Enhanced antiamyloidal activity of hydroxy cinnamic acids by enzymatic esterification with alkyl alcohols
Hazuki Kondo Haruka Sugiyama Shigeru Katayama∗ Soichiro Nakamura∗
Department of Bioscience and Biotechnology, Shinshu University, Minamiminowa, Nagano, Japan
Abstract Lipophilic derivatives of hydroxyl cinnamic acids (HCAs) including caffeic acid (CA), ferulic acid, sinapic acid (SA), and chlorogenic acid were synthesized by esterification with butanol, octanol, or hexadecanol catalyzed by the lipase from Candida antarctica to investigate the effect of lipophilicity on their antiamyloidal activity assessed by the inhibitory activities toward fibrillization of amyloid β (Aβ) peptide. Among them, CA showed the highest activity at 50 μM, reducing the amyloid fibril formation of Aβ to 34.4 ± 6.8%. The antiamyloidal effects of HCAs were enhanced by esterification with alkyl alcohols,
and the longer alkyl chain tended to be more effective except for SA. Aβ fibril formation was suppressed by the hexadecyl ester of CA, which was reduced to 8.8 ± 2.3%. In contrast, those of octyl and butyl esters were 19.3 ± 2.3% and 41.6 ± 6.1%, respectively. These results show that lipophilicity plays an important role in the antiamyloidal activities of esterified phenolic compounds. C 2013 International Union of Biochemistry and Molecular Biology, Inc. Volume 61, Number 4, Pages 401–407, 2014
Keywords: amyloid β, antiamyloidal activity, esterification, hydroxyl cinnamic acids, lipase, lipophilicity
1. Introduction Phenolic compounds occurring in nature are of considerable research interest because of their contribution to our health [1–4]. Their well-known beneficial effects include improvement in plasma antioxidant biomarkers, the effect on energy metabolism, contribution to bone health, and influence on carcinogenesis markers. As a current matter of first priority, phenolic compounds have been targeted for their ability to prevent amyloidosis, including Alzheimer’s disease, which is the major cause of dementia [5]. Naturally occurring phenolic
Abbreviations: Aβ, amyloid β; BHA, butylated hydroxyanisole; CA, caffeic acid; ChA, chlorogenic acid; Cur, curcumin; FA, ferulic acid; HCA, hydroxyl cinnamic acids; SA, sinapic acid; ThT, thioflavine T. ∗
Address for correspondence: Shigeru Katayama, PhD, Department of Bioscience and Biotechnology, Shinshu University, 8304, Minamiminowa, Nagano 399-4598, Japan. Tel.: +81 265 77 1603; Fax: +81 265 77 1603; e-mail: [email protected]; or Soichiro Nakamura, PhD, Department of Bioscience and Biotechnology, Shinshu University, 8304, Minamiminowa, Nagano 399-4598, Japan. Tel.: +81 265 77 1609; Fax: +81 265 77 1609; e-mail: [email protected] Received 29 July 2013; accepted 15 November 2013 DOI: 10.1002/bab.1182 Published online 12 May 2014 in Wiley Online Library (wileyonlinelibrary.com)
compounds have been demonstrated to prevent Alzheimer’s disease from onset via various pathways [6]. Because fibrillization of amyloid β (Aβ) is implicated with the onset of Alzheimer’s disease [5], inhibition of amyloid fibril formation of Aβ peptide is one of the major approaches. Thus, it has been demonstrated that various naturally occurring phenolic compounds such as myricetin, curcumin (Cur), ferulic acid (FA), and rosmarinic acid inhibit Aβ fibril formation [7–9]. Lipophilization of phenolic compounds has been developed by many researchers to functionally improve phenolic compounds because it has been reported to enhance both antioxidative and inhibitory activity on tumor proliferation [10–12]. To lipophilize phenolic compounds, various approaches have been investigated such as chemical/enzymatic esterification and chemical methylation [13–15]. Among them, enzymatic approaches are superior to chemical approaches in terms of regioselectivity and reducing unwanted side reactions [16]. The objective of this study was to investigate the effect of lipophilicity on the antiamyloidal activity of naturally occurring phenolic compounds by enzymatic lipophilization. As model compounds of lipophilization, we focused on hydroxyl cinnamic acids (HCAs) found in various fruits, coffee beverages, whole grain, and vegetables [17] because there is little known about the inhibitory activity of HCAs and their derivatives on amyloid fibril formation.
