MicroRNA-665-3p exacerbates nonalcoholic fatty liver disease in mice

ABSTRACT Oxidative stress and chronic inflammation are major culprits of nonalcoholic fatty liver disease (NAFLD). MicroRNA-665-3p (miR-665-3p) is implicated in regulating inflammation and oxidative stress; however, its role and molecular basis in NAFLD remain elusive. Herein, we measured a significant upregulation of miR-665-3p level in the liver and primary hepatocytes upon high fat diet (HFD) or 0.5 mmol/L palmitic acid plus 1.0 mmol/L oleic acid stimulation, and the elevated miR-665-3p expression aggravated oxidative stress, inflammation and NAFLD progression in mice. In contrast, miR-665-3p inhibition by the miR-665-3p antagomir significantly prevented HFD-induced oxidative stress, inflammation and hepatic dysfunction in vivo. Manipulation of miR-665-3p in primary hepatocytes also caused similar phenotypic alterations in vitro. Mechanistically, we demonstrated that miR-665-3p directly bound to the 3ʹ-untranslated region of fibronectin type III domain-containing 5 (FNDC5) to downregulate its expression and inactivated the downstream AMP-activated protein kinase alpha (AMPKα) pathway, thereby facilitating oxidative stress, inflammation and NAFLD progression. Our findings identify miR-665-3p as an endogenous positive regulator of NAFLD via inactivating FNDC5/AMPKα pathway, and inhibiting miR-665-3p may provide novel therapeutic strategies to treat NAFLD.


Introduction
Nonalcoholic fatty liver disease (NAFLD) has becoming one of the most common chronic liver diseases worldwide due to the increased prevalence of metabolic syndrome, and causes great personal suffering and socioeconomic burden [1][2][3][4]. Emerging studies have identified oxidative stress and chronic inflammation as major culprits of NAFLD, and inhibiting reactive oxygen species (ROS) overproduction and hepatic inflammation effectively ameliorates hepatocyte injury, fibrosis and hepatic dysfunction [5][6][7][8][9]. AMP-activated protein kinase alpha (AMPKα) is a central regulator of energy metabolism and serves as a promising therapeutic target to treat metabolic diseases [10][11][12]. In addition, AMPKα is also involved in controlling redox homeostasis and inflammatory response. It has been reported that AMPKα activation directly induces the expression of various antioxidant enzymes and reduces the generation of free radicals [13][14][15]. And AMPKα activation also suppresses the phosphorylation and nuclear translocation of nuclear factor kappa B (NF-κB), thereby preventing the transcription of downstream proinflammatory cytokines, including interleukin-1 beta (IL-1β), IL-6, monocyte chemotactic protein-1 (MCP-1) and tumor necrosis factor-alpha (TNF-α) [16]. As expected, Garcia et al demonstrated that hepatic steatosis and inflammation in high fat diet (HFD)stimulated mice were dramatically attenuated by genetic liver-specific AMPKα activation [17]. Therefore, it is reasonable to treat NAFLD via activating AMPKα.
The discovery of endogenous noncoding RNAs helps to establish a new field to investigate the mechanism of NAFLD progression. MicroRNAs (miRNAs) are a class of single-stranded, small noncoding RNAs that negatively regulate gene expression at the posttranscriptional level via binding to the 3ʹ-untranslated region (UTR) of target mRNAs [18][19][20][21][22]. Notably, the expressions of various miRNAs have been found to be dysregulated in NAFLD patients [23]. Recent studies have identified a critical role of miR-665-3p in regulating inflammation and oxidative stress. Li et al found that inhibition of miR-665-3p significantly suppressed systemic inflammation in intestinal ischemia/reperfusion (I/R) injury, and Guo et al also demonstrated an inhibitory role of the miR-665-3p antagomir on inflammation in fusarium solani-associated keratitis [24,25]. In contrast, Zhang et al showed that miR-665-3p overexpression blocked the activation of NF-κB signaling in microglial cells upon oxygen-glucose deprivation stimulation [26]. In addition, miR-665-3p expression was found to be upregulated in rat failing hearts, and miR-665-3p inhibition significantly reduced cardiac oxidative stress and inflammation [27]. And miR-665-3p downregulation by dexmedetomidine conferred protective effects on oxidative stress and I/R-induced cardiac injury [28]. Based on these findings, we herein aim to investigate the role of miR-665-3p in oxidative stress, inflammation and NAFLD progression by in vivo and in vitro studies, and elucidate whether these effects are mediated by AMPKα.

