Wasp Venom Ameliorates Scopolamine-Induced Learning and Memory Impairment in Mice

This study investigated the effects of wasp venom (WV) from the yellow-legged hornet, Vespa velutina, on scopolamine (SCO)-induced memory deficits in mice, as well as the antioxidant activity in HT22 murine hippocampal neuronal cells in parallel comparison with bee venom (BV). The WV was collected from the venom sac, freeze-dried. Both venoms exhibited free radical scavenging capabilities in a concentration-dependent manner. In addition, the venom treatment enhanced cell viability at the concentrations of ≤40 µg/mL of WV and ≤4 µg/mL of BV in glutamate-treated HT22 cells, and increased the transcriptional activity of the antioxidant response element (ARE), a cis-acting enhancer which regulates the expression of nuclear factor erythroid 2-related factor 2 (Nrf2)-downstream antioxidant enzymes. Concurrently, WV at 20 µg/mL significantly increased the expression of a key antioxidant enzyme heme oxygenase 1 (HO-1) in HT22 cells despite no significant changes observed in the nuclear level of Nrf2. Furthermore, the intraperitoneal administration of WV to SCO-treated mice at doses ranged from 250 to 500 µg/kg body weight ameliorated memory impairment behavior, reduced histological injury in the hippocampal region, and reduced oxidative stress biomarkers in the brain and blood of SCO-treated mice. Our findings demonstrate that WV possess the potential to improve learning and memory deficit in vivo while further study is needed for the proper dose and safety measures and clinical effectiveness.


WV Protected against Glutamate-Induced Cytotoxicity
Cytotoxicity assay showed that IC50 values of WV and BV were > 120 µg/mL and 6-12 µg/mL, respectively, indicating that WV was significantly less toxic than BV in HT22 mouse hippocampal neuronal cells (Figure 2A). When treated with glutamate, causing cytotoxicity in HT22 cells by raising intracellular oxidative stress [16][17][18][19], WV at the concentrations of ≤ 40 µg/mL and BV at ≤ 4 µg/mL inhibited glutamate-induced cellular toxicity ( Figure 2B). However, WV at 80 µg/mL or higher was not effective in alleviating glutamate-induced cytotoxicity, and rather found to be toxic.  Values not sharing a common alphabetic character represent a statistically significant difference among experimental groups (p < 0.05). Vit C, ascorbic acid (as a positive control); BV, bee venom; WV, wasp venom.

WV Protected against Glutamate-Induced Cytotoxicity
Cytotoxicity assay showed that IC50 values of WV and BV were > 120 µg/mL and 6-12 µg/mL, respectively, indicating that WV was significantly less toxic than BV in HT22 mouse hippocampal neuronal cells (Figure 2A). When treated with glutamate, causing cytotoxicity in HT22 cells by raising intracellular oxidative stress [16][17][18][19], WV at the concentrations of ≤ 40 µg/mL and BV at ≤ 4 µg/mL inhibited glutamate-induced cellular toxicity ( Figure 2B). However, WV at 80 µg/mL or higher was not effective in alleviating glutamate-induced cytotoxicity, and rather found to be toxic. . Statistical analysis was performed by one-way ANOVA, followed by Duncan's multiple range test. Values not sharing a common alphabetic character represent a statistically significant difference among experimental groups (p < 0.05). BV, bee venom; WV, wasp venom.

WV Increased Antioxidant Response Element (ARE)-Luciferase Activity
The transcriptional activity of ARE, a cis-acting enhancer which regulates the expression of Nrf2-downstream antioxidant enzymes, was examined using HT22-ARE cells harboring the luciferase reporter-encoding gene following the ARE sequence in the transduced vector. WV and BV at the concentrations nontoxic to HT22 cells were found to increase luciferase reporter activity in a concentration-dependent manner (Figure 3), indicating its potential to promote nuclear translocation of Nrf2 and induce the transcription Murine hippocampal neuronal cell line, HT22, was treated with various concentrations of BV (0.375-12 µg/mL) and WV (3.75-120 µg/mL) in the absence (A) and presence of 5 mM glutamate (B) for 24 h. The cytotoxicity was assayed using the CCK-8. Values and error bars are presented as mean ± SEM (n = 3). Statistical analysis was performed by one-way ANOVA, followed by Duncan's multiple range test. Values not sharing a common alphabetic character represent a statistically significant difference among experimental groups (p < 0.05). BV, bee venom; WV, wasp venom.

