Abstract
Metabolic disorder particularly diabetes is one of the leading causes of psychiatric or other neurodegenerative diseases. Previous clinical and pre-clinical studies indicate anti-diabetic drugs such as GLP- 1 analogs or GLP- 1 receptor (GLP- 1R) agonists could perform the neuroprotective effects with multiple molecular mechanisms. As one of natural compound to stimulate GLP- 1R,geniposide was reported could improve cognitive behaviors in diabetes associated Alzheimer’s disease rat model. Stimulating of GLP- 1R could act the crosstalk downstream like neurotrophic factor mediated cAMP-response element binding protein (CREB) would be activated and exert cellular events including promotion of adult neurogenesis, which is one of important treatment targets in antidepressant. Here in this study, we employed HDF in combined with corticosterone (CORT) treatment to create diabetes associated depression model. Geniposide treatment could not only correct the metabolic pattern but could also improve the cognitive dysfunctions and depressive/anxiety symptoms. In consistent with its pro-neurogenic effects, geniposide also enhanced the activity of CREB in hippocampal tissue. Moreover, blocking CREB activity with 666- 15 significantly compromised the effects of geniposide in promotion of neurogenesis and behavioral protective effects. In conclusion, this study expands the application of geniposide to treat diabetes associated depression subject and identified the underlying molecular mechanism for such effects.
Keywords:Geniposide, Diabetes, Depression, Neurogenesis, CREB
Introduction
Neurodegenerations and psychiatric diseases are currently serious healthy issue with heavy burden to individual and families. Patients with such disorders suffer multiple behavioral disabilities and suppressed brain functional plasticity. Neural plasticity is the functional base for maintaining brain’s ability to cope with environmental challengers (Muller et al., 2019). In adult brain,hippocampus plays the primary role in memory processing and emotional regulations (Chenani et al., 2019; Madan et al., 2018; Sanchez-Rodriguez et al., 2019). Synaptic
plasticity as well as adult neurogenesis offer the optimal functional and structural flexibility that enable the behavioral control of hippocampus (Gu et al., 2013). Continuous generation of new neurons in hippocampal dentate gyrus (DG) region can enhance the memory and environmental adaptability (Morcuende et al., 2003). Most memorial improvement drugs and antidepressants can improve the behaviors by promoting adult hippocampal neurogenesis (AHN) (Beckervordersandforth et al., 2017; Deyama and Duman, 2020). It was reported that AHN was significant declined during psychiatric diseases or process of neurodegenerations, like depression or Alzheimer’s disease (Valero et al., 2017; Yuan et al., 2015). Thus, improving AHN is one of effective strategies to promote neuronal regeneration and attenuate brain dysfunctions.
Metabolic disorders particularly diabetes is one of biggest risk factors associating with the development of neurodegeneration or psychiatric disorders (Habibi et al., 2017). Cell metabolism is highly involved with regulation of multiple neural functions. Mitochondrial functions and dynamics were reported plays the critical roles in regulating cell fate commitment of neural stem cells (NSCs) (Beckervordersandforth et al., 2017; Khacho et al., 2016). Memorial drug piracetam was also shown could improve mitochondrial oxidative phosphorylation
(OXPHOS) and enhance AHN in ageing brain (Beckervordersandforth et al., 2017). Regulating other types of metabolic factors, like insulin growth factor- 1 (IGF- 1), adiponectin or its downstream could promote AHN and improve brain functions (Gao et al., 2018; Sharma et al., 2016; Yau et al., 2018). Incretins is one group of important metabolic factors. Glucagon-like Peptide- 1 (GLP- 1) is an critical incretin participating in regulating glucose level and perform as the important drug target to treating diabetes (Aroda, 2018). Apart from other GLP- 1 analogs, natural product geniposide has been reported as the GLP- 1 agonist and acted neuroprotective effects in AD-like rat model (Gao et al., 2014). By ameliorate neuronal apoptosis via regulating GLP- 1R/AKT signaling pathway,geniposide could also improve depression impaired emotional dysregulations (Zhao et al., 2018). More importantly, geniposide could improve glucose homeostasis by inhibiting FoxO1/PDK4 signaling (Li et al., 2019). As G-protein coupled receptor, GLP- 1R could stimulate multiple signaling downstream for regulating hippocampal structural plasticity. Vildagliptin (Vilda), a dipeptidyl peptidase-4 (DPP-4) inhibitor, boosted striatal neurotrophic factors and performed neuroprotection with restored imbalance of neurotransmitters including GABA and glutamate (Sayed et al., 2019). This evidence indicates GLP- 1 may also participate in the activation of BDNF signaling. It is likely that geniposide could perform as the antidepressant effects in diabetic caused depression model via promoting AHN. However, whether BDNF/CREB signaling is the key mechanism underlying such functions of geniposide remain unknown.
