In vitro screening on amyloid beta modulation of aqueous extracts from plant seeds.

INTRODUCTION: Glycation process might contribute to both extensive protein cross-linking and oxidative stress in Alzheimer's disease (AD). The amyloid-like aggregation of glycated bovine serum albumin induces apoptosis in the neuronal cell. Dietary supplementation of antioxidants, vitamins, and polyphenols are beneficial to AD, and consumption of fruits and vegetables reduce the risk of AD. We conducted a screening of 14 aqueous extracts from plant seeds (PSAE) for inhibitory activity on amyloid beta (Aβ).
MATERIALS AND METHODS: To examine the effects of PSAE on the Aβ (1-42) concentration, PSAE were analyzed by Aβ (1-42) enzyme-linked immunosorbent assay. Furthermore, we carried out an antiglycation experiment of PSAE and an antiaggregation experiment of PSAE to confirm the modification mechanism of PSAE. PSAE were added to buffer containing D-ribose and albumins. The solutions were incubated at 37°C for 10 days. After incubation, the products were assayed on a fluorophotometer.
RESULTS: PSAE associated differential reduction in the levels of Aβ (1-42) (lettuce; 98.7% ± 2.4%, bitter melon; 95.9% ± 2.6%, and corn; 93.9% ± 2.1%), demonstrating that treatment with lettuce seeds extracts (LSE) effectively decreases Aβ (1-42) concentration. Among the 14 PSAE, LSE exhibited the second greatest potential for antiglycation. Inhibition of aggregates was not recognized in LSE.
CONCLUSION: These results suggest that LSE reduces the toxicity of Aβ by modifying Aβ.

One of the pathogenic hypotheses to explain Alzheimer's disease (AD) considers aggregated amyloid beta (Aβ),[1] to be the molecular triggers of a cascade of events causing the observed neuronal loss.[2] The major component of peptides is Aβ1–40 and Aβ1–42.[3] Mold et al. reported that the aggregation and deposition of Aβ1−42 in the brain is implicated in the etiology of AD.[4] Therefore, Aβ1−42 is very important in the research of AD. The aggregation of Aβ peptides starts with changes in their secondary structure leading to β-sheet formation, progresses with aggregation of the misfolded peptides into oligomers and culminates in the production of amyloid fibers that precipitate into the brain forming amyloid plaques. Aβ can be neurotoxic by a mechanism linked to peptide fibril formation. However, the mechanism by which Aβ produces brain dysfunction in patients with AD is largely unknown.

Edible plants are a source of various biologically active phytomolecules, including phenolic compounds, flavonoids, and vitamins.[5] The health properties of edible plants depend on its bioactive compounds and especially on polyphenols.[5] Much research has assessed the dietary role of polyphenols, and their characteristics, metabolic pathways, and biological effects.[6] Plant-derived polyphenols have been widely used to scavenge reactive oxygen species and treat a variety of diseases including AD.[7] We propose that extracts of plant seeds have significant antioxidant activity and protective effects against inflammation,[8] and suspect the extract might be useful in preventing AD.

Recently, it was proposed that the glycation process might contribute to both extensive protein cross-linking and oxidative stress in AD.[9] Nonenzymatic protein glycation is an endogenous process in which reducing the sugars that react with amino groups in proteins through a series of Maillard reactions forming reversible Schiff-base and Amadori compounds, produces a heterogeneous class of molecules, and collectively termed advanced glycation end products.[1011] A previous study discussed the globular amyloid-like deposits of D-ribose glycation of bovine serum albumin (BSA) aggregates. The amyloid-like aggregation of glycated BSA induces apoptosis in the neuronal cell. D-ribose saccharifies BSA, which then misfolds rapidly and forms globular amyloid-like aggregations, which play an important role in cytotoxicity of neuronal cells.[12] To this concern, in the present study we have investigated the effect of glycation on the aggregation pathways of BSA and lactalbumin (LAB). Although this reaction may not be directly related to an amyloid disease, it is thought to be a good representative model of proteins that intrinsically evolve toward the formation of amyloid aggregates. Therefore, we conducted a large-scale screening of 14 aqueous extracts from plant seeds (PSAE) for inhibitory activity on this reaction.

