THP-1 cells were grown in lipopolysaccharide-free complete RPMI medium containing 10% fetal bovine serum. HEK293 cells were grown in Dulbecco’s modified Eagle’s medium containing 10% fetal bovine serum. HUVEC cells were grown in Cascade Biologics Medium 200 (Gibco, Thermo Fisher Scientific, Waltham, MA, United States) plus Cascade Biologics Low Serum Growth Supplement (Gibco, Thermo Fisher Scientific, Waltham, MA, United States) on a gelatin-coated dish. Cells were grown at 37°C in an incubator with 5% CO₂. Uric acid (UA) was dissolved in 1 N sodium hydroxide and diluted in culture media to 3.5, 7, and 10.5 mg/dL. Media containing UA were adjusted to pH 7.4. MSU was dissolved in phosphate-buffered saline and diluted in culture media to 3.5, 7, and 10.5 mg/dL. Cells were seeded on a culture dish and treated with UA- or MSU-containing media for 48 h.

RNA was extracted from UA- or MSU-treated cells or from the buffy coat of patients with gout by using TRIzol Reagent (Thermo Fisher Scientific, Waltham, MA, United States) or the NucleoSpin RNA kit (Macherey-Nagel, Dueren, Germany), according to the manufacturer’s instructions. Two micrograms of RNA were subjected to reverse transcription polymerase chain reaction (RT-PCR) using the Applied Biosystems High-Capacity cDNA Reverse Transcription Kit (Thermo Fisher Scientific, Waltham, MA, United States), according to the manufacturer’s instructions. Real-time polymerase chain reaction (PCR) was performed on the Roche LightCycler 480 Real-Time PCR System using Universal ProbeLibrary System (Roche, Basel, Switzerland). The PCR process involved the following steps: initial denaturation at 94°C for 5 min, followed by 35 cycles of 94°C for 30 s, 60°C for 45 s, 72°C for 1 min, and final extension at 72°C for 7 min. The PCR products were separated by 3% agarose gel electrophoresis. The intensity of the PCR products was analyzed using LabWorks Image Acquisition and Analysis Software (UVP BioImaging Systems, Upland, CA, United States). The primers for RT-PCR and real-time PCR are listed in Supplementary Table 1.

Statistical analysis was performed using Microsoft Excel 2010 (Microsoft, Redmond, WA, United States) and GraphPad Prism 5.01 (GraphPad Software, San Diego, CA, United States). Student’s t-tests were conducted to evaluate differences between groups. A probability of less than 0.05 was considered to be statistically significant. *P < 0.05, **P < 0.01, and ***P < 0.0001.

Proteins were extracted from UA- or MSU-treated cells or from the buffy coat of patients with gout by using RIPA lysis buffer (50 mM Tris, pH 8.0, 150 mM NaCl, 1 mM EDTA, 1% NP40, 1% sodium deoxycholate, 0.1% SDS) and were analyzed through SDS-PAGE: the proteins were transferred onto polyvinylidene difluoride (PVDF) membranes (Bio-Rad, Hercules, CA, United States) after electrophoresis. The PVDF membranes were blocked with 5% bovine serum albumin (BSA) solution in Tris-buffered saline with Tween 20 (TBST) for 60 min. The PVDF membranes were incubated with the primary antibody in 1% BSA overnight and then incubated with horseradish peroxidase–conjugated secondary antibody for 60 min. After washing three times with TBST, the PVDF membranes were exposed to the ECL Plus substrate (GE Healthcare Amersham, Fisher Scientific, Göteborg, Sweden) for 5 min. Chemiluminescence signals were detected, and the expression levels were analyzed using LabWorks Image Analysis Software. The primary antibodies used were anti-PUS1 (Invitrogen, Thermo Fisher Scientific, Waltham, MA, United States), anti-PUS7 (Invitrogen, Thermo Fisher Scientific, Waltham, MA, United States), anti-NSUN3 (Abcam, Cambridge, United Kingdom), anti-NSUN4 (Abcam, Cambridge, United Kingdom), anti-METTL14 (Abcam, Cambridge, United Kingdom), anti-YTHDC1 (Abcam, Cambridge, United Kingdom), anti-YTHDF3 (Abcam, Cambridge, United Kingdom), anti-ADAR1 (Abcam, Cambridge, United Kingdom), anti-RED1 (also known as ADAR2, Abcam), anti-SRPK1 (BD Biosciences, Franklin Lakes, NJ, United States), anti-SRPK2 (BD Biosciences, Franklin Lakes, NJ, United States), anti-SF2/ASF (Zymed Laboratories, South San Francisco, CA, United States), anti-hnRNPA1 (Sigma-Aldrich), anti-Sam68 (Santa Cruz, Dallas, TX, United States), anti-histone H1 (GeneTex, Irvine, CA, United States), anti-ß-actin (GeneTex, Irvine, CA, United States), and anti-tubulin (GeneTex, Irvine, CA, United States).

