MicroRNA-126 regulates DNA methylation in CD4+ T cells and contributes to systemic lupus erythematosus by targeting DNA methyltransferase 1

Authors


Abstract

Objective

To identify microRNA genes with abnormal expression in the CD4+ T cells of patients with systemic lupus erythematosus (SLE) and to determine the role of microRNA-126 (miR-126) in the etiology of SLE.

Methods

MicroRNA expression patterns in CD4+ T cells from patients with SLE and healthy control subjects were analyzed by microRNA microarray and stem loop quantitative polymerase chain reaction (qPCR). Luciferase reporter gene assays were performed to identify miR-126 targets. Dnmt1, CD11a, and CD70 messenger RNA and protein levels were determined by real-time qPCR, Western blotting, and flow cytometry. CD11a, CD70, and EGFL7 promoter methylation levels were detected by bisulfite sequencing. IgG levels in T cell–B cell cocultures were determined by enzyme-linked immunosorbent assay.

Results

The expression of 11 microRNA was significantly increased or decreased in CD4+ T cells from patients with SLE relative to that in CD4+ T cells from control subjects. Among these, miR-126 was up-regulated, and its degree of overexpression was inversely correlated with Dnmt1 protein levels. We demonstrated that miR-126 directly inhibits Dnmt1 translation via interaction with its 3′–untranslated region, and that overexpression of miR-126 in CD4+ T cells can significantly reduce Dnmt1 protein levels. The overexpression of miR-126 in CD4+ T cells from healthy donors caused the demethylation and up-regulation of genes encoding CD11a and CD70, thereby causing T cell and B cell hyperactivity. The inhibition of miR-126 in CD4+ T cells from patients with SLE had the opposite effects. Expression of the miR-126 host gene EGFL7 was also up-regulated in CD4+ T cells from patients with SLE, possibly in a hypomethylation-dependent manner.

Conclusion

Our data suggest that miR-126 regulates DNA methylation in CD4+ T cells and contributes to T cell autoreactivity in SLE by directly targeting Dnmt1.

Systemic lupus erythematosus (SLE) is a chronic and potentially fatal autoimmune disorder characterized by T lymphocyte autoreactivity and the production of autoantibodies that cause widespread tissue damage. Although the mechanisms that initiate these manifestations remain unclear, it has been widely reported that epigenetic factors play a central role in the onset and progression of SLE (1–3).

The traditional mechanisms of epigenetic regulation include DNA methylation and histone modifications. DNA methylation involves the addition of a methyl group to the pyrimidinyl ring of cytosine, primarily within CpG pairs, and is catalyzed by DNA methyltransferases (Dnmt). Methylation of CpG islands in promoter regulatory regions is associated with transcriptional inactivation of the corresponding gene, while demethylation of these regions creates a permissive transcriptional environment (4). The T cells of patients with active lupus exhibit a global reduction in DNA methylation, and T cell DNA hypomethylation levels correlate with disease activity in patients with lupus (5, 6). Furthermore, treating human and murine CD4+ T cells with DNA-demethylating drugs such as 5-azacytidine, procainamide, and hydralazine can induce lupus-like autoreactivity in vitro and lupus-like symptoms if the cells are injected back into host mice (7–10), suggesting that changes in T cell DNA methylation can contribute to the pathogenesis of SLE (7, 9–12). Recent studies have also demonstrated an association between DNA hypomethylation and decreased enzymatic activity of Dnmt (especially Dnmt1) in patients with SLE, suggesting a mechanism by which T cell DNA becomes hypomethylated (13).

Our previous studies have shown that the promoter regions of TNFSF7 and ITGAL, genes encoding the autoimmune-related proteins CD70 and CD11a, are hypomethylated in SLE CD4+ T cells. This causes a concomitant increase in CD11a and CD70 levels (14–16), and the degree of overexpression is directly proportional to disease activity (16). CD70, the cellular ligand for the tumor necrosis factor receptor family member CD27, is transiently expressed on activated T cells and B cells, where it stimulates the synthesis of IgG by inducing B cell costimulatory functions. CD11a, also known as lymphocyte function–associated antigen 1, is a member of the integrin family of cell surface receptors and can strengthen the adhesion of T lymphocytes to other immune cells (17). Taken together, the results of these studies suggest that the autoimmune responses observed in SLE are caused by reduced methylation of TNFSF7 and ITGAL (and perhaps other immunity genes) that permits increased expression of CD11a and CD70, which leads to T cell and B cell autoreactivity. However, the reason why Dnmt1 levels are reduced in patients with lupus remains incompletely understood.

