MiR-92a inhibits fibroblast-like synoviocyte proliferationand migration in rheumatoid arthritis by targeting AKT2

FANG-YUAN YU, CONG-QIN XIE, CHANG-LIANG JIANG, JI-TONG SUN, HUI-CHENG FENG,CHAO LI and XUN-WU HUANG*

Department of Orthopaedics, 309th Hospital of Chinese PLA, Beijing 100091, People’s Republic of China

*Corresponding author (Email, huangxunwu_309@163.com)

MS received 3 March 2018; accepted 17 July 2018; published online 4 October 2018

Growing data have indicated that the miR-17–92 cluster is implicated in inflammatory response and rheumatoid arthritis(RA). This study was aimed to investigate the effects of miR-92a on the proliferation and migration of rheumatoid arthritisfibroblast-like synoviocytes (RA-FLSs). Our results showed that miR-92a was significantly down-regulated in RA synovialtissue and RA-FLSs, whereas the protein level of AKT2 is increased. Restoration of miR-92a suppressed the proliferationand migration of RA-FLSs. Down-regulation of miR-92a promotes proliferation and migration of normal human FLSs.Dual luciferase reporter gene assay showed that miR-92a could specifically bind with the 30UTR of AKT2 and significantlyrepressed the luciferase activity. Down-regulation or up-regulation of miR-92a significantly increased or decreased theprotein and phosphorylation levels of AKT2. siRNA-mediated down-regulation of AKT2 significantly prevented cellproliferation and migration of RA-FLSs, which were similar to the effects induced by overexpression of miR-92a.Moreover, AKT2 overexpression rescued miR-92a-mediated suppressive effect on proliferation and migration of RA-FLS.Thus, miR-92a could inhibit the proliferation and migration of RA-FLSs through regulation of AKT2 expression.

Keywords. AKT2; MiR-92a; rheumatoid arthritis; RA-FLS

1. Introduction

Rheumatoid arthritis (RA) is a kind of chronic joint inflam-matory autoimmune disease, characterized by abnormal syn-ovial hyperplasia, tissue destruction and functional disabilityinduced by inflammatory cytokines and proteases (McInnesand Schett 2011). RA that has incredibly high morbidity andmortality rates is affecting 0.5–1% adults in developing anddeveloped countries (Covic et al. 2012; Hajiabbasi et al. 2016).Many studies have reported that fibroblast-like synoviocytes(FLSs) are key effector cells in the occurrence and progressionof RA (Audo et al. 2015; Kaneko et al. 2016; Xing et al. 2016).FLSs from RA patients exhibit ‘tumor-like’ properties includ-ing hyperproliferation, decreased apoptosis and aberrantmigration (Karouzakis et al. 2009; Bottini and Firestein 2013).ActivatedRA-FLSsmigrate and invade into cartilage and bone,which causes synovial pannus formation and joint destruction(Mor et al. 2005; Bartok and Firestein 2010; McInnes andSchett 2011). Accumulated evidence suggests that inhibition ofproliferation and migration of FLSs may be promising for RAtreatment (McInnes and Schett 2011). In the current study, theelements that affect FLSs activities were investigated.

MicroRNAs (miRNAs) are small (* 22 nucleotide),highly conserved endogenous non-coding RNAs that

regulate about 30% of gene expression (Guo et al. 2010).Accumulating evidence indicates that miRNAs regulateactivation of FLSs and involved in pathological develop-ment of RA (Yang and Yang 2015; Mu et al. 2016). ThemiRNA cluster is a polycistronic gene which encodes sev-eral miRNAs in a single primary, or nascent, transcript(Sakai et al. 2017). The MiR-17–92 cluster is located in theC13orf25 (chromosome 13 open reading frame 25), whichencodes six distinct miRNAs in a single primary transcript(miR-17, miR-18a, miR-19a, miR-19b, miR-20a and miR-92a) (Philippe et al. 2013). Growing data have indicated thatmiR-17–92 cluster is implicated in inflammatory responseand RA (O’Connell et al. 2010; Philippe and Alsaleh 2013).For instance, miR-19a/b affects the production of inflam-matory cytokines such as IL-6 in LPS-activated RA synovialfibroblasts (Philippe et al. 2012). MiR-18a promotes chronicinflammation in the joint (Trenkmann et al. 2013). MiR-17suppresses expression of inflammatory cytokines andinflammatory responses in rheumatoid arthritis synovialfibroblasts (Akhtar et al. 2016). These studies suggest thatmiR-17–92 cluster is potential targets in RA treatment.

