Ethnomedicinal Plants with Protective Effects against Beta-Amyloid Peptide (Aβ)1-42 Indicate Therapeutic Potential in a New In Vivo Model of Alzheimer’s Disease

Alzheimer’s disease (AD) is a progressive neurodegenerative disorder with unmet medical need. This investigation consisted of testing a range of ethanolic ethnomedicinal plant extracts (n = 18) traditionally used in the treatment of disorders such as anxiety, delirium, and memory loss. They were then screened for in vitro inhibitory activity against acetylcholinesterase (AChE), butylcholinesterase (BuChE), beta-secretase 1/beta-site amyloid precursor protein (APP) cleaving enzyme 1 (BACE1), and antioxidant activities. Plants with potent activities were further characterised using a recently developed in vivo model of AD, Globodera pallida. The ability of phytoextracts to protect this organism against amyloid-beta Aβ (1-42) exposure was assessed by measuring chemosensing, survival rate, production of reactive oxygen species (ROS), and antioxidant responses. Extracts (n = 5) from Juglans regia (leaves), Ellettaria cardamomum (seeds), Cinnamomum zeylanicum (bark), Salvia officinalis (leaves/flowers), and Hypericum perforatum (flowers) exerted concentration-dependent inhibitory activities against AChE and BuChE. Three of these plant extracts (i.e., J. regia, E. cardamomum, and S. officinalis) possessed strong concentration-dependent inhibitory activity against BACE1. Furthermore, the five selected medicinal plant extracts not only enhanced significantly (p < 0.05) the nematode’s chemosensing, survival rate, and antioxidant responses (i.e., anti-ROS production, mitochondrial reductase activity, oxidized glutathione (GSSG) to reduced glutathione (GSH) ratio), but also greatly restored (p < 0.05) in a concentration-dependent manner the Aβ (1-42)-induced deleterious changes in these same parameters. In brief, this investigation highlights plant extracts with strong anti-AD activities which could be trialled as novel therapeutic supplements or undergo further biodiscovery research.


Introduction
Therapeutical options for Alzheimer's disease (AD) are currently quite limited. The current available medications (e.g., AchE inhibitors such as galantamine, rivastigmine, and donepezil) often provide symptomatic relief, and some may slow disease progression, but long-term use can be associated with side effects and diminished effectiveness. The development of novel, safe, and cost-effective therapies for AD is a global public health priority [1]. Natural products represent potential sources of novel lead compounds, and

AChE and BuChE Inhibitory Activities
AChE and BuChE inhibitory activities were measured by the Ellman method [16]. Briefly, 125 µL of the chromogen reagent 5,5 dithiobis nitro benzoic acid (3 mM DTNB) was added to wells of a 96-well plate, followed by either 25 µL acetylthiocholine iodide substrate (1.5 mM ATCI) or 25 µL s-butyrylthiocholine chloride substrate (15 mM BTCC). Next, 50 µL phytoextracts (n = 18; 5 mg/mL) were diluted in 1× PBS to concentrations of 25-500 µg/mL. The diluted phytoextracts were subsequently filtered using a 0.22 µm filter and placed in the prepared 96-well plate. Enzymatic reactions were commenced by the addition of 25 µL at the optimal unit amount of either AChE (0.4 U) or BuChE (0.04 U). Plates were then placed in a plate incubator (VWR ® symphony ™ , Lutterworth, Leicestershire, UK) at 37 • C for 30 min. The spectrophotometric absorbance was recorded at 405 nm using a microplate reader (Tecan Safire 2, Thermo Fisher Scientific, Enderby, Leicester, UK), and the inhibitory activities were expressed as percentages.

Antioxidant Activity
Antioxidant activity was measured in a 96-well plate by the commercial colorimetric acid 2,2 -azino-bis (3-éthylbenzothiazoline-6-sulphonique (ABTS) antioxidant assay kit (Sigma, UK) in accordance with the manufacturer's instructions. Trolox [6-Hydroxy-2,5,7,8tetramethylchroman-2-carboxylic acid], a water-soluble vitamin E analogue, served as a PC because it could inhibit the formation of the radical cation in a dose-dependent manner. Stock solutions of pre-screened phytoextracts (n = 7; 5 mg/mL) were diluted 1:2 with the kit assay buffer (2.5 mg/mL), and then serial diluted concentrations (4.3-140 µg/mL) were obtained. The seven selected phytoextracts were: J. regia, E. cardamomum, C. zeylanicum, S. officinalis, and H. perforatum, F. thunbergii, and P. cupana. Assay buffer was used as a NC. Reactions stood at room temperature (RT) for 5 min before 100 µL of Stop Solution was added. Thereafter, the absorbance was immediately measured (405 nm). The experiments were conducted in duplicate, and all results were interpolated from a Trolox standard curve (mM).

