Age-dependent responsiveness to interleukin-6 in B lymphocytes from a systemic lupus erythematosus-prone (NZB × NZW)F1 hybrid

CLINICAL

I~UNOL~Y

AND

I~~OPATHOLOGY

Vol. 62, No. 3, March, pp. 264-269, 1992

Age-Dependent Responsiveness to Interleukin-6 in B Lymphocytes from a Systemic Lupus Erythematosus-Prone (NZB x ~Z~)Fl Hybrid MARTA E. ALARC~N-RIQUELME,' Department

of Immunology,

Arrhenius

Laboratories

GORAN MOLLER, AND C~MEN for Natural

The cellular mechanisms for the production of IgG antiDNA antibodies were studied. Culture of T and B cells from old (NZB x NZW)FI mice led to the production of IgG antiDNA antibodies- We found that direct cell contact was partly necessary for the production of IgG anti-DNA antibodies. Fixation of the T cells showed that lymphokines were largely responsible for the antibody synthesis. Antibodies to mouse interieukin-6 (IL-61 inhibited the production of these antibodies in the T-B cell coculture. Human IL-6 could induce small “resting” B cells from the old (NZB x NZW)Fl mice to produce IgG anti-DNA antibodies in a dose-dependent fashion. The response was inhibited by an anti-human IL-6 monoclonal antibody. Large or small B cells from young (B/W)Fl mice or Balblc mice were not induced by IL-6 to antibody production. Therefore, the capacity of the B/W mice to produce the IgG anti-DNA antibodies correlated with the ability of the B cells to respond to IL-6 and with the age at which the mice begin to have signs of the disease. Q 1992 Academic Press, Inc.

INTRODUCTION The (NZB x NZW)Fl (B/WlFl mouse develops a syndrome very similar to human systemic lupus erythematosus (SLE) where IgG anti-DNA antibodies are produced in high amounts (1). The production of the IgG anti-DNA antibodies is initiated, for unknown reasons, when the mice reach 5 months of age, and progresses when more cells are engaged in antibody secretion. The mice die, presumably of immune complex-induced glomerulonephritis, at around 9-10 months of age (1). By 7 months, almost 100% of the mice show IgG anti-DNA antibodies in their serum, and antibodies are readily detected in the supernatants from cultured spleen cells. It is known that T cells are required for the production of IgG but not IgM anti-DNA antibodies (2). IgM anti-DNA antibodies are 1 Supported by the Consejo National de Ciencia y Tecnologia, CONACyT, Mtlxico and Stockholm University.

Stockholm

University,

Stockholm,

Sweden

present in normal mice and can be elicited by polyclonal stimulation of B cells (2, 3). Datta et al. (4) found that both T and B cells from old mice were required for the production of IgG anti-DNA antibodies in vitro. T and B cells from young mice if-4 months) cultured together or in combination with cells from old mice did not produce IgG anti-DNA antibodies. These data suggest that both ‘I’ and B cells from old mice are affected or could be in a stage of differentiation in which both can act in concordance to produce highaffinity IgG anti-DNA antibodies. An imbalance in lymphokine production and/or alterations at the cellular level could be responsible for this. In previous experiments (5) we tried to induce the production of IgG anti-DNA antibodies by combining the effects of a switching factor, interleukin-4 (IL-4)> and an expanding factor, IL-5, to increase the production of IgM anti-DNA antibodies by purified B cells 161. Large amounts of poly~lonal immunoglobulins and IgM anti-DNA antibodies were produced, but no IgG anti-DNA antibodies could be found, despite the extensive class switch elicited by IL-4. This could be explained if low-affinity IgG anti-DNA antibodies were produced as a consequence of the IL-4-induced switch, but were undetectable, while low-affinity IgM antiDNA antibodies, showing higher avidity due to their pentameric conformation (61, were thus readily shown in the culture supernatants. In addition, in vitro cultures of purified B cells did not provide the necessary selection for the production of high-affinity antibodies. Al~rnatively, lymphokines other than IL-4 and IL-5 might be involved, or direct collaboration between T and B cells could be an essential requirement for the production of pathogenic antibodies. In the present paper, we analyzed both the requirement for direct physical contact and the lymphokine involvement in the production of IgG anti-DNA antibodies. We found that a direct contact between T and B cells was required and that IL-6 stimulated purified B cells from old (B/W)Fl mice to IgG anti-DNA antibody production.

264 0090-1229192 $1.50 Copyright Q 1992by AcademicPress, Inc. Ail rights of reproductionin any form reserved.