Blood samples obtained from 67 patients with gout were approved by the China Medical University and Hospital Research Ethics Committee (CMUH108-REC2-051). Approximately 10 mL of blood was collected in an EDTA tube, and buffy coat was used for RNA and protein extraction.

That RNA was of high quality (RNA integrity number, RIN > 8) was confirmed using the Agilent Bioanalyzer 4200 (Agilent Technologies, Santa Clara, CA, United States). One microgram of RNA was used for library preparation of the TruSeq Stranded mRNA Library Prep kit (Illumina, San Diego, CA, United States), according to the manufacturer’s instructions. Briefly, mRNA was purified using poly-A magnetic beads and fragmented through enzyme treatment. Subsequently, double-strand cDNA synthesis, end-repair, adaptor ligation, and enrichment PCR were performed. Samples were subjected to 2 × 150-bp paired-end sequencing using the Illumina NovaSeq 6000 platform.

Base calling and quality scoring were performed with an updated implementation of Real-Time Analysis on the NovaSeq 6000 system. We used bcl2fastq Conversion Software (v2.20.0.422) to demultiplex data and convert BCL files to FASTQ files. To facilitate downstream analysis, a quality control (QC) step was performed that involved the use of Trimmomatic tools (v0.39) (19). Sequencing primers and adapters in the reads were trimmed off, and a read was dropped out if the average quality score was less than 20 or the length of the read was shorter than 100 bp. Transcriptome sequencing reads were aligned to human genome hg38 with reference annotation GENCODE version 30 by using STAR (v2.7.5a) (20). featureCounts (v35) (21) and edgeR (v 3.32.1) (22) were used to quantify and normalize the mRNA expression level. To identify alternative splicing events (ASEs), rMATS (v4.1.0) (23) was applied to alignment results from STAR. A total of five types of ASEs were observed, namely annotated, skipped exon (SE), mutually exclusive exons (MXE), alternative 3′ splice site (A3SS), alternative 5′ splice site (A5SS), and retained intron (RI). To analyze RNA editing, whole genome sequencing (WGS) data were obtained from NCBI SRA (HEK293: SRR2123657, THP-1: SRR8670675); fastp (v0.20.1) (24) was used for QC. WGS reads were aligned to human genome hg38 with reference annotation GENCODE version 30 by using BWA-MEN (v0.7.17) (25) and were then sorted to generate an index by using SAMtools (v 1.9) (26). RedITools (v 1.0) (27) plus the relevant protocol (28) was used for RNA editing. QC was performed by filtering out the editing events with read counts less than 30 and editing ratios less than 0.1.

Epigenomic regulations include DNA and RNA levels, and over 17 and 160 types of modifications have been detected in DNA and RNA, respectively (8, 29, 30). Previous studies have shown that RNA modifications modulate innate immune response and are linked to many human diseases (31, 32). We hypothesized that RNA modifications may be regulated in hyperuricemia and gout. To identify the effect of UA and MSU on RNA modifications, we treated the THP-1 monocyte cell line, HEK293 embryonic kidney cell line, and HUVEC umbilical vein epithelial cell line with UA or MSU for 48 h. These cell lines were selected because the evidence indicates that hyperuricemia and gout are involved in the development of chronic kidney disease and hypertension. In addition, MSU crystal-stimulated monocytes and macrophages secret IL-1 and initiate gouty inflammation. Subsequently, we measured the expression levels of enzymes involved in RNA modifications, including pseudouridine, 5-methylcytosine, and N6-methyladenosine; these enzymes have been associated with human disease-related RNA modifications. We determined that UA and MSU had diverse effects on RNA modifications in different cell lines (Figure 1). The expression of enzymes involved in 5-methylcytosine was affected in THP-1 cells treated with UA (Figure 1A). The effect of MSU on RNA modifications in THP-1 cells was ubiquitous in pseudouridine, 5-methylcytosine, and N6-methyladenosine (Figure 1B). The expressions of the enzymes involved in 5-methylcytosine in UA-treated HEK293 cells changed significantly (Figure 1C). The effect of MSU on RNA modifications in HEK293 cells was ubiquitous in pseudouridine, 5-methylcytosine, and N6-methyladenosine (Figure 1D). UA treatment of HUVEC cells altered the expressions of the enzymes involved in pseudouridine, 5-methylcytosine, and N6-methyladenosine (Figure 1E). MSU treatment of HUVEC cells affected the expressions of the enzymes involved in pseudouridine and N6-methyladenosine (Figure 1F). The results indicated that both UA and MSU affected RNA modifications. We further detected the protein expression levels of RNA modification enzymes that were affected by RNA levels. We found no significant changes of NSUN3 and NSUN4 protein expressions in THP-1 cells treated with UA (Figure 2A). The expression of METTL14 decreased in THP-1 cells treated with MSU; however, no changes were observed in the expressions of YTHDC1 and NSUN3 (Figure 2B). The expression of NSUN3 decreased in HEK293 cells treated with UA or MSU (Figures 2C,D). No significant changes in protein expressions of RNA modification enzymes were observed in HUVEC cells treated with UA or MSU (Figures 2E,F). The quantitative results are presented in Supplementary Figure 1.

