DNA Methylation of T Lymphocytes as a Therapeutic Target: Implications for Rheumatoid Arthritis Etiology

Rheumatoid arthritis (RA) is an autoimmune disease that can cause joint damage and disability. Epigenetic variation, especially DNA methylation, has been shown to be involved in almost all the stages of the pathology of RA, from autoantibody production to various self-effector T cells and the defects of protective T cells that can lead to chronic inflammation and erosion of bones and joints. Given the critical role of T cells in the pathology of RA, the regulatory functions of DNA methylation in T cell biology remain unclear. In this review, we elaborate on the relationship between RA pathogenesis and DNA methylation in the context of different T cell populations. We summarize the relevant methylation events in T cell development, differentiation, and T cell-related genes in disease prediction and drug efficacy. Understanding the epigenetic regulation of T cells has the potential to profoundly translate preclinical results into clinical practice and provide a framework for the development of novel, individualized RA therapeutics.


INTRODUCTION
Rheumatoid arthritis (RA) is a chronic and systemic inflammatory illness that causes persistent synovial inflammation and bone and joint damage. RA increases the risk of various diseases such as respiratory infections, osteoporosis, cardiovascular disease, urinary tract disease, and cancer (1). The incidence in women is higher than that in men. RA has a somewhat bimodal distribution of age of onset, with the largest fraction initially being diagnosed in the 20-40s, and then a smaller fraction being diagnosed roughly in the 60s, and it can appear at any stage (1). Typically, numerous autoimmune antibodies and cytokines are present in the pre-onset stage, and this molecular signature further advances with the clinical onset of synovial hyperplasia and bone/joint damage. In the absence of effective therapy, the disease typically progresses, resulting in impairment of both physical and mental health (2).
Many factors influence RA, including heredity, metabolism, environmental factors, and microbiota-and involve a complex interplay of genetic predisposition, activation of a variety of immune cells, and imbalance of pro-and anti-inflammatory mechanisms. This pathobiological immune state ultimately drives the failure to regulate immune tolerance, causing the development of autoimmune diseases targeting the synovium (3-5) (see Figure 1). Current research has discovered that numerous gene variants contribute to illness development, implying that each susceptible gene variant contributes to the disease to varying degrees (6). In addition to gene variations, the interaction between the environment and genes influences gene expression, which is heavily influenced by epigenetic modifications. This includes DNA methylation, histone posttranslational modification, miRNA, and, eventually, immune system regulation (6). DNA methylation refers to the addition of a methyl group to cytosine guanine (CG) dinucleotide (CpG). Two-thirds of CpGs in the human genome can be methylated (7). Under normal conditions, the 5-methylcytosine of CpG islands in DNA methylation prevents transcription factor complexes from binding to DNA through multiple processes, resulting in the inhibition of gene expression. On the other hand, DNA hypermethylation of gene promoters can enhance downstream gene transcription through a variety of mechanisms. Promoter hypermethylation, for example, blocks transcription inhibitors from binding to DNA (8). Hypermethylated advocates engage with enhancers to attract transcriptional activators and improve downstream gene transcription. Hypermethylation also promotes gene expression by binding to enhancers instead of insulators. Furthermore, in mammals, the hypermethylated promoter stimulates transcription of the first promoter of multiple variable promoters, resulting in increased target gene transcript output and decreased alternative spliceosome content (8).
