Antibody production in autoimmune BXSB mice. I. CD40L-expressing B cells need fewer signals for polyclonal antibody synthesis

Abstract

Male, but not female, BXSB mice develop severe lupus associated with multiple immune system defects. It was recently shown that one immunological abnormality found in male BXSB mice encompasses B cell expression of CD40 ligand (CD40L) by an expanded population of large B cells. The present study was undertaken to determine how the CD40L-expressing large B cells in male BXSB mice compared with size-matched B cells from female mice in terms of their ability to secrete antibody. It was shown that the large B cells from female mice, similar to the small B cells from either male or female mice, required CD40 signalling, immunoglobulin cross-linking and cytokines for optimal antibody synthesis. In contrast, large B cells from male BXSB mice produced high levels of antibody when stimulated with only two of the three signals, and made significantly more total IgM and IgG, and anti-ssDNA antibody than size-matched B cells from female mice when stimulated with IL-4/IL-5 alone, IL-4/IL-5 plus low levels of anti-IgD-dextran, or IL-4/IL-5 plus anti-CD40 MoAb. The ability of the large B cells from male mice to produce antibody under suboptimal stimulatory conditions correlated with their expression of CD40L, and was inhibited by CD40-immunoglobulin. Taken together, these findings suggested that large CD40L-expressing B cells from male BXSB mice may be able to bypass a need for CD40 signalling from T cells, thus contributing to autoimmune disease by promoting antibody production in the absence of cognate T cell help.

INTRODUCTION

Male BXSB mice, unlike female BXSB mice, die early in life from a severe lupus-like disease accelerated by the presence of the Yaa gene on the Y chromosome [1,2]. Although multiple immune system defects have been documented in the BXSB model, increasing evidence suggests that an intrinsic B cell defect constitutes one of the primary immune system abnormalities present in male BXSB mice. A recent study in our laboratory suggested that the unusual B cell activity observed in male BXSB mice was promoted by the fact that a large proportion of splenic B cells from male BXSB mice expressed significant levels of CD40L capable of stimulating B cell proliferation [3].

Disease pathology in the BXSB mouse model, as well as in human lupus, is mediated by the production of autoantibodies (autoAbs) reacting with various nuclear components such as chromatin, histones, and DNA (both single- and double-stranded). These autoAbs form immune complexes which in turn lead to fatal glomerulonephritis (for review see [4]). AutoAb production in male BXSB mice is believed to be at one level polyclonal. However, as the disease in male BXSB mice progresses, the autoAbs begin to display characteristics of an antigen-driven immune response, including class switching to IgG [5–7]. The presence of immunoglobulin class switching as evidenced by an age-related increase in the amount of total serum IgG, as well as the presence of pathogenic IgG autoAbs, suggests a role for T cell help in the development of lupus in male BXSB mice.

One of the most important sources of T cell help for antibody production is CD40L, a molecule transiently expressed on activated CD4⁺ T cells. CD40L on the T cell binds CD40 on the B cells, and provides a crucial signal for B cell expansion and immunoglobulin class switching, promoting both polyclonal and antigen-specific antibody production, as well as memory B cell formation [8–10]. Based on the importance of CD40L in stimulating antibody production, it seemed possible that the presence of CD40L on the B cells from male BXSB could not only promote expansion of the B cell population, but could also stimulate antibody production. The ability of CD40L-expressing B cells to promote antibody production has been described using polyclonally stimulated human peripheral blood B cells [11]. The present study was designed to test whether the propensity of B cells from male BXSB mice to express CD40L could similarly promote antibody synthesis, and to delineate the signals needed for antibody production by these B cells.

MATERIALS AND METHODS

Mice

BXSB breeding pairs were obtained from the Rodent Breeding Colony at The Scripps Research Institute (La Jolla, CA). BXSB mice were later bred and housed at the University of Arkansas for Medical Sciences (Little Rock, AR) animal care facility. The mice used in this study were between 3 and 5 months of age unless otherwise specified.