401
Biotechnology and Applied Biochemistry
2. Materials and Methods 2.1. Materials
˚ n-Butanol, n-octanol, n-hexadecanol, molecular sieves (4 A, beads, 8–12 mesh), tert-butanol, caffeic acid (CA), and sinapic acid (SA) were purchased from Sigma Aldrich (Tokyo, Japan). FA was purchased from LKT Laboratories (St. Paul, MN, USA). Chlorogenic acid (ChA), butylated hydroxyanisole (BHA), thioflavine T (ThT), and Cur were purchased from Wako Pure Chemical Industries (Tokyo, Japan). Lipase from Candida antarctica fixed on a macroporous resin (Novozym 435 ) was the courtesy of Novozymes (Bagsvaerd, Denmark). Aβ (1–42) peptide was purchased from Peptide Institute (Osaka, Japan). All reagents were of analytical grade. R
2.2. Lipophilization of HCAs Lipophilization of HCAs by enzymatic esterification was performed according to the esterification procedures of previous papers [18, 19] with slight modification. CA, FA, SA, and ChA were used as substrates for esterification, and n-butanol, n-octanol, and n-hexadecanol were used as conjugation counterparts. Novozym 435 was employed as the lipase for lipophilization. To improve synthesis efficiency, BHA and molecular sieves were added as an antioxidant and a removal reagent of the water generated by esterification, respectively. In a typical case, the reaction mixture consisted of 100 mg HCAs, 5 mL aliphatic alcohols, and 5 mL of tert-butanol. Then, 70 mg Novozym 435, 10 mg BHA, and 400 mg molecular sieves were added in the reaction mixtures. Reaction mixtures were incubated in sealed flasks and kept at 60 ◦ C for 1 week without shaking. The reaction temperature was chosen at 60 ◦ C, because it was the optimum temperature of Novozym 435 for esterification.
2.3. HPLC and TLC analysis Lipophilization of HCAs was monitored by HPLC analysis on Inertsil ODS-3 column (250 × 4.6 mm2 , 5 μm; GL Science, Tokyo, Japan) and normal phase TLC analysis. In HPLC analysis, PU-2089 Plus pumps (Jasco, Tokyo, Japan) at a flow rate of 1 mL/Min and solvents, (A) 2 mM H3 PO4 aqueous solution and (B) methanol, were utilized. Gradient elution was performed as follows: 0 Min (A:B = 95:5), 35 Min (A:B = 25:75), 45 Min (A:B = 0:100), and 60 Min (A:B = 0:100). Column temperature was kept at 40 ◦ C. The elution pattern was detected at 324 nm. TLC analysis was performed with TLC Silica gel 60 F254 (Merck Millipore Japan, Tokyo, Japan). The samples were developed using the following solvent system of chloroform–methanol– acetic acid/63:2:1. The detection of each spot was carried out using a UV lamp (Chromato-Vue lamp Model UVM-57; UVP, San Gabriel, CA, USA). Lipophilizing reaction efficiency was calculated by peak area of HPLC chromatograms. R
2.4. Purification of lipophilized HCAs After the lipophilizing reaction was complete, each reaction mixture was washed by saturated aqueous NaHCO3 solution to
402
remove unreacted HCAs. Then, lipophilic derivatives of HCAs were purified by a silica gel column with procedures developed by Jayaprakasam et al. [14]. For the mobile phase of column chromatography, chloroform–methanol was used to purify all esters. Eluted fractions with lipophilic derivatives of HCAs were collected and evaporated under reduced pressure.
2.5. Antiamyloid assay Aβ fibrillization assay was carried out in accordance with the procedure reported by previous researchers [20]. Aβ (1–42) peptide was dissolved in 0.1% ammonia aqueous solution at a concentration of 250 μM and stored at −80 ◦ C until use. Fibril formation of Aβ peptide was performed in 10 μL of reaction mixture consisting of 1 μL of phosphate buffer (500 mM, pH 7.5), 1 μL of NaCl solution (1 M), 1 μL of Aβ peptide solution, 0.25 μL of test compounds dissolved in dimethyl sulfoxide (a final concentration of 5 or 50 μM), and 6.75 μL of ultrapure water. Reaction mixtures were put into tubes and kept at 37 ◦ C for 24 H with a 2720 thermal cycler (Applied Biosystems, Carlsbad, CA, USA). The fibrillization degree of Aβ peptide was measured by ThT fluorescence assay, which probes Aβ for the presence of β-sheet-rich structures. Aliquots (6 μL) of each fibrillization mixture were collected after incubation and added to 1.2 mL of ThT solution (5 μM ThT in 50 mM glycine–NaOH buffer, pH 8.5). ThT fluorescence was observed on a Jasco FP-6200 spectrofluorometer (Jasco). Excitation and emission wavelength were 446 and 490 nm, respectively.
