Irisin attenuates ethanol-induced behavioral deficits in mice through activation of Nrf2 and inhibition of NF-κB pathways

This study aims to investigate the effect of irisin on ethanol-induced behavioral deficits and explore the underlying mechanisms. A mouse model of ethanol addiction/withdrawal was constructed through chronic ethanol administration. Depressive-like behaviors were evaluated by the tail suspension test and forced swimming test, and anxiety-like behaviors were evaluated by the marble-burying test and elevated plus maze test. The expression of Nrf2 was measured by western blotting. Levels of inflammatory mediators (NF-κB, TNF-α, IL-1β and IL-6) and oxidative stress factors (ROS, MDA, GSH and SOD) were detected by ELISA. The ethanol-induced PC12/BV2 cell injury model was used to elucidate whether the effect of irisin on ethanol-induced neurological injury was related to anti-inflammatory and antioxidant mechanisms. Ethanol-induced ethanol preference and emotional deficits were improved by chronic irisin treatment; however, these improvements were partly reversed by cotreatment with the Nrf2 inhibitor ML385. Further results implied that the improvement effect of irisin on behavioral abnormalities may be related to its anti-inflammatory and antioxidant effects. In detail, irisin inhibited ethanol-induced abnormal expression of ROS and MDA and upregulated the expression of GSH and SOD. Meanwhile, irisin treatment inhibited ethanol-induced overexpression of NF-κB, TNF-α, IL-1β and IL-6 in the hippocampus and cerebral cortex. The regulation of oxidative stress factors by irisin was reversed after ML385 treatment. In the in vitro study, overexpression of oxidative stress factors in ethanol-treated PC12 cells was inhibited by irisin treatment; however, the prevention was reversed after the knockdown of Nrf2 siRNA. Moreover, ethanol-induced overexpression of inflammatory mediators in BV2 cells was also inhibited by irisin treatment. Irisin improved depressive and anxiety-like behaviors induced by ethanol addiction/withdrawal in mice, and this protection was greatly associated with the NF-κB-mediated anti-inflammatory signaling pathway and Nrf2-mediated antioxidative stress signaling pathway.


Introduction
Moderate drinking is beneficial for health (Guzman et al. 2004;Orio et al. 2019). However, excessive drinking leads to the aggravation of various diseases and ultimately causes damage to the central nervous system (Jiang et al. 2020). For a long time, physical and drug therapies have been used to treat emotional disorders and withdrawal symptoms induced by ethanol addiction, but the curative effect isexpected (Bruijnzeel et al. 2004). Recent studies have suggested that voluntary exercise decreases ethanol preference and consumption, which facilitates the improvement of ethanol-induced injuries in the central nervous system (Gallego et al. 2015). However, the pathogenesis of how exerciseimproves alcoholism has not been clarified thus far.
Growing evidence suggests that exercise effectivelyinhibits oxidative stress and reverses inflammation in metabolic diseases such as. Azheimer's disease, chronic liver disease, and diabetes (Ji and Zhang 2014;Tu et al. 2018).
Relevant studies have found that the benefits of exercise are closely related to a skeletal muscle-derived myokine, irisin, which originates from fibronectin type III domaincontaining protein 5 (FNDC5) (Bostrm et al. 2012). The FNDC5/irisin system is stimulated upon physical exercise. FNDC5/irisin can be packaged into extracellular vesicles (EVs) and secreted into the circulatory system (Chi et al. 2022). Many studies have indicated that the FNDC5/irisin system offers multiple advantages in metabolic diseases, such as an anti-inflammatory response in Alzheimer's disease (AD) by decreasing amyloid β-protein (Aβ) (Noda et al. 2018;Kim and Song 2018), antioxidant effects in type 2 diabetes (Zhu et al. 2015), and antiapoptotic effects in lipopolysaccharide-induced liver injury (Li et al. 2021). Moreover, some studies have shown that irisin is beneficial to ethanolinduced liver diseases, such as alcoholic fatty liver disease (AFLD), primary biliary cholangitis (PBC) and alcoholic cirrhosis (AC) (Choi et al. 2014;Waluga et al. 2019). However, most studies still focus on the effect of irisin on peripheral metabolic diseases caused by ethanol. In recent years, an increasing number of studies have shown that irisin prevents neuroinflammation by inhibiting the inflammatory response in AD (Lourenco et al. 2020) and Parkinson's disease (Pignataro et al. 2021). Therefore, we wondered whether irisin plays an important role in chronic ethanol-induced central metabolic abnormalities.
Oxidative stress and inflammation are considered the major mechanisms in the pathogenesis of ethanol addiction and withdrawal (Schneider et al. 2017;Mohseni et al. 2021). Excessive exposure to ethanol induces overexpression of reactive oxygen species (ROS), resulting in abnormal expression of glutathione (GSH), malondialdehyde (MDA) and Superoxide dismutase (SOD) (Contreras et al. 2017), which is mainly conducted by the nuclear factor erythroid 2-related Factor 2 (Nrf2)-mediated antioxidative signaling pathway (Quintanilla et al. 2020). These pathological changes after ethanol exposure can result in neuronal apoptosis and peroxidation. Moreover, ethanol can upregulate the expression of proinflammatory cytokines, such as interleukin-1β (IL-1β) and interleukin-6 (IL-6), via the activation of nuclear factor kappaB (NF-κB) and tumor necrosis factor-α (TNF-α) (Obernier et al. 2002;Tilg et al. 2016). Excessive expression of neuroinflammation causes direct damage to the central nervous system. Therefore, therapeutics targeting oxidative stress and inflammation are potential therapeutic strategies to protect against ethanol addiction and withdrawal.
At present, studies addressing whether irisin protects against ethanol addiction and withdrawal and its underlying mechanisms are rare. Therefore, the primary aim of this study was to explore the effects of irisin on ethanol addiction. We found that exogenous irisin treatment can relieve ethanol addiction and withdrawal symptoms, which is largely related to its anti-inflammatory and antioxidative stress effects.

