Differential Role of Three Major New Zealand Black-Derived Loci Linked with Yaa-Induced Murine Lupus Nephritis
http://www.100md.com
免疫学杂志 2005年第2期
Abstract
By assessing the development of Y-linked autoimmune acceleration (Yaa) gene-induced systemic lupus erythematosus in C57BL/6 (B6) x (New Zealand Black (NZB) x B6.Yaa)F1 backcross male mice, we mapped three major susceptibility loci derived from the NZB strain. These three quantitative trait loci (QTL) on NZB chromosomes 1, 7, and 13 differentially regulated three different autoimmune traits: anti-nuclear autoantibody production, gp70-anti-gp70 immune complex (gp70 IC) formation, and glomerulonephritis. Contributions to the disease traits were further confirmed by generating and analyzing three different B6.Yaa congenic mice, each carrying one individual NZB QTL. The chromosome 1 locus that overlapped with the previously identified Nba2 (NZB autoimmunity 2) locus regulated all three traits. A newly identified chromosome 7 locus, designated Nba5, selectively promoted anti-gp70 autoantibody production, hence the formation of gp70 IC and glomerulonephritis. B6.Yaa mice bearing the NZB chromosome 13 locus displayed increased serum gp70 production, but not gp70 IC formation and glomerulonephritis. This locus, called Sgp3 (serum gp70 production 3), selectively regulated the production of serum gp70, thereby contributing to the formation of nephritogenic gp70 IC and glomerulonephritis, in combination with Nba2 and Nba5 in NZB mice. Among these three loci, a major role of Nba2 was demonstrated, because B6.Yaa Nba2 congenic male mice developed the most severe disease. Finally, our analysis revealed the presence in B6 mice of an H2-linked QTL, which regulated autoantibody production. This locus had no apparent individual effect, but most likely modulated disease severity through interaction with NZB-derived susceptibility loci.
Introduction
The New Zealand Black (NZB)3 x New Zealand White (NZW)F1 hybrid mice spontaneously develop a generalized autoimmune disorder resembling human systemic lupus erythematosus (SLE). The major known targets of the autoimmune responses in this model of SLE are chromatin (including its major constituent, DNA) and serum gp70, the major envelope glycoprotein of endogenous retrovirus (1, 2, 3, 4). Anti-chromatin, anti-DNA, and anti-gp70 autoantibodies have been implicated in the development of severe immune complex (IC)-mediated glomerulonephritis (GN), not only in these mice, but also in lupus-prone MRL-Faslpr and BXSB mice (5, 6, 7).
It is now well established that SLE is a polygenic disease, in which multiple, unlinked genes are operative in a threshold manner. Linkage analysis using microsatellite markers in crosses involving NZB or NZW mice have revealed a number of lupus-susceptibility loci (8, 9). Although F1 hybrids of NZW and C57BL/6 (B6) mice are essentially normal, these mice develop severe lupus-like nephritis closely associated with elevated serum levels of IgG anti-DNA autoantibodies and gp70-anti-gp70 IC (gp70 IC) when they carry the Yaa (Y-linked autoimmune acceleration) mutation, derived from male BXSB mice (10). Linkage analysis of B6 x (NZW x B6.Yaa)F1 backcross (BC) male mice provided evidence for a major quantitative trait locus (QTL) on NZW chromosome 7 controlling both the severity of GN and the production of IgG anti-DNA and gp70 IC (11). The results indicated that the genetic analysis involving Yaa represents a useful tool for dissecting the complex genetic interactions responsible for the development of murine SLE.
In the present study, we used B6 x (NZB x B6.Yaa)F1 BC male mice and congenic mice bearing mapped susceptibility intervals to identify critical NZB-derived lupus-susceptibility loci implicated in murine SLE. In this study, we report the mapping of three major QTL from the NZB strain: the first on chromosome 1 overlapping with the previously identified Nba2 (New Zealand black autoimmunity 2) locus (12); the second on chromosome 7, designated Nba5 (New Zealand black autoimmunity 5); and the third on chromosome 13 corresponding to the Sgp3 (serum gp70 production 3) locus of NZW mice (13). Moreover, we identified a QTL on chromosome 17 from the B6 strain contributing to autoantibody production.
Materials and Methods
Mice
NZB mice (H2d) were purchased from The Jackson Laboratory. B6 mice (H2b) bearing the Yaa mutation (B6.Yaa) were established by repeated backcrossing (n > 20), as described previously (14). (NZB x B6.Yaa)F1 and B6 x (NZB x B6.Yaa)F1 BC mice were obtained by local breeding in our animal facility. B6.NZB-Nba2 (B6.Nba2) and B6.NZW-Sgp3 congenic mice were generated, as described previously (13, 15). B6.NZB-Nba5 (B6.Nba5) and B6.NZB-Sgp3 congenic mice were generated by backcrossing the NZB-derived Nba5 or Sgp3 intervals onto the B6 background using marker-assisted selection, as described previously (13). After five or six generations of backcrossing, siblings were intercrossed to generate B6.Nba5 and B6.NZB-Sgp3 congenic mice homozygous for the respective NZB chromosome 7 and chromosome 13 intervals. Males of all congenic mice used in the present study carry the Yaa mutation. Blood samples were collected by orbital sinus puncture, and sera were stored at –20°C until use.
Serological assays
Serum levels of IgG autoantibodies against chromatin and heat-denatured DNA were determined by ELISA, as described previously (15, 16). Results are expressed in U/ml in reference to a standard curve derived from a serum pool of MRL-Faslpr mice. Concentrations of total gp70 in sera were determined by ELISA, as described previously (17). Serum levels of gp70 IC were quantified by the same ELISA combined with the treatment of sera with 10% polyethylene glycol (average m.w. 6000), which precipitates only Ab-bound gp70, but not free gp70, as described (17). Results are expressed as micrograms per milliliter of gp70 by referring to a standard curve obtained from a serum pool of NZB mice with known amounts of gp70.
Genotyping and statistical analysis
Genotypes were determined by PCR using 95 selected microsatellite markers either purchased from Research Genetics or Invitrogen Life Technologies. DNA from NZB, B6, (NZB x B6)F1, and BC mice were extracted from tail biopsies kept at –70°C before use. PCR amplification was conducted with RED TaqDNA polymerase (Sigma-Aldrich) using a GeneAmp PCR system 9700 thermal cycler (Applied Biosystems), as described (13). The positions of the microsatellite markers with respect to the centromere were obtained from the Mouse Genome Database at www.informatics.jax.org. The linkage program MAPMAKER/QTL was used to identify QTL (18). Autoantibody and gp70 levels were log10 transformed. The association of severe GN (grade 3) with marker loci was tested by a 2 test for goodness-of-fit against an expected 50:50 distribution using a standard (2 x 2) contingency matrix. A threshold for suggestive linkage was set at log-likelihood of the odds (LOD) > 1.9, p < 0.0034 (2 > 8.6, 1 degree of freedom), and for significant linkage was
LOD >3.3, p < 0.0001 (2 > 10.8, 1 degree of freedom), based on the recommendation of Lander and Kruglyak (19). Comparison of serum levels of anti-nuclear autoantibodies, gp70 IC, and gp70 between different groups of mice was performed with the Wilcoxon two-sample test. Probability values >5% were considered insignificant.
Histopathology
Kidney samples were collected when mice were moribund or at the end of the experiment at 15–18 mo of age. Histological sections were stained with periodic acid-Schiff reagent. The extent of GN was graded on a 0–4 scale based on the intensity and extent of histopathological changes, as described previously (14). GN with grade 3 or 4 was considered a significant contributor to clinical disease and/or death.
Results
Development of lupus-like autoimmune syndrome in (NZB x B6.Yaa)F1 male mice
The Yaa gene has been shown to be responsible for the accelerated development of lupus-like autoimmune syndrome in BXSB mice and in their F1 hybrids with NZW or NZB mice (10). In this study, we first analyzed whether (NZB x B6.Yaa)F1 males bearing the Yaa gene also spontaneously developed a lupus-like autoimmune syndrome. At 8 mo of age, sera from (NZB x B6.Yaa)F1 Yaa males displayed markedly elevated IgG anti-DNA and anti-chromatin autoantibodies, as well as gp70 IC, compared with B6.Yaa mice (p < 0.001; Fig. 1). Consequently, F1 male mice developed a lethal form of severe GN; 50% of them (n = 18) died of GN (grade 3) by 14 mo of age, and only 15% were alive at 18 mo of age. Their renal lesions were characterized by increased glomerular cellularity, obliteration of the glomerular architecture, and the presence of tubular cast formation. In contrast, B6.Yaa males showed only slight mesangial alterations at 15 mo of age (mean histological grades of 14 mice: 0.4 ± 0.5).
