Abstract
Aim
To evaluate the frequency, phenotype and the potential function of CD57+ T cell subsets in patients with pars planitis.
Methods
CD4+CD57+ and CD8+CD57+ T cells were quantitated in peripheral blood from 15 patients with pars planitis and 15 healthy controls. To evaluate the phenotype and potential function of CD57+ T cell subsets CCR7, CD27, CD28, CD45RA, CD45RO, intracellular IFN‐γ, IL‐4, perforin and granzyme‐A expression were assessed by flow cytometry.
Results
CD57+ T cells subsets were increased in patients with pars planitis (p = 0.002). The majority of CD4+CD57+ T cells were CCR7−CD27−CD28−CD45RO+, while the most CD8+CD57+ T cells were CCR7−CD27−CD28−CD45RA+. The number of cells positive for intracellular IFN‐γ and IL‐4 was higher in the CD57+ T cell populations. A greater number of CD8+CD57+ T cells than CD8+CD57− T cells were positive to perforin (p = 0.006) and granzyme‐A (p = 0.01).
Conclusions
CD57+ T cells had a phenotype associated with peripheral memory (CCR7−CD27−CD28−). Cytokine production by CD57+ T cells suggests that these cells may play a role in helper cell regulation. High expression of intracellular proteins involved in cytotoxicity suggests that CD8+CD57+ T cells may play an effector role. Taken together, this study proposes that CD57+ T cells function as memory‐effector T cell subsets during pars planitis pathogenesis.
Pars planitis is an idiopathic intermediate uveitis characterised by chronic and insidious intraocular inflammation involving anterior vitreous, peripheral retina, and pars plana. While the cause of pars planitis remains unknown, CD4+ T cells expressing the activation marker CD69 in either peripheral blood1 or the aqueous humor of patients with idiopathic uveitis2 support the hypothesis that T‐helper cells are involved in inducing and sustaining the inflammatory process. Little is known about the role of CD8+ T cells in pars planitis however, although CD8+CD56+ T cells are associated with Behçet's uveitis3 This issue is of particular interest because coexpression of functional NK receptors, like CD56 or CD57 on CD8+ T cells, is thought to be associated with effector function.3,4 In particular, CD57 expression on both CD4+ and CD8+ T cells, correlates with late immune responses5 and some pathologic conditions resulting from persistent stimulation of the immune system, including HIV, tuberculosis, colorectal cancer and arthritis rheumatoid.6,7,8,9,10 The aim of this work was to study the frequency, phenotype and potential function of CD57+ T cells subsets during a chronic eye disease such as active pars planitis.
Materials and methods
Patients
Thirty individuals (19 males and 11 females, mean age 9.5 years, range 6–14) with idiopathic pars planitis were studied. Pars planitis diagnosis was based on clinical history (mean disease duration 4 (SD 2) years), eye and physical examination. All patients were classified as having active intermediate uveitis according to the diagnostics standards of the SUN working group.11 Healthy age‐ and sex‐matched volunteers were used as controls. All participants gave their informed consent for blood sampling after written information was provided. This study adhered to the tenets of the Declaration of Helsinki and was approved by the Medical Investigation and Bioethics Committees of the Institute of Ophthalmology “Fundacion Conde de Valenciana” in Mexico City.
Monoclonal antibodies and reagents
Phycoerythrin‐labelled mouse IgG monoclonal antibodies (mAbs) against human CD28, IL‐4, granzyme‐A, and FITC‐labelled antibodies against human CD57, IFN‐γ, perforin, were purchased from BD PharMingen (San Diego, CA, USA). PE‐labelled mAbs against human CD57 were purchased from Santa Cruz Biotechnology, Inc (Santa Cruz, CA, USA). FITC‐labelled goat anti‐mouse IgM, FITC‐labelled mouse anti‐human CD4 and CD45RA and phycoerythrin‐labelled mouse anti‐human CD4 and CD45RO mAbs were purchased from Southern Biotech Inc (Birmingham, AL, USA). FITC‐labelled mAbs against human CD27, and phycoerythrin‐Cy5‐labelled mAbs anti‐CD8 and CD4, were obtained from e‐Biosciences (San Jose, CA, USA). FITC‐ and phycoerythrin‐labelled mAbs anti‐human CCR7 were purchased from RD Systems (Minneapolis, MN, USA) Lymphoprep (Ficoll 1.077 density) was obtained from Nycomed Pharma (Nyegaard, Oslo, Norway). Saponin, brefeldin‐A, RPMI‐1640 culture medium, PMA, ionomicyn, and salts were purchased from Sigma Chemical Co (St Louis, MO, USA). Sodium pyruvate, L‐glutamine, and 2‐mercaptoethanol were obtained from Gibco BRL (Rockville, MD, USA). Fetal calf serum was purchased from HyClone Labs (Logan, TU, USA).