T cells primarily engage with antigen-presenting cells and mix with antigen peptides provided by human leukocyte antigen (HLA) to create complexes that further mediate immunological responses (9). DNA methylation can influence various degrees of gene expression or selective splicing, both of which are associated with RA (10). The biological roles of the majority of the differential genes were shown to be associated with T cell development and activation by assessing gene expression and DNA methylation of CD4+ T cells in RA patients and healthy individuals (11). Many differentially methylated sites are found in T cell regulatory regions. Most gene variants were moderately or strongly correlated with methylation quantitative trait loci (meQTLs). Differentially variable positions (DVPs) are a type of CpG site. By studying the methylation of whole-genome DNA in monozygotic twins with discordant conditions, it was found that 1171 RA-related DVPs, of which 763 DVPs are in a highly variable state in RA twins, were mainly enriched in gene bodies and the 3'UTR of genes. Of the 763 DVPs, 563 were hypomethylated (12). Pathway analysis revealed that two of the top five were linked to T cells and cytokines (12). This suggests that the differential expression of RA genes may be partly driven by DNA methylation and are linked to variants of disease susceptibility genes that contribute to disease progression (11). In addition, there are several aberrant T-cell patterns in RA patients. In RA patients, both CD4+ naïve T cells and memory T cells exhibit premature aging and excessive telomere degradation (13). Furthermore, the ataxia telangiectasia mutated defect of T cells in RA patients leads to the destruction of the DNA repair mechanism (14). It enhances the sensitivity of T cells to apoptosis, while the apoptosis attrition of naive T cells ultimately accelerates immune senescence and the proliferation of autoimmune T cells (14). DNA methylation may cause aberrant activation and premature aging of T cells, contributing to atypical T-cell patterns in RA. In this article, we summarize and discuss the relationship between T cell DNA methylation and the pathological mechanism of RA in order to provide a theoretical foundation and reference for the development of novel clinical diagnosis and treatment programs from an epigenetic perspective (See Table 1).

THE POTENTIAL CONNECTION BETWEEN DNA METHYLATION OF CD4+ T CELLS AND RHEUMATOID ARTHRITIS
Several intriguing results were discovered using transcriptomic data, methylation data, and bioinformatics to analyze the differential genes and differentially methylated genes of diverse immune cells in RA (54). First, the presence of anti-lymphocyte antibodies may have resulted in a considerable reduction in the total number of T and B cells in the peripheral blood of patients with RA compared to healthy controls (54,55). Second, RA patient CD4+ T cells exhibit hypermethylated genes, including JUN, STAT1, PTEN, and CD44, compared to healthy controls, whereas hypomethylated genes include KRAS and ALB. Additionally, the expression levels of STAT5B, SOCS3, JUN, STAT1, KRAS, ALB, and CD44 were increased in RA. PTEN, FGFR2, and DICER1 expression levels were reduced. Gene ontology (GO) enrichment studies of differentially methylated genes are connected to T cell biological processes, suggesting that the regulation of CD4+ cells via DNA methylation plays an important role in RA (54). Finally, the analysis of the differentially methylated genes in the Kyoto Encyclopedia of Genes and Genomes (KEGG) showed that it was mainly related to thyroid hormone and estrogen signaling, hepatitis B virus (HBV), and the mitogen-activated protein kinase (MAPK) pathway (54). The results of the KEGG pathway analysis suggested various potential connections between gene DNA methylation and RA. Patients with RA have a higher prevalence of thyroid disease (56,57). It often shows a certain degree of hypothyroidism at onset, which is used as a sign of autoimmune thyroiditis (58). RA and HBV are interconnected and have been widely studied (59). For example, RA patients receiving tofacitinib can induce HBV reactivation in HBsAg+ and HBsAg-/HBcAb+ patients (60), and HBV exposure correlates with rheumatoid factor (RF) (61). The increased incidence of RA in postmenopausal women may be related to the lack of estrogen. Studies have found that estrogen can promote the degradation of acid-sensitive ion channel 1a protein and inhibit the cytotoxicity of acidosis through autophagy-lysosome-dependent pathways, thereby protecting chondrocytes (62,63).
