HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Snyder, M. R. Articles by Goronzy, J. J. Search for Related Content PubMed PubMed Citation Articles by Snyder, M. R. Articles by Goronzy, J. J.

Formation of the Killer Ig-Like Receptor Repertoire on CD4⁺CD28^null T Cells¹

Melissa R. Snyder^*, Lars-Olof Muegge^*, Chetan Offord, William M. O’Fallon, Zeljko Bajzer, Cornelia M. Weyand^* and Jörg J. Goronzy^2,*

Departments of ^* Medicine/Rheumatology and Immunology, Biochemistry and Molecular Biology, and Biostatistics, Mayo Clinic, Rochester, MN 55905

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
Killer Ig-like receptors (KIRs) are expressed on CD4⁺CD28^null T cells, a highly oligoclonal subset of T cells that is expanded in patients with rheumatoid arthritis. It is unclear at what stage of development these T cells acquire KIR expression. To determine whether KIR expression is a consequence of clonal expansion and replicative senescence, multiple CD4⁺CD28^null T cell clones expressing the in vivo dominant TCR -chain sequences were identified in three patients and analyzed for their KIR gene expression pattern. Based on sharing of TCR sequences, the clones were grouped into five clone families. The repertoire of KIRs was diverse, even within each clone family; however, the gene expression was not random. Three particular receptors, KIR2DS2, KIR2DL2, and KIR3DL2, had significant differences in gene expression frequencies between the clone families. These data suggest that KIRs are successively acquired after TCR rearrangement, with each clone family developing a dominant expression pattern. The patterns did not segregate with the individual from whom the clones were derived, indicating that peripheral selection in the host environment was not a major shaping force. Several models were examined using a computer algorithm that was designed to simulate the expression of KIRs at various times during T cell proliferation. The computer simulations favored a model in which KIR gene expression is inducible for a limited time during the initial stages of clonal expansion.

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
Rheumatoid arthritis (RA)³ is a chronic inflammatory disease characterized by the presence of mononuclear infiltrates in synovial tissue. Clinical manifestations are not limited to the joint, and it has been suggested that RA is a systemic immune disorder. Recent studies have identified abnormalities in the adaptive immune system of patients with RA that are highly reminiscent of the aging immune system (1). The TCR repertoire in patients with RA is markedly contracted; the contraction includes the memory as well as the naive compartments of CD4 T cells (2). The mechanism underlying the repertoire contraction appears to be related to a premature cessation of thymic function. In patients with RA, the number of recent thymic immigrants carrying TCR excision circle episomes is significantly decreased (3). Concurrently, the frequencies of CD4⁺CD28^null T cells expressing TCR - dimers is increased (4). CD4⁺CD28^null T cells are functionally distinct from normal CD4 T cells in that they secrete large amounts of IFN-, are resistant to apoptosis, and express perforin and granzyme B, which conveys cytotoxic capability (5, 6, 7, 8). Additionally, they express a variety of receptors belonging to the killer Ig-like receptor (KIR) family (9, 10).

The regulation of KIR expression on CD4 T cells is unclear. KIR expression on NK cells is dependent on IL-15 and starts after lineage commitment (11). The murine genome does not contain a gene family homologous to the human KIR cluster, but a sequential expression model has been suggested for the functionally similar Ly49 family (12). Under the proper conditions, murine NK1.1⁺Ly49A⁺Ly49G2/C/F/I/^- NK cells are capable of initiating expression of Ly49F, Ly49G2, and Ly49C/I, as well as maintaining expression of Ly49A. Under similar conditions, NK1.1⁺Ly49G2⁺Ly49A/C/F/I/^- NK cells were able to induce expression of Ly49C/I, although they were unable to induce expression of either Ly49A or Ly49F (13).

The close resemblance of CD4⁺CD28^null T cells to NK-T cells may suggest that they represent a separate lineage that has been expanded to compensate for thymic dysfunction. NK-T cells are a subset of T cells that express the NK receptor CD161 in addition to the TCR and that are reactive to microbial glycolipids presented by the nonpolymorphic CD1d molecule (14, 15, 16). The TCR repertoire of these cells is highly restricted, typically involving rearrangement of the AV24-AJ18 gene segments. In contrast to classic NK-T cells, CD4⁺CD28^null T cells do not express the conserved TCR that is characteristic of CD1d-restricted NK-T cells (9, 17). In the alternative model, expression of KIR molecules may be acquired by normal T cells during replicative senescence. Indeed, all CD4⁺CD28^null T cells express the memory phenotype, and TCR studies in RA have shown that this population of T cells is dominated by relatively few clonally expanded T cells (17).

The principle focuses of the current study were to understand how the KIR repertoire of the CD4⁺CD28^null T cell populations in RA is formed and whether KIR expression occurs early in T cell development or during peripheral clonal expansion. To address these, we obtained multiple members of CD4 clone families sharing identical TCR -chain sequences, determined the diversity of KIR gene expression, and compared the results with different models of KIR acquisition.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
Study population

PBMCs from patients with RA were isolated using Ficoll-Hypaque (Amersham Pharmacia Biotech, Arlington Heights, IL) density centrifugation. CD4⁺CD28^null T cell frequency was determined by flow cytometry using FITC-conjugated anti-CD4 and PE-conjugated anti-CD28 mAb (BD Biosciences, San Jose, CA). The protocol was approved by the Mayo Clinic Institutional Review Board (Rochester, MN) and all people gave written, informed consent. Three patients who displayed expanded CD4⁺CD28^null T cell populations were selected. TCR--chain variable region (BV) gene segment usage in the CD4⁺CD28⁺ and CD4⁺CD28^null populations was analyzed by three-color flow cytometry using FITC-conjugated BV-specific mAbs (anti-BV1, anti-BV2, anti-BV3, anti-BV11, anti-BV13S1, anti-BV13S6, anti-BV14, anti-BV16, anti-BV17, anti-BV18, anti-BV20, anti-BV21S3, anti-BV22 (Beckman Coulter, Miami, FL), anti-BV5S1, anti-BV5S2/3, anti-BV6S7, anti-BV8, anti-BV12 (Endogen, Cambridge, MA), and anti-BV7 (kindly provided by A. W. Boylston, St. James’ University Hospital, Leeds, U.K.)), PE-conjugated anti-CD28 mAb, and PerCP-conjugated anti-CD4 mAb (BD Biosciences). TCR-BV elements that were overrepresented in the CD4⁺CD28^null population were identified.

