Salvianolic acid B inhibits Aβ fibril formation and disaggregates preformed fibrils and protects against Aβ-induced cytotoxicty

Available online at www.sciencedirect.com

Neurochemistry International 52 (2008) 741–750 www.elsevier.com/locate/neuint

Salvianolic acid B inhibits Ab fibril formation and disaggregates preformed fibrils and protects against Ab-induced cytotoxicty Siva Sundara Kumar Durairajan a, Qiuju Yuan b, Lixia Xie a, Wing-Sai Chan a, Wan-Fung Kum a, Irene Koo a, Chenli Liu b, Youqiang Song b, Jian-Dong Huang b,*, William L. Klein c, Min Li a,** a

b

School of Chinese Medicine, Hong Kong Baptist University, Kowloon Tong, Hong Kong, China Department of Biochemistry, Li Ka Shing Faculty of Medicine, The University of Hong Kong, 21 Sassoon Road, Pokfulam, Hong Kong, China c Department of Neurobiology and Physiology, Northwestern University, 2205 Tech Drive, Evanston, IL 60208, USA Received 30 July 2007; received in revised form 27 August 2007; accepted 5 September 2007 Available online 16 September 2007

Abstract One of the major pathological features of Alzheimer’s disease (AD) is the appearance of senile plaques characterized by extracellular aggregation of amyloid b-peptide (Ab) fibrils. Inhibition of Ab fibril aggregation is therefore viewed as one possible method to halt the progression of AD. Salvianolic acid B (Sal B) is an active ingredient isolated from Salvia miltiorrhiza, a Chinese herbal medicine commonly used for the treatment of cardiovascular and cerebrovascular disorders. Recent findings show that Sal B prevents Ab-induced cytotoxicity in a rat neural cell line. To understand the mechanism of Sal B-mediated neuroprotection, its effects on the inhibition of Ab1–40 fibril formation and destabilization of the preformed Ab1–40 fibrils were studied. The results were obtained using Thioflavin T fluorescence assay and Ab aggregating immunoassay. We found that Sal B can inhibit fibril aggregation (IC50: 1.54–5.37 mM) as well as destabilize preformed Ab fibril (IC50: 5.00–5.19 mM) in a doseand time-dependent manner. Sal B is a better aggregation inhibitor than ferulic acid but less active than curcumin in the inhibition of Ab1–40 aggregation. In electron microscope study, Sal B-treated Ab1–40 fibrils are seen in various stages of shortening or wrinkling with numerous deformed aggregates of amorphous structure. Circular dichroism data indicate that Sal B dose dependently prevents the formation of b-structured aggregates of Ab1–40. Addition of preincubated Sal B with Ab1–42 significantly reduces its cytotoxic effects on human neuroblastoma SH-SY5Y cells. These results suggest that Sal B has therapeutic potential in the treatment of AD, and warrant its study in animal models. # 2007 Elsevier Ltd. All rights reserved. Keywords: Alzheimer’s disease; b-Amyloid fibrils; Salvianolic acid B; Thioflavine T; Ab aggregating ELISA; Electron microscopy; Circular dichroism spectroscopy; Cytotoxicity

1. Introduction Alzheimer’s disease (AD) is a neurodegenerative disease that mostly affects the elderly. Prevalence studies reveal that in

Abbreviations: AD, Alzheimer’s disease; Ab, amyloid b-peptide; MTT, 3(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide; HFIP, Hexafluoroisopropanol; DMSO, dimethyl sulphoxide; PBS, phosphate buffered saline; PC-12, pheochromocytoma cells; SH-SY5Y, human neuroblastoma cells; ThT, Thioflavin T; EM, electron microscopy; CD, circular dichroism; TCM, traditional Chinese medicine; Sal B, salvianolic acid B; TMP, tetramethylpyrazine; AS-IV, astragaloside-IV; IC50, effective concentrations at 50% value. * Corresponding author. Tel.: +852 2819 2810; fax: +852 2855 1254. ** Corresponding author. Tel.: +852 34112919; fax: +852 34112461. E-mail addresses: [email protected] (J.-D. Huang), [email protected] (M. Li). 0197-0186/$ – see front matter # 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.neuint.2007.09.006

2000 there were 25 million persons with AD globally, and this number is predicted to increase to 114 million by 2050 if new preventive or neuroprotective therapies do not emerge (Lleo´ et al., 2006). A recent local survey indicated that AD is accounts for 64.6% of the overall prevalence rate of 6% dementia in Hong Kong (Chan and Lam, 2005). The prevalence doubles every 5 years from the age of 65 (Chan and Lam, 2005). Existing treatments for AD cannot cure the disease, but can offer some people modest improvement in some symptoms with side effects (Lleo´ et al., 2006). Therefore, there is a need for alternative drugs and in particular for phytotherapy. It is accepted that abnormal production and aggregation of b-amyloid peptides (Ab) are initial pathogenic events in AD (Selkoe, 2004; Lleo´ et al., 2006). Substantial evidence suggests that Ab-induced oxidative stress plays a key role in the

