The Neurotrophic and Anti-Inammatory Effect of β-asarone on the DA-Induced the Pathology of Minimal Hepatic Encephalopathy

Background: The mechanism underlying the impaired cognitive function and memory loss in Minimal hepatic encephalopathy (MHE) remains unclear. Dopamine (DA) is reported to be associated with dementia. Methods: In this study, we investigated mechanism underlying DA-induced MHE pathology by immunoblotting, ELISA, FM4-64 and uorescence staining. Results: We observed that MHE brains showed the increased content of DA, after administration of anti-DA antibody, and cognitive loss in MHE rats was recovered to the normal level, indicating the involvement of DA in the pathogenesis of MHE. Moreover, DA (10 μM) treatment obviously induced the decrease in the production of GDNF/NGF and the increase in TNFα levels in primary cultured neurons, which were blocked by addition of β-asarone (βASA). We also demonstrated that DA stimulated the activation of ASK1/JNK1 pathway. and the addition of anti-TNFα antibody reversed the inactivation of Notch signaling, the downregulation of neurotrophins and synaptic loss. Conclusions: Overall, we suggested that DA stimulated abundant production and secretion of neuronal TNFα, which elicited progressive loss of neurotrophic factors, leading to cognitive disorder of MHE.


Introduction
Brain dopamine (DA) has been implicated in cognitive processes such as working memory and cognitive exibility [1] and the DA system is involved in cognitive decline [2]. Our previous studies have con rmed that the pathogenesis of MHE was caused by the abnormal elevation of intracranial DA [3,4,5,6,7,8]. It has been demonstrated as a novel regulatory mechanism that DA -induced astrocytic TNF-α release triggers progressive neurodegeneration, which leads to learning and memory decline in MHE [9]. Tumor necrosis factor-α (TNFα) signaling has been shown to regulate neuroin ammation, and plays roles in modulation of microglial activation [10], control of glutamatergic transmission and regulation of synaptic strength [11] in the central nervous system (CNS) [12]. Ample evidence suggests that elevated levels of TNFα have been implicated in some neurodegenerative disorders including Alzheimer's disease [13], Parkinson's disease [14] and HIV-associated dementia [15]. Hence, Neuroin ammatory conditions are often related with memory dysfunction [16,17,18],which suggests that DA may be in uenced on brain function involved with in ammatory factor. We further proposed that DA may regulate synaptic formation underlying long-term cognition with involvement of TNFα in MHE. Therefore, we hypothesized that the induction of cognitive decline by accumulation of DA levels in the MHE rats is associated with its ability to produce TNFα.
Here, we investigated (1) whether neuroin ammation was induced by DA in neurons in MHE; (2) whether neurons interference in disruption of synaptic formation triggered by DA in MHE. Moreover, β-asarone increased the expression of neurotrophin such as brain-derived neurotrophic factor (BDNF), glial cellderived neurotrophic factor, and ciliary neurotrophic factor in RSC96 cells [19]. β-asarone treatment also improved the spatial learning and memory in Aβ (1-42)-induced memory-impaired rats [20]; in transgenic model of Alzheimer's disease such as Senescence-accelerated prone-8 (SAMP-8) mice [21]; and AβPP/PS1 double transgenic mice [22,23],also in chronic exogenous corticosterone-induced memoryimpaired rats [24,25]. Then, we describe that the attribution of pathogenesis of MHE is upregulation of DA, followed by the production of TNFα from neurons, leading to the disruption of synaptic formation, which was diminished by βASA treatment.

MHE models and treatment
A total of 40 Sprague-Dawley (SD) rats (220-250 g) were used, and all guidelines of the care and use of animals for experimental procedures were laid down by the Ethics Committees of the A liated Hospital of Wenzhou Medical University regarding [26].
Before experimenting, all animals were subjected to behavioural tests: Y-maze (YM) and water-nding task (WFT

Behavioral test
Rats were individually placed at the end of one of arm of the three arms in the apparatus for YM, and were allowed to nd the arm entries freely for 8 min. Total arm entries and spontaneous alternation percentage (SA%, a ratio of successful choices to total choices minus two) were measured [27].
Rats were deprived of water for 24 h and individually placed at the near-right corner of the testing apparatus for WFT and allowed to explore the tube and drink the water in the cubic alcove freely for 3 min. The latency times of the rst entry into the alcove (entry latency, EL), of the rst touching/sni ng/licking of the water tube (contacting latency, CL) and of the initiation of drinking from the water tube (drinking latency, DL) were measured [28].

