Behçet’s uveitis, characterized by chronic recurrent uveitis and obliterating retinal vasculitis, frequently causes bilateral blindness. Intraocular infiltration of TCRαβ+CD8brightCD56+ cells was a distinct feature in Behçet’s uveitis. However, phenotypic natures and effector functions of the cells have remained elusive. This study was conducted to determine phenotypic and functional characteristics and cytotoxic mechanisms of CD8brightCD56+ T cells in Behçet’s uveitis. CD11b+CD27CD62L phenotypes of CD8brightCD56+ T cells were increased in patients with active Behçet’s uveitis compared with inactive Behcet’s patients and normal controls. Interestingly, CD45RAdimCD45RO phenotypes were expanded, and CD94 expression was markedly up-regulated in contrast to the down-regulation of NKG2D. Furthermore, these subsets were polarized to produce IFN-γ and contained high amounts of preformed intracellular perforin while exclusively expressing surface FasL upon PI stimulation. Moreover, the cytolytic functions of freshly isolated CD8brightCD56+ T cells were up-regulated against both K562 (NK-sensitive) and Raji (NK-resistant) cells, which were effectively inhibited by perforin inhibitor (concanamycin A). Their cytolytic activity against HUVECs was also increased and was effectively suppressed by Fas ligand inhibitor (brefeldin A) and partly by perforin inhibitor. Furthermore, cytolytic functions of PMA and ionomycin-stimulated CD8brightCD56+ T cells against HUVECs were greatly enhanced, by pretreatment of recombinant human IFN-γ on HUVECs. Therefore, CD8brightCD56+ T cells in Behçet’s uveitis are characterized by cytotoxic effector phenotypes with functional NK receptors and function as strong cytotoxic effectors through both Fas ligand-dependent and perforin-dependent pathways.

Behçet’s uveitis (BD),4 a prototype of chronic recurrent uveitis with retinal vasulitis, is accompanied by multisystemic manifestations and commonly results in bilateral blindness necessitating the use of potent immunosuppressive drugs. Various immune effectors such as γδ T cells, neutrophils, or Th1 cells have been recognized as important immunopathogenic cells of BD (1, 2, 3). However, immunopathogenic roles of NK-typed CD8bright T cells in BD remain elusive. NK receptors are constitutively expressed and inducible on CD8bright T cells upon Ag exposure or the cellular stress (4, 5). CD56, a NK marker, is expressed on NK cells and T lymphocytes that mediate MHC-unrestricted cytotoxicity (6). CD56 can also be induced on cytotoxic αβ T cells via TCR or cytokine stimulation, and CD56+ T cells were found to be more autoaggressive as effectors than their CD56 counterparts (7, 8). Furthermore, the cytolytic and cytokine-producing effector functions in circulating CD8+ T cells are closely correlated with CD56 expression (9, 10).

Recently, we reported that the intraocular infiltration and peripheral expansion of TCRαβ+CD8brightCD56+ cells were distinct developments in patients with Behçet’s uveitis in contrast to endogenous uveitis of other etiologies (11). In addition, these immune cells were down-regulated following combined low-dose cyclosporine and prednisone treatment (12). Because the immunological roles of CD8bright T cells are distinctly different in each subset, phenotypic and functional separation of CD8bright T cells is crucial to identify the immunopathogenic effectors (13). However, there is little evidence regarding the phenotypic natures and functional characteristics of TCRαβ+CD8brightCD56+ cells in Behçet’s uveitis.

Contact-dependent, cell-mediated cytotoxicity functions through non-MHC- or MHC-restricted mechanisms. MHC-restricted cytotoxicity is mainly mediated by CD8+ CTLs through two distinct perforin- and Fas-based pathways resulting in tissue destruction (14). Concanamycin A ((CMA) perforin inhibitor) and brefeldin A ((BFA) Fas ligand (FasL) inhibitor) are useful tools for evaluation of each pathway in CD8+ T cell-mediated cytotoxicity (15). IFN-γ, a cytokine associated with cell-mediated immunity, represents an additional effector molecule released by activated CD8+ CTL and promotes the cytolytic effector functions of CD8+ CTL (16). TCRαβ+CD8brightCD56+ cells in Behçet’s uveitis may act as potent cytolytic effectors upon intraocular infiltration. However, there is a paucity of reports concerning the cytotoxic mechanisms of TCRαβ+CD8brightCD56+ cells in Behçet’s uveitis.

This study was conducted to determine the phenotypic and functional characteristics of CD8brightCD56+ T cells in Behçet’s uveitis and to investigate cytotoxic mechanisms of these expanded subsets.

Forty-five patients with BD according to the international criteria (17) were recruited from the Uveitis Clinic of Seoul National University Hospital from November 2003 to December 2004. Other etiologies were excluded following a complete ocular and systemic examination. The patients were clinically divided into two groups according to the degree of uveitis and systemic disease. The active group consisted of 24 patients with panuveitis (anterior chamber cells > 2+, vitreous haziness > 2+, exudative chorioretinal lesion > 1 according to the criteria of the International Uveitis Study Group (18)) and angiographically proven retinal vasculitis at the time of study. They also showed concurrent systemic manifestations such as oral ulcer and skin lesion. Inactive group consisted of 21 patients who had not experienced attacks of uveitis for >6 mo with or without treatment. All inactive patients showed normal levels of C-reactive protein, white blood cell counts, and erythrocyte sedimentation rate without any major systemic symptoms such as orogenital ulcers or skin lesions. Twenty-four non-BD uveitis patients were included in this study. These patients were diagnosed with the following: 10 with HLA-B27-related anterior uveitis, 7 with idiopathic anterior uveitis, 3 with Vogt-Koyanagi-Harada syndrome, 2 with idiopathic panuveitis, and 2 with pars planitis. All patients had received no medications within 4 wk at the time of venipuncture. In addition, 10 insulin-dependent diabetes (IDDM) patients were recruited for the analysis of the proportions of CD8brightCD56+ or CD8brightCD56 cells in peripheral blood. Twenty age- and sex-matched healthy subjects were used as controls. All procedures adhered to the tenets of the Declaration of Helsinki, and local approval was received from the Investigational Review Board of the Seoul National University Clinical Research Institute. Informed consents were obtained from all patients and controls subjects. Demographic features in enrolled patients are described in Table I.

Table I.

Patients characteristics

Controls (n = 20)BDNon-BD Uveitis (n = 24)IDDM (n = 10)
Activea (n = 24)Inactivea (n = 21)
Age (years) 36 ± 8 37 ± 9 40 ± 12 35 ± 11 22 ± 4 
Sex (M:F) 13:7 16:8 14:7 14:10 6:4 
Disease duration (years)  4.8 ± 2.1 6.2 ± 2.4 3.2 ± 1.4 8.2 ± 2.4 
Medicationsb      
 Cyclosporine  18/24 8/21 2/24  
 Azathioprine  8/24 4/21 4/24  
 Prednisone  19/24 4/21 6/24  
 Colchicine  21/24 15/21   
 NSAIDs  0/24 3/21 10/24  
Controls (n = 20)BDNon-BD Uveitis (n = 24)IDDM (n = 10)
Activea (n = 24)Inactivea (n = 21)
Age (years) 36 ± 8 37 ± 9 40 ± 12 35 ± 11 22 ± 4 
Sex (M:F) 13:7 16:8 14:7 14:10 6:4 
Disease duration (years)  4.8 ± 2.1 6.2 ± 2.4 3.2 ± 1.4 8.2 ± 2.4 
Medicationsb      
 Cyclosporine  18/24 8/21 2/24  
 Azathioprine  8/24 4/21 4/24  
 Prednisone  19/24 4/21 6/24  
 Colchicine  21/24 15/21   
 NSAIDs  0/24 3/21 10/24  
a

The manifestations at the time of venipuncture.

b

The previous treatments that were stopped 4 wk before the time of venipuncture in both active and inactive patients.

