Thrombospondin1 Enhanced Neuronal Wnt7a and CNTF Expressions by TNFα Signaling Pathway

Background: Thrombospondin (TSP) is an astrocyte-secreted protein, well-known for its function as a modulator of synaptogenesis and neurogenesis. The mechanism underlying the effects of TSP1 on synaptic activity and formation involved in MHE pathogenesis remains unclear. Methods: The present study explored the effect of TSP1 on neurodegeneration, inammatory response and the activation of Wnt7a/CNTF signaling in the primary rat neurons and an MHE rat model. Results: When we treated neurons with TSP1, p38MAPK phosphorylation and TNFα expression was increased signicantly. Also, the exposure of TSP1 increased the expression and release of Wnt7a and CNTF, upregulated spinophilin, enhanced the interaction of Wnt7a/CNTF with spinophilin, and triggered synaptic activity through p38/TNFα signaling in PC12 cells and primary neurons. The hippocampal injection of TSP1 siRNA in mice decreased the interaction of Wnt7a/CNTF with spinophilin, while injection of Wnt7a and CNTF improved learning and memory dysfunctions. MHE brains showed decreased TSP1 expression. The overexpression of TSP1 in the hippocampus of MHE rats ameliorated the disrupted synaptogenesis, learning, and memory. Conclusions: Taken together, these results indicated that TSP1 is a potential synaptic factor related to the inammatory response and the activation of Wnt7a/CNTF signaling in MHE pathogenesis. interaction of Wnt7a/CNTF with spinophilin via p38/TNFα signaling and synaptic plasticity. The present study demonstrated that the disturbance of endogenous TSP1 generation is involved in the loss of activation of Wnt/CNTF signaling in the hippocampus, which leads to a decit in learning and memory. Therefore, preventing TSP1 might be a novel therapeutic strategy for MHE.


Introduction
Minimal hepatic encephalopathy (MHE) represents a mild neurocognitive disorder in patients with liver cirrhosis and portosystemic shunts. These subtle neurocognitive abnormalities primarily affect attention, information processing, motor abilities, and coordination that are not recognizable on standard neurological examinations and independent of sleep dysfunction or problems with overall intelligence [1][2][3][4]. However, the speci c molecular mechanism of MHE pathology is not yet understood.
Astrocyte-derived thrombospondins (TSPs) are large extracellular matrix proteins that have been identi ed as major contributors to astrocyte-regulated excitatory synapse formation [5]. The TSP family consists of two subfamilies, A and B, according to their organization and domain structure: A includes the trimeric TSP1 and TSP2, whereas B includes pentameric TSP3, TSP4, and TSP5 [6]. The expression of all TSPs has been detected in the brain, and studies using puri ed retinal ganglion cell cultures indicated that TSPs, especially TSP1 and TSP2, promote the formation of new excitatory (glutamatergic) synapses [7,8]. Emerging evidence demonstrated that TSPs expressed by immature and reactive astrocytes are responsible for the excitatory central nervous system (CNS) synaptogenesis or synapse formation [9,10].
Double TSP1 and TSP2 knockout mice show a reduced number of excitatory synapses in the cortex [7] and display dendritic spine irregularities [5]. Therefore, we focused our attention on the mechanisms underlying the involvement of TSP1 in the pathogenesis of MHE.
Overall, the expression of TSP1 and its effect on cognitive function is associated with Wnt7a/Ciliary Neurotrophic Factor(CNTF)/spinophilin interaction in MHE, which has not yet been investigated. The current study implied that increasing TSP1 levels in MHE would prevent the impairment of Wnt7a/CNTF with spinophilin interaction by inactivating p38/Tumor Necrosis Factor α (TNFα) signaling.

MHE models and treatment
A total of 40 Sprague-Dawley (SD) rats (Experimental Animal Center of the Chinese Academy of Sciences in Shanghai, China), weighing 220-250 g, were housed under controlled conditions of temperature (24 ± 1°C) and light (12 h light starting at 07:00 a.m.). All experiments were carried out in accordance with the guidelines laid down by the Ethics Committees of the A liated Hospital of Wenzhou Medical University regarding the care and use of animals for experimental procedures [11].
Before experiments, all animals underwent a series of behavioral tests: Y-maze (YM) and water-nding task (WFT). The normal values of these behavioral tests were obtained. Then, the rats were then randomly divided into two groups: control (n = 20) and thioacetamid (TAA) (n = 30). Liver cirrhosis was induced by intraperitoneal injection (i.p.) of TAA (200 mg/kg in normal saline, Sigma-Aldrich,Darmstadt, Germany) twice per week for 8 weeks. TAA-treated rats with symptoms were diagnosed as HE. The symptoms of HE were as follows: a subsequent decrease in motor activity, lethargy, and an eventual progression to coma. TAA-treated rats with no hepatic encephalopathy (HE) symptoms were again subjected to behavioral tests to con rm MHE. If TAA-treated rats met the criteria: a) values of YM were lower than mean ± 1.96 (SD); b) values of WFT were more than mean ± 1.96 (SD), the animals were included in the MHE group.
MHE rats were microinjected with TSP1 overexpression plasmid into hippocampus for 24 h.