Chemotaxis Studies in G. pallida
Culture, maintenance, hatching, and testing of G. pallida have been previously described [11]. Briefly, freshly hatched second-stage juvenile (J2s) nematodes were used in each assay. For chemosensory assays, about 100 J2s were washed in spring water, centrifuged (2500 rpm; 2 min), re-suspended in spring water, and transferred to 24-well plates. Aβ (1-42) (GL Biochem, Shanghai, China) was prepared at 100 µM as well as plant extracts diluted to the concentrations indicated in the diagrams with 1% DMSO in ddH 2 O, which was also used as vehicle control. G. pallida J2s were exposed to Aβ (1-42) (100 µM) and/or the selected plant extracts (n = 5; 10-100 µg/mL) in 24-well plates at 16 • C for 24 h. The five selected phytoextracts were: J. regia, E. cardamomum, C. zeylanicum, S. officinalis, and H. perforatum. Petri dish (3 cm) 'arenas' were prepared on agar, with either water or mannitol (chemotractant), as previously described [11]. After washings, centrifugation (2500 rpm; 2 min), and resuspension in spring water, the treated G. pallida were transferred from the 24-well plates to the centre of the petri dish 'arenas', which were then left for 2 h at RT on a Antioxidants 2022, 11, 1865 5 of 18 flat surface. Organisms on each side of the 'arena' were counted, and chemotaxis index (CI) was calculated as previously described [11].

Phytoprotective Effects against Aβ (1-42)-Induced Impairments in G. pallida
The survival rate (%) was determined by incubating, for 24 h at 16 • C, G. pallida J2s (50/well) in 24-well plates with 100 µM Aβ  at and/or 100 µg/mL plant extracts. The viability was estimated by naked-eye observation under a light microscope, according to a previously detailed method [11]. Live organisms had characteristic 'body bends' or any movement, whereas dead organisms had characteristic 'poker straight' shapes [11].
The mitochondrial reductase activity was determined, as previously detailed [11], using Alamar Blue (Life Technologies, Paisley, UK), in accordance with the manufacturer's instructions. Briefly, G. pallida J2s (100/well) were placed in 96-well plates containing 100 µM Aβ (1-42) and/or 100 µg/mL plant extracts. Alamar Blue (10 µL) was then added to each well, and the fluorescence (Ex: 530 nm; Em: 590 nm) was measured every 30 min for 24 h at 18 • C. Controls included a NC (medium control without cells-background absorbance), a vehicle control (untreated cells, which is shown in the diagrams), and a PC (100% reduced Alamar Blue).
The quantitative measurement of the total glutathione content within a sample (GSH/GSSG) was determined, as previously reported [11], using OxiSelect ™ Total Glutathione Assay Kit (Cell Biolabs, Inc., San Diego, CA, USA), in accordance with the manufacturer's instructions. Glutathione Reductase reduces oxidized glutathione (GSSG) to reduced glutathione (GSH) in the presence of nicotinamide adenine dinucleotide phosphate (NADPH). Subsequently, the chromogen reacts with the thiol group of GSH to produce a coloured compound that spectrophotometrically absorbs at 405 nm. The total glutathione content in unknown samples is determined by comparison with the predetermined glutathione standard curve. Briefly, 25 µL of the 1× Glutathione Reductase solution was added to each well to be tested in 96-well microtiter plates prior to the addition of 25 µL of the 1× NADPH solution to each well to be tested. Then, 100 µL of the prepared glutathione standards or samples to be tested in G. pallida J2s (1000/well) (i.e., 100 µM Aβ (1-42) and/or 100 µg/mL plant extracts) was added to each well and mixed thoroughly. Untreated cells were used as a vehicle control. Eventually, 50 µL of the 1× Chromogen was added and mixed quickly. The optical density (OD)/absorbance was immediately recorded at 1-2 min intervals for 10 min. The rate of chromophore production is proportional to the concentration of glutathione within the sample. The average of each standard, sample, and background absorbance value against incubation time was graphed. The total glutathione content was calculated by determining the slopes for each value from the linear portion of each curve, subtraction of the background, then plotting the net slopes of the GSSG standards against the micromolar concentration of GSSG. Unknown concentrations were determined from the interpolation of the standard curve.