Target cells purchased from American Type Culture Collection for ex vivo cytotoxicity assays were lymphoblast K562 (NK sensitive), Burkitt’s lymphoma Raji (NK resistant), human vascular endothelial cell lines from the umbilical vein (HUVECs). Each cell line was maintained in the specific medium according to the manufacturer’s recommendations (IMDM for K562; RPMI 1640 for Raji; Ham’s F12K medium for HUVEC). PMA and ionomycin (PI), CMA, BFA, and recombinant human IFN-γ (rhIFN-γ) were purchased from Sigma-Aldrich. All mAbs, including anti-CD8-CyChrome or -allophycocyanin (RTA-T8), anti-CD11b-PE (ICRF44), anti-CD25-CyChrome (M-A251), anti-CD27-PE (M-T271), anti-CD45RA-PE (HI100), anti-CD45RO-allophycocyanin (UCHL1), anti-CD56-FITC or -PE (NCAM 16.2), anti-CD62L-PE (Dreg 56), anti-CD94-FITC (HP-3D9), anti-CD95-FITC (DX2), anti-HLA-DR-PE (G46-6), anti-NKG2D-allophycocyanin (1D11), anti-perforin-PE (δG9), anti-IFN-γ-PE (25723.11), and anti-IL-4-PE (3010.211) were purchased from BD Pharmingen, except anti-FasL-PE mAb (MBL).

Ten microliters of heparinized venous blood was obtained from uveitis patients, IDDM patients, and normal controls. PBMCs were recovered at the buffy coat layer using a Ficoll-Hypaque (Pharmacia) gradient, and then washed twice to remove RBCs. Directly conjugated mAbs were added at predetermined optimal dilutions. After a 30-min incubation, cells were washed twice in FACS buffer. Phenotypic analysis of fresh PBMCs was performed using anti-CD8 mAb and anti-CD56 mAb in conjunction with a three- or four-color immunofluorescence test for the expression levels of the following molecules: CD11b, CD25, CD27, CD45RA, CD45RO, CD62L, CD94, NKG2D, and HLA-DR. In 21 active BD uveitis and 24 non-BD uveitis patients, phenotypic analysis of intraocular infiltrating cells were performed through aqueous humor sampling using a 30-gauge needle. One hundred to 200 μl of aqueous humor were aspirated and immediately placed into a microcentrifuge tube precoated with EDTA and subjected to four-color flow cytometry using anti-CD3, anti-CD4, anti-CD8, and anti-CD56 mAb. Data were acquired using a FACSCalibur flow cytometer and analyzed with the CellQuest software program. Expression levels are represented by the frequencies of positive cells and the mean fluorescence intensity (MFI) of gated cells.

To measure the IFN-γ and IL-4 production by circulating CD8bright T cells with or without CD56, flow cytometric measurement was performed as described previously (19). Briefly, fresh PBMCs (106/well) from BD patients were stimulated in a 96-well round-bottom plate for 6 h at 37°C containing PMA (30 ng/ml) and ionomycin (1 μM) in the presence of protein-secreting inhibitor BFA (10 μg/ml). Unstimulated control wells were incubated with BFA alone. This short-term stimulation did not alter the membrane phenotype with respect to the CD56 expression. Cells were fixed with 4% paraformaldehyde at the end of a 6-h incubation period. They were also permeabilized with buffer to allow intracellular access of anti-IFN-γ-PE and anti-IL-4-PE mAb. For the phenotype staining, cells were incubated for 30 min at 4°C with anti-CD8-CyChrome mAb and anti-CD56-FITC mAb. Cells were immediately analyzed on FACSCalibur and CellQuest software. Intracellular IFN-γ and IL-4 expression is represented by the frequencies of positive cells in gated cells. In both unstimulated and stimulated wells, analysis gates were set for lymphocytes using forward and side scatter properties, and the frequencies of cytokine-producing cells were acquired by phenotype gating using anti-CD8 and anti-CD56. We preliminarily confirmed that the frequencies of IFN-γ- and IL-4-producing CD8bright T cells reached a peak at 6 h stimulation in normal controls and that the specificity of IFN-γ and IL-4 was tested in patients with BD and atopic dermatitis.

To determine intracellular amounts of perforin in unstimulated CD8bright T cells with or without CD56, PBMCs were first stained with anti-CD8 mAb and anti-CD56 mAb, washed, and then fixed and permeabilized in 1 ml of perm/wash solution containing saponin (permeabilizing buffer) at room temperature for 40 min. Cells were washed twice with buffer and incubated with intracellular anti-perforin mAb for 40 min at 4°C. Cells were then washed and immediately analyzed on a FACSCalibur. PE-labeled mouse IgG1 was used as negative control. In some experiments, we measured the amounts of intracellular perforin after 5 h of PI stimulation. The amounts of intracellular perforin were represented by the frequencies of positive cells and the MFI of gated cells.

To investigate the expression levels of surface FasL, fresh PBMCs (106/well) were stained with anti-CD8, anti-CD56, and anti-FasL after 5 h of PI stimulation without BFA. Unstimulated wells and PE-labeled-mouse IgG1 were used as controls. Surface FasL expression is represented as the frequencies of positive cells in gated cells. In some experiments, the inhibitory effect of 0–500 nM CMA and 0–100 μM BFA on surface FasL expression was determined. The frequencies of membrane FasL-expressing CD8brightCD56+ T cells reached a peak at 5 h stimulation in healthy controls.

Fresh PBMCs from 60 ml of heparinized peripheral blood were stained with anti-CD8-CyChrome mAb and anti-CD56-FITC mAb. CD8brightCD56+ cells and CD8brightCD56 cells were purified by FACStarPlus (purity of each population was >97%). Cells of each population (>2 × 105 cells in RPMI 1640 supplemented with 10% FCS and antibiotics) were directly subjected to ex vivo cytotoxicity assays. We confirmed that the proportions of apoptotic cells in sorted cells were <10% using Annexin VFITC and propidium iodide.

To investigate the killing ability of sorted cells, K562 cells and Raji cells were labeled with 100 μCi Na251CrO4 (Amersham Biosciences) for 90 min at 37°C in each growth medium, washed twice with medium, and then subjected to cytotoxicity assays. The labeled targets (3 × 103 cells per 100 μl/well) were incubated with sorted effector cells (3 × 104 cells per 100 μl/well) (E:T ratio = 10:1) in 96-well round-bottom microtiter plates. The plates were centrifuged after incubation for 4 h, after which the 100-μl supernatant was harvested and counted with a gamma counter. Specific 51Cr release (percentage) was calculated using the following formula: (experimental release − spontaneous release)/(maximum release − spontaneous release) × 100. Maximum release was determined by the incubation of target cells with 2% SDS. The spontaneous release was <10% of the maximum release.

To determine the cytotolytic activities of sorted cells on vascular endothelial cells, HUVECs were labeled with 100 μCi Na251CrO4 for 8 h at 37°C in Ham’s F12K medium, washed twice with medium, and then subjected to 6 h 51Cr release assays with the E:T ratio of 10:1. Maximum release was determined by the incubation of target cells with 100 μl of 1 N hydrochloric acid. The spontaneous release was <10% of the maximum release. Specific 51Cr release (percentage) was calculated as described previously.

Target cell treatment.

To determine the promoting effect of IFN-γ on the cytotoxicity of HUVECs, HUVECs were preincubated with 100 U/ml rhIFN-γ for 24 h at 37°C in Ham’s F12K medium, extensively washed with medium, and then subjected to 6-h 51Cr release assays with the E:T ratio of 10:1. Furthermore, we investigated the effect of IFN-γ on the Fas and HLA-DR expression of HUVECs.