FIGURE 1. Serum levels of IgG anti-DNA and anti-chromatin autoantibodies, gp70 IC, and gp70 in B6.Yaa, (NZB x B6.Yaa)F1, and B6 x (NZB x B6.Yaa)F1 BC male mice. Each symbol represents individual animal of B6.Yaa (n = 10), (NZB x B6.Yaa)F1 (n = 20), and 154 BC Yaa males (n = 154). Serum levels of anti-DNA, anti-chromatin, and gp70 IC were determined at 8 mo of age, and serum gp70 concentrations were measured at 4 mo of age. Results are expressed as U/ml for anti-DNA and anti-chromatin autoantibodies, and as μg/ml for gp70 IC and gp70. The mean values are indicated by the horizontal line.
Linkage of marker loci to IgG anti-DNA and anti-chromatin autoantibodies in B6 x (NZB x B6.Yaa)F1 BC mice
To determine the NZB genetic contribution to the Yaa-induced acceleration of autoimmune disease in (NZB x B6.Yaa)F1 male mice, a total of 154 B6 x (NZB x B6.Yaa)F1 male BC mice was produced, and sera from 6-, 8-, and 10-mo-old BC mice were analyzed for IgG anti-DNA and anti-chromatin autoantibodies. Analysis of these two autoantibodies with MAPMARKER/QTL revealed significant linkage with an interval on NZB chromosome 1 directly overlapped with the Nba2 locus at 90–98 cM from the centromere (Fig. 2). The locus Fcgr2 (92.3 cM from the centromere) gave a LOD score of 7.08 (p = 1.13 x 10–8) for anti-DNA and 13.24 (p = 5.73 x 10 –15) for anti-chromatin autoantibodies. In addition to Nba2, a high LOD score for IgG anti-DNA (10.55, p = 3.14 x 10–12) and IgG anti-chromatin (3.02, p = 1.94 x 10–4) was observed at the H2 locus on chromosome 17 (Fig. 2). However, this susceptibility allele was inherited from the B6 strain.
FIGURE 2. Linkage of chromosome 1 and 17 markers with serum levels of IgG anti-DNA and anti-chromatin autoantibodies in B6 x (NZB x B6.Yaa)F1 BC male mice bearing the Yaa mutation. Autoantibody titers were determined at 6, 8, and 10 mo of age. LOD scores for each trait were generated with MAPMAKER/QTL, and the highest scores obtained at 8 mo of age are shown. The horizontal dotted line represents the threshold for suggestive linkage.
Linkage of marker loci to gp70 and gp70 IC in B6 x (NZB x B6.Yaa)F1 BC mice
Serum levels of gp70 are highly variable among different strains of mice, and NZB mice produce larger amounts of serum gp70 than B6 mice (2). Consistently, serum gp70 concentrations in 4-mo-old (NZB x B6.Yaa)F1 male mice (23.1 ± 3.9 μg/ml) were 10-fold higher than those in B6.Yaa male mice (2.3 ± 0.7 μg/ml; p < 0.001) (Fig. 1). QTL analysis was conducted to investigate the genetics of serum gp70 production in B6 x (NZB x B6.Yaa)F1 BC male mice at 4 mo of age. A single QTL of NZB origin was identified on midchromosome 13 and peaked close to the D13Mit193 marker (43 cM from the centromere), with a LOD score of 8.77 (p = 2.11 x 10–10; Fig. 3). The Sgp3 locus has been previously mapped to this position in B6 x (NZW x B6.Yaa)F1 and B10 x (B10 x BXSB)F1 BC mice (7, 11). The current results show that the Sgp3 locus is also the major determinant of quantitative variation of gp70 in NZB mice.
FIGURE 3. Linkage of chromosome 1, 7, 13, and 17 markers with serum levels of gp70 and gp70 IC in B6 x (NZB x B6.Yaa)F1 BC male mice bearing the Yaa mutation. Serum concentrations of gp70 were determined at 4 mo of age, and gp70 IC levels were measured at 6, 8, and 10 mo of age. LOD scores for each trait were generated with MAPMAKER/QTL, and the highest scores for gp70 IC obtained at 8 mo of age for chromosomes 1 and 13 and at 6 mo of age for chromosomes 7 and 17 are shown. The horizontal dotted line represents the threshold for suggestive linkage.
When the genotype data of BC mice were analyzed for loci linked with levels of gp70 IC, significant linkage with four regions was revealed (Fig. 3). As in the case of anti-nuclear autoantibody production, the NZB-derived Nba2 on chromosome 1 and the B6-derived H2 on chromosome 17 were strongly associated with gp70 IC (chromosome 1 with a LOD of 7.28, p = 7.10 x 10–9; chromosome 17 with a LOD of 8.90, p = 1.55 x 10–10). Two additional loci were associated with the formation of gp70 IC. One locus colocalized to the Sgp3 locus on chromosome 13, which contributed to gp70 production. The other locus peaked at the D7Nds5 marker (23.0 cM from the centromere) on proximal chromosome 7 and was designated Nba5. The Sgp3 and Nba5 loci gave LOD scores of 8.27 (p = 6.80 x 10–10) and 4.65 (p = 3.66 x 10–6), respectively. Notably, these two loci did not show any effect on IgG anti-DNA and anti-chromatin autoantibody production.
Linkage of marker loci to GN in B6 x (NZB x B6.Yaa)F1 BC mice
To assess the development of GN, a cohort of 121 B6 x (NZB x B6.Yaa)F1 BC mice was sacrificed either moribund or at the end of an 18-mo observation period, and glomerular lesions were evaluated histologically (grade 0–4). Approximately one-half of the BC mice (62 of 121) showed evidence of severe GN (grade 3). The comparison of genotypes of BC mice that were classified as positive or negative for severe GN was conducted by 2 analysis. Significant linkage was observed at two regions corresponding to Nba2 and Sgp3 (Table I). The highest association in each region was obtained at D1Mit36 and Fcgr2 on chromosome 1 (2 = 27.0, p = 2.01 x 10–7) and at D13Mit250 and D13Mit122 on chromosome 13 (2 = 8.94, p = 2.79 x 10–3). As in the case for autoantibody production, mice heterozygous at these loci developed significantly more severe GN than homozygous mice. No significant association was found with H2 on chromosome 17 and Nba5 on chromosome 7, despite their linkage with the production of autoantibodies.
Table I. Loci linked with severe GN in B6 x (NZB x B6.Yaa)F1 BC male mice bearing the Yaa mutation
Notably, BC mice developing severe GN had significantly higher titers of anti-DNA, anti-chromatin, and gp70 IC in sera at 8 mo of age (Fig. 4). Among these three different serological parameters, the strongest association was detected by the analysis of gp70 IC distribution (p = 6.60 x 10–6), followed by anti-DNA (p = 1.20 x 10–5) and anti-chromatin autoantibodies (p = 2.18 x 10–4). In contrast, serum levels of total gp70 were not significantly different between mice with or without severe GN (GN-positive group, 15.2 ± 9.0 μg/ml; GN-negative group, 12.8 ± 9.4 μg/ml; Fig. 4).
FIGURE 4. Serum levels of IgG anti-DNA and anti-chromatin autoantibodies, gp70 IC, and gp70 in B6 x (NZB x B6.Yaa)F1 BC male mice bearing the Yaa mutation. GN+, BC mice (n = 62) that developed severe GN (grade 3) by 18 mo of age. GN–, BC mice (n = 59) that failed to develop severe GN by 18 mo of age. Each symbol represents individual animal. Serum levels of anti-DNA, anti-chromatin, and gp70 IC were determined at 8 mo of age, and serum gp70 concentrations were measured at 4 mo of age. Results are expressed as U/ml for anti-DNA and anti-chromatin autoantibodies, and as micrograms per milliliter for gp70 IC and gp70. The mean values are indicated by the horizontal line.
Development of SLE in B6.Nba2 and B6.Nba5 congenic mice bearing the Yaa mutation
We have previously shown that B6 female mice bearing the Nba2 interval spontaneously produce increased levels of IgG anti-DNA and anti-chromatin autoantibodies, but fail to develop severe GN (15). To determine the effect of the Yaa mutation, B6.Yaa male mice bearing an Nba2 interval flanked by markers D1Mit47 and D1Mit461 (23-cM interval) were produced, and the development of lupus-like autoimmune syndrome was compared with that of B6.Nba2 female littermates and control B6.Yaa males. At 8 mo of age, B6.Nba2 Yaa males had higher titers of gp70 IC as well as IgG autoantibodies against DNA and chromatin than B6.Nba2 females (p < 0.001 for anti-DNA; p < 0.005 for anti-chromatin; p < 0.01 for gp70 IC) and control B6.Yaa males (p < 0.005 for anti-DNA; p < 0.001 for anti-chromatin and gp70 IC; Fig. 5). In marked contrast to B6.Nba2 female and control B6.Yaa male mice, B6.Nba2 Yaa males developed lethal GN (p < 0.001; Fig. 6), with a 50% mortality rate at 14 mo.