742

S.S.K. Durairajan et al. / Neurochemistry International 52 (2008) 741–750

pathogenesis or progression of AD (Butterfield et al., 2002; Butterfield, 2002). Recently, several studies have reported the potential of antioxidants in the inhibition of Ab deposition using in vitro assays and transgenic mouse model studies of AD, and these antioxidants currently are under development (Reddy, 2006; Ono et al., 2006). For example, curcumin, resveratrol, rosmarinic acid, nordihydroguaiaretic acid, ferulic acid, tannic acid and some polyphenols have been shown to inhibit Ab fibril formation as well as destabilize preformed Ab fibrils in vitro (Ono et al., 2006). However, few studies have illustrated the role of these antioxidants in the inhibition of Ab accumulation either in vitro or in vivo (Yang et al., 2005). Here we report the anti-amyloidogenic properties of salvianolic acid B (Sal B). Sal B is a water soluble polyphenolic caffeic acid derivative (Fig. 1) extracted from the root of Salvia miltiorrhiza (‘‘Danshen’’ in Chinese), and it is the most abundant and bioactive of the salvianolic acids in Danshen (Jiang et al., 2005). Sal B has been commonly used in traditional Chinese medicine (TCM) for the treatment of coronary artery disease and cerebrovascular diseases (Zhou et al., 2005). Danshen dripping pill (Fufang–Danshen–Diwan) is the first Chinese medicine approved for clinical trials by the Food and Drug Administration in the U.S. (Zhou et al., 2005). Recently, the neuroprotective potential of Sal B against Abinduced toxicity in PC-12 cells has been explored (Lin et al., 2006). Sal B is also an anti-apoptotic drug that can reduce Abinduced damage in PC-12 cells by decreasing prostate apoptosis response-4 expression, an apoptotic factor of the cells after they have been exposed to Ab (Tang and Zhang, 2001). However, the mechanism by which Sal B inhibits Ab fibril formation and aggregation of Ab fibrils has not yet been investigated. In the present study, we examined the effects of Sal B on formation of Ab aggregates and destabilization of

preformed Ab fibrils in vitro by using fluorescence spectroscopy with Thioflavin T (ThT), Ab aggregating immunoassay (ELISA), electron microscopy (EM) and circular dichroism (CD) spectroscopy. Finally we demonstrated the neuroprotective effect of Sal B by a 3-(4,5-dimethylthiazol-2-yl)-2,5diphenyl tetrazolium bromide (MTT) assay. 2. Materials and methods 2.1. Preparation of Ab and antioxidants for the inhibition and destabilization of Ab aggregates Ab1–40 and Ab1–42 peptides were purchased from Bachem (Switzerland) and American peptide (California, USA), respectively. Sal B and other pure compounds were purchased from the National Institute for the Control of Pharmaceutical and Biological Products (Beijing, China). Ab peptides were prepared for the assays as described previously (Munoz-Ruiz et al., 2005). Briefly, Ab peptides were dissolved in 100% hexafluoroisopropanol (HFIP) (Sigma–Aldrich) at room temperature for 2 h to produce monomeric Ab. The HFIP was dried under a gentle stream of nitrogen gas, and the resulting peptide thin film was stored at 80 8C. The peptides were then dissolved in DMSO (Sigma–Aldrich) at a concentration of 4.6 mM prior to assay. Ab aggregation was performed at 15 mM of Ab1–40 or Ab1–42 in 50 mM phosphate buffer pH7.4, and 100 mM NaCl (PBS) at 37 8C. In our preliminary experiments, we first screened Sal B, astragaloside-IV, and tetramethylpyrazine compounds for their ability to inhibit Ab1–40 fibril formation and to destabilize preformed Ab1–42 and then for their neuroprotective effects against Ab-induced cytotoxicity in rat pheochromocytoma (PC12) cells at concentrations of 10 and 100 mM. Among them, only Sal B showed significant inhibition of Ab aggregation. Subsequently, we evaluated the efficiency of Sal B on the inhibition Ab1–40 in detail. Our pilot data indicated that formation of Ab aggregates was time-dependent, so readings were taken on days 1–7 by both ThT fluorescence and ELISA absorbance. The maximal readings were reached on days 6 and 7, indicating the saturating level of Ab aggregation (Fig. 3A and B). For the inhibition of Ab aggregation assay, the reaction mixtures containing 15 mM aliquots of Ab1–40 with/without Sal B or ferulic acid at a final concentration of 0.01–100 mM were incubated for 3 and 7 days at 37 8C. Curcumin was prepared at a final concentration of 0.01–50 mM in PBS. All the reaction volumes contained less than 1% DMSO. For destabilization of preformed Ab aggregates, Ab1–40 alone was incubated for 4 days, and then further incubated for 3 days after addition of Sal B, ferulic acid or curcumin. Ferulic acid and curcumin were taken as reference compounds in the assay.

2.2. Thioflavin T fluorescence assay ThT-induced fluorescence changes were measured to quantify amyloid fibril formation by spectrofluorophotometer (Fusion Universal Microplate Analyzer) as described in (Munoz-Ruiz et al., 2005). To determine amyloid fibril formation, the solutions containing Ab1–40 with/without Sal B or ferulic acid were added to 50 mM glycine–NaOH buffer, pH 8.5, containing 3 mM ThT (Sigma– Aldrich) in a final volume of 150 ml. Each assay was run in triplicate and fluorescence intensities were measured at 450 nm (excitation) and 485 nm (emission) under time-resolved fluorescence mode.

2.3. ELISA for disaggregation and inhibition of Ab aggregates

Fig. 1. Chemical structures of salvianolic acid B, ferulic acid and curcumin.

To confirm the potency of Sal B in inhibiting Ab1–40 aggregation, aliquots of Ab with/without Sal B or curcumin were taken at the end of incubation and tested on a single-antibody double-sandwich ELISA. Specificity for polymerized Ab was accomplished by the use of 10G4 monoclonal antibody as the capture antibody and biotin-labeled 10G4 for detection. 10G4 is monoclonal to the amino acid 5–13 region of native human Ab1–40 (Yang et al., 1994). We also confirmed the identity of both these antibodies by staining the Ab plaques in the brain section of TgCRND8 mice and detecting the Ab1–40 polymers in Western blotting (data not shown). The ELISA was performed as mentioned

S.S.K. Durairajan et al. / Neurochemistry International 52 (2008) 741–750 earlier with some modifications (LeVine, 2004; Yang et al., 2005). Briefly, 96well plates were coated with 10G4 at 3 mg/ml in 0.1 M NaHCO3 (pH 9.6) and incubated at 4 8C overnight. The wells were then blocked with blocking buffer (PBS containing 2% BSA and 0.05% Tween 20) for 2 h at room temperature. Then 100 ml aliquots of the samples (Ab solutions were diluted to 65 nM in PBS) were added to each well and incubated at room temperature for 2 h with constant rotation at 30 rpm. After washing four times with PBST, 100 ml/well biotinylated 10G4 (1:1500) was added, and incubated for 2 h with shaking. The plates were then washed four times with PBST, incubated with 100 ml/well of streptavidin-HRP (DAKO) (1:4000) for 1 h. Finally the plates were washed four times with PBST before adding 100 ml/well of tetramethylbenzidine substrate (Pierce) for 30 min. Absorbance values at 450 nm were measured in duplicate wells after addition of 1 M H2SO4. All ELISA experimental data were from three different independent days.