MicroRNA are endogenous 21–24–nucleotide noncoding RNA molecules that regulate the expression of target genes by specifically binding to and interfering with their messenger RNAs (mRNA) (18, 19). Recent studies have shown that microRNA are involved in various immune responses (20) and are associated with autoimmune diseases (21) and could potentially serve as diagnostic biomarkers (22) or therapeutic targets (23). Several groups of investigators have demonstrated that microRNA function as both targets and effectors of aberrant DNA methylation (24, 25). However, it is not yet clear how microRNA dysregulation may contribute to the pathogenesis of autoimmune diseases such as SLE.

In the present study, we used microarray analysis to compare the microRNA expression profiles of CD4+ T cells from patients with SLE with those of CD4+ T cells from healthy control subjects and identified 11 unique microRNA that were either up-regulated or down-regulated in CD4+ T cells from patients with SLE. The up-regulation of one of these genes, microRNA-126 (miR-126), was confirmed by stem loop real-time polymerase chain reaction (PCR), and its involvement in SLE was studied. We observed that miR-126 specifically targets Dnmt1 mRNA through interactions with its 3′–untranslated region (3′-UTR). Expression of plasmid-encoded miR-126 in CD4+ T cells led to the down-regulation of Dnmt1 and to DNA hypomethylation. Furthermore, overexpression of miR-126 induced the up-regulation of CD11a and CD70 and caused CD4+ T cells to stimulate IgG production in cocultured B cells. Expression of an miR-126 inhibitor in CD4+ T cells from patients with SLE restored Dnmt1 levels and led to increased TNFSF7 and ITGAL promoter methylation, resulting in reduced CD11a and CD70 expression as well as T cell activity. The miR-126 host gene EGFL7 was also overexpressed in CD4+ T cells from patients with SLE, and miR-126/EGFL7 up-regulation was associated with a reduction in EGFL7 promoter methylation. These results suggest that miR-126 regulates DNA methylation in CD4+ T cells and contributes to T cell autoreactivity in SLE by directly targeting Dnmt1.

PATIENTS AND METHODS

Subjects.

Relevant information about the patients with SLE is shown in Table 1. Patients with SLE (n = 30; mean ± SD age 32.8 ± 8.3 years) were recruited from the outpatient dermatology clinic and the in-patient ward at the Second Xiangya Hospital, Central South University. All patients fulfilled at least 4 of the American College of Rheumatology criteria for the classification of SLE (26), and disease activity was assessed using the SLE Disease Activity Index (SLEDAI) (27). Active disease was defined as a SLEDAI score of ≥5, and inactive disease was defined as a SLEDAI score of <4. Healthy control subjects (n = 20; mean ± SD age 30.4 ± 5.8 years) were recruited from among the medical staff at the Second Xiangya Hospital. Patients and control subjects were matched for age and sex in all experiments, and T cell samples from each group were paired and studied in parallel. This study was approved by the human ethics committee of the Central South University Xiangya Medical College, and written informed consent was obtained from all participants.

Table 1. Patient demographics and medications
PatientAge/sexSLEDAI*Medication
  • *

    SLEDAI = Systemic Lupus Erythematosus Disease Activity Index; active disease was defined as a score of ≥5, and inactive disease was defined as a score of <4.

123/F20None
223/F10None
313/F8None
421/F8Prednisone 20 mg/day
528/F12Prednisone 15 mg/day
632/F10None
720/F0Prednisone 30 mg/day
834/F0None
918/F12None
1034/F10None
1138/F20None
1241/F12None
1325/F8Prednisone 5 mg/day
1440/F3None
1546/F12Prednisone 60 mg/day
1618/F2Prednisone 5 mg/day
1726/F4Prednisone 15 mg/day
1825/F2Prednisone 15 mg/day
1922/F0Prednisone 20 mg/day
2028/F4None
2113/F8None
2242/F2Prednisone 10 mg/day
2337/F10None
2416/F15None
2525/F14Prednisone 30 mg/day
2621/F6Prednisone 10 mg/day
2741/F8Prednisone 5 mg/day
2819/F8Prednisone 5 mg/day
2921/F12Prednisone 15 mg/day
3020/F14Prednisone 15 mg/day

T cell isolation, cultures, and transfection.

A total of 60 ml of venous peripheral blood was obtained from each subject and preserved in heparin. Peripheral blood mononuclear cells (PBMCs) were isolated by Ficoll-Hypaque density-gradient centrifugation (Shanghai Hengxin Chemical Reagent Co.). CD4+ T cells were isolated using magnetic beads (Miltenyi Biotec; the purity was generally >95%) and cultured in human T cell culture medium (Amaxa) containing 15% fetal bovine serum (FBS) and 100 μg/ml penicillin G and streptomycin. Cells were transiently transfected with plasmid-encoded human pSilencer-mir-126, pSilencer4.1CMV-negative (gifts from Dr. Duan Ma, Fudan University, China), miScript miR-126 inhibitor, or scrambled oligonucleotides (Qiagen), using Human T cell Nucleofector Kits and a nucleofector (Amaxa), and collected for analysis 72 hours later.