WV Increased Antioxidant Response Element (ARE)-Luciferase Activity
The transcriptional activity of ARE, a cis-acting enhancer which regulates the expression of Nrf2-downstream antioxidant enzymes, was examined using HT22-ARE cells harboring the luciferase reporter-encoding gene following the ARE sequence in the transduced vector. WV and BV at the concentrations nontoxic to HT22 cells were found to increase luciferase reporter activity in a concentration-dependent manner (Figure 3), indicating its potential to promote nuclear translocation of Nrf2 and induce the transcription of a set of antioxidant enzyme genes. Figure 2. Cytotoxicity of BV and WV in HT22 cells. Murine hippocampal neuronal cell line, HT22, was treated with various concentrations of BV (0.375-12 µg/mL) and WV (3.75-120 µg/mL) in the absence (A) and presence of 5 mM glutamate (B) for 24 h. The cytotoxicity was assayed using the CCK-8. Values and error bars are presented as mean ± SEM (n = 3). Statistical analysis was performed by one-way ANOVA, followed by Duncan's multiple range test. Values not sharing a common alphabetic character represent a statistically significant difference among experimental groups (p < 0.05). BV, bee venom; WV, wasp venom.

WV Increased Antioxidant Response Element (ARE)-Luciferase Activity
The transcriptional activity of ARE, a cis-acting enhancer which regulates the expression of Nrf2-downstream antioxidant enzymes, was examined using HT22-ARE cells harboring the luciferase reporter-encoding gene following the ARE sequence in the transduced vector. WV and BV at the concentrations nontoxic to HT22 cells were found to increase luciferase reporter activity in a concentration-dependent manner (Figure 3), indicating its potential to promote nuclear translocation of Nrf2 and induce the transcription of a set of antioxidant enzyme genes.   ), an ARE activator, was used as a positive control. Results are expressed as mean ± SEM (n = 3). Statistical analysis was performed by one-way ANOVA, followed by Duncan's multiple range test. Values not sharing a common alphabetic character represent a statistically significant difference among experimental groups (p < 0.05). SFN, sulforaphane; BV, bee venom; WV, wasp venom.

WV Upregulated Cytoplasmic HO-1 Level Downstream of Nrf2
Consistently, WV treatment at 20 µg/mL resulted in a significant induction of HO-1, a type of inducible antioxidant enzyme, although the nuclear level of Nrf2 was marginally affected ( Figure 4).

WV Upregulated Cytoplasmic HO-1 Level Downstream of Nrf2
Consistently, WV treatment at 20 µg/mL resulted in a significant induction of HO-1, a type of inducible antioxidant enzyme, although the nuclear level of Nrf2 was marginally affected ( Figure 4).

WV Decreased Cellular ROS Level
The DCF assay showed that treatment of HT22 cells with glutamate, a cytotoxic agent, increased intracellular ROS level as expected ( Figure 5). However, the WV or BV treatment significantly lowered the ROS level in glutamate-treated cells in a concentration-dependent manner.

WV Decreased Cellular ROS Level
The DCF assay showed that treatment of HT22 cells with glutamate, a cytotoxic agent, increased intracellular ROS level as expected ( Figure 5). However, the WV or BV treatment significantly lowered the ROS level in glutamate-treated cells in a concentration-dependent manner. The fluorescence intensity was mea using a fluorescence microplate reader and relatively quantified to the control. The results a pressed as means ± SEM (n = 3). Statistical analysis was performed by one-way ANOVA, fol by Duncan's multiple range test. Values not sharing a common alphabetic character indicate tistically significant difference among experimental groups (p < 0.05). BV, bee venom; WV venom.

WV Improved Learning and Memory in SCO-Treated Mouse Model
A total of 56 C57BL/6J mice were randomly allocated into 8 different groups ( 1). WV or BV were intraperitoneally administered in combination with SCO trea every day for 10 days ( Figure 6).  The fluorescence intensity was measured using a fluorescence microplate reader and relatively quantified to the control. The results are expressed as means ± SEM (n = 3). Statistical analysis was performed by one-way ANOVA, followed by Duncan's multiple range test. Values not sharing a common alphabetic character indicate a statistically significant difference among experimental groups (p < 0.05). BV, bee venom; WV, wasp venom.

