Inhibition of DNM1L and mitochondrial fission attenuates inflammatory response in fibroblast‐like synoviocytes of rheumatoid arthritis

Abstract Mitochondrial fission and fusion are important for mitochondrial function, and dynamin 1‐like protein (DNM1L) is a key regulator of mitochondrial fission. We investigated the effect of mitochondrial fission on mitochondrial function and inflammation in fibroblast‐like synoviocytes (FLSs) during rheumatoid arthritis (RA). DNM1L expression was determined in synovial tissues (STs) from RA and non‐RA patients. FLSs were isolated from STs and treated with a DNM1L inhibitor (mdivi‐1, mitochondrial division inhibitor 1) or transfected with DNM1L‐specific siRNA. Mitochondrial morphology, DNM1L expression, cell viability, mitochondrial membrane potential, reactive oxygen species (ROS), apoptosis, inflammatory cytokine expression and autophagy were examined. The impact of mdivi‐1 treatment on development and severity of collagen‐induced arthritis (CIA) was determined in mice. Up‐regulated DNM1L expression was associated with reduced mitochondrial length in STs from patients with RA and increased RA severity. Inhibition of DNM1L in FLSs triggered mitochondrial depolarization, mitochondrial elongation, decreased cell viability, production of ROS, IL‐8 and COX‐2, and increased apoptosis. DNM1L deficiency inhibited IL‐1β–mediated AKT/IKK activation, NF‐κBp65 nuclear translocation and LC3B‐related autophagy, but enhanced NFKBIA expression. Treatment of CIA mice with mdivi‐1 decreased disease severity by modulating inflammatory cytokine and ROS production. Our major results are that up‐regulated DNM1L and mitochondrial fission promoted survival, LC3B‐related autophagy and ROS production in FLSs, factors that lead to inflammation by regulating AKT/IKK/NFKBIA/NF‐κB signalling. Thus, inhibition of DNM1L may be a new strategy for treatment of RA.


| INTRODUC TI ON
Rheumatoid arthritis (RA) is a chronic autoimmune inflammatory disease and is characterized by progressive destruction of multiple joints and synovial. 1 Despite significant therapeutic advances, RA remains a huge threat for human health due to the high rate of patient disability. During the pathogenic process of RA, fibroblast-like synoviocytes (FLSs) in the intimal lining can produce high levels of pro-inflammatory cytokines and contribute to cartilage destruction. 2 FLSs have been considered therapeutic targets for RA interventions. However, how FLSs contribute to the pathogenesis of RA has not been clarified.
Hypoxia can alter mitochondrial morphology and lead to mitochondrial dysfunctions, with reduced levels of adenosine triphosphate (ATP) production, excessive reactive oxygen species (ROS) production, increasing mitochondrial DNA mutations and oxidative damages, which in turn enhance inflammation in the joint, contributing to the pathogenesis of RA. [3][4][5] Mitochondria are the main inducer and target of ROS. 6 A previous study has shown that excessive ROS induces mitophagy, 7 and feedback can remove damaged mitochondria to reduce ROS accumulation. 8 ROS can activate the NF-κB signalling to produce pro-inflammatory cytokines, promoting synovial inflammation. 9 It is notable that a defect in the mitochondrial respiratory chain (MRC) complex IV, and increased autophagy (mediated by LC3 and Beclin 1) are associated with the development of RA. 10,11 Inhibition of mitochondrial function can aggravate inflammatory responses in FLSs. 12 Inhibition of mitochondrial functions exacerbates inflammatory responses in the joint in vivo. 13 In addition, treatment with anti-tumour necrosis factor (TNF) α can suppress mitochondrial mutagenesis in inflammatory arthritis. 14 Mitochondrial membrane depolarization in peripheral blood lymphocytes has been reported in patients with RA. 15 Accordingly, protecting mitochondrial function may be an effective intervention for RA.
DNM1L is conservatively expressed in the cytoplasm of many types of cells and is crucial for mitochondrial fission. 24 Following its activation, DNM1L can be translocated into the mitochondrial outer membrane and bind to mitochondrial adaptor protein FIS1 to drive the process of mitochondrial fission. 24 Accumulated evidence has shown that excessive expression of DNM1L is associated with development of neurodegenerative diseases, diabetes, tumours, kidney diseases, sepsis and cardiac diseases. 17,[25][26][27][28][29][30] However, it is still unclear whether excessive expression of DNM1L-mediated mitochondrial fission is associated with the progression of RA.
In this study, DNM1L expression was examined in synovial tissues (STs) of RA and non-RA patients. The effects of DNM1L deficiency on RA FLSs and in mice with collagen-induced arthritis (CIA) were studied. The results indicated that up-regulated DNM1L expression in STs was associated with RA severity, and DNM1L deficiency altered mitochondrial morphology and reduced ROS production, inflammatory responses, and autophagy in FLSs by down-regulating the AKT/IKK/NFKBIA/NF-κB signalling. Therefore, DNM1L may be a new therapeutic target, and inhibition of mitochondrial fission may be an effective intervention for RA.

