The significance of m6A-circFOXK2 in CD4⁺T cells from patients with rheumatoid arthritis and circFOXK2 correlates with Th17 % and autophagy

Abstract

Objective

CD4⁺T cells are significant compositions of immune cells in rheumatoid arthritis (RA) and the aberrant function of CD4⁺T cells have been demonstrated to promote RA's progression. N6-methyladenosine (m6A) is a highly enriched modification found in CircRNAs. However, the role of m6A-methylated circRNA in regulation of CD4⁺T cells function in RA and its association with RA disease activity are presently unclear.

Methods

We used m6A-circRNA epitranscriptomic microarray analysis and m6A RNA immunocoprecipitation-quantitative polymerase chain reaction (MeRIP-qPCR) to screen and verify the differentially expressed m6A-methylated circRNAs in CD4⁺T cells from RA and HC.

Results

Many m6A-methylated circRNAs were differentially expressed in patients with new-onset RA. M6A-methylated circFOXK2 in the CD4⁺T cells of patients with new-onset RA was significantly decreased, the expression level of circFOXK2 in the CD4⁺T cells of patients with new-onset RA was significantly increased and correlated with treatment, Th17 %, autophagy. CircFOXK2 could function as competing endogenous RNAs to regulate miR-486-3p expression. In many m6A regulators, only a-ketoglutarate-dependent dioxygenase alkB homolog 5 (ALKBH5) was significantly increased in the CD4⁺T cells of patients with new-onset RA, and its level positively correlated with rheumatoid factor, the expression of circFOXK2. Moreover, the change of ALKBH5 could affect the level of m6A and circFOXK2 in Jurkat cell line.

Conclusion

We indicated a notable decrease in the m6A-methylated level of circFOXK2 in CD4⁺T cells from human RA, indicating a potential role of hypomethylated circFOXK2 in CD4⁺T cells function and RA pathogenesis. This may show valuable insight into the ALKBH5/m6A-circFOXK2/miRNA interaction network and the mechanism of RA.

Abbreviations

Anti-CCP

anti-cyclic citrullinated peptide antibodies

ALKBH5

a-ketoglutarate-dependent dioxygenase alkB homolog 5

CircRNAs

circular RNAs

CRP

C-reactive protein

DAS28

disease activity score

dNLR

derived neutrophil-lymphocyte ratio

ESR

erythrocyte sedimentation rate

FTO

fat mass and obesity-associated protein

HC

healthy controls

HCT

hematocrit

HGB

hemoglobin

L:

lymphocyte count

L%

lymphocyte percentage

LMR

lymphocyte-to-moncyte ratio

M

monocyte count

M%

monocyte percentage

m6A

N6-methyladenosine

m6A

immunocoprecipitation-quantitative polymerase chain reaction (MeRIP-qPCR)

METTL3

methyltransferase-like 3

METTL14

methyltransferase-like 14

MPV

mean platelet volume

N

neutrophil count

N%

neutrophil percentage

NLR

neutrophil-to-lymphocyte ratio

PCT

plateletcrit

PDW

platelet distribution width

PLR

platelet-to-lymphocyte ratio

PLT

platelet count

RA

rheumatoid arthritis

RBC

red blood cell

RF

rheumatoid factors

ROC:

receiver operating characteristic

RT-qPCR

Reverse-transcription quantitative PCR

SII

systemic immune-inflammation index

Th17

IL-17-producing helper T

WBC

white blood cell

WTAP

wilms tumor 1-associating protein

YTHDF2

YT521-B homology domains 2

1. Introduction

Rheumatoid arthritis (RA) is a long-term autoimmune condition with persistent synovitis in limb joints and destruction of the joints, resulting in significant morbidity [1]. RA affects approximately 0.5–1.0 % of adults in Europe and North America [2], while in China, the prevalence of RA is estimated to 0.28–0.45 % [3], which resulting in significant individual burden and substantial socioeconomic costs. However, despite the extensive research conducted on RA, the precise etiology and pathology remain unknown.

Research studies have demonstrated that the aberrant activation of CD4⁺T cells play a pivotal role in the pathogenesis of RA [1]. Hyperactive CD4⁺T cells associated with RA are more resistant to apoptosis, and these CD4⁺T cells with hyperactivation

and resistance to apoptosis can promote autoimmune process [4]. Importantly, increased autophagy in CD4⁺ T cells of RA patients results in T cell hyperactivation and apoptosis resistance [4]. In addition, IL-17-producing helper T (Th17) cells which are considered to be positive regulators of immune responses have important roles in RA development, because they produce pro-inflammatory cytokines, including IL-17A, IL-17F, and IL-22 [5].

Circular RNAs (CircRNAs), which are a kind of noncoding RNA with a covalent closed-loop structure and possess a relatively stable structure, play a crucial role in the progression of various diseases as RNA molecule “sponges”, RNA protein translators, and gene regulators [6]. Our previous studies and other researches have identified that differential expressions of circRNAs play a key role in the pathogenesis of RA [[7], [8], [9], [10], [11]]. However, the causal relationships between circRNAs and the function of CD4⁺T cells in RA remain unknown.

RNA N6-methyladenosine (m6A) is considered as the most common modification in eukaryotic organism [12]. M6A modification plays a critical role in regulating RNA processing, splicing, nucleation, translation, and stability, which is closely associated with numerous physiological and pathological processes [12]. CircRNAs can undergo m6A modification, which involves a methylase that adds the m6A modification, a demethylase that removes it, and an m6A reading protein that recognizes it. Previous studies have showed that m6A modification on human circRNAs can inhibit innate immunity [13]. ALKBH5, as a classic m⁶A demethylase, has been proved to regulate the expression of target genes by affecting the m⁶A level of JARID2 mRNA in fibroblast-like synoviocytes, thus participating in synovial hyperplasia and invasion in RA [14]. However, the potential mechanisms of m6A modification and circRNA regulation RA immune inflammatory response have not yet been elucidated.

We, therefore, aimed to explore the m6A-circRNA epitranscriptomic microarray analysis in CD4⁺T cells from human RA and healthy control (HC). We identified and validated differentially m6A-methylated circRNAs in human RA and HC. In addition, we investigated the clinical value of m6A-circFOXK2 from CD4⁺T cells in RA and found circFOXK2 may involve in Th17 response and autophagy in CD4⁺ T cells. Further, ALKBH5 could regulate the level of m6A and the expression of circFOXK2. Moreover, we constructed a ALKBH5/m6A-circFOXK2/miRNA interaction network. In this paper, we further explored m6A-circFOXK2 regulation of CD4⁺T cells function in RA, which leads to the occurrence and development of RA.

