Protective Effect of Bajijiasu Against β-Amyloid-Induced Neurotoxicity in PC12 Cells

Abstract

Beta-amyloid peptide (Aβ), a major protein component of senile plaques associated with Alzheimer’s disease (AD), is also directly neurotoxic. Mitigation of Aβ-induced neurotoxicity is thus a possible therapeutic approach to delay or prevent onset and progression of AD. This study evaluated the protective effect of Bajijiasu (β- d-fructofuranosyl (2–2) β- d-fructofuranosyl), a dimeric fructose isolated from the Chinese herb Radix Morinda officinalis, on Aβ-induced neurotoxicity in pheochromocytoma (PC12) cells. Bajijiasu alone had no endogenous neurotoxicity up to 200 μM. Brief pretreatment with 10–40 μM Bajijiasu (2 h) significantly reversed the reduction in cell viability induced by subsequent 24 h exposure to Aβ_25–35 (21 μM) as measured by MTT and LDH assays, and reduced Aβ_25–35-induced apoptosis as indicated by reduced annexin V-EGFP staining. Bajijiasu also decreased the accumulation of intracellular reactive oxygen species and the lipid peroxidation product malondialdehyde in PC12 cells, upregulated expression of glutathione reductase and superoxide dismutase, prevented depolarization of the mitochondrial membrane potential (Ψm), and blocked Aβ_25–35-induced increases in [Ca²⁺]_i. Furthermore, Bajijiasu reversed Aβ_25–35-induced changes in the expression levels of p21, CDK4, E2F1, Bax, NF-κB p65, and caspase-3. Bajijiasu is neuroprotective against Aβ_25–35-induced neurotoxicity in PC12 cells, likely by protecting against oxidative stress and ensuing apoptosis.

Introduction

Alzheimer’s disease (AD), the most common form of dementia in the elderly, is characterized by the progressive deterioration of learning, memory, and other cognitive functions. Senile plaques, neurofibrillary tangles, and extensive neuronal loss are the main histological hallmarks of the AD brain (Katzman and Saitoh 1991). Beta-amyloid peptide (Aβ), the major component of senile plaques, is neurotoxic and thus considered integral to the development and progression of AD (Hardy 1997; Selkoe 2000). The neurological deficits associated with AD are largely irreversible, and current therapies can only improve specific symptom but cannot halt disease progression. The incidence of AD is expected to increase in developed countries with aging populations. Beta-amyloid may trigger neurodegeneration by inducing oxidative stress, and indeed excessive reactive oxygen species (ROS) production and signs of oxidative stress are found in the AD brain (Crack and Taylor 2005; Li et al. 2008a, b; Zhang et al. 2008; Hu et al. 2010). Moreover, antioxidants and free radical scavengers have been shown to protect against Aβ-induced neurotoxicity both in vitro and in vivo (Heo et al. 2004). Therefore, therapeutic intervention with antioxidants may help prevent Aβ-induced neurotoxicity and improve neurological outcome in AD.

The β-amyloid fragment peptide 25–35 (Aβ_25–35) is the cytotoxic sequence of the parent peptide Aβ_1–39/42 (Behl et al. 1992; Yankner et al. 1990) and exhibits the same early neurotropic and late neurotoxic activities as Aβ (Iversen et al. 1995). The Aβ_25–35 fragment includes the 25–28 hydrophilic domain forming a putative β-turn (Bond et al. 2003; Laczkó et al. 1994) and the 29–35 hydrophilic domain (Pike et al. 1995). It has been shown using X-ray diffraction that Aβ_25–35 also adopts the cross-β motif (Bond et al. 2003), probably due to residues 29–35. In vivo and in vitro studies have shown that Aβ_25–35 can cause neuronal cell damage and disrupt both learning and memory (Young et al. 2013; Butterfield et al. 2011; Kaminsky et al. 2010; Tian et al. 2012; Diaz et al. 2012).

Currently, several types of drugs are used to maintain cognitive function in AD patients, including acetylcholinesterase inhibitors (AChEIs) (Cracon et al. 1998; Borroni et al. 2001; Wolff et al. 2005; Jay 2005), glutamate modulators (Knopman 2006), vasodilating agents, and the putative nootropic racetam drugs (Winnicka et al. 2005; Son et al. 2004). In addition, several alternative approaches, including anti-inflammatory agents, antioxidants, estrogens, and anti-Aβ-peptide agents have been used with varying degrees of success to ameliorate the symptoms of AD (Yamada and Nabeshima 2000; Krishnan et al. 2003; Wolff et al. 2005; Cummings 2001; Tariot et al. 2004; Wolfe 2001; Miller et al. 2003; Gracon et al. 1998).