RNA editing is another critical mechanism in gene regulation. RNA editing is different from other RNA modifications because it alters the cellular fate of RNA molecules and their sequence relative to the genome. Adenosine to inosine (A-to-I) RNA editing is the most common type of RNA editing in vertebrates (higher eukaryotes). This editing reaction is catalyzed by the adenosine deaminase acting on RNA (ADAR) protein family (33). A-to-I RNA editing has been determined to be involved in many biological functions; one of them is innate immune response (34–36). According to these studies, A-to-I RNA editing may be regulated in hyperuricemia and gout. To identify the effect of UA and MSU on A-to-I RNA editing, we detected the expression levels of enzymes involved in A-to-I RNA editing, including ADAR1 and ADAR2, and found no significant changes of ADAR1 and ADAR2 expressions in THP-1 and HUVEC cells treated with UA or MSU (Figures 3A,B,E,F). The expression of ADAR2 increased in HEK293 cells treated with UA (Figure 3C), and the expressions of ADAR1 and ADAR2 increased in HEK293 cells treated with MSU (Figure 3D). We also detected the protein expressions of ADAR1 and ADAR2 in three cell lines treated with UA or MSU and observed that the expression of ADAR1 increased in THP-1 cells treated with UA and MSU, but no change was observed in the expression of ADAR2 (Figures 4A,B). Increases in the expression of ADAR2 in HEK293 cells treated with UA were observed (Figure 4C), whereas no changes of ADAR1 and ADAR2 in HEK293 cells treated with MSU were observed (Figure 4D). No changes in the expressions of ADAR1 and ADAR2 were observed in HUVEC cells treated with UA (Figure 4E), but the expression of ADAR1 increased in HUVEC cells treated with MSU (Figure 4F). The quantitative results are presented in Supplementary Figure 2. These findings indicate that not only RNA modifications but also RNA editing was affected by UA or MSU treatment. To further explore RNA editing, we analyzed genome-wide RNA editing by transcriptome sequencing in UA- or MSU-treated THP-1 and HEK293 cells. We found 10,264 editing variants in THP-1 cells and 12,189 and 12,819 editing variants in MSU-treated THP-1 cells and UA-treated THP-1 cells, respectively (Figure 5); 6,845 editing variants were commonly observed in untreated THP-1 and MSU-treated THP-1 cells, and 3,419 and 5,344 unique editing variants were observed in untreated THP-1 and MSU-treated THP-1 cells, respectively. A total of 6,995 common variants and 3,269 and 5,824 unique variants were observed in untreated THP-1 and UA-treated THP-1 cells, respectively (Figure 5). A total of 4,222, 5,159, and 4,071 editing variants were observed in untreated, MSU-treated, and UA-treated HEK293 cells, respectively (Figure 6); 2,612 editing variants were common in untreated and MSU-treated HEK293 cells, and 1,610 and 2,547 unique editing variants were observed in untreated THP-1 and MSU-treated HEK293 cells, respectively. In total, 2,607 common variants and 1,615 and 1,464 unique variants were observed in untreated HEK293 and UA-treated HEK293 cells, respectively (Figure 6). Subsequently, we analyzed editing variants correlated with gene expression levels. Four genes exhibited expression level changes with RNA editing variants in MSU-treated THP-1 cells, namely GATD3B, GLO1, IL21R, and SIGLEC6. In UA-treated THP-1 cells, four genes exhibited changed expression levels with RNA editing variants, namely FADS2, IL21R, SIGLEC11, and SIGLEC6. In total, 78 and four genes exhibited changed expression levels in MSU- and UA-treated HEK293 cells, respectively (Supplementary Table 2). We further analyzed the functions of 78 genes in MSU-treated HEK293 cells through the KEGG pathway and found that amino acid metabolism was affected (Table 1). We also analyzed these genes using Enrichr (37–39). Five of the 26 tools in Enrichr showed that amino acid metabolism was affected (see Supplementary Figure 3; more details at https://maayanlab.cloud/Enrichr/enrich?dataset=4f1656f33465251920dc111be4070988). In addition to gene expression, we analyzed RNA splicing changes with RNA editing and found significant changes in alternative splicing of CARD8 with RNA editing on the splice site in MSU- or UA-treated THP-1 cells. Alternative splicing of non-coding RNA AC012313.3 with RNA editing on the splice site induced substantial changes in UA-treated HEK293 cells. These findings suggest that RNA editing was affected by MSU and UA treatment in a genome-wide manner.

