Anti-dfl TcR antibody in adjuvant arthritis

Eur. J. Immunol. 1990. 20: 2805-2808

2805

Short paper Shin YoshinoO, Eva Schlipkoterv Raimund Kinneoa, Thomas Hiinig" and Frank EmmrichO

Suppression and prevention of adjuvant arthritis in rats by a monoclonal antibody to the alp T cell receptor

Max-Planck-Society, Clinical Research Units for Rheumatology at the Institute for Clinical Immunology of the University of Erlangen-Numbergo, Erlangen and Genzentrum der Ludwig Maximilians Universitatv Miinchen, Martinsried

Adjuvant arthritis (AA) in rats is an experimentally induced autoimmune disease mediated by T lymphocytes specific for Mycobacterium tuberculosis. We raised the question whether Tcells carrying the y/6 Tcell receptor (TcR), reactive or not to mycobacterial antigens, are involved in the pathogenesis of A A . For this purpose,Tcells bearing the TcR a@ were depleted from circulation by treatment with a monoclonal antibody against the rat T c R d P (R73). This treatment efficiently suppressed existing disease. Even more efficient was pretreatment with R73 from birth, which prevented A A induction completely. In these a@+T cell-depleted animals an elevated level of cdp- Tcells (about 15% vs. 1% in normal rats) was evident, which was not significantly increased by Mycobacterium tuberculosis injection. We found no positive evidence that y/6+ T cells do contribute to A A induction. Moreover, treatment with an anti-TcR a@ monoclonal antibody may be very efficient in treatment of Tcell-mediated autoimmune diseases.

1 Introduction

2 Materials and methods

Adjuvant arthritis (AA) is induced by a single i.d. injection of CFA containing heat-killed Mycobacterium tuberculosis (MT; [I]). Ten to thirteen days after MT injection arthritis begins to develop in susceptible strains of rats (Lewis). Joint swelling and cellular infiltration peak at day 18 to 24 and decline thereafter while destruction of the joints increases until the process ends in ankylosis. The disease is transferable to naive recipients by LN cells from arthritic rats as well as by MT-specificTcell lines and clones [2-41. An arthritogenic clone has been reported to recognize cartilage proteoglycan as well as MT, suggesting that A A might be induced by an autoimmune response to tissue antigens in the joint [4].

2.1 Induction of AA

To induce A A , eight female Lewis rats (8 weeks old) per group were injected i.d. into their tail base with 0.1 ml of CFA containing 10 mg/ml of heat-killed MT (H73Ra: Difco, Detroit, MI).To evaluate the severityof arthritis, the lesions of the four paws were each graded from 0 to 4 according to increasing extent of erythema and edema of the periarticular tissue as described by Wood et al. [ 101.The maximum score was 16.

2.2 Treatment protocols Recently, a link between MTspecificity and theTcR y/6 has been suggested.TheTcRyRi is found on a minor population of peripheral blood Tcells not exceeding 5% [5] in normal individuals. MT-specific cells bearing the TcR y/6 seem to dominate the primary immune response against MT in mice [6]. Moreover, MT-reactive y/6+ T cells have also been isolated from synovial fluid of rheumatoid arthritis patients [7] and were found to be increased by a factor of 4 in human peripheral blood T cells stimulated with PPD, a mycobacterial antigen [8]. To investigate whether y/6+ T cells play a key role in the pathogenesis of rat A A , either y/6+ Tcells or a@+Tcells could be depleted by use of mAb. Since an anti-TcRy16 antibody is not yet available we used mAb R73 which is directed against a common epitope of the TcRa/P [9]. [I 86121 A

Supported by DFG grant Ki 329/2-1.

Correspondence: Frank Emmrich, Clinical Research Unit for Rheumatology/Immunology, Max-Planck-Society, Schwabachanlage 10, D-8520 Erlangen, FRG Abbreviation: AA: Adjuvant arthritis 0 VCH Verlagsgesellschaft mbH, D-6940 Weinheim, 1990

Two milligrams of R73 dissolved in PBS was injected i.p. on day 15, 18 and 21 after MT injection. The control group received 1 ml of PBS. To test the effect of dexamethasone (DxM) on A A , DxM (I mg/kg body weight) was suspended in 1 ml of 0.5% sodium carboxylmethyl cellulose (CMC) solution and was administered orally using a gastric tube. One milliliter of 0.5% CMC solution alone was given to the control group. In separate experiments, newborn rats were pretreated for 8 weeks by i.p. injections twice a week of 2 m g R73 dissolved in 0.5 ml PBS. The control group received PBS only.

