Apart from NK cells, TCRγδ and CD8+ T cells, killer cell Ig-like receptor (KIR) expression was described on a minor subset of CD4+ T cells. However, their functions remain to be elucidated in this latter lymphocyte population. We demonstrated that KIR2DL2/L3 (CD158b) and KIR2DS2 (CD158j) transcripts were synthesized by sorted CD4+CD158b/j+ T cells obtained from healthy individuals. In contrast, we observed that only the inhibitory or activating receptor was expressed at the cell surface according to the donor tested. In CD158b-expressing cells, KIR triggering leads to an inhibition of the CD3-induced cell proliferation and Erk activation, and the receptor exhibits an activation-dependent tyrosine phosphorylation and association with the Src homology 2-containing phosphatase 1. In CD158j-positive cells, KIR-engagement results in an enhanced CD3-mediated cell growth and Erk phosphorylation. Our results suggested that, in contrast to NK cells, the functions of KIR in CD4+ T lymphocytes might derive from a selective expression of their activating or inhibiting forms.
The killer cell Ig-like receptors (KIRs)4 belong to a family of structurally related cell membrane receptors that recognize MHC class I molecules. Although KIRs are mainly expressed by functional circulating NK cells, they were also found on CD8+ and TCRγδ T lymphocytes and on a marginal CD4+ T cell population (1, 2, 3). KIR molecules were classified according to their structure and functional properties into inhibitory and activating forms. Inhibitory receptors (KIR-L, for KIR “long”) are characterized by a long cytoplasmic tail where consensus ITIMs are located. Once phosphorylated, the latter allow the recruitment of downstream phosphatases, leading to the general shutdown of NK cell effector functions. In contrast, activating KIRs (KIR-S for “KIR short”) encompass a short cytoplasmic tail and mainly use the ITAM-containing adaptor molecule DAP12 to deliver activating signals. For a given receptor, both forms are usually expressed on NK cells and the balance between their respective signaling pathways then determines the ability of NK cells to lyse specific targets (4).
The expression of NK receptors on effector/memory CD4+ T cells was essentially described under chronic inflammatory conditions. Indeed, it has been reported that clonally expanded CD4+CD28null T cells from patients with rheumatoid arthritis (RA) expressed KIRs (5), with a preferential expression of the activating receptor KIR2DS2/CD158j (6). In these cells, which do not express DAP12, CD158j was found to selectively activate the DAP12-independent JNK signaling pathway, and was therefore thought to contribute to the T cell-mediated autoimmune disease (7). Similarly, KIR expression was reported on CD4+CD28null T cells from patients with acute coronary syndromes (ACS), with again a preference for the stimulatory variant CD158j (8). However, DAP12 was frequently transcribed by the ACS patient-derived clones, and CD158j-mediated cytotoxicity was exclusively observed in T cell clones that expressed both CD158j and DAP12. More recently, we identified KIR3DL2/CD158k as a phenotypic marker of Sézary cells (9). We further reported the expression of CD158a and CD158b on the circulating CD4+ malignant cells isolated from one Sézary syndrome (SS) patient. In this case, both KIRs were found to display a coactivatory function through the recruitment of the JNK-dependent signaling pathway, although the receptors were efficiently expressed under their activating and inhibitory forms at the cell surface (10).
The expression of KIRs on CD4+ T lymphocytes in healthy individuals was more controversial. It has been initially reported that among the CD3+KIR+ cells from healthy donors, 7 ± 3% expressed CD4 (11). However, a recent report by van Bergen et al. established that in healthy individuals an average of 0.2% of CD4+ T lymphocytes was stained with anti-KIR mAbs (3). In addition, KIR expression was first thought to be restricted to the CD4+CD28− lymphocyte subset (12) until it was shown that the CD4+CD28+ cells displayed a broader repertoire of KIR than the CD4+CD28− subpopulation (3). Finally, while the presence of KIR on CD4+ T lymphocyte subsets was better documented, the precise function of these receptors remained to be established. In this study we provide data regarding CD158b1/b2 (KIR2DL2/KIR2DL3) and CD158j (KIR2DS2) cell surface expression and function in normal CD4+ T lymphocytes.
