Puerarin ameliorated the behavioral deficits induced by chronic stress in rats

The present study aimed to investigate the mechanisms underlying the antidepressant-like effects of puerarin via the chronic unpredictable stress (CUS) procedure in rats. Similar to Sertraline (Ser), Chronic treatment of puerarin (60 and 120 mg/kg, i.g) elicited the antidepressant-like effects by reversing the decreased sucrose preference in sucrose preference test (SPT), by blocking the increased latency to feed in novelty-suppressed feeding test (NSFT) and the increased immobility time in forced swimming test (FST) without affecting locomotor activity. However, acute puerarin treatment did not ameliorate the antidepressant- and anxiolytic- like effects in FST and NSFT, respectively. In addition, enzyme linked immunosorbent assay (ELISA) and high performance liquid chromatography-electrochemical detection (HPLC-ECD) showed that chronic treatment of puerarin (60 and 120 mg/kg, i.g) reversed the decreased levels of progesterone, allopregnanolone, serotonin (5-HT) and 5-Hydroxyindoleacetic acid (5-HIAA) in prefrontal cortex and hippocampus of post-CUS rats. Furthermore, puerarin (60 and 120 mg/kg, i.g) blocked the increased corticotropin releasing hormone (CRH), corticosterone (Cort) and adrenocorticotropic hormone (ACTH). Collectively, repeated administration of puerarin alleviated the behavioral deficits induced by chronic stress which was associated with the biosynthesis of neurosteroids, normalization of serotonergic system and preventing HPA axis dysfunction.

Preparation of the chronically stressed rats. The CUS model was performed as described previously with some modifications 24 . Except for control, the animals were subjected to the following stressors (from Mondays to Saturadays, except Sundays): (1) low-intensity stroboscopic illumination (80 flashes/min); (2) overnight illumination; (3) soiled cage (180 mL water in 90 g sawdust bedding); (4) food or water deprivation (24 h); (5) forced swimming (5 min at 10 °C); (6) 45° cage tilt; (7) white noise (approx. 110 dB); (8) restraint (2 h); and (9) tail pinch (2 min). All of the stressors were applied randomly and continuously. The rats in the control group were left undisturbed in the home cages, except for the 14 h period of water deprivation prior to each sucrose test. The outline of design for CUS was shown in Table (Table 1).
Behavioral paradigms. After the acclimatization, rats were singly placed in cages and subjected to sucrose training and sucrose baseline test. The animals were divided into six statistically equivalent groups according to their sucrose intake in the final baseline test. After the 4-week CUS procedure, behavioral assessments including sucrose preference test (SPT) (from day 36 to 40), novelty-suppressed feeding test (NSFT) (from day 42 to 43), forced swimming test (FST) (from day 45 to 46), and open field test (OFT) (day 48) were performed on different days. The drugs were given to the rats from day 29 to the end of the behavioral tests. All the behavioral tests were performed 1 h after drugs administration. Control animals received 0.9% normal saline. The schedule of behavioral tests and drug treatment was outlined in Fig. 1.
SPT. SPT is widely applied to measure the depressive-like behavior and performed as described previously 24,29,30 .
Rats were placed in individual cages to train to consume 1% (w/v) sucrose solution for 48 h without food and water supply. Following 16 h water deprivation, one bottle of 1% sucrose solution was replaced by water. The position of the sucrose and water bottles for the SPT was counterbalanced. The SPT was performed for 1 h. During the test, rats could select between two preweighed bottles, one with 1% sucrose solution and the other with tap water. The sucrose preference was calculated as sucrose intake/(sucrose intake + water intake) ×100%.

NSFT.
The NSFT provides a sensitive and reliable measure of depressive-like behavior and motivation level in animals which mimics the situation in human 31,32 . The test was performed according to the literature with minor modifications 24 . Briefly, after fasting for 24 h, each rat was placed in the corner of the plastic box (76 × 76 × 46 cm) with several pallets placed in the center. The latency to begin eating within 5 min was recorded (defined as chewing or biting the pallet, instead of merely sniffing or toying with it). Moreover, home-cage food consumption in 5 min was immediately determined to assess effects of drugs on feeding drive.