Neurotrophins like BDNF as well as growth factor like EGF or VEGF are the important neurotrophic factors in promote synaptic plasticity and AHN (Bhattarai et al., 2020; Sjors
Dahlman et al., 2019). By stimulating the downstream signaling and activate cAMP-response element binding protein (CREB), growth factors or neurotrophic factors could induce the certain gene expression and enhance the neurogenesis (Ebrahimzadeh et al., 2019; Zhang et al., 2019b). Thus, CREB is the key transcriptional factor in response for promoting AHN. In this study, we detected the effects of geniposide to the behavioral functions and AHN by creating high-fat diet combining with chronic stress model. Moreover, we also tested the role of CREB signaling in neuroprotective effects of geniposide by inhibiting CREB signaling with chemical approach.
Material and Methods
Animal
Adult male 8 weeks old C57BL/6J mice were obtained from Laboratory Animal Unit, Shanxi Medical University. Mice was raided with ad libitum of tap water as well as food pellets with 12:12 h light/dark cycle during whole period. The animal experiment was approved by ethical committee in Shanxi Medical University in regard of Guidelines for the Care and Use of
Mammals in Neuroscience and Behavioral Research (2003). Temperature (21 ± 2 °C) and humidity (23 ± 2%) were monitored and maintained at constant level.The high-fat diet (HFD) consisted of 58% fat from lard, 25.6% carbohydrate, and 16.4% protein (total 23.4 kJ/g), whereas the normal diet contained 11.4% fat, 62.8% carbohydrate, and 25.8% protein (total 12.6 kJ/g) (Winzell andAhren, 2004). Food intake and body weight were measured once a week, and blood samples were taken at indicated time points from injection of glucose.
HFD was performed for 60 days, during which corticosterone (CORT; 70μg/ml, 0.45% beta-cyclodextrin, in drinking water) was administrated to mice from 30th day of HFD modelling.
Treatment groups with low (20 mg/kg, 4 mg/ml in saline) and high (100 mg/kg, 25 mg/ml in saline) by oral administration at the beginning of CORT treatment. CREB inhibitor 666- 15 (20 mg/kg, 4 mg/ml in saline) was treated to 100 mg/kg geniposide treatment group together by I.P. injection.
OGTT, plasma insulin and GLP-1 assay
For OGTT, mice were fasted for 18 h before the OGTT. Each mouse group was orally administered saline before glucose gavage (5 g/Kg). Blood glucose was measured from the tail vein using an Accu-Check Performa system (Roche Diagnostics, Mannheim, Germany) at 6 time points: 0 (before glucose gavage), 30 (after glucose gavage), 60, 120, 180, and 300 min.Mice were fasted for 18 h before the experiments. Each mouse group was orally administered saline before the glucose gavage (5 g/Kg). Collected blood samples were centrifuged at 1,000 × g for 20 min at 4 °C, and the plasma was carefully separated into fresh tubes. A multiplex assay (Mouse Diabetes panel: total GLP- 1 and insulin; Bio-Rad) was performed as described in the manufacturer’s instructions. The total GLP- 1 and insulin concentrations in each sample were measured using a Bio-Plex MAGPIX Multiplex reader (Bio-Rad) with sametime points as OGTT. The results were analyzed with Bio-Plex Manager software (Bio-Rad).
Behavioral Tests
Morris water maze
Mice was rested for 2 days after blood collection. Morris water maze (MWM) was started with acquisition task for assess the spatial learning behavior. Briefly, each mouse was put into water for 60 sec for freely searching the hidden platform and forming spatial memory in reference with the marks on insidewall of water tank. The escaping latency was recorded, and the acquisition task was performed for 4 consecutive days. Probe trail was acted in the 5th day by remove the platform. Mouse was then put into the water for a 60 sec free swim. Swimming time in original platform zone was recorded for evaluating the memory ability. Visible platform test was performed for avoid the interference from the visual and motor ability induced by treatments.
Sucrose preference test
Adhesion behavior was detected with sucrose preference test (SPT). 2 days after MWM, mice were single cage raised with two bottles of water (tape water and 2% sucrose with random sides). After 3 days of habituation, all water bottles were deprived for 18 h and the resupplied for 1 h. All sucrose preference index (sucrose consumption/total water consumption) each day was recorded.