AKT2 belongs to the serine/threonine kinase family, whichregulates many pathological processes such as RA develop-ment (Xu et al. 2017). The blockage of AKT pathway could

http://www.ias.ac.in/jbiosci 911

J Biosci Vol. 43, No. 5, December 2018, pp. 911–919 � Indian Academy of SciencesDOI: 10.1007/s12038-018-9803-0

inhibit RAprogression by suppressing IL-21 inducedmigrationof FLSs (Xing and Jin 2016). In addition, inhibition ofAKT2 inosteoblasts prevents osteoporosis in RA (Kaneshiro et al.2014). Moreover, activation of PI3 K/AKT pathway promotesRA-FLS proliferation and inflammation (Li et al. 2014). It alsohas been reported that miR-650 inhibits proliferation andmigration of rheumatoid arthritis synovial fibroblasts by tar-getingAKT2 (Xu andChen 2017). These results suggestAKT2as a potential therapeutic target of RA.

The aim of this study was to discovery miRNAs directlytargeting AKT2 and regulating activities of FLSs. Here, wefound that miR-92a directly targeted 30UTR of AKT2.Moreover, miR-92a inhibited proliferation and migration ofRA-FLSs through targeting AKT2.

2. Material and methods

2.1 Object of study

Synovial tissue samples were obtained from 19 RApatients (14 females and 9 males, 33–73 years old, mean51 years old) and 13 normal control patients (5 femalesand 8 males, 28–69 years old, mean 49 years old), whowere undergoing joint replacement surgery or synovec-tomy at department of Orthopaedics, 309th Hospital ofChinese PLA from 2012 to 2017. RA was diagnosedaccording to the American College of Rheumatology cri-teria for classification of disease (Arnett et al. 1988).Normal control patients who suffered joint trauma are freeof other diseases such as autoimmune disease, infectiousdisease and cancer. This study was approved by the EthicsCommittee of 309th Hospital of Chinese PLA. Informedwritten consent was obtained from all participants.

2.2 Cells and cell culture

Normal human FLSs and RA-FLSs were purchased from CellApplications (San Diego, CA, USA). Human embryonickidney 293 (HEK293) cells were obtained from the AmericanType Culture Collection (ATCC). RA-FLSs and HEK293were cultured in Dulbecco’s modified Eagle’s medium(DMEM) supplemented with 10% fetal bovine serum (FBS;PAALaboratories, Pasching, Austria) at 37�Cwithin 5%CO2.

2.3 Quantitative realtime polymerase chain reaction(qRTPCR) analysis

Total RNA, including miRNA, was extracted from synovialtissues and cells using TRIzol (Invitrogen, La Jolla, CA,USA) according to manufacturer’s instructions. For miRNAquantification, reverse transcription was performed using the

TaqMan microRNA reverse transcription kit (AppliedBiosystems, Foster City, CA, USA). miR-92a and AKT2expression was quantified using TaqMan microRNA Assays(Applied Biosystems, Foster City, CA, USA) according tothe manufacturer’s instructions. U6 snRNA was used as anendogenous control.

2.4 Protein extraction and Western blot

After transfection for 48 h, cells were lysed using RIPAbuffer (Beyotime Biotech Inc., China) on ice for 15 min.Protein concentrations were determined by the BCA Pro-tein Assay kit (Applygen, Beijing, China). Equal protein ineach sample was separated by 8% sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) andtransferred onto polyvinylidene fluoride (PVDF) membrane(Bio-Rad, USA). The membranes were blocked with 5%nonfat milk for 30 min, and then incubated with primaryantibodies including anti-AKT2 (1:1000 dilution, SantaCruz, USA), anti-p-AKT2(Ser474) (1:1000, Cell SignalingTechnology), anti-AKT1 (1:1000, Cell Signaling Technol-ogy), anti-AKT3 (1:1000, Cell Signaling Technology) andanti-GAPDH (1:2000 dilution, Santa Cruz, USA) at 4�Covernight. Each membrane was rinsed three times for10 min, then incubated with the secondary horseradishperoxidase-linked antibodies. The protein bands weredetected using anti-rabbit or mouse IgG conjugated withHRP, followed by use of the ECL Plus western blottingdetection system (GE Healthcare, UK) according to themanufacturer’s instructions. GAPDH was used as theloading reference for data analysis.

