CELLULAR

IMMUNOLOGY

133,456-467 (199 1)

Induction of Unresponsiveness to Heymann’s Nephritis: Inhibited by Monoclonal Antibody to CD4 but Not to CD8’ CESAR G. QUIZA, PETER

L. LEENAERTS, AND BRUCEM. HALLS

Division of Nephrology, Department of Medicine, Stanford University Medical Center, Stanford, California 94305 Received September 21, 1990; accepted October

16, 1990

The effects of mAb therapy to CD4 or CD8 on induction of unresponsivenessto Heymann’s nephritis by preimmunization with renal tubular antigen in IFA. Anti-CD4 mAbs (MRC 0x35) given for 2 weeks after RTA/IFA completely prevented the induction of resistanceto HN, all rats developing proteinuria as well as high titers of autoantibody and Ig and C deposits in glomeruli. Anti-CD8 mAbs (MRC 0x8) did not prevent induction of unresponsiveness,even though it totally depleted CD8+ cells. In control rats not preimmunized with RTA/IFA, mAb therapy did not suppress disease induction, but in the case of anti-CD4 therapy enhanced the severity of disease. Persistent depletion of T cell subsetsor complement components did not explain the effects of mAb therapy. These studies suggestthat CD4+ cells are critical for the induction of unresponsivenessto HN and that therapy with mAb to CD4 can prevent induction of tolerance to an antigen, which has implications for its use in the induction of tolerance. o 1991 Academic Press. Inc.

INTRODUCTION Treatment with monoclonal antibodies to T cell subsets has been widely used to study the role of different subsetsin a variety of immune responsesincluding autoimmune disease( 1-9) and allograft models (8, lo- 13). Anti-CD4 mAb therapy will not only ablate a variety of autoimmune diseases(l-7) and allograft rejection (9- 13), but induces subsequent unresponsiveness to the antigens (14-16). Anti-CD8 has a more limited effect on autoimmunity (2,9) and allograft rejection (8, 10) and has not been shown to induce tolerance. Unresponsiveness in autoimmune models can also be induced by preimmunization by specific antigen in incomplete Freund’s adjuvant (17-2 1). In this study we examined the effectsof mAb therapy on induction of tolerance to HN3 by preimmunization with renal tubular antigen in IFA. ’ This work was supported by the Stanford Institute of Biological and Clinical Investigation and NIH Grants AI26041 and DK 40800. MRC cloneswere a kind gill of Dr. A. Williams, MRC Cellular Immunology Unit, Sir William Dunn School of Pathology, Oxford, UK. 2To whom correspondenceshould be addressedat Division of Nephrology Room S-215,Stanford University Medical Center, Stanford, CA 94305. 3Abbreviations usedin this paper: HN, Heymann’s nephritis; RTA, renal tubular antigen; PBS, phosphatebuffered saline; PBST, tween-20 in PBS; IFA, incomplete Freund’s adjuvant; CFA, complete Freund’s adjuvant; mAb, monoclonal antibody; PBL, peripheral blood lymphocytes; FITC, fluorescein isothiocyanate; ELISA, enzyme-linked immunosorbent assay; BSA, bovine serum albumen; EAE, experimental allergic encephalitis. 456 0008-8749/9 1 $3.00 Copyright 0 1991 by Academic Press, Inc. All rights of reproduction in any form reserved.

