FGF21 restores hippocampual mitochondrial dysfunction via enhancing Nrf2/HO-1 and AMPK/SirT1/PGC-1α signaling pathways to alleviate chronic unpredictable mild stress induced depressive like behaviors in mice

because our study suggests that Fibroblast growth factor 21 (FGF21), a member of the FGF superfamily, exert anti-depressive roles on a chronic unpredictable mild stress (CUMS)- induced model of depression. The effects were consistent with improved hippocampual mitochondrial function, reflected by FGF21-induced increases in mitochondrial membrane potential (MMP), ATP concentration and decrease of reactive oxygen species (ROS) levels. At the same time, FGF21 ameliorated oxidative stress in CUMS-exposed mice by enhancing superoxide dismutase (SOD), glutathione peroxidase (GSH-Px) and catalase activities, and reducing malondialdehyde (MDA) level in the hippocampus. Mechanistically, we found that CUMS treatment decreased expression levels of mitochondrial fusion protein 1 (MFN1), and increased expression levels of mitochondrial dynamin-related protein 1 (DRP1). FGF21 administration increased expression of MFN1, and reduced expression of DRP1. Meanwhile, FGF21 treatment promoted the expressions of Nrf2, HO-1, phosphorylated AMPK, SirT1, PGC-1a in the hippocampus. This study revealed that FGF21 alleviates CUMS induced depressive like behaviors by restoring mitochondria function, improving oxidative stress and enhancing Nrf2/HO-1 and AMPK/SirT1/PGC-1α signaling pathways. It suggested that FGF21 would be a potential therapeutic agent in the management of depression. Abstract: 1 Mitochondrial dysfunction plays a key role in the pathogenesis of depression. Ample research 2 proves mitochondria are a promising target for depression. Fibroblast growth factor 21 (FGF21) 3 exerts roles in neuroprotection and could enhance mitochondria function. Here, the anti-depressive 4 effect of FGF21 was evaluated on a chronic unpredictable mild stress (CUMS)- induced model of 5 depression. The depressive-like behaviors were assessed using sucrose preference test (SPT), forced 6 swim test (FST) and tail suspension test (TST). The results showed that treatment of FGF21 7 significantly attenuated the decrease in SPT, and dramatically reduced the immobility time in the 8 TST and FST. These effects were associated with enhanced hippocampal mitochondrial function, 9 reflected by FGF21-induced increases in mitochondrial ATP concentration, mitochondrial 10 membrane potential (MMP), and decrease of reactive oxygen species (ROS) levels. At the same 11 time, FGF21 ameliorated oxidative stress in CUMS-exposed mice by enhancing superoxide 12 dismutase (SOD), glutathione peroxidase (GSH-Px) and catalase activities, and reducing 13 malondialdehyde (MDA) level in the hippocampus. Mechanistically, we found that CUMS 14 treatment decreased expression level of mitochondrial fusion protein 1 (MFN1), and increased 15 expression level of mitochondrial dynamin-related protein 1 (DRP1). FGF21 administration 16 increased expression of MFN1, and reduced expression of DRP1. Meanwhile, FGF21 treatment 17 promoted the expression levels of Nrf2, HO-1, phosphorylated AMPK, SirT1, PGC-1a in the 18 hippocampus. This study revealed that FGF21 alleviates CUMS induced depressive like behaviors 19 by restoring mitochondria function via enhancing Nrf2/HO-1 and AMPK/SirT1/PGC-1α signaling 20 pathways. It suggested that FGF21 would be a potential therapeutic agent in the management of 21 depression.


Animals and treatments 98
Male C57BL/6 mice (20±2 g) were obtained from Guangdong Medical Laboratory Animal Center 99 (Guangzhou, China). The mice were housed under standard animal room conditions (temperature 100 22 ± 1 °C, humidity 55-60%, 12/12 h light/dark cycle). Mice were randomly divided into three 101 groups with fifteen mice in each group: Control, CUMS and CUMS + FGF21 groups. The CUMS 102 + FGF21 group was administrated with FGF21 (0.1 mg/kg) by intraperitoneal injection once every 103 day for 35 days. The experimental procedures were illustrated in Fig. 1A. 104