2.4. Electron microscopy Aliquots (25 ml) of 50 mM Ab1–40 and preformed Ab1–40 were incubated with/without Sal B at a final concentration of 1–100 mM for 3–7 days at 37 8C. Following incubation, 5 ml of duplicate samples were placed on glow-discharged 300-mesh Formvar-coated copper grids for 1 min; excess samples were wicked off, and the grids were washed with distilled water and negatively stained with 1% uranyl acetate (Sigma–Aldrich) solution for 1 min. After 2 h of air drying, the specimens were examined in a Philips CM12 electron microscope at 80 kV (Eindhoven, Netherlands) at instrumental magnification of 44,000.

2.5. CD spectroscopy Modifications in the secondary structural content of Ab1–40 with/without Sal B for 3–7 days (as prepared for EM study) were recorded using a Jasco circular dichroism spectrometer (CD) model J-720. CD spectra were measured using a 1 cm quartz cell from 200 to 250 nm, with a step interval 0.5, 1 nm bandwidth, a scanning speed of 50 nm/min, and a resolution of 0.1 nm. Three scans each of duplicate samples were measured and averaged. The CD spectra of Ab1–40 and experimental samples were subtracted from the CD spectra of the blank (PBS buffer with 1% DMSO). The spectra were analyzed using CDpro software using the CONTIN program (Sreerama and Woody, 2004).

2.6. Cell culture and neurotoxicity assay For the neurotoxicity assay, SH-SY5Y (human neuroblastoma) cells were cultured, maintained, differentiated and assayed as previously described (Datki et al., 2003; Yang et al., 2005). SH-SY5Y cells were plated in a 24-well plate at a density of 2  105 cells/ml in Dulbecco’s modified Eagle/F12 medium (Gibco) medium and differentiated with 10 mM of all-trans retinoic acid (RA) (Sigma– Aldrich) for 7 days. The medium was replaced with 900 ml of new medium (with 1% FBS and N2 supplement) free from RA; this was done rapidly prior to assay. Ab1–42 (500 mM) seed samples with/without Sal B (5–1000 mM) in PBS were preincubated for 3–7 days for fibril aging at 37 8C (as prepared in aggregation studies). Preformed Ab1–42 fibrils were allowed to age for 4 days and Sal B was added at the concentration indicated above. At the end of incubation, 100 ml samples containing Sal B at final concentrations ranging between 0.5 and 100 mM, and 50 mM of Ab1–42 were dispensed into 900 ml media/SH-SY5Y cells. As a control, PBS containing 5% DMSO was diluted 10:1 with medium and added to some wells. The plates were incubated at 37 8C for 24 h and then 100 ml of MTT (Sigma–Aldrich) (4 mg/ml) was added to each well and further incubated for 4 h. The final concentration of DMSO in each well was less than 0.5%. The MTT solution was aspirated off and the cell crystals were washed once with PBS and then dissolved using 1 ml of 4:1 DMSO–EtOH mixture. Finally, color intensity was measured using an ELISA reader at 570 nm with the reference set to 620 nm.

2.7. Data analysis All values were represented as mean  S.D., or the S.E.M. Student–Newman–Keuls test was used to analyze differences among the groups and p values

743

less than 0.05 and 0.001 were taken as statistically significant. Sigmastat 3.5 software was used to analyze the data. The IC50 values were determined graphically by four parametric logistic function using Sigmaplot 10.

3. Results In searching Chinese medicine for neuroprotective agents with anti-oxidative properties, our preliminary data indicated that Sal B significantly inhibited Ab1–40 fibrillogenesis and destabilized preformed fibrils of Ab1–42 at concentrations of 10–100 mM in ThT fluorescence assay and also protected PC12 cells from Ab-induced toxicity (Fig. 2A–D). Sal B exhibited a direct inhibitory action on Ab fibril aggregation because the pre-incubated Sal B with preformed Ab1–42 fibril showed more protection against Ab-induced toxicity than the simultaneous addition of Sal B and preformed Ab1–42 fibril into PC-12 cells (Fig. 2D and E). In order to confirm our preliminary data on Sal B’s inhibition of Ab fibril formation along with the mechanism of its action in detail, we here report the dose- and time-dependent action of Sal B in the inhibition of Ab formation and the destabilization of preformed Ab fibrils along with neuroprotection in SH-SY5Y cells using ThT fluorescence assay, immunoassay, EM, CD spectroscopy and MTT assay. 3.1. Inhibition of Ab fibril formation assessed by ThT fluorescence binding assay and ELISA To exclude a possible interference of the data by the ThT dye, we confirmed the results using single antibody sandwich immunoassay for Ab aggregation (LeVine, 2004; Yang et al., 2005). It allows the detection of the Ab aggregation with different forms except the monomer (LeVine, 2004; Yang et al., 2005). The inhibitory effect of Sal B was found to be dosedependent in both assays (Fig. 3C–F). Three days of incubation showed higher inhibitory effect than 1 week according to the respective IC50 values (Table 1). The 3-day IC50 values of Sal B were 2.12 and 1.54 mM in ThT fluorescence and ELISA, respectively; these values are 2.5- and 1.9-fold less than those after 1 week of incubation. Significant inhibitory effects of Sal B were observed at concentrations above 0.1 mM, with maximum effect at 100 mM in ThT fluorescence binding assay and ELISA in both periods of incubation ( p < 0.001). Next we examined the effect of Sal B on Ab fibril elongation in the presence of preformed Ab aggregates assessed by ThT fluorescence binding and ELISA. Sal B not only inhibited amyloid aggregation but also exhibited a destabilizing effect on preformed Ab1–40 fibrils. ThT fluorescence and ELISA absorbance of preformed Ab1–40 were both reduced in relation to the amount of drugs added with a minimal inhibition concentration from 0.1 mM ( p < 0.001) (Fig. 3C–F) (IC50: 5.19 and 5.00 mM in the ThT fluorescence and ELISA, respectively). We then compared these results to two well-known antiamyloidogenic antioxidants curcumin and ferulic acid. Since the fluorescence emission spectrum of curcumin overlaps with that of ThT dye, the anti-Ab aggregation activity of curcumin