Statistical Analyses
All experiments were performed in triplicate. All data are expressed as mean ± standard error mean (SEM). All statistical analyses including linear/non-linear regression, enzyme kinetics (Km and Vmax), and one-way ANOVA (with Tukey's post hoc test) were undertaken with Graphpad Prism (GraphPad, California, USA). Statistical significance and insignificance are indicated in figures: ns (not significant) means # p > 0.05, whereas * p < 0.05, ** p < 0.01, and *** p < 0.001 represent significant differences between groups.

AChE inhibitory activity
BuChE inhibitory activity  Concentration-dependent effects of plant extracts on acetylcholinesterase (AChE) and butyrylcholinesterase (BuChE) and inhibitory activities. A total of 18 plant extracts were screened using both AChE and BuChE inhibitory assays. Di-nitro-thiobenzoate was added to wells of a 96-well plate, followed by either acetylthiocholine iodide (ATCI) or butyrylthiocholine chloride (BTCC) as substrates. Phytoextracts were tested in duplicate at concentrations ranging from 25-500 µg/mL. Mean percentage inhibition data is summarised in the heatmaps. Positive control: galantamine (18 µM in AChE experiments or 2 µM in BuChE experiments); negative control: sodium phosphate buffer.

Effects on Chemotaxis in G. pallida
The concentration-dependent effects of the selected medicinal plant extracts (n = 5, 10-100 µg/mL) on the chemotaxis index (CI), both in the presence and absence of Aβ (100 µM), are shown in Figure 4.
achieved almost complete restoration to normal after exposure of the nematodes to Aβ (1-42)-induced impairments. The responses behaved in a dose-dependent manner.

Phytoprotective Effects against Aβ (1-42)-Induced Impairments in G. pallida
The ability of the selected plant extracts (n = 5, 100 µg/mL) to exert protective effects against Aβ (1-42)-induced toxicity was investigated in G. pallida. The effects of these plant extracts (i.e., J. regia, E. cardamomum, C. zeylanicum, S. officinalis, and H. perforatum) were evaluated on cell survival rate and oxidative stress indexes (i.e., total ROS production, No extracts adversely affected CI at the highest concentration of 100 µg/mL, and all the plant extracts enhanced it. All the five selected plant extracts (i.e., J. regia, E. cardamomum, C. zeylanicum, S. officinalis, and H. perforatum) at all tested concentrations significantly (p < 0.05) ameliorated chemotaxis of untreated G. pallida (vehicle control) and achieved almost complete restoration to normal after exposure of the nematodes to Aβ (1-42)-induced impairments. The responses behaved in a dose-dependent manner.

Discussion
This study undertook in vitro screening of 18 medicinal plant extracts traditionally used in the treatment of neurological conditions within ethnomedicinal systems. The to reduced glutathione (GSH), which is a measure of cellular oxidative stress. It was determined by employing OxiSelect ™ Total Glutathione assay in the presence of nicotinamide adenine dinucleotide phosphate (NADPH) at 405 nm. The total glutathione content in unknown samples is determined by comparison with the predetermined glutathione standard curve. In each type of experiment, the nematodes were exposed to Aβ (1-42) alone (100 µM), plant extracts (n = 5; 100 µg/mL), or plant extracts (n = 5; 100 µg/mL) in combination with 100 µM Aβ (1-42). Control represents untreated cells (vehicle control); heat-killed G. pallida (not shown) were also performed as controls. Approximately 50 G. pallida J2 organisms per replicate were used in survival assays,~100 organisms were used in mitochondrial reductase and ROS assays, and glutathione studies involved~1000 organisms per replicate. Data are mean ± SEM (n = 3), * p < 0.05, ** p < 0.01, *** p < 0.001 (significance), and ns (not significant) for the statistical comparisons indicated (one-way analysis of variance).