Effector cell treatment.

To explore the cytotoxic mechanisms, the sorted effector cells were preincubated with 100 nM CMA or 20 μM BFA for 2 h in a final volume of 3 × 104 cells per 100 μl/well, mixed with 100 μl of target cells (E:T ratio = 10) in the presence of CMA or BFA, and then subjected to standard 4-h (K562, Raji) or 6-h (HUVECs) 51Cr release assay. Furthermore, PI-stimulated effector cells for 5 h were used against HUVECs in some experiments.

The MFI and frequencies of positive cells were compared using the Mann-Whitney U test. Values of p < 0.05 were considered significant.

In comparison to non-BD uveitis patients, intraocular infiltration and peripheral expansion of CD8brightCD56+ T cells were remarkable in BD uveitis patients (Fig. 1, A and B). These findings were more outstanding in aqueous humor than peripheral blood. The proportion of the CD8brightCD56 T cell was also increased in BD uveitis patients and was higher in aqueous humor than peripheral blood (Fig. 1,C). As shown in Fig. 1 D, the peripheral expansion of the CD8brightCD56+ T cell was not noted in IDDM patients.

FIGURE 1.

Intraocular and peripheral populations of CD8bright T cells with or without CD56 in uveitis patients and IDDM patients. Representative flow cytometric data of aqueous humors and peripheral blood using anti-CD8 mAb and anti-CD56 mAb are illustrated in patients with HLA-B27-related uveitis and with BD panuveitis (A). The percentages of each subset are shown in each upper right quadrant. Each proportion of CD8brightCD56+ T cells (B) and CD8brightCD56 T cells (C) in the eye was plotted vs the corresponding proportion in the peripheral blood from the same uveitis patients (BD uveitis, 21; non-BD uveitis, 24). The populations of CD8brightCD56+ T cells and CD8brightCD56 T cells were compared between the 24 active BD uveitis patients and the 10 IDDM patients (D).

FIGURE 1.

Intraocular and peripheral populations of CD8bright T cells with or without CD56 in uveitis patients and IDDM patients. Representative flow cytometric data of aqueous humors and peripheral blood using anti-CD8 mAb and anti-CD56 mAb are illustrated in patients with HLA-B27-related uveitis and with BD panuveitis (A). The percentages of each subset are shown in each upper right quadrant. Each proportion of CD8brightCD56+ T cells (B) and CD8brightCD56 T cells (C) in the eye was plotted vs the corresponding proportion in the peripheral blood from the same uveitis patients (BD uveitis, 21; non-BD uveitis, 24). The populations of CD8brightCD56+ T cells and CD8brightCD56 T cells were compared between the 24 active BD uveitis patients and the 10 IDDM patients (D).

Close modal

In active BD uveitis, CD8brightCD56+ T cells had a number of features compatible with primed effector phenotypes (Table II). Compared with inactive BD uveitis and normal controls, CD27 and CD62L of CD8brightCD56+ T cells were found to be down-regulated, whereas HLA-DR and CD11b were highly expressed (Fig. 2). CD8brightCD56 T cells showed the similar findings in active BD uveitis, but the changes were more remarkable on the CD8brightCD56+ T cells (Fig. 2). No significant difference was observed in CD25 expression levels. Furthermore, in active BD uveitis, the proportions of CD45RAlowCD45RO phenotypes were markedly increased in the CD8brightCD56+ T (Fig. 3). CD94 was expressed exclusively in active BD on the CD8brightCD56+ T cells. In contrast, NKG2D expression was significantly down-regulated on both the CD8brightCD56+ T cells and CD8brightCD56 T cells in BD compared with normal controls. However, NKG2D expression on CD8bright T cells with or without CD56 was up-regulated in active BD compared with inactive BD.

Table II.

Phenotypic characteristics of CD8bright T cells with or without CD56 in BD

Controls (n = 20)Active BD (n = 24)Inactive BD (n = 21)
CD8brightCD56+   
 % of PBMCs 2.1 ± 0.7 7.2 ± 2.4a 3.1 ± 0.9 
 CD11bb 68 ± 15 110 ± 12a 72 ± 18 
 CD62Lb 281 ± 73 134 ± 63a 298 ± 103 
 CD27c 47 ± 11 31 ± 6a 62 ± 14 
 HLA-DRb 25 ± 10 242 ± 104a 48 ± 14 
 CD25b 2 ± 0.2 2 ± 0.4 2 ± 0.5 
 CD45RAhighc 49 ± 7 54 ± 9 48 ± 7 
 CD45ROc 34 ± 14 5 ± 2a 12 ± 4d 
 NKG2Dc 82 ± 24 42 ± 10a 20 ± 7d 
 CD94c 10 ± 4 37 ± 11a 17 ± 5 
CD8brightCD56   
 % of PBMCs 23.6 ± 4.3 26.5 ± 5.8 25.1 ± 4.8 
 CD11bb 27 ± 10 35 ± 11 34 ± 7 
 CD62Lb 861 ± 93 360 ± 184a 896 ± 163 
 CD27c 86 ± 5 70 ± 9a 86 ± 8 
 HLA-DRb 29 ± 14 120 ± 12a 30 ± 12 
 CD25b 2 ± 0.3 2 ± 0.5 2 ± 0.4 
 CD45RAhighc 55 ± 14 60 ± 12 52 ± 15 
 CD45ROc 23 ± 9 11 ± 5 17 ± 8 
 NKG2Dc 65 ± 11 24 ± 4a 15 ± 6d 
 CD94c 0.7 ± 0.6 7 ± 2a 3 ± 1 
Controls (n = 20)Active BD (n = 24)Inactive BD (n = 21)
CD8brightCD56+   
 % of PBMCs 2.1 ± 0.7 7.2 ± 2.4a 3.1 ± 0.9 
 CD11bb 68 ± 15 110 ± 12a 72 ± 18 
 CD62Lb 281 ± 73 134 ± 63a 298 ± 103 
 CD27c 47 ± 11 31 ± 6a 62 ± 14 
 HLA-DRb 25 ± 10 242 ± 104a 48 ± 14 
 CD25b 2 ± 0.2 2 ± 0.4 2 ± 0.5 
 CD45RAhighc 49 ± 7 54 ± 9 48 ± 7 
 CD45ROc 34 ± 14 5 ± 2a 12 ± 4d 
 NKG2Dc 82 ± 24 42 ± 10a 20 ± 7d 
 CD94c 10 ± 4 37 ± 11a 17 ± 5 
CD8brightCD56   
 % of PBMCs 23.6 ± 4.3 26.5 ± 5.8 25.1 ± 4.8 
 CD11bb 27 ± 10 35 ± 11 34 ± 7 
 CD62Lb 861 ± 93 360 ± 184a 896 ± 163 
 CD27c 86 ± 5 70 ± 9a 86 ± 8 
 HLA-DRb 29 ± 14 120 ± 12a 30 ± 12 
 CD25b 2 ± 0.3 2 ± 0.5 2 ± 0.4 
 CD45RAhighc 55 ± 14 60 ± 12 52 ± 15 
 CD45ROc 23 ± 9 11 ± 5 17 ± 8 
 NKG2Dc 65 ± 11 24 ± 4a 15 ± 6d 
 CD94c 0.7 ± 0.6 7 ± 2a 3 ± 1 
a

Values of p < 0.05 compared with inactive BD and normal controls.

b

Data represents the values of MFI in gated cells.

c

Data represents the values of percentages of positive cells in gated cells.

d

Values of p < 0.05 compared with normal controls.

FIGURE 2.

