Neuroprotective role of Diosgenin, a NGF stimulator, against Aβ (1–42) induced neurotoxicity in animal model of Alzheimer’s disease

Diosgenin is a neurosteroid derived from the plants and has been previously reported for its numerous health beneficial properties, such as anti-arrhythmic, hypolipidemic, and antiproliferative effects. Although several studies conducted earlier suggested cognition enhancement actions of diosgenin against neurodegenerative disorders, but the molecular mechanisms underlying are not clearly understood. In the present study, we investigated the neuroprotective effect of diosgenin in the Wistar rats that received an intracerebroventricular injection of Amyloid-β (1–42) peptides, representing a rodent model of Alzheimer’s disease (AD). Animals were treated with 100 and 200 mg/kg/p.o of diosgenin for 28 days, followed by Amyloid-β (1–42) peptides infusion. Animals were assessed for the spatial learning and memory by using radial arm maze and passive avoidance task. Subsequently, animals were euthanized and brains were collected for biochemical estimations and histopathological studies. Our results revealed that, diosgenin administration dose dependently improved the spatial learning and memory and protected the animals from Amyloid-β (1–42) peptides induced disrupted cognitive functions. Further, biochemical analysis showed that diosgenin successfully attenuated Amyloid-β (1–42) mediated plaque load, oxidative stress, neuroinflammation and elevated acetylcholinesterase activity. In addition, histopathological evaluation also supported neuroprotective effects of diosgenin in hippocampus of rat brain when assessed using hematoxylin–eosin and Cresyl Violet staining. Thus, the aforementioned effects suggested protective action of diosgenin against Aβ (1–42) induced neuronal damage and thereby can serve as a potential therapeutic candidate for AD.


Introduction
Alzheimer's disease (AD) is the most prevalent, complex and devastating neurodegenerative disorder in the aging population and clinically characterized by gradual erosion of cognitive functions, particularly decline in perception and language ability, personality disturbance, inability of selfcare and death from opportunistic infections (Souza et al. 2016). Basal forebrain cholinergic neurons (BFCNs) are considered to be prime neurons responsible for learning, memory, and attention, through acetylcholine (ACh) innervation (Baxter and Chiba 1999). Abnormal formation of intracellular neurofibrillary tangles and extracellular deposition of amyloid beta (Aβ) peptides in the AD brain displayed neuritic atrophy and loss of synapses in BFCNs. Degeneration of BFCNs act as one of the major pathological diagnostic hallmark of AD and is severely affected due to disruption of communication with target neurons in the hippocampus and cortex (Fahnestock and Shekari 2019). With the age, these effects gets worsen leading to the cascade of pathogenic events causing neuroinflammation, glutamate excitotoxicity, metal dyshomeostasis, oxidative stress and mitochondrial dysfunction in specific parts of the brain (Souza et al. 2016;Lian et al. 2017). BFCNs are essentially dependent on neurotrophic factor, nerve growth factor (NGF), belonging to the neurotrophin family, for their neuronal growth, survival, maintenance, and plasticity by binding with tyrosine kinase A receptor (TrkA) and the p75 pan neurotrophin receptor (p75NTR) (Seiler and Schwab 1984;Lad et al. 2003). Growing evidence conveys that altered expression of TrkA/ p75 results in basal forebrain cholinergic metabolic dysfunction due to defective NGF processing and supports accumulation of Aβ (Iulita and Cuello 2014). Interestingly, accumulated Aβ, in AD pathology, in turn causes degeneration of BFCNs and enhances Aβ production by regulating p75NTR overexpression, NGF-induced signalling transduction and internalization of NGF receptors (Zhang et al. 2013).
In recent decades, a number of herbal derived drug compounds, used in traditional medicine, found to improve brain functions by reconstructing neuronal networks and remodelling the atrophic neurites by upregulating NGF expression. For instance, Huperazine A, a novel alkaloid, specific and reversible acetylcholinesterase (AChE) inhibitor, isolated from the Chinese herb Huperzia serrate, elevated NGF levels and thus ameliorated cognitive deficits in AD (Wang et al. 2006). Diosgenin is a one such herbal derived neurosteroid and a promising bioactive biomolecule which has a high industrial value and is the principal precursor compound in the production of therapeutically useful steroidal drugs, including sex hormones and corticosteroids (Masilamoni et al. 2008). It is a natural antioxidant usually derived from Dioscorea rhizome and other herbal drugs, such as Trigonella spp., Polygonatum spp. and Smilax spp with multiple medicinal properties, such as, hypolipidemic (Son et al. 2007), anti-inflammatory , anti-proliferative (Sethi et al. 2018), anti-arrhythmic (Chen et al. 2015), anti-diabetic (Hua et al. 2016), neuroprotective (Lee et al. 2018), and found to have protective effect against cardiovascular disease (Kalailingam et al. 2014) and skin aging . Diosgenin, was reported earlier as a potential NGF which efficiently regulates the expression of TrkA and p75 and activates the intracellular signaling cascades involving pathways that are dependent on extracellular signalregulated kinase (ERK) and phosphatidylinositol 3-kinase (PI3K), which act to prevent nuclear and mitochondrial cell death programs, resulting in neuronal survival and differentiation. In addition, diosgenin boosted memory and exhibited protection and regrowth of axons in Aβ-treated neurons by stimulating 1,25D 3 -membrane-associated, rapid response steroid-binding protein (1,25D3 -MARRS) in 5XFAD mice Tohda et al. 2012). Similarly, diosgenin produced cellular protection against oxidative stress by inhibiting ROS production in neuronal cells in D-galactose induced aging model (Choi et al. 2010;Chiu et al. 2011). All these aforementioned reports provided strong evidences of cognition enhancement property of diosgenin and thus convinced us that diosgenin might be able to attenuate cognitive impairment in the animal model of AD.
Multiple molecular pathogenesis such as tau hyperphosphorylation, Aβ deposition, metal dyshomeostasis, cholinergic hypofunction, glutamate excitotoxicity, neuroinf lammation, oxidative stress, and mitochondrial dysfunction act together in complex way in AD progression. Hence there is a need of a broader ''one-compound-multi-targets'' neuroprotective approach which can collectively target these pathological processes (Pi et al., 2012). In this regard, current study was designed to evaluate the neuroprotective effect of diosgenin on AD related complications in Aβ (1-42) induced Wistar rats, representing sporadic AD model. Based on satisfactory results from our studies we can say that diosgenin could be a promising neuroprotectant and therapeutic solution for AD rescue.

