Withania somnifera (L.) Dunal ameliorates neurodegeneration and cognitive impairments associated with systemic inflammation

3–4 months old disease free wistar strain male albino rats with body weight ranges from 130 to 150 g were purchased from NIPER, Mohali, Punjab, India and used in the present study. 3–4 animals were housed per cage with free access to food and water under controlled temperature (25 ± 2 °C) and humidity conditions with 12 h light/dark cycle. Protocols were duly approved by Institutional Animal Ethical Committee (IAEC) of the Guru Nanak Dev University, Amritsar, Punjab, India as per Committee for the Purpose of Control and Supervision of Experiments on Animals (CPSCEA) guidelines for care and use of laboratory animals laid down by same committee. All experimental procedures were conducted according to ARRIVE guidelines.

The animals were randomized in four groups:

12 h after LPS treatment, animals were assessed for their behavior using various behavioral tests like Rotarod, Narrow beam walking test, Novel object recognition test. For detailed biochemical and molecular studies, animals were anesthetized by injecting thiopentone intraperitoneally at the dosage of 1 unit/10 g and decapitated after cervical dislocation to isolate brains. Further, the brains were microdissected to isolate the hippocampus and pyriform cortex (PC) regions of brain.

Western blotting was performed according to protocol as described earlier (Gupta and Kaur 2018) [28]. Briefly the animals (n = 4–5/group) were anesthetized by injecting thiopentone intraperitoneally at the dosage of 1 unit/10 g and decapitated after cervical dislocation to isolate brains. Further, the hippocampus and pyriform cortex (PC) regions of brain were microdissected, homogenized and their protein lysates were prepared. Protein concentration was estimated using Bradford method followed by preparation of samples which were then electrophoretically resolved and transferred onto the membranes. After blocking the membranes, they were probed with monoclonal anti-PSA-NCAM (1:1500), anti-NCAM (1:2000), anti-CaMKIIα (1:2000), anti-CaN (1:2000) anti-phospho Akt (1:2000), anti-MAP2 (1:2000), anti-GAP43 (1:2500), (Sigma-Aldrich St. Louis, MO, USA), anti-TrkB (1:1000) (Trk A & B antibody sampler kit, Cell Signaling Technology, MA, USA), anti-Bcl-xL (1: 1500) for overnight at 4 °C. Further, the membranes were washed and incubated with their respective secondary antibodies (Genei) HRP-labeled anti-mouse IgM (1:2000) for PSA-NCAM, anti-rabbit IgG (1:5000) for pAkt1, anti-rabbit IgG (1:1500) for anti-TrkB, anti-mouse IgG (1:5000) for CaMKIIα, CaN, MAP2, GAP43 (Merck Millipore, USA) for 2 h at 25 °C. α-Tubulin was used as internal control for protein loading. ECL Prime Western Blotting Detection Reagent was used for visualizing immunoreactive bands using Image Quant LAS 4000 system (GE Healthcare, Little Chalfont, UK). For western blots analysis, the expression of each control (in triplicates) was taken as 100% and then percentage change in expression of protein of interest in test samples compared to corresponding control was calculated. The average and corresponding standard error was then calculated using MS Excel and Sigma Stat. Standard error bars were also plotted on the scale of 100. This way of western blot analysis is well accepted and reliable one. Researchers usually present the relative expression in their test samples either in ratio of control (at the scale of 1) [36‐38] or percent of control (at the scale of 100) [39‐43]. We have been presenting our western data in previous lab publications in this pattern [27, 28, 32, 33, 44]. Data was analyzed for change in protein expression.

mRNA expression analysis was done according to protocol as described earlier [28]. Briefly, the total RNA was extracted from brain tissues using TRI reagent (Sigma-Aldrich, St. Louis, MO, USA) according to manufacturer’s instructions (n = 4–5/group). Further, the RNA was reverse transcribed using M-MLV reverse transcriptase to prepare the cDNA. Amplification of gene of interest was then done from 50 μg of cDNA using StepOne Plus Real Time PCR system (Applied Biosystems, Foster City, CA, USA) according to SYBR chemistry. GAPDH was used as an internal control. The ‘Livak method’ was applied to calculate the relative gene expression and data was presented as 2 ^{− ΔΔCt} ± SEM.