WV Improved Learning and Memory in SCO-Treated Mouse Model
A total of 56 C57BL/6J mice were randomly allocated into 8 different groups (Table 1). WV or BV were intraperitoneally administered in combination with SCO treatment every day for 10 days ( Figure 6).
As a result of regular monitoring, administration of BV or WV and treatment with SCO did not have a significant effect on the BW of mice during the experimental period ( Figure 7).
The Morris water maze task demonstrated that the treatment with WV (500 µg/kg bw) or donepezil (5 mg/kg BW) improved SCO-induced spatial learning and memory impairment in both tested concentrations ( Figure 8A). The passive avoidance task showed that WV (50 µg/kg BW) or BV (5 µg/kg BW) ameliorated the SCO-induced associative learning and memory deficit ( Figure 8B). Moreover, administration of WV at 250 µg/kg BW or higher displayed similar memory improvement to donepezil, a positive control. The WV administration also improved short-term spatial memory as tested the willingness of mice to explore new environments in the Y-maze, which was diminished by the SCO treatment ( Figure 8C). The parts of the brain involved in such learning and memory include the hippocampus, basal forebrain, and prefrontal cortex [20].  Figure 6. Experimental schedule for behavioral study using C57BL/6J mice.
As a result of regular monitoring, administration of BV or WV and treatmen SCO did not have a significant effect on the BW of mice during the experimental ( Figure 7).

Figure 7. Effect of intraperitoneal injection of WV on mouse body weight.
After 1-week tion, mice were intraperitoneally injected with SCO at a dose of 1 mg/kg BW every single d total of 10 days and were intraperitoneally treated with BV (5 or 50 µg/kg BW) or WV (50 500 µg/kg BW) 15 h prior to SCO treatment. The average BW was monitored during the en perimental period. Results are expressed as means ± SD (n = 7). Statistical analysis was per by one-way ANOVA, followed by Duncan's multiple range test. SCO, scopolamine; donepezil; BV, bee venom; WV, wasp venom.
The Morris water maze task demonstrated that the treatment with WV (500 bw) or donepezil (5 mg/kg BW) improved SCO-induced spatial learning and memo pairment in both tested concentrations ( Figure 8A). The passive avoidance task s  . Experimental schedule for behavioral study using C57BL/6J mice.
As a result of regular monitoring, administration of BV or WV and treatment with SCO did not have a significant effect on the BW of mice during the experimental period ( Figure 7).

Figure 7. Effect of intraperitoneal injection of WV on mouse body weight.
After 1-week adaptation, mice were intraperitoneally injected with SCO at a dose of 1 mg/kg BW every single day for a total of 10 days and were intraperitoneally treated with BV (5 or 50 µg/kg BW) or WV (50, 250, or 500 µg/kg BW) 15 h prior to SCO treatment. The average BW was monitored during the entire experimental period. Results are expressed as means ± SD (n = 7). Statistical analysis was performed by one-way ANOVA, followed by Duncan's multiple range test. SCO, scopolamine; DONE, donepezil; BV, bee venom; WV, wasp venom.
The Morris water maze task demonstrated that the treatment with WV (500 µg/kg bw) or donepezil (5 mg/kg BW) improved SCO-induced spatial learning and memory impairment in both tested concentrations ( Figure 8A). The passive avoidance task showed that WV (50 µg/kg BW) or BV (5 µg/kg BW) ameliorated the SCO-induced associative learning and memory deficit ( Figure 8B). Moreover, administration of WV at 250 µg/kg BW or higher displayed similar memory improvement to donepezil, a positive control. The WV administration also improved short-term spatial memory as tested the willing- Figure 7. Effect of intraperitoneal injection of WV on mouse body weight. After 1-week adaptation, mice were intraperitoneally injected with SCO at a dose of 1 mg/kg BW every single day for a total of 10 days and were intraperitoneally treated with BV (5 or 50 µg/kg BW) or WV (50, 250, or 500 µg/kg BW) 15 h prior to SCO treatment. The average BW was monitored during the entire experimental period. Results are expressed as means ± SD (n = 7). Statistical analysis was performed by one-way ANOVA, followed by Duncan's multiple range test. SCO, scopolamine; DONE, donepezil; BV, bee venom; WV, wasp venom.

