Osteoarthritis year in review 2019: genetics, genomics and epigenetics.

Although osteoarthritis (OA) aetiology is complex, genetic, genomic and epigenetic studies published within the last decade have advanced our understanding of the molecular processes underlying this common musculoskeletal disease. The purpose of this narrative review is to highlight the key research articles within the OA genetics, genomics and epigenetics fields that were published between April 2018 and April 2019. The review focuses on the identification of new OA genetic risk loci, genomics techniques that have been used for the first time in human cartilage and new publicly available databases, and datasets that will aid OA functional studies. Fifty-six new OA susceptibility loci were identified by two large scale genome wide association study meta-analyses, increasing the number of genome-wide significant risk loci to 90. OA risk variants are enriched near genes involved in skeletal development and morphology, and show genetic overlap with height, hip shape, bone area and developmental dysplasia of the hip. Several functional studies of OA loci were published, including a genome-wide analysis of genetic variation on cartilage gene expression. A specialised data portal for exploring cross-species skeletal transcriptomic datasets has been developed, and the first use of cartilage single cell RNAseq analysis reported. This year also saw the systematic identification of all microRNAs, long non-coding RNAs and circular RNAs expressed in human OA cartilage. Putative transcriptional regulatory regions have been mapped in human chondrocytes genome-wide, providing a dataset that will facilitate the prioritisation and characterisation of OA genetic and epigenetic loci.

Discovery and analysis of methylation quantitative trait loci (mQTLs) mapping to novel osteoarthritis genetic risk signals.

OBJECTIVE: Osteoarthritis (OA) is polygenic with over 90 independent genome-wide association loci so far reported. A key next step is the identification of target genes and the molecular mechanisms through which this genetic risk operates. The majority of OA risk-conferring alleles are predicted to act by modulating gene expression. DNA methylation at CpG dinucleotides may be a functional conduit through which this occurs and is detectable by mapping methylation quantitative trait loci, or mQTLs. This approach can therefore provide functional insight into OA risk and will prioritize genes for subsequent investigation. That was our goal, with a focus on the largest set of OA loci yet to be reported. METHOD: We investigated DNA methylation, genotype and RNA sequencing data derived from the cartilage of patients who had undergone arthroplasty and combined this with in silico analyses of expression quantitative trait loci, epigenomes and chromatin interactions. RESULTS: We investigated 42 OA risk loci and in ten of these we identified 24 CpGs in which methylation correlated with genotype (false discovery rate (FDR) P-values ranging from 0.049 to 1.73x10-25). In silico analyses of these mQTLs prioritised genes and regulatory elements at the majority of the ten loci, with COLGALT2 (encoding a collagen galactosyltransferase), COL11A2 (encoding a polypeptide chain of type XI collagen) and WWP2 (encoding a ubiquitin ligase active during chondrogenesis) emerging as particularly compelling target genes. CONCLUSION: We have highlighted the pivotal role of DNA methylation as a link between genetic risk and OA and prioritized genes for further investigation.

The role of inflammation-related genes in osteoarthritis.

In this review article we examine the role of inflammation-related genes in osteoarthritis (OA) from the perspective of genetics, epigenetics and gene expression. There have been great strides in such genomic analyses of OA in recent years thanks to the study of adequately powered patient cohorts, the detailed analysis of candidate genes, and the application of genome-wide approaches. These have led to some unexpected and therefore exciting discoveries, implicating pathways that would not necessarily have been predicted to have a role in this common arthritis. Inflammatory-related genes sit firmly in the candidate camp based on prior observations that the OA disease process can have an inflammatory component. What is clear from the genetic studies published to date is that there is no compelling evidence that DNA variation in inflammatory genes is an OA risk factor. This conclusion may of course change as ever more powerful association studies are conducted. There is, however, compelling evidence that epigenetic effects involving inflammatory genes are a component of OA and that alteration in the expression of these genes is also highly relevant to the disease process. We may in fact be close to demonstrating, at the genomic level, a clear separation of OA patients into those in whom inflammation is a key driver of the disease and those in whom it is not. This has obvious implications for the design of trials of novel OA interventions and may also guide the intelligent re-purposing of anti-inflammatory therapies.

Identification and analysis of a SMAD3 cis-acting eQTL operating in primary osteoarthritis and in the aneurysms and osteoarthritis syndrome.

