Antioxidant Activity of a Novel Acetylcholinesterase Inhibitor: In Vivo and Ex vivo Studies

The acetylcholinesterase (AChE) inhibitors are the main drugs for symptomatic treatment of neurodegenerative disorders like Alzheimer’s disease. A recently designed, synthesized and tested hybrid compound between the AChE inhibitor galantamine (GAL) and the antioxidant polyphenol curcumin (CU) showed high AChE inhibition in vitro. Here, we describe tests for acute and short-term toxicity in mice as well as antioxidant tests on brain homogenates measured the levels of malondialdehide (MDA) and glutathione (GSH). Haematological and serum biochemical analyses were also performed. In the acute toxicity tests, the novel AChE inhibitor given orally in mice showed LD50 of 49 mg/kg. The short-term administration of 2.5 and 5 mg/kg did not show toxicity. In the ex vivo tests, the GAL-CU hybrid performed better than GAL and CU themselves. In a dose of 5 mg/kg, it demonstrates 25% reduction in AChE activity, 28% and 73% increase in the levels of MDA and GSH, respectively. No significant changes in blood biochemical data were observed. The GAL-CU hybrid is a novel non-toxic AChE inhibitor with high antioxidant activity which makes it a perspective multitarget drug candidate for treatment of Alzheimer’s disease.


Introduction
The neurodegenerative disorders like Alzheimer's and Parkinson diseases rise exponentially every ten years and during the next few decades will overtake cancer and become the second leading cause of death after cardiovascular diseases [1]. The successful treatment of these disorders requires reliable tests and biomarkers for early diagnosis, identification of translational drug targets and fast discovery and development of multitarget drugs and therapeutic strategies to tackle the complex pathogenesis of the neurodegeneration.
In the last 20 years, the main drugs used to cope with some of the symptoms of Alzheimer's disease (AD) are inhibitors of the enzyme acetylcholinesterase (AChE). The reversible AChE blockage increases the levels of acetylcholine (ACh) -the main neurotransmitter associated with the cognitive and motor brain functions. Galantamine (GAL) (Figure 1) is among the few AChE inhibitors approved for treatment of AD [2]. Additionally, GAL is an allosteric modulator of nicotinic and muscarinic ACh receptors [3,4], increases the phagocytosis of amyloid β (Aβ) peptide in rat microglia [5], exhibits a moderate scavenging effect in antioxidant studies [6].
The oxidative stress hypothesis in AD dates back to the late 1990s [7,8]. According to this hypothesis, the mammals' brain is very sensitive to oxidative stress because of the abundant presence of polyunsaturated fatty acids and transition metals like iron, copper and zinc [9] and the relative shortage of antioxidant ability to detoxify the free radicals [10]. Most of the clinical trials report associations between antioxidant use and better cognitive functions [11][12][13][14][15]. Curcumin (CU) (Figure 1) is a natural polyphenol with a powerful antioxidant activity [16] and ability to reduce oxidative stress and amyloid pathology in transgenic mice [17]. Additionally, it was found that CU binds to Aβ oligomers and fibrils and inhibits the β-sheet formation [18]. We simulated the primary nucleation of Aβ peptide by molecular dynamics and showed that CU molecules inhibit the process by intercalating among the Aβ chains [19]. Even more, CU is able to disintegrate preformed Aβ fibrils [18], reduce insoluble Aβ deposits [20] and amyloid plaques [21].
Recently, we designed a combinatorial library of GAL-CU hybrids, screened for optimal ADME properties and blood-brain permeability and docked on AChE [19]. The 14 best performing hybrids were synthesized and tested for neurotoxicity and AChE inhibition in vitro. Five of them showed less toxicity than GAL and CU and AChE inhibition between 41 and 186 times higher than GAL. Here, we describe the in vivo evaluation of acute and short-term toxicity and the ex vivo antioxidant properties of the best performing inhibitor -compound 4b. As the present study is a continuation of our previous research [19] and will be followed by other studies on this particular compound, we prefer to keep the same compound ID. As positive controls in the study are used GAL and CU.
The synthesis of 4b and detailed analytical data is already reported in the context of the synthesis of a series of GAL-CU hybrids [19]. A contribution of the herein reported synthetic procedures is upscale and optimization of the protocols.

