The seed oil of Paeonia ludlowii ameliorates Aβ25‐35‐induced Alzheimer's disease in rats

Abstract Paeonia ludlowii, a plant of the Paeoniaceae family, has abundant genetic diversity in different populations, and the seed oil can be used in a diverse number of activities. However, its neuroprotective effect is not clear. We investigated the memory‐improving effects and associated mechanisms of Paeonia ludlowii seed oil (PLSO) on amyloid beta (Aβ)25–35‐induced Alzheimer's disease (AD) in rats. The Morris water maze test was undertaken, and subsequently, the content of malondialdehyde (MDA), superoxide dismutase (SOD), glutathione (GSH), and acetylcholinesterase (ACHE) in the hippocampus was detected by biochemical analyses. To further study PLSO, we examined the pathologic structure and apoptosis of hippocampal tissue by staining. Immunohistochemical analysis was used to detect expression of IBA‐1 and GFAP in the hippocampus. Detection of proinflammatory factors was achieved by reverse transcription–quantitative polymerase chain reaction and Western blotting. High‐dose PLSO inhibited expression of GFAP and IBA‐1. We demonstrated that high‐dose PLSO can regulate activation of glial cells and mediate apoptosis of hippocampal cells, and significantly improve learning and memory deficits in AD rats. PLSO could be developed as a nutritional supplement and sold as a drug for AD prevention and/or treatment.


| INTRODUC TI ON
Alzheimer's disease (AD) dementia refers to a particular onset and course of cognitive and functional decline associated with age together with a particular neuropathology (Soria Lopez et al., 2019). It is the most common cause of dementia worldwide and reported to be the sixth leading cause of death for citizens in the USA (Masters et al., 2015). Aging is considered to be the main risk factor for AD.
The main manifestations of AD are memory loss, cognitive impairment, and personality change (Graham et al., 2017). The pathologic features of AD are the accumulation of "senile plaques" and intracellular neurofibrillary tangles, accompanied by the loss of synapses and neurons, leading to cognitive dysfunction and, eventually, dementia (Congdon et al., 2018;Kimura et al., 2016).
Five drugs have been approved for AD treatment. Four are acetylcholinesterase (ACHE) inhibitors (donepezil, tacrine, rivastigmine, and galantamine), and one is an antagonist of N-methyl-Daspartate receptors (memantine). Donepezil (Aricept ® ) is a centrally acting reversible ACHE inhibitor. It is first-choice AD treatment recommended by the National Institute for Health and Care Excellence in the UK. However, like other AD drugs, donepezil improves the cognition and behavior of AD patients without slowing down AD progression (Lee et al., 2015). In addition, donepezil has several side effects, such as loss of appetite, gastrointestinal upset, sleeping difficulty, and muscle cramp (Querfurth et al., 2010). Moreover, due to poor efficacy or severe adverse reactions, few clinical trials of AD drugs have been successful (Godyń et al., 2016). A new drug for AD treatment has not been approved since 2003. In spite of this grave situation, no effective treatments are available for AD (Scheltens et al., 2010). There is an urgent need to find and develop novel anti-AD drugs or nutritional supplements that are more effective and/ or less toxic.
Natural products have a significant role in the prevention of disease and boosting of health in humans (Rehman et al., 2019).
Some scholars have suggested that various oils and nutrients exhibit preventative effects against AD, for example, fish oil (Xiang et al., 2016;Xie et al., 2020), walnut oil (Liao et al., 2020), and Rhodiola rosea (Vasileva et al., 2016). The roots of Paeonia ludlowii have been used in Tibetan folk medicine for ~100 years for treatment of diseases (gynecological, skin, cardiovascular, cerebrovascular) (Jiang et al., 2017). Some studies have been carried out on its chemical composition, endophytic bacteria, and habitat profile (Jiang et al., 2016;Lu et al., 2020;Yang et al., 2007;Zeng et al., 2015). However, studies on improvement of memory impairment using Paeonia ludlowii seed oil (PLSO) have not been carried out. Based on previous study from our research group (zhang et al., 2020), it is speculated that the PLSO may have the effect of enhancing memory. This study addressed to investigate whether PLSO can improve ameliorating learning and memory impairment in Aβ-induced rat model of AD, as well as to clarify the possible mechanism briefly underlying the neuroprotective effect of PLSO in ameliorating.