Animal procurement
Forty adult male Wistar rats, aged 3 months, weighing 250-300 g were provided by central animal facility of JSS College of Pharmacy, Ooty, Tamilnadu, India. The study was carried out in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals (National Research Council (US)) and was approved by the Institutional Animal Ethics Committee (IAEC) (JSSCP/IAEC/PhD/Pharmacy Practice/01/2018-19). In order to adapt the environmental conditions of 12-hour light/dark cycle, animals were acclimatized for ten days to the laboratory conditions in polypropylene cages with pellet diet and water ad libitum prior to the commencement of the experiment. All animals were housed five per cage and were maintained under a controlled room temperature of 25 ± 2 °C and relative humidity of 60 ± 5% degree.

Surgical procedure
Rats were anesthetized with ketamine hydrochloride (91 mg/ kg, i.p.) and xylazine (9.1 mg/kg, i.p.). After fixing the animals into a stereotaxic apparatus, a burr hole was drilled through the skull above the bilateral hippocampal coordinates (anterior-posterior (AP) = − 3.5 mm, medial-lateral (ML) = ± 2.0 mm from the bregma and dorsal-ventral (DV) = 2.7 mm from the skull surface) after midline sagittal incision according to stereotaxic atlas. 2 μl containing 10 μg of Aβ (1-42) was subsequently injected as a single dose over 5 min through a microsyringe into the hole (Rahman et al. 2019). To prevent infection, animals were provided with post-operative care by applying antiseptic over the wound and then transferred to a thermo-regulated chamber to maintain normal body temperature until recovery.

