Quantitative and functional profiles of CD4⁺ lymphocyte subsets in systemic lupus erythematosus patients with lymphopenia

Abstract

Lymphopenia is a common clinical manifestation in patients with systemic lupus erythematosus (SLE). However, its physiopathogenic role and the contribution of different T cell subsets in this setting have not been addressed fully. The aim of this study was to characterize T cell subsets quantitatively and functionally and their association with lymphopenia and azathioprine treatment in SLE. We included 84 SLE patients and 84 healthy controls and selected 20 patients for a 6-month longitudinal analysis. Peripheral blood mononuclear cells were isolated, and T cell subsets were analysed by flow cytometry. Functional analyses included autologous and allogeneic co-cultures of T cells. Our data show persistently lower absolute numbers of CD4⁺CD25^high T cells [regulatory T cells (T_regs)] (1·9 versus 5·2, P < 0·01) and CD4⁺CD69⁺ T cells (3·2 versus 9·3, P = 0·02) and higher activity scores (4·1 versus 1·5, P = 0·01) in SLE patients with lymphopenia compared with those without lymphopenia. Lymphopenia increased the risk for decreased numbers of CD4⁺CD25^high cells (relative risk 1·80, 95% confidence interval 1·10–2·93; P = 0·003). In addition, azathioprine-associated lymphopenia was characterized by decreased absolute numbers of CD4⁺CD69⁺ and CD4⁺interleukin (IL)-17⁺ cells compared to disease activity-associated lymphopenia. Functional assays revealed that SLE effector T cells were highly proliferative and resistant to suppression by autologous T_regs. In summary, lymphopenia was associated with deficient numbers of CD4⁺CD25^high and CD4⁺CD69⁺ cells and resistance of effector T cells to suppression by T_regs, which could contribute to the altered immune responses characteristic of SLE. Furthermore, azathioprine treatment was associated with decreased numbers of CD4⁺CD69⁺ and CD4⁺IL-17⁺ cells and diminished T_reg suppressive activity.

Introduction

The maintenance and proliferation of T cells influence the magnitude of immune responses and are tightly regulated under physiological conditions. Homeostatic proliferation of T cells, also known as lymphopenia-induced proliferation (LIP) [1,2], is an important property of the adaptive immune system that occurs under conditions of T cell depletion and involves T cell expansion in response to both foreign and self-antigens [3].

LIP of T cells induced by chronic inflammatory diseases or therapeutic treatments (such as thymectomy or irradiation) is crucial for maintaining T cell numbers. Two forms of LIP have been described in terms of their kinetics and cytokine requirements: (a) homeostatic, which is slow and dependent on interleukin (IL)-7 and (b) spontaneous, which is rapid and IL-7-independent [2]. LIP differs qualitatively and quantitatively from normal T cell expansion, as it involves T cell proliferation in response to low-affinity self-antigens which, under normal conditions, do not induce T cell responses to the magnitudes observed in homeostatic proliferation [4].

Hyperresponsive T cells can contribute to the development of autoimmune pathology by breaking peripheral tolerance. Studies in murine models in which the endogenous antigen repertoire is limited have demonstrated that low-affinity self-antigens induce positive selection and antigen-specific proliferation only in lymphopenic mice. Furthermore, LIP contributes to autoimmune diseases, such as by promoting the onset of autoimmune gastritis and influencing T cell repertoire in rheumatoid arthritis [5]. The self-antigen repertoire to which T cells respond during lymphopenia is limited [6]. In transplantation, allospecific T cell expansion after lymphoablative treatments contributes to abrogation of tolerance [7]. However, the fact that only a minority of subjects, whether human or rodents, develop autoimmune disease during lymphopenia suggests that other factors contribute to the development of autoimmune pathology under lymphopenic conditions. Among these factors, regulatory T cell (T_reg) depletion [8,9] and local tissue inflammation are considered the most relevant [10,11].

Lymphopenia is a common clinical manifestation in patients with systemic lupus erythematosus (SLE) and has diagnostic and prognostic implications [12]. Lymphopenia may contribute to disease activity [13,14], but its role in the multiple immunological abnormalities displayed by lupus T cells has not been evaluated widely. Although we have shown previously that CD4⁺CD25^high T_reg numbers are decreased in patients with active SLE [15], T_reg defects and other abnormalities in T cell subsets have not been analysed in SLE in the context of lymphopenia.