By comparing the number of differentially methylated positions of memory CD4+ T cells and CD4+ naive T cells in RA patients, it was found that the number of the former was significantly greater than the latter. Most of the DNA methylation of differentially methylated positions (DMP) in RA patients with active disease is increased (64). Among the 101 RA genes recently identified (65), more than 30 overlapped with DMP (P=0.1 after 1000 permutations). One of these, UBASH3A, was found in CD4+ memory T cells. Its introns contain both single nucleotide polymorphisms (SNPs) that are correlated with RA risk r and two DMP and show a decrease in DNA methylation, which may increase gene expression (64). The protein encoded by UBASH3A, ubiquitinassociated and SH3 domain-containing A protein, primarily plays a role in the presentation of antigens to T cells and NA It is related to the pharmacological mechanism of MTX and TNF-a inhibitors (52,53) dampens T cell activity (41,42). Therefore, the expression of UBASH3A may increase the expression of genes through DNA methylation to promote antigen presentation to autoimmune T cells. UBASH3A SNPs (rs1893592, rs3788013), rheumatoid factor, cyclic citrullinated peptide (anti-CCP), disease activity score in 28 joints (DAS28), and C-reactive protein (CRP) were all significantly correlated with RA susceptibility (66,67). By studying peripheral blood samples from RA patients, researchers have found that CD4+ T memory cells have one hypermethylated CpG site, and CD4+ naïve T cells have 18 CpG sites; six are hypomethylated, and 12 are hypermethylated. The most important top 10 sites were located in TYK2, PRKAR1B, ABCC4, COMT, CAI2, MCF2L, GALNT9, C7orf50, and two intergenic regions (15). All 19 CpG sites were related to RA susceptibility. The hypomethylation status of CpG sites in the promoter regions of GATA3 and GATA3-AS1 (cg17566118 and cg15852223) was also found (15). The meQTL-mediated differentially methylated regions lead to the downregulation of GATA3 and CD83 expression (11). GATA3 is a transcription factor that drives the differentiation of antiinflammatory Th2 cells, promotes the enhancement of the interleukin (IL)-4 promoter, and regulates the expression of the IL-4 gene. The decrease in its expression may further lead to an imbalance in anti-inflammatory mechanisms (48,49). Oral administration of a-L-Guluronic acid, a non-steroidal antiinflammatory drug (NSAID), can significantly increase the expression of IL-4 and GATA3 in peripheral blood mononuclear cells (PBMCs) and improve the symptoms of related diseases in RA patients (68). CD4+ T cells lacking CD83 exhibit enhanced pro-inflammatory differentiation ability under in vitro stimulation; for example, more interferon-gamma (IFN-g) secretion and more Th1 and Th17 cell differentiation (31). The study detected genome-wide DNA methylation of CD4+ T cells from RA patients of Han Chinese ancestry. This reveals RA-related DNA methylation patterns in East Asian populations (18). There were 383 hypomethylation genes and 785 hypermethylated genes in CD4+ T cells. HDAC4, NXN, TBCD, and TMEM61 are the top four hypermethylated genes. HLA-DRB6, HLA-DQA1, HLA-E, ITIH3, TCN2, PRDM16, SLC1A5, and GALNT9 were hypomethylated. HLA-DQB1 is hypermethylated in the CpG island region, and the CpG framework is hypomethylated (18). In addition, the DNA methylation mechanism may also be a reason for the higher prevalence of RA in women. The hypomethylation status of CD40L in CD4+ T cells of female patients with RA is conducive to the expression of CD40L and partially explains the higher prevalence of RA in women (43). Consistent with this, the expression of 1,25(OH) 2 D 3 and the vitamin D receptor (VDR) in the peripheral blood of RA patients is reduced. 1,25(OH) 2 D 3 and VDR can inhibit the PKCd/ERK pathway and promote DNA methylation of CD11a, CD70, and CD40L to inhibit CD4+ T cell activation (44).
