1148: GWAS Identified New Genes in Synovial Fibroblasts Linked to Early Remission in RA

Marc Maurits¹, lydia Abasolo Alcazar², Erik van den Akker³, Johan Askling⁴, Anne Barton⁵, Stephan Blüml⁶, Stephan Böhringer⁷, Andrew Cope⁸, Paul Emery⁹, Stephen Eyre⁵, Priya Gaddi⁵, Isidoro Gonzalez¹⁰, Carl Goodyear¹¹, Annette van der Helm-van Mil¹², Xinli Hu¹³, Tom Huizinga³, John D Isaacs¹⁴, Scott Jelinsky¹³, Martina Johannesson⁴, Samantha Jurado Zapata³, Changlin Ke³, Lars Klareskog⁴, Dennis Lendrem¹⁵, Myles Lewis¹⁶, Mingdong Liu³, Paul Martin¹⁷, Iain B McInnes¹⁸, Raphael Micheroli¹⁹, Ann Morgan²⁰, Fraser Morton¹¹, Najib Naamane¹⁵, Gisela Orozco⁵, Caroline Ospelt²¹, Leonid Padyukov⁴, Caron Paterson¹¹, Darren Plant⁵, Duncan Porter²², Arthur Pratt²³, Soumya Raychaudhuri²⁴, Louise Reynard¹⁵, Luis Rodriguez-Rodriguez²⁵, Daniela Sieghart²⁶, Paul Studenic⁶, John Taylor²⁷, René Toes³, Marloes Verstappen³, Suzanne Verstappen⁵, Helga Westerlind⁴, Aaron Winkler¹³ and Rachel Knevel³, ¹LUMC, Leiden, Netherlands, ²IDISSC Hospital Clinico San Carlos, Madrid, Spain, ³Leiden University Medical Center, Leiden, Netherlands, ⁴Karolinska Institutet, Stockholm, Sweden, ⁵The University of Manchester, Manchester, United Kingdom, ⁶Medical University of Vienna, Vienna, Austria, ⁷Department of Biomedical Data Sciences, Leiden University Medical Center, Leiden, Leiden, Netherlands, ⁸King's College London, Surrey, United Kingdom, ⁹Leeds Institute of Rheumatic and Musculoskeletal Medicine, University of Leeds, Leeds, United Kingdom, ¹⁰Hospital Universitario de La Princesa, Madrid, Spain, ¹¹University of Glasgow, Glasgow, Scotland, United Kingdom, ¹²Leiden University Medical Center, Erasmus Medical Center, Leiden, Netherlands, ¹³Pfizer Inc., Cambridge, MA, ¹⁴Institute for Translational and Clinical Research, Newcastle University and Musculoskeletal Unit, Newcastle upon Tyne Hospitals, Newcastle upon Tyne, United Kingdom, ¹⁵Newcastle University, Newcastle, United Kingdom, ¹⁶Queen Mary University of London, London, United Kingdom, ¹⁷The University of Manchester, Oberhaching, Germany, ¹⁸Institute of Infection, Immunity and Inflammation, College of Medical, Veterinary and Life Sciences, University of Glasgow, Glasgow, Scotland, United Kingdom, ¹⁹University Hospital Zurich, Department of Rheumatology, Zürich, Switzerland, ²⁰University of Leeds, Leeds, United Kingdom, ²¹Center of Experimental Rheumatology, Department of Rheumatology, University Hospital Zurich, University of Zurich, Zürich, Switzerland, ²²Gartnavel General Hospital, Bearsden, United Kingdom, ²³Newcastle University, Newcastle upon Tyne, United Kingdom, ²⁴Brigham and Women's Hospital, Boston, MA, ²⁵Instituto de Investigación Sanitaria San Carlos, Rheumatology, Madrid, Spain, ²⁶Division of Rheumatology, Department of Internal Medicine III, Medical University of Vienna, Vienna, Austria, ²⁷University of Leeds, Euclid, OH

Background/Purpose: In order to understand the genetic factors that lead to early remission in RA, we performed a GWAS to uncover important biological pathways.

Methods: We collected 5,635 untreated patients with RA from 10 centers throughout Europe. We defined remission as DAS28CRP < 2.6 at 6 months. We estimated the genetic component of remission using LDAK-PCGC. Next, we ran a GWAS in each cohort, correcting for age, sex and population stratification, and combined the results with an inverse-variance weighted meta-analysis with fixed-effects. Positional and expression quantitative trait loci (eQTL) mapping was performed using FUMA. For transcriptional analyses, we used AMP-I [1] single cell RNAseq synovial biopsies (n=55 RA patients), Zurich cohort of 26 synovial tissues across 5 arthritides, PEAC synovial biopsies RNAseq (n=355 RA patients [2]) from unstimulated and stimulated fibroblast-like synoviocytes (FLS) and circulating CD4+ T-cell eQTL data from untreated RA patients (NEAC) [3].