Materials and Methods
FITC-conjugated anti-CD4 (clone 13B8.2) and PE-coupled anti-CD158b/j (clone GL183) and anti-CD28 (clone28.2) mAbs, purified anti-CD16 (clone 3G8) mAb, and their corresponding isotype-matched control mAbs were from Beckman Coulter. The specific anti-KIR2DL3/CD158j mAb (ECM41; IgM) was kindly provided by Prof. A. Moretta (University of Genoa, Genoa, Italy), while anti-CD3 mAb (ANA 3, IgG1) was produced locally.
Cell sorting and amplification
PBMC were isolated from heparinized venous blood from healthy volunteers by density gradient centrifugation. To isolate the CD4+CD158b/j+ subpopulation, cells were simultaneously stained with FITC-conjugated anti-CD4 mAb and PE-coupled anti-CD158b/j (clone GL183) mAb for 30 min at 4°C. CD4+GL183+ cells were further sorted using a MoFlo cell sorter (DakoCytomation). Cells were cultured in RPMI 1640 medium, supplemented with penicillin (100 IU/ml), streptomycin (100 μg/ml), 2 mM l-glutamine (all from Invitrogen Life Technologies), 10% heat-inactivated human serum (Institute of Biotechnology, Reims, France), and 200 IU/ml rIL-2 (a gift of Sanofi-Aventis). Mixed irradiated allogenic PBMC plus 2 μg/ml PHA (Sigma-Aldrich) were added at the start of the culture and subsequently at 7- to 10-day intervals.
Cells were stained with PE-labeled anti-CD158b/j or FITC-conjugated anti-CD4 or anti-CD28 mAbs using a standard protocol. Alternatively, anti-KIR2DL3 mAb was used followed by goat anti-mouse IgM PE-labeled Ig. Appropriate isotype-matched mAbs were used as controls. Cell analyses were performed on an Epics flow cytometer (Beckman Coulter).
Ninety six-well plates were coated with suboptimal to optimal concentrations of anti-CD3 together with anti-CD16 (as isotype-matched control) or anti-CD158b/j mAbs. Cells (7.5 × 104 in 200 μl of culture medium) were cultured for 4 days and pulsed with 1 μCi of [3H]TdR for an additional 16 h. Incorporation of [3H]TdR was measured in a liquid scintillation counter (Topcount, Packard Instruments). Results were expressed as the mean of triplicates ± SD.
KIR transcripts analysis
Total mRNA were extracted from 5 × 106 cells using the TRIzol reagent according to the manufacturer’s instructions (Invitrogen Life Technologies). Reverse transcription was performed using an oligo-dT(12–18) primer (Invitrogen Life Technologies) and the PowerScript reverse transcriptase (BD Clontech). Specific primers for the amplification of KIR2DL2, KIR2DL3, and KIR2DS2 cDNA, as well as amplification conditions, were as described previously (13). β-actin cDNA amplification was performed in parallel as an internal control.
Cell surface biotinylation, immunoprecipitation, and immunoblotting
To detect KIR surface expression, cells were biotinylated with sulfosuccinimidyl-2-(biotinamido)-ethyl-1,3-dithiopropionate (sulfo-NHS-LC-biotin) before lysis (Perbio Sciences) as previously reported (14). After washes, cells (2 × 107 per sample) were either left untreated or incubated with 1 mM sodium vanadate and 0.03% H2O2 for 2 min at room temperature. Cells were washed and resuspended in 1% Nonidet P-40 lysis buffer (20 mM Tris-HCl (pH 7.5), 150 mM NaCl, 1% Nonidet P-40, 1 mM sodium vanadate, 10 mM NaF, 1 mM PMSF, 1 μg/ml aprotinin, and 1 μg/ml leupeptin) for 1h at 4°C. After a preclearing step on protein A-Sepharose beads, lysates were subjected to immunoprecipitation with 5 μg of control mIgG1 or anti-CD158b/j GL183 mAb. Immunoprecipitates were collected on protein G-Sepharose beads, separated by SDS-PAGE, and transferred onto a nitrocellulose membrane. To detect biotinylated proteins, membranes were directly probed with streptavidin-peroxidase and revealed with an ECL system (Perbio Sciences). Alternatively, blots were probed in a series with an anti-phosphotyrosine mAb (clone 4G10; Upstate Biotechnology) and anti-Src homology 2-containing phosphatase 1 (SHP-1) Abs (Cell Signaling Technology), or anti-CD3ζ, anti-DAP10, and anti-DAP12 Abs (all from Santa Cruz Biotechnology) followed by the appropriate HRP-conjugated secondary Abs (Jackson ImmunoResearch Laboratories/Beckman Coulter).