In this section, we summarize the changes in CD4+ T cellrelated differential DNA methylation, such as JUN, PTEN, and STAT1. Next, we explored the potential connection between its related genes and RA. Cell survival and death of CD4+ T cells play an essential role in RA (46). The upregulation of c-Jun expression can activate the MAPK signaling pathway and regulate cell proliferation, cell cycle, and apoptosis (20). Both STAT1 and STAT3 are related to IL-6 inflammatory factors and are involved in the course of RA (21). Gp130/STAT1 may increase the joint infiltration of T cells in antigen-induced arthritis (AIA) animal models (69). PTEN is mainly involved in cell proliferation and pro-inflammatory factors. Overexpression of PTEN can significantly reduce the expression of pro-inflammatory factors, chemokines, vascular cell adhesion factor-1, and vascular endothelial growth factor-a in AIA experimental animals. Consistent with the previous, the expression of PTEN is regulated by DNA methylation and is accompanied by changes in the AKT signal (22), and the methylation of PTEN has been shown to promote inflammation and activation of RA fibroblast-like synoviocyte (FLS) (23). Shikonin significantly reduces angiogenesis in RA. The mechanism may involve up-regulating the expression of PTEN, downregulating the expression of tumor necrosis factor-a (TNF-a), vascular endothelial growth factor 2, and IL-1b, and inhibiting the phosphorylation and gene expression of ERK1/2, JNK1/2, and P38 induced by TNF-a (70). In addition, PTEN is regulated by various miRNAs. Tocilizumab can regulate the proliferation of RA FLS and the production of proinflammatory factors via the MIR31HG/miR-214/PTEN/AKT axis (71). Therefore, PTEN may be a vital factor in inflammation in RA. CD44 is mainly used as a cell-surface adhesion molecule (72). The anti-CD44 antibody can significantly increase the incidence of experimental arthritis mice induced by type II collagen (24). Studies have shown that the preclinical stage of the disease in the K/BxN mouse model reduced CD4+ T lymphocytes. However, in the early clinical setting, most of the CD4+ T cells begin to express CD44, which is regarded as a sign of cell homeostasis expansion (73). Selenium supplementation can improve the joint pathological performance of CIA mouse models, reduce the number of CD4+CD44+ receptor activators of nuclear factor (NF)-kB ligand (RANKL) +T cells, and inhibit the expression of RANKL in mouse Th17 cells in vitro (74). Osteopontin in the synovial fluid of patients with RA induces Th17 differentiation and promotes the production of proinflammatory factor IL-17 through the IL-6/STAT3 pathway and CD44, CD29, and retinoic acid-related orphan receptor (ROR). The mechanism may be CD44 binding domain of CD4+ T cells induces H3 acetylation of the IL-17A promoter and promotes interaction with ROR (75). The overexpression of KRAS and BRAF in CD4+ T cells lowers the activation threshold of T cells to respond more efficiently to autoantigens in RA. This is related to the level of phosphorylated ERK (38).

THE EFFECT OF DNA METHYLATION ON THE DIFFERENTIATION OF CD4+ T CELLS
T cells have cellular heterogeneity and can differentiate into different subsets under stimulation by different cytokines, such as Treg cells, Th1 cells, Th2 cells, Th17 cells, and T-follicular helper (Tfh) cells, which play different roles in RA (76). Th1 and Th17 cell subsets contribute to inflammation in RA. Many studies have shown that the inhibition of Th1 and Th17 cell subsets can alleviate RA (77)(78)(79). Th2 cell subsets can inhibit Th1 and Th17 cell differentiation and function by secreting IL-4 as a mechanism to antagonize inflammation (80). In early patients with RA, Th2 cells and their secreted cytokines are few, and the differentiation of CD4+ T cells is severely unbalanced, which significantly promotes the production of chronic inflammation (81). Studies have shown that T cells lacking DNA methyltransferase (DNMT) 3a are activated and produce IL-4 and IFNG site hypomethylation. When T cells with Th2 cell characteristics are biased to induce Th1 cells, they will express more IFN-g, indicating that DNA methylation plays a vital role in controlling T cell differentiation according to the microenvironment (82). Further research shows that DNMT3a is an essential element that controls the differentiation of Th2 cells, expresses IL-13, and maintains the function of Th2 cells to inhibit inflammation (40). There is a substantial body of literature on how epigenetic mechanisms, including DNA methylation, affect the differentiation of lymphoid T cells (26,83).