Forced swim test
Mice was put into the cylinder water tank (diameter 15 cm; height 30 cm) and allowed for free swim for 6 min. Immobility time in last 4 min exclude struggle and swim were recorded for
evaluating the depressive mood.
Open field test
The anxiety mood of mice was detected with open field. Mice were put into the open field (40 cm -40 cm -40 cm H -L -W) and allowed for free exploration for 10 min, during which the duration of mice in center region (20 cm-20 cm L-W) was recorded for evaluating anxiety mood.
Immunofluorescence
Mice after behavioral tests was then sacrificed with 4% PFA cardiac perfusion. Brain tissue was collected and embedded with OCT frozen medium. Hippocampal sections were prepared, and
tissue sections was retrieved with pH6.0 citrate acid buffer in microwave oven. Tissue sections was then blocked with 5% BSA buffer and 0.3% TritonX- 100. Primary antibody (Rabbit-DCX; 1:500; CST) was incubated overnight at 4°C. Afterward, sections was washed with PBST for 3 times (15 min per time), and secondary antibody (Goat-anti-Rabbit, Alexa 568; 1:1000; Thermo Fisher) was incubated with sections for 2 h. DAPI was incubated with section for 10 min.Prolonged antifade mountant medium (Thermo Fisher) was used for section mountant. Image was obtained with Nikon C2 confocal microscope.
Western blot
Fresh hippocampus was separated after anesthesia. Protein lysate was prepared with RIPA buffer with protease inhibitor and protein phosphatase inhibitor. SDS-PAGE was performed and primary antibodies (Rabbit-anti-CREB; 1:1000; Rabbit-anti-pCREB (Ser133); 1:1000) was incubated with PVDF membrane overnight at 4°C. Secondary antibody (Goat-anti-Rabbit-HRP; 1:5000) was
incubated with membrane at RT for 2 h. Western blot bands was visualized with chemoluminescence (GE, Healthcare Life Science), and quantified by Quantify One (Bio-rad).
Statistical analysis
All data was shown as mean±S.E.M. Two-way or one-way ANOVA was used for comparing multiple groups with two or one factor. Tukey’s post-hoc was followed for comparing difference among each condition. Two tailed unpaired student t-test was used for comparing two treatment groups. Probabilities of p<0.05 were considered as statistical significance.
Results
Geniposide reduced the metabolic symptoms in HFD mice
We first tested the plasma glucose, insulin and GLP- 1 level to evaluate whether Geniposide could attenuate HFD induced metabolic abnormalities. During the HFD administration, CORT was
simultaneously orally administrated to the mice from the 30th day of HFD for further establish the depression condition (Fig 1 A, N=10 mice per group). As the result indicated, HFD induced an impaired glucose tolerance with significant growing plasma glucose concentration after the glucose injection (Fig 1 B, two-way ANOVA; E, one-way ANOVA). Moreover, the plasma levels of insulin and GLP1 was significantly suppressed Bioprinting technique by HFD treatment (Fig 1 C, F, two-way ANOVA; D, G, one-way ANOVA). In comparing with HFD group, 100 mg/kg geniposide daily treatment decreased glucose level and increased plasma insulin and GLP- 1 concentration (Fig 1 B~G). 20 mg/kg geniposide treatment did not change the GLP- 1 level in significant matter (Fig 1, B~G). Therefore, geniposide treatment could reduce the HFD-associated metabolic abnormalities in a dose dependent manner.
Geniposide improved HFD-STRESS impaired cognitive and emotional functions
We further conducted the morris water maze (MWM) to detect the learning and memory behaviors of mice. As result shown, HFD combining with CORT treatment impaired spatial learning in
MWM navigation training phase. The escaping latency of HFD-STRESS group was dramatically prolonged in comparing with control (Fig 2 A, two-way ANOVA). Moreover, HFD-STRESS also resulted in the dramatical decreased time percentage in target zone in the probe trail, indicating the impaired spatial memory of mice (Fig 2 B, one-way ANOVA). Compared with HFD-STRESS model, geniposide especially high dosage treatment established shorter escape latency in navigation training and increased the time percentage in target zone during probe trail (Fig 2 A,
B). All groups showed no significance on escape latency invisible platform test, which helped to avoid the interference induced by locomotor and visible variations (Fig 2 C, one-way ANOVA).This result suggests geniposide could attenuate the HFD-STRESS impaired cognitive functions. We then conduced SPT for evaluating the anhedonia emotion. During the habituation and test phase, HFD-STRESS decreased the preference of mice to sucrose, indicating the anhedonia symptom (Fig 2 D, two-way ANOVA). While treatment of geniposide in high dosage restored the preference index during habituation and test phase (Fig 2 D). In FST, compared to HFD-STRESS, geniposide treatment in both dosages dramatically decreased the immobility of the mice in water tank, indicating the attenuated depressive emotion (Fig 2 E, two-way ANOVA). While in OFT,compared to HFD-STRESS, geniposide treatment in high dosage dramatically increased time in central region of mice, suggesting its anxiolytic effect (Fig 2 F, two-way ANOVA). Therefore,geniposide could improve HFD-STRESS impaired cognitive and emotional functions.
