1 Introduction

Major depressive disorder (MDD) is a mental disease that has become the major cause of disability worldwide [1, 2]. Symptoms of MDD often include loss of appetite and weight loss [3, 4], and these are also observed in some animal models [5]. However, the relationship between body weight and MDD is complex and multifaceted. While rodents often overeat due to acute psychological stress [6], chronic psychological stress reduces food intake, body weight, and adiposity in rodents [7, 8]. Melancholic depression is characterized by intense loss of appetite and weight, whereas atypical depression is the opposite, characterized by increased appetite and weight [9]. A systematic review and meta-analysis found that both underweight and obesity increase the risk of MDD [10]. Another study indicated that fat mass is a risk factor for MDD [11].

In recent years, autophagy has been the focus of much attention in studies of both body weight and MDD [12-14]. Autophagy is a vital intracellular pathway for cellular homeostasis through the lysosomal degradation of damaged macromolecules and organelles [15]. Neuronal cells are particularly vulnerable to autophagy deficiency [16]. Thus, autophagy has been linked to several psychiatric diseases including MDD [12, 16]. Previous studies showed increased expression of autophagy-related genes in mononuclear cells in patients with MDD [17]. In animal models, chronic mild unpredictable stress in mice has been reported to enhance hippocampal autophagy [18]. Antidepressants have been reported to induce autophagy, suggesting that autophagy may play an important role in the pathology and treatment of MDD [19, 20]. The pharmacological inhibition of Beclin1, an autophagy marker, abrogated the beneficial cellular and behavioral effects of antidepressants in mice [21].

In addition, it has been shown that autophagy is related to body weight regulation and eating behavior [13, 14, 22]. Fasting or calorie restriction can cause adaptive autophagy and prolong the longevity of eukaryotic cells [22]. Dysfunctional autophagy in the hypothalamus has been linked to obesity and related metabolic complications [13]. Deletion of FK-506 binding protein 51 (Fkbp51), the regulator protein of autophagy, in the mouse hypothalamus was a strong inducer of obesity, whereas its overexpression prevented high-fat diet-induced obesity [23]. Autophagy promotes an organism's resilience to weight gain caused by high-calorie diets [24].

Although weight loss is a well-known phenomenon as a symptom of psychological distress and MDD, its mechanism has not been clarified. This study aimed to determine how stress-induced weight loss altered autophagy in the hippocampus and midbrain of depression model mice. In this study, mice underwent a chronic immobilization stress (CIS) protocol for 14 days. Mice whose immobilization time was significantly longer in the TST than in the control group were considered depression model mice. Weight changes before and after the CIS protocol were compared between control and depression model mice. The expression of autophagy-related genes (Atg5, Atg7, Atg12, Becn1, Mmp9, Fkbp5, and Map1lc3b), which are reported to be upregulated in the brains of MDD patients [25-27], were examined in the hippocampus and midbrain of control and depression model mice. Furthermore, the correlations of the gene expressions with TST immobilization time and body weight change were investigated.

2 Materials and Methods

3 Results

3.1 CIS Induced Depressive-Like Behavior and Weight Loss

CIS response mice (n = 7) with longer than the average immobilization time in the control group mice were regarded as depression model mice. The data of TST immobility time and weight are shown in Figure 1 and Table S1. TST immobility time differed significantly between the control and CIS response mice (Student's t-test p value < 0.001; Figure 1A and Table S1). Although the weight of mice before the CIS protocol was not different between control and CIS response group mice (Student's t-test p value = 0.304, Table S1), the weight after the CIS protocol was significantly lower in the CIS response group mice than in the control group mice (Mann–Whitney U test p value < 0.001, Table S1). The weight change ratio in the CIS response group was downregulated compared with the control group (Student's t-test p value < 0.001; Figure 1B and Table S1). There was no significant difference in plasma cortisol levels between the CIS response group and the control group (Student's t-test p value = 0.111; Figure 1C and Table S1).