Primary rats neurons culture and treatments
Primary hippocampal rats neurons (PHNs) were obtained from freshly dissected hippocampus and cerebral cortex from 1-day-old SD rat pups by mechanical disruption in the presence of trypsin and DNase and then seeded in poly-L-lysine-precoated six-well plates at a density of 2 × 10 6 cells/well in Neurobasal® Medium (1X) supplemented with 0.5 mM GlutaMAX™-I, B-27®. Then neurons remained untreated or were stimulated with DA (1, 10 or 20 µmol/l) alone or after preincubation with JNK1 inhibitor SP600125, ASK1 inhibitor K811, 50 µM Notch inhibitor MK-0752, 100 ng/ml Notch activator Jagged-1 or anti-TNFα antibody (ATAB), or treated with βASA in the presence or absence of DA together with MK-0752/TNFα.

PC12 cells culture and treatments
The rat pheochromocytoma PC12 cell line were achieved from the American Type Culture Collection (ATCC, Manassas, VA

Measurement of TNFα release
TNFα level in the culture medium of neurons was measured using enzyme-linked immunosorbent assay (ELISA) kit (R&D systems) in the light of the manufacturer's protocols and a Thermo-Fisher Multiskan MCC plate reader.
Densitometry was indicated as ratio of proteins to β-actin levels.

Statistical analysis
All of the data were expressed as mean ± SD. Data comparisons were analyzed using one-way analysis of variance (ANOVA). All of the data were detected in normal distribution and equal variances, when Back-ofthe-envelope test was used to verify the normal distribution, and F-test was applied for determining the equality of variances. Dunnett's post hoc multiple comparison test was applied when signi cant differences were detected by the ANOVA model. Then P values were made for adjustment by Bonferroni correction. The level of signi cance was determined for P < 0.05 or P < 0.01. All analyses were performed with SPSS 18.0 (PASW Statistics 18.0).

DA inhibits the production of neurotrophic factors in neurons
First, we assessed the effect of DA on the production of neurotrophic factors in neurons. As indicated by the IB analysis, the decreased trend on the expressions of glial cell line-derived neurotrophic factor (GDNF)/Nerve growth factor (NGF) were observed after lower DA treatment, higher DA treatment signi cantly decreased the two proteins levels in PHNs ( Fig. 1A and B). PC12 cells in response to DA showed that the expressions of GDNF/NGF were also reduced in a dose-dependent fashion ( Fig. 1C and D). Time course analyses via IB analysis of PHNs treated with high dose of DA revealed steady decreases in GDNF/NGF expression up to 48 h ( Fig. 1E and F). Double IF staining of PHNs using GDNF antibody combined with MAP2 antibody con rmed that high DA treatment induced the weak expression of GDNF ( Fig. 1G). As analyzed by immunostaining, PC12 in response to DA displayed signi cant decrease in NGF expression (Fig. 1H). These results suggested that DA dramatically triggered the loss of neurotrophins.

DA stimulated TNFα secretion from neurons
Recent studies have reported that DA is involved in the process of in ammation [29,30,31,32].
Therefore, we examined the effect of DA on the production of proin ammatory factor TNFα in neurons. Signi cant upregulation of TNFα was induced by higher DA treatment, but not by lower DA treatment in PHNs using IB analysis ( Fig. 2A and B). Otherwise, IB analysis of PC12 cells showed that DA also signi cantly triggered TNFα expression in a dose-dependent manner ( Fig. 2C and D). As determined by ELISA, DA signi cantly stimulated the release of TNFα with dose-dependence in PHNs (Fig. 2E). Parallelly, as shown by IB analysis, PHNs exhibited high TNFα levels in incubation of DA with a time-dependent manner ( Fig. 2F and G). By ELISA, DA signi cantly elevated release of TNFα with time-dependence from PHNs (Fig. 2H). Double IF staining using TNFα antibody with MAP2 antibody con rmed that the expression of TNFα was obviously increased in PHNs after DA treatment (Fig. 2I). Immunostaining of PC12 cells in response to DA displayed the strong expression of TNFα (Fig. 2J). These results suggested that DA induced the production of proin ammatory factor TNFα in neurons in vitro.

