have resulted from specific stimulation by an antigen. Moreover, identical clones between blood and cerebrospinal fluid were observed in 3 of 9 patients with demyelinating disease, thus providing further evidence of an equilibrium between peripheral and central nervous system immune compartments. Lee SJ, Wucherpfennig KW, Brod SA, Benjamin D, Weiner HL, H d e r DA. Common T-cell receptor V, usage in oligoclonal T lymphocytes derived from cerebrospinal fluid and blood of patients with multiple sclerosis. Ann Neurol 1991;29:33-40
Multiple sclerosis is a chronic, inflammatory, demyelinating disease of the central nervous system (CNS) { 1, 23. Pathological studies have indicated that the majority of infiltrating cells at sites of active demyelination are activated T cells and macrophages { 3 ] . A major hypothesis regarding the pathogenesis of multiple sclerosis is that T cells reactive with myelin protein determinants migrate into the CNS and, either directly or by recruitment of effector cells, mediate damage to white matter tracts [4, 51. Thus, characterization of immune T cells in the CNS and blood of patients with multiple sclerosis is likely to be important in attempting to understand the pathogenesis of the disease. It is now known that each T cell expresses a single receptor that consists of either an a-p or y-6 heterodimer that recognizes antigen in association with major histocompatibility gene complex proteins. The highly diverse T-cell receptor (TCR) repertoire is generated
during T-cell ontogeny in the thymus by rearrangement of TCR genes, which consist of variable (V), diversity (D), junctional and constant (C) gene regions {b}.Whether two T cells are the same and thus derived from a common progenitor cell can be determined by analysis of these TCR gene rearrangements. That is, Southern blot analysis of DNA from T cell clones digested with different restriction enzymes and probed with different TCR genes can identify two clones that have a high probability of sharing the same TCR gene rearrangements and thus can be assumed to arise from the identical progenitor T cell [7]. If T-cell populations in the blood and cerebrospinal fluid (CSF) of patients with multiple sclerosis are derived from a large number of progenitor cells, examination of TCR genes of many T-cell clones would reveal different gene rearrangements. Alternatively, there may be more restricted expansion of T cells in
From the Center for Neurologic Disease, Division of Neurology, Department of Medicine, Brigham and Women’s Hospital, and Harvard Medical School, Boston, MA.
Address correspondence to Dr HaAer, Center for Neurologic Disease, Brigham and Women’s Hospital, 75 Francis Street, Boston, MA 02115.
Received Oct 19, 1989, and in revised form Apr Accepted for publicatiun Jd 9, 1990.
u),
9 and Jul9, 1990.
Copyright 0 1991 by the American Neurological Association
33
Table 1, Oligoclonal T Lymphocytes fvom Patients with Multiple ScEerosis T-cell Clonalitya Patient
Disease
Multiple sclerosis Gr ... Ka ... Mu ... Po ... Ca ... Re ... Da ... Sh ... Auf ... Other neurological diseases Ki Myasthenia sz SSPE Av Lyme Fe Stroke McK NPH Ga GBS Wi SSPE Ve HZV Normal controls Pw ... He ... Ka ... Ha ...
Age (yr)
Sex
34
F F M F M
39 38 51 22 55
26 40 5 30 20
60 65 65 52 16
63 42 28 30 36
M E
F F
F F M F F F F M
Duration of Disease
Oligoclonal Bands
10 yr 8 Yr 8 mo 8 yr 5 Yr 10 yr 6 Yr 15 y t 4 mo
3 Yr 8 Yr 3 mo 1 mo
6 mo 4 mo 3 Yr 9 mo
Blood Blood'
CSFb
& CSP
Totald
Yes No 1 Band No No No Yes No Yes
2126 016 014 1 2/10 0133
18128
2 0 0
20154' 5/33'
ND Yes NO No
0152 0119 0135 Oil 1 0114
NO
0116
0119 019 017
5127 0137 0130 0156 0140
0136 0118 2127 ND
019 018
012 012 5
No Yes No
ND
0126
0114
0133 0120
ND
F
...
...
I; M
...
...
0166" 0128
...
...
012 8
M
...
...
0113
0178'
3
3160
0 4 0 0 0
0189
4/56 0155 0127 2/34
ND 0 0 0 0
0152 0128 014 3
ND 0
0113 0139 0126 0168'
ND
0120'
"Total number of oligoclonal T cells: barno% blood or CSF, 'berween blood and CSF, 'total among blood and CSF. See Results section for description of clone patterns. cPrcviously described clones [S}. fPatietit with single episode of perivascular lymphocytic infiltrates and demyelination. Myasthenia = myasthenia gravis; SSPE subacute sclerosing panencephalitis; Lyme = Lyme meningoencephalitis; h'PH hydrocephalus; GBS = Guillain Bar& syndrome; ND = not done; HZV = herpes zoster virus meningoencephalitis.
the immune system of patients with multiple sclerosis resulting in either oligoclonal or possibly monoclonal T-cell populations. Southern blot analysis of a large number of T-cell clones derived from patients with multiple sclerosis would then show common TCR gene rearrangements. We investigated this in a previous study and found oligoclonal T-cell populations in the CSF and blood of some but not all patients with multiple sclerosis ES}. The observation that the T-cell repertoire may be altered in autoimmune disease with expansion of particular T-cell populations has subsequently been reported in other autoimmune diseases r9, 101. To determine whether oligoclonal T cells derived from different patients share similar specificities of their T-cell antigen receptor, Vp gene usage was determined by the polymerase chain reaction (PCR) by using a panel of TCR Vp primers specific for published Vp families. Identification of a common TCR Vp gene usage would suggest that the clonal expansion of these cells is due to specific stimulation through the T-cell
34 Annals of Neurology Vol 29 No 1 January 1991
=
normal pressure
antigen receptor and that these cells may be important in the iniciation or progression of the inflammatory response leading to demyelination in the CNS. Materials and Methods Subjects Patients with multiple sclerosis had clinically definite disease by using standard criteria [ l l )with confirmation on magnetic resonance imaging (MRI) scanning. The characteristics of the patients are presented in Table 1. Patients with other neurological disease are also shown in Table 1 and include patients with myasthenia gravis, subacute sclerosing panencephalitis (SSPE) 123, Lyme meningoencephalitis (with pleocytosis of 45 lymph~cyteslmm~ in CSF); Guillain-Barri. syndrome, acute thromboembolic vascular disease of the brain (CVA), normal pressure hydrocephalus, and varicella-zoster meningoencephalitis. Patients Gr, Ka, and Mu with multiple sclerosis and control subjects Wi, Ve, and Pw were previously reported [8].Chronic progressive multiple sclerosis was defined as a decline in the Kurtzke disability status scale in the 9 months before venous and lumbar punctures [12]. The 2 subjects with relapsing-remitting W R ) multiple sclerosis
had acute clinical relapses within 2 weeks of venous and lumbar punctures. Patient Au had a 4-month progressive course of gait difficulty and behavioral changes with multiple white matter lesions on MRI. CSF examination in this patient showed oligoclonal banding with normal immunoglobulin index, 19 lymph~cytesimm~, and normal protein and glucose. On brain biopsy, there was demyelination with perivascular infiltrates of lymphocytes and monocytes. Peripheral blood and CSF was obtained from all patients after informed consent.
Cell Preparation Peripheral blood mononuclear cells (PBMC) were isolated from heparinized venous blood by means of a Ficoll-Hypaque density gradient (Pharmacia Fine Chemicals, Piscataway, NJ), washed twice with Hanks balanced salt solution (GIBCO, Grand Island, NY), counted, and resuspended in standard media consisting of lo%, pooled human serum (GIBCO, Pel Freeze, Arkansas) in RPMI, 2% glutamine (GIBCO), and 1% penidhistreptomycin (GIBCO). CSF was obtained by lumbar puncture within 2 hours of blood drawing, and cells were prepared as previously described [S}.