Peripheral blood mononuclear cells
Whole heparinised peripheral blood was diluted 1:2 (vol/vol) in phosphate buffered saline (PBS), pH 7.2. Peripheral blood mononuclear cells (PBMCs) were separated on a Ficoll density gradient by centrifugation at 1700 rpm for 30 min at room temperature. After centrifugation, the interface cells were collected, washed twice, and counted using a haemocytometer to assess viability by trypan blue dye exclusion.
Immunofluorescence staining of cell surface markers
Triple‐colour staining was performed on PBMC by direct immunofluorescence, using FITC‐ or phycoerythrin‐mAb anti‐CD57 and phycoerythrin‐Cy5‐labelled mAbs against human CD4 or CD8, and either FITC‐ or phycoerythrin‐labelled mAbs against CD27, CD28, CD45RA, CD45RO or CCR7. Briefly, 2×105 cells were suspended in 20 μl PBS supplemented with 0.2% bovine serum albumin and 0.2% sodium azide (PBA), and incubated with fluorochrome‐labelled mAb for 30 min at 4°C. After incubation, the cells were washed twice with PBA, fixed with 1% p‐formaldehyde and analysed by flow cytometry.
Cell cultures
PBMC were cultured in 24‐well flat bottomed cell culture plates (Costar, Cambridge, MA, USA) at 1×106 cells/well in RPMI‐1640 medium supplemented with 1 mM sodium pyruvate, 2 mM L‐glutamine, 50 μg/ml gentamicin and 0.5% heat‐inactivated fetal calf serum, and incubated at 37°C in a 5% CO2 humidified chamber. After 24 h the culture medium was removed and fresh culture medium supplemented with 10% heat‐inactivated fetal calf serum, PMA/ionomicyn (5 ng/ml–0.2 μg/ml) and brefeldin‐A (10 μg/ml) were added. Four hours later, the cells were harvested and processed to measure intracellular protein expression by flow cytometry.
Immunofluorescence staining of intracellular proteins
Stimulated or unstimulated PBMCs were washed with PBA and stained with fluorescent labelled mAbs against CD57 and either CD4 or CD8 for 30 min. After washing, the cells were fixed with 4% p‐formaldehyde in PBS for 10 min at 4°C. The cells were washed twice with PBS and permeabilised with saponin buffer (0.1% saponin and 10% BSA in PBS) by shaking gently for 10 min at room temperature. The cells were then incubated with phycoerythrin‐labelled mAbs against human IL‐4 or granzyme A, and FITC‐labelled anti‐human IFN‐γ or anti‐perforin antibodies. In all cases isotype‐matched Ig/FITC and Ig/ phycoerythrin controls were used. After 30 min the cells were washed with PBS, fixed again with 1% p‐formaldehyde and analysed by flow cytometry.
Flow cytometric analysis
All cells were analysed for marker expression by collecting 5000 events using a FACScan flow cytometer (Becton Dickinson, CA, USA) and CellQuest Pro software. To analyse cell surface marker staining, a gate was drawn around the lymphocyte population based on their physical properties (forward and side scatter) and a second gate was drawn based on positive fluorescent antibody binding to CD4 and CD8. To analyse intracellular protein staining, positive and negative fluorescence staining of CD57+ T cells (forward scatter and fluorescence) were set manually based on the distribution of cells stained with the isotype controls. Data are presented as two‐dimensional dot plots or histograms. Control stains were performed using isotype‐matched mAb of unrelated specificity that were labelled with FITC‐, phycoerythrin‐Cy5‐ or phycoerythrin. Background staining was <1% and was subtracted from experimental values.
Statistical analysis
T tests or Mann–Whitney U tests were used to detect significant differences. The analysis was performed with Sigma‐Stat 3.1 and Sigma‐Plot 10.0 software. Differences were considered statistically significant when the test yielded p values less than 0.05.