HLA-C and KIR genotyping

Genomic DNA was isolated using the DNA Isolation Kit for Mammalian Blood (Roche Molecular Biochemicals/Boehringer Mannheim, Indianapolis, IN). KIR genotypes were analyzed by PCR using KIR-specific primers (18). HLA-C alleles were analyzed by PCR using the Dynal HLA-C "Low Resolution" SSP kit (Dynal, Lake Success, NY) according to the manufacturer’s instructions.

Cloning of CD4⁺CD28^null T cells

CD4⁺CD28^null T cells were sorted from PBMCs by FACS using FITC-conjugated anti-CD4 and PE-conjugated anti-CD28 mAb (BD Biosciences). cDNA was amplified with the appropriate TCR-BV/-chain constant region-specific primer pairs corresponding to the overrepresented populations previously determined by flow cytometry. Amplified products were radiolabeled and separated on 5% polyacrylamide gels. Dominant bands were eluted and directly sequenced (4). In parallel, the sorted cells were cloned using limiting dilution. The clones were maintained on 1.5 x 10⁵/ml irradiated EBV-transformed lymphoblastoid cells (treated with galactose oxidase/neuraminidase to enhance cellular interactions) (19), 25 ng/ml anti-CD3 mAb (Orthoclone OKT-3; Ortho Diagnostics, Raritan, NJ), and 50 U/ml recombinant human IL-2 (Proleukin; Chiron, Emeryville, CA).

TCR -chain sequence analysis and KIR phenotyping

CD4⁺CD28^null T cell clones were harvested and total RNA was extracted using TRIzol reagent (Invitrogen, Carlsbad, CA). cDNA was synthesized using an oligo(dT) primer. Clones were screened for expression of in vivo expanded TCR sequences by PCR amplification with the appropriate BV and -chain constant region gene segment primers and TCR sequencing. The clones from each individual were grouped into clone families based on the expression of identical TCR -chains. The KIR repertoire for each clone was analyzed by RT-PCR using KIR-specific primers described by Uhrberg et al. (18). PBMCs from other individuals expressing the appropriate KIRs were used as the positive controls.

Cytokine stimulation

Resting CD4⁺CD28^null T cell clones were stimulated with either 50 ng/ml OKT-3, 10 ng/ml IL-12 (R&D Systems, Minneapolis, MN), 50 ng/ml IL-15 (R&D Systems), 50 ng/ml IL-18 (PeproTech, Rocky Hill, NJ), or a combination of 10 ng/ml IL-12 and 50 ng/ml IL-18 over the course of seven days. At the indicated times, 1.0 x 10⁶ cells were harvested for RNA isolation and analyzed for expression of individual KIR transcripts.

Statistical analyses

Descriptive statistics were used to determine whether the five clone families could be distinguished based on the probability of expression of KIR2DL2, KIR2DS2, and KIR3DL2. Due to the small sample size, formal comparisons using all five groups were underpowered. The five clone families were partitioned into two larger groups, and univariate associations with the KIR2DL2, KIR2DS2, and KIR3DL2 were performed using Fisher’s exact test.

Computer modeling

The proliferation of CD4⁺CD28^null T cells was simulated as a stochastic branching process under the following assumptions: 1) simple cell division without cell death; 2) within each generation, there is a probability p, which is the same for all clone families, that a cell gains individual KIR expression; 3) in each generation, the cells are stochastically independent regarding the likelihood of KIR expression; and 4) KIR expression, once acquired, is not lost with further cell divisions.

Under these assumptions, the theoretical, or expected, proportion p of cells in the population with KIR expression after k generations at risk for gaining KIR expression is given by P = 1- (1-p)^k. This can be defined as the probability that a cell, drawn at random from the population of cells after the k generation, will demonstrate KIR expression.

Data to estimate P come from several small samples of cells (see Tables I and III). From these data, P is estimated as the ratio of the number of cells for which KIRs are observed divided by the total number of cells. This is defined as . Then, = 1-(1–p)^k, so = 1- (1-)^1/k.

View this table: [in this window] [in a new window]   Table I. TCR -chain sequences of clonally expanded T cells

For each of these models, a population corresponding to the last generation was simulated. Then, 20,000 samples of the sizes obtained in the experimental work were drawn from each simulated population. Thus, probability distributions for each model and sample size combination were obtained. These distributions were then used to calculate the likelihood function for each observed data set and appropriate models were compared according to the logs of the corresponding likelihood functions (20).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
Diverse repertoire of KIR gene expression in CD4 T cell clones in RA patients

Three patients with RA, who possessed expanded frequencies of CD4⁺CD28^null T cells (4.9, 29.0, and 34.6% for donors A, B, and C, respectively) were chosen for this study. To identify dominant clonal populations in the subset of CD4⁺CD28^null T cells, the BV gene segment usage in CD4⁺CD28⁺ T cells and CD4⁺CD28^null T cells were compared. Results from flow cytometric analysis using 19 BV-specific mAbs are shown in Fig. 1. In all three patients, the population of CD4⁺CD28⁺ cells displayed a diverse repertoire of TCR-BV gene segment usage, with each BV gene segment being represented in 1–10% of the total CD4⁺ population. Compared with the CD4⁺CD28⁺ population, T cells expressing particular BV gene segments were expanded in the CD4⁺CD28^null population and gained dominance. In donor A, T cells expressing BV14 and BV17 were expanded; donor B showed a dominance of BV13S1 and, to a lesser degree, BV18; donor C had an increased frequency of CD4⁺CD28^null T cells expressing BV18 and BV22. Sequence analysis of TCR -chain gene sequences of these expanded populations yielded dominant clonal sequences (Table I).