Reagents and antibodies

The anti-Thy-1.2 MoAb (58-A, mouse IgG2b) was purchased from Accurate Chemical and Scientific Corp. (Westbury, NY), and anti-CD40 MoAb (3T3, rat IgG2a) was purchased from Harlan Bioproducts for Science (Indianapolis, IN). For the ELISAs, capture MoAbs specific for mouse IgM MoAb (LO-MM, rat IgG2a), and IgG1 (LO-MG1, rat IgG1) were purchased from Caltag Labs (Burlingame, CA). The biotinylated detecting MoAbs specific for mouse IgG1 (A85-1, rat IgG1) and IgM (R6-60.2, rat IgG2a) were purchased from PharMingen (La Jolla, CA). The isotype standards (purified mouse myeloma IgM, IgG1), and the polyclonal alkaline phosphatase-conjugated goat anti-mouse IgG, IgA, and IgM (H&L) were all purchased from Zymed Labs (San Francisco, CA). Dr C. M. Snapper (Uniformed Services University of the Health Sciences, Bethesda, MD) provided the anti-IgD^b-dextran (AF3). The PE-conjugated anti-CD45R (B220) MoAb (RA3-6B2, rat IgG2a) was purchased from PharMingen. The FITC-conjugated streptavidin was purchased from Jackson ImmunoResearch (West Grove, PA). Biotinylated anti-CD40 ligand MoAb (MR1, hamster IgG) was provided by Dr R. J. Noelle (Dartmouth Medical School, Hanover, NH). Biotinylated hamster IgG (PharMingen) was used as the isotype control. The purified anti-syndecan-1 MoAb (281.2, rat IgG2a) was provided by Dr R. Sanderson (University of Arkansas for Medical Sciences, Little Rock, AR). Murine CD40-Ig and the control fusion protein MoAb, L6 (control-Ig), were supplied by Dr D. Hollenbaugh (Bristol-Myers Squibb Pharmaceutical Research Institute, New Brunswick, NJ).

Preparation of splenic B cells

Spleens were aseptically removed and gently teased in medium containing RPMI 1640. The resulting cell suspension was resuspended in cold erythrocyte-lysing solution made from 0.17 m Tris and 0.16 m NH₄CL, pH 7.2. The cells were washed, resuspended in RPMI 1640 + 10% fetal calf serum (FCS), and allowed to incubate in large flasks for 1 h at 37°C. In some experiments this adherence step was repeated. The non-adherent cells were then collected, and the T cells removed by complement-mediated lysis with anti-Thy-1.2 MoAb and rabbit complement (Accurate Chemical and Scientific Corp.). To obtain B cell populations of different sizes, the B cell preparation was subjected to Percoll gradient fractionation as described previously [3]. The resulting populations of small cells from both male and female mice consisted of 94–98% CD45R⁺ (B220⁺) cells as determined by flow cytometry. The resulting populations of large cells from both male and female mice consisted of 89–93% CD45R⁺ (B220⁺) cells. The non-B220⁺ cells in the populations of spleen cells enriched for large B cells expressed a low level of MHC class II molecules, which was increased upon stimulation with lipopolysaccharide (LPS) (data not shown), suggesting that these residual cells were macrophages.