| Reagents
The specific reagents used in this study included antibodies against

| Tissue collection and cell culture
Synovial tissues (STs) were obtained from patients with RA, who underwent a joint replacement (hit or knee) surgery in the Department The partial fresh STs were washed with PBS and cut into small pieces. Subsequently, the tissue pieces were digested with 4 mg/mL of collagenase I (Invitrogen) at 37°C for 2 hours and filtered through a cell strainer (70 µm), followed by centrifugation. The obtained cells were cultured in DMEM/F12 supplemented with 10% fetal bovine serum (FBS), 100 U/mL of penicillin and 100 µg/mL of streptomycin. The cells at passages of 3-6 were used as FLSs for experiments.
All patients signed the written informed consent agreements before participation. The experimental protocols were approved by the Ethics Committee of Tangdu Hospital, the Second Affiliated Hospital of Air Force Medical University.

| Immunohistochemistry (IHC)
Another part of STs was fixed in 10% formalin overnight and paraffin-embedded. The ST sections were deparaffinized, rehydrated and treated with 3% H 2 O 2 in methanol for 10 minutes, followed by antigen retrieval in a microwave. After being blocked with 5% BSA (AR0004; Boster), the sections were probed with anti-DNM1L (1:300) at 4°C overnight, and the bound antibodies were detected with horseradish peroxidase (HRP)-conjugated goat anti-rabbit antibodies (1:1000, AS014; ABclonal), followed by visualization using diaminobenzidine. Images were obtained under a light microscope (Olympus).

| Transfection of siRNA
Fibroblast-like synoviocytes (1 × 10 5 cells/mL) were cultured in 6-well plates overnight and then transfected with 100 nmol/L DNM1Lspecific or control siRNA using lipofectamine 3000 (Invitrogen), according to the manufacturer's instruction. Three days later, DNM1L expression was determined by qRT-PCR and Western blot assays.

| Quantitative real-time polymerase chain reaction (qRT-PCR)
Fibroblast-like synoviocytes (1 × 10 5 cells/mL) were treated with vehicle control, 50 µmol/L mdivi-1 or siRNA for 72 hours. Total RNA was extracted from the different groups of cells or fresh STs using TRIzol Reagent (Invitrogen). After removing genomic DNA using the Genomic DNA Removal Kit (ABM), each RNA sample (2 µg) was reversely transcribed into cDNA. The relative levels of target mRNA transcripts were determined by qRT-PCR using the SYBR Green PCR Kit (ABM).
Specific primers (Table 1) were used with the Rotor-Gene Q RT-PCR instrument (Qiagen GmbH, Hilden, Germany). All mRNA levels are expressed relative to the control (GAPDH) using the 2 −ΔΔCt method. 34

| Transmission electron microscopy (TEM) and confocal microscopy
The different groups of FLSs and STs were fixed in glutaraldehyde at 4°C for 2 hours and post-fixed in 1% citric acid-0.1 mol/L phosphate buffer (PB) for 2 hours, followed by embedded. The ultra-thin sections (70 nm) were stained with uranyl acetate and lead citrate.
Images were collected using TEM (Hitachi, HT7700). In addition, the mitochondrial morphology of different groups of FLSs was examined by confocal microscopy (Zeiss 800) after staining with fluorescent MitoTracker Green, according to the manufacturer's instructions. In particular, cells (1 × 10 4 cells/ well) were cultured in a chamber for 24 hours, stained with MitoTracker Green at 37°C in the dark for 40 minutes, and then examined under a confocal microscope. The different groups of cells were also stained with JC-1 using the JC-1 assay kit (Beyotime, C2006), according to the manufacturer's instruction.
The fluorescent signals from this stain were also examined under a confocal microscope to determine mitochondrial membrane potential.

| Cell viability and apoptosis assays
The cell viability was measured using the CCK-8 assay kit. Briefly, The absorbance at 450 nm was measured using a microplate reader.
The apoptosis of different groups of cells was determined by flow cytometry using the Annexin V-FITC apoptosis Detection Kit.
The percentages of apoptotic cells were determined by flow cytometry (Epics XL-MCL, Beckman). In addition, following transfection with DNM1L-specific miRNA or control scramble siRNA for 72 hours, cells viability and apoptosis were determined.

