Dynamic changes of serotonin levels induces depression-like behaviour in mice with high-fat diet

Introduction of SSRI as the rst-line pharmacotherapy options for depression, which purpose is to increase the concentrations of 5-HT in brain and help reduce depressive symptoms. However, there are still many limitations to the effectiveness of these treatment strategies, 30-40% of patients do not respond to treatment at all, and even suicide occurs. In the present study, we revealed that increasing brain 5-HT concentration is not the best treatment options. When we injected sucient amounts of 5-HT into the lateral ventricular region of mice, it actually caused depression-like behavior in the animals. The second study on high-fat fed induces depressive symptoms in mice demonstrated that 5-HT concentration in brain showed a tendency to increase rst and then decrease. Both experiments indicated that if SSRI are administrated to the patient in the acute phases of depression, it may lead to exacerbation of mental disorders or suicidal tendencies. a high-fat may to levels 5-HT injection the


Introduction
Major depression disorder (MDD) is a highly prevalent and debilitating mental illness that has a signi cant impact on more than 16% of the global population and is one of the major causes of the severe global health and socio-economic burden 1 . In middle-income countries (including China)accounts for an estimated three-quarters of the global burden these conditions. MDD patients have several clinical symptoms: persistent dark emotions and other problems, such as suicidal thoughts, sleep disorders, altered appetite, and cognitive impairment 2,3 . The complexity of the 5-HT system and the role of other neurochemical systems suggest that depression may be caused by a variety of pathological mechanisms, and the common denominator is undoubtedly a conduction disorder in the central 5-HT synapse 4 . Fourteen different serotonin receptor have been identi ed, and these receptors are subdivided into different families, labeled 1, 2, 3, 4, 5, 6, and 7. The subtypes in each family are marked with letters (for example, a, b, c). Some receptors are thought to be involved in the pathogenesis of various central nervous system disease 5 . Among these functional receptors, the serotonin receptor 5-HT 1A (5-HT 1A R) seems to play a key role in neuropathology depression 6, 7 . 5-HT 1A R is involved in the regulation of depression and anxiety 8 .5-HT 1A R de cient mice have been shown to exhibit increased anxiety-related responses, fear of classical conditioning, and increased freezing behavior 9 . Reduced 5-HT 1A R levels have been shown to cause anxiety and stress disorders in mice and primates 10,11 . In addition, studies have shown that lower 5-HTT(5-hydroxytryptamine transporter) expression in humans is associated with higher levels of generalized anxiety and depression 12,13 . Rodent studies have shown that 5-HTT knockout (Ko) rodents exhibit more anxiety-and depression-like behavior 14 .Based on this theory, selective 5-HT reuptake inhibitors (SSRIs) are widely prescribed as antidepressants, and becoming a rst-choice treatment for depression. Patients with depression usually require long-term treatment with antidepressant drugs and those with major depression often require polytherapy. However, most antidepressant drugs are far from ideal and they often possess undesirable side effects while lacking the ability to manage these conditions in the long term. For example, Selective 5-HT reuptake inhibitors (SSRIs), including paroxetine or sertraline are some well-known limitations of antidepressant drugs, and there are has been concern that used antidepressant drugs might be associated with an increased risk of suicidal ideation in pediatric patients 15 . Up to now, the cause of depression is not very clear, but it can be sure that some factors such as biology, psychology and social environment, participate in the process of depression. However, little is known about the role of dietary pattern in the development of depressive disorders. Several cross-sectional studies found a signi cantly lower prevalence of depression symptoms for persons with a low consumption of fast-food (hamburgers, sausages, pizza) 16 . Similarly, depressive symptoms were associated positively with consumption of fast-food, snacks and high-energy sweets 17 .Moreover, a recent analysis showed that a food pattern rich in processed or fried foods, re ned grains, sugary products and high-fat dairy products was signi cantly related to perceived stress 18 . In