MiR‐155 promotes interleukin‐1β‐induced chondrocyte apoptosis and catabolic activity by targeting PIK3R1‐mediated PI3K/Akt pathway

Abstract Osteoarthritis (OA) is a common joint disease characterized by progressive cartilage degradation, in which elevated chondrocyte apoptosis and catabolic activity play an important role. MicroRNA‐155 (miR‐155) has recently been shown to regulate apoptosis and catabolic activity in some pathological circumstances, yet, whether and how miR‐155 is associated with OA pathology remain unexplored. We report here that miR‐155 level is significantly up‐regulated in human OA cartilage biopsies and also in primary chondrocytes stimulated by interleukin‐1β (IL‐1β), a pivotal pro‐catabolic factor promoting cartilage degradation. Moreover, miR‐155 inhibition attenuates and its overexpression promotes IL‐1β‐induced apoptosis and catabolic activity in chondrocytes in vitro. We also demonstrate that the PIK3R1 (p85α regulatory subunit of phosphoinositide 3‐kinase (PI3K)) is a target of miR‐155 in chondrocytes, and more importantly, PIK3R1 restoration abrogates miR‐155 effects on chondrocyte apoptosis and catabolic activity. Mechanistically, PIK3R1 positively regulates the transduction of PI3K/Akt pathway, and a specific Akt inhibitor reverses miR‐155 effects on promoting chondrocyte apoptosis and catabolic activity, phenocopying the results obtained via PIK3R1 knockdown, hence establishing that miR‐155 promotes chondrocyte apoptosis and catabolic activity through targeting PIK3R1‐mediated PI3K/Akt pathway activation. Altogether, our study discovers novel roles and mechanisms of miR‐155 in regulating chondrocyte apoptosis and catabolic activity, providing an implication for therapeutically intervening cartilage degradation and OA progression.


| INTRODUC TI ON
Osteoarthritis (OA) is the most common chronic whole-joint disease featured by the progressive destruction of articular cartilage. 1 Other structural alterations including those occurring in subchondral bone, ligaments and synovial membrane also frequently appear in OA. 2 OA causes pain, deformity and reduction of motion 3 and represents a leading cause of physical disability and also imposes a great socioeconomic burden globally, inflicting approximate 10% of the elderly population aged over 60 years. 4,5 The evidence-based guidelines for OA treatment, such as prescribing capsaicin, cane and oral non-steroidal anti-inflammatory drugs, have been recommended by the OA Research Society International (OARSI), 6 but, and no effective therapy is available to actually modify disease progression. 7 This dilemma could be largely attributed to our limited understanding of the molecular mechanisms that underlie OA pathogenesis, which dismays the development of effective treatment approaches. 8 OA pathogenesis is very complex and involves mechanical, inflammatory and metabolic factors that ultimately destruct the synovial joint. 2 Although the augmented catabolic activity and the apoptosis of chondrocytes, the sole cell type residing this anatomic compartment to fulfil the cartilage development and cartilaginous extracellular matrix (ECM) maintenance, have been shown to play an important role in promoting articular cartilage degradation and OA pathogenesis, 9,10 much unidentified factors and signalling pathways regulating these pathogenic processes need to be unveiled, which holds potential significance to offer therapeutic targets for developing disease-modifying therapies.
Mechanistically, in OA cartilage, the increased production of catabolic factors in chondrocytes, such as the matrix-degrading enzymes, particularly the matrix metalloproteinase 3 (MMP3) and MMP13, is crucial for accelerating the destruction of core ECM molecules, including the collagen and aggrecan, leading to their functional loss that is closely related to OA progression. 11,12 Aside from this, the chondrocyte cell death through apoptosis also contributes to cartilage ECM degradation. 13 Various extracellular catabolic factors, like the pro-inflammatory cytokine interleukin (IL)-1β, can provoke the degradation of ECM molecules and apoptosis in chondrocytes, 14 and a growing body of evidence has shown that some microRNAs (miRNAs), a class of small and non-coding RNAs that post-transcriptionally suppress gene expression by targeting mRNAs, 15 play a role in regulating these pathologic processes. 16,17 The epigenetic roles of miRNAs in OA pathogenesis have also been recognized recently, 18,19 shedding new light on the association between miRNAs with OA pathogenesis, and also evokes an interest to engage in further relevant studies.
MiR-155 is a multifunctional miRNA playing key roles in numerous physiological and pathological processes, such as inflammatory disease and cancer. 20,21 In recent years, miR-155 has been reported to regulate apoptosis in various cell types, including immunologic and cardiac cells. 22,23 Moreover, miR-155 also affects catabolic activity in nucleus pulposus. 24,25 In the current study, we aimed to investigate whether miR-155 regulates chondrocyte apoptosis and catabolic activity. We demonstrate that miR-155 promotes IL-1β-induced chondrocyte apoptosis and catabolic activity through modulating the PI3K/Akt pathway by means of targeting and suppressing the expression of PIK3R1.