The aim of this work was to characterize T lymphocyte subsets, such as regulatory and activated T cells, quantitatively and functionally, and their potential association with lymphopenia, disease activity and azathioprine treatment in patients with SLE.

Materials and methods

Patients and controls

Eighty-four patients with SLE diagnosis according to the American College of Rheumatology revised classification criteria for SLE [16] were included in this study. Lymphopenia was defined as <1500 cells/µl of venous peripheral blood in two consecutive measurements. Eighty-four age- and gender-matched healthy controls (HC) were included. The study was approved by the institutional review board, and all participants signed informed consent forms.

Of the SLE patients, 57 were in clinical remission [Mexican Systemic Lupus Erythematosus Disease Activity Index (MEX-SLEDAI) score of 0 and had gone at least 12 months without treatment with any immunosuppressive drugs], and 27 had active disease and were untreated at the time of study. These patients had arthritis, mucosal ulcers, leukopenia and increased levels of double-stranded DNA (anti-dsDNA) autoantibodies. Five had nephropathy and two had thrombocytopenia.

Cell isolation and cultures

Peripheral blood mononuclear cells (PBMCs) were obtained from SLE patients and HCs by density-gradient centrifugation (Lymphoprep, Oslo, Norway). CD4⁺ T cells were isolated by magnetic depletion of non-CD4⁺ T cells (CD4 T Cell Isolation Kit II; MicroBeads Miltenyi Biotec, Auburn, CA, USA). CD25⁺ and CD25^- subsets were obtained by magnetic isolation of labelled CD25⁺ cells (CD25 MicroBeads Miltenyi Biotec).

Flow cytometry

PBMCs were stained with combinations of the following antibodies: anti-CD4 fluorescein isothiocyanate (FITC), anti-CD25 phycoerythrin (PE), anti-CD4 peridinin-chlorophyll (PerCP) and anti-CD69 PerCP (BD Pharmingen, San Jose, CA, USA). Intracellular staining was performed in freshly isolated PBMCs. After fixation and permeabilization (Cytofix/Cytoperm Kit; BD Biosciences), cells were stained with anti-IL-17A-PE (eBiosciences). Data were collected on a FACScan flow cytometer (BD Biosciences) and analysed using FlowJo software. The cut-off point for positive staining was set above the level of staining of unstained cells. Also, a gate that included 2% of cells with the highest CD25 expression was set to define CD25^high cells. Results are expressed as absolute numbers. Absolute numbers of CD4⁺ cells were obtained by multiplying the number of lymphocytes (cells/µl of venous peripheral blood) by the percentage of CD4⁺ cells. For CD4⁺CD25^high quantification, absolute numbers were calculated by multiplying the number of CD4⁺CD25⁺ cells by the percentage of CD4⁺CD25^high cells.

Forkhead box P3 (FOXp3) gene expression

RNA was extracted from 1 million CD4⁺CD25^- and CD4⁺CD25^high cells using the RNeasy Kit (Qiagen, Valencia, CA, USA). cDNA was obtained from 2 µg of RNA using Moloney murine leukaemia virus reverse transcriptase (Gibco-BRL, Douglasville, GA, USA). FOXP3 cDNA amplification was performed using specific primers (5′-TCA AGC ACT GCC AGG CG-3′ and 5′-CAG GAG CCC TTG GAT-3′) and Taq DNA polymerase (Gibco-BRL).

Proliferation and suppression assays

CD4⁺CD25^high T_regs and CD4⁺CD25^- effector T cells were stimulated alone or in co-cultures (effector : T_reg ratio of 1:1). Cultures were set up in triplicate. 5,6-Carboxy-succinimidyl-fluorescein ester (CFSE)-labelled cells were stimulated in 96-well plates coated with anti-CD3 (10 µg/ml, clone HIT3a) and soluble anti-CD28 (0·25 µg/ml, clone 28·2). After 96 h of culture, cells were harvested and proliferation was analysed by flow cytometry. Cell proliferation was measured by CFSE dilution and analysed using the cell proliferation index [17]. Results are expressed in arbitrary units.

