Ferulic Acid Suppresses Amyloid β Production in the Human Lens Epithelial Cell Stimulated with Hydrogen Peroxide

It is well known that oxidative stresses induce the production of amyloid β (Aβ) in the brain, lens, and retina, leading to age-related diseases. In the present study, we investigated the effects of ferulic acid on the Aβ levels in H2O2-stimulated human lens epithelial (HLE) SRA 01/04 cells. Three types of Aβ peptides (Aβ1-40, Aβ1-42, and Aβ1-43) were measured by ELISA, and the levels of mRNA for the expressed proteins related to Aβ production (APP, BACE1, and PS proteins) and degradation (ADAM10, NEP, and ECE1 proteins) were determined by quantitative real-time RT-PCR. H2O2 stimulation augmented gene expression of the proteins related to Aβ production, resulting in the production of three types of Aβ peptides. Treatment with 0.1 μM ferulic acid attenuated the augmentations of gene expression and production of the proteins related to the secretion of three types of Aβ peptides in the H2O2-stimulated HLE cells. These results provided evidence of antioxidative functions of ferulic acid for lens epithelial cells.


Introduction
A sequential proteolytic processing of amyloid precursor protein (APP) cleaved by -secretase ( site APP cleaving enzyme, BACE1) [1] and -secretase [a presenilin complex (PS)] [2] leads to the production of amyloid (A ) peptides. The cleavage of APP at different positions by -secretase mainly produces three A peptides, A 1-40 , A 1-42 , and A 1-43 peptides [3][4][5][6][7]. The most abundant variant of A is A 1-40 , but A 1-42 and A 1-43 , the longer forms, are more neurotoxic than A  . In addition, the neurotoxicity of A 1-43 is higher than that of A 1-42 , although the accumulation of A 1-43 is lower than that of A 1-42 [7]. However, the APP cleaved by -secretase (a disintegrin and metalloprotease domain protease 10, ADAM10) generates nonamyloidogenic soluble APP [8][9][10]. Neprilysin (NEP) [11] and endothelin converting enzyme (ECE) [12] degrade the A peptide. The production of A is a normal process; however, the overproduction of A , or an increased proportion of A 1-42 and A 1-43 , appears to cause an early onset of cataract and Alzheimer's disease [13,14].
Cataract represents a disease of increasing lens opacity, and it has been reported that A accumulation in the human lens causes lens opacification [15][16][17][18]. In addition, there have been several reports that oxidative stress can increase the accumulation and toxicity of A peptides in the lens, retina, and brain and that enhanced A accumulation leads to oxidative stress [14,[19][20][21][22][23][24]. Therefore, it is expected that drugs that inhibit the effect of A peptides and lower oxidative stress can be used as anticataract eye drops.
Ferulic acid (4-hydroxy-3-methoxycinnamic acid) is a widely distributed constituent of plants and is used as an antioxidant to prevent the oxidation of substrates such as lipids, proteins, DNA, and carbohydrates [25,26]. Ferulic acid has been shown to possess some scavenging activity toward hydroxyl radicals and peroxynitrite [27,28]. Recently, several studies have shown that ferulic acid has beneficial effects against Alzheimer's disease [29][30][31], diabetes [32], and cancer [33]. Alzheimer's disease is a neurodegenerative disorder, and it had been shown that one of the main pathophysiological features of Alzheimer's disease was the presence of extracellular senile plaques consisting essentially of A , a peptide thought to be a leading cause of neurotoxicity [34]. It was reported that ferulic acid could be a suitable molecule to specifically bind to A and inhibit fibril formation [35]. Furthermore, its compact structure could also be used for specific interactions with A mature fibrils, thereby possibly promoting their destabilization [35]. Ferulic acid inhibits A aggregation, destabilizes preformed A fibrils in vitro [36], and protects cultured neuronal cells against A -induced cytotoxicity [29]. Moreover, long-term administration of ferulic acid protects mice against A -induced learning and memory deficits in vivo [30,31]. Therefore, ferulic acid may be a useful molecule for the inhibition of A production and accumulation.
In this study, we investigated the changes in A production and accumulation in human lens epithelial (HLE) SRA 01/04 cells stimulated with hydrogen peroxide (H 2 O 2 ). In addition, we demonstrated the preventive effects of ferulic acid on A production and accumulation in HLE cells.

