Adjuvant-induced arthritis (AIA) in rats is a widely used autoimmune experimental model with many features similar to rheumatoid arthritis (RA). To identify potential genetic regulatory mechanisms in RA, we conducted genome-wide linkage analysis in F2 progeny of arthritis-susceptible Dark Agouti (DA) and relatively resistant Fischer 344 (F344) inbred rats. We compared the data with our previously reported investigation of collagen-induced arthritis (CIA), which was expanded in the follow-up study reported in this work. We found two quantitative trait loci (QTLs) in common, i.e., Aia1/Cia1 on chromosome 20, which includes the MHC, and Aia3/Cia3 on chromosome 4. We also identified a second unique QTL in AIA, Aia2, on chromosome 4. Interestingly, the QTL region on chromosome 4 (Aia3/Cia3), like the MHC, appears to be involved in several other autoimmune diseases in rats, including insulin-dependent diabetes, thyroiditis, and experimental autoimmune uveitis. Moreover, an analysis of conserved synteny among rats, mice, and humans suggested that Aia2 and Aia3/Cia3, like Aia1/Cia1, contain candidate genes for several autoimmune/inflammatory diseases in mice and humans, including diabetes, systemic lupus erythematosus, inflammatory bowel disease, asthma/atopy, multiple sclerosis, and RA. The rat models appear to provide a powerful complementary approach to identify and characterize candidate genes that may contribute to autoimmune diseases in several species.

Rheumatoid arthritis (RA)4 has a complex multifactorial etiopathogenesis, influenced by age, gender, hormonal, and environmental factors (1). Susceptibility to RA is also influenced by genetic factors, as indicated by many studies showing higher rates of disease concordance in monozygotic than dizygotic twins, higher incidence in offspring of RA patients, etc. (2, 3, 4). The genetic contribution to RA susceptibility is estimated to be as much as 60%, of which the HLA DRB1 locus is thought to account for 30 to 50% (5, 6). Identification of genetic loci regulating RA is complicated by genetic heterogeneity and incomplete penetrance, as well as the nongenetic factors noted above (7, 8). Genetic analysis of well-defined experimental models of erosive arthritis has the potential to markedly accelerate the genetic analysis of RA.

Experimental models of arthritis in rats, such as adjuvant-induced arthritis (AIA) (9, 10, 11, 12, 13, 14, 15, 16, 17) and collagen-induced arthritis (CIA) (18, 19, 20, 21, 22), have been used extensively in studying the roles of autoimmunity and inflammation in the pathogenesis of joint disease. Susceptibility/severity of disease varies significantly among inbred strains. For example, inbred Dark Agouti (DA) rats are markedly susceptible to CIA and AIA, as well as several other autoimmune diseases, whereas inbred Fischer 344 (F344) rats are relatively resistant (23, 24). Most studies indicate that both non-MHC genes and RT1 MHC class II genes are associated with the susceptibility to CIA and AIA in DA rats (25, 26, 27, 28). Recently, our groups, employing a genome-wide scan involving 98 polymorphic markers, localized five QTLs that regulate CIA in (DA × F344)F2 progeny. One QTL (Cia1) is located on chromosome 20 and includes the MHC. The other QTLs (Cia2-5) are located on chromosomes 1, 4, 7, and 10, respectively. In addition, a region on chromosome 8 is suggestive for linkage (29). We, therefore, were interested in determining whether AIA in (DA × F344)F2 progeny is regulated by the same loci.

To determine whether the QTLs previously identified in CIA were also operative in another autoimmune model of arthritis and to identify additional QTLs, we conducted a genome scan to identify QTLs that regulate AIA in (DA × F344)F2 progeny. We have also expanded our previous investigation of CIA (29), and compared the revised CIA analysis with our new AIA analysis. Although CIA and AIA have many clinical similarities, fundamental differences in their pathophysiology are clearly evident and provide an ideal opportunity for defining and comparing the genes that control arthritis susceptibility and severity. This comparison is particularly interesting because AIA predominantly involves T cell-mediated mechanisms, whereas CIA requires both humoral and cellular immunity (11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23). Identification of one or more loci common to both models would suggest the existence of genes that regulate the development of autoimmune inflammatory arthritis in response to diverse stimuli. We report the identification of a non-MHC locus on chromosome 4, in addition to the MHC, that regulates both AIA and CIA. We also discuss evidence that these loci may also be involved in multiple additional autoimmune diseases in rats, mice, and humans.

Specific pathogen-free Fischer 344 (F344/NHsd) rats were obtained from Harlan Sprague-Dawley (Indianapolis, IN), and Dark Agouti (DA/Bkl) rats from Bantin & Kingman (Fremont, CA). Using the convention of (female × male) to indicate strain parentage, F1 progeny were generated at the Veterinary Medical Unit, Salt Lake City VAMC from four mating pairs: two (F344 × DA) and two (DA × F344). F2 progeny were subsequently generated using eight mating cages (sib × sib) established with the F1 progeny of one (F344 × DA) mating pair and one (DA × F344) mating pair. The progeny derived from (DA × F344) and (F344 × DA) matings were evaluated for arthritis incidence and severity separately, but no statistically significant differences were detected. Analyses with the combined data are described.

For AIA, CFA was prepared by suspending heat-killed Mycobacterium butyricum (Difco, Detroit, MI) in incomplete Freund’s adjuvant (IFA) at 7.5 mg/ml. CFA-induced arthritis was stimulated by injection of 0.1 ml of the CFA emulsion intradermally at the base of the tail, as previously described (16, 17).

For CIA, the rats were injected intradermally on the back with 2 mg/kg bovine type II collagen, prepared as previously described (18, 19, 20, 21, 22). In brief, collagen was suspended in 0.1 M acetic acid at 1 mg/ml and emulsified in an equal volume in IFA (Difco). Seven days after the initial injection, a booster injection of 100 μg bovine type II collagen emulsified in IFA was injected intradermally at the base of the tail.