spite of many studies show a potentially detrimental association between the consumption of high-fat diet and the risk of depression. But, people would rather choose fast-food than 'healthy' dietary pattern, such as the Mediterranean diet, because they think consumption of fast-food is a lifestyle, especially there is signi cant proportion among teenagers who hold this view. While the determining factors of depression are complex, the mechanism by which a high-fat diet causes depression is not yet understood.Due to the fact that multiple studies have previously shown that high daily intake of saturated fats and re ned carbohydrates often leads to depression and cognitive impairment, and it has been linked to impair the serotonin system, which includes decreased tryptophan concentration, impaired 5-HT synthesis or release, and postsynaptic 5-HT receptor dysfunction 19,20,21 . Collectively, these ndings emphasize the existence of common molecular pathway of serotonin system between both high-fat diet and major depression. However, there is little research that explicitly demostrates whether the decrease of 5-HT level in the brain is not the only cause of depression-like behavior induced by high-fat diet. In this study, the authors rst intervened in mice with diet to induce the occurrence of depression-like behavior, determine the changes in 5-HT levels in the brain, and then set the brain stereotactic injection group to con rm that the excessive 5-HT content in the brain may also be another cause of depression-like behavior in mice. In addition to the behavioral assessment, the expression and other biochemical indicators of 5-HT-associated proteins and genes were also examined in each group. Research on epigenetics has made great strides over the past few decades 22 .Epigenetics is de ned as the stably heritable changes to the regulation of gene expression through modi cations in chromatin structure and DNA methylation without altering DNA sequences 23 . Epigenetic modi cations, including DNA methylation, histone modi cation, positioning of histone variants and gene regulation of miRNA, are potentially reversible. DNA methylation, one of the most important epigenetic modi cations, has been proofedproven to be associated with depression in many studies 24,25,26 .For this purpose, we evaluated the methylation levels of 5-HTT and 5-HT 1A in promoter region CPG islands in diet intervention and brain stereotaxic injection groups were evaluated in the study.

Results
Increases in bodyweight by HFD Exposure.To study the effects of metabolic disorder on depressive-like behaviours, mice were fed HFD for 18 weeks. Behavioral tests were performed on mice at weeks 3, 5, 11 and 17 (Fig.1a) and biochemical indicators were determined after testing.The timeline is shown in Fig.1a.
As expected, the HFD diet progressively increased body weight over the weeks (Fig.1b).
Promotes of depression behaviors by HFD exposure. HFD exposure promotes depression behaviors as demonstrated by behavioral tests. HFD-fed mice were tested for depressive-like symptoms using behavioural tests. There was a signi cant difference in behavior between the STD group and the high-fat group at the beginning of 5 weeks. In the OFT, 5 and 11 weeks of an HFD signi cantly decreased the time spent in this compartment. In FST and TST ( Fig.2c and 2d), as compared to STD groups ,the immobility time of HFD group increased at 5 and 11 weeks, respectively. At the 17th week, the sugar and water preference of the high fat group was signi cantly lower than that of the STD group (Fig.2a). In the open eld test (Fig. 2b) the percentage of central residence decreased , at the same time, the time of immobility in the tail suspension test (Fig.2c) and forced swimming test (Fig.2d) was signi cantly increased. At the behavioral level, mice showed signs of depression.
Dynamic changes of 5-HT level in the brain of HFD exposed group.The monoamine neurotransmitter 5-HT is believed to be directly or indirectly involved in regulating one's mood, and its dysfunction is related to the pathophysiology of depression. Studies have shown that the occurrence of depression-like behavior is caused by a reduction in 5-HT content in the brain. Interestingly, the debase changes was not observed in 5-HT levels in the HFD groups (Fig.3). The 5-HT levels of hippocampus from each group were detected. 5-HT levels of the HFD group were a dynamic change process as compared to the STD group. Levels of 5-HT signi cantly increased at 5 and 11 weeks and then decreased at 17th week in the HFD mice.