Statistical analysis

Descriptive data are expressed as median and interquartile range or mean and standard deviation, as needed. Differences among groups were analysed by independent-samples t-test or one-way analysis of variance (anova) with Dunnett T3 correction for multiple comparisons, as well as by Mann–Whitney U-test or Kruskal–Wallis test, as needed. Paired data were analysed by Wilcoxon's signed-rank or Friedman test. Correlations were evaluated using Spearman's rho. The association between categorical variables was analysed by χ² test, and 95% confidence interval (CI) were obtained. Statistical significance was defined as P < 0·05. Statistical analyses were performed using spss version 16·0.

Results

Differences among CD4⁺ T cell subsets between lymphopenic SLE patients and healthy controls

Demographic and clinical characteristics of the 84 SLE patients and 84 HCs are summarized in Table 1. Peripheral blood CD4⁺ T cell subsets (CD4⁺CD25^-, CD4⁺CD25⁺, CD4⁺CD25^high, CD4⁺CD69⁺, CD4⁺IL-17⁺) were measured by flow cytometry (Fig. 1).

Table 1.

Clinical characteristics and T lymphocyte subsets of systemic lupus erythematosus (SLE) patients with and without lymphopenia

	SLE patients with lymphopenia (n = 62)	SLE patients without lymphopenia (n = 22)	P
Gender (female/male)	60/2	21/1
Age (years)	33 ± 4	29 ± 6	n.s.
Years since SLE diagnosis	6 ± 3	8 ± 5	n.s.
MEX-SLEDAI	4·1 ± 3·8	1·5 ± 2·1	0·01
Leucocytes (absolute numbers)	3604 ± 1296	4668 ± 1683	0·03
Lymphocytes (absolute numbers)	582 ± 231	1992 ± 433	<0·01
CD4+CD25high (absolute numbers)	1·9 ± 1·2	5·2 ± 3·2	<0·01
CD4+CD69+ (absolute numbers)	3·2 ± 4·7	9·3 ± 9·1	0·02

MEX-SLEDAI, Mexican Systemic Lupus erythematosus Disease Activity Index; n.s., not significant.

Fig. 1.

Decreased CD4⁺CD25^high regulatory T cell (T_reg) numbers in SLE patients with lymphopenia compared to systemic lupus erythematosus (SLE) patients without lymphopenia. Peripheral blood mononuclear cells (PBMCs) were obtained from SLE patients with and without lymphopenia by density-gradient centrifugation. CD4⁺ T cells were isolated by magnetic depletion. CD25⁺ and CD25^- subsets were obtained by magnetic isolation of labelled CD25⁺ cells. The cut-off point for positive staining was set above the level of staining of unstained cells and isotype control. A gate that included the 2% of cells with the highest CD25 expression was set to define CD25^high cells. SLE patients without lymphopenia had higher (a) CD4⁺CD25⁺ and (b) CD4⁺CD25^high numbers than SLE patients with lymphopenia (c,d). Zebra plots from representative samples are depicted.

We evaluated FOXP3 mRNA expression to confirm the accuracy of using CD4⁺CD25^high expression as a marker for T_regs. As expected, FOXP3 mRNA was detected only in the CD25^high cell fraction (data not shown).

Patients with lymphopenia had higher MEX-SLEDAI scores (4·1 versus 1·5, P = 0·01), as well as decreased numbers of CD4⁺CD25⁺ (21·2 ± 17·2 versus 74·3 ± 53·5, P < 0·01), CD4⁺CD25^high (1·9 versus 5·2, P < 0·01) and CD4⁺CD69⁺ (3·2 versus 9·3, P = 0·02) cells in comparison to SLE patients without lymphopenia (Table 1).