Study design
The experiments in this study included two parts. The first part of the experiments lasted for 4 weeks and was designed to investigate the correlation between tissue levels of irisin and ethanol-induced behavioral abnormalities. To study the changes in irisin during the formation of ethanol-induced behavioral deficits, 24 mice were divided into three groups, each for one time point (0 week, 2nd week and 4th week), and received different concentrations of ethanol (5%, 10%, 20%, and 35%) in the 1st − 4th weeks (Jiang et al. 2020). Mice in each group received the behavior tests and were then sacrificed at the 0, 2nd, and 4th weeks. The second part of the experiments lasted for 8 weeks and was designed to explore the effects of irisin on ethanol-induced behavioral abnormalities. Mice received different concentrations of ethanol (5%, 10%, 20%, and 35%) in the 1st − 4th weeks. Mice continued to receive 35% ethanol from the 5th to 8th week. Sixty mice were divided into 5 groups: control, ethanol, irisin (5 µg/kg) + ethanol, irisin (10 µg/kg) + ethanol and Nrf2 inhibitor + irisin (10 µg/kg) + ethanol. There were two subgroups in each group (n = 6 per subgroup). One subgroup was used for depressive-like behavior tests. The other subgroup was used for anxiety-like behavior tests. Mice were fed with ethanol at a gradient concentration (5%, 10%, 20%, and 35%) for four weeks and then continued to receive ethanol at a concentration of 35% for three weeks. Mice in the irisin-treated groups received drug treatment during the 5th-8th weeks, and mice in the inhibitor cotreatment group received ML385 (Nrf2 inhibitor, 30 mg/kg, ip) for one week (7th-8th week, Fig. 1). Mice were sacrificed after the behavior test at the 8th week. In these two parts, the serum, hippocampus and cerebral cortex of mice were collected and stored at -80 °C.

Behavior tests
Each mouse received behavioral tests after ethanol withdrawal. The behavioral tests were conducted every two weeks.

Locomotor activity
The locomotor activity test was based on our previous research (Yu et al. 2016). Individual mice were placed in the testing chamber for 10 min. The testing software recorded the spontaneous activities of each animal and exported the results with the number of crawling grids.
Ethanol preference test Mice were given both 5% ethanol diluted in drinking water and normal drinking water for 1 h. The volumes of both solutions consumed by the mice were recorded during this hour. The ethanol preference behavior of the animals was evaluated as the ratio of the volume of ethanol solution consumption/the total liquid intake.

Tail Suspension Test
The tail suspension test (TST) was performed based on our previous research (Jiang et al. 2017a). The mice were fixed with tape at the posterior 1/3 of the tail and suspended on a stand with the head 50 cm above the ground. The test lasted for 6 min, and the immobility time of the mice was recorded and analyzed during the last 4 min.