Results
Increased frequency of CD57 T cells subsets in patients with pars planitis
We began by determining the percentage of CD57+ T cells in the peripheral blood of 15 patients with pars planitis and 15 healthy controls. While the percentage of CD4+ and CD8+ T cells was similar between both groups, the frequency of CD4+CD57+ T cells was 2.2 times higher in patients with pars planitis than healthy control individuals (p = 0.01). In contrast the frequency of CD8+CD57+ T cells was 5.3 times higher in patients with pars planitis than healthy controls (p = 0.0003) (fig 1, table 1).
Figure 1 Expression of CD4, CD8 and CD57 on peripheral blood mononuclear cells (PBMC). (A) CD4 and CD57 expression on PBMC; (B) CD8 and CD57 expression on PBMC. Representative images of one of 15 patients with pars planitis and one of 15 healthy controls.
Table 1 Frequency of cells analysed in lymphocyte gate.
| Patients with pars planitis | Healthy controls | |
|---|---|---|
| CD4+ | 35 (5) | 32 (7) |
| CD8+ | 20 (2) | 25 (2) |
| CD4+CD57+ | 6 (3)* | 3 (1)* |
| CD8+CD57+ | 15 (3)† | 3 (1)† |
Mean (standard error), n = 15.
The cells are represented in percentage of lymphocyte gated cells.
CD4+ and CD8+ are total percentage of lymphocyte gated cells.
*p = 0.01; †p = 0.003.
CD57 expression on differentiated effector memory T cell subsets
To determine whether CD57 was associated with central memory or effector memory T cell subsets,12 we measured CD27, CD28, CD45RA, CD45RO and CCR7 expression on CD57+ and CD57− T cell subsets from 15 patients with active pars planitis. The vast majority of CD4+CD57+ had 2.6‐fold less expression of CD27 (p = 0.003) (82.4 (SE 6) were CD27−) and 5.2 times lower expression of CCR7 (p = 0.0001) (84 (SE 8) were CCR7−) than CD4+CD57− T cells. Helper CD57+ T cells also showed 1.5‐fold less expression of CD28 than CD57− T cells (p = 0.02). The percentage of CD4+CD57+ T cells expressing CD45RO was 2.5 times higher than the percentage of CD4+CD57− T cells expressing this marker (p = 0.002). In contrast, CD45RA expression was 2.6‐fold higher on CD4+CD57− T cells than CD4+CD57+ T cells (p = 0.0005) (fig 2, table 2).
Figure 2 Phenotypic characterisation of CD4+CD57‐ and CD4+CD57+ T cells. Representative histograms of effector‐memory related markers on CD57− and CD57+ T helper subsets in patients with pars planitis.
Table 2 Phenotypic characterisation of CD57+ and CD57− T cells in patients with pars planitis.
| CD4+ | CD8+ | |||
|---|---|---|---|---|
| CD57− | CD57+ | CD57− | CD57+ | |
| CD27 | 53.2 (11)* | 17.6 (6)* | 55.6 (5)** | 21.5 (2)** |
| CD28 | 83.5 (10)* | 54.2 (9)* | 54.7 (8)†† | 24.3 (6)†† |
| CD45RA | 52.5 (6)†, § | 19.5 (6)†, † | 60.4 (7)‡‡ | 67.2 (8)§§ |
| CD45RO | 28.2 (3)‡, § | 72.3 (8)‡, † | 19.7 (5)‡‡ | 23.0 (5)§§ |
| CCR7 | 52.5 (11)¶ | 16.0 (8)¶ | 46.4 (4)¶¶ | 1.2 (1)¶¶ |
Mean (standard error), n = 15.
*p = 0.02; †p = 0.0005; ‡p = 0.002; §p = 0.009; ¶p = 0.03; **p = 0.001; ‡‡p = 0.01; ‡‡p = 0.0003; §§p = 0.0001; ¶¶p<0.001.
CD8+CD57+ T cells showed reduced expression of CD27 (p = 0.0009) (76.5 (SE 5) were CD27−) and 2.8‐fold reduced expression of CD28 (p = 0.004) (73.9 (SE 6) were CD28−) than CD8+CD57− T cells. Furthermore, 42.6‐fold fewer CD8+CD57+ T cells expressed CCR7 (p<0.001) (97.7 (SE 1) were CCR7−) than CD8+CD57− T cells (fig 2, table 2). In contrast to helper CD57+ T cells, CD8+ T cells were predominantly CD45RA+, independent of their CD57 status (fig 3, table 2).