Animals and treatments
C57BL/6 male mice aged 8-10 weeks were fed with a HFD (60% kcal fat, 20% kcal carbohydrates and 20% kcal protein) for 24 weeks to establish NAFLD model as previously described, and the matched control mice were fed with a normal diet (ND, 10% kcal fat, 70% kcal carbohydrates and 20% kcal protein) for 24 weeks [29]. To manipulate hepatic miR-665-3p levels, mice were intraperitoneally injected with the miR-665-3p agomir, miR-665-3p antagomir or matched negative controls at a dose of 100 mg/kg weekly for 6 consecutive weeks before the mice sacrificed. To inhibit AMPKα, mice were intraperitoneally injected with CpC (20 mg/kg) once two days for 8 consecutive weeks before the mice sacrificed [30]. To knock down hepatic FNDC5, mice received a single intravenous injection of shFNDC5 from the tail vein at a concentration of 1 × 10 11 viral genome per mouse for 4 weeks before the miR-665-3p antagomir treatment [31,32]. All animal procedures were approved by our institution's ethical guidelines and also in compliance with the Animal Research: Reporting of In Vivo Experiments guidelines.

Western blot
Total proteins were extracted from the livers or primary hepatocytes using RIPA lysis buffer, and the concentrations were quantified using Pierce™ BCA Protein Assay kit [33][34][35][36]. Then, 20 μg total proteins were separated by 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis, transferred onto polyvinylidene fluoride membranes and incubated with the primary antibodies at 4°C overnight after being blocked with 5% skimmed milk at room temperature for 1 h. On the next day, the membranes were incubated with the horseradish peroxidase-conjugated secondary antibodies at room temperature for an additional 1 h, and then visualized by LumiGLO chemiluminescent substrate. β-actin was used as the house-keeping gene and the data were analyzed using an Image Lab software.

Histologic analysis
The livers were quickly excised and fixed in 4% (w/v) paraformaldehyde, embedded in paraffin and sectioned to 5 μm slices. Next, paraffinembedded sections were exposed to dewaxing, hydratization and hematoxylin and eosin (HE) staining by standard procedures. Oil red O staining was performed on frozen tissue sections to measure lipid droplet accumulation in the liver.

Biochemical analysis
Fasting blood glucose (FBG) levels were detected using a Life Scan One Touch Ultra Easy glucometer (Wayne, PA, USA), and serum insulin levels were detected using an insulin mouse ELISA kit according to the manufacturer's instructions. And the homeostastic model assessmentinsulin resistance (HOMA-IR) index was calculated according to the following formula: FBG × fasting serum insulin/22.5 [41]. Blood-biochemical indicators, including serum TG, TC, alanine transaminase (ALT) and aspartate transaminase (AST), were measured using a Beckman automatic biochemistry analyzer (Palo Alto, CA, USA). The levels of TG, TC and NEFA in the liver or primary hepatocytes were detected using commercial kits according to the manufacturer's instructions. Hepatic hydroxyproline levels were determined to evaluate hepatic fibrosis using a commercial kit. Briefly, fresh liver homogenates were hydrolyzed in 12 mol/L hydrochloric acid at 120°C for 3 h, incubated with chloramine T and 4-(dimethylamino) benzaldehyde and then subjected to absorbance detection at 560 nm. The levels of inflammatory cytokines in the liver or cell culture were measured using commercial ELISA kits according to the manufacturer's instructions. LDH releases in the medium and hepatic irisin levels were measured using commercial kits following the manufacturer's instructions.

Primary hepatocytes isolation and treatments
Primary hepatocytes were isolated as previously described [29]. Briefly, the livers were harvested from 8-week-old C57BL/6 mice and perfused with collagenase buffer to isolate the hepatocytes. Next, the cell suspensions were centrifuged at 50 g for 2 min and purified with a 50% Percoll solution. To imitate NAFLD in vitro, primary hepatocytes were stimulated with 0.5 mmol/L palmitic acid plus 1.0 mmol/L oleic acid (PO) for 24 h, while cells in the control group were incubated with an equal amount of fatty acid free bovine serum albumin (BSA) [29]. To manipulate miR-665-3p levels in hepatocytes, cells were transfected with the miR-665-3p agomir, miR-665-3p antagomir or matched negative controls at a dose of 50 nmol/L using a Lipofectamine 6000 TM reagent for 24 h before PO stimulation. To inhibit AMPKα, cells were treated with CpC (10 μmol/L) for 24 h along with PO or BSA stimulation [42].