Forced Swimming Test
In the forced swimming test (FST), each mouse was placed in a water tank (containing a 20 cm depth of water), and the immobility time of each mouse was recorded for 6 min (Jiang et al. 2017b).

Marble-Burying Test
In the marble-burying test (MBT), mice were individually placed in a test chamber (40 cm × 25 cm × 20 cm) with corn cobs evenly laid at a depth of 5 cm. Nine glass marbles with a diameter of 2.5 cm were placed in the chamber. The number of glass marbles buried within 10 min was recorded, and the marbles were considered buried only when 2/3 or more of the volume of glass marbles were covered by corn cob bedding (Jiang et al. 2017b).

Elevated Plus Maze Test
In the elevated plus maze test (EPMT), each mouse was placed in the central area of the maze with its head facing the open arm. The number of entries in the closed arm and the time spent in the closed and open arms were recorded and analyzed for 10 min. The experimenter was at a distance of 1 m from the maze during the experiment. After each mouse was tested, the maze was wiped with 75% alcohol to eliminate the effect of animal odor on subsequent experiments. Mice were subjected to ethanol drinking for 8 weeks and received the first administration of irisin (5,10 µg/kg, i.v.) or vehicle after 4 weeks of drinking; behavior tests were performed at 0, 2, 4, 6, and 8 weeks after drinking. Each group contained control, ethanol, ethanol supplemented with different doses (5, 10 µg/kg) of irisin, and Nrf2 inhibitor + irisin (10 µg/kg) + ethanol. Each group was divided into two sets (n = 6 per set). The mice in the 1st set received locomotor activity at 9:00, an ethanol preference test at 11:00, a forced swimming test at 13:00, and a tail suspension test at 15:00. The mice in the 2nd set received locomotor activity at 9:00, an ethanol preference test at 11:00, a marble-burying test at 13:00, and an elevated plus maze test at 15:00 Cell experiment BV2 and PC12 cells (Cell Storage Center,Wuhan,China) were cultured using Dulbecco's modified Eagle's serum (DMEM; Life Technologies, Carlsbad, CA) supplemented with 10% fetal bovine serum (FBS; Life Technologies), GlutaMAX (Life Technologies), and antibiotic-antimycotic (Life Technologies).
Irisin treatment of ethanol treated cells: To test ethanol tolerance in PC12 cells, PC12 cells were seeded in 96 well plates and then treated with different concentrations of ethanol (50, 200, 400, 800 and 1200 mM) for 24 h.The cell survival rate was tested using a CCK-8 kit. In brief, cells were incubated in culture medium with 10 µl CCK-8 solution (Abcam, Shanghai, China) for 1 h. The absorbance of each well was measured at a wavelength of 450 nm with a reference wavelength of 650 nm by a spectrophotometer (Multiskan Spectrum, Thermo Scientific). To determine the effect of irisin on ethanol-induced damage in PC12 cells, PC12 cells were cultured with irisin (10, 20, and 40 nmol/L) for 12 h following 24 h of ethanol (800 mM) treatment. The levels of oxidative stress markers were then determined.
To explore the effect of irisin on ethanol-induced neuroinflammation in BV2 cells, BV2 cells were cultured with irisin (doses of 40 nmol/L) for 12 h following a 24-h ethanol (150 mM) treatment. The dose of ethanol (150 mM) was selected according to ethanol tolerance in BV2 cells.
Nrf2 Small interfering RNA (siRNA) transfections: Nrf2 siRNA was purchased from Santa Cruz Biotechnology. Lipofectamine 2000 (Invitrogen) was used as the transfection reagent according to the transfection requirements. BV2 cells were incubated with a final concentration of 20 nM siRNA in the presence of transfection reagent. After transfection, cells were collected for subsequent experiments.

Measurement of oxidative stress markers
For animal samples, the homogenate was prepared by sonication of each tissue. The supernatant was collected after centrifugation. For PC12 cell samples, the cells were collected and centrifuged, and the supernatant was used for detection. The levels of ROS, MDA and GSH and the activity of SOD were measured with reagent kits (Thermo Scientific). The SOD data are expressed as U/mg protein. The MDA results are expressed as nmol/mg protein. The GSH data are expressed as nmol/mg protein.

