Geniposide Alleviates Depression-Like Behavior Via Enhancing BDNF Expression in Hippocampus of Streptozotocin-Evoked Mice

Metab Brain Dis DOI 10.1007/s11011-016-9856-4

ORIGINAL ARTICLE

Geniposide alleviates depression-like behavior via enhancing BDNF expression in hippocampus of streptozotocin-evoked mice

Junming Wang1,2 & Peili Duan 2 & Ying Cui1,2 & Qing Li2 & Yanran Shi 2

Received: 4 February 2016 /Accepted: 9 June 2016 # Springer Science+Business Media New York 2016

Abstract Clinical and preclinical data suggest that diabetes is indicate that GP can alleviate depression-like behavior in often psychological complications such as depression. STZ-evoked diabetic mice, and suggest its mechanisms may Geniposide (GP), a major compound in Gardenia jasminoides partially be ascribed to up-regulating BDNF expression in Ellis with both medicinal and nutritional values, has been brain. previously confirmed to exert anti-diabetic and anti- depressive activities. The present study attempted to observe Keywords Antidepressant . Geniposide . Brain-derived anti-depressive mechanisms of GP in streptozotocin (STZ) neurotrophic factor . Diabetic complications . Forced evoked diabetic mice by involving brain-derived neurotrophic swimming test factor (BDNF), for the first time. Mice were given GP daily (50, and 100 mg/kg, ig) or reference drugs FHMH [fluoxetine hydrochloride (FH, 10 mg/kg, ig)combinedwithmetformin Introduction hydrochloride (MH, 100 mg/kg, ig)] for 3 weeks. The forced swimming test (FST) was performed to observe depression- Diabetes mellitus (DM) is one of the most commonly occur- like behavior, and serum and brain tissues were used for neu- ring chronic diseases with recent estimates suggesting 366 rochemical and fluorescent quantitative reverse transcription million people currently affected (Rhee 2015). DM is often PCR analyses. STZ induced excessively increased blood sug- associated with psychological complications such as depres- ar and immobility time in FST, in a manner attenuated by GP sion, which have severe impact on morbidity and mortality and FHMH administration. GP administration further elevated (Islam et al. 2015). A literature review has shown that about BDNF levels, and up-regulated the mRNA expression of 42 % of patients with diabetes have elevated levels of depres- BDNF and tropomyosin-related kinase B (TrkB) in hippo- sion symptoms (Gemeay et al. 2015). In addition, recent clin- campus of diabetic mice. In addition, STZ induced the exces- ical studies have shown the prevalence of depression in dia- sive level of serum corticosterone (CORT), while GP did not betic patients is much higher than the normal population influence on it in diabetic mice. Taken together, these findings (Zhang et al. 2015a, 2015b, 2015c), supporting the notion that there exists a strong link between diabetes and depression disorder. Coexisting depression in people with diabetes causes significantly decreased adherence to treatment, poor metabol- * Junming Wang [email protected] ic control, higher complication rates, decreased quality of life, increased healthcare use and cost, increased disability and lost productivity, and increased risk of death (Egede and Ellis 1 Collaborative Innovation Center for Respiratory Disease Diagnosis 2010). Depression in patients with diabetes is associated with and Treatment & Chinese Medicine Development of Henan Province, Henan University of Traditional Chinese Medicine, less frequent blood glucose monitoring and suboptimal glyce- Zhengzhou 450046, China mic control (Gemeay et al. 2015), which further worse the 2 College of Pharmacy, Henan University of Traditional Chinese diabetic condition. Medicine, East Jinshui Road & Boxue Road, Zhengzhou 450046, Streptozotocin- (STZ-) evoked diabetes is among the most China widely used preclinical models. Previous reports have shown Metab Brain Dis that STZ-induced diabetes results in depression-like behavior antioxidative enzyme heme oxygenase-1 (HO-1) (Yin et al. in various preclinical testing paradigms (Castillo-Gómez et al. 2010a), antidepressant-like effect on chronic unpredictable 2015). STZ-induced diabetes has demonstrated significantly mild stress-induced depressive rats by regulating the prolonged immobility time in tail suspension and forced hypothalamus-pituitary-adrenal axis (Cai et al. 2015)and swimming tests which are both the important behavioral tests anti-diabeticeffect(Yaoetal.2014), etc. However, as for its with strong predictive validity (Anjaneyulu et al. 2003;Kamei antidepressant-like effect on diabetes, there is no experimental et al. 2003). report until now. Furthermore, whether its mechanism is asso- Increasing evidence suggests that brain-derived neuro- ciated with BDNF has not been known. trophic factor (BDNF), as one of the neurotrophin families The present study was designed to assess the of growth factors with an important impact on the survival antidepressant-like effect of GP on STZ-evoked diabetic mice and synaptic of neurons in the adult central nervous system and the mechanism involved in BDNF by means of behavioral (CNS), may play important roles in the pathogenesis of dia- and pharmacological procedures. betes as well as depression, and contribute to the association of diabetes and depression (Alzoubi et al. 2013; Caletti et al. 2015; El-Marasy et al. 2014; Lenart et al. 2016; Wang and He Materials and methods 2015;Wangetal.2012;Zhouetal.2013). On the one hand, both animal and clinical experiments showed that BDNF has Experimental animals important influence on the pathobiology of diabetes, because BDNF modulates the secretion and actions of insulin, leptin, Male Kunming (KM) male mice (18–22 g) were purchased ghrelin, various neurotransmitters and peptides, and pro- from Experimental Animal Center of Henan Province inflammatory cytokines related to energy homeostasis (Rao (Zhengzhou, Henan Province, China). Animals were given et al. 2008). On the other hand, a strong association between rodent laboratory chow and water ad libitum, and maintained BDNF and depression has been demonstrated in both clinical under controlled conditions with a temperature of 22 ± 1 °C, and animal studies. For instance, animal experiments have relative humidity 60 ± 10 % and a 12/12 h light/dark cycle showed that BDNF decrease was associated with depressive (lights on at 7:00 a.m.). All the procedures were in strict ac- states, and clinical investigations revealed that BDNF levels in cordance with the P.R. China legislation on the use and care of depressed patients were lower than those in normal controls laboratory animals and with the guidelines established by and supplementary BDNF was able to reverse depressive sta- Institute for Experimental Animals of Henan University of tus (Haghighi et al. 2013;Lietal.2016). BDNF may play an Traditional Chinese Medicine and were approved by the uni- important role in the pathogenesis of depression and, there- versity committee for animal experiments. fore, might participate in the therapeutic actions of antidepressants (Kerman 2012). Reagents Geniposide (GP), an iridoid glycoside isolated from the Chinese herb Gardenia jasminoides (Zhizi), possesses diverse Geniposide (GP) was provided from Shanghai Jinsui bioactivities including neuroprotection in the MPTP mouse Biotechnology Co., Ltd. (Shanghai, China) with the purity model of Parkinson’s disease by enhancing growth factor sig- more than 98 % as assayed by high-performance liquid chro- naling and the reduction of apoptosis (Chen et al. 2015), ame- matography (HPLC). Fluoxetine hydrochloride (FH) was pur- liorating cerebral ischemia-reperfusion injury in rats by the chased from Changzhou Siyao Pharmaceuticals Co., Ltd. enhancement of BDNF and inhibition of caspase-3 (Zhang (Changzhou, Jiangsu Province, China). Metformin hydro- et al. 2006a; b), ameliorating learning memory deficits, reduc- chloride (MH) was purchased from Shanghai Xinyi ing tau phosphorylation and decreasing apoptosis via GSK3β Pharmaceuticals Co., Ltd. (Shanghai, China). Streptozotocin pathway in streptozotocin-induced alzheimer rat model (Gao (STZ), purchased from Sigma Chemical Co. (St. Louis, et al. 2014), attenuating insulin-deficiency-induced accelera- Missouri, USA), was dissolved in sodium citrate buffer tion of β-amyloidosis in an APP/PS1 transgenic model of (0.1 mol/L, pH 4.2) immediately before its administration. Alzheimer’s disease via enhancing insulin signaling (Zhang Mouse corticosterone (CORT) and BDNF enzyme-linked im- et al. 2015a, 2015b, 2015c), protecting human neuroblastoma munoassay (ELISA) kits (R&D system) were purchased from cells (SH-SY5Y cell line) against formaldehyde stress through R&D Systems China Co. Ltd. (Shanghai, China). modulating the expression of Bcl-2, P53, caspase 3 and caspase 9, and by increasing the activity of intracellular superox- Diabetes induction ide dismutase and glutathione peroxidase (Sun et al. 2013), antagonizing oxidative stress probably involving phos- Experimental diabetes was induced following an overnight phatidylinositol 3′-kinase (PI3K)- and nuclear factor-E2- fast by a single intraperitoneal (ip) injection of STZ at the dose related factor 2 (Nrf2)-mediated upregulation of the of 60 mg/kg freshly dissolved in sodium citrate buffer (0.1 M, Metab Brain Dis pH 4.2). Hyperglycemia was confirmed 72 h after STZ ad- Assay for Open Field Test (OFT) ministration by a portable Freestyle glucometer (Jiangsu Yuyue Medical Equipment & Supply Co. Ltd., Danyang, Locomotor activity was observed by the open field test (OFT) Jiangsu Province, China) in a blood sample obtained by tail through a modified method (Porsolt et al. 1977). Briefly, the prick and confirmed again at ending of the behavioral tests locomotor activity of the mice was measured using a box (ElBatsh 2015). Animals with fasting blood glucose levels (40 × 60 × 50 cm) with the floor divided into 12 identical ≥11.1 mmol/L (equal to 200 mg/dL) were maintained in the squares illuminated with light from the ceiling. Mice were study (ElBatsh 2015). placed in the central square, and horizontal locomotion (number of crossed squares) and the number of rears (posture sustained with hind-paws on the floor) were recorded for Experimental procedure 3 min. The open field arena was cleaned following each trial.