Measuring immunoglobulin levels in culture supernatants and serum

Small or large B cells (1 × 10⁵/ml) were cultured in 96-well plates in 0.2 ml medium (RPMI 1640 supplemented with 2 mml-glutamine, 1 mm non-essential amino acids, 1 mm sodium pyruvate, 100 U/ml penicillin, 100 μg/ml streptomycin, 5 × 10⁻⁵ m 2-mercaptoethanol (2-ME), and 10% FCS. IL-4 (0.5 ng/ml) and IL-5 (0.25 ng/ml, both from R&D Systems, Minneapolis, MN) were titrated into the wells in the presence or absence of increasing concentrations of anti-IgD^b-dextran ± anti-CD40 MoAb (1 μg/ml). Some wells received 25 μg/ml CD40-Ig or control-Ig at the initiation of culture. Cultures were incubated at 37°C in a 6% CO₂ humidified air incubator. Supernatants were collected at 120 h and analysed by ELISA. Briefly, 96-well culture plates were incubated for 24 h at 4°C with 2.5 μg/ml of either purified rat anti-mouse IgG1 or IgM MoAbs in binding buffer (0.05 m Tris, 0.15 m NaCl, pH 8.6). The undiluted culture supernatants (100 μl) were added to the wells for 24 h at 37°C, followed by extensive washing. For IgG1 and IgM determinations, the detecting antibody was then added at a concentration of 2.5 μg/ml to the washed plates for 1 h at room temperature. ExtrAvidin Alkaline phosphatase conjugate (Sigma, St Louis, MO) was added (1/4000) subsequently for 1 h at room temperature. The plates were subsequently developed with phosphatase substrate diluted in substrate buffer (0.05 m K₂CO₃, 0.001 m MgCl₂, pH 9.8). After 30 min, immunoglobulin levels were quantified by an ELISA reader (absorbance 405 nm). The concentration of immunoglobulin (μg/ml) was determined by comparison with a standard curve obtained using a series of serial two-fold dilutions of the purified mouse myeloma immunoglobulin standards. Immunoglobulin levels in the serum of male and female BXSB mice were measured similarly using 100 μl of diluted (1:100) serum.

Measuring autoantibodies (anti-ssDNA antibody)

Relative levels of anti-ssDNA antibody were determined by an ELISA as described above with a few modifications. Briefly, Immulon-1 plates (Dynatech, Alexandria, VA) were coated with 50 μg/ml poly-l-lysine (Sigma) for 1 h at 37°C. Meanwhile, 20 μg/ml calf thymus DNA (Calbiochem, La Jolla, CA) were boiled for 15 min and immediately placed on ice. The plates were washed free of poly-l-lysine and incubated with the denatured DNA at 4°C for 4 days to promote adhesion of the DNA. The plates were next washed, incubated with 100 μl of culture supernatants for 24 h at 37°C, washed, and detected with polyclonal alkaline phosphatase-conjugated goat anti-mouse IgG, IgA, and IgM (H&L), followed by the phosphatase substrate as described above. Immunoglobulin levels were quantified by an ELISA reader (optical density (OD) at 405 nm). Anti-ssDNA antibody levels in the serum of the mice were measured similarly using 100 μl of diluted (1:100) serum.

Immunofluorescent staining and flow cytofluorometric analysis

Cells for immunophenotypical analysis were stained in a buffer consisting of 1× PBS and 1% bovine serum albumin (BSA) (staining buffer). To determine surface CD40L, the B cells were harvested from the cultures after 0, 24, 48 72, 96 and 120 h, washed and resuspended in staining buffer as described previously [3]. To remove possible soluble CD40 that had bound to the CD40L, the B cells were incubated for 3 min in an acidic buffer consisting of 140 mm NaCl and 40 mm citrate, pH 4.0 [12]. The acid treatment did not significantly affect cell viability as determined by trypan blue exclusion and light microscopy. The cells were washed in staining buffer, incubated with biotinylated anti-CD40L MoAb for 30 min at 4°C, followed by incubation with PE–anti-CD45R (B220) MoAb and FITC–streptavidin. The stained cells were analysed by flow cytometry using a FACS IV flow cytometer (Becton Dickinson, Mountain View, CA). As a negative control in the staining for CD40L, a parallel sample was incubated with biotinylated hamster IgG isotype standard followed by PE–anti-CD45R (B220) and FITC–streptavidin. To measure CD40L expression on CD4⁺ T cells, a Th1 cell clone specific for human gammaglobulin (HGG) [13] was stimulated for 16 h at 37°C with plate-bound anti-CD3e MoAb (hamster IgG, clone 145-2C11 (PharMingen) added at 15 μg/ml MoAb to six-well plates and incubated at 37°C overnight prior to washing and adding the Th1 cells). The Th1 cells were then stained as described above for expression of CD40L.