DA modulates production of TNFα by ASK1/JNK1 pathway in neurons
In this experiment, we studied the mechanism underlying DA-induced TNFα production. ASK1 upregulates c-Jun N-terminal kinase (JNK1) signaling [33]. In ammatory responses are frequently associated with cJun NH2-terminal kinase (JNK1) signaling pathway [34]. Thus, we examined the association of DAinduced production of TNFα with ASK1/JNK1 pathway in neurons. As determined by IB analysis, the phosphorylation of ASK1 was signi cantly increased in PHNs in response to DA with dose-dependent manner ( Fig. 3A and B). DA treatment signi cantly elevated the phosphorylation of JNK1 in PHNs, which was reversed by ASK1 inhibitor K811 by IB analysis (Fig. 3C and D). DA-mediated the increase in TNFα expression in PHNs was signi cantly diminished by addition of K811 or JNK1 inhibitor SP600125 by IB analysis (Fig. 3E and F). The e ciency of ASK1 (Fig. 3G) or JNK1 siRNA (Fig. 3H) plasmid transfection was con rmed by qPCR in PC12 cells. Then we found that the increase in TNFα expression in response to DA were signi cantly diminished by ASK1 or JNK1 silencing in PC12 cells by IB analysis (Fig. 3I and J). ELISA of supernatants from PC12 cells exhibited that DA-induced release of TNFα was blocked by knockdown of ASK1 or JNK1 (Fig. 3K).

DA induced downregulation of neurotrophins depending on TNFα in neurons
In this experiment, we tried to con rm whether DA-induced synaptic loss was associated with ASK1/JNK1/TNFα pathway. Parallelly, as shown by IB analysis, PHNs exhibited lower GDNF/NGF levels after DA treatment, addition of K811, SP600125 or ATAB blocked the decrease in GDNF/NGF expression ( Fig. 4A and B). The e ciency of TNFα siRNA plasmid transfection into PC12 cells was con rmed by qPCR (Fig. 4C). Parallelly, as shown by IB analysis, PC12 cells exhibited the increase in GDNF/NGF levels after knockdown of ASK1/JNK1/TNFα ( Fig. 4D and E). Furthermore, by IB analysis, exposure of PHNs to DA also displayed the decreased expression of Homer/synaptotagmin which were blocked by addition of K811, SP600125 or ATAB ( Fig. 4F and G). Furthermore, by immunostaining, GDNF expression in PC12 cells with DA treatment was markedly decreased, while K811, SP600125 or ATAB signi cantly increased the protein expression (Fig. 4H). Double IF staining of PHNs with anti-vGluT1 (for staining dendritic spines) and anti-MAP2 (for staining microtubules) antibodies for probing dendritic spine density showed that vGluT1-positive signals in PHNs was markedly increased after DA treatment DA, whereas ATAB administration diminished the decrease in vGluT1-positive signals ( Fig. 4I and J). These results suggested that DA impaired the production of neurotrophins and synaptogenesis through the activation of ASK1/JNK1/TNFα signaling pathway.

DA induced inactivation of Notch pathway in neurons via TNFα
Notch is also required for memory consolidation, a process believed to require remodeling of existing neurons in adults [35]. Several studies have shown that Notch also plays an important role in memory formation in adults, both in mammals and Drosophila [36,37]. It is reported that Notch also plays an important role in memory formation in adults [36,37]. Then we addressed whether DA impacted on notch signaling via the involvement of TNFα. The expressions of NICD1/HES1 were markedly decreased in PHNs in response to higher dose of DA, whereas lower dose of DA had no signi cant effect on the two proteins by IB analysis (Fig. 5A and B). Parallelly, as shown by IB analysis, PHNs exposed to DA exhibited lower NICD1/HES1 levels, addition of ATAB blocked the decrease in NICD1/HES1 expression ( Fig. 5C and   D). ATAB PC12 cells after exposure to DA also displayed the decreased expression of NICD1/HES1, while knockdown of TNFα blocked the decrease in NICD1/HES1 expression PHNs (Fig. 5E and F). As determined by double IF staining using NICD1 antibody with MAP2 antibody, we con rmed that the decrease in NICD1 expression in PHNs in response to DA was diminished by ATAB (Fig. 5G). Immunostaining con rmed that addition of ATAB diminished the decrease in NICD1 expression in response to DA in PC12 cells (Fig. 5H). Parallelly, these data indicated that DA inactivated notch signaling depending on stimulation of TNFα secretion.

