Anti-TcR antibody inhibits collagen-induced arthritis
Eur. J. Immunol. 1991. 21: 1327-1330
1327
Short paper Tom J. Goldschmidton and Rikard HolmdahlO Department of Medical and Physiological Chemistry, University of UppsalaO and Department of Inflammation Research, Pharmacia Therapeuticso, UppsaIa
Anti-T cell receptor antibody treatment of rats with established autologous collagen-induced arthritis: suppression of arthritis without reduction of anti-type I1 collagen autoantibody levels* Activation of T cells is critical for the development of type I1 collagen (CI1)-induced arthritis (CIA). However, the relative importance of Tcells in their delivery of help to B cells, promoting autoantibody formation or acting as inflammatory initiating cells, is unclear. The effect of a monoclonal antibody directed to the a / p Tcell receptor (TcR) on the development of autologous CIA was studied. Two weeks after immunization with autologous CII the onset of severe arthritis occurred, followed by a chronic arthritis activity in the peripheral joints. Anti-TcR treatment before immunization suppressed the incidence of arthritis and the autoantibody response to CII. Treatment given immediately before the expected onset delayed the appearance of arthritis. Treatment given to already arthritic rats reduced the severity. In the latter two groups the serum levels of anti-CII autoantibodies were not affected. The duration of the ameliorating effect was limited and with the return of arthritis a concomitant antibody response towards the injected mouse anti-TcR antibody was observed. These results show that the role of Tcells in both the induction and perpetuation of CIA is essential and not limited to the triggering of production of pathogenic anti-CII autoantibodies.
1 Introduction
when applied before CII immunization but not when applied at a later stage of the disease. These findings have Type I1 collagen (CI1)-induced arthritis (CIA) is an organ- given support for the suggestion that the main role of Tcells specific autoimmune disease in experimental animals which in CIA is to give help to B cells producing arthritogenic shares certain clinical, histopathological and immunologi- anti-CII autoantibodies. However, although it is likely that cal features with rheumatoid arthritis (R4) in humans. CIA anti-CII autoantibodies play an important role in CIA, the has been induced in rats, mice and apes after immunization pathogenesis may be more complex. Histopathological with CII, the major protein of articular cartilage [1-3].The analysis show evidence for both immune complexdevelopment of CIA is dependent on the presence of mediated mechanisms and DTH-like inflammation. The functional T cells [4, 51 and is genetically associated with importance of cellular immunity has also been implied by the expression of certain MHC class11 molecules [6] transfer experiments [13]. Moreover, the occurrence and suggesting that the triggering of MHC-restricted CD4+ T severity of arthritic disease is not always correlated with cells is a critical step. Immunization with CII also triggers B high levels of anti-CII autoantibodies [14]. cells producing anti-CII autoantibodies [7] and a correlation between serum levels of anti-CII antibodies and The recent isolation of an antibody (R73) to the a / p antigen arthritic diseases has been suggested [8]. Direct pathoge- receptor on rat T cells [15] provided a new possibility for nicity of humoral components has been demonstrated by interfering critically with the function of activated T cells. the successful induction of arthritis after injection of Ameliorating effects of R73 have recently been demonanti-CII serum [9] and purified anti-CII serum antibodies strated in adjuvant arthritis [16]. In the present study the [lo] into naive rats. T cell-suppressive regimens, such as contribution of Tcells to different stages of autologous CIA anti-Tcell antibodies [ 5 , 111 or cyclosporin [12] treatment, in the DA rat was analyzed by performing prophylactic and inhibit both arthritis and the anti-CII antibody response therapeutic treatments with the anti-TcR a/@ mAb R73. Autologous CII was used to induce CIA since the autolo[I 91581 gous more than the heterologous CIA may represent an experimental model relevant for a self-perpetuative * The study was supported by grants from the Swedish Medical autoimmune disease such as RA [17]. Research Council, the Craaford Foundation, King Gustav V s 80-years Foundation, Riksforbundet mot Reumatism and Professor Nanna Svartz Research Foundation.
Correspondence: Tom Goldxhmidt, Department of Medical and Physiological Chemistry, University of Uppsala, Box 575, S-75123 Uppsala, Sweden
Abbreviations: CIA: Collagen-induced arthritis CII: S p e I1 collagen RA: Rheumatoid arthritis 0 VCH Verlagsgesellschaft mbH, D-6940 Weinheim, 1991
2 Materials and methods 2.1 Induction and evaluation of CIA DA rats originally obtained from Olac (Bicester, GB) were bred and kept at the Biomedical Center in Uppsala. All experiments were performed on male rats at an age of 0014-2980/91/0505-1327$3.50+ .25/0
1328
Eur. J. Immunol. 19Y1. 21: 1327-1330
T. J. Goldschmidt and R. Holmdahl
8-12 weeks. Native rat CII was prepared as earlier described [ 181. For immunization, native CII was dissolved in 0.1 M acetic acid at 4°C and emulsified 1: 1on ice with IFA (Difco, Detroit, MI) to a final concentration of 1 mg/ml [except for the thymectomy study where CFA (Difco) was used]. Rats were injected i.d. at the base of the tail with 150 ~l of the emulsion. The arthritis development was followed by a macroscopic scoring system for the four paws ranging from 0 to 3 (1 = swelling andor redness of one toe or finger joint, 2 = two or more joints involved and 3 = severe arthritis in the entire paw). Furthermore, to be able to evaluate the arthritis development in more detail, an additional scoring system was employed where the number of involved joints were further discriminated. Arthritic fingers and toes were given one point each, intercarpal and intertarsal joints up to four points and the wrist and ankle one point each so the maximal score per limb was ten. Adult thymectomy was done under pentobarbital anesthesia (Mebumal, ACO Lakemedel, Goteborg, Sweden)and the rats were allowed to recover for 2 weeks after surgery before the experiments.