744

S.S.K. Durairajan et al. / Neurochemistry International 52 (2008) 741–750

Fig. 2. Effects of Sal B, astragaloside IV (AS-IV), tetramethylpyrazine (TMP) preincubated with Ab on the inhibition of Ab1–40 fibrillogenesis (A) and preformed Ab1–42 aggregates (B) in ThT fluorescence assay (PI: propidium is the reference compound) and neuroprotection in PC-12 cells against Ab1–40 (C) and preformed Ab1–42 (D)-induced toxicity in MTT-viability assay. Effects of concomitant addition of compounds and preformed Ab1–42 into PC-12 cells (E) in MTT-viability assay. The values are the means  S.D. (n = 3). **p < 0.001; *p < 0.05.

was measured using ELISA. The IC50 values of ferulic acid to inhibit the fibril formation of Ab after 3 days and 1 week of incubation, and to destabilize preformed Ab fibrils are 3.42, 2.89 and 6.59 mM, respectively (Table 1). In comparison, the IC50 values of Sal B are nearly twofold less than those of ferulic acid both in the inhibition of Ab fibril formation after 3 days of incubation and in destabilization; however the ferulic acid showed less IC50 value than Sal B in the inhibition of Ab1–40 fibril formation after 1 week of incubation. Thus, Sal B is a better aggregation inhibitor in the short term (i.e., 3 days or less). In contrast, curcumin showed profound inhibition of Ab aggregation with a twofold lower IC50 values than Sal B in all tested durations of the inhibition and in the disaggregation of Ab1–40 aggregation.

3.2. Electron microscopy Fig. 4A, E and I shows the typical amyloid fibril formation from untreated aged Ab1–40 with apparently ordered filaments several microns in length. Compared to 3-day incubation (Fig. 4A), 1-week incubation (Fig. 4E) showed more Ab1–40 fibril formation. In the presence of Sal B at concentrations from 10 to 100 mM in 3- and 7-days incubations, the fibrils are converted into deformed aggregates of amorphous structure, with the degree of deformation dose-dependent (Fig. 4C, D, G and H). In 3- and 7-days incubations of Sal B at a concentration of 1 mM, only short fibrillar assemblies were observed (Fig. 4B and F). In contrast, Sal B fully converted the preformed Ab aggregates into an amorphous structure with no fibrils observed

S.S.K. Durairajan et al. / Neurochemistry International 52 (2008) 741–750

745

Fig. 3. Inhibition of Ab1–40 fibril formation and disassembly of preformed Ab1–40 fibril induced by inhibitor compounds monitored by ThT fluorescence and ELISA absorbance. Time-dependent Sal B-induced inhibition of Ab1–40 fibril formation by ThT fluorescence binding assay (A) and ELISA absorbance (B). Ab1–40 was incubated at final concentration of 15 mM at 37 8C in the presence of PBS control or 100 mM Sal B. The extent of Ab1–40 aggregation in the presence of inhibitors at various concentrations (0.01–100 mM for Sal B and ferulic acid; 0.01–50 mM for curcumin) was measured by the ThT fluorescence assay and ELISA absorbance. The dose-dependent inhibition of Ab1–40 aggregates by Sal B, ferulic acid and curcumin on 3 days and 1 week of incubation were monitored by ThT fluorescence assay (C) and ELISA absorbance (E), respectively. For destabilization of Ab1–40 aggregates, the preformed Ab1–40 were allowed for 4 days and then exposed to inhibitors for another 3 days at 37 8C at the above indicated concentrations. The dose-dependent destabilization effect of Sal B, ferulic acid and curcumin were identified by ThT fluorescence assay (D) and ELISA (F). The mean data were from the three individual experiments. The fits were illustrated as four parametric logistic curves, and the IC50 values are summarized in Table 1.

at concentrations up to 10 mM (Fig. 4K and L). At a concentration of 1 mM only few amorphous aggregates were seen (Fig. 4J). 3.3. Effect of Sal B on secondary structure of Ab1–40 To elucidate the modulation in the structural details of Ab upon Sal B treatment, we monitored the secondary structural conformation of Ab1–40 with/without Sal B at concentrations from 1 to 100 mM for 3- and 7-days incubations using CD spectroscopy. The CD spectra of Ab1–40 exhibit a strong absorption minimum at 218 nm, indicating a pattern of b-sheet

conformation (Fig. 5). However, this minimum flattens out with less pronounced negative ellipticity values after treatment with Sal B at concentrations of 10 and 100 mM to Ab1–40, indicating different structural modifications compared with those observed when Ab1–40 was incubated alone over periods of 3 and 7 days (Fig. 5). The b-sheet content of control Ab1–40 increased over time from 63.6% on day 3 to 71.4% after day 7 at 37 8C (Fig. 5). After the treatment with Sal B, there was a drastic decrease in the b-sheet population from around 10% for 1 mM to over 30% for 100 mM of Sal B, indicating their preventive effect on Ab1–40 from adopting a b-sheet conformation. There is direct correlation between the increases