2.6. Hydrophobicity analysis Hydrophobicity of HCAs and their lipophilic derivatives were assessed by calculated partition coefficient (clog P). Clog P values were calculated by ChemBioDraw Ultra 12.0 (CambridgeSoft, Cambridge, MA, USA), which calculates clog P based on the algorithm developed by Medicinal Chemistry Project and Biobyte (BioByte Corporation, Claremont, CA, USA), using the fragment-based method.
2.7. Electron microscopy Transmission electron microscopy was used to characterize the structural morphology of Aβ fibrils both in the presence and in the absence of CA16. Aβ (1–42) peptide (25 μM) was incubated with or without 5 μg/mL CFA at 37 ◦ C. Aliquots (5 μL) of the sample were spotted onto collodion-coated 400 mesh copper grids. The grids were negatively stained with 1% phosphotungstic acid and visualized with a JEM-1400 microscope (JEOL, Tokyo, Japan) operating with an accelerating voltage of 80 kV.
2.8. Statistical analysis
All results are indicated as mean ± SD. Statistical evaluation was carried out by an unpaired Student’s t-test (analysis of variance followed by Scheffe’s multiple range test) with Statcel software version 2.0 (OMS-Publishing, Saitama, Japan).
Antiamyloid Effect of Esterified Phenolic Acids
FIG. 1
(A) Chemical structures of CA, FA, SA, and their lipophilized derivatives. (B) Chemical structure of ChA and its lipophilized derivatives.
3. Results 3.1. Enzymatic lipophilization of HCAs The lipophilic derivatives of HCAs with different alkyl chain lengths were synthesized by lipase treatment. Lipophilized HCAs are named based on their alkyl chain length, for instance, lipophilized CA with a butyl chain is named CA4. Figure 1 shows the chemical structures of HCAs used in this study and their presumed lipophilic derivatives. Table 1 illustrates the behavior of HCAs and the lipophilic derivatives of HCAs on normal-phase TLC, their behaviors in reverse-phase HPLC, and the efficiency of the lipophilizing reaction catalyzed by lipase. In TLC analysis using a silica gel plate, lipophilized HCAs produced by lipase treatment were observed as new single spots exhibited higher Rf values than each HCA. The attachment of a longer alkyl chain tended to lead to higher Rf values. As shown in Fig. 2, lipophilized ChAs were observed as a new single HPLC peak eluting later than ChA, and the attachment of longer alkyl chain resulted in later elution. Similar shifts in HPLC retention time were observed in other lypophilized HCAs (Table 1). Increases in Rf values in TLC, and elongation of retention times in HPLC, imply that lipase treatment of HCAs escalated their lipophilicity, and esterification with alcohols having longer alkyl chains led to higher lipophilicity in HCAs. The clog P value was potentially used as an indicator of lipophilicity. As shown in Table 1, all HCA derivatives exhibited a high clog P value, and longer alkyl
Biotechnology and Applied Biochemistry
chains led to higher clog P. This tendency was consistent with the behavior of HCA derivatives in both normal-phase TLC and reverse-phase HPLC. The reaction efficiency of lipophilization to ChA was in the range of 23.4%–35.3% and was higher among HCA derivatives. In contrast, the reaction efficiency of CA, FA, and SA was far lower than that of ChA lipophilization, and their yield was in the range of 1.7%–10.3%.
3.2. Inhibitory activity of HCAs without lipophilization on Aβ (1–42) fibrillization To investigate whether HCAs have potent antiamyloidal activity, Aβ (1–42) peptide, which is implicated in the onset of Alzheimer’s disease, was employed as the model for amyloidogenic peptide to measure the degree of Aβ fibrillization with or without HCAs. Fibrillization degree of Aβ (1–42) was indicated as ThT fluorescence intensity. As shown in Fig. 3, CA, FA, and ChA significantly decreased Aβ fibrillization both at concentrations of 5 and 50 μM in a dose-dependent manner. In contrast, SA showed no significant inhibition at any concentration. Inhibitory activity of HCAs on Aβ (1–42) fibrillization was compared with Cur, which is known as a potent Aβ fibrilizing inhibitor [21]. Cur exhibited higher activity than FA, SA, and ChA at both 5 and 50 μM, and decreased Aβ fibrillization to 66.3 ± 3.0% and 38.0 ± 5.3%, respectively. In contrast, CA showed almost the same level of inhibition as Cur.
