Treatment with Akebia Saponin D Ameliorates Aβ1–42-Induced Memory Impairment and Neurotoxicity in Rats

Amyloid-β peptide (Aβ) is known to be directly associated with the progressive neuronal death observed in Alzheimer’s disease (AD). However, effective neuroprotective approaches against Aβ neurotoxicity are still unavailable. In the present study, we investigated the protective effects of Akebia saponin D (ASD), a typical compound isolated from the rhizome of Dipsacus asper Wall, on Aβ1–42-induced impairment of learning and memory formation and explored the probable underlying molecular mechanisms. We found that treatment with ASD (30, 90 or 270 mg/kg) significantly ameliorated impaired spatial learning and memory in intracerebroventricularly (ICV) Aβ1–42-injected rats, as evidenced by a decrease tendency in escape latency during acquisition trials and improvement in exploratory activities in the probe trial in Morris water maze (MWM). Further study showed that ASD reversed Aβ1–42-induced accumulation of Aβ1–42 and Aβ1–40 in the hippocampus through down-regulating the expression of BACE and Presenilin 2 accompanied with increased the expression of TACE, IDE and LRP-1. Taken together, our findings suggested that ASD exerted therapeutic effects on Aβ-induced cognitive deficits via amyloidogenic pathway.


Introduction
Alzheimer's disease (AD) is a multifaceted neurodegenerative disorder featured by progressive memory and cognitive impairment [1][2][3]. The pathological hallmarks in AD brains are extracellular amyloid plaques composed of aggregated-amyloid (Aβ) peptide [4] and abnormally phosphorylated microtubule-associated tau protein [5]. Neurotoxicity Aβ hypothesis of AD. Aβ is derived from sequential endoproteolytic cleavage of a transmembrane Aβ precursor protein (APP) by type 1 transmembrane protein β-site APP cleaving enzyme 1 (BACE1) and γ-secretase complex yielding two major Aβ C-terminal variants, Aβ  and Aβ 1-42 [6][7][8]. Aggregate Aβ of low molecular weight has been proved to exhibit the greatest neurotoxicity. Alternatively, APP can be subjected to a non-amyloidogenic processing. In the non-amyloidogenic pathway, α-secretase cleavage precludes Aβ generation, releasing a soluble protein called sAPPα [9], which has been reported to have neurotrophic and neuroprotective properties [10,11]. The past decade has witnessed the discovery of  To evaluate the effects of ASD on spatial learning and memory abilities in ICV Aβ 1-42 -injected rats, we assessed the performance of rats using Morris water maze. Regardless of different treatments, all the rats displayed progressive decrease in escape latency to reach the hidden platform. Results indicated that ICV Aβ 1-42 -injected pronouncedly increased escape latencies compared to the vehicle controls on sessions 2, 3, 4. Interestingly, rat treatment with ASD for four weeks significantly decreased escape latencies compared to the ICV Aβ 1-42 -injected group, especially at 90 mg/kg for session 3-4 and 70 mg/kg for session 4 [4 trials/rat/day for 4 days, effect of day, F (6, 69) = 264.68, p < 0.01; effect of group, F (6, 69) = 16.53, p < 0.001; effect of group-by-day interaction, F (6, 69) = 2.25, p < 0.001, Figure 1A]. In addition, the rats in ASD treatment groups also displayed progressive decrease in distance traveled compared to the ICV Aβ 1-42 -injected group [effect of day, F (6, 69) = 557.46, p < 0.001; effect of group, F (6, 69) = 17.59, p < 0.001; effect of group-by-day interaction, F (6, 69) = 2.92, p < 0.001, Figure 1B]. In the probe trial, the platform removed from the pool. The time spent in the IV quadrant and the number of platform location crossings was recorded. Rats treatment with ASD significantly increased the time s which were decreased obviously in the model group ( Figure 1C), and the distances traveled were significantly increased which were decreased obviously in the model group [effect of group, F (6, 69) = 13.61, Figure 1C; effect of group, F (6, 69) = 29.68, Figure 1D]. The strategy of searching for the hidden platform was also recorded ( Figure 1E). Taken together, our findings suggest that ASD may prevent significantly Aβ 1-42 -induced memory impairment in rats. The ratio of distance in the target quadrant to total moved distance during the probe trial test are presented 24 h after the last acquisition trial; (E) Representative swim paths during the spatial probe test are also shown. Values shown are expressed as means ± SEM, n = 10, * p < 0.05, ** p < 0.01, *** p < 0.001 vs. model group.

