Identification of two new arthritis severity loci that regulate levels of autoantibodies, interleukin-1β, and joint damage in pristane- and collagen-induced arthritis

Abstract

Objective: Cia3 is a locus on rat chromosome 4 that regulates severity and joint damage in collagen- and pristane-induced arthritis (CIA and PIA). This study was undertaken to refine the Cia3 gene-containing interval toward gene identification and obtain insights into its mode of action.

Methods: Five DA.F344(Cia3) subcongenic rat strains were generated and studied using the PIA and CIA models. Levels of antibodies against type II collagen (both allo- and autoantibodies) were measured. Joints and synovial tissue were collected 32 days after the induction of PIA (chronic stage) for histologic and quantitative polymerase chain reaction analysis of interleukin-1β (IL-1β) and matrix metalloproteinase (MMP) levels.

Results: Three subcongenic strains sharing the centromeric Cia3d interval were protected and 2 subcongenic strains sharing the telomeric Cia3g interval, which did not overlap with Cia3d, were also protected, developing significantly less severe CIA and PIA. Normal joint architecture was preserved in DA.F344(Cia3) and DA.F344(Cia3d) congenic rats with PIA, while DA rats had pronounced synovial hyperplasia, angiogenesis, inflammatory infiltration, and bone or cartilage erosions. The DA.F344(Cia3d) and DA.F344(Cia3g) strains had significantly lower synovial levels of IL-1β (5-fold and nearly 2-fold, respectively [the latter not reaching statistical significance]), MMP-1 (expressed predominantly in DA rats), MMP-3 (79-fold and 8-fold, respectively), and MMP-14 (21-fold and 1.4-fold, respectively) and reduced levels of pathogenic autoantibodies against type II collagen, compared with DA rats.

Conclusion: We have identified 2 new arthritis severity and articular damage loci within Cia3. These loci regulate pathogenic processes in 2 different models of rheumatoid arthritis, and the identification of these genes has the potential to generate new targets for therapies aimed at reducing disease severity and articular damage, and may additionally have prognostic value.

Figure 1. Markers used in the breeding of DA.F344(Cia3) congenic and subcongenics

Numbers indicate interval distance in megabases (Mb) (http://www.ensembl.org/Rattus_norvegicus/). Black filling indicates homozygous F344 alleles (F/F), white filling indicates homozygous DA alleles (D/D), and grey area indicates the region where recombination took place. Right side of the figure shows co-localizing arthritis loci. PIA=pristane-induced arthritis; CIA= collagen-induced arthritis; AIA=adjuvant-induced arthritis; OIA=oil-induced arthritis. P=protected; NT=not tested.

Figure 2. Arthritis severity clinical scores in DA and congenic/subcongenic rats during a 31-day observation period

Males and females had similar disease severity scores in each strain, and therefore were combined for analyses. Left column: Pristane-induced arthritis (PIA): (A) DA and DA.F344(Cia3); (B) DA and DA.F344(Cia3a); (C) DA and DA.F344(Cia3b); (D) DA and DA.F344(Cia3d), including rats heterozygous at the congenic interval; (E) DA and DA.F344(Cia3f); (F) DA and DA.F344(Cia3g); Right column: Collagen-induced arthritis (CIA): (G) DA and DA.F344(Cia3); (H) DA and DA.F344(Cia3a); (I) DA and DA.F344(Cia3d); (J) DA and DA.F344(Cia3g). *p<0.03; **p≤0.003, Mann-Whitney test; Data shown as mean ± SEM.

Figure 3. Clinical and histological characteristics of DA and DA.F344(Cia3) congenic and subcongenic strains

(A) DA rats developed severe arthritis with pronounced ankle and midfoot swelling and erythema (arrow), while (B) DA.F344(Cia3b), (C) DA.F344(Cia3d) and (D) DA.F344(Cia3g) developed very mild arthritis. (E–F) DA rats had extensive synovial hyperplasia, pannus formation, cartilage and bone invasion and erosion, contrasting with normal joint architecture in (G) DA.F344(Cia3) congenics, and (H) DA.F344(Cia3d) subcongenics. (Ankles were collected on day 32 post-pristane injection; hematoxylin-eosin staining; 100X magnification)

Figure 4. Synovial tissue levels of IL-1β, MMP-3 and MMP14 mRNA, allo and autoantibodies against type II collagen in DA and Cia3 strains

(A) qPCR analyses of synovial tissues from rats with PIA showing DA.F344(Cia3d) (n=8), and DA.F344(Cia3g) (n=9) rats with significantly lower levels of IL-1β, MMP-3 and MMP-14 mRNA compared with DA (n=12). Fold-differences were determined with the 2^−ΔΔCt method, and are shown compared with DA.F344(Cia3d) as reference. ΔCt results were used for comparisons with DA (all with p≤0.001, except for Cia3g IL-1β which had p=0.56; t-test). (B) Levels (measured at day 18 post-induction; μg IgG/mL) of antibodies against bovine type II collagen (anti-BII) and (C) autoantibodies against rat type II collagen (BII) were significantly lower in DA.F344(Cia3), DA.F344(Cia3d and in DA.F344(Cia3g) strains. (ANOVA on ranks with a pairwise multiple comparison procedure [Dunn’s method]; *p≤ 0.05; ** p≤ 0.01; † p=0.054); results shown as mean±S.E.M..