2.3 Immunocytology of blood smears A small volume of blood was withdrawn by a syringe from the tail vein on day 14, 19, 22, 26, 33, 36 and 42 for preparation of smears. Detection of TcR a@+Tcells with R73 was performed by the alkaline phosphatase antialkaline phosphatase (APAAP) method as described by Cordell et al. [ I l l . The blood smears were counterstained with hematoxylin and the percentage of R73+ leukocyte was determined. OO14-2980/9O/1212-2805$3.S0+ ,2510

2806

S.Yoshino, E. Schlipkoter, R. Kinne et al.

2.4 Histological examination The tissues of the hind feet were collected, fixed in 4% formalin, and decalcified in a solution of 3.1% HCI, 5% formic acid, and 7% aluminium chloride. Subsequently, the material was embedded in paraffin, sectioned at 5-8 pm, and stained with hematoxylin and eosin. Immunohistology was performed on undecalcified cryostat sections of the ankle joints from one treated rat.The treated animal received 2 mg of R73 at day 15, 18 and 21 and was killed at day 24. Methods were identical to those used for blood smears. 2.5 FCM

Nucleated cells (2 x lo5 cells) in 0.1 ml PBS/O.l% BSA/0.02% azide) were sequentially exposed for 15 min on ice to saturating concentrations of (a) mAb R73 (antiTcR a@), (b) PE-conjugated rat anti-mouse 1c (Becton Dickinson, Mountain View, CA), (c) normal mouse IgG and (d) FITC-OX19 (anti-CD5; Serotec Ltd., Oxford, GB). Isotype (IgGl)-matched negative control mAb were used to control specificity of direct and indirect labeling. Cells stained in this fashion were indistinguishable from unstained cells. FITC-labeled isotype-matched control mAb was also used to control the effectiveness of the blocking procedure (step c) which was found to be fully effective. Ten thousand cells were analyzed on a FACSan (Becton Dickinson) flow cytometer with light scatter gates set to exclude dead cells and erythrocytes.

3 Results and discussion R73 was given on day 15,18 and 21 after MT injection when the animals were developing arthritis as shown in Fig. 1A. We found that arthritis was drastically suppressed after R73 injection, whereas control animals developed severe arthritis. However, since no further injection was given after day 21, edema and swelling began to increase again after day 27 until the scoring of the control group was reached. R73 suppressed A A so effectivly that we compared the suppressive effect of R73 to that of DxM which is one of the most powerful steroidal anti-inflammatory drugs. We used 1 mg/kg of the drug, a dose which is rather high and exceeds the dose generally used for anti-inflammatory drug testing in A A [12, 131. DxM was given according to the protocol used for R73 treatment. As shown in Fig. lB, the suppressive effect of DxM was weaker than that of R73 with an earlier exacerbation of arthritis after the last injection. To investigate whether the suppression and re-emergence of AA correlated with the number of a#+ cells, we determined their frequencies in blood smears by the APAAP method. As shown in Fig. 2A.1, the relative number of a@+ cells in blood decreased rapidly after injection of R73 as compared to the control group, and increased thereafter, consistent with the clinical joint score after the treatment with R73 was terminated. No circulating cells coated with R73 were detected 1 day after treatment (data not shown). In addition, OX19 (CD5)positive T cells were determined in the blood smears by APAAF! CD5 is a pan-Tcell marker equally expressed on