T cell activation.

Purified CD4+ T cells from healthy control subjects were cultured in 24-well plates (1 × 106/ml) and stimulated with plate-bound anti-CD3 antibody (eBioscience), followed by the addition of soluble anti-CD28 antibody (eBioscience; 2 μg/ml each). Culture plates were incubated for 72 hours at 37°C.

Microarrays.

Total RNA was extracted from CD4+ T cells, using miRNeasy Mini Kits (Qiagen). RNA from 10 samples per group (SLE patients and healthy control subjects) were pooled, and RNA concentrations were equalized. The pooled samples were sent to LC Sciences for microRNA microarray analysis. RNA quality control, labeling, hybridization, and scanning were performed at LC Sciences, using the probes in Sanger miRBase microRNA database version 11.0 (http://www.ebi.ac.uk/). Total RNA (1 μg) from the pooled SLE lysate was labeled with Cy5, and total RNA (1 μg) from the pooled healthy control lysate was labeled with Cy3. Samples were then hybridized onto microRNA microarrays, and Cy3:Cy5 ratio images of the microarrays were generated (Figure 1A). From these images, in which imbalanced color intensities indicated differences in RNA levels, changes in microRNA expression in T cells from patients with SLE compared with that in T cells from healthy control subjects were assessed. Statistical analyses were performed at LC Sciences; microRNA products with at least a 2-fold difference in expression between the 2 groups and a P value less than 0.01 were considered significant.

Figure 1.

MicroRNA expression patterns in CD4+ T cells from patients with systemic lupus erythematosus (SLE) and from healthy control subjects. A, High-throughput analysis of microRNA expression. Each spot shows the Cy3:Cy5 ratio representing the expression of a single microRNA in SLE CD4+ T cells (Cy5; red) relative to healthy control CD4+ T cells (Cy3; green). MicroRNA with nonequivalent expression levels in the 2 groups (determined by an imbalance of Cy3 and Cy5 signals) are indicated in blue. B, Relative expression of the 11 microRNA up-regulated or down-regulated at least 2-fold in SLE CD4+ T cells relative to control CD4+ T cells. C and D, SLE-associated up-regulation of microRNA-126 (miR-126) (C) and down-regulation of miR-142-3p (D) measured by microRNA stem loop real-time polymerase chain reaction in CD4+ T cells from 20 patients with SLE and 20 age- and sex-matched healthy control subjects. Data are presented as box plots, where the boxes represent the 25th to 75th percentiles, the lines within the boxes represent the median, and the lines outside the boxes represent the 10th and 90th percentiles. E, Expression of CD69 in CD3/CD28-stimulated CD4+ T cells from 3 individual healthy donors. Cells were stained with fluorescein isothiocyanate (FITC)–labeled anti-CD69 antibody and analyzed by flow cytometry. ∗∗ = P < 0.01 versus control. F, Expression of miR-126 in CD3/CD28-stimulated CD4+ T cells from 3 individual healthy donors. Values in E and F are the mean ± SD.

Real-time quantitative PCR (qPCR).

For real-time qPCR, complementary DNAs were synthesized from 10 ng of total RNA using microRNA-specific primers and Hairpin-it miRNA qPCR Quantification Kits (GenePharma). Real-time PCR was performed using the Rotor-Gene 3000 real-time PCR instrument (Corbett Research). The cycle parameters for PCRs were 95°C for 5 minutes followed by 40 cycles of 95°C for 30 seconds and 60°C for 40 seconds. All reactions were run in duplicate. Expression levels of target microRNA were normalized to 18S ribosomal RNA and analyzed with Rotor-Gene Real-Time Analysis Software 6.0. Dnmt1, CD70, and CD11a mRNA were amplified by SYBR green real-time PCR using the One Step PrimeScript RT-PCR Kit (Takara Bio) and normalized to β-actin. The ΔCt value was calculated by subtracting the Ct value for 18S or β-actin from the Ct value for the gene of interest. The ΔΔCt value was calculated by subtracting the control ΔCt value from the SLE ΔCt value. The fold difference of expression between the level in control and SLE samples was calculated as 2math image.

Flow cytometric analysis.

CD4+ T cell suspensions (1 × 105 cells) were incubated with fluorescein isothiocyanate–conjugated anti-human CD69, CD70, or CD11a antibodies (Becton Dickinson) for 30 minutes at room temperature then washed with 2 ml of phosphate buffered saline (PBS) pH 7.4 containing 1% bovine serum albumin (BSA) and centrifuged at 400g for 5 minutes. Supernatants were discarded, and cells were resuspended in 0.5 ml PBS/BSA. Data were acquired with a FACSCalibur system (Becton Dickinson) and analyzed using CellQuest software (Becton Dickinson). The expression levels of CD69, CD70, and CD11a were evaluated by calculating the mean fluorescence intensity and the percentage of cells expressing each protein.

