Oral Virtual Presentation (Virtual only) ESA-SRB-ANZBMS 2021

Multiscale analysis of single cell transcriptomics, human genetic association studies, and knockout mouse models identifies genes and cell types associated with bone mineral density and height (#76)

Ryan C Chai 1 2 , Mischa Lundberg 3 4 5 , Monika Frysz 6 , Weng Hua Khoo 1 2 , Paul Baldock 1 2 , Duncan Bassett 7 , Graham R Williams 7 , Peter I Croucher 1 2 , John P Kemp 5 6
  1. Bone Biology, Garvan Institute of Medical Research, Sydney, NSW, Australia
  2. School of Clinical Medicine, Faculty of Medicine, University of New South Wales, Sydney, NSW, Australia
  3. The University of Queensland Diamantina Institute, The University of Queensland, Woolloongabba, QLD, Australia
  4. Faculty of Medicine, The University of Queensland, Brisbane, QLD, Australia
  5. Institute for Molecular Bioscience, The University of Queensland, Brisbane, QLD, Australia
  6. MRC Integrative Epidemiology Unit (IEU), University of Bristol, Oakfield House, Oakfield Grove, Bristol, United Kingdom
  7. Molecular Endocrinology Laboratory, Department of Medicine, Imperial College London, London, UK

Background: Bone mineral density (BMD) is a major determinant of bone strength and adult height reflects the outcome of skeletal growth. Genome-wide association studies (GWAS) have identified hundreds of genomic regions associated with BMD and height, however the underlying genes and cell types remain unclear. We hypothesised that integration of GWAS, single cell RNA sequencing (scRNA-seq) data and knockout mouse (KO) models would identify genes associated with BMD and height in a cellular context.

Methods: scRNA-seq was used to map the transcriptome of cells isolated from mouse femurs. Hypergeometric tests were used to identify cell types enriched for genes involved in monogenetic skeletal disorders. MAGMA gene-set analysis was used in conjunction with GWAS of 448,010 participants in the UK-Biobank Study to identify cell types that are enriched for BMD- and height-associated genes. Skeletal phenotyping of 1000 KO mouse lines was used to validate the functional roles of identified genes.

Results: scRNA-seq analysis of 133,942 cells identified 34 cell types with distinct transcriptional profiles. Osteoblasts and osteoclasts were enriched for genes that cause monogenetic low and high bone mass skeletal disorders respectively, whereas chondrocytes were enriched for genes involved in disorders of bone growth (P<3×10-7). MAGMA identified 3681 BMD- and 7382 height-associated genes (P<2.5×10-6). Osteoblasts, endothelial and vascular smooth muscle cells were enriched for BMD-associated genes, whereas chondrocytes were enriched for height-associated genes (P<5×10-4). In-depth phenotyping of 1000 unselected KO mouse lines showed that cell type specific BMD- and height-associated genes are more likely to result in skeletal abnormalities when deleted in mice (P<1×10-3). Novel BMD-associated genes were identified, including the endothelial-specific gene Slc9a3r2, which when deleted in mice resulted in reduced trabecular bone mass.

Conclusions: Our multiscale approach identified novel BMD- and height-associated genes, and the cellular context through which they may function to regulate bone homeostasis and growth.