Eur. J. Immunol. 1990. 20: 2805-2808 both, a@+and y/6+ Tcells [9]. It is exclusively expressed on OX52+ cells (data not shown), which comprise Tcells and NK cells but not B cells [14].We found that the depletion of CD5+ cells paralleled that of R73+ cells (Fig. 2A.2), indicating that no dramatic increase of y/6+ Tcells occurs during R73 therapy. Moreover, this finding indicates that modulation of TcRa@ was not the reason for disappearance of TcR a@+cells from circulation. CD5 is an independent T cell marker not co-modulated by R73 treatment (data not shown). Histological examination of the rat joints on day 27 after MT injection revealed only minor synovial membrane hyperplasia and marginal destruction of cartilage in the R73 treated animals (Fig. 2B.1). In contrast, there was formation of an aggressive pannus, massive infiltration by mononuclear and polymorphonuclear leukocytes, and severe destruction of cartilage and bone in control animals (Fig. 2B.2). The presence of T cells in the pannus tissue was investigated in cryostat sections of ankle joints at day24 after A A induction. Only very small numbers of Tcells could be detected by R73 and a similar number of CD5+ cells was stained with OX19 arguing against a significant increase of y/6+ cells in the inflammatory focus. Fig. 2A shows that depletion of a@+T cells in the adult arthritic rats was not complete because the number of R73+ cells did not fall below 8% of total leukocytes. Therefore, newborn rats were pretreated with i.p. injections of R73 twice weekly which depleted the a@+ Tcells more efficiently, i. e. down to 3% of nucleated cells. This low level of a@' T cells could be maintained for several months [15]. Under these conditions, the animals were completely resistant to induction of A A (Fig. 3), in contrast to a PBS-treated control group. Moreover, we found only marginal tissue reaction at the site of injection where necrosis and chronic inflammation is normally observed when the arthritis starts in the control group.

8-h-p 0

c ._

o

t

10

L

.

.

10

.

I

I

20

I n- .. -

1

--

.

I

PBS R73 I

30

40

+cMc

*

I

DxM I

20

30

40

Days after MT injection

Figure 1. Suppression of arthritis by R73 (A) in comparison to thc effect of dexamethasone (DxM, B). Arrows indicate R73 injection (2 mg/dose).

Anti-d[3 TcR antibody in adjuvant arthritis

Eur. J. Immunol. 1990. 20: 2805-2808

2807

Figure 2. Determination of T cell depletion upon R73 treatment by staining of blood smears with antibodies to OX19 ( A l ) and R73 (A2) and histological examination of their ankle joints (B1,2). Rats were killed on day 27 as indicated in A2. Groups of arrows indicated R73 treatment. R73-treated rats ( B l ) were compared with PBS-treated control rats (B2). Magnification X 100.

40

-I

h

ae

v

Q)

8

".

10

40

-I

s? 8

30

-

g .-

20-

I

I

I

20

30

40

h

Y

v)

a,

v)

0

0.

g

10-

U "

I

10

I

I

I

20

30

40

Days after MT injection To determine the number of y/S+ Tcells (CD5+,TcRa/fiP) in R73-pretreated naimals, a two-color cytofluorometric analysis of CD5 (0x19)and TcR a@ (R73) was performed on LN cells (Fig. 4). Depletion of a@+ T cells by R73 pretreatment was efficient as demonstrated in Fig. 4B, D, F and an increased percentage of 15%-20% a/(?- Tcells was observed in R73-pretreated animals. In addition, no significant increase of a/(?- T cells was detected upon MT injection at a dose normally used for A A induction (compare Fig. 4A with C, E and B with D, F). A difference of up to 4% in the two-color cytofluorometry between cells from MT-injected and non-injected rats or between inguinal and mesenteric L N cells was not considered to be significant for two reasons: (a) an increased level of dfi- Tcells (between 1% and 3% increase) was also observed in inguinal vs. mesenteric LN in R73-pretreated animals without MT injection (data not shown) and (b) a 2%-3% variation in separate experiments has to be considered.We conclude, that y/6+ Tcells were not or only slightly increased upon MT injection. We think it is important to stress that no symptoms of AA-like edema or

foot swelling, and no histological indication for synovitis or cartilage erosion (data not shown) were found after M T injection in these animals. Thus, it seems rather unlikely that y/6+ Tcells alone may contribute to the pathogenesis of AA.

15

-I

10

-

C

m

r"

5 Q

*

-, Y

10

20

30

PBS pretr. R73 pretr. I

40

Days after MT injection

Figure 3. Prevention of arthritis induction by pretreatment with R73.