746

S.S.K. Durairajan et al. / Neurochemistry International 52 (2008) 741–750

Table 1 The inhibition concentration (IC50)a of Sal B, curcumin and ferulic acid for Ab assembly inhibition and disassembly Compounds

Salvianolic acid B Ferulic acid Curcumin

In 3 daysb (mM)

In 7 daysb (mM)

Destabilizationc (mM)

ThT fluorescence assay

ELISA

ThT fluorescence assay

ELISA

ThT fluorescence assay

ELISA

2.12  0.03 3.42  0.04 nd

1.54  0.04 nd 0.64  0.07

5.37  0.04 2.89  0.04 nd

4.75  0.05 nd 2.58  0.07

5.19  0.04 6.59  0.05 nd

5.00  0.03 nd 4..11  0.04

nd: not determined. a IC50 values from the mean of three different independent experiments with standard errors. b The reaction solution containing 15 mM Ab1–40 with Sal B or ferulic acid at a final concentration of 0.01–100 mM or curcumin at a final concentration of 0.01– 50 mM in PBS pH 7.4 were incubated for 3 or 7 days at 37 8C. c The reaction solution containing 15 mM Ab1–40 in PBS was allowed to fibril for 4 days and 0.01–100 mM Sal B or ferulic acid or 0.01–50 mM curcumin in PBS, pH 7.4 were added and then incubated further for 3 days.

Fig. 4. Electron microscopy imaging of Ab1–40 aggregation with or without Sal B. EM analysis of Ab1–40 aggregates after 3 days when incubated alone (A), or with Sal B at 100 mM (B), 10 mM (C) and 1 mM (D); Ab1–40 aggregates after 1 week when incubated alone (E) or with Sal B at 100 mM (F), 10 mM (G) and 1 mM (H); preformed Ab1–40 aggregates after 4 days when incubated alone (I) or with 100 mM (J), 10 mM (K), and 1 mM (L) Sal B. Scale bars = 200 nm. Images were acquired using Philips CM12 electron microscope at 80 kV in with 44,000 magnification.

S.S.K. Durairajan et al. / Neurochemistry International 52 (2008) 741–750

747

Fig. 5. The influence of Sal B on the secondary structure components in Ab1–40 studied by CD. Ab1–40 peptides (50 mM) were incubated with/without different concentration of Sal B in PBS at 37 8C for 3 days (A) to 7 days (B). Data were collected as described in Section 2.

of the unordered population with the decrease in the b-sheet population. The changes observed in the CD spectra are also reflected in ThT fluorescence and ELISA absorbance and finally in direct evidence from EM studies. 3.4. Protective effect of Sal B on the cytotoxicity of Ab1–42 Exposure of SH-SY5Y cells to differently aged Ab1–42 fibrils at 37 8C showed maximal inhibition of cell viability 25.1  1.9 for 3 days and 28.1  1.8 for 7 days of fibrils formation (Fig. 6). The death rate increases with the aging time of Ab1–42 fibrils. The lowest cell viability 21.1  2.6 was observed with preformed Ab1–42 fibrils (Fig. 6). Ab-induced toxicity was attenuated when Ab1–42 co-incubated with Sal B

dose dependently. As shown in Fig. 4, Sal B at concentrations of 50 and 100 mM, significantly restored 70–88% viability of SHSY5Y cells against differentially aged Ab-induced cytotoxicity ( p < 0.001). The treatment of Sal B did not affect cell viability in all tested concentrations (results not shown). The cell protection effect of Sal B in 7-day incubation at all tested concentrations against Ab-induced neurotoxicity was less than that of the 3-day incubation with Ab1–42 and with preformed Ab1–42 at a corresponding concentration. Nevertheless, significant dose-dependent cytoprotective activity of Sal B was observed at concentrations above 0.5 mM ( p < 0.05) for all days of incubation with Ab1–42. Therefore, these results indicate that SH-SY5Y cell protection by Sal B due to its involvement in disaggregating neurotoxic Ab aggregates. These data are well corroborated with our preliminary findings on the protection of PC-12 cells by Sal B against Ab1–40- and Ab1–42-induced toxicity. PC-12 cells seem to be highly susceptible to Ab1–42 than SH-SY5Ycells because there is a significant cell death even at concentration of 15 mm of Ab1–42. 4. Discussion

Fig. 6. Protective effects of Sal B on Ab1–42- induced cytotoxicity in SHSY5Y cells. The preincubated mixtures of Ab1–42 (50 mM) with/without various concentrations of Sal B for 3–7 days or preformed Ab1–42 were added to the differentiated SH-SY5Y cells for 24 h. Cell viability was determined using the MTT and data represented as mean  S.D. of three independent experiments. Significant cytoprotective activity of Sal B was observed dose dependently at concentrations above 1 mM ( p < 0.001) in all days of incubation with Ab.

Sal B is believed to have multiple therapeutic and preventive effects against human vascular diseases, including atherosclerosis (Lay et al., 2003a,b; Li et al., 2004, 2005; Chen et al., 2006; Zhang and Wang, 2006; Lam et al., 2006; Lin et al., 2007). Recently Sal B has been proposed to have a protective role in amyloid-induced toxicity (Tang and Zhang, 2001; Lin et al., 2006). However, the disaggregation effect of Sal B on the Ab aggregates has not yet been well characterized. Our results show, for the first time, that Sal B significantly ( p < 0.001) inhibits the formation of amyloid fibrils in a concentration- and time-dependent manner (Figs. 2 and 3). Sal B not only inhibits Ab1–40 aggregation but also decreases preformed aggregation at concentrations as low as 0.1 mM. Sal B converts fibrils to an

