Abnormal T cell receptor V gene usage in myasthenia gravis: prevalence and characterization of expanded T cell populations

Abstract

The usage of T cell receptor (TCR) Vα/Vβ chains on cells from 38 patients with myasthenia gravis (MG) was determined by flow cytometry. There was a decreased number of cells expressing Vβ2 in CD8⁺ and Vβ3 in CD4⁺ cells in patients compared with healthy individuals. Abnormal expansions of T cells using particular TCR Vα/Vβ gene products were found in 18/38 patients. A significantly higher usage of Vβ13 was observed but there was no restriction with regard to other TCR Vα/Vβ. Expanded cells belonging to both CD4⁺ and CD8⁺ were present in MG patients while restricted to the CD8⁺ population in healthy individuals. To elucidate the role of the expanded populations, we studied characteristics of the expanded and non-expanded T cells from MG patients who had persistent T cell expansions over more than 2 years. The cells were analysed with regard to phenotype, cytokine secretion, cytokine mRNA expression and reactivity with the autoantigen, the acetylcholine receptor. The characteristics of the expanded populations in MG clearly differed from those found in healthy individuals. More cells in the CD4⁺ expanded populations expressed HLA-DR and there was also a tendency for higher expression of CD25, CD28 and CD57. The number of cells spontaneously secreting cytokines was higher in the expanded populations. A dominant Th1-type cytokine secretion and mRNA expression was noted. Autoantigen-reactive CD4⁺ T cells were largely restricted to the expanded populations.

INTRODUCTION

Expansions of T cells bearing a particular Vα/Vβ gene product are commonly found in the CD8⁺ population in healthy individuals, but are much more rare in the CD4⁺ population [1]. Abnormal T cell receptor (TCR) gene usage by CD4⁺ cells has been described in several autoimmune and inflammatory disorders, such as multiple sclerosis [2], rheumatoid arthritis (RA) [3], sarcoidosis [4] and vasculitis [5–7]. Expanded cell populations are also present in patients with myasthenia gravis (MG) [8–11] and its corresponding animal model, experimental autoimmune myasthenia gravis (EAMG) [12–15]. In EAMG a highly selected TCR Vβ6⁺ CD4⁺ cell expansion with a conserved glutamic acid residue in CDR3 has been reported [13]. These Vβ6⁺ CD4⁺ T cells responded to an immunodominant peptide of the acetylcholine receptor (AChR) α-subunit and the cells were therefore suggested to be directly involved in the pathogenesis of the disease [12]. However, both an in vitro immune response to AChR and development of EAMG could be achieved after depletion of TCR Vβ6⁺ cells [14]. Moreover, in TCR Vβ6⁺ knock-out mice the usage of other TCR Vβ genes in EAMG has been shown. These results suggest that several TCR Vβ might be involved in the pathogenesis of EAMG.

In the human disease, most earlier studies have shown expansions of both CD4⁺ and CD8⁺ cells. A restricted usage of Vβ12 [8], Vβ4 and Vβ6 [11], Vβ1, Vβ13.2, Vβ17 and Vβ20 [9] in both CD4⁺ and CD8⁺ cell populations and no restriction to certain Vα/Vβ have been described [10]. The functional role of these expanded populations in human diseases has not been elucidated.

In this study we investigate the prevalence of the expanded T cell populations and determine certain phenotypic and functional characteristics. The expanded cell populations in MG showed different characteristics when compared with such cells present in healthy individuals. The expanded cells were present in both CD4⁺ and CD8⁺ populations, and the CD4⁺ cells expressed in most cases more HLA-DR, CD25, CD28 and CD57. The number of cells spontaneously secreting cytokines was high, with a dominant Th1-type cytokine secretion and mRNA expression. Autoantigen-reactive CD4⁺ T cells were largely restricted to the expanded populations.