Cell surface marker analysis of CD8bright T cells with or without CD56. Fresh PBMCs were stained simultaneously with anti-CD8 mAb, anti-CD56 mAb in conjunction with anti-CD27 mAb, anti-CD62L mAb, anti-CD11b mAb, anti-HLA-DR, and anti-CD25. CD8brightCD56+ T cells (A) and CD8brightCD56 T cells (B) were gated and then analyzed for each cell surface marker expression. Expression levels are represented as the MFI and the frequencies of positive cells in gated cells.

FIGURE 2.

Cell surface marker analysis of CD8bright T cells with or without CD56. Fresh PBMCs were stained simultaneously with anti-CD8 mAb, anti-CD56 mAb in conjunction with anti-CD27 mAb, anti-CD62L mAb, anti-CD11b mAb, anti-HLA-DR, and anti-CD25. CD8brightCD56+ T cells (A) and CD8brightCD56 T cells (B) were gated and then analyzed for each cell surface marker expression. Expression levels are represented as the MFI and the frequencies of positive cells in gated cells.

Close modal
FIGURE 3.

Phenotypic profiles of CD45 and functional NK receptors of CD8bright T cells with or without CD56. Fresh PBMCs were stained simultaneously with anti-CD8 mAb, anti-CD56 mAb, anti-CD45RA mAb, and anti-CD45RO mAb (A) or with anti-CD8 mAb, anti-CD56 mAb, anti-NKG2D mAb, and anti-CD94 mAb (B). CD8brightCD56+ T cells and CD8brightCD56 T cells were gated and then the percentages of CD45RA/CD45RO subsets (A) or NKG2D/CD94 subsets (B) are shown in each upper right quadrant.

FIGURE 3.

Phenotypic profiles of CD45 and functional NK receptors of CD8bright T cells with or without CD56. Fresh PBMCs were stained simultaneously with anti-CD8 mAb, anti-CD56 mAb, anti-CD45RA mAb, and anti-CD45RO mAb (A) or with anti-CD8 mAb, anti-CD56 mAb, anti-NKG2D mAb, and anti-CD94 mAb (B). CD8brightCD56+ T cells and CD8brightCD56 T cells were gated and then the percentages of CD45RA/CD45RO subsets (A) or NKG2D/CD94 subsets (B) are shown in each upper right quadrant.

Close modal

Because IFN-γ-positive CD8bright T cells were <1% of unstimulated cells from patients and controls, we measured the capacities of IFN-γ production after in vitro stimulation. After 6 h of PI stimulation, the frequencies of IFN-γ-producing CD8brightCD56+ T cells were significantly higher in active BD than those in inactive BD and normal controls (Fig. 4). In contrast, the frequencies of IL-4-producing-CD8brightCD56+ T cells were lower in active BD uveitis. Similarly, the IFN-γ-productions of CD8brightCD56 T cells were also elevated in active BD uveitis but smaller than those of CD8brightCD56+ T cells (Tables III and IV; Fig. 4).

FIGURE 4.

Measurements of intracellular IFN-γ and IL-4 in PI-stimulated CD8bright T cells with or without CD56. Fresh PBMCs were stimulated for 6 h by PI in the presence of protein secreting inhibitor BFA. After surface staining with anti-CD8 mAb and anti-CD56 mAb, cells were fixed, permeabilized, and intracellular accumulated cytokines were detected with anti-IFN-γ or anti-IL-4 mAbs. The histograms show the cytokine staining in gated CD8brightCD56+ T cells (A) and CD8brightCD56 T cells (B). Thin line indicates unstimulated controls cells; thick line indicates 6-h PI-stimulated cells.

FIGURE 4.

Measurements of intracellular IFN-γ and IL-4 in PI-stimulated CD8bright T cells with or without CD56. Fresh PBMCs were stimulated for 6 h by PI in the presence of protein secreting inhibitor BFA. After surface staining with anti-CD8 mAb and anti-CD56 mAb, cells were fixed, permeabilized, and intracellular accumulated cytokines were detected with anti-IFN-γ or anti-IL-4 mAbs. The histograms show the cytokine staining in gated CD8brightCD56+ T cells (A) and CD8brightCD56 T cells (B). Thin line indicates unstimulated controls cells; thick line indicates 6-h PI-stimulated cells.

Close modal
Table III.

Functional characteristics of peripheral CD8bright T cells with or without CD56 in BD

Cytokine productionControls (n = 10)Active BD (n = 10)Inactive BD (n = 10)
CD8brightCD56+ T cells    
 IFN-γa 25 ± 6 65 ± 15b 34 ± 18 
 IL-4a 2.5 ± 1.4 0.9 ± 0.6b 3.1 ± 1.1 
CD8brightCD56 T cells    
 IFN-γa 16 ± 7 35 ± 7b 25 ± 14 
 IL-4a 1.0 ± 0.7 0.7 ± 0.3 0.9 ± 0.5 
Cytokine productionControls (n = 10)Active BD (n = 10)Inactive BD (n = 10)
CD8brightCD56+ T cells    
 IFN-γa 25 ± 6 65 ± 15b 34 ± 18 
 IL-4a 2.5 ± 1.4 0.9 ± 0.6b 3.1 ± 1.1 
CD8brightCD56 T cells    
 IFN-γa 16 ± 7 35 ± 7b 25 ± 14 
 IL-4a 1.0 ± 0.7 0.7 ± 0.3 0.9 ± 0.5 
a

Data represents the frequencies of positive cells in gated cells after 6 h of PI stimulation.

b

Values of p < 0.05 compared with inactive BD and normal controls.

Table IV.

Functional characteristics of peripheral CD8bright T cells with or without CD56 in BD

Cytotoxic moleculesControls (n = 12)Active BD (n = 15)Inactive BD (n = 15)
CD8brightCD56+ T cells    
 Perforina 30 ± 10 (15 ± 6) 66 ± 7 (71 ± 25)§ 33 ± 11 (22 ± 11) 
 FasLb 45 ± 13 86 ± 14§ 51 ± 16 
CD8brightCD56 T cells    
 Perforina 6 ± 2 (6 ± 2) 26 ± 9 (20 ± 7)§ 9 ± 5 (9 ± 3) 
 FasLb 1.5 ± 0.4 2.5 ± 0.7 1.8 ± 0.6 
Cytotoxic moleculesControls (n = 12)Active BD (n = 15)Inactive BD (n = 15)
CD8brightCD56+ T cells    
 Perforina 30 ± 10 (15 ± 6) 66 ± 7 (71 ± 25)§ 33 ± 11 (22 ± 11) 
 FasLb 45 ± 13 86 ± 14§ 51 ± 16 
CD8brightCD56 T cells    
 Perforina 6 ± 2 (6 ± 2) 26 ± 9 (20 ± 7)§ 9 ± 5 (9 ± 3) 
 FasLb 1.5 ± 0.4 2.5 ± 0.7 1.8 ± 0.6 
a

Data represents the frequencies of positive cells in gated cells without stimulation (MFI).

b

Data represents the frequencies of positive cells in gated cells after 5 h of PI stimulation.

Because CD8brightCD56+ T cells shed the preformed intracellular perforin after in vitro PI stimulation, we measured the amounts of intracellular perforin of CD8bright T cells without in vitro stimulation. CD8brightCD56+ T cells in active BD uveitis contained higher levels of intracellular perforin than those in inactive BD uveitis and normal controls (Table III; Fig. 5). The perforin-producing proportion was also increased in CD8brightCD56 T cells, although the level was not as high as that of CD56+ subsets. Because CD8brightCD56+ T did not express the surface FasL without stimulation, we measured the expression levels of surface FasL after 5 h of PI stimulation. Surface FasL was strongly and exclusively expressed in CD8brightCD56+ T cells of both patients and normal controls and was markedly up-regulated in active BD (Table III; Fig. 5).

FIGURE 5.