However, there is currently no safe and broadly effective treatment for delaying the progression of AD. During the past decade, the benefits of many herbs used in traditional medicine have been demonstrated in different AD-related experimental models as well as in clinical trials (Baum et al. 2008; Huang et al. 2008). Morinda officinalis (Indian Mulberry or Ba Ji Tian) root is a common medicinal herb in traditional Asian medicine used widely in China, Korea, and Japan, particularly in southern China. The raw root contains a number of potentially active compounds, including hexasaccharides and heptasaccharides that have been shown to ameliorate symptoms in an animal model of depression (Li et al. 2001; Li et al. 2004, 2008a, b; Deng et al. 2012). Previous studies (appearing mainly in Chinese) have demonstrated that Bajijiasu, a dimeric fructose isolated from M. officinalis (the chemical structure shown in Fig. 1, previous name: Bajisu), is able to enhance population spikes (PSs) and long-term potentiation (LTP) of the excitatory postsynaptic potential following high-frequency synaptic stimulation, ameliorate cognitive deficits induced by D-galactose in mice, and protect against ischemic neuronal damage. The aims of the present study were to investigate whether Bajijiasu has protective effects against Aβ-induced neurotoxicity in PC12 cells and to explore the underlying molecular mechanisms.

Fig. 1.

Chemical structure of Bajijiasu

Materials and Methods

Drugs and Reagents

Bajijiasu (purity ≥98 %, isolated from the Chinese herb M. officinalis How; Chinese patent: ZL03139998.3, 201010224513.2), β-amyloid peptide (Aβ_25–35), 3-(4, 5-dimethylthiazol-2-yl)-2, 5-dipheny-ltetrazolium bromide (MTT), and rhodamine 123 were purchased from Sigma-Aldrich (St. Louis, MO, USA). Trizol reagent and 2′,7′-dichlo-rofluorescin diacetate (DCFH-DA) were obtained from Invitrogen (Carlsbad, CA, USA) and Fura-2/AM from Biotium (Hayward, CA, USA). Dulbecco’s modified Eagle Medium (DMEM), fetal bovine serum (FBS), penicillin, and streptomycin were purchased from Gibco (Grand Island, NY, USA). Annexin V-EGFP was purchased from Nanjing Bibo Bioengineering Insititute (China). Assay kits for lactate dehydrogenase (LDH), malondialdehyde (MDA), superoxide dismutase (SOD), and glutathione reductase (GSH-Px) were purchased from Nanjing JianCheng Bioengineering Institute (China). All other reagents and chemicals used in the study were of analytical grade.

Preparation of Aggregated Aβ_25–35

The Aβ_25–35 peptide was dissolved in deionized distilled water at 1 mM and incubated for 4 days at 37 °C to induce aggregation (Li et al. 2008a, b; Labbé et al. 2013). After aggregation, the solution was stored at −20 °C until use. Raman spectroscopy was used to determine protein state after incubation (free or aggregated) and the thioflavin T (ThT) assay to monitor aggregation (fibril formation) over time. The Raman spectrum assay was performed as described previously (Labbé et al. 2013; Heldt et al. 2011) using the Via+Plus laser Micro-Raman spectroscopy system (Renishaw, UK). In brief, aggregated Aβ_25–35 was placed on a silicon slice for Raman scanning between 900 and 2,000 cm⁻¹ (instrument resolution 1 cm⁻¹). All the Raman spectra were recorded over 10 s and multiple samples were scanned 5 times each. Laser excitation at 785 nm and 20 % laser power were used. The total energy was much lower than the safe limit of exposure for Aβ_25–35. All spectral data were collected under the same conditions and the instrument was calibrated using silicon at 520 cm⁻¹.

The ThT assay was performed as described previously (Hudson et al. 2009). In brief, 10 μL of 1 mM Aβ_25–35 (incubated for different times) was added to 980 μL of 3 μM ThT (Sigma-Aldrich) dissolved in 2 mM Tris–HCl buffer (pH 7.4, with 150 mM NaCl), and the fluorescence emission (Em) intensity at 440–520 nm from 450 nm excitation (Ex) measured using a fluorescence microplate reader.

Cell Culture and Treatment

Pheochromocytoma (PC12) cells were obtained from the American Type Culture Collection (Rockville, MD, USA). Cells were seeded in 25 cm² flasks and maintained in DMEM supplemented with penicillin (100 U/mL), streptomycin (100 U/mL), and 5 % FBS at 37 °C in a humidified atmosphere of 95 % air and 5 % CO₂. Once flask cultures reached 80 % confluence, the cells were subcultured (at a density dependent on the assay to be performed) at 37 °C for 24 h in free-serum DMEM, and then incubated with different concentrations of Bajijiasu (usually 10, 20, or 40 μM) for 2 h. Then, Aβ_25–35 at 21 μM was added to the culture for an additional 24 h, followed by the assays described below. All the cells used in this study were undifferentiated.

Cell Viability Assay

Cell viability following treatment with Bajijiasu, Aβ_25–35, or both was measured by quantitative colorimetric MTT and LDH assays. The MTT method was described previously (Mao et al. 2011). In brief, cells were seeded onto 96-well culture plates at 2 × 10⁴ cells/well in serum-free DMEM. After drug treatment, 20 μL/well MTT solution (final concentration, 1 mg/mL) was added, and the cells were incubated at 37 °C for 4 h. The media was aspirated off, and the formazan crystals produced from MTT by viable cells were dissolved in 150 μL of DMSO. The optical density of each well was determined at 570 nm using a microplate reader (RT-2100C, USA). Cell viability was expressed as a percentage of untreated controls.