Figure 3 Phenotypic characterisation of CD8+CD57− and CD8+CD57+ T cells. Representative histograms of effector‐memory related markers on CD57− and CD57+ T CD8+ cells in patients with pars planitis.
Intracellular IFN‐γ and IL‐4 are more highly expressed on CD57+ than CD57− T cells
Polyclonal stimulation in vitro showed that CD57+ T cells were able to produce cytokines. As shown in table 3, IFN‐γ was expressed by 11.8‐fold more in CD4+CD57+ T cells than CD4+CD57− T cells (p<0.001). Similarly, IL‐4 was expressed by 8.4‐fold more in CD4+CD57− T cells than CD4+CD57− cells (p<0.001) (fig 3, table 3). Likewise, IFN‐γ was expressed by 3.9 more in CD8+CD57+ T cells in comparison to CD8+CD57− T cells. Interestingly, CD8+CD57+ T cells produced 2.6 times more IFN‐γ than IL‐4 (p = 0.002) (fig 4, table 3).
Table 3 Intracellular cytokine expression by CD57+ and CD57− T cells.
| CD4+ | CD8+ | |||
|---|---|---|---|---|
| CD57− | CD57+ | CD57− | CD57+ | |
| IFN‐γ | 5.5 (1)* | 64.8 (4)*,† | 6 (2)*,* | 23.4 (2)* |
| IL‐4 | 6.0 (3)* | 51.0 (6)*,† | 1 (0.2)‡,* | 9.0 (2)‡ |
Mean (standard error), n = 15.
*p<0.001; †p = 0.02; ‡p = 0.002.
Figure 4 Comparative histogram analysis of intracellular cytokines on CD57− and CD57+ T cells subsets. The x axis denotes the fluorescence intensity to IFN‐γ or IL‐4 on CD57− T cells (thin line) and on CD57+ T cells (thick line) on both CD4 gated cells (top) and CD8 gated cells (bottom).
Perforin and granzyme‐A expression on CD57+ and CD57− CD8+ T cells
To establish the potential functional involvement of the granule‐exocytosis pathway in CD57+ and CD57− T cells, we assessed the percentage of CD8+ T cells containing perforin and granzyme‐A. Histogram analysis showed that CD8+CD57+ T cells expressed 13.5 times more perforin, and 8.2 times more granzyme‐A than CD8+CD57− T cells (p<0.001) (fig 5, table 4).
Figure 5 Perforin and granzyme‐A expression on CD8+CD57− and CD8+CD57+ T cells. Representative histograms of perforin and granzyme‐A expression on CD57 subsets. The x axis denotes the fluorescence intensity to granzyme‐A (top) or perforin (bottom) on CD57− T cells and on CD57+ T cells.
Table 4 Perforin and granzyme‐A expression by CD57+ and CD57− T cells.
| CD8+ T cells | ||
|---|---|---|
| CD57− | CD57+ | |
| Perforin | 3.5 (2)* | 46.3 (4)* |
| Granzyme‐A | 10.5 (3)* | 84.3 (6)* |
Mean (standard error), n = 15.
*p<0.001.
Discussion
Systemic CD4+CD69+ T cells responses are associated with idiopathic intermediate uveitis1 a chronic and insidious eye disease. Numerous studies have associated CD57 expression with persistent antigenic stimulation and chronic diseases.6,7,8,9,10 In this study, we characterised a population of CD57+ T cells that produced cytokines and contain granules associated with perforin‐granzyme A cytotoxic pathway, and showed increased frequency of this population in patients with pars planitis. As previously reported,6 these cells were also found to express an effector/memory phenotype with reduced levels of CCR7, CD27 and CD28. CCR7 is a cysteine chemokine receptor that controls homing to secondary lymphoid organs and divides human memory T cells into two functionally distinct subsets CCR7− and CCR7+ T cells. CCR7− cells are memory cells that migrate to inflamed tissues and display immediate effector function (effector‐memory), while CCR7+ cells are memory cells that express lymph‐node homing receptors and lack immediate effector function (central‐memory).13 Likewise, the absence of CD27 and CD28 expression on T cells is associated with cellular memory or effector