2. Materials and methods

2.1. Subjects and samples

102 RA patients were recruited from the First Affiliated Hospital of Nanchang University as subjects in the this study. The RA patients consecutively enrolled herein all fulfilled the revised American College of Rheumatology (ACR) 2010 criteria for RA [15]. Among them, 65 patients were new-onset RA that first time diagnosis of RA and no history of immunosuppressive drugs or corticosteroids use before recruitment. In addition, 71 HC without other chronic disorders and without relation to patients of autoimmune disease were randomly recruited from from the First Affiliated Hospital of Nanchang University. Among, 12 new-onset RA patients and 12 age-matched and sex-matched HC were enrolled for microarray measurement after immunoprecipitate with anti-m6A antibody and m6A immunocoprecipitation-quantitative polymerase chain reaction (MeRIP-qPCR). Other new-onset RA patients, HC and 37 revisiting RA specimen were used to detect circRNA and m6A regulatory factors by reverse-transcription quantitative PCR (RT-qPCR). The participants characteristics are detailed in Table 1. There was no significant difference in the age and gender between the three groups. The study had approval from the Ethics Committee of the First Affiliated Hospital of Nanchang University ((2023)CDYFYYLK(01–035)) and complied with the Helsinki Declaration.

Table 1.

Clinical details of the patients with new-onset RA, revisiting RA and HC.

Clinical characteristic	new-onset RA (65)	HC (71)	revisiting RA (37)
Sex, female/male	48/17	50/21	26/11
Age, years	54.55 ± 11.06	52.04 ± 11.66	55.54 ± 10.95
Time from diagnosis, months	12 [3, 48]	–	–
Time from ongoing therapies, months			16 [11.88, 25.25]
DAS28	5.09 ± 1.00	–	3.49 ± 1.35#
RF, IU/ml	287.24 ± 638.34	–	–
Anti-CCP, U/ml	360.14 ± 1032.46	–	–
ESR, mm/h	40.84 ± 30.97	–	35.27 ± 21.00
CRP, mg/l	24.67 ± 29.18	–	22.38 ± 30.50
WBC, 109/l	7.40 ± 2.29	5.91 ± 1.10*	8.54 ± 2.50#
RBC, 1012/l	4.15 ± 0.47	4.66 ± 0.33*	4.11 ± 0.50
HGB, g/l	118.47 ± 16.61	140.70 ± 9.89*	122.36 ± 17.96
HCT, l/l	0.37 ± 0.05	0.42 ± 0.03*	0.38 ± 0.05
PLT, 109/l	313.48 ± 109.09	228.86 ± 51.20*	288.86 ± 89.05
MPV, fl	9.90 ± 0.97	10.54 ± 0.73*	9.47 ± 1.23
PCT, %	0.30 ± 0.10	0.24 ± 0.05*	0.27 ± 0.07
PDW, fl	14.46 ± 2.54	12.66 ± 1.72*	15.67 ± 1.55#
L, 109/l	1.68 ± 0.56	2.04 ± 0.46*	1.62 ± 0.83
L%	24.05 ± 7.25	34.19 ± 7.40*	19.66 ± 8.76#
M, 109/l	0.44 ± 0.16	0.39 ± 0.11	0.53 ± 0.22
M%	6.10 ± 1.60	6.65 ± 1.36*	6.39 ± 2.49
N, 109/l	5.12 ± 2.01	3.32 ± 0.89*	6.26 ± 2.17#
N%	67.75 ± 8.65	56.40 ± 6.98*	72.39 ± 9.91#
LMR	4.06 ± 1.46	5.42 ± 1.41*	3.35 ± 1.58#
NLR	3.35 ± 1.69	1.70 ± 0.55*	4.44 ± 1.99#
PLR	209.30 ± 109.46	116.53 ± 31.39*	204.42 ± 85.00
SII	1100.75 ± 819.86	390.48 ± 157.05*	1302.21 ± 762.28
dNLR	2.35 ± 1.03	1.34 ± 0.40*	3.03 ± 1.29#

∗P < 0.0500 new-onset RA compared to HC. ^#P < 0.0500 new-onset RA compared to revisiting RA.

Anti-CCP: anti-cyclic citrullinated peptide antibodies, CRP: C-reactive protein, DAS28: disease activity score, dNLR: derived neutrophil-lymphocyte ratio, ESR: erythrocyte sedimentation rate, HC: health control, HCT: hematocrit, HGB: hemoglobin, L: lymphocyte count, L%: lymphocyte percentage, LMR: lymphocyte-to-moncyte ratio, M: monocyte count, M%: monocyte percentage, MPV: mean platelet volume, N: neutrophil count, N%: neutrophil percentage, NLR: neutrophil-to-lymphocyte ratio, PCT: plateletcrit, PDW: platelet distribution width, PLR: platelet-to-lymphocyte ratio, PLT: platelet count, RA: rheumatoid arthritis, RBC: red blood cell, RF: rheumatoid factors, SII: systemic immune-inflammation index, WBC: white blood cell.

2.2. Isolation of peripheral blood CD4⁺T cells

Peripheral blood mononuclear cells (PBMC) was isolated from EDTA anticoagulant venous blood by a Ficoll-density gradient (Invitrogen Bio, Waltham, MA). Purified CD4⁺ T cells of PBMC from RA patients and HC were isolated by magnetic beads (Miltenyibiotec, Germany). Then, the CD4⁺T cells purity was analyzed by flow cytometry and the purity could reach 95 %.

2.3. m6A-circRNA epitranscriptomic microarray analysis

The sample preparation and microarray hybridization were performed based on the Arraystar's standard protocols. Briefly, the total RNAs were immunoprecipitated with anti-N6-methyadenosine (m6A) antibody. The modified RNAs were eluted from the immunoprecipitated magnetic beads as the “IP”. The unmodified RNAs were recovered from the supernatant as “Sup”. The “IP” and “Sup” RNAs were treated with RNase R, and then labelled with Cy5 and Cy3 respectively as cRNAs in separate reactions using Arraystar Super RNA Labeling Kit. The cRNAs were combined together and hybridized onto Arraystar Human circRNA epitranscriptomic Microarray (8 × 15K, Arraystar). After washing the slides, the arrays were scanned in two-color channels by an Agilent Scanner G2505C. Agilent Feature Extraction software (version 11.0.1.1) was used to analyze acquired array images. Raw intensities of IP (immunoprecipitated, Cy5-labelled) and Sup (supernatant, Cy3-labelled) were normalized with average of log2-scaled Spike-in RNA intensities. After Spike-in normalization, the probe signals having Present (P) or Marginal (M) QC flags in at least 3 out of 6 samples were retained for further “m6A methylation level” and “m6A quantity” analyses. “m6A methylation level” was calculated for the percentage of modification based on the IP (Cy5-labelled) and Sup (Cy3-labelled) normalized intensities. “m6A quantity” was calculated for the m6A methylation amount based on the IP (Cy5-labelled) normalized intensities. Differentially m6A-methylated circRNAs between two comparison groups were identified by filtering with the fold change and statistical significance (P-value) thresholds. Hierarchical Clustering was performed to show the distinguishable m6A-methylation pattern among samples. M6A-circRNA epitranscriptomic microarray analysis were executed by KangChen Bio-tech (Shanghai, China).

