Although systemic lupus erythematosus appears to be a humorally mediated disease, both Th1 and Th2 type responses have been implicated in its pathogenesis. The Th1 response, as exemplified by IFN-γ production, has been uniformly shown in mouse lupus models to be critical for disease induction. The role of Th2 type responses, however, is more complicated, with some studies showing detrimental and others beneficial effects of IL-4 in these models. To further address this issue, we generated and analyzed IL-4 gene-deficient BXSB mice. Mice homozygous for this deletion had significantly lower serum levels of total IgG1 compared with wild-type BXSB, consistent with the lack of IL-4. However, no significant differences were observed in mortality, spleen weight, severity of glomerulonephritis, levels of anti-chromatin and anti-ssDNA Abs, or frequency of activated (CD44high) CD4+ T cells. The anti-chromatin Ab isotype response was virtually all Th1 type in both the knockout and wild-type BXSB. These findings directly demonstrate that IL-4 and, by inference, Th2 cells are not obligatory participants in the induction and maintenance of lupus in this strain.

F1 hybrids among New Zealand Black (NZB), New Zealand White (NZW), and BXSB lupus-prone strains develop accelerated autoimmunity in both sexes regardless of the specific combination. To identify BXSB susceptibility loci in the absence of the Y chromosome accelerator of autoimmunity (Yaa) and to study the genetics of this complementation, genome-wide quantitative trait locus (QTL) mapping was performed on female (BXSB × NZW)F2 mice. Six QTL were identified on chromosomes 1, 4, 5, 6, 7, and 17. Survival mapped to chromosomes 5 and 17, anti-chromatin Ab to chromosomes 4 and 17, glomerulonephritis to chromosomes 6 and 17, and splenomegaly to chromosomes 1, 7, and 17. QTL on chromosomes 4 and 6 were new and designated as Lxw1 and -2, respectively. Two non-MHC QTL (chromosomes 1 and 4) were inherited from the BXSB and the rest were NZW-derived, including two similar to previously defined loci. Only two of 11 previously defined non-MHC BXSB QTL using male (Yaa+) crosses were implicated, suggesting that some male-defined BXSB QTL may require coexpression of the Yaa. Findings from this and other studies indicate that BXSB and NZB backgrounds contribute completely different sets of genes to complement NZW mice. Identification of susceptibility genes and complementing genes in several lupus-prone strain combinations will be important for defining the epistatic effects and background influences on the heterogeneous genetic factors responsible for lupus induction.

Imbalances of Th1- and Th2-type responses have been postulated to be a predisposing factor for both humoral and cellular mediated autoimmune diseases. To further define their roles in systemic autoimmunity, IL-4 and IFN-γ gene knockout mice were studied for susceptibility to the prototypic Th2-mediated mercury-induced autoimmunity. A predominant Th2-type response following HgCl2 treatment of wild-type B10.S mice was confirmed by the findings of a significant increase in splenic IL-4 and hypergammaglobulinemia primarily of the IgG1 isotype, without an increase in IFN-γ levels. Paradoxically, IL-4-deficient mice developed the characteristic anti-nucleolar autoantibodies and tissue deposition of immune complexes, while IFN-γ-deficient mice had very low autoantibody levels and essentially normal immunohistology. Studies to define defects in Ab responses of IFN-γ-deficient mice, using the T-dependent Ag (4-hydroxy-3-nitrophenyl)acetyl, revealed an attenuated IgG response to low and to a lesser extent high doses of (4-hydroxy-3-nitrophenyl)acetyl-hemocyanin, but maintenance of affinity maturation. These results indicate that Th1/Th2 imbalance does not directly play a role in susceptibility to mercury-induced autoimmunity, and suggest that the dependence on Th1-type responses in certain autoimmune diseases is due to the requirement for IFN-γ for Ab production to weakly antigenic self molecules.