YM test
Rats were individually placed at the end of an arm and allowed to explore the maze freely for 8 min. Total arm entries and spontaneous alternation percentage (SA%) were measured. SA% was de ned as the ratio of the arm choices that differed from the previous two choices ("successful choices") to total choices during the run ("total entry minus two" because the two entries could not be evaluated) [12].

WFT test
A rat was placed at the near-right corner of the apparatus and allowed to explore freely for 3 min. The rats that could not nd the tube within 3 min were omitted from the analysis. After the training session, the animals were deprived of water for 24 h. In the trial session, they were again individually placed at the same corner of the apparatus and allowed to nd and drink the water in the alcove. The elapsed time until the entry into the alcove (entry latency, EL) until the touching/sni ng/licking of the water tube (contacting latency, CL) and until the initiation of drinking from the water tube (drinking latency, DL) was measured [13]. TSP1 knockdown mice WT (wild-type) mice (Balb/c) were anesthetized with intramuscular xylazine (16 mg/kg), followed by ketamine (100 mg/kg). TSP1 small interfer ribonucleic acid (siRNA) was (10 or 40 µg/5µL) injected into the left hippocampus of mice for 24 h (anterior -posterior, + 0.3 mm; lateral, 1.0 mm; horizontal, 3.0 mm from the bregma) (n = 15).

Primary hippocampal neurons culture
Primary hippocampal rats neurons (PHNs) were prepared from 1-day-old SD rat pups and dissociated from freshly dissected hippocampus or cerebral cortex by mechanical disruption in the presence of trypsin and DNase and then plated in poly-L-lysine-precoated six-well plates. The cells were seeded at a density of 2×10 6 cells/well in Neurobasal® Medium (1X) supplemented with 0.5 mM GlutaMAX™-I and B-27® and incubated at 37°C, 5% CO 2 .

PC12 cell culture
The rat pheochromocytoma PC12 cell line was obtained from the American Type Culture Collection (ATCC, Manassas, VA, USA) and cultured in DMEM with 2 mM glutamine, 10% horse serum, and 5% heatinactivated fetal bovine serum (FBS) that was passaged one time per week.
Each well of cells received 0.25 µg of p38/TNFα/Wnt7a/CNTF siRNA (Santa Cruz, CA, USA) using Lipofectamine 2000 reagent (Invitrogen,Carlsbad, CA,USA), according to the manufacturer's protocol. An equivalent of 10 ng of The Silencer Negative Control number 1 siRNA (scrambled siRNA) was used as a control (Santa Cruz).
For coimmunoprecipitation, the lysates of tissues were incubated with antibodies overnight (4°C) and subsequently with protein G-agarose beads (Millipore, USA) for 5 h (4°C). Followed by washes with lysis buffer, the eluent was separated by SDS-PAGE and electroblotted to PVDF membrane using primary and secondary antibodies.

Dendritic spine density analysis in primary neurons
For dendritic spine analysis, we used immunocytochemistry with anti-microtubule-associated protein 2 (anti-MAP2) and anti-vesicular glutamate transporter 1 antibodies. After xation, the primary antibodies used were microtubule-associated protein 2B (MAP2B; 1:200; BD Transduction Laboratories, San Jose, CA, USA) and vesicular glutamate transporter 1 (vGlut1; 1:100; Neuromab, Davis, CA, USA). Primary antibodies were applied overnight at 4°C, followed by incubation with AlexaFluor-conjugated secondary antibody for 1 h (1:500; Life Technologies, Waltham, MA, USA) and glass cover slipping. A z stack of the optical sections was visualized on a confocal laser scanning microscope (FV10i-w; Olympus Corporation, Tokyo, Japan). At least 10 cultured neurons from two batches of cultures per group were used for quantitative analysis. Double-labeled uorescence staining Frozen brain sections or neurons (4mm) cultured on glass coverslips were xed with 4% paraformaldehyde for 30min and then treated with 0.1% Triton X-100 for 10 min at room temperature. Subsequently, the sections were blocked with 5% normal goat serum in phosphate-buffered saline (PBS) for 1 h at room temperature. Then, the sections were incubated overnight at 4°C with the following primary antibodies: TSP1, TNFR1, FZD2, CNTFR, spinophilin, and MAP2 (Abcam) that were detected by incubation for 30 min with uoresceine isothiocyanate (FITC) (green)/Alexa Fluor 594 (red)-conjugated secondary antibody. The images were captured using a Leica TCS SP2 (Leica Microsystems, Heerbrugg, Switzerland) confocal laser scanning microscope. The imaging data were analyzed and quanti ed using Image Pro Plus software.

