Role of the Costimulatory Molecule CD28 in the Development of Lupus in MRL/lpr Mice

Yoshifumi Tada, Kohei Nagasawa, Alexandra Ho, Fumitaka Morito, Syuichi Koarada, Osamu Ushiyama, Noriaki Suzuki, Akihide Ohta and Tak W. Mak
J Immunol September 15, 1999, 163 (6) 3153-3159;

Abstract

MRL/Mpj-lpr/lpr (MRL/lpr) mice develop autoimmune disorders, including lymphoproliferation, glomerulonephritis, autoantibody production, and hypergammaglobulinemia. To investigate the role of the costimulatory molecule CD28 in the development of these disorders, MRL/lpr mice lacking CD28 were generated by gene targeting. Compared with CD28+/+ MRL/lpr mice, CD28−/− MRL/lpr mice showed decreased lymphadenopathy but increased splenomegaly associated with the expansion of abnormal B220+ TCRαβ+ T cells. Although levels of IgM Abs were unchanged in CD28−/− MRL/lpr mice, the production of anti-DNA IgG Abs and IgG rheumatoid factors were suppressed. IgG deposition in the glomeruli was markedly decreased, and the development of glomerulonephritis was significantly retarded. Furthermore, renal vasculitis and arthritis were absent in CD28−/− MRL/lpr mice. These results indicate that, although CD28 is not required for the generation of the abnormal T cell population in MRL/lpr mice, it does play an important role in the development of autoimmune disease in these animals.

The MRL/Mpj-lpr/lpr (MRL/lpr)3 mouse spontaneously develops an autoimmune disorder resembling systemic lupus erythematosus in humans. The clinical features of the disease include autoantibody production, hypergammaglobulinemia, nephritis, arthritis, vasculitis, and lymphadenopathy (1, 2, 3). The molecular defect underlying the phenotype was identified as a mutation of the gene encoding Fas, a cell surface receptor belonging to the TNF receptor superfamily whose engagement induces apoptosis (4, 5). Impaired Fas expression and Fas-mediated apoptotic death of T cells is thought to be the main cause of the expansion of abnormal B220+ TCRαβ+ T cells in lymph nodes (LN) of MRL/lpr mice (6, 7).

The critical role of T cells in the development of autoimmune disorders in MRL/lpr mice has been demonstrated in various studies in which treatment with Abs directed against the T cell markers CD4 (8, 9, 10, 11), CD3 (12), or Thy 1.2 (13), or administration of cyclosporin A (14), have all mitigated the development of the lupus-like phenotype to some degree. Autoantibody levels and organ infiltration were decreased in CD4-deficient (15) and MHC class II-deficient (16) MRL/lpr mice, despite a degree of lymphadenopathy similar to that observed in standard MLR/lpr mice. In contrast, CD8-deficient (15) and MHC class I-deficient (17, 18, 19, 20) MRL/lpr mice showed a reduced accumulation of B220+ TCRαβ+ T cells in LN and either unchanged (15, 17, 18) or decreased (19, 20) serum IgG autoantibody levels, respectively. Finally, MRL/lpr mice lacking the TCRαβ showed reduced renal disease and lymphadenopathy but unchanged levels of anti-DNA Abs (21). Together, these studies indicate that each pathological feature of the MRL/lpr phenotype, including lymphoproliferation, autoantibody production, and organ disease, is subject to the influence of different genetic factors.

Successful T cell activation generally requires a primary signal through the TCR coupled with a secondary signal provided by costimulatory molecules (22). CD28 has been shown to provide the most effective costimulatory signal for T cell activation (23, 24, 25). Recent studies using the CTLA4-Ig to block the interaction of CD28 with its ligands CD80 and CD86 have revealed that various autoimmune diseases including (NZB × NZW)F1 lupus mice (26) can be prevented or treated by the inhibition of CD28-mediated costimulation (26, 27, 28, 29, 30, 31). These results indicate a critical role for CD28 in the development of autoimmune disease.

To further investigate the role of CD28 costimulation in the development of the lupus-like autoimmune disease in MRL/lpr mice, we generated gene-targeted CD28-deficient MRL/lpr mice. In this paper we show that the absence of CD28 in MRL/lpr mice results in an improved clinical picture, including markedly decreased autoantibody production, milder glomerulonephritis, and an absence of vasculitis and cellular infiltration in organs.

Materials and Methods

Mice

Homozygous MRL/lpr mice were purchased from The Jackson Laboratory (Bar Harbor, ME). CD28-deficient mice (32) were backcrossed to MRL/lpr mice for four generations and genotyped by PCR using DNA obtained by ear biopsy (32). The disrupted Fas locus in the mutant (the lpr locus) and intact Fas locus in the wild type were detected by PCR using the following primers: A, AAA AGG TTA CAA AAG GTC ACC C (Fas sense); B, GTT GCG ACA CCA GTT AT (Fas antisense); and C, AAC GCA GTC AAA TCT GCT (lpr antisense). Amplifications were performed under standard conditions for 35 cycles with temperatures of 94°C melting, 56°C annealing, and 72°C extension. PCR products were 170 bp (primers A and B) and 330 bp (primers A and C) for the wild-type Fas and mutant lpr genes, respectively. The CD28−/−lpr/lpr mice were originally bred and maintained in the Ontario Cancer Institute animal facility (Toronto, Canada), whereas progeny mice used in this study were born and maintained at the Saga Medical School animal facility (Saga, Japan). Care of animals was in accordance with guidelines of the Canadian Medical Research Council and the Saga Medical School guidelines for animal experimentation.

Flow cytometric analysis

Mice were sacrificed at 18 wk of age, the spleens removed and weighed, and the inguinal and axillar LN removed. Single cell suspensions were obtained by passing tissue through a mesh screen in DMEM (Life Technologies, Grand Island, NY). Cells (1 × 106) were suspended in PBS containing 1% BSA and 0.1% sodium azide and then incubated with various combinations of conjugated Abs for 30 min at 4°C. Anti-TCRαβ, anti-B220, anti-CD4, and anti-CD8 mAbs were purchased from Caltag (South San Francisco, CA). At least 104 stained cells/sample were analyzed using a FACScan (Becton Dickinson, Mountain View, CA) equipped with Lysis software (Becton Dickinson).

