Protective Effects of 2-Dodecyl-6-Methoxycyclohexa-2,5 -Diene-1,4-Dione Isolated from Averrhoa Carambola L. (Oxalidaceae) Roots on Neuron Apoptosis and Memory Deficits in Alzheimer's Disease

Alzheimer's disease (AD) is the most common human neurodegenerative disorder characterized by the progressive deterioration of cognition and memory [1-3]. Its etiology entails oxidative stress, deposition of extracellular amyloid beta (Ab) plaques, formation of intracellular neurofibrillary tangles (NFTs), metal mediated neurotoxicity, mutations in genes, neuro inflammation, hyperphosphorylation of tau and apoptosis [4-8].

Apoptosis, a biological process that plays an essential role in normal physiology to oust a few cells and contribute to the normal growth, when impaired or influenced by various factors such as Bcl2, Bax, caspases, amyloid beta, tumor necrosis factor-a, amyloid precursor protein intracellular C-terminal domain, reactive oxygen species, perturbation of enzymes leads to deleterious neurodegenerative disorders like AD [9, 10].

Averrhoa carambola L. is a perennial herb widely distributed in Southeast Asia. Its roots have been employed in Traditional Chinese Medicine (TCM) for thousands of years. Averrhoa carambola L. is commonly used to treat headaches, vomiting, coughing and hangovers [11]. Furthermore, it is used as an appetite stimulant, a diuretic, and as an antidiarrheal and febrifugal agent [12]. Our previous study demonstrated that the roots of Averrhoa carambola L. have beneficial anti-hyperglycemic effects [13], and the effective component (DMDD) that was extracted from the roots of Averrhoa carambola L. also has therapeutic potential for the treatment of diabetic nephropathy and inhibitory effects on the kidneys of diabetic mice [11].

In this study, we aim to investigate and evaluate the potential beneficial effects of DMDD on neuron apoptosis and memory deficits in APP/PS1 transgenic Alzheimer's disease mice in vivo and the mechanism of protection in vitro.

The chemical structure of DMDD is shown in Fig. 1. DMDD was isolated and purified from extraction of Averrhoa carambola L in 60% aq. EtOH using silica gel column chromatography as described previously [14]. DMDD was dissolved in distilled water (2.5 mg/L) before being administered to the mice.

APPswePSEN1dE9 transgenic mice (terms in AD, certificate no. SCXK 2015-0001) and nontransgenic littermate controls (terms in WT) were obtained from the Model Animal Research Centre of Nanjing University (Nanjing, China). All experiments were done with approval of the Institutional Animal Care and Use Committee of Guangxi Medical University. Mice had access to food and water ad libitum and were kept on a 12: 12 h light–dark cycle, 23 ± 3 °C of temperature and 50 ± 2 % humidity. All animals were housed in SPF-grade animal facility in Guangxi Medical University for 6 months prior to drug treatment, Morris water maze, and Y maze assessments.

Six-month-old male APP/PS1 transgenic AD mice were randomly divided into five groups and oral-gavaged one time for 21 days: AD model group (0.9% NaCl), AD-Huperzine A groups (0.03 mg/kg/d), and three different dose of AD-DMDD group (12.5, 25, and 50 mg/kg/d) (n=10 each). Age-matched WT male mice (WT group) were also oral-gavaged for 21 days with same volume of 0.9% NaCl.

All mice were blinded to the experimenter with respect to treatment status and genotype. The tasks were conducted in a large pool (diameter 100 cm) filled with 22 ± 1 °C water. Water was made opaque using non-toxic white paint. A hidden square platform (10 cm2) was placed 1 cm beneath the surface of the water in one quadrant of the pool. Visual cues were mounted on a screen surrounding the pool in fixed positions. The mice were pre-trained for 1 day to find and climb on to the hidden platform within 100 s of being placed in the water. If a mouse was unable to find the platform, it was placed there manually for 10 s. Mice were given up to three times to find the platform. Mice that were unable to find the platform during these training sessions were eliminated from the experiment. Mice were then tested for 5 successive days, with four 100-second trials per day, and escape time to platform (escape latency), swim speed and distance swum were recorded. Following the last test, the probe trial was carried out to test the spatial memory by allowing all animals to freely swim for 100 s in the maze with the platform removed. The time and times of each mouse spent on or cross the target area which previously contains an escape platform was recorded for spatial memory evaluation.