Brain injection of 5-HT leading to the occurrence of depression-like behavior.To further understand the relationship between 5-HT and the development of depressive-like behaviours, another model of stereotactic Infusions was used. In behavioral test experiments, depression-like behavior was found in the model group as compared to the control group. In the OFT (Fig.4a), there was decrease in time spent in the centre as compared to the control group and depression-like behavior . The TST (Fig.4b) and FST ( Fig.4c) were used to further determine the depression level in mice; the immobility time signi cantly increased in both tests in the model mice. The mice in the model group showed depressive-like symptoms.
Brain hippocampal neurons damage in high-fat diet and 5-HT mice.When compared with the normal group, the mice in the HFD group showed signi cant difference in some behavioral indexes in the 5th week. Therefore, the degree of hippocampal neuron injury was observed after 5, 11 and 17 weeks of ethology in the diet intervention group and brain-targeted injection group. Under the light microscope, the DG neurons in the STD group were arranged regularly, the cell layer was rich, the cell membrane and cytoplasm were intact, and the nucleus was round and large. In the HFD group, the number of hippocampal neurons decreased, the structure was disordered, the neurons shrank, the cell morphology was irregular, the intercellular space was enlarged, the nucleolus was not obvious, and some neurons were vacuolated, especially at 17 weeks ( Fig.5.a). In the stereotactic injection experiment, when compared with the Control Group, the number of hippocampal neurons in the Model group was signi cantly decreased, the neurons were disordered, the neurons were atrophied, the shapes were irregular, the intercellular spaces were enlarged and some neurons were vacuolated ( Fig.5.b).
High-fat diet leading to depression-like behavior because 5-HT pathway is destroyed in the brain.In order to determine whether the depression-like behavior induced by HFD in mice is due to the in uence of 5-HT pathway-related genes and proteins, 5-HTT and 5-HT 1A were measured and it was found that the expression of 5-HTT and 5-HT 1A was dynamic. At the level of gene and protein, the expression level of mRNA ( Fig. 6a) and protein (Fig. 6b) in the HFD group increased signi cantly in the 5th week, and then decreased in the 11th week until the 17th week, there was a signi cant difference.
An overdose of 5-HT in the brain leading to damage of the 5-HT nerves.In order to further investigate whether the single decrease of 5-HT in the brain is not the main cause of depression-like behavior in mice, the present study rst measured the level of 5-HT in the brain of the stereotactic injection group (Fig6.2a) and found that the level of 5-HT in the brain of the model group was higher than that of the control group. At the same time, the expression of 5-HTT and 5-HT 1A in the stereotactic brain injection group was determined at the gene and protein levels. The results showed (Fig6.2b and Fig6.2c) that when compared with the control group, the mRNA and protein expressions of 5-HTT and 5-HT 1A in the model group decreased with statistical signi cance, indicating that stereotaxic injection could damage 5-HT neural energy.
Higher methylation levels in the 5-HT-related gene promoter region resulting in abnormal protein expression.In order to further study the relationship between 5-HT related protein expression and epigenetics, DNA methylation was detected on the CPG islands of the promoter region of the 5-HT related gene. In the diet intervention experiment, as compared to the STD group(Fig7.a), the mean methylation level of 5-HT 1A and 5-HTT promoter regions in HFD mice at 11 weeks was higher than that of the STD mice and the same result was obtained at 17 weeks. It was hypothesized that the elevated mean methylation level in the promoter regions of both genes might be due to elevated 5-HT levels in the brain. Thus, the level of methylation in these two gene promoter regions in the stereotaxic mice was also measured, and it was found that the promoter region of 5-HTT gene in model group was hypermethylated, as compared to the control mice(Fig7.b).