FST. FST is a well-established measurement for evaluating the effects of antidepressants and the assessment is highly reliable to predict the validity of antidepressants 33,34 . The procedure comprised two sections (the pretest and the test) with identical apparatus and conditions (height 40 cm, diameter 20 cm, containing 25 cm of water maintained at 25 °C). During the pretest section, rats were forced to swim for 5 min. After 24 h, rats were placed in the same apparatus for 5 min and the section was designated as a test section. The duration of immobility during 5 min was measured.
OFT. To evaluate whether the reversion of depressive-like behavior by puerarin is dependent on an affect on locomotor activity, we assessed the number of crossings, rears, and fecal pallets in rats. The test was performed based on the literature with minor adjustments 35 . Each rat was placed in the corner of a plastic box (dimensions: 76 × 76 × 46 cm) that the base was divided into 16 equal squares for the 5-min acclimation period. Following that, the number of crossings (with all four paws placed into a new square), rears (with both front paws raised from the floor), and fecal pallets was rated by the observers who were blind to the grouping for 5 min by camera. The OFT was cleaned with a 5% ethanol/water solution after each test to remove any confounding olfactory cues and dried thoroughly between sections.
Neurosteroids measurement. Altered levels of progesterone and allopregnanolone have been implicated as one of the possible contributors to the development of depressive-like behavior 23,36 . The levels of neurosteroids in the antidepressant-like activity of puerarin were evaluated by enzyme linked immunosorbent assay (ELISA). The brain tissues preparation was based on a literature 24 . The brains were removed and carefully dissected to remove the prefrontal cortex and hippocampus at the end of OFT in 24 h after the last drug treatment. The brain regions were extracted by 1 mL extraction buffer per 100 mg tissue and then homogenized in the ice-cold lysis buffer containing 137 mM NaCl, 1% NP40, 10% glycerol, 20 mM Tris-HCl (pH 8.0), 1 μg/mL leupeptin, 1 mM PMSF 10 μg/mL aprotinin, and 0.5 mM sodium vanadate. The tissue homogenate solutions were centrifuged at 10,000 g for 25 min at 4 °C, and then the supernatants were collected. Progesterone and allopregnanolone were quantified by Enzyme Immunoassay kit  The determination of monoamine neurotransmitter levels. To further explore neurochemical mechanisms involved in antidepressant-like effect of puerarin, the levels of prefrontal cortex, hippocampus monoamine neurotransmitters and their metabolites were detected by high performance liquid chromatography-electrochemical detection (HPLC-ECD) after behavioral assessments. The detection was performed as described previously 7,28 . Also, the brain tissues were dissected at the end of behavioral tests in 24 h after the last drug treatment on ice by a binocular dissection microscope and homogenized in an ice-cold tissue lysis buffer containing 0.

Effects of puerarin on the CUS-induced behavioral deficits in SPT.
The effects of puerarin on CUS rats in SPT were shown in Fig. 2. Sucrose preference was significantly decreased in CUS rats. Consistent with Ser (15 mg/kg, i.g), puerarin (60 and 120 mg/kg, i.g) reversed the decreased sucrose preference (F (5,54) = 14.25, p = 0.0000; Fig. 2) in CUS rats. The results indicated that puerarin ameliorated the CUS-induced behavioral deficits in SPT.  Fig. 6D), respectively. These results indicated that puerarin attenuated the CUS-induced behavioral deficits were associated with the biosynthesis of progesterone and allopregnanolone in brain.  Fig. 7C) in serum were significantly increased. In accordance with Ser (15 mg/kg, i.p), these effects were significantly reversed by treatment with puerarin (60 and 120 mg/kg, i.g), respectively. These results indicated that the effects of repeated puerarin treatment on CUS-induced behavioral deficits were associated with decreased HPA stress hormone (Cort, CRH and ACTH) levels.   Table (Tables 2 and 3). After CUS exposure, levels of 5-HT and 5-HIAA in the prefrontal cortex and hippocampus were significantly decreased, respectively. Similar to Ser (15 mg/kg, i.g.), the decreased levels of 5-HT (F (5,30) = 3.538, p = 0.0124, for prefrontal cortex,  prefrontal cortex, Table 2; F (5,30) = 0.3393, p = 0.8850, for hippocampus, Table 3) in both brain regions were not significantly affected by CUS and puerarin. These results indicated that the effects of repeated puerarin treatment on CUS-induced behavioral deficits were associated with the normalized levels of 5-HT and 5-HIAA in the prefrontal cortex and hippocampus.

