IL-7Rα^low memory CD8⁺ T cells are significantly elevated in patients with systemic lupus erythematosus

Abstract

Objective. Human effector memory (EM) CD8⁺ T cells include IL-7Rα^high and IL-7Rα^low cells with distinct cellular characteristics, including the expression of cytotoxic molecules. Both NK cells and the NK cell-associated molecule 2B4 that is expressed on CD8⁺ T cells promote cytotoxicity. Here we analysed the expression of 2B4 on IL-7Rα^high and IL-7Rα^low EM CD8⁺ T cells and its contribution to cytotoxicity. We also analysed the frequency of IL-7Rα^high and IL-7Rα^low EM CD8⁺ T cells in patients with SLE or lupus and in healthy individuals given the potential role of cytotoxic CD8⁺ T cells in the pathogenesis of lupus.

Methods. We used flow cytometry to measure the expression of 2B4 on IL-7Rα^high and IL-7Rα^low EM CD8⁺ T cells as well as the frequency of these cell populations in the peripheral blood of healthy individuals and patients with SLE. Also, 2B4-mediated cytotoxicity was quantitated in IL-7Rα^high and IL-7Rα^low EM CD8⁺ T cells using target cells with CD48 antigen.

Results. We found that IL-7Rα^high EM CD8⁺ T cells had higher levels of 2B4 expression compared with IL-7Rα^low EM CD8⁺ T cells. Triggering 2B4 enhanced the cytotoxic function of IL-7Rα^low EM CD8⁺ T cells against target cells. We also noticed that patients with SLE had an increased frequency of IL-7Rα^low EM CD8⁺ T cells that correlated with disease manifestation.

Conclusion. Our findings show that SLE patients have increased IL-7Rα^low EM CD8⁺ T cells, possibly contributing to tissue damage through 2B4-mediated cytotoxicity.

Introduction

Memory CD8⁺ T cells recognize specific antigens in target tumours and virally infected cells, and are therefore vital for host defense. Although persistent antigenic stimulation may aid the maintenance of memory T cells [1], the cytokines IL-7 and IL-15 also play an important role in the generation and maintenance of memory CD8⁺ T cells by promoting cell survival and proliferation, respectively, even in the absence of antigen [2, 3]. Memory CD8⁺ T cells can be divided into central and effector memory (EM) populations based on their capacity to migrate to secondary lymphoid tissues such as lymph nodes and spleen. We recently identified two unique subsets of human EM CD8⁺ T cells that express high and low levels of IL-7 receptor alpha chain (IL-7Rα^high and IL-7Rα^low, respectively) in peripheral blood [4]. IL-7Rα^low EM CD8⁺ T cells are largely antigen-exposed (CD27⁻CD28⁻) cells with increased expression of the cytotoxic molecule perforin compared with IL-7Rα^high EM CD8⁺ T cells. Hence the development of IL-7Rα^low cells is likely driven by repetitive antigenic immune stimulation [4]. This finding is supported by other studies that have found increased IL-7Rα^low memory CD8⁺ T cells in individuals with chronic or latent viral infections such as CMV, EBV, HCV and HIV [5–8].

SLE or lupus is an autoimmune inflammatory disease of unknown aetiology. Both genetic and environmental factors are likely involved in the pathogenesis of SLE, leading to the development of autoantibodies and inflammation [9, 10]. CD4⁺ T cells are critical drivers of the B cell-dependent autoantibody response in lupus [9, 11]. Conversely, the role of CD8⁺ T cells in the development of lupus is less understood, despite the fact that these cells comprise a significant portion of peripheral lymphocytes and play a major role in immune responses via cytotoxicity and cytokine production. Interestingly, recent studies have documented expansion of cytotoxic CD8⁺ T cells in patients with SLE, which was correlated with disease activity [12, 13]; the presence of CD8⁺ T cells in the periglomerular area was also correlated with the presence of lupus nephritis [14]. These findings suggest a possible role of CD8⁺ T cells in the pathogenesis of SLE.