| Ethical approval of the study protocol
This study was carried out in strict accordance with the recom-

| Reagents and chemicals
Donepezil hydrochloride tablets (lot: 110822A) were purchased from Eisai China and prepared in a mixed suspension at 0.1 mg/ml. Aβ25-35 (lot: Y-0044) was obtained from Bioss Biotech. Streptavidinbiotin complex and terminal deoxynucleotidyl transferase dUTP nick-end labeling (TUNEL) apoptosis detection kits were purchased from Boster Biotech. A real-time reverse transcription-quantitative polymerase chain reaction (RT-qPCR) kit was obtained from TransGen Biotech. Antibodies against Bcl-2 Bax, and β-actin were purchased from Bio-Rad Laboratories. A brain stereotaxic apparatus (ZH-LAN-STAR, C type) was obtained from Huaibei Zhenhua Biological Equipment. An inverted microscope (S70) was purchased from Leica. A ChemiDoc XRS gel imaging and analysis system was obtained from Bio-Rad Laboratories. An SP-9000 immunohistochemistry kit was purchased from Beijing ZSGB Biological Engineering.
PCR amplification instrument (TC-512) was from Techen. Detection kits for malondialdehyde (MDA), superoxide dismutase (SOD), glutathione (GSH) and ACHE were purchased from Nanjing Jiancheng Bioengineering Research Institute. Rabbit anti-GFAP antibody (lot: 16825-1-AP) was obtained from Proteintech. Rabbit anti-IBA1 antibody (lot: bs-1363R) was from Bioss Biotech. University (Nyingchi City). The seed pretreatment and oil extraction process of Paeonia ludlowii was carried out as follows: first of all, the full, normal size, no mechanical damage, and no mildew Paeonia ludlowii seeds were picked out, then manually its black outer kernel coats were removed to obtain Paeonia ludlowii kernels, subsequently, which was placed in a drying oven at 65°C to a constant quality, and the press machine was set in motion, then the Paeonia ludlowii seed kernel was added into the feeder, preheated for 6 min, scent was set and pressed, and after pressing and extraction, the crude oil in the collecting bottle was filtered to obtain the Paeonia ludlowii seed oil.

| Experimental animals
Specific-pathogen-free male Sprague Dawley rats (6 months; 180-220 g) were purchased from Shanghai SIPPR-BK Laboratory Animals (certificate number 20170005014637). Animals were allowed to acclimatize for 1 week in an environmentally controlled breeding room at room temperature (25 ± 1°C) and a 12-hr light/dark cycle. Animals had access to food and water ad libitum.
Except for the control group, the AD model was established in all groups. Anesthesia was induced by 1% pentobarbital sodium (40 μg/g, i.p.). Then, rats were fixed on the brain stereotaxic instrument. According to the stereotaxic map of the rat brain, the anterior fontanel was the starting point, 3.5 mm behind the bregma was the puncture point, and the right side of the midline was opened 2 mm, and a posterior dental drill was used to open the skull (AP = −3.5 mm, Ml = 2.0 mm, DV = 3.0 mm). Then, 10 μg (1 μl) of Aβ1-40 was injected at both sides of the CA1 area of the hippocampus at a slow and uniform rate, the needle was retained for 5 min, and the wound was closed after needle withdrawal. Drug administration started 4 days after the model had been created. Rats were gavaged according to each dose of the drug, once a day for 2 weeks; the control group and model group were given physiologic (0.9%) saline.

| Morris water maze test (MWMT)
The spatial learning was investigated using the Morris water maze. The general testing procedure was described previously . Rats were removed from cages and placed rapidly into the water maze. When the rat entered the water through the monitoring system, the button on the video recorder was pressed immediately. When the two cooperate, the rat pressed the button when entered the water. The recording time was set to 90 s. When the rat arrived at the station platform and stayed there for 30 s, this was regarded as "finding the station," and video recording stopped automatically. If the rat did not swim on to the platform within 90 s, a wooden stick was used to guide the rat to the platform immediately.
The rat was removed after becoming familiar with the surrounding environment on the platform for 10 s; if the rat arrived on the platform or stayed for a specified time, he was taken away again after becoming familiar with the environment for 10 s. After removing the rat from the water maze, he was dried with a towel. After completing each set of experiments, the next round of experiments was carried out at 30-min intervals. Four rounds of experiments were completed. Positioning navigation experiment was used for 4 days. On day 5, remove the platform, select the opposite quadrant of the quadrant where the platform is located as the water entry point, only do it once, and the recording time is 60 s.

| Biochemical test
Biochemical test was performed according to protocol published in previous study (Engvall et al., 1971). Detection kits for MDA, SOD, GSH, and ACHE were operated according to the manufacturer's (Nanjing Jiancheng Bioengineering Research Institute) instructions.
The content of each compound was calculated according to the formula.

