Genetic and epigenetic variation within extracellular matrix genes as risk factors for human tendinopathy

Background and Aims Regular physical activities have shown to have various health benefits however there is always an accompanying risk of developing musculoskeletal soft tissue injuries. Indeed, damaged tendons account for 30-50% of sports-related injuries where the lifetime risk of Achilles tendon pathology (ATP) approaches 50% in runners, and that of patellar tendon pathology (PTP) is 21% in football players. The exact aetiology and mechanisms of tendon pathologies are still under investigation, however extrinsic and intrinsic risk factors (of which genetic) have been identified. Recent genetic association studies found that gene variants within the TNC, COL5A1, MMP3, GDF5 and CASP8 were associated with ATP in a Caucasian Australian and South African population. Furthermore, epigenetic mechanisms such as DNA methylation and microRNA (miRNA) activity have been implicated in a range of diseases but were never investigated for their role in human tendinopathy. Based on the aforementioned information, this thesis aimed at investigating novel candidate genes that may be associated with ATP and to replicate previously conducted studies in a newly recruited case-control population. Additionally, this thesis aimed at investigating potential differences in DNA methylation profiles and miRNA expression levels between healthy and damaged Achilles and patellar tendons. One hundred and thirty six UK Caucasian participants with clinically diagnosed ATP and 131 asymptomatic, unrelated, physically active control participants were recruited for this study. Furthermore, the previously recruited 173 clinically diagnosed ATP participants and 238 asymptomatic, unrelated, physically active control participants from Australia and South Africa (AUS+SA) were also included in the studies of this thesis. Participants within the combined AUS+SA were genotyped for the ELN rs2071307, FBN2 rs331079, ADAM12 rs3740199, ADAMTS2 rs1054480, ADAMTS5 rs226794, ADAMTS14 rs4747096, and TIMP2 rs4789932 variants, and the UK participants were genotyped for the COL5A1 rs71746744, FBN2 rs331079, GDF5 rs143833, MMP3 rs679620, TIMP2 rs4789932 variants using fluorescent based TaqMan® technology. Furthermore, the UK cohort was genotyped for the COL5A1 rs12722 variant using polyacrylamide gel electrophoresis. Moreover, 10 healthy and 10 diseased patellar tendon tissue samples in addition to 4 healthy, and 1 diseased Achilles tendon samples were obtained for the epigenetic studies. The DNA methylation profile within the TIMP2 and GDF5 promoter regions were analysed for all samples using pyrosequencing technology. Furthermore, the expression levels of TIMP2, miR-21, miR-155, and miR-191 were determined by RT-PCR using TaqMan technology. Results and Discussion The genetic association studies conducted showed that the FBN2 rs331079 GG genotype was over-represented among the tendinopathy (TEN) group and that the ELN rs2071307 AA genotype was over-represented in the rupture (RUP) group within the AUS+SA population. Furthermore, the COL5A1 rs12722 and rs71746744 were associated with RUP (TT genotype over-represented in the RUP group, p=0.004; OR=4.2; 95% CI 1.58-11.97) and ATP (DEL allele over-represented in the CON group, p=0.046; OR=1.61; 95% CI 1.01-2.56) respectively in the male UK cohort. The GDF5 rs143833, on the other hand, was not associated with ATP (p=0.538) and showed no sign of gender-specific association (female p=0.737; male p=0.319) in the UK population. Furthermore, the CT genotype for the TIMP2 rs4789932 variant was over-represented (p=0.004; OR=1.77; 95% CI 1.20 - 2.64) in the ATP group of the combined AUS+SA population and the CC genotype was over represented (p=0.016; OR=2.36; 95% CI 1.16 – 5.81) in ATP of the UK male cohort. It was also reported that the MMP3 rs679620 GG genotype was over represented (p=0.027; OR=2.51; 95% CI 1.11 – 5.64) in the UK RUP group. The ADAM12 rs3740199 (p=0.633), ADAMTS2 rs1054480 (p=0.316), ADAMTS5 rs226794 (p=0.342), and ADAMTS14 rs4747096 (p=0.849) gene variants were not associated with ATP in the AUS+SA population. Interestingly, individuals carrying the ADAMTS14 rs4747096 GG genotype within the AUS+SA population and the ELN rs2071307 AA genotype within the AUS population developed their injuries at a significantly (p=0.024; p=0.005, respectively) later stage than other participants. Moreover, UK males diagnosed with tendinosis and carrying the GG genotype at the MMP3 rs679620 locus developed significantly (p=0.003) thicker tendons than other participants with the AA, and AG genotypes. The preliminary epigenetic DNA methylation studies showed no differences in the average methylation profiles of the investigated regions within the TIMP2 (p=0.885) and GDF5 (p=0.333) genes in the patellar tendon samples. Moreover, no DNA methylation differences (p=0.617) were observed in the investigated region of the TIMP2 gene in the Achilles tendon samples. Interestingly, the single ATP sample showed a lower GDF5 average methylation profile than the CON samples. Furthermore, the expression of TIMP2 was up-regulated, and miR-191 was down-regulated in the ATP tissue sample compared to the CON group. The expression levels of miR-21 and miR-151, however, were not different between the two groups. Conclusion This thesis provides evidence that novel genes coding for structural and ECM regulatory enzymes are associated with ATPs in Caucasians. The findings of this thesis should to be replicated in new and larger cohorts from different ethnic backgrounds before being incorporated into multifactorial risk assessment models aiming at reducing the incidence of human tendinopathy.