For clinical evaluation of AIA and CIA, we used a scoring system as follows. The severity of arthritis in the wrist, midforepaw, ankle, and midfoot was scored for each extremity using the following scale: 0, no arthritis; 1, minimal swelling; 2, medium swelling; 3, severe swelling; and 4, severe swelling and non-weight bearing.

The presence of arthritis in the three joints of each of the lateral four digits, counting from lateral to medial (two interphalangeal, and metatarsophalangeal or metacarpophalangeal), was scored as: 0, swelling absent; 1, swelling present.

The total score for each extremity was calculated by summing the scores of the individual joints of each extremity with arthritis. The maximum score for each extremity was, therefore, 20, and the maximum total joint score was 80.

Scores were obtained two times per week by a single observer unaware of the genetic constitution of the animals. The maximum arthritis score for each animal was determined as the highest score during the 6-wk observation period. These scores are highly correlated with histologic or radiologic evaluations of disease severity. All experimental groups contained control rats of both parental strains. Scores in the text are indicated by mean ± SD in the text.

DNA was prepared by standard procedures (30), and genotypes were determined by PCR amplification of polymorphic DNA fragments containing simple sequence repeats. For the genome-wide QTL analysis, 150 markers were used to genotype animals from the phenotypic extremes (severely affected versus mildly affected animals) of the F2 progeny (Table I). The markers were selected to give a spacing interval of less than 20 cM. The map of these markers covered 1535 cM, which is estimated to be more than 95% of the rat genome. The sequences of the PCR primers and amplification protocols are described at ARB Rat Genetic Database (31), Wellcome Trust/Oxford University Rat Genetic Database (32), and Whitehead Institute/MIT Rat Genetic Database (33). Primers for most of these markers are available from Research Genetics (Huntsville, AL). Linkage maps were generated using the MAPMAKER/EXP version 3.0b (34, 35).

Table I.

Associations of markers with AIA in the (F344 × DA)F2 cross, experiment 1 (n = 546)

          
          
a

Distances between loci are described in centiMorgans (Kosambi).

b

F344 indicates the number of F344 alleles.

c

DA indicates the number of DA alleles.

d

P-values were calculated by a χ2 test of independence using a 2 × 2 contingency table.

e

Coat phenotype

Nonparametric analysis of disease susceptibility and severity was evaluated with the Mann-Whitney test. Linkage of individual simple sequence length polymorphism (SSLP) markers to disease susceptibility/severity was evaluated by a χ2 test of independence using a 2 × 2 contingency table comparing allele frequencies in the high and low arthritis score animals (36), i.e., homozygous DA progeny contributed two DA alleles to the allele frequency, homozygous F344 progeny contributed two F344 alleles, and heterozygous progeny contributed one DA and one F344 allele. Significant cosegregation of marker alleles with a QTL, which controlled arthritis severity, and its mode of inheritance were also evaluated with the MAPMAKER/QTL program, version 1.1b (37). We used the arcsin square root as a variance stabilizing transformation for normalizing the phenotypic distribution (38). As recently suggested (39), the threshold for suggestive linkage is a LOD score of 2 corresponding to a p value of 2.4 × 10−3, and the threshold for significant linkage is a LOD score of 3.4 corresponding to a p value of 7.2 × 10−4 in dominant inheritance mode. In free inheritance mode, the threshold for suggestive linkage is a LOD score of 2.8 corresponding to a p value of 1.6 × 10−3, and the threshold for significant linkage is a LOD score of 4.3 corresponding to a p value of 5.2 × 10−5.

The rat map was constructed from data obtained in this study. Locations of additional genes on this map were estimated from data in ARB Rat Genetic Database (31). The mouse map and the human cytogenetic map were extracted from data in Mouse Genome Database (40) and Genome Database (41), respectively. The most likely positions of various autoimmune disease susceptibility loci were deduced from our new data and review of the literature (42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53).

As shown in Figure 1, we compared the incidence and severity of CIA (as previously reported, 29) with AIA in parental DA, parental F344, F1, and F2 rats during 6 wk following immunization. DA rats were extremely susceptible to induction of severe CIA and AIA, exhibiting maximum arthritis scores (see Materials and Methods) of 62 ± 10 (mean ± SD) for CIA and 60 ± 22 for AIA. However, F344 rats were highly resistant to CIA and moderately resistant to AIA (mean maximum score: CIA = 0; AIA = 8 ± 6). The incidence of disease in F1 progeny in both models was 100%, but the mean severity was different in the F1 progeny than in the two parental strains (mean maximum arthritis score of F1 progeny: CIA = 40 ± 14; AIA = 50 ± 4). The F2 population showed a wide range of maximum arthritis scores: CIA (0–73, n = 502), AIA (0–78, n = 546). But, 23 and 5% of the F2 population failed to develop CIA and AIA, respectively.

FIGURE 1.

Arthritis score distribution in parental, F1, and F2 hybrid rats. The data are shown as histograms for CIA (a, c, e, g) and AIA (b, d, f, h). Parental DA rats (a, b), parental F344 rats (c, d), (DA × F344)F1 hybrid rats (e, f), and (DA × F344)F2 hybrid rats (g, h). The CIA and AIA F2 populations include 502 and 546 progeny, respectively.

FIGURE 1.

Arthritis score distribution in parental, F1, and F2 hybrid rats. The data are shown as histograms for CIA (a, c, e, g) and AIA (b, d, f, h). Parental DA rats (a, b), parental F344 rats (c, d), (DA × F344)F1 hybrid rats (e, f), and (DA × F344)F2 hybrid rats (g, h). The CIA and AIA F2 populations include 502 and 546 progeny, respectively.