Longitudinal study: T cell subsets during 6-month follow-up

Persistent deficiency in CD4⁺CD25^high and CD4⁺CD69⁺ T cell numbers in patients with lymphopenia

We conducted a longitudinal study that included 20 patients, 13 with lymphopenia and seven without lymphopenia. Samples were collected at 0 (basal), 1, 3 and 6 months. At time zero, all patients were untreated. Of the 13 patients with lymphopenia, seven patients were treated subsequently with immunosuppressive drugs (n = 6 for azathioprine, n = 4 for hydroxychloroquine, n = 3 for cyclophosphamide and n = 6 for prednisone). Three patients without lymphopenia received hydroxychloroquine. At baseline, patients with lymphopenia had decreased numbers of CD4⁺ (26·21 versus 85·84, P = 0·002), CD4⁺CD25⁺ (13·85 versus 48·6, P = 0·013) and CD4⁺CD25^high (1·48 versus 4·79, P = 0·049) cells compared to patients without lymphopenia. In addition, there were no significant differences or correlations in disease activity measurements (MEX-SLEDAI and anti-dsDNA) and lymphocyte subset numbers between both groups of SLE patients at time zero.

Over the course of the 6-month study, SLE patients with lymphopenia had higher disease activity indices (3·98 versus 1·74, P = 0·050) and decreased numbers of CD4⁺CD25^high (1·73 versus 5·07, P = 0·001) and CD4⁺CD69⁺ lymphocytes (3·2 versus 8·8, P = 0·049) (Fig. 2) compared to patients without lymphopenia. Decreased numbers of CD4⁺CD25^high and CD4⁺CD69⁺ lymphocytes were persistent during the follow-up, as depicted in Figs 3 and 4.

Fig. 2.

Decreased regulatory T cell (T_reg) and CD4⁺ CD69⁺ numbers in systemic lupus erythematosus (SLE) patients with lymphopenia during 6-month follow-up. Pooled data from the longitudinal study show decreased CD4⁺CD25^high (white bars) and CD4⁺CD69⁺ (black bars) numbers among SLE patients with lymphopenia versus those without lymphopenia. Error bars represent standard deviation.

Fig. 3.

Decreased regulatory T cell (T_reg) numbers in systemic lupus erythematosus (SLE) patients with lymphopenia during 6-month follow-up. Persistently decreased CD4⁺CD25^high T_reg numbers were found among SLE patients with lymphopenia (black line) versus those without lymphopenia (grey line) during the 6-month follow-up. Error bars represent standard deviation. *P < 0·05 for the comparison among patients with lymphopenia versus those without lymphopenia for each time-point.

Fig. 4.

Decreased activated T cell numbers in SLE patients with lymphopenia during 6-month follow-up. Systemic lupus erythematosus (SLE) patients with lymphopenia had reduced CD4⁺CD69⁺ T cell numbers compared to patients without lymphopenia during the follow-up period. Basal measurements (mean absolute cell numbers) are shown in black bars; measurements (mean absolute cell numbers) taken at 1, 3 and 6 months are shown in grey bars. Error bars represent standard deviation. *P < 0·05 for the comparison among patients with lymphopenia versus those without lymphopenia for each time-point.

Absolute lymphocyte numbers correlated with numbers of CD4⁺ (r = 0·509, P < 0·01), CD4⁺CD25^high (r = 0·523, P < 0·01) and CD4⁺CD69⁺ (r = 0·364, P < 0·01) cells.

Patients with lymphopenia had increased numbers of CD4⁺CD25^- cells at 0 and 3 months (26·22 at 0 months versus 30·62 at 3 months, P = 0·019) as well as at 6 months (26·22 at 0 months versus 74·36 at 6 months, P = 0·023). Patients without lymphopenia also had increased numbers of CD4⁺CD25^- cells at 6 months (85·84 at 0 months versus 188·75 at 6 months, P > 0·05).

Patients with lymphopenia showed increased numbers of CD4⁺CD69⁺ cells from 0 and 6 months (0·85 at 0 months versus 3·06 at 6 months, P = 0·023). Interestingly, in patients without lymphopenia, the numbers of CD4⁺CD25^high cells decreased over the 6-month period (4·87 at 0 months versus 2·91 at 6 months, P = 0·018). This did not correlate with MEX-SLEDAI scores (r = 0·012, P > 0·05). Furthermore, immunosuppressive drugs evaluated had no effect on numbers of each lymphocyte subset.