For the LDH assay, 100 μL incubation medium was collected from each well and added to an LDH assay solution. Lactate dehydrogenase activity was measured spectrophotometrically according to the manufacturer’s instructions (Nanjing Jiancheng Bioengineering Institute; Nanjing, Jiangsu, China).

Assessment of Apoptosis

Annexin V-EGFP plus propidium iodide double staining was used to detect apoptosis and necrosis following drug treatment. During the early stages of apoptosis, membrane phosphatidylserine (PS) is translocated from the inner lipid layer of the plasma membrane to the outer layer in many cell types, including PC12 cells. Once on the cell surface, PS can be easily detected by staining with Annexin, a protein with a strong affinity to PS. Annexin was conjugated to the highly photostable green fluorescent protein (EGFP). Late apoptosis was measured by conventional propidium iodide (PI) staining. The assay can be directly performed on live cells and the relative number of early and late apoptotic cells measured by flow cytometry. Following drug treatment, cells (2 × 10⁵ cells/plate) were washed with PBS, harvested, centrifuged at 2,000×g for 5 min, and resuspended into 1 mL buffer (10 mM HEPES/NaOH, pH 7.4, 140 mM NaCl, 2.5 mM CaCl₂). An aliquot (190 μL) of the cell suspension was mixed with 5 μL Annexin V-EGFP and 10 μL PI (20 μL/mL), and incubated for 10 min at room temperature under darkness. Staining was measured in a flow cytometer (CytomicsTM FC 500, Beckman Coulter, USA) at Ex = 488 nm and Em = 530 nm. The level of apoptosis was expressed as the percentage of control cultures incubated with PI + Annexin V-EGFP, but not treated with Bajijiasu or Aβ_25–35.

Measurement of Intracellular ROS Production

Intracellular ROS were measured using the redox-sensitive fluorescent dye 2′,7′-dichlorofluorescein diacetate (DCFH-DA). Conversion of non-fluorescent DCFH-DA to fluorescent dichlorofluorescein (DCF) in the presence of ROS was measured on a microplate reader. In brief, following drug treatment, cells were washed twice with D-Hanks solution, incubated with 10 μM DCFH-DA for 30 min at 37 °C in the dark, and washed twice with D-Hanks solution to remove the extracellular DCFH-DA. Fluorescence emission intensity of DCF (538 nm) was measured in response to 485 nm excitation on a fluorescence microplate reader (Corona MTP-601F, Japan). The level of intracellular ROS was expressed as a percentage of control cultures incubated in DCFH-DA, but not treated with Bajijiasu or Aβ_25–35.

Measurement of Mitochondrial Membrane Potential

The mitochondrial membrane potential (Ψm) was measured using rhodamine 123. Rhodamine 123 can enter the mitochondrial matrix and cause photoluminescent quenching that is dependent on mitochondrial transmembrane potential. After drug treatment, PC12 cells were incubated with 5 mg/L rhodamine 123 for 30 min at 37 °C in the dark and washed three times with PBS. Fluorescence emission intensity measured on a microplate reader (Ex = 488 nm, Em = 510 nm) and expressed as the percentage of control cells incubated in rhodamine 123, but not treated with Bajijiasu or Aβ_25–35.

Measurement of Intracellular Free Calcium Concentration ([Ca²⁺]_i)

The [Ca²⁺]_i was measured using Fura-2/AM and quantified by the formula (1)

$${[{\text{Ca}}^{2+}]}_i=K_{\text{d}}(F_0-F_{\text{min}})/\phantom{K_{\text{d}}(F_0-F_{\text{min}})(F_{\text{min}}-F_0)}(F_{\text{min}}-F_0)$$ 1

where F ₀ is the fluorescence emission intensity for samples, F _max is the fluorescence emission intensity at the Ca²⁺ concentration of 1 mM with 0.09 % TritonX-100. F _min is at EGTA concentration of thrice Ca²⁺, and the K _d is 224 nM.

Malondialdehyde (MDA), Superoxide Dismutase (SOD), and Glutathione Peroxidase (GSH-Px) Assays

The PC12 cells were seeded in 25 cm² culture flasks. Once the cells reached 80 % confluence, Bajijiasu (10, 20, and 40 μM) was added for 2 h, followed by incubation in 21 μM Aβ_25–35 for 24 h. At the end of drug treatment, cells were washed with D-Hanks solution, scraped from the plates into 1 mL ice-cold PBS (0.1 M, containing 0.05 mM EDTA), and homogenized. The homogenate was centrifuged at 4,000×g for 10 min at 4 °C. The supernatants were stored at −80 °C until analyses. The protein concentration in each supernatant sample was determined using the Coomassie Brilliant Blue G250 assay. The SOD and GSH-Px activities, level of MDA, and protein content were all determined using specific detection kits (Nanjing Jiancheng Bioengineering Institute) according to the manufacture’s instructions. Concentrations were normalized to the sample protein concentration expressed as a percentage of untreated control cultures.