Cell Culture and Treatment.
The HLE cell line SRA 01/04 was incubated in Dulbecco's Modified Eagle's Medium (DMEM) containing 10 g/L gentamicin and 10% (v/v) fetal bovine serum (FBS) under humidified air containing 5% CO 2 , at 37 ∘ C. Each treatment was usually carried out on the third day after seeding of 0.4 × 10 4 cells/cm 2 (at approximately 80% confluency), and the culture medium was changed to non-FBS medium 1 hour before each experiment. The experiment was initiated by changing the medium to fresh medium containing 0-100 M of H 2 O 2 and the HLE cells were cultured for 24 hours to produce A 1-40 , A 1-42 , and A 1-43 (H 2 O 2 -stimulated HLE cells). Ten M of H 2 O 2 and 0-30 M of ferulic acid included in the medium were used for 24-hour cultures to observe the effect of ferulic acid on A production (H 2 O 2 -stimulated HLE cells in the presence of ferulic acid).

Measurement of A .
The measurements of A 1-40 , A 1-42 , and A 1-43 levels were performed as previously described [14,38]. Approximately 2.5 × 10 6 HLE cells were collected with a cell scraper (Asahi Glass, Tokyo, Japan), homogenized in 250 L of 70% formic acid, and then centrifuged at 9,100 ×g for 15 min at 4 ∘ C. The supernatants were added to 4.75 mL of 1 M Tris base, and the mixtures were used for the measurements of the A 1-40 , A 1-42 , and A 1-43 peptides. The A 1-40 , A 1-42 , and A 1-43 levels were measured using the human amyloid (40) ELISA kit (dynamic range, 1-100 pM, Wako, Osaka, Japan), the human amyloid (42) ELISA kit (dynamic range, 0.1-20 pM, Wako), and the human amyloid (1-43) (FL) ELISA kit (dynamic range, 0.51-32.5 pM, IBL, Gunma, Japan) according to the manufacturer's instructions, respectively. The A levels were expressed as pmol/g of protein according to our previous reports. The protein levels in the samples used to determine A 1-40 , A 1-42 , and A 1-43 were assessed using a Bio-Rad Protein Assay kit (Bio-Rad, Hercules, CA, USA).

Measurement of Cell Viability. HLE cells with or without
added H 2 O 2 and ferulic acid treatments were incubated in 96-well microplates (Iwaki, Chiba, Japan). The viability of the HLE cells was calculated by the Cell Count Reagent SF kit (Nacalai Tesque, Kyoto, Japan) according to the manufacturer's instructions. The absorbance at 450 nm was measured, and the cell viability (%) was represented as the percentage of the absorbance measured for unstimulated HLE cells for each point (normal cells).

Statistical Analyses.
All values were presented as mean ± standard error (SE), and statistical differences were evaluated by one-way analysis of variance followed by Dunnett's multiple comparisons. P values of less than 0.05 were considered significant.    Figure 4 shows the effects of ferulic acid on the mRNA levels for proteins related to A production (APP, BACE1, PS1, and PS2) and degradation (ADAM10, NEP, and ECE1) in the H 2 O 2 -stimulated HLE cells. Ferulic acid decreased the mRNA levels of APP, BACE1, and PS1. However, the mRNA levels of PS2 and the mRNA levels for proteins related to A degradation (ADAM10, NEP, and ECE1) in HLE cells treated with ferulic acid were similar to the HLE cells without ferulic acid treatment. and the mRNA levels of proteins related to A production (APP, BACE1, PS1, and PS2) and degradation (ADAM10, NEP, and ECE1) were similar to those of the naïve HLE cells. On the other hand, the A 1-42 levels in HLE cells, treated with 0.1 M ferulic acid, were significantly lower than that without ferulic acid.

