Increased Expression of Down-Regulatory CTLA-4 Molecule onT Lymphocytes from Rheumatoid Synovial Compartment

M.-F. LIU*, C.-Y. YANG†, J.-S. LI*, K.-A. LAI†, S.-C. CHAO* & H.-Y. LEI‡

Departments of* Internal Medicine,†Orthopedics and‡Immunology, Medical College, National Cheng Kung University,Tainan, Taiwan, Republic of China

(Received 4 January 1999; Accepted in revised form 12 March 1999)

Liu M-F, Yang C-Y, Li J-S, Lai K-A, Chao S-C, Lei H-Y. Increased Expression of Down-Regulatory CTLA-4Molecule on T Lymphocytes from Rheumatoid Synovial Compartment. Scand J Immunol 1999;50:68–72

Since the CTLA-4 molecule expressed on activated T lymphocytes has recently been suggested to be animportant negative regulator in autoimmune diseases, this study was undertaken to investigate the expressionand function of CTLA-4 on synovial T cells from patients with rheumatoid arthritis. CTLA-4-expressingT cells were detected using a dual fluorescence flow cytometric method. Only a small percentage of peripheralblood T cells from patients with rheumatoid arthritis had detectable surface CTLA-4 expression (mean6 SD,1.896 1.92%). However, the levels of CTLA-4-positive T cells was increased significantly in rheumatoidsynovial fluids (5.446 4.96%) and synovial membranes (28.766 14.30%). To explore the role of CTLA-4molecule in the inflammation of rheumatoid joints, CTLA-4 was blocked with anti-CTLA-4 antibody to assessits effects on the production of tumour necrosis factora and interleukin 1b in synovial fluid mononuclear cellculture. The addition of anti-CTLA-4 antibody enhanced the production of tumour necrosis factora andinterleukin 1b in a dose-dependent manner. The data suggest that the expression of CTLA-4 plays a down-regulatory role in rheumatoid articular inflammation. We thus concluded that CTLA-4 was up-regulated onsynovial T cells from patients with RA, and the increased CTLA-4 expression might exert a down-regulationeffect on tumour necrosis factora and interleukin 1b production.

Ming-Fei Liu MD, Department of Internal Medicine, National Cheng Kung University Hospital, 138 Sheng LiRoad, Tainan, Taiwan, Republic of China

INTRODUCTION

B7 family molecules on antigen-presenting cells (APC) can bindto CD28 on T cells and provide a potent costimulatory signal forT-cell activation and proliferation [1–3]. CTLA-4 is a structuralhomologue of CD28, possibly due to gene duplication during theevolutionary process [4, 5]. CTLA-4 and CD28 bind the sameligands of B7-1 (CD80) and B7-2 (CD86), but with distinctkinetics [6, 7]. CD28 is constitutively expressed on all T cells,whereas CTLA-4 is expressed transiently on activated T cells [8].Initially, CTLA-4 was thought to deliver a costimulation signalas does CD28. However, it has recently been suggested thatCD28 and CTLA-4 molecules mediate opposite functions, e.g.CTLA-4 exerts an inhibitory rather than a costimulatory effect onT-cell activation [9–12].

Rheumatoid arthritis (RA) is an immune-mediated chronicinflammatory polyarthritis of unknown aetiology. Previous studieshave shown that the expression of costimulatory molecules CD80

and CD86 on APC was markedly increased in the RA synovium,which might interact with constitutively expressed CD28 oninfiltrating T cells, and contribute to the sustained inflammationin rheumatoid joints [13, 14]. Since CTLA-4 has recently beensuggested to be a potentially important negative regulator ofautoimmune diseases [15], the study of CTLA-4 expression inhuman RA is mandatory. In the present study, we explored theexpression and function of the CTLA-4 molecule on T cells inrheumatoid synovial compartment.

MATERIALS AND METHODS

Patients. Eighteen patients who fulfilled at least four criteria for RA setby the American College of Rheumatology were enrolled in the study.They were 16 females and two males with an age range of 37–67 years(mean 51 years). Each patient had active synovitis as evidenced byintense inflammatory effusion over the knee or shoulder joints. Thesynovial fluids (SF) were aspirated therapeutically and their peripheral

Scand. J. Immunol.50, 68–72, 1999

q 1999 Blackwell Science Ltd

blood (PB) was simultaneously drawn. The paired SF and PB were usedto analyse CTLA-4 expression on T lymphocytes. Eighteen age- and sex-matched normal persons were used as a control group. In addition, ninesamples of synovial membranes were obtained from different RApatients during therapeutic operation on knees or hips.