UNRESPONSIVENESS

TO HEYMANN’S

NEPHRITIS

457

HN is a rat autoimmune diseasemodel, that is thought to resemble membranous glomerulonephritis in man (22, 23), and is induced by immunizing rats with a crude filtrate of renal cortical tissue,the renal tubular antigen, emulsified in complete Freund’s adjuvant (24). The gp 300 molecule, found in the proximal tubules is thought to be the most important molecule for the induction of HN (25). HN has a typical course; after immunization there is a disease-freeperiod, before autoantibodies to RTA appear in the serum by week 4 and bind to glomerular epithelial cells where they activate complement and cause glomerular barrier damage (22, 23). In the following weeks, severeproteinuria develops and by 12 weeks all animals suffer from a nephrotic syndrome (26). Immunization with RTA in IFA does not induce autoantibodies or HN but results in a specific form of unresponsiveness(17, 1827-29). When theserats are rechallenged with RTA/CFA, the onset of HN is either completely prevented or only minor proteinuria will develop (17, 18). A similar unresponsivenesshas been reported in other immune diseasemodels such as experimental allergic encephalitis (19), experimental autoimmune thyroiditis (20), and interstitial nephritis (21) after the injection of the autoantigen emulsified in IFA. The unresponsivenessin HN is mediated by antigenspecific thymus-derived suppressorcells (27), and Litwin et al. (18) demonstrated that a T cell fraction of spleen and lymph node cells of pretreated rats could adoptively transfer unresponsiveness to naive hosts. Later studies showed that the CD8 cells were responsible for the resistance to HN (28) and to interstitial nephritis (21). In contrast, studies by our group (29) showed that in the first week after preimmunization with RTA/IFA unfractionated T cells can transfer suppression while CD8+ cells alone cannot. In experimental allergic encephalitis CD4+ suppressorcells have been identified (30) and therapy with anti-CD8 has not affected development of such resistance (9). It has recently been demonstrated that anti-CD4 therapy, and not anti-CD8 therapy, prevented tolerance induction in an experimental autoimmune thyroiditis model (3 1). In this study we examine the effect of anti-CD4 and -CDS mAb therapy during the immediate period after RTA/IFA pretreatment, which is the induction phase of unresponsiveness.Rats were later rechallenged with RTA/CFA and were then monitored for the development of HN. Therapy was with MRC 0x8 which nearly totally depletes the CD8+ population in blood and peripheral lymphoid tissue (9) and with MRC 0x35 which causesonly partial depletion of CD4’ T cells but is the most potent antiCD4 mAb to ablate the rejection response in rat (10, 11). Anti-CD4 mAb therapy prevented induction of HN, whilst anti-CD8 mAb therapy had no effect. MATERIALS AND METHODS Animals Inbred male Lewis rats (Harlan-Spraque-Dawley Inc., Indianapolis, IN) weighing I50 g were used and housed in the Department of Laboratory Animal Medicine (Stanford). The rats were maintained on standard chow and water ad libitum. Antigen Preparation RTA was prepared as described by Edgington (24). Briefly, rat kidney cortices were dissectedand pushed through a 150 mesh steel sieve. The filtrate was centrifuged twice at 4008 for 10 min. The supernatants that contained the RTA were then washed three

458

QUIZA,

LEENAERTS,

AND

HALL

times with distilled water and ultracentrifuged at 78,680g (20,000 rpm) for 45 min. The remaining pellet was lyophilized and stored at -20°C. Induction of HN Rats were inoculated behind their pinna with 10 mg RTA emulsified in 200 ~1CFA containing 1 mg mycobacterium tuberculosis HRa37 (Difco, Detroit, MI). A booster dose of 5 mg was administered 2 weeks later. Induction of Unresponsivenessto HN An identical immunization protocol was used as above but RTA was now emulsified in IFA (RTA/IFA) (Difco). Monoclonal Antibody Therapy The clones MRC 0x35 (mIgG2a) and MRC 0x8 (mIgG1) which are directed against the rat CD4 and CD8 antigens, respectively (32), were produced in ascitesas described (11). Seven milligrams per kilogram body weight of each of the mAbs were given intraperitoneally to the rats on the day of inoculation with RTA/IFA and again 4, 7, and 10 days later. This dose regime of MRC 0x35 has been shown to induce unresponsiveness to heart allografts (10, 1l), and that of MRC 0x8 depletes blood and peripheral lymphoid tissues of CD8+ cells (9). Peripheral blood lymphocytes were monitored throughout the study by FACS using the mAbs W3/25 (anti-CD4 not blocked by MRC 0x35), MRC 0x35, MRC 0x8, MRC 0x19 (anti-CDS, T cells), MRC Ox 12 (K chain of rat Ig), and second step reagent goat anti-mouse Ig FITC rat absorbed (Caltag Laboratories, South San Francisco, CA) as described (11). Monitoring of HN Autoantibody determination. Autoantibody titers were determined by ELISA at weeks 4, 8, and 12 post-RTA/CFA challenge as described (29). Briefly, Immulon I ELISA microtiter plates (Dynatech Laboratories, Alexandria, VA) were coated overnight with 100 ~1solubilized RTA. The plates were washed twice with 0.05% Tween 20 in PBS and blocked with 2% BSA for 1 hr at room temperature followed again by two washeswith PBST. One hundred microliters of a dilution seriesof a known strongly positive serum and two dilutions (1:8 and 1:64) of the test sera were added in quadruplicates to the plates. A 1:8 diluted irrelevant rat serum was the negative control. These plates were incubated for 1 hr at 37°C. After the plates were washed three times with PBST, 100 ~1 of goat anti-rat Ig Alk. Phos. (Tago Inc., Burlinghame, CA) dilute 1:500 in 2% BSA was added to each well and incubated for 1 hr at 37°C. The plates were again washed three times with PBST and once with carbonate-bicarbonate buffer, and then 100 ~10.5% p-nitrophenyl phosphate (Sigma Chemical Co., St. Louis, MO) in carbonate-bicarbonate buffer was added for 15 min. The color reaction was read at 405 nm on an ELISA reader (Dynatech Laboratories). The results were expressed in log, titers. Urinaryprotein determination. All rats were individually housed in metabolic cages (Lab Products Inc., Maywood, NJ) and 24-hr urine samples were collected at 2, 4, 6, 8, 10, and 12 weeks post-RTA/FCA challenge. Proteinuria was assayedwith a colorimetric method (Bio-Rad, Richmond, CA).