748

S.S.K. Durairajan et al. / Neurochemistry International 52 (2008) 741–750

amorphous structure as evidenced from EM and CD spectral studies, thereby supporting ThT fluorescence and ELISA data with a completely different method. These results suggest that the protective effect of Sal B may arise from anti-Ab fibrillogenesis in solution. The decrease in the efficacy of Sal B after 1 week of incubation with Ab1–40 might be due to oxidation or to some other chemical transformation of the polyphenolic group of Sal B that renders them less active. In comparison with ferulic acid, Sal B shows twofold less activity than ferulic acid after 1 week of incubation. Nonetheless, Sal B was shown to be a more effective inhibitor of Ab1–40 fibril formation at various concentrations than ferulic acid in 3 days of incubation (Fig. 3C and E; Table 1). Our IC50 values for ferulic acid’s inhibition of Ab1–40 fibril formation are nearly similar to the results of Ono et al. (2005), despite differences in the assay procedures. Like Sal B, ferulic acid is a phenolic antioxidant compound shown to be neuroprotective against Abinduced toxicity in primary neural culture (Sultana et al., 2005). Sal B also showed significant neuroprotection in differentiated SH-SY5Y cells against Ab1–42-induced toxicity at concentrations above 1 mM (Fig. 5) and the protection was found to be similar to that conferred by ferulic acid at concentrations of 10 and 50 mM (Sultana et al., 2005). In contrast curcumin is twofold more potent than Sal B in disaggregation and inhibition of Ab aggregation (Fig. 3D and F). The IC50 values of curcumin from this study are 4.11 and 2.58 for the disaggregation and inhibition of Ab1–40 aggregation, respectively (Table 1), and these values are not similar to the IC50 values from Yang et al. (2005). This contradiction might be due to the sensitivity of antibodies or detection method used in their ELISA system, 6E10 antibody which recognize 1–17 amino acid residue of native human Ab was employed by Yang et al. (2005), but in our experiment we used 10G4 and 10G4-b antibodies, which recognize 5–13 amino acid residues of Ab (Yang et al., 1994). Although curcumin is more active than Sal B, the oral availability of this drug is in question because a recent phase I clinical trial of curcumin in AD patients showed oral administration of curcumin neither decreased Ab accumulation nor improved cognitive impairment (Braun et al., 2007) despite its successful clearance in the Tg2567 mouse model (Yang et al., 2005). In the context of cell viability, the neuroprotective efficacy of Sal B seems to be similar to that of curcumin and ferulic acid because it significantly protects the neural cells as curcumin and ferulic does at concentrations above 1 mM (Kim et al., 2001; Sultana et al., 2005; Yang et al., 2005). The relative order of magnitude of the overall activity of Ab aggregate inhibitors from the ThT fluorescence assay and ELISA is curcumin > Sal B > ferulic acid. Sal B, ferulic acid and curcumin are structurally similar in that they are all caffeic acid derivatives with different categories of caffeic acid units (Fig. 1) (Kim et al., 2001; Butterfield et al., 2002; Jiang et al., 2005; Ono et al., 2005; Yang et al., 2005). Sal B is a tetramer of caffeic acid units whereas ferulic acid is a monomer and is one of the degradation products of curcumin (Jiang et al., 2005; Ono et al., 2005). The hydroxyl groups of caffeic acid have been shown to be essential for the pharmacological activity in heptoprotection (Perez-Alvarez et al., 2001).

EM images and CD spectra were used to investigate the possible structural changes induced by Sal B. The presence of relatively large insoluble amorphous particles in all the experimental solutions containing 10 and 100 mm Sal B indicates that these Ab1–40 aggregates represent sites of disturbed fibrillogenesis (Fig. 4). Fibril formation has been linked to structural transition in Ab peptides from random structures to organized b-sheet conformation (Kirschner et al., 1986). Sal B showed dose- and time-dependent structural conversion of b-sheet rich fibrillar Ab1–40 to random structure (Fig. 3), indicating the capability of initiating random structural links with inhibition of Ab fibril formation. Moreover, the toxicity of Ab has been associated with its ability to form bsheet secondary structures (Lorenzo and Yankner, 1994). Even though there is a debate on the neurotoxicity of Ab fibrils and oligomers, in our study we found that Sal B significantly attenuated Ab fibril-induced toxicity on different neural cell lines. In addition, we have also tested the effect of Sal B on the Ab oligomer formation using NU-1 oligomer specific antibody in a dot blot assay (Lambert et al., 2007). Sal B failed to inhibit the Ab oligomers formation in all tested concentrations, however curcumin showed significant inhibition of oligomer formation at concentrations above 10 mM (results not shown). Therefore, it seems Sal B only have anti-fibrillization effects without anti-oligomer activity. Although many Chinese herbal components have been tested, few are reported to have anti-amyloidogenic effects (Fujiwara et al., 2006). In particular, an extract of Uncaria rhynchophylla showed a significant destabilization of preformed Ab1–40 and Ab1–42 fibrils among the tested Chinese herbal components (Fujiwara et al., 2006). Similarly, in our experiment, Sal B significantly showed both inhibition of Ab1–40 fibril formation and destabilization of the preformed Ab1–42, even more than astragaloside-IV and tetramethylpyrazine (Fig. 2). Another well-known TCM plant Ginkgo biloba’s extract EGb761 provides a combination of antioxidative, anti-amyloidogenic, and anti-apoptotic effects (Luo et al., 2002). This plant extract has also been shown to block an age-dependent decline in spatial cognition in a transgenic mouse model of AD (Stackman et al., 2003). Because of the combined anti-oxidative and anti-amyloidogenic activities exhibited by Ginkgo extract for the improvement of AD, the same or similar effects may be attained by using other Chinese herbs that are a source of antioxidants such as Sal B. Like other phenolic antioxidants, Sal B has many hydroxyl groups (Fig. 1) that might interact with the peptide side chain to inactivate fibril aggregation. Further research is needed to elucidate the exact mechanism by which Sal B distorts the Ab fibrils. These newly found anti-amyloidogenic properties of Sal B support the hypothesis that it has neuroprotective effects with regard to Ab-induced activity. Because of the link between oxidative stress and AD, it was initially thought that the neuroprotective effects of Sal B were mostly due to its intracellular antioxidant activities (Tang and Zhang, 2001; Lin et al., 2006). However, the results presented here indicate that the anti-amyloidogenic properties are dependant on structural