The molecule 2B4 (CD244), which is associated with NK cells, belongs to the signalling lymphocyte activation molecule (SLAM/CD150) family [15, 16]. In addition to NK cells, 2B4 is expressed on other immune cells, including γδ T cells and subsets of CD8⁺ T cells. Upon binding its natural ligand, CD48, 2B4 may serve as a co-stimulatory molecule promoting NK and CD8⁺ T cell function such as degranulation, cytokine production and proliferation [17–25]. However, recent studies have also reported an inhibitory role of 2B4, suggesting that this molecule may have a dual function [26, 27]. This study explored the biological significance of 2B4 expression on IL-7Rα^high and IL-7Rα^low EM CD8⁺ T cells in human peripheral blood. Specifically, we investigated whether IL-7Rα^high and IL-7Rα^low EM CD8⁺ T cells differentially expressed 2B4, and whether it served as an activator or inhibitor of cytotoxicity. We also determined whether the frequency of such cell subsets was altered in SLE, an inflammatory condition with chronic immune stimulations similar to those observed in latent and chronic viral infections.

Materials and methods

Patients and healthy subjects

This work was approved by the institutional review board of Seoul National University Hospital and the institutional review committee of Yale University. Peripheral blood was drawn from healthy volunteers and patients with SLE after obtaining informed consent. Disease activity was evaluated using the SLEDAI score according to the criteria of the ACR, revised in 1982 [28]. Patients with SLE and healthy individuals were matched for gender (all females) and age [mean (s.d.) 35.8 (9.3) vs healthy 33.3 (7.0)]. Patients were taking the following medications: prednisone (n = 12), HCQ (n = 16), MTX (n = 5), AZA (n = 4) and MMF (n = 5).

Reagents and flow cytometry

Baf/3 cells stably expressing either green fluorescent protein (GFP, Baf/3-GFP) or human CD48 (Baf/3-CD48) (generously provided by Dr Carsten Watzl) were used as target cells in a cytotoxicity assay [15]. As previously described [4], Peripheral blood mononuclear cells (PBMCs) were purified and stained with goat anti-human IL-7Rα antibodies (Abs) (R&D Systems, Minneapolis, MN, USA), then stained with donkey anti-goat IgG Abs (Santa Cruz Biotech, Santa Cruz, CA, USA) and Abs to CD8 and CCR7 (BD Pharmingen, San Diego, CA, USA). Cells were sorted into IL-7Rα^high and IL-7Rα^low EM (CCR7⁻CD45RA^+/−) CD8⁺ T cells using a FACSAria (BD Immunocytometry Systems, San Jose, CA, USA). Some PBMCs were stained with Abs to IL-7Rα, CD8, CCR7, CD45RA, CD27, CD28, perforin, granzyme B, and 2B4 (BD Pharmingen) or isotype control, and analysed on an LSRII flow cytometer (BD Immunocytometry Systems). Collected data were analysed using FlowJo software (Tree Star, Ashland, OR, USA). In cytotoxicity and degranulation (CD107a) assays [29], sorted CD8⁺ T cells were stimulated for 2 days with anti-CD3 (clone HIT3a, 5 µg/ml, BD Pharmingen) and CD28 Abs (clone CD28.2, 2 µg/ml, BD Pharmingen) in the presence of IL-15 (5 ng/ml, R&D Systems).

Cell-mediated cytotoxicity assay

Target cells were incubated with 100 µl Na₂⁵¹CrO₄ (1 mCi/ml; PerkinElmer Life Science, Boston, MA, USA) for 1.5 h at 37°C. Then the ⁵¹Cr-labelled cells were washed twice with RPMI media containing 2% FBS. In the chromium release assay, activated CD8⁺ (IL-7Rα^high or IL-7Rα^low) T cells were incubated for 6 h with the indicated percentage of ⁵¹Cr-labelled target cells in a 96-well round-bottom plate. The ⁵¹Cr released was measured using a Packard Cobra gamma counter (GMI, Albertville, MN, USA). To block cytotoxicity, Abs to 2B4 (clone eBioPP35, 2.5 µg/ml; eBioscience, San Diego, CA, USA), CD48 (clone eBio156-4H9, 2.5 µg/ml; eBioscience) or the isotype control were added. Results were expressed as the percentage of specific lysis determined at each effector-to-target (E:T) ratio by the following formula: per cent cytotoxicity = [(mean cpm experimental release − mean cpm spontaneous release)/(mean cpm maximal release − mean cpm spontaneous release)] × 100.