For the detection of Erk phosphorylation, CD4+GL183+ T cells were incubated with control mouse IgG1, anti-CD3, or GL183 mAb alone or in combination and then cross-linked with rabbit anti-mouse IgG Abs (Jackson ImmunoResearch Laboratories). Activation was carried out for 20 min at 37°C. Cells were harvested immediately after stimulation and lysed. Following SDS-PAGE and transfer to nitrocellulose membrane, the blot was probed with an anti-phosphorylated Erk1/2 mAb (Cell Signaling Technology), stripped, and reprobed with an anti-Erk1/2 Ab (Cell Signaling Technology). Quantification of a phosphorylated Erk immunoblot was performed by densitometry analysis and a relative densitometric intensity value was assigned to each band, the relative densitometric intensity of 1 being assigned to the maximal density value obtained under optimal CD3-mediated activation.
Results and Discussion
GL183 mAb triggers activating or inhibitory signals on CD4+GL183+ T lymphocytes from healthy individuals
By performing initial KIR expression analysis on sorted CD4+ T lymphocytes from healthy individuals, we observed a preferential cell staining with the anti-CD158b/j mAb GL183, as reported elsewhere (Ref. 3 and data not shown). To further establish the function of CD158b/j on normal CD4+ T cells, the CD4+GL183+ cell subset, which represented 0.05–0.2% of the circulating CD4+ T lymphocytes (not shown), was isolated from the blood of six healthy donors and the whole double-positive population from each individual was further expanded in vitro for no longer than 4 wk. Immunostaining analyses confirmed that the sorted and amplified cells stably expressed CD4 and CD158b/j, as exemplified by the data obtained on cells arising from two representative donors (Fig. 1). Furthermore, among the CD4+GL183+ population, CD28 expression seemed to vary from one donor to another. Thus, the amplified CD4+GL183+ cell subset may consist of a mix of CD28+ and CD28− cells or exclusively in CD28dim cells (Fig. 1). Finally, we ensured that the in vitro derived cells were representative of the entire circulating CD4+CD158b/j+ T lymphocyte population and therefore did not correspond to a clonal cell expansion by performing cell staining analyses using specific anti-TCRVβ mAbs. This led to the detection of several subclones exhibiting a distinct TCRVβ rearrangement for all donors tested (data not shown).
The influence of CD158b/j on the proliferation of CD4+GL183+ T lymphocytes was then evaluated for each individual. As demonstrated by the results obtained from two representative donors, the in vitro amplified CD4+CD158b/j+ cells showed a dose-dependent proliferation upon CD3 triggering (Fig. 2,A). However, we observed a significant but opposite effect of CD158b/j engagement using GL183 mAb on the CD3-mediated proliferation of the cells tested. Thus, a dramatic inhibition of the CD3-induced cell proliferation was detected upon CD158b/j coligation on sorted cells corresponding to one set of donors (n = 2 of 6; Fig. 2,A, group 1). In contrast, CD158b/j triggering resulted in a significant increase in the CD3-mediated cell growth of the CD4+GL183+ subpopulation isolated from a second set of individuals (n = 4 of 6; Fig. 2 A, group 2). Thus, depending on the donor, GL183-mediated KIR targeting up-regulated or down-regulated the CD3-mediated proliferation of CD4+GL183+ T lymphocytes.