Preparation of peripheral blood mononuclear cells, synovial fluidmononuclear cells and synovial tissue cells. Peripheral blood mono-nuclear cells (PBMC) were prepared from heparinized fresh blood with astandard Ficoll-Hypaque density gradient solution. To prepare synovialfluid mononuclear cells (SFMC), the synovial fluids collected from thetherapeutic aspiration of inflammed joints of RA patients were firstincubated with 15 U/ml hyaluronidase for 30 min at 378C, and thenpassed through a wire mesh. The cells were spun down, and resuspendedin RPMI-1640. The mononuclear cells were separated using standardFicoll-Hypaque density centrifugation. In isolating synovial tissue cells(STC), synovial membranes obtained from the operating theatre wereimmediately trimmed to remove fatty tissue and minced into small pieces.The tissues were incubated in 0.5 mg/ml of collagenase, 0.15 mg/ml ofDNAse and 15 U/ml heparin for 4 h at 378C. The digests were then passedthrough a wire mesh. The cells were washed twice with phosphate-buffered saline (PBS).

Immunostaining and flow cytometry. Dual fluorescence staining withphycoerythrin (PE)-conjugated anti-CTLA-4 monoclonal antibody(MoAb; BNI 3.1, IgG2a) plus fluorescein isothiocyanate (FITC)-con-jugated anti-CD3 MoAb (UCHT1, IgG1) was performed to analysesurface CTLA-4 expression in freshly prepared PBMC, SFMC andSTC. Both MoAbs were purchased from PharMingen (San Diego, CA,USA). Either 5× 105 or 10× 105 cells were used in each staining. Asample of 1× 106 cells was incubated with 20ml PE-conjugated anti-CTLA-4 MoAb plus 20ml FITC-conjugated anti-CD3 MoAb for 30 minon ice in the dark. When the studied cell population was 5×105 cells,10ml fluorescence-conjugated MoAbs were used. After washing twice,the stained cells were analysed using a FACsort flow cytometer andassociated software program (Becton Dickinson, Lysis II). FITC- andPE-conjugated mouse IgG MoAbs staining as isotype controls wereincluded in each sample.

Cell culture. Tumour necrosis factor (TNF)-a and interleukin (IL)-1bwere measured in the supernatants derived from RA SFMC culture in theabsence or presence of anti-CTLA-4 MoAb. SFMC were placed at aconcentration of 1×105 cells/well in duplicate in a 96-well culture platefor 2 days in a CO2 incubator at 378C. Three different concentrations ofanti-CTLA-4 MoAb (0.1, 1 and 10mg/ml) were used to block CTLA-4.After 2 days’ culture, the supernatants were collected and frozen at¹208C until the quantitation of TNF-a and IL-1b.

TNF-a and IL-1b assay. The levels of TNF-a and IL-1b were bothmeasured using commercially available enzyme-linked immunosorbantassay (ELISA) kits (QuantikineTM, R&D Systems, Minneapolis, MN,USA). The assay employed the quantitative sandwich enzyme immuno-assay technique. Briefly, 200ml of standard or sample were addedto react with specific anti-TNF-a or anti-IL-1b MoAbs precoated ontoa microplate for 2 h at room temperature. After washing, horseradish-peroxidase-conjugated polyclonal antibody against TNF-a or IL-1b wasadded to react with the bound cytokines for 1 h. Then, 200ml of substratesolution was added and incubated for 20 min, followed by 50ml of stopsolution. The optical density of each well was determined immediatelyusing a microplate reader set to a wavelength of 450 nm. The concentra-tions of cytokines in each sample were calculated from the standardcurves.

Statistical analysis. Statistical analysis of the data was performed byWilcoxon rank-sum test in the present study.

RESULTS

Increased percentages of CTLA-4-expressing T cells inrheumatoid synovial fluids and synovial membranes

Only a very minor percentage of PB T cells from normal controlsand patients with RA had detectable levels of CTLA-4 expres-sion, and the fluorescence intensity of CTLA-4 expression wasrather low. The percentages of CTLA-4þ T cells were similarbetween normal controls and RA patients (1.886 1.63% versus1.896 1.92%). However, the levels of CTLA-4þ T cells in SF ofRA patients was increased significantly, compared with theirpaired PB T cells (5.446 4.96% versus 1.896 1.92%,P<0.05).Two representative CTLA-4 stainings in paired PB and SF areshown in Fig. 1. The levels of CTLA-4þ T cells between pairedPB and SF in each individual patient with RA are shown in Fig. 2.An overwhelming majority of patients (15/18) had apparentlyincreased CTLA-4þ T cells in SF. The degree of increase rangedfrom 1.04 to 16.24% with a mean of 4.60%. The expression ofCTLA-4 on T cells from synovial membranes was investigated innine different patients with RA. The result showed that a signifi-cant population of T cells expressed the CTLA-4 molecule. Theyranged form 7.79 to 43.21% (mean6 SD, 28.766 14.30%) asshown in Fig. 3.