Diabetic mice were randomly selected to receive daily doses Serum CORT and hippocampus BDNF levels of 50 or 100 mg/kg of GP, positive drug FHMH (10 mg/kg of determination FH combined with 100 mg/kg of MH) or vehicle 0.5 % sodium carboxyl methyl cellulose (CMC-Na) (0.2 mL/10 g) Serum CORT and hippocampus BDNF levels were deter- (n = 10 mice per group) orally for 21 consecutive days by mined by commercially enzyme-linked immunoassay intragastrical administration (ig)(Caietal.2015). Non- (ELISA) kits (R&D Systems China Co. Ltd., China) accord- diabetic control mice (CTR) (n = 40) received vehicle on the ing to the manufacturer’s protocols. day of diabetes induction and were subsequently allocated to receive 0.5 % CMC-Na, or the same doses of GP or FHMH as Fluorescent Quantitative Reverse Transcription-PCR the STZ mice, following the same protocol as before. At 1 h (FQ-RT-PCR) after the last administration, the animals were submitted to the openfieldtest(OFT),followedbyforcedswimmingtest Total RNAwas extracted from hippocampus using Trizol reagent (FST). Thirty minutes after the FST and immediately after following the manufacturer’s instructions. Reverse transcription the last determination of glycemia, blood samples were col- (RT) was performed using a cDNA synthesis kit according to the lected by eyeball removal exsanguinations and centrifuged at manufacturer’s instructions. The house-keeping gene glyceralde- 3500 rpm for 15 min to obtain serum for detections of corti- hydes-3-phosphate dehydrogenase (GAPDH) was used as an costerone (CORT). Thereafter, the animals were euthanized internal control. Sequences of the PCR primers were as follows: by decapitation and the brain rapidly removed and placed on BDNF Forward 5′-CAGGGGCATAGACAAAAG-3′, Reverse ice. The whole hippocampuses in each hemisphere were dis- 5′-CTTCCCCTTTTAATGGTC-3′ (153 bp product), TrkB sected, weighed and kept frozen until the analysis of brain- Forward 5′-TTTCCGCCACCTTGACTTG-3′, Reverse 5′- derived neurotrophic factor (BDNF) level, and the mRNA ACAGGAACACGTGAACGGATT-3′ (61 bp product), and expression of BDNF and its cognate receptor tropomyosin- GAPDH Forward 5′-CAAGGTCATCCATGACAACTTTG-3′, related kinase B (TrkB). Reverse 5′-GGGCCATCCACAGTCTTCTG-3′ (90 bp product). FQ-RT-PCR was performed using Hot Start Fluorescent PCR Core Reagent Kits (SYBR Green I) on a real-time PCR Assay for Forced Swimming Test (FST) instrument (ABI StepOnePlus, Applied Biosystems). PCR ther- mal cycling parameters were as follows: the denaturing step at Independent groups of mice were submitted to the FST as 94 °C for 4 min, followed by 40 cycles annealing step at 94 °C described by Porsolt et al. (1978) with minor modifications. 30 s, 60 °C 30 s, and 72 °C 30 s. All amplifications and detec- The test was conducted in 2 sessions. First, in the pre-test tions were carried out in a MicroAmp optical 96-well reaction session, mice were individually placed into glass cylinders plate with optical adhesive covers. Relative expression of (height, 25 cm; diameter, 10 cm) containing water (10 cm mRNA (%) =2-ΔCT(1–2) × 100 %, where CT represents thresh- deep; 22 ± 1 °C) for 15 min. Twenty-four hours later, animals old cycle, Δ CT1 = CT( BDNF) -CT( GAPDH) , and weresubmittedtoa5minsessionoftheFST.During ΔCT2 = CT(TrkB)-CT(GAPDH). this session, total time of immobility (the minimum movement required to keep the head above water) and Statistical analysis the latency to the first immobility episode were recorded with a stopwatch (Castagné et al. 2011). After each All experimental data were expressed as mean ± standard error session (pre-test and test session), the animals were removed of mean (SEM). Significant differences among experimental and allowed to dry in a separate cage before being returned to groups were compared by one-way Analysis of Variance their home cages. (ANOVA) followed by Least Significant Difference (LSD) Metab Brain Dis