Animals
Male and female ICR mice (6 weeks old, 25-35 g) obtained from the National Breeding Center, Sofia, Bulgaria were used in the experiments. As a more sensitive sex [22], 18 females were used in the acute toxicity test and 30 males in the short-term toxicity test. Mice were housed in Plexiglas cages (6 per cage) in a 12/12 light/dark cycle, under standard laboratory conditions (ambient temperature 20°C ± 2°C and humidity 72% ± 4%) with free access to water and standard pelleted food № 53-3, produced in accordance with ISO9001:2008. Prior to the start of the experiment, the mice were acclimatized under vivarium conditions for seven days and their health was monitored daily by a veterinarian. The vivarium (registration certificate № 15320139/01.08.2007) was inspected by the Executive Agency for Medicines in order to verify the conditions for keeping laboratory animals (№ A-16-0532/14.10.2016). All experiments strictly followed the principles set out in the European Convention for the Protection of Vertebrate Animals Used for Experimental and Other Scientific Purposes (ETS 123) (Council of Europe, 1991). Efforts have been made to minimize animal suffering. The Animal Care Ethic Committee approved the study protocol and Ethic clearance (№ 273 from 02/06/2020) was issued for the study by the Bulgarian Agency for Food Safety.

Acute Toxicity Protocol
Acute toxicity test was performed on female mice after oral (po) administration of 4b using a simplified Lorke's method [23]. Three animals per dose were used, starting with a dose of 10 mg/kg and gradually increasing it every 3 h until the first lethal case. Compound 4b was solubilized with Tween 80 (0.1%) in distilled water and administered orally in a volume of 0.1 ml/ 10 g. The LD50 is calculated using the following equation: where D0 is the highest dose that does not cause mortality and D100 is the lowest dose that results in death.
Surviving animals were observed for 24 hours and then for up to 14 days. On day 14, they were euthanized after anesthesia with ketamine/xylazine (80/10 mg/ kg, i.p.) and an examination of the internal organs for possible macroscopic abnormalities (organ color, consistency, neoplasms, etc.) was made.

Short-Term Toxicity Protocol
The short-term toxicity test was performed on male ICR mice at the same age of 6 weeks and approximately 30-35 g. Three drugs (GAL, CU and 4b) were administered daily for 14 days orally with a dosing needle at approximately the same time of day, 11.00 h. Based on the LD50 value ≈ 50 mg/ kg, derived in the acute toxicity test, two doses of 2.5 mg/kg and 5 mg/kg (1/20 and 1/10 of the LD50) were selected for multiple administration of 4b. The animals were divided into 5 experimental groups of 6 mice (n = 6). Group 1 was the control group treated with physiological saline via oral gavage (0.1 ml/10 g bw); group 2 was treated orally with the positive control GAL in dose of 3 mg/kg dissolved in distilled water [24]; group 3 was treated orally with the second positive control CU at a dose of 25 mg/kg [25]; group 4 was treated orally with 4b at a dose of 2.5 mg/kg (1/20 LD50); group 5 was treated orally with 4b at a dose of 5 mg/kg (1/10 LD50).
Animals were observed daily for behavioral changes and signs of toxicity. On day 15, after 14-day treatment, the animals were anesthetized with ketamine/xylazine (80 mg/10 kg ip) and decapitated. Blood was taken in vacutainers for complete blood count and biochemistry measurment. The brains were dissected, measured and prepared for the assessment of AChE inhibition, MDA and GSH levels. Protein content of brain homogenates was measured by the method of Lowry [26] using bovine serum albumin as a standard.

Measurement of Acetylcholinesterase (AChE) Inhibition in Brain Homogenate
Brains were homogenized with 0.1 M, phosphate buffer, pH 7.4. Aliquots of brain homogenates from different groups were used to measure AChE activity for 10 min by the method of Ellman [27]. AChE activity was calculated and expressed as nmol/min/mg protein using a molar absorption coefficient of 13,600 M -1 cm -1 .

Measurement of Malondialdehyde (MDA) Levels in Brain Homogenate
Brains were homogenized with 0.1 M phosphate buffer and EDTA, pH 7.4. Aliquots of the homogenates were heated for 20 minutes on a water bath (100°C) with thiobarbituric acid. The amount of thiobarbituric acid-formed reactive species (TBARS) (expressed as MDA equivalents) was measured spectrophotometrically by the method of Deby and Goutier [28] at a wavelength of 535 nm. The concentration of MDA was calculated using a molar absorption coefficient of 1,56 × 10 5 M -1 cm -1 and expressed in nmol/g tissue.

Measurement of Glutathione (GSH) Levels in Brain Homogenate
GSH was evaluated by measuring non-protein sulfhydryls after trichloroacetic acid (TCA) protein precipitation by the method described by Bump et al. [29]. Brains were homogenized in 5% TCA (1:10) and centrifuged for 20 min at 4 000 × g. The reaction mixture contained 0.05 mL supernatant, 3 mL 0.05M phosphate buffer (pH = 8), and 0.02mL DTNB reagent. Absorption was determined at a wavelength of 412 nm and the results were expressed as nmol/g tissue.

Statistical Analysis
The MEDCALC statistical program was used to analyze the in vivo data. Results are expressed as mean ± SD for six animals in each group. The significance of the data was assessed by a nonparametric Mann-Whitney U test. Values of p ≤ 0.05 are considered as statistically significant.