Microglial enriched cultures were obtained from primary mixed glial cultures from 1-day-old pups using mild trypsinization method [45] with slight modification as described earlier [27]. Isolated microglial cells were seeded in multiwell plates by trypsinization with 0.25% trypsin-EDTA for 10 mins and treated after 24 h of seeding. For preparation of the microglial conditioned medium, primary microglial cells were pretreated with 0.2% and 10 μg/mL of ASH-WEX and FIV, respectively for 2 h prior to activation with 100 ng/mL LPS [LPS (rough strains) from E. coli F583 (Rd mutant)]. The microglial conditioned medium was harvested after 36 h of treatment, centrifuged to make cell free and used freshly for different experiments. Conditioned medium from different groups were designated as CM-Control, CM-ASH-WEX, CM-FIV, CM-LPS, CM-LPS + ASH-WEX, CM-LPS + FIV.

Primary neuronal cultures were prepared from 1-day-old (cortical), 2-day-old (hippocampal) and 7-day-old (cerebellar) pups, respectively using the procedure followed by Frandsen and Schousboe (1990) with some modifications [46]. Briefly, the pup brain was dissected to collect cortex, hippocampus and cerebellum regions; their meninges were carefully removed and the tissue was digested with 0.025% (w/v) trypsin-EDTA solution containing 0.1% DNase for 15 mins at 37 °C. Activity of trypsin was stopped by adding equal amount of trypsin inhibitor solution or media and centrifuged (1000 rpm, 4 mins). The tissue was titurated and the cells were resuspended in the Neurobasal medium. Then the cells were seeded on poly-L-lysine coated 24 well culture plates and maintained in Neurobasal medium (supplemented with bFGF and B12 supplement) at 37 °C in 5% humidified atmosphere.

For primary microglial–cerebellar neurons co-culture, firstly the primary microglia cultures were prepared from 1-day-old pups as described above. Microglial cells obtained were seeded at a density of 15,000 cells/mL on PLL coated coverslips. After 24 h, microglial cells were pretreated with ASH-WEX and FIV for another 24 h. Then ASH-WEX and FIV containing medium was removed and cerebellar neurons were cultured from 7-day-old pups as detailed above and seeded on the top of the microglial monolayer in a ratio of 2:1 (neuronal: microglial ratio). After 4 h of seeding cerebellar neurons, microglial-neuronal cultures were challenged with 100 ng/ml LPS for 36 h. Microglial-neuronal co-cultures were then used for MAP2 immunostaining and DAPI nuclear staining to determine the percentage apoptotic cell death. Images captured for MAP2 immunostaining were used to determine the percentage neurite outgrowth. Experiments were performed in triplicate. Neurite lengths were determined from at least 100 cells per group using Image Pro Plus software (Media Cybernetics, Silver Spring, USA) and then analysis was done using Microsoft excel and Sigma Stat for windows (version 3.5).

The control and treated cells were washed with 1X PBS thrice, fixed with 1:1 acetone and methanol followed by permeabilization with 0.3% Triton X-100 in PBS (PBST). After blocking with 2% BSA in PBS, cells were incubated with anti-MAP2 (1:200, Sigma-Aldrich) monoclonal antibody diluted in blocking solution for 24 h at 4 °C in a moist chamber. Then the cells were washed twice or thrice with 0.1% PBST followed by incubation with anti-mouse IgG 488 secondary antibody (diluted at a concentration of 1:500 in 2% BSA) for 2 h at room temperature. For nuclear staining, cells were incubated with 1:5000 dilution of DAPI in 1X PBS for 10 mins. Cells were then mounted with anti-fading reagent (Fluoromount, Sigma-Aldrich). Nikon AIR Confocal Laser Microscope was used for capturing the images. Experiments were performed in triplicate. For quantifying MAP2 relative optical intensity, ROIs were selected from different cells (at least 100 cells per group) and analyzed for the intensity using NIS elements AR analysis software version 4.11.00. For neurite outgrowth analysis, at least 100 cells per group were measured for length of neurites using the Image Pro Plus software (Media Cybernetics, Silver Spring, USA) and percentage neurite outgrowth per group was calculated using Microsoft excel and Sigma Stat for windows (version 3.5).