Discussion
In the present study, we preliminarily evaluate the effects of puerarin on CUS-induced behavioral deficits via the CUS model and its possible mechanism. The CUS-induced behavioral deficits were significantly ameliorated by puerarin without affecting locomotor activity in rats. In addition, together with the results of neurosteroids biosynthesis, serotonergic system and hormones of HPA axis, we found that the effects of repeated puerarin treatment on CUS-induced behavioral deficits were associated with biosynthesis of neurosteroids, normalization of serotonergic system and preventing HPA axis dysfunction.
Depressive-like behavior is a serious mental health problem associated with psychiatric morbidity and has been increased worldwide from past several decades 1, 2 . The CUS model mimics the depressive-like symptoms and is widely used in the preclinical antidepressants evaluation which offers a closer resemblance to chronic stress 24,38 . CUS induced behavioral changes that resemble clinical depressive-like behavior, such as reduced sucrose intake and responsiveness to rewarding stimuli 24,39 . In present study, CUS induced a significant reduction of sucrose preference in SPT (Fig. 2) and the increased immobility time in FST (Fig. 4B), two indicators of the core symptoms of depressive-like behavior. Furthermore, it has been documented that the CUS model may also induce anxiogenic-like symptoms, which is evidenced by the increased latency to feed (Fig. 3). The NSFT is initially used to examine the effects of anxiolytic agents, and also sensitive to chronic treatment with antidepressants 40,41 . These findings were further supported by the successful experimental protocol of the CUS procedure.
The CUS-induced behavioral parameters can be reversed by the chronic administration of antidepressants 7, 24 . Our results showed that similar to Ser (15 mg/kg i.g.), puerarin (60 and 120 mg/kg i.g.) reversed the decreased sucrose preference to normal levels in SPT (Fig. 2) and the increased immobility time in FST (Fig. 4B). Moreover, the increased latency to feed in NSFT was blocked by puerarin (60 and 120 mg/kg i.g.) (Fig. 3A) without affecting home-cage food consumption (Fig. 3B), suggesting that puerarin attenuated the CUS-induced behavioral deficits in rats. The effective doses of puerarin (60 and 120 mg/kg i.g.) were almost confirmed among SPT (Fig. 2), FST (Fig. 4B) and NSFT (Fig. 3A) and in line with depressive-like behavioral deficits that were reversed by puerarin with the similar doses 16 . Moreover, the OFT was also performed to evaluate whether the reversion of chronic stress behavioral deficits by puerarin was dependent on locomotor activity. Consistent with the previous findings and Ser (15 mg/kg i.g.) 28 , the present study also showed that the locomotor activity (e.g line crossings, rears and fecal pallets) was not affected by puerarin in OFT (Fig. 5). In addition, these findings were also consistent with our observations found that the anxiolytic activities of puerarin determined by EPMT which the total time (Fig. 8E) and entries (Fig. 8F) in arms were not significantly altered by puerarin. The data indicated that the effects of puerarin on the CUS-induced behavioral deficits were not mediated by affecting the locomotor activity in rats.   Ser was delivered 1 h before behavioral experiments and exerted the anti-stress activities with its chronic treatment 27,42 . Therefore, puerarin had the similar the treatment schedule with Ser and showed that both drugs elicited the antidepressant-like effects via behavioral tests in present study. The study showed that the behavioral experiments were performed after the drug treatment at least 1 week (chronic/subchronic treatment). Moreover, our studies also found that the single treatment of puerarin 1 h before the behavioral experiments in rodents  Table 3. The effects of puerarin on hippocampal monoamine neurotransmitter levels in CUS rats. ## p < 0.01 vs. vehicle-treated CUS (−) group; * p < 0.05, ** p < 0.01 vs. vehicle-treated CUS (+) group (n = 6). did not produce the anxiolytic-and antidepressant-like effects (Fig. 8). The findings were supported by the antidepressant-like effects of puerarin with repeated treatment 16 . Thus, it indicated that the effects of puerarin might be induced by its chronic treatment effect. It was reported that dysfunction of the prefrontal cortex or hippocampus is implicated in the pathogenesis of depressive-like behavior 34,43 . Both brain regions play an important role in fear conditioning, emotional processing and explicit memory that are relevant to anxiogenic-and depressive-like behavior [43][44][45] . Thus, to confirm the role of neurosteroids in the effects of puerarin on the CUS-induced behavioral deficits, we then measured levels of endogenous neurosteroids and monoamine neurotransmitters in the above brain regions.