3.3. Inhibitory activity of HCAs lipophilic derivatives on Aβ fibrillization Figure 4 shows the effect of lipophilized derivatives of HCAs on Aβ fibrillization. Lipophilization of HCAs resulted in increased inhibition of Aβ fibrillization, although the inhibitory effect
403
Biotechnology and Applied Biochemistry
TABLE 1
Summary of TLC and HPLC analysis with lipophilicity parameters
The inhibitory effect on Aβ fibril formation and morphology were further studied by transmission electron microscopy. When Aβ alone was incubated for 24 H, clear amyloid fibrils were observed (Fig. 5A). In contrast, incubation of Aβ with 5 μg/mL CA16 for 24 H resulted in a significant decrease in fibril formation (Fig. 5B).
TLC Rf valuea
HPLC retention time (Min)b
CA
0.06
20.8
0.94
CA4
0.19
37.8
3.00
1.7%
4. Discussion
CA8
0.21
46.7
5.02
3.2%
CA16
0.22
52.5
8.72
4.4%
FA
0.36
26.8
1.25
FA4
0.65
42.7
3.30
5.6%
FA8
0.69
48.4
5.32
10.3%
FA16
0.74
54.2
8.86
7.6%
SA
0.36
24.5
1.27
SA4
0.68
41.9
3.32
2.2%
SA8
0.69
48.9
5.34
4.7%
SA16
0.71
52.6
8.87
8.1%
ChA
0.00
18.9
− 0.45
ChA4
0.04
32.4
1.60
35.3%
ChA8
0.05
40.8
3.62
27.3%
ChA16
0.06
49.5
7.77
23.4%
Previous studies have examined the relationship between structures and inhibitory activity on amyloid fibril formation of phenolic compounds. Many researchers have demonstrated that the number or position of aromatic ring and substituents on aromatic rings should act as a critical factor of the inhibitory activity toward amyloid fibril formation [9, 22]. They have suggested that these factors contribute to noncovalent interaction between amyloidgenic peptides and inhibitors of amyloid fibril formation through amyloidgenic aromatic residues. However, little is known about the contribution of lipophilization to antiamyloidal activity of phenolic compounds. In the present study, we have demonstrated that the inhibitory activity of HCAs toward Aβ fibrillization was enhanced by lipophilization. In addition, as a longer alkyl chain was attached, lipophilized HCAs, except for SA, tended to exhibit higher inhibition of Aβ fibrillization. Sawaya et al. [23] have reported that a hydrophobic segment of Aβ peptide plays important roles in Aβ fibril extension. They assume that the hydrophobic segment intermediates binding between an incoming Aβ monomer and Aβ fibril. Therefore, it is possible that lipophilized HCAs can interact with the hydrophobic segment in Aβ via hydrophobic interaction, resulting in lipophilized HCAs, possibly preventing incoming Aβ monomer contact with fibril. From these findings, it was considered that there is a close correlation between lipophilicity and Aβ fibrillization inhibitory activity of HCAs and that the antiamyloidal activity of HCAs increases along with its lipophilicity. Among HCAs, CA and ChA possessing catechol structure showed high antiamyloidal activity (Fig. 3), and the activity of CA was comparable to Cur, which is known as a potent antiamyloidal compound. CA, ChA, and Cur have two hydroxylic groups on the aromatic ring, whereas FA and SA possess only a single hydroxylic group on their aromatic ring. Consistent with this result, previous reports also have demonstrated that phenolic compounds having more hydroxylic groups, such as myricetin, morin, and quercetin, on their aromatic rings exhibit higher antiamyloidal activity [7, 22]. Thus, they assumed that the critical factor for antiamyloidal activity of phenolic compounds is the presence of a hydroxylic group on the aromatic ring, which enhances interactions between amyloidgenic peptide and phenolic compounds by forming hydrogen bonds. ChA, the ester of CA with quinic acid, showed lower activity than CA. In addition, antiamyloidal activities of ChA esters were lower than those of CA esters having the same alkyl chain length. The hydrophilicity of quinic acid might affect the antiamyloid activity of ChA. On the other hand, a methoxy group of phenolic
Compound
clog Pc
Reaction efficiencyd –e
–e
–e
–e
a Rf values in normal phase TLC with chloroform–methanol–acetic acid (63:2:1). b Observed
on ODS column with H3 PO4 aqueous solution–methanol as mobile phase.