FIGURE 5. Serum levels of IgG anti-DNA and anti-chromatin autoantibodies, and gp70 IC in 8-mo-old B6, B6.Nba2, and B6.Nba5 male and female mice. Closed symbols, male mice bearing the Yaa mutation; open symbols, female mice lacking the Yaa mutation. Each symbol represents individual animal (10–16 mice in each group). Results are expressed as U/ml for anti-DNA and anti-chromatin autoantibodies, and as micrograms per milliliter for gp70 IC. The mean values are indicated by the horizontal line.
FIGURE 6. Development of GN in B6, B6.Nba2, B6.Nba5, and B6.NZB-Sgp3 male mice bearing the Yaa mutation. The intensity of glomerular lesions was scored on a 0–4 scale. Results from individual mice, sacrificed either moribund or at the end of a 15-mo observation period, are shown (10–16 mice in each group). Incidence of severe GN (grade 3) in both B6.Nba2 and B6.Nba5 Yaa males was significantly increased, as compared with B6 and B6.NZB-Sgp3 Yaa males (p < 0.001), and difference between B6.Nba2 and B6.Nba5 Yaa males was significant (p < 0.005).
To assess the pathogenic effect of the Nba5 locus on the development of SLE, we produced B6.Yaa mice bearing an Nba5 interval flanked by markers D7Mit154 and D7Mit194 (23-cM interval). As expected from the results obtained from BC mice, the presence of the Nba5 locus did not significantly enhance serum levels of IgG anti-DNA and anti-chromatin autoantibodies in B6.Yaa mice at 8 mo of age (Fig. 5). In contrast, serum gp70 IC levels in B6.Nba5 Yaa male mice were substantially increased, as compared with control B6.Yaa males and B6.Nba5 female littermates (p < 0.001). One-third (5 of 15) of Nba5 Yaa male mice died of GN by 15 mo of age (Fig. 6), although the incidence of severe GN in these mice was less than that of B6.Nba2 Yaa males (p < 0.005). However, none of B6.Nba5 female littermates died of GN within this period of time.
High serum gp70 levels in B6.NZB-Sgp3 congenic mice
We recently showed that B6 mice carrying the Sgp3 locus derived from the NZW strain have increased basal levels of serum gp70 (13). Because the corresponding region of the NZB strain was also linked with high serum levels of gp70 in B6 x (NZB x B6.Yaa)F1 BC mice, we generated a B6.NZB-Sgp3 congenic strain, which harbored a 20-cM NZB-derived interval encompassing markers D13Mit250 and D13Mit73. As observed in B6.NZW-Sgp3 congenic mice (13), 3- to 4-mo-old B6.NZB-Sgp3 congenic Yaa male mice had 10-fold higher levels of serum gp70 (19.6 ± 4.1 μg/ml), as compared with B6.Yaa males (1.9 ± 0.4 μg/ml; p < 0.001), but comparable to those of B6.NZW-Sgp3 Yaa males (18.8 ± 6.7 μg/ml; Fig. 7A). Notably, serum levels of gp70 in B6.Nba2 (2.0 ± 0.5 μg/ml) and B6.Nba5 Yaa males (2.1 ± 0.4 μg/ml) were not higher than those of B6.Yaa males.
FIGURE 7. Serum levels of gp70 and gp70 IC in B6, B6.NZB-Sgp3, and B6.NZW-Sgp3 male mice bearing the Yaa mutation. A, Serum concentrations of gp70 were determined at 3–4 mo of age. Each symbol represents individual animal (13 mice in each group). Results are expressed as μg/ml. B, Serum concentrations of gp70 IC were measured in B6.NZB-Sgp3 Yaa male mice at 2 and 8 mo of age (8 mice at 2 mo, and 16 mice at 8 mo). Results are expressed as μg/ml. The mean values are indicated by the horizontal line.
The linkage analysis of BC mice also showed the association of gp70 IC levels and GN with the NZB-Sgp3 locus. To determine whether this locus is able to promote anti-gp70 autoimmune responses in the presence of the Yaa mutation, serum levels of gp70 IC were compared between young (2-mo-old) and aged (8-mo-old) B6.NZB-Sgp3 Yaa male mice. Notably, 8-mo-old mice displayed only slightly increased titers of gp70 IC in sera, as compared with 2-mo-old mice (Fig. 7B). Furthermore, no elevated production of IgG anti-DNA and anti-chromatin autoantibodies was observed in 8-mo-old B6.NZB-Sgp3 Yaa male mice (data not shown), and none of them developed severe lethal GN by 15 mo of age (Fig. 6).
Discussion
Using a cohort of B6 x (NZB x B6.Yaa)F1 BC male mice, we mapped and characterized the NZB-derived lupus-susceptibility loci linked with Yaa-induced lupus-like disease. Our analysis identified three major loci, on NZB chromosomes 1, 7, and 13, which were implicated in anti-nuclear autoantibody production, nephritogenic gp70 IC formation, and lupus nephritis. Studies of B6.Yaa mice bearing each of the NZB-derived susceptibility intervals confirmed the significance of these loci, which differentially contributed to the development of SLE. In addition, our analysis of BC mice also demonstrated the presence of a B6-derived lupus-susceptibility locus closely linked to the MHC locus.
Comparative assessment of three different B6.Yaa mice congenic for each NZB susceptibility interval, together with the analysis of BC mice, clearly showed a major contribution of the Nba2 locus, located on the distal portion of NZB chromosome 1, to spontaneous production of anti-nuclear and anti-gp70 autoantibodies and subsequent development of lupus-like GN. These results are consistent with the previously demonstrated linkage of Nba2 with lupus traits in multiple different BC (12, 20, 21, 22, 23, 24, 25). Among a number of candidate lupus-susceptibility genes in the Nba2 interval, we recently showed that greatly increased expression of the IFN-inducible p202 (Ifi202) gene is mediated by the NZB vs B6 allele (15). In addition, the Fcgr2 allele present in the NZB strain has also been suggested as an additional candidate gene, because the expression of FcRIIB, a negative regulator BCR signaling, is defective in B cells in germinal centers of NZB mice (23, 26). In fact, we observed that partial FcRIIB deficiency, i.e., heterozygous level of FcRIIB expression, markedly promoted the production of autoantibodies in B6 mice by the presence of the Yaa mutation (27). Exploration of the respective contributions of the Ifi202 and Fcgr2 genes to the Nba2-linked autoimmune traits is currently in progress by the generation and analysis of Nba2 subcongenic mice.
It is significant that B6.Yaa males bearing the Nba5 locus (derived from the NZB chromosome 7) were able to develop severe GN. A lower incidence of severe GN in these congenic males than in B6.Nba2 Yaa males is most likely related to the fact that the effect of the Nba5 locus is selective only for anti-gp70 autoantibody production, while the Nba2 locus promotes not only the formation of gp70 IC, but also anti-DNA and anti-chromatin autoantibody production. However, it should be stressed that B6.Nba5 Yaa males significantly developed more severe glomerular lesions than control B6.Yaa male mice. Because serum levels of anti-DNA and anti-chromatin autoantibodies were comparable between B6.Nba5 and control Yaa male mice, this strongly suggests a major role for gp70 IC in the development of GN occurring in B6.Nba5 Yaa mice. Relatively low incidence and delayed onset of GN in B6.Nba5 Yaa males could be in part due to low serum concentrations of gp70 in these mice because of the difference of the Sgp3 allele. B6.Nba5 Yaa mice express much lower levels of serum gp70, compared with lupus-prone (NZB x NZW)F1, MRL, and BXSB mice (2). Because the linkage analysis in BC mice showed the association of the Sgp3 locus with high serum levels of gp70 and GN, it is of interest to determine whether the onset and incidence of GN can be accelerated in B6.Nba5 and B6.Nba2 Yaa male mice bearing the NZB-derived Sgp3 allele.
The present study confirmed and extended our previous observation that the Sgp3 locus on midchromosome 13 controls basal levels of serum gp70 production (7, 11, 13, 28). The analysis of sera from B6 mice bearing the NZB-derived Sgp3 interval provided evidence that the NZB-Sgp3 allele was associated with elevated serum levels of gp70, which were comparable to those obtained in B6 mice bearing the NZW-Sgp3 interval (13). This indicates that both NZB and NZW mice most likely share the same Sgp3 allele, consistent with the fact that these mice have similarly high serum levels of gp70 (2, 5). However, serum concentrations of gp70 in these two Sgp3 congenic mice were still lower than those seen in NZB and NZW mice, indicating the presence of other loci controlling the production of serum gp70. In contrast to earlier studies (29), genome-wide linkage analyses involving NZB, NZW, and BXSB mice had failed to show linkage of serum gp70 levels with the H2 locus (7, 11, 28). In contrast, we recently noted the linkage of serum gp70 levels with a locus on distal chromosome 4 of both NZB and NZW mice (28, 30). Preliminary studies have shown that B6 mice bearing the NZB-derived Nba1 interval on distal chromosome 4 have significantly increased concentrations of serum gp70. It is worth noting that Nba1 is linked with GN, but not with the production of lupus autoantibodies (31).