Acute Toxicity in Mice
The median lethal dose (LD50) of 4b in mice was 49 mg/kg. No serious toxic effects or mortality were observed in mice at doses of 10 and 20 mg/kg. At dose of 40 mg/kg, tachypnea and mild tremor for up to 2 hours were observed with no other visible signs of toxicity. At dose of 60 mg/kg, one fatal outcome was observed 3 h post-dosing with accelerated breathing, piloerction, mild tremor and seizures. Forthteen days later, the animals were euthanized and a macroscopic examination of the internal organs was performed. No changes in the size, color and consistency of the lungs, liver, heart, kidneys, stomach, spleen, and intestines were observed. No abnormalities in the morphology of the gonads and brain were detected.

Short-term Toxicity in Mice
The short-term toxicity of 4b was assessed by daily administration of 2.5 mg/kg (1/20 of LD50) and 5 mg/kg (1/10 of LD50) for 14 days, according to the protocol described in Materials and Methods. As positive contols were used GAL in dose of 3 mg/kg (1/10 of LD50) and CU in dose of 25 mg/kg (1/10 of LD50). A control (placebo) group also was included in the test. Animals were examined daily and no behavioral changes and signs of toxicity were observed. On day 15, the animals were anesthetized and decapitated. Blood from each group was collected for haematological and serum biochemical analuses. Brains from each group were collected as described in Material and Methods and prepared for measurements of AChE inhibition, MDA and GSH levels.

AChE Inhibition in Mice Brain Homogenate
After 14-day treatment of mice by GAL, CU, 4b or placedo, the AChE activity was measured for 10 min by the Ellman's method as described in Materials and Methods. The results are given in Figure 2. CU caused a mild decrease in AChE activity, followed by GAL and 4b. The AChE decrease of 4b is dose-dependent. The dose of 5 mg/kg causes 25% inhibition of enzyme activity compared to the control group. Figure 2. AChE activity measured for 10 min in brain homogenates derived from mice treated 14 days by GAL, CU, 4b in two doses (2.5 mg/kg and 5 mg/kg) or placebo (control group). p < 0.05 vs. control.

Antioxidant Activity in Mice Brain Homogenate
The antioxidant activity of the tested compounds after 14-day treatment, was assessed by measuring the levels of malondialdehyde (MDA) and gluthathione (GSH) in mice brain homogenates as described in Materials and Methods. MDA is a specific marker for lipid peroxidation. The thiobarbituric acid (TBA) test is widely used to measure the formation of red pigment, which is an adduct of TBA with aldehydes, MDA, ketones and others. The effects of the tested compounds on MDA levels are given in Figure 3a. All compounds caused increase of MDA levels: CU -by 9%, GAL -by 26%, 4b in dose 2.5 mg/kg -by 24% and 4b in dose 5 mg/kg -by 28% compared to control. The effect of 4b on MDA levels is dose-dependent.
The oxidative stress causes decrease in GSH levels [9]. All tested compounds in the present study elevated the GSH levels ( Figure 3b) showing antioxidant activity. The most prominent effect has 4b in dose of 5 mg/kg (73%), followed by 4b in dose of 2.5 mg/kg (67%), CU (52%) and GAL (35%). Here again, the effects of 4b on GSH levels is dose-dependent.  . Antioxidant activity measured by MDA (a) and GSH levels (b) in brain homogenates derived from mice treated 14 days by GAL, CU, 4b in two doses (2.5 mg/kg and 5 mg/kg) or placebo (control group). p < 0.05 vs. control.

