Unraveling the Neuroprotective Effect of Tinospora Cordifolia in Parkinsonian Mouse Model Through Proteomics Approach.

Background Stress-induced dopaminergic (DAergic) neuronal death in the midbrain region is the primary cause of Parkinson’s disease (PD). Approximately 2% of the global population aged above 65 years is affected with PD. Various factors are responsible for the death of DAergic neurons, among which mitochondrial dysfunction, oxidative stress, misfolded protein aggregation and neuroinammation are the primary factors. From the discovery of L-dopa, multiple drugs were discovered to improve lifestyle of PD patients, but they failed due to their multiple side effects. Tinospora cordifolia (Tc), a medicinal herb has been used in traditional medicines to treat neurodegenerative diseases. In our previous study, the neuroprotective role of Tc against MPTP-intoxicated Parkinsonian mice was reported. Here, we further explore the neuroprotective molecular mechanisms of Tc in Rotenone (ROT) intoxicated mouse model through proteomics approach.

nigrostriatal region of midbrain of mice [27]. Using this method, quanti cation of the differentially expressed proteins (DEPs) of control vs. Parkinsonian group, and Parkinsonian vs. Tc group was performed. Our results revealed that the Tc treatment improves the behavioral abnormalities in PD mice and in label free proteome analysis, we tried to report multiple proteins in both groups, which were differentially expressed. From these results, we have selected 20 proteins (Supplementary le 2, Table  1)whose expression were signi cantly modulated in PD as well as treatment group and performed bioinformatics analysis, gene ontology (GO), functional pathway enrichment analysis along with proteinprotein interaction (PPI) network and gene-gene interaction, to investigate the correlation with previously reported protein and gene prominent pathway in uenced by Tc for neuroprotective activity. These results may assist to gure out the new therapeutic approaches for prevention and treatment of PD.

Plant Extract Preparation
Tc stem was procured from Botanical garden, Institute of Science, Banaras Hindu University, Varanasi, India in month of January 2019. Tc stems were dried for 7-10 days in shade and pulverized with the help of an electric grinder. The dried sample was extracted with ethanol and water solvent at a ratio of 70: 30 (200 gm powder into 1000 mL) at 40 °C for 16 h in Soxhlet apparatus. Consequently, the plant extract was ltered through 0.45 µm lter and the residue was concentrated under reduced pressure using rotary vacuum evaporator. The plant extract are expressed in terms of dry weight [28][29][30].

Ethical Statement
The care and maintenance of the experimental animals was carried out in strict accordance with the Animal Ethics procedure and guidelines of the Institutional Animal Ethical Committee of the Banaras Hindu University, Varanasi, India (Ref no. BHU/DoZ/IAEC/2018-19/032). All the efforts were made to ease suffering of mice used in the study.
For conducting the experiment male Swiss albino mice (25 ± 5 g) was obtained from the animal facility atInstitute of Medical Science, Banaras HinduUniversity, Varanasi (India). The mice were housed in cleanpolypropylene cages with constant light-dark cycles (12/12 h)prior to the experiment. Until the dosing was completed, mice were supplied with water and standard diet pellets ad libitum. The experimental protocol was established in accordance with the guidelines of the Animal Ethics Committee,Banaras Hindu University, Varanasi, India.

Animal Dosing
Animals were randomly assigned into three groups: Control (vehicle group), Rotenone (2 mg/kg body wt. per day) subcutaneously for 35 days and Rotenone + Tc (200 mg/kg body wt. per day) orallyadministered for 7 days, then concurrent administration of Tc (21 days) and ROT-intoxication (35 days) were done similar to that Rotenone group (n = 8/group). In accordance with our previous study, dosing was administered with minor changes [19,31]. Upon completion of the dosing, behavioral analysis was carried out, and the mice were sacri ced to isolate the brains for the proteome analysis. All the experiments were carried out in triplicate.

Behavioral Analysis
To determine the effect of ROT-intoxication on the motor activity in Parkinsonian mouse model, behavioral parameters including the catalepsy test, rotarod test, pole test, and foot-print assay were performed.