Measurement of anti-DNA Ab and rheumatoid factor levels

Anti-DNA Abs were detected by ELISA. Flat-bottom plates (Nunc, Roskilde, Denmark) were coated with 1.5 mg/ml of native calf thymus DNA (Life Technologies) in buffer containing 0.1 M sodium bicarbonate and 0.05 M citric acid at 4°C overnight. After washing with PBS containing 0.05% Tween 20, plates were blocked with PBS containing 1% BSA for at least 1 h at room temperature. Serum samples were serially diluted (starting at 1/200) and added to the plates for a 1-h incubation at 37°C. The plates were then washed four more times, and peroxidase-conjugated goat anti-mouse IgG1, IgG2a, IgG2b, IgG3, or IgM (Southern Biotechnology Associates, Birmingham, AL) was added and incubated for 1 h at 37°C. Ab binding was visualized using orthophenylenediamine (Sigma, St. Louis, MO), and plates were read at an absorbance of 490 nm. Twofold dilutions of a mixture of sera from aged female MRL/lpr mice (constituting the “standard serum”) were added to each plate as an internal control, and a standard curve was calculated. The standard serum was defined as 100 U, and the Ab titers of experimental serum samples were calculated with Nippon Intermed software (Osaka, Japan) from the standard curve. Although this method detected predominantly anti-dsDNA Abs, it may also detect low levels of anti-ssDNA Abs. To determine the sensitivity of this method for anti-ssDNA, we measured Ab levels of anti-native DNA and anti-ssDNA (boiled for 10 min) from reference sera of patients with connective tissue diseases known to have only anti-ssDNA. The results showed that >10 times higher concentrations of reference sera from ssDNA-coated ELISA were required to give equivalent OD values as shown in native DNA-coated ELISA. For the measurement of IgG and IgM rheumatoid factor, human IgG (Chemicon International, Temecula, CA) or rabbit IgG (Zymed, San Francisco, CA) was coated onto plates at 10 μg/ml or 5 μg/ml in carbonate buffer, respectively, and the same procedures were followed as described above.

Measurement of serum Igs by ELISA

Serum samples were assayed for IgG1, IgG2a, IgG2b, IgG3, and IgM levels by ELISA using a clonotyping system and a mouse Ig standard panel (Southern Biotechnology Associates) according to manufacturer’s instructions.

Histopathology and immunofluorescence

Kidneys, livers, lungs, and ankles were fixed in buffered 10% formalin for 48 h, and ankles were further decalcified in 10% EDTA. Tissues were embedded in paraffin, sectioned, and stained with hematoxylin and eosin. For evaluation of glomerular lesions, >20 glomeruli/mouse were photographed. Each glomerular lesion was scored for severity in a blinded fashion as follows: 0, normal; 0.5, minimal; 1, mild; 2, moderate; and 3, severe. The average severity grade was calculated and defined as the renal score of the mouse. Renal vascular lesion was scored in a similar fashion. At least six interlobular arteries were scored for perivascular infiltration, and the average scores were calculated. For immunofluorescence, kidneys were embedded in Tissue-tek medium (Miles, Elkhart, IN) and snap-frozen in hexane chilled with liquid nitrogen. Sections were cut using a cryostat, fixed in acetone, rinsed with PBS, and incubated with F(ab′)2 fluorescein-conjugated anti-mouse IgG (Biosource International, Camarillo, CA) at 4°C overnight. Glomerular IgG deposits were evaluated in a similar manner.

Statistics

Mann-Whitney’s U test was used to determine the statistical significance of differences between groups. Survival of female MRL/lpr mice was analyzed by the Kaplan-Meier’s method, and the significance of differences was determined by the Log-rank test.

Results

Lymphadenopathy, splenomegaly, and cell populations in CD28−/−MRL/lpr mice

CD28+/+lpr/lpr and CD28−/−lpr/lpr mice were sacrificed at 18 wk of age and examined for spleen weights, LN cell numbers, and populations of T and B lymphocytes in spleens and LN. As shown in Fig. 1⇓, the spleens of CD28−/−lpr/lpr mice were increased in size compared with those of CD28+/+lpr/lpr mice (0.63 ± 0.07 g vs 0.36 ± 0.05 g, p < 0.005, Fig. 1⇓A). LN cell counts were significantly decreased in CD28−/−lpr/lpr mice compared with CD28+/+lpr/lpr mice (20.0 ± 3.1 × 107 vs 41.1 ± 8.3 × 107, p < 0.05), although they were still 20-fold greater than those of CD28−/−lpr/+ (Fig. 1⇓B). FACS analysis of the total spleen cell population revealed that B220+ TCRαβ+ cells and TCRαβ+CD4−CD8− cells were increased by 2.5- to 3-fold in CD28−/−lpr/lpr mice compared with CD28+/+lpr/lpr mice (p < 0.001, Table I⇓). In contrast, B220+ TCRαβ− B cells were decreased in CD28−/−lpr/lpr mice (p < 0.01). Furthermore, although total CD4+ T cells were decreased, the CD4+B220+ T cell population was higher in CD28−/−lpr/lpr mice (p < 0.001), suggesting more activated CD4+ T cells in the spleen. These data indicate that the accelerated splenomegaly in CD28−/−lpr/lpr mice was due to the increased accumulation of B220+ TCRαβ+ cells. However, the numbers of B220+ TCRαβ+ cells, CD4+ T cells, CD8+ T cells, and normal B cells were similar in LN of CD28−/−lpr/lpr and CD28+/+lpr/lpr mice (Table I⇓). We further analyzed the phenotype of LN T cells from CD28−/−lpr/lpr mice, which showed similar surface phenotype as those from CD28+/+lpr/lpr mice. Most T cells showed CD44high, CD2−, and TCRαβint (data not shown).

View this table:
Table I.

Analysis of lymph node cells and spleen cells of CD28−/−lpr/lpr micea

FIGURE 1.
FIGURE 1.