| Reactive oxygen species (ROS) detection
The contents of cellular ROS were determined using ROS fluores-

| Immunofluorescence staining
The NF-κBp65 nuclear accumulation was examined by immunofluorescence. Briefly, FLSs (0.5 × 10 4 cells/well) were cultured in 24well culture plates and treated with vehicle or the DNM1L inhibitor for 24 hours. The cells were fixed with 4% paraformaldehyde and blocked with confining liquid (containing Triton X-100 and 5% BSA).
After being washed, the cells were probed with anti-NF-κBp65 (1: 50) at 4°C overnight, stained with Cy3-conjugated secondary antibody (1:500), and then, the nuclei were stained using DAPI. The fluorescent signals were observed in an inverted fluorescence microscope.

| Clinical and histological assessment of arthritis
The mice were monitored for arthritis signs every other day posttreatment. The severity of arthritis was scored from 0 to 4. 35 The paw thickness of individual mice was measured using a Vernier caliper. On day 50, the mice were killed, and their knee joints, liver and renal tissues were collected and fixed in 10% formalin. The tissue sections (4 µm) were stained with haematoxylin and eosin (H&E) or toluidine blue. In addition, their STs were used for analysing the expression of cytokine mRNAs. The ROS contents in crystal synovial membrane sections were determined by DHE staining. 13

| Statistical analysis
Data are presented as the means ± SD of three or more separate experiments. The difference between two groups was analysed by Student's t test. Correlation was determined from Pearson's correlation coefficient. All statistical analyses were performed using the

| Enhanced mitochondrial fission in STs of RA patients is correlated with disease severity
To investigate the mitochondrial dynamics in STs during the pathogenic process of RA, the morphologic changes of mitochondria were investigated in STs from RA and non-RA patients and examined their

| DNM1L deficiency alters mitochondria morphology and mitochondrial membrane potential in FLSs
DNM1L is crucial for mitochondrial fission. Next, we examined how DNM1L deficiency could affect mitochondrial morphology and membrane potential in FLSs. In comparison with the control cells transfected with scramble siRNA, transfection with DNM1L-specific siRNA significantly decreased DNM1L protein expression by about 30% and DNM1L mRNA transcripts by 55% in FLSs (Figure 2A,B). We also measured GTPase activity of DNM1L after mdivi-1 treatment. The results indicate that mdivi-1 inhibited the GTPase activity of DNM1L in a dose-dependent manner ( Figure 2C). Following staining with MitoTracker Green, treatment of mdivi-1 (50 µmol/L) or DNM1L silencing by transfection with DNM1L-specific siRNA significantly increased the length of mitochondrial in FLSs ( Figure 2D). Following JC-1 staining, treatment of mdivi-1 or DNM1L silencing reduced the ratios of red to green fluorescent signals, a hallmark of depolarization in FLSs ( Figure 2E). Collectively, such data indicated that DNM1L deficiency altered mitochondrial morphology and induced mitochondrial membrane depolarization in FLSs. In addition, treatment with 50 µmol/L mdivi-1 or DNM1L silencing significantly increased the percentages of apoptotic FLSs ( Figure 3D).

| DNM1L deficiency in FLSs reduces their viability and production of pro-inflammatory cytokines and increases apoptosis
Thus, DNM1L deficiency reduced the viability of FLSs and their production of pro-inflammatory cytokines by triggering apoptosis.

| DNM1L deficiency reduces ROS production and LC3B-related autophagy in FLSs
To investigate the impact of DNM1L deficiency on ROS production, FLSs were treated with mdivi-1 (50 µmol/L) or transfected with AKT expression was less than that of IL-1β-induced AKT activation in FLSs ( Figure 4B). Moreover, treatment with mdivi-1 or DNM1L silencing significantly decreased the ratio of LC3B-II to LC3B-I and the IL-1βincreased ratios of LC3B-II to LC3B-I in FLSs ( Figure 4C). The results suggest that ROS and LC3B-related autophagy may play a key role in DNM1L-mediated proliferation, inflammation and apoptosis in RA FLS.