(P < 0.05) using the Statistics Package for Social Science Effect of geniposide on locomotor activity in OFT (SPSS) program Version 13.0. To exclude the possibility that GP induces changes to locomotory behaviour, we investigated the effects of GP on loco- Results motor activity. Mice treated with GP (50 or 100 mg/kg) and the positive control FHMH (10 mg/kg of FH combined with Effect of geniposide on fasting blood glucose 100 mg/kg of MH) for 21 days showed no effect on the number of crossed squares or rearing frequency in either the VEH Figure 1a shows significant differences in fasting blood glu- controls or the STZ-mice (P >0.05;Fig.2), indicating that the cose between the vehicle (VEH) controls and the STZ-mice. antidepressant-like effects of GP are not attributable to a stim- Elevated fasting blood glucose level was observed in the STZ- ulatory effect on locomotory function. mice by comparison with VEH control mice. Treatment with GP (50 or 100 mg/kg) and FHMH only significantly de- Effect of geniposide on serum corticosterone level creased blood glucose in the STZ-mice (P <0.05,P <0.05 and P < 0.01, respectively). Figure 3 shows significant differences in serum CORT levels between the VEH controls and the STZ-mice. Elevated serum Effect of geniposide on the immobility time in FST CORT level was observed in the STZ-mice by comparison with the VEH control mice. GP (50 or 100 mg/kg) did not Our results demonstrated that diabetic mice had significantly change serum CORT level in either the VEH controls or the higher immobility duration by comparison with the VEH con- STZ-mice (P > 0.05; Fig. 3). trol (P < 0.01), and GP attenuated this depressive-like behavior in the STZ-mice at doses of 50 and 100 mg/kg (P <0.05 Effect of geniposide on hippocampal BDNF level and P < 0.01). This effect was comparable with that of the positive control FHMH [the classical antidepressant FH Hippocampal BDNF activity in the STZ-mice at the time of (10 mg/kg, oral) combined with the hypoglycemic MH sacrifice were significantly decreased (10.2 ± 1.9 pg/mg protein) (100 mg/kg, oral)] (Fig. 1b). compared with the non-diabetic VEH control (22.3 ± 4.1 pg/mg

Fig. 1 Effects of geniposide (GP) 25 on fasting blood glucose and A immobility time in the forced ## 20 swimming test (FST) in STZ- evoked mice. Data are presented as mean ± SEM (n =10). 15 ##, Significant differences compared * with vehicle (VEH) control were #, ** designated as #P <0.05and 10 ## **

P < 0.01, and with (mmol/l) Glycemia streptozotocin (STZ) control as 5 *P <0.05and**P <0.01