Discussion
This study undertook in vitro screening of 18 medicinal plant extracts traditionally used in the treatment of neurological conditions within ethnomedicinal systems. The phytoextracts were screened for their ability to enhance the antioxidant activity and survival rate and inhibit three enzymatic targets (i.e., AChE, BuChE, and BACE-1) in G. pallida (pretreated or not with Aβ (1-42) peptide) related to the progression and development of AD. To the best of our knowledge, this is the first time that this nematode model has been employed in drug discovery studies [11,12]. Several plant candidates with promising activities were identified by in vitro testing, necessitating further in vivo investigations using G. pallida. Indeed, five potent extracts acted strongly against the cholinergic enzymes AChE and BuChE; four potent extracts acted against BACE-1; and seven plants elicited extraordinarily strong antioxidant activities. Extracts obtained from J. regia, E. cardamomum, C. zeylanicum, S. officinalis, and H. perforatum exerted the most potent and multi-faceted activities to protect against Aβinduced impairment in G. pallida (Table S4). In addition to their greatest AchE and BuChE inhibitory activities, the effects of J. regia and E. cardamomum extracts in restoring the organism's chemosensing and improving survival were extremely encouraging.
As a species, J. regia L., commonly known as walnut, as previously been reported to possess neuroprotective effects in in vitro [17,18] and in vivo studies [19][20][21][22][23]. However, the major distinction between the present study and such studies is that these previous studies almost exclusively focused on the edible fruit of this plant (walnut), and not the leaves. Indeed, the studies arose from epidemiological evidence associating diets rich in walnuts with lower incidence of neurodegenerative disease. However, leaves of J. regia have been used in traditional medicine for thousands of years in the treatment of many conditions. Flower and leaf aqueous 'tea infusions' of J. regia were reported to inhibit AChE by up to 45% [18], an action that may be attributed to gallic acid, ellagic acid, and other related compounds [24,25]. The present study, using a ethanol:water extraction, shows that much greater levels of activity can be achieved, reaching up to 73% inhibition against AChE and 90% inhibition against BuChE. Such levels of activity can be compared very favourably with the purified galantamine. Other studies investigating AChE inhibitory activity of J. regia [17,26] have had mixed levels of success, indicating that the measurement methods and the plant part extracts are also important. There has been a pre-ponderance of studies investigating the anti-amyloidogenic activity of 'walnut extracts'; however, these were compositionally very different from leaf extracts [27,28]. Herein, we show for the first time that J. regia leaves possess compounds not only with BACE-1 inhibitory activity but also protective action against Aβ-induced damage (as demonstrated by several different assessments). Along with the other four extracts, G. pallida's chemosensing was enhanced by J. regia in the absence of Aβ, but comparatively to other selected phytoextracts, J. regia extracts achieved this activity at much lower concentrations. Furthermore, in the presence of Aβ, leaves extract of J. regia (i) significantly improved the nematode's chemosensing at all concentrations tested, restoring up to 70% of lost chemosensory capability; (ii) improved the survival rate of the nematode to the greatest extent, which was restored from a level of 74% to 92%, almost reaching the levels of control incubations; (iii) led to a strong reduction in ROS production, which, again, was almost completely normalised; and (iv) significantly augmented oxidative stress defence responses. The detailed mechanisms underlying these protective effects remain to be unpicked. The main source of ROS biologically is NADPH oxidase, and oligomeric Aβ  induces ROS production by activating this enzyme [29]. Therefore, we can hypothesize that NADPH oxidase activity was impacted by compounds from J. regia and other shortlisted plants that were shown to possess a strong antioxidant activity, which is known to counteract Aβ toxicity [30,31]. What is clear is that both the glutathione and mitochondrial reductase systems were augmented when G. pallida was exposed to leaf extract of J. regia and improved the principal defensive capabilities of the organism against Aβ. Further investigations could examine whether heat shock proteins (HSPs), transcription factors, and/or other antioxidant enzymes in G. pallida are altered, since these molecules have been investigated in other model systems [30,32].
The second very promising ethanolic extract was E. cardamomum (L.), commonly known as cardamom, which inhibited AChE and BuChE to a similar extent to galantamine, and to a greater magnitude compared to other reports [33]. E. cardamomum is cultivated commercially in India and Sri Lanka. Its traditional use is in the treatment of gastrointestinal disorders, but it is also used as an anti-depressant which, over a prolonged period, can enhance cognitive function [33,34]. In vivo studies suggest that alcoholic extracts of E. cardamomum have anxiolytic, anti-depressant, and cognition-enhancing properties, possibly due to its flavonoid content [34,35]. The anxiolytic effects of E. cardamomum have been particularly attributed to the high flavonoid content and quercetin levels [34]. In many respects, the protective actions of E. cardamomum against Aβ were like that of J. regia, with a remarkable ability to restoring the chemosensing of G. pallida, and so this plant extract should also be prioritised for further investigations.