| Histopathology of the hippocampal neurons of rats
Anesthesia was induced with pentobarbital sodium (1.0%, i.p.).
Brains were removed from anesthetized rats and weighed. Brain tissue was fixed in 10% neutral buffered formalin, dehydrated, and embedded in paraffin. Coronal sections were obtained using a microtome, followed by staining with hematoxylin and eosin (H&E) for pathology studies under light microscopy (200×), according to protocol published in previous study (Alghamdi et al., 2018).

| Neuron apoptosis by the TUNEL assay
The apoptotic study was carried out with TUNEL analysis as previously described (Fang et al., 2005). TUNEL staining was used to assess apoptosis. Tissue blocks were placed in 10% formalin for 48 hr.
After washing with phosphate-buffered saline (PBS), staining with 3,3-diaminobenzidine was done for 3-10 min. This action was followed by rinsing with tap water, counterstaining with hematoxylin, and differentiation with 1% hydrochloric acid in alcohol. After observation under a microscope, the degree of staining was controlled.
Dehydration with different concentrations of alcohol was undertaken for 3 min each time. Sealing with neutral gum was completed.
Images were captured under an inverted microscope (S70; Leica) and the apoptosis rate was calculated.

| Immunohistochemical (IHC) analyses
The reactive microglia and astrocytes in the brain samples were visualized by immunohistochemistry (IHC) that was performed as described in the previous study with slight modifications (Delcambre et al., 2016). Tissue blocks were placed in 10% formalin for 48 hr.
Fixed tissue was washed with running water to remove residual fixative and impurities. Ethanol of different concentrations enabled dehydration in a step-by-step manner. Blocks were cleared, soaked in wax, and embedded in paraffin. Slides were placed in a constanttemperature oven at 65°C for 30 min. Then, they were placed in xylene I solution and allowed to soak for 15 min. Next, they were placed in xylene II solution for 15 min to dewax and rehydrate.
Sodium citrate buffer solution (0.01 M) was employed for high-pressure heat repair for 15 min. After natural cooling, slides were washed thrice with PBS (0.02 M) for 3 min each time. Microwave treatment in sodium citrate (0.01 mol/L) for 15 min was carried out. EDTA (pH 8.0) or 0.01 M sodium citrate buffer solution (pH 6.0) was added to boiling water for high-pressure heat repair. Next, slides were placed in 3% H 2 O 2 and allowed to incubate for 10 min in a wet box to eliminate derived peroxidase activity. Then, sections were incubated with primary antibody (CHOP; 1:100 dilution; GRP78, 1:50; caspase-3, 1:50; Proteintech) at 4°C overnight. Finally, sections were incubated with horseradish peroxidase-conjugated secondary antibodies (Santa Cruz Biotechnology) and counterstained with hematoxylin.
All tests were carried out on at least three sections of each sample.
Images were recorded through a microscope, the relevant parts of the sample observed, and the positive area calculated.

| RT-qPCR
Reverse transcription-quantitative polymerase chain reaction was carried out according to protocol published in previous study (Yang et al., 2017). Total RNA was extracted with TRIzol ® Reagent (Invitrogen) from hippocampal tissues, reverse-transcribed, and amplified by RT-qPCR according to manufacturer instructions. The primers used for qRT-PCR are shown in Table 1. Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) served as the housekeeping gene. Experiments were carried out in triplicate.

| Statistical analyses
Data are the mean ± standard deviation of at least three independent experiments. Differences between groups were calculated using one-way analysis of variance (ANOVA). Statistical analyses were undertaken using SPSS v19.0 (IBM). Differences between experimental groups were considered significant at p < .05, very significant at p < .01, and extremely significant at p < .001.