A Y-type electric maze was used to test learning and memory in the mice. Three arms were randomly designated as; ‘start’ arm (with electric stimulus only), ‘safety’ arm (with signal lamp only) and ‘other’ arm (the same as start arm), and the role of three arms during the detection process alternates randomly, dependent on animals staying arm. The animals were monitored and the rate of learning to escape to the safety arm of the maze was determined for each animal. Direct escape from the start arm to the safety arm was considered to be the correct reaction upon electrical stimulus. Each animal was put in the Y-maze apparatus to adapt the condition for 5 min and trained before formal testing. Animals were considered to have learned the maze with 7/8 correct choices during continuous training. After 24 h, the animals were put in the Y-type electric maze again to test active avoidance learning and memory ability, which was expressed as the correct rate. The correct rate was calculated as the ratio of the number of correct reactions to total number of electrical stimulations multiplied by 100 as shown in the following equation: correct rate (%)=(the number of correct reaction/total number of electrical stimulation) × 100.

After the behavioral tests were completed, mice were decapitated under halothane anesthesia. One hemibrain of each mouse was fixed in 4% paraformaldehyde solution in phosphatebuffered saline (PBS) and cryoprotected in 30% sucrose solution in PBS containing 0.01% sodium azide. The slices were stained with H&E (hematoxylin and eosin) and observed using a light microscope at 400X. The total numbers of neurons in DG, CA1, and CA3 region were counted according previous protocol [15].

Ultrastructure changes of cells were observed using a transmission electron microscope (Hitachi, Tokyo, Japan). Transmission electron microscopy analysis was conducted as previously described [16].

PC-12 cells, purchased from the Committee on Type Culture Collection of Chinese Academy of Sciences (CTCCCAS, Shanghai, China), were maintained in Dulbecco’s Modified Eagle Medium (DMEM) (GIBCO, New York, USA) supplemented with 10% fetal bovine serum (FBS) (GIBCO, New York, USA) and 1% penicillin– streptomycin (Sigma-Aldrich, St. Louis, Missouri, USA) in a humidified incubator at 37 °C with 5% CO₂.

The cytotoxicity of DMDD was first measured by using a Cell Counting Kit-8 (CCK8) (Dojindo Laboratory, Kumamoto, Japan). Briefly, PC-12 cells were seeded into a 96-well plate overnight and treated with concentrations (0 μmol/l, 5 μmol/l, 10 μmol/l, 20 μmol/l, 40 μmol/l, and 60 μmol/l) of DMDD or (0 μmol/l, 5 μmol/l, 10 μmol/l, 20 μmol/l, 40 μmol/l, and 60 μmol/l) of Aβ1-42. After incubation for 24 h, cell survival was measured by using a CCK8 as described previously [17].

The apoptosis of cells was measured by using the Annexin V-FITC/ PI Kit (Life Technologies, Carlsbad, CA, USA), and analyzed on a FACStar-Plus flow cytometer (Becton Dickinson, Franklin Lakes, NJ, USA) as previously described [18].

Protein isolation and western blot was conducted as previously described [19]. Proteins were detected with primary antibodies to CleavedCaspase-3, -9, Bid, Bim, Bax, and Bcl-2 (all from Cell Signaling Technology, Boston, MA, USA).

Total RNA was extracted by using Trizol (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s instructions. Complementary DNA (cDNA) was synthesized using the PrimeScript RT reagent kit with gDNA Eraser. RT-PCR was conducted using a 7300 real-time PCR detection system (Applied Biosystems, Foster City, CA, USA). cDNA samples (equivalent to 1 μg of total RNA) were used as templates with the corresponding gene primers (Table 1). The real-time PCR results were calculated by the 2-△△Ct method and normalized to β-actin with arbitrary units [20].

All the experimental data were analyzed by SPSS 16.0 software (SPSS lnc., USA) and are presented as the mean ± SD. Statistical significance was assessed using one-way analysis of variance (ANOVA). Multiple comparisons between the groups were performed using S-N-K method. The data are presented as the mean ± SD. *P ≤ 0.05; **P ≤ 0.01; ***P ≤ 0.001.

To determine the protective effect of DMDD on spatial learning and memory deficit in APP/PS1 mice, Morris water maze (MWM) test was performed. The latency of finding the hidden platform was significantly increased in APP/PS1 (AD) mice compared with the wild type control (WT) mice (Fig. 2A). Compared to the untreated AD mice, the administration of DMDD at low- middle- and high doses significantly shortened the latency of finding the hidden platform (Fig. 2A). Probe trials were conducted to assess the memory retention in the last training session. APP/PS1 mice showed less numbers of crossings in the target quadrant (where the platform had been located) and spent less time on the target area which previously contains an escape platform compared to the WT group (Fig. 2B and C). The numbers of crossings where the platform located of DMDD-treated APP/PS1 mice were more than those of the untreated APP/PS1 mice (Fig. 2B). DMDD-treated APP/PS1 mice spent more time on the target area which previously contains an escape platform relative to untreated APP/PS1 mice (Fig. 2C). Fear memory was evaluated by Y-type electric maze tests. As illustrated in Fig. 1D, DMDD treatment greatly increased the correct rate of escaping electrical stimulation in APP/PS1 mice compared with untreated APP/PS1 mice. Taken together, these data indicated that DMDD treatment significantly improved the learning and memory deficit in APP/PS1 mice.