Close modal

We have previously reported highly significant cosegregation of CIA susceptibility/severity with several MHC markers, including Tnfa (a marker for TNF: Cia1, LOD score = 78.5), which is located within the class III region of the rat MHC (29). For our new study of AIA, we genotyped 80 severely (arthritis index >64) and 80 mildly affected (arthritis index <17) rats from the (DA × F344)F2 progeny (n = 546, Expt. 1) with SSLP markers spanning rat chromosome 20 (Table I). Cosegregation of genotype with phenotype was noted over a 30-cM interval on this chromosome, with the strongest association between markers D20Arb8 and D20Arb2 (RT1.N1) (Table I). This interval includes MHC class I, II, and III genes. Interval analysis using MAPMAKER/QTL confirmed a strong QTL in this region (LOD score = 17.9 at Tnfa). Its inheritance was consistent with dominant mode, i.e., the DA allele acted dominantly to promote disease severity. We have named this locus Aia1. To facilitate comparison of AIA with CIA, we extended our previous analysis of CIA (29) by analyzing the same SSLP markers spanning rat chromosome 20 that we used for the AIA analysis. We cannot separate Aia1 from Cia1 (Fig. 2, a and b).

FIGURE 2.

Log-likelihood plots for CIA and AIA QTLs on chromosomes 20 and 4 (n = 656). Log-likelihood values for the presence of a QTL were determined by an interval mapping technique (MAPMAKER/QTL). The most likely position for each QTL, determined by its two LOD support interval (99% confidence interval), is indicated by the closed bar above the plot. a, Cia1 on chromosome 20; b, Aia1 on chromosome 20; c, Cia3 on chromosome 4; d, Aia2 and Aia3 on chromosome 4.

FIGURE 2.

Log-likelihood plots for CIA and AIA QTLs on chromosomes 20 and 4 (n = 656). Log-likelihood values for the presence of a QTL were determined by an interval mapping technique (MAPMAKER/QTL). The most likely position for each QTL, determined by its two LOD support interval (99% confidence interval), is indicated by the closed bar above the plot. a, Cia1 on chromosome 20; b, Aia1 on chromosome 20; c, Cia3 on chromosome 4; d, Aia2 and Aia3 on chromosome 4.

Close modal

To evaluate the effects of the MHC on susceptibility and severity in both AIA and CIA, the variance in arthritis score at the Tnfa locus was examined. Because the entire MHC region is within a 3-cM genetic interval (54), recombination within the MHC is infrequent. Therefore, the Tnfa-α genotype closely reflects the genotype of the entire MHC region. The MHC region influenced susceptibility differently in these two models. For CIA, in (DA × F344)F2 progeny, the MHC genotype had a relatively strong effect on disease susceptibility. For CIA, 43.4, 94.6, and 93.7% of rats with 0, 1, or 2 DA alleles at the Tnfa locus developed arthritis (Fig. 3,a, c, and e). For AIA, however, the MHC genotype had much weaker effects on disease susceptibility, i.e., 90.4, 96.2, and 96.8% of rats with 0, 1, or 2 DA alleles at the Tnfa locus developed arthritis (Fig. 3, b, d, and f). On the other hand, MHC genotype clearly correlated with disease severity in both models. For CIA, the mean severity scores of arthritis-affected rats with 0, 1, and 2 DA alleles at the Tnfa locus were 12 ± 15, 35 ± 16, and 42 ± 16, respectively. For AIA, the mean severity scores of arthritis-affected rats with 0, 1, and 2 DA alleles at the Tnfa locus were 34 ± 20, 48 ± 16, and 50 ± 17, respectively. In F2 rats with 0 DA alleles, the susceptibility and severity in AIA were significantly greater than those of CIA (Mann-Whitney test: p < 0.01). In other words, F2 progeny with 0 DA MHC alleles typically developed mild disease, but the range of arthritis scores was wide in AIA and was more restricted in CIA than in AIA. Furthermore, F2 progeny with 1 or 2 DA MHC alleles developed more severe disease than progeny that were homozygous for the F344 MHC alleles in both AIA and CIA, but the range of arthritis scores was wide. In summary, the MHC genotype clearly influenced disease expression for both AIA and CIA, but this genotype information alone for any individual F2 rat was an imprecise prognostic factor for determining disease susceptibility or severity.

FIGURE 3.

The effect of DA MHC alleles on CIA and AIA severity. The data are shown as histograms for CIA (a, c, e) and AIA (b, d, f). (DA × F344)F2 hybrid rats with 0 Tnfa DA alleles (a, b), 1 Tnfa DA allele (c, d), and 2 Tnfa DA alleles (e, f). The CIA and AIA F2 populations include 502 and 546 progeny, respectively.

FIGURE 3.

The effect of DA MHC alleles on CIA and AIA severity. The data are shown as histograms for CIA (a, c, e) and AIA (b, d, f). (DA × F344)F2 hybrid rats with 0 Tnfa DA alleles (a, b), 1 Tnfa DA allele (c, d), and 2 Tnfa DA alleles (e, f). The CIA and AIA F2 populations include 502 and 546 progeny, respectively.

Close modal

To complete the genome scan of AIA, we examined 80 severely and 80 mildly affected F2 rats with AIA for chromosomes 1, 4, 7, 8, and 10, which contain Cia2–5, as well as a suggestive locus on chromosome 8. Maximum arthritis scores were greater than 64 in 80 severely affected rats and were lower than 17 in 80 mildly affected rats. For the remainder of the genome scan, we evaluated 40 severely (arthritis index >66) and 40 mildly affected (arthritis index <6) F2 rats. For markers (e.g., chromosomes 3, 6, 15, and 19) that showed trends for nonrandom genotype distributions (p < 0.05) after genotypic analysis of 40 severely and 40 mildly affected F2 rats, additional genotyping was done to include the 80 severely and 80 mildly affected F2 rats. We found evidence of genotype-phenotype cosegregation with multiple chromosome 4 SSLP markers (Table I). The strongest association was with chromosome 4 markers between D4Arb26 and D4Arb16 (Ampp) (D4 Mgh15, p = 2.4 × 10−3) (Table I). Since this significance level, by definition (39), is suggestive for linkage, we conducted a replication experiment with a second group of (DA × F344)F2 progeny (n = 110, Expt. 2). In this replication experiment, we genotyped severely affected (n = 28, arthritis index >64) and mildly affected (n = 24, arthritis index <17) rats that had arthritis scores equivalent to those in experiment 1. We again detected cosegregation with chromosome 4 markers, but the strongest association was with D4Arb24 (p = 4 × 10−4) (Table II). Because the chromosomal segment associated with AIA extended over multiple markers and an interval of approximately 50 cM, we suspected that this chromosome contained more than 1 QTL. We, thus, further analyzed our combined data from experiment 1 and 2 (n = 656) for chromosome 4 using MAPMAKER/QTL (Fig. 2,d). An interval between D4Arb26 and D4Arb30 showed significant linkage with a LOD score of 5.8 in the dominant inheritance mode. We have named this locus Aia2. Aia2 is a unique locus for AIA because additional genotyping with markers spanning the genomic interval containing Aia2 of the phenotypically highest and lowest 15% of the CIA F2 population with at least 1 DA MHC allele did not reveal a corresponding CIA regulatory locus (Fig. 2, c and d). A second interval between D4Arb30 and D4Arb16 had a maximum LOD score of 3.9 in free inheritance mode. Since this QTL was detected in both experiments 1 and 2 (Tables I and 2), we have defined this locus as Aia3. Aia3, like Cia3, was strongly linked to D4Arb24 (Fig. 2, c and d), suggesting that this region of chromosome 4 contains a QTL that regulates both AIA and CIA. Both Aia3 and Cia3 appear to act additively.