To exclude the possibility that the enhanced CTLA-4 expres-sion on T cells from synovial compartment represented anin vitroartefact, i.e. was induced by hyaluronidase or collagense during

CTLA-4 on Rheumatoid Synovial T Cells69

q 1999 Blackwell Science Ltd,Scandinavian Journal of Immunology, 50, 68–72

Fig. 1. The detection of CTLA-4 expression on CD3þ T cells usingdual fluorescent flow cytometric analysis. Two representativeexamples of cell-surface CTLA-4 staining on anti-CD3þ T cells inpaired PB and SF from patients with RA. The grey peaks representisotype controls. Compared with PB, SF had 6.52% and 6.18% higherlevels of CTLA-4þ T cells, respectively.

the procedure of preparing SFMC or STC, PBMC from bothnormal controls and RA patients were either untreated or treatedwith hyaluronidase for 30 min or collagense for 4 h, and CTLA-4-expressing T cells were then analysed. We did not find asignificant difference in the percentages of CTLA-4þ T cellsbetween untreated and treated cells (data not shown). The controlexperiment confirms that increased CTLA-4 expression onsynovial T cells is anin vivo induced phenomenon.

Increased TNF-a and IL-1b production in cultured SFMC byblocking CTLA-4

To explore the role of CTLA-4 in the inflammation of rheuma-toid joints, CTLA-4 was blocked with anti-CTLA-4 antibody toassess its effects on the production of TNF-a and IL-1b in SFMCculture. Variable amounts of TNF-a and IL-1b were spon-taneously released into the supernatants during thein vitro cultureof SFMC derived from RA patients. Production of each wasenhanced by anti-CTLA-4 antibody. As shown in representativecases, anti-CTLA-4 antibody dose-dependently enhanced thespontaneous production of TNF-a and IL-1b in SFMC (Fig.4A,B). The percentage increase in TNF-a and IL-1b performedin six or four patients is shown in Fig. 4C and D, respectively.

DISCUSSION

The CTLA-4 molecule, a structural analogue of CD28, hasrecently been suggested to play an inhibitory, rather thancostimulatory role, in immune responses and may be an impor-tant negative regulator in autoimmune diseases [15–17]. Asdemonstrated in CTLA-4-deficient mice, decreased expressionof CTLA-4 may promote autoimmunity [18]. In addition, block-age of CTLA-4 resulted in accelerated and exacerbated diseasein murine experimental allergic encephalomyelitis and diabetesmodels [19, 20]. To date, the potential role of the CTLA-4molecule in the pathogenesis of human autoimmune diseasesremains unexplored, although CTLA-4 gene polymorphism wasfound to be associated with organ-specific autoimmune diseasesincluding Graves’ disease and type-1 diabetes in the Caucasianpopulation [21, 22]. There exists the possibility that patients withautoimmune diseases may have abnormalities in the expressionand regulation of CTLA-4, which lead to a failure to terminateongoing autoimmune responses. In the present study, we showedthat CTLA-4 expression was significantly up-regulated in syno-vial T cells in the majority of patients with RA. The result wasconcordant with an earlier study by Venwilghenet al. [23]. Theyreported higher expression (0–36%, average 15%) of the CTLA-4molecule on SF T cells than on paired PB T cells in RA patients. Inthe present study, we further examined the expression of CTLA-4on synovial tissues and showed that synovial membranesobviously contained a higher percentage of CTLA-4þ T cellsthan PB or SF. These results argue against an apparentlydefective expression of CTLA-4 within the inflammatory jointsof RA.

The proposed inhibitory function of the B7–CTLA-4

70 M.-F. Liu et al.

q 1999 Blackwell Science Ltd,Scandinavian Journal of Immunology, 50, 68–72

Fig. 2. The comparisons of CTLA-4þ T cells between paired PB andSF in 18 patients with RA. Among them, 15 patients had higherpercentages of CTLA-4þ T cells in SF than PB. The amplitudes ofincrease ranged from 1.04 to 16.24% with an average of 4.60%.

