Salvianolic Acids Alleviate Chronic Mild Stress-Induced Depressive-Like Behaviors in Rats

Background : Salvianolic acids possess anti-inflammatory properties. This study investigated the therapeutic effect of salvianolic acids on chronic mild stress (CMS)-induced depressive-like behaviors in rats and the involvement of toll-like receptor 4 ( TLR4 ) and myeloid differentiation factor 88 ( MyD88 ). Methods : Eighty male Sprague-Dawley rats were randomly subjected to CMS or non-CMS protocol for 6 weeks. Starting 3 weeks after CMS exposure, the rats in each group were administered saline, fluoxetine (positive control), salvianolic acids, or salvianolic acids + fluoxetine daily for 3 weeks. The body weight change, sucrose preference, and immobility duration in forced swimming were examined before and after drug treatment. The rats were sacrificed at 3 weeks after drug treatment. Quantitative real-time PCR was performed to measure the mRNA levels of TLR4 and MyD88 in the prefrontal cortex and hippocampus of rats. Results : Compared with non-CMS rats, CMS rats had significantly reduced weight gains and sucrose preference, along with significantly increased immobility durations and elevated mRNA levels of TLR4 and MyD88 in both the prefrontal cortex and hippocampus. Treatment with fluoxetine and salvianolic acids, alone or in combination, facilitated weight gains, alleviated depressive-like behaviors, and reduced cerebral TLR4/MyD88 mRNA levels in CMS rats. Besides, fluoxetine and salvianolic acids additively suppressed TLR4/MyD88 mRNA expression in the prefrontal cortex of rats. Furthermore, TLR4 mRNA levels in both hippocampus and prefrontal cortex positively correlated with MyD88 mRNA expression, inflammatory cytokine secretion, and immobility duration but negatively correlated with sucrose preference. Conclusions : Thus, salvianolic acids alleviate depressive-like behaviors, possibly by suppressing TLR4/MyD88 -mediated inflammatory signaling in the brain.


Introduction
Depression is a mood disorder characterized by a persistent feeling of sadness and loss of interest, representing one of the leading causes of disability worldwide [1]. Major depressive disorder (MDD) is a highly prevalent type of depression that has been projected by WHO as the first cause of the burden of disease globally by 2030 [2]. Approximately 30% of patients with MDD are resistant to conventional treatment for depression [3]. Therefore, it is urgently needed to better understand the development of depression to identify a more effective treatment.
Accumulating evidence has suggested that inflammation contributes to the pathogenesis of depression [4][5][6]. Numerous studies have shown that microglia mediate inflammatory signaling that regulates mood and that microglial activation is responsible for depression symptoms [7][8][9]. The innate immune receptor Toll-like receptor 4 (TLR4) is highly expressed on the surface of microglia, serving as a first-line defense against invading microbes [10]. Together with its signaling adaptor myeloid differentiation factor 88 (MyD88), TLR4 plays a critical role in microglial activation [11]. Hines et al. [12] have reported that inflammatory stimulus lipopolysaccharide (LPS) activates TLR4/MyD88 signaling and triggers cytokine production in mouse microglia, resulting in depressive behaviors in mice. Blocking TLR4/MyD88 interaction may prevent LPS-induced microglial activation, cytokine production, and depressive behaviors. Besides, TLR4 signaling is upregulated in peripheral blood mononuclear cells of untreated patients with MDD. Psychotherapy reduces the expression of TLR4 and inflammatory markers, displaying a positive correlation with the improvement of depressive symptoms [13]. Thus, targeting TLR4/MyD88 signaling represents a promising therapeutic strategy in depression treatment. Salvia miltiorrhiza (Danshen) is a traditional Chinese medicine that is widely used in patients with cardiovascular diseases and acute ischemic stroke due to its function in promoting blood circulation. Salvianolic acids are the most abundant water-soluble components in S. miltiorrhiza, including salvianolic acid A and salvianolic acid B [14]. Salvianolic acid B can cross the blood-brain barrier and alleviate chronic mild stress (CMS)-induced depressive behaviors in animal models [15][16][17][18]. However, the underlying mechanism remains unclear. Recent studies have shown that salvianolic acid B suppresses TLR4/MyD88 signaling in primary cortical neurons and white adipose tissue in rodents [19,20]. Thus, we hypothesized that the suppression of TLR4/MyD88 signaling is involved in the antidepressant effect of salvianolic acids.
To test our hypothesis, we examined the antidepressant effect of salvianolic acids and the involvement of TLR4/MyD88 signaling in a CMS rat model that is widely used in depression research [21]. FDA-approved antidepressant drug fluoxetine was used as a positive control [22]. Our results suggest that the antidepressant effect of salvianolic acids is associated with the disturbance of TLR4/MyD88-mediated inflammatory signaling in the brain.