Expression of cytotoxic molecules (perforin and FasL) in CD8bright T cells with or without CD56. A, Flow cytometric measurement of intracellular perforin was performed in fresh PBMCs without in vitro stimulation. After surface staining with anti-CD8 mAb and anti-CD56 mAb, cells were fixed, permeabilized, and intracellular-preformed perforin were detected with anti-perforin mAb. The histograms show the intracellular perforin staining in gated CD8brightCD56+ T cells and CD8brightCD56 T cells. Thin line indicates the negative isotype controls. B, Flow cytometric analysis of surface FasL expression was performed in fresh PBMCs after 5 h of PI stimulation. Cells were stained with anti-CD8 mAb and anti-CD56 mAb in conjunction with anti-FasL mAb. The histograms show the surface FasL staining in gated CD8brightCD56+ T cells and CD8brightCD56 T cells. Dotted line indicates the negative controls; thin line indicates the unstimulated cells; thick line indicates the 5-h PI-stimulated cells.

FIGURE 5.

Expression of cytotoxic molecules (perforin and FasL) in CD8bright T cells with or without CD56. A, Flow cytometric measurement of intracellular perforin was performed in fresh PBMCs without in vitro stimulation. After surface staining with anti-CD8 mAb and anti-CD56 mAb, cells were fixed, permeabilized, and intracellular-preformed perforin were detected with anti-perforin mAb. The histograms show the intracellular perforin staining in gated CD8brightCD56+ T cells and CD8brightCD56 T cells. Thin line indicates the negative isotype controls. B, Flow cytometric analysis of surface FasL expression was performed in fresh PBMCs after 5 h of PI stimulation. Cells were stained with anti-CD8 mAb and anti-CD56 mAb in conjunction with anti-FasL mAb. The histograms show the surface FasL staining in gated CD8brightCD56+ T cells and CD8brightCD56 T cells. Dotted line indicates the negative controls; thin line indicates the unstimulated cells; thick line indicates the 5-h PI-stimulated cells.

Close modal

CD8brightCD56+ T cells in active BD uveitis exhibited potent cytolytic activities against NK-sensitive K562 compared with the CD8brightCD56 T cell (Fig. 6). NK-like cytolytic functions of CD8brightCD56+ T cells in active BD were comparable to CD56+ (NK) cells (data not shown). Both CD8brightCD56+ T cells and CD8brightCD56 T cells in patients with active BD demonstrated strong cytotoxicity against NK-resistant Raji (Fig. 6). These findings indicated that, in active BD uveitis, NK-like cytolytic activities were additionally up-regulated in CD8brightCD56+ T cells. Furthermore, these cytotoxic activities were effectively inhibited by CMA pretreatment, which indicated that NK-like and MHC-dependent cytotoxic mechanisms of CD8brightCD56+ T cells in BD might be mediated through the perforin-dependent pathway (Fig. 6).

FIGURE 6.

Cytotoxic capacities of CD8bright T cells with or without CD56. Unstimulated fresh PBMCs from patients with BD and normal controls were sorted into CD8brightCD56+ T cells and CD8brightCD56 T cells, and then cytotoxicity was directly analyzed against K562 (A) and Raji (B) in a 4-h 51Cr release assay at the E:T ratio (10:1). Horizontal bar indicates the mean of three independent experiments. To determine the cytotoxic pathways of CD8brightCD56+ T cells, the cytotoxicity assay against K562 and Raji was performed in the presence of 100 nM CMA (C). Spontaneous release of all experiments was <10%.

FIGURE 6.

Cytotoxic capacities of CD8bright T cells with or without CD56. Unstimulated fresh PBMCs from patients with BD and normal controls were sorted into CD8brightCD56+ T cells and CD8brightCD56 T cells, and then cytotoxicity was directly analyzed against K562 (A) and Raji (B) in a 4-h 51Cr release assay at the E:T ratio (10:1). Horizontal bar indicates the mean of three independent experiments. To determine the cytotoxic pathways of CD8brightCD56+ T cells, the cytotoxicity assay against K562 and Raji was performed in the presence of 100 nM CMA (C). Spontaneous release of all experiments was <10%.

Close modal

The cytotoxic capacities of unstimulated CD8brightCD56+ T cells against HUVECs were up-regulated in active BD uveitis (Fig. 7). PI stimulation of CD8brightCD56+ T cells caused the discharge of prestored perforin from the cells and elevated surface FasL expression, which was suppressed by BFA (Fig. 7). IFN-γ pretreatment of HUVECs elevated the surface expression of both Fas and HLA-DR (Fig. 7). Moreover, the cytolytic capacities against HUVECs were enhanced by PI stimulation of the effector cells and/or by IFN-γ pretreatment of HUVECs, which were successfully inhibited by BFA (Fig. 7). In contrast, CMA only partially reduced the cytotoxicity of unstimulated CD8brightCD56+ T cells against HUVECs irrespective of IFN-γ pretreatment of HUVECs and did not affect the cytolytic activities of PI-stimulated CD8brightCD56+ T cells.

FIGURE 7.

Cytotoxic capacities of CD8brightCD56+ T cells against HUVECs. A, In active BD, CD8brightCD56+ T cells exhibited the potent cytolytic functions against HUVECs. Cytolytic activities of freshly sorted CD8brightCD56+ T cells and CD8brightCD56 T cells against HUVECs were measured by standard 6-h 51Cr release assay with E:T ratio of 10:1. B, Amounts of intracellular perforin in CD8brightCD56+ T cells were reduced after 5 h of PI stimulation. Dotted line indicates isotype negative control Ab, thin line indicates unstimulated status, and thick line indicates 5-h PI-stimulated status. BFA down-regulated surface FasL expression of 5-h PI-stimulated CD8brightCD56+ T cells and CMA did not. C, rhIFN-γ pretreatment for 24 h up-regulated the surface expression of Fas and HLA-DR on HUVECs. Dotted line indicates unrelated surface mAb (CD3), thin line indicates Fas expression of HUVECs without IFN-γ pretreatment, and thick line indicates Fas expression of HUVECs with IFN-γ pretreatment. D, PI stimulation of CD8brightCD56+ T cells and/or IFN-γ pretreatment of HUVECs up-regulated the cytotoxicity against HUVECs, which were effectively inhibited by BFA. In contrast, CMA partly inhibited the cytolytic activities of unstimulated CD8brightCD56+ T cells.

FIGURE 7.

Cytotoxic capacities of CD8brightCD56+ T cells against HUVECs. A, In active BD, CD8brightCD56+ T cells exhibited the potent cytolytic functions against HUVECs. Cytolytic activities of freshly sorted CD8brightCD56+ T cells and CD8brightCD56 T cells against HUVECs were measured by standard 6-h 51Cr release assay with E:T ratio of 10:1. B, Amounts of intracellular perforin in CD8brightCD56+ T cells were reduced after 5 h of PI stimulation. Dotted line indicates isotype negative control Ab, thin line indicates unstimulated status, and thick line indicates 5-h PI-stimulated status. BFA down-regulated surface FasL expression of 5-h PI-stimulated CD8brightCD56+ T cells and CMA did not. C, rhIFN-γ pretreatment for 24 h up-regulated the surface expression of Fas and HLA-DR on HUVECs. Dotted line indicates unrelated surface mAb (CD3), thin line indicates Fas expression of HUVECs without IFN-γ pretreatment, and thick line indicates Fas expression of HUVECs with IFN-γ pretreatment. D, PI stimulation of CD8brightCD56+ T cells and/or IFN-γ pretreatment of HUVECs up-regulated the cytotoxicity against HUVECs, which were effectively inhibited by BFA. In contrast, CMA partly inhibited the cytolytic activities of unstimulated CD8brightCD56+ T cells.