2.4. m6A immunocoprecipitation-quantitative polymerase chain reaction (MeRIP-qPCR)

MeRIP-qPCR was executed by KangChen Bio-tech (Shanghai, China). Briefly, the purified polyadenylated mRNA from CD4⁺T cells was incubated with anti-m6A antibody (Synaptic Systems) or rabbit IgG in immunoprecipitating (IP) wash buffer for 2 h at 4 °C. Protein A beads (Synaptic Systems) were added and incubated at 4 °C for 2 h. After incubation, the immunoprecipitated beads-m 6A antibody-mRNA complex was extensively washed with IP wash buffer. TRIzol reagent (Sigma) was added to elute m6A nucleotide. The SuperScriptTM III Reverse Transcriptase (Invitrogen Bio, Waltham, USA) was used for cDNA synthesis. All qPCRs were run on ViiA 7 Real-time PCR System (Applied Biosystems) using 2 × PCR master mix (Arraystar:AS-MR-006-5).

2.5. Reverse-transcription quantitative PCR (RT-qPCR)

After evaluating the CD4⁺T cells purity of sample, RNA was extracted from CD4⁺T cells using 0.75 mL TRIzol reagent (Invitrogen Bio, Waltham, USA) according to the manufacturers' instructions. The purity and concentration of isolated total RNA was evaluated in NanoDrop ND-1000 spectrophotometer (Invitrogen Bio, Waltham, USA) by A260/A230, A260/A280 ratios. cDNA was obtained by PrimeScript™ RT reagent kit (Takara Bio, Tokyo, Japan) following the manufacturers’ instructions. The expression levels of hsa_circ_0001931, circFOXK2 (hsa_circ_0000817), and m6A regulatory factors were quantified by SYBR Green detection method (Takara Bio, Tokyo, Japan) and analyzed onto ABI 7500 Real-time PCR System (Invitrogen Bio, Waltham, USA). The primers of genes in this study were listed as follows:

hsa_circ_0001931, forward, 5′ AACTGGAAGAAAGTGTGGGCAAT 3′, reverse, 5′AATGTAGTCTAGCTGCAAACACC3′;

circFOXK2 (hsa_circ_0000817), forward, 5′TACTACAGGACTGCGGACAAGG3′, reverse, 5′ATCTTGATGTTTGTGCTCGGG3′;

methyltransferase-like 3 (METTL3), forward, 5′AAGCTGCACTTCAGACGAAT3′, reverse, 5′GGAATCACCTCCGACACTC3′;

methyltransferase-like 14 (METTL14), forward, 5′AGAAACTTGCAGGGCTTCCT3′, reverse, 5′ TCTTCTTCATATGGCAAATTTTCTT3′;

wilms tumor 1-associating protein (WTAP), forward, 5′GGCGAAGTGTCGAATGCT3′, reverse, 5′ CCAACTGCTGGCGTGTCT3′;

ALKBH5, forward, 5′CCCGAGGGCTTCGTCAACA3′, reverse, 5′ CGACACCCGAATAGGCTTGA3′;

fat mass and obesity-associated protein (FTO), forward, 5′TGGGTTCATCCTACAACGG3′, reverse, 5′CCTCTTCAGGGCCTTCAC 3′;

YT521-B homology domains 2 (YTHDF2), forward, 5′GGCAGCACTGAAGTTGGG3′, reverse, 5′CTATTGGAAGCCACGATGTTA 3′;

GAPDH, forward, 5′TGCACCACCAACTGCTTAGC3′, reverse, 5′GGCATGGACTGTGGTCATGAG3′. The results of above genes were counted using the 2^−ΔΔCt method [16], using GAPDH as the housekeeping gene.

2.6. Flow cytometry analysis

After CD4⁺T cells from clinical participants were collected, fluorescent-labelled CD4 antibody (Beckman Coulter, Brea, CA, USA) was used to label cells for isolation by flow cytometry (Beckman Coulter, Brea, CA, USA).

For detection of IL-17, cells were incubated with Cell Stimulation Cocktail (BD Bioscience, USA) for 5 h, resuspended in fixation/permeabilization solution (BD Bioscience, USA) and stained with IL-17 (BD Bioscience, USA). Labelled cells were isolated by flow cytometry or measured on CYTOMICS FC 500 flow cytometer (Beckman Coulter, Brea, CA, USA) and analyzed with CXP software programs.

For detection of autophagy, PBMC were incubated with Autophagy Green™ working solution (AAT Bioquest bio. California, USA) in 37 °C, 5 % CO₂ incubator for 40 min. Wash the cells with Wash Buffer for 3 times and PBS for 1 times, then incubated with 2 μl APC-labelled CD4 (Beckman Coulter, Brea, CA, USA) in 4 °C for 20 min. All flow samples were analyzed with a CYTOMICS FC 500 flow cytometer (Beckman Coulter, Brea, CA, USA) and associated software programs (CXP).

2.7. Western blot

The expressions of ALKBH5, and β-actin proteins were analyzed by Western blot. CD4⁺T cells were lysed in RIPA lysis buffer (Solarbio, Beijing, China). Protein concentrations were measured using a BCA protein assay kit (Solarbio, Beijing, China). Equal amounts of protein (30 μg) were loaded onto 10 % SDS-PAGE gels. Separated proteins were transferred to PVDF membranes membrane (Millipore, Billerica, USA); subsequently, the membranes were blocked with 5 % non-fat milk for 2 h and probed with antibodies against ALKBH5 (Abcam, Cambridge, USA), or β-actin (immunoway, Texas, USA) overnight at 4 °C. Thereafter, the membranes were washed with TBS-Tween buffer and further incubated with an HRP-conjugated goat anti-rabbit IgG Ab (absin, Shanghai, China) at room temperature for 2 h. Protein bands were visualized on luminescent imager (BIO-RAD, USA) by an enhanced chemiluminescence detection kit (Thermo Fisher Scientific, Waltham, USA) and evaluated the band density on Image J software.

2.8. Cell culture

Jurkat cell line was cultured in RPMI 1640 medium with 10 % FBS (Sigma-Aldrich, USA), 100 U/mL penicillin and 100 μg/mL streptomycin and incubated at 37 °C in a humidified atmosphere with 5 % CO₂. The third generation of Jurkat cell line was used to perform experiment.