UNRESPONSIVENESS TO HEYMANN’S

NEPHRITIS

459

Immunojluorescent studies. Kidney biopsies were taken at 4, 8, and 12 weeks using a standard surgical procedure. Tissues were snap frozen in liquid nitrogen, sectioned, air dried, and fixed in acetone (-20°C) overnight. The sections were then incubated for 30 min with FITC-conjugated rabbit anti-rat Ig. Fluorescence was studied with the fluorescence microscope and the degree of fluorescence pattern was scored as previously described (29): 0 = none, 0.5 = small granular deposits only apparent with the 100X objective, 1 = weak granular, 2 = moderate granular, 3 = intense granular, 4 = pseudolinear. Complement determination. Functional complement was determined by measuring the hemolytic titer (CH50 units/ml) in fresh serum as described (33). Briefly, 5 X lo8 antibody-sensitized sheepred blood cells were incubated for 1 hr at 37°C with a series of rat serum dilutions. Hemolysis was measured by reading the optical density of the supernatants at 541 nm (OD,,,). The serum amount that resulted in 50% hemolysis was calculated by applying the formula Y/l - Y = 1, wherein Y = (OD54, test sample tube-OD54, blank)/(OD54, 100%lysis-OD,,, blank). The hemolytic titer was then calculated as the reciprocal of this dilution. A control rat serum containing 52 CH50 units/ml (Sigma Chemical Co., St. Louis, MO, lot 67F9476) was run simultaneously and used as an internal standard. Experimental design. Seven groups of rats were studied (seefirst column Table 1). All animals were immunized with RTA/CFA on Day 0. Three weeks earlier three groups (B, D, F) were preimmunized with IFA alone and another three groups (C, E, G) with RTA/IFA. Group A received no prior immunization and was the control group for induction of HN. Groups D and E received MRC 0x8 therapy, and groups G and F received MRC 0x35 therapy. Proteinuria was monitored every 2 weekswhile autoantibody titers, kidney deposits, and lymphocyte subpopulations were determined every 4 weeks. The study was terminated at 12 weeks post-RTA/CFA immunization. Statistical analysis. All data are expressedas means and standard deviations. Results of various groups were compared to each other using the unpaired Student’s t test. A P value < 0.05 was considered significant. RESULTS The results are summarized in Table 1 and are shown in more detail in Fig. I. The first three groups of animals were controls. Normal RTA/CFA-immunized rats (group A) developed diseasewith proteinuria as early as week 8 (49 ? 22 mg/day), and by week I2 ( 146 t- 65 mg/day) all animals had severeproteinuria or nephrotic syndrome. In all rats at 12 weeks glomeruli had intense granular or pseudolinear deposits. The autoantibody titer reached a maximal value by week 8 and declined by week 12. Animals preimmunized 3 weeks earlier with -/IFA (group B), showed a similar evolution of HN and no significant difference was found between groups A and B, demonstrating that -/IFA pretreatment did not induce unresponsiveness.However, in the animals of group C which received RTA/IFA, after 12 weeks only one animal had developed severe (110 mg/day) and one moderate proteinuria (22 mg/day) while the three other rats had less than 10 mg/day urinary protein. Peak autoantibody titer in this group was four- to eightfold lower than in groups A and B, and glomerular deposits score differed by at least 1 point at all time points. The results for all three monitored parameters were significantly different (P < 0.05) from the ones obtained with the control groups A and B, indicating that RTA/IFA markedly reduced the severity of

460

QUIZA,

LEENAERTS,

AND

v-l

VI

c-4

0

VI

u-l

m

N

m

CI

0

0

0

v-l

ri

0

0

0

0

0

N

-

0

0

z

HALL

UNRESPONSIVENESS

TO HEYMANN’S

NEPHRITIS

HN and protected three out of five rats from developing proteinuria, feature of glomerular injury.