In addition to chemical modifications, some other mechanisms are included in epigenomic regulations. For example, mRNA alternative splicing is critical for all cellular processes, as over 90% of all human genes undergo alternative splicing. Many studies have shown that alternative splicing is involved in apoptosis, inflammation, autoimmune diseases, and immune response regulations (40–45), indicating the possibility that alternative splicing is regulated in hyperuricemia and gout. We detected alternative splicing changes of BCL-x, SMAC, and HIPK3 in UA- or MSU-treated cell lines; these genes were involved in apoptosis regulation. We found significant changes of BCL-x splicing in THP-1 cells treated with UA (Figure 7A). The splicing patterns were also affected in UA- or MSU-treated cell lines, although no significance was observed owing to variation in the patterns (Figures 7B–F). We then detected the expression of splicing factors that had been reported to affect alternative splicing in UA- or MSU-treated cell lines and found that SRPK1 decreased in UA-treated THP-1 cells (Figure 8A). The expression of SRPK1, ASF/SF2, and hnRNPA1 increased in MSU-treated THP-1 (Figure 8B). The expressions of SRPK1 and hnRNPA1 decreased in UA-treated HEK293 cells (Figure 8C), and those of the detected splicing factors did not change significantly in MSU-treated HEK293 cells (Figures 8D,F). The expression of hnRNPA1 decreased and that of Sam68 increased in UA-treated HUVEC cells (Figure 8E). The quantitative results are presented in Supplementary Figure 4. These results indicated that the expressions of splicing factors were affected by UA or MSU treatment and subsequently caused alternative splicing changes. We suspected that the effects of UA or MSU on RNA splicing were universal in the genome. Therefore, we analyzed genome-wide splicing changes by transcriptome sequencing in UA- or MSU-treated THP-1 and HEK293 cells. The ASEs were divided into five groups: SE, MXE, A3SS, A5SS, and RI. We found that 447 splicing events significantly changed in UA-treated THP-1 cells and 431 splicing events changed in MSU-treated THP-1 cells. Similarly, 455 and 435 splicing events significantly changed in UA- and MSU-treated HEK293 cells, respectively (Table 2). These results indicate that alternative splicing was affected by UA or MSU treatment universally in the genome.

We observed changes in RNA splicing and the expression of RNA modifications and RNA editing enzymes in patients with gout compared with healthy controls. The patients were divided into three groups: acute gout, intercritical gout, and chronic tophaceous gout. The clinical characteristics of patients with gout are presented in Table 3 and Supplementary Figure 5. The expressions of RNA modification enzymes significantly differed between the three patient groups compared with normal controls (Figure 9). The expressions of RNA editing enzymes ADAR1 and ADAR2 also decreased significantly in patients with gout (Figure 10). Alternative splicing of BCL-x changed in the intercritical gout group. Alternative splicing of SMAC and HIPK3 did not differ significantly between the various patients with gout. We further detected the alternative splicing of TP53 and NFKB1 and identified differences in patients with acute and intercritical gout (Figure 11). The quantitative results are presented in Supplementary Figure 6. These results indicate that RNA modifications, editing, and splicing were also affected in gout and may influence the mechanisms of hyperuricemia and gout.