Catalepsy Test
Stiffness in the muscles of mice can be estimated by placing the forelimbs of mice on an elevated bar and the hind limbs on the ground. Catalepsy intensity was measured by recording the time when the mice moves their hind limbs from a wooden platform (3 cm in height) to correct their posture. The was recoded in second and the test was discontinued when the time exceeded 180 sec [32].

Pole Test
Initially, for two consecutive days, the mice were trained. The performance was recorded after the last MPTP injection. Mice were placed head-up on the top of a rod (10 mm in diameter, 52 cm in height, with rough surface). T-turn (time taken to orient downward) and T-descend (the time required for the mouse to step down the rod length) were recorded in second and experiment was discontinued after 300 sec [33].

Rotarod Test
For the rotarod test, each mouse was trained in a rotarod before the experiment for three consecutive days, at a set speed of 15 rpm. The time was recorded till the mouse fall from the rotarod and reading were recorded up to 5 min. For each mouse, the test was repeated ve times and the mean time was recorded. The difference in fall time observed between the PD group and the Tc-treated group is an indicator of muscle relaxation [34].

Footprint Assay
In the footprint test, the mice were trained to walk on a white paper for three consecutive days. The length of the stride was calculated by immersing the mouse's forefoot in blank ink and measuring the gap between the steps, from the middle toe ( rst step) to the heel (second step) on the same side of the body [35]. Foot print assay was conducted thrice for each animal. Stride forepaw length were recorded in cm.

Tissue Collection
After behavioral tests, mice were sacri ced by cervical dislocation followed by decapitation to ensure minimal pain. The brain of each mouse was isolated and instantly frozen. Later, the brain was dissected under ice cold conditions to isolate the nigrostriatal tissue from the midbrain and was stored at -80 °C until the further experiments were performed.

Sample Preparation for LFQ Analysis
Protein extraction was done following the previous method of Gupta et. al. (2010) with slight modi cations [32]. The brain tissue was thawed, and homogenized using the lysis buffer (RIPA, and protease cocktail inhibitor) and incubated for 90 min on ice. The homogenate was centrifuged at 12,000 rpm at 4 ℃ for 20 min, and the supernatant was collected in a fresh eppendorf tube. The proteins were quanti ed by Bradford assay and expressed in terms of mg/ml/gm. tissue [36]. Quantitative assessment of the protein samples was performed by SDS-PAGE. 100 μg of the protein sample was taken from each group for digestion. The samples were diluted with 50 mM NH 4 HCO 3 protein followed by treatment with 100 mM dithiothreitol (DTT) at 95 ℃ for 1 h. 250 mM iodoacetamide (IDA) was added to the RT in dark and incubated for 45 min. The samples were then digested with Trypsin (Trypsin gold promega, USA) and incubated overnight at RT. The resulting sample was vacuum dried and dissolved in 50 µl of 0.1% formic acid (FA) in water, followed by centrifugation at 10,000 rpm and the supernatant was collected into a separate tube. 10 µL injection volume was used on BEH C18 UPLC column for separation of peptides, the mobile phase contained 0.1% FA in water (buffer A) and 0.1% FA in Acetonitrile (98:2 v/v for 1-30 min, 50:50 v/v for 30-40 min, 20:80 v/v for 40-50 min then 98:2 v/v for the next 30 min), and the ow rate was set at 0.3 ml/min. The glutamate level was examined at a wavelength of 472 nm, and the concentration was determined using a standard curve. Three runs per sample were carried out for LFQ. The separated peptides on the column were directed to Waters Synapt G2 Q-TOF instrument for MS and MS/MS analysis. The raw data was processed by MassLynx 4.1 WATERS. The individual peptides MS/MS spectra were matched to the database sequence on PLGS software (ProteinLynx Global Server Software Scores each protein based on the signi cance of the protein being identi ed based on MS pattern and MS/MS pattern matching for each peptide, as well as the coverage), WATERS for protein identi cation. The instrument used for acquiring Mass Spec Data was connected with Waters Synapt G2 (QTOF). The instrument parameters used for identi cation were Peptide Mass Tolerance at MS1 level: 50 ppm and Fragment Mass Tolerance at MS2 level: 100 ppm. During processing of the sample cysteine sites were modi ed to carbamidomethylated cysteine, and the methionine sites being prone to oxidation, was considered as a variable modi cation to the mass ( Figure  6) [37,38].