PATIENTS AND METHODS

Patients

Thirty-eight MG patients were included in this study, comprising 22 female and 16 male patients. The age of the patients ranged from 24 to 93 years (mean 61 years). The clinical evaluation was done using the Osserman–Oosterhuis classification [16]. Two patients had only ocular symptoms (stage I), 25 had mild generalized disease (stage IIA), nine had severe generalized disease (stage IIB) and three were in complete remission (A). Twenty-eight of the patients were thymectomized and five had normal thymic histology, 14 hyperplasia and nine thymomas.

The normal usage of different TCR Vα and Vβ gene products was determined in cells from 16 to 57 healthy individuals. The age of these controls ranged from 27 to 64 years (mean 45 years).

Analysis of lymphocyte subsets by flow cytometry

EDTA blood was collected by Vacutainers (Becton Dickinson, Mountain View, CA). The blood was kept at room temperature and staining of lymphocytes was performed within 24 h of collection of the blood. The following MoAbs were purchased from Becton Dickinson: CD3, CD45RA and CD57 (FITC-conjugated); CD8, CD25, CD28 and HLA-DR (PE-conjugated); CD4 and CD8 (pyridine chlorophyll protein (PerCp)-conjugated). MoAbs against CD45RO (FITC-conjugated) were purchased from Dakopatts A/S (Glostrup, Denmark). The TCR Vα- and Vβ-specific MoAbs included were: Vα2.3, Vα12.1, Vβ3, Vβ5.1, Vβ5.2 + 3, Vβ5.3, Vβ6.7, Vβ8 and Vβ12 purchased from T Cell Sciences Inc. (Cambridge, MA), Vβ2, Vβ13, Vβ19 and Vβ21 from Immunotech S.A. (Marseille, France) and Vβ9 from Pharmingen (San Diego, CA).

A triple-staining technique was used. Blood (50 μl) was incubated on ice with a saturating amount of unlabelled TCR MoAb for 30 min. After the first incubation, cells were washed twice with PBS (containing 0.2% bovine serum albumin (BSA) and 0.01% sodium azide) and then incubated with FITC-conjugated F(ab)₂ fragments from rabbit anti-mouse immunoglobulin for 30 min. After washing the cells three times in PBS, normal mouse serum (NMS) was added for 10 min to block remaining rabbit anti-mouse immunoglobulin, and the cells were washed once again. Thereafter, a cocktail of PE-conjugated CD4 and PerCp-conjugated CD8 was added, and after 30 min the erythrocytes were lysed at room temperature with FACS lysing solution (Becton Dickinson). The cells were finally washed three times with PBS, and a total of 10⁴ cells per sample was analysed in a FACScan flow cytometer (Becton Dickinson) by a Hewlett Packard 300 computer using the Lysis II program. Lymphocytes were gated by forward and side scatter. Optimal compensation was set for green (FITC), orange (PE) and red (PerCp) fluorescence. For estimation of the absolute percentage of CD4⁺ or CD8⁺ T cells, cells were stained with FITC-conjugated CD3, PE-conjugated CD4 and PerCp-conjugated CD8, and the CD4⁺ lymphocytes were identified as CD3⁺CD4⁺ cells, and the CD8⁺ lymphocytes as CD3⁺CD8⁺ cells. Each population was analysed separately to ensure that a change in one subpopulation of T cells would not be masked by the other. The presence of a T cell expansion was defined as a value > 3 times the median value for that seen in the control population.

Cells from six MG patients showing persistent TCR expansions over 2 years or more were further studied for expression of cell surface markers, cytokine mRNA expression and cytokine secretion.

Phenotypic characterization of the expanded/non-expanded cell populations by FACS

A triple staining was done with the MoAbs to TCR, CD4/CD8 and one of the following: HLA-DR, CD45RA, CD45RO, CD28, CD25 and CD57. The cells were stained and analysed in a FACScan flow cytometer using the Lysis II software. The expanded populations were analysed by gating the CD4⁺ or CD8⁺ population, which was positive for one of the following markers: CD45RA, CD45RO, HLA-DR, CD25, CD28 or CD57. This population was further analysed to see if the subset of cells was positive for any particular TCR Vα/Vβ gene segment.

