Title Altered microRNA expression profile with miR-146a upregulation in CD4 + T cells from patients with rheumatoid arthritis

Introduction: Increasing evidence indicates that microRNAs (miRNAs) play a critical role in the pathogenesis of inflammatory diseases. The aim of the study was to investigate the expression pattern and function of miRNAs in CD4+ T cells from patients with rheumatoid arthritis (RA). Methods: The expression profile of miRNAs in CD4+ T cells from synovial fluid (SF) and peripheral blood of 33 RA patients was determined by microarray assay and validated by qRT-PCR analysis. The correlation between altered expression of miRNAs and cytokine levels was determined by linear regression analysis. The role of miR-146a overexpression in regulating T cell apoptosis was evaluated by flow cytometry. A genome-wide gene expression analysis was further performed to identify miR-146a-regulated genes in T cells. Results: miRNA expression profile analysis revealed that miR-146a expression was significantly upregulated while miR363 and miR-498 were downregulated in CD4+ T cells of RA patients. The level of miR-146a expression was positively correlated with levels of tumor necrosis factor-alpha (TNF-α), and in vitro studies showed TNF-α upregulated miR-146a expression in T cells. Moreover, miR-146a overexpression was found to suppress Jurkat T cell apoptosis. Finally, transcriptome analysis of miR-146a overexpression in T cells identified Fas associated factor 1 (FAF1) as a miR-146aregulated gene, which was critically involved in modulating T cell apoptosis. Conclusions: We have detected increased miR-146a in CD4+ T cells of RA patients and its close correlation with TNF-α levels. Our findings that miR-146a overexpression suppresses T cell apoptosis indicate a role of miR-146a in RA pathogenesis and provide potential novel therapeutic targets. Introduction Rheumatoid arthritis (RA) is a common chronic inflammatory disease characterized by radiographic joint destruction with severe functional deterioration and high mortality. A hallmark pathological feature of RA is the infiltration and accumulation of T cells in the synovium of joint [1]. As the shared epitope in human leukocyte antigen-DR genes is found in about 80% of RA patients, dysregulated CD4+ T cell activation and function have been investigated based on the available evidence of genetic predisposition [2,3]. T cells isolated from joint tissue and synovial fluid (SF) show an activated and memory phenotype and appear to respond poorly to stimulation with mitogen or antigens in vitro [4,5]. These T cells are unusually resistant to apoptosis in SF that contains a significant amount of proapoptotic factors such as bioactive FasL, TRAIL and TNF-α [6,7]. In addition, studies on a murine model of proteoglycan-induced arthritis also showed that CD4+ T cells failed to undergo apoptosis [8]. All these findings from patients and animal models suggest that the inhibition of T cell death may result in the persistence and accumulation of T cells in synovium, as well as the accumulation of T cells in the periphery. The long-term sur* Correspondence: [email protected] 1 Institute of Immunology, PLA, Third Military Medical University, 30# Gaotanyan Street, District Shipingba, Chongqing 400038, PR China † Contributed equally Full list of author information is available at the end of the article BioMed Central © 2010 Li et al.; licensee BioMed Central Ltd. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduc Li et al. Arthritis Research & Therapy 2010, 12:R81 http://arthritis-research.com/content/12/3/R81 Page 2 of 12 vival of CD4+ T cells has been shown to affect the behavior of synovial fibroblasts through the cell-to-cell contact and the secretion of proinflammatory factors such as Th1 and Th17 cytokines [9,10], ultimately contributing to the maintenance and exacerbation of inflammation in RA [11]. Although elevated levels of antiapoptotic proteins such as the Bcl-2 family have been found in these T cells [12], the possible mechanism underlying the impaired apoptosis of T cells in RA remains largely unclear. MicroRNAs (miRNAs) are about 22 nucleotide (nt) non-coding RNA that regulate mRNA expression at the posttranscriptional level for degradation or translational repression, which have been found to control cell division, differentiation and death [13]. To date, thousands of miRNAs have been identified in mammalian genomes, and up to 30% of human genes are regulated by them [14]. Recently, miRNAs have been recognized as a novel player in normal immune function and inflammation [15]. In particular, T cell-mediated immune responses are associated with changes in the expression of specific miRNAs. CD4+ T cells have also been found to express different miRNAs subsets that are linked to cell differentiation, maturation, activation and function [16-19]. Notably, a growing number of reports have revealed that deregulation of miRNA expression contributes to human autoimmune diseases including psoriasis and systemic lupus erythematosus [20,21], in which expression of a set of altered miRNAs are identified. In RA patients, increased expression of miR-146a, miR-155, miR-132 and miR-16 have been found in peripheral blood mononuclear cells (PBMCs) [22]. Recently, analysis of miRNA expression profile has revealed that miR-223 is overexpressed in peripheral T cells of RA patients [23]. Moreover, there is evidence that proliferation of fibroblast-like synoviocytes is regulated by miR-124a [24]. The abnormal expression of miR-146 and miR-155 is also found in the synovium of RA patients, while the expression of these miRNAs is markedly upregulated in SF after stimulation with TNF-α and IL-1β [25,26], which indicate that the inflammatory milieu may alter the miRNAs expression in joint tissue. Although a set of altered expression miRNAs have been identified in synovial tissue or PBMC of patients with RA, the functions of these dysregulated miRNAs remain largely unclear. In particular, both the expression profile and the roles of miRNAs in CD4+ T cells of RA patients have not been characterized. In this study, we revealed that miR-146a expression was upregulated in CD4+ T cells from RA patients, exhibiting a close correlation with increased levels of TNF-α. Furthermore, we demonstrated that the increased expression of miR-146a suppressed T cell apoptosis by regulating Fas-associated factor 1 (FAF1). These findings suggest that miR-146a in CD4+ T cells may play an important role in RA pathogenesis. Materials and methods Patients and control samples All patients in the study who first visited at Rheumatology Department of Southwest Hospital fulfilled the American College of Rheumatology criteria for the classification of RA. All patients in this study had been diagnosed without other complications. The patients had never been treated with oral or intra-articular corticosteroids, disease-modifying antirheumatic drugs (DMARDs) or other immunosuppressive drugs. However, nonsteroidal anti-inflammatory drugs and other symptomatic treatments were given to some patients. Samples from healthy donors were collected from 12 volunteers, half of which were men and half women. Serum and peripheral blood samples were collected from RA patients and healthy donors. SF from RA patients was collected by routine knee joint paracentesis. At least 10 ml of SF and 20 ml of peripheral blood anticoagulated by heparin were collected. All study protocols and consent forms were approved by the Institutional Medical Ethics Committee of Third Military Medical University. Written permission was obtained from all subjects who participated in the study. Microarray analysis CD4+ T cells were purified from the SF of two RA patients and peripheral blood of one healthy donor using microbeads (MACS, Miltenyi Biotec, Bergisch Gladbach, Germany). Total RNA of these CD4+ T cells was prepared using the mirVana miRNA Isolation Kit (Ambion, Austin, Texas, USA) for miRNA microarray analysis and realtime RT-PCR assay. Hybridization was carried out using a miRCURYTM Array microarray kit (Exqion, Vedbaek, Denmark), which contained 461 mature miRNA probes. The Affymetrix GeneChip (Human Genome U133 Plus 2.0 Array, Affymetrix, Santa Clara, CA) was used to analyze over 47,000 transcripts of Jurkat T cells before and after transfection with FUGW-FF3, FUGW-miR-146a or FUGW-miR-146a sponge. The expression data and gene annotations were stored in NCBI's Gene Expression Omnibus (GEO) database, accession no. [GEO:GSE21118] and [GEO:GSE21132] [27]. Quantitative RT-PCR analysis Quantitative RT-PCR assays were performed using a TaqMan® MicroRNA Assays kit (Applied Biosystems, Carlsbad, CA, USA) for the mature miRNA according to manufacturer's instructions. RUN6B small nuclear RNA was quantified as a control to normalize differences in total RNA levels. Specific primers to FAF1 for real-time Li et al. Arthritis Research & Therapy 2010, 12:R81 http://arthritis-research.com/content/12/3/R81 Page 3 of 12 PCR were designed using Roche's Applied Science Universal Probe Library Assay Design Center (Roche, Indianapolis, Indiana, USA) [28]; β-actin was quantified as the control. The primer sequences for FAF1 were as follows: 5'-CAGCGGGAGTACAACCTGA-3' (forward) and 5'AAGGTCATACACATTTCTCTTTACCTC-3' (reverse). The quantitative RT-PCR was performed by using Opticon-2 Detection System (Applied Biosystems, Carlsbad, CA, USA). All