HE Staining
HE staining revealed no remarkable neuronal abnormalities in the hippocampus of rats in the control group. The pyramidal cells in the CA1 region were arranged neatly and tightly, and no cell loss was found. Additionally, for the control group, cells were round and intact with nuclei stained clear, dark blue ( Figure 2B). However, obvious hippocampal histopathological damage was observed in the model group. The pyramidal layered structure was disintegrated, and neuronal loss was found in the CA1 region. Neurons with pyknotic nuclei and with shrunken or irregular shape were also observed ( Figure 2C). These abnormalities were attenuated by DON, GLT and ASD treatment. The cells in ASD groups had better cell morphology and were more numerous than those in the Model groups, but were overall worse than those in the control group. The average number of Figure 1. Effects of ASD on the spatial learning and memory deficits in Aβ 1-42 -induced rats evaluated by Morris water-maze test. (A) Changes in escape latency to reach the hidden platform during the 4-day acquisition trails and (B) distances traveled; (C) The times of former platform location crossings and (D) The ratio of distance in the target quadrant to total moved distance during the probe trial test are presented 24 h after the last acquisition trial; (E) Representative swim paths during the spatial probe test are also shown. Values shown are expressed as means˘SEM, n = 10, * p < 0.05, ** p < 0.01, *** p < 0.001 vs. model group.

HE Staining
HE staining revealed no remarkable neuronal abnormalities in the hippocampus of rats in the control group. The pyramidal cells in the CA1 region were arranged neatly and tightly, and no cell loss was found. Additionally, for the control group, cells were round and intact with nuclei stained clear, dark blue ( Figure 2B). However, obvious hippocampal histopathological damage was observed in the model group. The pyramidal layered structure was disintegrated, and neuronal loss was found in the CA1 region. Neurons with pyknotic nuclei and with shrunken or irregular shape were also observed ( Figure 2C). These abnormalities were attenuated by DON, GLT and ASD treatment. The cells in ASD groups had better cell morphology and were more numerous than those in the Model groups, but were overall worse than those in the control group. The average number of healthy cells was highest in the control group, lower in the treated groups, and lowest in the Model group [effect of group, F (6, 20) = 17.30, p < 0.001, Figure 2I].  Figure 2I]. In Model and DON groups, cells in the hippocampal CA1 region appeared decreased in number. Furthermore, the remnants of the pyramidal cells were arranged irregularly and some exhibited shrunken and irregular shape. The cells in ASD-H group were more numerous with better cell morphology and were more numerous than those in Model and DON groups, *** p < 0.001 vs. model group.

ASD Blocked Aβ1-42-Induced Production of Aβ
To examine the potential mechanisms of ASD, the levels of Aβ1-42 and Aβ1-40 in hippocampus were measured by ELISA assays. As the result exhibited, there was a rise in concentration of Aβ1-40 and Aβ1-42 in ICV Aβ1-42-injected group compared to the corresponding vehicle control group. Of note, ASD treatment significantly decreased the level of Aβ1-42 and Aβ1-40 compared with model group [effect of group, F (6, 69) = 290.0, p < 0.001, Figure 3A; effect of group, F (6, 69) = 24.23, p < 0.001, Figure 3B]. As demonstrated by these results, ASD might regulate the generation of Aβ1-40 and Aβ1-42.  In Model and DON groups, cells in the hippocampal CA1 region appeared decreased in number. Furthermore, the remnants of the pyramidal cells were arranged irregularly and some exhibited shrunken and irregular shape. The cells in ASD-H group were more numerous with better cell morphology and were more numerous than those in Model and DON groups, *** p < 0.001 vs. model group.

ASD Blocked Aβ1-42-Induced Production of Aβ
To examine the potential mechanisms of ASD, the levels of Aβ1-42 and Aβ1-40 in hippocampus were measured by ELISA assays. As the result exhibited, there was a rise in concentration of Aβ1-40 and Aβ1-42 in ICV Aβ1-42-injected group compared to the corresponding vehicle control group. Of note, ASD treatment significantly decreased the level of Aβ1-42 and Aβ1-40 compared with model group [effect of group, F (6, 69) = 290.0, p < 0.001, Figure 3A; effect of group, F (6, 69) = 24.23, p < 0.001, Figure 3B]. As demonstrated by these results, ASD might regulate the generation of Aβ1-40 and Aβ1-42.