0 VEH 0 VEH GP 50 VEH GP 100 VEH FHMH STZ 0 STZ GP 50 STZ GP 100 STZ FHMH Normoglycemic Diabetic 120 B ## 100 ##, * 80 ** ** 60 # 40

Immobility time (seconds) 20

0 VEH 0 VEH GP 50 VEH GP 100 VEH FHMH STZ 0 STZ GP 50 STZ GP 100 STZ FHMH Normoglycemic Diabetic Metab Brain Dis

Fig. 2 Effects of geniposide on 70 locomotor activity in open field A test in STZ-evoked mice. Data are 60 presented as mean ± SEM 50 (n =10) 40

20 Number of crossed squares crossed of Number 10

0 VEH 0 VEH GP 50 VEH GP 100 VEH FHMH STZ 0 STZ GP 50 STZ GP 100 STZ FHMH Normoglycemic Diabetic 16 B 14

6 Rearing frequency Rearing 4

0 VEH 0 VEH GP 50 VEH GP 100 VEH FHMH STZ 0 STZ GP 50 STZ GP 100 STZ FHMH Normoglycemic Diabetic protein; P < 0.01). GP (50 and 100 mg/kg) administration sig- whether BDNF and TrkB participates in anti-depressive nificantly elevated hippocampal BDNF levels to 20.5 ± 3.8 pg/ mechanisms of GP. As shown in Fig. 5,themRNA mg protein and 25.4 ± 4.7 pg/mg protein in STZ-mice, respec- expression of BDNF and TrkB in STZ-mice was signif- tively (both P < 0.01). Moreover, treatment with FHMH also icantly downregulated in compared with the non- increased BDNF level to 24.2 ± 4.6 pg/mg protein in STZ- diabetic VEH controls (both P < 0.01). Administration mice (P < 0.01) (Fig. 4). of GP at the doses of 50 and 100 mg/kg significantly upregulated the hippocampal mRNA expression of Effect of geniposide on hippocampal BDNF and TrkB BDNF and TrkB in STZ-mice (BDNF P < 0.05, mRNA expression P < 0.01; TrkB P < 0.05, P < 0.01). Moreover, treatment with FHMH also enhanced BDNF and TrkB The mRNA expression of BDNF and its cognate recep- mRNA expression in STZ-mice (both P <0.05) tor TrkB in the hippocampus were assessed to explore (Fig. 5).

Fig. 3 Effects of geniposide on 50 ## serum corticosterone (CORT) 45 ## levels in STZ-evoked diabetic # 40 mice. Data are presented as mean ± SEM (n = 10). Significant 35 differences compared with 30 * vehicle (VEH) control were 25 designated as #P <0.05and ##P < 0.01, and with 20 streptozotocin (STZ) control as 15 Serum CORT (ng/ml) *P <0.05 10 5 0 VEH 0 VEH GP 50 VEH GP 100 VEH FHMH STZ 0 STZ GP 50 STZ GP 100 STZ FHMH Normoglycemic Diabetic Metab Brain Dis

Fig. 4 Effects of geniposide on 45 hippocampal brain-derived 40 # neurotrophic factor (BDNF) levels in STZ-evoked diabetic 35 mice. Data are presented as 30 ** ** mean ± SEM (n = 10). Significant differences compared with 25 ** vehicle (VEH) control were 20 designated as #P <0.05and 15 ##P < 0.01, and with ## streptozotocin (STZ) control as 10 **P <0.01 5 Hippocampal BDNF (pg/mg protein) BDNF Hippocampal 0 VEH 0 VEH GP 50 VEH GP 100 VEH FHMH STZ 0 STZ GP 50 STZ GP 100 STZ FHMH Normoglycemic Diabetic

Discussion Hence, it seems advantageous to use GP, an active ingredient from GJ, for treatment of diabetes patients In Traditional Chinese Medicine (TCM), herb compounds de- complicated with depression. However, there is still a lack of coction containing G. jasminoides fruits (GJ) have been wide- evidence. ly used for the treatment of diabetes mellitus and depression in Indeed, in the previous report, Xue et al. (2013)havefound East Asian countries for hundred years. Diabetes and depres- that acute administration of ethanol extract of Yueju pill sion are common co-morbid conditions, and patients with (YJP), a herb compounds containing GJ, rapidly attenuated diabetes have also elevated levels of depression symptoms. depressive-like symptoms in learned helpless paradigm, and

Fig. 5 Effects of geniposide on hippocampal mRNA expression 1.6 A # of BDNF and TrkB in STZ- 1.4 evoked diabetic mice. Data are (%) mRNA presented as mean ± SEM 1.2 (n = 10). Significant differences BDNF ** #, * compared with vehicle (VEH) 1 * control were designated as ## #P <0.05and##P <0.01,and 0.8 with streptozotocin (STZ) control as *P < 0.05 and **P <0.01 0.6