Western blot analysis
The tissues were weighed and lysed by adding RIPA containing protease inhibitors, and the supernatant was extracted after centrifugation. The protein concentration of each sample was determined by a BCA kit (Thermo Scientific). First, polyacrylamide gels were prepared, and 48 µg total protein was added to each well for electrophoresis and membrane transfer. After that, blots were blocked with blocking buffer for 2 h and then incubated with primary antibodies (anti-Nrf2 1:1,000, purchased from Abcam; β-actin 1:1,000, purchased from Santa Cruz) for 2 h. After washing with Tris-buffered saline containing 0.1% Tween 20 (TBST), blots were incubated with secondary antibodies. Finally, blots were imaged by a fluorescence scanner (LI-COR Biotechnology, Inc. South San Francisco, CA, USA).

Enzyme linked immunosorbent assay (ELISA)
The levels of NF-κBp65, IL-6, IL-1β, and TNF-α in the brain tissues and BV2 cells were tested by an ELISA kit (Thermo Scientific). Irisin levels were measured with an ELISA kit purchased from Phoenix Pharmaceuticals, Burlingame, CA, USA. Briefly, diluted protein standards and samples were added to a 96-well ELISA plate, followed by biotinylated antibodies. After washing with wash buffer, the prepared solution of avidin and horseradish peroxidase-conjugated complex was added to each well. Finally, the reaction was terminated by the stopping solution. For irisin, IL-6, IL-1β, and TNF-α, the OD values were measured at 450 nm. For NF-κBp65, the OD value was assessed at 405 nm.

Data analysis
Data are presented as the mean ± S.E.M. Statistical analysis of data was performed by two-way or one-way analysis of variance (ANOVA) via SPSS software (IBM, USA). For two-way ANOVA, ethanol and irisin treatments were taken as between-group factors. When needed, the time factor (0, 2, 4, 6 and 8 weeks) was taken as a within-subject factor. The Newman-Keuls test was used for post hoc comparisons. For one-way ANOVA, Dunnett's test was used for post hoc comparisons. P < 0.05 was regarded as a significant difference.

Ethanol addiction decreases the level of irisin in the serum and brain
In the ethanol drinking mouse model, four weeks of ethanol drinking significantly decreased the level of irisin in the serum (p < 0.001, Fig. 2A) compared to the 0-week group (control group). In the brain, four weeks of ethanol drinking caused a decrease in irisin expression in both the hippocampus (p < 0.001, Fig. 2B) and the cerebral cortex (p < 0.001, Fig. 2C) compared to the control group. To study the relationship between ethanol drinking-induced emotional behavior and irisin expression, we analyzed the correlation between depressive/anxiety-like behavior and irisin levels. The results showed that ethanol-induced behavioral deficits, including ethanol preference (p < 0.001, Fig. 2D), anxiety-like behavior (MBT) (p < 0.001, Fig. 2E) and depressive-like behavior (FST) (p < 0.001, Fig. 2F), had a strong negative correlation with irisin expression in the serum. A negative correlation between behavioral deficits and irisin expression was also found in the hippocampus and cerebral cortex (p < 0.001 in all cases, Fig. 2G-I). These results indicated that irisin expression was strongly associated with ethanol-induced behavioral deficits, including depressive and anxiety-like behaviors.