| OA cartilage collection
OA cartilage tissues tested in this study were collected from OA patients who underwent total knee joint replacement surgery in our hospital (n = 18). Normal cartilage tissues from 8 patients without OA who underwent the amputation were used as negative controls (n = 8), due to our limited accessibility to clinical resources. There is no statistical significance in clinical characteristics of age and gender between OA and control patients. The ethical approval was obtained from the Institutional Ethics Committee of The Second Hospital of Hebei Medical University, which conforms to the principles of the Declaration of Helsinki. The written informed consent was also obtained from each patient prior to study. After collection, OA cartilage tissues were examined using the Safranin O-fast green staining to evaluate the cartilage deterioration, and the severity of histological changes was graded according to a modified Mankin scale. 26 The control tissues were also examined histologically and confirmed without early pathologic OA changes.

| Human chondrocyte isolation and culture
The isolation and following culture of human primary chondrocytes were performed as described previously. 27 Briefly, normal human articular cartilage tissues were processed in accordance with the approved guidelines and then minced and digested using 0.2% collagenase II (Sigma) prepared in Dulbecco's modified Eagle's medium (DMEM, Invitrogen: Carlsbad, CA, USA). After filtration and centrifugation and washing with sterile phosphate-buffered saline (PBS), the extracted chondrocytes were maintained in DMEM containing 10% foetal bovine serum (FBS, Invitrogen) at 37°C with 5% CO 2 in a humidified incubator. Medium was replaced every two days. The first-passage chondrocytes were obtained after two weeks. All experiments performed with primary chondrocytes were conducted within one week thereafter.

| Cell treatment and transfection
Primary chondrocytes were cultured with the complete DMEM medium containing 0, 2, 5 or 10 µg/mL human recombinant IL-1β

| Quantitative real-time PCR (qRT-PCR) analysis
Total RNA from cartilage tissues and primary chondrocytes was ex-

| Western blot analysis
Cartilage tissues and primary chondrocytes were lysed to obtain protein extracts, which were subjected to SDS-PAGE (8%-12% gel) and

| TdT mediated dUTP nick end labelling (TUNEL) assay
The apoptosis of chondrocytes was detected in situ through using a TUNEL assay according to the manufacturer's instructions (Roche).
Briefly, primary chondrocytes plated on cover slides in a 6-well plate were washed with PBS, fixed with 4% paraformaldehyde for 20 min and blocked with 5% bovine serum albumin (BSA) for 1 hour. Then, cover slides were immersed with TUNEL reaction mixture for 1 hour at 37°C and covered with fluorescence mounting medium (Zhongshan Golden Bridge Biotechnology, Beijing, China) in the darkness. Apoptotic cells were visualized under a microscope (LSM 510; Zeiss: Jena, Germany).
Fifteen random fields in each group were analysed to calculate the percentage of TUENL-positive cells (apoptotic).

| Luciferase reporter assay
The wild-type (wt) or mutant (mut) 3′-UTR of human PIK3R1 containing potential binding sites for miR-155 as predicted by the Each treatment was performed in 5 replicates.