Separation of the expanded T cell population and the remaining CD4⁺ and/or CD8⁺ T cells

Heparinized whole blood was separated on Ficoll–Hypaque gradient [17]. Peripheral blood mononuclear cells (PBMC) were washed twice in PBS and then resuspended in RPMI 1640 medium (Gibco, Paisley, UK).

The cell populations were separated using the MiniMacs separation system (MiniMacs; Miltenvi Biotec GmbH, Bergisch Gladbach, Germany). Three cell populations were obtained from each patient: (i) the expanded T cell population with a certain Vα/Vβ, (ii) the remaining CD4⁺ cells, and (iii) the remaining CD8⁺ cells. Briefly, PBMC was incubated with the TCR Vα/Vβ-specific MoAb at 4°C for 20 min, washed three times with PBS and thereafter incubated with anti-mouse IgG-coated magnetic beads at 4°C for 20 min. The expanded T cells were collected by positive selection. After elimination of the expanded cell population, the non-expanded CD4⁺ or CD8⁺ cells were separated by using anti-CD4⁺ or CD8⁺ antibody-coated magnetic beads subsequently.

Reverse transcription-polymerase chain reaction

Total RNA was extracted from the expanded T cells, CD4⁺ or CD8⁺ T cells using the acid guanidinium thiocyanate phenol chloroform method as described [18].

The cytokines interferon-gamma (IFN-γ), IL-2, IL-3, IL-4, IL-5, IL-10 and tumour necrosis factor-alpha (TNF-α) were analysed, and β-actin was used to verify addition of equal amounts of RNA. Reverse transcription (RT) took place with a specific primer for each cytokine analysed, using the recombinant Thermus thermophilus DNA polymerase (Perkin-Elmer, Branchburg, NJ). Thereafter the polymerase chain reactions (PCR) were performed by adding the 5′ primer and changing the buffer conditions. The primer sequences used in the PCR are listed in Table 1. The PCR profile used was: denaturation at 94°C for 15 s, annealing at different temperatures for 30 s and extension at 70°C for 60 s. Annealing temperatures and number of PCR cycles for the different cytokines were: IFN-γ, 52°C, 35 cycles; IL-2, 53°C, 38 cycles; IL-3, 64°C, 38 cycles; IL-4, 60°C, 42 cycles; IL-5, 59°C, 38 cycles; IL-10, 64°C, 38 cycles; and TNF-α, 60°C, 38 cycles.

Table 1.

Primer sequences

The PCR products were visualized on a 1.5% agarose gel, stained with ethidium bromide.

