Sandoval-Plata, Gabriela
(2022)
An investigation into inflammasome-gene expression profile and genome-wide genotype data in gout.
PhD thesis, University of Nottingham.
Abstract
Background: Gout is the most common form of inflammatory arthritis, with a prevalence of 2.49% in the United Kingdom. It is characterised by episodes of acute inflammation (flares) caused by shedding of monosodium urate (MSU) crystals that deposit within and around joints when serum urate (SU) increases above its solubilisation point. Despite the central role of hyperuricaemia in the pathogenesis of gout, not everyone with elevated SU develop symptomatic gout. The reasons underlying the transition from asymptomatic hyperuricaemia, with our without MSU crystal deposits, to symptomatic gout remain poorly understood.
Objectives: (1) To explore expression of inflammasome and toll-like receptors (TLRs)-associated genes and cytokine levels in different stages of the pathogenesis of gout: normal SU, hyperuricaemia without MSU crystal deposits, hyperuricaemia with asymptomatic MSU crystal deposits, intercritical gout and gout flares. (2) To generate a polygenic risk score (PRS) to distinguish gout cases from controls regardless of SU. (3) To generate and validate a genome-wide association study (GWAS) of asymptomatic hyperuricaemia controls vs. gout cases. (4) To generate a PRS to distinguish gout cases from controls with asymptomatic hyperuricaemia.
Methods: Gene expression and cytokine profiling: Participants recruited for three clinical studies conducted at the Department of Academic Rheumatology were classified according to their SU, presence of MSU crystals, and/or gout stage. A total of 108 were included in the gene expression analysis, which evaluated the relative expression of 86 genes associated to inflammasome and TLRs pathways, using the QIAGEN RT2 RNA PCR Arrays. 185 participants
were included in the cytokine measurements, conducted using latex agglutination and Meso-Scale Discovery by Affinity Biomarkers Labs. mRNA and cytokine levels were compared among groups using the Kruskal-Wallis H Test with Bonferroni post-hoc for independent measurements, and Wilcoxon signed-rank rest or Friedman test with Bonferroni post-hoc for repeated measurements. P-values were corrected for multiple testing using a false discovery rate of 5%.
GWAS and PRS studies: Phenotype and genotype data from participants of the UK Biobank resource were used to derive two cohorts: (1) gout cases vs. non-gout controls (regardless SU levels), and (2) gout cases vs. asymptomatic hyperuricaemia controls. The second cohort was divided into the discovery dataset (comprised by 70% of the original cohort) and the replication dataset (comprised by 30% of the original cohort). Genotype data was quality controlled using PLINK v1.9 to remove participants with non-European ethnicity, third degree relatives, sex mismatches, low call-rate and heterozygosity outliers, and markers deviating from Hardy-Weinberg equilibrium and high missingness. Discovery and replication association analyses for gout vs. asymptomatic hyperuricaemia were conducted in PLINK using an additive logistic regression, with age at recruitment, sex and the first 10 principal components as covariates. Genome-wide associations were annotated using FUMA-SNP2GENE resource. PRS for gout vs. controls were calculated with PRSice v2.0, using the GUGC GWAS summary statistics as the base dataset and the UK Biobank genotype data as the target dataset. PRS for gout vs. asymptomatic hyperuricaemia were generated using the discovery GWAS summary statistics as the base dataset and the replication cohort as the target dataset. Predictive ability of the PRS, the demographic and combined models were assessed using the area under the receiving operating characteristic curve (AUROC).
Results: Gene expression profiling: BIRC2, CD40LG, CXCL1, IL-1β, MEFV, NLRP12, PANX1, TNFSF14, TXNIP and XIAP showed a significant upregulation in participants with hyperuricaemia with asymptomatic MSU crystal deposits, compared to participants with normouricaemia. CFLAR, NAIP, NFBIA, NLRC4, NLRP6 and TLR2 were downregulated in participants during the intercritical stage, compared to the gout flare; however, these differences were not significant after correcting for multiple testing. When these 16 genes were compared among the whole spectrum of normouricaemia to acute gout, CD40LG, PANX1 and TNFSF14 showed a downregulation in participants with acute gout, compared to the groups of participants with asymptomatic hyperuricaemia.
Cytokine profiling: In the comparison of intercritical gout vs. gout flare, the levels of VEGF-α and hsCRP showed significant differences. Cytokine levels did not show significant differences among participants with normouricaemia, hyperuricaemia without MSU crystals and hyperuricaemia with asymptomatic deposits of MSU crystals. When cytokines were compared among all groups, GRO-α, IL-1β, IL-6, IL-8, IP-10, MCP-1, TNF-α and hsCRP were greater in the intercritical gout group, compared to the normouricaemia and asymptomatic hyperuricaemia groups.
PRS for gout: The best-fit PRS was generated from 10 SNPs, and showed a significant association with gout. The mean PRS for gout cases and controls were significantly different (0.016 compared to 0.0019, respectively). The predictive ability of the PRS alone was of 62% that increased to a 75% when added to the demographics model.
GWAS and PRS for gout vs. asymptomatic hyperuricaemia: The GWAS revealed 13 independent SNPs located in or near ABCG2, SLC2A9, SLC22A11,
GCKR, MEPE, ADH1B, and the non-coding regions PPM1K-DT and LOC105377323 These loci associated with the transition from asymptomatic hyperuricaemia to gout, and replicated successfully. The PRS generated to distinguish gout from asymptomatic hyperuricaemia was generated from association data of 17 SNPs and gave a predictive ability of 58.2% and 69.2% when combined with the demographics model.
Conclusions: Inflammasome-associated gene expression and cytokine measurements suggest the activation of immune mechanisms in people with asymptomatic deposition of MSU crystals and subclinical systemic inflammation during intercritical gout. The GWAS findings revealed novel loci associated with gout, and confirms the importance of urate transporters and metabolic genes in SU variation and its central role in the pathogenesis of gout. Validation studies in independent datasets are required to confirm gene expression and genome-wide genotype results.
Actions (Archive Staff Only)
 |
Edit View |
Tools
Tools