Spleen size and LN cellularity in CD28+/+ and CD28−/− MRL/lpr mice. Spleen weights (A) and cell numbers of bilateral axillar and inguinal LN (B) of 18-wk-old CD28+/+lpr/lpr (n = 10), CD28−/−lpr/lpr (n = 9), CD28+/+lpr/+ (n = 3), and CD28−/−lpr/+ (n = 6) mice are shown. Values represent the mean ± SEM. ∗, p < 0.05; and ∗∗, p < 0.005 as compared with CD28+/+lpr/lpr mice.

Serum Ig levels in CD28−/− MRL/lpr mice

When serum Ig levels were measured at 18 wk of age, CD28−/−lpr/lpr mice were found to produce significantly decreased levels of IgG1, IgG2a, and IgG3 compared with CD28+/+lpr/lpr mice (p < 0.0001), whereas levels of IgG2b and IgM were not different (Fig. 2⇓). In fact, IgG1 levels in CD28−/−lpr/lpr mice were comparable with those in control CD28+/+lpr/+ mice. In addition, IgG1 levels were significantly decreased in CD28−/−lpr/lpr mice compared with IgG2a levels in the same mice (mean ± SEM, 3.4 ± 0.5 mg/ml vs 9.3 ± 1.3 mg/ml, p = 0.0002). Although control CD28+/+lpr/lpr mice produced slightly decreased levels of IgG1 compared with IgG2a (17.5 ± 2.7 mg/ml vs 25.5 ± 2.1 mg/ml, p = 0.02), the ratio of serum IgG1/IgG2a was lower in CD28−/−lpr/lpr mice compared with CD28+/+lpr/lpr mice (0.37 vs 0.69). These results indicate that the absence of CD28 suppresses serum IgG1 levels more severely than IgG2a levels in MRL/lpr mice.

FIGURE 2.
FIGURE 2.

Serum Ig levels in CD28+/+ and CD28−/− MRL/lpr mice. Levels of IgG isotypes (A--D) and IgM (E) were measured by ELISA in serum from 18-wk-old mice. ∗, p < 0.0001 as compared with CD28+/+lpr/lpr mice.

Autoantibody levels in CD28−/− MRL/lpr mice

Levels of serum anti-DNA Abs of various isotypes and IgG and IgM rheumatoid factors were measured at 18 wk of age. As shown in Fig. 3⇓, anti-DNA Abs of the IgG1, IgG2a, IgG2b, and IgG3 subclasses were significantly decreased in CD28−/−lpr/lpr mice compared with CD28+/+lpr/lpr mice (p = 0.0005, p < 0.0001, p = 0.0001, p = 0.0005, respectively). Six of 13 CD28−/−lpr/lpr mice showed undetectable levels of IgG1 anti-DNA Abs, whereas the rest showed low to medium (10–100 U) levels. In contrast, levels of IgG1 anti-DNA Abs in CD28+/+lpr/lpr mice ranged from low (<10 U) to high (>100 U). However, two CD28−/−lpr/lpr mice showed levels of IgG2a and IgG3 anti-DNA Abs that were higher than median seen in CD28+/+lpr/lpr mice: IgG2a was increased in 1 of 13 CD28−/−lpr/lpr mice, whereas IgG3 was elevated in 2 of 13 mice. These results indicate that although IgG anti-DNA Ab levels are generally suppressed in CD28−/−lpr/lpr mice, some animals can continue to produce elevated levels of IgG anti-DNA Abs even in the absence of CD28. In contrast, IgM anti-DNA Ab levels were not different between CD28−/−lpr/lpr and CD28+/+lpr/lpr mice. Furthermore, although both mice produced similar IgM rheumatoid factor levels, CD28−/−lpr/lpr mice showed significantly decreased IgG rheumatoid factor levels (Fig. 3⇓F). These data indicate that isotype switching of autoantibodies is impaired in MRL/lpr mice in the absence of CD28.

FIGURE 3.
FIGURE 3.

Autoantibody levels in CD28+/+ and CD28−/− MRL/lpr mice. Levels of anti-DNA Abs of IgG (A–D) and IgM (E) isotypes, and rheumatoid factors of IgG (F) and IgM (G), were measured by ELISA in serum from 18-wk-old mice. ∗, p = 0.0005; ∗∗, p < 0.0001; and ∗∗∗, p = 0.0007 as compared with CD28+/+lpr/lpr mice.

Renal disease and mortality in CD28−/− MRL/lpr mice

Renal histology was assessed in CD28−/−lpr/lpr and CD28+/+lpr/lpr mice. Glomerular lesions were scored for severity as described in Materials and Methods. Glomerular lesions and vascular lesions in CD28−/−lpr/lpr mice were less severe than those observed in CD28+/+lpr/lpr mice (Fig. 4⇓). Mean score ± SEM of CD28−/−lpr/lpr and of CD28+/+lpr/lpr mice were 0.91 ± 0.07 and 1.81 ± 0.19, respectively (p < 0.0005). Whereas CD28+/+lpr/lpr mice showed moderate to severe mesangial proliferation (Fig. 5⇓A), and some mice showed marked hyalinization of glomeruli or crescent formation (2 of 11 mice, Fig. 5⇓B), CD28−/−lpr/lpr mice showed only mild mesangial proliferation (Fig. 5⇓C). Interestingly, the two CD28−/−lpr/lpr mice that showed relatively high levels of IgG3 anti-DNA Ab (Fig. 3⇑D) also exhibited higher glomerular pathology scores in the CD28−/−lpr/lpr group: one showed the highest score and the other showed the third highest (1.31 and 0.98, respectively). These results suggest that high IgG3 anti-DNA levels correlate with advanced renal pathology among CD28−/−lpr/lpr mice. Most strikingly, in contrast to the dense perivascular and interstitial infiltration of inflammatory cells observed in CD28+/+lpr/lpr mice, no cellular infiltration was observed in CD28−/−lpr/lpr mice (Figs. 4⇓ and 5⇓).

FIGURE 4.
FIGURE 4.