Lymphopenia as a risk factor for decreased numbers of CD4⁺CD25^high cells in SLE patients

We analysed pooled data from the longitudinal study to evaluate the association between lymphopenia and decreased numbers of CD4⁺CD25^high cells. We first analysed the numbers of CD4⁺CD25^high cells in HC and found that the mean absolute number ± standard deviation for this group was 4·78 ± 0·73. We used this number as the cut-off value (<3·32) for low versus normal CD4⁺CD25^high cell numbers. We found that lymphopenia increased the risk for decreased numbers of CD4⁺CD25^high cells [relative risk (RR) 1·804, 95% CI 1·10–2·93; P = 0·003].

Azathioprine-associated lymphopenia versus disease activity-associated lymphopenia

To evaluate differences between disease activity- and azathioprine-associated lymphopenia, we conducted an analysis of the following four groups (n = 10 per group): (1) SLE patients with lymphopenia secondary to disease activity (lymphocyte count <1500 cells/µl in two consecutive measurements, MEX-SLEDAI score >5, untreated at the time of study); (2) SLE patients without lymphopenia; (3) SLE patients with azathioprine-associated lymphopenia (lymphocyte count <1500 cells/µl in two consecutive measurements in patients treated with azathioprine in the 6 months prior to the time blood samples were collected, increased mean corpuscular volume (>100) and other cytopenias (e.g. thrombocytopenia) defined by the attending rheumatologist as myelotoxic from azathioprine treatment); and (4) age- and gender-matched HCs.

We found decreased numbers of CD4⁺CD69⁺ (0·46 versus 1·04, P = 0·028) and CD4⁺IL-17⁺ (0·24 versus 0·67, P = 0·019) cells in SLE patients with azathioprine-associated lymphopenia versus patients with disease activity-associated lymphopenia (Fig. 5). A negative correlation was found between azathioprine treatment duration and absolute numbers of CD4⁺CD25^high cells (r = −0·633, P = 0·049). No correlation was found between the daily dosage of azathioprine and number of T cells in each subset analysed.

Fig. 5.

T cell subsets in systemic lupus erythematosus (SLE) patients with azathioprine-associated lymphopenia versus SLE patients with disease activity-associated lymphopenia. SLE patients with lymphopenia had decreased CD4⁺CD69⁺ (P = 0·028, black bars) and CD4⁺interleukin (IL)-17⁺ (P = 0·019, white bars) cell numbers compared to patients with disease activity-associated lymphopenia. Error bars represent standard deviation.

Functional abnormalities of effector T cells and T_regs in lymphopenic SLE patients with or without azathioprine treatment

To analyse further the potential functional relevance of the T cell number abnormalities found in SLE patients with lymphopenia, we performed co-culture assays between CD4⁺CD25^- effector T cells and CD4⁺CD25^high T_regs from 10 patients (five with disease activity-associated lymphopenia and five with azathioprine-associated lymphopenia). Patients with disease activity-associated lymphopenia were untreated at the time of the study, and patients with azathioprine-associated lymphopenia were treated only with azathioprine.

We first assessed effector T cell and T_reg proliferation after in vitro stimulation with plate-bound anti-CD3 and anti-CD28. Effector T cells and T_regs from SLE patients with disease activity-associated lymphopenia showed higher proliferation in comparison to T cells from SLE patients with azathioprine-associated lymphopenia (1079 ± 196 versus 474 ± 92 for effector T cells, P = 0·001; 481 ± 137 versus 237 ± 61 for T_regs, P = 0·012) (Fig. 6).

Fig. 6.

Abnormal effector cell and regulatory T cell (T_reg) proliferation in systemic lupus erythematosus (SLE) patients with lymphopenia. SLE patients with disease activity-associated lymphopenia showed increased proliferation of CD4⁺CD25^- effector T cells (black bars) and CD4⁺CD25^high T_regs (white bars) in response to in vitro stimulation with plate-bound anti-CD3 and soluble anti-CD28 compared to patients with azathioprine (AZA)-associated lymphopenia (1079 ± 196 versus 474 ± 92, P = 0·001 for effector T cells and 481 ± 137 versus 237 ± 61, P = 0·012 for T_regs). Bars represent mean proliferation cell index (arbitrary units). Error bars represent standard deviation.