Enumeration of IFN-γ- and IL-4-secreting T cells

Three of six MG patients with persistent TCR expansion of CD4⁺ populations were included in this test. PBMC were separated into adherent and non-adherent cells by incubation in Petri dishes at 37°C for 1 h. The non-adherent cells were recovered and the adherent cells scraped off and used as antigen-presenting cells (APC). Expanded CD4⁺ T cells or CD4⁺ T cells without expansion were separated by magnetic beads. A preliminary study showed that binding to the beads did not affect the measured functions of the cells. ELISPOT assay was performed as described [19]. Briefly, plates (Millititer-HAM; Millipore Co., Bedford, MA) were coated with 100-μl aliquots of mouse anti-human IFN-γ (200 U/ml) or IL-4 (10 μg/ml) MoAbs (Genzyme Corp., Boston, MA) at 4°C overnight. After washings with PBS, the expanded CD4⁺ T cells or control CD4⁺ T cells, all preincubated with 10% monocytes/macrophages and human muscle AChR for 20 h, were transferred to the coated wells and incubated at 37°C with 5% CO₂ for another 24 h. Thereafter 100 μl of rabbit polyclonal anti-human IFN-γ (1/400) or IL-4 (5 μg/ml) antibodies (Genzyme) were added and incubated at 37°C for 2 h. After washing, biotinylated anti-rabbit IgG diluted 1:1000 (Vector Labs, Burlingame, CA) was added followed by incubation with avidin–biotin–peroxidase complex diluted 1:200 (ABC Vectastain-Elite kit; Vector Labs) for 50 min. After peroxidase staining, using the substrate 3-amino-9-ethyl carbazol (Sigma, St Louis, MO), spots corresponding to cells that had secreted IFN-γ or IL-4 were enumerated under a dissection microscope. Cells incubated in parallel in complete medium were used to detect spontaneous cytokine secretion, and mitogen-induced stimulation was evaluated in cultures stimulated with concanavalin A (Con A; 20 μg/ml). All assays were done in triplicate. The plates were read double-blinded by two experienced researchers. The intra-assay coefficient of variation was 8% for IFN-γ and 13% for IL-4.

Statistical analysis

Mann–Whitney U-test and χ² test were used to compare the values between the groups. P < 0.05 was considered to be significant.

RESULTS

TCR usage

The percentages of CD4⁺ and CD8⁺ T cells expressing specific Vα or Vβ gene products in MG patients and healthy individuals are presented in Fig. 1a,b. Decreased numbers of cells expressing Vβ2 in the CD8⁺ population and Vβ3 in the CD4⁺ population were found in the patients. Ten patients had Vβ2-expressing cells and 17 patients had Vβ3-expressing cells below 10% of that found in the healthy group.

Fig. 1.

(a) Percentage of the various Vα/Vβ T cell receptor (TCR)-bearing CD4⁺ cells in myasthenia gravis (MG) patients and healthy controls (HC). The lower, mid and upper horizontal lines of the boxes represent 25th, 50th and 75th percentiles, respectively; the vertical lines from the 10th to the 90th percentile. (b) Percentage of the various Vα/Vβ TCR-bearing CD8⁺ cells in MG patients and HC. The lower, mid and upper horizontal lines of the boxes represent 25th, 50th and 75th percentiles, respectively; the vertical lines from the 10th to the 90th percentile.

An expansion of CD4⁺ and/or CD8⁺ cell populations was found in 18/38 MG patients, a total of 43 expansions. Of the 18 MG patients with T cell expansions, 14 had more than one expanded population. One patient had an expansion in only CD4⁺ cells, 10 in only CD8⁺ cells and seven patients in both the CD4⁺ and the CD8⁺ populations. Extremely high values for Vβ3 (40% and 41%) and Vβ19 (50% and 26%) were found in the CD8⁺ populations of three patients. The expanded populations expressed different Vα or Vβ gene products including Vα2.3 (n = 5), Vα12.1 (n = 3), Vβ2 (n = 1), Vβ3 (n = 3), Vβ5.1 (n = 2), Vβ5.2 + 3 (n = 4), Vβ5.3 (n = 3), Vβ6.7 (n = 3), Vβ8 (n = 2), Vβ12 (n = 2), Vβ13 (n = 10), Vβ19 (n = 3) and Vβ21 (n = 2).

T cell expansions were also found in 15/57 healthy individuals. A total of 20 expanded cell populations expressed Vα2.3 (n = 3), Vβ5.1 (n = 1), Vβ5.2 + 3 (n = 5), Vβ5.3 (n = 4), Vβ6.7 (n = 3), Vβ8 (n = 2), Vβ12 (n = 1) and Vβ19 (n = 1). In contrast to what was seen in the MG patients, all the expanded cells in the healthy individuals were CD8⁺.

There was a significantly increased number of expansions expressing TCR Vβ13 in the MG patients compared with healthy individuals (P < 0.05).