WV Protected Hippocampal Region from SCO-Induced Damage
While SCO treatment caused histological injury in the CA1 region of the hippocampal area, a treatment with WV (250 and 500 µg/kg BW) or BV (50 µg/kg BW) attenuated the SCO-induced hippocampal damage ( Figure 9). More specifically, pyramidal cell arrange-

WV Protected Hippocampal Region from SCO-Induced Damage
While SCO treatment caused histological injury in the CA1 region of the hippocampal area, a treatment with WV (250 and 500 µg/kg BW) or BV (50 µg/kg BW) attenuated the SCO-induced hippocampal damage ( Figure 9). More specifically, pyramidal cell arrangement in CA1 district was noticeably disrupted in the hippocampus of SCO-treated mice while the treatment with WV or BV significantly improved the abnormality.

WV Activated the Nrf2/HO-1 Axis
Mice treated with WV at 500 µg/kg BW showed increased nuclear translocation of Nrf2 and subsequent transcriptional activation of its downstream gene, HO-1 ( Figure 10). However, BV did not increase the nuclear level of Nrf2 and the expression of the HO-1 gene relative to the SCO control ( Figure 10).

WV Activated the Nrf2/HO-1 Axis
Mice treated with WV at 500 µg/kg BW showed increased nuclear translocation of Nrf2 and subsequent transcriptional activation of its downstream gene, HO-1 ( Figure 10). However, BV did not increase the nuclear level of Nrf2 and the expression of the HO-1 gene relative to the SCO control ( Figure 10).

WV Protected Hippocampal Region from SCO-Induced Damage
While SCO treatment caused histological injury in the CA1 region of the hippocampal area, a treatment with WV (250 and 500 µg/kg BW) or BV (50 µg/kg BW) attenuated the SCO-induced hippocampal damage (Figure 9). More specifically, pyramidal cell arrangement in CA1 district was noticeably disrupted in the hippocampus of SCO-treated mice while the treatment with WV or BV significantly improved the abnormality.

WV Activated the Nrf2/HO-1 Axis
Mice treated with WV at 500 µg/kg BW showed increased nuclear translocation of Nrf2 and subsequent transcriptional activation of its downstream gene, HO-1 ( Figure 10). However, BV did not increase the nuclear level of Nrf2 and the expression of the HO-1 gene relative to the SCO control ( Figure 10).  The expression levels of nuclear Nrf2 and cytoplasmic HO-1 in the hippocampus were measured by western blot analysis and normalized by lamin B and β-actin, respectively. Results are expressed as means ± SD (n = 5). Statistical analysis was performed by one-way ANOVA, followed by Duncan's multiple range test. Values not sharing a common alphabetic character represent a significant difference among experimental groups (p < 0.05). SCO, scopolamine; DONE, donepezil; BV, bee venom; WV, wasp venom.

WV Decreased Scopolamine-Induced Oxidative Stress Biomarkers
WV treatment significantly reduced the levels of oxidative stress markers in SCOtreated mice (Figure 11). WV administration suppressed the MDA level (lipid peroxidation marker) in cortical homogenates ( Figure 11A) and plasma 8-OHdG level (overall DNA damage marker; Figure 11B) in a dose-dependent manner. BV also lowered the levels of those biomarkers of which levels were significantly increased by SCO treatment. multiple range test. Values not sharing a common alphabetic character represent a significant difference among experimental groups (p < 0.05). SCO, scopolamine; DONE, donepezil; BV, bee venom; WV, wasp venom.

WV Decreased Scopolamine-Induced Oxidative Stress Biomarkers
WV treatment significantly reduced the levels of oxidative stress markers in SCOtreated mice (Figure 11). WV administration suppressed the MDA level (lipid peroxidation marker) in cortical homogenates ( Figure 11A) and plasma 8-OHdG level (overall DNA damage marker; Figure 11B) in a dose-dependent manner. BV also lowered the levels of those biomarkers of which levels were significantly increased by SCO treatment.

Figure 11. Effect of BV and WV on the levels of oxidative stress markers. MDA level in cortex homogenate (A) and 8-OHdG level in the plasma (B).
Results are expressed as means ± SD (n = 3 for cortical MDA level; n = 6 for plasma 8-OHdG level). Statistical analysis was performed by one-way ANOVA, followed by Duncan's multiple range test. Values not sharing a common alphabetic character represent a significant difference among experimental groups (p < 0.05). SCO, scopolamine; DONE, donepezil; BV, bee venom; WV, wasp venom.