Although not as potent in restoring chemosensing abilities, some of other plant extracts tested had interesting properties. We observed that the ethanolic flowers extracts of H. perforatum (L.), known as St. John's Wort, and S. officinalis (L.), also called sage, had greater propensity to inhibit BuChE than AChE, albeit with a degree of selectivity for BuChE, suggesting that there might have compounds that should be purified from these two extracts. This appears to be a new observation, but previous studies using methanolic extracts of H. perforatum observed similar maximal levels of AChE inhibition indicating a good data reproducibility [36][37][38]. The inhibitory activity of S. officinalis against AChE has been previously reported, and other studies have also reported the ability of various sage species in enhancing cognitive function [39][40][41][42]. Traditionally, S. officinalis has a very longstanding reputation for improving memory and cognitive function. Many bioactive phytochemicals have been isolated from S. officinalis, mainly phenolic diterpenes, which possess diverse biological activities [39]. It is thought that the ability of S. officinalis extracts to improve memory and cognitive functions relates to its inhibition of AChE [40,43,44], and possibly its neuroprotective effect against Aβ-induced toxicity [45]. Isorosmanol and 7-methoxyrosmanol isolated from S. officinalis have been previously reported to inhibit AChE by 50-65% at concentrations of 500 µM [39,46]. Cytoprotective effects of sage against Aβ in neuronal cells have been reported [45], and oral administration of its essential oil to patients improves cognition and memory with no adverse effects [47,48]. Interestingly, we also found that, from all the extracts tested herein, S. officinalis achieved the highest BACE-1 inhibitory activity (up to 76%). In vitro and in vivo studies have reported that salvia species contain a large array of active compounds that may enhance cognitive activity, protect against neurodegenerative diseases, and may be beneficial in preventing the onset of age-related dementia [40,49]. Compounds present in salvia plants (which may offer either cellular protection against Aβ or inhibition of BACE-1) include rosmarinic acid, salvianolic acid, carnosic acid, quercetin, and tanshinone [45,[50][51][52][53][54][55]. It is also interesting that S. officinalis protects mice from Aβ-induced neurotoxicity by inhibiting increases of tumour necrosis factor-α (TNF-α) levels, because this suggests that other mechanisms (such as re-balancing cytokines and neurotrophins levels) are important in preventing learning and memory deficits [56].
The ethanolic extract with most potent antioxidant activity was from the bark of C. zeylanicum (Linn.), known as cinnamon, which has been used as a spice and traditional medicine for many centuries [57]. Cinnamon significantly inhibits tau protein aggregation, promotes the disassembly of recombinant tau filaments, and substantially alters the morphology of paired-helical filaments isolated from AD brain [58]. Furthermore, oral administration of cinnamon reduces Aβ oligomerization and improves the cognitive behaviour of mouse models of AD pathology [58]. C. zeylanicum can improve the activity of antioxidants enzymes in vivo, lowering malondialdehyde (MDA) levels in humans [59][60][61][62]. Evidence suggests that the bioactivity of cinnamon is largely ascribed to phenolic compounds, including cinnamic acid, cinnamate, cinnamaldehyde, and proanthocyanidins [63,64].
In short, this study identified several promising plant extracts with pleiotropic functions representing a huge therapeutic potential in AD. Moreover, the recently described G. pallida model system was employed for the first time in drug screening, demonstrating its utility in drug discovery research. The strengths of this investigation include the (i) selection of plants based on ethnomedicinal use, (ii) stepwise use of in vitro and in vivo approaches, (iii) detailed concentration-dependent testing, and (iv) probing of enzyme inhibitory mechanisms. One limitation was that the broad screening approach made it impractical to optimize, standardize, and determine the reproducibility of the extraction procedure. Nonetheless, this study has brought forward a range of new information; for instance, BACE-1 inhibition has not previously been assessed in many of the plants herein reported, except for C. zeylanicum and C. longa [65,66].
Future studies should characterize the biochemical composition of J. regia and E. cardamomum extracts using techniques such as HPLC and perform pre-clinical and clinical testing to evaluate them as alternative or adjunct therapies. Inexpensive and effective treatments based on medicinal plants remain a viable way of reducing the global burden of AD in developing countries. Furthermore, it will be necessary to (i) extract, fractionate, and isolate candidate compounds before subjecting them to repeated rounds of screening in G. pallida and (ii) discover novel compounds/chemical scaffolds with protective effects against Aβ-induced toxicity.