2.2 Antibody treatment The hybridoma R73, a mouse anti-rat TcR a / p antibody,was a kind gift of Dr. Thomas Hunig, Miinchen, FRG [15], Hy-2.15, a mouse IgGl which recognizeTNP was used as a control antibody in the in viva treatments [19]. The antibodies were affinity purified on protein A-Sepharose (Pharmacia AB, Uppsala, Sweden), dialyzed with PBS and sterile filtered. The protein content was quantified by absorbance measurement at 280 nm. The antibodies were injected i.p. in doses of 500 yg given on two consecutive days. This treatment was repeated in 6-day intervals.
2.3 Quantification of antibodies in serum Sera were collected individually and stored at -20°C until assayed. For the quantification of anti-CII-reactive autoantibodies in serum, a modified standard ELISA technique was used [20]. In brief, micro-ELISA plates (Dynatech, Plochingen, FRG) were coated with 10 pg/ml of native rat CII. All tests were carried out in duplicate, and the standard deviations did not exceed 10%. The amount of bound antibody was estimated after incubation with a mouse serum adsorbed goat anti-rat IgG (H + L) affinitypurified antibody conjugated to alkaline phosphatase (Jackson Immunoresearch Laboratories,West Grove, PA). The subsequent quantification of bound enzyme was performed with a paranitrophenol containing substrate buffer in a Titertek Multiscan spectrophotometer. To estimate the amount of anti-CII-reactive antibodies present in the serum samples, affinity-purified rat anti-CII-reactive antibodies were used as standards.
3 Results 3.1 Both prophylactic and therapeutic anti-TcRalp treatment suppressed development of arthritis
DA male rats were immunized with autologous CII and developed arthritis 12-13 days after immunization and
reached full incidence (100%)within 1week after onset. The rats were treated with anti-TcR antibodies at three different stages of arthritis, starting: (a) immediately after immunization (injections given day 0, 1, 6, 7, 12, 13, 18, 19); (b) immediately before the expected onset of arthritis (injections given day 12, 13,18, 19,24,25); or (c) when all rats had become arthritic (injections given day 15, 16, 21, 22, 27, 28, 34, 35) (Fig. 1). The induction of arthritis was completely blocked by anti-TcR treatment given from immunization. A similar efficient blocking of arthritis development was achieved even when the treatment started immediately before the expected onset of arthritis, which is after the point of time when both anti-CII Tcell activation and antibody production can be first seen [7, 11, 171. However, the effect was only transient and a delayed onset of severe arthritis developed after termination of the anti-TcR antibody treatment. When anti-TcR treatment was initiated after the arthritis had been fully established the arthritic severity was reduced. A significant reduction of severity was already seen a few days after the first injection of anti-TcR antibodies. Involvement of new joints was blocked and recovery was seen in nearly half the number of the inflamed joints, both 100
80 u) c
m
.0 _
60
L .L
r, m
40
20
0 0
10
20
30
40
50
Days after immunization
Figure 1. Incidence (A) and severity (B) of arthritis in male rats after immunization with rat CII and treatment with the anti-TcR antibody R73 or control antibody Hy2.15 at different stages of disease (CIA). ( 0 )Represents the rats treated with 500 pg R73 or (0)Hy2.15 as i.p. injections day (d) 0, 1, 6, 7, 12, 13 and 18, 19 ( n = 5) (0)R73 or (A) Hy2.15 d 12, 13, 18, 19 and 24,25 ( n = 5); (m) R73 treatment 15, 16, 21, 22, 27, 28 and 33, 34 ( n = 9). The mean arthritic score in the group R73 d 15-34 was significantly lower than in Hy2.15 d 12-25; p < 0.05 d 16,p < 0.01 d 18-41, and in Hy2.15 d 0-19; p < 0.05 d 21, p < 0.01 d 24-33, as evaluated by Mann-Whitney test.
Anti-TcR antibody inhibits collagen-induced arthritis
Eur. J. Immunol. 1991. 21: 1327-1330
mild and severely affected (Fig. 2), while in the control group no recovery at all was seen during this period.Two of the arthritic individuals were totally recovered from visible arthritis.This therapeutical blockage of further progression
1329
of the inflammatory process was observed both as diminished swelling and redness, but also as an improved functional appearance. However, in several joints where the erythema disappeared, a considerable deformation persisted, probably because the arthritis already had caused deformation of the connective tissues. The persisting tissue deformations was nevertheless scored and probably resulted in an overestimated degree of severity.