c Calculated
by the algorithm developed by Medicinal Chemistry Project and Biobyte.
d Percentage e Without
of HCAs converted to ester.
lipase treatment.
was different depending on the HCA compounds. The relative fluorescence intensity of CA4 was 41.6 ± 6.1%, which shows a similar level to that of CA; however, the inhibitory activity of CA8 and CA16 increased with the attachment of longer chains and decreased Aβ fibrillization to 19.3 ± 2.3% and 8.8 ± 2.3%, respectively (Fig. 4A). Similarly, all lipophilic derivatives of FA and ChA exhibited higher activity than intact FA and ChA (Figs. 4B and 4D). Lipophilization with a hexadecyl chain showed the most inhibition of Aβ fibrillization, followed by octyl and butyl chains. Although SA itself showed no significant activity without lipophilization, lipophilized SA inhibited Aβ fibrillization significantly (Fig. 4C). However, the decrease in the inhibitory effect was not observed in lipophilization of SA with a longer alkyl chain.
404
Antiamyloid Effect of Esterified Phenolic Acids
FIG. 2
FIG. 3
UV chromatograms of ChA and purified ChA lipophilization. (A) ChA only, (B) purified ChA4, (C) purified ChA8, and (D) purified ChA16.
Effect of HCAs and curcumin on Aβ (1–42) fibrillization. CA, FA, SA, ChA, or Cur were incubated with 25 μM Aβ at a concentration of 5 or 50 μM for 24 H. Aβ fibrillization was assessed by ThT fluorescence intensity. Values with different superscripts are significantly different (P < 0.05).
compounds might attenuate the antiamyloidal activity because the inhibitory activity of SA containing two methoxy groups was lowest among HCAs. Also, lipophilization of SA with short alkyl chains resulted in an increase in antiamyloidal activity, whereas the introduction of longer alkyl chain decreased the antiamyloidal activity. It seems that the introduction of the alkyl chain contributes to the enhancement of interaction with amyloidogenic peptide. However, hydrophobicity of the methoxy
Biotechnology and Applied Biochemistry
group on their aromatic ring of SA would interfere when longer alkyl chain was introduced to phenolic compounds. Further studies are needed to reveal the relationship between hydrophilic and hydrophobic groups on lipophilized HCAs and their antiamyloidal activity. To fully understand the antiamyloidal activity of lipophilized HCAs, in vivo investigation is obviously necessary. Murakami et al. [24] reported that esterification of FA with 2-ethylhexanol remarkably improved cellular intake at human promyelocytes cell line HL-60. Generally, the cell permeability depends on lipophilicity, and lipophilization of HCAs might be thus expected as the way to improve cell permeability. It has been reported that the optimal clog P range for cell permeability is 1–3 in terms of pharmacokinetics [25]. Clog P of CA4 and ChA4 reached this range, whereas that of CA and ChA was less than 1 (Table 1). On the other hand, CA16, FA16, and SA16 exhibited clog P higher than 8. Even if CA16 showed the highest value of the antiamyloidal activity, the esterified phenolic compounds with too high clog P might not show potent activity in vivo because of low cell permeability. The stability of newly synthesized lipophilic derivatives of HCAs is an important parameter for further in vivo investigation. The lipophilized HCAs collected from column chromatography was evaporated under reduced pressure; however, no hydrolytic reaction was observed in HPLC chromatograms (data not shown), indicating that the lipophilic derivatives were stable during this process. In conclusion, we demonstrated that CA and ChA possessing catechol structure inhibited Aβ fibrillization to a greater extent. Furthermore, higher antiamyloidal activities of HCA derivatives except for SA are associated with having a longer alkyl chain. These results suggest the importance of lipophilicity and catecholic structure on antiamyloidal activity of HCAs and provide understanding about the relationship between
405
Biotechnology and Applied Biochemistry
FIG. 4
Effect of lipophilized HCAs on Aβ (1–42) fibrillization. (A) Lipophilic derivatives of CA. (B) Lipophilic derivatives of FA. (C) Lipophilic derivatives of SA. (D) Lipophilic derivatives of ChA. HCAs, or their lipophilized derivatives (50 μM) were incubated for 24 H with 25 μM Aβ. Aβ fibrillization was assessed by ThT fluorescence intensity. Values with different superscripts are significantly different (P < 0.05).
antiamyloidal activity of phenolic compounds and their lipophilicity. Our results will contribute to the development of antiamyloidal agents derived from naturally occurring phenolic compounds.