2.5 Plasmid construction and cell infection

FLS were seeded 60–70% confluency in a medium withoutantibiotics 24 h prior to transfection. MiRNA control mimic(miR-Control), miR-92a mimic, miRNA control inhibitor(anti-NC), and miR-92a inhibitor (anti-miR-92a) (AppliedBiosystems, USA) were transfected at a final concentrationof 100 nM with using Lipofectamine RNAiMAX transfec-tion reagent (Invitrogen, USA) according to the manufac-turer’s instruction. For AKT2 overexpression, the wild-typeAKT2 cDNA sequence without the 30UTR was subclonedinto pLenti6/V5 (Invitrogen, USA). The production andpurification of lentivirus and cell infection were performedas previously described (Zheng et al. 2012; Yuan et al.2013). Empty lentiviral vectors (Vector) were used as controlof lenti-AKT2. SiRNAs of AKT2 were obtained from Santacruz (St. Louis, Missouri, USA) and were transfected intocells using Lipofectamine 2000 reagent (Invitrogen, USA).The cells were used for next experiments after transfectionfor 48 h.

912 Fang-Yuan Yu et al.

2.6 Dual luciferase reporter assays

The human AKT2 30UTR (WT) or mutated (AKT2-MUT)fragment of AKT2 30UTR that contained the putative miR-92a binding sites were cloned into pIS2 luciferase reportervector (Addgene) according to a previous description (Zhanget al. 2011). For the luciferase reporter assay, FLS wereseeded onto 24-well plates at a density of 1 9 105 cells/welland co-transfected with 100 ng of the luciferase reportervectors and 20 pmol of miR-control or miR-92a mimic. At48 h after transfection, the cells were lysed and measured forluciferase activities using the Dual-luciferase Reporter AssaySystem (Promega). The activity of firefly luciferase wasnormalized to that of Renilla luciferase.

2.7 Cell proliferation

Cell proliferation was detected by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay. Aftertransfection for 48 h, FLS were seeded onto 96-well plates(5 9 103 cells/well) and incubated for 24, 48 or 72 h. TheMTT solution (20 ml, 5 mg/mL, Sigma-Aldrich, USA) wasadded and incubated for additional 4 h at 37�C incubatorwith 5% CO2. Dimethyl sulphoxide (DMSO, 300 ll) wasthen added to dissolve the crystals. Absorbance of sample at570 nm was read using a spectrophotometer.

2.8 Cell migration assays

After transfection for 48 h, cell migration assay was per-formed using Transwell chambers with 8-lm pores (Mill-pore, USA). FLS (5 9 104 cells per chamber) weresuspended in serum-free DMEM in the upper chambers intriplicate. DMEM containing 1% FBS was added to thelower chamber as a chemoattractant. The chambers wereincubated at 37�C within 5% CO2 for 6 h. After incubation,the cells were fixed in methanol for 10 min at room tem-perature and stained with 0.1% crystal violet. The non-mi-grating cells on the upper surface of the filter membranewere removed by a cotton swab. The stained cells on thebottom side of the membrane were observed, photographed,and counted in five representative microscopic fields(100 9 magnification).

2.9 Statistical analyses

Data were present as means ± standard deviation (SD). Thesoftware of SPSS19.0 software (Chicago, IL) was used forstatistical analysis. For comparisons of three ormoregroups, theanalyses were performed by one-way analysis of variance(ANOVA) with Dunnett’s post hoc test. Student’s t-test was

performed to assess differences between two groups. P-valuesless than or equal to 0.05 were considered significant.

3. Results

3.1 Expression of miR-92a in synovial tissues and FLSs

To investigate the effects of miR-92a on RA-FLSs andrheumatoid arthritis, qRT-PCR was performed to detect theexpression of miR-92a in synovial tissue from 19 RApatients and 13 healthy controls. The results showed that theexpression of miR-92a was significantly down-regulated inRA patient compared to healthy controls (P\ 0.01, fig-ure 1A). Oppositely, the expression of AKT2 was increasedin RA patients (P\ 0.01, figure 1B). Moreover, theexpression of miR-92 was lower in human RA-FLSs thanthat in normal human FLSs (P\ 0.05, figure 1C), whereasthe protein level of AKT2 is increased in human RA-FLSs(P\ 0.05, figure 1D). These data suggested that down-regulation of miR-92a and up-regulation of AKT2 might beinvolved in the pathogenesis of RA.