Analysis of oxidative stress
Intracellular ROS level was measured using a DCFH-DA probe according to previous studies [43][44][45]. Briefly, the livers or hepatocytes were lysed, incubated with 50 μmol/L DCFH-DA at 37°C for 1 h, and then the absorbance was measured at the excitation/emission wavelength of 485/535 nm. Hepatic H 2 O 2 and MDA levels were detected using commercial kits according to the manufacturer's instructions.

Statistical analysis
All results were expressed as the means ± standard deviations, and statistical analyses were performed using a SPSS software (Version 19.0). Differences between 2 groups were compared using an unpaired two-tailed Student's t-test, while differences among 3 or more groups were compared using the one-way analysis of variance followed by Tukey post-hoc test. P-values less than 0.05 were considered statistically significant.

Results
In the present study, we aim to investigate the role of miR-665-3p in oxidative stress, inflammation and NAFLD progression by in vivo and in vitro studies, and elucidate whether these effects are mediated by AMPKα. A significant upregulation of miR-665-3p level was measured in the liver and primary hepatocytes upon HFD or PO stimulation, and the elevated miR-665-3p expression aggravated oxidative stress, inflammation and NAFLD progression in mice. In contrast, miR-665-3p inhibition by the miR-665-3p antagomir significantly prevented HFD-induced oxidative stress, inflammation and hepatic dysfunction in vivo. Manipulation of miR-665-3p in primary hepatocytes also caused similar phenotypic alterations in vitro. Mechanistically, we demonstrated that miR-665-3p directly bound to the 3ʹ-UTR of FNDC5 to downregulate its expression and inactivated the downstream AMPKα pathway, thereby facilitating oxidative stress, inflammation and NAFLD progression.

miR-665-3p antagomir alleviates NAFLD progression in mice
To determine the involvement of miR-665-3p in NAFLD progression, we first measured whether miR-665-3p expression was altered during NAFLD progression. As shown in Figure 1(a), hepatic miR-665-3p level was significantly increased in HFD-treated mice. In addition, the level of miR-665-3p was also elevated in PO-stimulated primary hepatocytes (Figure 1(b)). The dramatic increase of miR-665-3p level during NAFLD progression promoted us to explore whether inhibiting miR-665-3p could alleviate HFD-induced NAFLD. As shown in Figure S1A, hepatic miR-665-3p expression was inhibited by the miR-665-3p antagomir. Intriguingly, treatment with the miR-665-3p antagomir significantly reduced body weight in HFD mice, and the weights of epididymal and inguinal fat pads were also decreased (Figure S1B-C). However, food intake in mice with either ND or HFD stimulation was unaffected by the miR-665-3p antagomir ( Figure S1D). As shown in Figure S1E-G, the miR-665-3p antagomir dramatically restored HFD-induced glycometabolic disorder, as evidenced by the decreased FBG, serum insulin and HOMA-IR levels. Meanwhile, hyperlipemia in HFD mice was also improved in those treated with the miR-665-3p antagomir ( Figure S1H). Consistent with the alterations of systemic metabolism, the miR-665-3p antagomir also decreased lipid deposition in the liver upon HFD stimulation, as evidenced by HE/Oil red O staining and the decreased hepatic TG, TC and NEFA contents (Figure 1(c-d)). Accordingly, HFD-induced increases in liver weight and liver weight/tibial length were prevented by the miR-665-3p All results were expressed as the means ± standard deviations, n = 6 for each group, and *P < 0.05 was considered statistically significant. n.s. indicated no significance.
antagomir (Figure 1(e-f)). Fibrosis is a critical pathology of NAFLD, and our findings indicated that the miR-665-3p antagomir dramatically ameliorated HFD-induced hepatic fibrosis, as evidenced by the decreased hepatic hydroxyproline content and mRNA levels of fibrotic markers, including Col1α1, Col3α1, CTGF and TGF-β1 (Figure 1(g) and Figure S1I). Consistently, serum levels of ALT and AST, two biomarkers of liver injury, were significantly lower in the miR-665-3p antagomir-treated mice upon HFD stimulation (Figure 1(h)). These findings indicate that the miR-665-3p antagomir alleviates NAFLD progression in mice.