An NZB locus on chromosome 13 overlapping with the Sgp3 locus was also associated with high serum levels of gp70 IC and GN, but had no effect on anti-DNA and anti-chromatin autoantibody production. Similar results were also obtained by the analysis of B6 x (NZW x B6.Yaa)F1 and BXSB x (B10 x BXSB)F1 BC mice bearing the Yaa mutation (7, 11, 32) as well as (NZB x B6)F2 mice (28). However, B6.Yaa NZB-Sgp3 congenic mice showed only minimal age-dependent increases in serum levels of gp70 IC and failed to develop GN. This indicates that the Sgp3 locus itself regulates the production of serum gp70, but not autoimmune responses. Accordingly, the association of gp70 IC with this locus in BC mice can be a consequence of increased production of serum gp70 in liver, which is the major organ to synthesize and secrete gp70 into sera (33). It is worth noting that the Gv1 (gross virus Ag 1) gene that overlaps with the Sgp3 locus controls the expression of thymic GIX gp70 Ag (34, 35), the expression of which is closely correlated to serum levels of gp70 (36, 37). As Gv1 most likely regulates in trans the expression of multiple endogenous retroviral transcripts in different tissues, including the liver (38), it is reasonable to assume that Gv1 and Sgp3 are identical or related genes regulating the expression of endogenous retrovirus.
Our present BC study also revealed a QTL of B6 origin on chromosome 17, which showed a strong linkage with anti-DNA production and gp70 IC formation. This locus overlapped with the H2 complex, and the development of lupus-like disease in our BC mice was strongly associated with H2b/b homozygosity (vs H2b/d heterozygosity, with H2b and H2d of B6 and NZB origin, respectively). We and others have previously shown that lupus susceptibility is more closely linked to the H2b haplotype than to the H2d haplotype in BXSB, (NZB x BXSB)F1, and B6-Faslpr mice (17, 39, 40). The autoimmune inhibitory effect of H2d may be in part related to the expression of I-E molecules, because H2b mice fail to express I-E due to the deletion of the promoter region of the Ea gene encoding the I-E -chain (41). It has been previously shown that the introduction of two copies of the functional Ea transgene in lupus-prone BXSB mice (H2b) is sufficient to prevent the development of SLE (42). Although the precise mechanism of the protection of SLE by I-E has not fully been elucidated, our recent results support the hypothesis that formation of I-E -chain-derived peptides, which display a high affinity to I-A molecules, in B cells may decrease the use of I-A molecules for the presentation of pathogenic self peptides, thereby limiting the activation of autoreactive T and B cells (43).
Past genome-wide linkage analyses involving different lupus-prone mice have revealed >30 lupus-susceptibility loci in lupus-prone NZB, NZW, BXSB, and MRL mice (8, 9). However, it should be stressed that our studies in B6 x (NZW x B6.Yaa)F1 and B6 x (NZB x B6.Yaa)F1 BC show that a relatively small number and selective group of critical genes may be sufficient to cause full expression of SLE in the presence of the Yaa gene. Because both (NZW x B6)F1 and (NZB x B6)F1 mice lacking the Yaa gene are essentially normal, Yaa is capable of interacting with NZW and NZB lupus-susceptibility genes, which are by themselves not sufficient, to trigger lupus-like autoimmune responses. Moreover, the NZW- and NZB-derived genes involved in the development of SLE through interaction with Yaa are mostly different in each strain and only a small subset of those previously described. Notably, studies in B6 x (NZW x B6.Yaa)F1 BC mice identified a single locus on the NZW chromosome 7, which predisposes (NZW x B6)F1 mice to Yaa gene-induced SLE (11). However, this NZW locus is distinct from Nba5, as the former promotes overall autoantibody production, while the latter is selective on gp70 IC formation. In addition and in contrast to the NZB-derived Nba2, these past BC studies did not identify the important NZW lupus-susceptibility genes encoded on distal chromosome 1 (44).
A congenic dissection approach to convert a polygenic system into a series of monogenic systems has proved to be a powerful tool to dissect the complex genetic interactions implicated in murine SLE (13, 15, 25, 44, 45, 46). In addition, analysis of phenotypes in mono- and bicongenic strains carrying the susceptibility loci should help define key epistatic interactions in the development of SLE (47). The construction of further subcongenic B6.Yaa mice carrying the Nba2, Nba5, or Sgp3 locus should allow fine mapping and positional cloning of the gene(s) underlying the NZB predisposition to murine SLE. The eventual identification of mouse lupus-susceptibility genes will let us directly address the relevance of their human counterparts, and has obvious and promising implications for diagnostic, prognostic, and therapeutic approaches in SLE and related autoimmune diseases.
Acknowledgments
We thank Dr. L. Reininger for his critical reading of the manuscript; Drs. S. Hirose and T. Winkler for providing us information for microsatellite markers polymorphic between NZB and B6 mice; and G. Leyvraz, G. Brighouse, S. Jacquier, and G. Celetta for their excellent technical assistance.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This work was supported by a grant from the Swiss National Foundation for Scientific Research and Grant AR 37070 from the National Institutes of Health. L.F.-J. is a recipient of a fellowship from the Arthritis Research Campaign, U.K.
2 Address correspondence and reprint requests to Dr. Shozo Izui, Department of Pathology and Immunology, Centre Médical Universitaire, 1211 Geneva 4, Switzerland. E-mail address: Shozo.Izui@medecine.unige.ch
3 Abbreviations used in this paper: NZB, New Zealand Black; BC, backcross; GN, glomerulonephritis; Gv1, gross virus Ag 1; IC, immune complex; LOD, log-likelihood of the odds; Nba, NZB autoimmunity; NZW, New Zealand White; QTL, quantitative trait locus; Sgp3, serum gp70 production 3; SLE, systemic lupus erythematosus; Yaa, Y-linked autoimmune acceleration.
Received for publication September 27, 2004. Accepted for publication November 2, 2004.
References
Lambert, P. H., F. J. Dixon. 1968. Pathogenesis of the glomerulonephritis of NZB/W mice. J. Exp. Med. 127:507.
Izui, S., P. J. McConahey, A. N. Theofilopoulos, F. J. Dixon. 1979. Association of circulating retroviral gp70-anti-gp70 immune complexes with murine systemic lupus erythematosus. J. Exp. Med. 149:1099.
Maruyama, N., F. Furukawa, Y. Nakai, Y. Sasaki, K. Ohta, S. Ozaki, S. Hirose, T. Shirai. 1983. Genetic studies of autoimmunity in New Zealand mice. IV. Contribution of NZB and NZW genes to the spontaneous occurrence of retroviral gp70 immune complexes in (NZB x NZW)F1 hybrid and the correlation to renal disease. J. Immunol. 130:740.
Vyse, T. J., R. K. Halterman, S. J. Rozzo, S. Izui, B. L. Kotzin. 1999. Control of separate pathogenic autoantibody responses marks MHC gene contributions to murine lupus. Proc. Natl. Acad. Sci. USA 96:8098.
Yoshiki, T., R. C. Mellors, M. Strand, J. T. August. 1974. The viral envelope glycoprotein of murine leukemia virus and the pathogenesis of immune complex glomerulonephritis of New Zealand mice. J. Exp. Med. 140:1011.
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.
Haywood, M. E. K., T. J. Vyse, A. McDermott, E. M. Thompson, A. Ida, M. J. Walport, S. Izui, B. J. Morley. 2001. Autoantigen glycoprotein 70 expression is regulated by a single locus, which acts as a checkpoint for pathogenic anti-glycoprotein 70 autoantibody production and hence for the corresponding development of severe nephritis, in lupus-prone BXSB mice. J. Immunol. 167:1728
Vyse, T. J., B. L. Kotzin. 1998. Genetic susceptibility to systemic lupus erythematosus. Annu. Rev. Immunol. 16:261.
Wakeland, E. K., K. Liu, R. R. Graham, T. W. Behrens. 2001. Delineating the genetic basis of systemic lupus erythematosus. Immunity 15:397.
Izui, S., M. Iwamoto, L. Fossati, R. Merino, S. Takahashi, N. Ibnou-Zekri. 1995. The Yaa gene model of systemic lupus erythematosus. Immunol. Rev. 144:137.