Table II.

Associations of markers on chromosome 4 with AIA in the (F344 × DA)F2 cross (replication study), experiment 2 (n = 110)

Chr MarkersGene SymbolDistancea (cM)Severely affectedMildly affectedχ2p valued
GenotypeGenotype
F344bDAcF344bDAc
D4Arb13  12.1 17 11 16 2.9 0.09 
D4Arb18  11.4 15 13 15 0.136 0.94 
D4Arb26  11.7 13 15 17 2.23 0.135 
D4Mgh15  2.6 12 16 18 4.2 0.04 
D4Arb11  6.2 12 16 18 4.2 0.04 
D4Arb30  12.9 11 17 18 5.3 0.021 
D4Arb21 Fabp1 1.3 12 16 20 7.3 0.0068 
D4Arb24  13.5 10 18 21 12.3 0.0004 
D4Arb16 Ampp 13.4 19 18 7.9 0.005 
D4Arb4 Caln1a1 11.5 20 16 6.1 0.014 
D4Arb2 Prp 11.7 14 14 10 14 0.1 0.7 
D4Arb1 lapp — 13 15 11 13 0.06 0.8 
Chr MarkersGene SymbolDistancea (cM)Severely affectedMildly affectedχ2p valued
GenotypeGenotype
F344bDAcF344bDAc
D4Arb13  12.1 17 11 16 2.9 0.09 
D4Arb18  11.4 15 13 15 0.136 0.94 
D4Arb26  11.7 13 15 17 2.23 0.135 
D4Mgh15  2.6 12 16 18 4.2 0.04 
D4Arb11  6.2 12 16 18 4.2 0.04 
D4Arb30  12.9 11 17 18 5.3 0.021 
D4Arb21 Fabp1 1.3 12 16 20 7.3 0.0068 
D4Arb24  13.5 10 18 21 12.3 0.0004 
D4Arb16 Ampp 13.4 19 18 7.9 0.005 
D4Arb4 Caln1a1 11.5 20 16 6.1 0.014 
D4Arb2 Prp 11.7 14 14 10 14 0.1 0.7 
D4Arb1 lapp — 13 15 11 13 0.06 0.8 
a

Distances between loci are described in centiMorgans (Kosambi).

b

F344 indicates the number of F344 alleles.

c

DA indicates the number of DA alleles.

d

p values were calculated by a χ2 test of independence using a 2 × 2 contingency table.

Our analysis of AIA also suggested phenotype:genotype cosegregation (p < 0.05, Table I) with one SSLP marker on chromosome 3, and two markers on chromsome 15. The genomic interval on chromosome 15 containing these markers had a maximum LOD score of 3 in the free inheritance mode, while the chromosome 3 interval had a LOD score of 2.7. Thus, the genomic interval on chromosome 15, but not the Chr3 interval, meets criteria for suggestive linkage (39). Interestingly, the putative disease-enhancing allele on chromosome 15 is derived from the F344 strain.

Since our previously reported genomic scan analysis of CIA involved only 98 SSLP markers and incompletely spanned the rat genome (29), all markers not previously evaluated in CIA, which were evaluated in the study of AIA, were also analyzed in CIA F2 progeny to insure that the genomic scans of AIA and CIA were comparable. In other words, 150 markers, which we estimate cover more than 95% of the rat genome, were analyzed for both CIA and AIA. No additional CIA QTLs, however, were identified with this additional analysis. In particular, markers on chromosomes 3 and 15 showed no evidence for linkage in CIA (data not shown).

To examine homologies between mouse and human chromosomal regions and the chromosomal regions associated with Aia2 and Aia3/Cia3 on rat chromosome 4, we constructed comparative maps of these three species. The region including Aia2 and Aia3/Cia3 on rat chromosome 4 was homologous to mouse chromosome 6 and human chromosomes 7p, 7q, 2p, 3p, 3q, 10q, and 12p (Fig. 4).

FIGURE 4.

Comparative mapping of genomic regions containing Aia2, Aia3, and Cia3 among rats, mice, and humans. The rat map is a composite map constructed from data obtained in this study and additional gene locations estimated from ARB Rat Genetic Database (31). The mouse map is based on Mouse Genome Database (40). Distances (in centimorgans) from centromere for the mouse loci are indicated. The human cytogenetic map is based on Genome Database (41). The bars indicate the most likely position for each autoimmune disease-susceptibility locus, as follows: Aia, adjuvant-induced arthritis in rat; AS, bronchial asthma (42); Bphs, Bordetella pertussis-induced histamine sensitization (43); Cia, collagen-induced arthritis (29); CD, Crohn’s disease (44); Eau, experimental autoimmune uveitis (45); Iddm, insulin-dependent diabetes mellitis (46–49); Lbw, systemic lupus erythematosus (53); Lyp, lymphopenia (47); MS, multiple sclerosis (50–52); RA, rheumatoid arthritis (46); and UC, ulcerative colitis (44).