3.2 The anti-CII antibody response was reduced only when anti-TcR antibody treatment started at immunization but not later The levels of anti-rat CII IgG in serum was measured by ELISA technique day 21 after immunization, during the period of arthritis inhibition, and found to be impaired only if the anti-TcR antibody treatment were started at immunization (Fig. 3). In the same sera, a strong anti-R73 response was registered in all R73-treated groups as measured by ELISA (data not shown). '
Days after immunization
3.3 Adult thymectomy did not prolong the anti-TcR a / p inhibition of CIA
Figure 2. Histogram showing a detailed scoring of newly involved or recovered individual joints for rats given anti-TcR antibody from day 15 after CII immunization, the same individuals as presented in Fig. 1 (n = 9). Each toe was given one point, the intratarsal and intracarpal joints 0-4 and ankle and wrist one point each. The maximal score was 10 points per paw. For intervals spanning over more than 1 day the scoring points were equally spread to the days within the interval.The start of the anti-TcR treatment is indicated by the arrow. Among the rats receiving control antibodies (Hy2.15) from day 0 (n = 9) or 12 (n = 9), no recovery was seen during this period.
To investigate whether the delayed appearance of arthritis in anti-TcR-treated rats was dependent on newly matured T cells migrating from thymus, rats were thymectomized before the next experiment. Treatment with anti-TcR or control antibodies were given on individual onset day and repeated 1week later. In this case the rats were immunized with CII emulsified in CFA, resulting in a stronger adjuvant effect and a more severe disease. Also in this experiment the inhibition was only transient showing that new thymus emigrants were not important for the reappearance of arthritis (Fig. 4).
Hy do-19 Hy do-13
R73 do-19
E
R73 do-13
2
%
Hy d l 2-25
.-c
Hy d12-19
R73 d12-25 R72 d12-19 R73 d15-34
' r 00
AA
Treatments
Figure 3. The anti-rat CII antibodies concentrations in serum day (d) 21 after immunization with native rat CII as measured with ELISA. The groups treated with anti-RR (R73) or control antibodies (Hy2.15) d 0-19 or d 12-25 (n = 5) and R73 d15-34 (n = 9) are the same as presented in Fig. 1. For R73 d 0-13 or d 12-19, n = 5 and for Hy2.15 d 0-13 or d 12-19, n = 4. * Indicates significant lower anti-CII levels than in the groups treated with Hy2.15 from day 0 (pSO.O5), as tested by Student's t-test.
5
10
15
20
AA
Days after onset
Figure 4. Severity of arthritis in thymectomized rats. Male DA rats, 8-10 weeks old, were thymectomized 2 weeks prior to immunization with rat CII. Intraperitoneal injections with anti-TcR (R73) (0)or control antibodies (Hy2.15 ( 0 )in doses of 500 pg were given on day 0, 1 , 6 and 7 after individual onset day, indicated by arrows. R73 day after onset (d) 0-14, n = 14; d 16, n = 11; d 19, n = 10 and Hy2.15 d 0-14, n = 14; d 16, n = 13;d 19, n = 12. d 2-16 p < 0.01 except for day 3 p < 0.05.
1330
T. J. Goldschmidt and R. Holmdahl
4 Discussion In the present investigation anti-TcR a$ antibodies were used to study the role of Tcells at different stages of CIA. As expected, treatment during the immunization period blocked both arthritis development and production of anti-CII autoantibodies. Of importance for the understanding of the role of Tcells in the CIA model is, however, the finding that anti-TcR treatment initiated immediately before the expected onset of arthritis was similarly effective in blocking arthritis but did not affect the levels of anti-CII autoantibodies. In addition, antibody treatment commenced at day 15,when all rats had severe arthritis, rapidly decreased the severity of arthritis. The potent effect of the anti-TcR antibody treatment was more likely due to functional blockage of the receptor than to elimination of the target cells. By FCM analyses we have found that, although large numbers of T cells are eliminated, significant numbers remain carrying anti-TcR antibodies on their surface (data not shown). Functional blocking is further indicated in the thymectomy experiment, in which the subsequent progression of the arthritis did not seem to be dependent onTcells newly matured after the inhibition period. An increased resistance to antibodymediated elimination of activated T cells may explain the difficulties in treating autoimmune disease therapeutically in experimental models [21]. In agreement with this, a refractory behavior of activated T cells has indeed been observed in the CIA model. In the mouse we recently found that CII-specific CD4+ T cells were resistant to anti-CD4 mAb if the treatment was given after, but not before, immunization (manuscript in preparation). Combined treatment with anti-CD4 and anti-Thy-1 antibodies may, however, have some effects on the long-term progression of established arthritis in mice as reported by Horn et al. [22] and it seemed as if a nearly complete elimination of Tcells was crucial in that experiment. In the present study functional blockage of T cells could explain the transient effect which was followed by a rapid appearance of arthritis onset.The duration of the anti-TcR antibodies was probably shortened by the high titers of antibodies reactive with the mouse anti-TcR antibodies that appeared in serum (data not shown). The inflammatory tissues in affected joints contains only few Tcells and the dominating inflammatory cells consist of MQ, fibroblasts and infiltrating granulocytes [23,24]. Moreover, the autoreactiveTcel1 response has been difficult to measure directly since it gives a poor proliferative response in vitro [25].Nevertheless, in the present study we show for the first time that functional blockage of a/@Tcells in an autologous CIA model effectively inhibits arthritis both prophylactically and therapeutically, indicating that T cells reactive with autologous CII play a crucial role both for initiation and maintenance of arthritis. The study does not, however, rule out an important enhancing role of CII-reactive B cells or anti-CII autoantibodies in the disease process. Experiments in both rats and mice have suggested a cooperative effect of anti-CII autoantibodies and Tcells [26,27]. In fact, a plausible explanation for the ease with which CII autoimmunity leads to chronic arthritis may be that anti-CII reactive B cells function as APC, thereby breaking the tolerance of autoreactive Tcells [28]. Besides the mechanistic information obtained from the
Eur. J. Immunol. 1991. 21: 1327-1330
usage of anti-TcR treatment of CIA in rats, the treatment could also provide a promising approach for future therapy of RA. We thank Christina Kvick f o r excellent technical assistance. Received December 27, 1990.