T-cell Cloning PBMC were directly cloned at less than one cell per well with lo5 autologous irradiated (5,000 rads) PBMC and Phytohemagglutin P (PHA.P) (1.0 Fglml) (Wellcome, Dartford, UK) in 96 well V-bottom plates. Forty-eight hours later, 0.1 ml of media containing either 10% interleukin (1L)-2 (delectinated T-cell growth factor that is column purified, from ABI, Columbia, MD) and, in some patients, an additional 50 Uirnl of recombinant interleukin 4 (rIL-4) (Genzyme, Cambridge, MA) were added to each well. rIL-4 was used in more recent clones to increase growth rate although cloning efficiency was not changed f12). Cultures were fed with IL-2 and rIL-4 every 3 to 5 days until approximately day 12, when all the wells are passed to 96 well U-bottom plates. Growthpositive wells were scored macroscopically by an inverted microscope and transferred into V-bottom plates with 5,000 cloned T cells per well with lo5 allogeneic irradiated (5,000 rads) mononuclear cells (from leukopacs) and PHA.P (1.0 pg/ml) with IL-2 plus rIL-4. When there were approximately 20 to 100 x lo3 T cells per V-bottom well (usually 3-5 days), clones were transferred to flasks at a cell concentration of 0.5 x lo6 cellslml. T-cell clones were restimulated every 10 to 14 days, as previously described, in V-bottom plates with 5,000 cloned T cells per well, 10' allogeneic irradiated mononuclear cells, PHA.P, IL-2, and rIL-4.
Southern Blot Analysis Genomic DNA extractions were performed according to standard methodology El3). DNA was digested with the restriction enzyme EcoRI or HindIII, size-fractioned by electrophoresis through an 0.8% agarose gel, and transferred onto nitrocellulose by the method of Southern blot analysis. Filters were hybridized to nick-translated 32P-labeledprobes of the TCR p chain gene and y chain gene and were washed at 60°C in 0.1% sodium dodecylsulfate, 0.015 M sodium chloride, and 0.0015 M sodium citrate before autoradiography. DNA probes were restriction enzyme DNA frag-
ments of previously cloned germline T-cell antigen receptor p chain genes. J p t is a 2.6-kilobase (kb) HindII-Nd fragment containing the J p l gene segment cluster; Jp2 are contiguous 1.9-kb PvtlII-PtwII and 1.5-kb PvtlII-EcoRV fragments containing the 3' half of the Jp2 gene segment cluster. The two probes do not cross-hybridize. Jv is a 0.8-kb HindIIT-EroRI fragment containing Jvl.It hybridizes to both the Jvl and Jy2 gene segments CS}.
PCR Analysis PCR amplification of genomic D N A (Patients [with multiple sclerosis), Po and Ka) and cDNA (Patient {with multiple sclerosis], Re) was performed as previously described 1141. Briefly, RNA was extracted from T-cell clones by extraction with guanidium-isothiocyanateiphenolchloroform and isopropanol precipitation. Single-stranded complementary DNAs (cDNAs) were synthesized from 1 pg RNA by using oligo-deoxythymidine (Sigma Chemical, St Louis, MO) and Avian Myeloblastosis virus-reverse transcriptasc (Bethesda Research Laboratories, Inc, Gaithersburg, MD). PCR amplification was done with a panel of 15 oligonucleotides corresponding to the CDR2 region of the TCR p chain (Vpl-16) and a Cp primer (cDNA). For amplification of genomic DNA, Vp primers were used in combination with both a J p l and a Jp2 primer, which were located at the 3' end of the J p l and Jp2 gene cluster, respectively. Amplifications were done for thirty cycles (94°C for 1 minute, 55°C for 2 minutes, 72°C for 3 minutes) with 1 pg of each primer in 50-1-1.1reactions. Amplified products were separated in 1% agarose gels, transferred to nitrocellulose, and hybridized with an internal oligonucleotide probe. Probes were end-labeled with Y-'~Padenosine triphosphate and T4 polynucleotide kiriase (BRL) to a specific activity of lo8 cpmipg and hybridized. Blots were washed at a final stringency of 6 x SSCi70"C and autoradiographed for 2 to 18 hours. Sequences for J primers are as follows:J p l primer, 5' CCC CCG AGT CAA GAG TGG AGC CCC CAT ACC 3'; Jp2 primer, 5' CCG AGG GGC TGG AAG GTG GGG AGA CGC CCG 3'; J p l probe, 5' CCT GGT CCC ATT CCC AAA GTG GAG GGG TGA 3'; and Jp2 probe, 5' TGA CCG TGA GCC TGG TGC CCG GCC CGA AGT 3'.
Phenotyping Cytofluographic analysis of T cells was performed by means of direct immunofluorescence with fluorescein-conjugated anti-CD3 monoclonal antibody (mAb), anti-CD4 mAb, and phycoerythrin-conjugated anti-CD8 mAb at a dilution of 1 : 20 (kindly supplied by Coulter Immunology, Hialeah, FL) as previously reported {IS]. Flow cytometric analysis was performed by using an Epics C flow cytometer (Coulter Electronics, Hialeah, FL).
Results T cells were directly cloned before other in vitro manipulation and stimulated with mitogen and growth factors. By using this technique, a very high cloning efficiency could be obtained with T-cell clones that were representative of t h e original populations of cells in the CSF or blood [S, 161. In these experiments, the
Lee et al: T-cell Receptors in Multiple Sclerosis 35
cloning efficiency of the plated T cells was always greater than 78%. A total of 1,017 independent longterm (> 2 months) T-cell cultures were derived from the different patient groups. The pattern of TCR p and y chain gene rearrangements for each of the cultures was determined as previously described by Southern blot analysis with D N A probes specific for the Jpl, Jp2, and J, gene segment clusters {8]. Nine hundred ten T-cell clones had 2 or fewer different rearranged fragments hybridizing to either the Jpl or Jp2 probes, and these were used for analysis of common T-cell clonality. The other cultures were assumed to have had more than one T cell originally plated and thus were not examined. The pattern of p chain and y chain gene rearrangements of the 910 T-cell clones were determined by Southern blotting from a total of 9 patients with demyelinating disease, 8 patients with other neurological diseases, and 4 normal subjects. Three hundred eighteen of these clones were presented in an initial report IS] and are combined here for analysis. The genomic DNA isolated from a number of T-cell clones among some patients with demyelinating disease had identical restriction fragment patterns after digestion with the enzymes EcoRI or Hind111 followed by hybridization in separate experiments with the TCR gene probes J p l , Jp2, or J,. Representative Southern blots after an initial screening of genomic DNA digested with EcoRI and probed with a combination of Jpl and Jp2 probes is shown in Figures 1A and 2A. Each lane represents a different T-cell clone; experimentally, the restriction fragment pattern of each clone was compared with the pattern produced by the other clones. As is seen, the majority of clones have different restriction fragment patterns. In contrast, some clones, such as Po.B, Po.Q, and Po.1 in Figure lA, and Re.2 and R e . 0 in Figure 2A had similar restriction fragment patterns and were selected for further analysis with other enzyme and TCR probe combinations. For two clones to be considered identical, they had to have the same patterns of non-germ line rearrangements after digestion with both EcoRI and HzndIII restriction enzymes and probing with J p l , Jp2, and J, gene probes (Figs lB,C,2B,C,D). The data from 15 patients in the present study and 6 subjects from the initial report are presented in Table 1. Specifically, subject Pw had 3 clones with identical restriction fragment patterns of which 2 were from the blood and 1 was from the CSF. Patient Re had 2 pairs of clones with identical restriction fragment patterns; each pair comprised of 1 clone from the CSF and the other from blood. Patient Au with childhood demyelinating disease had 1 pair of identical clones in the CSF. Patients Ka and G r have been previously reported; there was 1 clone in the blood of Patient G r with identical restriction fragment patterns in the CSF. None of the T-cell clones from 36 Annals of Neurology
Vol 29
No l January 1991
Fig 1 , (A)A n EcoRl digest of genomic D N A from clones of Patient Po, who has multiple slcerosis, probed with both J p l and J$ probes (Repeat individual Southern blots with these and J., probes are not shown). D N A from these clones was also digested with HindIII and probed with J p l , Jp2, and J y probes; shown are Southern blots (B and Cj aftev a HindIII digest of genomic D N A with Jp2 andJ, probes. Clones P0.B. Po.Q, and Po.1 have the same restriction fragment rearrangement patterns.