View larger version (18K): [in this window] [in a new window]   FIGURE 1. Frequency of BV gene segments on CD4+ T cells. The TCR -chain repertoire was analyzed using BV-specific mAbs and flow cytometry. The BV gene segment frequencies were analyzed in three patients with RA on the CD4+CD28+ () and the CD4+CD28null () T cell populations. Several BV gene segments were found to be overrepresented on the CD4+CD28null T cells. TCR -chain spectrotyping and sequence analysis demonstrated dominant clonal sequences in these families (Table I).

All three patients were characterized for their genomic representation of KIR genes. Results of the KIR and HLA-C genotyping are shown in Table II. Donor B lacked KIR2DS1, donor C was negative for KIR2DS3 and KIR3DS1, and donor A possessed the gene for each of the different isotypes.

View larger version (40K): [in this window] [in a new window]   FIGURE 2. KIR expression in CD4+CD28null T cell clone families. T cell clones expressing identical TCR -chains were analyzed by RT-PCR for expression of 11 KIRs. Results are shown for members of two clone families: A, Clone family B1 from donor B; and B, Clone family C2 from donor C. cDNA from PBMCs of individuals expressing the appropriate KIR gene was used as the positive control for each receptor.

View larger version (40K): [in this window] [in a new window]   FIGURE 3. Frequency of KIR expression in five clone families obtained from three people with RA. KIR expression was analyzed in a total of 38 clones from five clone families using RT-PCR. The frequency is expressed as the percentage of clones within a family that expressed each individual receptor.

The repertoire analysis of KIR gene expression was done on established T cell clones; therefore, it was important to consider whether the observed diversity was generated during in vitro culture. Several clones were randomly chosen, and the KIR transcription pattern was analyzed at several time points during culture. At each time point, the clones displayed identical patterns of gene expression (data not shown). To analyze whether cytokines or TCR-mediated activation could regulate the transcription of the KIRs in CD4⁺CD28^null T cells, a T cell clone that transcribed mRNA for KIR2DL4, KIR2DL3, and KIR3DL2 was selected and cultured with selected cytokines. Previous results have shown that the expression of C-type lectin receptors on NK cells and CD8 T cells can be regulated in vitro by IL-15 (22). IL-15 also induces KIR expression on progenitor cells during in vitro NK cell differentiation (11). Moreover, it has been shown that KIR expression can be modulated by TCR-mediated activation (23). IL-12 was chosen because preliminary data suggested that CD4⁺CD28^null T cells expressed high cell surface levels of the IL-12 receptor and that IL-12 could modify gene expression in CD4⁺CD28^null T cells (24). IL-12 is also known to function synergistically with IL-18 (25, 26). The clone was stimulated with anti-CD3, IL-12, IL-15, IL-18, or a combination of IL-12 and IL-18 over the course of a week. During the in vitro culture, all clones were also maintained in IL-2. On days 3 and 7, the cells were harvested, and the KIR gene expression pattern was analyzed. Representative results are shown for KIR2DL3, KIR3DL2, KIR2DL1, and KIR3DS1 (Fig. 4). After in vitro culture of this clone with the various cytokines, as well as anti-CD3 mAb, there was no change in the gene expression pattern of these particular receptors. There was no down-regulation of KIR2DL3 and KIR3DL2, which were previously transcribed by the clone, nor did the cytokines induce transcription of KIR2DL1 or KIR3DS1. Therefore, these data would suggest that under in vitro culture conditions, the gene expression pattern in CD4⁺CD28^null T cells is not altered, and that the significant degree of diversity in the gene expression of KIRs is acquired during in vivo clonal expansion.

View larger version (78K): [in this window] [in a new window]   FIGURE 4. Stability of KIR gene expression patterns in CD4+CD28null T cell clones during in vitro culture. A CD4+CD28null T cell clone was selected that expressed KIR2DL3 and KIR3DL2, but not KIR2DL1 and KIR3DS1. This clone was stimulated in vitro with 50 ng/ml OKT3 (lanes 4 and 9), 10 ng/ml IL-12 (lanes 5 and 10), 50 ng/ml IL-15 (lanes 6 and 11), 50 ng/ml IL-18 (lanes 7 and 12), or a combination of 10 ng/ml IL-12 and 50 ng/ml IL-18 (lanes 8 and 13) for 3 days (lanes 4–8) or 7 days (lanes 9–13). The KIR expression pattern after stimulation was compared with that before stimulation (lane 3). PBMCs were used as a positive control (lane 1); cDNA was omitted from the negative control (lane 2).

Results in Fig. 2 document the enormous diversification in KIR gene expression for each clone family. To address the issue of whether this diversification was random or biased, the KIR expression pattern within each CD4⁺CD28^null T cell clone family was compared. These results are summarized in Fig. 3. It is apparent that certain KIRs are transcribed more frequently than others. KIR2DL3 and KIR2DL4 are transcribed by a majority of T cell clones irrespective of the TCR family grouping. In contrast, there are other receptors, specifically KIR2DL1, KIR3DL1, KIR2DS1, KIR2DS3, KIR2DS4, and KIR3DS1, that are not frequently transcribed by clones in any of these TCR families. Therefore, we concentrated on the transcription pattern of the three remaining receptors, KIR2DL2, KIR2DS2, and KIR3DL2, which are found with wide variation among T cell clone families. The frequency distribution of these receptors allowed for the distinction of three patterns. The association of these patterns with the different clone families was statistically significant (Table III). The clone families A1 and B1 were characterized by frequent transcription of each of these three receptors. Transcription of these genes was infrequent or totally absent in the B2 family. In contrast, in the remaining clone families, C1 and C2, transcription of KIR3DL2 was detected in 50% of the clones, while transcription of KIR2DL2 and KIR2DS2 was again infrequent. Taken together, despite the enormous diversification in KIR transcription, three KIR isoforms, KIR2DL2, KIR2DS2, and KIR3DL2, can be used to define a consensus pattern for a particular clone family. It is of interest to note that these dominant patterns were clone-specific and not necessarily donor-specific. Although clone families C1 and C2 (both derived from donor C) had a similar pattern of transcription, the clone families B1 and B2 from donor B were clearly distinct. In the clone family B1, KIR3DL2 was most frequently transcribed (80% of all clones tested), followed by KIR2DL2 (67%), and KIR2DS2 (67%). In contrast, KIR2DL2 and KIR3DL2 were completely absent in the clone family B2, and only KIR2DS2 was found at a low frequency of 20%.