Real-Time PCR

Total RNA was isolated by TRIZOL Reagent (Invitrogen) and reverse transcribed into cDNA using RT-PCR kits (Invitrogen) according to the manufacturer’s instructions. The synthesized cDNA was amplified by quantitative real-time PCR on a ABI Prism 9700 system (Applied Biosystems, Singapore) using SYBR Green Real-time PCR Master Mix Reagent (Toyoba, Japan). Expected RT-PCR product sizes and the primers used in this study are presented in Table 1. Samples were inactivated for 2 min at 94 °C prior to hot-start amplification. The amplification cycle was 94 °C for 1 min, 28 cycles of 58 °C for 45 s, and 72 °C for 45 s, and a final extension for 10 min at 72 °C. Data from the reactions were collected and analyzed by the complementary computer software. Relative gene expression was calculated using the 2^−ΔΔCt method and normalized to GAPDH expression in each sample.

Table 1.

Primers used for quantitative real-time polymerase chain reaction analysis of GAPDH, p21, CDK4, E2F1, Bax, NF-κB p65, and caspase-3 transcript levels

cDNA/product sizes	Sequence (5′–3′)
NF-κB/115 bp	Forward primers: TGAATGCGGTCCGCCTCTGC
	Reverse primers: GCAGTGTTGGGGGCACGGTT
CDK4/492 bp	Forward primers: CGGGCCCAGATAAAGGGCCACC
	Reverse primers: TGGCTTCAGGTCCCGGTGAACA
p21/80 bp	Forward primers: TTGCGATGCGCTCATGGCGA
	Reverse primers: CCAGTGGCGTCTCAGTGGCG
caspase-3/280 bp	Forward primers: GGT ATT GAG ACA GAC AGTGG
	Reverse primers: CAT GGG ATC TGT TTC TTT GC
E2F1/417 bp	Forward primers: GTGCAGAAACGACGCATCTA
	Reverse primers: CTCAGGGC ACAGGAAAACAT
Bax/166 bp	Forward primers: TGGTTGCCCTTTTCTACTTTG
	Reverse primers: GAAGTAGGAA AGGAGGCCATC
GAPDH/517 bp	Forward primers: GGTGAAGGTCGGTGTCAACG
	Reverse primers: GAGCCCTTCCA CGATGCCAA

Western Blotting

The PC12 cells were seeded onto 6-well culture plates at 5 × 10⁶ cells/well. Following drug treatment, cells were washed twice with D-Hanks solution, harvested, and lysed in lysis buffer. The protein concentration in each lysate sample was determined using the Coomassie Brilliant Blue G250 assay kit. Protein samples were separated on SDS-PAGE gels for 1.5 h at 90 V and transferred to PVD membranes using a transblotting apparatus (Bio-Rad Laboratories, USA) for 90 min at 90 V. Membranes were blocked with 5 % (w/v) non-fat milk in Tris-buffer saline containing 0.1 % Tween-20 (TBS-T) at room temperature for 30 min and subsequently incubated at 4 °C overnight with primary antibodies (all from Santa Cruz) against p21(sc-44271), CDK4 (sc-260), E2F1(sc-35274), Bax (sc-29213), NF-κBp65 (sc-372), caspase-3 (sc-98785), and GAPDH(sc-35449). The immunolabeled membranes were washed with TBS-T for 15 min once followed by three separate washes (5 min/wash) and then probed with a horseradish peroxidase-conjugated secondary antibody at 4 °C for 1 h. To verify equal loading of samples on gel lanes, the membranes were incubated with a monoclonal GAPDH antibody, followed by a horseradish peroxidase-conjugated goat anti-mouse IgG at 37 °C for 2 h. The membrane was washed three times with TBS-T and protein bands visualized by ECL Western blotting detection reagents (Amersham Biosciences, Buckinghamshire, UK). The intensity of each target protein band was analyzed using Image J software (NIH Image, Bethesda, MD, USA) and expressed relative to GAPDH density.

Statistical Analysis

Data are expressed as mean ± SD. Multiple group comparisons were performed using one-way analysis of variance (ANOVA) followed by Dunnett’s test for pair-wise comparisons between groups. A P < 0.05 was considered statistically significant. Statistical Package for the Social Science (SPSS 17.0, SPSS Inc.) was used for all statistical analyses in this study.

Results

Variations of Aggregated Aβ_25–35 Detected

Following Aβ_25–35 aggregation (4 days at 37 °C), the aggregated forms were analyzed by Raman spectroscopy. For comparison of samples, all spectra were normalized to the intensity of the phenylalanine band (1,002 cm⁻¹). At least 5 spectra were obtained from each sample. The spectra were then baseline corrected by R2.8.1 software (Renishaw), smoothed, and averaged by Origin 8.0 (OriginLab Crop., Northampton, MA). As shown in Fig. 2, the Raman peaks at 1,617, 1,640, and 1,660 cm⁻¹, which belong to amide 1, fell as the incubation time increased (Labbé et al. 2013; Heldt et al., 2011), suggesting that amyloid fibrils were formed.

Fig. 2.