In response to IL-12 stimulation, CD4+ T cells differentiate into Th1 cells and produce IFN-g. When activated for the first time, intracellular IFNG can be expressed under the stimulus of TCR signals and can bind to the transcription factor Tbet or STAT4. IL-12 can also reactivate the IFNG of memory Th1 cells for a second time (84)(85)(86). Studies have shown that the hypomethylation pattern of the IFNG promoter region and the CNS-1 region may contribute to the differentiation of Th1 and establish a stable epigenetic imprint, which is conducive to the reactivation of Th1 and promotes inflammation (39). Many meaningful results have been obtained by detecting the differential DNA methylation sites of monocytes, memory CD4 + T cells, and CD4+ naïve T cells in patients with RA (87). First, 1047 and 913 differentially methylated CpGs were detected in CD4+ naïve T cells and memory CD4+ T cells, respectively. Second, CD4+ T cells in early RA patients seem to develop in the direction of Th17 cells. Differential methylation of TBX21 (T-bet in mice) may lead to a decrease in Th1 cell differentiation (85). At the same time, the IL-17 and IL-17R of CD4+naïve T cells undergo differential methylation modification, which tend to differentiate into pro-inflammatory Th17 cells and interact with TNF. Differential methylation of IFN-related genes affects gene expression (87). As previously reported, it may indicate the characteristics of chronic inflammation and predict the progression of arthritis in early RA (88,89). Interestingly, when T cells were treated with DNA methyltransferase inhibitors, the expression of STAT4 was significantly upregulated. The shortening of the methylation site of the proximal regulatory element of its promoter greatly enhanced transcriptional activity, indicating that STAT4 in T cells is affected by DNA methylation regulation and is not regulated by promoter polymorphism (90). Decitabine, a DNA methylation inhibitor, can significantly improve the clinical symptoms of experimental arthritis mice induced by type II collagen, inhibit pro-inflammatory cytokines of Th1 and Th17 cells, reduce the production of anti-type II collagen antibodies (91), and promote the production of cytokines by Th2 cells (92). In addition, Tfh cells mainly promote B cell antibody maturation and development (93). The binding of BCL6, an essential molecule for Tfh cell differentiation, is associated with a decrease in 5-hydroxymethylcytosine (5hmC), implying a link between Tfh and DNA methylation (94). Hence, DNA methylation can regulate the differentiation of various T cell subsets via different mechanisms that affect the development of patients with RA.

THE DEFECTS OF DNA METHYLATION IN TREG CELLS PROMOTE INFLAMMATION
Treg cells are defined as CD4+CD25+FOXP3+; their expression defects contribute to the production of autoimmunity because their functions can inhibit the proliferation and differentiation of autoimmune T cells (95). Treg cells also have cellular heterogeneity, mainly divided into resting-state CD45 + FOXP3 low Treg cells, activated CD45RA -FOXP3 hi , and CD45 -FOXP3 low Treg cells, and a subset of Treg cells expressing HLA-DR with early contact-dependent inhibition (96,97). FOXP3 is the most commonly used marker for identifying Treg cells and is an essential transcription factor for the development and maintenance of suppressive function in Treg cells (26). There is a differential methylation region (DMR) upstream of the Foxp3 promoter in RA Treg cells, with enhancer activity sensitive to methylation-induced silencing. Both the expression of DMR and DNMT1/3 in RA Treg cells were downregulated, and the methylation of DMR was negatively correlated with the mRNA expression of Foxp3. RFnegative and-positive Treg cells also express Foxp3 differently. RF-positive Treg cells express lower Foxp3 levels and have a higher DMR methylation status (98). Tumor necrosis factor receptor 2 (TNFR2) is an important molecule that regulates Foxp3 expression. It can inhibit DNA methylation of the Foxp3 promoter by maintaining its expression in Treg cells. The lack of TNFR2 leads to the deterioration of experimental animal arthritis models, accompanied by a decrease in the number and a defect in the suppressive function of Treg cells. It leads Treg cells to inflammatory phenotypic differentiation of Th17 cells, which attempts to suppress inflammatory cells but fails (99). Consistent with this, the number of a subset of Treg cells expanded during the active stage of RA, similar to the pathogenic T cell phenotype (100). Similarly, the degree of methylation