Santiago, M. L., C. Mary, D. Parzy, C. Jacquet, X. Montagutelli, R. M. E. Parkhouse, R. Lemoine, S. Izui, L. Reininger. 1998. Linkage of a major quantitative trait locus to Yaa gene-induced lupus-like nephritis in (NZW x C57BL/6)F1 mice. Eur. J. Immunol. 28:4257.[Medline](Shuichi Kikuchi, Liliane )
By assessing the development of Y-linked autoimmune acceleration (Yaa) gene-induced systemic lupus erythematosus in C57BL/6 (B6) x (New Zealand Black (NZB) x B6.Yaa)F1 backcross male mice, we mapped three major susceptibility loci derived from the NZB strain. These three quantitative trait loci (QTL) on NZB chromosomes 1, 7, and 13 differentially regulated three different autoimmune traits: anti-nuclear autoantibody production, gp70-anti-gp70 immune complex (gp70 IC) formation, and glomerulonephritis. Contributions to the disease traits were further confirmed by generating and analyzing three different B6.Yaa congenic mice, each carrying one individual NZB QTL. The chromosome 1 locus that overlapped with the previously identified Nba2 (NZB autoimmunity 2) locus regulated all three traits. A newly identified chromosome 7 locus, designated Nba5, selectively promoted anti-gp70 autoantibody production, hence the formation of gp70 IC and glomerulonephritis. B6.Yaa mice bearing the NZB chromosome 13 locus displayed increased serum gp70 production, but not gp70 IC formation and glomerulonephritis. This locus, called Sgp3 (serum gp70 production 3), selectively regulated the production of serum gp70, thereby contributing to the formation of nephritogenic gp70 IC and glomerulonephritis, in combination with Nba2 and Nba5 in NZB mice. Among these three loci, a major role of Nba2 was demonstrated, because B6.Yaa Nba2 congenic male mice developed the most severe disease. Finally, our analysis revealed the presence in B6 mice of an H2-linked QTL, which regulated autoantibody production. This locus had no apparent individual effect, but most likely modulated disease severity through interaction with NZB-derived susceptibility loci.
Introduction
The New Zealand Black (NZB)3 x New Zealand White (NZW)F1 hybrid mice spontaneously develop a generalized autoimmune disorder resembling human systemic lupus erythematosus (SLE). The major known targets of the autoimmune responses in this model of SLE are chromatin (including its major constituent, DNA) and serum gp70, the major envelope glycoprotein of endogenous retrovirus (1, 2, 3, 4). Anti-chromatin, anti-DNA, and anti-gp70 autoantibodies have been implicated in the development of severe immune complex (IC)-mediated glomerulonephritis (GN), not only in these mice, but also in lupus-prone MRL-Faslpr and BXSB mice (5, 6, 7).
It is now well established that SLE is a polygenic disease, in which multiple, unlinked genes are operative in a threshold manner. Linkage analysis using microsatellite markers in crosses involving NZB or NZW mice have revealed a number of lupus-susceptibility loci (8, 9). Although F1 hybrids of NZW and C57BL/6 (B6) mice are essentially normal, these mice develop severe lupus-like nephritis closely associated with elevated serum levels of IgG anti-DNA autoantibodies and gp70-anti-gp70 IC (gp70 IC) when they carry the Yaa (Y-linked autoimmune acceleration) mutation, derived from male BXSB mice (10). Linkage analysis of B6 x (NZW x B6.Yaa)F1 backcross (BC) male mice provided evidence for a major quantitative trait locus (QTL) on NZW chromosome 7 controlling both the severity of GN and the production of IgG anti-DNA and gp70 IC (11). The results indicated that the genetic analysis involving Yaa represents a useful tool for dissecting the complex genetic interactions responsible for the development of murine SLE.
In the present study, we used B6 x (NZB x B6.Yaa)F1 BC male mice and congenic mice bearing mapped susceptibility intervals to identify critical NZB-derived lupus-susceptibility loci implicated in murine SLE. In this study, we report the mapping of three major QTL from the NZB strain: the first on chromosome 1 overlapping with the previously identified Nba2 (New Zealand black autoimmunity 2) locus (12); the second on chromosome 7, designated Nba5 (New Zealand black autoimmunity 5); and the third on chromosome 13 corresponding to the Sgp3 (serum gp70 production 3) locus of NZW mice (13). Moreover, we identified a QTL on chromosome 17 from the B6 strain contributing to autoantibody production.
Materials and Methods
Mice
NZB mice (H2d) were purchased from The Jackson Laboratory. B6 mice (H2b) bearing the Yaa mutation (B6.Yaa) were established by repeated backcrossing (n > 20), as described previously (14). (NZB x B6.Yaa)F1 and B6 x (NZB x B6.Yaa)F1 BC mice were obtained by local breeding in our animal facility. B6.NZB-Nba2 (B6.Nba2) and B6.NZW-Sgp3 congenic mice were generated, as described previously (13, 15). B6.NZB-Nba5 (B6.Nba5) and B6.NZB-Sgp3 congenic mice were generated by backcrossing the NZB-derived Nba5 or Sgp3 intervals onto the B6 background using marker-assisted selection, as described previously (13). After five or six generations of backcrossing, siblings were intercrossed to generate B6.Nba5 and B6.NZB-Sgp3 congenic mice homozygous for the respective NZB chromosome 7 and chromosome 13 intervals. Males of all congenic mice used in the present study carry the Yaa mutation. Blood samples were collected by orbital sinus puncture, and sera were stored at –20°C until use.
Serological assays
Serum levels of IgG autoantibodies against chromatin and heat-denatured DNA were determined by ELISA, as described previously (15, 16). Results are expressed in U/ml in reference to a standard curve derived from a serum pool of MRL-Faslpr mice. Concentrations of total gp70 in sera were determined by ELISA, as described previously (17). Serum levels of gp70 IC were quantified by the same ELISA combined with the treatment of sera with 10% polyethylene glycol (average m.w. 6000), which precipitates only Ab-bound gp70, but not free gp70, as described (17). Results are expressed as micrograms per milliliter of gp70 by referring to a standard curve obtained from a serum pool of NZB mice with known amounts of gp70.
Genotyping and statistical analysis
Genotypes were determined by PCR using 95 selected microsatellite markers either purchased from Research Genetics or Invitrogen Life Technologies. DNA from NZB, B6, (NZB x B6)F1, and BC mice were extracted from tail biopsies kept at –70°C before use. PCR amplification was conducted with RED TaqDNA polymerase (Sigma-Aldrich) using a GeneAmp PCR system 9700 thermal cycler (Applied Biosystems), as described (13). The positions of the microsatellite markers with respect to the centromere were obtained from the Mouse Genome Database at www.informatics.jax.org. The linkage program MAPMAKER/QTL was used to identify QTL (18). Autoantibody and gp70 levels were log10 transformed. The association of severe GN (grade 3) with marker loci was tested by a 2 test for goodness-of-fit against an expected 50:50 distribution using a standard (2 x 2) contingency matrix. A threshold for suggestive linkage was set at log-likelihood of the odds (LOD) > 1.9, p < 0.0034 (2 > 8.6, 1 degree of freedom), and for significant linkage was
LOD >3.3, p < 0.0001 (2 > 10.8, 1 degree of freedom), based on the recommendation of Lander and Kruglyak (19). Comparison of serum levels of anti-nuclear autoantibodies, gp70 IC, and gp70 between different groups of mice was performed with the Wilcoxon two-sample test. Probability values >5% were considered insignificant.
Histopathology
Kidney samples were collected when mice were moribund or at the end of the experiment at 15–18 mo of age. Histological sections were stained with periodic acid-Schiff reagent. The extent of GN was graded on a 0–4 scale based on the intensity and extent of histopathological changes, as described previously (14). GN with grade 3 or 4 was considered a significant contributor to clinical disease and/or death.
Results
Development of lupus-like autoimmune syndrome in (NZB x B6.Yaa)F1 male mice
The Yaa gene has been shown to be responsible for the accelerated development of lupus-like autoimmune syndrome in BXSB mice and in their F1 hybrids with NZW or NZB mice (10). In this study, we first analyzed whether (NZB x B6.Yaa)F1 males bearing the Yaa gene also spontaneously developed a lupus-like autoimmune syndrome. At 8 mo of age, sera from (NZB x B6.Yaa)F1 Yaa males displayed markedly elevated IgG anti-DNA and anti-chromatin autoantibodies, as well as gp70 IC, compared with B6.Yaa mice (p < 0.001; Fig. 1). Consequently, F1 male mice developed a lethal form of severe GN; 50% of them (n = 18) died of GN (grade 3) by 14 mo of age, and only 15% were alive at 18 mo of age. Their renal lesions were characterized by increased glomerular cellularity, obliteration of the glomerular architecture, and the presence of tubular cast formation. In contrast, B6.Yaa males showed only slight mesangial alterations at 15 mo of age (mean histological grades of 14 mice: 0.4 ± 0.5).
