Background

Involving research users in setting priorities for research is essential to ensure research outcomes are patient-centred and to maximise research value and impact. The Musculoskeletal (MSK) Disorders Research Advisory Group Versus Arthritis led a research priority setting exercise across MSK disorders.

This paper reports on a proof of concept project funded by the UK National Council for the Replacement, Refinement and Reduction of Animals in Research (NC3Rs), with the aim of developing an in vitro model to recapitulate the human osteoarthritic joint, based on a multiple human cell type co-culture system, for research and drug development in OA. The targets were: (i) the development of a cell culture platform that could produce a mixed stable cell culture of cell types that represent the key components of the human joint: synoviocytes – type I and type II; osteoblasts; osteoclasts; chondrocytes/cartilage or cartilage-like matrix; adipocytes; and immune cells. (ii) demonstration of cell phenotype stability and viability for at least 72 hours. In order to establish the cell culture platform we have developed an eight-channel cell printer, capable of accurately and reliably printing the required cell types to create osteochondral and synovial cell types within a transwell system. Two different sets of cells have been developed and processed using the cell printer: a set based on using an immortalised hTERT MSC line to create osteoblasts, chondrocytes and adipocytes, with commercial cells lines providing the other cell types, and a set obtained from tissue excised during orthopaedic surgery. This gives both a repeatable set of cells with which to undertake mode of action studies, and a bank of cell sets which will be representative of different stages of osteoarthritis. The co-cultures have been immunohistochemically assessed in order to demonstrate maintenance of phenotype.

Introduction: Bone phenotype, such as osteoarthritis (OA) pattern and development of osteolysis or heterotopic ossification (HO) after THA, may be governed by genetic and environmental factors. We investigated whether single nucleotide polymorphisms within the gene encoding secreted-Frizzled Related Protein-3, FRZB Arg200Trp and FRZB Arg324Gly influence bone phenotype.

Methods: Genomic DNA was extracted from 609 subjects at a mean of 11 years following cemented THA for idiopathic osteoarthritis. Pre-operative OA was defined using The American College of Rheumatology criteria and post operative HO after primary THA was assessed using Brooker’s classification

Results: For FRZB Arg200Trp, minor allele carriage (MAC) was greater in subjects with pre-operative pelvic osteophytes (n=267) versus those without osteophytes (n=34) (MAC 27.9% versus 6.3%, Fisher’s exact test p=0.037). There were no associations with other radiographic criteria of OA. MAC was also higher in HO+ve subjects (n=291) versus HO-ve subjects (n=341), (MAC 21.7% versus 12.0%, χ² test p=0.063). Finally MAC was 14.2% in osteolysis +ve subjects (n=268) and 21.7% in osteolysis –ve subjects (n=341) (χ² test p=0.041).

The adjusted odds ratios for pelvic osteophytes and HO with carriage of the rare FRZB 200 variant were 4.34 (1.01–18.7 p=0.048) and 1.64 (1.05 to 2.54, p=0.028) respectively. The adjusted odds ratio for osteolysis was 0.62 (0.38 to 0.99 p=0.049).

There were no bone phenotype associations with the FRZB Arg324Gly variants.

Discussion: Carriage of the FRZB 200Trp allele is positively associated with osteophyte and HO formation and negatively associated with osteolysis, suggesting this locus may be a marker for pro-osteoblastic activity.

In-vitro evidence suggests that wear debris can alter osteoblast function resulting in decreased bone matrix production and negative remodelling balance. FRZB encodes for Secreted Frizzled-Related Protein 3 which may play a role in bone formation and osteoarthritis. This study was undertaken to investigate whether the recently described single nucleotide polymorphisms (SNPs) at positions [+6] and [+109] of the FRZB gene are associated with osteolysis after THA.

Genomic DNA was extracted from 481 North European Caucasians at a mean of 12 years following cemented THA for idiopathic osteoarthritis. The control group consisted of 267 subjects and the osteolysis group 214 subjects. The [+6] and [+109] FRZB SNPs were genotyped using standard techniques.

For the FRZB [+6] SNP, the rare T allele was significantly over-represented in control versus the osteolysis group (χ² test for trend, p=0.02,). The odds ratio for osteolysis associated with carriage of the [+6] T-allele versus the [+6] C-allele was 0.58 (95%CI 0.36 to 0.94), p=0.03. The odds ratio for osteolysis associated with carriage of the [+109] G-allele versus the [+109] C-allele was 0.66 (0.38 to 1.12), p=0.15. A number of covariates have previously been described in this cohort and after adjustment for the effects of these covariates, the odds ratio for osteolysis with carriage of the [+6] T-allele was 0.69 (0.42–1.16).

We found that the FRZB [+6] T-allele is less common in subjects with osteolysis after THA versus controls, suggesting that allelic variants of genes associated with bone formation pathways may have a role in modulating the risk of osteolysis. However its loss of significance after correction for other factors suggests an interaction between this allele and other risk factors in osteolysis.