The Raman spectral shift of Aβ_25–35 after incubation for 0, 24, 36, and 48 h. Each trace is the average of 5 spectra

The ThT assay is typically performed by adding a solution of amyloid protein in a base and to a buffered ThT solution. Since ThT fluoresces only when bound to aggregated fibrils, a hypochromic shift is indicative of amyloid fibril aggregation (Ban et al. 2003). The ThT signal fell with increasing incubation time (Fig. 3a) and with increasing Bajijiasu concentration after incubation for 12 h (Fig. 3b), confirming formation of amyloid fibrils and demonstrating that Bajijiasu can partially block fibril formation.

Fig. 3.

The time-dependence of Aβ_25–35 aggregation (a) and blockade of amyloid fibril formation by Bajijiasu (b). a The thioflavin T signal increased with Aβ_25–35 incubation time, confirming fibril formation. b Bajijiasu disrupted thioflavin T binding to Aβ_25–35 and this effect was dependent on the Bajijiasu concentration. Values expressed as mean ± SD (n = 3)

Effect of Bajijiasu on PC12 Cell Viability

Cell viability was assessed using the MTT reduction assay (Fig. 4). Treatment of PC12 cells with up to 200 μM Bajijiasu alone for 24 h was non-toxic. In fact, the MTT reduction, which is linearly correlated with cell number, increased compared to untreated controls (to 145 % of the control value at 100 μM). Only at 400 μM (the highest concentration tested) did Bajijiasu reduce viable cell number (to about 90 % of control).

Fig. 4.

Bajijiasu was non-toxic to PC12 cells below 400 μM. Cell viability was assessed by MTT reduction assay and expressed relative to untreated control cultures (100 % viability). Values are mean ± SD (n = 5)

Effect of Bajijiasu on Aβ_25–35-Induced Cytotoxicity in PC12 Cells

Treatment of PC12 cells with 21 μM Aβ_25–35 for 24 h reduced estimated viable cell number to 44.25 % of control (Fig. 5a). When the cells were pretreated for 2 h with Bajijiasu (10, 20, and 40 μM), however, Aβ_25–35-induced cytotoxicity was significantly reduced to 64 % of control viability at 10 μM, 76 % at 20 μM, and 88 % at 40 μM).

Fig. 5.

Bajijiasu partially reversed Aβ_25–35-induced cytotoxicity. a Cell viability was assessed by the MTT reduction assay. b Cell death was determined by LDH release into the culture media during Aβ_25–35 treatment. Values are expressed as mean ± SD (n = 5), ^# P < 0.001 compared to the control group; *P < 0.05 and **P < 0.001 compared to the Aβ_25–35-treated group

Measurement of LDH released into the culture media from dead/dying cells confirmed that Bajijiasu partially blocked Aβ_25–35-induced cytotoxicity. Compared to the control group, Aβ_25–35 exposure significantly increased LDH leakage (P < 0.001), and this release was significantly attenuated by Bajijiasu (P < 0.05) (Fig. 5b).

Effect of Bajijiasu on Aβ_25–35-Induced Apoptosis

Annexin V-EGFP and PI double staining can distinguish healthy cells (PI-negative, Annexin-negative) from early apoptotic (Pi-negative, Annexin-positive), late apoptotic (PI-positive, Annexin-positive), and necrotic (PI-positive, Annexin-negative) cells. The apoptotic rate after Aβ_25–35 was 52.6 %, but was reduced by Bajijiasu pretreatment to 39 % at 10 μM, 36 % at 20 μM, and 18 % at 40 μM (P < 0.05 compared to Aβ_25–35 alone) (Fig. 6).

Fig. 6.

Bajijiasu reduced Aβ_25–35-induced apoptosis. Shown are FACS histograms for a control PC12 cells, b PC12 cells exposed to 21 μM Aβ_25–35, c 21 μM Aβ_25–35 + 10 μM Bajijiasu, d 21 μM Aβ_25–35 + 20 μM Bajijiasu, and e 21 μM Aβ_25–35 + 40 μM Bajijiasu. The decrease in propidium iodide (PI)-positive and annexin-positive cells (lower right quadrant) is indicative of reduced apoptosis. f Summary. Values expressed as mean ± SD (n = 5), ^# P < 0.001 compared to the control group; *P < 0.05 and **P < 0.001 compared to the Aβ_25–35-treated group

Effect of Bajijiasu on Aβ_25–35-Induced Oxidative Stress in PC12 Cells

Oxidative stress induced by Aβ_25–35 was assessed by microfluorometric measurements of intracellular ROS using the redox-sensitive dye DCFH-DA (Fig. 7a), measurement of the lipid peroxidation product MDA (Fig. 7b), and by expression of the antioxidant enzymes GSH-Px (Fig. 7c) and SOD (Fig. 7d). After exposure of PC12 cells to 21 μM Aβ_25–35 for 24 h, intracellular ROS and MDA levels were 222 and 342 % of control, while GSH-Px and SOD expression levels were attenuated to 53 and 76 % of control. Thus, Aβ_25–35 induced marked oxidative stress and reduced endogenous antioxidant capacity. Pretreated with Bajijiasu at 10, 20, and 40 μM for 2 h significantly reduced ROS accumulation in response to subsequent 24 h exposure to 20 μM Aβ_25–35 (to 152 % of control at 10 μM, 104 % at 20 μM, and 97 % at 40 μM), as well as MDA accumulation (to 215, 166, and 104 % of control, respectively). In addition, Bajijiasu pretreatment (10, 20, and 40 μM for 2 h) enhanced both GSH-Px activity (to 70, 73, and 95 % of control, respectively) and SOD activity (to 81, 90, and 95 % of control, respectively). These results indicate that Bajijiasu acts an antioxidant, either by directly scavenging free radicals, by reducing amyloid fibril formation (Fig. 2b), by enhancing endogenous antioxidant capacity, or a combination of these effects.