Close modal

Previous studies examining the role of T cell-mediated autoimmunity in BD focused on CD4+ T cells and found that T cells were autoreactive to heat shock proteins and retinal S Ag in BD patients, especially with ocular involvement (20, 21). However, extrinsic infectious pathogens and HLA-B51 are strongly associated with disease susceptibility (22, 23). A recent study suggested that HLA-B51-restricted cytotoxic CD8+ T cells might be involved in the pathogenesis of BD (24). Because the immune reactions evoked by certain infectious agents and by the possible autoantigen presentation with HLA-B51 may occur primarily in the CD8+ T cell compartment, we investigated the immunological alterations of CD8+ T cell population of Behcet’s uveitis patients.

In this study, the intraocular infiltration and peripheral expansion of CD8brightCD56+ T cells were unique features of BD uveitis in contrast to other etiologies of uveitis. This is consistent with previous reports that CD8+ T cells are the predominant intraocular infiltrating cells in BD uveitis and that CD8bright CD56+ T cells are expanded in the uveitis patients (11, 12). However, the skin biopsy of BD patients demonstrated that CD4+ T cells were predominant and that CD8+ T cells and CD56+ cells were much less predominant (25, 26). This discrepancy may reflect the unique intraocular immunomodulatory environment and more destructive inflammatory nature of BD uveitis compared with skin lesions (27, 28). Furthermore, IDDM patients, a representative immune-mediated inflammatory disease, did not show those characteristics. Therefore, CD8brightCD56+ T cells may function as particular immune effectors pertinent to chronic uveitis like BD.

This study demonstrates that the activated effector phenotypes (CD27CD62LCD11b+HLA-DR+) of CD8bright T cells are expanded in patients with active Behçet’s uveitis. The phenotypic natures of CD8bright T cells in BD were remarkable in the CD56+ subsets. CD56 is induced on CTLs from human peripheral blood lymphocytes in a Th1 dominant environment or via chronic antigenic stimulation (7, 8). Strong Th1 polarization and chronic autoantigenic stimulation have been observed in BD (2, 24). Previous studies also showed that CD27 and CD62L expression of human CD8+ T cells is inversely related to the effector functions (29, 30) and that CD8+CD28CD11b+ T cells are expanded in BD (31). Therefore, CD8brightCD56+ T cells in active BD uveitis may be induced from CD8brightCD56 T cells.

In this study, CD45RAdimCD45RO phenotypes of CD8brightCD56+ T cells were increased in active BD uveitis. CD45RA can be up-regulated on CD45RACD8+ T cell effectors during the late immune response following cytokine-driven proliferation (32, 33). Thus, in active BD uveitis, the CD45RAdimCD45RO effector phenotypes of CD8brightCD56+ T cells may acquire the CD45RA or CD45RO following the repetitive stimulation. Interestingly, CD94 was markedly up-regulated on CD8brightCD56+ T cells in active BD uveitis and supports the hypothesis that CD8brightCD56+ T may play an immunopathogenic role in the CD8+ T cell-mediated immune response of BD. CD94, a functional NK receptor for HLA-E, is inducible on activated effector CTL phenotypes in vivo and may act as an inhibitory NK receptor for CD8+ T cells (34, 35). Inhibitory function of CD94 on CD8+ T cells only occurs during chronic inflammation that needs to be controlled (36, 37). Therefore, it is plausible that CD94 on CD8brightCD56+ T cells in active BD uveitis may be induced for the regulation of effector functions (4, 38). Moreover, we found that NKG2D was considerably down-regulated on CD8bright T cells with or without CD56 in BD compared with controls, suggesting that the activation of CD8bright T cells may be actively regulated in BD. NKG2D, a functional NK receptor for MHC class I-related chain A (MICA), is constitutively expressed on CD8+ T cells and may function as a costimulatory receptor of CD8+ T cells (39, 40). Because MICA gene polymorphism is associated with the risk of BD (41), it is possible that aberrant expression of NKG2D on CD8bright T cells in BD may be related to the abnormal expression of MICA on target tissues.

In line with previous reports (2, 12, 42), this study confirms that CD8brightCD56+ T cells in active BD uveitis are polarized to produce large amounts of IFN-γ upon stimulation compared with the inactive BD and normal controls. Furthermore, these IFN-γ-producing capacities of CD8brightCD56+ T are 2-fold greater than those of CD56 counterparts in active BD uveitis. Thus, CD8brightCD56+ T cells may contribute to Th1 polarization of BD.

Cytotoxic CD8+ T cells destroy their target cells using two distinct cytotoxic molecules such as secreted lytic protein perforin and surface FasL. This study demonstrates that CD8brightCD56+ T cells in active BD uveitis are heavily armed with both cytotoxic molecules. Recent studies in normal human also showed that in vitro-stimulated CD8brightCD56+ T cells contained high amounts of intracellular perforin (9, 10). In this study, CD8brightCD56+ T cells in active BD contained comparable levels of intracellular perforin without in vitro stimulation and exclusively expressed surface FasL upon stimulation. Paralleled with the results of surface markers and cytokine production, both cytotoxic molecules were predominant in CD8brightCD56+ T cells compared with CD56 counterparts.

The perforin-dependent cytotoxic pathway dominates in CD8+ CTLs and NK cells (43). NK cells, part of the innate immune system, can kill target cells within minutes of the first stimulation of activating receptors and respond rapidly to challenge with limited capacity for Ag loads (44). In contrast, CD8+ CTLs, part of the adaptive immune system, can kill multiple target cells after TCR activation and exhibit the slower response with higher capacity for Ag loads (45). We found that MHC-mediated cytotoxicities of freshly isolated CD8bright T cells were increased in active BD uveitis. Interestingly, unstimulated CD8brightCD56+ T cells in active BD uveitis showed remarkable NK-like cytolytic activities. Moreover, the cytolytic functions of unstimulated CD8brightCD56+ T cells were mediated through the perforin-dependent pathway. Previous studies also reported that CD8+CD56+ T cells in normal human showed strong NK-like cytotoxicity after in vitro anti-CD3 or cytokine stimulation (9, 10, 45). Therefore, this suggests that CD8brightCD56+ T cells in active BD uveitis may acquire the NK-like cytotoxic activity in addition to the cytolytic functions of conventional CD8+ CTLs.

The essential retinal finding of Behçet’s uveitis is an obliterative, necrotizing retinal vasulitis resulting in ischemic retinal damages. Previous studies suggested that γδ T cells, neutrophils, CD4+ T cells, or anti-endothelial cell Ab might cause vascular endothelial cell damage of BD (1, 2, 3, 46). However, there has been insufficient evidence regarding the immunopathogenic roles of CD8bright T cells in vascular endothelial cell damage. In this study, the cytolytic activities of CD8bright T cells against HUVECs were increased in active BD uveitis, particularly in the subsets of CD8brightCD56+ T cells. Furthermore, the cytotoxic mechanisms of CD8brightCD56+ T cells against HUVECs were mediated through FasL-dependent and perforin-dependent pathways. Human CD8+ CTLs can use both FasL and perforin pathways (47). The important differences between the FasL- and perforin-initiated pathways are the speed of the cytotoxic event and the half-life of the cytotoxic molecules. Once formed, perforin granules are reoriented and released within minutes after TCR stimulation. In contrast, FasL is scarcely stored, even in activated effectors, requiring induction over 1–2 h post-TCR stimulation, and has a long half-life of 2–3 h in the absence of TCR stimulation (44). Similar findings were discovered in PI-stimulated CD8brightCD56+ T cells by flow cytometric detection of cytotoxic molecules and by cytotoxicity assay against HUVECs using inhibitors of both pathways. Thus, it is possible that CD8brightCD56+ T cells in active BD uveitis may damage vascular endothelial cells initially through perforin-initiated pathway after TCR stimulation and persistently through FasL pathway after repetitive chronic stimulation or even without TCR stimulation.