Discussion
While BV has been widely utilized as an acupuncture agent for the treatment of rheumatoid arthritis and osteoarthritis in Korea, WV is not well-studied for its clinical usefulness [21]. Although hymenoptera venoms can elicit both local and systemic allergic reactions, including life-threatening anaphylaxis, venom immunotherapy remains the most effective treatment reducing the risk of systemic reactions in individuals with hymenoptera venom allergy [22]. In addition, phospholipase A2 which is also found in BV was reported to have the potential to inhibit the progression of AD in the 3 × Tg AD mouse model presumably through the increase in regulatory T cell population [23].
WV has been reported to possess some pharmacological effects in the treatment of pain, inflammatory and neurodegenerative diseases [7,24,25]. Our previous study showed that WV exhibited strong antiinflammatory activity, although the effective dose was higher than the dose given of BV [6]. The present study is a follow-up to our previous work, and aimed to examine if WV could attenuate SCO-induced learning and memory impairment through antioxidant enzyme induction in the mouse model.
Multiple studies have indicated that Nrf2, a master regulator of inducible antioxidant enzymes, is involved in suppressing inflammatory responses and ROS generation in the brain, and thereby its activation can be utilized as a preventive measure from neurodegenerative disorders [8,26]. In particular, hallmarks of AD include an accumulation of senile plaques mainly consisting of fibrillary Aβ peptide, dystrophic neurites, and neurofibrillary tangles composed of hyperphosphorylated tau protein in the brain, leading to dysfunction and loss of synapses and eventual neuronal death [27,28]. Although the etiology of AD remains unknown, oxidative stress is presumed to play a key role in initiation and

Discussion
While BV has been widely utilized as an acupuncture agent for the treatment of rheumatoid arthritis and osteoarthritis in Korea, WV is not well-studied for its clinical usefulness [21]. Although hymenoptera venoms can elicit both local and systemic allergic reactions, including life-threatening anaphylaxis, venom immunotherapy remains the most effective treatment reducing the risk of systemic reactions in individuals with hymenoptera venom allergy [22]. In addition, phospholipase A2 which is also found in BV was reported to have the potential to inhibit the progression of AD in the 3 × Tg AD mouse model presumably through the increase in regulatory T cell population [23].
WV has been reported to possess some pharmacological effects in the treatment of pain, inflammatory and neurodegenerative diseases [7,24,25]. Our previous study showed that WV exhibited strong antiinflammatory activity, although the effective dose was higher than the dose given of BV [6]. The present study is a follow-up to our previous work, and aimed to examine if WV could attenuate SCO-induced learning and memory impairment through antioxidant enzyme induction in the mouse model.
Multiple studies have indicated that Nrf2, a master regulator of inducible antioxidant enzymes, is involved in suppressing inflammatory responses and ROS generation in the brain, and thereby its activation can be utilized as a preventive measure from neurodegenerative disorders [8,26]. In particular, hallmarks of AD include an accumulation of senile plaques mainly consisting of fibrillary Aβ peptide, dystrophic neurites, and neurofibrillary tangles composed of hyperphosphorylated tau protein in the brain, leading to dysfunction and loss of synapses and eventual neuronal death [27,28]. Although the etiology of AD remains unknown, oxidative stress is presumed to play a key role in initiation and progression of the disease [29]. In this regard, our study investigated whether WV can promote antioxidant response by activating the Nrf2/HO-1 axis in the hippocampus and thus improve oxidative stress-induced cognitive damage in the SCO-treated mouse model.
Results from the behavioral tasks demonstrated that the administration of WV to mice at the doses of 50 to 500 µg/kg BW ameliorated SCO-induced spatial and associative learning and memory impairments. It has been known that oxidative stress raised in the mammalian brain contributes to cognitive impairment in experimental animals as well as human [13], and that SCO treatment can provoke oxidative stress in the brain and result in learning and memory deficits [14,15]. Thus, SCO-treated mice or rats are often used as a dementia model.
Interestingly enough, WV induced ARE-luciferase reporter activity in a concentrationdependent manner, suggesting that the presence of WV activated the Nrf2 signaling pathway and its downstream antioxidant enzyme genes like HO-1. Consistently, the expression of HO-1 protein was upregulated in HT22 cells and hippocampal tissue in mice. According to the literature, ROS production was inversely proportional to the expression level of HO-1 [30]. HO-1 catalytically converts heme to carbon monoxide and biliverdin, and biliverdin is subsequently metabolized to bilirubin which works as a strong biological antioxidant in mammalian cells. Therefore, we speculate that the protective effect of WV from oxidative stress-induced hippocampal neuronal cell death and memory disruption was mediated by Nrf2-dependent induction of antioxidant enzymes.
The effective dose of WV on antioxidant activity and cognitive function was approximately 10-fold higher than BV as assessed by cultured cell and mouse models. In addition, the cytotoxicity of WV was about 10-fold lower than BV. Therefore, it is presumed that bioactive ingredient(s) in WV is diluted by about 10-fold compared to BV. However, further study is needed to determine appropriate dosage regimens for the clinically beneficial effects in human health.
Although this study supports the memory-improving effect of WV in a mouse model, the active compounds responsible for cognitive enhancement remains unclear. WV reportedly contains a variety of biologically active constituents, including biogenic amines, enzymes, allergens, bioactive peptides, and many others [31]. Similar to BV, it is most likely that small molecule(s) in WV would exert a neuroprotective effect against SCO insult in the mouse brain and hippocampus via activation of the Nrf2 signaling pathway. We attempted to purify the WV component(s) involved in cognitive improvement using bioassay (ARE-luciferase reporter assay in HT22-ARE cells)-guided fractionation and succeeded in identifying serotonin as a potential bioactive component. Serotonin was found to efficiently suppress ROS production induced by tert-butyl hydroperoxide (tBHP) in mouse hippocampal HT22 cells (Supplementary Figure S1). It is consistent with the findings from a study by Liu and colleagues that WV exhibited antioxidant activity in human keratinocyte against oxidative stress, and that serotonin was identified as the major compound [32].
However, it is generally believed among scientists that serotonin rarely cross the blood-brain barrier (BBB), and therefore, it is uncertain yet whether serotonin found in WV is one of the bioactive compounds primarily responsible for improving the learning and memory function in mice. Based on previous research, serotonin may convert to BBBpermeable metabolite(s) such as 5-hydroxy tryptophan or melatonin in tissue or in blood before being transported to the brain [33,34] and consequently the metabolite(s) might have produced the effect. Another possibility is an indirect effect of serotonin-containing WV which was intraperitoneally administered to mice. Serotonin in mammals is mainly produced by enterochromaffin cells in the gut (about 90% of total serotonin) while the remaining part is synthesized in the brain [35,36]. Peripheral serotonin is actively taken up by platelets and released on their activation in blood and, moreover, contributes to a variety biological functions including immune responses and energy balance through the gut-brain axis [37,38]. Thus, it is conceived that WV-derived serotonin or its metabolites would indirectly influence the memory function in the brain via yet unknown mechanism(s).