Modeling of KIR gene expression by computer simulation

The diverse repertoire of KIR transcription in each T cell clone family clearly indicated that KIR transcription occurred subsequent to TCR expression. It was also obvious from our data that the likelihood of transcription for each KIR isoform was not identical. Certain KIR isoforms, such as KIR2DL4, were frequently transcribed, while others, such as KIR2DL1, KIR3DL1, or KIR2DS1, were only infrequently found in any of the clone families. In a similar pattern, KIR2DL4 was expressed by essentially all KIR⁺ NK cells (27) and CD8 T cells (28). In addition, other receptors, including KIR3DL1, KIR2DS1, and KIR2DL1, were only infrequently found on KIR⁺ NK cells (27) and CD8 T cells (28). Although minor influences of HLA type cannot be excluded, peripheral selection does not appear a likely reason for these irregular expressions because the patients included in this study expressed different HLA-C variants. The KIR patterns, however, were not completely random, and each clone family possessed a consensus pattern. This observation raised the issue of whether KIR expression occurs during the early stages of T cell expansion and the dominant pattern represents a founder effect, or whether the clone-specific patterns could also be explained by a cumulative expression during the late stages of clonal expansion. Based on the frequencies of CD4⁺CD28^null T cells expressing particular BV elements in the peripheral blood, the size of each clone was on the order of 1% of all T cells. Therefore, the clonal size could be estimated to be 1 x 10⁹ cells, suggesting that a clone underwent proliferation through a minimum of 30 generations.

As described in Materials and Methods, a computer model was developed to simulate the stochastic branching process defined by repeated cell divisions. We simulated the clonal proliferation of a single T cell clone over the course of up to 30 cell divisions. Although previous data suggest that expression of Ly49 receptors on NK cells occur in a sequential manner (12), our data for KIR gene expression was more consistent with a stochastic model. Therefore, at various times during the proliferation, KIR2DS2, KIR2DL2, KIR3DL2, KIR2DL3, and KIR3DL1 expression was initiated with a given fractional probability over a defined number of cell divisions (Table IV). We initially based the models on four basic assumptions. First, the probability of KIR expression was gene-specific, but identical for the different clone families and constant for a given number of cell cycles. Second, all clone families had the same number of generations during which de novo KIR expression could be induced. Third, KIR expression was not lost after it had been expressed. Fourth, selective forces that may favor or disfavor the proliferation of KIR-expressing cells were ignored. Several models were tested using these assumptions (models 1–7, Table IV). In models 1, 2, and 3, KIR expression occurred only in the early stages of clonal expansion, between the second and fifth, third and sixth, or second and sixth cell divisions, respectively. In contrast, models 4, 5, and 6 induced KIR expression at the later stages of proliferation, between generations 20 and 25, 23 and 27, or 18 and 27, respectively. In model 7, KIR expression occurred over the entire course of clonal expansion, from generations 1–25. Based on the simulations, a probability was calculated that reflected the ability of each model to generate KIR expression patterns identical to those found in the experimental data. The models were then compared using a likelihood ratio test. Results of one representative comparison (model 3 vs 5) is given in Table V. Likelihood ratios that are significantly >1 (or log likelihood ratios >0) indicate that the model being analyzed is less likely to approximate the experimental data than the model with which it is being compared. The simulations suggested that the model in which KIR expression occurred during the first several generations of clonal expansion was significantly more likely to result in the experimentally observed expression patterns of KIR2DL2 and KIR3DL1. A similar trend was seen for KIR2DS2. For KIR3DL2, both models described the experimental results equally well. KIR2DL2, KIR2DS2, and KIR3DL2 allowed for the distinction of the different clone families (Table III). In contrast, the simulations favored the late expression model for KIR2DL3, a receptor that was equally expressed in all clone families.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

 
KIRs are rarely found on human T cells, and their expression is particularly infrequent on CD4⁺ T cells. However, KIRs are regularly found on CD4⁺CD28^null T cells, a subset of T cells that is expanded in patients with RA (29) and patients with unstable angina (30), but is only infrequently found in healthy individuals. It is unclear at what stage of development these T cells acquire KIR expression. CD4⁺CD28^null T cells are highly oligoclonal; however, they express a variety of TCR - and -chains and do not show any TCR preference, a characteristic for CD1d-restricted NK-T cells (17). Moreover, typical CD4⁺ NK-T cells may express several NK cell-related receptors, such as CD161, but generally not KIRs (15). This suggests that CD4⁺CD28^null T cells are distinct from the CD1d-restricted NK-T cell. CD4⁺CD28^null T cells express markers of previous activation, including CD45RO, CD44, and CD57 (4). Therefore, it could be argued that the expression of KIRs represents a late developmental stage in T cells undergoing chronic stimulation, as has been argued for the subset of CD8⁺KIR⁺ T cells (31). In vitro conditions that induce KIR expression have not been found, neither for CD8 nor CD4 T cells. Our studies provide clear evidence that KIR transcription in CD4⁺CD28^null T cells occurs after termination of TCR rearrangement. T cell clones expressing identical TCR -chain transcripts showed a diverse repertoire of KIR expression. These results parallel previously published data on CD8 T cells that share identical TCR - and -chains and also express varying repertoires of KIRs (32). In vitro culture of CD4⁺CD28^null T cells was not associated with further diversification, suggesting that KIR transcription occurred during clonal expansion in vivo. Surprisingly, transcription of particular KIRs in individual clone families, including KIR2DL2, KIR2DS2, and KIR3DL2 was not random, and each family possessed a preferred family-specific pattern. Based on computer simulations, such patterns can emerge if KIR expression is limited to the first generation cycles of an expanding T cell clone family. Such a founder effect would indicate that commitment for KIR expression occurs at an early, and not a late, developmental stage of selected CD4 T cells. Alternatively, such patterns may emerge if clone families differ largely in the number of generation cycles during which induction of KIR expression is possible.