5 References 1 Trentham, D. E.,Townes, A. S. and Kang, A. H., J. Exp. Med. 1977. 146: 857. 2 Courtenay, J. S., Dallman, M. J., Dayan, A. D., Martin, A. and Mosedal, B., Nature 1980. 283: 666. 3 Cathcart, E. S., Hayes, K. C., Gonnerman,W. A., Lazzari, A. A. and Franzblau, C., Lab. Invest. 1986. 54: 26. 4 Klareskog, L., Holmdahl, R., Larsson, E. and Wigzell, H., Clin. Exp. lmmunol. 1983. 51: 117. 5 Ranges, G . E., Sriram, S. and Cooper, S. M., J. Exp. Med. 1985. 162: 1105. 6 Griffiths, M. M., Eichwald, E. J., Martin, J. M., Smith, C. B. and DeWitt, C. W., Arthritis Rheum. 1981. 24: 781. 7 Trentham, D. E.,Townes, A. S., Kang, A. H. and David, J. R., J. Clin. Invest. 1978. 61: 89. 8 Stuart, J. M., Townes, A. S. and Kang, A. H., Annu. Rev. lmmunol. 1984. 2: 199. 9 Stuart, J. M., Cremer, M. A.,Townes, A. S. and Kang, A. H., J. Exp. Med. 1982. 155: 1. 10 Kerwar, S. S., Englert, M. E., McReynolds, R. A., Landes, M. J., Lloyd, J. M., Oronsky, A. L. and Wilson, F. J., Arthritis Rheum. 1983. 26: 1120. 11 Brahn, E. and Trentham, D. E., Cell. lmmunol. 1984. 86: 421. 12 Kaibara, N., Hotokebuchi, T. ,Takagishi, K. and Katsuki, I., J. Exp. Med. 1983. 158: 2007. 13 Trentham, D. E., Dynesius, R. A. and David, J. R., J. Clin. Invest. 1978. 62: 359. 14 Carlson, R. P., Blazek, E. M., Datko, L. J. and Lewis, A. J., J. lmmunopharmacol. 1984. 6: 379. 15 Hiinig, T., Wallny, H.-J., Hartley, J. K., Lawetzky, A. and Tiefenthaler, G., J. Exp. Med. 1989. 169: 73. 16 Yoshino, S., Schlipkoter, E., Kinne, R., Hiinig,T. and Emmrich, F., Eur. J. lmmunol. 1990. 20: 2805. 17 Larsson, €?,Kleinau, S., Holmdahl, R. and Klareskog, L., Arthritis Rheum. 1990. 33: 693. 18 Miller, E. J. and Rhodes, R.K.,Methods Enzymol. 1982. 82: 33. 19 Brideau, R. J., Carter, €! B., McMaster,W. R., Mason, D.W. and Williams, A., Eur. J. lmmunol. 1980. 10: 609. 20 Larsson, €? and Holmdahl, R., Scand. J. lmmunol. 1987. 27: 579. 21 Ranges, G. E., Cooper, S. M. and Sriram, S., Cell. lmmunol. 1987. 106: 163. 22 Hom, J.T., Butler, L. D., Riedl, €? E. and Bendele, A. M., Eur. J. lmmunol. 1988. 18: 881. 23 Holmdahl, R., Rubin, K., Klareskog, L., Dencker, L., Gustafsson, G. and Larsson, E., Scand. J. lmmunol. 1985. 21: 197. 24 Quinn Dejoy, S., Ferguson, K. M., Oronsky, A. L. and Kerwar, S. S., Cell. lmmunol. 1988. 113: 117. 25 Andersson, M. and Holmdahl, R., Eur. J. lmmunol. 1990.20: 1061. 26 Taurog, J. D., Kerwar, S. S., McReynolds, R. A., Sandberg, G. I?, Leary, S. L. and Mahowald, M. L., J. Exp. Med. 1985.162: 962. 27 Seki, N., Sudo, Y., Yoshioka, T., Sugihara, S., Fujitsu, T.. Sakuma, S., Ogawa,T., Hamaoka,T., Senoh, H. and Fujiwara. H., J. lmmunol. 1988. 140: 1477. 28 Holmdahl, R., Andersson, M., Go1dschmidt.T. J., Gustavsson, K., Jansson, L. and Mo, J. A., lmmunol. Rev. 1990. 118: 193.