IFN-γ was reported to promote cytotoxic CD8+ T cell-mediated target cell destruction through the up-regulation of MHC and Fas expression (16, 48). This study shows that IFN-γ activates HUVECs and promotes CD8brightCD56+ T cell-mediated cytotoxicity against HUVECs through the FasL-dependent pathway. A recent study also reported that IFN-γ augmented CD95 expression resulting in sensitization of HUVECs to Fas-mediated apoptosis (49). Because CD8brightCD56+ T cells exhibited higher capacities of IFN-γ production after PI stimulation, secreted IFN-γ of these cells might be further harmful to vascular endothelial cells.

In summary, CD8brightCD56+ T cells in active BD are cytotoxic effector phenotypes with functional NK receptors and exhibit up-regulated NK-like cytotoxicity. These effectors exert cytolytic functions against vascular endothelial cells through both FasL- and perforin-dependent pathways. Therefore, CD8brightCD56+ T cells may function as strong cytotoxic effectors upon ocular infiltration resulting in visual loss of BD uveitis.

We thank Young Joo Kim for technical assistance.

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.

1

This study was supported in part by Seoul National University Hospital Clinical Research Institute Grant 03-2002-060, 061, 063.

4

Abbreviations used in this paper: BD, Behçet’s disease; CMA, concanamycin A; BFA, brefeldin A; FasL, Fas ligand; IDDM, insulin-dependent diabetes; PI, PMA and ionomycin; rhIFN-γ, recombinant human IFN-γ; MFI, mean fluorescence intensity; MICA, MHC class I-related chain A.