As shown in Fig. 3(A), pyramidal neurons in hippocampal area CA1, CA3 and DG region of APP/PS1 mice were occasionally pyknotic and/ or exhibited clumps of condensed chromatin and irregular membrane structure. The total number of neurons in CA1, CA3 and DG regions of APP/PS1 mice was significantly decreased compared with WT mice (Fig. 3B). DMDD treatment greatly increased the total number of neurons in CA1, CA3 and DG regions of APP/PS1 mice, although still couldn’t fully reverse the loss of neurons in CA3 region (Fig. 3B). As shown in Fig. 3 (C), pyramidal neurons in hippocampal area displayed cell shrinkage, irregular nuclear outline, condensed chromosomes, a loose endoplasmic reticulum, morphologically abnormal mitochondrial structures, and cytoplasm vacuolization. The ultrastructural changes of the DMDD-treated group cells appeared to be improved compared with untreated group.

As shown in Fig. 4 (A), DMDD showed a non-toxic range at concentrations of 0–20 μmol/l for 24 h. As shown in Fig. 4 (B), cell viability was significantly lower in cells only treated with Aβ1-42. Therefore, we used the low–, medium–, and high–concentrations of DMDD (5 μmol/l, 10 μmol/l, and 20 μmol/l) and 40 μmol/l of Aβ1-42 for the following studies in cultured PC-12 cells. The representative images for AnnexinV-FITC/PI flow cytometry analysis were presented in Fig. 4 (C). As shown in Fig. 4 (D), the rate of apoptosis was significantly higher in cells only exposed to Aβ1-42, and this increase was prevented by pretreating cells with DMDD. As shown in Fig. 4 (E, F and G), the expressions of Cleaved-Caspase-3, and -9 proteins were up-regulated in cells exposed to Aβ1-42. However, the expressions of these proteins were down-regulated in cells pretreated with DMDD.

To further explore the molecular mechanisms underlying DMDD prevents apoptosis in PC-12 cells exposed to Aβ1-42, we first checked the change of mitochondrial membrane potential by JC1 staining. As shown in Fig. 5 (A and C), Aβ1-42 treatment resulted in reduction of the mitochondria membrane potential, DMDD significantly reversed the reduction. Additionally, we analyzed the mitochondria pathway apoptosis-relevant proteins Bim, Bid, Bax, and Bcl-2. As shown in Fig. 5 (B, D, and E), the expressions of Bax mRNA and protein were up-regulated whereas the expressions of Bcl-2 mRNA and protein were down-regulated in Aβ1-42 group cells. However, the expressions of Bax mRNA and protein were down-regulated whereas the expressions of Bcl-2 mRNA and protein were up-regulated in cells pretreated with DMDD. The Bcl-2/Bax ratio was significantly lower in cells only exposed to Aβ1-42, and this decrease was prevented by pretreating cells with DMDD. In the meantime, although protein level of Bim was up-regulated in cells exposed to Aβ1-42, pretreatment of DMDD only slightly reverse this induction (Fig. 5B). Bid remained unchanged throughout all Aβ1-42 alone or combination treatment with DMDD (Fig. 5B).

As shown in Fig. 6 (A, B and C), the mRNA and protein level of Bcl-2 was significantly decreased in APP/ PS1 mice compared with littermate WT mice, whereas Bax expression was induced. Similar with in vitro cell based result, DMDD treatment greatly reversed the reduction and induction of Bcl-2 or Bax at both mRNA and protein level, as well as the protein ration of Bcl-2/ Bax Fig. 6 (D).