Although the pathological factor of depressive-like behavior have been widely researched, the exact factors involved are still not discovered. More evidences demonstrate that the biosynthesis of neurosteroids (e.g. progesterone and allopregnanolone) has been implicated as one of the possible contributors in the development of depressive-like behavior 46,47 . The present findings showed that similar to Ser, both decreased levels of neurosteroids were significantly reversed by puerarin in the prefrontal cortex ( Fig. 6A and B) and hippocampus ( Fig. 6C  and D), respectively. The results indicated that the effects of puerarin on the CUS-induced behavioral deficits were associated with the biosynthesis of progesterone and allopregnanolone in brain. Our present study was also confirmed by the finding that altered levels of progesterone affected the levels of metabolite steroids, such as allopregnanolone, and progesterone withdrawal in animals also dramatically reduces levels of allopregnanolone in brain 48 .
Progesterone serves as the main precursor molecule for 3β-pregnane neuroactive steroids which possessed antidepressant-like activities 49 . The beneficial effect of progesterone may come after its conversion to allopregnanolone and through that metabolite's agonistic action at GABA (γ-aminobutyric acid) A receptors 47 . Allopregnanolone is the brain neurosteroid that acts GABAA receptors and is the selective positive endogenous modulator of the action of GABAA at GABAA receptors in brain 49,50 . More clinical studies have shown that depressive-like behavior is closely associated with the biosynthesis of neurosteroids. For instance, decreased allopregnanolone in peripheral blood or cerebrospinal fluid (CSF) is found to relevant to MDD, anxiety disorders, premenstrual dysphoric disorders, negative symptoms in schizophrenia, or impulsive aggression 51 . These might be related to the effects of levels of allopregnanolone on the regulation of GABAA receptor function, which plays a role in the pathophysiology of depressive-like behavior. The GABAA agonist modulator interacted on the brain by changing the expression of GABAA receptor subunit to produce the neuroprotective effects on the depressive-like behavior 52 .
Moreover, the hyperactivity of the HPA axis, which is commonly seen in patients with depressive-like symptoms, is the most commonly observed neuroendocrine abnormality in MDD 53 . Here, we found that the increased levels of Cort (Fig. 7A), CRH (Fig. 7B) and ACTH (Fig. 7C) in serum of post-CUS rats. The findings were supported by that the increased levels of CRH, Cort and ACTH in menopause depressive ovariectomized rats under chronic unpredictable mild stress 54 . The results were also consistent with the previous study that the role of allopregnanolone as an endogenous negative regulator of HPA axis activity, and plasma Cort was elevated concomitantly with decreased levels of allopregnanolone in chronically stressed rats 55 . Interestingly, study also demonstrated that the hormones of HPA axis above in post-CUS rats were reversed by puerarin (Fig. 7), which indicated that the normalization of brain neurosteroid levels and the subsequent prevention HPA axis dysfunction may underlie the activities of puerarin on the CUS-induced behavioral deficits.