Animals and CMS Model
Eighty Sprague-Dawley male rats weighing 180-220 g were obtained from the Experimental Animal Center of Hebei Medical University (Shijiazhuang, Hebei, China; Certificate No. 1411040). The rats were housed at 23 ± 2°C under a 12: 12-h light: dark cycle, with free access to food and water. All experiments were conducted following the guidelines of the Animal Care and Use Committee of Hebei Medical University (#2014116).
The rats were randomly divided into CMS and non-CMS groups (n = 40/group). Rats in the CMS group were exposed to mild stressors daily for 6 weeks, as previously described [23]. The mild stressors included cage shaking for 1 h, water deprivation for 24 h, soiled cage for 24 h, reverse of light/dark cycle, food deprivation for 24 h, cage tilting for 24 h, swimming in 45°C water for 5 min, wrap restraint at 4°C for 1 h, tail clamping for 1 min, and wrap restraint for 1 h (Table 1). Rats in the non-CMS group were housed in a separate room under identical conditions without stress.

Sucrose Preference Test
The sucrose preference test was carried out in the animal's home cage as previously described [26,27]. The rats were given 1% sucrose (13-201-00107, Tianjin Baishi Chemical Co. Ltd, Tianjin, China) solution for acclimation for 1 day, and the sucrose water was replaced with pure water on the second day. On the day of the test, the rats were fasted with no water for 10 h, provided with 1% sucrose solution and pure water bottles, and the consumption of sucrose was measured after 2 h. Preference % = [consumption of sucrose water/(consumption of sucrose water + consumption of pure water) × 100%].

Forced Swimming Test
The forced swimming test was performed as previously described using a glass cylinder (20 cm in diameter and 50 cm high) (XR-XQ202, Shanghai Xinruan Information Technology Co. Ltd, Shanghai, China) filled with tap water (20 cm deep) at 25 ± 1°C [28]. Briefly, each rat was placed in the cylinder for 5 min, and the duration of immobility was recorded during the last 4 min. The water was refreshed between each test.

Quantitative Real-Time PCR
Rats were sacrificed at 3 weeks after drug treatment (24 h after the final stressor exposure). The whole prefrontal cortex and hippocampus of each rat were immediately harvested and snapped frozen in liquid nitrogen. Quantitative real-time PCR was performed to measure the mRNA levels of TLR-4 and MyD88 in the prefrontal cortex and hippocampus tissue samples. Total RNA was isolated using TRIzol (DP405-02, Tiangen Biotech Co. Ltd., Beijing, China), followed by cDNA synthesis (RR047B, TaKaRa, Tokyo, Japan). PCR was performed using SYBR green (RR82LR, TaKaRa, Tokyo, Japan) and the primers ( Table 2). β-actin was used as an internal reference. The relative expression was calculated using the 2 −∆∆Ct method.

Statistical Analysis
Data were expressed as the mean ± standard deviation. Statistical analysis was performed using Prism 5 (GraphPad Software Inc., San Jose, CA, USA) and SPSS 21.0 (IBM, Armonk, NY, USA). The treatment/time interaction was assessed using two-way ANOVA. The intergroup comparison was conducted using a one-way analysis of variance. A pairwise comparison was performed using the LSD method. The correlation of TLR4 expression with MyD88 expression, inflammatory cytokine secretion, or behavioral performance was evaluated using Spearman correlation analysis. p < 0.05 was considered statistically significant.

Salvianolic Acids Facilitate Weight Gains in CMS Rats
Because body weight changes reflect the overall impact of a chronic stressful situation [29], we examined the effect of salvianolic acids treatment on body weight changes. As shown in Fig. 1, at 3 weeks post-stress, CMS rats had reduced weight gain rates compared with non-CMS rats (all p < 0.05), suggesting that the animal model was successfully generated. At 3 weeks post-treatment, the weight gain rates of CMS rats treated with fluoxetine and salvianolic acids, alone or in combination, were higher than  those treated with saline (all p < 0.05). No significant difference was observed in the weight gain rates among the drug-treated CMS rats or among all four subgroups of non-CMS rats (all p > 0.05). Two-way ANOVA revealed significant treatment/time interaction (F (14,144) =10.258, p < 0.001) and treatment effect (F (7,72) = 29.013, p < 0.001).