Fig. 3. The percentage distribution of CTLA-4þ T cells in peripheralblood (PB), synovial fluid (SF) and synovial tissue (ST). Values areexpressed as percentages in anti-CD3þ cell population. There werestatistically significant increases in the levels of CTLA-4þ T cells inSF and ST, compared with those in PB.

interaction has not been demonstrated directly [17]. Venwilghenet al. [23] reported that CTLA-4 functions as a costimulatorymolecule because blockage of CTLA-4 was able to partiallyinhibit the primary allogenic mixed lymphocyte reaction. Incontrast, we found that CTLA-4 might play a down-regulatoryrole in the inflammation of rheumatoid joints. As is well known,TNF-a and IL-1b are two major pro-inflammatory cytokinesinvolved in the pathogenesis of rheumatoid synovitis and aremainly derived from macrophages [24]. Since anti-CTLA-4antibody prevented the interaction of CTLA-4 with the consti-tutively expressed B7 family molecules on macrophages, theobservation that blocking CTLA-4 molecules with anti-CTLA-4antibody enhanced the TNF-a and IL-1b production in theinvitro SFMC culture implied that the CTLA-4 molecule mightsuppress the production of TNF-a and IL-1b cytokines throughits interaction with B7 molecules on macrophages, and wasbeneficial to the patients in terms of resolution of inflammation.

CTLA-4 is known to be transiently expressed on activatedT cells. The increased expression of CTLA-4 on rheumatoidsynovial T cells, but not paired PB T cells, indicates that thestimuli inducing CTLA-4 expression took place locally withinthe rheumatoid synovium. However, the possibility that CTLA-4-expressing T cells were preferentially attracted into joints couldnot be completely excluded. The substances or factors thatenhance CTLA-4 expression in the rheumatoid compartmentremain to be explored. It is probable that these substances orfactors might have potential clinical applications in suppressingsynovitis in the future. In addition to the down-regulation ofTNF-a and IL-1b production, the increased expression ofCTLA-4 on synovial T cells might contribute to the improvement

of synovitis by mediating the apoptosis of synovial T cells.Gribbenet al. [25] reported that cross-linking of the CTLA-4molecule could mediate apoptosis of activated human T cells inan antigen-restricted manner through the T-cell receptor. There-fore, we speculate that manipulation of CTLA-4 will eitherdelete pathogenic T cells or down-regulate TNF-a and IL-1b

production. This provides a new approach to the immunotherapyof autoimmune diseases with fewer side-effects.

In conclusion, we have shown that the CTLA-4 molecule wasup-regulated in synovial T cells from patients with RA, and thatincreased CTLA-4 expression down-regulated TNF-a and IL-1b

production.

ACKNOWLEDGMENTS

This work was supported by grant NSC87-2314-B006-039 fromthe National Science Council, Taiwan, Republic of China.

REFERENCES

1 Allison JP. CD28–B7 interaction in T-cell activation. Curr OpinImmunol 1994;6:414–19.

2 Linsley PS, Ledbetter JA. The role of the CD28 receptor during T cellresponses to antigen. Annu Rev Immunol 1993;11:191–212.

3 Lenschow DJ, Walunas TL, Bluestone JA. CD28/B7 system of T cellcostimulation. Annu Rev Immunol 1996;14:233–58.

4 Brunet JF, Denizot F, Luciani MF, Roux-Dosseto M, Suzan M,Mattei MG, Golstein P. A new member of the immunoglobulinsuperfamily – CTLA-4. Nature 1987;328:267–70.

5 Harper K, Balzano C, Rouvier E, Matti MG, Luciani MF, Golstein P.CTLA-4 and CD28 activated lymphocyte molecules are closely

CTLA-4 on Rheumatoid Synovial T Cells71

q 1999 Blackwell Science Ltd,Scandinavian Journal of Immunology, 50, 68–72

Fig. 4. Anti-CTLA-4 antibody enhanced thespontaneous TNF-a and IL-1b production insynovial fluid mononuclear cells culture.Representative data are shown in panels Aand B. The enhancement effect appeared tobe dose dependent. The percentages ofincreased TNF-a and IL-1b production in sixor four patients were shown in panels C andD, respectively.

related in both mouse and human as to sequence, message expression,gene structure, and chromosomal location. J Immunol 1991;147:1037–44.