5. Acknowledgements This study was supported by a grant-in-aid from the Ministry of Education, Culture, Sports, Science and Technology of Japan (23580161). We also appreciate Mr. Takahiro Ohno for his assistance with electron microscopy.
6. References
FIG. 5
406
Electron micrographs after 24 H of incubation of Aβ (1–42) in the absence (A) and presence (B) of CA16. CA16 (50 μM) were incubated for 24 H with 25 μM Aβ. The samples were negatively stained with 1% phosphotungstic acid. The scale bar indicates 200 nm.
[1] Williamson, G., and Manach, C. (2005) Am. J. Clin. Nutr. 81, 243S–255S. [2] Aggarwal, B. B., Kumar, A., and Bharti, A. C. (2003) Anticancer Res. 23, 363–398. [3] Baur, J. A., Pearson, K. J., Price, N. L., Jamieson, H. A., Lerin, C., Kalra, A., Prabhu, V. V., Allard, J. S., Lopez-Lluch, G., and Lewis, K. (2006) Nature 444, 337–342. [4] Korkina, L. (2007) Cell Mol. Biol. 53, 15–25. [5] Blennow, K., Leon, M. d., and Zetterberg, H. (2006) Lancet 368, 387–403. [6] Ono, K., Yoshiike, Y., Takashima, A., Hasegawa, K., Naiki, H., and Yamada, M. (2003) J. Neurochem. 87, 172–181. [7] Ono, K., Hamaguchi, T., Naiki, H., and Yamada, M. (2006) Biochim. Biophys. Acta 1762, 575–586. [8] Ono, K., Hirohata, M., and Yamada, M. (2005) Biochem. Biophys. Res. Commun. 336, 444–449. [9] Porat, Y., Abramowitz, A., and Gazit, E. (2006) Chem. Biol. Drug Des. 67, 27–37.
Antiamyloid Effect of Esterified Phenolic Acids
[10] Garrido, J., Gaspar, A., Garrido, E. M., Miri, R., Tavakkoli, M., Pourali, S., Saso, L., Borges, F., and Firuzi, O. (2012) Biochimie 94, 961–967. [11] Salem, J. H., Humeau, C., Chevalot, I., Harscoat-Schiavo, C., Vanderesse, R., Blanchard, F., and Fick, M. (2010) Proc. Biochem. 45, 382–389. [12] Serafim, T. L., Carvalho, F. S., Marques, M. P., Calheiros, R., Silva, T., Garrido, J., Milhazes, N., Borges, F., Roleira, F., Silva, E. T., Holy, J., and Oliveira, P. J. (2011) Chem. Res. Toxicol. 24, 763–774. [13] Chebil, L., Humeau, C., Falcimaigne, A., Engasser, J. M., and Ghoul, M. (2006) Proc. Biochem. 41, 2237–2251. [14] Jayaprakasam, B., Vanisree, M., Zhang, Y., Dewitt, D. L., and Nair, M. G. (2006) J. Agric. Food Chem. 54, 5375–5381. [15] Walle, T. (2007) Mol. Pharm. 4, 826–832. [16] Figueroa-Espinoza, M. C., and Villeneuve, P. (2005) J. Agric. Food Chem. 53, 2779–2787. [17] Clifford, M. N. (1999) J. Sci. Food Agr. 79, 362–372.
Biotechnology and Applied Biochemistry
[18] Compton, D. L., Laszlo, J. A., and Berhow, M. A. (2000) J. Am. Oil Chem. Soc. 77, 513–519. [19] Stevenson, D. E., Parkar, S. G., Zhang, J., Stanley, R. A., Jensen, D. J., and Cooney, J. M. (2007) Enzyme Microb. Tech. 40, 1078–1086. [20] Katayama, S., Ogawa, H., and Nakamura, S. (2011) J. Agric. Food Chem. 59, 12691–12696. [21] Ono, K., Hasegawa, K., Naiki, H., and Yamada, M. (2004) J. Neurosci. Res. 75, 742–750. [22] Reinke, A. A., and Gestwicki, J. E. (2007) Chem. Biol. Drug Des. 70, 206–215. [23] Sawaya, M. R., Sambashivan, S., Nelson, R., Ivanova, M. I., Sievers, S. A., Apostol, M. I., Thompson, M. J., Balbirnie, M., Wiltzius, J. J. W., McFarlane, H. T., Madsen, A. O., Riekel, C., and Eisenberg, D. (2007) Nature 447, 453–457. [24] Murakami, A., Kadota, M., Takahashi, D., Taniguchi, H., Nomura, E., Hosoda, A., Tsuno, T., Maruta, Y., Ohigashi, H., and Koshimizu, K. (2000) Cancer Lett. 157, 77–85. [25] Waring, M. J. (2010) Expert Opin. Drug Dis. 5, 235–248.