The association of CD4⁺CD28^null T cells with certain disease states has given rise to the hypothesis that these cells directly contribute to disease manifestations. In RA, frequencies of CD4⁺CD28^null T cells correlate with extraarticular manifestations. Patients with nodular RA, and in particular patients with vasculitic complications of RA, have grossly expanded CD4⁺CD28^null T cells (29). Also, in patients with early RA, the frequency of CD4⁺CD28^null T cells is a prognostic marker that correlates with the risk of subsequent joint erosions (J. J. Goronzy and C. M. Weyand, unpublished observations). In patients with coronary artery disease, the frequency of CD4⁺CD28^null T cells correlated with the risk of having acute coronary syndromes (30). Acute coronary syndromes develop if the atherosclerotic plaque is inflamed and develops a fissure or ulceration with subsequent thrombosis. Indeed, clonal expansion of CD4⁺CD28^null T cells was found in the inflamed plaque of such patients (33). These findings raised the question of whether the involvement of these cells in tissue injury in certain diseases reflects unique regulatory or effector functions.

Gene expression profiling showed that these cells, in contrast to normal CD4⁺ T cells, express perforin and granzyme B (6). CD4⁺CD28^null T cells isolated from patients with coronary artery disease exhibited cytotoxic activity toward endothelial cells in a perforin-dependent manner (34). In addition to their cytotoxic abilities, they express a set of regulatory molecules that are usually found on NK cells. CD4⁺CD28^null T cells express CD8 -chain dimers and not the CD8 -chain, reminiscent of T cells that have undergone extrathymic T cell maturation (4). In addition, MHC class I-recognizing receptors of the KIR family are frequently found, particularly those recognized by the mAb GL183 (9). KIRs comprise a multigenic, multifunctional family and, depending upon their cytoplasmic domains, are subdivided into inhibitory and activating isoforms (18). GL183 recognizes the inhibitory isoforms, KIR2DL2 and KIR2DL3, and the stimulatory isoform, KIR2DS2. Analysis of KIR isoforms in CD4 T cell clones from patients with RA demonstrated a preferential expression of the stimulatory receptor KIR2DS2, often in the absence of the inhibitory KIR2DL2 and KIR2DL3 isoforms. This prompted the hypothesis that a preponderance of stimulatory receptors on CD4⁺CD28^null T cells may predispose a person to autoimmune manifestations (9). Indeed, the KIR2DS2 gene was found to be genetic risk factor of vasculitic manifestations in patients with RA (35).

The mechanism through which KIR expression is regulated is currently unclear. In contrast to CD94/NKG2 gene transcription, KIR expression cannot be induced on mature T cells. In our experiments, KIR transcription was stable in vitro and could not be changed by the addition of cytokines that are known to influence NKG2 expression or have been shown to be important for KIR and NKG2 expression during NK cell maturation (11, 36).

In the absence of an experimental system that allows for the induction of KIR expression on T cells, it remains debatable at which stage of development KIR expression is induced. The multitude of different markers on CD4⁺CD28^null T cells that are usually not expressed on normal CD4 T cells may favor the interpretation that CD4⁺CD28^null T cells are a special lineage and that KIR expression may occur early during development. However, the loss of CD28 expression, which is one of the hallmarks of CD4⁺CD28^null T cells, has been shown to occur with replicative senescence (37). The transcriptional mechanism that leads to the down-regulation of CD28 during replication is identical to the transcriptional defect in CD4⁺CD28^null T cells, raising the possibility that KIR expression is part of a senescence program. Accordingly, it has been postulated for CD8 T cells that KIR expression is a late event during clinical expansion. Vely et al. (32) and Uhrberg et al. (28) isolated CD8 T cell clone families with identical TCR-/TCR- rearrangement. KIR expression typing in one clone family allowed for the definition of 18 different KIR phenotypes, clearly demonstrating that KIR expression was initiated subsequent to TCR rearrangement (32). Vely et al. (32) sorted CD158b⁺CD8⁺ and CD158b^-CD8⁺ T cells and found identical productive and nonproductive TCR-/TCR- transcripts in both subsets. Young et al. (31) have provided evidence supporting a sequential program of gene expression encoded in the human leukocyte receptor complex on chromosome 19q13.4. In their model, genes belonging to the leukocyte Ig-like receptors were expressed on CD8 T cells after activation at the stage of effector cells, while KIR gene transcription was activated in the minor proportion of activated T cells that survived clonal downsizing and became long-term memory T cells.