Electrophysiological Analysis
Rats were anesthetized with iso urane, decapitated, and the hippocampi were cut into 400-mm thick transverse slices with a vibratome. After incubation at room temperature in a-CSF for 60-90 min, the slices were placed in a recording chamber on the stage of an upright microscope (Olympus CX-31) and perfused at a rate of 3 mL/min with a-CSF (containing 1 mM MgCl 2 ) at 23-24°C. A 0.1-MΩ tungsten monopolar electrode was used to stimulate the Schaffer collaterals. The eld excitatory postsynaptic potentials (fEPSPs) were recorded in CA1 stratum radiatum by a glass microelectrode lled with a-CSF with resistance of 3-4 MΩ. Field potential input-output curves were constructed by measuring fEPSP slope responding to the stimulus intensity increased from 1 to 7 V, with a 0.5 V increment. The long-term potentiation (LTP) of fEPSPs was induced by three theta-burst stimulations (TBSs), i.e., four pulses at 100 Hz, repeated three times at a 200-ms interval. Paired-pulse facilitation (PPF) was examined by applying pairs of pulses, separated by 20-500-ms intervals. The magnitudes of LTP were expressed as the mean percentage of the baseline fEPSP initial slope.

Statistical analysis
Data are presented as mean ± SD. The statistical signi cance between-group comparisons was determined by one-way analysis of variance (ANOVA). P < 0.05 or P < 0.01 was considered to be statistically signi cant.