Experimental protocol and drug treatment
Animals were randomly divided into 4 groups, containing 10 rats each. Group I was kept as sham operated (SO), where rats in this group received cerebral injection of aCSF (2 μl) bilaterally. Rest of the animals from groups II, III and IV were bilaterally infused with 2 μl/10 μg of Aβ (1-42) peptides dissolved in aCSF, all at the same time. Group II was designed as negative control (NC), indicating AD model and received 0.5% CMC orally for 28 days. Similarly, group III (Dio 100) and IV (Dio 200) were considered as treatment groups, where the test animals were administered with 100 mg/kg/d and 200 mg/kg/d p.o. of diosgenin suspended in 0.5% CMC respectively for 28 days. After the treatment protocol, on 28 th day, all the animals from group II III and IV were bilaterally injected with Aβ (1-42) peptides once as depicted in Fig. 1. Before initiation of behavioural assessment, animals were housed in their respective cages for a week for recovery (Fig. 1). The dose of the diosgenin selected in the present investigation is based on studies conducted earlier (Zhang et al. 2017).

Behavioural cognitive assessments
Radial arm maze (RAM) task RAM was used as a standard method to assess reference and working memory errors in the animals. RAM apparatus consists of 8 identical arms originating from the centre of the elevated platform, height of about 28 cm. Before the task, animals were kept on restricted diet with water being available ad libitum. Animals were trained for 5 consecutive days by simultaneously placing in the radial maze prior to the beginning of drug treatment followed by training once per week during experiment protocol (Fig. 1). Animals were allowed to explore for 5 min and take food freely. Rats were given one training trial per day to run to the end of the arms and consume the bait. During the experiment, four arms were randomly baited with food pellets, provided with a number of visual clues and the training trial continued until all the four baits had been consumed or until 5 min has elapsed. The number of entries into an unbaited arm (referred as reference memory errors) and number of reentries into baited arm (referred as working memory errors) within 5 min were recorded during the investigation. If the animal visited all the arms before 5 min then the trial was said to be accomplished (Sehgal et al. 2012). An arm entry was counted when all four limbs of the rat were within an arm. After each trial the arms were cleaned with 70% of alcohol. At the end of the experimental protocol, i.e. on 28 th day, animals were evaluated for spatial memory using RAM task.

Passive avoidance test
Passive avoidance test was used to examine the long term memory based on negative reinforcement. The test apparatus consists of a small chamber (illuminated with a 50 W bulb) connected to a dark chamber separated by a guillotine door. All the animals were placed in the bright compartment with opened guillotine door. As the rats entered the dark chamber, door was closed and provided with a low intense foot shock (1 Hz, 5 s, 1/5 MA). The total time taken by the animal to transfer from the bright to dark compartment was recorded as step-through latency. After 24 h, retention test was conducted where animals were placed in the illuminated compartment and total time spent in the dark chamber was recorded. (Barkur and Bairy 2015).

Brain homogenate preparation
After completion of neurobehavioural assessment, animals were sacrificed under ether anaesthesia. Brains were extracted and cleaned with ice saline. The cerebral hemispheres were separated and the left hemisphere was fixed in neutral buffered formalin (10% v/v) for histological examination. The hippocampus was dissected from the right hemisphere and 10% (w/v) homogenate of brain sample (0.03 M sodium phosphate buffer, ph-7.4) was prepared by using Teflon glass homogenizer. The homogenate was centrifuged at 1000 rpm at 4 °C for 3 min. Supernatant was separated and used for biochemical studies. Hippocampal protein was measured by the method of Lowry et al. (1951) using bovine serum albumin (1 mg/ml) as a standard.

Determination of hippocampal AChE activity
AChE activity was evaluated in rat hippocampus by using Ellman's method (Ellman et al. 1961). The assay mixture consisted of 0.05 ml of supernatant, 3 ml of 0.01 M sodium phosphate buffer (pH 8), and 0.1 ml of 0.2 mM 5,5-dithiobis-(2-nitrobenzoic acid) (DTNB, Ellman's reagent). About 0.1 ml of acetylthiocholine iodide was added and the change in the absorbance was measured at the 30 s interval for 5 min at 412 nm. The enzyme activity was expressed in micromoles of acetylthiocholine iodide hydrolysed per min per g of protein. AChE activity was calculated using an extinction coefficient of 13.6 mM −1 cm −1 (Khalil and Abass 2017).