Renal pathology in CD28+/+ and CD28−/− MRL/lpr mice. Glomerular lesions and renal vascular lesions were scored as described in the Materials and Methods. ∗, p < 0.0005; and ∗∗, p < 0.0001 as compared with CD28+/+lpr/lpr mice.

FIGURE 5.
FIGURE 5.

Renal histology in CD28+/+ and CD28−/− MRL/lpr mice. Glomerulonephritis with mesangial proliferation (A) and with lobulation and crescent formation (B) in CD28+/+lpr/lpr mice. Mild mesangial proliferation in CD28−/−lpr/lpr mice (C). Dense perivascular infiltration in CD28+/+lpr/lpr mice (D) and absence of infiltration in CD28−/−lpr/lpr mice (E). All sections were prepared from mice at 18 wk of age. All gels were stained with hematoxylin and eosin; magnifications, ×40 (A–C) and ×20 (D and E).

Immunofluorescence studies also confirmed these results. Although there were moderate to severe IgG deposits in the glomeruli of CD28+/+lpr/lpr mice (Fig. 6⇓A), only mild deposits were observed in the mesangium of CD28−/−lpr/lpr mice (Fig. 6⇓B). We examined renal pathology of CD28−/−lpr/lpr mice (>24 wk, n = 3) to assess the progression of nephritis. Perivascular and interstitial infiltration were not detected in kidneys of these mice. However, glomerular lesions such as mesangial proliferation, hyalinization, or tuft atrophy did develop in these old CD28−/−lpr/lpr mice. One mouse showed moderate glomerular lesion (score = 1.7, 25 wk), and two other mice showed global sclerosis of glomeruli, indistinguishable from those observed in CD28+/+lpr/lpr mice (score = 2.5 and 2.7, 28 wk and 31 wk, respectively). These results indicate that the vasculitis and interstitial infiltration typical in MLR/lpr mice are completely abolished in the absence of CD28, whereas the progression of glomerulonephritis is only retarded. The inflammatory infiltration observed in livers, lungs, and ankle joints of CD28+/+lpr/lpr mice was not observed in similar organs of CD28−/−lpr/lpr mice (not shown). As a result of these ameliorated pathologies, the mortality rate of female CD28−/−lpr/lpr mice was decreased compared with that of CD28+/+lpr/lpr mice. At 20 wk of age, 100% (10/10) of CD28−/−lpr/lpr mice compared with 61.5% (8/13) of CD28+/+lpr/lpr mice were alive, whereas at 25 wk of age, 70.0% (7/10) and 30.8% (4/13), respectively, had survived (p < 0.05, Log-rank test).

FIGURE 6.
FIGURE 6.

Immunofluorescence of renal Ig deposits in CD28+/+ and CD28−/− MRL/lpr mice. IgG deposits in CD28+/+lpr/lpr mice (A) and in CD28−/−lpr/lpr mice (B) at 18 wk of age. Magnification, ×40.

Discussion

In this study, we have investigated the role of the costimulatory molecule CD28 in the development of lupus-like autoimmune disease in MRL/lpr mice. The results have demonstrated that CD28 is differentially required for the development of various autoimmune disease features in MRL/lpr mice. The role of CD28 in the accumulation of abnormal B220+ TCRαβ+ T cells in MRL/lpr mice is minimal. Although the number of abnormal T cells was reduced by 50% in LN of CD28−/−lpr/lpr mice, splenomegaly was accelerated by the expansion of this population. Thus, CD28 is either not required or can be substituted for by other molecules during the generation and expansion of B220+ T cells. The more accelerated splenomegaly in CD28−/−lpr/lpr mice may be the result of preferential homing to the spleen, local accumulation, or extrathymic development and proliferation in the spleen. In CD28-deficient mice, there are no gross abnormalities in the sizes of the LN and the spleen (32). Yet in MRL/lpr mice, treatment with anti-L-selectin Ab prevented lymphadenopathy by blocking the homing of B220+ TCRαβ+ T cells. These mice also have massive accumulation of the abnormal T cells in the spleens and the sizes of the Peyer’s patches are similar (33). These results indicate that local factors and adhesion molecules play a role in the accumulation of B220+ TCRαβ+ T cells. Although the role of CD28-CD80/CD86 interaction in homing of T cells was unclear, it is possible that the loss of CD28 altered the local factors and the distribution of B220+ TCRαβ+ T cells. Accelerated splenomegaly with increased accumulation of B220+ TCRαβ+ T cells has also been observed in CD4-deficient MRL/lpr mice (15). However, in contrast to CD28-deficient MRL/lpr mice, no reduction in LN cell numbers were observed in these mice.

CD28-deficient MRL/lpr mice showed decreased serum levels of IgG1, IgG2a, and IgG3 but unchanged levels of IgM, suggesting that polyclonal IgG production is regulated by CD28 costimulation. In CD28-deficient MRL/lpr mice, the level of serum IgG1 (an isotype generally associated with Th2 responses) was much lower than that of IgG2a (associated with Th1 responses). IgG1/IgG2a ratio was lower in CD28−/−lpr/lpr mice compared with control MRL/lpr mice (0.37 vs 0.69), showing more severe reduction in IgG1 levels. This result is consistent with previous reports in which CD28-deficient mice or mice injected with the CD28 inhibitor CTLA4-Ig showed a specific decrease in the production of Th2-induced IgG1 Ab compared with Th1-induced IgG2a Ab (32, 37). We therefore conclude that the absence of CD28 severely impairs Th2 responses in MRL/lpr mice.