TSP − 1 inactivated p38MAPK/TNFα signaling in vitro
TSP1 is responsible for CNS synaptic formation [7,10]. The P38MAPK pathway activation results in impaired synaptic plasticity [14], and increased systemic TNF-α may have roles in hippocampal neurodegeneration [15]. Next, we measured whether TSP1 stimulated the inactivation of p38MAPK/TNFα signaling in neurons. IB analysis of cell lysates showed that TSP1 gradually decreased the phosphorylation of p38MAPK ( Fig. 1A and B) and the expression of TNFα ( Fig. 1C and D) in PC12 cells in a dose-dependent manner. Similarly, TSP1 treatment signi cantly decreased the release of TNFα from PC12 cells compared to the unstimulated cells in a dose-dependent manner, as determined by ELISA (Fig. 1E). Moreover, the expression of TNFα in response to TSP1 was decreased in a time-dependent manner ( Fig. 1F and G). In addition, TSP1 treatment signi cantly decreased the release of TNFα from PC12 cells in a time-dependent manner, as determined by ELISA (Fig. 1H). Simultaneously, TSP1 treatment decreased the expression of TNFα/TNFR1 that was blocked by p38MAPK activator anisomycin ( Fig. 1I and J), as well as the release of TNFα from PHNs, which was diminished by anisomycin, as determined by ELISA (Fig. 1K). The cell-bound TNFα was further decreased in PHNs with TSP1 treatment, as determined by immunostaining, which was diminished by anisomycin (Fig. 1L). These data indicated that TSP1 treatment reduced TNFα production via dephosphorylation of p38 in neurons. TSP1 activated Wnt7a signaling via p38/TNFα signaling Wnt signaling also plays a relevant role in the presynaptic and postsynaptic structure [16]. Herein, we evaluated the effect of TSP1 on the activity of Wnt7a/β-catenin signaling in neurons. PC12 cells treated with TSP1 did not show any change in the Wnt3a and Wnt5a levels, but a signi cant increase was noted in the expression of Wnt7a in a dose-dependent manner, as assessed by immunoblotting ( Fig. 2A and B). Similarly, TSP1 treatment signi cantly increased the release of Wnt7a from PC12 cells compared to the unstimulated cells in a dose-dependent manner, as determined by ELISA (Fig. 2C). Also, PC12 cells exposed to TSP1 showed a signi cant increase in the expression of Wnt7a in a time-dependent manner by immunoblotting analysis (Fig. 2D and E) and ELISA (Fig. 2F). Simultaneously, PHNs were transfected with p38 ( Fig. 2G) or TNFα (Fig. 2H) siRNA plasmids, and the results of transfection e ciency assessed by qPCR showed the p38 or TNFα mRNA level was signi cantly decreased, while the content of Wnt7a was increased in response to TSP1 post-transfection of PC12 cells with p38 or TNFα siRNA ( Fig. 2I and J). Simultaneously, PHNs were transfected with p38 ( Fig. 2K) or TNFα (Fig. 2L) overexpression plasmid, and the results of transfection e ciency showed p38 or TNFα mRNA level was signi cantly increased by qPCR. Immunoblotting analysis showed that TSP1 treatment signi cantly increased the expression of FZD2 in PC12 cells, which was diminished by p38 or TNFα overexpression ( Fig. 2M and N). Also, Wnt7a was released from PHNs after TSP1 treatment, which was inhibited by anisomycin, as determined by ELISA (Fig. 2O). Moreover, Wnt7a expression was markedly increased by TSP1 treatment, as assessed by immunoblotting, which could be blocked by p38 activator anisomycin and TNFα ( Fig. 2P and Q), and SB202190 or TNFα inhibitor, lenalidomide, enhanced the ability of TSP1 to stimulate the protein expression in PHNs ( Fig. 2R and S). An increase in cell-bound pGSK3β was observed in PHNs after TSP1 treatment; this phenomenon was reversed by the addition of TNFα, as determined by immunostaining (Fig. 2T). Together, these data indicated that TSP1 treatment induced Wnt7a production via dephosphorylation of p38 and downregulation of TNFα in neurons. TSP1 activated CNTF signaling via p38/TNFα CNTF exerts neuroprotective effects [17][18][19]. Next, we determined the effect of TSP1 on the activity of CNTF signaling associated with Wnt7a signaling in primary neurons. The immunoblotting of PC12 cell lysates showed a maximal increase in the expression of CNTF and CNTFR by middle dose of TSP1 in (Fig. 3A and B). The TSP1 treatment signi cantly increased the release of CNTF from PC12 cells in a dose-dependent manner, as determined by ELISA (Fig. 3C). The cells exposed to TSP1 exhibited a signi cant increase in the expression of CNTF in a time-dependent manner by immunoblotting analysis (Fig. 3D and E). Similarly, TSP1 treatment signi cantly increased the release of CNTF from PC12 cells compared to unstimulated cells in a time-dependent manner, as determined by ELISA (Fig. 3F). An increase in the content of CNTF in response to TSP1 was ampli ed by p38 or TNFα siRNA in PC12 cells ( Fig. 3G and H). Immunoblotting analysis showed that TSP1 treatment signi cantly increased the expression of CNTFR in PC12 cells, which was diminished by p38 or TNFα overexpression ( Fig. 3I and J). TSP1treatment signi cantly increased CNTF expression, which was blocked by p38 activator anisomycin or Wnt inhibitor XAV939 ( Fig. 3K and L); CNTFR expression was also markedly increased by TSP1 treatment, SB202190, lenalidomide, or Wnt7a; these molecules enhanced the ability of TSP1 to stimulate the protein expression ( Fig. 3M and N) in PHNs. An increase in cell-bound CNTFR was observed in PHNs after TSP1 treatment; the addition of XAV939 reversed the effect, as determined by immunostaining (Fig. 3O). These data indicated that TSP1 treatment induced CNTF production through the inactivation of p38/TNFα signaling and activation of Wnt7a signaling in neurons.