The mean ± SEM of values calculated from minimum of three independent experiments were used to express data. Student’s t-test, one-way ANOVA (Holm-Sidak post hoc method) and two-way ANOVA have been employed to assess the level of significance by using Sigma Stat software for Windows (version 3.5). Values showing p ≤ 0.05 were acceptable as statistically significant. Only where the significance level comes ≤ 0.05 while comparing different groups, the symbol representing significance has been entered in the graphs.

Animals were examined for their motor coordination using rotarod. LPS injected rats exhibited maximum number of falls (Fig. 1a; p ≤ 0.01) and spent significantly less time (Fig. 1b; p ≤ 0.001) on rotating rod in comparison to other three groups. However, LPS + ASH-WEX- and ASH-WEX alone-treated rats exhibited significantly lower number of falls (Fig. 1a; p ≤ 0.01) and spent more time (Fig. 1b; p ≤ 0.001) on rotating rod as compared to LPS alone-treated group. To further confirm the improvement in locomotor activity by ASH-WEX, animals were assessed for the time taken to transverse the narrow beam and the number of paw slippage using narrow beam walk test. LPS-treated rats exhibited maximum number of paw slippages (Fig. 1d; p ≤ 0.05) and also took more time (Fig. 1c; p ≤ 0.01) to transverse the beam compared to control group. LPS + ASH-WEX- and ASH-WEX alone-treated rats swiftly crossed the beam (Fig. 1c; p ≤ 0.01) with least number of paw slippages (Fig. 1d).

Further, the animals were tested for their recognition learning and memory using NOR test (Fig. 2). LPS injected rats were observed to show minimum frequency of episodes and total time spent in exploring novel object over the old one implying their impaired memory in recognizing novel object in comparison to other three groups (Fig. 2a and b). In contrast, LPS + ASH-WEX- and ASH-WEX alone-treated rats exhibited normal and intact memory as evident from more frequency of episodes and time spent in exploring novel object over the old one (Fig. 2a and b). Moreover, preference index (PI) score of these animals was more than 0.5 indicating their higher preference for novel object over the old one (Fig. 2c; p ≤ 0.001), further approving their normal working and recognition memory. While, PI score for the LPS-treated animals was less than 0.5 suggesting their preference for old over the new object (Fig. 2c; p ≤ 0.001).

Moreover, LPS alone-treated rats exhibited longer latency to start grooming with reduced number of grooming bouts and spent less time in grooming during NOR test. Impaired grooming behavior probably could be due to sickness induced by LPS (Fig. 2d and e; p ≤ 0.005). LPS + ASH-WEX- and ASH-WEX alone-treated rats exhibited normal grooming behavior with more duration and frequency of grooming bouts comparable to control rats. Similarly, LPS-treated animals exhibited less number of rearing than control group animals, while, LPS + ASH-WEX- and ASH-WEX alone-treated animals exhibited maximum number of rearing implying their interest in exploration of empty box and reduction in LPS-induced stress by ASH-WEX administration (Fig. 2f; p ≤ 0.001).

LPS treatment upregulated the expression of PSA-NCAM as compared to control rats, whereas its expression in LPS + ASH-WEX- and ASH-WEX alone-treated rats were near to control level in both hippocampus (Fig. 3a and b; p ≤ 0.05) and PC (Fig. 3a and c; p ≤ 0.05) regions of the brain. Similar changes were also observed in the expression of NCAM in both regions (Fig. 3a-c; p ≤ 0.05). LPS treatment also decreased the expression of neuronal growth related protein GAP43 expression up to 87.2 and 90% in hippocampus and PC regions, respectively as compared to control rats (taken as 100%). Whereas, its expression was significantly upregulated up to 116% (hippocampus) and 107% (PC) in LPS + ASH-WEX-treated rats (Fig. 3a-c; p ≤ 0.05). GAP43 expression was also near control level in both brain regions of ASH-WEX alone-treated rats. Furthermore, the expression of the neuronal structural protein MAP2 was normalized by ASH-WEX supplementation which was downregulated by LPS treatment in both brain regions (Fig. 3d-f; p ≤ 0.05).