Fig. 7.

Bajijiasu reduced Aβ_25–35-induced oxidative stress. Oxidative stress was assessed by measuring intracellular ROS (a), MDA (b), GSH-Px (c), and SOD (d). Values expressed as mean ± SD (n = 5), ^# P < 0.001 compared to the control group; *P < 0.05 and **P < 0.001 compared to the Aβ_25–35-treated group

Effect of Bajijiasu on Aβ_25–35-Induced [Ca²⁺]_i and Mitochondrial Membrane Potential in PC12 Cells

Exposure to 21 μM Aβ_25–35 for 24 h increased [Ca²⁺]_i to 162.76 % of control (Fig. 8a) and depolarized the mitochondrial membrane potential (86.88 % loss of Ψm compared to controls, Fig. 8b), two signs of impending apoptosis or necrosis. Pretreatment with Bajijiasu partially reversed both effects in response to subsequent treatment with 21 μM Aβ_25–35 for 24 h ([Ca²⁺]_i: 144.78 % at 10 μM, 129.79 % at 20 μM, and 102.78 % at 40 μM compared to controls; Ψm: 94.83 % at 10 μM, 96.56 % at 20 μM, and 98.25 % at 40 μM compared to controls).

Fig. 8.

Bajijiasu suppressed Aβ_25–35-induced [Ca²⁺]i elevations (a) and mitochondrial membrane depolarization (b). a [Ca²⁺]i was measured by loading PC12 cells with Fura-2. b Reduced rhodamine 123 fluorescence is indicative of mitochondrial membrane depolarization. Values expressed as mean ± SD (n = 5), ^# P < 0.001 compared to the control group; *P < 0.05 and **P < 0.001 compared to the Aβ_25–35-treated group

Effect of Bajijiasu on Aβ_25–35-Induced Expression of Pro-apoptotic Proteins and Associated Signaling Proteins

Finally, we examined the effects of Bajijiasu on Aβ_25–35-induced changes in the expression of a battery of pro-apoptotic factors (Bax, caspase-3) and associated signaling molecules (p21, CDK4, E2F1, and NF-κB p65) using real-time PCR (Figs. 9, 10) and Western blot analysis (Figs. 11, 12). Exposure to 21 μM Aβ_25–35 for 24 h caused significant upregulation of CDK4 mRNA (145 % of the control value), E2F1 mRNA (152 %), Bax mRNA (149 %), caspase-3 mRNA (145 %), and NF-κB p65 mRNA (132 %), as well as the corresponding proteins (CDK4: 241 %, E2F1:165 %, Bax:190 %, caspase-3:163 %, NF-κB p65:169 %, all relative to control cultures). In contrast, Aβ_25–35 decreased both p21 mRNA and protein expression (to 46 and 38 % of control, respectively). As expected from the anti-apoptotic and cytoprotective effects of Bajijiasu, pretreatment with Bajijiasu for 2 h at 10, 20, or 40 μM partially or completely reversed these responses to subsequent 24 h exposure to 21 μM Aβ_25-35 (CDK4 mRNA:122, 114, and 101 %; E2F1 mRNA: 98, 105, and 124 %; Bax mRNA: 122, 109, and 102 %; caspase-3 mRNA: 158, 123, and 108 %; NF-κB p65 mRNA: 143, 116, and 111 %; p21 mRNA: 57, 67, and 83 %, all relative to expression in untreated controls) (Figs. 9, 10). Estimation of the corresponding protein expression levels by Western blotting (Figs. 11, 12) mirrored these RT-PCR results (CDK4:185, 166, and 111 %; E2F1: 136, 114, and 93 %; Bax:149, 147, and 112 %; NF-κB p65:154, 137, and 114 %; caspase-3: 131, 108, and 94 %; p21: 72, 83, and 98 %). In sum, these results unambiguously demonstrate the ability Bajijiasu to protect PC12 cells from Aβ_25–35-induced apoptosis.

Fig. 9.

Bajijiasu pretreatment reversed changes in p21 (top), CDK4 (middle), and E2F1 (bottom) mRNA expression induced by Aβ_25–35 in PC12 cells as revealed by real-time PCR. Data calculated according to the comparative Ct method and using expression of the GAPDH gene for normalization. Values expressed as means ± SD (n = 4), (^# P < 0.001 compared to the control group; *P < 0.05, **P < 0.001 compared to the Aβ_25–35-treated group)

Fig. 10.