Salvianolic Acids Alleviate Immobility in CMS Rats
The forced swimming test is commonly used to evaluate behavioral despair in animal models [30]. As shown in Fig. 2, before treatment, CMS rats showed significantly increased immobility durations compared with non-CMS rats (all p < 0.05). After 3 weeks of treatment, the immobility durations of CMS rats treated with saline remained significantly longer than those of non-CMS rats. However, compared with CMS rats administered saline, CMS rats administered fluoxetine and salvianolic acids, alone or in combination, exhibited decreased immobility durations (all p < 0.05). No significant difference was observed in the immobility durations among the drug-treated CMS rats or among all four subgroups of non-CMS rats (all p > 0.05). Two-way ANOVA revealed significant treatment/time interaction (F (7,70) = 5.390, p < 0.001) and treatment effect (F (7,70) = 33.251, p < 0.005).

Salvianolic Acids Improve Sucrose Preferences in CMS Rats
To examine the effect of salvianolic acids on anhedonia, we performed the sucrose preference test [26]. As shown in Fig. 3, sucrose preference was reduced in CMS rats compared with that in non-CMS rats before drug treatment (all p < 0.001), suggesting that CMS induces anhedonia in model rats. After 3 weeks of treatment, the sucrose preferences of CMS rats administered fluoxetine and salvianolic acids, alone or in combination, were elevated to comparable levels of non-CMS rats (all p > 0.05), whereas the sucrose preferences of CMS rats treated with saline remained lower than non-CMS rats treated with saline (p < 0.001). No significant difference was observed in the sucrose preferences among the drug-treated CMS rats or

Salvianolic Acids Reverse CMS-Induced Upregulation of TLR4 and MyD88 mRNA Expression in Rat Brain
Considering the involvement of TLR4 and MyD88 in the development of depression [12], we determined the mRNA levels of TLR4 and MyD88 in the hippocampus and prefrontal cortex of rats. As shown in Fig. 4A,B, before treatment, the mRNA levels of TLR4 and MyD88 in both the left and right hippocampus of saline-treated CMS rats were higher than those of saline-treated non-CMS rats, suggesting that CMS induces upregulation of TLR4/MyD88 expression in the brain. Compared with saline treatment, 3 weeks of treatment with fluoxetine and salvianolic acids, alone or in combination, reduced TLR4 and MyD88 mRNA in CMS rats (all p < 0.05) at comparable levels of those in non-CMS rats (all p > 0.05). Similar results were observed in both left and right prefrontal cortexes of rats (Fig. 4C,D).
Of note, fluoxetine and salvianolic acids cotreatment outperformed fluoxetine alone in suppressing TLR4 mRNA expression in the right prefrontal cortex of rats (p < 0.05). Fluoxetine and salvianolic acids cotreatment also outperformed salvianolic acids alone in suppressing TLR4 and MyD88 mRNA expression in the left prefrontal cortex of rats (both p < 0.05).

TLR4 mRNA Expression in the Brain is Correlated with MyD88 Expression, Inflammatory Cytokine Secretion, and Behavioral Performance of CMS Rats
Our previous data have shown that CMS enhances interleukin-1β (IL-1β) and tumor necrosis factor-α (TNFα) production in the hippocampus and prefrontal cortex of rats [31]. Compared with saline treatment, 3 weeks of treatment with fluoxetine and salvianolic acids, alone or in combination, reduced IL-1β and TNF-α secretion in the brain of CMS rats at comparable levels of non-CMS rats. Thus, we further assessed the correlation of TLR4 mRNA expression with MyD88 mRNA expression, inflammatory cytokine secretion, and behavioral performance. Our results showed that TLR4 mRNA levels in both hippocampus and prefrontal cortex were positively correlated with MyD88 mRNA expression, TNF-α secretion, and immobility duration but negatively correlated with sucrose preference. TLR4 mRNA expression was also positively correlated with IL-1β secretion in the prefrontal cortex of rats (all p < 0.05; Table 3).