Estimation of superoxide dismutase (SOD) activity
The activity of SOD in brain hippocampus was analysed by the method suggested by Kakkar et al. (1984). An assay mixture was prepared by adding 1.2 ml of 0.052 M sodium pyrophosphate buffer (pH 8.3), 0.1 ml of 186 μM phenazonium methosulphate, 0.3 ml of 300 μM nitroblue tetrazolium and 0.1 ml brain homogenate. To the mixture, 0.2 ml of 780 μM nicotinamide adenine dinucleotide (NADH) was added to begin the reaction. Reaction mixture was incubated for 90 s at 30 °C followed by addition of 0.1 ml of glacial acetic acid to stop the reaction. 4 ml of n-butanol was added to the mixture and stirred vigorously and centrifuged at 4000 rpm for 10 min. The absorbance of organic layer was read at 560 nm. SOD activity was expressed as units/ min/mg protein.

Estimation of catalase (CAT) activity
Catalase activity was measured using spectrophotometric procedure according to the method of Aebi (1984). A mixture of 0.1 ml of brain homogenate and 1.9 ml of 50 mM phosphate buffer was incubated at 25 ºC for 30 min. To this 1.0 ml of freshly prepared 30 mM H 2 O 2 was added to initiate the reaction. The change in absorption was taken for 3 min at 240 nm at an interval of 30 s. CAT activity was expressed as μmol/ min/g protein.

Estimation of glutathione reductase (GSR) activity
GSR activity was estimated by the method proposed by Carlberg and Mannervik (1975). The reaction mixture was prepared by adding 0.1 ml of brain homogenate, 1.65 ml of 0.1 M phosphate buffer (pH 7.6), 0.1 ml of 0.1 mM nicotinamide adenine dinucleotide phosphate (NADPH), 0.05 ml of 1 mM oxidized glutathione and 0.1 ml of 0.5 mM ethylenediamine tetraacetic acid (EDTA). The absorbance of each sample was noted at 340 nm. The GSR activity was expressed as nmol/min/mg protein.

Histopathological analysis of brain
The left hemisphere of the brain was stored overnight in neutral buffered formalin (10% v/v) followed by embedment in paraffin for 4 h. The hippocampus was sectioned at 5 μL thickness using a Leica RM 2135 microtome. Sections were mounted and washed in xylene for deparaffinization followed by rehydration in graded ethanol and finally stained with hematoxylin-eosin (H & E). Another batch of samples were stained with cresyl violet (CV) acetate. Hippocampal neurons were observed and captured at 40 X magnification under light microscopy.

Statistical analysis
The results were analysed using Graph Pad Prism 8.0.2 (263) and values were expressed as mean ± standard deviation (SD). Statistical analysis between groups was measured using one way analysis of variance (ANOVA) followed by Tukey's post hoc multiple comparisons. Values were considered to be significant when p < 0.05.

Effect of diosgenin on RAM task
In RAM task, animals from NC group showed poor ability in learning the task and made high number of reference (p < 0.001, F (3, 36) = 45.3) and working (p < 0.001, F (3, 36) = 61.5) memory errors as compared to SO group. These results showed a significant deficit in spatial cognition after i.c.v. infusion of Aβ (1-42) peptides in NC animals, indicating successful induction of AD like pathology. In contrast, administration of diosgenin significantly reduced reference memory errors at 100 mg/kg (p = 0.004) and 200 mg/kg (p < 0.001) when compared to NC rats ( Fig. 2A). Likewise, working memory errors was also turned down remarkably in Dio 100 (p = 0.003) and Dio 200 (p < 0.001) animals (Fig. 2B).

Effect of diosgenin on passive avoidance task
Results from passive avoidance task implicated compromised aversive associative learning in NC rats by significantly reducing step-through latency (p < 0.001, F (3, 36) = 86) and increasing total time spent in the dark chamber (p < 0.001, F (3, 36) = 38) when compared to SO animals. However, rats administered with diosgenin at 100 mg/kg (p = 0.007) and 200 mg/kg (p < 0.001) significantly enhanced the step-through latency when compared with NC (Fig. 3A). On the contrary, total time spent in the dark chamber was dose dependently reduced in Dio100 (p = 0.009) and Dio 200 (p < 0.001) animals (Fig. 3B).

Effect of diosgenin on hippocampal AChE activity
Effect of diosgenin on brain AChE activity is depicted in Table 1. Biochemical analysis of brain hippocampus indicated high AChE activity in NC when compared to SO group (p < 0.001, F (3, 36) = 19.8). Animals that received diosgenin at 100 mg/kg (p < 0.01) and 200 mg/kg (p < 0.01) manifested enhanced nootropic effect by exhibiting significant decline in AChE activity as compared to NC rats.  (Table 1).