| DNM1L deficiency attenuates the IL-1βinduced NF-κB activation in FLSs
To further explore the role of DNM1L in RA FLS inflammation, we examined the effect of DNM1L deficiency on the IL-1β-induced AKT and IKK activation and NFKBIA expression in FLSs. We found that treatment with mdivi-1 or DNM1L silencing mitigated the IL-1β-induced AKT and IKK activation, but enhanced the IL-1β-up-regulated NFKBIA expression in FLSs ( Figure 5A). Further immunofluorescence studies revealed that treatment with mdivi-1 attenuated the IL-1β-induced NF-κBp65 nuclear translocation in FLSs ( Figure 5B). Together, such data indicated that DNM1L deficiency attenuated the IL-1β-induced NF-κB activation in FLSs by down-regulating AKT/IKK/NFKBIA signalling.  Figure 6E). While control CIA mice displayed joint damages, treatment with mdivi-1 preserved the intact joint architecture ( Figure 6F). In addition, treatment with mdivi-1 also decreased ROS production in the STs of mice ( Figure 6H). However, histological examination revealed that treatment with mdivi-1 did not change the morphology of the liver and kidney tissues in mice (Figure5G). Therefore, treatment with mdivi-1 to impaired DNM1L activity and ameliorated the severity of CIA in mice by modulating the pro-inflammatory and anti-inflammatory responses, but had no effect on non-target tissues (liver and kidney).

| D ISCUSS I ON
In the present study, we found that up-regulated DNM1L expression in STs from patients with RA was associated with the severity of RA. Such findings support the notion that up-regulated mitochondrial fission is associated with the development of different types of diseases, including inflammatory diseases. 4,17,[25][26][27][28][29][36][37][38][39][40] To the best of our knowl-  12 We also found that the ratio of DNM1L-to-MFN1 transcripts correlated with the level of serum anti-CCP antibodies, DAS28 and ESR values in patients with RA. Such data suggest that DNM1L-related mitochondrial fission may contribute to the pathogenesis of RA.
Mdivi-1 inhibits DNM1L GTPase activity and is a potent inhibitor of mitochondrial fission. Previous studies demonstrated that mdivi-1 can inhibit myocardial infarction, stroke, neurodegenerative diseases, and cancers. 22,25,26,36,[43][44][45][46] In the present study, we found that mdivi-1 inhibited DNM1L GTPase activity in a dose-dependent manner, consistent with previous studies. 43,47 Furthermore, treatment with mdivi-1 significantly reduced the severity of CIA in mice by decreasing pro-inflammatory cytokine production, but increasing anti-inflammatory IL-10 production. These, together with its inhibition on ROS and pro-inflammatory COX-2 and IL-8 production in FLSs, suggest that inhibition of mitochondrial fission may modulate the balance of pro-inflammatory and anti-inflammatory responses in joint lesions. Given that mitochondria are crucial for nutritional metabolisms, it is possible that inhibition of mitochondrial fission may differentially modulate glucose and lipid metabolisms to affect pathogenic and regulatory T-cell responses during the pathogenic process of RA.
It is well known that mitochondrial dysfunction can induce excessive ROS production, which promotes autophagy 17 and exacerbates synovial inflammation by activating the NF-κB signalling to promote the expression of pro-inflammatory cytokines. 12,48 Furthermore, mitochondrial dysfunction can activate PI3K/AKT signalling, which is crucial for inflammation in RA FLSs. 17,49 In addition, the activated AKT can induce IKK phosphorylation and NF-κB, which is down-regulated by NFKBIA. 50,51 In this study, we found that treatment with mdivi-1 or DNM1L silencing effectively decreased ROS production and LC3B-related autophagy in FLSs, extending the previous observation in osteoblasts. 42 Moreover, DNM1L deficiency also mitigated the IL-1β-induced AKT and IKK phosphorylation and NF-κBp65 nuclear translocation, but increased NFKBIA expression in RA FLS.
Therefore, anti-oxidants and inhibition of mitochondrial fission to attenuate the AKT/IKK/NF-κB signalling may be promising strategies for treatment of RA.
This study had limitations, including a small sample size and the lack of studies on how inhibition of mitochondrial fission alters mitochondrial DNA and glucose and lipid metabolisms that affect pathogenic and regulatory T-cell responses. Hence, future investigations in a bigger population are warranted to determine molecular mechanisms by which the mitochondrial dynamic factors regulate the pathogenesis of RA.
In conclusion, our results indicated that up-regulated DNM1L expression in STs was associated with the severity of RA.
Treatment with mdivi-1 or DNM1L silencing altered mitochondrial morphology, promoted mitochondrial membrane depolarization and apoptosis, and decreased cell viability, ROS, LC3B-related autophagy and pro-inflammatory cytokine production in FLSs.
The same treatment also mitigated the IL-1β-mediated LC3B-

CO N FLI C T O F I NTE R E S T
The authors declare no competing interests.

DATA AVA I L A B I L I T Y S TAT E M E N T
All data generated or analysed during this study are included in this article.