OBJECTIVE: The TGF-ß pathway plays a central role in joint development with polymorphism in TGF-ß pathway genes implicated in osteoarthritis susceptibility. One association is to rs12901499, within intron 1 of SMAD3. Since rs12901499 is not in linkage disequilibrium with a non-synonymous polymorphism, it is likely the association is operating by influencing expression of SMAD3. DESIGN: Using tissues from the joints of primary osteoarthritis patients who had undergone joint replacement we measured the overall expression of SMAD3 by quantitative real-time PCR. We also measured allelic expression of SMAD3 using these tissues and vascular smooth muscle cells from patients with aneurysms and osteoarthritis syndrome, a rare condition featuring early-onset osteoarthritis. We tested the functional effect of SNPs in vitro using luciferase assays and assessed association with osteoarthritis using a large osteoarthritis case-control dataset. RESULTS: We observed that genotype at rs12901499 did not correlate with overall SMAD3 expression or allelic expression. However, genotype at a 3'UTR SNP, rs8031440, did correlate with SMAD3 expression in cartilage (P = 0.005) which was supported by allelic expression data showing that the G allele correlated with decreased SMAD3 expression in joint tissues and vascular smooth muscle cells. This G allele was underrepresented in osteoarthritis cases vs controls (P = 0.027, odds ratio = 0.921). rs8031440 is in perfect linkage disequilibrium with five other SMAD3 3'UTR SNPs and our luciferase analysis identified rs3743342 and rs12595334 as being functional. CONCLUSION: SMAD3 is subject to cis-acting regulatory polymorphism in the tissues of relevance to both primary osteoarthritis and the aneurysms-osteoarthritis syndrome.

Gene expression analysis reveals HBP1 as a key target for the osteoarthritis susceptibility locus that maps to chromosome 7q22.

OBJECTIVES: An osteoarthritis (OA) susceptibility locus has been mapped to chromosome 7q22, to a region of high-linkage disequilibrium encompassing six genes: PRKAR2B, HBP1, COG5, GPR22, DUS4L and BCAP29. The authors assessed whether these genes were subject to cis-acting regulatory polymorphisms that are active in joint tissues and which could contribute to the association signal. METHODS: Using joint tissues from 156 patients with OA, and control cartilage from 25 patients who had neck of the femur fractures, the authors measured the overall gene expression by quantitative PCR and the allelic expression of the genes, using an assay that can distinguish mRNA output from each allele of a transcript single nucleotide polymorphism. RESULTS: Five of the genes were expressed in joint tissues, the exception being GPR22, which the authors could not detect. In OA cartilage compared with control cartilage, significantly reduced expression levels were observed for these five genes. Carriers of the OA-associated alleles showed a significant reduction in expression of HBP1 in cartilage (p=0.0002) and synovium (p=0.02), and of DUS4L in fat pad (p=0.04). HBP1 and DUS4L also demonstrated allelic expression imbalance across a range of different joint tissues, with carriers of the associated allele showing an HBP1 allelic expression imbalance profile that was significantly different from non-carriers (p=0.008). CONCLUSION: Cis-acting regulatory polymorphisms acting on HBP1 contribute to the OA association signal at chromosome 7q22. HBP1 codes for a transcription factor and studies by the authors have enabled them to prioritise this gene for further investigation.

Deep sequencing of GDF5 reveals the absence of rare variants at this important osteoarthritis susceptibility locus.

OBJECTIVE: The common single nucleotide polymorphism (SNP) rs143383 in the 5' untranslated region (5'UTR) of growth and differentiation factor 5 (GDF5) is strongly associated with osteoarthritis (OA) and influences GDF5 allelic expression in vitro and in the joint tissues of OA patients. This effect is modulated in cis by another common SNP, also located within the 5'UTR, whilst a common SNP in the 3'UTR influences allelic expression independent of rs143383. DNA variants can be common, rare or extremely rare/unique. To therefore enhance our understanding of the allelic architecture of this very important OA susceptibility locus we sequenced the gene for potentially functional and novel rare variants. METHOD: Using the Sanger method we sequenced GDF5 in 992 OA patients and 944 controls, with DNA changes identified by sequencing software. We encompassed the protein-coding region of the two GDF5 exons, both untranslated regions and approximately 100 bp of the proximal promoter of the gene. RESULTS: We detected 13 variants. Six were extremely rare with minor allele frequencies (MAFs) of ≤ 0.0006. One is in a predicted transcription factor binding site in the GDF5 promoter whilst two substitute conserved amino acids. The remaining seven variants were common and are previously known variants, with MAFs ranging from 0.025 to 0.39. There was a complete absence of variants with frequencies in-between the extremely rare (n=6) and the common (n=7). CONCLUSIONS: This is the first report of the deep sequencing of an OA susceptibility locus. The absence of rare variants informs us that within the regions of the gene that we have sequenced GDF5 does not harbour any novel variants that are able to contribute, at a population level, to the OA association signal mediated by rs143383 nor does it harbour, at a population level, any novel variants that can influence OA susceptibility independent of rs143383.