S.S.K. Durairajan et al. / Neurochemistry International 52 (2008) 741–750

interactions of Sal B with Ab rather than on antioxidant properties exclusively. The results of our anti-Ab aggregation studies confirm that Sal B exhibits anti-amyloidogenic effects. The next step will be to confirm its efficacy in an animal model of AD; then it can proceed to human trials. We are interested to see whether the anti-amyloidogenic and neuroprotective effects of Sal B can influence Ab clearance and be helpful to overcome the memory deficits caused by Ab in AD. Acknowledgements This work was supported by research grants FRG/05-06/II24, FRG/06-07/I-07 and FRG/06-07/II-43 from Hong Kong Baptist University, and also partly supported by grant HKU 7636/05M from the Research Grant Council of Hong Kong to Dr. J.D. Huang and by grant EYS/05-06/01 from Eu Yan Sang (Hong Kong) Limited. 10G4 antibodies were donated by Drs. Fusheng Yang and Greg Cole, University of California, USA. We thank Dr. H.Z. Sun, Department of Chemistry, The University of Hong Kong, for his help in performing CD spectroscopy. References Braun, L., Mok, V., Cheung, K.K.S., 2007. Phase I clinical trial of curcumin for Alzheimer’s disease in Chinese Communities. In: World Congress on Ageing & Dementia in Chinese Communities. p. 1. Butterfield, D., Castegna, A., Pocernich, C., Drake, J., Scapagnini, G., Calabrese, V., 2002. Nutritional approaches to combat oxidative stress in Alzheimer’s disease. J. Nutr. Biochem. 13, 444–461. Butterfield, D.A., 2002. Amyloid beta-peptide (1–42)-induced oxidative stress and neurotoxicity: implications for neurodegeneration in Alzheimer’s disease brain, a review. Free Radic. Res. 36, 1307–1313. Chan, C.C., Lam, P.T., 2005. Update on dementia. Part 1. Mild cognitive impairment, screening and diagnostic assessment. HK Pract. 27, 235–241. Chen, Y.L., Hu, C.S., Lin, F.Y., Chen, Y.H., Sheu, L.M., Ku, H.H., Shiao, M.S., Chen, J.W., Lin, S.J., 2006. Salvianolic acid B attenuates cyclooxygenase-2 expression in vitro in LPS-treated human aortic smooth muscle cells and in vivo in the apolipoprotein-E-deficient mouse aorta. J. Cell Biochem. 98, 618–631. Datki, Z., Juhasz, A., Galfi, M., Soos, K., Papp, R., Zadori, D., Penke, B., 2003. Method for measuring neurotoxicity of aggregating polypeptides with the MTT assay on differentiated neuroblastoma cells. Brain Res. Bull. 62, 223–229. Fujiwara, H., Iwasaki, K., Furukawa, K., Seki, T., He, M., Maruyama, M., Tomita, N., Kudo, Y., Higuchi, M., Saido, T.C., Maeda, S., Takashima, A., Hara, M., Ohizumi, Y., Arai, H., 2006. Uncaria rhynchophylla, a Chinese medicinal herb, has potent antiaggregation effects on Alzheimer’s betaamyloid proteins. J. Neurosci. Res. 84, 427–433. Jiang, R.W., Lau, K.M., Hon, P.M., Mak, T.C., Woo, K.S., Fung, K.P., 2005. Chemistry and biological activities of caffeic acid derivatives from Salvia miltiorrhiza. Curr. Med. Chem. 12, 237–246. Kim, S., Park, S.Y., Kim, J.K., 2001. Curcuminoids from Curcuma longa L. (Zingiberaceae) that protect PC12 rat pheochromocytoma and normal human umbilical vein endothelial cells from Ab (1–42) insult. Neurosci. Lett. 303, 57–61. Kirschner, D.A., Abraham, C., Selkoe, D.J., 1986. X-ray diffraction from intraneuronal paired helical filaments and extraneuronal amyloid fibers in Alzheimer disease indicates cross-beta conformation. Proc. Natl. Acad. Sci. U.S.A. 83, 503–507. Lam, F.F., Yeung, J.H., Kwan, Y.W., Chan, K.M., Or, P.M., 2006. Salvianolic acid B, an aqueous component of danshen (Salvia miltiorrhiza), relaxes rat

