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 &lt; 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+.抗原诱导性关节炎（Antigen-induced arthritis, AIA）通过甲基化牛血清白蛋白（methylated bovine serum albumin, mBSA）免疫C57BL6小鼠，并随后向其关节腔内注射mBSA进行造模。非免疫（Non-immunized, NI）小鼠在所有时间点均注射磷酸盐缓冲液（phosphate buffered saline, PBS）。关节腔内注射10天后，采集膝关节，通过向关节腔注射IV型胶原酶分离滑膜细胞，随后使用BD FASCAria IIu流式细胞仪，通过荧光激活细胞分选术（fluorescence-activated cell sorting, FACS）分选出目标细胞：从NI小鼠中获取200~500个TER119–CD31–CD45–CD51+CD200+CD105–细胞（NI 200+，样本1~4），从AIA模型小鼠中分选出200~500个TER119–CD31–CD45–CD51+CD200+CD105–细胞（AIA 200+，样本5~9），以及AIA模型小鼠中的TER119–CD31–CD45–CD51+CD200–CD105+细胞（AIA 105+，样本11~14）。对于每个样本，将200~500个DAPI阴性活细胞直接分选加入Smartseq v4 Ultra® 低起始量RNA测序试剂盒（Takara, 日本京都）的细胞裂解液中。向裂解后的细胞样本中加入ERCC RNA Spike-In Mix（Invitrogen, 赛默飞世尔科技, 美国马萨诸塞州沃尔瑟姆）。使用SmartSeq v4 Ultra® 低起始量RNA测序试剂盒（Takara）构建cDNA扩增子，并通过Nextera XT DNA文库制备试剂盒（Illumina, 美国加利福尼亚州圣地亚哥）构建测序文库。使用NextSeq 500（Illumina）测序仪对文库进行测序，原始数据已上传至GSE148130。对原始reads进行质控后，使用cutadapt（Martin, EMBnet Journal, 2011）、HISAT2（Kim, Nature Methods, 2015）与StringTie（Pertea, Nature Biotechnology, 2015）对reads进行修剪、比对、组装与定量，并通过edgeR软件包（Robinson, Bioinformatics, 2010）完成标准化与过滤。基因集富集分析（Gene Set Enrichment Analysis, GSEA）采用ClusterProfiler软件包的gseGO函数（Yu, OMICS, 2012），针对生物过程（biological processes, BP）、细胞组分（cell component, CC）及分子功能（molecular function, MF）三类GO基因集开展分析。基因按照经Benjamini-Hochberg校正后的P值的符号化对数（log10）进行排序，其中倍数变化为正的基因赋予正号，倍数变化为负的基因赋予负号。以Benjamini-Hochberg校正后P值<0.05的基因集视为显著富集，将其富集得分（enrichment score, ES）、标准化富集得分（normalized enrichment score, NES）、P值、校正后P值（Benjamini-Hochberg）、q值、最大富集得分出现的排名、前沿分析结果及各基因集的核心富集基因一并列于表格中。表格同时包含基因集ID、本体论分类（BP、CC或MF）、基因集描述与基因集大小。表格1为NI 200+与AIA 200+的比较分析，表格2为NI 200+与AIA 105+的比较分析，表格3为AIA 200+与AIA 105+的比较分析。