FIGURE 4.

Comparative mapping of genomic regions containing Aia2, Aia3, and Cia3 among rats, mice, and humans. The rat map is a composite map constructed from data obtained in this study and additional gene locations estimated from ARB Rat Genetic Database (31). The mouse map is based on Mouse Genome Database (40). Distances (in centimorgans) from centromere for the mouse loci are indicated. The human cytogenetic map is based on Genome Database (41). The bars indicate the most likely position for each autoimmune disease-susceptibility locus, as follows: Aia, adjuvant-induced arthritis in rat; AS, bronchial asthma (42); Bphs, Bordetella pertussis-induced histamine sensitization (43); Cia, collagen-induced arthritis (29); CD, Crohn’s disease (44); Eau, experimental autoimmune uveitis (45); Iddm, insulin-dependent diabetes mellitis (46–49); Lbw, systemic lupus erythematosus (53); Lyp, lymphopenia (47); MS, multiple sclerosis (50–52); RA, rheumatoid arthritis (46); and UC, ulcerative colitis (44).

Close modal

This study was intended to compare genetic loci regulating susceptibility/severity in two experimental autoimmune arthritis models in which the strain susceptibility patterns are similar, but the pathogenetic mechanisms appear to differ. Using genome-wide linkage analytic techniques, including 150 SSLP markers, we identified both MHC (Aia1) and non-MHC (Aia2, Aia3) loci that regulated AIA in F2 progeny of DA and F344 strains. We also found an interval on chromosome 15, which was suggestive for linkage. Interestingly, the putative disease-enhancing QTL on chromosome 15 was derived from the F344 genetic background. We compared these data with our previously reported investigation of CIA (29), which was also extended in the current study to 150 SSLP markers (98 markers were examined in the earlier report), to facilitate comparison with AIA. Aia2 on chromosome 4 and the suggestive locus on chromosome 15 appear to be unique to AIA. Aia1 and Aia3, however, mapped to the same locations on chromosomes 20 and 4, respectively, that Cia1 and Cia3 were previously mapped (29). The chromosomal regions containing Cia2, Cia4, Cia5, and Cia6* (∗, region on chromosome 8 associated with suggestive linkage) do not appear to be involved in regulating AIA.

Although the Ags that initiate RA remain elusive, several subtypes of HLA DR4, including the shared epitope in many ethnic groups, are associated with RA (55, 56). The effect of HLA class II genes on the expression of RA is controversial. Some studies support an association with susceptibility and others with severity (55, 56, 57, 58, 59). We examined the effects of the MHC on susceptibility and severity in both AIA and CIA. Our study shows that the MHC region influences susceptibility and severity differently in these two models. For CIA, in (DA × F344)F2 progeny, MHC genotype had a relatively strong effect on disease susceptibility. For AIA, however, MHC genotype had a much weaker effect on disease susceptibility. Moreover, MHC genotype clearly exerted strong regulatory effects on disease severity in both models. But, in F2 rats with 0 DA alleles, CIA severity was milder than that of AIA. In both models, we noted many rats that were F344 homozygous at the MHC locus developed arthritis, and, conversely, many rats that were DA homozygous at the MHC locus failed to develop arthritis. In other words, our data demonstrate that non-MHC genotypes also contributed significantly to determining susceptibility and severity. Other studies have also indicated that DA non-MHC genes enhance arthritis susceptibility (25, 26, 27, 28, 29).

Among the non-MHC QTLs, Aia3 and Cia3 are especially interesting because they map to overlapping genomic regions and may be the same locus. Moreover, these loci are potentially of more general interest. Iddm1 and Lyp (lymphopenia), in the progeny of diabetic BB rats, cosegregate with this genomic region (47). The Lyp/Iddm1 locus is also associated with the risk of development of autoimmune thyroiditis, independent of diabetes (60). It is also interesting that RA, diabetes, and thyroid disease cluster in families of some RA patients (61, 62). Aia3/Cia3 also appear to have a homologue in experimental autoimmune uveitis in the rat. In (F344 × LEW)F2 rats, a locus very close to Aia3/Cia3 on chromosome 4 has been identified (45). It remains to be definitively demonstrated whether or not Aia3, Cia3, Iddm1, Lyp, and the chromosome 4 QTL in experimental autoimmune uveitis are allelic, but in any event, rat chromosome 4, in addition to the MHC, appears to contain important genes that regulate several forms of autoimmune disease.

Since the gene map of rat chromosome 4 is still relatively sparse, we generated comparative maps among rats, mice, and humans, to examine the possibility of identifying overlaps with autoimmune diseases in other species and to identify candidate genes (Fig. 4). Rat chromosome 4 and mouse chromosome 6 show strong linkage conservation. Interestingly, the region homologous to Aia3 in the rat includes loci on mouse chromosome 6 controlling Ab production (63) and Bordetella pertussis histamine sensitization (43), which cosegregates with susceptibility to experimental autoimmune orchitis (64). In addition, 2 loci regulating diabetes in nonobese diabetic mice (D6Nds1/Idd and D6Mit52/Idd), and a locus regulating systemic lupus in mice (D6Mit25/Lbw4) have been mapped to this region (48, 49, 53). The human homologues of the rat Aia2 and Aia3 regions are located on chromosomes 2, 3, 7, 10, and 12 (Fig. 4). Of significant interest, these homologous regions contain candidate genes for many other autoimmune/inflammatory diseases in humans, such as multiple sclerosis (50, 51, 52), inflammatory bowel disease (44), and atopy/bronchial asthma (42). Moreover, the Aia3 homologous region is also linked to recently described candidate loci on human chromosome 3q for rheumatoid arthritis (RA2) and insulin-dependent diabetes mellitis (IDDM9) (46). Human chromosome 12 may also have a RA candidate locus (46). Numerous candidate genes are predicted to map in this region of rat chromosome 4 and include neuropeptide Y, TCR Vβ loci, the κ-light chain of Ig, CD8, max dimerization protein, TGF-α, IL-5R α-chain, histamine receptor H1, TNF receptor 1, CD27, CD4, CD9, and the CD69 Ag (Fig. 4).