CD28-deficient MRL/lpr mice also exhibited decreased production of anti-DNA Ab of all IgG subclasses and IgG rheumatoid factor, consistent with the reduction in IgG Ab production observed in CD28-deficient C57BL/6 mice (32). However, levels of IgM anti-DNA Ab and IgM rheumatoid factor were comparable between CD28−/− and CD28+/+ MRL/lpr mice. The critical role of T cells in autoantibody production has been established by experiments in which CD4+ T cells were depleted by Ab treatment (8, 9, 10, 11) and by analyses of CD4-deficient (15) and TCRαβ+ T cell-deficient (21) MRL/lpr mice. Our results imply that T cell-dependent isotype switching in MRL/lpr mice is a function of CD28 signaling, even though B cells in MRL/lpr mice are intrinsically defective (34, 35, 36). However, a few CD28−/− MRL/lpr mice appeared to be able to circumvent the requirement for CD28, because their production of anti-DNA Ab of the IgG1, IgG2a, or IgG3 isotypes or serum IgG2b levels were unchanged. It should be noted that the two CD28−/− MRL/lpr mice that produced relatively high levels of IgG3 anti-DNA Ab nevertheless showed the same low serum IgG3 concentrations as other CD28−/− littermates. These results imply that anti-DNA Ab production may be selectively induced in some cases, independently of polyclonal IgG3 production. Alternatively, anti-DNA Ab production in some MRL/lpr mice may be less dependent on CD28 function compared with the production of other IgG Abs. Although some CD28−/− MRL/lpr mice showed comparable serum IgG2b levels, IgG2b anti-DNA Ab was not elevated in these mice. These data suggest that dysregulated expansion of Ab-producing cells can occur in the absence of CD28 costimulation, and it might differ in types of Abs or IgG subtypes.

Glomerulonephritis and renal insufficiency are important clinical features of the disease affecting MRL/lpr mice. Glomerular lesions such as mesangial proliferation and crescent formation were consistently observed in the CD28+/+ MRL/lpr mice examined in this study. In contrast, most of the CD28−/− MRL/lpr mice showed much milder glomerular lesions and lower levels of glomerular IgG deposits. This pathological profile correlates with the low levels of anti-DNA IgG Abs in CD28−/− MRL/lpr mice and is consistent with previous studies showing the critical role of autoantibodies and B cells in the development of glomerulonephritis (38, 39, 40, 41). It has been previously shown that IgG2a, IgG2b, and IgG3 activate the complement system (42), and in addition, IgG3 cryoglobulins are important nephritogenic factors (43, 44, 45, 46). In our study, two CD28−/− MRL/lpr mice that showed high IgG3 anti-DNA Ab levels exhibited relatively severe glomerular pathology among CD28−/− MRL/lpr mice. Although global comparison between anti-DNA levels and renal pathology was not available because of the limited numbers of mice, these results suggest a correlation between high IgG3 anti-DNA Ab levels and advanced glomerular pathology in CD28−/− MRL/lpr mice. CD28−/− MRL/lpr mice of advanced age still showed significant glomerular lesions, and some mice displayed global sclerosis and atrophy of glomeruli. Thus, although the development of glomerulonephritis is significantly retarded in the absence of CD28, it nevertheless slowly proceeds, a finding corroborating previous reports showing that glomerulonephritis was able to develop in the absence of TCRαβ+ (21) or CD4+ T cells (47). On the other hand, vasculitis and interstitial infiltration were completely abolished in CD28−/− MRL/lpr mice. There was no cellular infiltration in the liver, lung, and joints of not only young but also aged CD28−/− MRL/lpr mice, indicating that CD28 is critical for the development of cellular infiltration, vasculitis, and arthritis in MRL/lpr mice.

The dissociation of renal vasculitis and glomerulonephritis, i.e., diminished vasculitis but unmitigated glomerulonephritis, has also been observed in TCRαβ+ T cell-deficient (21) and NO synthase type 2-deficient MRL/lpr mice (48). Glomerulonephritis was mild in the former but severe in the latter. Taken together, these results indicate that different pathogeneses underlie vasculitis and glomerulonephritis. It is possible that the development of vasculitis may be more dependent than that of glomerulonephritis on the activation of T cells. Histological studies have shown that the earliest sign of vascular disease in MRL/lpr mice is perivascular “cuffing” of T cells, followed by an adventitial infiltration of reactive inflammatory cells (49). In contrast, glomerulonephritis is histologically characterized by the deposition of immune complexes (3). It has also been shown that glomerulonephritis, but not skin vasculitis, can develop following administration of IgG3 rheumatoid factor (43). In addition, Nose et al. (50, 51) used a genetic approach to show that vasculitis, glomerulonephritis, and arthritis could be dissociated in congenic strains of MRL/lpr mice. Finally, in CD28-deficient MRL/lpr and TCRαβ+ T cell-deficient MRL/lpr mice, variable levels of anti-DNA Abs are produced, and significant but decreased Ig deposits are observed in glomeruli (Fig. 6⇑ and Ref. 21). We conclude that although CD28 is critically required for the development of vasculitis and arthritis, it is important, but not absolutely essential, for the development of glomerulonephritis.

Differential involvement of CD28 has been observed in other mouse disease models. It has been shown that nonobese diabetes (NOD) mice experience accelerated diabetes of increased severity in the absence of CD28 (52). In NOD mice, disease development occurs in several distinct stages, and the loss of CD28 in the early phase is thought to preferentially block Th2 differentiation, leading to a Th1/Th2 imbalance, which results in accelerated disease (52). In contrast, blocking of CD28 signaling at 5–7 wk of age leads to inhibition of the disease (27). CD28 thus plays different roles in the initiation and progression phases in NOD mice. Furthermore, T cell activation and expansion proceed unchecked in CD28-deficient NOD mice, suggesting that a loss of CD28 costimulation can be overcome or substituted for in pathogenic T cells. It is not known whether the disease affecting MRL/lpr mice develops in a multistage fashion. Autoimmunity develops spontaneously in these mice, perhaps resulting in the continuous exposure of T cells to autoantigens from birth. Because the congenital loss of CD28 suppressed disease progression in MRL/lpr mice of all ages, the activation threshold beyond which pathogenic T cells induce autoimmune pathology in MRL/lpr mice may be higher than is the case in NOD mice. Alternatively, more efficient substitution of CD28 signaling may occur in MLR/lpr mice than in NOD mice.