patients with other neurological diseases and none of the normal controls had common restriction fragment patterns. Oligoclonal T-cell populations derived from 3 patients with chronic progressive multiple sclerosis (Po, Ka, Re) were further analyzed for TCR Vp gene usage. PCR amplification was performed by use of a panel of TCR Vp primers, which were specific for published Vp families (Table 2). This analysis confirmed that T-cell clones, which shared the same TCR rearrangement pattern, were indeed clonally related by demonstrating usage of the same Vp gene segment. When TCR Vp gene usage was compared among 4 oligoclonal T-cell populations derived from 3 patients with chronic progressive multiple sclerosis, common usage of the Vp12 gene segment was found among all T-cell clones (Fig 3). Also, T-cell clones derived from Patients Po and Ka could be examined by PCR amplification of genomic DNA for assessment of rearrangement patterns. Both oligoclonal T-cell populations from these 2 pa-
Fig 2. (A) An EcoRl digest of genomic D N A from clones of patient Re, who has multiple sclemis, p d e d with both,JB1and probes (repeat indizidual Southern blots with these and J , probes are not shown). D N A from these clones was also digested with HindIII and probed with J p l and Jp2, and J , probes: shown are Southern blots after a Hind111 digest of genomic D N A with J$ (B),J$ ICj, and.Jy ID) probes. Clones Re.2 and Re.0 have the same restriction fragment reawangemeizt patterns.
tients shared a similar rearrangement to the Jp2 gene cluster. T-cell clones were phenotyped for the expression of CD4 and CD8 cell surface molecules. Eighty percent of the blood clones expressed high densities of CD4+ with either low density or no expression of CD8 determinants, whereas 209% were high density CD8+ and CD4 [12). In the CSF, 9 195 of the clones expressed high densities of CD4+ with either low density or no expression of CD8 determinants, whereas 99h were high density CD8+ and CD4K. All of the oligoclonal T cells expressed the CD4 determinant. None of the oligoclonal T cells proliferated in standard proliferation assays to myelin basic protein, tetanus toxoid, measles virus, or mumps virus.
Table 2. T-cell Receptor Vp Gene Usage among Oligoclonal T Cells from Patients with Multiple Sclerosis Oligoclonal Pair
Vp Gene Usage
Po.1 (CSF) Po.B (blood) Ka.1 (CSF) Ka.3 (CSF) Re.18 (CSF) Re.Q (blood) Re.2 (CSF) R e . 0 (blood)
Vp 12-Jp2 Vpl 2-Jp2 Vp12-Jp2 Vp12-Jp2 vp12
vp12 ND vp12
T-cell receptor V, gene usage was determined by polymerase chain reaction of genomic DNA (Patients Po and Ka) or complementary DNA (cDNA) (Patient Re) by use of a panel of T-cell receptor V, primers in combination with a C , primer (cDNA) or a Jgl and Jp2 primer (Renomic DNA). Amplified DNA was separated on agarose gels, and Southern blots were hybridized with internal probes. N D = nor done.
+
Discussion T cells were directly cloned from the CSF and blood of patients with multiple sclerosis and other neurological diseases before other in vitro manipulation and were compared by Southern blot analysis of the rearrangement patterns of their TCR f3 chain and y chain genes. This allowed the determination of whether two
T-cell clones shared the same TCR and thus arose from identical, clonally expanded progenitor T cells. As an extension of previous studies, oligoclonal T-cell clones representing restricted, clonally expanded Tcell populations were identified among both blood and CSE; populations in 5 of 9 patients with demyelinating disease. Moreover, identical clones between blood and CSF were observed in 3 of those 9 patients. In contrast, oligoclonal T cells were not observed in 8 patients with other neurological diseases or in the blood of 4 normal control patients.
Lee et al: T-cell Receptors in Multiple Sclerosis
37
Fig 3.Southern blot anahisis of7'-cell receptor i'TCR) Vp gene usage for oligoclonal T celh derivedfrnm 3 patients with multiple sclemszs. Pohimerase chain veclction of genomzc D N A (Patients Po and Ka) and of complementaary DNAs (cDNAs)(Patient Re) zoas done by using a panel of TCR V, primers (Vpl-l 6)in combination uith,Jplandj$ primen (geriomic D N A ) or a Cpprimer (cDNA).Oligoclonal T' ceflr from Patients Po and Ka were all found t o have a Ve12d$ rear-
rangement, whereas no amplification products were obtained using VB-Jpl primers (data not shown). Southern blots were bybridzzed by using internal TCK probes and autoradiographed. Data from these Soathern blot.( are summarized in Table 2. bji
Only a minority of T cells were clonally expanded in either blood or CSF of some but not all patients with multiple sclerosis. Nevertheless, this appears to represent an extraordinary degree of clonal expansion. That is, it is estimated that there are over lo6 different TCR gene rearrangements, and of these, we estimate that this methodology allows us to determine if 2 T cells are identical with a very high level of confidence (p > 0.001) [S}. This is based on the resolution of Southern blotting, with the use of different probe and restriction enzyme combinations, each of which gyves an independent assessment of clonality. Thus, considering there are likely to be at least lo4 different T-cell clones, this methodology could detect IS], finding two identical Tcell clones among 20 to 60 randomly selected clones represents a significant expansion of the circulating T-
38 Annals of Neurology Vol 29 No 1 January 1991
cell population. Moreover, direct single cell cloning will only detect high degrees of clonal expansion. Patients and controls negative for oligoclonal T cells may have more subtle clonal expansion that could only be detected with the investigation of an order of magnitude of more T-cell clones. Nevertheless, the present investigation extends our initial work, which was the first to demonstrate that clonally expanded T cells can exist in an immune compartment in an autoimmune disease. Recent reports indicate that oligoclonal T cells may be found in other immune compartments, such as the synovial fluid in rheumatoid arthritis [9] and the liver in primary biliary cirrhosis {lo] during the course of an inflammatory autoimmune disease process. As T cells regulate immunoglobulin synthesis, it is also possible that these oligoclonal T-cell populations are directly responsible for the oligoclonal bands observed in multiple sclerosis, and this is presently under investigation. We have previously demonstrated oligoclonal T-cell populations in the CSF of 2 patients with chronic progressive multiple sclerosis (Patients Gr and &), whereas a patient with acute fatal multiple sclerosis (Patient Mu) did not show oligoclonal T-cell populations [8). In the present study, an additional 3 patients with well-defined chronic progressive disease were studied and 2 of the patients had oligoclonal T cells.