Discussion
Depression is characterized by persistent sadness and loss of interest, along with cognitive and physical symptoms such as feelings of worthlessness, sleep disturbances, and lack of energy [1]. In this study, we used a wellestablished CMS protocol to induce depressive symptoms in rats [21]. By monitoring the body weight changes and assessing the results of forced swimming and sucrose preference tests that are common behavioral tests in depressionlike behavior in rodents [32], we found that three weeks of CMS exposure significantly reduced weight gains and sucrose preference while increasing immobility duration in rats. This is consistent with the results of previous studies [33,34], suggesting that the CMS rat model is successfully generated. We further found that treatment with fluoxetine and salvianolic acids, alone or in combination, effectively facilitated weight gains and alleviated depressive-like behaviors in CMS rats. Importantly, treatment with fluoxetine and salvianolic acids, alone or in combination, significantly reduced the mRNA levels of TLR4 and MyD88 in the hippocampus and prefrontal cortex of CMS rats. Moreover, TLR4 mRNA expression in the brain positively correlated with MyD88 expression, inflammatory cytokine production, and behavioral performance in CMS rats. These results suggest that salvianolic acids exhibited a comparable antidepressant effect to fluoxetine possibly by suppressing TLR4/MyD88 signaling in the brain.
Microglial activation is involved in the development of depression. Significant changes are noted in the number, morphology, and activity of microglia in depression [35][36][37]. Kreisel et al. [38] have demonstrated that in a chronic unpredictable stress mouse model, following an initial 2-3 days of stress-induced microglial proliferation and activation, some microglia underwent apoptosis, dystrophy, and decline in numbers within the hippocampus, but not in other brain regions. Pharmaceutical blockade of the initial stress-induced microglial activation rescued the microglial disturbance as well as the depressive-like behavior, suggesting that the dynamic microglial alteration has an etiological role in chronic stress-induced depression. TLR4 expressed in the prefrontal cortex and hippocampus plays an important role in stress-induced depression [39]. Upon recognizing specific pathogen-associated molecular patterns, TLR4 initiates innate immune responses through MyD88 or TRIF to activate the transcription factor nuclear factor κB (NF-κB), leading to neuroinflammatory responses [40]. It is well accepted that activation of TLR4/MyD88 signaling induces phosphorylation of NF-κB and subsequent expression of inflammatory mediators, such as IL-1β, IL-6, and TNF-α in the brain, contributing to the development of depression [31,41]. Consistently, our results showed that CMS exposure resulted in significant upregulation of TLR4 and MyD88 mRNA expression in the frontal cortex and the hippocampus of rats. Of note, TLR4 mRNA expression in the brain positively correlated with MyD88 expression, inflammatory cytokine secretion, and behavioral performance in CMS rats. These findings suggest that CMS activates TLR4/MyD88 signaling and enhances the production of inflammatory cytokines, leading to the development of depressive-like behavior.
Nonsteroidal anti-inflammatory drugs, alone or in combination with other antidepressants, are promising therapeutic agents for depression [42,43]. Studies have shown that salvianolic acids attenuate inflammation in different organs and tissues, including the brain [19,44]. A recent study has shown that salvianolic acid B ameliorates CMS-induced depressive-like behaviors and inhibits CMS-induced neural apoptosis and microglial activation in the hippocampus and cortex of mice [18]. Consistent with these reports, our results showed that salvianolic acids alleviated depressivelike behaviors in CMS rats, comparable to antidepressant fluoxetine. No additive or synergistic effect was observed in depressive-like behavior alleviation when fluoxetine and salvianolic acids were administered in combination. Furthermore, salvianolic acids reversed CMS-induced upregulation of TLR4 and MyD88 mRNA expression in the hippocampus and front cortex of rats, consistent with a previous study [19]. Interestingly, fluoxetine and salvianolic acids cotreatment outperformed fluoxetine alone in suppressing TLR4 mRNA expression in the right prefrontal cortex of rats and outperformed salvianolic acids alone in suppressing TLR4 and MyD88 mRNA expression in the left prefrontal cortex of rats, suggesting that fluoxetine and salvianolic acids have an additive effect on disturbing cerebral TLR4/MyD88 signaling. However, the underlying mechanism remains unknown and needs further investigation.
This study has several limitations. First, we only examined the mRNA levels of TLR4 and MyD88 in rat brains. The protein expression of these two genes and other components of TLR4 signaling will be investigated in future studies. Second, we did not examine the cerebral production of inflammatory cytokines in this study. The correlation assay was based on the results of our previous study. Third, further investigation is required to establish a causal link between salvianolic acid treatment and the suppression of TLR4 signaling in microglia.

Conclusions
In conclusion, we demonstrated that salvianolic acids were comparable to fluoxetine in alleviating CMS-induced depressive-like behaviors in rats while suppressing mRNA expression of TLR4 and MyD88 in the prefrontal cortex and hippocampus of rats. Salvianolic acids and fluoxetine showed an additive effect on suppressing TLR4/MyD88 signaling in the prefrontal cortex of rats. Our results suggest that salvianolic acids are promising antidepressants targeting TLR4/MyD88 signaling in the brain.

Availability of Data and Material
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Author Contributions
FZ and LY conceived and coordinated the study, designed, performed and analyzed the experiments, wrote the paper. LW, WL, SW, XW and CA carried out the data collection, data analysis, and revised the paper. All authors contributed to editorial changes in the manuscript. All authors read and approved the final manuscript. All authors have participated sufficiently in the work and agreed to be accountable for all aspects of the work.

Ethics Approval and Consent to Participate
All experimental protocols were carried out in accordance with the guidelines of the Animal Care and Use Committee of The First Hospital of Hebei Medical University (#2014116).