Thus, our data raise the possibility that like Aia1/Cia1, both Aia2 and Aia3/Cia3 and their homologues in other species will be important in many forms of autoimmune disease. In humans, autoimmune disease frequency is increased significantly in first degree relatives of patients with idiopathic inflammatory myopathies (65). A recent review of published mapping studies also suggested that autoimmune disease regulatory loci tend to cluster within and among species (66). In other words, it appears increasingly likely that many autoimmune disorders share genes that together act as polygenic risk factors.

In summary, our data indicate that at least three genetic loci contribute to disease development in AIA. Their homologues may also be involved in multiple autoimmune diseases in humans, including RA. Aia1 and Aia3 are likely to be identical to Cia1 and Cia3, respectively, and probably influence multiple additional autoimmune diseases. The development of QTL congenic strains involving these loci will provide a powerful approach to identify and characterize the biologic function of the underlying genes. Moreover, our data suggest the possibility of ultimately developing therapies directed at biochemical pathways common to multiple related autoimmune diseases.

We are appreciative of the excellent technical assistance of Shawna McCall.

1

This work was partially supported by funds from the Department of Veterans Affairs Medical Research, the Nora Eccles Treadwell Foundation, and the Arthritis Foundation.

4

Abbreviations used in this paper: RA, rheumatoid arthritis; AIA, adjuvant-induced arthritis; CIA; collagen-induced arthritis; DA, Dark Agouti rat; F344, Fischer 344 rat; LOD, logarithm of the likelihood ratios; QTLs, quantitative trait loci; SSLP, simple sequence length polymorphism.