0.4

0.2

0 VEH 0 VEH GP 50 VEH GP 100 VEH FHMH STZ 0 STZ GP 50 STZ GP 100 STZ FHMH Relative expression of hippocampal of expression Relative Normoglycemic Diabetic 1.2 B #

mRNA (%) mRNA 1 TrkB 0.8 ** #, #, * 0.6 * ## 0.4

0.2

0 VEH 0 VEH GP 50 VEH GP 100 VEH FHMH STZ 0 STZ GP 50 STZ GP 100 STZ FHMH Relative expression of hippocampal of expression Relative Normoglycemic Diabetic Metab Brain Dis the antidepressant-like effects were sustained for at least 24 h spontaneous activity in the OFT, suggesting that the decreases in tail suspension test of ICR mice. Additionally, YJP, like in immobility in the FST were not caused by motor stimula- ketamine, an antidepressant by blocking NMDA receptors for tion but rather by increases in active movements, such as glutamate, rapidly increased the expression of BDNF in the struggling and swimming. hippocampus of mice. YJP rapidly reduced the phosphoryla- The effect of GP on locomotory activity was not in agree- tion of eukaryotic elongation factor 2 (eEF2), leading to de ment with Cai et al. (2015), who reported that GP (25, 50 and suppression of BDNF synthesis (Xue et al. 2013;Zhangetal. 100 mg/kg) improved spontaneous locomotor activity in the 2015a, 2015b, 2015c). However, until now, the bioactive com- OFT of rats undergoing chronic unpredictable mild stress pounds of YJP in antidepressant-like effects are not known. (CUMS). The discrepancy of the data may depend on the Our study confirms that GP can be one of the main bioactive differences between species and (or) the experimental proce- compounds of GJ even or YJP in antidepressant-like effects in dures used. diabetic mice. In fact, that YJP has antidepressant-like effects Diabetes and depression are common co-morbid condi- provided our study for experimental bases and backgrounds tions. As previously mentioned, depression complicated with instead of inhibiting our innovation. Furthermore, our study diabetes severely harms human health and life, as evidenced mainly observed antidepressant-like effects of GP in diabetic by causing significantly decreased adherence to treatment, pathological conditions, while other reported studies mainly did poor metabolic control, higher complication rates, decreased in normal physiological conditions. Considering that Chinese quality of life, increased healthcare use and cost, increased medicine is usually used on patients instead of normal people in disability and lost productivity, and increased risk of death clinic, thus our designed study has more innovation than except (Egede and Ellis 2010). However, until now, the mechanism for different animal model. on depression complicated with diabetes is not quite clear. This study has provided the relatively complementary ev- In recent years, increasing evidence indicates that BDNF idence. As stated in the article, GP at the doses of 50 and may play important roles in the pathogenesis of diabetes as 100 mg/kg by ig not only can decrease blood glucose levels, well as depression, and contribute to the association of diabe- but also alleviate depression-like behavior in STZ-evoked di- tes and depression (Alzoubi et al. 2013; Caletti et al. 2015;El- abetic mice. To the author’sknowledge,thisisthefirststudy Marasy et al. 2014;Lenartetal.2016;WangandHe2015; to address the anti-depressant effects of GP in STZ-evoked Wang et al. 2012;Zhouetal.2013). Therefore, the present diabetic mice. study further evaluated the anti-depressant mechanisms of GP STZ-evoked diabetes is a well-established animal model of involved in BDNF on STZ-evoked diabetic mice. type 1 diabetes that results in marked hyperglycaemia, proba- STZ-evoked diabetic mice demonstrated decreased BDNF bly through the destruction of pancreatic β cells and lack of levels in hippocampus by comparison with the normal mice. insulin secretion (Ho et al. 2012). In the present study, GP This finding is consistent with a previous study by Pachauri attenuated STZ-evoked hyperglycaemia in a manner similar et al. (2013). In the present study, GP effectively reversed to that of FHMH. This finding is in accordance with the pre- STZ-evoked decreases in BDNF levels, and up-regulated the vious studies of Liu et al. (2013) and Wu et al. (2009), and mRNA expression of BDNF and its cognate receptor TrkB in suggests that the anti-depressant effect of GP may be, in part, hippocampus of diabetic mice. In fact, the BDNF-enhanced due to its attenuation of hyperglycaemia in diabetic mice. effects of G. jasminoides (GJ) or GP had been reported in STZ-evoked diabetic mice demonstrated elevated immo- previous studies. For one example, 95 % ethanol extract of bility duration in the FST by comparison with the normal GJ produced rapid antidepressant activity from 2 h after a single mice. This finding is in line with previous studies by oral administration at the dosage of 0.8 g/kg in the TST, and the Anjaneyulu et al. (2003); ElBatsh (2015); El-Marasy et al. underlying mechanism was associated with up-regulation of (2014) and Gupta et al. (2014), and indicates the development BDNF expression in the hippocampus of mice (Zhang et al. of depressive-like behavior resulting from diabetes. In the 2015a, 2015b, 2015c). Besides, genipin, an important bioactive present study, GP effectively reversed STZ-evoked prolonga- compound from GJ, was demonstrated to possess tion of immobility duration in the FST. This effect was similar antidepressant-like effects with a CUMS-induced depression to that produced by the standard anti-depressant drug FH com- model in rats, and the possible mechanism, at least in one part, bined with anti-diabetic drug MH, and suggests the anti- resulted from monoaminergic neurotransmitter system and the depressant effect of GP in diabetic mice. This finding is in potential dysfunctional regulation of the post-receptor signaling accordance with a previous study by Cai et al. (2015), who pathway, which particularly affected the levels of BDNF as imparted anti-depressant-like effect of GP in a rat FST. well as 5-hydroxytryptamine 1 A receptor (5-HT1A)R and 5- To exclude the possibility that the antidepressant-like ef- HT(2 A)R in hippocampus (Wang et al. 2014). In addition, the fects of GP are attributable to stimulatory effects on locomo- BDNF-enhanced effect of GP had been reported in a previous tory function, we assessed GP-induced spontaneous activity in studybyZhangetal.