Bioinformatics Analysis
Gene Ontology and Pathway enrichment analysis The biological signi cance of DEGs was investigated using WebGestalt, as it is a free online WEB-based gene set analysis toolkit for functional enrichment analysis in different biological levels. In GO and PATHWAY enrichment analysis, we obtained more valuable knowledge about cellular, molecular and biological processes. A p-value < 0.05 was considered a signi cant enrichment [39].

Protein-Protein Interaction Networks Analysis
Search tool for the retrieval of interacting genes database (STRING) is a database and web resource for forecasting PPI networks. To investigate the relationship between DEGs, the STRING database was used.
Our active interaction source of data was text mining, experiments, database, co-expression, gene fusion and co-occurrence. Minimum required score is set 0.7 (higher con dence). By analyzing the predicted interaction networks, we can propose new directions for future experimental research and come up with cross-species forecasts for systematic interconnecting mapping [40].
Investigation through KEGG Mapper tool and Genemania KEGG mapper tool is used for cellular or biological interpretation of large-scale data sets like genome or meta genome sequence [41]. It comprises of three different databases, Brite, Pathway and Module which, further contains experimental information from the already published literature and represented in the form of these three different pathways. Bioinformatics tools such as Genemania was used for imaging and visualizing the different interacting networks between different genes [42,43]. The interaction between genes may be physical, correlation, colocalize and genetic interactions. It also provides the information about the interaction with unknown genes from already published literature.

Statistical Analysis.
Statistical analysis of behavioral data was done by one-way analysis of variance (ANOVA) using the Student-Newman-Keuls test by GraphPad Prism 6.0. For gene expression, we applied Student's t-test to unequal variances in order to nd the p-value for each gene. The results are expressed as the mean ± SEM. p values < 0.05 were considered statistically signi cant.

Neurobehavioral Parameters
After ROT injection, stiffness was observed in mice by catalepsy test. The latency in the ROT group tended to increase over time (p <0.001), and the observed data was signi cantly different from those in the control group, but Tc-treated mice showed a shorter latency time (p <0.001), instead of the ROTintoxicated group (Fig. 1a). In the pole test, ROT-intoxicated mice showed prolonged T-turn in comparison to control group mice (p <0.01). The T-descend duration in Tc pre-treated mice was signi cantly reduced (p <0.01) (Figure 1b). Rotarod test was performed to check the balance, grip, and movement adjustment of the mice. Compared to control, ROT-intoxicated group had signi cantly reduced retention time (p <0.001) on rotarod, and it was observed that mice lost motor coordination and grip strength. Compared to ROT-intoxicated mice, Tc treatment signi cantly increased retention time on rotarod (p <0.01) (Fig. 1c).
Footprint analysis showed differences in average stride length between the control group, ROT group, and Tc-treated group. The stride difference in the ROT-intoxicated group was signi cantly reduced compared to the control group (p <0.01), whereas the gait disturbance was signi cantly improved on Tc treatment (p <0.05) (Fig. 1d).