TSP1 increased the interaction of Wnt7a/CNTF /spinophilin and synaptic activity via p38/TNFα signaling in vitro Spinophilin is related to neuronal plasticity, which plays a role in learning. [20][21][22] , [23]. Herein, we investigated whether TSP1 could stimulate the interaction among Wnt7a/CNTF with spinophilin. Immunoblotting from cell lysates showed that a maximal increase in the expression of spinophilin was achieved by the middle dose of TSP1 in PC12 cells ( Fig. 4A and B). Time course analyses of PC12 cells treated with TSP1 showed a steady increase in spinophilin expression up to 48 h via immunoblotting ( Fig. 4C and D).
Simultaneously, PHNs were transfected with Wnt7a (Fig. 4E) or CNTF (Fig. 4F) siRNA plasmids, and the results of transfection e ciency showed Wnt7a or CNTF mRNA level was signi cantly decreased by qPCR. Immunoblotting analysis showed that an increase in the content of spinophilin in response to TSP1 was abated by Wnt7a or CNTF siRNA in PC12 cells (Fig. 4G and H). Simultaneously, PHNs were transfected with Wnt7a (Fig. 4I) or CNTF (Fig. 4J) overexpression plasmids, and the results of transfection e ciency showed that Wnt7a or CNTF mRNA level was signi cantly decreased by qPCR. Wnt7a or CNTF overexpression enhanced the ability of TSP1 to stimulate spinophilin expression (Fig. 4K and L).
Next, we examined the effect of TSP1 on Wnt7a/CNTF with spinophilin interaction in neurons using coimmunoprecipitation. As shown in Fig. 4M, Wnt7a was immunoprecipitated, TSP1 treatment increased the content of CNTF, and spinophilin coimmunoprecipitated with Wnt7a in PHNs, which was enhanced by SB202190 or lenalidomide. CNTF was also immunoprecipitated, and TSP1 elevated the level ofWnt7a, while spinophilin was coimmunoprecipitated with CNTF, which was further ampli ed by SB202190 or lenalidomide. TSP1 increased the level of Wnt7a and CNTF coimmunoprecipitated with spinophilin, which was further ampli ed by SB202190 or lenalidomide.
Using FM4-64 dye for probing activity-dependent synaptic vesicle recycle, TSP1 treatment markedly increased the synaptic activity, which was blocked by the addition of TNFα (Fig. 4N and O). Double immuno uorescence (IF) staining with anti-vGluT1 (for staining dendritic spines) with anti-MAP2 (for staining microtubules) antibodies, and TSP1 treatment signi cantly increased vGluT1-positive signals in PHNs, which was remarkably diminished by the addition of XAV939 (Fig. 4P and Q). These data indicated that TSP1 treatment induced spinophilin expression via interaction with Wnt7a/CNTF and inactivation of p38/TNFα signaling in neurons. TSP1 knockdown decreased spinophilin expression and synaptic activity via Wnt7a/CNTF signaling in vivo TSP1 knockdown impaired Wnt7a/CNTF with spinophilin interaction associated with cognitive loss in vivo. Then, we examined the association of cognitive function/LTP with Wnt/CNTF in TSP1 knockdown mice. First, TSP1 siRNA plasmid was injected into the hippocampus of mice, and the results of transfection e ciency showed that the TSP1 mRNA level in the hippocampus was signi cantly decreased, as assessed by qPCR (Fig. 5A). In YM, a signi cant reduction in SA% was found in the hippocampus of mice after TSP1 siRNA injection but was restored to the normal level by the administration of Wnt7a or CNTF (Fig. 5B). In WFT, TSP1 knockdown markedly increased the level of EL, CL, and DL in mice, which was reversed by Wnt7a or CNTF (Fig. 5C). Then, we tested LTP and the basal synaptic transmission at CA1 synapses by measuring the input/output (I/O) curves. As shown in Fig. 5D and E, TSP1 knockdown in mice signi cantly decreased the LTP magnitude and impaired the basal synaptic transmission, and administration of Wnt7a or CNTF rescued this LTP de cit and the impairment of basal synaptic transmission. Also, the interaction of Wnt7a/CNTF with spinophilin was examined in TSP1 knockdown mice. As shown in Fig. 5F, Wnt7a was immunoprecipitated, and TSP1 knockdown decreased the content of CNTF and spinophilin coimmunoprecipitated with Wnt7a in the hippocampus of mice, which was diminished by lenalidomide. Moreover, CNTF was immunoprecipitated, and TSP1 knockdown induced a signi cant decrease in Wnt7a and spinophilin coimmunoprecipitated with CNTF, which was further blocked by lenalidomide. In addition, TSP1 knockdown decreased the level of Wnt7a and CNTF coimmunoprecipitated with spinophilin, which was abated by lenalidomide. A decrease in cellbound spinophilin was observed in the hippocampus of mice after TSP1 knockdown, and the administration of Wnt7a or CNTF reversed the level of this protein, as determined by immunostaining (Fig. 5G). These data indicated that the disruption of Wnt7a/CNTF and spinophilin interaction impaired the cognitive function and LTP in TSP1 knockdown mice.