Conclusions
In conclusion, we found that WV restored SCO-induced learning and memory impairment partly through activation of the Nrf2/HO-1 signaling pathway and subsequently increased antioxidant potential. However, further studies on the identification and working mechanism of bioactive component(s) in WV which are responsible for memory-enhancing effect in the SCO-induced amnesic mouse model are needed.

Preparation of WV and BV
V. velutina colonies were collected in South Korea during August and October of 2019, and were stored at −80 • C until needed. The venom sample was filtered through a Spin-X 0.45-µm cellulose acetate centrifuge tube filter (Corning Inc., Salt Lake City, UT, USA) after manual removing the venom sac from each wasp. The filtrate was then freeze-dried. The details and yield of WV sample preparation are described in our previous report [6]. BV powder was purchased from Chung Jin Biotech Co., Ltd. (Ansan, South Korea). The lyophilized WV and BV were dissolved in dimethyl sulfoxide (DMSO; Dongin Biotech, Seoul, South Korea) at a stock concentration of 100 mg/mL for further examinations.

Cell Viability
HT22 cells were plated at a density of 3 × 10 3 cells per well in a 96-well culture plate and maintained in 10% (v/v) FBS-containing DMEM for 24 h. Cells were then treated with WV and BV for 24 h in the absence and presence of 5 mM glutamate. Cell viability was determined by the cell counting kit-8 (CCK-8; Dojindo Laboratories, Kumamoto, Japan) as per the protocol supplied by the manufacturer. The relative cell viability is presented as the percentage of untreated cells.