Eur. J. Immunol. 1991. 21: 1327-1330
1327
Short paper Tom J. Goldschmidton and Rikard HolmdahlO Department of Medical and Physiological Chemistry, University of UppsalaO and Department of Inflammation Research, Pharmacia Therapeuticso, UppsaIa
Anti-T cell receptor antibody treatment of rats with established autologous collagen-induced arthritis: suppression of arthritis without reduction of anti-type I1 collagen autoantibody levels* Activation of T cells is critical for the development of type I1 collagen (CI1)-induced arthritis (CIA). However, the relative importance of Tcells in their delivery of help to B cells, promoting autoantibody formation or acting as inflammatory initiating cells, is unclear. The effect of a monoclonal antibody directed to the a / p Tcell receptor (TcR) on the development of autologous CIA was studied. Two weeks after immunization with autologous CII the onset of severe arthritis occurred, followed by a chronic arthritis activity in the peripheral joints. Anti-TcR treatment before immunization suppressed the incidence of arthritis and the autoantibody response to CII. Treatment given immediately before the expected onset delayed the appearance of arthritis. Treatment given to already arthritic rats reduced the severity. In the latter two groups the serum levels of anti-CII autoantibodies were not affected. The duration of the ameliorating effect was limited and with the return of arthritis a concomitant antibody response towards the injected mouse anti-TcR antibody was observed. These results show that the role of Tcells in both the induction and perpetuation of CIA is essential and not limited to the triggering of production of pathogenic anti-CII autoantibodies.
1 Introduction
when applied before CII immunization but not when applied at a later stage of the disease. These findings have Type I1 collagen (CI1)-induced arthritis (CIA) is an organ- given support for the suggestion that the main role of Tcells specific autoimmune disease in experimental animals which in CIA is to give help to B cells producing arthritogenic shares certain clinical, histopathological and immunologi- anti-CII autoantibodies. However, although it is likely that cal features with rheumatoid arthritis (R4) in humans. CIA anti-CII autoantibodies play an important role in CIA, the has been induced in rats, mice and apes after immunization pathogenesis may be more complex. Histopathological with CII, the major protein of articular cartilage [1-3].The analysis show evidence for both immune complexdevelopment of CIA is dependent on the presence of mediated mechanisms and DTH-like inflammation. The functional T cells [4, 51 and is genetically associated with importance of cellular immunity has also been implied by the expression of certain MHC class11 molecules [6] transfer experiments [13]. Moreover, the occurrence and suggesting that the triggering of MHC-restricted CD4+ T severity of arthritic disease is not always correlated with cells is a critical step. Immunization with CII also triggers B high levels of anti-CII autoantibodies [14]. cells producing anti-CII autoantibodies [7] and a correlation between serum levels of anti-CII antibodies and The recent isolation of an antibody (R73) to the a / p antigen arthritic diseases has been suggested [8]. Direct pathoge- receptor on rat T cells [15] provided a new possibility for nicity of humoral components has been demonstrated by interfering critically with the function of activated T cells. the successful induction of arthritis after injection of Ameliorating effects of R73 have recently been demonanti-CII serum [9] and purified anti-CII serum antibodies strated in adjuvant arthritis [16]. In the present study the [lo] into naive rats. T cell-suppressive regimens, such as contribution of Tcells to different stages of autologous CIA anti-Tcell antibodies [ 5 , 111 or cyclosporin [12] treatment, in the DA rat was analyzed by performing prophylactic and inhibit both arthritis and the anti-CII antibody response therapeutic treatments with the anti-TcR a/@ mAb R73. Autologous CII was used to induce CIA since the autolo[I 91581 gous more than the heterologous CIA may represent an experimental model relevant for a self-perpetuative * The study was supported by grants from the Swedish Medical autoimmune disease such as RA [17]. Research Council, the Craaford Foundation, King Gustav V s 80-years Foundation, Riksforbundet mot Reumatism and Professor Nanna Svartz Research Foundation.
Correspondence: Tom Goldxhmidt, Department of Medical and Physiological Chemistry, University of Uppsala, Box 575, S-75123 Uppsala, Sweden
Abbreviations: CIA: Collagen-induced arthritis CII: S p e I1 collagen RA: Rheumatoid arthritis 0 VCH Verlagsgesellschaft mbH, D-6940 Weinheim, 1991
2 Materials and methods 2.1 Induction and evaluation of CIA DA rats originally obtained from Olac (Bicester, GB) were bred and kept at the Biomedical Center in Uppsala. All experiments were performed on male rats at an age of 0014-2980/91/0505-1327$3.50+ .25/0
1328
Eur. J. Immunol. 19Y1. 21: 1327-1330
T. J. Goldschmidt and R. Holmdahl
8-12 weeks. Native rat CII was prepared as earlier described [ 181. For immunization, native CII was dissolved in 0.1 M acetic acid at 4°C and emulsified 1: 1on ice with IFA (Difco, Detroit, MI) to a final concentration of 1 mg/ml [except for the thymectomy study where CFA (Difco) was used]. Rats were injected i.d. at the base of the tail with 150 ~l of the emulsion. The arthritis development was followed by a macroscopic scoring system for the four paws ranging from 0 to 3 (1 = swelling andor redness of one toe or finger joint, 2 = two or more joints involved and 3 = severe arthritis in the entire paw). Furthermore, to be able to evaluate the arthritis development in more detail, an additional scoring system was employed where the number of involved joints were further discriminated. Arthritic fingers and toes were given one point each, intercarpal and intertarsal joints up to four points and the wrist and ankle one point each so the maximal score per limb was ten. Adult thymectomy was done under pentobarbital anesthesia (Mebumal, ACO Lakemedel, Goteborg, Sweden)and the rats were allowed to recover for 2 weeks after surgery before the experiments.