TSP1 was downregulated in MHE rats
Next, we examined the expression of TSP-1 in MHE brains. As shown in Fig. 6A and B, decreased TSP1 level was observed in the hippocampus of MHE rats by IB analysis similar to that in the cortexes of MHE rats. The qPCR showed that TSP1 mRNA transcription was weak in the hippocampus and cortexes of MHE rats (Fig. 6C). IF staining revealed that pronounced that astrocytic TSP1 level was signi cantly decreased in the cortex of MHE rats compared to the controls (Fig. 6D). These observations indicated a decreased expression of TSP1 in the MHE brain. TSP1 overexpression alleviated the impaired Wnt7a/CNTF and spinophilin interaction in the cognitive decline in MHE rats. We also tested the effect of TSP1 overexpression on the cognitive de cit in MHE rats. Flag-TSP1 overexpression plasmid was used for injection into the hippocampus of MHE rats. The results of transfection e ciency showed the TSP1 mRNA level in the hippocampus was signi cantly increased by qPCR compared to that in MHE rats (Fig. 7A). In YM, a signi cant reduction was detected in SA% in MHE rats, which was recovered to the normal level by TSP1 overexpression (Fig. 7B). In WFT, MHE rats showed a marked increase in EL, CL, and DL, and the signi cant delay in EL, CL, and DL was reversed by TSP1 overexpression in the hippocampus (Fig. 7C).
Furthermore, we tested the LTP and the basal synaptic transmission at CA1 synapses by measuring the I/O curves. As shown in Fig. 7D and E, TSP1 knockdown in MHE rats showed decreased LTP magnitude and impaired basal synaptic transmission, and overexpression of TSP1 rescued this LTP de cit and the impairment of the basal synaptic transmission. Therefore, we explored whether TSP1 overexpression is involved in Wnt7a/CNTF and spinophilin interaction in MHE brains. As determined by immunostaining, a decrease in cell-bound spinophilin (Fig. 7F) was observed in the hippocampus of MHE rats, and TSP1 overexpression reversed the level of this protein. The effect of TSP1 overexpression on Wnt7a/CNTF and spinophilin interaction was evaluated using coimmunoprecipitation. As shown in Fig. 7G, Wnt7a was immunoprecipitated, and the hippocampus of MHE rats showed a decreased content of CNTF, and spinophilin coimmunoprecipitated with Wnt7a was diminished by TSP1 overexpression. CNTF was also immunoprecipitated, and the hippocampus of MHE rats displayed a signi cant decrease in Wnt7a, and spinophilin coimmunoprecipitated with CNTF was further blocked by TSP1 overexpression. Spinophilin was also immunoprecipitated, and the hippocampus of MHE rats displayed the obvious decrease in Wnt7a, and CNTF coimmunoprecipitated with spinophilin, which was further abated by TSP1 overexpression.
These data indicated that TSP1 overexpression rescues cognitive function and LTP associated with Wnt7a/CNTF and spinophilin interaction in the MHE brain.