Sciences,

FERN~NDEZ

MECHANISMS

OF AUTOANTIBODY

MATERIALS AND METHODS

265

PRODUCTION

kindly provided by Dr. Ulf Andersson versity).

(Stockholm

Uni-

Animals

(NZB x NZW)Fl hybrids were produced in our laboratory. Analyses for viral infections are done routinely in our animal house. A pool of three to five female mice of 7 months of age was used for each experiment. Young (NZB x NZW)Fl hybrids had no splenomegaly and no IgG anti-DNA antibodies in their serum. Only IgM antibodies with DNA-binding properties could be detected. Young mice were used at 2 months of age and Balb/c mice, when used, were 8-12 weeks old. Interleukins

A supernatant from the human bladder cell carcinoma line T24 was used as a source of IL-6 (7) (kindly provided by Dr. Staffan Pauli, Stockholm University). Cell Lines

IL-6 activity was determined by measuring the proliferation of the IL-6-dependent B45 cell line developed by Dr. F. Melchers and Dr. Karasuyama (8) and kindly provided by Dr. Staffan Pauli (Stockholm University). Cells were cultured at 5 x lo3 cells per well, and proliferation induced by IL-6 was measured by tritiated thymidine incorporation on the third day. One unit of IL-6 was defined as the amount required for the induction of 50% of the maximal proliferation of the IL-6dependent B45 cell line. To measure IL-2 production, supernatants from concanavalin A (Con-A&stimulated T cells were tested for their capacity to induce proliferation of the standard IL-2-dependent CTLL assay (9). Antibodies

The hybridomas HO-13-4 (lo), GK1.5 (111, and H02.2 (12) produce antibodies against Thy.l.2, CD4, and CD8, respectively, and were obtained from ATCC. Supernatants were used in a cocktail with rabbit complement (Buxted Rabbit Co., Sussex, UK) for anti-T cell cytotoxicity at the proper dilutions. For the ELISA test, both unconjugated and alkaline phosphatase-conjugated anti-IgG were purchased from Southern Biotechnology Associates (Birmingham, AL). The H205 anti-DNA monoclonal antibody (IgG2a) was kindly provided by Dr. Syamal K. Datta (4) (New England Medical Center, Boston, MA) and was used as a standard for the IgG anti-DNA ELISA at 10 p,g/ml. The anti-mouse IL-6 monoclonal antibody 6B4 (13) was a generous gift of Dr. Jacques Van Snick (Ludwig Institute for Cancer Research, Brussels Branch, Brussels) and was used as a culture supernatant. The antihuman IL-6 monoclonal antibody BSF-166 (14) was

T and B Cell Enrichment

Mice were killed by cervical dislocation and the spleens were aseptically removed and teased. Red blood cells were lysed using Gey’s solution. Macrophage depletion was accomplished by adherence of the spleen cells to petri dishes for 1 hr. After this treatment, the nonadherent cells were removed by gentle whirling, counted again, and treated for B and T cell purification as follows: T cells were purified using a cell affinity chromatography column of CNBr-activated Sepharose 6MB (Pharmacia, Uppsala, Sweden) coupled with goat anti-mouse immunoglobulins and was used according to the manufacturers instructions. Enriched T cells were more than 90% positive for staining by anti-Thyl.2. B cells were enriched using a cocktail of anti-T cell antibodies and complement as mentioned above and incubated for 1 hr at 37°C. Viability was determined by trypan blue exclusion. The enriched B cells were 85% sIg+ and less than 3% Thyl.2+. Despite the adherence step, we noticed that macrophages and other adherent cells (dendritic cells) were always found in the bottom of the culture plates by the end of the 6-day culture. Cell Cultures Culture for IgG anti-DNA antibody production. T and B cells from old and young mice were cultured for 6 days at a 1:l ratio with a total concentration of 4 x lo6 cells in 3 ml of RPM1 (supplemented with 10% FCS, 1 mM sodium pyruvate, and lop5 2 ME) in 6-well Costar Culture Plates (Cambridge, MA). In some experiments, microculture plates were used with a volume of 0.2 ml to keep the same cell ratio. The supernatants were collected, filtered, and tested. When physical contact between T and B cells was to be avoided, membranes with 0.4 pm-size pores were used (Costar Transwell No. 3408): T cells were left on the top of the membrane with 1.5 ml of culture medium, whereas the B cells were in the lower compartment in 1.5 ml of medium. The membranes were previously tested for lymphokine passage. In another set of experiments, T cells were treated in various ways in order to avoid lymphokine secretion. The T cells were treated by irradiation (2000 rad), with mitomycin C (50 I*g per 50 X lo6 cells for 45 min), or fixed (0.05% glutaraldehyde on 5 x lo6 cells/ml in BSS with 100 mM Hepes buffer). Fixation was stopped after 30 set by the addition of 0.2 M glycine and the cells were washed three times with BSS and cultured with B cells as described above. Effect of lymphokines

tured at a concentration

Cells were culof lo5 B cells/well in 96-well

on B cells.