2.2 Antibody treatment The hybridoma R73, a mouse anti-rat TcR a / p antibody,was a kind gift of Dr. Thomas Hunig, Miinchen, FRG [15], Hy-2.15, a mouse IgGl which recognizeTNP was used as a control antibody in the in viva treatments [19]. The antibodies were affinity purified on protein A-Sepharose (Pharmacia AB, Uppsala, Sweden), dialyzed with PBS and sterile filtered. The protein content was quantified by absorbance measurement at 280 nm. The antibodies were injected i.p. in doses of 500 yg given on two consecutive days. This treatment was repeated in 6-day intervals.
2.3 Quantification of antibodies in serum Sera were collected individually and stored at -20°C until assayed. For the quantification of anti-CII-reactive autoantibodies in serum, a modified standard ELISA technique was used [20]. In brief, micro-ELISA plates (Dynatech, Plochingen, FRG) were coated with 10 pg/ml of native rat CII. All tests were carried out in duplicate, and the standard deviations did not exceed 10%. The amount of bound antibody was estimated after incubation with a mouse serum adsorbed goat anti-rat IgG (H + L) affinitypurified antibody conjugated to alkaline phosphatase (Jackson Immunoresearch Laboratories,West Grove, PA). The subsequent quantification of bound enzyme was performed with a paranitrophenol containing substrate buffer in a Titertek Multiscan spectrophotometer. To estimate the amount of anti-CII-reactive antibodies present in the serum samples, affinity-purified rat anti-CII-reactive antibodies were used as standards.
3 Results 3.1 Both prophylactic and therapeutic anti-TcRalp treatment suppressed development of arthritis
DA male rats were immunized with autologous CII and developed arthritis 12-13 days after immunization and
reached full incidence (100%)within 1week after onset. The rats were treated with anti-TcR antibodies at three different stages of arthritis, starting: (a) immediately after immunization (injections given day 0, 1, 6, 7, 12, 13, 18, 19); (b) immediately before the expected onset of arthritis (injections given day 12, 13,18, 19,24,25); or (c) when all rats had become arthritic (injections given day 15, 16, 21, 22, 27, 28, 34, 35) (Fig. 1). The induction of arthritis was completely blocked by anti-TcR treatment given from immunization. A similar efficient blocking of arthritis development was achieved even when the treatment started immediately before the expected onset of arthritis, which is after the point of time when both anti-CII Tcell activation and antibody production can be first seen [7, 11, 171. However, the effect was only transient and a delayed onset of severe arthritis developed after termination of the anti-TcR antibody treatment. When anti-TcR treatment was initiated after the arthritis had been fully established the arthritic severity was reduced. A significant reduction of severity was already seen a few days after the first injection of anti-TcR antibodies. Involvement of new joints was blocked and recovery was seen in nearly half the number of the inflamed joints, both 100
80 u) c
m
.0 _
60
L .L
r, m
40
20
0 0
10
20
30
40
50
Days after immunization
Figure 1. Incidence (A) and severity (B) of arthritis in male rats after immunization with rat CII and treatment with the anti-TcR antibody R73 or control antibody Hy2.15 at different stages of disease (CIA). ( 0 )Represents the rats treated with 500 pg R73 or (0)Hy2.15 as i.p. injections day (d) 0, 1, 6, 7, 12, 13 and 18, 19 ( n = 5) (0)R73 or (A) Hy2.15 d 12, 13, 18, 19 and 24,25 ( n = 5); (m) R73 treatment 15, 16, 21, 22, 27, 28 and 33, 34 ( n = 9). The mean arthritic score in the group R73 d 15-34 was significantly lower than in Hy2.15 d 12-25; p < 0.05 d 16,p < 0.01 d 18-41, and in Hy2.15 d 0-19; p < 0.05 d 21, p < 0.01 d 24-33, as evaluated by Mann-Whitney test.
Anti-TcR antibody inhibits collagen-induced arthritis
Eur. J. Immunol. 1991. 21: 1327-1330
mild and severely affected (Fig. 2), while in the control group no recovery at all was seen during this period.Two of the arthritic individuals were totally recovered from visible arthritis.This therapeutical blockage of further progression
1329
of the inflammatory process was observed both as diminished swelling and redness, but also as an improved functional appearance. However, in several joints where the erythema disappeared, a considerable deformation persisted, probably because the arthritis already had caused deformation of the connective tissues. The persisting tissue deformations was nevertheless scored and probably resulted in an overestimated degree of severity.