Statistical analysis

All data are expressed as mean (s.d.) or mean (s.e.m.). Data were compared using the two-tailed Student’s t-test or Mann–Whitney U test. P < 0.05 was considered significant. The Pearson’s correlation analysis test was applied to SLEDAI and the frequency of IL-7Rα^low EM CD8⁺ T cells. All statistical analyses were performed using GraphPad Prism 5.01 (GraphPad Software, La Jolla, CA, USA) and SPSS 18.0 (SPSS, Chicago, IL, USA).

Results

Human IL-7Rα^low EM CD8⁺ T cells express high levels of 2B4 compared with IL-7Rα^high EM CD8⁺ T cells.

We recently identified two unique subsets of EM CD8⁺ T cells (CCR7⁻CD45RA^+/−) that express IL-7Rα^high and IL-7Rα^low in human peripheral blood [4] (Fig. 1A). Compared with IL-7Rα^high EM CD8⁺ T cells, IL-7Rα^low EM CD8⁺ T cells showed more expression of cytotoxic perforin, a molecule also expressed in terminally differentiated CD8⁺ T cells (Fig. 1B). In addition to perforin, the expression of granzyme B was higher in IL-7Rα^low EM CD8⁺ T cells than in IL-7Rα^high EM CD8⁺ T cells. Previous human and mouse studies have shown that terminally differentiated effector CD8⁺ T cells express NK-associated molecules such as CD57 and NKG2D [4, 30]. Thus we measured the expression of the NK cell-associated molecule 2B4, which promotes cytotoxicity, in IL-7Rα^high and IL-7Rα^low EM CD8⁺ T cells using flow cytometry (Fig. 1C, representative dot plot). The expression levels of 2B4 were significantly higher in IL-7Rα^low EM CD8⁺ T cells than in IL-7Rα^high EM CD8⁺ T cells [Fig. 1D, mean fluorescence intensity (MFI) (s.e.m.) 15.24 (5.61) vs 5.24 (0.91) and P = 0.002; mean % (s.e.m.) 17.73 (2.72) vs 47.74 (5.93) and P < 0.001]. Next, we determined whether TCR triggering with IL-15 would up-regulate the expression of 2B4 in purified IL-7Rα^high and IL-7Rα^low CD8⁺ T cells. IL-15 was used along with TCR triggering to restore impaired TCR-mediated proliferation of the IL-7Rα^low EM CD8⁺ T cells [31]. The combination of anti-CD3/CD28 Abs and IL-15 increased 2B4 expression in both IL-7Rα^high and IL-7Rα^low EM CD8⁺ T cells (Fig. 1E), but the levels of 2B4 in the latter remained above those in the former.

Fig. 1.

2B4 is differentially expressed in IL-7Rα^high and IL-7Rα^low EM CD8⁺ T cells in human peripheral blood.