461 the cardinal

Eflect of Anti-CD4 mAb Therapy on the Induction of Unresponsiveness Treatment with MRC 0x35 mAbs during the preimmunization period without administering RTA/IFA did not prevent the induction of HN. All animals in group F developed a manifest nephrotic syndrome with proteinuria up to 747 t- 272 mg/ day at week 12 and all the animals became very edematous (see Table 1 and Fig. 1). In fact, pretreatment with anti-CD4 mAbs worsened the disease. A significant difference in level of proteinuria was seen between group F and each of the other groups not pretreated with RTA/IFA (P < 0.05). The rats which received RTA/IFA and antiCD4 mAb therapy (group G) also developed HN and no difference was found between this group and the groups not pretreated with RTA/IFA, (groups A, B, D) indicating that therapy with this mAb restored the inducibility to HN. The proteinuria of the rats of group F was more pronounced than those of the animals of group G (P < 0.05) and a significant difference in autoantibody titer was also found at week 4 (P < 0.05). Anti-CD4 therapy did not remove the CD4+ population completely as can be seen from Table 2. At the end of the two-week treatment course, approximately 17% of the PBL could be coated with anti-mouse Ig or with W3/25, an anti-CD4 mAb that recognizes a different epitope to MRC 0x35 (data not shown). Afterward, the CD4 cells progressively repopulated and reached near to normal values of 37% by week 4.

Efect of Anti-CD8 mAb Therapy on the Induction of Unresponsiveness Animals treated with anti-CD8 mAb therapy (groups D and E) behaved indistinguishably from their control counterparts (see Fig. 2). RTA/IFA pretreatment prevented the onset of disease, while preimmunization with -/IFA did not. When compared to each other, group E and D had significantly different proteinuria (P < 0.05), autoantibody titer (P < 0.05) and glomerular deposit score (P < 0.05). MRC 0x8 therapy was effective in depleting CDS+ cells, in that on the last day of therapy none of these cells are detected in blood, but there was partial recovery 11 days later, which was the day of challenge with RTA/CFA (see Table 2). There was recovery to near normal values 4 weeks later. Removal of CD8+ cells resulted in a proportional increase in CD4+ cells (see Table 2) and B cells (data not shown). Taken together, these results suggest that CDS+ cells are not required for induction of unresponsiveness to HN. Although it is assumed that cell depletion by mouse mAbs occurs primarily by the phagocytic system and not by complement (8), it cannot a priori be excluded that certain mAbs cause significant complement consumption. Thus anti-CDS mAb therapy may have been equally effective in restoring the inducibility to HN as anti-CD4 therapy, but complement levels may have been depleted thus preventing HN development. The CH50 value was not impaired in any of the animals at the time of RTA/CFA immunization and at 4 weeks after the challenge, however. Animals in group B had a CH50 value of 156 -t 30 and 118 & 38 at week 0 and week 4, respectively, as compared to animals in group E 117 _t 24 and 102 + 36 and to group G which are 142*22and 119_t37. DISCUSSION In this study we examined the effects of therapy with mAb to T cell subsets on the induction of unresponsiveness to HN following preimmunization with RTA/IFA.

462

QUIZA, LEENAERTS, AND HALL 800 ,

0

2 4 8 8 Weeks after RTAlCFA

10

12

14

challenge

FIG. 1. Influence of MRC 0x35 treatment on the evolution of proteinuria, autoantibody titer, and glomerular deposits in four groups; the control groups B primed with IFA alone (0) and C primed with RTA/IFA (D) and the MRC 0x35 treated groups F primed with IFA alone (A) and G primed with RTA/IFA (A). Group F developed proteinuria which was significantly higher than the proteinuria found in any other group. MRC 0x35 treatment restored induction to HN in group G which developed diseasethat was indistinguishable from group B receiving therapy and primed with IFA alone. SD were all less than 20% and were not represented for clarity.

Anti-CD4 mAb therapy given for 2 weeks abolished the capacity of RTA/IFA pretreatment to induce unresponsiveness to HN. Further, in animals not preimmunized with RTA/IFA, anti-CD4 therapy administered prior to immunization worsened the severity of the disease,MRC 0x8 therapy had no effect at all on the induction of HN, even though this antibody was very effective in depleting CD8+ cells. This depletion

MRC 0x35 (n = 10)

29.2 f 6.4

18.3 + 5.4

71.3 f 8.4

ND

16.3 +_6.2

Induction of unresponsiveness to Heymann's nephritis: inhibited by monoclonal antibody to CD4 but not to CD8.

The effects of mAb therapy to CD4 or CD8 on induction of unresponsiveness to Heymann's nephritis by preimmunization with renal tubular antigen in IFA...
746KB Sizes 0 Downloads 0 Views