Western blotting.

CD4+ T cells were lysed, and proteins were extracted and separated by 8% sodium dodecyl sulfate–polyacrylamide gel electrophoresis. Proteins were then transferred onto a polyvinylidene difluoride membrane (Millipore). Membranes were blocked in Tris buffered saline−Tween containing 5% nonfat dry milk and blotted with Dnmt1 (1:200; Abcam) or β-actin (1:2,000; Santa Cruz Biotechnology) antibodies. Relative expression levels were quantified using Quantity One software (Bio-Rad).

Luciferase activity assay.

A 120-bp sequence from the DNMT1 3′-UTR containing the putative binding sites for miR-126 was amplified by PCR from human CD4+ T cell genomic DNA using the following primers: forward 5′-TTACTAGTCTTCTTCAGCACAACCGTCA-3′ (underline indicates the Spe I site, and italicized bases represent the miR-126 binding site), reverse 5′-ATAAGCTTGCCACAAACACCATGTACCA-3′ (underline indicates the Hind III site). The same procedure was used to generate reporter constructs with mutations in the DNMT1 3′-UTR with the exception that the reverse primer was substituted for 5′-ATAAGCTTGCCACAAACACCATTGACCA-3′ (dashed line indicates nucleotide substitutions) (see Figure 2E). The DNMT1 3′-UTR sequences were inserted into pMIR-REPORT luciferase microRNA Expression Reporter Vector (Ambion) using Spe I and Hind III. The inserts were confirmed by DNA sequencing.

Figure 2.

Effect of miR-126 on Dnmt1 expression in CD4+ T cells, and identification of Dnmt1 as an miR-126 target. A, Dnmt1 expression in CD4+ T cells from patients with SLE was significantly decreased compared with that in cells from healthy control subjects. B, MicroRNA-126 transcript levels were negatively correlated with Dnmt1 protein levels in SLE CD4+ T cells. C, The expression of miR-126, miR-142-5p, and Dnmt1 mRNA and protein was analyzed after transfection with miR-126–encoding plasmid DNA or negative control plasmid DNA. D, MicroRNA-126, miR-142-5p, and Dnmt1 expression levels were analyzed after transfection with miR-126 inhibitor or control scrambled oligonucleotide. E, A schematic representation of the DNMT1 luciferase reporter construct is shown. The sequence of the miR-126 binding site in the 3′–untranslated region (3′-UTR) of DNMT1 (grey box) is shown on the right. Mutated residues are shown in red. F, Relative firefly luciferase activity in Jurkat cells cotransfected with an empty vector (negative control) or an miR-126–expressing construct, together with luciferase reporter constructs containing either a wild-type (WT) or a mutated (Mut) DNMT1 3′-UTR are shown. Values in A, C, D, and F are the mean ± SD results from 3 independent experiments. ∗ = P < 0.05 versus control in A, C, and F; versus scrambled in D. See Figure 1 for other definitions.

Jurkat cells were cultured in RPMI 1640 with 10% FBS. Cells were plated in a 6-well plate at a density of 2 × 106/well. After overnight incubation, cells were cotransfected with 5 μg of firefly luciferase reporter vector containing the wild-type or mutant oligonucleotides, 10 μg of miR-126–encoding plasmid (pSilencer-mir-126), or negative control (pSilencer4.1CMV-negative) by electroporation, using the Gene Pulser II (Bio-Rad). Each sample was cotransfected with 0.05 μg pRL-TK plasmid expressing Renilla luciferase to monitor transfection efficiency (Promega). Another 48 hours later, cells were washed twice, suspended in 500 μl reporter lysis buffer (Promega), and firefly luciferase activity was measured using the Dual-Luciferase Reporter Assay System (Promega) with a GloMax 20/20 luminometer (Promega), according to the manufacturer's protocol. Relative luciferase activity was normalized to Renilla luciferase activity for each transfected well. The experiments were performed in triplicate in 3 independent experiments.

Bisulfite sequencing.

Genomic DNA was isolated from CD4+ T cells using the TIANamp Genomic DNA Kit (Tiangen Bio). Bisulfite conversion was performed using the EpiTect Bisulfite Kit (Qiagen). The 294-bp (−581 to −288) CD70, 310-bp (−1289 to −979) CD11a, and 193-bp (−291 to –98) EGFL7 promoter fragments were amplified by nested PCR and cloned into pGEM-T vector (Promega). Ten independent clones were sequenced for each of the amplified fragments.

B cell/CD4+ T cell cocultures for costimulation assays.