2.9. Cell transfection

The double-strand oligonucleotides corresponding to the target sequences were synthesized by Jikai Gene (Shanghai, China). The following sequences were targeted to human ALKBH5 small interfering RNA (siRNA): ALKBH5-1: 5′-CATCGTGTCCGTGTCCTTCTT-3′; ALKBH5-2: 5′-TAGCTTCAGCTCTGAGAACTA-3′; ALKBH5-3: 5′-ACGGATCCTGGAGATGGACAA-3′ and NC (negative control) siRNA: 5′-TTCTCCGAACGTGTCACGT-3′. The lentiviral vectors expressing shRNA targeting ALKBH5 and NC (named sh-ALKBH5 and sh-NC), and the ALKBH5 or NC-lentiviral expression vector (named oe-ALKBH5 or oe-NC) were provided by Jikai Gene (Shanghai, China). Jurkat cell line were seeded onto plates for 24 h and then transfected using the above lentivirals following the standard protocol.

2.10. m⁶A RNA methylation analysis

Total RNA that was isolated from sh-ALKBH5, sh-NC, oe-ALKBH5, and oe-NC was used to detect m⁶A RNA methylation by The EpiQuik™ m⁶A RNA Methylation Quantification Kit (Colorimetric) according to the manufacturer's protocol.

2.11. Statistical analysis

All data are presented in forms of means ± standard error. For the analyses, statistical comparisons were performed using unpaired t-test or Mann-Whitney U test among two groups according to the normality. The Spearman method was used for correlation analysis. And receiver operating characteristic (ROC) curve was used to evaluate the predictive value for RA. All data were analyzed using SPSS v.16.0 (SPSS Bio, Chicago, USA) and Prism v.5.0 (GraphPad Bio, San Diego, USA) software. Prism v.5.0 (GraphPad Bio, San Diego, USA) software was also used to plot figures. P-value ≤0.05 was considered to be a statistically significant difference.

3. Results

3.1. Dysregulated m6A-methylated circRNAs expression profiling in CD4⁺T cells from new-onset RA patients

To explore the differentially expressed m6A-methylated circRNAs in RA, CD4⁺T cells samples from 12 patients with new-onset RA and 12 age- and sex-matched HC were selected to perform immunoprecipitate with anti-m6A antibody and microarray analysis using an Arraystar Human circRNA epitranscriptomic Microarray (8 × 15K, Arraystar). Differentially m6A-methylated circRNAs based on "m6A methylation level" and "m6A quantity" passing fold change (>1.5) and statistical significance cutoffs (P < 0.0500) were identified and compiled (Fig. 1A and B), 71 m6A-methylated circRNAs (5 were upregulated and 66 were downregulated) based on "m6A methylation level" were differentially expressed between the patients with new-onset RA and HC, and 204 m6A-methylated circRNAs (195 were upregulated and 9 were downregulated) based on "m6A quantity" were differentially expressed between the patients with new-onset RA and HC (Supplementary data). The top 10 dysregulated m6A-methylated circRNAs were listed in Table 2. Pathway Analysis was constructed to group the m6A-methylated circRNAs based on "m6A methylation level" and "m6A quantity" among the samples, respectively (Fig. 1C and D), and pathway analysis of these discrepant m6A-circRNAs showed they focused on cGMP-PKG signaling pathway, RNA transport et al.

Fig. 1.

Determination of the m6A-methylated circRNA expression profiles in CD4⁺T cells from 12 new-onset RA patients and 4 HCs by microarray analysis.

(A) Heatmap of differentially expressed m6A-methylated circRNAs. (B) Volcano Plot visualization of the distributions of a dataset for the m6A-methylated circRNAs profiles. (C) Pathway Analysis was constructed to group the m6A-methylated circRNAs based on "m6A methylation level". (D) Pathway Analysis was constructed to group the m6A-methylated circRNAs based on "m6A quantity". HC, healthy controls; m6A, N6-methyadenosine; RA, rheumatoid arthritis.

Table 2.

The top 10 dysregulated m6A-methylated circRNAs based on "m6A methylation level" and "m6A quantity".

items	circRNA	Alias	Spliced length	Fold change	P value	source	circRNA_type
m6A methylation level (Down)	hsa_circRNA_008640	hsa_circ_0008640	855	0.38	0.0211	circBase	exonic
	hsa_circRNA_404215	/	/	0.42	0.0037	25242744	exonic
	hsa_circRNA_005322	hsa_circ_0005322	329	0.43	0.0286	circBase	exonic
	hsa_circRNA_102239	hsa_circ_0000817	490	0.50	0.0172	circBase	exonic
	hsa_circRNA_100644	hsa_circ_0019083	756	0.52	0.0492	circBase	exonic
	hsa_circRNA_003653	hsa_circ_0003653	951	0.54	0.0298	circBase	exonic
	hsa_circRNA_101930	hsa_circ_0006105	451	0.55	0.0287	circBase	exonic
	hsa_circRNA_104874	hsa_circ_0088058	603	0.55	0.0161	circBase	exonic
	hsa_circRNA_101226	hsa_circ_0029634	246	0.55	0.0287	circBase	exonic
	hsa_circRNA_103657	hsa_circ_0069977	308	0.55	0.0120	circBase	exonic
m6A methylation level (Up)	hsa_circRNA_102097	hsa_circ_0044195	3672	1.91	0.0488	circBase	exonic
	hsa_circRNA_405048	/	/	1.73	0.0003	25070500	exonic
	hsa_circRNA_001931	hsa_circ_0001931	390	1.64	0.0158	circBase	exonic
	hsa_circRNA_405427	/	/	1.57	0.0490	25070500	intronic
	hsa_circRNA_001596	hsa_circ_0001596	84	1.51	0.0477	circBase	sense overlapping
m6A quantity (Down)	hsa_circRNA_401849	/	/	0.44	0.0194	25242744	exonic
	hsa_circRNA_003022	hsa_circ_0003022	632	0.61	0.0204	circBase	exonic
	hsa_circRNA_080887	hsa_circ_0080887	1264	0.61	0.0162	circBase	exonic
	hsa_circRNA_021417	hsa_circ_0021417	189	0.63	0.0316	circBase	exonic
	hsa_circRNA_004738	hsa_circ_0004738	354	0.64	0.0366	circBase	exonic
	hsa_circRNA_400094	hsa_circ_0092347	400	0.64	0.0150	circBase	intronic
	hsa_circRNA_005322	hsa_circ_0005322	329	0.65	0.0413	circBase	exonic
	hsa_circRNA_104935	hsa_circ_0003362	379	0.65	0.0041	circBase	exonic
	hsa_circRNA_406741	/	/	0.66	0.0039	25070500	exonic
m6A quantity (UP)	hsa_circRNA_407033	/	/	4.88	0.0379	25070500	exonic
	hsa_circRNA_007211	hsa_circ_0007211	4517	4.38	0.0306	circBase	sense overlapping
	hsa_circRNA_000425	hsa_circ_0001563	355	3.23	0.0213	circBase	sense overlapping
	hsa_circRNA_001931	hsa_circ_0001931	390	2.91	0.0032	circBase	exonic
	hsa_circRNA_406535	/	/	2.82	0.0013	25070500	sense overlapping
	hsa_circRNA_104160	hsa_circ_0077495	594	2.72	0.0330	circBase	exonic
	hsa_circRNA_102097	hsa_circ_0044195	3672	2.68	0.0337	circBase	exonic
	hsa_circRNA_009181	hsa_circ_0009181	11334	2.59	0.0016	circBase	exonic
	hsa_circRNA_104223	hsa_circ_0078277	476	2.58	0.0370	circBase	exonic
	hsa_circRNA_103393	hsa_circ_0001313	468	2.52	0.0040	circBase	exonic