(2006b), who reported that GP and baicalin the OFT. GP-treated mice exhibited no alterations in produced a synergy effect on the enhancement of BDNF Metab Brain Dis mRNA expression in hippocampus of rats undergoing focal Although GP can be hydrolyzed into genipin by β-D- cerebral ischemia-reperfusion injury. These results suggest that glucosidases in the intestine (Akao et al. 1994), previous stud- GP partially elicits antidepressant effects by the enhancement ies had still indicated that both GP and genipin exerted of BDNF expression. However, the present study is limited as antidepressant-like (Cai et al. 2015; Peng et al. 2014; Tian lack of other potentially mediated parameters observed particu- et al. 2013) and hypoglycemic (Wu et al. 2009; Zhang et al. larly involved in the monoaminergic neurotransmitter system 2006a; b) effects, suggesting that the prototype substance and as above mentioned. its metabolite can both be likely to participate in such actions. In addition, considering the recent report (Cai et al. 2015) As for the specific and exact participation between GP and its on antidepressant-like mechanism of GP probably involved in metabolite genipin, it needs further study. We have supple- its regulation of the hypothalamus-pituitary-adrenal (HPA) mented the exploration according to the reviewer. axis, we measured serum biomarker corticosterone (CORT), In conclusion, the present study indicates that GP produces an index of hypothalamic–pituitary–adrenal (HPA) axis hy- antidepressant-like effect in STZ-evoked diabetic mice, and peractivation. In this study, STZ-evoked mice demonstrated the underlying mechanism, at least in one part, may be in- elevated levels of serum CORT in accordance with a previous volved in the up-regulation of BDNF expression in hippocam- study by Revsin et al. (2008). GP did not change serum CORT pus, for the first time. level in either the VEH controls or the STZ-mice, at least from one part, indicating that the antidepressant effect of GP in Acknowledgments This work was financially supported by Funding diabetic mice may be not associated with HPA axis. This Scheme for Young Key Teachers of Colleges and Universities in Henan Province (2014GGJS-072), Zhengzhou Scientific and Technological finding is not in agreement with a previous study by Cai Attack Project (20150309), Provincial Fundamental Research Fund in et al. (2015), who reported the antidepressant-like effect of Henan University of Chinese Medicine (2014KYYWF-QN01), and GP on CUMS-induced depressive rats by regulating the Program for Science & Technology Innovation Talents in Universities hypothalamus-pituitary-adrenal axis. This discrepancy of the of Henan Province (16HASTIT032). data may depend on the differences between species and (or) Compliance with ethical standards the experimental procedures used. Moreover, other authors also verified that the ethanolic or Conflict of interest There is no conflict of interest to disclose. methanolic extract of GJ could slightly inhibit monoamine oxidase A and B (MAO-A/B) and dopamine β-Hydroxylase (DBH) activities in vitro assays, and the action of the extract can been verified by oral administration of rats (Kim et al. References 2012). Further research also showed that iridoids compounds, geniposide and genipin, are considered as major MAO-B in- Akao T, Kobashi K, Aburada M (1994) Enzymic studies on the animal hibitors because it was showed significantly selective MAO-B and intestinal bacterial metabolism of geniposide. Biol Pharm Bull 17(12):1573–1576 inhibition (Qu et al. 2014). However, until now, it is not known Alzoubi KH, Khabour OF, Alhaidar IA, Aleisa AM, Alkadhi KA (2013) whether GP can produce anti-depressant activity in STZ- Diabetes impairs synaptic plasticity in the superior cervical gangli- induced diabetic model by inhibiting MAO-A/B and DBH ac- on: possible role for BDNF and oxidative stress. J Mol Neurosci – tivities, which is likely to be as a next direction to our further 51(3):763 770 Anderberg RH, Richard JE, Hansson C, Nissbrandt H, Bergquist F, research. Skibicka KP (2016) GLP-1 is both anxiogenic and antidepressant; Besides, other researches indicated that glucagon-like pep- divergent effects of acute and chronic GLP-1 on emotionality. tide-1 (GLP-1) could improve depression (Anderberg et al. Psychoneuroendocrinology 65:54–66 2016) and GP was GLP-1 receptor angonist (Yin et al. Anjaneyulu M, Chopra K, Kaur I (2003) Antidepressant activity of quer- 2010b). As a receptor agonist of GLP-1R, which is a newer cetin, a bioflavonoid, in streptozotocin-induced diabetic mice. J Med Food 6(4):391–395 drug class to treat type 2 diabetes mellitus (T2DM), GP shows Cai L, Li R, Tang WJ, Meng G, Hu XY, Wu TN (2015) Antidepressant- clear effects in inhibiting pathological processes underlying like effect of geniposide on chronic unpredictable mild stress- Alzheimer’s disease (AD), such as promoting neurite out- induced depressive rats by regulating the hypothalamus-pituitary- growth (Liu et al. 2015). Besides, GP increases insulin secre- adrenal axis. Eur Neuropsychopharmacol 25(8):1332–1341 tion through GLP-1 receptors in rat INS-1 insulinoma cells Caletti G, Almeida FB, Agnes G, Nin MS, Barros HM, Gomez R (2015) Antidepressant dose of taurine increases mRNA expression of (Guo et al. 2012). Therefore, we speculate that GLP-1 maybe GABAA receptor α2 subunit and BDNF in the hippocampus of plays a critical role in antidepressant-like effects of GP in diabetic rats. Behav Brain Res 283:11–15 T2DM complicated with depression. The present study did Castagné V, Moser P, Roux S, Porsolt RD (2011) Rodent models of not design experiment to observe the mechanism of GP from depression: forced swim and tail suspension behavioral despair tests in rats and mice. Curr Protoc Neurosci Chapter 8:Unit 8.10 A. the perspective of GLP-1 receptor in antidepressant activity on Castillo-Gómez E, Coviello S, Perez-Rando M, Curto Y, Carceller H, STZ-induced diabetic animal model, which is likely to be as Salvador A, Nacher J (2015) Streptozotocin diabetic mice display another next direction to our further research. depressive-like behavior and alterations in the structure, Metab Brain Dis

