Rheumatoid arthritis: identifying and characterising polymorphisms using rat models

Abstract

Rheumatoid arthritis is a chronic inflammatory joint disorder characterised by erosive inflammation of the articular cartilage and by destruction of the synovial joints. It is regulated by both genetic and environmental factors, and, currently, there is no preventative treatment or cure for this disease. Genome-wide association studies have identified ∼100 new loci associated with rheumatoid arthritis, in addition to the already known locus within the major histocompatibility complex II region. However, together, these loci account for only a modest fraction of the genetic variance associated with this disease and very little is known about the pathogenic roles of most of the risk loci identified. Here, we discuss how rat models of rheumatoid arthritis are being used to detect quantitative trait loci that regulate different arthritic traits by genetic linkage analysis and to positionally clone the underlying causative genes using congenic strains. By isolating specific loci on a fixed genetic background, congenic strains overcome the challenges of genetic heterogeneity and environmental interactions associated with human studies. Most importantly, congenic strains allow functional experimental studies be performed to investigate the pathological consequences of natural genetic polymorphisms, as illustrated by the discovery of several major disease genes that contribute to arthritis in rats. We discuss how these advances have provided new biological insights into arthritis in humans.

Keywords: Chronic inflammation; Congenic mapping; Genetics; Rat models; Rheumatoid arthritis; Susceptibility genes.

Fig. 1.

The key stages of rheumatoid arthritis. In this schematic, inflammation is plotted against the lifetime of an individual, who has or has not received treatment upon clinical diagnosis. The pathogenic autoimmune process passes through several stages. First, unknown environmental triggers (potentially including smoking, peridontisis and the gut microbiome) activate immune responses in genetically susceptible individuals many years before clinical onset. These responses can be identified by the production of autoantibodies, such as rheumatoid factor (RF) or anticitrullinated protein antibodies (ACPAs) in the serum. This stage is followed by a joint-specific inflammatory reaction, which leads to clinical onset and which can be perceived by individuals, leading to clinical diagnosis. In the last stage, the disease develops into a chronic phase.

Fig. 2.

Positional cloning of arthritis-linked genes in rats. The identification of arthritis-associated genes using rat models involves several steps. First, genetic linkage analysis is performed. This identifies genetic markers that correlate with the arthritis trait that segregates in a population, leading to the identification of the quantitative trait locus (QTL) for that particular phenotype. The QTL concerned is then isolated in a congenic strain. The chromosomal region that associates with the phenotype is narrowed down to the smallest region possible, preferably containing only one gene, in a process known as positional cloning. Isolating a single gene in a congenic strain is usually not possible. Therefore, other methods, such as transgenesis, genetic engineering or functional assays, are used to identify the causative gene. Once the causative gene is identified, in-depth molecular pathway analysis and comparative studies with humans can be performed. Hypotheses generated from human studies can be tested in animal models to improve our understanding of the human disease. LOD, logarithm of odds.

Fig. 3.

Rat genetic lines used to study arthritis. A schematic illustration of the generation of (A) heterogeneous stock, (B) advanced intercross lines, (C) F1 backcross, (D) F2 intercross and (E) congenic strains, which are used to identify candidate disease loci and genes in rats.

Fig. 4.

The positional identification of genes underlying quantitative trait loci (QTLs) linked to arthritis using congenic rat strains. (A) Inbred rat strains used for linkage analysis and/or construction of congenic strains. (B) Rat chromosomes are depicted as solid vertical blocks. On the right are exemplar logarithm of odds (LOD) likelihood plots for chromosome 20, 12 and 4 for clinical arthritis scoring, and for chromosome 11 for IgM rheumatoid factor (RF) (PIA, day 49). The lines inside the plots represent experiment-wide significance levels. The chromosomal regions with LOD above significance levels (Pia1, Pia4, Pia7 and Rf1) were then further isolated and studied in congenic strains. (C) The genomic intervals of Pia1, Pia4, Pia7 and Rf1 QTL in the DA background were gradually narrowed through stepwise recombinations. For Pia1, Pia4 and Pia7, the green bars denote PIA-protective/promoting strains, and grey bars denote strains with no effect on PIA. For Rf1, green bars denote congenic strains with RF phenotype; grey bars denote congenic strains with no RF phenotype; dark blue bars denote loci that correlate with RF-Igλ production in AIL strains (see Fig. 3); light blue bars denote loci with no RF-Igλ production. The location and size of genomic regions that link to arthritis and RF are indicated. (D) A list of genes in each arthritis- or RF-linked region. The genes in bold are linked to arthritis susceptibility. Note, not all of the congenic strains used in these studies are shown here, and maps of the congenic strains are not to scale. The data shown are derived from previous studies (Haag et al., 2015; Lorentzen et al., 2007; Olofsson and Holmdahl, 2003; Olofsson et al., 2003c; Rintisch et al., 2008; Tuncel et al., 2014; Wernhoff et al., 2003; Yau et al., 2016).