| Statistical analysis
All data were presented as the means ± SEM from at least three independent assays. The statistical significance was calculated using the unpaired Student's t test or analysis of variance (ANOVA) test with GraphPad Prism 5 software. P < 0.05 was considered to be statistically significant.

| MiR-155 level is elevated in OA cartilage and IL-1β-treated chondrocytes
The OA articular cartilage tissues manifest signs of chondrocyte apoptosis, up-regulation of matrix-degrading enzymes and enhanced degradation of extracellular matrix (ECM). 10,29 Consistently, we also observed that compared with normal counterparts (n = 8), the expression levels of cleaved caspase-3, a typical apoptosis marker, as well as the matrix metalloproteinase 3 (MMP3) and MMP13, were all markedly increased, and conversely, those of collagen Ⅱ and aggrecan were decreased in the articular cartilage tissues collected from the knee joint of OA patients (n = 18) ( Figure 1A). To seek whether there exists a possible connection between miR-155 and OA pathology, we initially checked whether miR-155 expression is altered in OA articular cartilage tissues by applying the quantitative real-time PCR (qRT-PCR) analysis. As a result, we found that miR-155 level was significantly up-regulated in OA cartilage tissues in contrast with normal control samples ( Figure 1B, P < 0.01). More pathologically relevant, miR-155 level was positively correlated well with the severity of OA ( Figure 1C), as graded by a modified Mankin scale, 26 further implicating that miR-155 may be involved in OA progression. The exposure of chondrocytes to a pro-inflammatory cytokine interleukin (IL)-1β can induce pathological phenotypes akin to those emerging during OA progression, such as increased apoptosis and catabolic activities. 30,31 These pathologic changes were reproducible in primary chondrocytes stimulated with IL-1β in our study, as detected by the expression of corresponding molecular markers ( Figure 1D).
Noticeably, upon IL-1β treatment, the expression level of miR-155 in primary chondrocytes was drastically induced ( Figure 1E) in a dosedependent manner, which is reminiscent of the result observed in OA cartilage tissues ( Figure 1B). Together, these in vivo and in vitro data suggest that miR-155 is associated with OA pathogenesis.

| MiR-155 functions to exaggerate IL-1β-induced chondrocyte apoptosis and catabolic activity in vitro
Next, to interrogate the potential roles of miR-155, we carried out lossof-function studies and examined the effects on chondrocytes treated with IL-1β. The inhibition of miR-155 was accomplished via transfection of synthetic sequence-specific oligonucleotides targeting miR-155 (miR-155 inhibitors). 28 Consistent with Figure 1D, IL-1β treatment caused obvious apoptosis and increased catabolic activity in primary

| MiR-155 targets PIK3R1 in chondrocytes
MiRNAs perform versatile biological functions through suppressing their mRNA targets. 32 To elucidate the underlying mechanism by which miR-155 regulates chondrocyte apoptosis and catabolic activity, we sought its potential targets via an in silico algorithm (http:// www.targe tscan.org) ( Figure 3A). Among these putative targets, the PIK3R1 (p85α regulatory subunit of phosphoinositide 3-kinase (PI3K)) has been reported to impact chondrocyte survival and matrix synthesis. 33 Hence, we asked whether PIK3R1 is a downstream target that mediates miR-155 function. We first utilized luciferase reporter assay to confirm whether PIK3R1 is a bona fide target of miR-155 in chondrocytes. As shown, MiR-155 inhibition increased ( Figure 3B Figure 5D,E) and TUNEL assay ( Figure 5F), photocopying the results obtained by PIK3R1 knockdown (Figure 4C,D).
Overall, this study demonstrates that the PI3K/Akt pathway downstream of miR-155/PIK3R1 axis dictates the cell fate of chondrocytes insulted with IL-1β exposure (Figure 6), providing a mechanistic insight to understand miR-155 roles in exacerbating IL-1β-induced chondrocyte apoptosis and catabolic activity.