In summary, we have found that CD28 costimulation is essential for several features of the autoimmune disease manifested in MRL/lpr mice. In the absence of CD28, vasculitis, arthritis, and cellular infiltration into organs are abolished, glomerulonephritis is significantly retarded, and the production of autoantibodies is suppressed, but the accumulation of abnormal B220+ T cells is almost unchanged. These results provide convincing evidence that CD28 is a key regulator of many aspects of autoimmune disease induction.

Acknowledgments

We thank Anna Tafuri-Bladt for critical reading of the manuscript, M. Fujisaki for technical assistance and animal care, Y. Tsugitomi for technical help with histology and immunofluorescence, and M. Saunders for scientific editing of the manuscript.

Footnotes

  • ↵1 This study was supported in part by the Medical Research Council of Canada.

  • ↵2 Address correspondence and reprint requests to Dr. Tak W. Mak, Amgen Institute, 620 University Avenue, Suite 706, Toronto, Ontario, Canada M5G 2C1. E-mail address: tmak{at}oci.utoronto.ca

  • ↵3 Abbreviations used in this paper: MLR/lpr, MRL/Mpj-lpr/lpr; LN, lymph node.

  • Received December 1, 1998.
  • Accepted July 7, 1999.

References

  1. ↵
    Cohen, P. L., R. A. Eisenberg. 1991. lpr and gld: single gene models of systemic autoimmunity and lymphoproliferative disease. Annu. Rev. Immunol. 9: 243
    OpenUrlCrossRefPubMed
  2. ↵
    Theofilopuolos, A. N., F. J. Dixon. 1985. Murine models of systemic lupus erythematosus. Adv. Immunol. 37: 269
    OpenUrlCrossRefPubMed
  3. ↵
    Andrews, B. S., R. A. Eisenberg, A. N. Theofilopoulos, S. Izui, C. B. Wilson, P. J. McConahey, E. D. Murphy, J. B. Roths, F. J. Dixon. 1978. Spontaneous murine lupus-like syndromes: clinical and immunopathological manifestations in several strains. J. Exp. Med. 148: 1198
    OpenUrlAbstract/FREE Full Text
  4. ↵
    Watanabe-Fukunaga, R., C. I. Brannan, N. G. Copeland, N. A. Jenkins, S. Nagata. 1992. Lymphoproliferation disorder in mice explained by defects in Fas antigen that mediates apoptosis. Nature 356: 314
    OpenUrlCrossRefPubMed
  5. ↵
    Watson, M. L., J. K. Rao, G. S. Gilkeson, P. Ruiz, E. M. Eicher, D. S. Pisetsky, A. Matsuzawa, J. M. Rochelle, M. F. Seldin. 1992. Genetic analysis of MRL-lpr mice: relationship of the Fas apoptosis gene to disease manifestation and renal disease-modifying loci. J. Exp. Med. 176: 1645
    OpenUrlAbstract/FREE Full Text
  6. ↵
    Russell, J. H., B. Rush, C. Weaver, R. Wang. 1993. Mature T cells of autoimmune lpr/lpr mice have a defect in antigen-stimulated suicide. Proc. Natl. Acad. Sci. USA 90: 4409
    OpenUrlAbstract/FREE Full Text
  7. ↵
    Bossu, P., G. G. Singer, P. Andres, R. Ettinger, A. Marshak-Rothstein, A. K. Abbas. 1993. Mature CD4+ T lymphocytes from MRL/lpr mice are resistant to receptor-mediated tolerance and apoptosis. J. Immunol. 151: 7233
    OpenUrlAbstract
  8. ↵
    Snatoro, T. J., J. P. Portanova, B. L. Kotzin. 1988. The contribution of L3T4+ T cells to lymphoproliferation and autoantibody production in MRL-lpr/lpr mice. J. Exp. Med. 167: 1713
    OpenUrlAbstract/FREE Full Text
  9. ↵
    O’Sullivan, F. X., C. J. Ray, Y. Takeda, G. C. Sharp, S. E. Walker. 1991. Long-term anti-CD4 treatment of MRL/lpr mice ameliorates immunopathology and lymphoproliferation but fails to suppress rheumatoid factor production. Clin. Immunol. Immunopathol. 61: 421
    OpenUrlCrossRefPubMed
  10. ↵
    Jabs, D. A., C. L. Burek, Q. Hu, R. C. Kuppers, B. Lee, R. A. Prendergast. 1992. Anti-CD4 monoclonal antibody therapy suppresses autoimmune diseases in MRL/Mp-lpr/lpr mice. Cell. Immunol. 141: 496
    OpenUrlCrossRefPubMed
  11. ↵
    Merino, R., M. Iwamoto, L. Fossati, S. Izui. 1993. Polyclonal B cell activation arises from different mechanisms in lupus-prone (NZB × NZW)F1 and MRL/MpJ-lpr/lpr mice. J. Immunol. 151: 6509
    OpenUrlAbstract
  12. ↵
    Henrickson, M., E. H. Giannini, R. Hirsch. 1994. Reduction of mortality and lymphoadenopathy in MRL-lpr/lpr mice treated with nonmitogenic anti-CD3 monoclonal antibody. Arthritis Rheum. 37: 587
    OpenUrlCrossRefPubMed
  13. ↵
    Seaman, W. E., D. Wofsy, J. S. Greenspan, J. A. Ledbetter. 1983. Treatment of autoimmune MRL/lpr mice with monoclonal antibody to Thy-1.2: a single injection has sustained effects on lymphoproliferation and renal disease. J. Immunol. 130: 1713
    OpenUrlAbstract
  14. ↵
    Mountz, J. D., H. R. Smith, R. L. Wilder, J. P. Reeves, A. D. Steinberg. 1987. CS-A therapy in MRL-lpr/lpr mice: amelioration of immunopathology despite autoantibody production. J. Immunol. 138: 157
    OpenUrlAbstract