Exogenous irisin improves ethanol-induced depressive and anxiety-like behaviors in mice
There was no significant difference in the locomotor activities in mice after ethanol drinking with and without irisin treatment (Fig. 3A). Four weeks of ethanol drinking led to a significant ethanol preference in mice, and this preference lasted until the 8th week with continued ethanol feeding (p < 0.001, Fig. 3B). For depressive-like behavior, four weeks of ethanol drinking caused an increase in immobility time in the FST (p < 0.001, Fig. 3C) and TST (p < 0.001, Fig. 3D), and these phenomena lasted until the 8th week compared to the control group. For anxiety-like behavior, four weeks of drinking increased the number of buried marbles in the MBT in mice (p < 0.001, Fig. 3E). In the EPMT, ethanol drinking obviously increased the time spent in the closed arms and the number of entries into the closed arms (p < 0.001, Fig. 3F and G). Importantly, four weeks of 10 µg/kg irisin treatment (5 to 8 weeks) decreased ethanol-induced preference for ethanol (p < 0.001, Fig. 3B) and depressive and anxiety-like behaviors (p < 0.001 in all cases, Fig. 3C-G) in mice. More importantly, cotreatment with the Nrf2 inhibitor ML385 partly reversed the gains attributed to irisin in diminishing ethanol preference (p < 0.05, Fig. 3B), depressive-like behaviors (p < 0.05 for FST and p < 0.01 for TST, Fig. 3C-D) and anxietylike behaviors (p < 0.05 for MBT and p < 0.01 for EPMT, Fig. 3E-G). These data suggested that exogenous irisin administration can ameliorate ethanol-induced behavioral deficits, which may be related to Nrf2 activation. Effects of exogenous irisin on the expression of nuclear Nrf2 and NF-κB p65 in ethanol-exposed mice To further explore whether irisin's amelioration of drinkinginduced behavioral deficits is related to inflammation and oxidative stress, the expressions of Nrf2 and NF-κB p65 in the hippocampus and cerebral cortex were determined by Western blot or ELISA. In our study, the expression of nuclear Nrf2 was significantly decreased in the hippocampus and cerebral cortex of the drinking mice (p < 0.001 in all cases, Fig. 4A-C). Moreover, NF-κB p65 protein levels were increased in ethanol-treated mice compared with control mice (p < 0.001 in all cases, Fig. 4D-E). In contrast, treatment with 10 µg/kg irisin markedly increased Nrf2 expression (p < 0.01 for the hippocampus and cerebral cortex, Fig. 4B-C) and decreased NF-κB p65 expression (p < 0.01 for the hippocampus and p < 0.001 for the cerebral cortex, Fig. 4D-E) compared with ethanol drinking. Furthermore, ML385 cotreatment reversed the effect of irisin on Nrf2 expression in the hippocampus (p < 0.05, Fig. 4B) and cerebral cortex (p < 0.01, Fig. 4C). These results confirmed that irisin's amelioration of ethanol drinking-induced behavioral deficits can be attributed to the upregulation of Nrf2 and the suppression of NF-κB p65 expression.

Effects of exogenous irisin on Nrf-2-mediated oxidative stress markers (ROS, MDA, GSH, and SOD levels) in ethanol-exposed mice
To investigate the effect of irisin on oxidative stress injury in ethanol-treated mice, the levels of the key antioxidative factor Nrf-2, including ROS, MDA, GSH, and SOD, were measured in our study. Ethanol exposure led to a significant increase in ROS production and MDA levels (p < 0.001 in all cases, Fig. 5A-D), as well as a significant decrease in GSH levels and SOD activity in the hippocampus and cerebral cortex of the drinking mice (p < 0.001 in all cases, Fig. 6A-D) compared with the normal group. Irisin treatment at 10 µg/kg significantly reversed these changes. However, cotreatment with ML385 eliminated the improvements associated with irisin on ROS, MDA, GSH, and SOD in the hippocampus (p < 0.01 in all cases, Figs. 5A, C, 6A, C) and cerebral cortex (p < 0.01 for ROS, Fig. 5B; p < 0.05 for MDA, Fig. 5D; p < 0.05 for GSH, Fig. 6B; p < 0.01 for SOD, Fig. 6D). These results indicated that irisin's ameliorative effect on ethanol-induced behavioral deficits could be attributed to its regulation of oxidative stress-associated factors, including an increase in ROS and MDA levels and a decrease in GSH and SOD

Fig. 3 Effects of irisin on locomotor activity (A), ethanol preference test (B), forced swimming test (C), tail suspension test (D), marbleburying test (E) and elevated plus maze test (F, G) in drinking mice
at the 0-8th week. Values are expressed as the mean ± S.E.M. with 6 mice in each group. **p < 0.01 and ***p < 0.001 compared with the control group; # p < 0.05, ## p < 0.01 and ### p < 0.001 compared with the water drinking group. $ p < 0.05 and $$ p < 0.01 compared with the irisin-treated drinking group with 6 mice in each group. ***p < 0.001 compared with the control group; ## p < 0.01 and ### p < 0.001 compared with the water drinking group. $ p < 0.05 and $$ p < 0.01 compared with the irisin-treated drinking group activities, which may be closely associated with the Nrf2 pathway.