RESULTS

Large B cells from male BXSB mice need fewer signals than B cells from female BXSB mice in order to produce polyclonal antibody

The culture conditions needed to demonstrate the requirement for CD40 signalling in polyclonal antibody production in our system were first delineated using small resting B cells from non-autoimmune C57Bl/10 mice of the same haplotype (H-2^b) as the BXSB mice. It was shown that the B cells produced no IgM when unstimulated, or when stimulated with IL-4 and IL-5 (1 ng/ml and 0.5 ng/ml, respectively) alone. Cross-linking the surface IgD on the B cells by a multivalent antibody conjugated to dextran (anti-IgD-dextran), in the presence of IL-4 and IL-5 induced a low level IgM production (440 ng/ml), which was increased (2296 ng/ml) following the addition of anti-CD40 MoAb. The stimulatory effect of CD40 signalling was not observed if the B cell density was too high (10⁶/ml instead of 10⁵/ml), or if the levels of IL-4 and IL-5 were too low (0.25 ng/ml and 0.125 ng/ml, respectively). Based on these results, subsequent experiments to examine antibody production by B cells from BXSB mice were conducted using 10⁵ B cells/ml and IL-4/IL-5 at a concentration of 0.5 ng/ml and 0.25 ng/ml, respectively. It was expected that this concentration of cytokines would not be stimulatory alone, but would promote antibody production in association with anti-CD40 MoAb and anti-IgD-dextran.

In order to compare B cells from male and female BXSB mice in terms of their respective signalling requirements for antibody production, small and large B cells from male and female BXSB mice were incubated with IL-4 and IL-5, together with increasing concentrations of anti-IgD-dextran antibody in the presence or absence of anti-CD40 MoAb. The results demonstrated that large B cells from male BXSB mice made significantly more IgM and IgG1 than size-matched B cells from female BXSB mice when stimulated with (i) IL-4/IL-5 alone, (ii) IL-4/IL-5 in combination with low concentrations of anti-IgD-dextran, or (iii) IL-4/IL-5 in combination with anti-CD40 MoAb (Fig. 1). Large B cells from female mice made the most IgG1 and IgM when stimulated with a combination of IL-4/IL-5, anti-CD40 MoAb, 3 μg/ml anti-IgD-dextran. In contrast, the combination of these three stimuli did not enhance antibody production by large B cells from male mice beyond that induced by cytokines and anti-CD40 MoAb alone. The addition of a higher concentration of anti-IgD-dextran to cultures stimulated with IL-4/IL-5 and anti-CD40 MoAb inhibited antibody production by all groups of B cells. However, it appeared that the large B cells from male mice were especially sensitive to the inhibitory effects of high concentrations of anti-IgD-dextran.

Fig. 1.

Small and large B cells from male BXSB mice have fewer requirements for total antibody production than small and large B cells from female BXSB mice. Small and large B cells from male (▪) and female (□) BXSB mice were cultured either with IL-4 and IL-5 alone, or in the presence of increasing concentrations of anti-IgD-dextran ± anti-CD40. Culture supernatants (six wells/group) were collected at 120 h and analysed for the presence of IgM and IgG1 by ELISA. The values presented are the means ± s.d. from three experiments. *Significantly different (P < 0.05) by Student's unpaired t-test from that of control wells (supernatants from cultures of B cells from female mice).

Small B cells from female mice, similar to large B cells from female mice, needed all three signals, i.e. cytokines, CD40 cross-linking and immunoglobulin cross-linking, for optimal antibody production. Although small B cells from male mice were fairly similar to small B cells from female mice in their requirements for antibody production, they did, at a couple of data points, produce more antibody then their female counterparts when activated with only two of the three stimuli.