PBMCs from healthy individuals were stained with Abs to CD8, CCR7, CD45RA, IL-7Rα and 2B4 (A, C and D), perforin, and granzyme B (B). The stained cells were analysed on a flow cytometer. (A) Representative dot plot and histogram showing naïve, CM and EM CD8⁺ T cell subsets as well as IL-7Rα expression by the individual CD8⁺ T cell subsets. (B) Representative dot plots showing the expression of perforin, granzyme B and IL-7Rα by EM CD8⁺ T cells. These data were gathered from three individuals (A and B). (C) Representative dot plot showing the expression of 2B4 and IL-7Rα by EM CD8⁺ T cells. (D) MFI of 2B4 expression by IL-7Rα^high and IL-7Rα^low EM CD8⁺ T cells (n = 7, left panel). The frequency of 2B4 positive cells by IL-7Rα^high and IL-7Rα^low EM CD8⁺ T cells in healthy individuals (n = 7) as measured by flow cytometry (right panel). (E) Increased expression of 2B4 by IL-7Rα^high and IL-7Rα^low EM CD8⁺ T cells in response to a combination of anti-CD3/CD28 Abs and IL-15. CD8⁺ T cells were sorted into IL-7Rα^high and IL-7Rα^low EM CD8⁺ T cells using a FACSAria. Sorted cells were cultured for 2 days with Abs to CD3 and CD28 (each at 5 µg/ml) in the presence of IL-15 (5 ng/ml), followed by staining with anti-2B4 or isotype Abs. Results are representative of two independent experiments. P values were obtained using the Mann–Whitney U test. Numbers in dot plots indicate the percentage of cells in each quadrant.

2B4-mediated cytotoxicity is elevated in IL-7Rα^low EM CD8⁺ T cells compared with IL-7α^high EM CD8⁺ T cells

We investigated whether the differential expression of 2B4 in IL-7Rα^high and IL-7Rα^low EM CD8⁺ T cells would have any functional implication by inducing cytotoxicity using target cells that expressed CD48 (Baf/3-CD48). The target cells were co-cultured with purified IL-7Rα^high and IL-7Rα^low EM CD8⁺ T cells that had first been stimulated with a combination of anti-CD3/CD28 Abs and IL-15. At low E:T ratios, the levels of cell lysis were higher in target cells co-cultured with IL-7Rα^low EM CD8⁺ T cells than in those co-cultured with IL-7Rα^high EM CD8⁺ T cells (Fig. 2A). However, similar levels of cell lysis were found when target cells were co-cultured with IL-7Rα^high or IL-7Rα^low EM CD8⁺ T cells at a high E:T ratio (Fig. 2A). IL-7Rα^high and IL-7Rα^low EM CD8⁺ T cells barely induced target cell lysis when they were co-cultured with Baf/3 cells expressing control GFP (Fig. 2B, IL-7Rα^high cell, data not shown). To further determine the specific role of the 2B4 and CD48 interaction in killing target cells, Baf/3-CD48 target cells were co-cultured with IL-7Rα^low EM CD8⁺ T cells in the presence of Abs to 2B4, CD48, both or an isotype control. Target cells treated with anti-2B4 or anti-CD48 Abs or both had less cell lysis than the same cells treated with the isotype control (Fig. 2C). Although the blocking effect of anti-2B4 Abs appeared to be weaker than that of anti-CD48 Abs, the combination of the two Abs synergistically increased the effect of blocking on target cell lysis. We next measured the expression of CD107a, a lysomal-associated membrane protein-1, by IL-7Rα^low EM CD8⁺ T cells because this molecule is mobilized to the cell membrane when the cytotoxic molecules perforin and granzyme B are released from cytotoxic cells [29]. IL-7Rα^low EM CD8⁺ T cells had increased CD107a expression when co-cultured with target cells expressing CD48. This was blocked by adding anti-CD48 Abs during the culture (Fig. 2D). Overall, these findings suggest that IL-7Rα^low EM CD8⁺ T cells with high levels of 2B4 expression have greater 2B4 and CD48-mediated cytotoxicity compared with IL-7Rα^high EM CD8⁺ T cells.

Fig. 2.

Enhanced cytotoxicity of activated IL-7Rα^low EM CD8⁺ T cells.