In this study, five different T cell clone families from three different patients with RA were included. In these five clone families, KIR gene transcription was diverse, which is clearly consistent with the model that KIR expression occurred after completion of TCR rearrangement. However, KIR gene transcription was not random. Certain KIR molecules were found more frequently than others. KIR2DL4 was expressed by virtually all T cell clones. Transcripts for KIR2DL3 were also consistently found in all five clone families, although at a lower frequency. The receptors KIR2DL2, KIR2DS2, and KIR3DL2 were variably transcribed in the different clone families. In some clone families, most clones were positive for each of these receptors. This was in contrast to other clone families, in which fewer of the T cell clones transcribed these receptors. The remaining receptors, KIR2DL1, KIR3DL1, KIR2DS1, KIR2DS3, KIR2DS4, and KIR3DS1, were infrequently transcribed. In general, the hierarchy of gene expression was similar to that previously reported for CD8⁺ T cells (28) and NK cells (27). Further evidence that the expression pattern was not random came from statistical analyses of gene expression in the different clone families. These analyses suggested that at least three different dominant patterns could be distinguished between the five clone families. We considered several models to explain this bias. It is possible that the repertoire of KIR expression was shaped by selective forces, depending on the MHC class I polymorphisms expressed by the original host. We have previously shown that stimulatory KIRs were able to promote proliferative responses of T cell clones, suggesting that the recognition of self-MHC class I molecules may lead to clonal expansion (9). Ugolini et al. (38) have used a KIR2DL3-HLACw3 transgenic mouse system and have found that KIR2DL3 is involved in T cell survival. In contrast, several studies have not found any evidence that KIR cell surface expression is directly affected by the nature or expression levels of HLA class I molecules (11, 39). In the study described here, two T cell clone families that showed a clear difference in KIR expression patterns were derived from the same patient, suggesting that peripheral selection is not a dominant force imposing KIR expression patterns.

Alternatively, the dominant pattern may reflect a founder effect and de novo KIR expression may be limited to early T cell expansion. We simulated KIR expression during early or late stages of clonal expansion for varying numbers of cell cycles. There was no indication in our experimental data that expression occurs in an ordered, sequential process. Therefore, the models were built on the assumption that KIR expression occurs as a stochastic process with a given probability for each KIR gene. The simulations indicated that the experimental data were consistent with a model in which KIR expression occurred after TCR rearrangement but during the early generation cycles of T cell clones undergoing expansion. If this model is correct, CD4⁺CD28^null T cells would likely be a separate lineage that acquires KIRs during early clonal expansion after Ag encounter. Alternatively, clone families may have grossly differed in the number of generation cycles during which they were amenable to KIR expression. In these simulations, we could not clearly distinguish between models of early and late gene expression to fit the experimental data.

In summary, our data demonstrate that CD4⁺CD28^null T cells acquire a diverse repertoire of KIR expression. Diversification within different clone families sharing the same TCR -chain is consistent with a model of accumulative KIR expression after completion of TCR rearrangement. The profile of KIR expression on individual T cell clone families is not random. If KIR expression patterns reflect a founder effect, then stochastic events in KIR expression during the first generations of T cell clonal expansion may have significant downstream effects and determine the behavior of this clone during pathological immune responses.

	Footnotes

 
¹ This work was supported by the National Institutes of Health (Grants R01 AR41974, R01 AR42527, and R01 AG15043) and the Mayo Foundation. M.R.S. is supported by a National Arthritis Foundation Fellowship (AF21).

² Address reprint requests and correspondence to Dr. Jörg J. Goronzy, Department of Medicine and Immunology, Mayo Clinic and Foundation, 401 Guggenheim Building, 200 First Street Southwest, Rochester, MN 55905. E-mail address: goronzy.jorg{at}mayo.edu

³ Abbreviations used in this paper: RA, rheumatoid arthritis; BV, -chain variable region; KIR, killer Ig-like receptor.

Received for publication December 5, 2001. Accepted for publication February 13, 2002.

	References

Top Abstract Introduction Materials and Methods Results Discussion References

 