In spite of major research efforts, current therapies, including the preferred drug Aricept and other cholinesterase inhibitors, can only partially control the symptoms or delay the development of AD, and adverse effects and toxicity often occur, leading to limited clinical application. Without novel therapeutic improvements, it is estimated that the number of affected people worldwide will rise to 100 million by 2050 [21]. Therefore, broad clinical application prospects for exploring new drug targets and developing better efficacy with safer and tolerable drug candidates for the treatment of AD. In our study we used the aged APP1/PS1 mice to evaluate if the therapeutic effects of DMDD on learning and memory deficits at the first time. Similar to human subjects with AD, these mice demonstrate deficits in spatial learning and memory and fear memory, as evidence by increased latency of finding the hidden platform assessed on Morris water maze or decreased percentage of correct rate of escape from electrical stimulation in Y-type electric maze tests [22, 23]. Surprisedly, our results showed that DMDD treatment restored learning and memory deficits in APP/PS1 mice in all behavior tests of Morris water maze and Y-type electric maze. In short, the present study clearly demonstrated that effective component (DMDD) in roots of Averrhoa carambola L. (Oxalidaceae) had protective effects on APP/PS1 transgenic AD mice.

It is now generally accepted that massive neuronal death due to apoptosis is a common characteristic in the brains of patients suffering from neurodegenerative diseases, and apoptotic cell death has been found in neurons and glial cells in AD [24]. Consistent with previous report [25], loss and apoptosis of pyramidal neurons in hippocampal area CA1, CA3 and DG region of APP/PS1 mice was confirmed in our study by H&E and transmission electron microscopy analysis. Available data presents some abnormalities in the metabolism of amyloid precursor protein (APP) as a causative factor, which can results in mitochondrial dysfunction and eventually cell apoptosis. Whenever there is an overexpression of APP, its metabolite, Aβ peptide, will overload [26]. Indeed, treating the homo sapiens bone marrow neuroblast PC-12 cells with Aβ1-42 in vitro, clearly resulted in activation of Caspase -9 and -3 but Caspase-8 (data not shown), as well as loss of mitochondria membrane potential, which is the hallmark of intrinsic apoptosis [20]. In the meantime, DMDD significantly attenuated the pro-apoptotic events and prevented Aβ1-42 induced apoptosis in PC-12 cells. With the accumulating evidences, our study supported the Aβ induced neuronal apoptosis can be a potential therapeutic target for AD treatment.

The Bcl-2 family proteins are the central regulators of mitochondria-mediated apoptosis [27-29]. The BH3-only family members are proximal signaling molecules that respond to distinct as well as overlapping signals, and consist of at least 10 members. Several of them, such as Bid and Bim, are potent inducers of apoptosis by activating Bax/Bak following the neutralization of all five known anti-apoptotic Bcl-2 family members [27-29], and promote mitochondrial dysfunction and caspase activation [30-32]. Our work indicated that DMDD affected the intrinsic apoptotic pathway by suppressing Cleaved- Caspase-3 and Cleaved-Caspase-9, and increasing the Bcl-2/Bax ratio. This result is highly consistent with our more recent work showing DMDD protected Min6 cells against PA-induced dysfunction by attenuating apoptosis by reversing the Bcl-2/Bax ratio [33], and suggested that regulating Bcl-2 family protein is a common mechanism to DMDD or similar structure chemicals.

We provide evidence that DMDD has potential benefit on treating learning and memory deficit in APP/PS1 transgenic AD mice, and its effects may be associated with reversing the apoptosis of neuron via inhibiting Bax/Bcl-2 mediated mitochondrial membrane potential loss. DMDD may be a promising candidate for the treatment of Alzheimer's disease.

DMDD (2-dodecyl-6-methoxycycyclohexa-2, 5-1, 4-dione); Ab (amyloid beta); NFTs (neurofibrillary tangles); TCM (Traditional Chinese Medicine); PBS (phosphatebuffered saline); H&E (hematoxylin and eosin); DMEM (Dulbecco’s Modified Eagle Medium); FBS (fetal bovine serum); CCK8 (Cell Counting Kit-8); cDNA (Complementary DNA); MWM (Morris water maze); APP (amyloid precursor protein).

This work was supported by the National Natural Science Foundation of China (No. 81460205, 81360129, 81760665, and 81160533), the State Key Laboratory Cultivation Base for the Chemistry and Molecular Engineering of Medicinal Resources of the Ministry of Science and Technology of China (CMEMR2014-B01), and the Postdoctoral Science Foundation of China (No.2017M613271XB), Nanning Science and Technology Research and Production of new products (No.201102084C). Xiaojie Wei and Xiaohui Xu are major contributors in performing experiments, data analysis and manuscript preparation. Zhenfeng Chen, Tao Liang, Qingwei Wen, Ni Qin, Wansu Huang, Xiang Huang, Yuchun Li, Juman Li, Junhui He performed the research. Jinbin Wei and Renbin Huang conceived or designed the study. All authors have read and approved the final manuscript.

The authors declare that there is no conflict of interests.

X. Wei and X. Xu contributed equally to this work.