Figure 1. miR-92a is down-regulated in RA synovial tissues andFLSs. (A and B) The expression of miR-92a and AKT2 insynovial tissues from RA patients (n = 19) and healthy controls(n = 13) was detected by qRT-PCR analysis. (C) qRT-PCRanalysis of miR-92a levels in normal human FLSs and RA-FLSs.(D) Western blot analysis of AKT2 protein levels in normal humanFLSs and RA-FLSs. **P\ 0.01 ***P\ 0.001.

Inhibitory effect of miR-92a on rheumatoid arthritis 913

3.2 Restoration of miR-92a suppressesthe proliferation and migration of RA-FLSs

To further elucidate the function of miR-92a in RA-FLSsand rheumatoid arthritis, its expression was restored bytransfection with miR-92a mimic in RA-FLS. A 3.3-foldincrease in miR-92a level was observed after transfection for48 h (figure 2A). Moreover, up-regulation of miR-92a sig-nificantly prevented cell proliferation and migration of RA-FLSs (figure 2B and C). These data suggested that miR-92awas a growth and migration suppressor of human RA-FLSs.

3.3 Down-regulation of miR-92a promotesproliferation and migration of normal human FLSs

To further confirm the inhibitory effects of miR-92a onproliferation and migration of RA-FLSs, we detected theproliferation and migration of normal human FLSs in whichmiR-92a were down-regulated by transfection with miR-92ainhibitor. The expression of miR-92a significantly decreasedafter transfection for 48 h (figure 3A). As expectedly, down-regulation of miR-92a markedly promoted the proliferationand migration of normal human FLSs (figure 3B and C),which confirmed the inhibitory effects of miR-92 on cellproliferation and migration of FLSs.

3.4 AKT2 is a direct target of miR-92a

Using targetsan database, AKT2 was identified as a potentialtarget gene of miR-92a. To verify the binding site, the30UTR of AKT2 containing the wild-type (AKT-WT) ormutated (AKT-MUT) seed-sequence of miR-92a shown infigure 4A was cloned into a firefly luciferase reporter vector.Then, dual luciferase assays were performed. Compared

with the miR-control group, miR-92a significantly inhibitedrelative luciferase activity when co-transfected with theAKT-WT reporter plasmid. However, miR-92a-mediatedinhibitory effect was not observed in the AKT-MUT reportertransfected cells (figure 4B). These data suggested that miR-92a directly bound to the predicted binding site that ismutated in the study. In addition, the protein level andactivity of AKT2 was decreased in RA-FLSs transfectedwith miR-92a mimic, whereas increased in normal FLSstransfected with miR-92a inhibitor (figure 4C–F). However,the expression changes of miR-92a did not affect theexpression levels of AKT1 and AKT3 (figure 4C–F). Theseresults demonstrated that miR-92a may inhibited AKT2expression by directly binding to the 30UTR of AKT2.

3.5 Knocking down of AKT2 inhibits the proliferationand migration of RA-FLSs

To further determine the role of AKT2 in RA-FLSs andrheumatoid arthritis, AKT2 was knocked down in RA-FLSsand cell proliferation and migration were detected by MTTassay and transwell migration assay, respectively. As shownin figure 5A, delivery of AKT2 siRNA 1 and 2 significantlydecreased the expression of AKT2. siRNA-mediated down-regulation of AKT2 significantly prevented cell proliferationand migration of RA-FLSs, which were similar to the effectsinduced by miR-92a overexpression.

3.6 AKT2 overexpression rescues miR-92a-mediatedsuppressive effect on proliferation and migration of RA-FLS

The rescue experiments were further performed to evaluatewhether overexpression of AKT2 could reverse the miR-

Figure 2. Restoration of miR-92a suppresses the proliferation and migration of RA-FLSs. (A) qRT-PCR analysis of miR-92a levels inRA-FLSs transfected with the miR-92a mimic or miR-control. (B) MTT assay was performed to test the viability of RA-FLSs transfectedwith the miR-92a mimic or miR-control after culturing for 24, 48 or 72 h. (C) Cell migration was determined by transwell migration assay.Bar graphs represent data from three independent experiments. Data are expressed as mean ± SD. ***P\ 0.001.