Discussion
Effective management of obesity and its associated emotional complications is a challenging task unless underlying neurobiological mechanisms are fully understood 32,33 . In this study, mice were fed either a high-fat diet or a normal diet for 17 weeks, and since the serotonin system is a major factor in mood regulation 34 , therefore, at 3, 5, 11, 17 weeks, ethology and 5-HT neuroenergy analysis were carried out to determine the relationship between the high-fat diet and the serotonin system. Recent studies have shown that some 5-HT receptor subtypes, of which 5-HT 1A , play an important role in anxiety and depression relief 34 . Glucose preference test, open eld test, forced swimming, tail suspension test and pathological observation were used to determine depression-like features and hippocampal injury, and 5-HT content in brain was assessed by ELISA Kit. The results suggest that a high-fat diet leads to depressive-like behavior and damage to hippocampal neurons, which may be due to more than a decrease in 5-HT in the brain. It is a dynamic process of rising and then falling. To further investigate whether depression-like presentation in mice on a high-fat diet may be due to elevated levels of 5-HT in the presenile brain, a brain stereotactic injection group was set up, and the data demonstrated that the depression-like appearance in the model group showed that the hippocampal neurons were damaged, when compared with control mice. Interestingly, these results are different from the 5-HT theory 35 proposed by earlier scholars, and it allow us to speculate that 5-HT levels in patients is not reduced but substantially increased in the early stages of depression. Elevated serotonin levels makes postsynaptic 5-HT receptor 5-HT 1A insensitive. At the same time, the continuous increase of serotonin concentration in synaptic cleft will enhance of 5-HTT transport rate and elicits a dose-dependent increase of serotonin uptake or recycles in presynaptic membrane, which eventually triggers a negative feedback mechanism that leads to the decrease of serotonin secretion. This explains the reason why the patient with depression take the 5-HT inhibitors in the early stage of depression can increase the amount of 5-HT in the synaptic cleft and further damage the 5-HTT, leading to the suicidal tendency of patients.
It is now recognized that there is a physiological and pharmacological relationship between the 5-HT 1A receptor and 5-HTT in the regulation of 5-HT neurotransmission (31) .Membrane serotonin transporter (5-HTT) is the main regulator of 5-HT neurotransmission 36 in the brain and controls the reuptake of 5-HT in the synaptic cleft. Studies have shown that 5-HTT gene knockout leads to desensitization 37 and a decrease in the density and gene expression of 5-HT 1A receptors in the Dorsal raphe nucleus (31) . To further determine the relationship between the high-fat diet and the serotonin system, this study also measured the expression of 5-HT 1A R and 5-HTT genes and proteins in the brains of mice from each group. In the diet intervention group, when compared with the normal group, the expression of 5-HT 1A R and 5-HTT in the cerebral cortex of the high-fat group increased rst and then decreased in the whole feeding period. In combination with the previous high-fat group, the 5-HT level in the brain of mice showed a tendency to increase at rst and then decrease. In order to maintain the homeostasis of 5-HT, it was transported back to the vesicles or decomposed, and 5-HTT expression was increased. However, the 5-HT 1A receptor in the postsynaptic membrane was stimulated by 5-HT, and the 5-HT 1A receptor expression was increased. At 11 weeks, 5-HT levels were still high in the brains of mice that were still on a high-fat diet. It is the long-term excessive 5-HT stimulation that makes 5-HTT and 5-HT 1A to be activated all the time, leading to desensitization, resulting in the decrease of 5-HTT and 5-HT 1A expression and the destruction of the 5-HT nervous energy system in the brain. There is abundant evidence that the 5-HT 1A receptor plays an important role in the regulation of 5-HT release 36, 38, 39 . Since the 5-HT nervous system was destroyed at 11 weeks, the 5-HT release decreased and the 5-HT level in the brain decreased in the 17 week high-fat group, and the expression of 5-HTT and 5-HT 1A receptor decreased. The data from the stereotactic experiment also showed that the expression of 5-HTT and 5-HT 1A R decreased in the gene and protein levels in the model group when compared with the control group.
A recent review of 67 studies concluded that there is evidence that DNA methylation variants are associated with depression risk 40 , when exploring the speci c causes of the decline in the expression of 5-HTT and 5-HT 1A R genes. In order to explore the relationship between the decrease of the front protein expression and the DNA methylation, the DNA methylation of 5-HTT and 5-HT 1A R in the cerebral cortex of each group of mice was determined. It was found that in the diet intervention group, the hypermethylation level of 5-HTT and 5-HT 1A in the brain of the hyperlipidemic group at the 11th and 17th week was in line with the trend of protein expression as compared with the normal group, it is suspected that the previous decline in the expression of 5-HTT and 5-HT 1A was due to the hypermethylation of their corresponding genes. In the brain stereotactic injection group, it is noteworthy that the level of DNA methylation of the gene corresponding to the down-regulation of 5-HTT expression was increased in the model group when compared with the control group, but the level of DNA methylation of the gene corresponding to the down-regulation of 5-HT 1A expression was not increased. It is hypothesized in the present study that this may be due to the fact that the high-fat diet is a chronic stimulus, but the brain-targeted injection is an acute stimulus, and the mice respond to acute and chronic stress differently. Of course, this result is open to further discussion.