reactions were run in duplicate and repeated three times. A threshold cycle was determined in the exponential phases of amplification, and the comparative threshold cycle method was calculated with 2ΔΔT and used to calculate the relative gene expression. The value of each control sample was set at one and was used to calculate the fold change in target genes. Cloning of miR-146a gene and construction of vectors A 452 bp sequence containing the miR-146a hairpin (nucleotides 4721752-4722203 of chromosome5-EMBL reference) was amplified from human genomic DNA by PCR. The primer sequences for primary miR-146a were as follows: 5'-AATGCGGCCGCTCAAGAGATCCACCCACATC-3' (forward) and 5'-CCGACGCGTGCTACTTGGAACCCTGCTTA-3' (reverse). The miR146a-coding genomic DNA fragment was cloned downstream from the GFP gene of the lentiviral vector FUGW by digesting with BsrGI and EcoRI. Similarly, FAF1 cDNA (Open Biosystems, Huntsville, Alabama, USA) was cloned into FUGW. The plasmids were verified by sequencing. Lentivirus production and transfection Lentivirus particles derived from HIV were produced by transient co-transfecting lentiviral expression constructs into 293FT cells with pFIV-PACK Lentiviral Packaging Kit (System Biosciences, Mountain View, CA, USA) following Ca3(PO4)2 transfection protocol. The production of HIV replication incompetent lentiviral particles was performed by simultaneously delivering lentiviral transfer vectors and packaging plasmids (pSPAX2 and pMD2.G) into 293FT cells. Pseudo-viral particles generated by 293 FT cells within 48 hours were centrifuged at 100,000 g for two hours and frozen at -70°C for later experiments. Jurkat T cells were seeded into 24-well plates and lentivirus was used to transduce at a multiplicity of infection (MOI) of 10 for 48 hours. Cells were analyzed by FACs and quantitative RT-PCR. Cytokines assay The BDTM Cytometric Bead Array Human Th1/Th2 Cytokine kit (BD Biosciences, San Jose, CA, USA) was used to quantitatively measure the levels of IL-2, IL-4, IL6, IL-10, TNF-α and interferon (IFN)-γ in serum and SF according to manufacturer's instructions. CD4+ T cells from healthy donors were transfected with FUGW-FF3 and FUGW-mi-R146a, then stimulated with Phorbol myristate acetate (PMA) (50 ng/ml) and ionomycin (1 μg/ ml). Levels of IL-2, IL-17A, TNF-α and IFN-γ expression were detected by intracellular staining. Cytoplasmic expression of these cytokines was determined in the green fluorescent protein (GFP+) population by FACs. All samples were analyzed on a FACS Aria flow cytometer and all data were analyzed by FlowJo software (BD Biosciences, San Jose, CA, USA). Cell stimulation CD4+ T cells were sorted using microbeads (MACS, Miltenyi Biotec, Bergisch Gladbach, Germany). The quantity of CD4+ T cells should be at least half million and the purity should be at least 90%. Jurkat cell line was purchased from ATCC (Manassas, VA, USA). SF from 10 patients were mixed and centrifuged. The supernatant of SF mixture was collected. Then Jurkat cells and normal CD4+ T cells were stimulated by SF diluted to 1:2 or TNFα for 48 hours. Evaluation of cell proliferation and apoptosis Jurkat T cells transfected with FUGW-FF3 or FUGWmiR-146a were stimulated with PHA (30 μg/ml) for the indicated time. Cell proliferation was assayed with a cell counting kit-8 [29] (CCK-8, Dojindo, Kumamoto, Japan). Briefly, 10 μl of CCK-8 solution was added to each well of the plate, and the plate was incubated at 37°C for one hour. The absorbance was measured at 450 nm as an indicator of cell viability. All experiments were independently repeated three times. FUGW-FF3, FUGW-miR-146a and FUGW-miR-146a-FAF1 Jurkat T cells were grown in RPMI with 3% FBS and stimulated with anti-Fas antibody (100 ng/ml) for six hours. Cells were stained with Annexin V and propidium iodide (PI) for flow cytometric analysis. Statistical analysis All data were analyzed by SPSS10.0 software (SPSS, Chicago, IL, USA). Microarray data were analyzed by Cluster Analysis. All data were represented by mean values ± standard deviation. P values less than 0.05 were considered statistically significant. Independent simple T test was used to compare with different groups. The relation between miRNA and clinical demographics was analyzed by Pearson Correlation. Results Increased miR-146a expression in CD4+ T cell from SF and peripheral blood of RA patients The expression profiles of miRNAs in CD4+ T cells of SF from two RA patients were determined by miRNA microarray analysis. The clinical features and serological Li et al. Arthritis Research & Therapy 2010, 12:R81 http://arthritis-research.com/content/12/3/R81 Page 4 of 12 parameters of patients with RA were shown in Table 1. Compared with normal