Effects of ASD on the Expression of Aβ Generation Proteins
To explore the underlying mechanisms of ASD against Aβ-induced neurotoxicity by down-regulating the generation of Aβ1-42 and Aβ1-40, we examined the expression of TACE, BACE, presenilin 1 and presenilin 2 by Western blot analyses. As the results exhibited, the expression of TACE was significantly increased in the hippocampus of rats treated with ASD [effect of group, F (6, 20) = 9.960, p < 0.001, Figure 4B]. The expression of BACE and presenilin 2 was decreased [effect of group, F (6, 20) = 13.38, p < 0.001, Figure 4C; effect of group, F (6, 20) = 13.38, p < 0.001, Figure 4E], the expression of presenilin 1 had no significant change [effect of group, F (6, 20) = 0.1369, p = 0.9889, Figure 4D]. Quantification of TACE (B); BACE (C); Presenilin 1 (D) and Presenilin 2 (E) was represented as the ratio (in percentage) of the sham group. β-actin was used as a loading control. The data were expressed as mean ± SEM of three independent experiments, * p < 0.05, ** p < 0.01, *** p < 0.001 vs. hippocampus of vehicle plus Aβ1-42 group.

Effects of ASD on the Expression of Aβ Degradation-Related and Transshipment-Related Proteins
To further explore the underlying mechanisms of ASD on Aβ degradation, we examined the expression of IDE and LRP-1 by Western blot analyses. As shown in Figure 5, ASD significantly increased the expression of IDE and LRP-1 in the hippocampus of rats (p < 0.05 or p < 0.01, Figure5A,B).

Effects of ASD on the Expression of Aβ Degradation-Related and Transshipment-Related Proteins
To further explore the underlying mechanisms of ASD on Aβ degradation, we examined the expression of IDE and LRP-1 by Western blot analyses. As shown in Figure 5, ASD significantly increased the expression of IDE and LRP-1 in the hippocampus of rats (p < 0.05 or p < 0.01, Figure 5A,B).

Animals and Housing
A total of 84 Sprague-Dawley rats of adult male, weighing between 250 and 300 g, were procured from B & K Laboratory Animal Corp. Ltd. Shanghai, China. Throughout the study period, the rats were maintained in accordance with the guidelines of National Institutes of Health Guide for Care and Use of Laboratory Animals (publication no. 85-23, revised 1985). Twelve rats were housed in each cage during the study period and were maintained under standardized conditions (22 ± 1 °C, 60% ± 10% humidity, 12 h light/dark cycle). Water was provided ad libitum.

Drugs and Materials
ASD ( Figure 6) was prepared by Professor Zhong-Lin Yang. Isolated from Dipsacus asperoides at the Key Laboratory of Modern Chinese Medicines, China Pharmaceutical University, and and MALDI-TOF mass spectrometry revealed >93.4% purity. Commercially Aβ1-42 (Invitrogen, Waltham, MA, USA) was stored at −20 °C, then the peptide was dissolved in sterile normal saline at concentration of 2 μg/μL, this solution was incubated at 37 °C for 7 days before the surgery [32].

Animals and Housing
A total of 84 Sprague-Dawley rats of adult male, weighing between 250 and 300 g, were procured from B & K Laboratory Animal Corp. Ltd. Shanghai, China. Throughout the study period, the rats were maintained in accordance with the guidelines of National Institutes of Health Guide for Care and Use of Laboratory Animals (publication no. 85-23, revised 1985). Twelve rats were housed in each cage during the study period and were maintained under standardized conditions (22˘1˝C, 60%˘10% humidity, 12 h light/dark cycle). Water was provided ad libitum.

Drugs and Materials
ASD ( Figure 6) was prepared by Professor Zhong-Lin Yang. Isolated from Dipsacus asperoides at the Key Laboratory of Modern Chinese Medicines, China Pharmaceutical University, and and MALDI-TOF mass spectrometry revealed >93.4% purity. Commercially Aβ 1-42 (Invitrogen, Waltham, MA, USA) was stored at´20˝C, then the peptide was dissolved in sterile normal saline at concentration of 2 µg/µL, this solution was incubated at 37˝C for 7 days before the surgery [32].
The day of ICV Aβ1-42 injection was designated as Day 0. One day after the surgery, rodent feed and sterile distilled water were provided ad libitum. The test group, ASD (30, 90, 270 mg/kg), donepezil (0.5mg/kg) [34,35] or Ginkgo Biloba Extract (25 mg/kg) were intragastrically administered once daily for consecutive 30 days. In control and model groups the same volume of vehicle as ASD were received.