FIGURE 1. Serum levels of IgG anti-DNA and anti-chromatin autoantibodies, gp70 IC, and gp70 in B6.Yaa, (NZB x B6.Yaa)F1, and B6 x (NZB x B6.Yaa)F1 BC male mice. Each symbol represents individual animal of B6.Yaa (n = 10), (NZB x B6.Yaa)F1 (n = 20), and 154 BC Yaa males (n = 154). Serum levels of anti-DNA, anti-chromatin, and gp70 IC were determined at 8 mo of age, and serum gp70 concentrations were measured at 4 mo of age. Results are expressed as U/ml for anti-DNA and anti-chromatin autoantibodies, and as μg/ml for gp70 IC and gp70. The mean values are indicated by the horizontal line.
Linkage of marker loci to IgG anti-DNA and anti-chromatin autoantibodies in B6 x (NZB x B6.Yaa)F1 BC mice
To determine the NZB genetic contribution to the Yaa-induced acceleration of autoimmune disease in (NZB x B6.Yaa)F1 male mice, a total of 154 B6 x (NZB x B6.Yaa)F1 male BC mice was produced, and sera from 6-, 8-, and 10-mo-old BC mice were analyzed for IgG anti-DNA and anti-chromatin autoantibodies. Analysis of these two autoantibodies with MAPMARKER/QTL revealed significant linkage with an interval on NZB chromosome 1 directly overlapped with the Nba2 locus at 90–98 cM from the centromere (Fig. 2). The locus Fcgr2 (92.3 cM from the centromere) gave a LOD score of 7.08 (p = 1.13 x 10–8) for anti-DNA and 13.24 (p = 5.73 x 10 –15) for anti-chromatin autoantibodies. In addition to Nba2, a high LOD score for IgG anti-DNA (10.55, p = 3.14 x 10–12) and IgG anti-chromatin (3.02, p = 1.94 x 10–4) was observed at the H2 locus on chromosome 17 (Fig. 2). However, this susceptibility allele was inherited from the B6 strain.
FIGURE 2. Linkage of chromosome 1 and 17 markers with serum levels of IgG anti-DNA and anti-chromatin autoantibodies in B6 x (NZB x B6.Yaa)F1 BC male mice bearing the Yaa mutation. Autoantibody titers were determined at 6, 8, and 10 mo of age. LOD scores for each trait were generated with MAPMAKER/QTL, and the highest scores obtained at 8 mo of age are shown. The horizontal dotted line represents the threshold for suggestive linkage.
Linkage of marker loci to gp70 and gp70 IC in B6 x (NZB x B6.Yaa)F1 BC mice
Serum levels of gp70 are highly variable among different strains of mice, and NZB mice produce larger amounts of serum gp70 than B6 mice (2). Consistently, serum gp70 concentrations in 4-mo-old (NZB x B6.Yaa)F1 male mice (23.1 ± 3.9 μg/ml) were 10-fold higher than those in B6.Yaa male mice (2.3 ± 0.7 μg/ml; p < 0.001) (Fig. 1). QTL analysis was conducted to investigate the genetics of serum gp70 production in B6 x (NZB x B6.Yaa)F1 BC male mice at 4 mo of age. A single QTL of NZB origin was identified on midchromosome 13 and peaked close to the D13Mit193 marker (43 cM from the centromere), with a LOD score of 8.77 (p = 2.11 x 10–10; Fig. 3). The Sgp3 locus has been previously mapped to this position in B6 x (NZW x B6.Yaa)F1 and B10 x (B10 x BXSB)F1 BC mice (7, 11). The current results show that the Sgp3 locus is also the major determinant of quantitative variation of gp70 in NZB mice.
FIGURE 3. Linkage of chromosome 1, 7, 13, and 17 markers with serum levels of gp70 and gp70 IC in B6 x (NZB x B6.Yaa)F1 BC male mice bearing the Yaa mutation. Serum concentrations of gp70 were determined at 4 mo of age, and gp70 IC levels were measured at 6, 8, and 10 mo of age. LOD scores for each trait were generated with MAPMAKER/QTL, and the highest scores for gp70 IC obtained at 8 mo of age for chromosomes 1 and 13 and at 6 mo of age for chromosomes 7 and 17 are shown. The horizontal dotted line represents the threshold for suggestive linkage.
When the genotype data of BC mice were analyzed for loci linked with levels of gp70 IC, significant linkage with four regions was revealed (Fig. 3). As in the case of anti-nuclear autoantibody production, the NZB-derived Nba2 on chromosome 1 and the B6-derived H2 on chromosome 17 were strongly associated with gp70 IC (chromosome 1 with a LOD of 7.28, p = 7.10 x 10–9; chromosome 17 with a LOD of 8.90, p = 1.55 x 10–10). Two additional loci were associated with the formation of gp70 IC. One locus colocalized to the Sgp3 locus on chromosome 13, which contributed to gp70 production. The other locus peaked at the D7Nds5 marker (23.0 cM from the centromere) on proximal chromosome 7 and was designated Nba5. The Sgp3 and Nba5 loci gave LOD scores of 8.27 (p = 6.80 x 10–10) and 4.65 (p = 3.66 x 10–6), respectively. Notably, these two loci did not show any effect on IgG anti-DNA and anti-chromatin autoantibody production.
Linkage of marker loci to GN in B6 x (NZB x B6.Yaa)F1 BC mice
To assess the development of GN, a cohort of 121 B6 x (NZB x B6.Yaa)F1 BC mice was sacrificed either moribund or at the end of an 18-mo observation period, and glomerular lesions were evaluated histologically (grade 0–4). Approximately one-half of the BC mice (62 of 121) showed evidence of severe GN (grade 3). The comparison of genotypes of BC mice that were classified as positive or negative for severe GN was conducted by 2 analysis. Significant linkage was observed at two regions corresponding to Nba2 and Sgp3 (Table I). The highest association in each region was obtained at D1Mit36 and Fcgr2 on chromosome 1 (2 = 27.0, p = 2.01 x 10–7) and at D13Mit250 and D13Mit122 on chromosome 13 (2 = 8.94, p = 2.79 x 10–3). As in the case for autoantibody production, mice heterozygous at these loci developed significantly more severe GN than homozygous mice. No significant association was found with H2 on chromosome 17 and Nba5 on chromosome 7, despite their linkage with the production of autoantibodies.
Table I. Loci linked with severe GN in B6 x (NZB x B6.Yaa)F1 BC male mice bearing the Yaa mutation
Notably, BC mice developing severe GN had significantly higher titers of anti-DNA, anti-chromatin, and gp70 IC in sera at 8 mo of age (Fig. 4). Among these three different serological parameters, the strongest association was detected by the analysis of gp70 IC distribution (p = 6.60 x 10–6), followed by anti-DNA (p = 1.20 x 10–5) and anti-chromatin autoantibodies (p = 2.18 x 10–4). In contrast, serum levels of total gp70 were not significantly different between mice with or without severe GN (GN-positive group, 15.2 ± 9.0 μg/ml; GN-negative group, 12.8 ± 9.4 μg/ml; Fig. 4).
FIGURE 4. Serum levels of IgG anti-DNA and anti-chromatin autoantibodies, gp70 IC, and gp70 in B6 x (NZB x B6.Yaa)F1 BC male mice bearing the Yaa mutation. GN+, BC mice (n = 62) that developed severe GN (grade 3) by 18 mo of age. GN–, BC mice (n = 59) that failed to develop severe GN by 18 mo of age. Each symbol represents individual animal. Serum levels of anti-DNA, anti-chromatin, and gp70 IC were determined at 8 mo of age, and serum gp70 concentrations were measured at 4 mo of age. Results are expressed as U/ml for anti-DNA and anti-chromatin autoantibodies, and as micrograms per milliliter for gp70 IC and gp70. The mean values are indicated by the horizontal line.
Development of SLE in B6.Nba2 and B6.Nba5 congenic mice bearing the Yaa mutation
We have previously shown that B6 female mice bearing the Nba2 interval spontaneously produce increased levels of IgG anti-DNA and anti-chromatin autoantibodies, but fail to develop severe GN (15). To determine the effect of the Yaa mutation, B6.Yaa male mice bearing an Nba2 interval flanked by markers D1Mit47 and D1Mit461 (23-cM interval) were produced, and the development of lupus-like autoimmune syndrome was compared with that of B6.Nba2 female littermates and control B6.Yaa males. At 8 mo of age, B6.Nba2 Yaa males had higher titers of gp70 IC as well as IgG autoantibodies against DNA and chromatin than B6.Nba2 females (p < 0.001 for anti-DNA; p < 0.005 for anti-chromatin; p < 0.01 for gp70 IC) and control B6.Yaa males (p < 0.005 for anti-DNA; p < 0.001 for anti-chromatin and gp70 IC; Fig. 5). In marked contrast to B6.Nba2 female and control B6.Yaa male mice, B6.Nba2 Yaa males developed lethal GN (p < 0.001; Fig. 6), with a 50% mortality rate at 14 mo.