Chronic unpredictable mild stress (CUMS) induced depression model in Mice 105
CUMS protocol was performed as described previously (Zhao et al. 2021). In brief, , mice of 106 CUMS group and CUMS + FGF21 group were treated with the following mild stressors for 63 days 107 randomly: 4℃ for 1.5 h; swimming in 18℃ cold water for 10 min; bondage for 1.5 h; 45° cage 108 tilting for 12 h; strobe light for 12 h; inversion of the light/dark cycle for 12 h; water shortage for 109 12 h; food shortage for 12 h; wet bedding for 12 h; different bedding for 12 h; cage shaking for 2.5 110 h (150 rpm). These eleven stressors were used continuously and individually, the same stressor does 111 not reuse within three days. Control group mice were not received any kind of stressors and were 112 raised in normal conditions. 113

Sucrose preference test (SPT) 115
The SPT was done to operationally define the anhedonia behavior as previously described (Zhao 116 et al. 2021). In the adaption phase, mice were placed individually in cages supplied with one bottle 117 of 1% sucrose water (w/v) and one bottle of pure water for 48 h. Then, mice were deprived of food 118 and water for 24 h. Then, each mouse was allowed to drink water from the two bottles freely for 4 119 h, one with pure water and the other with 1 % sucrose water. After 4 h, the consumption of sucrose water and pure water was calculated, and the sucrose preference was measured as follows: (sucrose 121 water consumption/total water consumption] × 100%). 122

Forced swimming test (FST) 123
The FST was performed as previously described (Zhao et al. 2021). The experiment was carried 124 out within 2 days. On day 1, mice were individually placed in a swimming tank (50-cm height, 20-125 cm diameter) filled with water (temperature 25 ℃ ± 1 ℃) to a depth of 35 cm for 10 min. On day 126 2, the mice were individually placed in a swimming tank for 6 min, and the immobility time was 127 recorded during the last 5 mins by experimenters blinded to the design. 128

Tail suspending test (TST) 129
In the TST, each mouse was suspended individually by its tail's tip at a height of 50 cm above the 130 floor. The test was performed for a period of 5 min, and the total duration of immobility was scored 131 by an observer blinded to the experimental conditions. 132

Isolation of hippocampal mitochondria 133
Mitochondria in the hippocampus blocks were isolated using the tissue mitochondria isolation 134 kit according to the instructions. All centrifugation procedures were performed at 4 °C. Briefly, 135 fresh tissues were homogenized using a handheld Tissue-Tearor homogenizers using an isolation 136 buffer (1:10, w/v). The tissue homogenizations were centrifuged at 1000 × g for 5 min. 137 Supernatants were transferred to separate tubes and centrifuged at 11000 × g for 10 min. The 138 sediment consisted of mitochondria. The sediments were resuspended with appropriate amounts of 139 mitochondrial stock solution and the protein concentrations were measured by BCA assay kit.

Statistical analysis 182
Results are expressed as the mean ± standard error of the mean (SEM). Statistical analysis was 183 determined by one-way analysis of variance (ANOVA) followed by Tukey's test when analyzing 184 more than two groups by the GraphPad Prism software version 8.0 (GraphPad Software, Inc., La 185 Jolla, CA, USA). p < 0.05 was considered statistically significant. 186