The clinical and laboratory data of the MG patients with T cell expansions are shown in Table 2. There was no relation of the TCR Vα/Vβ expansions to certain HLA-A, HLA-B, DQA1, DQB1 alleles [10]. However, the patients with expanded T cell populations were significantly older, with a median age of 68 years, than patients without expansions who had a median age of 47 years (P < 0.05). Patients who were not thymectomized had a higher prevalence of expanded T cell populations than others (P < 0.02) [10].

Table 2.

Clinical data of myasthenia gravis (MG) patients with T cell expansions

Phenotypic characteristics of the expanded T cell populations

The expression of the cell surface markers CD45RA/RO, HLA-DR, CD25, CD28 and CD57 in both the expanded and non-expanded T cell populations from six patients with persistent cell expansions for > 2 years is shown in Table 3. HLA-DR was expressed on significantly more expanded CD4⁺ cells (P < 0.03). CD45RA was expressed on more CD4⁺ cells belonging to the expanded population in 3/4 cases, CD25 in 3/4, CD28 in 3/4 and CD57 in 3/4 cases, compared with the non-expanded population. There was no significant difference with regard to the expression of these phenotypic markers on the expanded and non-expanded populations belonging to CD8⁺ populations.

Table 3.

Phenotypic characterization of expanded T cell populations with a duration of > 2 years in myasthenia gravis (MG) patients

Expression of cytokine mRNA

The mRNA expression of IFN-γ, IL-2, IL-3, IL-4, IL-5, IL-10 and TNF-α by CD4⁺, CD8⁺ and expanded T cells is shown in Table 4. Most patients showed a Th1 type of cytokine profile. Thus, IFN-γ was expressed in all cell populations and IL-2 as well as TNF-α in most of them. Neither IL-4 nor IL-10 was detected and IL-5 was expressed in a few cell populations. A Th0-type cytokine profile of cells expressing IFN-γ, IL-2, IL-3 and IL-5 was found in CD4⁺ cells of patient 2. There was no difference between expanded and non-expanded populations. However, this RT-PCR is not quantitative and thus the amount of cytokine mRNA expression of different cell populations could not be accurately determined.

Table 4.

Cytokine gene expression in different cell subsets in myasthenia gravis (MG) patients

IFN-γ and IL-4 secretion

The spontaneous, AChR- and Con A-induced numbers of IFN-γ- or IL-4-secreting cells of both expanded and non-expanded CD4⁺ cell populations are shown in Fig. 2a,b.

Fig. 2.

(a) Number of IFN-γ-secreting cells per 10⁵ cells from both expanded and non-expanded populations of three myasthenia gravis (MG) patients with CD4⁺ cell expansions. The bars represent the mean value and s.d. from triplicate analyses. (b) Number of IL-4-secreting cells per 10⁵ cells from both expanded and non-expanded populations of three MG patients with CD4⁺ cell expansion. The bars represent the mean value and s.d. from triplicate analyses. AChR, Acetylcholine receptor; Con A, concanavalin A.

More cells with spontaneous secretion of IFN-γ were present in expanded than non-expanded T cell populations from patient 16 and patient 2. The AChR induced more IFN-γ-secreting cells in the expanded populations than in the non-expanded in all three patients, although a marked difference was present in only patients 16 and 11. With regard to Con A stimulation, there was no consistent difference between expanded and non-expanded populations. Thus, in 2/3 cases there was a higher spontaneous and autoantigen-induced IFN-γ secretion in the expanded cell populations compared with the non-expanded.

Comparatively low numbers of IL-4-secreting cells were demonstrated in both expanded and non-expanded populations. A higher number of cells with spontaneous secretion was present in two expanded populations from patient 16. In this patient, autoantigen-reactive cells producing IL-4 were all found in the expanded population.