These patients did not have the same degree of clonally expanded CSF T cells as observed in Patient Gr. As all the clones were established by using essentially the same technique, it seems unlikely that this difference is due to a methodological problem. The patient groups we studied were also similar. The possibility that cross-contamination of T-cell clones occurred at an early stage of the cloning procedure in the first patient we studied (Gr), however, cannot be totally excluded. It was important to determine if clonal T-cell expansion is an early or late event in the disease. Two patients with relapsing remitting multiple sclerosis and l patient (Au) with a monophasic demyelinating episode were studied. Patient Au, with biopsy-confirmed early inflammatory demyelinating disease, had oligoclonal T cells in the CSF. This patient is potentially instructive in suggesting that clonal expansion of T cells may occur early in the course of a demyelinating disease episode. The relation between this potentially single episode of demyelinating disease (acute penvenous encephalomyelitis) and adult multiple sclerosis, however, is as yet unknown. Analysis of TCR V, gene usage among oligoclonal T-cell populations from 3 patients with multiple sclerosis demonstrated a common usage of the V,12 gene segment among 4 sets of oligoclonal T cells. These data raise the possibility that oligoclonal T cells may share similar antigenic specificities and that the clonal expansion of these cells may be the result of specific stimulation with a myelin antigen, although an antigen reactivity could not be found for these clones. This is particularly interesting because of recent observations indicating that myelin basic protein reactive T-cell clones that induce an experimental model of multiple sclerosis, experimental allergic encephalomyelitis, use restricted TCR Vp and V, genes in both mice and rats {17-191. Furthermore, we have recently deSCrlbed a shared usage of TCR Vp genes among myelin basic protein reactive T cells in patients with multiple sclerosis, and these T cells frequently use VP17 or V,12 1141. The TCR VP12 gene segment identified among both MBP (84-102) reactive T-cell lines and oligoclonal T cells is homologous to the Vp8.2 gene segment described among the majority of encephalitogenic, MBP-reactive T cells in both mice and rats. Examination of a large series of T-cell clones in the present study demonstrates common T-cell clonotypes between the blood and CSF of patients with chronic progressive multiple sclerosis, suggesting the presence of an equilibrium between T cells in the blood and CSF immune compartments. This is consistent with previous studies using monoclonal antibodies to label circulating T cells in vivo, which indicated there is rapid traffic of T cells from the blood to CSF in patients with multiple sclerosis {20]. Other investiga-
tors have suggested there can be rapid movement of immune B cells into the CNS comparcment after systemic immunization. Specifically, Sandberg-Wollheim and co-workers [21} demonstrated that CSF B cells stimulated with T cells/monocytes plus pokeweed mitogen synthesized anti-tetanus toxoid antibodies several weeks after a booster injection, but CSF cells stimulated before the booster did not. In total, it appears that unlike immunoglobulin, which is locally synthesized in the CNS and generally does not cross the blood-brain barrier, close relations exist among certain clonally expanded T-cell populations and B-cell populations between blood and CSF. In summary, clonal T-cell expansion was observed among T cells cloned from blood ,and CSF of patients with multiple sclerosis. Common T-cell clones found between blood and CSF provides further evidence for a close relation between T-cell populations in blood and CSF immune compartments in patients with multiple sclerosis. Common TCR V, gene usage among oligoclonal T cells between different patiems suggests that these cells may share similar specificities and that clonally expanded T cells bearing specific TCRs play an important role in the inflammatory response leading to CNS demyelination. Supported by National Institutes of Health (NIH) Grants NS17182 and NS-00981, granrs from the National Multiple Sclerosis Society, and the Albert J. and Diane E. Kaneb Charitable Lead Trust. S.J.L. and S.A.B. are the recipienrs of N I H National Research Service Awards. K.W.W. is the Susan Furbacher Conroy Fellow in Multiple Sclerosis of the Center for Neurologic Disease, Brigham & Women’s Hospital, Boston, MA. We acknowledge the helpful comments of Dr J.G. Seidman and the technical assistance of Ms Meghan Purvee. Dr Cynthia J. Rutherford and t h e Blood Bank of the Brigham and Women’s Hospital are gratefully appreciated for rheir assistance in supplying blood products.
References 1. McFarlin DE, McFarland H . Multiple sclerosis. N Engl J Med 1983;307:1181-1188 2. H d e r DA, Weiner HL. MS: a CNS and systemic autoimmune disease. lmmunol Today 1989;lO:104-107 3. Hofman FM, von Hanwehr RI, Dinarello CA, et al. Immunoregulatory molecules and IL 2 receptors identified in multiple sclerosis brain. J Immunol 1986;136:3239-3245 4. Ota K, Matsui M, Milford E, et al. T cell recognition of an immunodominant myelin basic protein epitope in multiple sclerosis. Nature 1990;346:183-187 i. Martin R, Jarayuemada D, Flerlage M, et al. Fine specificity and HLA restriction of myelin basic protein-specific cytotoxic T cell lines from multiple sclerosis patients and healrhy individuals. J Immunol 1990;145:540-548 6. Toyonaga B, Ytoshikai Y, Vdasz V, et al. Organization and sequences of the diversity, joining, and constant region genes of the human T cell receptor beta chain. Proc Narl Acad Sci USA 1985;82:8624-8628 7 Waldman TA, Davis MM, Rongivanni KF, Korsmeyer SL. Rearrangements of genes for the antigen receptor on T cells as
Lee et al: T-cell Keceptors in Multiple Sclerosis 39
8.
9. 10.
11.
12. 13.
14.
15.
markers of lineage and clonality in human lymphoid neoplasms. N Engl J Med 1985;313:776-783 Hafler DA, Duby AD, Lee SJ, et al. Oligoclonal T lymphocytes in the cerebrospinal fluid of patients with multiple sclerosis. J Exp Med 1988;167 :I 3 13-1 322 Sramenkovic I, Stegagno M, Wright KA, et al. Clonal dominance among T-lymphocyte infiltrates in arthritis. Proc Natl Acad Sci USA 1988;85:1179-1183 Mosbius U, Meuer SC, Hess G, er al. T ccll receptor gene rearrangements of T cells infiltrating the liver in chronic active hepatitis and primary biliary cirrhosis. Proc 7th lnt Congress Imunol 1989;7:567 (Abstract) Kurtzke JK. Course of exacerbations of multiple sclerosis in hospitdzed patients. Arch Neurol Psychiatry 1956;76:175184 Brad SA, Purvee M, Hafler DA. IL-4 induces CD8 expression on human T cell clones. Cell Immunol 1990;125:426-436 Southern EM. Detection of specific sequences among DNA fragments separated by gel electrophoresis. J Mol Biol 1975; 981503-511 Wucherpfennig KW, Ota K, Endo N, et al. Shared human T cell receptor V, usage to immunodominant regions of myelin basic protein. Science 1990;248:1016-1019 Hafler DA, Fox D, Schlossman SF, et al. In vivo activated Tlymphocytes in the peripheral blood and cerebrospinal fluid of
40 Annals of Neurology Vol 29 No 1 January 1991
patients with multiple sclerosis. N Engl J Med 1985;312:14051411 16. Moretta A, Pantaleo G, Moretta LJ, et al. Direct demonstration of the clonogenic potential of every human peripheral blood cell. J Exp Med 1983;157:743-753 17. Urban JL,Kumar V, Kono DH, er al. Restricted use of T cell receptor V genes in murine autoimmune encephalomyeliris raises possibilities for antibody therapy. Cell 1988;54:577-592 18. Acha-Orbea H , Mitchell DJ, Timmermann L, et al. Limited heterogeneity of T cell receptors from lymphocytes mediating autoimmune encephalomyelitis allows specific immune intervention. Cell 1988;54:263-273 19. Burns FR, Li X, Shen N , et al. Both rat and mouse T cell receptors specific for the encephalitogenic determinant of myelin basic protein use similar V, and V, chain genes even though the major histocompatibility complex and encephalitogenic determinants being recognized are different. J Exp Med 1989;169:27-39 20. Hafler DA, Weiner HL. In vivo labeling of blood T cells: rapid traffic into cerebrospinal fluid in multiple sclerosis. Ann Neurol 1987;21:89-93 21. Sandberg-Wollheim M, Zweiman B, Levinson AI, Lisak RP. Humoral immune responses withn the human central nervous system following systemic immunization. J Neuroimmunol 1986;11:205-214
