Bench to Bedside: Modelling Inflammatory Arthritis

Abstract

Inflammatory arthritides such as rheumatoid arthritis are a major cause of disability. Pre-clinical murine models of inflammatory arthritis continue to be invaluable tools with which to identify and validate therapeutic targets and compounds. The models used are well-characterised and, whilst none truly recapitulates the human disease, they are crucial to researchers seeking to identify novel therapeutic targets and to test efficacy during preclinical trials of novel drug candidates. The arthritis parameters recorded during clinical trials and routine clinical patient care have been carefully standardised, allowing comparison between centres, trials, and treatments. Similar standardisation of scoring across in vivo models has not occurred, which makes interpretation of published results, and comparison between arthritis models, challenging. Here, we include a detailed and readily implementable arthritis scoring system, that increases the breadth of arthritis characteristics captured during experimental arthritis and supports responsive and adaptive monitoring of disease progression in murine models of inflammatory arthritis. In addition, we reference the wider ethical and experimental factors researchers should consider during the experimental design phase, with emphasis on the continued importance of replacement, reduction, and refinement of animal usage in arthritis research.

Keywords: 3Rs; Rheumatoid arthritis; TNF; clinical score; murine models of arthritis.

Graphical Abstract

Figure 1:

Example score sheet for inflammatory arthritis monitoring. (A) First, behaviour and coat condition are observed in home cage and prior to any handling. (B) Subsequently, weighing of individual animal followed by transfer to a new, clean cage (clear of any housing or bedding) to allow observation of mobility. Evidence of “grimace” as described in the Mouse Grimace Scale [53] can be recorded at this stage, if not already. (C) Finally, each paw of the restrained mouse is observed, and swollen regions shaded on the paw schematic. A score is then given of between 0 – 3, based on the number of joints/regions shaded (0, represents no visible swelling, 1 = 1 or 2 affected joints; 2 = multiple affected joints; 3 = generalised swelling across the paw). Whilst restrained, calliper measurements can be taken of the front and rear footpads and of the rear ankles (hock joint). (D) The “arthritic paw score” or “clinical score” can then be calculated. The sum of scores from each of A, B, and C give the “global score”, thus comprising both clinically evident arthritis and the clinically evident extra-articular manifestations of disease.

Figure 2.

Comparison of scoring parameter outputs between two inducible models of inflammatory polyarthritis. Left: K/BxN serum transfer arthritis, induced in 8–10-week-old male C57BL/6J via two 100μl intraperitoneal injections of K/BxN serum. N = 17. Right: Collagen-induced arthritis, induced in 8–10-week-old male DBA1 mice via the protocol described in Brand et al. 2007 [49]. N = 17. In each case, scoring was performed using the parameters detailed above (figure 1). (A). Clinical score: A measure of the number of swollen joints. (B). Mobility: A measure of effect of arthritis on mobility. (C). Weight change from baseline, shown as percentage change. (D). Grimace scale: Evidence of grimace identified using the ‘Mouse Grimace Scale’ developed by Langford et al. 2010 [53] and scored as demonstrated on the score sheet in figure 1. (E). Global score: A composite measure combining the scores from all aspects of the scoring system described in figure 1. All animal experiments were performed in accordance with U.K. laws (Animal [Scientific Procedures] Act 1986) and with the approval of the local ethics committees at the University of Birmingham.

Figure 3.

Current replacement options for IA models. (A) Use of freely available large datasets (RNAseq, proteomics, metabolomics, CyToF) from patient materials to identify the pathways, genes, or processes of interest to the research question being investigated (e.g. reviewed in Buckley et al. 2021 [3]). (B) In vitro analysis of patient material in simple 2D culture systems (e.g., culturing fibroblasts on tissue culture plastic [99]), more complex 3D culture systems involving the incorporation of stromal cells or immune cells into gel structures (e.g., collagen or hydrogel) or the formation of 3D organoids [e.g. [100, 101],]. These types of culture systems are progressing towards more whole tissue models including the creation of mini-bones within tissue culture [102] or the use of organ-on-a-chip style microfluidics channels (e.g., reviewed in [98]). Figure created in Biorender.com.

Figure 4.

Refinement considerations regarding environment, housing, and choice of cohort. (A). Social housing with same sex and appropriate cage mates promotes social exploration and natural behavioural activities such as digging, but also provide social support during stressful situations. However, incompatible mice can lead to aggression, stress and injury which is more common in males. (B). Standard mouse housing conditions generally have room temperatures of between 20-24°C (68-75F) and are as stable as possible. Consideration could be given to increasing these temperatures for arthritic mice, as studies have shown that warmer temperatures are most preferred (described in detail in Hawkins et al. 2015 [55]). (C). Environmental enrichments provide sensory and motor stimulation. Soft, non-tangling nesting material, as well as soft litter reduce pain on walking, and cushion sore joints. (D). Easy access to food and water is necessary to cater for any disability in movement. This can be achieved by using bottles with long spouts and placing soft palatable food on the cage floor. (E). When handling animals avoid catching them by the tail, a practice known to induce a profound stress response. Instead, mice should be restrained with cupped hands or encouraged to enter handling tunnels [79, 80], this reduces stress and discomfort, which can be a potential source of variation within studies, while increasing willingness of the mice to interact with the observer. (F). Daily calliper measurement and weight measurements ensure that mice are carefully monitored and that disease course for each model is thoroughly understood by researchers and animal care staff. Figure created in Biorender.com.