DA impairs the production of neurotrophins by the inactivation of Notch pathway in vitro
Hence, we examined whether DA had effect on the expressions of GDNF/NGF through notch signaling in vitro. As indicated by IB analysis, we also found that DA-induced the decreases in the expressions of GDNF/NGF in PHNs were signi cantly enhanced by addition of Notch inhibitor MK-0752 and reversed by addition of Notch activator Jagged-1 ( Fig. 6A and B). Notch overexpression plasmid transfection into PC12 cells was further con rmed by western blot using the anti-Flag antibody (Fig. 6C). IB analysis revealed that the decreases in GDNF/NGF levels in PC12 cells in response to TNFα were abolished by Notch overexpression (Fig. 6D and E). Notch siRNA transfection into PC12 cells was further con rmed by western blot using the anti-Flag antibody (Fig. 6F). Furthermore, DA -mediated the decreases in Homer/synaptotagmin expression were enhanced by Notch silencing in PC12 cells ( Fig. 6G and H). These data suggested that DA impaired the production of neurotrophins and synaptic formation through the inactivation of notch signaling in vitro.

βASA inhibits the DA-induced the loss of neurotrophins in vitro
βASA has been reported to have neuroprotective and synaptoprotective effects on MHE [38]. We therefore examined the effect of βASA on the DA-mediated downregulation of GDNF/NGF. Treatment with βASA abolished DA-mediated the increase in the expression of TNFα in a dose-dependent manner in PHNs ( Fig. 7A and B). As determined by ELISA, DA-stimulated the release of TNFα was abated after βASA treatment from PHNs with dose-dependence (Fig. 7C). And the addition of βASA abrogated DA-induced the decrease in the notch expression with dose-dependence in PHNs ( Fig. 7D and E). We then examined whether the effect of βASA on the DA-induced the loss of neurotrophins was associated with TNFα. IB analysis indicated that DA-induced the downregulation of GDNF/ synaptotagmin levels were reversed by the addition of βASA in PHNs, whereas, the effect of βASA was then inhibited by addition of TNFα or MK-0752 ( Fig. 7F and G). Using FM4-64 dye to probe activity-dependent synaptic vesicle recycling in PHNs revealed that synaptic activity displayed obvious reduction in response to DA, which was blocked by addition of βASA, however, the effect of βASA was reversed by treatment of TNFα or MK-0752 ( Fig. 7H and I). These data suggested that βASA diminished DA-mediated the loss of neurotrophins required for downregulation of TNFα and activation of notch signaling.

βASA improves DA-induced memory impairment in MHE rats
Then, we con rmed the effect of βASA on TNFα level, neurotrophic expression, and synaptic formation in vivo. It is clear from IF staining of cortices of MHE rats using TNFα combined with MAP2, pronounced increased levels in neurons in cortices of MHE rats were blocked by administration of βASA (Fig. 8A), we con rmed that the administration of βASA signi cantly diminished the decrease in expression of GDNF in neurons in cortices (Fig. 8B) and in expression of synaptotagmin in neurons in hippocampi (Fig. 8C) in MHE rats. Furthermore, the decreases in the NGF/synaptotagmin expressions were detected in hippocampus of MHE rats, which were ampli ed by DA and recovered to the normal level by administration of βASA, furthermore, the enhancement of DA was blocked by βASA addition (Fig. 8D and E). Then we addressed whether βASA impacted on the memory decline. In YM test, SA% in MHE rats, which were signi cantly lower (P < 0.01) than that of control rats, was enhanced by DA and reversed by βASA administration, furthermore, the enhancement of DA on the SA% of MHE rats was blocked by βASA addition (Fig. 8F). These data suggested that βASA inhibited the impairment of synaptic formation and memory function in vivo.