Histology of Hippocampus
H&E and CV staining were used to examine the neuroprotective activity of the diosgenin in rat hippocampus. In the current study, H& E staining in SO rats displayed intact neurons with prominent nucleoli in CA1 region of hippocampus. Aβ (1-42) peptides infusion in NC animals resulted in significant (p < 0.001, F (3,36) = 53.1) increase in protruding eosinophilic cytoplasm, swelling of neurons, pyknotic nuclei, pyramidal cells shrinkage and dispersed vacuolization identified from H& E staining as compared to SO group (Fig. 4). Diosgenin led to the significant reversal of the condition with mild neuronal toxicity, reduced pyknotic nuclei with lesser number of eosinophilic stained neurons and vacuolization at 100 mg/kg (p = 0.001) and 200 mg/kg (p < 0.001) when compared to NC. Similarly, on comparison with SO group, extensive neurodegeneration in the hippocampus of NC rats was observed as number of Nissl bodies with blue staining was reduced drastically when assayed by using CV stain, indicating disrupted cell viability (p < 0.001, F (3,36) = 22.9). CV positive neurons were dense and easily detected with more blue-stained Nissl bodies in CA1 region of SO hippocampus. Hence, from our experiment, we can convey that neuronal integrity and the number of CV-positive neurons in diosgenin treated groups were significantly improved at 100 mg/kg (p = 0.014) and 200 mg/kg (p < 0.001) as compared to NC (Fig. 5).

Discussion
One of the biggest challenge in the world is how to minimize and control the number of patients suffering from AD which is raising day by day (Mahdy et al. 2012). On this context, in the past decades, several clinical trials were carried out in the hunt of new alternative therapeutics but were failed due to serious adverse effects or no significant therapeutic efficacy (Rahman et al. 2019). As complications associated with AD have complexed the development of the new curative, hence, it is crucial to search for the other strategies. Present study aimed to investigate the neuroprotective effects of diosgenin on Aβ (1-42) peptide induced oxidative stress, plaque accumulation, neuroinflammation and cholinergic dysfunction.
Intracerebroventricular injection of aggregated Aβ (1-42) in rat brain is a conventional AD animal model. Incubation of Aβ (1-42) peptides in aCSF for several days resulted in aggregation of peptides due to conformational transformation from random coil to β-sheet causing enhanced peptide neurotoxic potency (Nakamura et al. 2001). Earlier it was The values were expressed as mean ± SD (n = 10) and were tested by one-way ANOVA (df1 = 3, df2 = 36) followed by Tukey post hoc test. The subscript a (p < 0.001) and b (non-significant) denotes statistical significance versus SO group; c (p < 0.05), d (p < 0.01) and e (p < 0.001) versus NC group; f (p < 0.01) and g (non-significant) versus Dio group