Geniposide could improve adult hippocampal neurogenesis in HFD-STRESS mice
We further conducted immunofluorescent staining to detect the AHN profile in different groups. Doublecortin (DCX) was used for labelling developing neurons (Fig 3 A). To evaluate the
generation and growth of developing neurons, we recorded the somatic density of DCX+ cells as well as the enrichment of DCX+ fibers in molecular layer (ML) region of hippocampus. As the result indicated, HFD-STRESS dramatically decreased the somatic number of DCX+ cells in hippocampal DG region. While high dosage of geniposide administration remarkably restored the number of Tissue Slides DCX+ cells (Fig 3 A, B, one-way ANOVA), indicating the capacity of geniposide to improve the neuronal generation in adult DG. Moreover, HFD-STRESS also decreased enrichment of DCX+ dendrites in ML. Treatment of geniposide in both dosages increased number of DCX+ dendrites compared with HFD-STRESS (Fig 3 A, C, one-way ANOVA). Thus,geniposide can also enhance the dendritic enrichment and promote the growth or maturation of developing neurons in adult hippocampus.
CREB-mediated neurogenic mechanism promotes behavioral effects of geniposide
We first conducted immune-blot assay for evaluating effect of geniposide to CREB activity.Western blot showed that treatment of geniposide at high dosage significantly restored the
phosphorylation level of CREB at Ser133 (Fig 4 A, B, one-way ANOVA). The result indicates CREB might be one of molecular target of geniposide. We then administrated the specific CREB
inhibitor 666- 15 to geniposide treated HFD-STRESS model (Fig 4 C). Administration of 666- 15 prohibited the density of immature neurons including the DCX+ immature neuronal density as
well as the enrichment of DCX+ neuronal fiber in ML (Fig 4 D,H, I, student t-test). Noteworthy,666- 15 treatment dramatically suppressed the spatial learning and memorial capacity in MWM
(Fig 4 E, two-way ANOVA; F, student t-test). By comparing the immobility in FST, we found 666- 15 prohibited the effects of geniposide in attenuating depressive mood (Fig 4 G, student t-test).Collectively, inhibition of CREB could compromise the pro-neurogenic effects of geniposide and thereby prohibited its behavioral function including cognitive and antidepressant patterns.
Discussion
Metabolic factors were widely reported could serve as the regulatory target to promote AHN for exerting antidepressant effects. GLP- 1 receptor agonist geniposide has been described to perform the neuroprotective effects in neurodegenerative model. In this study, we identified the antidepressant effects of geniposide to diabetes associated depression mice model. Geniposide treatment cannot only attenuate the metabolic dysfunctions but can also improve psychiatric adaptability. Furthermore, our investigation suggested that CREB-mediated AHN is the biological underpinning of geniposide for its antidepressant effects.Geniposide can serve as the metabolic regulator. As the agonist of GLP- 1R, treatment of geniposide attenuates the glucose tolerance, and normalized the peripheral insulin and GLP- 1 level. Long-term exposure to a HFD causes glucotoxicity and lipotoxicity in islet β cells and leads to the development of diabetes. In HFD model, GLP- 1 analog liraglutide attenuated the glucose tolerance, improved insulin release, and glucose-dependent insulinotropic polypeptide level (Hao et al., 2017). In this study, we first created the HFD induced diabetic condition to mimic diabetic associated depression. Streptozotocin (STZ) ventricular injection induced the neural toxicity for evaluating pharmacological functions in diabetic related neurodegeneration (Gao et al., 2014). As such notion, to establish the inhibitory status of neurogenesis instead of neural damage, we employed HFD for creating diabetic condition. In HFD-CORT combined mice model, treatment of geniposide improved the glucose regulation and restored the insulin and GLP- 1 production (Fig 1). In consistency, via regulating AMPK signaling and Sirt1, geniposide could also ameliorate HDF-induced cardiac injury (Ma et al., 2018). Thus, our study identified the effects of geniposide in normalizing metabolic profile in diabetic model.