Selective expression of CD158b and/or CD158j on CD4+ T lymphocytes
As a result of the high sequence identity existing within the extracellular domain of the inhibitory (KIR-L) and the activating (KIR-S) forms of a given KIR, the commercially available anti-KIR mAb usually recognize both structures. Thus, the generic anti-CD158b/j mAb GL183 efficiently targets KIR2DL2 (CD158b1), KIR2DL3 (CD158b2), and KIR2DS2 (CD158j) receptors. To establish whether the detection of an activating or inhibitory function of CD158b/j on the CD3-induced cell proliferation was related to a donor-dependent expression of the KIR-L or KIR-S molecules, the analysis of their mRNA synthesis was first performed. A similar, but distinct, pattern of CD158b/j transcript synthesis was observed according to the donor group tested. Indeed, KIR2DL2, KIR2DS2, and KIR2DL3 cDNA were detected in the CD4+GL183+ T cells isolated from a representative donor belonging to group 1 (n = 2; Fig. 2,B). Moreover, only KIR2DL2 and KIR2DS2 transcripts were amplified from the cells corresponding to a donor representative of group 2 (n = 4; Fig. 2 B). These results established that, regardless of the donor, the in vitro derived cells possessed the transcriptional machinery for the concomitant expression of an inhibitory (CD158b1 and/or b2) and an activating (CD158j) KIR.
To assess which of the three KIRs was efficiently expressed at the surface of CD4+GL183+ T lymphocytes, cells were surface biotinylated and further subjected to an immunoprecipitation procedure using GL183 mAb. The immunoprecipitates obtained on cells corresponding to individuals from group 1 revealed an exclusive detection of inhibitory molecule(s) presenting an apparent molecular mass of 58 kDa (Fig. 2,C, upper panel, group 1). Further flow cytometry analyses performed with a specific anti-KIR2DL3/CD158b2 mAb (15) showed that this inhibitory receptor was expressed at the cell surface (Fig. 2,C, bottom panel). However, the lack of a specific anti-KIR2DL2 mAb did not allow us to establish whether detection of the KIR2DL2/CD158b1 transcript also corresponded to protein expression. In contrast, in CD4+GL183+ T cells corresponding to donors from group 2 only the activating KIR2DS2 receptor was efficiently immunoprecipitated by GL183 mAb, as assessed by the detection of a unique 50-kDa protein (Fig. 2 C, group 2). These data demonstrated that, according to the donor and despite the simultaneous synthesis of both KIR-L and KIR-S mRNA, either an inhibitory CD158b or an activating CD158j receptor is expressed at the cell surface of CD4+GL183+ T lymphocytes. This therefore distinguishes CD4+KIR+ T cells from NK cells, because in the latter cell population a parallel synthesis of the KIR-L and KIR-S forms was usually reported for a given KIR. One can further hypothesize that in CD4+ T cells the function of a KIR might be driven by the selective and regulated expression of its KIR-L or KIR-S form rather than by a differential affinity of each receptor form for a common ligand, as previously suggested for KIR2DS1/L1 and KIR2DS2/L2 (16).