914 Fang-Yuan Yu et al.

Figure 3. Down-regulation of miR-92a promotes the proliferation and migration of normal human FLSs. Normal human FLSs weretransfected with the miR-92ap inhibitor (anti-miR-92a) or corresponding control (anti-NC) for 48 h. (A) qRT-PCR analysis of miR-92alevels in normal human FLSs transfected with the anti-miR-92a or anti-NC. (B) MTT assay was performed to test the viability of normalhuman FLSs transfected after culturing for 24, 48 or 72 h. (C) Cell migration was determined by transwell migration assay. Bar graphsrepresent data from three independent experiments. Data are expressed as mean ± SD. ***P\ 0.001.

Figure 4. Identification of AKT2 as a direct target of miR-92a. (A) Bioinformatic analysis revealed that the 30UTR of AKT2 mRNAcontained a putative miR-92a target site. (B) Luciferase reporter assay. 293T cells were co-transfected with AKT2 30UTR reporterconstructs harbouring wild-type or mutated miR-92a target site (shown in A) and miR-92a mimic or miR-control. Firefly luciferase activitywas normalized to Renilla luciferase activity. (C–D) RA-FLSs were transfected with miR-92a mimic (miR-92a) or miR-control for 48 h.(C) qPCR was used to detect the expression of miR-92a. (D) Western blot analysis of AKT2 protein levels in RA-FLSs transfected with themiR-92a or miR-control. (E–F) Normal human FLSs were transfected with anti-miR-92a or anti-NC for 48 h. (E) qPCR was used to detectthe expression of miR-92a. (F) Western blot analysis of AKT2 protein levels in Normal human FLSs transfected with anti-miR-92a or anti-NC. Bar graphs represent data from three independent experiments. Data are expressed as mean ± SD. **P\ 0.01, ***P\ 0.001.

Inhibitory effect of miR-92a on rheumatoid arthritis 915

92a-mediated suppressive effects on proliferation andmigration of RA-FLSs. We enforced the expression ofAKT2 in miR-92a-overexpressing RA-FLSs by infectionlentivirus vectors containing AKT2 (AKT2). The overex-pression of AKT2 did not affect the expression of miR-92a(figure 6A). Moreover, AKT2 is overexpressed in miR-92aoverexpressed RA-FLSs (figure 6A and B). As shown infigure 6B, the expression level of AKT2 in miR-92a andAKT2 co-overexpressed group was similar with that in miR-control transfected group. In addition, AKT2 overexpressionrestored the inhibitory effect of miR-92a on cell proliferationand migration of RA-FLSs (figure 6C and D). These resultssuggested that miR-92a exerted its inhibitory role in RA-FLSs proliferation and migration by targeting AKT2.

4. Discussion

RA is an inflammatory autoimmune disease that are char-acterized by joint synovial inflammation (McInnes andSchett 2011). MiRNAs can regulate post-transcriptionalmRNA expression, which are involved in many pathologicalprogression including RA development (Filkova et al.2012). Accumulating data have indicated that miRNAsmediate the proliferation and migration of RA-FLSs (Yangand Yang 2015). It has reported that miRNAs from miR-17–92 cluster play an important role in RA development andinflammatory regulation (O’Connell and Rao 2010; Philippeand Alsaleh 2013). However, the role of miR-92a in RAremains unclear. Here, our data suggested that miR-92a wassignificantly decreased in RA synovial tissue and FLSs. Thedown-regulation of miR-92a in RA patients suggested itscritical role in RA development.

As amajor cell population in synovial tissues, RA-FLSs areinvolved in RA development. Activations of RA-FLSs

including proliferation and migration contribute to pannusformation and tissue damage during RA progression (Lefevreet al. 2009). Previous study has demonstrated that inhibition ofRA-FLSs protects against joint destruction in RA (Lao et al.2016). It has been considered that RA-FLSs may be a noveltherapeutic target for RA (Li et al. 2015). Therefore, we furtherinvestigate the regulatory role of miR-92a in RA-FLSs. Ofnote, our results have shown that miR-92a suppresses theproliferation andmigration of RA-FLSs, which suggests that itplays an inhibitory role in RA progression.

