Gene set enrichment analysis of RNA sequencing data from osteoprogenitor populations in synovial joints of mice with antigen-induced arthritis
Antigen-induced arthritis (AIA) was induced in C57BL6 mice by immunization with methylated bovine serum albumin (mBSA) and subsequent intra-articular injection of mBSA. Non-immunized (NI) mice were injected with phosphate buffered saline at all timepoints. Ten days after intra-articular injection knee joints were harvested and synovial cells were released by collagenase type IV injection into joint cavitied, and fluorescence-activated cell sorting (FACS) was used to sort 200-500 TER119–CD31–CD45–CD51+CD200+CD105– cells from NI mice (NI 200+, Sample 1-4) and mice with AIA (AIA 200+, Sample 5-9) and, TER119–CD31–CD45–CD51+CD200–CD105+ cells from mice with AIA (AIA 105+, Sample 11-14) using BD FASCAria IIu. For each sample, 200-500 live (DAPI–) cells were sorted directly into cell lysis buffer from Smartseq v4 Ultra® Low Input RNA Kit for Sequencing (TakaRa, Kyoto, Japan). ERCC RNA Spike-In Mix (Invitrogen, Thermo Fisher Scientific, Waltham, MA, USA) were added to lysed cell samples. cDNA amplicons were created using SmartSeq v4 Ultra® Low Input RNA Kit for Sequencing (TakaRa) and libraries were prepared using Nextera XT DNA Library Preparation Kit (Illumina, San Diego, CA, USA). Libraries were sequenced using NextSeq 500 (Illumina) instrument and raw files are available at GSE148130. After quality control of raw reads, reads were trimmed, aligned, assembled and quantified using cutadapt (Martin, EMBnet Journal, 2011), HISAT2 (Kim, Nat Methods, 2015) and StringTie (Pertea, Nat Biotecnol, 2015), and normalized and filtered in egdeR package (Robinson, Bioinformatics, 2010). Gene set enrichment analysis (GSEA) was conducted by ClusterProfiler package using gseGO function (Yu, OMICS, 2012) on GO gene sets from biological processes (BP) cell component (CC) and molecular function (MF) categories. Genes were preranked according to signed logarithm (log10) of Benjamini-Hochberg adjusted p value, with positive or negative sign given to genes with positive or negative fold change, respectively. Gene sets with Benjamini-Hochberg adjusted p value < 0.05 were considered significantly enriched and are listed in the tables, together with their enrichment score, normalized enrichment score (NES), p value, adjusted p value (Benjamini-Hochberg) and q value, rank at which the maximum enrichment score occurred, leading edge analysis and core enriched genes for each gene set. The table also includes gene set ID, ontology category (BP, CC or MF), description and size of gene sets. Table 1 contains analysis of comparison of NI 200+ and AIA 200+, Table 2 NI 200+ and AIA 105+ and Table 3 AIA 200+ and AIA 105+.