266

AL~~~N-RIQ~ELME,

MGLLER,

microtiter plates in 0.2 ml of supplemented culture medium for 6 days with various concentrations of the different interleukins. Supernatants were collected and tested for the presence of anti-DNA antibodies. In some instances, I3 cells were separated into large (low density) and small (high density) cells by Percoll gradients (Pharmacia, Uppsala, Sweden) as previously described 6). ELISA for the quantitation

ofIgG anti-DNA

antibod-

ies. Briefly, Immunolon 2 plates (Dynatech Laboratories Inc.) were coated with 100 ~1 of calf thymus DNA (10 pg/ml) (Serva, Heidelberg, Germany), dissolved in phosphate-buffered saline, and incubated overnight at 4°C. Only ELISA for double-stranded DNA was performed. Nitrophenyl phosphate was used as the developing substrate (Sigma, St. Louis, MO). For total IgG determination, coating of the plates was done with the class-specific antiserum (Southern Biotechnology Associates) and developed with the respective alkaline phosphatase-conjugated antisera.

AND F~RN~~DE~

a 50% reduction in IgG anti-DNA antibody production. B cells from young mice did not produce IgG anti-DNA antibodies. Production of IgG Anti-DNA Antibodies Is Dependent on L~~phoki~e Production

We have shown that physical proximity was necessary for the production of IgG anti-DNA antibodies, but that collaboration between T and B cells might involve also the secretion of lymphokines. We treated the T cells by irradiation, mitomycin C, or glutaraldehyde to determine whether lymphokine secretion was abrogated by such treatments. Treated and untreated cells were cultured in the presence (added 2 hr in advance) of Con A, and proliferation and IL-2 secretion were measured. Figure la shows that proliferation was abolished by the three treatments while secretion of IL-2 was not completely abrogated unless the cells were fixed with glutaraldehyde. a

cpmx lo+

RESULTS

Production of IgG Anti-DNA Direct T-B Cell Contact

111

T old (Irradiated)

T old alone T ‘Id + ‘OnA

100

m

T old (Fixed)

Antibodies Requires

* ConA

T old (Mytomicj~~

* Cons

* CnnA

Collaboration between T and B cells from old B/W mice has been previously found to be required for the production of IgG anti-DNA antibodies (4). To determine if such a collaboration is dependent on direct cellular (physical) contact and/or lymphokine production, we separated T and B cells by means of a transwell membrane as described under Materials and Methods. The results of four independent experiments are shown in Table 1. The separation of T and B cells resulted in b TABLE

ugiml

1

Production of IgG Anti-DNA Antibodies” in Cultures of T and B Cells from Old and Young (NZB x NZW)Fl Mice IgG Anti-DNA Culture

Membrane& Exp 1 _._..,_.-. _-..~_-. NO 0.735 NO 0.050

(pgirnl) .~._.. Exp 2 Exp 3 Exp 4 _,____~~ ~~_ __._.x. I-___ 0.460 0.550 0.710 0.050 0.050 0.050

i

U old alone T old * B aid T old (Irradiated)

* I3 old

T old (Mytomicin T old (Fixed)

Antibodies

B old B young Told + B old No 1.40 0.980 1.250 1.100 T old/B old Yes 0.800 0.500 0.720 0.800 Told iB young No 0.050 0.100 0.230 0.160 T old/B young Yes ND” ND 0.120 0.060 ---___..--l_--___ * IgG anti-DNA antibodies were measured in the supernatants by ELISA afker 6 days of culture as described under Materials and Methods. b T and B cells were cultured together or separated by a transwell membrane which permitted passive diffusion of the factors (previously tested). Yes means that the membrane is present; No, that it is not. ’ Not done.

m

ci + R old

* F3 old

050

0.25

I&i anti-DNA FIG. 1. ia) T cells from old (B/W)Fl mice were treated with mitomycin C, irradiated, or fixed and tested for IL-2 production after Con A stimulation (done previously). The figure shows proliferation after 72 hr of culture and IL-2 pr~uction in the same supernatants as measured by the induction of CTLL cell proliferation, (b) Old mouse T cells treated as in (a) were cultured with B celis from old mice. Production of IgG anti-DNA antibodies in the supernatants (kg/ml) is shown. 0