Further, the expression of Ca²⁺ and calmodulin dependent serine threonine protein phosphatase (calcineurin) and protein kinase (CAMKIIα) were studied in hippocampus and PC regions of brain which play an integral role in synaptic transmission modulation. LPS significantly reduced CAMKIIα expression (up to 71%) as compared to control rats (taken as 100%) while its expression was restored up to 86% in LPS + ASH-WEX-treated rats compared to LPS alone group in hippocampus region (Fig. 3d and e; p ≤ 0.05). In PC region, no significant variations were found in CAMKIIα expression amidst all the four group of animals (Fig. 3d and f; p ≤ 0.05). Its expression remained normal in both brain regions of ASH-WEX alone-treated animals. In contrast, the expression of calcineurin was not altered among all groups in hippocampus (Fig. 3d and e) and PC (Fig. 3d and f) regions of brain.

The neurotrophin BDNF plays important role in neurons proliferation and survival by binding to TrkB receptor to initiate signaling pathway. At transcriptional level, LPS + ASH-WEX- and ASH-WEX alone-treated rats showed near control level expression of BDNF and TrkB in both hippocampus and PC regions which was significantly downregulated after LPS treatment (Fig. 4d and e; p ≤ 0.05). mRNA expression of BDNF was significantly upregulated up to 1.7 fold in PC region of LPS + ASH-WEX-treated animals (Fig. 4e; p ≤ 0.05). Expression of TrkB was also seen to be normalized significantly by ASH-WEX supplementation at translational level in both hippocampus (Fig. 4a and b; p ≤ 0.05) and PC (Fig. 4a and c; p ≤ 0.05) regions. The phosphotyrosine residues of TrkB receptor act as binding sites for specific cytoplasmic signaling and scaffolding proteins that in turn activate PI3K-Akt pathway. The expression of PI3K although did not change in LPS alone-treated animals compared to control group but its expression was found to be markedly upregulated in LPS + ASH-WEX and ASH-WEX alone rats from both hippocampus (Fig. 4d) and PC (Fig. 4e) regions. ASH-WEX supplementation also normalized the mRNA expression of total Akt which was significantly reduced by LPS treatment in hippocampus region (Fig. 4d; p ≤ 0.05). However, Akt mRNA expression did not altered among all groups in PC region (Fig. 4e). Similarly, LPS-induced decrease in expression of phosphorylated Akt was also recovered in LPS + ASH-WEX- and ASH-WEX alone-treated rats from both brain regions (Fig. 4a-c; p ≤ 0.05).

Akt further binds to BAD protein (BCL2 associated death promoter) and causes its inactivation. BAD protein complexes with 14-3-3 proteins in cytoplasm which prevents association with mitochondrially localized Bcl-xL and therefore inhibits apoptosis. LPS treatment marginally reduced Bcl-xL expression up to 95 and 82% from hippocampus and PC regions, respectively in comparison to control rats (taken as 100%), whereas, ASH-WEX administration restored the expression of Bcl-xL up to 107 and 99% in hippocampus and PC regions, respectively (Fig. 4a-c; p ≤ 0.05). ASH-WEX alone-treated rats exhibited near control level expression of Bcl-xL in both brain regions. mRNA expression of Bcl-xL was observed to be near control level in LPS + ASH-WEX- and ASH-WEX alone-treated animals but was significantly reduced with LPS treatment from hippocampus region (Fig. 4d; p ≤ 0.05). In PC region, no significant variations were found in mRNA levels of Bcl-xL among all the groups (Fig. 4e; p ≤ 0.05).

BDNF-TrkB pathway also activates PLCγ signaling pathway which further mediates the signals through IP3 by binding to its receptor and causing rapid release of Ca²⁺ involved to regulate neuronal activity and synaptic transmission. In PC region, ASH-WEX treatment showed 2-fold increase in expression of PLCγ which was decreased significantly in LPS alone-treated animals (Fig. 4e; p ≤ 0.05). mRNA expression of PLCγ remained unaffected among all groups in hippocampus region (Fig. 4d). LPS treatment significantly reduced the mRNA levels of IP3R gene in both hippocampus and PC regions, whereas, its expression was significantly upregulated by ASH-WEX administration (Fig. 4d and e; p ≤ 0.05).