3.2 CIS Changed Autophagy-Related Gene Expressions in the Hippocampus

Fkbp5 expression levels were higher in the CIS response mice than in the control group mice (Student's t-test p value < 0.001; Figure 2A and Table 1). Mmp9 expression levels were higher in CIS response mice than in control group mice (Student's t-test p value = 0.048; Figure 2B and Table 1). Map1lc3b expression levels were significantly higher in CIS response mice than in control group mice (Student's t-test, p value = 0.008; Figure 2C and Table 1). These increased expression gene changes (Fkbp5, Mmp9, and Map1lc3b) in the CIS response group were consistent with the results in postmortem brains of MDD patients [25-27]. Atg5, Atg7, Atg12, and Becn1 expression levels were not different between the two groups (Table 1).

TABLE 1. Expression levels of autophagy-related genes in hippocampus.

mRNA expression	Control (n = 9)	CIS response (n = 7)	p
Fkbp5	1.00 ± 0.18	1.40 ± 0.08	< 0.001*
Mmp9	1.00 ± 0.19	1.22 ± 0.19	0.048*
Atg5	1.00 ± 0.06	1.05 ± 0.06	0.133
Atg7	1.00 ± 0.07	1.06 ± 0.05	0.114
Atg12	1.00 ± 0.05	0.97 ± 0.07	0.288
Becn1	1.00 ± 0.10	1.03 ± 0.06	0.511
Map1lc3b	1.00 ± 0.11	1.14 ± 0.04	0.008*

• Note: Values are means ± standard deviation.
• Abbreviation: CIS, chronic immobilization stress.
• ^*p value < 0.05, Student's t-test or Mann–Whitney U test.

3.3 CIS Changed Autophagy-Related Gene Expressions in the Midbrain

Fkbp5 expression levels were higher in the CIS response mice than in the control group mice (Student's t-test p value = 0.016; Figure 3A and Table 2). Mmp9 expression levels in CIS response mice were higher than in control group mice (Student's t-test p value = 0.039; Figure 3B and Table 2). These increased expression gene changes (Fkbp5 and Mmp9) in the CIS response group were consistent with the results in the hippocampus and the results in the postmortem brains of MDD patients [25, 27]. Map1lc3b, Atg5, Atg7, Atg12, and Becn1 expression levels were not different between the two groups (Table 2).

TABLE 2. Expression levels of autophagy-related genes in midbrain.

mRNA expression	Control (n = 9)	CIS response (n = 7)	p
Fkbp5	1.00 ± 0.16	1.28 ± 0.24	0.016*
Mmp9	1.00 ± 0.22	1.27 ± 0.26	0.039*
Atg5	1.00 ± 0.08	0.96 ± 0.05	0.321
Atg7	1.00 ± 0.07	0.99 ± 0.08	0.725
Atg12	1.00 ± 0.07	0.93 ± 0.09	0.111
Becn1	1.00 ± 0.07	1.00 ± 0.07	0.904
Map1lc3b	1.00 ± 0.06	1.05 ± 0.06	0.108

• Note: Values are means ± standard deviation.
• Abbreviation: CIS, chronic immobilization stress.
• ^*p value < 0.05, Student's t-test.

3.4 Negative Correlation Between Fkbp5 mRNA Expression and the Weight Change Ratio

The weight change ratio was significantly negatively correlated with Fkbp5 mRNA expression in the hippocampus of CIS response group mice (Pearson's correlation analysis: r = −0.766, p = 0.045; Figure 4). TST immobility time and Fkbp5 mRNA expression were not significantly correlated in CIS response group mice (Pearson's correlation analysis: r = 0.054, p = 0.908). Other autophagy-related gene expressions in the hippocampus did not show any significant correlations with the weight change ratio or TST immobility time. In the midbrain, none of the autophagy-related gene expressions showed significant correlations with the weight change ratio or TST immobility time.