  15. ↵
    Koh, D.-R., A. Ho, A. Rahemtulla, W.-P. Fung-Leung, H. Griesser, T. W. Mak. 1995. Murine lupus in MRL/lpr mice lacking CD4 or CD8 T cells. Eur. J. Immunol. 25: 2558
    OpenUrlCrossRefPubMed
  16. ↵
    Jevnikar, A. M., M. J. Grusby, L. H. Glimcher. 1994. Prevention of nephritis in major histocompatibility complex class II-deficient MRL-lpr mice. J. Exp. Med. 179: 1137
    OpenUrlAbstract/FREE Full Text
  17. ↵
    Ohteki, T., M. Iwamoto, S. Izui, H. R. MacDonald. 1995. Reduced development of CD4−CD8−B220+ T cells but normal autoantibody production in lpr/lpr mice lacking major histocompatibility complex class I molecule. Eur. J. Immunol. 25: 37
    OpenUrlCrossRefPubMed
  18. ↵
    Mixter, P. F., J. Q. Russell, F. H. Durie, R. C. Budd. 1995. Decreased CD4−CD8− TCRαβ+ cells in lpr/lpr mice lacking β2-microglobulin. J. Immunol. 154: 2063
    OpenUrlAbstract
  19. ↵
    Maldonado, M. A., R. A. Eisenberg, E. Roper, P. L. Cohen, B. L. Kotzin. 1995. Greatly reduced lymphoproliferation in lpr mice lacking major histocompatibility complex class I. J. Exp. Med. 181: 641
    OpenUrlAbstract/FREE Full Text
  20. ↵
    Christianson, G. J., R. L. Blankenburg, T. M. Duffy, D. Panka, J. B. Roths, A. Marshak-Rothstein, D. C. Rooopenian. 1996. β2-Microglobulin dependence of the lupus-like autoimmune syndrome of MRL-lpr mice. J. Immunol. 156: 4932
    OpenUrlAbstract
  21. ↵
    Peng, S. L., M. P. Madaio, D. P. M. Hughes, I. N. Crispe, M. J. Owen, L. Wen, A. C. Hayday, J. Craft. 1996. Murine lupus in the absence of αβ T cells. J. Immunol. 156: 4041
    OpenUrlAbstract
  22. ↵
    Bretscher, P.. 1992. The two-signal model of lymphocyte activation twenty-one years later. Immunol. Today 13: 74
    OpenUrlCrossRefPubMed
  23. ↵
    Linsey, P. S., J. A. Ledbetter. 1993. The role of the CD28 receptor during T-cell responses to antigen. Annu. Rev. Immunol. 11: 191
    OpenUrlCrossRefPubMed
  24. ↵
    June, C. H., J. A. Bluestone, L. M. Nadler, C. B. Thompson. 1994. The B7 and CD28 receptor families. Immunol. Today 15: 321
    OpenUrlCrossRefPubMed
  25. ↵
    Harding, F. A., J. G. McArthur, J. A. Gross, D. H. Raulet, J. P. Allison. 1992. CD28-mediated signalling co-stimulates murine T cells and prevents induction of anergy in T-cell clones. Nature 356: 607
    OpenUrlCrossRefPubMed
  26. ↵
    Finck, B. K., P. S. Linsley, D. Wolfy. 1994. Treatment of murine lupus with CTLA4Ig. Science 265: 1225
    OpenUrlAbstract/FREE Full Text
  27. ↵
    Lenschow, D. J., S. C. Ho, H. Sattar, L. Rhee, G. Gray, N. Nabavi, K. C. Herold, J. A. Bluestone. 1995. Differential effects of anti-B7-1 and anti-B7-2 monoclonal antibody treatment on the development of diabetes in the nonobese diabetic mouse. J. Exp. Med. 181: 1145
    OpenUrlAbstract/FREE Full Text
  28. ↵
    Cross, A. H., T. J. Girard, K. S. Giacoletto, R. J. Evans, R. M. Keeling, R. F. Lin, J. L. Trotter, R. W. Karr. 1995. Long-term inhibition of murine experimental autoimmune encephalomyelitis using CTLA-4-Fc supports a key role for CD28 costimulation. J. Clin. Invest. 95: 2783
  29. ↵
    Arima, T., A. Rehman, W. F. Hickey, M. W. Flye. 1996. Inhibition by CTLA4Ig of experimental allergic encephalomyelitis. J. Immunol. 156: 4916
    OpenUrlAbstract
  30. ↵
    Knoerzer, D. B., R. W. Karr, B. D. Schwartz, L. J. Mengle-Gaw. 1995. Collagen-induced arthritis in the BB rat. Prevention of diseases by treatment with CTLA-4-Ig. J. Clin. Invest. 96: 987
  31. ↵
    Webb, L. M. C., M. J. Walmsley, M. Feldmann. 1996. Prevention and amelioration of collagen-induced arthritis by blockade of the CD28 co-stimulatory pathway: requirement for both B7-1 and B7-2. Eur. J. Immunol. 26: 2320
    OpenUrlCrossRefPubMed
  32. ↵
    Shahinian, A., K. Pfeffer, K. P. Lee, T. M. Kündig, K. Kishihara, A. Wakeham, K. Kawai, P. S. Ohashi, C. B. Thompson, T. W. Mak. 1993. Differential T cell costimulatory requirements in CD28-deficient mice. Science 261: 609
    OpenUrlAbstract/FREE Full Text
  33. ↵
    Mountz, J. D., W. C. Gause, F. D. Finkelman, A. D. Steinberg. 1988. Prevention of lymphadenopathy in MRL-lpr/lpr mice by blocking peripheral lymph node homing with Mel-14 in vivo. J. Immunol. 140: 2943
    OpenUrlAbstract
  34. ↵
    Ishigatsubo, Y., A. D. Steinberg, D. M. Klinman. 1988. Autoantibody production is associated with polyclonal B cell activation in autoimmune mice which express the lpr or gld genes. Eur. J. Immunol. 18: 1089
    OpenUrlCrossRefPubMed
  35. ↵
    Perkins, D. L., R. M. Glaser, C. A. Mahon, J. Michaelson, A. Marshak-Rothstein. 1990. Evidence for an intrinsic B cell defect in lpr/lpr mice apparent in neonatal chimeras. J. Immunol. 145: 549