Materials and Methods

Preparation of aqueous extracts from plant seeds

The 14 types of plant seeds were collected from a market in Japan [Table 1]. The preparation of PSAE was performed using our method as previously explained.[13] PSAE was stored at under −20°C until use. When used in assays, each sample was returned to ambient temperature, followed by filtration through a membrane filter (pore size 0.2 μm).

Table 1

Yield and TPC of PSAE extracted from seeds

Determination of total phenolic content

We measured total phenolic content (TPC) with a modified version of the Folin–Ciocalteu method[13] using 0–30 mg/L chlorogenic acid as a standard. The determination of TPC was done with the colorimetric method as previously described.[13] Briefly, 100: L of sample or standard was combined with 100: L of Folin-Ciocalteu reagent and 100: L of 2% Na2CO3 solution. We allowed the mixture to sit for 60 min before reading absorbance at 750 nm using an Ultrospec Visible Plate Reader II 96 (GE Healthcare Ltd., England) and calculating the concentration of PSAE as chlorogenic acid equivalents per gram of dried plant seed.

Assessment of amyloid beta (1–42) concentration

Levels of Aβ (1–42) in mixtures (10 µM Aβ (1–42) 55 μL and PSAE 55 μL) were determined with human enzyme-linked immunosorbent assay (ELISA) assay (Wako, Osaka, Japan). The mixtures (110 μL) were added to microplate wells. The mixtures were then incubated at room temperature for 24 h. After incubation, we obtained 100 μL of sample, and distributed some of the samples into each well coated with the human Aβ. Detected Aβ (1–42) was analyzed. The control ratio (%) was expressed as a percentage of the untreated control as follows: % Control ratio = (A450 nm of treated cells/A450 nm of untreated cells) ×100.

In vitro glycation of bovine serum albumin and lactalbumin induced by D-ribose

After sterilization, using a Millex GV filter (Millipore, Cork, Ireland) to prevent bacterial growth, BSA and LAB were dissolved in 20 mM Tris-HCl (pH 7.4) to yield a stock solution of 20 mg/mL. PSAE (4 or 8 µL) were added to Tris-HCl containing 1M D-ribose and either BSA (167 µL) or LAB (167 µL) to acquire final concentrations of 10 mg/mL. Then, the solutions were incubated at 37°C for up to 10 days. After incubation, the fluorescent reaction products were assayed on a fluorophotometer (λex 360 nm/λem 465 nm; multimode microplate reader infinite F200,[12] Tecan Trading AG, Switzerland). BSA or LAB, in the presence of D-ribose, was used as a control. Each experimental condition was performed in triplicate.

In vitro aggregation of bovine serum albumin and lactalbumin induced by D-ribose

PSAE (4 or 8 µL) were added to 20 mM Tris-HCl containing 1M D-ribose and either BSA (167 µL) or LAB (167 µL) to acquire final concentrations of 10 mg/mL. The solutions were then incubated at 37°C for up to 10 days. After incubation, thioflavin T (ThT, 30 µM), commonly used to detect protein aggregations, was added to the solution to investigate whether any amyloid-like deposits formed at 37°C. After incubation for 10 min, the fluorescent reaction products were assayed on a fluorophotometer (λex 430 nm/λem 465 nm). BSA or LAB, in the presence of D-ribose, was used as a control.

Statistical analysis

We present all data as the mean ± standard deviation of the three measurements. A statistical comparison between the groups was carried out using either ANOVA or Student's t-test. P <0.05 was considered as statistically significant.