B cells from male BXSB mice secreted more anti-ssDNA than B cells from female BXSB mice

Experiments were conducted to determine whether the propensity to express CD40L enhanced the ability of the large B cells from male BXSB mice to secrete autoAbs as well as total immunoglobulin. The results demonstrated that in the presence of IL-4/IL-5 alone, the large B cells from male BXSB mice synthesized significantly more anti-ssDNA antibody than large B cells from female BXSB mice (Fig. 2). Although adding anti-IgD-dextran to the cultures increased the level of anti-ssDNA antibody produced by both groups of large B cells, the large B cells from male BXSB mice still produced significantly more anti-ssDNA antibody than large B cells from female BXSB mice. A further addition of anti-CD40 MoAb to the cultures increased the anti-ssDNA antibody produced by large B cells from female BXSB mice, but did not increase antibody production by large B cells from male BXSB mice. Small B cells from male and female BXSB mice for the most part produced similar levels of anti-ssDNA antibody, with the exception of cultures stimulated with cytokines and anti-IgD-dextran, in which the small B cells from male mice exhibited a small but significantly increased capacity to secrete autoAbs compared with small B cells from female BXSB mice.

Fig. 2.

Small and large B cells from male BXSB mice secrete more anti-ssDNA antibody in the absence of exogenous CD40L than size-matched B cells from female BXSB mice. Small and large B cells from male (▪) and female (□) BXSB mice were cultured with IL-4 and IL-5 alone, or in the presence of 0.3 ng/ml anti-IgD-dextran, ± anti-CD40 MoAb. Culture supernatants (six wells/group) were collected at 120 h and analysed for the presence of anti-ssDNA by ELISA. The values presented are the means ± s.d. from three experiments. *Significantly different (P < 0.05) by Student's unpaired t-test from that of control wells (supernatants from B cell cultures from female BXSB mice).

Suboptimally stimulated B cells from male BXSB mice expressed more CD40L than B cells from female BXSB mice

To examine further the correlation between B cell CD40L expression and non-specific antibody production, the next set of experiments was conducted to examine CD40L expression on B cells incubated under the culture conditions used to induce antibody synthesis. As a positive control, CD40L expression was also examined on anti-CD3 MoAb-stimulated Th1 cells. When the kinetics of IL-4- and IL-5-induced CD40L expression was examined on large B cells from male and female BXSB mice, the greatest percentage of B cells expressing CD40L was found in cultures of large B cells from male BXSB mice following activation for 3 days (Fig. 3). The large B cells from male mice also expressed increased levels of CD40L at 4 days. In comparison, the expression of CD40L on the large B cells from female BXSB mice remained at background levels at every time point tested. The small B cells from male and female BXSB mice expressed similarly low levels of CD40L at all time points tested after stimulation with IL-4 and IL-5 (data not shown). Adding anti-IgD-dextran or anti-CD40 MoAb to the cultures of large B cells from male BXSB mice did not enhance B cell expression of CD40L in comparison with cultures stimulated with IL-4 and IL-5 alone (data not shown).

Fig. 3.

B cells from male BXSB mice express higher levels of CD40L than B cells from female BXSB mice. (a) B cells from male and female BXSB mice were incubated with IL-4 and IL-5 alone. The cells were harvested after 1, 2, 3 or 4 days of culture and two-colour stained with PE–anti-CD45R (B220) MoAb and either biotinylated anti-CD40L or biotinylated hamster IgG (control) followed by FITC–streptavidin, and analysed by flow cytometry. Quadrants in the dot plots were set using biotinylated hamster IgG control and PE–anti-CD45R (B220). The numbers in the right-hand corners of the dot plots represent the percentage of double-positive cells. Similar results were obtained in a second experiment. (b) Murine Th1 cell clone 12-11 specific for human gammaglobulin (HGG) was left untreated (resting), or stimulated for 16 h with plate-bound anti-CD3 MoAb (activated) and stained with anti-CD40L MoAb or isotype control immunoglobulin as described above.