2808

Eur. J. Immunol. 1990. 20: 2805-2808

S. Yoshino, E. Schlipkoter, R. Kinne et al.

However,Tcells may need the help provided by a@+Tcells 4 Concluding remarks which were lacking in our animals. Previous experiments have shown that proliferative responses to allogeneic cells Although not comparable in all features, rat A A has some or an antibody response to a soluble antigen like KLH characteristics in common with human rheumatoid arthritis could not be elicited in R73-pretreated rats [15],very likely [16] as for instance the massive Tcell infiltration, synovial indicating a missing helper capacity.Therefore, it could also hyperplasia and cartilage and bone destruction. An’ be argued that failure of y/6+ Tcells to induce A A could be increase of y/6+ Tcells has been found in some but not all the result of missing help by a/@+ T cells. However, this is synovial tissue samples of patients with this disease [ 171and unlikely because of the very efficient A A treatment by their role remains elusive. However, the curative effects of anti-TcR a l p antibody without complete alp+ Tcell deple- anti-TcRdp treatment in rat AA described in this report tion (see Fig. 2a), a situation in which helper function is may support therapeutical concepts aimed at suppression probably present. Since no anti-yI6 antibody is yet availa- .or deletion of a / p T cells in human rheumatoid arthritis. ble, the “missing help” hypothesis cannot be formally tested at the moment. 69.4%

10.2%

I

3.1%

We thank Dr. F! Matzinger for helpful discussion and suggestions regarding our experiments and this manuscript. We also thank Ulrike Vorderwiilbecke for the preparation of the histological sections and Barbara Thompson for her secretarial assistance.

Received June 8, 1990; in revised form August 20, 1990.

5 References

68.0%

2% n

3.6%

0.7%

@ d

& g -

>- I

11.2%

1

87.7%

17.4%

I

75.8% 3.9%

CD5 log fluorescence

Figure 4. Two-color FCM analysis of LN cells from MT-treated vs. untreated rats injected from birth with PBS or anti-TcR alp mAb. Correlation of CD5 with TcRa/@ expression on nucleated cells from pooled mesenteric and superficial LN (A-D) or from draining (inguinal) LN (E, F) of 8-week-old rats treated from birth with PBS (A, C, E ) or mAb R73 (B, D, F). A , B: No MT injection, C-F: injection of MT 29 days before death.

1 Pearson, C. M., Proc. SOC. Exp. Med. 1956. 91: 95. 2 Holoshitz, J., Naparstek.Y., Ben-Nun. A. and Cohen, I. R.. Science 1983. 219: 56. 3 Holoshitz. J.. Matitiau, A. and Cohen. I. R.. J. Clin. Invest. 1984. 73: 211. 4 Van Eden, W., Holoshitz, J., Nero, Z., Frenkel, A , , Klajman, A. and Cohen, I. R.. Proc. Natl. Acad. Sci. USA 1985. 82: 5117. 5 Raulet, D. H., Annu. Rev. fmmunol. 1989. 7: 175. 6 Janis, E. M., Kaufmann. S. H. E., Schwartz, R. H . and Pardoll, D. M., Science 1989. 244: 713. 7 Holoshitz, J., Koning, F., Coligan, J. E., Bruyn, D. J. and Strober. S., Nature 1989. 339: 226. 8 Harcgowoin, A., Soman, G., Horn, R. C. and Finberg, R. W., Nature 1989. 340: 309. 9 Hunig, T., Wallny, H.-J., Hartley, J. K., Lawetzky, A. and Tiefenthaler. G.. J. Exp. Med. 1989. 169: 73. 10 Wood, F. D., Pearson, C. M. and Tanaka, A., Int. Arch. Allergy 1969. 35: 456. 11 Cordell, J. L., Falini, B., Erber, W. N.. Ghosh, A. K.. Abdulazis, Z., MacDonald, S.. Pulford, K. A. F., Stein, H. and Mason, D. Y., J. Histochem. Cytochem. 1984. 32: 219. 12 Newbould, B. B., Br. J. Pharmacol. 1963. 21: 127. 13 Winter, C. A. and Nuss, G. W., Arthritis Rheum. 1966.9: 394. 14 Robinson, A. I?, Puklavec, M. and Mason, D., Immunology 1986. 57: 527. 15 Hunig, T., Tiefenthaler, G.. Lawetzky, A., Kubo, R. and Schlipkoter, E., Cold Spring Harbor Symposium 1989. 54: 61. 16 Pearson. C. M. and Wood, F. D., A m . J. Pathol. 1963. 42: 73. 17 Brennan, F. M., Londai, M.. Jackson, A. M., Hercend,T.. Brenner, M. B., Maini, R. N. and Feldmann. M., J. Autoimmun. 1988. I: 319.