Comparative Protein Pro ling Between control vs. PD and PD vs. Treatment Group
The protein pro ling was analyzed through PGLC software (Masslynux 4.1 Water, USA) and compared between control vs. Parkinsonian and Parkinsonian vs. Tc mice group. On the basis of fold change, the signi cance of differences in protein expression was also categorized. The protein whose fold change value was below (≤0.5) were consider downregulated and those protein whose value was greater than 2.0 fold were considered upregulated ( Table 1 and Table 2). The label-free LC-MS/MS results demonstrated that control vs. PD group contained 800 DEPs of which 67 were downregulated (Table 1A) and 101 were upregulated (Table 1B). In PD vs. TC group, 133 DEPs were obtained of which 11 (Table 2A) were downregulated and 13 were upregulated (Table 2B). Besides this, some proteins which were signi cantly expressed in the PD group were normalized by Tc (fold change between (0.5-2.0)) (Supplementary le 1, Figure 2). From this analysis, we identi ed 20 proteins whose expression was signi cantly modulated in ROT-intoxicated Parkinsonian group and was e ciently restored with the treatment of Tc. Some prominent upregulated proteins included Short transient receptor potential channel, glutamate receptor inotropic early endosome antigen 1, Serine/threonine-protein phosphatase 2A, transcription factor A, aryl hydrocarbon receptor nuclear translocator, eukaryotic translation initiation factor 5B, prosaposin receptor GPR37, RAC-alpha serine/threonine-protein kinase and neutrophil cytosol factor 4 (Table 2A). whereas ubiquitin carboxyl-terminal hydrolase, DNA (cytosine-5)-methyltransferase 1, phosphatidylinositol 3,4,5-trisphosphate 5-phosphatase 2, ribosome-releasing factor 2, phospholipase D1, receptor-type tyrosine-protein phosphatase, centromere protein I, mothers against decapentaplegic homolog 7, lysine-speci c demethylase 3A, T-lymphoma invasion and metastasis-inducing protein 1 were the prominently downregulated proteins (Table 2B).

Gene Ontology and Functional Pathway Enrichment Analysis
To comprehend the potential functions of DEPs in PD, these DEPs were examined. GO analysis  Tables 2,  3 and 4). This analysis revealed that dopamine metabolic process, catechol-containing compound metabolic process, catecholamine metabolic process, adherens junction assembly, mitochondrial gene expression, phenol-containing compound metabolic process, oligodendrocyte differentiation, adherens junction organization, positive regulation of protein modi cation by small protein conjugation or removal, response to toxic substance were found to be signi cantly enriched in BP. Furthermore, NMDA selective glutamate receptor complex, neuron spine, cation channel complex, extrinsic component of plasma membrane, cell-cell junction, cell junction, cell surface, trans-membrane transporter complex, ion channel complex, extrinsic component of membrane were signi cantly enriched in CC. Moreover, TGF-β receptor, cytoplasmic mediator activity, ligand-gated ion channel activity involved in regulation of presynaptic membrane potential, Ca 2+ -trans-membrane transporter activity, β-catenin binding, heat shock protein binding, NMDA glutamate receptor activity, δ-catenin binding, Ca 2+ channel activity was signi cantly enriched in MF. The criteria for p-values <0.05 was considered signi cant for enriched GO analysis. To investigate the enriched pathways associated with PD was performed using software Webgestalt (http://www.webgestalt.org). In this study we analyze various signi cant pathways through Wikipathway (https://www.wikipathways.org/index.php /WikiPathways) which has been shown in (Table 3). Enriched pathway analysis demonstrated that DEPs were highly involved in PD Pathway (Gpr37), Alzheimer's disease pathway, hypothetical network for drug addiction (Grin2a), mitochondrial gene expression (Tfam), lung brosis, TGF-β signaling pathway, Hfe effect on hepcidin production, ESC pluripotency (Smad7) and chemokine signaling pathway (Tiam 1).

In silico Analysis through PPI Network
PPIs play a crucial role in understanding the molecular function of DEPs, responsible for the PD onset. As the outcome, hub genes like AKT1, PARK2, MTOR, TFAM, PINK1, TH, SNCA, PID1, PARK7, LRRK2 and App were found with the highest degrees of connection in the network ( Figure 5A). In a constructed network, DEPs connected to each other, generally have analogous functions and can be considered as the functional genes. Gene-gene interaction network was constructed to investigate the interaction of signi cant proteins in the network ( Figure 5B).