407
Hazuki Kondo Haruka Sugiyama Shigeru Katayama∗ Soichiro Nakamura∗
Department of Bioscience and Biotechnology, Shinshu University, Minamiminowa, Nagano, Japan
Abstract Lipophilic derivatives of hydroxyl cinnamic acids (HCAs) including caffeic acid (CA), ferulic acid, sinapic acid (SA), and chlorogenic acid were synthesized by esterification with butanol, octanol, or hexadecanol catalyzed by the lipase from Candida antarctica to investigate the effect of lipophilicity on their antiamyloidal activity assessed by the inhibitory activities toward fibrillization of amyloid β (Aβ) peptide. Among them, CA showed the highest activity at 50 μM, reducing the amyloid fibril formation of Aβ to 34.4 ± 6.8%. The antiamyloidal effects of HCAs were enhanced by esterification with alkyl alcohols,
and the longer alkyl chain tended to be more effective except for SA. Aβ fibril formation was suppressed by the hexadecyl ester of CA, which was reduced to 8.8 ± 2.3%. In contrast, those of octyl and butyl esters were 19.3 ± 2.3% and 41.6 ± 6.1%, respectively. These results show that lipophilicity plays an important role in the antiamyloidal activities of esterified phenolic compounds. C 2013 International Union of Biochemistry and Molecular Biology, Inc. Volume 61, Number 4, Pages 401–407, 2014
Keywords: amyloid β, antiamyloidal activity, esterification, hydroxyl cinnamic acids, lipase, lipophilicity
1. Introduction Phenolic compounds occurring in nature are of considerable research interest because of their contribution to our health [1–4]. Their well-known beneficial effects include improvement in plasma antioxidant biomarkers, the effect on energy metabolism, contribution to bone health, and influence on carcinogenesis markers. As a current matter of first priority, phenolic compounds have been targeted for their ability to prevent amyloidosis, including Alzheimer’s disease, which is the major cause of dementia [5]. Naturally occurring phenolic
Abbreviations: Aβ, amyloid β; BHA, butylated hydroxyanisole; CA, caffeic acid; ChA, chlorogenic acid; Cur, curcumin; FA, ferulic acid; HCA, hydroxyl cinnamic acids; SA, sinapic acid; ThT, thioflavine T. ∗
Address for correspondence: Shigeru Katayama, PhD, Department of Bioscience and Biotechnology, Shinshu University, 8304, Minamiminowa, Nagano 399-4598, Japan. Tel.: +81 265 77 1603; Fax: +81 265 77 1603; e-mail: [email protected]; or Soichiro Nakamura, PhD, Department of Bioscience and Biotechnology, Shinshu University, 8304, Minamiminowa, Nagano 399-4598, Japan. Tel.: +81 265 77 1609; Fax: +81 265 77 1609; e-mail: [email protected] Received 29 July 2013; accepted 15 November 2013 DOI: 10.1002/bab.1182 Published online 12 May 2014 in Wiley Online Library (wileyonlinelibrary.com)
compounds have been demonstrated to prevent Alzheimer’s disease from onset via various pathways [6]. Because fibrillization of amyloid β (Aβ) is implicated with the onset of Alzheimer’s disease [5], inhibition of amyloid fibril formation of Aβ peptide is one of the major approaches. Thus, it has been demonstrated that various naturally occurring phenolic compounds such as myricetin, curcumin (Cur), ferulic acid (FA), and rosmarinic acid inhibit Aβ fibril formation [7–9]. Lipophilization of phenolic compounds has been developed by many researchers to functionally improve phenolic compounds because it has been reported to enhance both antioxidative and inhibitory activity on tumor proliferation [10–12]. To lipophilize phenolic compounds, various approaches have been investigated such as chemical/enzymatic esterification and chemical methylation [13–15]. Among them, enzymatic approaches are superior to chemical approaches in terms of regioselectivity and reducing unwanted side reactions [16]. The objective of this study was to investigate the effect of lipophilicity on the antiamyloidal activity of naturally occurring phenolic compounds by enzymatic lipophilization. As model compounds of lipophilization, we focused on hydroxyl cinnamic acids (HCAs) found in various fruits, coffee beverages, whole grain, and vegetables [17] because there is little known about the inhibitory activity of HCAs and their derivatives on amyloid fibril formation.