ARE-Luciferase Reporter Assay
The HT22 cells (passage 8) transduced with pGL4.37[luc2P/ARE/Hyg] vector (Promega Corp., Madison, MA, USA) were stably established and named as HT22-ARE cell line as previously described [42][43][44]. Briefly, HT22 cells were plated at a density of 5 × 10 4 cells per well of 6-well culture plate and transfected with 100 ng of the vector containing luciferaseencoding gene following the ARE sequence using Lipofectamine TM 2000 Transfection Reagents. The transfectant cells were then selected by clonal growth in the maintenance medium containing 400 µM hygromycin B (Sigma-Aldrich).
For ARE-luciferase reporter assay, HT22-ARE cells (passages between 12-18 after transfection) were plated at a density of 5 × 10 5 cells per well in a 6-well plate and treated with WV and BV for 24 h. Sulforaphane (1 µM) was used as a positive control. The cells were then harvested and subjected to luciferase assay system (Promega Corp., Madison, WI, USA) as per the manufacturer's instruction. Briefly, the harvested cells were lysed using the provided lysis buffer. The lysates were mixed with the luciferase assay substrate, luciferin. The reaction mixture was transferred to each well of a 96-well plate. The luminescence was determined using a Glomax 96 microplate luminometer (Promega Corp.). Each value was normalized to its corresponding total protein. The results were averaged and calculated relative to the control.

Determination of Intracellular ROS Level
The cell permeant dye, 2',7'-dichlorodihydrofluorescein diacetate (H 2 DCFDA; Sigma-Aldrich), can be deacetylated by cellular esterase and later oxidized by ROS to produce dichlorofluorescein (DCF) which is highly fluorescent. The intracellular ROS level can thus be assessed based on the level of fluorescence after treatment with H 2 DCFDA [15,43]. To measure the intracellular ROS level, HT22 cells were plated at a density of 3 × 10 3 cells per well in a 96-well black polystyrene plate (Nunc, Rochester, NY, USA) or at a density of 3 × 10 4 cells per well in a 24-well transparent plate containing a 12-mm coverglass precoated with 10% poly-L-lysine solution. After treatment with 5 mM glutamate and/or venom sample for 6 h, the cells were loaded with 20 µM H 2 DCFDA at 37 • C for 30 min. The fluorescence of intracellular DCF was quantified using a fluorescence microplate reader (Tecan, Grödig, Austria) at excitation and emission wavelengths of 485 nm and 535 nm, respectively. The fluorescence for each condition was expressed as a fold change relative to the control. The cells placed on the coverglass were mounted onto the microscope slide (Thermo Fisher Scientific), and fluorescent images were then taken by a fluorescence microscope (Eclipse TE2000-U; Nikon, Tokyo, Japan).

Western blotting
HT22 cells were plated at a density of 5 × 10 5 cells in a 100-mm dish and treated with BV (1 and 2 µg/mL) or WV (10 and 20 µg/mL) for 24 h. The harvested cells were lysed and subjected to fractionation of nuclear and cytoplasmic proteins using NE-PER™ Nuclear and Cytoplasmic Extraction Reagents (Thermo Fisher Scientific). The extracted fractions were quantified by Bradford assay and the equal amount of proteins was loaded and electrophoretically separated onto 10% sodium dodecyl sulfate polyacrylamide gel. The proteins were then transferred to polyvinylidene fluoride (PVDF) membranes (Millipore, Burlingron, MA, USA). Protein bands on the PVDF membranes were allowed to react sequentially with primary antibodies (Abcam, Cambridge, UK) against HO-1 and β-actin in cytoplasmic fraction, and those for Nrf2 and lamin B in nuclear fraction and the appropriate secondary antibodies conjugated with horseradish peroxidase (Thermo Fisher Scientific). The antibody-bound proteins were visualized using the SuperSignal TM West Femto PLUS Chemiluminescent Substrate Kit (Thermo Fisher Scientific) and ImageQuant LAS 4000 mini (GE Healthcare Life Sciences, Little Chalfont, UK). Intensities of protein bands were determined by Image Studio Lite version 5.2 (LI-COR Biotechnology, Lincoln, NE, USA).
For the preparation of protein samples from the animal tissues, the whole brain was dissected from the sacrificed mice and the hippocampal area was removed. The hippocampal tissues were homogenized and subsequently processed for fractionation and Western blotting as mentioned above.