B cells were enriched with CD19 magnetic beads (Miltenyi Biotec) and cultured in RPMI 1640 medium with 10% FBS, 100 units/ml penicillin, and streptomycin for T cell and B cell costimulation assays. Forty-eight hours after transfection, CD4+ T cells were cocultured with autologous B cells (4 × 105) at a ratio of 1:4 (14). The cells were cultured for 8 days in 24-well round-bottomed plates (Costar) containing a total volume of 250 μl and supplemented with 250 μl of medium on day 4.

IgG enzyme-linked immunosorbent assays (ELISAs).

IgG concentrations in the supernatants of T cell and B cell cultures were measured by Universal One-step IgG Quantification ELISA Kit-H (Columbia Bio). All experiments were performed in quadruplicate. Optical density values were read at 405 nm using an ELx800 Absorbance Microplate Reader (BioTek).

Statistical analysis.

Results are expressed as the mean ± SD. Data were analyzed by analysis of variance followed by Student's unpaired t-test for multiple comparisons. Spearman's rank test was used for correlation studies. All analyses were performed with SPSS 16.0 software. P values less than or equal to 0.05 were considered significant.

RESULTS

Up-regulation of miR-126 and down-regulation of miR142-3p in CD4+ T cells from patients with SLE.

We used Sanger miRBase microRNA database version 11.0 to examine the expression of microRNA in pooled CD4+ T cell lysates isolated from 10 patients with SLE and pooled CD4+ T cell lysates from age- and sex-matched healthy control subjects (see Patients and Methods). Highly distinct patterns of expression were observed between the 2 groups (Figure 1A). Based on the analysis performed at LC Sciences, 11 of the 873 distinct microRNA screened showed a >2-fold difference in expression between the 2 groups (P < 0.05). Among these, the levels of miR-1246, miR-574-5p, miR-1308, miR-638, miR-7, and miR-126 were increased by ≥2-fold, and the expression of miR-142-5p, miR-142-3p, miR-31, miR-186, and miR-197 was reduced to less than half in SLE CD4+ T cells compared with healthy control CD4+ T cells (Figure 1B). Interestingly, according to the miRBase microRNA database, 4 of these microRNA (miR-638, miR-126, miR-142-3p, and miR-142-5p) are predicted to target genes associated with SLE (28–31), suggesting that aberrant microRNA expression in CD4+ T cells is a factor in SLE pathogenesis.

The microarray results for these 4 potentially SLE-associated microRNA were then confirmed by stem loop real-time PCR on samples from an additional 20 patients with SLE and 20 healthy control subjects. We observed that miR-126 was significantly up-regulated (P < 0.01) (Figure 1C), and miRNA-142-3p was significantly down-regulated in SLE CD4+ T cells compared with controls (P < 0.01) (Figure 1D). The differences in the expression of miR-638 and miR-142-5p were not statistically significant between the 2 groups (data not shown). To investigate whether the alterations observed in samples from patients with SLE could be ascribed to drug treatment, we compared untreated patients with SLE (n = 10) with treated patients (receiving low- or medium-dose corticosteroids; n = 10) (see Table 1). No significant differences were found in miR-126 and miRNA-142-3p expression between the 2 groups (Figures 1C and D), thus suggesting that medication does not impact on miR-126 and miR-142-3p expression. Moreover, we did not observe any correlation between miR-126 levels and disease activity, as assessed by the SLEDAI (data not shown).

Because T lymphocyte activation has been reported to affect microRNA expression (32), we questioned whether the overexpression of miR-126 in patients with SLE is a consequence of T cell hyperactivity. CD4+ T cells from healthy control subjects were stimulated with anti-CD3/CD28 antibodies, and the expression of CD69, a marker of T lymphocyte activation, was measured by flow cytometry. We observed that CD69 expression was significantly increased in activated CD4+ T cells (mean ± SD 47.324 ± 6.736 versus 7.231 ± 1.353; P < 0.01) (Figure 1E); however, there was no significant change in miRNA-126 expression in response to the stimulation (P > 0.05) (Figure 1F). This suggests that miR-126 up-regulation in CD4+ T cells is a potential cause of autoimmunity in SLE and not simply a consequence of increased lymphocyte activity.

Identification of miR-126 target mRNA in CD4+ T cells from patients with SLE.

The SLE-associated methyltransferase enzyme Dnmt1 (33) is a predicted target of miR-126, according to the miRBase microRNA database. To analyze the relationship between miR-126 and Dnmt1 expression, we plotted miR-126 transcript levels from individual SLE CD4+ T cell lysates (n = 20), as measured by stem loop real-time PCR (Figure 1C), against Dnmt1 protein levels from the same samples, as measured by Western blotting (Figure 2A). A strong inverse correlation was seen between the 2 values (r = −0.558, P = 0.016, by Spearman's rank correlation test) (Figure 2B).