Morris Water Maze Task
The Morris water maze task test was conducted to evaluate the spatial learning and memory abilities, which consisted of 4-day training and a probe trial on day 5, carried out from Day 26 to Day 30. This was performed as descried previously [36,37]. Rats were individually trained in a black circular pool (150 cm in diameter and 50 cm in height) filled to a depth of 30 cm water with temperature maintained at 25 ± 1 °C. The circular pool was divided into four imaginary quadrants (I, II, III and IV). On each training day (Day 26 to Day 29), the platform always in the middle of IV quadrant (10 cm in diameter) was submerged 1 cm below water and hidden from the rats' view. The animals were subjected to four trials with a 1 h interval between trials. Each trial lasted for 90 s unless the animal reached the platform. If an animal failed to find the platform within 90 s, the test would be ended with the animal being gently navigated to the platform by hand for 30 s. On day 5 (Day 30), the platform was removed and the probe trial was started, during which animals had 90 s to search for the platform. The time spent in the target quadrant and the number of target crossings (i.e., the quadrant where the platform was previously located) was recorded. Data of the escape latency, the time spent in the target quadrant, the number of target crossings and swimming speed were collected by the video tracking equipment and processed by a computer equipped with an analysis-management system (Viewer 2 Tracking Software, Ji Liang Instruments, Shanghai, China).

Hematoxylin-Eosin Staining of Neurons in the Hippocampus of Rats
On Day 28, following the last treatment, 3 rats of each group were decapitated after injection of a lethal dose of chloral hydrate. The brains were rapidly removed and immersed in 4% formaldehyde for 48 h, then removed and dehydrated in a dehydrator overnight before being embedded in paraffin. Brains were section at a thickness of 5-μm sections and then baked for 20 min. Subsequently, sections were dewaxed with xylene three times followed by washed with 100%, 95%, 90% and 85% ethyl
The day of ICV Aβ 1-42 injection was designated as Day 0. One day after the surgery, rodent feed and sterile distilled water were provided ad libitum. The test group, ASD (30, 90, 270 mg/kg), donepezil (0.5 mg/kg) [34,35] or Ginkgo Biloba Extract (25 mg/kg) were intragastrically administered once daily for consecutive 30 days. In control and model groups the same volume of vehicle as ASD were received.

Morris Water Maze Task
The Morris water maze task test was conducted to evaluate the spatial learning and memory abilities, which consisted of 4-day training and a probe trial on day 5, carried out from Day 26 to Day 30. This was performed as descried previously [36,37]. Rats were individually trained in a black circular pool (150 cm in diameter and 50 cm in height) filled to a depth of 30 cm water with temperature maintained at 25˘1˝C. The circular pool was divided into four imaginary quadrants (I, II, III and IV). On each training day (Day 26 to Day 29), the platform always in the middle of IV quadrant (10 cm in diameter) was submerged 1 cm below water and hidden from the rats' view. The animals were subjected to four trials with a 1 h interval between trials. Each trial lasted for 90 s unless the animal reached the platform. If an animal failed to find the platform within 90 s, the test would be ended with the animal being gently navigated to the platform by hand for 30 s. On day 5 (Day 30), the platform was removed and the probe trial was started, during which animals had 90 s to search for the platform. The time spent in the target quadrant and the number of target crossings (i.e., the quadrant where the platform was previously located) was recorded. Data of the escape latency, the time spent in the target quadrant, the number of target crossings and swimming speed were collected by the video tracking equipment and processed by a computer equipped with an analysis-management system (Viewer 2 Tracking Software, Ji Liang Instruments, Shanghai, China).

Hematoxylin-Eosin Staining of Neurons in the Hippocampus of Rats
On Day 28, following the last treatment, 3 rats of each group were decapitated after injection of a lethal dose of chloral hydrate. The brains were rapidly removed and immersed in 4% formaldehyde for 48 h, then removed and dehydrated in a dehydrator overnight before being embedded in paraffin. Brains were section at a thickness of 5-µm sections and then baked for 20 min. Subsequently, sections were dewaxed with xylene three times followed by washed with 100%, 95%, 90% and 85% ethyl alcohol solutions for 1 min each and washed with water. Next, sections were stained with hematoxylin, differentiated with hydrochloric acid alcohol solution, treated with 1% ammonia, and stained with eosin for 3 s. After being washed with water, and then 85%, 90%, 95% and 100% ethyl alcohol for 1 min each, sections were cleared in xylene for 2 min and cover-slipped with resin.