characteristics. Fritsch RD et al14 characterised three distinct memory CD4 subpopulations that are distinguished by expression of CCR7 and the TNFR family member CD27. Telomere length also differed in these subsets (CCR7+CD27+ > CCR7−CD27+ > CCR7−CD27−), suggesting that CCR7−CD27− cells are more differentiated than CCR7−CD27+ cells. Tomiyama H et al showed that CD8+ T cells expressing CD27−CD28−CD45RA+/− phenotype are classified as effector cells.15 As previously suggested, the vast majority of CD4+CD57+ T cells had memory phenotype characterised by CD45RO expression.7 In contrast CD8+CD57+ T cells were predominantly CD45RA. The CD45RA isoform is expressed on naive T cells and a subset of effector cells and it has been proposed that CD45RA memory cells (CD28−CD27−CCR7−) represent the most differentiated memory cells.12,16,17
PBMCs from patients with sarcoid intermediate uveitis are polarised towards a Th1 response.1 Thus we measured intracellular cytokine expression in CD57+ T cells following activation with a polyclonal stimulus. Our results demonstrated a significant increase in the frequency of CD57+ T cells expressing either IFN‐γ or IL‐4. Th1 polarisation was only observed in the CD8+CD57+ T cell subset (Tc1 response). Interestingly, some findings have suggested that idiopathic intermediate uveitis is not associated with a Th1 immune response1 while others have shown a systemic Th1 response is found in both Behçet's uveitis and idiopathic intermediate uveitis.18 Our results support the hypothesis about Th1 polarisation. However it is critical to measure the levels of other regulatory cytokines to better define the real polarity of these cell subsets.
CD8+ T cell effector function is also defined by the expression of cytotoxic granules like perforin and granzyme‐A.19 These cytotoxic granules influence T cell‐mediated effector function against endogenous antigens.20 Interestingly, we found that a significant number of CD57+ T cells expressed perforin and granzyme A as shown by Sada‐Ovalle et al,21 supporting the notion that CD8+CD57+ T cells are effector cells.
The increase in frequency of CD57+ T cells has already associated to systemic infections6,21 and autoimmune diseases such as rheumatoid arthritis.10,22,23 Hingorani R et al22 demonstrated that CD8+CD57+ T cells from patients with rheumatoid arthritis were clonally expanded in the synovial fluid. Tan LC et al10 confirmed this data, but they showed specificity for viral antigens and not for auto‐antigens. Although it is not clear if these virus‐specific T cells play a role in the maintenance of local inflammation, Le Priol Y et al23 recently suggested that CD8+ CD57+ cells from HIV‐infected individuals are destined to migrate to non‐lymphoid tissues to exert its cytotoxic activities. In this regard, it will be interesting to analyse if CD57 expansion is related to infection in patients with pars planitis.
In summary, our results suggest that the CD57+ T cells are terminally differentiated cell subsets that probably act as memory‐effector T cells during pars planitis pathogenesis. Whether CD57+ T cell subsets are directly involved in the eye inflammation is still unknown and needs further investigation.
Acknowledgements
The authors thank Verónica Romero Martínez for technical assistance.
Abbreviations
PBMC - peripheral blood mononuclear cell
PBS - phosphate buffered saline
Footnotes
Funding: This work was supported by a grant from National Council of Science and Technology (CONACYT‐SALUD 2005‐14108).
Competing interests: None.