Results

Yield and total phenolic content estimation

Yield was expressed as a gram of dry matter/mL. Yield was in the range of 0.137–0.246 g/mL. TPC was expressed as mg of chlorogenic acid equiv/g of dry matter [Table 1]. Significant differences were observed for TPC among the 14 plant seed varieties. TPC was in the range of 0.74–3.32 mg. Crown daisy shows the highest phenolic content (3.32 mg) while the lowest content was observed in Bitter melon (0.74 mg).

Lettuce seeds extracts reduces amyloid beta (1–42) concentration

To examine the effects of PSAE on the Aβ (1–42) concentration, PSAE mixtures were analyzed by Aβ (1–42) ELISA. Figure 1 illustrates the PSAE associated differential reduction in the levels of Aβ (1–42) (lettuce; 98.7% ± 2.4%, bitter melon; 95.9% ± 2.6%, corn; 93.9% ± 2.1%, and crown daisy; 68.3% ± 1.9%), demonstrating that treatment with lettuce seeds extracts (LSE) effectively decreases Aβ (1–42) concentration. LSE treatment demonstrated the strongest Aβ-inhibition potential.

Figure 1

Inhibitory effects of aqueous extracts from plant seeds on amyloid beta (1–42). Levels of amyloid beta (1–42) in mixtures (10 μM amyloid beta (1–42) 55 μL and aqueous extracts from plant seeds 55 μL) were determined with human enzyme-linked immunosorbent assay. The mixtures (110 μL) were incubated at room temperature for 24 h. After incubation, amyloid beta (1–42) was analyzed. The A450 nm of amyloid beta (1–42) treatment and amyloid beta (1–42) + aqueous extracts from plant seeds treatments are indicated by unshaded and shaded columns, respectively. Data represents the percentage of amyloid beta (1–42) + aqueous extracts from plant seeds treated cells relative to amyloid beta (1–42) treated cells: % Control ratio = (A450 nm of amyloid beta (1–42) + aqueous extracts from plant seeds treated cells/A450 nm of amyloid beta (1–42) treated cells) ×100. Values are the mean ± standard deviation of the three measurements. **P < 0.01, *P < 0.05 compared with the controls

On the other hand, the correlations between Aβ (1–42) levels and yield and TPC were analyzed. The correlation coefficient for Aβ (1–42) levels was found to be smaller than 0.2 (r = 0.388) which proved that the yield could not be attributed to the Aβ (1–42) levels of the 14 PSAE. No significant correlation (r = 0.121) was observed between Aβ (1–42) levels and TPC among our experimental samples. These results proved that the yield and TPC of these plants could not be clearly attributed to their Aβ-inhibition potential.

Effects of aqueous extracts from plant seeds on D-ribose induced glycation of bovine serum albumin or lactalbumin

In this study, we investigated the effects of PSAE on glycation of BSA or LAB induced by D-ribose [Figure 2]. Inhibition of glycation was recognized in PSAE samples except for in 4 µL Bell pepper in BSA and Japanese radish in LAB. Among the 14 PSAE, crown daisy seeds extracts (CDSE) demonstrated the strongest antiglycation potential. LSE exhibited the second greatest potential for antiglycation. In Figure 2a, fluorescence assay results showed that BSA glycation levels significantly decreased in the CDSE-loaded treatments relative to the control; inhibition of BSA glycation by 4 and 8 μL CDSE decreased 46.2% ± 2.85% and 58.9% ± 3.39%, respectively. Similarly, BSA glycation levels significantly decreased in the LSE-loaded treatments relative to the control; inhibition of BSA glycation by 4 and 8 μL LSE decreased 45.9% ± 2.68% and 50.5% ± 2.79%, respectively.