T-cell populations were investigated in the blood and cerebrospinal fluid of patients with multiple sclerosis and other neurological diseases. Individual T cells were directly cloned from the cerebrospinal fluid and blood before in viuo expansion, and their clonotypes were compared by Southern blot analysis of the rearrangement patterns of their T-cell receptor p chain and y chain genes. This allowed the determination of whether two T cell clones shared the same Tcell receptor and thus arose from identical, clonally expanded (oligoclonal) progenitor T cells. As an extension of previous studies, oligoclonal T-cell clones were identified in both cerebrospinal fluid and blood populations in 5 of 9 patients with inflammatory demyelinating disease among a total of 486 blood and cerebrospinal fluid T-cell clones. In contrast, no clonally expanded T-cell populations were found among a total of 424 clones derived from either blood of 4 normal control subjects or blood and cerebrospinal fluid of 8 patients with other neurological diseases. Analysis of Tcell receptor V, genes among 4 oligoclonal T-cell populations derived from 3 patients with multiple sclerosis demonstrated common usage of the Vp12 gene segment. These data suggest that oligoclonal T cells share similar specificities and that clonal expansion may have resulted from specific stimulation by an antigen. Moreover, identical clones between blood and cerebrospinal fluid were observed in 3 of 9 patients with demyelinating disease, thus providing further evidence of an equilibrium between peripheral and central nervous system immune compartments. Lee SJ, Wucherpfennig KW, Brod SA, Benjamin D, Weiner HL, H d e r DA. Common T-cell receptor V, usage in oligoclonal T lymphocytes derived from cerebrospinal fluid and blood of patients with multiple sclerosis. Ann Neurol 1991;29:33-40
Multiple sclerosis is a chronic, inflammatory, demyelinating disease of the central nervous system (CNS) { 1, 23. Pathological studies have indicated that the majority of infiltrating cells at sites of active demyelination are activated T cells and macrophages { 3 ] . A major hypothesis regarding the pathogenesis of multiple sclerosis is that T cells reactive with myelin protein determinants migrate into the CNS and, either directly or by recruitment of effector cells, mediate damage to white matter tracts [4, 51. Thus, characterization of immune T cells in the CNS and blood of patients with multiple sclerosis is likely to be important in attempting to understand the pathogenesis of the disease. It is now known that each T cell expresses a single receptor that consists of either an a-p or y-6 heterodimer that recognizes antigen in association with major histocompatibility gene complex proteins. The highly diverse T-cell receptor (TCR) repertoire is generated
during T-cell ontogeny in the thymus by rearrangement of TCR genes, which consist of variable (V), diversity (D), junctional and constant (C) gene regions {b}.Whether two T cells are the same and thus derived from a common progenitor cell can be determined by analysis of these TCR gene rearrangements. That is, Southern blot analysis of DNA from T cell clones digested with different restriction enzymes and probed with different TCR genes can identify two clones that have a high probability of sharing the same TCR gene rearrangements and thus can be assumed to arise from the identical progenitor T cell [7]. If T-cell populations in the blood and cerebrospinal fluid (CSF) of patients with multiple sclerosis are derived from a large number of progenitor cells, examination of TCR genes of many T-cell clones would reveal different gene rearrangements. Alternatively, there may be more restricted expansion of T cells in
From the Center for Neurologic Disease, Division of Neurology, Department of Medicine, Brigham and Women’s Hospital, and Harvard Medical School, Boston, MA.
Address correspondence to Dr HaAer, Center for Neurologic Disease, Brigham and Women’s Hospital, 75 Francis Street, Boston, MA 02115.
Received Oct 19, 1989, and in revised form Apr Accepted for publicatiun Jd 9, 1990.
u),
9 and Jul9, 1990.
Copyright 0 1991 by the American Neurological Association
33
Table 1, Oligoclonal T Lymphocytes fvom Patients with Multiple ScEerosis T-cell Clonalitya Patient
Disease
Multiple sclerosis Gr ... Ka ... Mu ... Po ... Ca ... Re ... Da ... Sh ... Auf ... Other neurological diseases Ki Myasthenia sz SSPE Av Lyme Fe Stroke McK NPH Ga GBS Wi SSPE Ve HZV Normal controls Pw ... He ... Ka ... Ha ...
Age (yr)
Sex
34
F F M F M
39 38 51 22 55
26 40 5 30 20
60 65 65 52 16
63 42 28 30 36
M E
F F
F F M F F F F M
Duration of Disease
Oligoclonal Bands
10 yr 8 Yr 8 mo 8 yr 5 Yr 10 yr 6 Yr 15 y t 4 mo
3 Yr 8 Yr 3 mo 1 mo
6 mo 4 mo 3 Yr 9 mo
Blood Blood'
CSFb
& CSP
Totald
Yes No 1 Band No No No Yes No Yes
2126 016 014 1 2/10 0133
18128
2 0 0
20154' 5/33'
ND Yes NO No
0152 0119 0135 Oil 1 0114
NO
0116
0119 019 017
5127 0137 0130 0156 0140
0136 0118 2127 ND
019 018
012 012 5
No Yes No
ND
0126
0114
0133 0120
ND
F
...
...
I; M
...
...
0166" 0128
...
...
012 8
M
...
...
0113
0178'
3
3160
0 4 0 0 0
0189
4/56 0155 0127 2/34
ND 0 0 0 0
0152 0128 014 3
ND 0
0113 0139 0126 0168'
ND
0120'
"Total number of oligoclonal T cells: barno% blood or CSF, 'berween blood and CSF, 'total among blood and CSF. See Results section for description of clone patterns. cPrcviously described clones [S}. fPatietit with single episode of perivascular lymphocytic infiltrates and demyelination. Myasthenia = myasthenia gravis; SSPE subacute sclerosing panencephalitis; Lyme = Lyme meningoencephalitis; h'PH hydrocephalus; GBS = Guillain Bar& syndrome; ND = not done; HZV = herpes zoster virus meningoencephalitis.
the immune system of patients with multiple sclerosis resulting in either oligoclonal or possibly monoclonal T-cell populations. Southern blot analysis of a large number of T-cell clones derived from patients with multiple sclerosis would then show common TCR gene rearrangements. We investigated this in a previous study and found oligoclonal T-cell populations in the CSF and blood of some but not all patients with multiple sclerosis ES}. The observation that the T-cell repertoire may be altered in autoimmune disease with expansion of particular T-cell populations has subsequently been reported in other autoimmune diseases r9, 101. To determine whether oligoclonal T cells derived from different patients share similar specificities of their T-cell antigen receptor, Vp gene usage was determined by the polymerase chain reaction (PCR) by using a panel of TCR Vp primers specific for published Vp families. Identification of a common TCR Vp gene usage would suggest that the clonal expansion of these cells is due to specific stimulation through the T-cell
34 Annals of Neurology Vol 29 No 1 January 1991
=
normal pressure
antigen receptor and that these cells may be important in the iniciation or progression of the inflammatory response leading to demyelination in the CNS. Materials and Methods Subjects Patients with multiple sclerosis had clinically definite disease by using standard criteria [ l l )with confirmation on magnetic resonance imaging (MRI) scanning. The characteristics of the patients are presented in Table 1. Patients with other neurological disease are also shown in Table 1 and include patients with myasthenia gravis, subacute sclerosing panencephalitis (SSPE) 123, Lyme meningoencephalitis (with pleocytosis of 45 lymph~cyteslmm~ in CSF); Guillain-Barri. syndrome, acute thromboembolic vascular disease of the brain (CVA), normal pressure hydrocephalus, and varicella-zoster meningoencephalitis. Patients Gr, Ka, and Mu with multiple sclerosis and control subjects Wi, Ve, and Pw were previously reported [8].Chronic progressive multiple sclerosis was defined as a decline in the Kurtzke disability status scale in the 9 months before venous and lumbar punctures [12]. The 2 subjects with relapsing-remitting W R ) multiple sclerosis
had acute clinical relapses within 2 weeks of venous and lumbar punctures. Patient Au had a 4-month progressive course of gait difficulty and behavioral changes with multiple white matter lesions on MRI. CSF examination in this patient showed oligoclonal banding with normal immunoglobulin index, 19 lymph~cytesimm~, and normal protein and glucose. On brain biopsy, there was demyelination with perivascular infiltrates of lymphocytes and monocytes. Peripheral blood and CSF was obtained from all patients after informed consent.