In suppression assays, autologous co-cultures of effector T cells and Tregs revealed low T_reg suppressive activity in both patient groups (20·29 ± 6·4% for cells from patients with disease activity-associated lymphopenia versus 29·09 ± 20% for cells from patients with azathioprine-associated lymphopenia, P = 0·40). We found higher T_reg suppressive capacity in allogeneic co-cultures of effector T cells from healthy donors with T_regs from SLE patients with disease activity-associated lymphopenia compared to co-culture with T_regs from patients with azathioprine-associated lymphopenia (78 ± 7·22% versus 8·31 ± 6·8%, P = 0·001) (Fig. 7).

Fig. 7.

Lymphopenia-associated resistance of SLE effector T cells to suppression by regulatory T cells (T_regs) in autologous co-cultures, and azathioprine-associated decrease in suppressive capacity of systemic lupus erythematosus (SLE) T_regs in allogeneic co-cultures. Low T_reg suppressive activity was observed in autologous co-cultures of T cells from both patients with disease activity- and azathioprine-associated lymphopenia (20·29% and 29·09%, respectively; P = 0·40). Decreased T_reg suppressive activity was observed in allogeneic co-cultures of effector T cells from healthy donors and T_regs from SLE patients with azathioprine-associated lymphopenia compared to those from patients with disease activity-associated lymphopenia (8·31% versus 78%, P = 0·001). Bars represent mean percentage of suppression of cell proliferation index for autologous and allogeneic co-cultures. Error bars represent standard deviation.

Discussion

LIP has been associated previously with the development of both systemic and organ-specific autoimmune diseases [2,3]. In SLE, lymphopenia is recognized widely as a diagnostic and prognostic factor. In addition, other factors, such as T_reg deficiency, have been suggested as necessary for the development of autoimmunity in lymphopenic hosts [3].

Lymphopenia has been reported to occur in as many as 81·9% of Hispanic SLE patients and, consistent with our findings, has been associated independently with higher disease activity [14].

The relationship between lymphopenia and the development and activity of autoimmune disease has been challenging to dissect. In SLE, disease activity is a function of numerous immunological abnormalities, such as decreased numbers [15] and defective suppressive function [18] of T_regs. However, controversy still exists regarding these defects and other abnormalities, such as resistance of effector cells to the suppressive effects of T_regs, as reported previously by our group and others [17,19]. More importantly, none of these studies examined the role of lymphopenia in the aforementioned abnormalities.

We observed persistently decreased absolute numbers of CD4⁺CD25^high T_regs in SLE patients with lymphopenia compared to those without lymphopenia that correlated with disease activity, suggesting that lower T_reg numbers contribute to the maintenance of systemic autoimmunity. T_regs inhibit spontaneous proliferation of naive T cells from lymphopenic hosts via their suppressive functions [20] or by enhancing apoptosis and impairing differentiation of naive T cells [21]. However, our functional assays revealed that effector T cells from lymphopenic SLE patients displayed increased proliferation upon in vitro stimulation and in autologous co-cultures. Deficient T_reg numbers might be caused by increased conversion to effector cells. Lupus T cells have multiple defects in IL-2 production [22,23], and diminished IL-2 production in FoxP3^gfp KI mice is attributed to conversion of T_regs into effector pathogenic T cells during lymphopenia [24]. Consistent with these studies, we found that effector T cell numbers increased significantly in lymphopenic patients during the course of the 6-month follow-up. In addition, T_reg proliferation under lymphopenic conditions depends on effector T cells [25]. Hence, the steady increase in effector T cell numbers in patients with lymphopenia might serve as a counter-regulatory mechanism to promote T_reg proliferation. To test this hypothesis, longer follow-up is required. Therefore, our findings suggest that multiple lymphopenia-associated quantitative and functional defects in effector and T_regs are important for maintaining autoimmune pathology.