CD40-Ig blocks total immunoglobulin and autoAb production by B cells from male BXSB mice

We had previously shown that the CD40-Ig fusion protein blocked the spontaneous proliferation of large B cells from male BXSB mice, as well as the ability of these CD40L⁺ B cells to induce DNA synthesis in small, resting B cells [3]. To confirm similarly that the CD40L expressed by the B cells from male BXSB mice promoted their increased antibody synthesis under suboptimal stimulatory conditions, large B cells from the male BXSB mice were activated with IL-4/IL-5 ± anti-IgD-dextran, in the presence or absence of CD40-Ig. The addition of CD40-Ig to cultures of B cells stimulated with IL-4/IL-5 and anti-IgD-dextran inhibited by 88% the production of total IgG1 by large B cells from male BXSB mice (Fig. 4). IgM production by similarly stimulated large B cells was also significantly inhibited (65%). In addition to total immunoglobulin, the level of anti-ssDNA antibody produced by the large and small B cells from male BXSB mice was also significantly inhibited by CD40-Ig.

Fig. 4.

CD40-Ig blocks total immunoglobulin and anti-ssDNA production by large B cells from male BXSB mice. (a) Large B cells from male BXSB mice were isolated and cultured with IL-4/IL-5 in the presence or absence of 0.3 ng/ml anti-IgD-dextran. CD40-Ig (□, 25 μg/ml) or control immunoglobulin (▪) was added at the initiation of culture. Culture supernatants were collected at 120 h and analysed for the presence of IgM and IgG1 by ELISA. (b) Large B cells from male BXSB mice were isolated and cultured with IL-4/IL-5 in the presence of 0.3 ng/ml anti-IgD-dextran. CD40-Ig (□, 25 μg/ml) or control immunoglobulin (▪) was added at the initiation of culture. Culture supernatants were collected at 120 h and analysed for the presence of anti-ssDNA antibody by ELISA. *Significantly different (P < 0.05) by Student's unpaired t-test from that of controls (B cells treated with control-Ig). Similar results were achieved in a second experiment.

DISCUSSION

Based on our early finding that B cells from male BXSB mice expressed CD40L, and were also more sensitive to CD40 signalling [3], we now sought to determine how these characteristics affected the ability of these B cells to secrete non-specific antibody. While cytokines alone, such as IL-4, can induce the transcription of immature RNA for heavy chain isotypes, additional signals such as LPS, immunoglobulin receptor cross-linking, or signals through CD40 are necessary for induction of immunoglobulin synthesis. Snapper et al. showed that a multivalent, but not bivalent, form of anti-IgD synergized with CD40L and cytokines (IL-4 and IL-5) for immunoglobulin production and isotype switching in cultures of B cells from non-autoimmune mice [14]. These findings suggested that although two signals, if strong enough, are sufficient for antibody production, three signals, including immunoglobulin cross-linking, T cell-derived cytokines, and CD40 cross-linking are needed under more physiological conditions. Given the importance of CD40 signalling on B cell antibody production, it seemed possible that the CD40L expressed on the B cells of male BXSB mice, together with the increased sensitivity of these B cells to CD40 signalling, could to a certain extent circumvent the need for exogenous CD40L, and thereby promote a polyclonal increase in both total immunoglobulin, and autoAbs.

The results demonstrate that the propensity of the large B cells from male BXSB mice to express CD40L apparently enabled them to secrete significantly more IgM and IgG1 upon stimulation with IL-4/IL-5 alone, or in combination with low levels of anti-IgD-dextran, compared with size-matched B cells from female BXSB mice. The large B cells from male mice were also better than size-matched B cells from female mice in secreting IgM and IgG1 when stimulated with IL-4/IL-5 and anti-CD40 MoAb, suggesting that the increased sensitivity to CD40 signalling observed in the former population of B cells also promoted their polyclonal antibody production. In contrast, optimal antibody production by large B cells from female BXSB mice, as well as by small B cells from female mice, required stimulation with a combination of IL-4/IL-5, anti-CD40 MoAb and low to moderate levels of anti-IgD-dextran. These results suggest that B cells, especially the subset of large B cells, from male BXSB mice had fewer requirements for polyclonal antibody production than size-matched B cells from female BXSB mice.