To determine whether the down-regulation of Dnmt1 was a direct consequence of miR-126–mediated inhibition, we transfected primary CD4+ T cells from healthy donors with either empty plasmids (pSilencer4.1CMV-negative; negative control) or plasmid-encoded miR-126 (pSilencer-mir-126). Three days after transfection with pSilencer-mir-126, miR-126 transcripts were increased by 4.68-fold, while those of the unrelated miR-142-5p gene remained unchanged. Furthermore, levels of Dnmt1 protein, but not mRNA, were significantly decreased relative to negative controls (Figure 2C). Consistently, transfecting an miR-126 inhibitor into SLE CD4+ T cells induced a 3.12-fold decrease in miR-126, while miR-142-5p expression remained unchanged and significantly increased the level of Dnmt1 protein, but not mRNA, whereas a negative control (scrambled oligonucleotide) had no effect (Figure 2D).

To confirm that DNMT1 is a direct target of miR-126, we generated a firefly luciferase reporter plasmid fused downstream to a segment of the DNMT1 3′-UTR containing either the wild-type putative miR-126–binding sequence (DNMT1WT-luciferase), or the miR-126–binding sequence containing 2 point mutations (DNMT1Mut-luciferase) (Figure 2E). The constructs were then cotransfected into Jurkat cells with pSilencer-mir-126 or pSilencer4.1CMV-negative, and luciferase activity was measured 48 hours later. MicroRNA-126 significantly reduced DNMT1WT-luciferase activity (P < 0.05) but failed to inhibit DNMT1Mut-luciferase activity (Figure 2F). Taken together, these data strongly suggest that miR-126 up-regulation contributes to the reduction of Dnmt1 protein levels in SLE CD4+ T cells.

Role of miR-126 overexpression in CD11a and CD70 promoter methylation and CD4+ T cell autoreactivity.

Down-regulation of Dnmt1 in CD4+ T cells contributes to lupus autoreactivity because it leads to hypomethylation-dependent de-repression of autoimmunity genes, including CD11a and CD70 (33). To test whether miR-126 up-regulation was sufficient to induce DNA hypomethylation and lupus-like autoreactivity in vitro, we measured the methylation status of the ITGAL (CD11a) and TNFSF7 (CD70) promoters in the presence of transgenic miR-126. We transfected pSilencer-mir-126 or pSilencer4.1CMV-negative into primary CD4+ T cells from 3 healthy donors and harvested the cells 72 hours later. Bisulfite sequencing was then performed to determine the methylation status of CpG pairs within the CD11a and CD70 promoters, and qPCR and flow cytometric analyses were performed to correlate methylation levels with expression levels. Up-regulating miR-126 decreased the methylation of both CD11a and CD70 promoters compared with negative controls (Figures 3A–C) and caused a concomitant increase in their mRNA and protein levels (Figures 3D and E).

Figure 3.

Overexpression of microRNA-126 (miR-126) promotes hypomethylation of ITGAL (CD11a) and TNFSF7 (CD70) promoter DNA, leads to increased CD11a and CD70 expression, and induces B cell stimulation. A–C, CD4+ T cells from healthy control subjects were transfected with an miR-126 expression plasmid or an empty vector (negative control), and bisulfite sequencing was performed 72 hours later. A, The mean methylation status of each CpG pair in the 294-bp (−581 to −288) fragment of the CD70 promoter is shown. B, Relative methylation levels of the CpG pairs in the 310-bp (−1289 to −979) fragment of the CD11a promoter are shown. C, Overall average methylation levels of the CD70 and CD11a promoter regions were significantly decreased in CD4+ T cells 72 hours after transfection with the miR-126 expression plasmid relative to control. D, Relative CD70 and CD11a mRNA levels were determined in control and miR-126–overexpressing CD4+ T cells. E, Transfected cells were stained with fluorescein isothiocyanate (FITC)–conjugated anti-CD11a or FITC-conjugated anti-CD70 antibody and analyzed by flow cytometry. F, Relative IgG production in B cells stimulated with CD4+ T cells transfected with the miR-126–expressing plasmid or with negative control plasmid is shown. Values in CF are the mean ± SD. ∗ = P < 0.05; ∗∗ = P < 0.01, versus control.

To determine whether elevating miR-126 expression in healthy T cells is sufficient to induce B cell overstimulation similar to that in lupus T cells, we cocultured control CD4+ T cells overexpressing miR-126 with purified autologous B cells (1:4 ratio) for 8 days. Supernatants were then collected, and an ELISA reaction was used to quantify IgG levels. The results showed that the presence of high levels of miR-126 stimulated significantly more robust IgG synthesis (Figure 3F). These findings indicate that miR-126 up-regulation in healthy CD4+ T cells induces a hypomethylation-dependent increase in CD11a and CD70 expression, likely due to its effect on Dnmt1 translation, leading to CD4+ T cell activation and autoreactivity.