Discussion
Discovery of L-dopa was a major step forward in the treatment of PD, but its major drawback was the generation of various side effects. In addition, many combinational drugs were reported to reduce the side effects, but the success rate was very low [44]. Besides, the researchers also focused on other molecular targets as well as the alternative treatment options to improve the lifestyle of PD patients. PD attributed by mitochondrial dysfunctions, impairment of protein degradation, oxidative stress and death of DAergic neurons in SN, but the continuous progression of molecular events leading to cell death remains unclear [45]. Prior to designing any preventive intervention, in-depth knowledge of the molecular mechanisms underlying neurodegeneration in PD is required.
ROT, is a well-known pesticide acting selectively on mitochondrial complex I, causing its inhibition like MPP + . A growing body of proofs suggest that the insecticide model offers more advantages over different experimental models as it will effectively mimic the behavioral and neuropathological symptoms of the illness through the selective degeneration of DAergic neurons [46,47].
PD have various motor (tremor, slowness of movement, rigidity, falls and dizziness, freezing, postural imbalance) [48] and non-motor symptoms (pain, low blood pressure, fatigue, restless legs, swallowing and saliva control, sleep, skin and sweating, bowel and bladder problems) [49]. Various studies reported that non motor symptoms generally develop at the initial stage of PD progression so that it is not able to differentiate PD patients from others. Zahra et al. (2020) reported the characteristic motor impairment in ROT-intoxicated Parkinsonian mice [31]. To check the behavioral de cit in mice, we performed various behavioral tests such as catalepsy, rotarod, pole, and footprint, which highlighted severe motor abnormalities in ROT-intoxicated PD mouse model [42][43][44][45] (Fig 1). Indistinguishable co-ordinates and deterioration of motor skills were also observed in our result due to ROT toxicity, whereas the behavioral tests demonstrates that Tc improves the neurobehavioral shortfall as compared with previous studies [50].
Various medicinal herbs are used for therapeutic functions since ancient times in India, China, and other Asian countries. Medicinal plants are used to treat various neurodegerative diseases [51] and showed very promising results with minimal side effects. Herbs like Bacopa monnieri, Withania somnifera, Mucuna pruriens, Tc, are tested for its therapeutic potential against neurodegenerative diseases via redox balancing and immunomodulation activities [18,52,53]. In our previous study, we reported that Tc has neuroprotective role via suppressing oxidative stress as well as neuroin ammation [19].
With the application of proteomic pro ling analysis, the quantitative and qualitative analysis of thousands proteins from the tangled mixtures along with the demonstrations of their PTMs could be e ciently performed. Based on our ndings, we have identi ed various new molecules and pathways as alternate treatment approach. Here, we used comparative proteome pro ling through LFQ/LC-MS/MS to investigate DEP in nigrostriatal region of mice brain. Out of total 800 DEPs, 101 were upregulated and 76 were downregulated in ROT-intoxicated mice compared to control; whereas, out of total 133 DEPs in Tctreated mice, 13 were upregulated and 11 were downregulated as compared to ROT. Some proteins which are signi cantly expressed in PD group were reversed by Tc treatment. On the basis of expression and corelation with PD marker proteins, we selected 20 proteins (Table 2A, Table 2B). Further in-silico analysis suggested that some of DEPs (Smad7, Tfam, Grin2a, Gpr37, and Tiam7) play a key role in PD pathogenesis via regulating different signaling pathways (Table 3)  Gpr37 is a substrate of the E3 ligase Parkin, and henceforth is also called Parkin-related endothelin-like receptor (Pael-R) [66]. Missense type of mutation in PARKIN causes aggregation of Gpr37 in brain of PD patients [67]. Besides, overexpression of Gpr37 prompts its amassing in totals causes ER stress, and neuronal death [66,68]. On other hand, it has been proposed that native Gpr37 shows neuroprotective property by binding prosaptide and prosaposin [69]. Interestingly, Lundius et al. (2013) reported that overexpression of Gpr37 was robustly protected against ROT, MPP+ or 6-OHDA-induced cytotoxicity in N2a cells [70]. Furthermore, it has been suggested that Gpr37 regulates oligodendrocyte separation and myelination by means of ERK signaling [71]. However, Tc-treatment to the PD mice suppresses the expression of Gpr37 in ROT-intoxicated mice model. Here, we present that Gpr37 plays a broader role in neuroprotective and glioprotective activity of Tc.
Tiam1 (T lymphoma invasion and metastasis 1) is a Rac-speci c GEF (Dbl family member), potentially involved in neurodegenerative diseases (AD and PD). It is a Ras effector molecule, stimulated by the Ca 2+dependent activation of the NMDA receptor (NMDAR) [72]. Tiam1 controls the activation of Rho GTPase and plays an important role in PD by regulating oxidative stress and neuroin ammation [73]. Tiam1 controls neurite extension and DAergic neurons differentiation [74,75] . After exposing hippocampal neurons to amyloid-β peptide, Tiam1 is activated and mobilized to the membrane, which may affect the pathology of AD [76]. In the present study, expression of Tiam1 is downregulated by Tc extract treatment against ROT-intoxication. Interestingly, Smith et al. (2017) also reported that, an elevated expression of Tiam1 contributes to neuronal damage [77]. In contrast, Cajanek et al. (2013) showed, Tiam1 as a positive regulator of DAergic neuron differentiation [74]. Together, these studies indicate that ROT-induced neurotoxic effects are linked with an upregulated expression of Tiam1 and Tc reverts the ROT-induced neurotoxicity via downregulating Tiam1 expression.
Akt, a serine/threonine kinase [also known as protein kinase B (PKB)] is an important molecule necessary for neuronal survival as it plays a major role in phosphorylating its substrates, including GSK3, NF-κB, BAD, and forkhead proteins [78]. Downregulation of Akt signaling are seen in PD [79]. Previous studies suggested that PD-inducing neurotoxins including 6-OHDA and MPP+ decreases the expression of pAkt [80].  (Thr308 and Ser473). Similarly, we have also reported a downregulated expression of Akt1 in ROT-induced parkinsonian mice. Growing evidence also suggested the importance of mTOR pathway in autophagy and apoptosis which can lead to neuronal death, but later it was found that was the inhibition of the Akt phosphorylation, rather than mTOR activation that eventually led to neuronal loss [81]. Raghu et al. (2009) reported, activation of Akt by G1-4A (a polysaccharide from Tc) [82]. In an attempt to unravel the anti-apoptotic action of Tc against ROTintoxication, our study revealed an increased expression of Akt1, also supported by Salama et al. (2020) [83,84].
The present study showed that Tc signi cantly reversed ROT-induced downregulation of Grin2A expression and exhibits neuroprotective effect against ROT-induced neurotoxicity. Moreover, downregulation of Grin2A (NR2A, a glutamate ionotropic NR type subunit 2A) expression resulted in Ca 2+mediated neurotoxicity. NRs (N-methyl-D-aspartate or NMDA receptors) play a crucial role in the pathogenesis of neurological disorders [85]. Reduced expression of the NRs was found in the patient's brain with neurodegenerative diseases [86,87]. NRs require proper complex formation between GRIN1 and GRIN2 (A-D) subunits to permit Ca 2+ in ux into the cell [88]. GRIN2A, regulates excitatory neurotransmission in the brain and thus plausibly in uence the course of PD. Kong et al. (2015) con rmed that Grin2 is downregulated signi cantly in PD ies against α-synuclein neurotoxicity [89].