References
- 1.Murphy C C, Duncan L, Forrester J V.et al Systemic CD4(+) T cell phenotype and activation status in intermediate uveitis. Br J Ophthalmol 200488412–416. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Calder V L, Shaer B, Muhaya M.et al Increased CD4+ expression and decreased IL‐10 in the anterior chamber in idiopathic uveitis. Invest Ophthalmol Vis Sci 1999402019–2024. [PubMed] [Google Scholar]
- 3.Ahn J K, Chung H, Lee D S.et al CD8brightCD56+ T cells are cytotoxic effectors in patients with active Behcet's uveitis. J Immunol 20051756133–6142. [DOI] [PubMed] [Google Scholar]
- 4.Tarazona R, DelaRosa O, Alonso C.et al Increased expression of NK cell markers on T lymphocytes in aging and chronic activation of the immune system reflects the accumulation of effector/senescent T cells. Mech Ageing Dev 200012177–88. [DOI] [PubMed] [Google Scholar]
- 5.d'Angeac A D, Monier S, Pilling D.et al CD57+ T lymphocytes are derived from CD57‐ precursors by differentiation occurring in late immune responses. Eur J Immunol 1994241503–1511. [DOI] [PubMed] [Google Scholar]
- 6.Palmer B E, Blyveis N, Fontenot A P.et al Functional and phenotypic characterization of CD57+CD4+ T cells and their association with HIV‐1‐induced T cell dysfunction. J Immunol 20051758415–8423. [DOI] [PubMed] [Google Scholar]
- 7.Jimenez‐Martinez M C, Linares M, Baez R.et al Intracellular expression of interleukin‐4 and interferon‐gamma by a Mycobacterium tuberculosis antigen‐stimulated CD4+ CD57+ T‐cell subpopulation with memory phenotype in tuberculosis patients. Immunology 2004111100–106. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Okada T, Iiai T, Kawachi Y.et al Origin of CD57+ T cells which increase at tumour sites in patients with colorectal cancer. Clin Exp Immunol 1995102159–166. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Maeda T, Yamada H, Nagamine R.et al Involvement of CD4+, CD57+ T cells in the disease activity of rheumatoid arthritis. Arthritis Rheum 200246379–384. [DOI] [PubMed] [Google Scholar]
- 10.Tan L C, Mowat A G, Fazou C.et al Specificity of T cells in synovial fluid: high frequencies of CD8(+) T cells that are specific for certain viral epitopes. Arthritis Res 20002154–164. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Jabs D A, Nussenblatt R B, Rosenbaum JT; Standardization of Uveitis Nomenclature (SUN) Working Group Standardization of uveitis nomenclature for reporting clinical data. Results of the First International Workshop. Am J Ophthalmol 2005140509–516. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Sallusto F, Geginat J, Lanzavecchia A. Central memory and effector memory T cell subsets: function, generation, and maintenance. Annu Rev Immunol 200422745–763. [DOI] [PubMed] [Google Scholar]
- 13.Sallusto F, Lenig D, Forster R.et al Two subsets of memory T lymphocytes with distinct homing potentials and effector functions. Nature 1999401708–712. [DOI] [PubMed] [Google Scholar]
- 14.Fritsch R D, Shen X, Sims G P.et al Stepwise differentiation of CD4 memory T cells defined by expression of CCR7 and CD27. J Immunol 20051756489–6497. [DOI] [PubMed] [Google Scholar]
- 15.Tomiyama H, Takata H, Matsuda T.et al Phenotypic classification of human CD8+ T cells reflecting their function: inverse correlation between quantitative expression of CD27 and cytotoxic effector function. Eur J Immunol 200434999–1010. [DOI] [PubMed] [Google Scholar]
- 16.Faint J M, Annels N E, Curnow S J.et al Memory T cells constitute a subset of the human CD8+CD45RA+ pool with distinct phenotypic and migratory characteristics. J Immunol 2001167212–220. [DOI] [PubMed] [Google Scholar]
- 17.Hamann D, Kostense S, Wolthers K C.et al Evidence that human CD8+CD45RA+CD27‐ cells are induced by antigen and evolve through extensive rounds of division. Int Immunol111027–1033. [DOI] [PubMed] [Google Scholar]
- 18.Guenane H, Hartani D, Chachoua L.et al Production of Th1/Th2 cytokines and nitric oxide in Behcet's uveitis and idiopathic uveitis. J Fr Ophtalmol 200629146–152. [DOI] [PubMed] [Google Scholar]
- 19.Takata H, Takiguchi M. Three memory subsets of human CD8+ T cells differently expressing three cytolytic effector molecules. J Immunol 20061774330–4340. [DOI] [PubMed] [Google Scholar]
- 20.Kaech S M, Hemby S, Kersh E.et al Molecular and functional profiling of memory CD8 T cell differentiation. Cell 2002111837–851. [DOI] [PubMed] [Google Scholar]
- 21.Sada‐Ovalle I, Torre‐Bouscoulet L, Valdez‐Vazquez R.et al Characterization of a cytotoxic CD57+ T cell subset from patients with pulmonary tuberculosis. Clin Immunol 2006121314–323. [DOI] [PubMed] [Google Scholar]
- 22.Hingorani R, Monteiro J, Furie R.et al Oligoclonality of V beta 3 TCR chains in the CD8+ T cell population of rheumatoid arthritis patients. J Immunol 1996156852–858. [PubMed] [Google Scholar]
- 23.Le Priol Y, Puthier D, Lecureuil C.et al High cytotoxic and specific migratory potencies of senescent CD8+CD57+ cells in HIV‐infected and uninfected individuals. J Immunol 20061775145–5154. [DOI] [PubMed] [Google Scholar]