Discussion
In the cataractous lens, an accumulation of A peptide was observed, and lens opacification occurred via oxidative stress [14,15]. Therefore, the prevention of A accumulation in the lens is important for cataract therapy. In this study, we investigated whether treatments with ferulic acid prevented the A production and accumulation in the human lens and showed that ferulic acid attenuated the increase in the mRNA levels of proteins related to A production (APP, BACE1, and PS1) and prevented the accumulation of A 1-40 , A 1-42 , and A 1-43 in H 2 O 2 -stimulated HLE cells. Previous reports showed that the A levels in the retina and brain were enhanced by oxidative stress [19][20][21][22][23][24]. In addition, we also reported that H 2 O 2 in the rat lens induced lipid peroxidation and led to the accumulation of A  in the lens epithelium [14]. Therefore, in our current study, the HLE cells were exposed to H 2 O 2 to enhance the A production. The results clearly suggested that 10 M of H 2 O 2 induced A production, whereas 0.01-0.1 M of ferulic acid significantly decreased the mRNAs (APP, BACE1, and PS1) related to A production ( Figure 4)  reported that oxidative stress increased the accumulation of A in the lens, retina, and brain [14,[19][20][21][22][23][24]. In addition, ferulic acid has been shown to possess some scavenging activity for hydroxyl radicals and peroxynitrite [27,28] and to inhibit the oxidation of substrates. Taken together, our results suggested that 0.1 M and 1 M of ferulic acid may prevent H 2 O 2 effects, resulting in a decrease of specific mRNAs and in A accumulation. However, the mRNA levels of proteins related to A production (APP, BACE1, and PS) in HLE cells incubated with 30 M of ferulic acid were similar to that of the H 2 O 2 -stimulated HLE cells incubated with 0.1 M of ferulic acid ( Figure 4). However, the A 1-40 , A 1-42 , and A 1-43 levels in the HLE cells stimulated with 30 M of ferulic acid were elevated when compared with those of the H 2 O 2stimulated HLE cells incubated with 0.1 M of ferulic acid ( Figure 3). It was reported that ferulic acid undergoes specific interactions with A mature fibrils, possibly promoting their destabilization [35]. Moreover, ferulic acid inhibited A aggregation and destabilized preformed A fibrils [36]. From these findings, the destabilization of A fibrils by ferulic acid may be related to the difference in the levels of A in the HLE cells treated with 0.1 M and 30 M of ferulic acid. In this study, the A monomer in/on the cells would be measured, since the HLE cells were collected, and all protein fractions were precipitated with 70% formic acid. We tried to measure the A levels in medium; however the A levels were not detected by using ELISA method. We hypothesize that both the fibrils and monomeric A are present in the medium but were below the detection level of the ELISA assay. For the certification of this hypothesis, there is a need to measure the levels of A in the medium by using high resolution mass spectrometry in future studies. On the other hand, 0.1 M of ferulic acid shows different effects between A 1-40 and A 1-42 levels ( Figure 5). It was known that the A 1-42 was easy to aggregate in comparison with A 1-40 ; however, the amount of A 1-42 production was lower than that of A 1-40 [7]. Therefore, the destabilization of A aggregation by ferulic acid may strongly affect the A 1-42 in comparison with A 1-40 . Further studies are needed to elucidate the precise mechanisms for the prevention of A production and the accumulation by ferulic acid in the lens. In addition, it is important to measure the ratio of A fibrils and A monomers to clarify the molecular mechanism involved in the downregulation of APP, BACE1, and PS1 mRNAs in ferulic acid treated cells. Therefore, we are now investigating the protein expression and aggregation of A in H 2 O 2 -stimulated HLE cells incubated with 0.1 M and 30 M of ferulic acid by using western blotting and immunostaining methods. Moreover, we are developing a drug delivery system for ferulic acid to the lens. In the future, we will determine the in vivo effects of a ferulic acid ophthalmic formulation on lens opacification and A accumulation in the UPL rat model.

Conclusions
We have shown that stimulation with H 2 O 2 leads to increased mRNA levels of proteins related to A production (APP, BACE1, and PS) and to an enhanced accumulation of A

Conflicts of Interest
The authors declare no conflicts of interest. The authors alone are responsible for the content and writing of the paper.