In conclusion, the ndings of the current study suggest that the mechanism of depression-like behavior induced by high-fat diet in mice is not only due to the decrease of 5-HT levels in the brain, indeed, it may be due to a dynamic process in which the level of 5-HT in the brain increases rst and then decreases. At the same time, it is thought that long-term high level of 5-HT stimulation will lead to DNA methylation of 5-HT neuroenergy related genes, and then damage 5-HT neuroenergy. This study can explain the low e cacy of monoamine drugs to some extent, and provide a theoretical basis for the development of antidepressant drugs.

Materials And Methods
Animinals.The experiment was divided into two parts. The rst was divided into dietary intervention experiment. In order to prove that the depression-like characterization in mice in the dietary intervention experiment is not only due to the 5-HT level reduction in the brain, the brain stereotactic experiment was established. Speci c-pathogen-free C57BL/6J male mice (n = 70), approximately 3-4 weeks old.This study was approved by the Ethics Committee of Zhejiang University of Technology Animal Care. Male mice were housed under a 12-h light/dark cycle at 23 ± 2°C.After one week of adaptation, the C57BL/6J were randomly divided into Standard Diet Group (STD, n = 24) and high fat diet group (HFD, n = 24). The duration of dietary intervention was 18 weeks. In the brain stereotactic experiment , the C57BL/6J were randomly divided into Control group (saline injection, n = 10) and Model Group (5-HT injection, n = 10) for 14 days.In the dietary intervention experiment, body weight was monitored weekly and the degree of depression was assessed using behavioral tests and other biochemical indicators at 3, 5, 11, and 17 weeks. In the brain stereotactic experiment, the degree of depression was assessed using behavioral tests and other biochemical indicators.

Methods
Stereotactic Infusions.In this study, 5-hydroxytryptamine was injected into the lateral ventricular region of C57BL/6J mice. In brief, each mouse was immobilized in a stereotactic frame (RWD) after being anesthetized with 0.5% pentobarbital. The following coordinates were adopted: anteroposterior = -0.58mm, mediolateral = +1mm, dorsoventral = -2.2mm. The custom Cannula (RWD) was implanted into the lateral ventricle area of each mouse according to the coordinates, and the incision was sutured. After 7 days of recovery, 5-HT and 0.9% sodium chloride (0.9% NaCl) were injected into each mouse at a rate of 1 ml/min for 14 days, respectively.

Behavioral assays
Behavioral tests were conducted to assess the depression phenotype in the following sequence: sucrose preference test, open eld, tail suspension and forced swim test. During behavioral test, researchers kept quiet, removed each animal's excrement on time, and tested each group alternately.
Open eld test OFT .Since OFT was used for assessment of anxiety-like behavior by recording the time across the central square, OFT was carried out to assess anxiety-like behavior in this study. Each mouse was placed in the center of the open eld arena of white box (45 cm × 45 cm) and locomotor activity was recorded for 5 min 30s by an overhead camera. Mice showed higher levels of depression and less central residence time. After OFT of each mouse, alcohol was sprayed to remove the peculiar smell of the previous mouse and avoid affecting the next experiment results 27 .