3.2. Validation of m6A-methylated circRNAs expression

To verify the microarray data after immunoprecipitate with anti-m6A antibody, 4 m6A-methylated circRNAs (hsa_circ_0001931, hsa_circ_0001313, hsa_circ_0005322 and circFOXK2) were selected for validation via MeRIP-qPCR in 12 patients with new-onset RA and 12 HC. The 4 selected circRNAs were listed in the top 10 (fold change≥2, statistical significance cutoff P < 0.0300) dysregulated m6A-methylated circRNAs, the circRNA_type for 4 selected circRNAs was exonic, and the spliced length for these circRNAs was less than 500bp (Table 2). Since there were no primers that met the requirements, hsa_circ_0005322 was not amplified. Consistent with the results of the microarray analysis after immunoprecipitate with anti-m6A antibody, the m6A level of hsa_circ_0001931 in the CD4⁺T cells of patients with new-onset RA was significantly increased compared with those of the HC (P = 0.0173; Fig. 2A), the m6A level of circFOXK2 in the CD4⁺T cells of patients with new-onset RA was significantly decreased compared with those of the HC (P = 0.0322; Fig. 2B), while the m6A level of hsa_circ_0001313 did not exhibit any remarkable differences between patients with new-onset RA and the HC (P = 0.3011; Fig. 2C).

Fig. 2.

Determination of the relative m6A levels of circRNAs in CD4⁺T cells from 12 patients with RA and 12 HC by MeRIP-qPCR. (A) The m6A level of hsa_circ_0001931 in the CD4⁺T cells of patients with RA was significantly increased compared with those of the HC. (B) The m6A level of circFOXK2 in the CD4⁺T cells of patients with RA was significantly decreased compared with those of the HC. (C) The m6A level of hsa_circ_0001313 did not exhibit any remarkable differences between patients with RA and the HC. (D) SRAMP-predicted mammalian m6A site of hsa_circ_0001931. (E) SRAMP-predicted mammalian m6A site of circFOXK2. (F) SRAMP-predicted mammalian m6A site of hsa_circ_0001313. HC, healthy controls; m6A, N6-methyadenosine; MeRIP-qPCR, m6A RNA immunocoprecipitation-quantitative polymerase chain reaction; RA, rheumatoid arthritis.

Evidences from previous studies have shown that m6A predominantly occurs on the consensus motif sites “RRm6ACH” (R = G or A; H = A, C, or U) in the mammalian transcriptome [17]. SRAMP was utilized to explore the mammalian m6A site of the three m6A-methylated circRNA transcripts including hsa_circ_0001931, circFOXK2, and hsa_circ_0001313. Our analysis indicated moderatly confident GGm6ACT sites in the m6A hypermethylated hsa_circ_0001931 (Fig. 2D), very highly confident GGm6ACT sites and highly confident GGm6ACA sites in the m6A hypermethylated circFOXK2 (Fig. 2E), highly confident GAm6ACT/AAm6ACT/TGm6ACT/GGm6ACC sites in the m6A hypermethylated hsa_circ_0001313 (Fig. 2E).

3.3. The expression levels of hsa_circ_0001931, circFOXK2 in CD4⁺T cells from RA patients and the clinical value of circFOXK2

The aforementioned results demonstrate that the m6A levels of hsa_circ_0001931 and circFOXK2 in the CD4⁺T cells were differentially expressed between the patients with new-onset RA and HC. Considering that m6A modification of circRNA can affect the expression level of circRNA, the expression levels of hsa_circ_0001931 and circFOXK2 in the CD4⁺T cells from patients with new-onset RA and HC were explored. Data showed that the expression level of hsa_circ_0001931 did not exhibit any remarkable differences between patients with new-onset RA and the HC (P = 0.3914; Fig. 3A), while the expression level of circFOXK2 in the CD4⁺T cells of patients with new-onset RA was significantly increased compared with those of the HC (P = 0.0169; Fig. 3B). Whereafter, we analyzed the expression level of circFOXK2 in the CD4⁺T cells from patients with revisiting and new-onset RA, and result indicated that the expression level of circFOXK2 in the CD4⁺T cells of patients with new-onset RA was significantly increased compared with those of the revisiting RA (P = 0.0009; Fig. 3C).

Fig. 3.

The expression levels of hsa_circ_0001931, circFOXK2 in CD4+T cells from RA patients and the clinical value of circFOXK2 (A) The expression level of hsa_circ_0001931 did not exhibit any remarkable differences between patients with new-onset RA and the HC. (B) The expression level of circFOXK2 in the CD4⁺T cells of patients with new-onset RA was significantly increased compared with those of the HC. (C) The expression level of circFOXK2 in the CD4⁺T cells of patients with new-onset RA was significantly increased compared with those of the revisiting RA. (D) ROC analysis of circFOXK2 for distinguishing new-onset RA from HC. (E) ROC analysis of circFOXK2 for distinguishing new-onset RA from revisiting RA. HC, healthy controls; RA, rheumatoid arthritis; ROC, receiver operating characteristic.

Next, we investigated whether circFOXK2 could be used as a new diagnostic marker of new-onset RA using the ROC curve analysis. Our data indicated that the area under the ROC curve (AUC) for distinguishing new-onset RA from HC was up to 0.631 [95 % CI = 0.528–0.734; P = 0.0168], with a cutoff value of >0.9918, a sensitivity of 69.81 %, and a specificity of 52.54 % (Fig. 3D). The AUC for distinguishing new-onset RA from revisiting RA was up to 0.706 [95 % CI = 0.594–0.818; P = 0.0009], with a cutoff value of >0.7873, a sensitivity of 84.91 %, and a specificity of 56.76 % (Fig. 3E).

