Clinical samples were collected from patients diagnosed with obesity or chronic appendicitis and treated by laparoscopic appendectomy at our hospital between January and May 2021. Patients with a body mass index (BMI) ≥ 32.5 kg/m2 were included in an obese group and those with a BMI ≤ 20 kg/m2 were included in a normal-weight group. About 100 milligrams of VAT (Figure S1) was obtained during surgery, histologically confirmed, immediately frozen in liquid nitrogen, and stored at −80℃ until RNA extraction. Thirty patients were enrolled in an obesity group and 30 in a normal-weight group. Differences in BMI, abdomen circumference, systolic pressure, diastolic pressure, fasting blood glucose, cholesterol, and triglycerides between the two groups were significant (all P < 0.001, Table S1). The study was approved by the ethics committee of our hospital (2021-scientific-367) and all patients gave their informed consent.
Whole transcriptome RNA-seq
Six pairs of VAT samples from obese patients and four pairs from normal-weight patients were selected for RNA-seq, which was performed by Biomarker Technologies (Beijing, China). Total RNA was extracted with TRIzol (Invitrogen, Carlsbad, CA, USA) following the manufacturer’s instructions. The RNA quality and purity were determined with a Nanodrop 2000 Spectrophotometer (Thermo Fisher Scientific, Wilmington, DE, USA). The cDNA libraries for circRNA, lncRNA, and small RNA were constructed separately. Briefly, 1.5 μg RNA samples were used as input material for rRNA removal using Ribo-Zero rRNA Removal Kits (Epicentre, Madison, WI, USA). Sequencing libraries were generated with Next Ultra Directional RNA Library Prep kits (Illumina, New England Biolabs, USA) following the manufacturer’s instructions. First-strand cDNA was synthesized using random hexamer primer and reverse transcriptase. After adenylation of the 3′ ends of DNA fragments, Nebnext hairpin loop structure adapters were ligated to prepare for hybridization. The library fragments were purified with AMPure XP Beads (Beckman Coulter, Beverly, MA, USA) to select insert fragments of 150–200 bp in length. The cDNA was treated with 3 μL uracil DNA glycosylase and DNA glycosylase-lyase endonuclease VIII (USER) enzyme (New England Biolabs, USA) at 37°C for 15 min before PCR. PCR was performed with Phusion High-Fidelity DNA polymerase, universal PCR primers and Index(X) Primer. After cluster generation, the library preparations were sequenced on an Illumina Hiseq platform at Biomarker Technologies (Beijing, China), and paired-end reads were generated.
The transcriptome was assembled using Cufflinks v2.1.1 and Scripture , based on the reads mapped to the reference genome (GRCh38). Differential expression analysis of the two groups was performed using the DESeq R package v.1.10.1 (https://bioconductor.org/packages/release/bioc/html/DESeq2.html). RNAs found by RNA-seq with an adjusted P-value < 0.05 and an absolute value of log2 fold change of >1 were considered as DE.
GSEA, GO and KEGG enrichment analysis
Gene sets were downloaded from the MsigDB database , and broad hallmarks and KEGG symbols were included for GSEA, and pathways with corrected P-values of < 0.05 were considered significantly enriched. GO and KEGG enrichment analysis were conducted by DAVID (https://david.ncifcrf.gov) and KOBAS (http://kobas.cbi.pku.edu.cn/kobas3). GO includes three items, biologic processes (BPs), molecular function (MF), and cellular components (CCs). The DE mRNAs that were targets of DE miRNAs, DE lncRNAs, or DE circRNAs were enriched by GSEA, GO or KEGG analysis. The top ten enriched GO items and KEGG pathways were selected and visualized by SangerBox (http://sangerbox.com).
Protein-protein interaction (PPI) analysis
STRING (https://www.string-db.org) was used to construct PPI networks that showed the gene interactions involved in selected KEGG pathways. An interaction score > 0.4 was used as the extraction cutoff for the PPI pairs. CytoHubba (https://apps.cytoscape.org/apps/cytohubba) was used to select the top ten hub genes in the PPI network. The relative results were visualized by Cytoscape v3.7.0 .
lncRNA-miRNA-mRNA and circRNA-miRNA-mRNA networks were constructed using current hypothetical models of ceRNA regulation and the RNA-seq results. DE miRNA targets of DE lncRNAs were predicted using TargetScan (http://www.targetscan.org). DE miRNA targets of DE circRNAs were predicted by miRanda (http://www.mirbase.org). The target DE genes of DE miRNAs were predicted by the miRBase database (http://www.mirbase.org). Finally, one downregulated lncRNA and four dysregulated circRNAs were selected for inclusion in the ceRNA network, and the results were visualized by Cytoscape v3.7.0.
Quantitative real-time (qRT)-PCR validation
Total RNA was extracted from VAT with TRIzol (Invitrogen, Carlsbad, CA, USA) following the manufacturer’s instructions and then reverse transcribed to cDNA with FastKing RT Kits and gDNase (Tiangen Biotech, Beijing, China). qRT-PCR was performed with SuperReal PreMix Plus (SYBR Green; Tiangen Biotech, Beijing, China) on an ABI 7500 RT-PCR system (Applied Biosystems, Foster city, CA, USA) following the protocol provided by the manufacturer. Relative RNA expression was calculated by the 2−ΔΔCt method, and 18s rRNA (for circRNA), GAPDH (for mRNA), and U6 (for miRNA) were used as internal controls. The primers that were used are listed in Table S2.
SPSS 25.0 (IBM Corp., Armonk, NY, USA) was used for the statistical analysis and GraphPadPrism7.0 (GraphPad Software, La Jolla, CA, USA) was used to draw the figures. Normally distributed variables were reported as means ± standard deviation and categorical data were reported as numbers and percentages (%). Between-group differences were compared by paired t-tests or chi-square tests. P-values < 0.05 were considered statistically significant