1
Feldmann, M., F. M. Brennan, R. N. Maini.
1996
. Rheumatoid arthritis.
Cell
85
:
307
2
Aho, K., M. Koskenvuo, J. Tuominen, J. Kaprio.
1986
. Occurrence of rheumatoid arthritis in a nationwide series of twins.
J. Rheumatol.
13
:
899
3
Silman, A. J., A. J. MacGregor, W. Thomson, S. Holligan, D. Carthy, A. Farhan, W. E. R. Ollier.
1993
. Twin concordance rates for rheumatoid arthritis: results from a nationwide study.
Br. J. Rheumatol.
32
:
903
4
Worthington, J., W. E. R. Ollier, M. K. Leach, I. Smith, E. M. Hay, W. Thomson, L. Pepper, D. Carthy, A. Farhan, S. Martin, P. Dyer, J. Davison, S. Bamber, A. J. Silman.
1994
. The Arthritis and Rheumatism Council’s national repository of family material: pedigrees from the first 100 rheumatoid arthritis families containing affected sibling pairs.
Br. J. Rheumatol.
33
:
970
5
Ollier, W., W. Thomson.
1992
. Population genetics of rheumatoid arthritis.
Rheum. Dis. Clin. North Am.
18
:
741
6
Ollier, W. E. R., A. MacGregor.
1995
. Genetic epidemiology of rheumatoid disease.
Br. Med. Bull.
51
:
267
7
Vyse, T. J., J. A. Todd.
1996
. Genetic analysis of autoimmune disease.
Cell
85
:
311
8
Theofilopoulos, A. N..
1995
. The basis of autoimmunity. II. Genetic predisposition.
Imunnol. Today
16
:
150
9
Stoerk, H. C., T. C. Bielinski, T. Budzilovich.
1954
. Chronic polyarthritis in rats injected with spleen in adjuvants.
Am. J. Pathol.
30
:
616
(Abstr.).
10
Pearson, C. M..
1956
. Development of arthritis, periarthritis and periostitis in rats given adjuvant.
Proc. Soc. Exp. Biol. Med.
91
:
95
11
Cohen, I. R., J. Holoshitz, W. van Eden, A. Frenkel.
1985
. T lymphocyte clones illuminate pathogenesis and affect therapy of experimental arthritis.
Arthritis Rheum.
28
:
841
12
Whitehouse, M. W..
1988
. Adjuvant induced polyarthritis in rats. R. A. Greenwald, and H. S. Diamond, eds.
Handbook of Animal Models for Rheumatic Disease
3
CRC Press, Boca Raton.
13
Van Eden, W., J. Holoshitz, Z. Nevo, A. Frenkel, A. Klajman, I. R. Cohen.
1985
. Arthritis induced by a T-lymphocyte clone that responds to Mycobacterium tuberculosis and to cartilage proteoglycan.
Proc. Natl. Acad. Sci. USA
82
:
5117
14
Johnson, W. J., K. A. Muirhead, P. C. Meunier, B. J. Votta, T. C. Schmitt, M. J. DiMartino, N. Hanna.
1986
. Macrophage activation in rat models of inflammation and arthritis: systemic activation precedes arthritis induction and progression.
Arthritis Rheum.
29
:
1122
15
Van Eden, W., J. E. R. Thole, R. van der Zee, A. Noordzij, J. D. A. van Embden, E. J. Henson, I. R. Cohen.
1988
. Cloning of the mycobacterial epitope recognized by T lymphocytes in adjuvant arthritis.
Nature
331
:
171
16
Cannon, G. W., D. S. Harper, F. Clayton, M. M. Griffiths.
1993
. Passive transfer of adjuvant-induced arthritis into irradiated DA recipient rats.
Autoimmunity
15
:
267
17
Cannon, G. W., S. J. Openshaw, J. B. Hibbs, Jr, J. R. Hoidal, T. P. Huecksteadt, M. M. Griffiths.
1996
. Nitric oxide production during adjuvant-induced and collagen-induced arthritis.
Arthritis Rheum.
39
:
1677
18
Trentham, D. E., A. S. Townes, A. H. Kang.
1977
. Autoimmunity to type II collagen: an experimental model of arthritis.
J. Exp. Med.
146
:
857
19
Griffiths, M. M..
1988
. Immunogenetics of collagen-induced arthritis in rats.
Int. Rev. Immunol.
4
:
1
20
Crofford, L. J., R. L. Wilder.
1997
. Arthritis and autoimmunity in animals. W. J. Koopman, Jr, ed. In
Arthritis and Allied Conditions
Vol. 13
:
565
Williams & Wilkins, Baltimore.
21
Wooley, P. H., J. M. Chapedelaine.
1987
. Immunogenetics of collagen-induced arthritis.
CRC Crit. Rev. Immunol.
8
:
1
22
Wooley, P. H..
1991
. Animal models of rheumatoid arthritis.
Curr. Opin. Rheumatol.
3
:
407
23
Griffiths, M. M., C. W. DeWitt.
1984
. Genetic control of collagen-induced arthritis in rats: the immune response to type II collagen among susceptible and resistant strains and evidence for multiple gene control.
J. Immunol.
132
:
2830
24
Battisto, J. R., R. N. Smith, K. Beckman, M. Sternlicht, W. L. Welles.
1982
. Susceptibility to adjuvant arthritis in DA and F344 rats.
Arthritis Rheum.
25
:
1194
25
Lorentzen, J. C., L. Klareskog.
1996
. Susceptibility of DA rats to arthritis induced with adjuvant oil or rat collagen is determined by genes both within and outside the major histocompatibility complex.
Scand. J. Immunol.
44
:
592
26
Griffiths, M. M., C. W. DeWitt.
1981
. Immunogenetic control of experimental collagen-induced arthritis in rats. II. ECIA susceptibility and immune response to type II collagen (CALF) are linked to RT1.
J. Immunogenet.
8
:
463
27
Griffiths, M. M., M. A. Cremer, D. S. Harper, S. McCall, G. W. Cannon.
1992
. Immunogenetics of collagen-induced arthritis in rats: both MHC and non-MHC gene products determine the epitope specificity of immune response to bovine and chick type II collagens.
J. Immunol.
149
:
309
28
Griffiths, M. M., G. W. Cannon, P. A. Leonard, V. R. Reese.
1993
. Induction of autoimmune arthritis in rats by immunization with homologous rat type II collagen is restricted to the RT1av1 haplotype.
Arthritis Rheum.
36
:
254
29
Remmers, E. F., R. E. Longman, Y. Du, A. O’Hare, G. W. Cannon, M. M. Griffiths, R. L. Wilder.
1996
. A genome scan localizes five non-MHC loci controlling collagen-induced arthritis in rats.
Nat. Genet.
14
:
82
30
Ausubel, F. M., R. Brent, R. E. Kingston, D. D. Moore, J. G. Seidman, J. A. Smith, K. Struhl.
1994
.
Current Protocols in Molecular Biology
John Wiley and Sons, New York.
31
In The ARB Rat Genetic Database. Available at http://www.nih.gov/niams/scientific/ratgbase/index.htm.
32
In Wellcome Trust/Oxford University Rat Genetic Database. Available at http://www.well.ox.ac.uk/∼bihoreau/key.html.
33
In The Whitehead Institute/MIT Rat Genetic Database. Available at http://www.genome.wi.mit.edu/rat/public.
34
Lander, E. S., P. Green, J. Abrahamson, A. Barlow, M. J. Daly, S. E. Lincoln, L. Newburg.
1987
. MAPMAKER: an interactive computer package for constructing primary genetic linkage maps of experimental and natural populations.
Genomics
1
:
174
35
Lincoln, S. E., E. S. Lander.
1992
. Systematic detection of errors in genetic linkage data.
Genomics
14
:
604
36
Suzuki, D. T., A. J. F. Griffith, J. H. Miller, R. C. Lewontin.
1986
. In
An Introduction to Genetic Analysis
Vol. 4
: W. H. Freeman and Company, New York.
37
Lander, E. S., D. Botstein.
1989