Discussion
In the present study, we demonstrated that TSP1 expression was decreased in the brain of MHE rats. The injection of Flag overexpression plasmid into the hippocampus of MHE rats improves their learning and memory. Our results suggested that the disturbance of hippocampal and cortical TSP1 generation might be involved in the de cit in learning and memory under the interaction of Wnt7a/CNTF with spinophilin.
Recent evidence suggested that astrocytes regulate synapse formation and integrity through the secretion of the extracellular matrix proteins TSPs, especially TSP1 [7,24]. TSPs, critical astrocytesecreted proteins, are large oligomeric extracellular matrix proteins that have been previously shown to play major roles in excitatory CNS synaptogenesis or synapse formation [9,10]. TSP1, a member of a family of astrocyte-secreted extracellular matrix proteins, participates in synaptogenesis [5,7].
Notably, our data presented that shortage of astroglia-derived TSP1 contributes to synaptic dysfunction in MHE rats and TSP1 knockdown mice, suggesting that decreased level of TSP1 is a common feature of MHE disease. Considering the involvement of TSP1 in cognitive and memory disorder, the current data suggested that TSP1 release in the hippocampus of rats play a major role in impaired learning and memory in MHE rats. Reportedly, TSP1 serves as a protective signaling molecule by preventing synaptic loss and acts as a neuro-modulator by facilitating the formation of hippocampal long-term potentiation (LTP) and the major cellular mechanisms underlying learning and memory. Another potential mechanism is the reduction in TSP1 levels, which might impact neuronal integrity with respect to the pathogenesis of MHE.
The MAPK family, including extracellular signal-regulated kinase (ERK), p38, and jun N-terminal kinase (JNK), mediates in ammation, cell proliferation, and apoptosis [25]. p38MAPK is a member of the MAPK family, which also includes extracellular signal-regulated kinase, big MAP kinase1, and c-Jun NH2terminal kinase [26]. P38 MAPK pathway contributes to cognitive damage in pentylenetetrazole (PTZ)induced epilepsy. [27] Inhibiting the p38 pathway alleviated Aβ 1−42 -induced cognitive de cits and neuronal loss and death [28], while activation of p38-MAPK signaling pathway contributes to the formation disorder of cortical synapses [ [29]]. A local increase in TNFα in the hippocampal dentate gyrus activates TNF receptor type 1 (TNFR1), which in turn, results in persistent functional impairment of hippocampal excitatory synapses [30]. Moreover, TNF-α confers susceptibility to schizophrenia and cognitive dysfunction [31], mediates cognitive de cits [32], and the synaptic action interferes with brain circuits pertaining to learning and cognition, which contributes to excitotoxicity and neurodegeneration [33]. Furthermore, TNF is a proteolytically cleaved transmembrane protein that exerts its activity via TNFR1 [34]. This cytokine is a major regulator of synapse function implicated in synaptic transmission and homeostatic synaptic scaling [35,36]. Neurons in individuals are hypersensitive to TSP1 and exhibit pronounced dysregulation in the levels of p38MAPK/TNFα. Thus, upregulated TSP1 in neurons might downregulate neuronal p38MAPK/TNFα protein levels. Moreover, the increase in the p38MAPK/TNFα content in neurons was induced by transfection with TSP1 siRNA, while reduced TSP1 levels in astrocytes may stimulate both p38MAPK/TNFα signaling pathway in neurons.
Recent studies have proposed a critical role for Wnt pathway in the adult hippocampus involved in neurogenesis and synaptic plasticity [37][38][39], indicating a role of Wnt signaling also in maintaining the network connectivity, memory formation, and cognitive exibility [40][41][42][43]. Canonical Wnt/b-catenin signaling is altered or involved in the regulation of adult hippocampal neurogenesis and self-renewal of neural stem/progenitor cells [44,45], the pathophysiology of Alzheimer's disease (AD), and other neurodegenerative disorders [46]. Speci cally, postmortem studies have described reduced levels and altered activity of the intracellular component b-catenin in AD brains [47,48]. Thus, aberrant Wnt signaling has been a proposed mechanism underlying the onset and progression of the pathological hallmarks in MHE [49].
Wnt ligands have been linked to the assembly of structural components in presynaptic compartments. Wnt7a induces axonal spreading and incremental growth of cone size and branching, leading to the accumulation of synaptic proteins [50,51]. Then, in the cerebellum, Wnt7a acts as a retrograde signal from granular cells to induce presynaptic differentiation in mossy bers, serving as a synaptogenic factor [51,52]. Wnt7a also increases the clustering of postsynaptic proteins, such as postsynaptic density protein-95 [37]. Additionally, the use of FM dyes showed that TSP-1 stimulates recycling and accelerates exocytosis of synaptic vesicles. Wnt7a increases the amplitude of eld excitatory postsynaptic potentials (fEPSP) and decreases the rate of paired-pulse facilitation in the CA3-CA1 synapse [37]. TSP1 rapidly activates Wnt in primary neurons, decreasing the p38/TNFα content by transfection with overexpression or siRNA plasmid or by stimulation of synthesis with inhibitors or activators. These processes prevent the deleterious effects of downregulated TSP1 on Wnt7a signaling. These ndings implicated the TNFα/Wnt7a signaling cascade in the synaptic effects of TSP1. Therefore, we proposed that the p38/TNFα signaling is involved in decreased TSP1-induced hippocampal inactivation of Wnt signaling in MHE rats.
CNTF belongs to the IL-6 family of cytokines and exerts neuroprotective effects, thereby serving as a survival factor for sympathetic, sensory, hippocampal, and motor neurons in vitro as well as in vivo. Exogenous CNTF protects mature neurons from degeneration arising from multiple etiologies, including Huntington's disease, amyotrophic lateral sclerosis, or retinal degeneration, and its e cacy is tested in several clinical trials [17][18][19]. CNTF led to full recovery of cognitive functions associated with the stabilization of synaptic protein levels in the Tg2576 AD mice model. In vitro as well as in vivo, CNTF prevented synaptic and neuronal degeneration. These preclinical studies suggested that CNTF and/or CNTF receptor-associated pathways might exhibit AD-modifying activity through protection against progressive Aβ-related memory de cits. [53] CNTF enhances nicotinic synaptic transmission via both presynaptic and postsynaptic mechanisms [54]. TSP1 abolishes the effects of p38/TNFα on CNTF signaling, decreasing the p38/TNFα content in neurons by transfection with overexpression or siRNA plasmid or by stimulation of its synthesis with inhibitors or activators, thereby preventing the deleterious effects of the downregulation of TSP1 on CNTF signaling. Next, TSP1 increased the neuronal content of Wnt7a protein critically involved in the activation of CNTF signaling. These ndings implicated the TNFα/Wnt7a/CNTF signaling cascade in the synaptic effects of TSP1. Therefore, we proposed that the disruption in Wnt signaling is involved in decreased TSP1-induced hippocampal inactivation of CNTF signaling in MHE rats. Spinophilin expression was low during mice development but increased markedly after birth and persisted in the adult brain [55]. The immunolabeling approach revealed that spinophilin was found predominantly in dendritic spines in adult macaque prefrontal cortex [56]. Further studies showed that the protein is enriched in the cerebral cortex, caudatoputamen, hippocampal formation, and cerebellum [57], as well as at the synapse and in cadherin-based cell-cell adhesion sites [58]. Spinophilin is an optimal candidate to serve as a link between excitatory synapse transmission and changes in spine morphology and density. In initial experiments, spinophilin was implicated in the stabilization of the actin cytoskeleton in dendritic spines and in lipodia [59,60]. It also regulates excitatory synaptic transmission and plasticity [57]. Furthermore, spinophilin is a postsynaptic marker, primarily localized in the heads of dendritic spines. The decrease in spinophilin immunoreactivity was signi cantly related to both Braak neuro brillary tangle (NFT) staging and clinical severity but not Abeta deposition staging [61] It also has a critical role in learning in vivo and could be related to neuronal plasticity and learning [20][21][22].
Dysregulated spinophilin function might be implicated in synaptic failure and cognitive dysfunction linked to MHE. This evidence supported that TSP1 induced synaptic activity that was directly associated with spinophilin levels; our ndings implicated the interaction of Wnt7a/CNTF with spinophilin in the synaptic effects of TSP1. In the brain, TSP1 promotes structural and functional recovery of synapses compromised by MHE via the mechanism involving the interaction of Wnt7a/CNTF with spinophilin.
Therefore, enhanced interaction among Wnt7a/CNTF and spinophilin mediated by TSP1 has been shown to confer neuroprotection against synaptic impairment of MHE.

