Combining segmental bulk- and single-cell RNA-sequencing to define the chondrocyte gene expression signature in the murine knee joint

Abstract

Due to the small size of the murine knee joint, extracting the chondrocyte transcriptome from articular cartilage (AC) is a major technical challenge. In this study, we demonstrate a new pragmatic approach of combining bulk RNA-sequencing (RNA-seq) and single cell (sc)RNA-seq to address this problem. We propose a new cutting strategy for the murine femur which produces three segments with a predictable mixed cell population, where one segment contains AC and growth plate (GP) chondrocytes, another GP chondrocytes, and the last segment only bone and bone marrow. We analysed the bulk RNA-seq of the different segments to find distinct genes between the segments. The segment containing AC chondrocytes was digested and analysed via scRNA-seq. Differential expression analysis using bulk RNA-seq identified 350 candidate chondrocyte gene in the AC segment. Gene set enrichment analysis of these genes revealed biological processes related- and non-related to chondrocytes, including, cartilage development (adj. P-value: 3.45E-17) and endochondral bone growth (adj. P-value 1.22E-4), respectively. ScRNA-seq of the AC segment found a cluster of 131cells containing mainly chondrocytes. This cluster had 759 differentially expressed genes which enriched for extracellular matrix organisation (adj. P-value 7.76E-40) and other joint development processes. The intersection of the gene sets of bulk- and scRNA-seq contained 75 genes. Based on our results, we conclude that the combination of the two RNA-seq methods is necessary to precisely delineate the chondrocyte transcriptome and to study the disease phenotypes of chondrocytes in murine OA models in the future.