Bajijiasu pretreatment reversed Aβ_25–35-induced changes in the expression of BAX (top), caspase-3 (middle), and NF-κB (bottom) mRNA expression in PC12 cells as measured by real-time PCR. Data calculated and expressed as in Fig. 9 (n = 4), (^# P < 0.001 compared to the control group; *P < 0.05, **P < 0.001 compared to the Aβ_25–35-treated group)

Fig. 11.

Bajijiasu pretreatment reversed the changes in p21 (top), CDK-4 (middle), and E2F1 (bottom) protein expression induced by Aβ_25–35 as measured by Western blots. Densitometry values were normalized to GAPDH expression and expressed as means ± SD (n = 4) relative to untreated control PC12 cells (^# P < 0.001 compared to the control group; *P < 0.05, **P < 0.001 compared to the Aβ_25–35-treated group)

Fig. 12.

Bajijiasu pretreatment reversed the effect of Aβ_25–35 on BAX (top), caspase-3 (middle), and NF-κB (bottom) protein expression in PC12 cells as revealed by Western blot analysis. Data calculated and expressed as in Fig. 11 (n = 4), (^# P < 0.001 compared to the control group; *P < 0.05, **P < 0.001 compared to the Aβ_25–35-treated group)

Discussion

Studies on the chemistry and neurotoxic activity of various Aβ fragments have revealed that the peptide fragment Aβ_25–35 retains the capacity for self-aggregation and the toxicity of the complete peptide in vitro and in vivo. This toxicity involves oxidative stress due to the production of free radicals, leading to DNA, protein, and lipid damage, disruption of calcium homeostasis, and eventually apoptosis. Indeed, i.c.v. administration of Aβ_25–35 into the rodent brain-induced histological and biochemical sequela of oxidative stress and neurodegeneration, as well as memory deficits (Liu et al. 2009a, b; Meunier et al. 2006). Here we demonstrate that Bajijiasu (β- d-fructofuranosyl (2–2) β- d-fructofuranosyl), an active component of M. officinalis, a herb used for centuries in traditional Asian medicine, reverses Aβ_25–35-induced apoptosis in PC12 cell, likely by preventing Aβ_25–35 aggregation, Aβ_25–35-induced oxidative stress, calcium dysregulation, loss of mitochondrial membrane potential (Ψm), expression of pro-apoptotic factors such as effector caspase-3, or by some combination of these effects.

We used confocal micro-Raman spectroscopy, a combination of Raman spectroscopy and confocal microscopy, to demonstrate that Bajijiasu prevents Aβ_25–35 self-aggregation, which is necessary for peptide toxicity. Raman spectroscopy is based on the principle of inelastic scattering (Sánchez et al. 2012). The unique pattern of changes in the vibrational state of a molecule under laser excitation in a specific environment constitutes a chemical “fingerprint” to identify and characterize biomolecules in various conformations or states. Raman spectroscopy has proven extremely versatile and has a vast array of applications in biomedical science (Frost et al. 2013; Kotchey et al. 2013; Richard-Lacroix and Pellerin 2013; Zhuang et al. 2012; Pudlas et al. 2010; Won-in et al. 2011; Smith et al. 2013). ThT rapidly binds to amyloid fibrils, accompanied by a dramatic increase in fluorescence at around 485 nm when excited at 455 nm, making ThT a useful probe to detect the formation of amyloid fibrils in real time. Using ThT fluorescence, we demonstrated time-dependent Aβ_25–35 self-aggregation and dose-dependent disruption of this aggregation by Bajijiasu, suggesting a possible therapeutic mechanism for this compound.

Oxidative stress, defined as a disturbance in the balance between the production of ROS and antioxidant defense capacity, has been implicated in the neuronal injury induced by Aβ (Li et al. 2008a, b; Zhang et al. 2008). Evidence from both experimental models and human AD brain studies has implicated oxidative stress in the pathogenesis of AD (Chauhan and Chauhan 2006). β-Amyloid treatment caused a significant increase in ROS (Li et al. 2008a, b; Peng et al. 2009). Excessive ROS production is known to cause oxidative damage to cellular macromolecules, including DNA, lipids, and proteins, thereby disrupting cellular functions and structural integrity (Gardner et al. 1997; Fiers et al. 1999). In this study, Aβ_25–35 caused a marked accumulation of ROS and the lipid peroxidation product MDA, and reduced both GSH-Px and SOD expression. These effects were partially or completely reversed by Bajijiasu pretreatment, suggesting that Bajijiasu may directly scavenge free radicals, induce endogenous anti-oxidant defenses, or both. Further studies are required to distinguish among these possible neuroprotective mechanisms.