To further investigate the mechanism by which miR-92ainhibits the proliferation, migration and invasion of RA-FLSs, we sought to identify its direct target genes. It ispredicted by targetscan database that miR-92a potentiallytargeted 30UTR of AKT2. This prediction was confirmed byluciferase reporter assay that suggested the binding betweenmiR-92a and the 30UTR of AKT2. In addition, enforcedexpression of miR-92a significantly decreased the levels ofendogenous AKT2 in RA-FLSs. We further observed thatAKT2 was up-regulated in RA patients and RA-FLSs, whichis consistent with previous study (Xu and Chen 2017).

AKT comprises three isoforms including Akt1, Akt2 andAkt3 that aremammalian serine/threonine kinases (Cohen 2013;Yu et al. 2015). Since AKT2 exhibits regulatory role in cellgrowth and migration in different types of cancer cells (Shenet al. 2013; Rensing et al. 2014; Yu and Littlewood 2015;Wanget al. 2016). We hypothesized that miR-92a inhibited cell pro-liferation and migration by down-regulating AKT2. In line withthe hypothesis, our data demonstrated that silencing down-reg-ulation of AKT2 phenocopied the effects of miR-92a overex-pression on RA-FLSs, which caused a significant reduction ofRA-FLSs proliferation and migration. Rescue experimentsusing a miRNA-resistant AKT2 variant also confirmed the linkbetween miR-92a and AKT2 in the regulation of RA-FLSproliferation andmigration. In agreement, previous studies have

Figure 5. Silencing of AKT2 inhibits cell proliferation and migration of RA-FLSs. RA-FLSs were transfected with control (scramble) orAKT2-targeting siRNAs (AKT2 siRNA 1 and 2) for 48 h. (A) Western blot analysis of AKT2 protein levels in RA-FLSs. (B) MTT assaywas performed to detect cell proliferation after culturing for 24, 48 or 72 h. (C) Cell migration was determined by transwell migration assay.Bar graphs represent data from three independent experiments. *P\ 0.05 versus cells transfected with scramble. Data are expressed asmean ± SD. ***P\ 0.001.

916 Fang-Yuan Yu et al.

reported the promoting effect of AKT2 on cell proliferation andmigration (Shen and Zhang 2013; Rensing and de Jager 2014;Yu and Littlewood 2015; Wang and Wu 2016). It has beenreported that PI3 K/AKT2 signalling pathway inhibits osteo-blast differentiation in the pathogenesis of RA (Kaneshiro andEbina 2014). Besides, AKT2 increased the expression ofPGLYRP1, IL36A and MMP13 in RAFLSs, which led to pro-motion of joint destruction (Hartkamp et al. 2014). Moreover,the inhibition of PI3 K/Akt/mTOR pathway could lessen bonedestruction in RA (Kim et al. 2012). Thus, miR-92a may sup-pressed the proliferation and migration of RA-FLSs by pre-venting AKT2-mediated production of MMPs and cytokines.

In conclusion, miR-92a is down-regulated in RA synovialtissues and RA-FLSs. Besides, restoration of miR-92a sup-presses cell proliferation and migration of RA-FLSs by tar-geting AKT2. These findings suggest that the miR-92a/AKT2 pathway may play an important role in the

pathogenesis of RA and represent a novel target for pre-vention and treatment of RA.

Acknowledgements

This study was supported by National Science Foundation ofChina (No. 30801159), the 12th Five Year Programs for theMedical Research of Chinese PLA (CWS12J014), TheCapital Health Research and Development of Special grant(39770714 and 2016-3-5071).

References

Akhtar N, Singh AK and Ahmed S 2016 MicroRNA-17 SuppressesTNF-alpha signaling by interfering with TRAF2 and cIAP2

Figure 6. Enforced expression of AKT2 counteracts the biological effects of miR-92a on RA-FLSs. RA-FLSs were infected andtransfected with lentivirus vector carrying AKT2 lacking 30UTR (AKT2) along and miR-92a- or miR-control and empty vector control.(A) qPCR was used to explore the expression of miR-92a. (B) Western blot analysis was conducted to determine the protein level of AKT2.(C) MTT assay was used to detect the proliferation of RASFs after culturing for 24, 48 or 72 h. (D) Transwell migration assay were used todetected RA-FLSs migration. Bar graphs represent data from three independent experiments. *P\ 0.05 ***P\ 0.001.