Groups
AChE activity (μM/ min/g protein)  reported that single acute i.c.v injection of aggregated Aβ (1-42) in the rat mimics amyloid load as observed in the brains of AD patients, in both biochemical and histopathological studies. In addition, it also manifested memory dysfunction along with long-lasting disruption of both spatial and contextual fear memories, as well as short-term working memory leading to the neuronal death in rodents (Wang et al. 2014;Fu et al. 2006). Based on these observations, in the present study, we employed aggregated Aβ (1-42) infusion in rat brain as sporadic AD model. In order to investigate the effect of diosgenin on Aβ (1-42) induced cognition impairment, we performed RAM and passive avoidance task, two behavioural paradigm in the present study. RAM task was used to evaluate learning ability and memory in the rats by measuring the number of spatial memory errors made during specified period. Likewise, passive avoidance task was conducted in order to learn the effect of diosgenin on memory functions by measuring avoidance response of diosgenin treated rats on unpleasant stimulus. Reports from the earlier studies documented that Aβ (1-42) mediated hampered ACh release ensues initiation of cognitive inadequacy in AD pathology, associated with altered memory performances (Xu et al. 2016). Further, Aβ (1-42) peptides infusion in rat brain induces neurobehavioural deficits and underpins the amyloid accumulation by degenerating BFCNs (Turnbull et al. 2018;Baker-Nigh et al. 2015). Similar response was observed in our study. Animals injected with Aβ (1-42) peptides in the brain exhibited deterioration in the short term and long term memory, as significant decline in right choices and latency was observed in RAM and passive avoidance task respectively. These results are in consistent with the former experiments (Yamada et al. 1999;Postu et al. 2018). Diosgenin administration in rats contributed in improvement of spatial learning behaviour task, demonstrating intact learning and memory functions. Few studies conducted earlier represented reversal of impaired learning and memory capacity by diosgenin in D-galactose induced aging rats and in 5XFAD mice model of AD (Chiu et al. 2011;Wang et al. 2018;Tohda et al. 2012). Our results are in congruous with these earlier observations. Previously, scientists have confirmed that increase in NGF biosynthesis initiates signalling cascades accountable for cell survival, growth and release of ACh through the cortico-hippocampal projections (Iulita andCuello 2014, Mitra et al. 2019). As reported earlier, diosgenin being a potential NGF stimulator, might have produced protective effect on behavioural performances of cognitive impaired rats by improving ACh innervation to the cortex and hippocampus ). In addition, decline in the Aβ (1-42) levels, AChE activity, oxidative stress and neuroinflammatory cytokines by diosgenin treated animals as seen in the current study might be collectively responsible for improved neurobehavioural deficits.
Elevated levels of Aβ in brain, majorly neurotoxic Aβ (1-42) is the primary reason of cognitive decline in AD (Andreeva et al., 2017). Previous studies have demonstrated Fig. 5 Histopathological examination of CA1 region of hippocampus stained with Cresyl Violet (40X). (A) SO represents healthy neurons with prominent nuclei and densely stained, (B) NC infused group representing neuronal damage with lesser number of stained neuron indicating less viable neurons (C) diosgenin (100 mg/ kg) and (D) diosgenin (200 mg/ kg) treated groups representing mild neuronal injury with higher number of stained neurons, (E) quantitative assessment of number of healthy neurons in hippocampal sections of rats brain. The values were expressed as mean ± SD (n = 10) and were tested by oneway ANOVA (df1 = 3, df2 = 36) followed by Tukey post hoc test. The subscript a (p < 0.001) and b (non-significant) denotes statistical significance versus SO group; c (p < 0.05) and d (p < 0.001) versus NC group; e (non-significant) versus Dio group E that direct administration of aggregated Aβ (1-42) into the brain of rodents had caused memory deficits similar to that of AD patients by inducing plaque load (Yamada et al. 1999;Rahman et al. 2019). Our results are in agreement with these reports. After a series of investigation, it was reported that Caprospinol, analog of diosgenin, dramatically restored cognitive decline by reducing amyloid deposits in FAB (Fe 2+ , Aβ 42 and Buthionic sulfoxime) infused rat model (Lecanu et al. 2004;Lecanu et al. 2009;Lecanu et al. 2010;Papadopoulos and Lecanu 2012). Results from these studies support our findings as we noticed elevated plaque load when investigated after direct administration of amyloid beta (1-42) fibrils in rat brain when compared with SO group. However, rats from treatment groups showed reduced level of Aβ (1-42) when compared to NC group, indicating clearance of plaque load. In recent years, it has been clear that amyloid deposits in the AD brain contributes to neurodegeneration by directly accelerating the membrane-associated oxidative stress and inflammatory responses, leading to the abnormal energy metabolism and the loss of synaptic functions (Tönnies and Trushina 2017). Hence, in addition to plaque clearance, reduced oxidative stress and neuroinflammatory markers after diosgenin treatment might owed to its anti-amyloidogenic effect by preventing further production and accumulation of neurotoxic Aβ (1-42). However, further studies are needed to be conducted which can clarify the role of diosgenin on other major contributing factors behind Aβ accumulation in brain such as β-secretase, γ-secretase, Aβ-accumulation inhibitors and Aβ-degrading enzymes (Burg et al. 2013).