    OpenUrlAbstract
  36. ↵
    Sobel, E. S., T. Katagiri, K. Katagiri, S. C. Morris, P. L. Cohen, R. A. Eisenberg. 1991. An intrinsic B cell defect is required for the production of autoantibodies in the lpr model of murine systemic autoimmunity. J. Exp. Med. 173: 1441
    OpenUrlAbstract/FREE Full Text
  37. ↵
    Linsley, P. S., P. M. Wallace, J. Johnson, M. G. Gibson, J. L. Greene, J. A. Ledbetter, C. Singh, M. A. Tepper. 1992. Immunosuppression in vivo by a soluble form of the CTLA-4 T cell activation molecule. Science 257: 789
    OpenUrlAbstract/FREE Full Text
  38. ↵
    Dang, H., R. J. Harbeck. 1984. The in vivo and in vitro glomerular deposition of isolated anti-double stranded-DNA antibodies in NZB/W mice. Clin. Immunol. Immunopathol. 30: 265
    OpenUrlCrossRefPubMed
  39. ↵
    Madaio, M. P., J. Carlson, J. Cataldo, A. Ucci, P. Migliorini, O. Pankewycz. 1987. Murine monoclonal anti-DNA antibodies bind directly to glomerular antigens and form immune deposits. J. Immunol. 138: 2883
    OpenUrlAbstract
  40. ↵
    Vlahakos, D. V., M. H. Foster, S. Adams, M. Katz, A. A. Ucci, K. J. Barrett, S. K. Datta, M. P. Madaio. 1992. Anti-DNA antibodies from immune deposits at distinct glomerular and vascular sites. Kidney Int. 41: 1690
    OpenUrlCrossRefPubMed
  41. ↵
    Shlomchik, M. J., M. P. Madaio, D. Ni, M. Trounstein, D. Huszar. 1994. The role of B cells in lpr/lpr-induced autoimmunity. J. Exp. Med. 180: 1295
    OpenUrlAbstract/FREE Full Text
  42. ↵
    Germann, T., M. Bongartz, H. Dlugonska, H. Hess, E. Schmitt, L. Kolbe, E. Kolsch, F. J. Podlaski, M. K. Gately, E. Rüde. 1995. Interleukin-12 profoundly up-regulates the synthesis of antigen-specific complement-fixing IgG2a, IgG2b and IgG3 antibody subclasses in vivo. Eur. J. Immunol. 25: 823
    OpenUrlCrossRefPubMed
  43. ↵
    Reininger, L., T. Berney, T. Shibata, F. Spertini, R. Merino, S. Izui. 1990. Cryoglobulinemia induced by a murine IgG3 rheumatoid factor: skin vasculitis and glomerulonephritis arise from distinct pathogenic mechanisms. Proc. Natl. Acad. Sci. USA 87: 10038
    OpenUrlAbstract/FREE Full Text
  44. ↵
    Takahashi, S., M. Nose, J. Sasaki, T. Yamamoto, M. Kyogoku. 1991. IgG3 production in MRL/lpr mice is responsible for development of lupus nephritis. J. Immunol. 147: 515
    OpenUrlAbstract
  45. ↵
    Berney, T., T. Fulpius, T. Shibata, J. L Reininger, H. Van-Snick, M. Shan, A. Marshak Weigert, -R othstein, S. Izui. 1992. Selective pathogenicity of murine rheumatoid factors of the cryoprecipitable IgG3 subclass. Int. Immunol. 4: 93
    OpenUrlAbstract/FREE Full Text
  46. ↵
    Lemoine, R., T. Berney, T. Shibata, T. Fulpius, Y. Kyogoku, H. Shimada, S. Sawada, S. Izui. 1992. Induction of “wire-loop” lesions by murine monoclonal IgG3 cryoglobulins. Kidney Int. 41: 65
    OpenUrlCrossRefPubMed
  47. ↵
    Chesnutt, M. S., B. K. Finck, N. Killeen, M. K. Connolly, H. Goodman, D. Wofsy. 1998. Enhanced lymphoproliferation and diminished autoimmunity in CD4-deficient MRL/lpr mice. Clin. Immunol. Immunopathol. 87: 23
    OpenUrlCrossRefPubMed
  48. ↵
    Gilkeson, G. S., J. S. Mudgett, M. F. Seldin, P. Ruiz, A. A. Alexander, M. A. Misukonis, D. S. Pisetsky, B. Weinberg. 1997. Clinical and serologic manifestations of autoimmune disease in MRL-lpr/lpr mice lacking nitric oxide synthase type 2. J. Exp. Med. 186: 365
    OpenUrlAbstract/FREE Full Text
  49. ↵
    Moyer, C. F., J. D. Strandberg, C. L. Reinisch. 1987. Systemic mononuclear-cell vasculitis in MRL/Mp-lpr/lpr mice: a histologic and immunocytochemical analysis. Am. J. Pathol. 127: 229
    OpenUrlPubMed
  50. ↵
    Nose, M., M. Nishimura, M. Kyogoku. 1989. Analysis of granulomatous arteritis in MRL/Mp autoimmune disease mice bearing lymphoproliferative genes: the use of mouse genetics to dissociate the development of arteritis and glomerulonephritis. Am. J. Pathol. 135: 271
    OpenUrlPubMed
  51. ↵
    Nose, M., M. Nishimura, M. R. Ito, J. Toh, T. Shibata, T. Sugisaki. 1996. Arteritis in a novel congenic strain of mice derived from MRL/lpr lupus mice: genetic dissociation from glomerulonephritis and limited autoantibody production. Am. J. Pathol. 149: 1763
    OpenUrlPubMed
  52. ↵
    Lenschow, D. J., K. C. Herold, L. Rhee, B. Patel, A. Koons, H.-Y. Qin, E. Fuchs, B. Singh, C. B. Thompson, J. A. Bluestone. 1996. CD28/B7 regulation of Th1 and Th2 subsets in the development of autoimmune diabetes. Immunity 5: 285
    OpenUrlCrossRefPubMed
Back to top

In this issue

The Journal of Immunology: 163 (6)
The Journal of Immunology
Vol. 163, Issue 6
15 Sep 1999
Print
Download PDF
Article Alerts
Email Article
Citation Tools

Jump to section