miR-665-3p antagomir reduces HFD-induced hepatic oxidative stress and inflammation
Emerging studies have identified oxidative stress and chronic inflammation as major culprits of NAFLD; therefore, we detected whether treatment with the miR-665-3p antagomir could reduce HFD-induced hepatic oxidative stress and inflammation. As shown in Figure 2(a-b), hepatic ROS and H 2 O 2 levels were significantly decreased by the miR-665-3p antagomir upon HFD stimulation. And the miR-665-3p antagomir also reduced hepatic levels of MDA, a product of lipid peroxidation and a biomarker for oxidative stress (Figure 2(c)). In addition, the level of antiinflammatory cytokine (IL-10) was increased, whereas the levels of proinflammatory cytokines (IL-1β, IL-6, MCP-1 and TNF-α) were decreased in the miR-665-3p antagomir-treated livers upon HFD stimulation (Figure 2(d-e)). Collectively, our results reveal that the miR-665-3p antagomir reduces HFD-induced hepatic oxidative stress and inflammation.

miR-665-3p agomir facilitates NAFLD progression in mice
Next, we determined whether elevating miR-665-3p levels contributed to the progression of NAFLD. As shown in Figure S2A, hepatic miR-665-3p level was significantly increased by the All results were expressed as the means ± standard deviations, n = 6 for each group, and *P < 0.05 was considered statistically significant. miR-665-3p agomir. HFD-induced increases of body weight and adipose tissue weight were further enhanced in mice treated with the miR-665-3p agomir ( Figure S2B-C). And systemic metabolic disorders of glucose and lipid were also amplified by the miR-665-3p agomir upon HFD stimulation, as evidenced by the increased levels of FBG, serum insulin, HOMA-IR, serum TG and TC ( Figure S2D-G). In addition, treatment with the miR-665-3p agomir further aggravated lipid accumulation in the liver from HFD mice, and increased liver weight (Figure 3(a-c)). As expected, hepatic fibrosis and injury were exacerbated in the miR-665-3p agomir-treated HFD mice, as evidenced by the increased levels of hepatic hydroxyproline and serum ALT and AST (Figure 3 (d-e)). Taken together, our findings suggest that the miR-665-3p agomir facilitates NAFLD progression in mice.

miR-665-3p modulates oxidative stress, inflammation and cellular injury in PO-treated primary hepatocytes
Next, we tried to validate the role of miR-665-3p in primary hepatocytes in vitro. As shown in Figure 5 (a), the miR-665-3p antagomir significantly reduced LDH releases in PO-treated hepatocytes. And intracellular accumulations of TG and TC were also decreased in the miR-665-3p antagomir-treated hepatocytes upon PO stimulation ( Figure 5(b)). As expected, treatment with the miR-665-3p antagomir dramatically suppressed PO-stimulated oxidative stress in primary hepatocytes, as evidenced by the Serum ALT and AST levels. All results were expressed as the means ± standard deviations, n = 6 for each group, and *P < 0.05 was considered statistically significant. decreased cellular ROS and MDA levels ( Figure 5  (c-d)). And PO-induced increases of IL-1β, IL-6, MCP-1 and TNF-α, and decrease of IL-10 were significantly attenuated by the miR-665-3p antagomir (Figure 5e). In contrast, PO-induced elevation of LDH releases in primary hepatocytes was further increased by the miR-665-3p agomir (figure 5(f)). And treatment with the miR-665-3p agomir also facilitated intracellular TG and TC accumulations ( Figure 5(g)). Consistent with the phenotypic alterations, the miR-665-3p agomir also aggravated oxidative damage and inflammatory response in POtreated primary hepatocytes ( Figure 5(h-j)). These results implicate that miR-665-3p modulates oxidative stress, inflammation and cellular injury in POtreated primary hepatocytes.