4 Discussion

In recent years, autophagy has been highlighted in the study of MDD [12, 16]. There are substantially consistent results that autophagy markers are increased in MDD patients [17, 26]. Autophagy is a beneficial process of the stress response, but under certain conditions, excessive and chronic activation can be harmful [34, 35]. For example, exaggerated autophagy causes type II cell death [36]. Short-term caloric restriction, one of the most effective triggers of autophagy, has been reported to have antidepressant effects in humans and mice [37, 38], whereas the effects of long-term caloric restriction are detrimental [38]. It has been suggested that acute stress activates lytic autophagy, which has an initial beneficial effect by increasing neuroplasticity and cognitive alertness, while prolonged or chronic stress may induce neuroinflammation [35]. In addition, a previous report suggested that excessive autophagy reduces BDNF expression in neurons and disrupts hippocampal neurogenesis in corticosterone-induced depression model mice [39].

Fkbp51, which is encoded by the Fkbp5 gene, negatively regulates the HPA axis by inhibiting the interaction between GR and glucocorticoid hormone [40]. A previous study suggested that the inhibition of Fkbp51 by SAFit2, the selective antagonist of FKBP51, promotes stress resilience to chronic psychological stress in mouse models and might be a novel antidepressant [41]. Fkbp51 is also known to be an important upregulator protein of autophagy [42, 43]. It has been shown that upregulation of FKBP51 mediated glucocorticoid-induced autophagy [42]. FKBP51 induces autophagy and exacerbates damage in cerebral ischemic stroke [43]. A previous study suggested that the deletion of Fkbp51 in the mouse hypothalamus strongly induced obesity by decreasing autophagy, whereas its overexpression prevented obesity by increasing autophagy [23].

Microtubule-associated protein 1A/1B light chain 3B (Map1lc3b) is a major protein in the autophagy pathway and is most widely known as a marker of autophagosomes [44]. A previous report suggested that Map1lc3b mRNA expression in the brain was higher in MDD patients than in healthy controls [26]. Another report found that Map1lc3b mRNA expression was increased in the peripheral blood mononuclear cells of patients with MDD [17].

Matrix metallopeptidase 9 (Mmp9) protein is a matrixin involved in the degradation of the extracellular matrix in normal physiological processes [45] and is essential for autophagy [46]. A recent study showed that Mmp9 protein was enhanced by stress-induced secretory autophagy [35]. It has also been thought that Mmp9 might also be involved in the pathogenesis of MDD [25, 47, 48]. A previous study has shown that Mmp9 mRNA expression and protein levels were upregulated in the blood of MDD patients [47]. Another study reported that increased MMP9 enzymatic activity was observed in the hippocampus of postmortem brain samples from MDD patients [48]. Mmp9 protein levels were increased in the postmortem brains of MDD patients treated with antidepressants compared with controls [25].

In the present study, Fkbp5, MMP9, and Map1lc3b mRNA expressions were upregulated in the depression model mice compared with the control mice. These results were consistent with the results seen in postmortem brains of MDD patients [25-27]. These results also coincide with the hypothesis that chronic and excess autophagy activation is harmful and leads to psychiatric disease [34-36, 39]. Although it is well-studied that antidepressants also increase autophagy, it is thought that starting a process of neuronal reorganization that ultimately constitutes the transition from disease to health [49]. These differences in the effects of autophagy may be due to the duration and pathway [35, 49], but the mechanism remains unclear.

On the other hand, although the gene expressions of initial autophagy signaling (Atg5, Atg7, Becn1, and Atg12) in CIS mice were not significantly upregulated in the present study, most of these genes showed a tendency to increase compared to control mice. These could be explained by type II errors due to the small sample size or could indicate an abnormality in the autophagy pathway of depression and therefore require further analysis.