To further validate the neuroprotective effect of ASH-WEX and FIV against LPS-induced neuroinflammation, cell free conditioned medium (CM) of microglial cultures was prepared and tested (as per details in material and methods section). Direct treatment of LPS at the concentration of 100 ng/mL was found to be non toxic to neurons, whereas, neuronal cultures treated with CM-LPS showed significant downregulation in expression of MAP2 (Fig. 5a-c column II) as compared to CM-Control (Fig. 5a-c column II) and LPS alone-treated neurons (Fig. 5a-c column I). In contrast, treatment of neuronal cultures with CM-LPS + ASH-WEX and CM-LPS + FIV (Fig. 5a-c column II) showed near normal expression of MAP2 as compared to LPS + ASH-WEX- and LPS + FIV-treated neurons (Fig. 5a-c column I) as depicted by MAP2 immunostaining (Fig. 5a-c) and their intensity plots (Fig. 5d-f; p ≤ 0.05). MAP2 expression remained near control level in neurons treated with CM-ASH-WEX, CM-FIV, ASH-WEX and FIV alone. Since, neuroinflammation impairs the axonal regrowth and contributes to lack of neural regeneration, so confocal images of MAP2 immunostaining were further analysed for percentage neurite outgrowth (Fig. 5g-i). CM-LPS treatment group showed significant degeneration of neurite outgrowth as compared to CM-Control- and LPS alone-treated cortical, hippocampal and cerebellar neuronal cultures (Fig. 5g-i; p ≤ 0.05). CM-LPS + ASH-WEX and CM-LPS + FIV treatment also restored the neurite outgrowth as shown in histograms (Fig. 5 g-i).

Apoptosis was determined by analyzing nuclear morphology using 4′,6′-diamidino-2-phenylindole (DAPI) staining (Fig. 6). 4, 6-Diamidino-2-phenylindole (DAPI) is a DNA binding dye and is commonly used assay for detection of apoptosis in cells. Apoptotic cells have more permeability for dye so these cells show more intense blue fluorescence due to chromosomal condensation, whereas, normal cells show uniformly stained round nucleus with clear margins. So DAPI staining can be used to observe differences between apoptotic and non-apoptotic cells based on the intensity of fluorescence and the conformation of nucleus. Neurons treated with CM-LPS for 24 h showed both disintegration and condensation of nuclei resembling apoptotic cell death (Fig. 6a-c) and caused significant apoptosis of neurons as indicated by highest percentage of apoptotic cells (Fig. 6d-f; p ≤ 0.05). CM-LPS + ASH-WEX and CM-LPS + FIV treatment effectively suppressed apoptosis of neurons. CM-ASH-WEX, CM-FIV, ASH-WEX and FIV alone treatment did not affect the survival of the neurons. However, CM-LPS + ASH-WEX and CM-LPS + FIV treatment restored the neuronal cell viability thus suggesting the neuroprotective effect of ASH-WEX and FIV (Fig. 6).

To further investigate whether microglia directly participate in neurotoxicity via cell to cell contact or only through secreted chemical mediators of inflammation, we performed the primary microglia-cerebellar neurons co-culture and analyzed for MAP2 neuronal protein expression, percentage neurite outgrowth and percentage neuronal apoptotic cell death (Fig. 7). Microglial-neuronal co-cultures treated with LPS alone exhibited less number of MAP2 positive neurons as well as (up to 67%) reduced MAP2 expression (Fig. 7c; p ≤ 0.05) and showed neuronal degeneration with retracted neurite outgrowth (Fig. 7b) as compared to control co-cultures. Further, the proportion of the cerebellar neurons showing apoptosis like nuclear condensation reached up to 66% (Fig. 7d; p ≤ 0.05) of the total neuronal population after 36 h of co-culture. In contrast, ASH-WEX and FIV treatment normalized MAP2 expression, sustained neuronal viability with increase in length of neurite outgrowth (31% for LPS + ASH-WEX and 52% for LPS + FIV) and reduced the percentage of apoptotic neurons up to 15% (ASH-WEX) and 11% (FIV) as compared to LPS alone-treated co-cultures (Fig. 7b-d; p ≤ 0.05).