Discussion
In the present study, we performed in vivo and ex vivo studies of a newly synthesized hybrid between GAL and CU aiming to combine the AChE inhibitory properties of GAL with the powerful antioxidant properties of CU. Both properties have beneficial effect on the delay of neurodegeneration. As a result, an AChE inhibitor was generated, 186 times more potent than GAL in vitro (IC50 of compound 4b = 20 nM vs. IC50 of GAL = 3.52 μM) [19].
The LD50 of 4b registered in the acute toxicity test on mice is 49 mg/kg orally. The LD50 of GAL in mice is 10 mg/kg i.p. [32] and between 15 and 45 mg/kg (median 30 mg/kg) orally [33]. The oral LD50 of CU is more than 2000 mg/kg (at highest given dose 2000 mg/kg) [34,35]. Obviously, the presence of CU fragment in the molecule of 4b reduces its overall toxicity. The symptoms of toxic damage that were observed, expressed in breathing difficulty, initially rapid and then severely delayed, ataxia, lack of coordinated movements, tremor and tonic-clonic seizures, are characteristic of acetylcholine (ACh) neurotoxicity due to inhibition of AChE. In mammals, respiratory failure caused by inhibition of cerebral AChE is recognized as a cause of death [36]. In fish, the relationship between AChE inhibition and mortality is unclear, as some species are able to survive a high rate (90-95%) of brain enzyme inhibition [37][38][39]. In this regard, some authors have found that 50% of AChE inhibition may indicate intoxication or poisoning [40]. No changes in the morphology of the internal organs (lungs, liver, heart, kidneys, stomach, spleen, intestines, gonads and brain) were observed.
The daily administration of 4b in doses of 2.5 mg/kg (1/20 of LD50) and 5 mg/kg (1/10 of LD50) for 14 days did not show any sign of toxicity in animals' behavior. The ex vivo AChE activity measured after this period showed that 4b inhibits the enzyme in a dose-dependent manner: 21% inhibition by the dose of 2.5 mg/kg and 25% -by the dose of 5 mg/kg compared to the control group ( Figure 2). Such extent of AChE inhibition is not enough to cause toxic effects or mortality, but it leads to moderate increase in the levels of brain ACh. Any elevation of ACh levels in the brain are associated with improved cognitive effects [41], which was the purpose of the design of GAL-CU hybrids. The protective effect of GAL on mild to moderate Alzheimer's disease is due to its inhibitory effect on AChE, which was confirmed in the present study by the 16% reduction in AChE activity in mice brain homogenates. The neuroprotective and AChE inhibitory effects of curcumin and other curcuminoids is well known [42][43][44]. Our study also confirmed these effects: CU showed 12% reduction in the AChE activity in the tested brain homogenates. Being a congener of GAL and CU, 4b accumulates and fortifies these effects and exerts more powerful AChE inhibition than both leads.
Tested in vitro 4b showed 186 times higher inhibitory effect on AChE than GAL [19]. In the present ex vivo studies, the difference in the reductions of AChE activity by 4b and GAL is only 5-9% in healthy animals. Some authors report reactivation of AChE, or even overproduction of an enzyme in response to inhibition [45]. The mild difference in the ex vivo AChE inhibition between 4b and GAL might be indicative for such recovery of AChE levels due to high spontaneous reactivation of the enzyme associated with rapid synthesis and release of a new enzyme from the liver. This possible mechanism requires further investigation.
The AChE inhibition in vivo is associated with increased oxidative stress [46,47]. In response, the increased oxidative stress mediates an increase in the AChE activity. The central nervous system is very sensitive to oxidative stress. The brain has many areas of high iron content, an activator of oxidative reactions, and neuronal mitochondria generate large amounts of hydrogen peroxide [9]. Neuronal membranes are rich in polyunsaturated fatty acids, which are also particularly susceptible to oxidative stress [9]. Due to the oxidative stress, the cholinergic neurons undergo a degeneration leading to impairments in cognition and memory. One of the strategies for prevention and treatment of Alzheimer's disease is focused on the suppression of oxidative damage [7,8].
The association between AChE inhibition and increased oxidative stress also was observed in our MDA test. MDA is a biomarker of the oxidative stress relating to the lipid peroxidation. In the present study, we found that all tested AChE inhibitors elevated the MDA levels ( Figure 3a). As stronger is the AChE inhibition, as bigger is the increase of MDA levels. In the same time, it was found in our GSH test that all tested compounds increased the GSH levels ( Figure 3b). GSH is an important endogenous cellular protector and antioxidant. The increased GSH levels might be interpreted as a compensatory response of the brain to overcome the lipid peroxidation induced by the decreased AChE activity. Kaur и Sandhir [48] conducted acute and chronic toxicity studies in Wistar rats with two different oral doses of insecticide (AChE inhibitor). In this study, rats showed a decrease in GSH levels after acute pesticide exposure and an increase in GSH levels during chronic repeated administration. The authors suggest that GSH elevation is an adaptive response to increased oxidative stress.
GAL has an antioxidant activity related with the enolic OH group [6]. One of the mechanisms by which GAL improves the mental function in Alzheimer's dementia is thought to be its antioxidant effect [49][50][51]. CU also is a proven antioxidant and brain protector [52][53][54]. These effects were also confirmed in the present study, where CU and GAL significantly increased GSH levels by 52% and 35%, respectively, compared to the control group. As a hybrid of GAL and CU, 4b accrues and augments the antioxidant effects of its leads.
The analyses of the haematological and serum biochemical data point to some concern about the elevated values of ASAT and ALAT by 4b. ASAT levels are slightly higher than the upper limit but ALAT levels rise by 20% above the upper limit of the mice reference range. CU and 4b decrease the serum levels of uric acid. CU inhibits xanthine oxidase and increases uricosuric activity [55,56]. As a CU congener, 4b might be also involved in the same mechanism.