Conclusion
This study explored the neuroprotective effect of Tc in ROT-induced PD mouse model. Besides, we have also tried to explore the underlying mechanism of action in restoring the level of various protein molecules including Akt1, Smad7, Gpr37, Grin2A, Tiam1, Tfam, Dnmt1 and Kdm3a etc. In conclusion, the overall ndings of this study provides a new insight that Tc exerts therapeutic effect through the regulation of various signaling pathways by protecting the DAergic neurons and restoring mitochondrial function. However, further studies are needed to explore the bioactive components of Tc that play essential role in regulation of the disease.

Competing interests
The authors declare that they have no competing interests.

Funding
The author(s) received partial nancial support from Banaras Hindu University, Varanasi for this research.

Authors' Contributions
HB and SPS conceived, performed and designed experiments. HB, SPS, CK and HD wrote the manuscript.
HB, SSS, WZ, ASR, RS processed mice tissue for proteomic analysis and perform behavior study. MR and HB perform bioinformatics analysis. All authors approved the nal manuscript.    Venn diagram reprsenting the unique and common proteins between control vs PD and PD vs Tc groups.    Graphical representation of methodology used for LFQ analysis of Control, PD and treatment groups. Proteins were isolate and digested with multiple proteases followed by LC-MS/MS. Each sample were processed in triplicate.