749

coronary artery by inhibition of calcium channels. Eur. J. Pharmacol. 553, 240–245. Lambert, M.P., Velasco, P.T., Chang, L., Viola, K.L., Fernandez, S., Lacor, P.N., Khuon, D., Gong, Y., Bigio, E.H., Shaw, P., De Felice, F.G., Krafft, G.A., Klein, W.L., 2007. Monoclonal antibodies that target pathological assemblies of Abeta. J. Neurochem. 100, 23–35. Lay, I.S., Hsieh, C.C., Chiu, J.H., Shiao, M.S., Lui, W.Y., Wu, C.W., 2003a. Salvianolic acid B enhances in vitro angiogenesis and improves skin flap survival in Sprague–Dawley rats. J. Surg. Res. 115, 279–285. Lay, I.S., Chiu, J.H., Shiao, M.S., Lui, W.Y., Wu, C.W., 2003b. Crude extract of Salvia miltiorrhiza and salvianolic acid B enhance in vitro angiogenesis in murine SVR endothelial cell line. Planta Medica 69, 26–32. LeVine, H., 2004. Alzheimer’s b-peptide oligomer formation at physiologic concentrations. Anal. Biochem. 335, 81–90. Li, M., Zhao, C., Wong, R.N.S., Goto, S., Wang, Z.M., Liao, F.L., 2004. Inhibition of shear-induced platelet aggregation in rat by tetramethylpyrazine and salvianolic acid B. Clin. Hemorheol. Microcirc. 31, 97–103. Li, M., Zhao, M.Q., Zhang, H.X., Goto, S., Wu, W.K., Liao, F.L., 2005. Antiapoptosis effects of tetramethylpyrazine and salvianolic acid B on rat cerebral microvascular endothelial cells under high shear stress. Biorheology 42, 57. Lin, S.J., Lee, I.T., Chen, Y.H., Lin, F.Y., Sheu, L.M., Ku, H.H., Shiao, M.S., Chen, J.W., Chen, L., 2007. Salvianolic acid B attenuates MMP-2 and MMP-9 expression in vivo in apolipoprotein-E-deficient mouse aorta and in vitro in LPS-treated human aortic smooth muscle cells. J. Cell Biochem. 100, 372–384. Lin, Y.H., Liu, A.H., Wu, H.L., Westenbroek, C., Song, Q.L., Yu, H.M., Ter Horst, G.J., Li, X.J., 2006. Salvianolic acid B, an antioxidant from Salvia miltiorrhiza, prevents Abeta (25–35)-induced reduction in BPRP in PC12 cells. Biochem. Biophys. Res. Commun. 348, 593–599. Lleo´, A., Greenberg, S.M., Growdon, J.H., 2006. Current pharmacotherapy for Alzheimer’s disease. Ann. Rev. Med. 57, 513–533. Lorenzo, A., Yankner, B.A., 1994. b-Amyloid neurotoxicity requires fibril formation and is inhibited by Congo red. Proc. Natl. Acad. Sci. U.S.A. 91, 12243–12247. Luo, Y., Smith, J.V., Paramasivam, V., Burdick, A., Curry, K.J., Buford, J.P., Khan, I., Netzer, W.J., Xu, H., Butko, P., 2002. Inhibition of amyloid b aggregation and caspase-3 activation by the Ginkgo biloba extract EGb761. Proc. Natl. Acad. Sci. U.S.A. 99, 12197–12220. Munoz-Ruiz, P., Rubio, L., Garcia-Palomero, E., Dorronsoro, I., del MonteMillan, M., Valenzuela, R., Usan, P., de Austria, C., Bartolini, M., Andrisano, V., Bidon-Chanal, A., Orozco, M., Luque, F.J., Medina, M., Martinez, A., 2005. Design, synthesis, and biological evaluation of dual binding site acetylcholinesterase inhibitors: new disease-modifying agents for Alzheimer’s disease. J. Med. Chem. 48, 7223–7233. Ono, K., Hirohata, M., Yamada, M., 2005. Ferulic acid destabilizes preformed beta-amyloid fibrils in vitro. Biochem. Biophys. Res. Commun. 336, 444–449. Ono, K., Hamaguchi, T., Naiki, H., Yamada, M., 2006. Anti-amyloidogenic effects of antioxidants: implications for the prevention and therapeutics of Alzheimer’s disease. Biochem. Biophys. Acta 1762, 575–586. Perez-Alvarez, V., Bobadilla, R.A., Muriel, P., 2001. Structure-hepatoprotective activity relationship of 3,4-dihydroxycinnamic acid (caffeic acid) derivatives. J. Appl. Toxicol. 21, 527–531. Reddy, P.H., 2006. Amyloid precursor protein-mediated free radicals and oxidative damage: implications for the development and progression of Alzheimer’s. J. Neurochem. 96, 1–13. Selkoe, D.J., 2004. Cell biology of protein misfolding: the examples of Alzheimer’s and Parkinson’s diseases. Nat. Cell Biol. 6, 1054–1061. Sreerama, N., Woody, R.W., 2004. Computation and analysis of protein circular dichroism spectra. Methods Enzymol. 383, 318–351. Stackman, R., Eckenstein, F., Frei, B., Kulhanek, D., Nowlin, J., Quinn, J.F., 2003. Prevention of age-related spatial memory deficits in a transgenic mouse model of Alzheimer’s disease by chronic Ginkgo biloba treatment. Exp. Neurol. 18, 510–520. Sultana, R., Ravagna, A., Mohmmad-Abdul, H., Calabrese, V., Butterfield, D.A., 2005. Ferulic acid ethyl ester protects neurons against amyloid

750

S.S.K. Durairajan et al. / Neurochemistry International 52 (2008) 741–750

beta-peptide (1–42)-induced oxidative stress and neurotoxicity: relationship to antioxidant activity. J. Neurochem. 92, 749–775. Tang, M., Zhang, J., 2001. Prostate apoptosis response-4 involved in the protective effect of salvianolic acid B against amyloid beta peptide-induced damage in PC12 cells. Jpn. J. Pharmacol. 88, 422–427. Yang, F., Mak, K., Vinters, H.V., Frautschy, S.A., Cole, G.M., 1994. Monoclonal antibody to the C-terminus of b-amyloid. NeuroReport 5, 2117–2120. Yang, F., Lim, G.P., Begum, A.N., Ubeda, O.J., Simmons, M.R., Ambegaokar, S.S., Chen, P.P., Kayed, R., Glabe, C.G., Frautschy, S.A., Cole, G., 2005.

Curcumin inhibits formation of amyloid beta oligomers and fibrils, binds plaques, and reduces amyloid in vivo. J. Biol. Chem. 280, 5892–5901. Zhang, H.S., Wang, S.Q., 2006. Salvianolic acid B from Salvia miltiorrhiza inhibits tumor necrosis factor-alpha (TNF-alpha)-induced MMP-2 upregulation in human aortic smooth muscle cells via suppression of NAD(P)H oxidase-derived reactive oxygen species. J. Mol. Cell Cardiol. 41, 38–48. Zhou, L., Zuo, Z., Chow, M.S., 2005. Danshen: an overview of its chemistry, pharmacology, pharmacokinetics, and clinical use. J. Clin. Pharmacol. 45, 1345–1359.