Figure 2

Changes in the fluorescence of bovine serum albumin or lactalbumin + D-ribose treated with aqueous extracts from plant seeds. Bovine serum albumin or lactalbumin (final concentration 10 mg/mL) in the presence of D-ribose (final concentration 1M) was kept at 37°C in Tris-HCl buffer (pH 7.4). Aqueous extracts from plant seeds (4 μL: Blue columns and 8 μL: Yellow columns) was mixed with samples of bovine serum albumin (a) or lactalbumin, (b) +D-ribose for up to 10 days. The fluorescence intensity of glycation was recorded (λex 360 nm; λem 465 nm). Bovine serum albumin (or lactalbumin) and D-ribose were used as a control. Aliquots were taken for measurements of fluorescence spectra (λex = 360 nm). Values are the mean ± standard deviation of the three measurements. **P < 0.01, *P < 0.05 compared with the controls

The correlations between BSA glycation inhibition and yield and TPC were analyzed. The correlation coefficient for BSA glycation inhibition was found to be smaller than 0.2 (4 µL PSAE; r = 0.119, 8 µL PSAE; r = 0.0490) which proved that the yield could not be attributed to BSA glycation inhibition at sample concentrations of 4 or 8 μL. Similarly, no significant correlations were observed between BSA glycation inhibition and TPC among our experimental samples. Neither the TPC level in 4 µL PSAE (r = 0.171) nor the TPC level in 8 µL PSAE (r = 0.423) correlated with BSA glycation inhibition. These results proved that the yield and TPC of these plants could not be clearly attributed to the BSA glycation inhibition of the 14 PSAE.

In Figure 2b, fluorescence assay results showed that LAB glycation levels significantly decreased in the CDSE-loaded treatments relative to the control; inhibition of LAB glycation by 4 and 8 μL CDSE decreased 41.3% ± 2.85% and 43.3% ± 3.39%, respectively. Similarly, fluorescence assay results showed that LAB glycation levels significantly decreased in the LSE-loaded treatments relative to the control; inhibition of LAB glycation by 4 and 8 μL LSE decreased 36.7% ± 2.07% and 40.6% ± 2.19%, respectively. The correlations between LAB glycation inhibition and yield and TPC were analyzed. The correlation coefficient for LAB glycation inhibition was found to be smaller than 0.2 (4 µL PSAE; r = 0.175, 8 µL PSAE; r = 0.184) which proved that the yield could not be attributed to the LAB glycation inhibition of the 14 PSAE. Neither TPC level in 4 µL PSAE (r = 0.128) nor TPC level in 8 µL PSAE (r = 0.278) correlated with LAB glycation inhibition. These results proved that the yield and TPC of these plants could not be clearly attributed to their LAB glycation inhibition.

Effects of aqueous extracts from plant seeds on aggregates of D-ribose-glycated bovine serum albumin or lactalbumin

We added ThT to test whether PSAE are inhibitors of amyloid-like aggregates [Figure 3a]. Fluorescence of ThT significantly increased in the presence of BSA incubated with D-ribose. Fluorescence intensity showed about 32266 ± 2748 counts in BSA + D-ribose. Inhibition of aggregates was not recognized in PSAE samples except for in 4 µL Rapeseed, 4 µL Bitter melon, and 8 µL Japanese Honeywort in LAB. Among the 14 PSAE, Japanese Honeywort seeds extracts (4 μL JHSE decreased 12.5%) demonstrated the strongest anti-aggregates potential. The correlations between LAB aggregation inhibition and yield and TPC were analyzed. The correlation coefficient for LAB aggregation inhibition was found to be smaller than 0.1 (4 µL PSAE; r = 0.0047, 8 µL PSAE; r = 0.0026) which proved that the yield could not be attributed to the LAB aggregation inhibition of the 14 PSAE. Neither TPC level in 4 µL PSAE (r = 0.0200) nor TPC level in 8 µL PSAE (r = 0.114) correlated with LAB aggregation inhibition. These results proved that the yield and TPC of these plants could not be clearly attributed to their LAB aggregation inhibition.