of Foxp3 in Treg cells of TNFR2 knockout mice increased. Under in vitro culture conditions, TNF upregulates Foxp3 expression in Treg cells through TNFR2 signaling, and TNFR2+ Treg cells increase in patients with RA receiving anti-TNF-a therapy for three months (101). Regular expression of CD83, SMAD7, and CTLA4 can maintain the functional integrity of Treg cells. Mice lacking CD83 cause abnormal Treg cell differentiation and promote inflammation (31,32). The down-regulation of CD83 and SMAD7 may be caused by meQTL-mediated differentially methylated regions in RA (11). SMAD7 is significantly reduced in the synovial tissue of patients with RA, and the TGF-b/SMAD3 signal is enhanced considerably, further intensifying the pro-inflammatory response of Th1 and Th17 cells (102). Experimental arthritis mice lacking SMAD7 developed severe arthritis, including joint swelling, synovial hyperplasia, bone destruction, and immune cell infiltration. The mechanism may be related to the imbalance of inhibitory function in Treg cells, the inflammatory response of Th1 and Th17 cells, and the overactivation of the TGF-b/ SMAD3/IL-6 pro-inflammatory signaling pathway (102). In addition, the Treg inhibitory function of patients with RA is impaired, and the expression of CTLA4 is downregulated, which may be due to methylation of the CTLA4 promoter. It causes Treg cells to fail to induce the expression of tryptophandegrading enzyme indoleamine 2,3 -dioxygenase (IDO) and the activation of the immunomodulatory kynurenine pathway (103). Decitabine, a DNA methylation inhibitor, can induce the expansion of RA Treg cells through an IDO-dependent pathway and promote the apoptosis of Th1 and Th17 cells to improve RA (104). DNA methylation may affect Treg cell function, further influencing autoimmune T cells and inflammation in RA.

THE DNA METHYLATION OF T CELLS AS A POTENTIAL BIOMARKER FOR RHEUMATOID ARTHRITIS
The identification of DNA methylation in RA holds promise for augmenting and improving the clinical diagnosis process. At present, biomarkers for the diagnosis of RA include anticitrullinated protein antibody (ACPA) and RF. The sensitivity is insufficient, and new markers are still needed to assist ACPA and RF, enabling more accurate diagnoses (105). The addition of other molecular tests to the markers of seropositivity offers an opportunity to increase the sensitivity of the combined test. Prime candidates for this expansion of the diagnostic utility of RA include genetic markers and differentially methylated regions. There are 1951 differentially methylated CpGs of T lymphocytes in patients with early RA, 60% of which were hypermethylated, representing 1216 genes. Of the top 15 genes, C6orf47, AGPAT9, DUSP22, SLC25A13, and SNORA15 were hypermethylated. MAGI2, C11orf58, POMZP3, ZNF708, and CLEC2L were hypomethylated (28). Similarly, the study identified 509 differentially methylated CpGs of T lymphocytes in RA. The top 15 with gene hypermethylation included DUSP22, C1orf106, ZFP57, ARSB, LSM5, and HIST1H4D, and hypomethylation included TMEM9, CCS, FBRSL1, GALNT9, PDHB, MGMT, DDO, KLF6, and RDH12 (29). DUSP22 has been shown many times to be DNA hypermethylated, which is closely related to the regulation of immunity and inflammation. Its reduced expression can promote CD4+ T cells to differentiate into Th1 and Th17 cells, thereby promoting inflammation (106). Similarly, the inhibition of DUSP22 promotes the activation, proliferation, and differentiation of CD4+T cells into Th1/Th17 cells, thus promoting an inflammatory response (107).
In addition, serum DUSP22 levels were negatively correlated with ESR, CRP, and DAS28, and increased in patients with RA receiving DMARDs treatment (108). These results suggest that DUSP22 could potentially be used as a biomarker of RA. In addition, the detection of DNA methylation of the whole genome of CD4+ T cells revealed that the changes in the methylation levels of many genes are related to different aspects of RA. For example, OR5A2 (cg02981094), ALDH9A1 (cg03984859), and C5orf32 (cg02070114) are related to the course of RA. ZC3H11A (cg02337583) is related to RF. OAS2 (cg00085448) was related to the patient's global assessment. C16orf71 (cg04705084) and LOC100129716 (cg00598143) were correlated with DAS28. SLC38A8 (cg01740650), C18orf19 (cg00448482), COL18A1 (cg04760448), BAT3 (cg05649229), and PLD3(cg07071106) were associated with erythrocyte sedimentation rate. HSPA12A (cg06942850) is associated with TJC (18). Further experiments are required to verify the accuracy of implementing these interesting results.