Animal Experiment
The animal study was conducted according to the guidelines of the Institutional Animal Care and Use Committee of Kyungpook National University (approval number: KNU 2021-007). Each group was subjected to histological analysis (n = 2) and biochemical assays (n = 5) otherwise stated. C57BL/6J mice (6-week old, male) were obtained from Orient Bio Inc. (Seongnam, South Korea). After a week of acclimation, a total of 56 mice were allocated into 8 treatment groups (7 mice per group) as follows: (1) a group received vehicle only; (2) a group received SCO (Sigma-Aldrich) alone at a dose of 1 mg/kg body weight (BW); (3) a group received SCO and Donepezil (Sigma-Aldrich) at 5 mg/kg BW as a positive control; (4) a group received SCO and BV at 5 µg/kg BW; (5) a group received SCO and BV at 50 µg/kg BW; (6) a group received SCO and WV at 50 µg/kg BW; (7) a group received SCO and WV at 250 µg/kg BW; and (8) a group received SCO and WV at 500 µg/kg BW. Lyophilized BV and WV were dissolved in DMSO. Vehicle was normal saline (Sigma-Aldrich) containing 0.5% (v/v) DMSO and 5% (v/v) Tween ® 80. SCO and venom samples were all intraperitoneally injected every day. BV and WV were injected 15 h prior to the SCO injection on a daily basis for the whole experimental period (refer to Figure 6). Behavioral tests were conducted 30 min after SCO treatment. At termination of scheduled experiments, all mice were sacrificed by asphyxiation in a CO 2 chamber and dissected for brain and liver tissues and blood.

Behavioral Test
The learning and memory impairment behavior was tested by the Y-maze, passive avoidance, and Morris water maze tasks according to the procedures described in our previous reports [15,42,[44][45][46][47][48].
The passive avoidance test was performed in the testing apparatus (Gemini Avoidance System, San Diego, CA, USA) composed of two chambers and a guillotine door. On the first day (Experimental Day 2; refer to Figure 6), every mouse was adapted in the apparatus by placing it in the bright chamber and allowing to move back and forth to the dark chamber for 1 min. On the second day, each mouse was placed in the bright chamber. When the mouse moved to the dark chamber, an electrical food shock (0.5 mA) was delivered for 3 s after the door closed. On the third day (Experimental Day 4), each mouse was again placed in the bright chamber and the latency time for a mouse to stay in the bright chamber was acquired. The latency over 5 min was clocked as 300 s.
The Morris water maze test was performed in a circular water pool (90 cm in diameter and 45 cm in height; colored with nontoxic paint). On the first day (Experimental Day 5), each mouse was allowed to freely swim for 60 s. On the next day, a platform was submerged in one of the pool quadrants. For a consecutive three days (Experimental Day 6-8), mice were given three trials per session per day to search for the platform in place. If the mice did not locate the platform in 60 s, it was guided to place on the platform and allowed to stay for 10 s. On the fifth day (Experimental Day 9), the swimming time until a mouse arrived the platform was recorded.
The Y-maze test was performed in the Y-shaped maze having three arms on Experimental Day 10. Each mouse was placed in the A arm, and its alternations among the arms were monitored and recorded. Spontaneous alternations were defined as consecutive triplets of different arm entries. The percentage of alternations was calculated as follows: Spontaneous alternation (%) = [number of alternations/(total arm entries −2)] × 100.

Histological Analysis by Hematoxylin and Eosin (H&E) Staining
Collected brain tissues were immediately fixed in formalin solution (Sigma-Aldrich) and embedded in paraffin, as previously described [49]. The paraffin blocks were then sectioned at a thickness of 5 µm using a microtome (RM-2025 RT; Leica, Nussloch, Germany). The sections including parts of the hippocampi were placed on Superfrost PLUS microscope slides (Marienfeid, Lauda-Konigshfen, Germany), air-dried at 37 • C for 12 h, and stored at 4 • C before being processed for H&E staining as previously described [15,42,48].

Measurement of Plasma 8-hydroxy-2 -deoxyguanosine (8-OHdG) Level
To determine the level of 8-OHdG in the plasma, a biomarker for oxidative DNA damage, whole blood was collected from the mice into microcentrifuge tube coated with 10 unit of heparin (Sigma) and centrifuged at 2500× g for 15 min using a centrifuge (Gyrogen, Gimpo, South Korea). The plasma was subjected to quantification of 8-OHdG using an ELISA kit (Cat# ADI-EKS-350; Enzo Life Sciences International Inc., Plymouth Meeting, PA, USA) according to the manufacturer's instructions.