Figure 3

Changes in the thioflavin T fluorescence of bovine serum albumin or lactalbumin + D-ribose treated with aqueous extracts from plant seeds. Bovine serum albumin or lactalbumin (final concentration 10 mg/mL) in the presence of D-ribose (final concentration 1M) was kept at 37°C in Tris-HCl buffer (pH 7.4). Thioflavin T (final concentration 30 μM) was mixed with samples of bovine serum albumin (a) or lactalbumin, (b) +D-ribose + aqueous extracts from plant seeds (4 μL: Blue columns and 8 μL: Yellow columns), as described in materials and methods. The fluorescence intensity of thioflavin T was recorded (λex 430 nm; λem 465 nm). Bovine serum albumin (or lactalbumin) and D-ribose were used as a control. Aliquots were taken for measurements of fluorescence spectra (λex = 430 nm). Values are the mean ± standard deviation of the three measurements. *P < 0.05 compared with the controls

Discussion

As a result of having examined modifications to Aβ using Aβ (1–42) ELISA experiments with 14 samples of edible plant seed, LSE showed the most remarkable results [Figure 1]. In other words, LSE may interact with Aβ. Therefore, we examined the Aβ modification mechanism of LSE.

Yao et al.[14] showed that the terpenoid and flavonoid constituents of the Ginkgo biloba extract EGb 761 (EGb) are responsible for rescuing the neuronal cells from Aβ-induced cell death; their mechanism of action being distinct from their antioxidant properties. Because pre- and post-treatment with EGb did not protect the cells from Aβ-induced neurotoxicity, they examined whether EGb interacts directly with Aβ. In vitro reconstitution studies demonstrated that EGb inhibits the formation of Aβ-derived diffusible neurotoxic soluble ligands, suggested to be involved in the pathogenesis of AD.

There was report revealing the presence of terpenoids (diterpenoid) in the lettuce seeds.[15] In addition, Ajibade reported that phytochemical screening revealed the presence of terpenes and flavonoids in the extract of the seeds (Moringa oleifera).[16] Judging from the mechanism of G. biloba extract, it is likely that LSE might act in similar ways to G. biloba extract.

A full understanding of the pathogenesis of AD has remained elusive, and an increasing amount of evidence is confirming that AD is a disease with numerous contributing factors. It has been proposed that a chemical process known as glycation may contribute to extensive protein cross-linking in AD.[17] Nonenzymatic protein glycation is an endogenous process in which reducing sugars react with amino groups in proteins through a series of Maillard reactions forming reversible Schiff-base. These reactions increase the misfolding of proteins such as Aβ in AD. Thus, glycation may have a causal role in AD.

Taking the statement above into account, we examined the glycation inhibition ability of PSAE. In this study, we used the modification model of BSA or LAB and D-ribose.[12] We considered whether PSAE might be inhibiting BSA or LAB glycation. Our results demonstrated that some PSAE efficiently inhibited glycation of BSA or LAB. As a result, we found LSE significantly inhibited glycation of BSA in concentration-dependence [Figure 2a]. A similar result was provided for the different protein, LAB [Figure 2b]. On the other hand, significant results were not evident in BSA or LAB aggregation except for in Rapeseed, Bitter melon, and Japanese Honeywort in LAB [Figure 3a and b]. LSE had the ability to inhibit glycation but did not have the ability to inhibit aggregation of the protein. In other words, reactions to protein and sugar were inhibited by LSE, but LSE did not inhibit aggregation of the cross-linked, structure producing protein. It is thought likely that in the stage before the protein is modified by sugar, LSE bind to the protein, and disturbs protein modification by sugar. In other words, we propose that the sugar-protein binding site may also be the binding site for the LSE-protein and that LSE may compete with D-ribose. Therefore, LSE may contribute to a meaningful delay in the pathological progress of diseases such as AD. In conclusion, LSE reduces the toxicity of Aβ by modifying Aβ.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.