6 Linsley PS, Brady W, Urnes M, Grosmaire LS, Damle NK, LedbetterJA. CTLA-4 is a second receptor for the B cell activation antigen B7.J Exp Med 1991;174:561–9.

7 Linsley PS, Greene JL, Brady W, Bajorath J, Ledbetter JA, Peach R.Human B7-1 (CD80) and B7-2 (CD86) bind with similar aviditiesbut distinct kinetics to CD28 and CTLA-4 receptors. Immunity1994;1:793–801.

8 Linsley PS, Greene JL, Tan P, Bradshaw J, Ledbetter JA, Anasatti C,Damle NK. Coexpression and functional cooperation of CTLA-4 andCD28 on activated T lymphocytes. J Exp Med 1992;176:1595–604.

9 Walunas TL, Lenschow DJ, Bakker CY, Linsley PS, Freeman GJ,Green JM, Thompson CB, Bluestone JA. CTLA-4 can function as anegative regulator of T cell activation. Immunity 1994;1:405–13.

10 Krummel MF, Allison JP. CD28 and CTLA-4 deliver opposingsignals which regulate the response of T cells to stimulation. J ExpMed 1995;182:459–66.

11 Krummel MF, Allison JP. CTLA-4 engagement inhibits IL-2 accu-mulation and cell cycle progression upon activation of resting T cells.J Exp Med 1996;183:2533–40.

12 Walunas TL, Bakker CY, Bluestone JA. CTLA-4 ligation blocksCD28-dependent T cell activation. J Exp Med 1996;183:2541–50.

13 Lui M-F, Kohsaka H, Sakurai H, Azuma M, Okumura K, Saito I,Miyasaka N. The presence of costimulatory molecules CD86 andCD28 in rheumatoid arthritis synovium. Arthritis Rheum 1996;39:110–14.

14 Ranheim EA, Kipps TJ. Elevated expression of CD80 (B7/BB1):and other accessory molecules on synovial fluid mononuclear cellsubsets in rheumatoid arthritis. Arthritis Rheum 1994;37: 1637–46.

15 Karandikar NJ, Vanderlugt CL, Walunas TL, Miller SD, Bluestone

JA. CTLA-4: a negative regulator of autoimmune disease. J Exp Med1996;184:783–8.

16 Liu Y. Is CTLA-4 a negative regulator for T-cell activation?Immunol Today 1997;18:569–72.

17 Thompson CB, Allison JP. The emerging role of CTLA-4 as animmune attenuator. Immunity 1997;7:445–50.

18 Waterhouse P, Penninger JM, Timms E, Wakeham A, Shahinian A,Lee KP, Thompson CB, Griesser H, Mak TW. Lymphoproliferativedisorders with early lethality in mice deficient in CTLA-4. Science1995;270:985–8.

19 Perrin PJ, Maldonado JH, Davis TA, June CH, Racke MK. CTLA-4blockage enhances clinical disease and cytokine production duringexperimental allergic encephalomyelitis. J Immunol 1996;157:1333–6.

20 Luhder F, Hoglund P, Allison JP, Benoist C, Mathis D. Cytotoxic Tlymphocyte-associated antigen 4 (CTLA-4) regulates the unfoldingof autoimmune diabetes. J Exp Med 1998;187:427–32.

21 Yanagawa T, Hidaka Y, Guimaraes V, Soliman M, Degroot LJ.CTLA-4 gene polymorphism associated with Graves’ disease in aCaucasian population. J Clin Endocrinol Metab 1995;80:41–5.

22 Donner H, Rau H, Walfish PG, Braun J, Siegmund T, Finke R,Herwig J, Usadel KH, Badenhoop K. CTLA-4 analine-17 confersgenetic susceptibility to Graves’ disease and to type 1 diabetesmellitus. J Clin Endocrinol Metab 1997;82:143–6.

23 Verwilghen J, Lovis R, De Bor M, Linsley PS, Haines GK, Koch AE,Pope RM. Expression of functional B7 and CTLA-4 on rheumatoidsynovial T cells. J Immunol 1994;153:1378–85.

24 Feldmann M, Brennan FM, Maini RN. Role of cytokines in rheu-matoid arthritis. Annu Rev Immunol 1996;14:387–440.

25 Gribben JG, Freeman GJ, Boussiotis VAet al. CTLA-4 mediatesantigen-specific apoptosis of human T cells. Proc Natl Acad Sci USA1995;92:811–15.

72 M.-F. Liu et al.

q 1999 Blackwell Science Ltd,Scandinavian Journal of Immunology, 50, 68–72