miR-665-3p antagomir prevents NAFLD progression via activating AMPKα in vivo and in vitro
AMPKα is a promising therapeutic target to treat NAFLD; therefore, we investigated whether the miR-665-3p antagomir prevented NAFLD progression via activating AMPKα. As shown in Figure 6(a-b), the miR-665-3p antagomir restored, while the miR-665-3p agomir further decreased AMPKα phosphorylation in the liver from HFD mice. To validate the necessity of AMPKα, HFD mice were pretreated with CpC to inhibit AMPKα. As shown in Figure 6(c-e), the miR-665-3p antagomir-induced inhibition on oxidative stress and inflammation in HFD mice were completely abolished by CpC treatment. The miR-665-3p antagomir significantly reduced hepatic lipid accumulation in HFD mice, but had no effect in those treated with CpC ( figure 6(f-g)). Accordingly, AMPKα inhibition also abrogated the protective effects of the miR-665-3p antagomir against HFD-induced hepatic fibrosis and injury, as evidenced by the increased hepatic hydroxyproline content and serum ALT and AST levels ( Figure 6(h-i)). In line with the in vivo results, the inhibitory effects of the miR-665-3p antagomir on PO-induced LDH releases and intracellular lipid accumulations were completely abolished by CpC (Figure 7(a-b)). In addition, CpC treatment also blocked the antioxidant and antiinflammatory effects of the miR-665-3p antagomir in POstimulated primary hepatocytes (Figure 7(c-e)). Together, these findings suggest that the miR-665-3p antagomir prevents NAFLD progression via activating AMPKα in vivo and in vitro.

miR-665-3p antagomir activates AMPKα via increasing FNDC5 expression
Finally, we explored the underlying mechanism through which the miR-665-3p antagomir activated AMPKα. As predicted by the TargetScan in the miR-665-3p agomir-treated hepatocytes. All results were expressed as the means ± standard deviations, n = 6 for each group, and *P < 0.05 was considered statistically significant. online software, FNDC5 was identified as a potential candidate for its role in regulating NAFLD progression and AMPKα activation [49,50]. A predicted binding site of miR-665-3p was found in FNDC5 3ʹ-UTR (Figure 8(a)). In addition, we observed that the miR-665-3p antagomir increased, whereas the miR-665-3p agomir decreased hepatic FNDC5 mRNA and protein levels in HFD mice (Figure 8(b-c)). Meanwhile, hepatic levels of irisin, the cleaved and active form of FNDC5, were also increased by the miR-665-3p antagomir, but decreased by the miR-665-3p agomir (Figure 8(d)). Next, luciferase reporter assay was performed to determine the direct interaction between miR-665-3p and FNDC5. As shown in Figure 8(e), the miR-665-3p agomir significantly reduced the luciferase activities in cells transfected with WT FNDC5 3ʹ-UTR, but not in those transfected with TRU FNDC5 3ʹ-UTR. To validate the involvement of FNDC5, mice were intravenously injected with shFNDC5 to knock down endogenous FNDC5 expression in the liver (Figure 8(f)). As shown in Figure 8(g), FNDC5 silence blocked AMPKα activation in the miR-665-3p antagomir-treated HFD mice. Accordingly, the miR-665-3p antagomir-mediated hepatoprotection was also abrogated by FNDC5 knockdown, as evidenced by the increased hepatic hydroxyproline content and serum ALT and AST levels (Figure 8(h-i)). In conclusion, our results show that the miR-665-3p antagomir activates AMPKα via increasing FNDC5 expression.