Cell Preparation Peripheral blood mononuclear cells (PBMC) were isolated from heparinized venous blood by means of a Ficoll-Hypaque density gradient (Pharmacia Fine Chemicals, Piscataway, NJ), washed twice with Hanks balanced salt solution (GIBCO, Grand Island, NY), counted, and resuspended in standard media consisting of lo%, pooled human serum (GIBCO, Pel Freeze, Arkansas) in RPMI, 2% glutamine (GIBCO), and 1% penidhistreptomycin (GIBCO). CSF was obtained by lumbar puncture within 2 hours of blood drawing, and cells were prepared as previously described [S}.
T-cell Cloning PBMC were directly cloned at less than one cell per well with lo5 autologous irradiated (5,000 rads) PBMC and Phytohemagglutin P (PHA.P) (1.0 Fglml) (Wellcome, Dartford, UK) in 96 well V-bottom plates. Forty-eight hours later, 0.1 ml of media containing either 10% interleukin (1L)-2 (delectinated T-cell growth factor that is column purified, from ABI, Columbia, MD) and, in some patients, an additional 50 Uirnl of recombinant interleukin 4 (rIL-4) (Genzyme, Cambridge, MA) were added to each well. rIL-4 was used in more recent clones to increase growth rate although cloning efficiency was not changed f12). Cultures were fed with IL-2 and rIL-4 every 3 to 5 days until approximately day 12, when all the wells are passed to 96 well U-bottom plates. Growthpositive wells were scored macroscopically by an inverted microscope and transferred into V-bottom plates with 5,000 cloned T cells per well with lo5 allogeneic irradiated (5,000 rads) mononuclear cells (from leukopacs) and PHA.P (1.0 pg/ml) with IL-2 plus rIL-4. When there were approximately 20 to 100 x lo3 T cells per V-bottom well (usually 3-5 days), clones were transferred to flasks at a cell concentration of 0.5 x lo6 cellslml. T-cell clones were restimulated every 10 to 14 days, as previously described, in V-bottom plates with 5,000 cloned T cells per well, 10' allogeneic irradiated mononuclear cells, PHA.P, IL-2, and rIL-4.
Southern Blot Analysis Genomic DNA extractions were performed according to standard methodology El3). DNA was digested with the restriction enzyme EcoRI or HindIII, size-fractioned by electrophoresis through an 0.8% agarose gel, and transferred onto nitrocellulose by the method of Southern blot analysis. Filters were hybridized to nick-translated 32P-labeledprobes of the TCR p chain gene and y chain gene and were washed at 60°C in 0.1% sodium dodecylsulfate, 0.015 M sodium chloride, and 0.0015 M sodium citrate before autoradiography. DNA probes were restriction enzyme DNA frag-
ments of previously cloned germline T-cell antigen receptor p chain genes. J p t is a 2.6-kilobase (kb) HindII-Nd fragment containing the J p l gene segment cluster; Jp2 are contiguous 1.9-kb PvtlII-PtwII and 1.5-kb PvtlII-EcoRV fragments containing the 3' half of the Jp2 gene segment cluster. The two probes do not cross-hybridize. Jv is a 0.8-kb HindIIT-EroRI fragment containing Jvl.It hybridizes to both the Jvl and Jy2 gene segments CS}.
PCR Analysis PCR amplification of genomic D N A (Patients [with multiple sclerosis), Po and Ka) and cDNA (Patient {with multiple sclerosis], Re) was performed as previously described 1141. Briefly, RNA was extracted from T-cell clones by extraction with guanidium-isothiocyanateiphenolchloroform and isopropanol precipitation. Single-stranded complementary DNAs (cDNAs) were synthesized from 1 pg RNA by using oligo-deoxythymidine (Sigma Chemical, St Louis, MO) and Avian Myeloblastosis virus-reverse transcriptasc (Bethesda Research Laboratories, Inc, Gaithersburg, MD). PCR amplification was done with a panel of 15 oligonucleotides corresponding to the CDR2 region of the TCR p chain (Vpl-16) and a Cp primer (cDNA). For amplification of genomic DNA, Vp primers were used in combination with both a J p l and a Jp2 primer, which were located at the 3' end of the J p l and Jp2 gene cluster, respectively. Amplifications were done for thirty cycles (94°C for 1 minute, 55°C for 2 minutes, 72°C for 3 minutes) with 1 pg of each primer in 50-1-1.1reactions. Amplified products were separated in 1% agarose gels, transferred to nitrocellulose, and hybridized with an internal oligonucleotide probe. Probes were end-labeled with Y-'~Padenosine triphosphate and T4 polynucleotide kiriase (BRL) to a specific activity of lo8 cpmipg and hybridized. Blots were washed at a final stringency of 6 x SSCi70"C and autoradiographed for 2 to 18 hours. Sequences for J primers are as follows:J p l primer, 5' CCC CCG AGT CAA GAG TGG AGC CCC CAT ACC 3'; Jp2 primer, 5' CCG AGG GGC TGG AAG GTG GGG AGA CGC CCG 3'; J p l probe, 5' CCT GGT CCC ATT CCC AAA GTG GAG GGG TGA 3'; and Jp2 probe, 5' TGA CCG TGA GCC TGG TGC CCG GCC CGA AGT 3'.
Phenotyping Cytofluographic analysis of T cells was performed by means of direct immunofluorescence with fluorescein-conjugated anti-CD3 monoclonal antibody (mAb), anti-CD4 mAb, and phycoerythrin-conjugated anti-CD8 mAb at a dilution of 1 : 20 (kindly supplied by Coulter Immunology, Hialeah, FL) as previously reported {IS]. Flow cytometric analysis was performed by using an Epics C flow cytometer (Coulter Electronics, Hialeah, FL).
Results T cells were directly cloned before other in vitro manipulation and stimulated with mitogen and growth factors. By using this technique, a very high cloning efficiency could be obtained with T-cell clones that were representative of t h e original populations of cells in the CSF or blood [S, 161. In these experiments, the
Lee et al: T-cell Receptors in Multiple Sclerosis 35
cloning efficiency of the plated T cells was always greater than 78%. A total of 1,017 independent longterm (> 2 months) T-cell cultures were derived from the different patient groups. The pattern of TCR p and y chain gene rearrangements for each of the cultures was determined as previously described by Southern blot analysis with D N A probes specific for the Jpl, Jp2, and J, gene segment clusters {8]. Nine hundred ten T-cell clones had 2 or fewer different rearranged fragments hybridizing to either the Jpl or Jp2 probes, and these were used for analysis of common T-cell clonality. The other cultures were assumed to have had more than one T cell originally plated and thus were not examined. The pattern of p chain and y chain gene rearrangements of the 910 T-cell clones were determined by Southern blotting from a total of 9 patients with demyelinating disease, 8 patients with other neurological diseases, and 4 normal subjects. Three hundred eighteen of these clones were presented in an initial report IS] and are combined here for analysis. The genomic DNA isolated from a number of T-cell clones among some patients with demyelinating disease had identical restriction fragment patterns after digestion with the enzymes EcoRI or Hind111 followed by hybridization in separate experiments with the TCR gene probes J p l , Jp2, or J,. Representative Southern blots after an initial screening of genomic DNA digested with EcoRI and probed with a combination of Jpl and Jp2 probes is shown in Figures 1A and 2A. Each lane represents a different T-cell clone; experimentally, the restriction fragment pattern of each clone was compared with the pattern produced by the other clones. As is seen, the majority of clones have different restriction fragment patterns. In contrast, some clones, such as Po.B, Po.Q, and Po.1 in Figure lA, and Re.2 and R e . 0 in Figure 2A had similar restriction fragment patterns and were selected for further analysis with other enzyme and TCR probe combinations. For two clones to be considered identical, they had to have the same patterns of non-germ line rearrangements after digestion with both EcoRI and HzndIII restriction enzymes and probing with J p l , Jp2, and J, gene probes (Figs lB,C,2B,C,D). The data from 15 patients in the present study and 6 subjects from the initial report are presented in Table 1. Specifically, subject Pw had 3 clones with identical restriction fragment patterns of which 2 were from the blood and 1 was from the CSF. Patient Re had 2 pairs of clones with identical restriction fragment patterns; each pair comprised of 1 clone from the CSF and the other from blood. Patient Au with childhood demyelinating disease had 1 pair of identical clones in the CSF. Patients Ka and G r have been previously reported; there was 1 clone in the blood of Patient G r with identical restriction fragment patterns in the CSF. None of the T-cell clones from 36 Annals of Neurology
Vol 29
No l January 1991
Fig 1 , (A)A n EcoRl digest of genomic D N A from clones of Patient Po, who has multiple slcerosis, probed with both J p l and J$ probes (Repeat individual Southern blots with these and J., probes are not shown). D N A from these clones was also digested with HindIII and probed with J p l , Jp2, and J y probes; shown are Southern blots (B and Cj aftev a HindIII digest of genomic D N A with Jp2 andJ, probes. Clones P0.B. Po.Q, and Po.1 have the same restriction fragment rearrangement patterns.