MECHANISMS

OF AUTOANTIBODY

Pretreated T cells (without Con A) were cultured with B cells for 6 days. T cells irradiated or treated with mitomycin C (Fig. lb) were just as capable as untreated cells in inducing B cells to IgG anti-DNA antibody production. Fixed T cells were unable to induce the production of IgG anti-DNA antibodies under all experimental conditions (Fig. lb), indicating that secretion of a T cell-derived factor was also required. Glutaraldehyde used under the present conditions did not modify membrane structures (17). The inability of the fixed T cells to stimulate B cells was not due to toxicity by the possible presence of glutaraldehyde in the medium, presumably despite the extensive washing of the cultures. We tested this by combining treated and untreated cells and culturing them with Con A (T cells) or lipopolysaccharide (B cells). Proliferation was unaffected when compared with the untreated cells cultured alone (data not shown). Effect of Anti-IL-6 Antibodies on the Production of IgG Anti-DNA by T and B Cells

Apparently, soluble factors were required for the production of IgG anti-DNA antibodies. We had previously tested IL-4 and IL-5 on purified B cells from the (B/W)Fl mice, without any effect (5). Therefore, we were interested in testing the ability of various antibodies directed against other lymphokines to inhibit IgG anti-DNA antibody production in our system. Antibodies to interferon-y (IFN-y) (F3), IL-4 (llBll), and IL-6 (6B4) were tested. Antibodies to IFN-?/ and IL-4 did not show any effect on IgG anti-DNA antibody production (data not shown). However, a supernatant from the anti-mouse IL-6 hybridoma 6B4 was able to inhibit the production of IgG anti-DNA antibodies in the T-B collaboration system. Two individual experiments were performed and inhibition was found to be dose-dependent, although a background of IgG production remained (data not shown). Effects of Exogenous Human IL-6 on the Production of IgG Anti-DNA Antibodies by B Cells from (NZB x NZW)Fl Mice

In view of the above findings, we next tested the effect of exogenous human IL-6 on enriched B cells from old and young B/W mice. The IL8-containing supernatant from the human bladder carcinoma cell line T24 was serially diluted. The means (&SD) from three experiments are shown in Fig. 2. IL-6 was fully capable of substituting for T cells to induce IgG anti-DNA antibody synthesis in B cells from old mice. Total IgG polyclonal antibody synthesis was also increased by the addition of IL-6, but not IgM production (data not shown). No effect was observed on B cells from young mice or from Balb/c mice after IL-6 stimulation. The effect of exogenous human IL-6 was also tested on the B cell populations separated by Percoll gradients. B cells were separated into large (low density) B cells, considered to be in viuo-preactivated B cells, and

PRODUCTION

267

1.00

,750

.500 = ,250

B Old

c’ B young

0

40

80

I50

312

625

1250

WY1

Units/ml IL+ T24 supematant

FIG. 2. Culture of enriched B cells young (open squares) (NZB x NZW)Fl the human IL-6 containing supematant ies were measured in the supematants

from mice T24. after

old (dark squares) and with serial dilutions of IgG anti-DNA antibod6 days.

small (high density) cells, or resting B cells. As shown in Table 2 it was the small, high-density B cell population that produced increased levels of IgG anti-DNA antibodies in response to IL-6. Large B cells were already activated and produced IgG anti-DNA antibodies in the absence of IL-6. Large and small B cells from young mice and Balb/c mice were completely unresponsive to IL-6 with regard to production of IgG anti-DNA antibodies and polyclonal antibodies (not shown). No B cell proliferation was induced after stimulation with IL-6 (not shown). The effects of 250 U/ml of human IL-6 on old (B/W)Fl mice B cells could be inhibited by a anti-human IL-6 antibody (BSF-166) in a dosedependent fashion (Fig. 3). DISCUSSION

The animal model for human systemic lupus erythematosus, the (NZB x NZW)Fl mouse, is characterized by the production of IgG anti-DNA antibodies that TABLE Response of Large

2

and Small B Cells to 250 U/ml of IL-6”** B cells

Strain (NZB x NZW)Fl old (NZB x NZW)Fl young Balb/c

IL-6

Large

Small

+ -

0.208 f 0.049 0.200 2 0.010

0.865 k 0.235 0.100 * 0.020

+ + -

co.090 -co.090 co.090