DISCUSSION

In the present study significantly lower expressions of Vβ2 in CD8⁺ and Vβ3 in CD4⁺ in MG patients were observed. CD4⁺ cells in 17 MG patients had a Vβ3 antibody reactivity below 10% of that seen in the healthy group. This might indicate that the described Vβ3 allelic polymorphism [20] is an important factor in affecting the TCR repertoire. Expanded T cells expressing Vβ13 were significantly more common in MG and an expansion of Vβ13.2 in MG has been earlier suggested to be induced by superantigen [9]. Our findings of a lower usage of Vβ2 and Vβ3 might be consistent with an antigen- or superantigen-induced deletion of certain populations of T cells. In this case, the deletions might be the primary event and the expansions just compensatory to these deletions.

Of the 45 expansions, 11 were found in the CD4⁺ population. This is a much higher prevalence than in healthy individuals [1]. CD8 expansions frequently occur in healthy individuals and are usually CD28⁻, HLA-DR⁻, CD57⁺ with a variable degree of expression of CD45RA or RO [1,21,22]. Expansions in the CD4⁺ populations are rarely found in healthy individuals. CD4⁺ expanded cells have been described to be predominantly CD28⁻, CD45RO⁺ and CD57⁺ [22]. In MG patients who had persistent CD4⁺ T cell expansions over > 2 years, the expanded populations had a significantly higher expression of HLA-DR. Also, the phenotypic markers CD25, CD28 and CD57 showed a tendency to be expressed by more cells belonging to the expanded populations. The expression of these activation markers did not change with the clinical condition of the patients. Thus, even when the disease appears to be clinically inactive, the pathogenic mechanisms may still be active. This indicates that in MG there is a discordance between clinical disease activity and in vivo activation status of the expanded T cells.

The expanded populations expressed both the naive CD45RA and memory CD45RO markers, a profile which is similar to what is seen on effector cells [23]. Thus, the expanded populations in MG might represent effector cells. The expression of CD28, an important costimulatory molecule known to prevent anergy, shows that these cells are capable of inducing and maintaining a T cell activation [24–26]. We have earlier reported expanded cell populations in Takayashu's arteritis showing a similar phenotype with high expression of activation markers CD45RA, CD45RO and CD28 [7].

A significant percentage of the CD4⁺ expanded populations were positive for CD57. The CD57 marker was initially described on large granular CD8⁺ lymphocytes [27,28]. This marker has been suggested to be important in cell–cell or cell–matrix interactions [29,30]. Increased numbers of CD8⁺CD57⁺ T cells are associated with previous human cytomegalovirus infection in normal healthy subjects [31,32], HIV infection [33,34], RA [35] and lymphoproliferative diseases [36,37]. T cells of this phenotype are more frequent in germinal centres than in peripheral blood and normally account for only a small percentage of peripheral blood CD4⁺ lymphocytes [38]. CD4⁺CD57⁺ expansions are present in RA [39] and chronic lymphocytic leukaemia [37]. In our study, three of the four CD4⁺ expanded populations constituted > 50% of the CD4⁺CD57⁺ subsets and the remaining constituted 42%. The biological significance of this remains to be elucidated.

Of particular interest is the finding that autoantigen-reactive cells were concentrated in the expanded populations. It is thus possible that the T cell reaction with the AChR is the force driving the expansions. The cytokine mRNA expression showed a predominant Th1 type of cytokine profile and this was also the case with regard to the cytokine secretion pattern as determined by ELISPOT technique. The dominant Th1-type cytokine expression and secretion might suggest an involvement of cytotoxic cellular immune responses induced by the AChR in MG.

Our data show the characteristics of the expanded T cell populations in MG patients. CD4⁺ expanded cell populations, not found in healthy individuals, expressed more cell surface activation markers. CD57, normally expressed only on CD8⁺, was also highly expressed on the expanded CD4⁺ cells. The cytokine profile revealed a Th1 profile both at the mRNA and at the protein level. The autoantigen reactivity was largely restricted to the expanded cells.