3.2 The anti-CII antibody response was reduced only when anti-TcR antibody treatment started at immunization but not later The levels of anti-rat CII IgG in serum was measured by ELISA technique day 21 after immunization, during the period of arthritis inhibition, and found to be impaired only if the anti-TcR antibody treatment were started at immunization (Fig. 3). In the same sera, a strong anti-R73 response was registered in all R73-treated groups as measured by ELISA (data not shown). '
Days after immunization
3.3 Adult thymectomy did not prolong the anti-TcR a / p inhibition of CIA
Figure 2. Histogram showing a detailed scoring of newly involved or recovered individual joints for rats given anti-TcR antibody from day 15 after CII immunization, the same individuals as presented in Fig. 1 (n = 9). Each toe was given one point, the intratarsal and intracarpal joints 0-4 and ankle and wrist one point each. The maximal score was 10 points per paw. For intervals spanning over more than 1 day the scoring points were equally spread to the days within the interval.The start of the anti-TcR treatment is indicated by the arrow. Among the rats receiving control antibodies (Hy2.15) from day 0 (n = 9) or 12 (n = 9), no recovery was seen during this period.
To investigate whether the delayed appearance of arthritis in anti-TcR-treated rats was dependent on newly matured T cells migrating from thymus, rats were thymectomized before the next experiment. Treatment with anti-TcR or control antibodies were given on individual onset day and repeated 1week later. In this case the rats were immunized with CII emulsified in CFA, resulting in a stronger adjuvant effect and a more severe disease. Also in this experiment the inhibition was only transient showing that new thymus emigrants were not important for the reappearance of arthritis (Fig. 4).
Hy do-19 Hy do-13
R73 do-19
E
R73 do-13
2
%
Hy d l 2-25
.-c
Hy d12-19
R73 d12-25 R72 d12-19 R73 d15-34
' r 00
AA
Treatments
Figure 3. The anti-rat CII antibodies concentrations in serum day (d) 21 after immunization with native rat CII as measured with ELISA. The groups treated with anti-RR (R73) or control antibodies (Hy2.15) d 0-19 or d 12-25 (n = 5) and R73 d15-34 (n = 9) are the same as presented in Fig. 1. For R73 d 0-13 or d 12-19, n = 5 and for Hy2.15 d 0-13 or d 12-19, n = 4. * Indicates significant lower anti-CII levels than in the groups treated with Hy2.15 from day 0 (pSO.O5), as tested by Student's t-test.
5
10
15
20
AA
Days after onset
Figure 4. Severity of arthritis in thymectomized rats. Male DA rats, 8-10 weeks old, were thymectomized 2 weeks prior to immunization with rat CII. Intraperitoneal injections with anti-TcR (R73) (0)or control antibodies (Hy2.15 ( 0 )in doses of 500 pg were given on day 0, 1 , 6 and 7 after individual onset day, indicated by arrows. R73 day after onset (d) 0-14, n = 14; d 16, n = 11; d 19, n = 10 and Hy2.15 d 0-14, n = 14; d 16, n = 13;d 19, n = 12. d 2-16 p < 0.01 except for day 3 p < 0.05.
1330
T. J. Goldschmidt and R. Holmdahl
4 Discussion In the present investigation anti-TcR a$ antibodies were used to study the role of Tcells at different stages of CIA. As expected, treatment during the immunization period blocked both arthritis development and production of anti-CII autoantibodies. Of importance for the understanding of the role of Tcells in the CIA model is, however, the finding that anti-TcR treatment initiated immediately before the expected onset of arthritis was similarly effective in blocking arthritis but did not affect the levels of anti-CII autoantibodies. In addition, antibody treatment commenced at day 15,when all rats had severe arthritis, rapidly decreased the severity of arthritis. The potent effect of the anti-TcR antibody treatment was more likely due to functional blockage of the receptor than to elimination of the target cells. By FCM analyses we have found that, although large numbers of T cells are eliminated, significant numbers remain carrying anti-TcR antibodies on their surface (data not shown). Functional blocking is further indicated in the thymectomy experiment, in which the subsequent progression of the arthritis did not seem to be dependent onTcells newly matured after the inhibition period. An increased resistance to antibodymediated elimination of activated T cells may explain the difficulties in treating autoimmune disease therapeutically in experimental models [21]. In agreement with this, a refractory behavior of activated T cells has indeed been observed in the CIA model. In the mouse we recently found that CII-specific CD4+ T cells were resistant to anti-CD4 mAb if the treatment was given after, but not before, immunization (manuscript in preparation). Combined treatment with anti-CD4 and anti-Thy-1 antibodies may, however, have some effects on the long-term progression of established arthritis in mice as reported by Horn et al. [22] and it seemed as if a nearly complete elimination of Tcells was crucial in that experiment. In the present study functional blockage of T cells could explain the transient effect which was followed by a rapid appearance of arthritis onset.The duration of the anti-TcR antibodies was probably shortened by the high titers of antibodies reactive with the mouse anti-TcR antibodies that appeared in serum (data not shown). The inflammatory tissues in affected joints contains only few Tcells and the dominating inflammatory cells consist of MQ, fibroblasts and infiltrating granulocytes [23,24]. Moreover, the autoreactiveTcel1 response has been difficult to measure directly since it gives a poor proliferative response in vitro [25].Nevertheless, in the present study we show for the first time that functional blockage of a/@Tcells in an autologous CIA model effectively inhibits arthritis both prophylactically and therapeutically, indicating that T cells reactive with autologous CII play a crucial role both for initiation and maintenance of arthritis. The study does not, however, rule out an important enhancing role of CII-reactive B cells or anti-CII autoantibodies in the disease process. Experiments in both rats and mice have suggested a cooperative effect of anti-CII autoantibodies and Tcells [26,27]. In fact, a plausible explanation for the ease with which CII autoimmunity leads to chronic arthritis may be that anti-CII reactive B cells function as APC, thereby breaking the tolerance of autoreactive Tcells [28]. Besides the mechanistic information obtained from the
Eur. J. Immunol. 1991. 21: 1327-1330
usage of anti-TcR treatment of CIA in rats, the treatment could also provide a promising approach for future therapy of RA. We thank Christina Kvick f o r excellent technical assistance. Received December 27, 1990.