3.4. The expression level of circFOXK2 in CD4⁺T cells from RA patients correlated with Th17 % and autophagy

To investigate the effect of circFOXK2 on CD4⁺T cells function, we detected the percentage of Th17 (Th17 %), autophagy level of CD4⁺T cells in RA patients using flow cytometry (Fig. 4A and D) and assessed the correlation between Th17 %, autophagy level and the expression level of circFOXK2 in CD4⁺T cells from RA patients. Data showed that Th17 % (P = 0.0060; Fig. 4B) and autophagy level of CD4⁺T cells (P < 0.0001; Fig. 4C) in RA patients were significantly high than that in HC, and the expression levels of circFOXK2 in CD4⁺T from RA patients positively correlated with Th17 % (r_s = 0.5087, P = 0.0094; Fig. 4E) and autophagy level of CD4⁺T cells (r_s = 0.4939, P = 0.0030; Fig. 4F).

Fig. 4.

The increased level of circFOXK2 in CD4⁺T cells from RA patients correlated with Th17 % and autophagy level (A) Representative dot plots of population gating and Th17 % in RA patients and HC by flow cytometry analysis. (B) The Th17 % in RA patients was significantly high than that in HC. (C) The expression levels of circFOXK2 in CD4⁺T cells from RA patients positively correlated with Th17 %. (D) Representative dot plots of population gating and autophagy level of CD4⁺T cells in RA patients and HC by flow cytometry analysis. (E) The autophagy level in RA patients was significantly high than that in HC. (F) The expression levels of circFOXK2 in CD4⁺T cells from RA patients positively correlated with autophagy level. HC, healthy controls; RA, rheumatoid arthritis.

3.5. CircFOXK2/miRNA interaction analysis

It is well-known that circRNAs function as competing endogenous RNAs to regulate miRNA expression, resulting in the occurrence and progression of disease. In this study, to confirm the function of circFOXK2, potential miRNA targets of circFOXK2 were predicted by an online database (RegRNA 2.0, http://regrna2.mbc.nctu.edu.tw/). MiR-486-3p, miR-484, miR-410-5p, miR-409-3p, miR-34c-5p, miR-34b-5p, miR-34a-5p, miR-335-3p, miR-23a-5p, and miR-762, these ten putative miRNAs that could potentially bind to circFOXK2 were predicted. Then, CircInteractome (https://circinteractome.nia.nih.gov/) was used to verified the ten putative miRNAs that could potentially bind to circFOXK2, and results showed miR-486-3p, miR-409-3p may be the targets of circFOXK2. And miR-486-3p has been reported to be involved in the regulation of RA [[18], [19], [20]].

3.6. The expression level of ALKBH5 in CD4⁺T cells from RA patients and its association with the activity

M6A regulatory factors include m6A methyltransferases (METTL3, METTL14, WTAP), m6A demethylases (ALKBH5, FTO) and m6A "readers" (YTHDF2), thus these m6A regulatory factors in CD4⁺T cells were analyzed by RT-qPCR. Data showed that the expression level of ALKBH5 was significantly elevated in patients with new-onset RA compared to HC (P = 0.0001; Fig. 5A). No significant difference was observed in expression level of FTO, WTAP, METTL3, METTL14, or YTHDF2 in CD4⁺T cells between new-onset RA individuals and HC (P > 0.0500; Fig. 5B–F). In addition, Western blot showed that the expression of ALKBH5 in patients with new-onset RA was significantly elevated than that in HC (P = 0.0357; Fig. 5G and H).

Fig. 5.

The expression level of ALKBH5 in CD4⁺T cells from RA patients and its association with the activity (A) The expression level of ALKBH5 was significantly elevated in patients with new-onset RA compared to HC. (B) No significant difference was observed in expression level of FTO in CD4⁺T cells between new-onset RA individuals and HC. (C) No significant difference was observed in expression level of WTAP in CD4⁺T cells between new-onset RA individuals and HC. (D) No significant difference was observed in expression level of METTL3 in CD4⁺T cells between new-onset RA individuals and HC. (E) No significant difference was observed in expression level of METTL14 in CD4⁺T cells between new-onset RA individuals and HC. (F) No significant difference was observed in expression level of YTHDF2 in CD4⁺T cells between new-onset RA individuals and HC. (G) The protein expression level of ALKBH5 in CD4⁺T cells from RA individuals and HC was detected by western blot. (H) The expression level of ALKBH5 in CD4⁺T cells from RA individuals was significantly increased than that in HC. (I) The expression level of ALKBH5 in CD4⁺T cells was positively associate with RF. (J) The expression level of ALKBH5 in the CD4⁺T cells of patients with new-onset RA was significantly increased compared with those of the revisiting RA. ALKBH5, a-ketoglutarate-dependent dioxygenase alkB homolog 5; FTO, fat mass and obesity-associated protein; HC, healthy controls; METTL3, methyltransferase-like 3; METTL14, methyltransferase-like 14; RA, rheumatoid arthritis; WTAP, wilms tumor 1-associating protein; YTHDF2, YT521-B homology domains 2.

In addition, the association between the expression level of ALKBH5 in the CD4⁺T cells and the activity indicators including DAS28, ESR, CRP, CCP, RF, treatment was explored. Result demonstrated that the expression level of ALKBH5 in CD4⁺T cells was positively associate with RF in patients with new-onset RA (r_s = 0.3235, P = 0.0302; Fig. 5I), and the expression level of ALKBH5 in the CD4⁺T cells of patients with new-onset RA was significantly increased compared with those of the revisiting RA (P = 0.0006; Fig. 5J).

3.7. The expression level of circFOXK2 in CD4⁺T cells from RA patients correlated with m6A regulatory factors

Considering the decreased m6A level of circFOXK2 in the CD4⁺T cells of patients with new-onset RA and the fact that ALKBH5 can reduce the m6A level of RNA and affect the expression level of RNA, the correlation between the expression level of circFOXK2 and ALKBH5, YTHDF2 was explored. Data showed that the expression level of circFOXK2 positively correlated with the expression level of ALKBH5 (r_s = 0.5458, P < 0.0001; Fig. 6A) and YTHDF2 (r_s = 0.5090, P = 0.0002; Fig. 6B) in the CD4⁺T cells of patients with new-onset RA.

Fig. 6.

The expression level of circFOXK2 in CD4⁺T cells from RA patients correlated with m6A regulatory factors (A) The expression level of circFOXK2 positively correlated with the expression level of ALKBH5 in the CD4⁺T cells of patients with new-onset RA. (B) The expression level of circFOXK2 positively correlated with the expression level of YTHDF2 in the CD4⁺T cells of patients with new-onset RA. ALKBH5, a-ketoglutarate-dependent dioxygenase alkB homolog 5; HC, healthy controls; RA, rheumatoid arthritis; YTHDF2, YT521-B homology domains 2.