1. Goronzy, J. J., C. M. Weyand. 2001. Thymic function and peripheral T-cell homeostasis in rheumatoid arthritis. Trends Immunol. 22:251.[Medline]
2. Wagner, U. G., K. Koetz, C. M. Weyand, J. J. Goronzy. 1998. Perturbation of the T cell repertoire in rheumatoid arthritis. Proc. Natl. Acad. Sci. USA 95:14447.[Abstract/Free Full Text]
3. Koetz, K., E. Bryl, K. Spickschen, W. M. O’Fallon, J. J. Goronzy, C. M. Weyand. 2000. T cell homeostasis in patients with rheumatoid arthritis. Proc. Natl. Acad. Sci. USA 97:9203.[Abstract/Free Full Text]
4. Schmidt, D., J. J. Goronzy, C. M. Weyand. 1996. CD4⁺CD7^-CD28^- T cells are expanded in rheumatoid arthritis and are characterized by autoreactivity. J. Clin. Invest. 97:2027.[Medline]
5. Park, W., C. M. Weyand, D. Schmidt, J. J. Goronzy. 1997. Co-stimulatory pathways controlling activation and peripheral tolerance of human CD4⁺CD28^- T cells. Eur. J. Immunol. 27:1082.[Medline]
6. Namekawa, T., U. G. Wagner, J. J. Goronzy, C. M. Weyand. 1998. Functional subsets of CD4 T cells in rheumatoid synovitis. Arthritis Rheum. 41:2108.[Medline]
7. Schirmer, M., A. N. Vallejo, C. M. Weyand, J. J. Goronzy. 1998. Resistance to apoptosis and elevated expression of Bcl-2 in clonally expanded CD4⁺CD28^- T cells from rheumatoid arthritis patients. J. Immunol. 161:1018.[Abstract/Free Full Text]
8. Vallejo, A. N., M. Schirmer, C. M. Weyand, J. J. Goronzy. 2000. Clonality and longevity of CD4⁺CD28^null T cells are associated with defects in apoptotic pathways. J. Immunol. 165:6301.[Abstract/Free Full Text]
9. Namekawa, T., M. R. Snyder, J. H. Yen, B. E. Goehring, P. J. Leibson, C. M. Weyand, J. J. Goronzy. 2000. Killer cell activating receptors function as costimulatory molecules on CD4⁺CD28^null T cells clonally expanded in rheumatoid arthritis. J. Immunol. 165:1138.[Abstract/Free Full Text]
10. Warrington, K. J., S. Takemura, J. J. Goronzy, C. M. Weyand. 2001. CD4⁺CD28^- T cells in rheumatoid arthritis patients combine features of the innate and adaptive immune systems. Arthritis Rheum. 44:13.[Medline]
11. Miller, J. S., V. McCullar. 2001. Human natural killer cells with polyclonal lectin and immunoglobulinlike receptors develop from single hematopoietic stem cells with preferential expression of NKG2A and KIR2DL2/L3/S2. Blood 98:705.[Abstract/Free Full Text]
12. Roth, C., J. R. Carlyle, H. Takizawa, D. H. Raulet. 2000. Clonal acquisition of inhibitory Ly49 receptors on developing NK cells is successively restricted and regulated by stromal class I MHC. Immunity 13:143.[Medline]
13. Dorfman, J. R., D. H. Raulet. 1998. Acquisition of Ly49 receptor expression by developing natural killer cells. J. Exp. Med. 187:609.[Abstract/Free Full Text]
14. Elewaut, D., M. Kronenberg. 2000. Molecular biology of NK T cell specificity and development. Semin. Immunol. 12:561.[Medline]
15. Seaman, W. E.. 2000. Natural killer cells and natural killer T cells. Arthritis Rheum. 43:1204.[Medline]
16. Schaible, U. E., S. H. Kaufmann. 2000. CD1 molecules and CD1-dependent T cells in bacterial infections: a link from innate to acquired immunity?. Semin. Immunol. 12:527.[Medline]
17. Schmidt, D., P. B. Martens, C. M. Weyand, J. J. Goronzy. 1996. The repertoire of CD4⁺CD28^- T cells in rheumatoid arthritis. Mol. Med. 2:608.[Medline]
18. Uhrberg, M., N. M. Valiante, B. P. Shum, H. G. Shilling, K. Lienert-Weidenbach, B. Corliss, D. Tyan, L. L. Lanier, P. Parham. 1997. Human diversity in killer cell inhibitory receptor genes. Immunity 7:753.[Medline]
19. Fleischer, B.. 1988. Non-specific propagation of human antigen-dependent T lymphocyte clones. J. Immunol. Methods 109:215.[Medline]
20. Miller, J. S.. 2001. Data Analysis for Scientists and Engineers Peer Management Consultants, Evanston.
21. Arstila, T. P., A. Casrouge, V. Baron, J. Even, J. Kanellopoulos, P. Kourilsky. 1999. A direct estimate of the human T cell receptor diversity. Science 286:958.[Abstract/Free Full Text]
22. Mingari, M. C., M. Ponte, S. Bertone, F. Schiavetti, C. Vitale, R. Bellomo, A. Moretta, L. Moretta. 1998. HLA class I-specific inhibitory receptors in human T lymphocytes: interleukin 15-induced expression of CD94/NKG2A in superantigen- or alloantigen-activated CD8⁺ T cells. Proc. Natl. Acad. Sci. USA 95:1172.[Abstract/Free Full Text]
23. Huard, B., L. Karlsson. 2000. KIR expression on self-reactive CD8⁺ T cells is controlled by T-cell receptor engagement. Nature 403:325.[Medline]
24. Warrington, K. J., S. Takemura, A. N. Vallejo, C. M. Weyand, J. J. Goronzy. 2001. IL-12-mediated functional restoration of CD4⁺CD28^null T cells in rheumatoid arthritis. Arthritis Rheum. 44:S378. (Abstr.).
25. Lauwerys, B. R., N. Garot, J. C. Renauld, F. A. Houssiau. 2000. Cytokine production and killer activity of NK/T-NK cells derived with IL-2, IL-15, or the combination of IL-12 and IL-18. J. Immunol. 165:1847.[Abstract/Free Full Text]
26. Lauwerys, B. R., J. C. Renauld, F. A. Houssiau. 1999. Synergistic proliferation and activation of natural killer cells by interleukin 12 and interleukin 18. Cytokine 11:822.[Medline]
27. Valiante, N. M., M. Uhrberg, H. G. Shilling, K. Lienert-Weidenbach, K. L. Arnett, A. D’Andrea, J. H. Phillips, L. L. Lanier, P. Parham. 1997. Functionally and structurally distinct NK cell receptor repertoires in the peripheral blood of two human donors. Immunity 7:739.[Medline]
28. Uhrberg, M., N. M. Valiante, N. T. Young, L. L. Lanier, J. H. Phillips, P. Parham. 2001. The repertoire of killer cell Ig-like receptor and CD94:NKG2A receptors in T cells: clones sharing identical TCR rearrangement express highly diverse killer cell Ig-like receptor patterns. J. Immunol. 166:3923.[Abstract/Free Full Text]
29. Martens, P. B., J. J. Goronzy, D. Schaid, C. M. Weyand. 1997. Expansion of unusual CD4⁺ T cells in severe rheumatoid arthritis. Arthritis Rheum. 40:1106.[Medline]
30. Liuzzo, G., S. L. Kopecky, R. L. Frye, W. M. O’Fallon, A. Maseri, J. J. Goronzy, C. M. Weyand. 1999. Perturbation of the T-cell repertoire in patients with unstable angina. Circulation 100:2135.[Abstract/Free Full Text]
31. Young, N. T., M. Uhrberg, J. H. Phillips, L. L. Lanier, P. Parham. 2001. Differential expression of leukocyte receptor complex-encoded Ig-like receptors correlates with the transition from effector to memory CTL. J. Immunol. 166:3933.[Abstract/Free Full Text]
32. Vely, F., M. Peyrat, C. Couedel, J. Morcet, F. Halary, F. Davodeau, F. Romagne, E. Scotet, X. Saulquin, E. Houssaint, et al 2001. Regulation of inhibitory and activating killer cell Ig-like receptor expression occurs in T cells after termination of TCR rearrangements. J. Immunol. 166:2487.[Abstract/Free Full Text]
33. Liuzzo, G., J. J. Goronzy, H. Yang, S. L. Kopecky, D. R. Holmes, R. L. Frye, C. M. Weyand. 2000. Monoclonal T-cell proliferation and plaque instability in acute coronary syndromes. Circulation 101:2883.[Abstract/Free Full Text]
34. Nakajima, T., S. Schulte, S. L. Kopecky, R. L. Frye, J. J. Goronzy, and C. M. Weyand. C-reactive protein sensitizes endothelial cells to T-cell mediated cytotoxicity. Circulation. In press.
35. Yen, J. H., B. E. Moore, T. Nakajima, D. Scholl, D. J. Schaid, C. M. Weyand, J. J. Goronzy. 2001. Major histocompatibility complex class I-recognizing receptors are disease risk genes in rheumatoid arthritis. J. Exp. Med. 193:1159.[Abstract/Free Full Text]
36. Raulet, D. H., R. E. Vance, C. W. McMahon. 2001. Regulation of the natural killer cell receptor repertoire. Annu. Rev. Immunol. 19:291.[Medline]
37. Vallejo, A. N., C. M. Weyand, J. J. Goronzy. 2001. Functional disruption of the CD28 gene transcriptional initiator in senescent T cells. J. Biol. Chem. 276:2565.[Abstract/Free Full Text]
38. Ugolini, S., C. Arpin, N. Anfossi, T. Walzer, A. Cambiaggi, R. Forster, M. Lipp, R. E. Toes, C. J. Melief, J. Marvel, E. Vivier. 2001. Involvement of inhibitory NKRs in the survival of a subset of memory-phenotype CD8⁺ T cells. Nat. Immunol. 2:430.[Medline]
39. Gumperz, J. E., N. M. Valiante, P. Parham, L. L. Lanier, D. Tyan. 1996. Heterogeneous phenotypes of expression of the NKB1 natural killer cell class I receptor among individuals of different human histocompatibility leukocyte antigens types appear genetically regulated, but not linked to major histocompatibililty complex haplotype. J. Exp. Med. 183:1817.[Abstract/Free Full Text]