neurotransmission and plasticity of medial prefrontal cortex inter- Peng GJ, Shi BY, Tian JS, Gao S, Qin XM (2014) 1H NMR based neurons. Brain Res Bull 116:45–56 metabonomics study on the antidepressant effect of genipin in rat Chen Y, Zhang Y, Li L, Hölscher C (2015) Neuroprotective effects of hippocampus. Yao Xue Xue Bao 49(2):209–216 geniposide in the MPTP mouse model of Parkinson’sdisease.EurJ Porsolt RD, Bertin A, Jalfre M (1977) Behavioral despair in mice: a Pharmacol 768:21–27 primary screening test for antidepressants. Arch Int Pharmacodyn Egede LE, Ellis C (2010) Diabetes and depression: global perspectives. Ther 229:327 Diabetes Res Clin Pract 87(3):302–312 Porsolt RD, Bertin A, Jalfre M (1978) "Behavioural despair" in rats and ElBatsh MM (2015) Antidepressant-like effect of simvastatin in diabetic mice: strain differences and the effects of imipramine. Eur J rats. Can J Physiol Pharmacol 93(8):649–656 Pharmacol 51(3):291–294 El-Marasy SA, Abdallah HM, El-Shenawy SM, El-Khatib AS, El- Qu K, Zhao L, Luo X, Zhang C, Hou P, Bi K, Chen X (2014) An LC-MS Shabrawy OA, Kenawy SA (2014) Anti-depressant effect of hesper- method for simultaneous determination of five iridoids from Zhi-zi- idin in diabetic rats. Can J Physiol Pharmacol 92(11):945–952 chi decoction in rat brain microdialysates and tissue homogenates: Gao C, Liu Y, Jiang Y, Ding J, Li L (2014) Geniposide ameliorates towards an in depth study for its antidepressive activity. J – learning memory deficits, reduces tau phosphorylation and de- Chromatogr B Anal Technol Biomed Life Sci 965:206 215 creases apoptosis via GSK3β pathway in streptozotocin-induced Rao AA, Sridhar GR, Srinivas B, Das UN (2008) Bioinformatics analysis alzheimer rat model. Brain Pathol 24(3):261–269 of functional protein sequences reveals a role for brain-derived neu- Gemeay EM, Moawed SA, Mansour EA, Ebrahiem NE, Moussa IM, rotrophic factor in obesity and type 2 diabetes mellitus. Med – Nadrah WO (2015) The association between diabetes and depres- Hypotheses 70(2):424 429 sion. Saudi Med J 36(10):1210–1215 Revsin Y, van Wijk D, Saravia FE, Oitzl MS, De Nicola AF, de Kloet ER Guo LX, Xia ZN, Gao X, Yin F, Liu JH (2012) Glucagon-like peptide 1 (2008) Adrenal hypersensitivity precedes chronic hypercorticism in – receptor plays a critical role in geniposide-regulated insulin secre- streptozotocin-induced diabetes mice. Endocrinology 149(7):3531 tion in INS-1 cells. Acta Pharmacol Sin 33(2):237–241 3539 Rhee EJ (2015) Diabetes in Asians. Endocrinol Metab (Seoul) 30(3): Gupta D, Kurhe Y, Radhakrishnan M (2014) Antidepressant effects of 263–269 insulin in streptozotocin induced diabetic mice: modulation of brain serotonin system. Physiol Behav 129:73–78 Sun P, Chen JY, Li J, Sun MR, Mo WC, Liu KL, Meng YY, Liu Y, Wang F, He RQ, Hua Q (2013) The protective effect of geniposide on Haghighi M, Salehi I, Erfani P, Jahangard L, Bajoghli H, Holsboer- human neuroblastoma cells in the presence of formaldehyde. BMC Trachsler E, Brand S (2013) Additional ECT increases BDNF- Complement Altern Med 13:152 levels in patients suffering from major depressive disorders com- Tian JS, Shi BY, Xiang H, Gao S, Qin XM, Du GH (2013) 1H-NMR- pared to patients treated with citalopram only. J Psychiatr Res based metabonomic studies on the anti-depressant effect of genipin 47(7):908–915 in the chronic unpredictable mild stress rat model. PLoS One 8(9): Ho N, Balu DT, Hilario MR, Blendy JA, Lucki I (2012) Depressive e75721 phenotypes evoked by experimental diabetes are reversed by insu- Wang J, He M (2015) Levels of serum brain-derived neurotrophic factor lin. Physiol Behav 105(3):702–708 in type 2 diabetes mellitus patients with and without depressive Islam SM, Rawal LB, Niessen LW (2015) Prevalence of depression and symptoms. Acta Biochim Biophys Sin Shanghai 47(2):137–138 its associated factors in patients with type 2 diabetes: a cross- – Wang J, Zhao X, He M (2012) Is BDNF biological link between depres- sectional study in Dhaka, Bangladesh. Asian J Psychiatr 17:36 41 sion and type 2 diabetes mellitus? Med Hypotheses 79(2):255–258 Kamei J, Miyata S, Morita K, Saitoh A, Takeda H (2003) Effects of Wang QS, Tian JS, Cui YL, Gao S (2014) Genipin is active via modulat- selective serotonin reuptake inhibitors on immobility time in the tail ing monoaminergic transmission and levels of brain-derived neuro- suspension test in streptozotocin-induced diabetic mice. Pharmacol trophic factor (BDNF) in rat model of depression. Neuroscience – Biochem Behav 75(2):247 254 275:365–373 Kerman IA (2012) New insights into BDNF signaling: relevance to major Wu SY,Wang GF, Liu ZQ, Rao JJ, Lü L, Xu W, Wu SG, Zhang JJ (2009) depression and antidepressant action. Am J Psychiatry 169(11): Effect of geniposide, a hypoglycemic glucoside, on hepatic regulat- – 1137 1140 ing enzymes in diabetic mice induced by a high-fat diet and Kim JH, Kim GH, Hwang KH (2012) Monoamine oxidase and dopamine streptozotocin. Acta Pharmacol Sin 30(2):202–208 β -hydroxylase inhibitors from the fruits of Gardenia jasminoides. Xue W, Zhou X, Yi N, Jiang L, Tao W, Wu R, Wang D, Jiang J, Ge X, – Biomol Ther (Seoul) 20(2):214 219 Wang Y, Wu H, Chen G (2013) Yueju pill rapidly induces Lenart L, Hodrea J, Hosszu A, Koszegi S, Zelena D, Balogh D, et al. antidepressant-like effects and acutely enhances BDNF expression (2016) The role of sigma-1 receptor and brain-derived neurotrophic in mouse brain. Evid Based Complement Alternat Med 2013: factor in the development of diabetes and comorbid depression in 184367 streptozotocin-induced diabetic rats. Psychopharmacology. Yao DD, Shu L, Yang L, Jia XB (2014) Advance in studies on anti- doi:10.1007/s00213-016-4209-x diabetic mechanism of Gardeniae Fructus and its active ingredient Li CF, Chen XM, Chen SM, Mu RH, Liu BB, Luo L, Liu XL, Geng D, geniposide. Zhongguo Zhong Yao Za Zhi 39(8):1368–1373 Liu Q, Yi LT (2016) Activation of hippocampal BDNF signaling is Yin F, Liu J, Zheng X, Guo L, Xiao H (2010a) Geniposide induces the involved in the antidepressant-like effect of the NMDA receptor expression of heme oxygenase-1 via PI3K/Nrf2-signaling to en- antagonist 7-chlorokynurenic acid. Brain Res 1630:73–82 hance the antioxidant capacity in primary hippocampal neurons. Liu J, Zhang Y, Deng X, Yin F (2013) Geniposide decreases the level of Biol Pharm Bull 33(11):1841–1846 Aβ1-42 in the hippocampus of streptozotocin-induced diabetic rats. Yin F, Liu JH, Zheng XX, Guo LX (2010b) GLP-1 receptor plays a Acta Biochim Biophys Sin Shanghai 45(9):787–791 critical role in geniposide-induced expression of heme oxygenase- Liu W, Li G, Hölscher C, Li L (2015) Neuroprotective effects of 1 in PC12 cells. Acta Pharmacol Sin 31(5):540–545 geniposide on Alzheimer’s disease pathology. Rev Neurosci 26(4): Zhang CY, Parton LE, Ye CP, Krauss S, Shen R, Lin CT, Porco JA Jr, 371–383 Lowell BB (2006a) Genipin inhibits UCP2-mediated proton leak Pachauri SD, Verma PR, Dwivedi AK, Tota S, Khandelwal K, Saxena JK, and acutely reverses obesity- and high glucose-induced beta cell Nath C (2013) Ameliorative effect of noni fruit extract on dysfunction in isolated pancreatic islets. Cell Metab 3(6):417–427 streptozotocin-induced memoryimpairmentinmice.Behav Zhang ZJ, Li P, Wang Z, Li PT, Zhang WS, Sun ZH, Zhang XJ, Wang YY Pharmacol 24(4):307–319 (2006b) A comparative study on the individual and combined Metab Brain Dis