FGF21 alleviated CUMS-induced depressive-like behaviors 188
To explore antidepressant effects of FGF21, mice were subjected to the CUMS stimuli for 63 189 days and treated with FGF21 for 35 days (Fig. 1A). At the end of treatments, behavioral experiments 190 were performed. Results in the SPT (Fig. 1B) showed an overall significantly reduced preference 191 for sucrose in CUMS-induced mice when compared to that of control mice, which was partially 192 reversed by FGF21 treatment (p < 0.01). 193 In the FST (Fig. 1C), a significant increase in immobility time was found in CUMS-induced mice 194 compared to that of control mice (p < 0.01), which was dramatically reversed by administration of 195 FGF21 (p < 0.01). The administration of FGF21 markedly reversed CUMS-induced depressive-like 196 behavior in mice. 197 In the TST (Fig. 1D), CUMS-induced mice exhibited significantly longer immobility time 198 compared with those of control group (p < 0.01), which was significantly reversed by FGF21 199 administration (p < 0.01, Fig.1D). The pathogenesis of depression is strongly related to mitochondrial dysfunction in the 210 hippocampus. To assess the effects of FGF21 on mitochondrial function in the hippocampus, brain 211 mitochondrial ATP production, mitochondrial membrane potential changes and mitochondrial ROS 212 production were measured. Mitochondrial ATP production and mitochondrial membrane potential 213 were decreased markedly ( Fig. 2A, B), whereas ROS production was significantly increased in the 214 hippocampus obtained from the CUMS group mice compared with those from the control group 215 mice (Fig. 2C). FGF21 treatment remarkably reversed the declining mitochondrial ATP production 216 and membrane potential ( Fig. 2A, B, p < 0.05 vs CUMS group), and decreased mitochondrial ROS 217 production (Fig. 2C, p < 0.01 vs CUMS group) in the hippocampus of the CUMS-induced 218 depressive mice. These findings suggested that FGF21 treatment led to restored brain mitochondrial 219 function in CUMS-induced depressive mice. Mitochondrial dysfunction disrupts respiratory chain function and accelerates ROS production. 228 The changes of oxidative stress related parameters in the hippocampus of mice were detected. 229 Results showed that SOD, CAT and GSH-Px activities were markedly declined in the hippocampus 230 of CUMS mice, whereas FGF21 treatment restored these changes (p < 0.05, Fig. 3A-C). The content 231 of MDA in the hippocampus was markedly elevated in the hippocampus of CUMS mice. After 232 treatment with FGF21, MDA levels decreased significantly compared to that of CUMS group (p < 233 0.01, Fig. 3D). 234  Fig. 4A and B), and upregulated DRP-1 expression (p < 0.01, Fig. 4A and C) in the 245 hippocampus compared with that in the control group. Administration with FGF21 reversed these 246 changes dramatically (p < 0.05, Fig. 4A-C). The results suggested that FGF21 treatment restored 247 mitochondrial dynamics imbalance induced by CUMS in mice. The results showed that CUMS treatment significantly decreased the protein expression of Nrf2 and 259 HO-1 in hippocampal tissues of mice compared with the control group (p < 0.01, Fig. 5). 260 Administration with FGF21 ameliorated these changes dramatically (p < 0.05, Fig. 5). 261 study, we confirmed that CUMS treatment in mice induced hippocampal mitochondrial dysfunction 307 that was reflected by decreased ATP production and mitochondrial membrane potential as well as 308 increased mitochondrial ROS production. These changes were reversed by FGF21 treatment. The 309 results suggest that modifying of mitochondrial dysfunction plays a pivotal role in effects of FGF21 310 on CUMS-induced depressive behavior. 311 Oxidative stress refers to the imbalance between production of ROS and antioxidants. functions such as OXPHOS and membrane polarity, which increase oxidative stress and apoptosis, 317 may precede the development of depressive symptoms. Here, we found that SOD, CAT and GSH-Px 318 activities were decreased, and MDA content was increased, in the hippocampus of mice exposed to 319 CUMS. These changes were reversed by FGF21 treatment. Growing evidence shows that AMPK and SirT-1 enhance mitochondrial biogenesis and oxidation 347 by regulating PGC-1α, and prevent mitochondrial dysfunction. AMPK, a serine/threonine protein 348 kinase, is a pivotal protein in regulating mitochondrial biogenesis and oxidation (Herzig and Shaw 349 2018). AMPK can indirectly activate SirT1 by regulating the level of NAD + . SirT1, an NAD + -dependent protein histone deacetylase, has been shown to be involved in many physiological 351 processes in brain, including control of gene expression, metabolism, senescence, neurogenesis and 352 synaptic plasticity. SirT1 activation reportedly improves mitochondrial function and attenuates brain 353 injury by increasing the level of PGC-1α (Lu et al. 2018). Pharmacological intervene of the 354 AMPK/SirT1/PGC-1α signaling pathway may be of benefit in neuroprotection by restoration of 355 mitochondria dysfunction and reducing mitochondria-mediated oxidative stress (Yang et al. 2020). 356 In the present study, we found that CUMS treatment decreased expression of p-AMPK, SirT1 and 357 PGC-1α. After treatment with FGF21, expression levels of p-AMPK, SirT1 and PGC-1α increased, 358 which demonstrated that FGF21 exerts anti-depressive effect through enhancing AMPK/SirT-359 1/PGC-1α signal pathway to restore mitochondrial dysfunction. 360 In conclusion, this study shows that FGF21 reverses CUMS-induced depressive-like behaviors. 361 The anti-depressive effects of FGF21 may be realized through enhancing of Nrf2/HO-1 signaling 362 and AMPK/SirT-1/PGC-1α signaling to restore hippocampal mitochondrial dysfunction partly. The 363 study advances a novel understanding and the underlying mechanisms of FGF21 in depression and 364 provides a theoretical foundation for the use of FGF21 in depression. Further work will be needed 365 to understand its specific molecular mechanism of action and clinical applications.  Declarations: The article is original, has been written by the stated authors who are all aware of its 376 content and approve its submission, has not been published previously, it is not under consideration 377 for publication elsewhere, in whole or in part. The authors declare that there are no financial or other 378 relationships that might lead to a conflict of interest of the present article.