FIGURE 5. Serum levels of IgG anti-DNA and anti-chromatin autoantibodies, and gp70 IC in 8-mo-old B6, B6.Nba2, and B6.Nba5 male and female mice. Closed symbols, male mice bearing the Yaa mutation; open symbols, female mice lacking the Yaa mutation. Each symbol represents individual animal (10–16 mice in each group). Results are expressed as U/ml for anti-DNA and anti-chromatin autoantibodies, and as micrograms per milliliter for gp70 IC. The mean values are indicated by the horizontal line.
FIGURE 6. Development of GN in B6, B6.Nba2, B6.Nba5, and B6.NZB-Sgp3 male mice bearing the Yaa mutation. The intensity of glomerular lesions was scored on a 0–4 scale. Results from individual mice, sacrificed either moribund or at the end of a 15-mo observation period, are shown (10–16 mice in each group). Incidence of severe GN (grade 3) in both B6.Nba2 and B6.Nba5 Yaa males was significantly increased, as compared with B6 and B6.NZB-Sgp3 Yaa males (p < 0.001), and difference between B6.Nba2 and B6.Nba5 Yaa males was significant (p < 0.005).
To assess the pathogenic effect of the Nba5 locus on the development of SLE, we produced B6.Yaa mice bearing an Nba5 interval flanked by markers D7Mit154 and D7Mit194 (23-cM interval). As expected from the results obtained from BC mice, the presence of the Nba5 locus did not significantly enhance serum levels of IgG anti-DNA and anti-chromatin autoantibodies in B6.Yaa mice at 8 mo of age (Fig. 5). In contrast, serum gp70 IC levels in B6.Nba5 Yaa male mice were substantially increased, as compared with control B6.Yaa males and B6.Nba5 female littermates (p < 0.001). One-third (5 of 15) of Nba5 Yaa male mice died of GN by 15 mo of age (Fig. 6), although the incidence of severe GN in these mice was less than that of B6.Nba2 Yaa males (p < 0.005). However, none of B6.Nba5 female littermates died of GN within this period of time.
High serum gp70 levels in B6.NZB-Sgp3 congenic mice
We recently showed that B6 mice carrying the Sgp3 locus derived from the NZW strain have increased basal levels of serum gp70 (13). Because the corresponding region of the NZB strain was also linked with high serum levels of gp70 in B6 x (NZB x B6.Yaa)F1 BC mice, we generated a B6.NZB-Sgp3 congenic strain, which harbored a 20-cM NZB-derived interval encompassing markers D13Mit250 and D13Mit73. As observed in B6.NZW-Sgp3 congenic mice (13), 3- to 4-mo-old B6.NZB-Sgp3 congenic Yaa male mice had 10-fold higher levels of serum gp70 (19.6 ± 4.1 μg/ml), as compared with B6.Yaa males (1.9 ± 0.4 μg/ml; p < 0.001), but comparable to those of B6.NZW-Sgp3 Yaa males (18.8 ± 6.7 μg/ml; Fig. 7A). Notably, serum levels of gp70 in B6.Nba2 (2.0 ± 0.5 μg/ml) and B6.Nba5 Yaa males (2.1 ± 0.4 μg/ml) were not higher than those of B6.Yaa males.
FIGURE 7. Serum levels of gp70 and gp70 IC in B6, B6.NZB-Sgp3, and B6.NZW-Sgp3 male mice bearing the Yaa mutation. A, Serum concentrations of gp70 were determined at 3–4 mo of age. Each symbol represents individual animal (13 mice in each group). Results are expressed as μg/ml. B, Serum concentrations of gp70 IC were measured in B6.NZB-Sgp3 Yaa male mice at 2 and 8 mo of age (8 mice at 2 mo, and 16 mice at 8 mo). Results are expressed as μg/ml. The mean values are indicated by the horizontal line.
The linkage analysis of BC mice also showed the association of gp70 IC levels and GN with the NZB-Sgp3 locus. To determine whether this locus is able to promote anti-gp70 autoimmune responses in the presence of the Yaa mutation, serum levels of gp70 IC were compared between young (2-mo-old) and aged (8-mo-old) B6.NZB-Sgp3 Yaa male mice. Notably, 8-mo-old mice displayed only slightly increased titers of gp70 IC in sera, as compared with 2-mo-old mice (Fig. 7B). Furthermore, no elevated production of IgG anti-DNA and anti-chromatin autoantibodies was observed in 8-mo-old B6.NZB-Sgp3 Yaa male mice (data not shown), and none of them developed severe lethal GN by 15 mo of age (Fig. 6).
Discussion
Using a cohort of B6 x (NZB x B6.Yaa)F1 BC male mice, we mapped and characterized the NZB-derived lupus-susceptibility loci linked with Yaa-induced lupus-like disease. Our analysis identified three major loci, on NZB chromosomes 1, 7, and 13, which were implicated in anti-nuclear autoantibody production, nephritogenic gp70 IC formation, and lupus nephritis. Studies of B6.Yaa mice bearing each of the NZB-derived susceptibility intervals confirmed the significance of these loci, which differentially contributed to the development of SLE. In addition, our analysis of BC mice also demonstrated the presence of a B6-derived lupus-susceptibility locus closely linked to the MHC locus.
Comparative assessment of three different B6.Yaa mice congenic for each NZB susceptibility interval, together with the analysis of BC mice, clearly showed a major contribution of the Nba2 locus, located on the distal portion of NZB chromosome 1, to spontaneous production of anti-nuclear and anti-gp70 autoantibodies and subsequent development of lupus-like GN. These results are consistent with the previously demonstrated linkage of Nba2 with lupus traits in multiple different BC (12, 20, 21, 22, 23, 24, 25). Among a number of candidate lupus-susceptibility genes in the Nba2 interval, we recently showed that greatly increased expression of the IFN-inducible p202 (Ifi202) gene is mediated by the NZB vs B6 allele (15). In addition, the Fcgr2 allele present in the NZB strain has also been suggested as an additional candidate gene, because the expression of FcRIIB, a negative regulator BCR signaling, is defective in B cells in germinal centers of NZB mice (23, 26). In fact, we observed that partial FcRIIB deficiency, i.e., heterozygous level of FcRIIB expression, markedly promoted the production of autoantibodies in B6 mice by the presence of the Yaa mutation (27). Exploration of the respective contributions of the Ifi202 and Fcgr2 genes to the Nba2-linked autoimmune traits is currently in progress by the generation and analysis of Nba2 subcongenic mice.
It is significant that B6.Yaa males bearing the Nba5 locus (derived from the NZB chromosome 7) were able to develop severe GN. A lower incidence of severe GN in these congenic males than in B6.Nba2 Yaa males is most likely related to the fact that the effect of the Nba5 locus is selective only for anti-gp70 autoantibody production, while the Nba2 locus promotes not only the formation of gp70 IC, but also anti-DNA and anti-chromatin autoantibody production. However, it should be stressed that B6.Nba5 Yaa males significantly developed more severe glomerular lesions than control B6.Yaa male mice. Because serum levels of anti-DNA and anti-chromatin autoantibodies were comparable between B6.Nba5 and control Yaa male mice, this strongly suggests a major role for gp70 IC in the development of GN occurring in B6.Nba5 Yaa mice. Relatively low incidence and delayed onset of GN in B6.Nba5 Yaa males could be in part due to low serum concentrations of gp70 in these mice because of the difference of the Sgp3 allele. B6.Nba5 Yaa mice express much lower levels of serum gp70, compared with lupus-prone (NZB x NZW)F1, MRL, and BXSB mice (2). Because the linkage analysis in BC mice showed the association of the Sgp3 locus with high serum levels of gp70 and GN, it is of interest to determine whether the onset and incidence of GN can be accelerated in B6.Nba5 and B6.Nba2 Yaa male mice bearing the NZB-derived Sgp3 allele.
The present study confirmed and extended our previous observation that the Sgp3 locus on midchromosome 13 controls basal levels of serum gp70 production (7, 11, 13, 28). The analysis of sera from B6 mice bearing the NZB-derived Sgp3 interval provided evidence that the NZB-Sgp3 allele was associated with elevated serum levels of gp70, which were comparable to those obtained in B6 mice bearing the NZW-Sgp3 interval (13). This indicates that both NZB and NZW mice most likely share the same Sgp3 allele, consistent with the fact that these mice have similarly high serum levels of gp70 (2, 5). However, serum concentrations of gp70 in these two Sgp3 congenic mice were still lower than those seen in NZB and NZW mice, indicating the presence of other loci controlling the production of serum gp70. In contrast to earlier studies (29), genome-wide linkage analyses involving NZB, NZW, and BXSB mice had failed to show linkage of serum gp70 levels with the H2 locus (7, 11, 28). In contrast, we recently noted the linkage of serum gp70 levels with a locus on distal chromosome 4 of both NZB and NZW mice (28, 30). Preliminary studies have shown that B6 mice bearing the NZB-derived Nba1 interval on distal chromosome 4 have significantly increased concentrations of serum gp70. It is worth noting that Nba1 is linked with GN, but not with the production of lupus autoantibodies (31).