Effects of exogenous irisin on ethanol-induced changes in the Nrf2 signaling pathway in PC12 cells
First, we determined the optimal ethanol concentration to establish the cellular injury model in PC12 cells. The cell viability reached 44.8 ± 5.1% under 800 mM ethanol treatment, and the cell viability reached 21.3 ± 5.7% under 1200 mM ethanol treatment (Fig. 7A). Therefore, 800 mM ethanol was used in further studies. Our data found that 40 nM irisin treatment mitigated ethanol-induced cell damage (p < 0.01, Fig. 7B). Pretreatment with Nrf2 siRNA partly reversed irisin's protective effect on ethanol-induced cell damage. The nuclear Nrf2 expression was decreased after 800 mM ethanol treatment (p < 0.001, Fig. 7C-D), which was upregulated by treatment with 40 nM irisin (p < 0.01, Fig. 7D). In addition, 800 mM ethanol induced increases in ROS production (p < 0.001, Fig. 7E) and MDA levels (p < 0.001, Fig. 7F) and decreases in GSH levels (p < 0.01, Fig. 7G) and SOD activity (p < 0.001, Fig. 7H) in PC12 cells. These changes were reversed after irisin treatment (p < 0.01 for ROS, MDA and SOD; p < 0.05 for GSH, Figure E-H). Furthermore, pretreatment with Nrf2 siRNA reversed irisin's antioxidant effect on ethanol-induced abnormal expression of cellular oxidative stress markers (p < 0.05 for ROS, MDA, GSH, Figure E-H). These results were consistent with the protective effect of irisin on ethanol drinking mice, demonstrating that irisin's antioxidant effect on ethanol-induced cellular oxidative stress is mainly regulated by the Nrf2-mediated antioxidative stress pathway.