Oxidative phosphorylation depends on maintenance of the membrane potential across the inner mitochondrial membrane, and loss of Ψm (mitochondrial uncoupling) can cause oxidative stress and exacerbate oxidative stress from other sources (Perez and Cederbaum 2003). In turn, loss of Ψm can trigger apoptosis through activation of the permeability transition pore and release of the caspase-3 activator cyctochrome c from the mitochondrial matrix. Mitochondrial dysfunction has been demonstrated in cells treated with Aβ (Zhang et al. 2008), AD transgenic mice (Eckert et al. 2008), platelets from AD patients (Parker et al. 1990), and in postmortem brains of AD patients (Devi et al. 2006). Mitochondrial dysfunction is also a prominent feature of Aβ-induced neuronal toxicity in AD patients (Chen and Yan 2007) and is central to the development of oxidative stress because mitochondria are the primary source of endogenous free radicals (Beal 2005). Pretreatment of PC12 cells with Bajijiasu reversed the Ψm depolarization induced by Aβ_25–35, suggesting that Bajijiasu prevents cell death by blocking activation of the mitochondrial apoptotic pathway.

To further examine the molecular signaling pathways involved in Aβ-triggered apoptosis and Bajijiasu-mediated cytoprotection, we measured the mRNA and protein expression levels of p21, CDK4, E2F1, Bax, NF-κB p65, caspase-3 by RT-PCR and Western blotting. The kinase p21 is an inhibitor of cyclin-dependent kinase and thus critical for growth arrest in mitotic cells (by blocking progression beyond G₁). The HMG-CoA reductase inhibitor simvastatin inhibits cell cycle progression at the G1/S checkpoint in immortalized lymphocytes from AD patients independently of lower p21 and p27 expression (Sala et al. 2008). The CyclinD1/CDK4 complex is a key regulator of the G₁/S transition; once activated, nerve cells will leave the G0 phase and re-enter the cell cycle. The present study showed that p21 mRNA and protein expression was reduced by Aβ_25–35, whereas CDK4 expression was enhanced, as was expression of the downstream transcription factor E2F1. Overexpression of E2F1 mediated neuronal apoptosis in Down’s syndrome brains with dementia of Alzheimer-type triggered by Aβ deposition. Treatment with Aβ_25–35 also increased expression of the pro-apoptotic brain Bax, and this effect was also reversed by Bajijiasu. ROS produced by Aβ can activate the ubiquitous stress response nuclear factor kappa B (NF-κB), which contributes to cell apoptosis and development of AD (Siebenlist et al. 1994; Boehrer et al. 2000; Baichwal and Baeuerle 1997; Lezoualc’h et al. 1998), and regulates apoptosis by acting on caspase-3 (Bonavia et al. 2001; Allen et al. 2001). Caspase-3 also plays an important role in AD pathogenesis by participating in the decomposition of APP (Weidemann et al. 1999), PS-1, and PS-2 (Walter et al. 1998). Treatment with Aβ_25–35 increased both NF-κB p65 and caspase-3 expression levels, and again these responses were reversed by Bajijiasu. Whether these changes in protein expression are directly involved in neurotoxicity and neuroprotection or reflect these responses remains to be determined. By lowering expression of NF-κB p65 mRNA, Bajijiasu likely inhibits nuclear localization of NF-κB and the downstream stress response gene expression cascades that are generally regarded as protective. Thus, Bajijiasu likely blocks the stress response induced by Aβ_25–35.

Intracellular calcium overload can also promote mitochondria release of cytochrome c, activation of caspase-9 and caspase-3, and ultimately apoptosis (McCollum et al. 2002; Paszty et al. 2002; Petersen et al. 2000; Reiter 1998; Huang et al. 2000). Bajijiasu suppressed Aβ_25–35-induced calcium overload, possibly by suppressing ROS generation and the ensuing damage to calcium regulatory proteins.

Drugs can be designed to target specific molecules on the cell surface or particular cell types and tissues by chemical modification with certain carbohydrates. Indeed, carbohydrate recognition is important in many natural processes involving cell–cell recognition and signaling, such as fertilization, cell growth, differentiation, immune responses, bacterial infection, and tumor metastasis (Wang 2001). These drugs may have enormous potential. In fact, another marine sulfated oligosaccharide, HSH-971, has been proposed as a potential AD treatment.

Morinda officinalis has long been used in Chinese medicine to treat dementia, stroke, depression, and other neurodegenerative diseases (Yang et al. 2010; Li et al. 2004, 2008a, b). Bajijiasu, a dimeric fructose isolated from the radix of this plant, is a promising therapeutic agent for the treatment of various neurodegenerative diseases. It is worth noting that M. officinalis contains about 1.08 % Bajijiasu. Although M. officinalis has been the subject of many pharmacological investigations, our study is the first to demonstrate the neuroprotective efficacy of Bajijiasu against Aβ-mediated cytotoxicity. Treatment with Bajijiasu significantly increased cell viability and mitochondrial membrane potential, while reducing oxidative stress and apoptosis.

In conclusion, Bajijiasu exerts a protective effect against Aβ_25–35-induced neurotoxicity in PC12 cells. The protective efficacy of Bajijiasu most likely results from suppression of oxidative stress, either by direct scavenging or by interfering with Aβ fibril formation, resulting in stabilization of Ψm and [Ca²⁺]_i, and blockaid of mitochondria-dependent apoptosis. These observations highlight the potential of M. officinalis as a source of compounds for the treatment of neurological disease and underscore the promise of Bajijiasu as a natural compound worthy of further development as a pharmaceutical therapy for AD.