An NZB locus on chromosome 13 overlapping with the Sgp3 locus was also associated with high serum levels of gp70 IC and GN, but had no effect on anti-DNA and anti-chromatin autoantibody production. Similar results were also obtained by the analysis of B6 x (NZW x B6.Yaa)F1 and BXSB x (B10 x BXSB)F1 BC mice bearing the Yaa mutation (7, 11, 32) as well as (NZB x B6)F2 mice (28). However, B6.Yaa NZB-Sgp3 congenic mice showed only minimal age-dependent increases in serum levels of gp70 IC and failed to develop GN. This indicates that the Sgp3 locus itself regulates the production of serum gp70, but not autoimmune responses. Accordingly, the association of gp70 IC with this locus in BC mice can be a consequence of increased production of serum gp70 in liver, which is the major organ to synthesize and secrete gp70 into sera (33). It is worth noting that the Gv1 (gross virus Ag 1) gene that overlaps with the Sgp3 locus controls the expression of thymic GIX gp70 Ag (34, 35), the expression of which is closely correlated to serum levels of gp70 (36, 37). As Gv1 most likely regulates in trans the expression of multiple endogenous retroviral transcripts in different tissues, including the liver (38), it is reasonable to assume that Gv1 and Sgp3 are identical or related genes regulating the expression of endogenous retrovirus.
Our present BC study also revealed a QTL of B6 origin on chromosome 17, which showed a strong linkage with anti-DNA production and gp70 IC formation. This locus overlapped with the H2 complex, and the development of lupus-like disease in our BC mice was strongly associated with H2b/b homozygosity (vs H2b/d heterozygosity, with H2b and H2d of B6 and NZB origin, respectively). We and others have previously shown that lupus susceptibility is more closely linked to the H2b haplotype than to the H2d haplotype in BXSB, (NZB x BXSB)F1, and B6-Faslpr mice (17, 39, 40). The autoimmune inhibitory effect of H2d may be in part related to the expression of I-E molecules, because H2b mice fail to express I-E due to the deletion of the promoter region of the Ea gene encoding the I-E -chain (41). It has been previously shown that the introduction of two copies of the functional Ea transgene in lupus-prone BXSB mice (H2b) is sufficient to prevent the development of SLE (42). Although the precise mechanism of the protection of SLE by I-E has not fully been elucidated, our recent results support the hypothesis that formation of I-E -chain-derived peptides, which display a high affinity to I-A molecules, in B cells may decrease the use of I-A molecules for the presentation of pathogenic self peptides, thereby limiting the activation of autoreactive T and B cells (43).
Past genome-wide linkage analyses involving different lupus-prone mice have revealed >30 lupus-susceptibility loci in lupus-prone NZB, NZW, BXSB, and MRL mice (8, 9). However, it should be stressed that our studies in B6 x (NZW x B6.Yaa)F1 and B6 x (NZB x B6.Yaa)F1 BC show that a relatively small number and selective group of critical genes may be sufficient to cause full expression of SLE in the presence of the Yaa gene. Because both (NZW x B6)F1 and (NZB x B6)F1 mice lacking the Yaa gene are essentially normal, Yaa is capable of interacting with NZW and NZB lupus-susceptibility genes, which are by themselves not sufficient, to trigger lupus-like autoimmune responses. Moreover, the NZW- and NZB-derived genes involved in the development of SLE through interaction with Yaa are mostly different in each strain and only a small subset of those previously described. Notably, studies in B6 x (NZW x B6.Yaa)F1 BC mice identified a single locus on the NZW chromosome 7, which predisposes (NZW x B6)F1 mice to Yaa gene-induced SLE (11). However, this NZW locus is distinct from Nba5, as the former promotes overall autoantibody production, while the latter is selective on gp70 IC formation. In addition and in contrast to the NZB-derived Nba2, these past BC studies did not identify the important NZW lupus-susceptibility genes encoded on distal chromosome 1 (44).
A congenic dissection approach to convert a polygenic system into a series of monogenic systems has proved to be a powerful tool to dissect the complex genetic interactions implicated in murine SLE (13, 15, 25, 44, 45, 46). In addition, analysis of phenotypes in mono- and bicongenic strains carrying the susceptibility loci should help define key epistatic interactions in the development of SLE (47). The construction of further subcongenic B6.Yaa mice carrying the Nba2, Nba5, or Sgp3 locus should allow fine mapping and positional cloning of the gene(s) underlying the NZB predisposition to murine SLE. The eventual identification of mouse lupus-susceptibility genes will let us directly address the relevance of their human counterparts, and has obvious and promising implications for diagnostic, prognostic, and therapeutic approaches in SLE and related autoimmune diseases.
Acknowledgments
We thank Dr. L. Reininger for his critical reading of the manuscript; Drs. S. Hirose and T. Winkler for providing us information for microsatellite markers polymorphic between NZB and B6 mice; and G. Leyvraz, G. Brighouse, S. Jacquier, and G. Celetta for their excellent technical assistance.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This work was supported by a grant from the Swiss National Foundation for Scientific Research and Grant AR 37070 from the National Institutes of Health. L.F.-J. is a recipient of a fellowship from the Arthritis Research Campaign, U.K.
2 Address correspondence and reprint requests to Dr. Shozo Izui, Department of Pathology and Immunology, Centre Médical Universitaire, 1211 Geneva 4, Switzerland. E-mail address: Shozo.Izui@medecine.unige.ch
3 Abbreviations used in this paper: NZB, New Zealand Black; BC, backcross; GN, glomerulonephritis; Gv1, gross virus Ag 1; IC, immune complex; LOD, log-likelihood of the odds; Nba, NZB autoimmunity; NZW, New Zealand White; QTL, quantitative trait locus; Sgp3, serum gp70 production 3; SLE, systemic lupus erythematosus; Yaa, Y-linked autoimmune acceleration.
Received for publication September 27, 2004. Accepted for publication November 2, 2004.
References
Lambert, P. H., F. J. Dixon. 1968. Pathogenesis of the glomerulonephritis of NZB/W mice. J. Exp. Med. 127:507.
Izui, S., P. J. McConahey, A. N. Theofilopoulos, F. J. Dixon. 1979. Association of circulating retroviral gp70-anti-gp70 immune complexes with murine systemic lupus erythematosus. J. Exp. Med. 149:1099.
Maruyama, N., F. Furukawa, Y. Nakai, Y. Sasaki, K. Ohta, S. Ozaki, S. Hirose, T. Shirai. 1983. Genetic studies of autoimmunity in New Zealand mice. IV. Contribution of NZB and NZW genes to the spontaneous occurrence of retroviral gp70 immune complexes in (NZB x NZW)F1 hybrid and the correlation to renal disease. J. Immunol. 130:740.
Vyse, T. J., R. K. Halterman, S. J. Rozzo, S. Izui, B. L. Kotzin. 1999. Control of separate pathogenic autoantibody responses marks MHC gene contributions to murine lupus. Proc. Natl. Acad. Sci. USA 96:8098.
Yoshiki, T., R. C. Mellors, M. Strand, J. T. August. 1974. The viral envelope glycoprotein of murine leukemia virus and the pathogenesis of immune complex glomerulonephritis of New Zealand mice. J. Exp. Med. 140:1011.
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.
Haywood, M. E. K., T. J. Vyse, A. McDermott, E. M. Thompson, A. Ida, M. J. Walport, S. Izui, B. J. Morley. 2001. Autoantigen glycoprotein 70 expression is regulated by a single locus, which acts as a checkpoint for pathogenic anti-glycoprotein 70 autoantibody production and hence for the corresponding development of severe nephritis, in lupus-prone BXSB mice. J. Immunol. 167:1728
Vyse, T. J., B. L. Kotzin. 1998. Genetic susceptibility to systemic lupus erythematosus. Annu. Rev. Immunol. 16:261.
Wakeland, E. K., K. Liu, R. R. Graham, T. W. Behrens. 2001. Delineating the genetic basis of systemic lupus erythematosus. Immunity 15:397.
Izui, S., M. Iwamoto, L. Fossati, R. Merino, S. Takahashi, N. Ibnou-Zekri. 1995. The Yaa gene model of systemic lupus erythematosus. Immunol. Rev. 144:137.
Santiago, M. L., C. Mary, D. Parzy, C. Jacquet, X. Montagutelli, R. M. E. Parkhouse, R. Lemoine, S. Izui, L. Reininger. 1998. Linkage of a major quantitative trait locus to Yaa gene-induced lupus-like nephritis in (NZW x C57BL/6)F1 mice. Eur. J. Immunol. 28:4257.[Medline](Shuichi Kikuchi, Liliane )