Discussion
Due to the dramatic changes in diets and lifestyles, NAFLD has becoming one of the leading chronic liver diseases, affecting about 1.7 billion individuals worldwide, and the national prevalence of NAFLD in China is approximately 29.2% [1,51,52]. The spectrum of this disease ranges from simple hepatic steatosis to advanced nonalcoholic steatohepatitis that eventually progresses to end-stage liver diseases such as cirrhosis and hepatocellular carcinoma. Moreover, a large proportion of NAFLD patients also suffer from a high prevalence of extrahepatic complications (e.g., cardiovascular diseases and extrahepatic cancers). Yet, there are no effective strategies to manage NAFLD and its related comorbidities. In the present study, we measure a significant upregulation of miR-665-3p level in the liver and primary hepatocytes upon HFD or PO stimulation, and the elevated miR-665-3p expression aggravates oxidative stress, inflammation and NAFLD progression in mice. In contrast, miR-665-3p inhibition by the miR-665-3p antagomir significantly prevents HFD-induced oxidative stress, inflammation and hepatic dysfunction in vivo. Manipulation of miR-665-3p in primary hepatocytes also causes similar phenotypic alterations in vitro. Mechanistically, we demonstrate that miR-665-3p directly binds to the 3ʹ-UTR of FNDC5 to downregulate its expression and inactivates the downstream AMPKα pathway, thereby facilitating oxidative stress, inflammation and NAFLD progression ( Figure 9). Overall, our findings identify miR-665-3p as an endogenous positive regulator of NAFLD via inactivating FNDC5/AMPKα pathway, and inhibiting miR-665-3p may provide novel therapeutic strategies to treat NAFLD.
The liver is a key organ to control systemic glycolipid metabolism and the whole-body energy homeostasis, while its function is evidently compromised in the development of NAFLD due to the excessive lipid deposition and hepatic fibrosis. Accumulation of NEFA in hepatocytes interferes the electron transport chain and mitochondrial function, causing electron leakage and uncontrolled ROS generation. In addition, NEFA and intracellular substances derived from the injured hepatocytes activate intrahepatic Kupffer cells to aggravate hepatic inflammation [17,53,54]. Unfortunately, oxidative stress and inflammation form a vicious cycle to further accelerate NAFLD progression. Moreover, previous findings by us (unpublished) and the others have determined a compromised antioxidant capacity in the liver from HFD mice [7,54]. Hepatic fibrosis is a key feature of NAFLD, and closely correlates with the antagomir-treated hepatocytes with or without CpC incubated upon PO stimulation. All results were expressed as the means ± standard deviations, n = 6 for each group, and *P < 0.05 was considered statistically significant. liver function and prognosis in NAFLD patients. During NAFLD progression, hepatic stellate cells are activated and transform into proliferative, fibrogenic and contractile myofibroblasts, eventually resulting in excessive synthesis and deposition of extracellular matrix [55,56]. Herein, we demonstrated that the miR-665-3p antagomir significantly inhibited oxidative stress and inflammation, thereby attenuating HFD-induced hepatic fibrosis and dysfunction. The liver also plays critical roles in maintaining systemic glycolipid metabolism. In line with the improved NAFLD and hepatic function, HFD-induced glycometabolic disorder (e.g., adipose tissue weight, FBG, serum insulin, and HOMA-IR levels) was also alleviated by systemic injections of the miR-665-3p antagomir, yet exacerbated by the miR-665-3p agomir. These data suggest that inhibiting miR-665-3p may provide beneficial effects on both hepatic function and systemic glycometabolic disorder in NAFLD patients.
FNDC5 is a widely distributed type I transmembrane glycoprotein, and can be proteolytically cleaved and released as irisin, an exerciseresponsive polypeptide myokine [57,58]. Emerging studies have revealed a fact that FNDC5 and its cleaved form, irisin, play critical roles in metabolic diseases, including NAFLD [49,59]. Zhang et al previously reported that serum irisin levels were decreased in NAFLD patients, and negatively correlated with intrahepatic TG contents [60]. And Canivet et al demonstrated that the liver and hepatocytes are critical sources of FNDC5 and irisin, and that local FNDC5 in the liver stimulated significant benefits against hepatic steatosis and hepatocyte apoptosis [49]. In addition, FNDC5 and irisin are also implicated in the regulation of oxidative stress and inflammation in different organs, including the All results were expressed as the means ± standard deviations, n = 6 for each group, and *P < 0.05 was considered statistically significant.
liver. Zhang et al recently found that FNDC5 suppressed protein kinase B-dependent nuclear export and degradation of nuclear factor erythroid-2-related factor-2, thereby alleviating doxorubicininduced oxidative stress in the heart [61]. Bi et al showed that treatment with exogenous irisin protected against I/R-induced mitochondrial injury and oxidative stress in the liver [62]. Moreover, serum irisin levels were found to be independently associated with portal inflammation and oxidative status in NAFLD progression [63,64]. A very recent study by Li et al proved that FNDC5 upregulation by nicotinamide riboside significantly prevented HFDinduced NAFLD [65]. AMPKα is identified as a critical downstream target of FNDC5 and irisin, and mediates their antioxidant and antiinflammatory effects upon different stimulations [50,66].

Conclusion
In summary, we for the first time demonstrate the involvement of miR-665-3p in promoting NAFLD via inactivating FNDC5/AMPKα pathway, and inhibiting miR-665-3p may provide novel therapeutic strategies to treat NAFLD.

Disclosure statement
No potential conflict of interest was reported by the author(s).

Funding
The present work was supported by grants from the Fundamental Research Funds for the Central Universities [No. 2042020kf0085].

Data availability
The data that support the findings of this study are available from the corresponding author upon reasonable request.