Role of miR-126 down-regulation in restoring CD11a and CD70 promoter methylation levels and reversing autoreactivity in SLE CD4+ T cells.

To determine whether miR-126 up-regulation is necessary for DNA hypomethylation and autoimmune reactivity in patients with SLE, we transfected SLE CD4+ T cells with an miR-126 inhibitor (see Patients and Methods). Compared with the methylation levels in control-transfected CD4+ T cells from patients with SLE, those in the ITGAL and TNFSF7 promoters were significantly increased 72 hours after transfection with the inhibitor (Figures 4A–C). Consistently, we also observed that CD11a and CD70 protein and mRNA levels were down-regulated in miR-126 inhibitor–transfected CD4+ T cells from patients with SLE (Figures 4D and E). As expected, the suppression of CD11a and CD70 was accompanied by a relative decrease in IgG secretion when autologous B cells were cocultured with miR-126 inhibitor–expressing SLE CD4+ T cells compared with negative control–transfected SLE CD4+ T cells (Figure 4F). These data suggest that miR-126 is both necessary and sufficient for T cell autoreactivity and B cell hyperstimulation in patients with SLE.

Figure 4.

Inhibition of miRNA-126 reverses CD11a and CD70 DNA demethylation and suppresses CD11a and CD70 overexpression and B cell stimulation. CD4+ T cells from patients with systemic lupus erythematosus (SLE) were transfected with plasmid encoding an miR-126 inhibitor or a scrambled oligonucleotide (negative control). A, Relative methylation levels of CpG pairs in the 294-bp (−581 to −288) fragment of the CD70 promoter are shown. B, Relative methylation levels of the CpG pairs in the 310-bp (−1289 to −979) fragment of the CD11a promoter are shown. C, Overall average methylation levels of the CD70 and CD11a promoter regions were significantly increased in SLE CD4+ T cells 72 hours after transfection with miR-126 inhibitor. D, Relative CD70 and CD11a mRNA levels were determined in control and miR-126 inhibitor–transfected CD4+ T cells. E, Transfected cells were stained with FITC-conjugated anti-CD11a or FITC-conjugated anti-CD70 antibody and analyzed by flow cytometry. F, Relative IgG production in B cells stimulated with SLE CD4+ T cells transfected with miR-126 inhibitor or with negative control plasmid is shown. Values in CF are the mean ± SD. ∗ = P < 0.05; ∗∗ = P < 0.01, versus scrambled. See Figure 3 for other definitions.

Hypomethylation of the EGFL7 promoter in SLE CD4+ T cells.

To explore the mechanism by which miR-126 is up-regulated in cell samples obtained from patients with SLE, we analyzed the methylation status and expression of EGFL7. MicroRNA-126 is an intronic microRNA, located within the seventh intron of the EGFL7 locus, an intron containing 29 CpG pairs, and mature miRNA-126 is produced from the processing of EGFL7/miR-126 pre–RNA transcript rather than from its own promoter (34) (Figure 5A). Real-time PCR analysis revealed that EGFL7 expression was up-regulated in SLE CD4+ T cells compared with healthy controls (Figure 5B) and positively correlated with miR-126 expression in SLE CD4+ T cells (r = 0.538, P = 0.015) (Figure 5C). The methylation status of CpG pairs in intron 7 was then determined in 10 SLE and 10 control samples by bisulfite genomic sequencing analysis. No difference between samples from patients with SLE and samples from healthy control subjects was observed (data not shown), so we analyzed the methylation status of the CpG island within the EGFL7 promoter region and observed that the average methylation level in SLE CD4+ T cells was lower than that in healthy controls (Figures 5D–F). This suggests that the up-regulation of miR-126 in SLE CD4+ T cells is associated with promoter hypomethylation and up-regulation of EGFL7.

Figure 5.

Concomitant up-regulation of microRNA-126 (miR-126) and its host gene EGFL7, and EGFL7 promoter hypomethylation. A, Diagram depicting the EGFL7/miR-126 locus. The large grey box indicates the CpG-rich region of the EGFL7 promoter, the arrow indicates the transcription start site of EGFL7, and the white box indicates exon 1 of the EGFL7 transcript. The small grey box indicates the seventh intron, containing 29 CpG dinucleotide pairs. B, Real-time polymerase chain reaction analysis of EGFL7 mRNA normalized to GAPDH in systemic lupus erythematosus (SLE) CD4+ T cells. C, Positive correlation between miR-126 levels and EGFL7 mRNA levels in SLE CD4+ T cells. D and E, Mean methylation status of CpG pairs in the 193-bp (−291 to –98) fragment of the EGFL7 promoter in control and SLE CD4+ T cells, respectively. F, Significant decrease in the overall average methylation status of the EGFL7 promoter region in SLE CD4+ T cells compared with control CD4+ T cells. Values in B and F are the mean ± SD. ∗ = P < 0.05; ∗∗ = P < 0.01, versus control.