5 References 1 Trentham, D. E.,Townes, A. S. and Kang, A. H., J. Exp. Med. 1977. 146: 857. 2 Courtenay, J. S., Dallman, M. J., Dayan, A. D., Martin, A. and Mosedal, B., Nature 1980. 283: 666. 3 Cathcart, E. S., Hayes, K. C., Gonnerman,W. A., Lazzari, A. A. and Franzblau, C., Lab. Invest. 1986. 54: 26. 4 Klareskog, L., Holmdahl, R., Larsson, E. and Wigzell, H., Clin. Exp. lmmunol. 1983. 51: 117. 5 Ranges, G . E., Sriram, S. and Cooper, S. M., J. Exp. Med. 1985. 162: 1105. 6 Griffiths, M. M., Eichwald, E. J., Martin, J. M., Smith, C. B. and DeWitt, C. W., Arthritis Rheum. 1981. 24: 781. 7 Trentham, D. E.,Townes, A. S., Kang, A. H. and David, J. R., J. Clin. Invest. 1978. 61: 89. 8 Stuart, J. M., Townes, A. S. and Kang, A. H., Annu. Rev. lmmunol. 1984. 2: 199. 9 Stuart, J. M., Cremer, M. A.,Townes, A. S. and Kang, A. H., J. Exp. Med. 1982. 155: 1. 10 Kerwar, S. S., Englert, M. E., McReynolds, R. A., Landes, M. J., Lloyd, J. M., Oronsky, A. L. and Wilson, F. J., Arthritis Rheum. 1983. 26: 1120. 11 Brahn, E. and Trentham, D. E., Cell. lmmunol. 1984. 86: 421. 12 Kaibara, N., Hotokebuchi, T. ,Takagishi, K. and Katsuki, I., J. Exp. Med. 1983. 158: 2007. 13 Trentham, D. E., Dynesius, R. A. and David, J. R., J. Clin. Invest. 1978. 62: 359. 14 Carlson, R. P., Blazek, E. M., Datko, L. J. and Lewis, A. J., J. lmmunopharmacol. 1984. 6: 379. 15 Hiinig, T., Wallny, H.-J., Hartley, J. K., Lawetzky, A. and Tiefenthaler, G., J. Exp. Med. 1989. 169: 73. 16 Yoshino, S., Schlipkoter, E., Kinne, R., Hiinig,T. and Emmrich, F., Eur. J. lmmunol. 1990. 20: 2805. 17 Larsson, €?,Kleinau, S., Holmdahl, R. and Klareskog, L., Arthritis Rheum. 1990. 33: 693. 18 Miller, E. J. and Rhodes, R.K.,Methods Enzymol. 1982. 82: 33. 19 Brideau, R. J., Carter, €! B., McMaster,W. R., Mason, D.W. and Williams, A., Eur. J. lmmunol. 1980. 10: 609. 20 Larsson, €? and Holmdahl, R., Scand. J. lmmunol. 1987. 27: 579. 21 Ranges, G. E., Cooper, S. M. and Sriram, S., Cell. lmmunol. 1987. 106: 163. 22 Hom, J.T., Butler, L. D., Riedl, €? E. and Bendele, A. M., Eur. J. lmmunol. 1988. 18: 881. 23 Holmdahl, R., Rubin, K., Klareskog, L., Dencker, L., Gustafsson, G. and Larsson, E., Scand. J. lmmunol. 1985. 21: 197. 24 Quinn Dejoy, S., Ferguson, K. M., Oronsky, A. L. and Kerwar, S. S., Cell. lmmunol. 1988. 113: 117. 25 Andersson, M. and Holmdahl, R., Eur. J. lmmunol. 1990.20: 1061. 26 Taurog, J. D., Kerwar, S. S., McReynolds, R. A., Sandberg, G. I?, Leary, S. L. and Mahowald, M. L., J. Exp. Med. 1985.162: 962. 27 Seki, N., Sudo, Y., Yoshioka, T., Sugihara, S., Fujitsu, T.. Sakuma, S., Ogawa,T., Hamaoka,T., Senoh, H. and Fujiwara. H., J. lmmunol. 1988. 140: 1477. 28 Holmdahl, R., Andersson, M., Go1dschmidt.T. J., Gustavsson, K., Jansson, L. and Mo, J. A., lmmunol. Rev. 1990. 118: 193.