Discussion
An increasing number of studies have indicated that irisin plays a critical role in some metabolic diseases and alcoholic liver disease. However, the functions of irisin in neuroinjury after ethanol addiction are still unknown. In the present study, we found a negative correlation between ethanol drinking-induced behavioral deficits and irisin levels in ethanol drinking mice. Further study showed that exogenous irisin can significantly ameliorate ethanol-induced emotional deficits such as depressive and anxiety-like symptoms. Irisin's protective function against ethanol-induced neurological injury was attributed to the suppression of neuroinflammatory factor expression mainly via the NF-κB pathway, and the inhibition of oxidative stress was largely mediated via the Nrf2 pathway.
Behavioral abnormalities in traditional animal models of ethanol drinking are not obvious compared to clinical symptoms (Bertola et al. 2013;Sanchez-Alavez et al. 2019). This may be caused by the absence of an ethanol withdrawal procedure in the animal. In the present study, a withdrawal procedure was added to the animal model to simulate the process of ethanol withdrawal and return to drinking in clinical patients. Our results showed that model mice showed obvious behavioral deficits after four weeks of ethanol drinking and withdrawal, indicating that the animal model of drinking is effective. This model successfully mimics the disease progression of drinking in the clinic andshortens the modeling time when compared to a previous model (Crews et al. 2006;Tajuddin et al. 2014).
Recently, studies have demonstrated that exercise can protect against drinking-induced injury in the central nervous system (Coiro et al. 2007;Paulucio et al. 2018); however, its mechanism remains unclear. In our study, a significant reduction in irisin levels was observed in the serum, hippocampus and cerebral cortex of mice after four weeks of drinking. This reduction had a strong negative correlation with drinking-induced behavioral deficits. In fact, irisin, a skeletal muscle-derived myokine expressed after exercise, has been shown to be associated with antioxidant, antiinflammatory and anti-apoptotic effects in various diseases (Askari et al. 2018;Mahalakshmi et al. 2020). Previous studies have found that irisin is beneficial to the development of alcoholic liver disease, AD, and diabetes (Zhu et al. 2015;Kim and Song 2018;Li et al. 2021). Our present study first explored the role of irisin in treating ethanol drinkinginduced behavioral deficits and its mechanism.
There are multiple mechanisms associated with the development of ethanol drinking-induced behavioral deficits, and oxidative stress is an essential mechanism during the progression of ethanol drinking (Chopra and Tiwari 2012;Nascimento et al. 2020). Prooxidants and antioxidants are in balance under normal biological systems; however, an increase in oxidants and a decrease in antioxidants can cause imbalance, eventually leading to oxidative stress. Chronic ethanol intake suppresses the antioxidant defense system (e.g., SOD and GSH) and increases oxidants (e.g., ROS and MDA), leading to peripheral and central oxidative stress (Wu and Cederbaum 2009;Li et al. 2020b).  Ethanol intake stimulates the production of ROS and thus leads to central oxidative stress, which is the pathogenesis of drinking-induced behavioral abnormalities such as depression and anxiety (Li et al. 2020a). The excessive intake of ethanol also leads to an increase in MDA levels, which is the end product of lipid peroxidation and can indirectly reflect the degree of free radical damage (Ganie et al. 2011). In contrast, ethanol intake suppresses the protective effect of GSH and SOD, which can scavenge lipid peroxides and oxygen-free radicals. In the present study, we found that irisin's function of alleviating ethanol-induced behavioral deficits is associated with its antioxidative stress mechanism, in which the activities of ROS, MDA, GSH and SOD were regulated.
The related protein nuclear factor Nrf2 is an important transcription factor of the antioxidant stress pathway, encoding downstream genes of antioxidant enzymes/proteins in the oxidative stress responses (Tonelli et al. 2018). Under normal states, Nrf2 activity is inhibited by binding to its repressor, Kelch-like ECH-associated protein 1 (Keap1). Under oxidative stress states, Nrf2 dissociates from the repressor, is transferred into the nucleus, and combines with antioxidant response elements (ARE), promoting the transcription of many antioxidant genes, such as SOD and GSH (Bellezza et al. 2018). Previous studies indicate that ethanol reduces nuclear abundance of Nrf2 (Kumar et al. 2011;Amirshahrokhi and Niapour 2022), as well as DNA binding activity of Nrf2 to ARE sequence compared with control (Amirshahrokhi and Niapour 2022). In our study, the decrease in nuclear Nrf2 levels were also found after drinking, and this decrease was reversed by irisin treatment in mice, and direct upregulation of the Nrf2 pathway by irisin in neuroinjury was also observed in PC12 cells. Taken together, these results suggest that the antioxidative stress effects of irisin on ethanol addiction/withdraw-induced behavioral deficits are largely dependent on the Nrf2 pathway.
In addition to oxidative stress mechanisms, the inflammatory response is another major pathological mechanism of ethanol drinking. Inflammation plays a crucial role in the occurrence and development of ethanol drinking-induced behavioral deficits. NF-κB, a nuclear transcription factor, regulates various proinflammatory cytokines, such as IL-1β, IL-6 and TNF-α (Maraslioglu et al. 2013;Wang et al. 2020). Ethanol and its metabolites increase NF-κB expression and further stimulate the release of inflammatory mediators, causing central and peripheral inflammation (Maraslioglu et al. 2013;Nkpaa et al. 2019). A malignant increase in these genes can cause behavioral abnormalities, mood and sleep fluctuation, and learning and memory deterioration (Chastain and Sarkar 2014). In our study, the increase in NF-κB, IL-1β, IL-6 and TNF-α levels induced by drinking was reversed by irisin treatment, suggesting that irisin could protect against ethanol-induced inflammation by suppressing inflammatory mediators by regulating the NF-κB pathway.
Notably, it was recently found that the antidepressant effect of irisin may also be related to the modulation of the brain-derived neurotrophic factor (BDNF) pathway (Jo and Song 2021), BDNF is known to have anti-inflammatory effect by regulating microglia activation and polarization Fig. 9 Molecular mechanisms involved in the protective effects of irisin against ethanol-induced neuroinflammation and neurotoxicity. The inhibition of NF-κB and proinflammatory cytokines might be involved in the antineuroinflammatory effect of irisin against ethanol-induced depressive and anxiety-like behavior. The activation of Nrf2 and improvement of oxidative stress (e.g., decrease of MDA and ROS, increase of SOD and GSH) might be involved in the antineurotoxicity effect of irisin against ethanol-induced depressive and anxiety-like behaviors with binding its receptor tropomyosin receptor kinase B (TrkB) (Qin et al. 2016). In addition to its anti-inflammatory effects, in the brain, BDNF activates the Nrf2 pathway to improve stress-induced central oxidative stress levels . Related study also found that depressive symptoms caused by ethanol exposure are associated with reduced BDNF levels (Yao et al. 2022), so whether the effect of irisin on ethanol withdrawal-induced depressive and anxiety symptoms found in the present study is related to BDNF still needs to be confirmed by further experiments.

Conclusion
This study indicated that irisin can ameliorate ethanol drinking-induced behavioral deficits, and the protection is attributed to its promotion of the oxidative stress response via the Nrf2-mediated antioxidative stress pathway and the significant suppression of inflammatory mediators via regulation of the NF-κB pathway (Fig. 9). Irisin can be considered a potential therapeutic target in the treatment of ethanol drinking addiction in the future.