Sucrose preference test SPT .Sucrose preference was used to measure depressive-like behavior, where decreased sucrose preference was considered as anhedonia 28,29 . In this test, the mice were fed in a single cage. Two bottles of 1% sucrose solution per cage were given to the test rodents for 24 h, next, two bottles, one bottle with 1% sucrose solution, and the other bottle with water fed for 24 h, and then deprived of water for 24 h. After 24 h, 1% sucrose solution and puri ed water were respectively placed in each cage, fed for 24 h, and the positions of the two bottles were changed every 12 h to prevent the potential in uence of position preference. During the experiment, the weight of the water bottle was weighed, and the formula for calculating sucrose preference (%) = sucrose intake/ (sucrose intake + water intake) × 100% .
Tail suspension test and forced swimming test (TST and FST).Depressive-like behavior was assessed by tail suspension and forced swimming tests 30,31 . Mice with increased depression showed increased immobility time.In the rst test, mice were suspended by the tail on a horizontal bar (at around 50 cm from the oor) with adhesive tape, and immobility time (in seconds) was recorded during the last 4 min of the 6 min test. Mice were considered motionless when they did not exhibit escape-oriented behavior and passively hung without any body movements. The forced swimming test, was performed using glass drums (20 cm diameter, 45 cm height). Before the test, the cylinders were lled with 15 cm of water (25± 1°C), and the test mice were individually placed in the cylinders during the last 4 min of the 6 min test. The immobility period was considered when animal oated in the water without behavior oriented to nd an exit or made only the necessary movements to keep his head out of the water. The time of immobility was recorded for 6 min. In this condition, mouse developed escape-orientated behaviors interspersed with temporally enhancement bouts of immobility.
Sample Collection and Preparation.After all the behavior tests, brains from each mouse were collected, the samples were transferred to EP tubes and stored at -80°C for later analysis.
CORT ELISA.The content of 5-HT in mice brain tissue was determined using an ELISA kit according to the instruction provided by Shanghai Meilian Biotechnology Co, Ltd (China).
Morphological observations.Mice were transcardially perfused with saline, then followed by a 4% neutral paraformaldehyde solution. The brain was post-xed for 24 h at 4°C and after that, they were embedded in para n and cut into 4 μm thick slices. Sections of cortex and hippocampus were stained with hematoxylin and eosin (HE) for histopathological evaluation.
Total RNA isolation and quantitative RT-PCR.All the kits used in this work were from Shanghai Biyuntian Biotechnology Co., Ltd. Total RNA was extracted and puri ed using an RNAeasy™ kit (Beyotime, Shanghai), following the manufacturer's protocol. The cDNA was ampli ed using the BeyoFast™ SYBR Greeline by Primer 3 Plus. The comparative threshold cycle (Ct) method was employed for quanti cation of transcripts according to the manufacture's protocol (Applied Biosystems). Measurement of ΔCt was performed at least three times per sample. Ampli cation of the single product was con rmed by monitoring the dissociation curve. A control gene (glyceraldehyde-3-phosphate dehydrogenase) (GAPDH) was used for normalization to control for possible uctuations of the target transcripts. The target genes were 5-HTT and 5-HT 1A (Table 1). Western blotting analysis.The total protein was prepared with brain tissue, 40 μg of protein was loaded into SDS-PAGE gels and transferred onto PVDF membranes. After blocked with 5% skim milk in TBST buffer for 1 h at RT, the membranes were subsequently incubated with primary antibodies at 4°C overnight. Primary antibodies used are listed as follows in the following: β-actin and 5-HTT purchased from protein (1:500); 5-HT 1A R obtained from abcam (1:1000). Then, it was rinsed with TBST buffer 3 times, incubated with Erkang at room temperature for 1 h then rinsed with TBST buffer 3 times and exposed to Ecl developer. Membranes were exposed by chemiluminescence developing agents. The intensity of the targeting bands was quanti ed using ImageJ

Declarations
Data availability Source data are provided as a Source Data.xlsx. Source data are provided with this paper.
Author contributions Weiguang Shan and Yan Chen contributed to the conception of the study; Lifang Chen performed the experiment; Tingting Gong contributed signi cantly to analysis and manuscript preparation; Yan Chen and Yongyong Zhang performed the data analyses and wrote the manuscript; Yunxia Fang and Shengfeng Wang helped perform the analysis with constructive discussions.