Measurement of Aβ 1-40 and Aβ
Standard curves were prepared with synthetic Aβ 1-42 (Bachem, Torrance, CA, USA) diluted in extracts of non-transgenic mouse forebrain prepared in parallel as described above. Each sample was analyzed in duplicate.

Statistical Analysis
All experimental data shown are expressed as means˘SD. All the data, unless specified, were analyzed with one-way ANOVA which were followed by Newman-Keuls tests for multiple comparisons using SPSS 11.5 software (SPSS China, Shanghai, China). Statistical significance was considered when p < 0.05.

Discussion
A primary pathological hallmark of AD is the accumulation of Aβ in brain [38], which is generated by sequential cleavage from the proteolytic processing of β-amyloid precursor protein (APP). APP generates various peptide species including Aβ 1-42 , a more toxic form prone to oligomerize leading to neuroinflammation [39][40][41], synaptic dysfunction [42,43], apoptosis [44,45] and memory impairment [46], all of which are observed during the progression of AD. One of the important strategies for preventing or treating AD is to interfere with the toxic Aβ species.
Here, we demonstrated that ASD prevented Aβ 1-42 -induced spatial learning and memory impairments, as evidenced by the decrease in escape latency during acquisition trials and improvement in exploratory activities in the probe trial in the MWM task, coupled with modified expression generation-related proteins (TACE, BACE, Presenilin 2), degradation-related proteins (IDE) and transshipment-related proteins (LRP-1).
In this study, rats exposed to 10µg amyloid beta (1-42) into the bilateral hippocampus presented a significant impairment in learning and memory ability. In addition, insulin degrading enzyme (IDE), an enzyme involved in Aβ clearance, can regulate the level of Aβ in hippocampus. As the results showed, treatment with ASD (90 mg/kg, 270 mg/kg) significantly increased the levels of secreted IDE (* p < 0.05, *** p < 0.001) and dose-dependently improved the learning and memory deficits induced by Aβ  . In agreement with the experiment data, we hypothesized that ASD might ultimately enhance the levels or activities of secreted IDE which led to induced clearance of Aβ in the hippocampus. Moreover, Aβ 1-40 and Aβ 1-42 are produced by the proteolytic processing of APP. Proteases referred to as secretase cleave at the N-and C-termini of Aβ within APP to generate Aβ. A previous study showed that the levels of Aβ 1-40 is much higher than Aβ 1-42 after APP cleavage, we investigated the levels of Aβ 1-40 in hippocampus at the same time.
The results indicated that ASD can significantly reduce the levels of Aβ 1-40 in hippocampus, along with changes in the levels of Aβ generate e related enzyme. In conclusion, we hypothesis that ASD regulate the proteolytic processing of APP, mainly regulate the α-secretase, β-secretase or Presenilin 2 to reduce the levels of Aβ 1-40 in hippocampus.
In conclusion, we demonstrated that ASD, a saponin separated from Dipsacus asperwall, significantly improved the learning and memory abilities impaired by Aβ injury, protected neurons and markedly regulating Aβ generate-related proteins (TACE, BACE, Presenilin 2) expression. In addition, we found that ASD might promoted the Aβdegradation by regulating Aβdegradation-related proteins (IDE) expression. These findings indicate that ASD might have a therapeutic potential in AD. In the future, In the future, much work requires to be done to demonstrate the effects of ASD on the deposition of amyloid plaques, neurofibrillary tangles and other distinct neuropathological features in AD.

Conclusions
In conclusion, the present study has demonstrated for the first time that ASD exerted potently therapeutic effects on Aβ-induced cognitive deficits via amyloidogenic pathway, as evidenced by a decrease tendency in escape latency during acquisition trials and improvement in exploratory activities in the probe trial in Morris water maze (MWM). Moreover, we firstly found that ASD reversed Aβ 1-42 -induced accumulation of Aβ 1-42 and Aβ 1-40 through down-regulating the expression of BACE and Presenilin 2 accompanied with increased the expression of TACE, IDE and LRP-1.