DISCUSSION

The results of previous studies have shed light on a potentially central role of disrupted Dnmt1 enzymatic activity in the etiology of lupus disorders (12, 13, 35). Some evidence suggests that reduced ERK pathway signaling in lupus CD4+ T cells, which is thought to play a role in autoimmunity, leads to decreases in Dnmt expression (13, 29, 35). However, the “trigger” that leads to the reduction of Dnmt1 protein levels in the CD4+ T cells of patients with SLE remains unclear. In this study, we presented compelling evidence suggesting that the Dnmt1-targeting microRNA gene miR-126 is up-regulated in SLE CD4+ T cells, thereby reducing Dnmt1 translation and leading to the hypomethylation-dependent de-repression of autoimmune-related genes.

MicroRNA are implicated in the pathology of SLE. For example, Dai et al identified several microRNA that are dysregulated in the PBMCs of lupus patients (23). Another study showed that down-regulation of miR-146a disrupts normal type I interferon pathway signaling in the PBMCs of lupus patients (36). By using microarray analysis in the present study, we identified 11 microRNA with an expression profile in SLE CD4+ T cells that was significantly different from that in healthy donor CD4+ T cells. The expression of miR-126 was shown to be up-regulated in SLE CD4+ T cells. This up-regulation is not likely to be a downstream consequence of increased T lymphocyte activity experienced by patients with SLE, because the expression of miR-126 in CD4+ T cells from healthy control subjects was not affected by CD3/CD28 stimulation. Furthermore, the degree of increase in miR-126 expression did not correlate with SLEDAI scores in our patient population. These results suggest that miR-126 up-regulation is not a consequence but rather is a potential cause of SLE disease activity.

Several groups of investigators have recently reported on microRNA that are capable of regulating DNA methylation by targeting Dnmt (25, 37, 38). Based on the predicted targets of miR-126, we hypothesized that increased levels of miR-126 might repress Dnmt1 translation. We confirmed that miR-126 can reduce Dnmt1 levels via interactions with the 3′-UTR of Dnmt1 mRNA. Furthermore, we observed that miR-126 transcript levels were inversely correlated with Dnmt1 protein levels in lupus T cells. We demonstrated that the overexpression of miR-126 in CD4+ T cells from healthy control subjects induced demethylation of the CD11a and CD70 gene loci and up-regulation of the associated proteins. As expected, the up-regulation of CD11a and CD70 in miR-126–expressing CD4+ T cells increased T cell activity and B cell stimulation. In contrast, knocking down miR-126 in SLE CD4+ T cells reduced their autoimmune activity and their stimulatory effect on IgG production in cocultured B cells. Taken together, our findings revealed that miR-126 regulates DNA methylation by targeting Dnmt1 and plays an important role in the pathogenesis of SLE.

The expression of certain microRNA genes is regulated by the methylation of their promoters (39, 40), and knocking down the expression of Dnmt1 and Dnmt3b in vitro disrupts normal microRNA expression (41). MicroRNA located in introns are transcribed in tandem with their host gene (42) and spliced from the host gene mRNA (43). MicroRNA-126 and its host gene EGFL7 can be regulated by DNA methylation, and histone deacetylation inhibitors induce the up-regulation of both gene products in cancer cells (34). Herein, we presented evidence that miR-126 and EGFL7 are concomitantly up-regulated in SLE CD4+ T cells and revealed that the EGFL7 promoter is hypomethylated in SLE patient CD4+ T cells compared with control CD4+ T cells. These data suggest that EGFL7 promoter hypomethylation may lead to miR-126 de-repression in SLE CD4+ T cells.

The results of the present study demonstrate that miR-126 plays an important role in the aberrant demethylation of CD4+ T cell DNA observed in patients with SLE. These findings raise the possibility that medical strategies aimed at causing controlled alterations to the level of autoimmune gene–targeting microRNA, such as miR-126, could be effective therapies for immune-related conditions. MicroRNA are particularly good candidates for this type of therapeutic intervention, because rather than simply turning a gene on or off, microRNA influence the level of translation of target genes and thereby serve to fine-tune the cellular response to external stimuli.

AUTHOR CONTRIBUTIONS

All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Lu had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study conception and design. S. Zhao, Wang, Liang, M. Zhao, Long, Ding, Yin, Lu.

Acquisition of data. S. Zhao, Wang, Liang, M. Zhao, Long, Ding, Yin, Lu.

Analysis and interpretation of data. S. Zhao, Wang, Liang, M. Zhao, Long, Ding, Yin, Lu.

Ancillary