3.8. ALKBH5 could affect the level of m6A and circFOXK2 in Jurket cell line

To further identify the role of ALKBH5 in regulating the level of m6A and circFOXK2, we generated RNAi against ALKBH5 (sh-ALKBH5) and over-expression ALKBH5 (oe-ALKBH5) to infect Jurkat cell line. RT-qPCR and Western blot showed that sh-ALKBH5 significantly disturbed ALKBH5 expression, while oe-ALKBH5 significantly enhanced ALKBH5 expression (P < 0.0100, Fig. 7A and B). Then, we used sh-ALKBH5 and oe-ALKBH5 to explore the role of ALKBH5 in regulating the level of m6A and circFOXK2, and results showed that overexpression of ALKBH5 significantly declined the level of m6A and enhanced the expression of circFOXK2, while depletion of ALKBH5 markedly increased the level of m6A and reduced circFOXK2's expression (P < 0.0500, Fig. 7C and D).

Fig. 7.

ALKBH5 could affect the expression of circFOXK2 in Jurkat cell line (A) RT-qPCR showed that sh-ALKBH5 significantly disturbed ALKBH5 expression, while oe-ALKBH5 significantly enhanced ALKBH5 expression. (B) Western blot showed that sh-ALKBH5 significantly disturbed ALKBH5 expression, while oe-ALKBH5 significantly enhanced ALKBH5 expression. (C) Overexpression of ALKBH5 significantly enhanced the expression of circFOXK2, while depletion of ALKBH5 markedly reduced this expression. (D) overexpression of ALKBH5 significantly declined the level of m6A, while depletion of ALKBH5 markedly increased the level of m6A. ALKBH5, a-ketoglutarate-dependent dioxygenase alkB homolog 5; m6A, N6-methyladenosine; Reverse-transcription quantitative PCR (RT-qPCR).

4. Discussion

The present study determined the levels and quantities of m6A methylated circRNAs in CD4⁺T cells from human RA and HC by m6A-CircRNA epitranscriptomic microarray. We identified and validated a significant decrease in the m6A-methylation of circFOXK2 in RA. Moreover, we found an obvious rise in the level of circFOXK2 and the increased expression of circFOXK2 was associated with treatment. Correlation analysis indicated a positive relationship between the level of circFOXK2 and CD4⁺T cells autophagy, Th17 %. Additionally, m6A methylation level was altered upon ALKBH5 intervention. ALKBH5 may be involved in the methylation process of circFOXK2 in CD4⁺T cells from RA. The overexpression or interference of ALKBH5 could increase or decrease circFOXK2 expression. Further, online database (RegRNA 2.0 and CircInteractome) analysis predicted an RA interaction network of circFOXK2/miR-486-3p or miR-409-3p. The findings from our study suggest that the hypomethylation of circFOXK2 may affect the progression of RA through the ALKBH5/circFOXK2/miR-486-3p or miR-409-3p/CD4⁺T cells autophagy and Th17 % axis.

Recently, a fraction of investigations into the impact of m6A modification and circRNA on the pathogenesis of RA have primarily focused on regulation of circRNA by m6A modification and regulation of m6A modification by circRNA. Evidence from Luo et al. found the hypermethylation of hsa_circ_0007259 may impact the progression of RA through the hsa_circ_0007259/has_miR-21-5p/STAT3 axis in synovial tissues [21]. Wan et al. indicated the level of m6A-related enzyme WTAP was altered upon circ_0066715 intervention, and suggested circ_0066715 affects the WTAP methylation process and the expression of the downstream transcriptional gene ETS1, influencing macrophage polarization in RA [22]. Although the above researches have validated circRNA and m6A modification could regulate each other, the role and mechanism of this relationship still needs to be better understood.

It is well-known that aberrant function of CD4⁺T cells was found in patients with RA and played a critical role in this disease's development. Thus, we evaluated the differences in m6A modification in circRNAs between peripheral blood CD4⁺T cells of RA cases and HC. We found that CD4⁺T cells from RA group contained 275 discrepant circRNAs modified with m6A compared to HC, indicating that m6A-circRNAs is abundant in CD4⁺T cells of RA cases. Pathway analysis of these discrepant m6A-circRNAs showed they focused on cGMP-PKG signaling pathway, RNA transport et al., in addition, cGMP-PKG signaling [23,24] and RNA transport [25,26] were found to be connected to RA, which suggest that discrepant m6A-circRNAs in CD4⁺T cells may be functionally related in human RA.

We revealed a significant increased m6A-hsa_circ_0001931 and a significant decreased m6A-circFOXK2 in the CD4⁺T cells from RA cases. Considering that m6A modification can affect the stability of circRNA, and thus regulate the expression of circular RNA, we explored the level of hsa_circ_0001931 and circFOXK2 in the CD4⁺T cells between RA cases and HC, and results indicated only circFOXK2 was found to be elevated in RA cases, while no statistically significant difference was observed in the level of hsa_circ_000193 between the two groups. Is there any additional regulatory mechanism that modulates the expression level of hsa_circ_0001931 and potentially counteracts the regulatory effect of m6A on its expression? This warrants further discussion. CircFOXK2 (hsa_circ_0000817) is an exonic circRNA composed of the 2nd, 3rd, and 4th exon of the FOXK2 gene with a length of 490 nucleotides [27], which is different from circFOXK2 (hsa_circ_0000816) [28], circFOXK2 (hsa_circ_0046430) [29] and circFOXK2 (hsa_circRNA_102241) [30]. Evidences from Zheng et al. demonstrated the back-splicing junction site of circFOXK2 (hsa_circ_0000817) and its circular characterizations [27]. CircFOXK2 (hsa_circ_0000817) was firstly explored in the progression of RA and was found to be associatied with treatment. Although the regulatory mechanism of circFOXK2 has not been reported, we determined circFOXK2 in RA cases could involve in differentiation of Th17 cells and autophagy of CD4⁺T cells. Moreover, these results were consistent to pathway analysis of the discrepant m6A-circRNAs that cGMP-PKG signaling pathway mostly affect cell cycle, differentiation of Th17 cell [23].