effects of baicalin and jasminoidin on focal cerebral ischemia- associated with upregulation of BDNF expression in the hippocam- reperfusion injury. Brain Res 1123(1):188–195 pus. Evid Based Complement Alternat Med 2015:761238 Zhang W, Xu H, Zhao S, Yin S, Wang X, Guo J, Zhang S, Zhou H, Wang Zhang Y, Yin F, Liu J, Liu Z, Guo L, Xia Z, Zidichouski J (2015c) F, Gu L, Zhu L, Yu H, Qu Z, Tian D (2015a) Prevalence and Geniposide attenuates insulin-deficiency-induced acceleration of influencing factors of co-morbid depression in patients with type 2 β-amyloidosis in an APP/PS1 transgenic model of Alzheimer’sdis- diabetes mellitus: a general hospital based study. Diabetol Metab ease. Neurochem Int 89:7–16 Syndr 7:60 Zhou JX, Li HC, Bai XJ, Chang BC, Li CJ, Sun P, Chen LM (2013) Zhang H, Xue W, Wu R, Gong T, Tao W, Zhou X, Jiang J, Zhang Y, Functional Val66Met polymorphism of brain-derived neurotrophic Zhang N, Cui Y, Chen C, Chen G (2015b) Rapid antidepressant factor in type 2 diabetes with depression in Han Chinese subjects. activity of ethanol extract of Gardenia jasminoides Ellis is Behav Brain Funct 9:34 本文献由“学霸图书馆-文献云下载”收集自网络，仅供学习交流使用。

学霸图书馆（www.xuebalib.com）是一个“整合众多图书馆数据库资源，

提供一站式文献检索和下载服务”的24 小时在线不限IP 图书馆。图书馆致力于便利、促进学习与科研，提供最强文献下载服务。

图书馆导航：

图书馆首页文献云下载图书馆入口外文数据库大全疑难文献辅助工具