PBMCs were sorted into IL-7Rα^high and IL-7Rα^low EM (CCR7⁻CD45RA^+/−) CD8⁺ T cells using a FACSAria. To generate effector cells, sorted cells were cultured for 2 days with Abs to CD3 (5 µg/ml) and CD28 (2 µg/ml) in the presence of IL-15 (5 ng/ml). IL-7Rα^high (A and B) and IL-7Rα^low (B) EM CD8⁺ T cells were used as effector cells (E) in a 6-h chromium release assay against target cells (T) expressing human CD48 (Baf/3-CD48) (A and B) or GFP (Baf/3-GFP, control protein) (B). The percentage of specific lysis was calculated after subtracting the medium-only background. Circles and bars indicate the mean of triplicate counts. The data were compiled from five independent experiments using PBMCs from five healthy individuals. (C) The results of a chromium release assay where IL-7Rα^low EM CD8⁺ T cells (effector cells, E) were incubated with Baf/3-CD48 cells (target cells, T) at an E:T ratio of 40:1 in the presence of control Abs or Abs to 2B4, CD48, or both. Bars and error bars indicate the mean and s.d. of triplicate counts. Results are representative data from two (C) to four (A and B) independent experiments using PBMCs of five individuals. (D) Flow cytometric analysis of CD107a expression by IL-7Rα^low EM CD8⁺ T cells that were stimulated for 2 days with a combination of anti-CD3 (5 µg/ml)/CD28 (2 µg/ml) Abs and IL-15 (5 ng/ml) followed by co-culturing with target cells (Baf/3-CD48 or Baf/3-GFP) at an E:T ratio of 40:1 in the presence or absence of anti-CD48 Abs. Numbers in histograms indicate the percentage of cells that stained positive for antibody. P values were obtained using a two-tailed Student’s t-test (*P < 0.05, **P < 0.01, ***P < 0.001). The data represent two independent experiments.

Patients with SLE have an increased frequency of IL-7Rα^low EM CD8⁺ T cells that are correlated with disease activity

Although the exact role of CD8⁺ T cells in lupus pathogenesis is yet to be determined, a recent study reported the expansion of perforin-expressing CD8⁺ T cells in the peripheral blood of patients with SLE [12]. Hence we measured the frequency of IL-7Rα^low EM CD8⁺ T cells with the expression of perforin, granzyme B and 2B4 in patients with SLE and healthy controls. Similar to healthy controls, IL-7Rα^low EM CD8⁺ T cells from lupus patients expressed higher levels of perforin, granzyme B and 2B4 than IL-7Rα^high EM CD8⁺ T cells (Fig. 3A and B). The incidence of IL-7Rα^low EM CD8⁺ T cells was much higher in patients with SLE than in age- and gender-matched healthy controls (Fig. 3C) [mean frequency (%) (s.e.m.) 32.33% (3.86) and 22.13% (2.41), respectively; P = 0.027]. Furthermore, the incidence of this cell subset in patients with SLE was directly proportional to the activity of the disease as measured by SLEDAI (Fig. 3D) (r = 0.483, P = 0.026).

Fig. 3.

Increased frequency of IL-7Rα^low EM CD8⁺ T cells in patients with SLE.

(A and B) Flow cytometric analysis of 2B4, perforin, granzyme B and IL-7Rα expression by EM (CD45RA⁻CCR7^+/−) CD8⁺ T cells in a patient with SLE. Numbers in dot plots indicate the percentage of cells in each quadrant. (C) The frequency of IL-7Rα^low EM CD8⁺ T cells in patients with SLE (n = 21) and healthy individuals (HC, n = 23) as measured by flow cytometry. (D) Correlation of the frequency of IL-7Rα^low EM CD8⁺ T cells with SLE disease activity index (SLEDAI). The data were compiled from three individuals (A and B). The P value was obtained using the two-tailed Student’s t-test (C) or the Pearson’s correlation (D).