patients with other neurological diseases and none of the normal controls had common restriction fragment patterns. Oligoclonal T-cell populations derived from 3 patients with chronic progressive multiple sclerosis (Po, Ka, Re) were further analyzed for TCR Vp gene usage. PCR amplification was performed by use of a panel of TCR Vp primers, which were specific for published Vp families (Table 2). This analysis confirmed that T-cell clones, which shared the same TCR rearrangement pattern, were indeed clonally related by demonstrating usage of the same Vp gene segment. When TCR Vp gene usage was compared among 4 oligoclonal T-cell populations derived from 3 patients with chronic progressive multiple sclerosis, common usage of the Vp12 gene segment was found among all T-cell clones (Fig 3). Also, T-cell clones derived from Patients Po and Ka could be examined by PCR amplification of genomic DNA for assessment of rearrangement patterns. Both oligoclonal T-cell populations from these 2 pa-
Fig 2. (A) An EcoRl digest of genomic D N A from clones of patient Re, who has multiple sclemis, p d e d with both,JB1and probes (repeat indizidual Southern blots with these and J , probes are not shown). D N A from these clones was also digested with HindIII and probed with J p l and Jp2, and J , probes: shown are Southern blots after a Hind111 digest of genomic D N A with J$ (B),J$ ICj, and.Jy ID) probes. Clones Re.2 and Re.0 have the same restriction fragment reawangemeizt patterns.
tients shared a similar rearrangement to the Jp2 gene cluster. T-cell clones were phenotyped for the expression of CD4 and CD8 cell surface molecules. Eighty percent of the blood clones expressed high densities of CD4+ with either low density or no expression of CD8 determinants, whereas 209% were high density CD8+ and CD4 [12). In the CSF, 9 195 of the clones expressed high densities of CD4+ with either low density or no expression of CD8 determinants, whereas 99h were high density CD8+ and CD4K. All of the oligoclonal T cells expressed the CD4 determinant. None of the oligoclonal T cells proliferated in standard proliferation assays to myelin basic protein, tetanus toxoid, measles virus, or mumps virus.
Table 2. T-cell Receptor Vp Gene Usage among Oligoclonal T Cells from Patients with Multiple Sclerosis Oligoclonal Pair
Vp Gene Usage
Po.1 (CSF) Po.B (blood) Ka.1 (CSF) Ka.3 (CSF) Re.18 (CSF) Re.Q (blood) Re.2 (CSF) R e . 0 (blood)
Vp 12-Jp2 Vpl 2-Jp2 Vp12-Jp2 Vp12-Jp2 vp12
vp12 ND vp12
T-cell receptor V, gene usage was determined by polymerase chain reaction of genomic DNA (Patients Po and Ka) or complementary DNA (cDNA) (Patient Re) by use of a panel of T-cell receptor V, primers in combination with a C , primer (cDNA) or a Jgl and Jp2 primer (Renomic DNA). Amplified DNA was separated on agarose gels, and Southern blots were hybridized with internal probes. N D = nor done.
+
Discussion T cells were directly cloned from the CSF and blood of patients with multiple sclerosis and other neurological diseases before other in vitro manipulation and were compared by Southern blot analysis of the rearrangement patterns of their TCR f3 chain and y chain genes. This allowed the determination of whether two
T-cell clones shared the same TCR and thus arose from identical, clonally expanded progenitor T cells. As an extension of previous studies, oligoclonal T-cell clones representing restricted, clonally expanded Tcell populations were identified among both blood and CSE; populations in 5 of 9 patients with demyelinating disease. Moreover, identical clones between blood and CSF were observed in 3 of those 9 patients. In contrast, oligoclonal T cells were not observed in 8 patients with other neurological diseases or in the blood of 4 normal control patients.
Lee et al: T-cell Receptors in Multiple Sclerosis
37
Fig 3.Southern blot anahisis of7'-cell receptor i'TCR) Vp gene usage for oligoclonal T celh derivedfrnm 3 patients with multiple sclemszs. Pohimerase chain veclction of genomzc D N A (Patients Po and Ka) and of complementaary DNAs (cDNAs)(Patient Re) zoas done by using a panel of TCR V, primers (Vpl-l 6)in combination uith,Jplandj$ primen (geriomic D N A ) or a Cpprimer (cDNA).Oligoclonal T' ceflr from Patients Po and Ka were all found t o have a Ve12d$ rear-
rangement, whereas no amplification products were obtained using VB-Jpl primers (data not shown). Southern blots were bybridzzed by using internal TCK probes and autoradiographed. Data from these Soathern blot.( are summarized in Table 2. bji
Only a minority of T cells were clonally expanded in either blood or CSF of some but not all patients with multiple sclerosis. Nevertheless, this appears to represent an extraordinary degree of clonal expansion. That is, it is estimated that there are over lo6 different TCR gene rearrangements, and of these, we estimate that this methodology allows us to determine if 2 T cells are identical with a very high level of confidence (p > 0.001) [S}. This is based on the resolution of Southern blotting, with the use of different probe and restriction enzyme combinations, each of which gyves an independent assessment of clonality. Thus, considering there are likely to be at least lo4 different T-cell clones, this methodology could detect IS], finding two identical Tcell clones among 20 to 60 randomly selected clones represents a significant expansion of the circulating T-
38 Annals of Neurology Vol 29 No 1 January 1991
cell population. Moreover, direct single cell cloning will only detect high degrees of clonal expansion. Patients and controls negative for oligoclonal T cells may have more subtle clonal expansion that could only be detected with the investigation of an order of magnitude of more T-cell clones. Nevertheless, the present investigation extends our initial work, which was the first to demonstrate that clonally expanded T cells can exist in an immune compartment in an autoimmune disease. Recent reports indicate that oligoclonal T cells may be found in other immune compartments, such as the synovial fluid in rheumatoid arthritis [9] and the liver in primary biliary cirrhosis {lo] during the course of an inflammatory autoimmune disease process. As T cells regulate immunoglobulin synthesis, it is also possible that these oligoclonal T-cell populations are directly responsible for the oligoclonal bands observed in multiple sclerosis, and this is presently under investigation. We have previously demonstrated oligoclonal T-cell populations in the CSF of 2 patients with chronic progressive multiple sclerosis (Patients Gr and &), whereas a patient with acute fatal multiple sclerosis (Patient Mu) did not show oligoclonal T-cell populations [8). In the present study, an additional 3 patients with well-defined chronic progressive disease were studied and 2 of the patients had oligoclonal T cells.