peripheral blood CD4+ T cells, pairwise significance analysis of microarray data indicated that the expression levels of eight miRNAs (miR363, miR-512-5P, miR-345-MM, miR-146a, miR-146b, miR-296, miR-133b and miR-150) were increased more than two-fold by a change in CD4+ T cells of SF from RA patients whereas eight miRNAs (miR-331, miR-29a, miR26a, miR-498, miR-129, let-7a, let-7d and miR-21) were significantly downregulated (Figure 1a). According to the results of microarray analysis, we focused on verifying the expression of those miRNAs whose expression levels were altered in both RA patients (Figure 1b). To confirm the microarray data, eight miRNAs with altered expression levels in both patients (miR-512-5P, miR-363, miR-345-MM1, miR-146a, miR-146b, miR-150, miR-498 and miR-129) were analyzed in CD4+ T cells from 33 patients by quantitative RT-PCR analysis. Both clinical features and serological parameters of these patients were shown in Table 2. Compared with CD4+ T cells from peripheral blood of healthy donors, miR-146a expression was significantly increased in CD4+ T cells from both peripheral blood and SF of RA patients, while miR-498 was only down-regulated in SF T cells from patients (Figure 1c). Positive correlation of increased miR-146a expression in T cell with elevated TNF-α levels To determine the possible correlation of these altered miRNAs with secreted cytokines in RA patients, we measured the levels of IL-2, IL-4, IL-6, IL-10, IFN-γ and TNFα in both SF and serum of RA patients and healthy donors [Supplemental figure S1 in Additional file 1], among which the levels of TNF-α were found to be significantly higher in SF and serum of RA patients than healthy donors. The linear regression analysis showed that only miR-146a expression was positively correlated with the levels of TNF-α in both peripheral blood and SF of RA patients, but not correlated with the levels of other cytokines (Figure 2a) and disease activity indexes such as rheumatoid factor, anti-cyclic citrullinated peptide antibody, erythrocyte sedimentation rate, C-reactive protein and disease activity score of 28 joints (DAS28) scores (data not shown). Next, Jurkat T cells and normal CD4+ T cells were stimulated with SF and TNF-α in vitro, and expression of miR-146a was found to be upregulated in a dose-dependent fashion (Figure 2b). Moreover, the expression of miR-498, miR-363 and miR-150 did not show any correlation with levels of cytokine production [Supplemental figure S2 in Additional file 2]. Thus, these data indicate that miR-146a may play a significant role in inflammation of RA. Overexpression of miR-146a suppresses T cell apoptosis but does not alter cytokine production To investigate the potential role of miR-146a in T cell functions, we amplified the coding region of miR-146a and inserted it into the lentivirus vector FUGW, in which the ubiquitin promoter drove the expression of GFP and miR-146a. Moreover, the vector FUGW-FF3 containing GFP and miR30-shRNA served as a negative control. Upon transfection with high efficiency, Jurkat T cells were found to express miR-146a 30 times more than controls (Figure 3a). PHA stimulation, Jurkat T cells with miR-146a overexpression showed a similar proliferative response as their controls (Figure 3b). However, miR146a-overexpressing Jurkat T cells displayed significantly reduced apoptosis in culture (Figure 3c). To further verify the effect of miR-146a overexpression on cytokine production in primary T cells, we transfected freshly prepared CD4+ T cells from peripheral blood of healthy donors and analyzed the profile of cytokine production after PMA stimulation. Flow cytometric analysis did not detect any significant alteration in the expression levels of IFN-γ, TNF-α, IL-2 and IL-17A in normal CD4+ T cells transfected with miR-146a [Supplemental figure S3 in Additional file 3]. miR-146a overexpression suppresses T cell apoptosis possibly by regulating FAF1 expression To search for the potential target genes of miR-146a, we used the miRNA sponge technique employing miRNA specific decoy targets to suppress endogenous miRNA activity without affecting its transcription. We constructed the miR-146a sponge by inserting two miR-146a binding sites into the lentivirus vector FUGW with a reporter gene-eGFP driven by the ubiquitin promoter. The sponge had an imperfect binding site for miR-146a seed region, with a bulge at positions 9 to 12, which can prevent endonucleolytic cleavage (Figure 4a). We transfected the miRNA sponge into Jurkat T cells and compared the gene expression profiles when miR-146a expression was upand down-regulated. Together, six genes were found to be regulated by altered levels of miR146a expression (Figure 