Discussion
Neurons regulate synaptic transmission via monitoring neuronal signaling [39], and they release in ammatory cytokines, such as TNFα [40]. TNFα is one of the major in ammatory cytokines [41] and a vital acute injury associated cytokine. Elevated baseline and systemic TNFα may have roles in lower hippocampal volume [42] and hippocampal neurodegeneration [43] that triggers cognitive changes [44]. DA directly changed the synaptic formation and elevated the expression and release of TNFα in neurons in vitro. TNFα was reported to impair memory function. Neuron with DA treatment, displayed the elevation of TNFα level in media, the impairment of synaptic formation and activity and the decrease of dendrite spine, and addition of ATAB reversed synaptic loss. MHE rats also showed the increase in the expression of TNFα, neutralization of DA blocked expression of TNFα, indicating the induction of TNFα by DA. And neutralization of TNFα diminished the synaptic formation. Furthermore, neurons exposed to DA exhibited increased production and release of TNFα levels, suggesting that neuronal TNFα in turn induced disruption of synaptogenesis. Our ndings strongly suggest that the increase in DA could be responsible for enhancing synaptic loss and cognitive decline by increasing TNFα expression and secretion.
Apoptosis signal-regulating kinase1 (ASK1) is one of the mitogen-activated protein kinase (MAPK) kinase kinase family members. c-Jun N-terminal kinase (JNK) is a member of MAPK family that is upregulated by MAPK kinase kinase family members. ASK plays an essential role in determining cell fate, including death, survival, differentiation, and mitosis [33]. The ASK1 signaling pathway is involved in the modulation of expression of pro-in ammatory cytokines through JNK1 signaling pathway [34,45]. In ammatory responses are frequently mediated by the JNK signaling pathway that causes production of pro-in ammatory cytokines TNFα, and IL-6 [46]. DA which appears to be implicated in in ammatory response is its ability to induce the activation of the ASK1/JNK1 signaling pathway. Notably, evidence from DA-treated primary cultured neurons suggested that DA-induced TNFα production might be due to activation of ASK1/JNK1 pathway. Findings from TNFα-treated primary cultured neurons showed that TNFα had decreased activity of Notch signaling leading to the decreased expression of GDNF and NGF. The inhibition of ASK1/JNK1 pathway, blocked DA-induced the increase in TNFα release from neurons, inactivation of Notch pathway and impairment of synaptogenesis in neurons.
Notch is a 320 kDa polypeptide that is cleaved in the trans-Golgi network by a furin-like convertase to generate a functionally active receptor consisting of a 200 kDa extracellular ligand-binding domain and a 120 kDa transmembrane domain [47,48]. Activation of the Notch-1 receptor by the Jagged or Delta membrane bound ligands results in cleavage of the Notch-1 transmembrane domain to yield the Notch intracellular signaling domain (NICD). This 110 kDa domain of Notch translocates to the nucleus and induces expression of basic helix-loop-helix transcription repressors, such as hairy/enhancer of split (Hes), in particular Hes-1 [49,50]. Notch is also required for memory consolidation, a process believed to require remodeling of existing neurons in adults [35]. Several studies have shown that Notch also plays an important role in memory formation in adults, both in mammals and Drosophila [36,37]. Moreover, mice with antisense-reduced hippocampal Notch-1 mRNA and protein levels fail to sustain long-term potentiation but exhibit no defect in paired-pulse facilitation, a short-term form of synaptic plasticity [51]. Dysregulation of the Notch signaling pathway has been shown to contribute to a number of neurological diseases including Alzheimer's disease (AD) [52], cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy (CADASIL) [53], and cortical dysplasia [54]. Our study suggests that DA induced the dysregulation of the Notch to inhibit synaptic formation through TNFα, which was responsible for the memory impairment of MHE. Counteracting of TNFα, abated DA-induced inactivation of Notch pathway and impairment of synaptogenesis in neurons. Remarkably, DA-induced impairment of synaptogenesis was reversed by activation of Notch pathway and was enhanced by the inhibition of Notch pathway in neuron. These ndings suggested that DA elicited synaptic loss via ASK1/JNK1/TNFα/Notch pathway in neuron, which suggested that this cellular pathway was associated with pathogenesis of MHE.

Conclusion
In this study, we are the rst demonstrating that DA levels are increased under MHE conditions and DA can exert the learning impairment. the data in vitro presents show that DA plays a role in directly triggering TNFα production and accumulation in neurons, and resulting in synaptic loss, thus re ecting that DA burden modulates dendritic spine generation and synaptic formation via TNFα release. And the signals from DA mediated via ASK1/JNK1 pathway to production of TNFα operate linked to the inactivation of Notch pathway that are sensitive to the inhibition of synaptic formation in neurons.
Notably, the increase in DA was targeted to study the possible therapeutic effects of βASA on the associated increase in synaptic function and decrease in in ammation in MHE. The datasets generated and/or analyzed during the current study are available from the corresponding author on reasonable request.

Ethics approval
All experiments were conducted in accordance with the guidelines for the Care and Use of Laboratory Animals of the National Institutes of Health and Institutional Animal Care and Use Committee of Wenzhou Medical University.

Consent for publication
Not applicable     MRGD, merged image.