| D ISSCUSS I ON
It is widely recognized that chondrocyte apoptosis and dysregulated catabolism, two key features in OA cartilage, contribute to disruption of cartilage homeostasis and ECM maintenance, leading to progressive tissue degeneration that underlies OA development and progression. 9,39 Therefore, preventing chondrocyte apoptosis and inhibiting catabolic activity appear to be promising therapeutic approaches for interfering OA deterioration. 40,41 In the current study, we reveal that miR-155 acts as a positive regulator of apoptosis and catabolic activity in chondrocytes exposed to IL-1β and also demonstrate that the targeted PIK3R1-mediated PI3K/Akt pathway is the critical underlying molecular mechanism. Hence, according to these novel discoveries, we not only highlight the significant role of PI3K/ Akt pathway in modulating chondrocyte apoptosis and catabolic activity, but also offer miR-155 as a potential new molecular target for reducing chondrocyte apoptosis and cartilage catabolism in future OA therapy.
The features of increased apoptosis and catabolism in OA cartilage tissues were confirmed in clinical samples enrolled in our study. Meanwhile, miR-155 level was also increased in these OA cartilage tissues and correlated positively with disease severity.
These observations suggest that miR-155 level is very possibly F I G U R E 4 The targeted PIK3R1 mediates miR-155 promotive effects on chondrocyte apoptosis and catabolic activity. A and B, Primary chondrocytes were transfected with NC inhibitors (NC inh) or miR-155 inhibitors (miR-155 inh) (A), or NC mimics or miR-155 mimics (B), and cultured for further 24 hours in the presence or absence of 10 µg/mL IL-1β. PIK3R1 expression was detected by Western blot analysis. Images represent three independent assays. C and D, Primary chondrocytes were transfected with NC inhibitors or miR-155 inhibitors along with siRNA targeting negative control (siNC) or PIK3R1 (siPIK3R1) as indicated and cultured with 10 µg/mL IL-1β for further 24 hours. The expression of PIK3R1, cleaved caspase-3, collagen Ⅱ, aggrecan, MMP3 and MMP13 was analysed by Western blot assay. GAPDH is the loading control. C, Images represent three independent assays. D, The relative protein expression was analysed by ImageJ software. Means ± SEM ANOVA test (n = 3), **P < 0.01. NS, not significant. E and F, Primary chondrocytes were transfected with NC mimics or miR-155 mimics along with pcDNA vector or pcDNA-PIK3R1 as indicated and cultured with 10 µg/mL IL-1β for further 24 hours. The analysis of protein expression was performed as in ( The most well-established roles of miR-155 are its oncogenic activities documented in several types of cancers. 48 Recently, miR-155 has been shown to suppress autophagy in chondrocytes through inhibiting some key autophagy components, such as ULK1, ATG14 and ATG5. 49 Autophagy is considered as a protective mechanism against OA development, 50 and its activation reduces OA severity in experimental models. 51 Presumably, miR-155 may have adverse effects, if any, on OA pathology when autophagy is inhibited. In our study, we uncover a new function of miR-155 in promoting apoptosis and catabolic activity in chondrocytes under a pathologic condition, ie IL-1β exposure. Perhaps, we provide another clue indicating a detrimental role of miR-155 in OA. Nevertheless, direct in vivo evidence is F I G U R E 5 MiR-155 promotes chondrocyte apoptosis and catabolic activity through targeting PIK3R1-mediated PI3K/Akt pathway. A, Primary chondrocytes were transfected with NC inhibitors (NC inh) or miR-155 inhibitors (miR-155 inh) along with siNC or siPIK3R1 as indicated and cultured with 10 µg/mL IL-1β for further 24 hours. The expression of PIK3R1, p-Akt and Akt was analysed by Western blot assay. GAPDH is the loading control. B, Primary chondrocytes were transfected with NC mimics or miR-155 mimics along with pcDNA vector or pcDNA-PIK3R1 as indicated and cultured with 10 µg/mL IL-1β for further 24 hours. The protein expression was analysed as in (A). C and E, Primary chondrocytes were transfected with NC inh or miR-155 inh along with DMSO or 1 µM Akt inhibitor MK-2206 as indicated and cultured with 10 µg/mL IL-1β for further 24 hours. C, The protein expression of PIK3R1, p-Akt and Akt was analysed as in (A). (D-E) The protein expression of cleaved caspase-3, collagen Ⅱ, aggrecan, MMP3 and MMP13 was also analysed. The representative images (D) were shown, and the relative protein expression (E) was analysed by ImageJ software. Means ± SEM ANOVA test (n = 3), **P < 0.01. NS, not significant. F, Primary chondrocytes were treated as in (C). Cell apoptosis was detected by TUNEL staining. The representative images (left, ᵡ 200 magnification) and quantification of TUNEL positive cells (right) are shown. Means ± SEM ANOVA test (n = 15), **P < 0.01. NS, not significant lacking and experiments performed with OA animal models are warranted in future. For instance, it would be intriguing to test whether miR-155 inhibition in cartilage tissues is able to alleviate chondrocyte apoptosis and reduce cartilage degradation in animals with experimental OA. It also should be noted that we only conducted in vitro assays, and more elaborated techniques for detecting apoptosis and animal experiments would be necessary to strengthen our findings.
Addressing this issue is helpful to advance our understanding of the physiological significance of miR-155 in OA pathogenesis.
Previous studies have reported that PIK3R1 is a target of miR-155 in diffuse large B cell lymphoma 52 and breast cancer. 53 In accordance with these, we also identified PIK3R1 as a direct target and suppressed by miR-155 in chondrocytes. This regulation of PIK3R1 may exist in human cartilage tissues as well, as contrary to miR-155 up-regulation, PIK3R1 expression exhibits decrease in OA cartilage, which implies that aside from miR-155, PIK3R1 may also be involved in OA pathogenesis. This is very possible, because PIK3R1 gene loci were identified to be associated with hip OA. 54,55 PIK3R1 functions as a positive regulatory subunit for PI3K. 56 The PI3K/Akt pathway promotes chondrocyte survival and ECM synthesis, and its activation ameliorates cartilage damage. 33 We found that both PIK3R1 knockdown and pharmaceutical inhibition of PI3K/Akt pathway activation could rescue miR-155 effects on IL-1βinduced chondrocyte apoptosis and catabolic activity, thus establishing that the targeted PIK3R1-mediated PI3K/Akt pathway by miR-155 is the downstream signalling that modulates chondrocyte survival and catabolism in response to IL-1β insults.