In mammals, the HPA axis and the monoamines system closely interact in central nervous system (CNS) (particularly in prefrontal cortex and hippocampus) and are greatly involved in stress-related disorders 56,57 . Based on these evidences, the role of monoamines in the effects of puerarin on the CUS-induced behavioral deficits were evaluated. Our findings demonstrated after CUS exposure, levels of 5-HT and 5-HIAA in both brain regions were significantly decreased (Tables 2 and 3). The results were confirmed with decreased serotonin levels (e.g 5-HT and 5-HIAA) in a CUS animal model 7 . As the pathophysiological theory of depressive-like behavior, the monoamine hypothesis holds that lowered levels of 5-HT in the CNS are closely associated with depressive-like behavior 35 . However, similar to Ser, the decreased levels of 5-HT and 5-HIAA were significantly reversed by puerarin (Tables 2 and 3), indicating that the effects of puerarin on the CUS-induced behavioral deficits were associated with the normalized levels of 5-HT and 5-HIAA in brain. Substantial evidences indicated that depressive-like behavior was due to an absence of brain monoaminergic activity 7,35 , and the levels of monoamine neurotransmitters (e.g 5-HT) in brain were increased in the experimental groups after the antidepressant treatments 58 .
In neurological study, reports showed that puerarin elicited the potential effects on attenuating memory and learning disorders, including Parkinson's disease, Alzheimer's disease, ischaemic stroke et al. [59][60][61] . However, not many studies on the antidepressant-like effect of puerarin and its possible mechanism. It is possible that puerarin normalizes the depressive-like behavior by other alternative mechanisms, such as activation of brain-derived neurotrophic factor (BDNF) and glutamate receptor. Recently, study showed puerarin ameliorates depressive-like behaviors and chronic pain in mice with SNI which markedly promoted the activation of CAMP-response element binding protein (CREB) pathway and induced BDNF expression 16 . BDNF plays an important role on neuronal survival, neurogenesis, differentiation of neurons and synapses, learning and memory 62 . Interestingly, BDNF was decreased in hippocampus and prefrontal cortex of patients with depressive-like symptoms and other psychiatric disorders 63 . Increased BDNF expression could largely improve depressive disorders 64 . Thus, BDNF was recently suggested as an important biomarker for successful treatment of depressive-like behaviors. Puerarin profoundly induced the phosphorylation of CREB and expression of BDNF 16 . Therefore, it is possible that puerarin promotes the cross-talks between BDNF-CREB signaling pathways.
Moreover, the glutamate receptors have also been examined as potential therapeutic targets for depressive-like behavior 65 . Puerarin ameliorated learning and memory deficits in mice through normalizing the glutamatergic/ GABAergic system and causing synaptic structural modifications in the hippocampus 66 . Previous study showed that neuroprotective effect of puerarin was potentially mediated through the inhibition of glutamate-induced activation of mitochondrial-dependent signaling pathway and calmodulin-dependent protein kinase II (CaMKII)-dependent apoptosis signal-regulating kinase 1(ASK-1)/c-Jun N-terminal kinase (JNK)/p38 signaling pathway 67,68 .
Taken together, the present study showed that puerarin exerted the ameliorative activities via the CUS-induced stress and various behavioral tests. The possible mechanism may be associated with biosynthesis of neurosteroids, normalization of serotonergic system and preventing HPA axis dysfunction, which may account for the molecular and cellular mechanism underlying the effects of puerarin on the CUS-induced behavioral deficits. Based on the previous and present study, it indicates that not only advance our knowledge of the theories of MDD, but also have clinical implications for puerarin that may serve as a novel compound for the therapeutics of depressive-like behavior. Although we preliminarily evaluated the effects of puerarin on CUS-induced behavioral deficits and its possible mechanism, the effect of chronic puerarin on behavior and molecular readouts was not fully performed. Thus, it is possible that partial activity may not be specific to CUS. More work should be conducted the chronic puerarin in rats to provide more evidences for the chronic activity of puerarin on CUS-induced behavioral deficits and molecular pathways/targets. In addition, the pharmacodynamics of puerarin in CNS towards the clinical translation are also needed.