Inhibitory effect of miR-92a on rheumatoid arthritis 917

association in rheumatoid arthritis synovial fibroblasts. J.Immunol. 197 2219–2228

Arnett FC, Edworthy SM, Bloch DA, et al. 1988 The AmericanRheumatism Association 1987 revised criteria for the classifi-cation of rheumatoid arthritis. Arthritis Rheum. 31 315–324

Audo R, Hegglin A, Severac D, Dantec C, Combe B, Hahne M andMorel J 2015 Identification of genes regulating TRAIL-inducedapoptosis in rheumatoid arthritis fibroblasts-like synoviocytes.Genes Immun. 16 462–469

Bartok B and Firestein GS 2010 Fibroblast-like synoviocytes: keyeffector cells in rheumatoid arthritis. Immunological Rev. 233233–255

Bottini N and Firestein GS 2013 Duality of fibroblast-likesynoviocytes in RA: passive responders and imprinted aggres-sors. Nat. Rev. Rheumatol. 9 24–33

Cohen MM Jr. 2013 The AKT genes and their roles in variousdisorders. Am. J. Med. Genet. Part A. 161A 2931–2937

Covic T, Cumming SR, Pallant JF, Manolios N, Emery P,Conaghan PG and Tennant A 2012 Depression and anxiety inpatients with rheumatoid arthritis: prevalence rates based on acomparison of the Depression, Anxiety and Stress Scale (DASS)and the hospital, Anxiety and Depression Scale (HADS). BMCPsychiatry 12 6

Filkova M, Jungel A, Gay RE and Gay S 2012 MicroRNAs inrheumatoid arthritis: potential role in diagnosis and therapy.BioDrugs Clin. Immunother. Biopharm. Gene Therapy 26131–141

Guo H, Ingolia NT, Weissman JS and Bartel DP 2010 MammalianmicroRNAs predominantly act to decrease target mRNA levels.Nature 466 835–840

Hajiabbasi A, Shenavar Masooleh I, Alizadeh Y, Banikarimi ASand Ghavidel Parsa P 2016 Secondary Sjogren’s syndrome in 83patients with rheumatoid arthritis. Acta Medica Iranica. 54448–453

Hartkamp LM, van Es IE, Malvar Fernandez B, Tak PP, ReedquistKA2014TheAGCkinases protein kinase B (PKB) and serum andglucocorticoid kinase (SGK) differentially regulate the metabolicactivity and inflammatory activation of rheumatoid arthritisfibroblast-like synoviocytes and humanmacrophages.AnnRheumDis. https://doi.org/10.1136/annrheumdis-2014-eular.5096

Kaneko S, Kondo Y, Yokosawa M and Sumida T 2016 Rheumatoidarthritis and cytokines. Nihon Rinsho. 74 913–918

Kaneshiro S, Ebina K, Shi K, et al. 2014 IL-6 negatively regulatesosteoblast differentiation through the SHP2/MEK2 and SHP2/Akt2 pathways in vitro. J. Bone Mineral Metab. 32 378–392

Karouzakis E, Gay RE, Gay S and Neidhart M 2009 Epigeneticcontrol in rheumatoid arthritis synovial fibroblasts. Nat. Rev.Rheumatol. 5 266–272

Kim TH, Choi SJ, Lee YH, Song GG, Ji JD 2012 Combinedtherapeutic application of mTOR inhibitor and vitamin D(3) forinflammatory bone destruction of rheumatoid arthritis. MedHypotheses 79 757–760

Lao M, Shi M, Zou Y, et al. 2016 Protein inhibitor of activatedSTAT3 regulates migration, invasion, and activation of fibrob-last-like synoviocytes in rheumatoid arthritis. J. Immunol. 196596–606

Lefevre S, Knedla A, Tennie C, et al. 2009 Synovial fibroblastsspread rheumatoid arthritis to unaffected joints. Nat. Med. 151414–1420

Li GF, Qin YH and Du PQ 2015 Andrographolide inhibits themigration, invasion and matrix metalloproteinase expression ofrheumatoid arthritis fibroblast-like synoviocytes via inhibition ofHIF-1alpha signaling. Life Sci. 136 67–72