Results: In our data, 16% of phenotypic variance was explained by genetics (h2). Polygenic risk score analysis showed this was not attributable to established RA susceptibility variants. Our GWAS uncovered 2 loci of interest; a genome-wide significant hit (locus A, rs16866400, Chr2:179448911, P = 2.36 x 10-8) and a suggestive hit (locus B, rs11645657, Chr16:11223454, P = 1.13 x 10-7) (Figure 1). Locus A was positionally mapped to TTN(-AS1), a gene, expressed in muscle, with an eQTL for fk506 (tacrolimus) binding protein (FKBP7) in fibroblasts. Our data show that this variant affects transcription in FLS and that FKBP7 is differentially expressed in FLS. In the PEAC biopsies we demonstrate that the remission-associated allele increases transcription of FKBP7 in synovia, which correlates with lower baseline disease activity and better treatment response. We also observed a decrease in FKBP7 expression in FLS when treated with IL4 and an increase when treated with TNF (Figure 2). The leading SNP in locus B is likely itself regulatory, given its high regulatory scores (regulomeDB rank = 2B, CAD score = 1.5) and the lack of strong LD to alternative SNPs, suggesting a self-regulatory nature. Based on the ChromHMM and eQTL data, this SNP affects enhancement in FLS and is an eQTL in early RA CD4+ T-cells for CLEC16A, a gene with variants associated with multiple autoimmune disorders, ranging from MS to seronegative RA [4]. Additionally, CLEC16A KO mice exhibit an autoimmune phenotype reversed by JAK/STAT inhibition [5] (Figure 3).

Conclusion: We have identified two genes whose expression is associated with remission in patients with early RA: FKBP7 and CLEC16A. The FKBP7 pathway seems to be FLS mediated. We are performing further experimental studies to investigate both genes. Based on our current work we postulate a broadening of the conventional B- and T-cell focus in RA treatment to also include FLS.

References:
1. Zhang et al, 2019, Nat Imm.
2. Lewis et al, 2019, Cell Rep.
3. Thalayasingam et al, 2018, A&R.
4. Skinningsrud et al, 2010, ARD.
5. Pandey et al, 2021, Sci. Rep.

Acknowledgments
We would like to acknowledge support by RTCure, the IMI2 JU (grant 777357), RTCure, ZonMW (grant 90719069), MRC/Versus Arthritis MATURA Consortium, Versus Arthritis and NIHR.
Figure 1. GWAS results in the form of (A) LocusZoom Manhattan plot, dashed line indicates the genome-wide significance threshold of 5 x 10-8, (B) forest plot of locus A (FKBP7) effect size across datasets, and (C) forest plot of locus B (CLEC16A) effect size across datasets. Several cohorts are split across datasets.

Figure 2. Overview of locus A with (A) number (proportion) of RA patients with at least one effect allele vs reaching DAS28CRP remission at 6 months post-diagnosis, (B) association in early inflammatory arthritis patients of FKBP7 expression in blood and DAS28CRP (response) (PEAC synovial biopsies RNAseq in 355 RA patients [2]), (C) expression levels of FKBP7 across fibroblasts, monocytes, T-cells and B-cells in rheumatoid arthritis, osteoarthritis and healthy participants (AMP-I scRNAseq synovial biopsies [3]), (D) FKBP7 expression in fibroblast-like synoviocytes of RA patients under various stimuli, and (E) expression of FKBP7 across various cell types in arthritis patients (Zurich cohort of 26 synovial tissues across 5 arthritides).
FPKM, fragments per kilobase of transcript per million mapped reads; CPM, counts per million

Figure 3. Overview of locus B with (A) number (proportion) of RA patients with at least one effect allele vs reaching DAS28CRP remission at 6 months post-diagnosis, (B) top 10 GWAS catalogue reported trait associations of CLEC16A , and (C) local LD, CADD score and regulomeDB score plots.
Disclosures: M. Maurits, None; l. Abasolo Alcazar, None; E. van den Akker, None; J. Askling, None; A. Barton, Galapagos, Bristol-Myers Squibb(BMS), Roche Chugai; S. Blüml, None; S. Böhringer, None; A. Cope, None; P. Emery, AbbVie, Amgen, Bristol Myers Squibb, Celltrion, Eli Lilly, Gilead, Novartis, Pfizer, Roche, Samsung; S. Eyre, None; P. Gaddi, None; I. Gonzalez, Gebro Pharma; C. Goodyear, None; A. van der Helm-van Mil, None; X. Hu, None; T. Huizinga, None; J. Isaacs, AbbVie/Abbott, Bristol-Myers Squibb(BMS), GlaxoSmithKlein(GSK), Janssen, Eli Lilly, Gilead, Pfizer, Roche; S. Jelinsky, None; M. Johannesson, None; S. Jurado Zapata, None; C. Ke, None; L. Klareskog, None; D. Lendrem, None; M. Lewis, None; M. Liu, None; P. Martin, None; I. McInnes, Bristol-Myers Squibb (BMS), Janssen, Novartis, UCB, Pfizer, AbbVie, Celgene, AstraZeneca, Boehringer Ingelheim, EveloBio, LEO, Lilly; R. Micheroli, None; A. Morgan, None; F. Morton, None; N. Naamane, None; G. Orozco, None; C. Ospelt, None; L. Padyukov, None; C. Paterson, None; D. Plant, None; D. Porter, None; A. Pratt, None; S. Raychaudhuri, Mestag, Inc, Rheos Medicines, Janssen, Pfizer, Biogen; L. Reynard, None; L. Rodriguez-Rodriguez, None; D. Sieghart, None; P. Studenic, AbbVie/Abbott; J. Taylor, None; R. Toes, None; M. Verstappen, Bristol-Myers Squibb(BMS); S. Verstappen, None; H. Westerlind, None; A. Winkler, None; R. Knevel, None.