Interestingly, in non-lymphopenic SLE patients, T_reg numbers decreased after 6 months, although the final numbers were similar to that of HC. Hence it seems that, in this group of patients, an increased basal T_reg number is required to maintain normal lymphocyte counts but is dispensable after expansion of effector cells with time.

We also observed persistently lower CD4⁺CD69⁺ numbers in patients with lymphopenia. We used CD69 as an activation marker, which has not been evaluated previously in the context of LIP. CD69 expression might be down-regulated as a counter-regulatory mechanism to make autoreactive effector cells resistant to activation. To evaluate this possibility, T cell receptor (TCR) repertoire analysis must be performed. We hypothesize that such an analysis would corroborate previous findings regarding limited repertoire diversity and skewing during lymphopenia [26].

The defects in T cell numbers have functional implications, as our functional assays revealed that effector and T_regs from SLE patients with disease activity-associated lymphopenia displayed higher proliferative responses to in vitro stimulation. This is in contrast to the poor proliferative responses shown previously with lupus T cells, although these studies did not evaluate T cell proliferation specifically in lymphopenic patients [1,27]. We have also confirmed previous findings by our laboratory and others showing that effector T cells from SLE patients are resistant to suppression by T_regs[17,19]. Moreover, we found that azathioprine-associated lymphopenia preferentially affects T_reg proliferation and suppressive capacity, although the molecular mechanisms remain elusive.

Lymphopenia in SLE patients can be caused by disease activity or by immunosuppressive drugs, such as azathioprine. To our knowledge, this is the first study to evaluate differences in T cell subsets between disease activity- and azathioprine-associated lymphopenia. We observed decreased numbers of CD4⁺CD69⁺ and CD4⁺IL-17⁺ cells in SLE patients with azathioprine-associated lymphopenia. Azathioprine functions by inhibiting Rac1 activation downstream of CD28 co-stimulation, thereby suppressing the expression of target genes such as MEK, nuclear factor-kappa B (NF-κB) and bcl-x and inducing T cell apoptosis [28].

Our data suggest that azathioprine-associated lymphopenia might involve decreased CD28 co-stimulation from apoptosis of activated autoreactive T cells. Interestingly, the number of CD4⁺IL-17⁺ cells was also decreased in patients with azathioprine-associated lymphopenia. Decreased IL-17 production might result in reduced activation of Rac1 [29], which is a molecular target of azathioprine. In contrast, disease activity-associated lymphopenia was characterized by increased numbers of CD4⁺CD69⁺ and CD4⁺IL-17⁺ cells; higher expression of activation markers such as CD69 and proinflammatory cytokines such as IL-17 might reflect persistent disease activity [30,31]. Further studies will be needed to test these hypotheses, dissect the molecular mechanisms involved, and apply these parameters to the clinical diagnosis of lymphopenia in SLE patients.

We found that HC and SLE patients without lymphopenia displayed higher numbers of CD4⁺IL-17⁺ cells in comparison to lymphopenic SLE patients. Recent studies using Il17f^rfp reporter mice [32] have shown that transfer of T helper type 17 (Th17) cells into lymphopenic hosts causes these cells to lose IL-17 expression and become converted into Th1 cells. It is tempting to speculate that lymphopenic SLE patients, regardless of the aetiology of their lymphopenia, might have decreased numbers of CD4⁺IL-17⁺ cells as a result of increased conversion of Th17 cells into Th1 cells. This subject is currently being investigated in our laboratory.

In summary, our findings suggest that diminished absolute numbers of CD4⁺CD25^high T_regs and CD4⁺CD69⁺ activated T cells are related to lymphopenia. Functionally, these defects translated to increased resistance of effector T cells from lymphopenic SLE patients to suppression by T_regs, regardless of the aetiology of their lymphopenia. In addition, azathioprine treatment correlated with decreased absolute numbers of CD4⁺CD69⁺ and CD4⁺IL-17⁺ cells and diminished suppressive capacity of T_regs in lymphopenic patients.

Globally, these defects might contribute to the altered immune responses characteristic of SLE patients and extend the potential role of lymphopenia and T cell profile shaped by lymphopenia in the maintenance of systemic autoimmunity.

Disclosure

The authors declare no conflicts of interest.