Analysis of CD158b- and CD158j-mediated signaling pathways when expressed in CD4+ T lymphocytes
To definitely demonstrate that the exclusive expression of either KIR2DL2/L3 or KIR2DS2 dictates the KIR function in CD4+ T lymphocytes, CD158b- or CD158j-dependent signaling pathways were analyzed. Inhibitory KIRs were shown to exert their activity through the phosphorylation of their intracellular ITIM(s), which subsequently allows their interaction with the phosphatase SHP-1 (17). We therefore investigated the tyrosine phosphorylation status of CD158b in CD4+GL183+ T cells isolated from a healthy donor classified in group 1. To this end, cells were left untreated or incubated with vanadate and subjected to immunoprecipitation using GL183 mAb. Immunoblot analysis of the anti-CD158b precipitates with an anti-phosphotyrosine mAb allowed the detection of an activation-dependent phosphorylation of the KIR-L isoforms (Fig. 3,A, left panel, group 1). Consequently, an interaction of the phosphorylated receptor with the phosphatase SHP-1 was evidenced in vanadate-treated cells (Fig. 3,A, left panel, group 1). A similar experiment was performed on CD4+ T cells expressing CD158j (Fig. 3,A, right panel, group 2). Because stimulatory receptors were shown to initiate intracellular signals through their association with ITAM-containing adaptor molecules, among them DAP10, DAP12, and CD3ζ (17), a possible interaction of CD158j with one of these molecules was investigated. However, neither DAP10 nor DAP12 was found expressed in CD4+ T cells or detected in CD158j immunoprecipitates (Fig. 3 A, right panel). Furthermore, no coprecipitation of CD3ζ with CD158j was observed following cell activation. Thus, it seems that in KIR2DS2-expressing cells the activating function of the receptor relies on its interaction with an unusual adaptor molecule, the identity of which remains to be defined. Interestingly, a similar observation was made in CD4+ T cell clones isolated from patients with RA or SS. Thus, in these diseases a KIR2DS2-mediated DAP12-independent cell activation was also described (8, 10).
To ascertain that the exclusive expression of a KIR-L or KIR-S molecule functionally influences the CD4+ T cell activation process, the effect of CD158b or CD158j engagement on the Erk signaling pathway was investigated. To this aim, CD4+ T cells were stimulated with control mouse IgG1, anti-CD3, anti-CD158b/j mAb, or a combination of anti-CD3 and anti-CD158b/j mAb. The corresponding cell lysates were analyzed by Western blotting for the detection of phosphorylated Erk (Fig. 3,B). Note that the concentration of anti-CD3 mAb used for activation (1 and 0.2 μg/ml for group 1 and group 2 donors, respectively) was the one previously showed to allow the detection of a maximal effect of KIR engagement on CD3-mediated cell proliferation (see Fig. 2,A). In KIR2DL2/L3-expressing cells, the crosslinking of CD3 with an optimal concentration of mAb (1 μg/ml) together with the KIR allowed the detection of phosphorylated Erk, the phosphorylation level being lower than the one reached upon optimal CD3-mediated activation (Fig. 3,B, group 1). In addition, the use of a suboptimal concentration of anti-CD3 mAb (0.2 μg/ml) on KIR2DS2-positive cells resulted in a discrete Erk activation that is increased in the presence of the GL183 mAb (Fig. 3 B, group 2). However, in these cells the triggering of KIR2DS2/CD158j alone was not sufficient by itself to promote Erk activation, as previously reported (18). Thus, in normal CD4+ T lymphocytes CD158b or CD158j fulfills a coreceptor function and efficiently modulates the CD3-mediated Erk activation when engaged.
In conclusion, we showed that KIRs are expressed as functional coreceptors in CD4+ T lymphocytes from healthy individuals. We further demonstrated that, in this lymphocyte population, the detection of an inhibitory or activating KIR function might correspond to an exclusive and regulated expression of a KIR-L or KIR-S form. In this context, the nature of the delivered KIR-mediated signal would not relay on a commonly proposed model involving the receptors binding affinities to a common ligand (16) but rather reflects the structural features of the expressed molecule. In our study, the overall CD4+KIR+ T cell subset isolated from each donor, while not representing a clonal cell population, was found to homogeneously express a single form of the receptor and consequently to uniformly answer to KIR triggering in terms of cell proliferation and Erk activation. One can therefore postulate that the preferential expression of an activating KIR on CD4+ T lymphocytes might render a single individual more susceptible to inappropriate stimuli leading to the proliferation and growth escape of these cells, as observed in patients suffering from RA, ACS, or SS.
The authors have no financial conflict of interest.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
This work was supported by grants from the Institut National de la Santé et de la Recherche Médicale and Paris XII University.
Abbreviations used in this paper: KIR, killer-cell immunoglobulin like receptor; ACS, acute coronary syndrome; KIR-L, KIR (long) inhibitory form; KIR-S, KIR (short) activating form; RA, rheumatoid arthritis; SHP-1, Src homology 2-containing phosphatase 1; SS, Sézary syndrome.