A large volume of evidence has described about the relationship between Aβ infusion and cholinergic dysfunction in selected AD brain regions. Aβ (1-42) accumulation is associated with loss of cortical synapse and atrophy of BFCNs which is responsible for hampered cholinergic transmission and enhanced AChE activity due to altered NGF metabolism (Counts and Mufson 2005;Mufson et al. 1995;Niewiadomska et al. 2011). Also, reports have confirmed that presence of aggregated Aβ (1-42) in the brain enhances the AChE expression and forms a stable complex with Aβ (Aβ-AChE). These complexes accelerate Aβ toxicity, Aβ deposition, neuronal cell loss, mitochondrial dysfunction and influences astrocyte hypertrophy (Hu et al. 2003). Our study confirmed escalated AChE activity after i.c.v. infusion of Aβ (1-42) in rats which is corresponding to a previous report (Rahman et al. 2019). However, from our research we noticed that administration of diosgenin before AD induction in rats attenuated the elevation of AChE activity dose dependently. Previous study by Koh et al. (2016) demonstrated that diosgenin stimulated NGF biosynthesis via regulating TrkA mediated NGF signalling cascades, resulting in neuronal survival and differentiation of cholinergic neurons. Thus, the probable reason behind diosgenin mediated AChE activity inhibition could be due to enhanced NGF stimulation. Also, reduction in the levels of oxidative stress and neuroinflammatory markers by diosgenin treated animals as observed in the current study could be other contributing factors associated with decreased AChE activity.
Increased plaque load after Aβ (1-42) infusion in the AD brain activates microglial cells which generates pro-inflammatory cytokines (including IL-1β and TNF-α), as well as reactive oxygen species (ROS) and produces neuroinflammation, causing detrimental effect on neurons (Calsolaro and Edison 2016;Shal et al. 2018). Further, chronic neuroinflammation in turn exacerbate AD complication by providing neurotoxic environment (Essawy et al. 2019). Our study indicated increase in pro-inflammatory cytokines levels (IL-1β and TNF-α) after Aβ (1-42) infusion in the rat brain. Diosgenin treatment before induction of AD like condition, provided a protective effect against Aβ (1-42) induced neuroinflammation by arresting the cytokines levels (IL-1β and TNF-α), strengthening the evidence of anti-inflammatory property of diosgenin. These results are comparable with the previous studies which displayed the ability of diosgenin and its analog in suppressing microglia mediated neuroinflammation in different models by inhibiting the Toll-like receptor (TLR)/nuclear factor kappa B (NF-κB) pathway (Hirai et al. 2010;Zhao et al. 2017;Li et al. 2018, Binesh et al. 2018) and phosphorylated-p38 mitogen activated protein kinase (MAPK) signalling pathways Cai et al. 2019;Zhao et al. 2017). Hence, from these reports we can assume that by acting on the similar pathways, diosgenin was able to attenuate elevated cytokines levels in AD like pathogenesis. Moreover, modulation of ACh mediated cholinergic-anti-inflammatory pathway by stimulating NGF biosynthesis by diosgenin as reported from previous studies might support the desired activity (Mitra et al. 2019;Koh et al. 2016). Also, ROS scavenging property of diosgenin as illustrated in this study could be the other possible molecular explanation.
Amyloid accumulation along with the degenerated BFCNs after aggregated Aβ (1-42) infusion in rodents promotes free radical release causing oxidative stress (Boyd-kimball et al. 2004). It is exhibited by impairment in the functions of various endogenous antioxidant enzymes involved in maintenance of ion homeostasis such as superoxide dismutase, glutathione reductase, thioredoxins, and catalase, leading to neuropathological alterations (Feng and Wang 2012;Boyd-kimball et al. 2004). Although, the precise mechanism by which such an oxidative effect occurs is still unclear. It was proposed that senile plaques has high levels of copper which promotes reduction of peptide-bound Cu 2+ to Cu + and form hydrogen peroxide (H 2 O 2 ) and further reacts with H 2 O 2 to form highly reactive hydroxyl free radicals (Butterfield et al. 2013). Generated oxidative stress triggers apoptosis of neuronal cells and hinders the brain functions, resulting in impaired learning and memory deficits (Lennon et al. 1991;Alzoubi et al. 2013;Serrano and Klann 2004;Tuzcu and Baydas 2006). As the brain cells are vulnerable to oxidative