1
Hasan, A., F. Fortune, A. Wilson, K. Warr, T. Shinnick, Y. Mizushima, R. van der Zee, M. R. Standford, J. Sanderson, T. Lehner.
1996
. Role of γδ T cells in the pathogenesis and diagnosis with Behçet’s disease.
Lancet
347
:
789
.-794.
2
Frassanito, M. A., R. Dammacco, P. Cafforio, F. Dammacco.
1999
. Th1 polarization of the immune response in Behçet’s disease: a putative pathogenic role of interleukin-12.
Arthritis Rheum.
42
:
1967
.-1974.
3
Takeno, M., A. Kariyone, N. Yamashita, M. Takeuchi, Y. Mizushima, H. Kaneko.
1995
. Excessive function of peripheral blood neutrophils from patients with Behçet’s disease and from HLA-B51 transgenic mice.
Arthritis Rheum.
38
:
426
.-433.
4
McMahon, C. W., D. H. Raulet.
2001
. Expression and function of NK receptors in CD8+ T cells.
Curr. Opin. Immunol.
13
:
465
.-470.
5
Snyder, M. R., C. M. Weyand, J. J. Goronzy.
2004
. The double life of NK receptors: stimulation or co-stimulation?.
Trends Immunol.
25
:
25
.-32.
6
Lanier, L. L., R. Testi, J. Bindl, J. H. Phillips.
1989
. Identification of Leu-19 (CD56) leukocyte differentiation antigen and neural cell adhesion molecule.
J. Exp. Med.
169
:
2233
.-2238.
7
Ou, D., D. L. Metzger, X. Wang, P. Pozzilli, A. J. Tingle.
2002
. β Cell antigen-specific CD56+ NKT cells from type 1 diabetic patients: autoaggressive effector T cells damage human CD56+ β cells by HLA-restricted and non-HLA-restricted pathways.
Hum. Immunol.
63
:
256
.-270.
8
Satoh, M., S. Seki, W. Hashimoto, K. Ogasawara, T. Kobayashi, K. Kumagai, S. Matsuno, S. Takeda.
1996
. Cytotoxic γδ or αβ T cells with a natural killer cell marker, CD56, induced from human peripheral blood lymphocytes by a combination of IL-12 and IL-2.
J. Immunol.
157
:
3886
.-3892.
9
Pittet, M. J., D. E. Speiser, D. Valmori, J. Cerottini, P. Romero.
2000
. Cutting edge: Cytolytic effector function in human circulating CD8+ T cells closely correlates with CD56 surface expression.
J. Immunol.
164
:
1148
.-1152.
10
Ohkawa, T., S. Seki, H. Dobashi, H. Dobashi, Y. Koike, Y. Habu, K. Ami, H. Hiraide, I. Sekine.
2001
. Systematic characterization of human CD8+ T cells with natural killer cell markers in comparison with natural killer cells and normal CD8+ T cells.
Immunology
103
:
281
.-290.
11
Yu, H. G., D. S. Lee, J. Seo, J. K. Ahn, Y. S. Yu, W. J. Lee, H. Chung.
2004
. The numbers of CD8+ T cells and NKT cells increases in the aqueous humor of patients with Behçet’s uveitis.
Clin. Exp. Immunol.
137
:
437
.-443.
12
Ahn, J. K., J. M. Seo, J. Yu, F. S. Oh, H. Chung, H. G. Yu.
2005
. Down-regulation of IFN-γ-producing CD56+ T cells after combined low dose cyclosporine and prednisone treatment in active Behcet’s uveitis patients.
Invest. Ophthalmol. Vis. Sci.
46
:
2458
.-2464.
13
Hamann, D., P. A. Baars, B. Hooibrink, S. R. Kerkhof-Garde, M. R. Klein, R. A. van Lier.
1997
. Phenotypic and functional separation of memory and effector human CD8+ T cells.
J. Exp. Med.
186
:
1407
.-1418.
14
Kagi, B., K. Ledermann, R. Burki.
1994
. Molecular mechanisms of lymphocyte-mediated cytotoxicity and their role in immunological protection and pathogenesis in vivo.
Annu. Rev. Immunol.
14
:
207
.-232.
15
Kataoka, T., N. Shinohara, H. Takayama, K. Takaku, S. Kondo, S. Yonehara, K. Nagai.
1996
. Concanamycin A, a powerful tool for characterization and estimation of contribution of perforin- and Fas-based lytic pathways in cell-mediated cytotoxicity.
J. Immunol.
156
:
3678
.-3686.
16
Roth, E., H. Pircher.
2004
. IFN-γ promotes Fas ligand- and perforin-mediated liver cell destruction by cytotoxic CD8 T cells.
J. Immunol.
172
:
1588
.-1594.
17
International Study Group for Behçet’s disease.
1990
. Criteria for diagnosis of Behçet’s disease.
Lancet
335
:
1078
.-1080.
18
Bloch-Michel, E., R. B. Nussenblatt.
1987
. International Uveitis Study Group recommendations for the evaluation of intraocular inflammatory disease.
Am. J. Ophthalmol.
103
:
235
.-236.
19
Jung, T., U. Schauer, C. Heusser, C. Neumann, C. Rieger.
1993
. Detection of intracellular cytokines by flow cytometry.
J. Immunol. Methods
159
:
197
.-207.
20
Pervin, K., A. Childerstone, T. Shinnick, Y. Mizushima, R. van der Zee, A. Hasan, R. Vaughan, T. Lehner.
1993
. T cell epitope expression of mycobacterial and homologous human 65-kilodalton heat shock protein peptides in short term cell lines from patients with Behçet’s disease.
J. Immunol.
151
:
2273
.-2282.
21
de Smet, M. D., M. Dayan.
2000
. Prospective determination of T cell responses to S-antigen in Behçet’s disease patients and controls.
Invest. Ophthalmol. Vis. Sci.
41
:
3480
.-3484.
22
Lehner, T., E. Lavery, R. Smith, R. van der Zee, Y. Mizushima, T. Shinnick.
1991
. Association between the 65-kilodalton heat shock protein, Streptococcus sanguis, and the corresponding antibodies in Behçet’s disease.
Infect. Immunol.
59
:
1434
.-1441.
23
Mizuki, N., M. Ota, Y. Katsuyama, K. Yabuki, H. Ando, G. D. Palimeris, A. Kaklamini, M. Accoritini, P. Pivetti-Pezzi, S. Ohno, H. Inoko.
2000
. Localization of the pathogenic gene of Behçet’s disease by microsatellite analysis of three different populations.
Invest. Ophthalmol. Vis. Sci.
41
:
3702
.-3708.
24
Yasuoka, H., Y. Okazaki, Y. Kawakami, M. Hirakata, H. Inoko, Y. Ikeda, M. Kuwana.
2004
. Autoreactive CD8+ cytotoxic T lymphocytes to major histocompatibility complex class I chain-related gene A in patients with Behçet’s disease.
Arthritis Rheum.
50
:
3658
.-3662.
25
Mochizuki, M., E. Morita, S. Yamamoto, S. Yamana.
1997
. Characteristics of T cell lines established from skin lesions of Behçet’s disease.
J. Dermatol. Sci.
15
:
9
.-13.
26
Gul, A., S. Esin, N. Dilsen, M. Konice, H. Wigzell, P. Biberfeld.
1995
. Immunohistology of skin pathergy reaction in Behcet’s disease.
Br. J. Dermatol.
132
:
901
.-907.
27
Streilein, J. W..
1999
. Immunoregulatory mechanisms of the eye.
Prog. Retin. Eye Res.
18
:
357
.-370.
28
Direskeneli, H..
2001
. Behçet’s disease: infectious etiology, new autoantigens, and HLA-B51.
Ann. Rheum. Dis.
60
:
996
.-1002.
29
Oehen, S., K. Brduscha-Riem.
1998
. Differentiation of naïve CTL to effector and memory CTL: correlation of effector function with phenotype and cell division.
J. Immunol.
161
:
5338
.-5346.
30
Tomiyama, H., T. Matsuda, M. Takiguchi.
2002
. Differentiation of human CD8+ T cells from a memory to memory/effector phenotype.
J. Immunol.
168
:
5538
.-5550.
31
Houman, H., A. Hamzaoui, I. Ben Ghorbal, M. S. Khanfir, M. Feki, K. Hamzaoui.
2004
. Tc1/Tc2 ratio in the inflammatory process in patients with Behçet’s disease.
Mediators Inflamm.
13
:
247
.-253.
32
Hislop, A. D., N. E. Annels, N. H. Gudgeon, A. M. Leese, A. B. Rickinson.
2002
. Epitope-specific evolution of human CD8+ T cell responses from primary to persistent phases of Epstein-Barr virus infection.
J. Exp. Med.
195
:
893
.-905.
33
Geginat, J., A. Lanzavecchia, F. Sallusto.
2003
. Proliferation and differentiation potential of human CD8+ memory T cell subsets in response to antigen or homeostatic cytokines.
Blood
101
:
4260
.-4266.
34
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
.-1177.
35
Speiser, D. E., M. J. Pittet, D. Valmori, R. Dunbar, D. Rimoldi, D. Lienard, H. R. MacDonald, J. C. Cerottini, V. Cerundolo, P. Romero.
1999
. In vivo expression of natural killer cell inhibitory receptors by human melanoma-specific cytolytic T lymphocytes.
J. Exp. Med.
190
:
775
.-782.
36
McMahon, C. W., A. J. Zajac, A. M. Jamieson, L. Corral, G. E. Hammer, R. Ahmed, D. H. Raulet.
2002
. Viral and bacterial infections induce expression of multiple NK cell receptors in responding CD8+ T cells.
J. Immunol.
169
:
1444
.-1452.
37
Miller, J. D., M. Peters, A. E. Oran, G. W. Beresford, L. Harrington, J. M. Boss, J. D. Altman.
2002
. CD94/NKG2A expression does not inhibit cytotoxic effector function of lymphocytic choriomeningitis virus-specific CD8+ T cells.
J. Immunol.
169
:
693
.-701.
38
Moser, J. M., J. Gibbs, P. E. Jensen, A. E. Lukacher.
2002
. CD94-NKG2A receptors regulate antiviral CD8+ T cell responses.
Nat. Immunol.
3
:
189
.-195.
39
Bauer, S., V. Groh, J. Wu, A. Steinle, J. H. Philips, L. L. Lanier, T. Spies.
1999
. Activation of NK cells and T cells by NKG2D, a receptor for stress-inducible MICA.
Science
285
:
727
.-729.
40
Groh, V., R. Rhinehart, J. Randolph-Habecker, M. S. Topp, S. R. Riddell, T. Spies.
2001
. Costimulation of CD8 αβ T cells by NKG2D via engagement by MIC induced on virus-infected cells.
Nat. Immunol.
2
:
255
.-260.
41
Mizuki, N., M. Ota, M. Kimura, S. Ohno, H. Ando, Y. Katsuyama, M. Yamazaki, K. Watanabe, K. Goto, S. Nakamura, et al
1997
. Triplet repeat polymorphism in the transmembrane region of the MICA gene: a strong association of six GCT repetitions with Behçet’s disease.
Proc. Natl. Acad. Sci. USA
94
:
1298
.-1303.
42
Sugi-Ikai, N., M. Nakazawa, S. Nakamura, S. Ohno, M. Minami.
1998
. Increased frequencies of interleukin-2 and interferon γ producing T cells in patients with active Behçet’s disease.
Invest. Ophthalmol. Vis. Sci.
39
:
996
.-1004.
43
McDyer, J. F., T. J. Goletz, E. Thomas, C. H. June, R. A. Seder.
1998
. CD40 ligand/CD40 stimulation regulates the production of IFN-γ from human peripheral blood mononuclear cells in an IL-12 and/or CD28-dependent manner.
J. Immunol.
160
:
1701
.-1707.
44
Russell, J. H., T. J. Ley.
2002
. Lymphocyte-mediated cytotoxicity.
Annu. Rev. Immunol.
20
:
323
.-370.
45
Takayama, E., Y. Koike, T. Ohkawa, T. Majima, M. Fukasawa, N. Shinomiya, T. Yamaguchi, M. Konishi, H. Hiraide, T. Tadakuma, S. Seki.
2003
. Functional and Vβ repertoire characterization of human CD8+ T cell subsets with natural killer cell markers, CD56+CD57 T cells, CD56+CD57+ T cells and CD56CD57+ T cells.
Immunology
108
:
211
.-219.
46
Lee, K. H., H.-S. Chung, H. S. Kim, S.-H. Oh, M.-K. Ha, J.-H. Baik, S. Lee, D. Bang.
2003
. Human α-enolase from endothelial cells as a target antigen of anti-endothelial cell antibody in Behçet’s disease.
Arthritis Rheum.
48
:
2025
.-2035.
47
Ju, S. T., H. Cui, D. J. Panka, R. Ettinger, A. Marshak-Rothstein.
1994
. Participation of target Fas protein in apoptosis pathway induced by CD4+ Th1 and CD8+ cytotoxic T cells.
Proc. Natl. Acad. Sci. USA
91
:
4185
.-4189.
48
Mullbacher, A., M. Lobigs, R. T. Hla, T. Tran, T. Stehle, M. M. Simon.
2002
. Antigen-dependent release of IFN-γ by cytotoxic T cells up-regulates Fas on target cells and facilitates exocytosis-independent specific target cell lysis.
J. Immunol.
169
:
145
.-150.
49
Li, J. H., M. S. Kluger, L. A. Madge, L. Zheng, A. L. M. Bothwell, J. S. Pober.
2002
. Interferon γ augments CD95APO-1/Fas and procaspase-8 expression and sensitizes human vascular endothelial cells to CD95-mediated apoptosis.
Am. J. Pathol.
161
:
1485
.-1495.