4b), among which the FAF1 expression was negatively correlated with the levels of miR-146a as further confirmed by quantitative RT-PCR analysis (Figure 4c). Next, FAF1 was cloned into lentivirus vector FUGW and plasmid FUGW-miR-146a and transfected into Jurkat T cells. We further evaluated Fasinduced apoptosis of Jurkat T cells when transfected with either miR-146a alone or together with FAF1. It was found that miR-146a-overexpressing T cells showed significantly reduced apoptosis while co-overexpressing FAF1 abrogated this effect, indicating that miR-146a supLi et al. Arthritis Research & Therapy 2010, 12:R81 http://arthritis-research.com/content/12/3/R81 Page 5 of 12 Figure 1 The altered expression profile of miRNAs in CD4+ T cells from SF of RA patients. (a) Compared with normal CD4+ T cells, altered expression of 16 microRNAs (miRNAs) with fold change (>2) was found in CD4+ T cells of rheumatoid arthritis (RA) by miRNA microarray analysis. (red: upregulation; green: downregulation) (b) Expression of eight miRNAs was altered in both RA patients by miRNA microarray analysis. Six of them were upregulated and two of them were downregulated. (c) Compared with healthy controls, levels of miRNAs miR-498, miR-363, miR-146a, miR-150, miR512-5P, miR-345-MM1, miR-146b and miR-129 expression in peripheral blood (PB; n = 20) and synovial fluid (SF; n = 22) of RA patients and PB (n = 12) of healthy controls were determined by Taqman quantitative RT-PCR analysis. Horizontal bars represent the mean values. Ctl-PB: CD4+ T cells from PB of healthy controls; RA-PB: CD4+ T cells from PB of RA patients; RA-SF: CD4+ T cells from SF of RA patients. **P < 0.01, *P < 0.05. Li et al. Arthritis Research & Therapy 2010, 12:R81 http://arthritis-research.com/content/12/3/R81 Page 6 of 12 presses apoptosis possibly via regulating the expression of FAF1 gene in T cells (Figure 5). Discussion Increasing evidence indicates that aberrant expression of miRNAs is implicated in the pathogenesis of autoimmune diseases. Accumulating data have suggested that the proper regulation of miRNA expression is important in the maintenance of normal immune functions and prevention of autoimmunity. A number of miRNAs have been found to show a tissue-specific pattern during cancer development and viral infection. Previous studies have shown that increased miR-17-92 expression in lymphocytes leads to the development of lymphoproliferative disorder and autoimmune disease in mice [30]. Recently, upregulated miR-146a has been found in activated primary T cells and memory T cells [31]. Moreover, increased miR-146a expression has been detected primarily in PBMC and synoviocytes from RA patients [22,25]. Although deregulation of miRNA expression has been observed in human autoimmune diseases, it is still largely unclear how miRNAs affect autoimmune pathogenesis in patients. Although a set of altered expression miRNAs are recently identified in both PBMC and synovial tissue from RA patients, neither miRNAs expression profile nor their roles have been fully characterized in CD4+ T cells of RA patients. In this study, we have selected the RA patients who did not have any DMARDs therapy because it is known that drug treatment can affect expression of miRNAs [32]. We have found CD4+ T cells from both SF and peripheral blood of RA patients exhibiting a specific miRNAs expression profile. Although miR-146a expression was not found to be associated with the disease activity index in the current study, our results have shown a significant positive correlation between miR-146a and TNF-α in both peripheral blood and SF by dependability statistical analysis. Furthermore, we have found that miR-146a expression is upregulated in CD4+ T cells in response to TNF-α or SF stimulation in vitro. Thus the inflammatory milieu might alter miR-146a expression in infiltrated T cells. TNF-α is a critical mediator of the inflammatory pathway in the rheumatoid joints. As TNF-α inhibition therapy appears to dramatically reduce markers of inflammation and slow joint structural damage. Our findings of overexpression of miR-146a in T cells may have functional implications in eliciting joint inflammation of RA patients. In addition to its increasingly recognized function in modulating innate immunity, miR-146a has been shown to be involved in Th1/Th2 polarization and regulatory T cell development [33-36], indicating a potential role for miR-146a in autoimmune response. Notably, miR-146a was found to be up-regulated in skin lesions of psoriasis, PBMC and synovial tissue of RA, whereas it was downregulated in PBMC of systemic lupus erythematosus [20Table 1: Demographics clinical features of RA patients Parameter Patient 1 Patient 2