This article has been cited by other articles:

				I. E. Dumitriu, E. Trallero Araguas, C. Baboonian, and J. C. Kaski CD4+CD28null T cells in coronary artery disease: when helpers become killers Cardiovasc Res, November 20, 2008; (2008) cvn248v3. [Abstract] [Full Text] [PDF]

				B. Zal, J. C. Kaski, J. P. Akiyu, D. Cole, G. Arno, J. Poloniecki, A. Madrigal, A. Dodi, and C. Baboonian Differential Pathways Govern CD4+CD28- T Cell Proinflammatory and Effector Responses in Patients with Coronary Artery Disease J. Immunol., October 15, 2008; 181(8): 5233 - 5241. [Abstract] [Full Text] [PDF]

				S. M. Studer, M. P. George, X. Zhu, Y. Song, V. G. Valentine, M. W. Stoner, J. Sethi, C. Steele, and S. R. Duncan CD28 Down-Regulation on CD4 T Cells Is a Marker for Graft Dysfunction in Lung Transplant Recipients Am. J. Respir. Crit. Care Med., October 1, 2008; 178(7): 765 - 773. [Abstract] [Full Text] [PDF]

				G. Li, C. M. Weyand, and J. J. Goronzy Epigenetic mechanisms of age-dependent KIR2DL4 expression in T cells J. Leukoc. Biol., September 1, 2008; 84(3): 824 - 834. [Abstract] [Full Text] [PDF]

				M. Thewissen, V. Somers, N. Hellings, J. Fraussen, J. Damoiseaux, and P. Stinissen CD4+CD28null T Cells in Autoimmune Disease: Pathogenic Features and Decreased Susceptibility to Immunoregulation J. Immunol., November 15, 2007; 179(10): 6514 - 6523. [Abstract] [Full Text] [PDF]

				G. Henel, K. Singh, D. Cui, S. Pryshchep, W.-W. Lee, C. M. Weyand, and J. J. Goronzy Uncoupling of T-cell effector functions by inhibitory killer immunoglobulin-like receptors Blood, June 1, 2006; 107(11): 4449 - 4457. [Abstract] [Full Text] [PDF]

				N. Ortonne, D. Huet, C. Gaudez, A. Marie-Cardine, V. Schiavon, M. Bagot, P. Musette, and A. Bensussan Significance of circulating T-cell clones in Sezary syndrome Blood, May 15, 2006; 107(10): 4030 - 4038. [Abstract] [Full Text] [PDF]

				M. W. L. Teng, M. H. Kershaw, Y. Hayakawa, L. Cerutti, S. M. Jane, P. K. Darcy, and M. J. Smyth T Cells Gene-engineered with DAP12 Mediate Effector Function in an NKG2D-dependent and Major Histocompatibility Complex-independent Manner J. Biol. Chem., November 18, 2005; 280(46): 38235 - 38241. [Abstract] [Full Text] [PDF]

				E. Amyes, A. J. McMichael, and M. F. C. Callan Human CD4+ T Cells Are Predominantly Distributed among Six Phenotypically and Functionally Distinct Subsets J. Immunol., November 1, 2005; 175(9): 5765 - 5773. [Abstract] [Full Text] [PDF]

				J. Xu, A. N. Vallejo, Y. Jiang, C. M. Weyand, and J. J. Goronzy Distinct Transcriptional Control Mechanisms of Killer Immunoglobulin-like Receptors in Natural Killer (NK) and in T Cells J. Biol. Chem., June 24, 2005; 280(25): 24277 - 24285. [Abstract] [Full Text] [PDF]

				R. Gniadecki and A. Lukowsky Monoclonal T-Cell Dyscrasia of Undetermined Significance Associated With Recalcitrant Erythroderma Arch Dermatol, March 1, 2005; 141(3): 361 - 367. [Abstract] [Full Text] [PDF]