Oral Presentation ANZBMS-MEPSA-ANZORS 2022

Identifying genes involved in hip osteoarthritis and describing the cells in which they are differentially expressed (#17)

Kaitlyn A Flynn 1 , Ryan C Chai 2 3 , Weng Hua Khoo 2 3 , Paul Baldock 2 , Genetics of Osteoarthritis Consortium 4 , Duncan Bassett 5 , Graham R Williams 5 , Peter I Croucher 2 3 , John P Kemp 1 6
  1. Institute for Molecular Bioscience, The University of Queensland, Brisbane, QLD, Australia
  2. Bone Biology, Garvan Institute of Medical Research, Sydney, NSW, Australia
  3. St. Vincent's Clinical School, University of New South Wales, Sydney, NSW, Australia
  4. https://www.genetics-osteoarthritis.com/home/index.html, Germany
  5. Molecular Endocrinology Laboratory, Imperial College London, London, England, United Kingdom
  6. MRC Integrative Epidemiology Unit (IEU) , University of Bristol, Bristol, England, United Kingdom

Background: Hip osteoarthritis (HipOA) is a complex disorder that adversely affects joint function. Genome-wide association studies (GWAS) have identified 45 HipOA-associated loci. Many genes underlying these associations remain unknown, impairing understanding of disease pathogenesis. We hypothesise that HipOA susceptibility genes can be systematically identified by integrating single cell-RNA sequencing (scRNA-seq), GWAS, and knockout mouse models.

Methods: Bone and marrow cells were isolated from mouse femora, and transcriptomes were mapped by scRNA-seq. Seurat analysis identified sets of differentially expressed genes (DEGs) that distinguished each cell type from all other cells. Gene set analysis (GSA) was used in conjunction with the largest HipOA GWAS to test whether DEGs that defined each cell type were enriched with HipOA-associated genes. HipOA-associated genes that were differentially expressed in enriched cell types were followed up in skeletal dysplasia databases.

Results: scRNA-seq analysis of 133,942 cells identified 34 cell types, each defined by a set of DEGs. GSA showed that DEGs that defined osteoblasts (P<3.9×10-4), chondrocytes (P<9×10-5) and pre-neutrophils (P<7×10-6) were more strongly associated with HipOA, than genes that were not differentially expressed in these cells. Seventeen HipOA-associated genes were differentially expressed in enriched cell types. Nine resulted in abnormal skeletal phenotypes when mutated. Col11a1 (differentially expressed in osteoblasts and chondrocytes relative to all other cells), Ltbp3 (osteoblasts), and Mdfi (osteoblasts and chondrocytes) caused osteoarthritis in humans and/or mice. Tacc3 (pre-neutrophils) and Creb3l2 (chondrocytes) resulted in abnormal cartilage morphology. Eight had no known skeletal annotations. Tenascin-C (Tnc) represented a compelling candidate that was robustly associated with HipOA (P<2×10-18); expression by osteoblasts was >4 fold higher than other cells, and previous studies show that Tnc is upregulated in response to mechanical loading, and under pathological conditions caused by inflammation.

Conclusions: We highlighted several HipOA-associated genes that are differentially expressed in cells that may be involved in osteoarthritis pathogenesis.