stress because of its high oxygen consumption and impaired antioxidant enzymes activities, therefore, regulation of antioxidant enzymatic activities is crucial for developing new therapeutics for the management of neurodegenerative diseases (Bayrakdar et al. 2014). In the present study, reduced activities of antioxidant enzymes such as SOD, GSR and CAT was noticed in negative control rats. Our results are in consistent with the earlier reports (Ledezma et al. 2021;Bayrakdar et al. 2014) and indicates successful induction of oxidative stress after Aβ (1-42) infusion. Known for potent anti-oxidant effect (Son et al. 2007;Jagadeesan et al. 2012), diosgenin ameliorated hippocampal antioxidant activities in diseased rats by decreasing the elevated oxidative stress markers (SOD, CAT, and GSR). Several reports from previous studies stated that NGF attenuates oxidative stress by regulating PI3K, Akt, MAPK-JNK, p53, and NF-κB signalling pathways (Salinas et al. 2003;Chiu et al. 2011;Kaplan and Miller 2000). As diosgenin was earlier reported as a potential NGF regulator , hence, regulation of these signalling pathways by diosgenin could be the possible explanation behind its antioxidant potential. Besides, brain inflammatory response to the deposition of extracellular amyloid plaques mediated activated microglia is another source of ROS production (Tönnies and Trushina 2017). As evident from the present study, diosgenin has produced protective effect against Aβ (1-42) induced neuroinflammation, which could be another reason behind the radical scavenging property of diosgenin.
Furthermore, neuroprotective role of diosgenin was further supported by histopathological analysis. One of the most profoundly affected region in AD brain is the hippocampus. The hippocampus is comprised of dentate gyrus, CA1, CA2, CA3, and subiculum subregions which are interconnected but play distinct roles in memory and are differentially affected by disease. However, CA1 is the first effected hippocampal areas in AD brain and constitutes the primary output of the hippocampus (Masurkar 2018). CA1 neurons are responsible for encoding spatial memory such as in vivo physiological responses to changing locations (Hartley et al. 2013) as well as establishing non-spatial memories such as fear (Izquierdo et al. 2016). Also it plays an important roles in anxiety, goal-directed behavior (Ciocchi et al. 2015) and social memory (Okuyama et al. 2016). The number of neurons determines the functional capacity of the brain or any particular neural structure. Previously, several researchers demonstrated significant decline of the neuronal density in CA1 area, as compared to all other areas in AD (West et al. 1994;Padurariu et al. 2012;Spires et al. 2006). Thus, in the present study, we investigated only CA1 region of hippocampus. H & E staining revealed Aβ mediated neuronal damage in CA1 region of the hippocampus of Aβ (1-42) induced AD rats and demonstrated pyknotic nuclei, disorganization of neurons, nuclei swelling and neuronal shrinkage which is in accordant with the previous report (Rahman et al. 2019). Additionally, in cresyl violet staining we observed neuron swelling, vacuolization and apoptotic cells upon Aβ (1-42) induction in CA1 region of the hippocampus of NC rats. Diosgenin dose dependently protected the neuronal cells from the deteriorating effect of Aβ (1-42) and intensified the number of healthy neurons with prominent nuclei. These improved histopathology outcomes by diosgenin are in consistent with our behavioural and biochemical assessments.

Conclusion
The present work strongly suggests neuroprotective property of diosgenin which was in accordance with the available evidences against AD. The i.c.v. infusion of Aβ (1-42) peptides in Wistar rats, demonstrated behavioural disturbance together with plaque load, neuroinflammation, oxidative stress and cholinergic dysfunction. After behavioural, biochemical and histopathological evaluation, our study indicated that oral administration of diosgenin has improved the neurobehavioural performances of Aβ (1-42) peptides infused AD rats. Likewise, diosgenin also attenuated the Aβ (1-42) peptides induced elevated AChE activity and produced protective effect against plaque load, oxidative stress and neuroinflammation mediated neuronal damage. NGF stimulation along with cumulative defensive effects against different AD pathologies by diosgenin could possibly be responsible for its effectiveness against memory impairment in AD rats. However, possibility of introducing diosgenin for the management of AD and its related complications demands further investigation to elucidate the impact of diosgenin on the other neurochemicals involved in AD and identification of molecular mechanism behind its neuroprotective effects.
Data availability The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.

Declarations
Ethical approval All experiments were conducted in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals (National Research Council (US)) and was approved by the Institutional Animal Ethics Committee (IAEC) (JSSCP/IAEC/PhD/ Pharmacy Practice/01/2018-19).