Discussion

CD8⁺ T cells have been shown to express NK-cell associated molecules that affect cell function [20, 23–25, 32–34]. We recently reported the presence of two unique subsets of EM CD8⁺ T cells with high and low levels of IL-7Rα expression in human peripheral blood [4]. To better understand the biological significance of IL-7Rα^high and IL-7Rα^low EM CD8⁺ T cells, we investigated the expression of the NK-associated molecule 2B4 in these cell subsets as well as the impact of such molecular expression on cytotoxicity. In addition, we determined the frequency of IL-7Rα^low EM CD8⁺ T cells in patients with SLE and correlated it with disease severity. The results indicate that 2B4 is differentially expressed in IL-7Rα^high and IL-7Rα^low EM CD8⁺ T cells, contributing to cytotoxicity. In addition, patients with SLE had an increased frequency of IL-7Rα^low EM CD8⁺ T cells that correlated with disease activity compared with healthy controls. These findings provide new insight into understanding the function and biological significance of IL-7Rα^low EM CD8⁺ T cells in health and disease.

The molecule 2B4 is expressed in NK cells and subsets of CD8⁺ T cells such as memory and activated cells [35]. We found that IL-7Rα^low EM CD8⁺ T cells had increased expression of 2B4 compared with IL-7Rα^high EM CD8⁺ T cells. IL-7Rα^low EM CD8⁺ T cells are known to have terminally differentiated phenotypes with decreased CD27 and CD28 expression and limited TCR repertoires [4]. These findings suggest that the expression of 2B4 could be affected by the differentiation status of memory CD8⁺ T cells. In fact, we noted up-regulation of 2B4 in both IL-7Rα^high and IL-7Rα^low EM CD8⁺ T cells upon TCR and IL-15 stimulation. In general, triggering 2B4 augments cytotoxicity and IFN-γ production in NK cells and CD8⁺ T cells [20–25], although high levels of 2B4 expression could function as an inhibitory receptor [26, 27]. We noticed that 2B4, which is expressed at different levels by IL-7Rα^high and IL-7Rα^low EM CD8⁺ T cells, up-regulated cytotoxicity rather than inhibiting it. Upon antigenic triggering of 2B4 with CD48, IL-7Rα^low EM CD8⁺ T cells showed increased cytotoxicity than IL-7Rα^high EM CD8⁺ T cells. This finding could be secondary in part to the differential expression of 2B4 and cytotoxic molecules (e.g. perforin and granzyme B) by the two cell subsets.

CD8⁺ T cells, possessing potent cytotoxicity, play a pathogenic role in organ-specific autoimmune diseases such as type I diabetes mellitus [36, 37]. However, the role of CD8⁺ T cells in the pathogenesis or regulation of SLE is less understood. CD8⁺ T cell depletion prevents or mitigates lupus in experimental animals [38–40], suggesting that CD8⁺ T cells contribute to the pathogenesis of the disease. This notion is supported by human studies that have shown the expansion of memory CD8⁺ T cells in the peripheral blood of patients with SLE [12, 13]. In particular, the expansion of perforin- and/or granzyme B-positive CD8⁺ T cells has been found in patients with SLE, correlating with disease activity [12]. This observation is consistent with our results showing the expansion of IL-7Rα^low EM CD8⁺ T cells with the expression of perforin and granzyme B as well as the correlation of such cell expansion with the degree of disease. The expansion of cytotoxic CD8⁺ T cells causes tissue damage which could possibly lead to increased production of soluble nucleosomes, target antigenic structures in lupus, from target cells [12]. Interestingly, IL-7Rα^low EM CD8⁺ T cells had increased expression of 2B4, which could promote their cytotoxicity. This raises the possibility of enhanced tissue damage in lupus by 2B4 expressed by IL-7Rα^low EM CD8⁺ T cells through an interaction with CD48 expressed by other cells including lymphocytes, macrophages, dendritic cells and epithelial cells [15, 16].

In conclusion, we found that the NK cell-associated molecule 2B4 is differentially expressed in IL-7Rα^high and IL-7Rα^low EM CD8⁺ T cells in human peripheral blood, and that it activates cytotoxicity. We also noticed that patients with SLE had an increased frequency of IL-7Rα^low EM CD8⁺ T cells that correlated with disease activity. Our findings suggest the possible involvement of 2B4 in SLE-associated tissue damage by the expansion of IL-7Rα^low EM CD8⁺ T cells.