THE PHARMACOLOGICAL MECHANISM OF DRUGS FOR RHEUMATOID ARTHRITIS INVOLVES THE REGULATION OF DNA METHYLATION
Methotrexate (MTX) is a first-line drug for RA treatment. The recommended dose was 25 mg/week. It can be combined with glucocorticoids to achieve better clinical remission (109). DNA methylation analysis of CD4+ T cells purified from the peripheral blood of juvenile idiopathic arthritis revealed 145 differentially methylated sites. When removing four patients receiving MTX treatment, only 11 differentially methylated sites remained, strongly suggesting that the regulatory mechanism of MTX includes regulation of DNA methylation of CD4+ T cells (110). Studies have shown that MTX can affect one-carbon metabolism during the methyl transfer process of CpG island methylation through antifolate metabolism. Onecarbon metabolism requires the participation of folic acid (111).. Through further gene pathway analysis of 11 differentially methylated sites, most of them are in the "cellular growth and proliferation, hematological system development and function, hematopoiesis" network centered on TNF-a (110). The study also found that methylated IL-32 and MRPL28 are attractive potential targets (110). The pro-inflammatory factor TNF-a induces the expression of IL-32, which together constitute a chronic autoinflammatory cycle. Anti-TNF-a treatment can significantly reduce IL-32 protein levels in the synovial membrane of patients with RA (45). MRPL28 is mainly involved in encoding the mitochondrial ribosomal protein L28, and mitochondrial dysfunction may cause CD4+T cells to be sensitive to multiple forms of cell death. Mitochondria are also involved in the energy metabolism of cells, suggesting that the methylation of CD4+ T cells can affect RA in part by regulating the function of mitochondria, which may be a potential mechanism for MTX treatment (46,47). Significantly, MTX may affect gene expression by regulating the DNA methylation levels of multiple genes in T cells and further regulating multiple immune mechanisms to exert pharmacological effects. T cells in patients with RA who have just started receiving diseasemodifying antirheumatic drug (DMARD) treatment show an overall DNA hypomethylation pattern. The expression of DNA methyltransferase 1 (DNMT1) is low, the expression of demethylases (ten-eleven translocation (TET)1, TET2, and TET3) is increased, and the expression of growth arrest and DNA-damage-inducible protein 45A (GADD45A) was downregulated. After MTX treatment, the overall DNA methylation level and DNMT1 levels increased (112). By comparing the methylation levels of whole blood samples of patients with RA who had a good or bad response to MTX therapy for weeks, it was found that the differential methylation sites related to MTX therapy were mainly near GATA3 ( c g 2 7 4 2 7 5 8 1 ) , RP H3 AL ( c g 2 1 0 4 0 0 9 6 ) , a n d W D R 2 7 (cg09894276) (52). As mentioned above, GATA3 mainly promotes the differentiation of Th2 cells, indicating that MTX may partially regulate the methylation and gene expression of GATA3 to regulate the differentiation of CD4+ T cells to treat RA. Interestingly, the whole longitudinal genome of Japanese patients with RA receiving anti-TNF-a therapy showed that the gene expression of WDR27, MAP3K7, BACH2, and GFRA1 might be related to the therapeutic effect, further implying a correlation between DNA methylation, gene expression regulation, and various clinical therapies (53). As mentioned earlier, ALB in CD4+ T cells of patients with RA is hypomethylated (54). ALB is involved in the molecular mechanism of MTX treatment in RA (113). In addition, after receiving MTX treatment, the inhibitory function of RA Treg cells was restored, and the mechanism involved increased Foxp3 and CTLA-4 expression and reduced the methylation of the upstream enhancer of Foxp3 (27).