| CON CLUS ION
Taken together, our study establishes an unanticipated role and mechanism of miR-155 in chondrocyte apoptosis and catabolism with in vitro investigations, which may provide a molecular basis for targeting the miR-155/PIK3R1 axis and PI3K/Akt pathway in intervening OA progression. Relevant in vivo studies performed with OA animal models are required to collaborate our findings, which would provide more direct and physiological connection between miR-155 and OA pathology.

ACK N OWLED G EM ENT
The study was supported by grants from Hebei Provincial Health and Family Planning Commission (20130192).

CO N FLI C T O F I NTE R E S T
None declared.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.

Hua Zhang
https://orcid.org/0000-0002-6714-0847 F I G U R E 6 A schematic diagram of the mechanism underlies miR-155 function in chondrocyte apoptosis and catabolic activity. In chondrocytes, the PI3K/Akt pathway normally functions to suppress apoptosis and catabolism once upon stimulation of IL-1β through signalling transduction and indirect transcriptional control of collagen Ⅱ, aggrecan, MMP3 and MMP13. However, the IL-1β secreted in the surroundings binds to its cognate receptor IL-1R on the membrane of chondrocytes, leading to up-regulation of miR-155 via an unknown mechanism, which in turn inhibits PI3K/Akt pathway through targeting PIK3R1, one regulatory subunit of PI3K complex, and thus exacerbating IL-1β adverse effects to induce chondrocyte apoptosis and catabolism