| PLSO improved the escape latency and dwelling time on platform quadrants in AD rats
We administered the MWMT to evaluate the influence of PLSO on the behavioral memory of AD rats. With increasing training time and time, the escape latency of the model group did not change significantly ( Figure 1). For the AD model group, compared with the control group, from day 2 to day 4, the escape latency was extremely significantly longer (p < .001). Compared with the model group, the positive group (donepezil) had an extremely significantly shorter escape latency from day 2 to day 4 (p < .001). The low-dose PLSO group showed no significant difference compared with that of the model group (p > .05). On the fourth day of training, the middle-dose PLSO group was extremely significantly different from the model group (p < .001). The high-dose PLSO group, compared with the model group, had an extremely significantly shorter escape latency from day 2 to day 4 (p < .001). In the medium-dose PLSO group, compared with the low-dose PLSO group, on day-4, the escape latency was extremely significantly different (p < .001). Compared with the low-dose PLSO group, the high-dose PLSO group had an extremely significantly shorter escape latency (p < .001). Compared with the medium-dose PLSO group, the high-dose PLSO group had an extremely significantly shorter escape latency (p < .001). Compared with the model group, the activity time of the low-dose PLSO group was not significantly different (p < .05); medium-dose group was extremely significantly longer (p < .001); and high-dose group was extremely significantly longer (p < .001). Compared with the low-dose PLSO group, the activity time of the medium-dose group was extremely significantly longer (p < .001); high-dose PLSO group was extremely significantly longer (p < .001). Compared with the medium-dose PLSO group, the activity time of the high-dose group was extremely significantly longer (p < .001).
The results stated above suggest that the escape latency of rats in the high-dose PLSO group (10 g/kg) was shorter, which reflects their better memory. After platform removal, rats were more active in the quadrant where the platform was located, which also reflected their memory. In summary, the high-dose PLSO group (10 g/ kg) could improve the memory of rats significantly and improve AD.

| PLSO reversed the pathologic morphology of hippocampal tissue in AD rats
Next, we undertook pathologic H&E staining on the CA1 area of hippocampal tissue (Figure 3). Figure 3a shows the characteristic However, the neuronal cells in AD rats treated with physiologic saline (Figure 3b) displayed a loose arrangement with reduced numbers, irregular morphology, decreased cell volume from cytoplasm condensation, and karyopyknosis. Interestingly, the high-dose PLSO group (10 g/kg) revealed a significant relieving effect.

| TNF-α
The model group had extremely significantly higher expression than that of the normal group (p < .001

| IL-6
The model group showed higher expression than that of the control group, and the difference was extremely significant (p < .001). The positive group showed lower expression than that of the model group, and the difference was extremely significant (p < .001). The medium-dose PLSO group had lower expression than that of the model group, and the difference was significant (p < .05). The high-dose PLSO group had lower expression than that of the model group, and the difference was very significant (p < .01).
The results stated above suggest that expression of proinflammatory factors was low in the high-dose PLSO group.

| PLSO alleviated oxidant stress of hippocampal tissue in AD rats
The action of proinflammatory factors affects the antioxidant capacity of cells. Biochemical methods were used to measure expression of the oxidation-related moieties MDA, SOD, GSH and ACHE ( Figure 5).

| MDA
Compared with the control group, the model group had higher content, and the difference was extremely significant (p < .001).
Compared with the model group, the positive group, medium-dose PLSO group, and high-dose PLSO group had lower content, and the difference was extremely significant (p < .001).

| SOD
Compared with the control group, the model group had lower content, and the difference was extremely significant (p < .001).
Compared with the model group, the positive group and high-dose PLSO group had higher content, and the difference was extremely significant (p < .001).

| ACHE
Compared with the control group, the model group had higher content, and the difference was extremely significant (p < .001).
Compared with the model group, the positive group and high-dose PLSO group had lower content, and the difference was extremely significant (p < .001).
These results suggested that PLSO could inhibit activation of microglia and astrocytes, inhibit expression of proinflammatory factors, and, thus, affect the antioxidant capacity of cells.

| PLSO-activated glial cells in the hippocampal tissue of AD rats
To study the specific effect of PLSO on AD, we tested microglia and astrocytes in hippocampal tissues. Expression of the microglia marker, IBA1, and astrocyte marker, GAFP, was measured by IHC analyses (Figure 6b).
For GAFP, compared with the control group, the model group had significantly higher expression (p < .05). The high-dose PLSO group (10 g/kg) had significantly lower expression than that in the model group (p < .05).
For IBA1, the model group had significantly higher expression than that of the control group (p < .05). The medium-dose group (2 g/kg) of PLSO had significantly lower expression than that of the model group (p < .05). The high-dose group of PLSO (10 g/kg) had very significantly lower expression than that of the model group (p < .01).

| PLSO alleviated inflammation via inhibition of the nuclear factor-kappa B (NF-κB) signaling pathway
Paeonia ludlowii seed oil mediates expression of proinflammatory factors and then activates the NF-κB signaling pathway, thereby affecting apoptosis and reducing neuronal-cell damage. To further study PLSO, Western blotting was done for p65 (one of the five components of the NF-κB signaling pathway) and p-p65 in the NF-κB signaling pathway (Figure 7).