Conclusion
The current data provided comprehensive evidence of TSP1 involvement in the generation of Wnt7a/CNTF and the interaction of Wnt7a/CNTF with spinophilin via p38/TNFα signaling and synaptic plasticity. The present study demonstrated that the disturbance of endogenous TSP1 generation is involved in the loss of activation of Wnt/CNTF signaling in the hippocampus, which leads to a de cit in learning and memory. Therefore, preventing TSP1 might be a novel therapeutic strategy for MHE.

Data availability statement
The data that support the ndings of this study are available from the corresponding author upon reasonable request.     siRNA plasmid together with lenalidomide for the association among Wnt7a, CNTF and spinophilin. (G) Immunostaining of free-oating hippocampal sections from WT mice hippocampally injected with TSP1 siRNA plasmid together with Wnt7a or CNTF using antibodies against spinophilin (red) and MAP2 (green). Data are shown as mean ± SD. *P < 0.05, **P < 0.01 vs. Scr siRNA-knockdown mice; #P < 0.05, ##P < 0.01 vs. TSP1 siRNA-knockdown mice. Scr, scrambled. Scale bar, 25 μm. MRGD, merged image. homogenates from MHE rats using antibodies against TSP1 and β-actin, followed by densitometry analysis. (C) qPCR analysis of TSP1 mRNA in lysates from hippocampus or cerebral cortices of MHE rats. (D) Immunostaining of free-oating cortical sections from MHE rats using antibodies against TSP1 (red) and MAP2 (green).Data are shown as mean ± SD. *P < 0.05, **P < 0.01 vs. control rats. Scale bar, 25 μm. MRGD, merged image. Figure 7