Li X, Zhang Z, Peng A, et al. 2014 Effect of CD95 oninflammatory response in rheumatoid arthritis fibroblast-likesynoviocytes. Cell. Immunol. 290 209–216

McInnes IB and Schett G 2011 The pathogenesis of rheumatoidarthritis. New Engl. J Med. 365 2205–2219

Mor A, Abramson SB and Pillinger MH 2005 The fibroblast-likesynovial cell in rheumatoid arthritis: a key player in inflamma-tion and joint destruction. Clin. Immunol. 115 118–128

Mu N, Gu J, Huang T, et al. 2016 A novel NF-kappaB/YY1/microRNA-10a regulatory circuit in fibroblast-like synovio-cytes regulates inflammation in rheumatoid arthritis. Sci. Rep. 620059

O’Connell RM, Rao DS, Chaudhuri AA and Baltimore D 2010Physiological and pathological roles for microRNAs in theimmune system. Nat. Rev. Immunol. 10 111–122

Philippe L, Alsaleh G, Bahram S, Pfeffer S and Georgel P 2013The miR-17 approximately 92 cluster: A key player in thecontrol of inflammation during rheumatoid arthritis. FrontImmunol. 4 70

Philippe L, Alsaleh G, Suffert G, et al. 2012 TLR2 expression isregulated by microRNA miR-19 in rheumatoid fibroblast-likesynoviocytes. J. Immunol. 188 454–461

Rensing KL, de Jager SC, Stroes ES, et al. 2014 Akt2/LDLr doubleknockout mice display impaired glucose tolerance and developmore complex atherosclerotic plaques than LDLr knockoutmice. Cardiovasc. Res. 101 277–287

Sakai A, Saitow F, Maruyama M, et al. 2017 MicroRNA clustermiR-17-92 regulates multiple functionally related voltage-gatedpotassium channels in chronic neuropathic pain. Nat. Commun.8 16079

Shen YH, Zhang L, Ren P, et al. 2013 AKT2 confers protectionagainst aortic aneurysms and dissections. Circulation Res. 112618–632

Trenkmann M, Brock M, Gay RE, Michel BA, Gay S and HuberLC 2013 Tumor necrosis factor alpha-induced microRNA-18aactivates rheumatoid arthritis synovial fibroblasts through afeedback loop in NF-kappaB signaling. Arthritis Rheum. 65916–927

Wang SH, Wu XC, Zhang MD, Weng MZ, Zhou D and Quan ZW2016 Long noncoding RNA H19 contributes to gallbladdercancer cell proliferation by modulated miR-194-5p targetingAKT2. Tumour Biol. 37 9721–9730

Xing R, Jin Y, Sun L, et al. 2016 Interleukin-21 induces migrationand invasion of fibroblast-like synoviocytes from patients withrheumatoid arthritis. Clin. Exp. Immunol. 184 147–158

Xu X, Chen H, Zhang Q, Xu J, Shi Q and Wang M 2017 MiR-650inhibits proliferation, migration and invasion of rheumatoidarthritis synovial fibroblasts by targeting AKT2. Biomed.Pharmacother. 88 535–541

918 Fang-Yuan Yu et al.

Yang S and Yang Y 2015 Downregulation of microRNA221decreases migration and invasion in fibroblastlike synoviocytesin rheumatoid arthritis. Mol. Med. Rep. 12 2395–2401

Yu H, Littlewood T and Bennett M 2015 Akt isoforms in vasculardisease. Vasc. Pharmacol. 71 57–64

Yuan Y, Shen Y, Xue L and Fan H 2013 miR-140 suppresses tumorgrowth and metastasis of non-small cell lung cancer by targetinginsulin-like growth factor 1 receptor. PLoS One 8 e73604

Zhang M, Liu Q, Mi S, et al. 2011 Both miR-17-5p and miR-20aalleviate suppressive potential of myeloid-derived suppressorcells by modulating STAT3 expression. J. Immunol. 1864716–4724

Zheng F, Liao YJ, Cai MY, et al. 2012 The putative tumoursuppressor microRNA-124 modulates hepatocellular carcinomacell aggressiveness by repressing ROCK2 and EZH2. Gut 61278–289

Corresponding editor: ULLAS KOLTHUR-SEETHARAM

Inhibitory effect of miR-92a on rheumatoid arthritis 919