In addition to an increased capacity to secrete total immunoglobulin under suboptimal conditions, the large B cells from male BXSB mice also secreted more anti-ssDNA compared with size-matched B cells from female BXSB mice when stimulated with IL-4/IL-5 alone, or in the presence of moderate levels of anti-IgD-dextran. Large B cells from female mice secreted high levels of anti-ssDNA antibody only when stimulated with a combination of IL-4/IL-5, anti-CD40 MoAb and anti-IgD-dextran. The role of CD40L in the ability of the large B cells from male mice to secrete antibody in the absence of exogenous CD40 signalling was demonstrated by the fact that the specific blocker, CD40-Ig, but not the control-Ig, significantly inhibited both total immunoglobulin production, and anti-ssDNA antibody production by large B cells from male BXSB. Taken together, these results suggest that the propensity of the large B cells from male mice to express CD40L promoted non-specific antibody production.

The increased antibody production observed in cultures of large B cells from male BXSB mice correlated with an increased expression of CD40L compared with large B cells from female BXSB mice. The expression of CD40L by the B cells from male BXSB mice appeared to be primarily dependent upon IL-4 and IL-5, and was not further up-regulated by anti-IgD-dextran and/or anti-CD40 MoAb. This finding is consistent with a previous report showing that CD40L expression on B cells was enhanced with soluble anti-immunoglobulin, but was not up-regulated with anti-CD40 MoAb or anti-immunoglobulin conjugated to Sepharose [15]. CD40L appears to be easier to detect at the functional rather than phenotypic level: at best, CD40L was expressed on only 24% of the cultured B cells from male BXSB mice in the present study. In addition, measurable levels of CD40L have been shown in this and a previous study to be low (0–7%) on freshly isolated B cells [3]. CD40L has similarly been difficult to detect on freshly isolated T cells: even when stimulated with immobilized anti-CD3 antibody, only 14% of purified splenic CD4⁺ T cells expressed CD40L [16]. It has also been shown that CD40L expression on T cells is considerably impaired in the presence of CD40⁺ B cells, which can cause rapid down-regulation and endocytosis of CD40L upon engagement by CD40 [17]. It seems likely that the ability of CD40 on B cells to down-regulate CD40L expression could complicate detecting CD40L on CD40-expressing B cells. Regardless of the difficulties in detecting transient CD40L expression, the amount of CD40L expressed on the B cells from male BXSB mice was sufficient to promote antibody production.

In addition to their ability to express CD40L, and their hyper-responsiveness to CD40 signalling, it is also possible that the large B cells from male BXSB mice are hyper-responsive to IL-4 and/or IL-5. In the present study it was demonstrated that the large B cells from male BXSB mice have the capacity to produce immunoglobulin in the presence of IL-4 and IL-5 alone. Whether this effect represents an intrinsically increased capacity for antibody production, or an increased responsiveness to IL-4, is not known. It is known that CD40 ligation can increase B cell responsiveness to IL-4 by increasing the expression of IL-4 receptors (IL-4R) [18]. Perhaps the B cells from male BXSB mice, because of CD40L–CD40 interactions, express increased levels of IL-4R, thereby promoting increased responsiveness to IL-4. However, this possibility seems unlikely, since a preliminary examination of constitutive and activation-induced expression of IL-4R on B cells from male and female BXSB mice showed little difference.

The capacity of CD40L-expressing B cells to promote antibody production could provide an important mechanism for the development of lupus in both mice and humans. Similar to our finding that B cell CD40L expression in male BXSB mice promoted antibody synthesis, it has been shown that B cells from patients with active, but not inactive lupus constitutively express high levels of CD40L capable of stimulating antibody synthesis [19]. It has also been shown that CD40L expressed by B cells is important in promoting immunoglobulin production by non-autoimmune B cells [11]. We have shown that the CD40L-expressing B cells from male BXSB mice have an increased capacity to secrete polyclonal antibody. Since lupus is primarily antibody-driven, delineating a role for CD40L⁺ B cells in promoting antibody production should help expand our understanding of the immune pathology of this disease, and should confirm CD40/CD40L interactions as a target for immunotherapy.