Biological agents are another treatment option for patients with RA who have a poor response to MTX treatment. TNF-a inhibitors are the most widely used. Various biological agents may exert pharmacological effects by regulating DNA methylation (109). By studying the differential gene expression and differential methylation of PBMCs, monocytes, and CD4+ T cells in patients with RA receiving adalimumab (ADA) or etanercept (ETN) before and after treatment, it was found that more than 100 differential genes had DNA methylation changes. Some CD4+T cells of patients with RA who respond to ADA therapy have the characteristics of upregulation of TNF signaling pathway-related genes, such as CTLA4, TNFSF13B, TNFRSF1B, TNFSF4, IRF1, and IL18R1. In CD4+ T cells of patients with RA responding to ETN therapy, Foxo signaling pathways, such as FOXO3, FOXO4, TGFBR1, and USP7, were upregulated. NOD-like receptor signaling pathways, such as GSDMD, RIPK3, CASP4, and JAK/STAT signaling pathways, such as CISH, SOCS2, and PIM1, were downregulated (114). In addition to anti-TNF-a therapy, the DNA methylation changes of IL-6 significantly affect the gene expression of early RA, which can promote inflammation through the JAK1/STAT3 pathway (87). The clearance of apoptotic cells by phagocytes has an immunosuppressive effect of preventing the induction of autoimmunity and inflammation. If it cannot be eliminated effectively, it may be harmful. The DNA of apoptotic T cells is demethylated, which may further promote inflammation and stimulate the production of pro-inflammatory factor IL-6 in macrophages by interacting with TLR in RA (50). This further suggests that the existing IL-6R therapy and JAK inhibitor therapy should be applied as soon as possible, and the mechanism may involve DNA methylation in RA. Some drugs or natural components could regulate DNA methylation or hydroxymethylation in diseases like cancer (115,116). Furthermore, while some of these may have a therapeutic effect in RA, they may not be active directly in treating RA by modulating the DNA methylation status. We still believe some drugs or natural components have therapeutic potential for RA, given some association outcomes. Table 2 shows updates and summaries thereof. (See Table 2)

CONCLUSION
There are several unresolved issues regarding the role of DNA methylation in RA. First, a cell-specific investigation of the link between DNA methylation and RA is required to fully understand the impact of differential methylation on RA pathophysiology. Many studies have focused solely on the DNA methylation properties of the total cell population, ignoring the variability of cells within distinct cell subsets. There may be several mechanisms by which DNA methylation yields effects, and these effects may be substantially different across cell types. As a result, cell purification is required to properly evaluate DNA methylation and its link with gene expression and regulation of biological functions. There are some shortcomings, including the fact that there is little direct research-based evidence about other T cell subsets (CD8+ T cells, gamma delta T cells, and MAIT T cells) related to DNA methylation regulation. It is undeniable that T cells are heterogenous cell populations. Studying and clarifying the correlation between the DNA methylation patterns of other T cell subsets and RA is very important, and is a potential future research direction. Second, whether DNA methylation begins before the commencement of the disease or varies continually over the course of the ailment requires more well-designed clinical studies for sample collection to clarify its function. The detection of DNA methylation before illness onset can assist in identifying high-risk patients, allowing us to realize the objective of precision medicine. Finally, a distinction should be made between cell-specific alterations in DNA methylation in early and late RA. When comparing various study findings, we should keep the distinctions between chips and detecting systems in mind. Additionally, we mapped the protein interaction network based on the genes that undergo DNA methylation in T cells discussed in this article. Some of the proteins have interactions, potentially focusing on future research for further exploration (see Figure 2). In conclusion, integrating cell-specific DNA methylation data with other data, such as primary clinical data, gene expression data, proteomics data, and metabolomics data, would enable us to deliver a more powerful RA-related clinical treatment choice in the future.

AUTHOR CONTRIBUTIONS
JZ is responsible for the collection, collation and writing of the original manuscript. KW, CC, LX and PJ is responsible for the collection of the original manuscript. SG, SS, and DH are responsible for the revision and review of the manuscript. All authors reviewed and accepted with the final version.