In the present study, Fkbp5 mRNA expression was correlated with stress-stimulated weight loss in depression model mice. Considering that Fkbp5 is a regulator of body weight via autophagy [23], this result suggests that the increase in Fkbp5 mRNA expression due to chronic psychological stress may cause weight loss through the promotion of autophagy. As the central brain region for metabolic and weight control, the hypothalamus has been well-studied [13, 23, 50, 51]. On the other hand, the hippocampus and midbrain are well-known as the central region for MDD [18, 52-54] and are also thought to be involved in weight and metabolic control nowadays [55-57]. Therefore, we chose the hippocampus and midbrain as the research region for the connection between weight loss and depression. Several studies have implicated the autophagy pathway in the hypothalamus involved in weight regulation [14, 23, 50] and it has been suggested that the dysregulation of autophagy in the hypothalamus may influence body weight through impaired leptin signaling [58], glucose intolerance [59], and impaired sympathetic control of the peripheral metabolic organs [60]. It is possible that autophagy in the hippocampus and midbrain affects body weight by mechanisms similar to those in the hypothalamus. To elucidate these mechanisms, further research is needed.

In our present study, there was no significant difference in plasma cortisol levels between the CIS response group and the control group. Based on the previous study [31], we assumed that plasma cortisol would remain at a high level during CIS, but the actual result was different. This result could be attributed to the short duration of CIS in our study. In the previous study [31], the plasma cortisol levels decreased once after 2 weeks of CIS and increased again after 3 weeks of CIS. There may have been a significant difference in plasma cortisol levels between the CIS response group and the control group if 3 weeks of CIS had been performed. The CIS protocols are often conducted for 2–4 weeks [61]. In this study, we chose the shortest duration of 2 weeks due to ethical considerations for the mice. Future studies will need to consider the duration of CIS.

The present study has some limitations. First, the small sample size may lead to type II errors. Second, it is difficult to determine the function of autophagy by autophagy-related gene mRNA expressions, since assessing autophagic flux or turnover of long-lived proteins is complicated in mice [44, 49]. Third, although Fkbp5 mRNA expression was correlated with stress-stimulated weight loss in depression model mice, it does not allow us to postulate a direct causal relationship between Fkbp5 and weight loss because decreased appetite and food intake may be confounding factors. Due to technical limitations, the present study could not assess changes in food intake, body temperature, and activity of the mice. Further studies are needed to clarify the correlation between autophagy and stress-induced weight loss.

5 Conclusions

The results of the present study show that the expressions of autophagy-related genes (Fkbp5, Mmp9, and Map1lc3b) were increased in the hippocampus of a mouse model of depression that exhibited weight loss. The increased gene expression in the mouse model of depression was consistent with that observed in the postmortem brains of MDD patients. In particular, Fkbp5 mRNA expression was correlated with stress-induced weight loss in the depression model mice. These results suggest that enhanced autophagy may be related to the pathogenesis of MDD and that Fkbp5 mediates stress-induced weight loss.

Author Contributions

Conceptualization: H.M. and J.-i.I. Methodology: H.M., Y.Y., and J.-i.I. Investigation: H.M., M.O., Y.F., H.K., Y.Y., and S.O. Writing – original draft preparation: H.M. Writing – review and editing: Y.Y., S.-i.U., J.-i.I., and H.M. Supervision: S.-i.U. Funding acquisition: S.-i.U. and J.-i.I. All the authors contributed to and approved the final manuscript.

Ethics Statement

Approval of the research protocol by an Institutional Reviewer Board: All animal experiments were approved by the Animal Experiment Committee of Ehime University (Approval number; 05TU231) and were performed under the Guidelines for Animal Experiments at Ehime University.

Animal Studies: This study complies with to Act on Welfare and Management of Animals and Ministry of Education, Culture, Sports, Science and Technology (MEXT)'s Fundamental Guidelines for Proper Conduct of Animal Experiment and Related Activities in Academic Research Institutions.

Consent

The authors have nothing to report.

Conflicts of Interest

The authors declare no conflicts of interest.