In the study, the expression levels of circFOXK2 in CD4⁺T from RA patients positively correlated with Th17 % (r_s = 0.5087, P = 0.0094) and autophagy level of CD4⁺T cells (r_s = 0.4939, P = 0.0030). The correlation coefficient between circFOXK2 and Th17 %, autophagy was approximately 0.5, which was not particularly ideal. The possible reason for this situation is the used of total CD4⁺T cells instead of naive/memory CD4⁺T cells, in other words, the low correlation (r_s = 0.5) with Th17 % and autophagy may be indirect and related to differential expression in memory T cell subsets including Th17. Thus, the expression levels of circFOXK2 in different T cell subsets and the function of circFOXK2 in different T cell subsets is worth exploring in further.

The m6A regulators primarily include methyltransferase (METTL3, METTL14, WTAP) that creates m6A mark, demethylase (ALKBH5, FTO) that erases m6A, and m6A RNA-binding proteins (YTHDF2). Substantial evidence has showed a close relationship between circRNAs and methyltransferase, demethylase in the development of disease. Sun et al. revealed that METTL3 downregulation significantly decreased m6A modification levels of circCREBBP, which result in diminishing IGF2BP3's binding affinity to circCREBBP and then promoting radioresistance in esophageal squamous cell carcinoma [31]. Shu et al. discovered that METTL3 and YTHDF1 were involved in the m6A modification of circRPS6KC1 and the stabilization, and suppression of circRPS6KC1 contributed to cellular senescence in prostate cancer [32]. Wang et al. observed that ALKBH5-mediated m6A modification of circFOXP1 and circFOXP1 promoted GC progression by regulating SOX4 expression and sponging miR-338-3p in GC cells [33]. This study confirmed that ALKBH5 may be the major upstream dominator of m6A-circFOXK2 in CD4⁺T cells from RA cases. The expression level of circFOXK2 positively correlated with the expression levels of ALKBH5 and YTHDF2 in the CD4⁺T cells of patients with new-onset RA. ALKBH5 overexpression could decrease m6A level and increase circFOXK2 expression, whereas ALKBH5 knockdown increased m6A level and decreased circFOXK2 expression in Jurkat cell line. Thus, our results demonstrated the m6A methylation level of circFOXK2 and the stabilization were regulated by ALKBH5. Unfortunately, the present study did not investigate the expression patterns of ALKBH5 across distinct CD4⁺T cell subsets or its potential influence on the functional characteristics of these subsets.

The present results from online database showed miR-486-3p, miR-409-3p may be the targets of circFOXK2. However, no association between miR-409-3p and RA has been reported. In both groups of peripheral blood mononuclear cells, the expression level of miR-486-3p was lower in the RA group than in the normal control group, and knockdown of miR-486-3p affected inflammatory and oxidative factors by increasing the expression levels of MDA, IL-11, IL-17, PD-L1 and decreasing the expression of IL10, SOD, TAOC [18]. In addition, study from Ouboussad et al. [19] indiated miR-486-3p in serum associated with progression from systemic autoimmunity to RA inflammation and when high-risk individuals progress to RA, the expression of miR-486-3p in serum decreases. Moreover, bioinformatics analysis identified ITGB2-miR-486-3p-SNHG3 involved in RA pathogenesis [20]. Evidences from above studies demonstrates the reliability of miR-486-3p as target to a certain extent. Thus, we speculate that circFOXK2 (hsa_circ_0000817) may act as a ceRNA to regulate the Th17 cell differentiation and autophagy of CD4⁺T cells through the circFOXK2-miR-486-3p axis. Further researches are needed to demonstrate its specific mechanism in the pathogenesis of RA.

There are some limitations in the present study. Firstly, circFOXK2's clinical value needs to be verified by polycentric prospective researches. Secondly, further studies are needed to characterize m6A reading molecules and the specific mechanism of ALKBH5/m6A-circFOXK2/miRNA in regulating differentiation of Th17 cells and autophagy of CD4⁺T cells in RA. Thirdly, CD4⁺T cells are a highly heterogeneous population, the function of circFOXK2 in different CD4⁺T cell subsets, the expression of ALKBH5 and the mechanism of ALKBH5/m6A-circFOXK2/miRNA in different CD4⁺T cell subsets are worth exploring in further.

5. Conclusion

The current study firstly measured the levels and quantities of m6A methylated circRNAs in CD4⁺T cells from human RA, HC and found differentially expressed m6A-circRNA in CD4⁺T cells of RA cases. In addition, we identified a significant decrease in the m6A-methylation of circFOXK2 and an obvious rise in the level of circFOXK2 in RA, and the increased expression of circFOXK2 was associated with treatment, CD4⁺T cells autophagy, Th17 %. MiR-486-3p and miR-409-3p may be the targets of circFOXK2. Moreover, increased m6A regulator ALKBH5 in CD4⁺T cells correlated with disease activity of new-onset RA, the level of circFOXK2, and the level of m6A and circFOXK2 were altered upon ALKBH5 intervention. However, the specific mechanism of ALKBH5/m6A-circFOXK2/miRNA in regulating function of CD4⁺T cells in RA still need to investigate. Therefore, the role of m6A-circFOXK2 in RA may provide insights into the pathogenesis of RA.

CRediT authorship contribution statement

Qing Luo: Writing – review & editing, Writing – original draft, Validation, Supervision, Software, Resources, Project administration, Methodology, Investigation, Funding acquisition, Formal analysis, Data curation, Conceptualization. Zhiwei Wu: Software, Methodology, Investigation, Formal analysis, Data curation. Qiuyun Xiao: Software, Methodology, Investigation, Formal analysis, Data curation. Mengfan Lan: Software, Methodology, Formal analysis, Data curation. Shiqian Wang: Software, Methodology, Data curation. Peng Fu: Software, Methodology, Formal analysis. Biqi Fu: Software, Investigation, Formal analysis. Zikun Huang: Writing – review & editing, Writing – original draft, Validation, Supervision, Software, Resources, Project administration, Methodology, Investigation, Funding acquisition, Formal analysis, Data curation, Conceptualization. Junming Li: Writing – review & editing, Writing – original draft, Validation, Supervision, Software, Resources, Project administration, Methodology, Investigation, Funding acquisition, Formal analysis, Data curation, Conceptualization.

Declaration of ethics approval and consent to participate

The research protocol complied with the principles outlined in the Declaration of Helsinki and were approved by the Ethics Committee of The First Affiliated Hospital of Nanchang University (approval no.(2023)CDYFYYLK(01–035)).

Informed consent was not required for this study because this study approved to exempt informed consent.

Availability of data and materials

All data generated or analyzed during this study are included in this published article.

Funding

This work was sponsored by National Natural Science Foundation of China (Nos. 82160307, 82160308), Jiangxi Provincial Natural Science Foundation of China (Nos. 20224BAB216038, 20242BAB23077, 20212ACB216006).

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Handling Editor: Y Renaudineau

Appendix A. Supplementary data

The following is the Supplementary data to this article:

Data availability

Data will be made available on request.