| PLSO inhibited apoptosis
Expression of proinflammatory factors was low in the high-dose group of PLSO. Compared with the model group, expression of p65 and Bax was lower in the high-dose PLSO group, whereas Bcl-2 expression was higher in the high-dose group of PLSO.
Furthermore, the apoptosis rate was measured (Figure 8).
Compared with the control group, the model group had a higher apoptosis rate, and the difference was very significant (p < .01). The positive group had a lower apoptosis rate than that of the model group, and the difference was very significant (p < .001). Compared with the model group, the apoptosis rate of the high-dose PLSO group was low, and the difference was very significant (p < .01).
Paeonia ludlowii seed oil could inhibit expression of proinflammatory factors, thereby affecting the expression and phosphorylation of p65, reducing Bax content, and increasing the Bcl-2 content. In this way, PLSO reduced damage to neuronal cells and apoptosis.

| D ISCUSS I ON
The drugs used for AD treatment can only relieve the symptoms of patients, but cannot slow the progression or cure AD (Lee et al., 2015). To discover and develop more efficacious and less toxic novel anti-AD drugs or nutrient supplements, we used P. ludlowii seeds as raw materials to obtain PLSO through cold pressing. We established an AD model in rats to evaluate the anti-AD efficacy of PLSO and compared it with that of donepezil (a classic drug for AD treatment).
We demonstrated that PLSO improved cognitive function significantly in AD rats by shortening the escape latency (Figure 1) F I G U R E 7 Effect of PLSO on the protein expression of proinflammatory factors and p65 and p-p65 and increasing the dwelling time on platform quadrants ( Figure 2) compared with that of AD rats treated with physiologic saline.
To study the specific effect of PLSO on AD, we used IHC detection of a microglia marker (IBA1) and astrocyte marker (GAFP) in the hippocampus ( Figure 6). Next, we conducted RT-qPCR Increasing evidence from basic research in animal models of AD emphasizes that the key factor of neuronal injury/loss is due to neuronal apoptosis (Crews et al., 2010). Apoptosis plays a critical part in the development of the nervous system and in many chronic neurodegenerative diseases (including AD).
The ratio of pro-apoptosis proteins to anti-apoptosis proteins directly determines the degree of opening of various channels in the mitochondrial outer membrane and regulation of apoptosis (Danial et al., 2003).
Bcl-2 and Bax proteins have crucial roles in signal transduction during apoptosis. We showed that neuronal apoptosis plays a key part in AD pathogenesis and that PLSO inhibited neuronal apoptosis significantly (Figure 8b). mRNA expression in rat neuronal tissue and protein expression of Bax and Bcl-2 in PC12 cells was also analyzed. We found that PLSO could be even more efficacious against AD than donepezil in our AD model.
Liu and colleagues reported that the seed coat extract of peony could improve cognition defects, and they determined its active constituents (Liu et al., 2020). However, it is not known which type of drug component/molecule type caused the protective effect of PLSO upon rat neurons, or the related cell experiments, toxicity testing, and concentration gradients that need to be investigated further. However, the effects of PLSO on expression of Bax and Bcl-2 suggest that the anti-AD effect of PLSO may also be related to mitochondrial permeability, cytC, and the phosphoinositide 3-kinase/ protein kinase B signaling pathway. Therefore, further studies are needed to ascertain the mechanistic action of PLSO on the associated molecular pathways of neuronal apoptosis in the neurodegeneration seen in AD.
Although this new finding is promising, PLSO cannot be used as AD treatment. Animal testing is the first step in demonstrating its therapeutic potential. Human diseases are complex, and it is uncertain at which dose PLSO can be employed safely and efficaciously.
This study provides a scientific basis for the research and development of AD food products, which is helpful to reduce the use of a large number of AD drugs, promote the physiological health of Chinese people, improve people's quality of life, and reduce the medical burden related to AD to a certain extent.

| CON CLUS IONS
We demonstrated that PLSO improved cognitive function significantly by inhibiting apoptosis of hippocampal neuronal cells. Highdose PLSO could inhibit expression of proinflammatory cells by inhibiting activation of glial cells. Secretion of inflammatory factors causes production of free radicals and other substances, which leads to a reduction in the antioxidant capacity of cells, and PLSO could enhance the antioxidant capacity of cells.
Paeonia ludlowii seed oil was even more efficacious against AD than donepezil in our AD system. Furthermore, PLSO could downregulate expression of Bax protein and, simultaneously, upregulate expression of Bcl-2 protein, by inhibiting activation of the NF-κB pathway. Therefore, PLSO could be developed as a novel drug or nutrient supplement for the prevention and/or treatment of AD.

CO N FLI C T O F I NTE R E S T
The authors declare no conflicts of interest.