These patients did not have the same degree of clonally expanded CSF T cells as observed in Patient Gr. As all the clones were established by using essentially the same technique, it seems unlikely that this difference is due to a methodological problem. The patient groups we studied were also similar. The possibility that cross-contamination of T-cell clones occurred at an early stage of the cloning procedure in the first patient we studied (Gr), however, cannot be totally excluded. It was important to determine if clonal T-cell expansion is an early or late event in the disease. Two patients with relapsing remitting multiple sclerosis and l patient (Au) with a monophasic demyelinating episode were studied. Patient Au, with biopsy-confirmed early inflammatory demyelinating disease, had oligoclonal T cells in the CSF. This patient is potentially instructive in suggesting that clonal expansion of T cells may occur early in the course of a demyelinating disease episode. The relation between this potentially single episode of demyelinating disease (acute penvenous encephalomyelitis) and adult multiple sclerosis, however, is as yet unknown. Analysis of TCR V, gene usage among oligoclonal T-cell populations from 3 patients with multiple sclerosis demonstrated a common usage of the V,12 gene segment among 4 sets of oligoclonal T cells. These data raise the possibility that oligoclonal T cells may share similar antigenic specificities and that the clonal expansion of these cells may be the result of specific stimulation with a myelin antigen, although an antigen reactivity could not be found for these clones. This is particularly interesting because of recent observations indicating that myelin basic protein reactive T-cell clones that induce an experimental model of multiple sclerosis, experimental allergic encephalomyelitis, use restricted TCR Vp and V, genes in both mice and rats {17-191. Furthermore, we have recently deSCrlbed a shared usage of TCR Vp genes among myelin basic protein reactive T cells in patients with multiple sclerosis, and these T cells frequently use VP17 or V,12 1141. The TCR VP12 gene segment identified among both MBP (84-102) reactive T-cell lines and oligoclonal T cells is homologous to the Vp8.2 gene segment described among the majority of encephalitogenic, MBP-reactive T cells in both mice and rats. Examination of a large series of T-cell clones in the present study demonstrates common T-cell clonotypes between the blood and CSF of patients with chronic progressive multiple sclerosis, suggesting the presence of an equilibrium between T cells in the blood and CSF immune compartments. This is consistent with previous studies using monoclonal antibodies to label circulating T cells in vivo, which indicated there is rapid traffic of T cells from the blood to CSF in patients with multiple sclerosis {20]. Other investiga-
tors have suggested there can be rapid movement of immune B cells into the CNS comparcment after systemic immunization. Specifically, Sandberg-Wollheim and co-workers [21} demonstrated that CSF B cells stimulated with T cells/monocytes plus pokeweed mitogen synthesized anti-tetanus toxoid antibodies several weeks after a booster injection, but CSF cells stimulated before the booster did not. In total, it appears that unlike immunoglobulin, which is locally synthesized in the CNS and generally does not cross the blood-brain barrier, close relations exist among certain clonally expanded T-cell populations and B-cell populations between blood and CSF. In summary, clonal T-cell expansion was observed among T cells cloned from blood ,and CSF of patients with multiple sclerosis. Common T-cell clones found between blood and CSF provides further evidence for a close relation between T-cell populations in blood and CSF immune compartments in patients with multiple sclerosis. Common TCR V, gene usage among oligoclonal T cells between different patiems suggests that these cells may share similar specificities and that clonally expanded T cells bearing specific TCRs play an important role in the inflammatory response leading to CNS demyelination. Supported by National Institutes of Health (NIH) Grants NS17182 and NS-00981, granrs from the National Multiple Sclerosis Society, and the Albert J. and Diane E. Kaneb Charitable Lead Trust. S.J.L. and S.A.B. are the recipienrs of N I H National Research Service Awards. K.W.W. is the Susan Furbacher Conroy Fellow in Multiple Sclerosis of the Center for Neurologic Disease, Brigham & Women’s Hospital, Boston, MA. We acknowledge the helpful comments of Dr J.G. Seidman and the technical assistance of Ms Meghan Purvee. Dr Cynthia J. Rutherford and t h e Blood Bank of the Brigham and Women’s Hospital are gratefully appreciated for rheir assistance in supplying blood products.
References 1. McFarlin DE, McFarland H . Multiple sclerosis. N Engl J Med 1983;307:1181-1188 2. H d e r DA, Weiner HL. MS: a CNS and systemic autoimmune disease. lmmunol Today 1989;lO:104-107 3. Hofman FM, von Hanwehr RI, Dinarello CA, et al. Immunoregulatory molecules and IL 2 receptors identified in multiple sclerosis brain. J Immunol 1986;136:3239-3245 4. Ota K, Matsui M, Milford E, et al. T cell recognition of an immunodominant myelin basic protein epitope in multiple sclerosis. Nature 1990;346:183-187 i. Martin R, Jarayuemada D, Flerlage M, et al. Fine specificity and HLA restriction of myelin basic protein-specific cytotoxic T cell lines from multiple sclerosis patients and healrhy individuals. J Immunol 1990;145:540-548 6. Toyonaga B, Ytoshikai Y, Vdasz V, et al. Organization and sequences of the diversity, joining, and constant region genes of the human T cell receptor beta chain. Proc Narl Acad Sci USA 1985;82:8624-8628 7 Waldman TA, Davis MM, Rongivanni KF, Korsmeyer SL. Rearrangements of genes for the antigen receptor on T cells as
Lee et al: T-cell Keceptors in Multiple Sclerosis 39
8.
9. 10.
11.
12. 13.
14.
15.
markers of lineage and clonality in human lymphoid neoplasms. N Engl J Med 1985;313:776-783 Hafler DA, Duby AD, Lee SJ, et al. Oligoclonal T lymphocytes in the cerebrospinal fluid of patients with multiple sclerosis. J Exp Med 1988;167 :I 3 13-1 322 Sramenkovic I, Stegagno M, Wright KA, et al. Clonal dominance among T-lymphocyte infiltrates in arthritis. Proc Natl Acad Sci USA 1988;85:1179-1183 Mosbius U, Meuer SC, Hess G, er al. T ccll receptor gene rearrangements of T cells infiltrating the liver in chronic active hepatitis and primary biliary cirrhosis. Proc 7th lnt Congress Imunol 1989;7:567 (Abstract) Kurtzke JK. Course of exacerbations of multiple sclerosis in hospitdzed patients. Arch Neurol Psychiatry 1956;76:175184 Brad SA, Purvee M, Hafler DA. IL-4 induces CD8 expression on human T cell clones. Cell Immunol 1990;125:426-436 Southern EM. Detection of specific sequences among DNA fragments separated by gel electrophoresis. J Mol Biol 1975; 981503-511 Wucherpfennig KW, Ota K, Endo N, et al. Shared human T cell receptor V, usage to immunodominant regions of myelin basic protein. Science 1990;248:1016-1019 Hafler DA, Fox D, Schlossman SF, et al. In vivo activated Tlymphocytes in the peripheral blood and cerebrospinal fluid of
40 Annals of Neurology Vol 29 No 1 January 1991
patients with multiple sclerosis. N Engl J Med 1985;312:14051411 16. Moretta A, Pantaleo G, Moretta LJ, et al. Direct demonstration of the clonogenic potential of every human peripheral blood cell. J Exp Med 1983;157:743-753 17. Urban JL,Kumar V, Kono DH, er al. Restricted use of T cell receptor V genes in murine autoimmune encephalomyeliris raises possibilities for antibody therapy. Cell 1988;54:577-592 18. Acha-Orbea H , Mitchell DJ, Timmermann L, et al. Limited heterogeneity of T cell receptors from lymphocytes mediating autoimmune encephalomyelitis allows specific immune intervention. Cell 1988;54:263-273 19. Burns FR, Li X, Shen N , et al. Both rat and mouse T cell receptors specific for the encephalitogenic determinant of myelin basic protein use similar V, and V, chain genes even though the major histocompatibility complex and encephalitogenic determinants being recognized are different. J Exp Med 1989;169:27-39 20. Hafler DA, Weiner HL. In vivo labeling of blood T cells: rapid traffic into cerebrospinal fluid in multiple sclerosis. Ann Neurol 1987;21:89-93 21. Sandberg-Wollheim M, Zweiman B, Levinson AI, Lisak RP. Humoral immune responses withn the human central nervous system following systemic immunization. J Neuroimmunol 1986;11:205-214