. Mapping mendelian factors underlying quantitative traits using RFLP maps.
Genetics
121
:
185
38
Fox, E., J. Kuo, L. Tilling, C. Ulich.
1994
.
SigmaStat: User’s Manual, Revision SSW 1.0
12
Jandel Scientific Software, San Rafael.
39
Lander, E., L. Kruglyak.
1995
. Genetic dissection of complex traits: guidelines for interpreting and reporting linkage results.
Nat. Genet.
11
:
241
40
In The Mouse Genome Database. Available at http://www.informatics.jax.org/mgd.html.
41
In The Genome Database. Available at http://gdbwww.gdb.org/.
42
Daniels, S. E., S. Bhattacharrya, A. James, N. I. Leaves, A. Young, M. R. Hill, J. A. Faux, G. F. Ryan, P. N. leSouef, G. M. Lathrop, A. W. Musk, W. O. C. M. Cookson.
1996
. A genome-wide search for quantitative trait loci underlying asthma.
Nature
383
:
247
43
Sudweeks, J. D., J. A. Todd, E. P. Blankenhorn, B. B. Wardell, S. R. Woodward, N. D. Meeker, S. S. Estes, C. Teuscher.
1993
. Locus controlling Bordetella pertussis-induced histamine sensitization (Bphs), an autoimmune disease-susceptibility gene, maps distal to T cell receptor β-chain gene on chromosome 6.
Proc. Natl. Acad. Sci. USA
90
:
3700
44
Satsangi, J., M. Parkes, E. Louis, L. Hashimoto, N. Kato, K. Welsh, J. D. Terwilliger, G. M. Lathrop, J. I. Bell, D. P. Jewell.
1996
. Two stage genome-wide search in inflammatory bowel disease provides evidence for susceptibility loci on chromosome 3, 7 and 12.
Nat. Genet.
14
:
199
45
Sun, S. H., P. B. Silver, Y. Du, R. R. Caspi, R. L. Wilder, E. F. Remmers.
1997
. Identification of a locus on rat chromosome 4 associated with EAU susceptibility.
Invest. Ophthalmol. Vis. Sci.
38
:
S436
(Abstr.).
46
Cornélis, F., S. Fauré, M. Martinez, J.-F. Purd’homme, P. Fritz, C. Dib, H. Alves, P. Barrera, N. de Vries, A. Balsa, et al. for the European Consortium on Rheumatoid Arthritis Families. 1997. Rheumatoid arthritis genome scan and putative autoimmunity locus. Arthritis Rheum. 40:S329 (Abstr.).
47
Jacob, H. J., A. Pettersson, D. D. Wilson, Y. Mao, Å. Lernmark, E. S. Lander.
1992
. Genetic dissection of autoimmune type 1 diabetes in the BB rat.
Nat. Genet.
2
:
56
48
De Gouyon, B., E. Melanitou, M. F. Richard, M. Requarth, I. H. Hahn, J. L. Guenet, F. Demenais, C. Julier, G. M. Lathrop, C. Boitard, P. Avner.
1993
. Genetic analysis of diabetes and insulitis in an interspecific cross of the nonobese diabetic mouse with Mus spretus.
Proc. Natl. Acad. Sci. USA
90
:
1877
49
McAleer, M. A., Reifsnyder P., Palmer S. M., Prochazka M., Love J. M., Copeman J. B., Powell E. E., Rodrigues N. R., Prins J.-B., Serreze D. V., et al
1995
. Crosses of NOD mice with the related NON strain: a polygenic model for IDDM.
Diabetes
44
:
1186
50
Sawcer, S., H. B. Jones, R. Feakes, J. Gray, N. Smaldon, J. Chataway, N. Robertson, D. Clayton, P. N. Goodfellow, A. Compston.
1996
. A genome screen in multiple sclerosis reveals susceptibility loci on chromosome 6p21 and 19q22.
Nat. Genet.
13
:
464
51
Multiple Sclerosis Genetics Group. J., Haines, M. Pericak-Vance, E. Seboun, S. Hauser.
1996
. A complete genome screen for multiple sclerosis underscores a role for the major histocompatibility complex.
Nat. Genet.
13
:
469
52
Ebers, G. C., Kukay K., Bulman D. E., Sadovnick A. D., Rice G., Anderson C., Armstrong H., Cousin K., Bell R. B., Hader W., et al
1996
. A full genome search in multiple sclerosis.
Nat. Genet.
13
:
472
53
Kono, D. H., R. W. Burlingame, D. G. Owens, A. Kuramochi, R. S. Balderas, D. Balomenos, A. N. Theofilopoulos.
1994
. Lupus susceptibility loci in New Zealand mice.
Proc. Natl. Acad. Sci. USA
91
:
10168
54
T. J., Gill, III, Salgar S. K., Yuan X.-J., Kunz H. W..
1997
. Current status of the genetic and physical maps of the major histocompatibility complex in the rat.
Transplant. Proc.
29
:
1657
55
Stastny, P..
1978
. Association of the B-cell alloantigen DRw4 with rheumatoid arthritis.
N. Engl. J. Med.
298
:
869
56
Gregersen, P. K., J. Silver, R. J. Winchester.
1987
. The shared epitope hypothesis: an approach to understanding the molecular genetics of susceptibility to rheumatoid arthritis.
Arthritis Rheum.
30
:
1205
57
Weyand, C. M., K. C. Hicok, D. L. Conn, J. J. Goronzy.
1992
. The influence of HLA-DRB1 genes on disease severity in rheumatoid arthritis.
Ann. Intern. Med.
117
:
801
58
Suarez-Almazor, M. E., S. Tao, F. Moustarah, A. S. Russell, W. Maksymowych.
1995
. HLA-DR1, DR4, and DRB1 disease related subtypes in rheumatoid arthritis: association with susceptibility but not severity in a city wide community based study.
J. Rheumatol.
22
:
2027
59
Wagner, U., S. Kaltenhauser, H. Sauer, S. Arnold, W. Seidel, H. Hantzschel, J. R. Kalden, R. Wassmuth.
1997
. HLA markers and prediction of clinical course and outcome in rheumatoid arthritis.
Arthritis Rheum.
40
:
341
60
Pettersson, A., D. Wilson, T. Daniels, S. Tobin, H. J. Jacob, E. S. Lander, X. C. Lernmark.
1995
. Thyroiditis in the BB rat is associated with lymphopenia but occurs independently of diabetes.
J. Autoimmun.
8
:
493
61
Thomas, D. J. B., A. Young, A. N. Gorsuch, G. F. Bottazzo, A. G. Cudworth.
1983
. Evidence for an association between rheumatoid arthritis and autoimmune endocrine disease.
Ann. Rheum. Dis.
42
:
297
62
Taneja, V., R. R. Singh, A. N. Malaviya, C. Anand, N. K. Mehra.
1993
. Occurrence of autoimmune diseases and relationship of autoantibody expression with HLA phenotpes in multicase rheumatoid arthritis families.
Scand. J. Rheumatol.
22
:
152
63
Puel, A., P. C. Groot, M. G. Lathrop, P. Demant, D. Mouton.
1995
. Mapping of genes controlling quantitative antibody production in Biozzi mice.
J. Immunol.
154
:
5799
64
Teuscher, C..
1985
. Experimental allergic orchitis in mice. II. Association of disease susceptibility with the locus controlling Bordetella pertussis-induced sensitivity to histamine.
Immunogenetics
22
:
417
65
Ginn, L. R., J. P. Lin, P. H. Plotz, S. J. Bale, R. L. Wilder, A. Mbauya, F. W. Miller.
1998
. Familial autoimmunity in pedigrees of idiopathic inflammatory myopathy patients suggests common genetic risk factors for many autoimmune diseases.
Arthritis Rheum.
41
:
400
66
Becker, K. G., R. M. Simon, J. E. Bailey-Wilson, B. Friedlin, W. E. Biddison, H. F. McFarland, J. M. Trent.
1998
. Clustering of non-major histocompatibility complex susceptibility candidate loci in human autoimmune diseases.
Proc. Natl. Acad. Sci. USA
95
:
9979