Diabetes is the risk factor to induce depression and impaired AHN. Increased activity of hypothalamus-pituitary-adrenal (HPA) axis could result in over level of glucocorticoid in
depression subjects. Increased concentration of CORT in mice could result in inhibitory growth of NSCs (Lui et al., 2017). 4 months of HFD treatment causes disrupted intracellular cascades
involved in synaptic plasticity and insulin signaling/glucose homeostasis and thereby result in depressive and anxiety like behaviors (Dutheil et al., 2016). In our study, 2 months of HFD
administration in combining with 1 months of CORT treatment induced depressive and anxiety behaviors (Fig 2). Geniposide treatment reserved the depressive and anxiety symptoms induced by diabetic associated depression (Fig 2). In chronic unpredictable mild stress (CUMS) depression model, penta-acetyl geniposide presented the effects to increase sucrose intake, increase total crossing and rearing numbers, improve central activity and reduce immobility time (Cai et al.,2020).The combination of geniposide and eleutheroside B showed a certain antidepressant-like effect by decreasing the immobility in FST and TST (Zhang et al., 2019a). In consistent with such evidence, our study further expands the clinical application of geniposide in regulating psychiatric functions in depression model associated with diabetic conditions.
High-fat diet (HFD) has been demonstrated to induce an inflammatory response and to alter neurogenesis in the hypothalamus and functional outcome measures, e.g. body weight (Klein et
al., 2019). In our study, we detected that HFD combined CORT could inhibit AHN by decreasing immature neuronal density (Fig 3). Moreover, dendritic fiber of DCX+ neurons expanded into ML
subregion was also enriched after treatment of geniposide, indicating that geniposide could not only promote the neuronal differentiation of NSCs, but could also enhance the growth during
maturation of newborn neurons (Fig 3). Previous study showed that geniposide ameliorated fluoxetine-suppressed neurite outgrowth in cultural neuroblastoma cells, indicating geniposide treatment could help to avoid the side effects of classic antidepressants by promoting neuronal growth (Chen et al., 2019). Fostamatinib inhibitor Enhancing AHN could improve the antidepressant behaviors in
depression model (Park, 2019). As one part of important structural plasticity in hippocampus,immature neuronal density is critical for regulating pattern separation behaviors and clear panic memory (Miller and Sahay, 2019). Thus, we focused on evaluating the density of DCX+ immature neurons in treatment groups. By enhancing neuronal differentiation and growth of neuronal fiber, geniposide exerts the antidepressant effects in depression model.AHN is rigorously regulated by complex mechanisms including growth factors, neurotrophic factors, metabolic factors and neurotransmitters (Gao and Shen, 2017). In this study, geniposide may activate the GLP- 1R and serve as the metabolic factor to stimulate the down-streams of above-mentioned neurogenic factors by cell stimulating signaling crosstalk. CREB is the downstream transcript-factor that could be stimulated by different factors including BDNF and EGFR (Kim et al., 2020; Liu et al., 2019). Liraglutide, the analog of GLP- 1, could also exhibit anti-inflammatory activity through the activation of the PKA/CREB pathway (Wang et al., 2019). In our study, we detected treatment of geniposide could also increase the activity of CREB in hippocampal tissue. To further confirm the role of the CREB in geniposide induced antidepressant effects, we treated the geniposide administration group with specific inhibitor 666- 15. Inhibition of CREB blocked the effects of geniposide in promoting AHN and improving antidepressant behaviors (Fig 4). To our knowledge, this is the first identified molecular mechanism underlying the antidepressant and neurogenic promotion effects of geniposide. Since BDNF signaling mediated CREB could also enhance the synaptic plasticity and improve cognitive behavior,geniposide could be also serve as the drug to treat neurodegeneration (Rafa-Zablocka et al., 2019). As such, combination of geniposide with other psychiatric medication may also be effective strategy for neurological disorders.
In conclusion, our study identified the new function of geniposide in diabetes associated depression mice model. CREB-mediated adult neurogenesis promotion serves as the critical
mechanism underlying the effects of geniposide in antidepressant function. This study provides the new application of natural compound geniposide in treating metabolic disorder induced
depression. Moreover, geniposide could also be used as the drug to promote neural regeneration and simultaneously regulate metabolic profile. Further researches are worthwhile for exploring the effects of geniposide in other neurological model such as neural injury and sophisticated neural disorders e.g. Amyotrophic Lateral Sclerosis (ALS), schizophrenia and Alzheimer’s disease.