401
Biotechnology and Applied Biochemistry
2. Materials and Methods 2.1. Materials
˚ n-Butanol, n-octanol, n-hexadecanol, molecular sieves (4 A, beads, 8–12 mesh), tert-butanol, caffeic acid (CA), and sinapic acid (SA) were purchased from Sigma Aldrich (Tokyo, Japan). FA was purchased from LKT Laboratories (St. Paul, MN, USA). Chlorogenic acid (ChA), butylated hydroxyanisole (BHA), thioflavine T (ThT), and Cur were purchased from Wako Pure Chemical Industries (Tokyo, Japan). Lipase from Candida antarctica fixed on a macroporous resin (Novozym 435 ) was the courtesy of Novozymes (Bagsvaerd, Denmark). Aβ (1–42) peptide was purchased from Peptide Institute (Osaka, Japan). All reagents were of analytical grade. R
2.2. Lipophilization of HCAs Lipophilization of HCAs by enzymatic esterification was performed according to the esterification procedures of previous papers [18, 19] with slight modification. CA, FA, SA, and ChA were used as substrates for esterification, and n-butanol, n-octanol, and n-hexadecanol were used as conjugation counterparts. Novozym 435 was employed as the lipase for lipophilization. To improve synthesis efficiency, BHA and molecular sieves were added as an antioxidant and a removal reagent of the water generated by esterification, respectively. In a typical case, the reaction mixture consisted of 100 mg HCAs, 5 mL aliphatic alcohols, and 5 mL of tert-butanol. Then, 70 mg Novozym 435, 10 mg BHA, and 400 mg molecular sieves were added in the reaction mixtures. Reaction mixtures were incubated in sealed flasks and kept at 60 ◦ C for 1 week without shaking. The reaction temperature was chosen at 60 ◦ C, because it was the optimum temperature of Novozym 435 for esterification.
2.3. HPLC and TLC analysis Lipophilization of HCAs was monitored by HPLC analysis on Inertsil ODS-3 column (250 × 4.6 mm2 , 5 μm; GL Science, Tokyo, Japan) and normal phase TLC analysis. In HPLC analysis, PU-2089 Plus pumps (Jasco, Tokyo, Japan) at a flow rate of 1 mL/Min and solvents, (A) 2 mM H3 PO4 aqueous solution and (B) methanol, were utilized. Gradient elution was performed as follows: 0 Min (A:B = 95:5), 35 Min (A:B = 25:75), 45 Min (A:B = 0:100), and 60 Min (A:B = 0:100). Column temperature was kept at 40 ◦ C. The elution pattern was detected at 324 nm. TLC analysis was performed with TLC Silica gel 60 F254 (Merck Millipore Japan, Tokyo, Japan). The samples were developed using the following solvent system of chloroform–methanol– acetic acid/63:2:1. The detection of each spot was carried out using a UV lamp (Chromato-Vue lamp Model UVM-57; UVP, San Gabriel, CA, USA). Lipophilizing reaction efficiency was calculated by peak area of HPLC chromatograms. R
2.4. Purification of lipophilized HCAs After the lipophilizing reaction was complete, each reaction mixture was washed by saturated aqueous NaHCO3 solution to
402
remove unreacted HCAs. Then, lipophilic derivatives of HCAs were purified by a silica gel column with procedures developed by Jayaprakasam et al. [14]. For the mobile phase of column chromatography, chloroform–methanol was used to purify all esters. Eluted fractions with lipophilic derivatives of HCAs were collected and evaporated under reduced pressure.
2.5. Antiamyloid assay Aβ fibrillization assay was carried out in accordance with the procedure reported by previous researchers [20]. Aβ (1–42) peptide was dissolved in 0.1% ammonia aqueous solution at a concentration of 250 μM and stored at −80 ◦ C until use. Fibril formation of Aβ peptide was performed in 10 μL of reaction mixture consisting of 1 μL of phosphate buffer (500 mM, pH 7.5), 1 μL of NaCl solution (1 M), 1 μL of Aβ peptide solution, 0.25 μL of test compounds dissolved in dimethyl sulfoxide (a final concentration of 5 or 50 μM), and 6.75 μL of ultrapure water. Reaction mixtures were put into tubes and kept at 37 ◦ C for 24 H with a 2720 thermal cycler (Applied Biosystems, Carlsbad, CA, USA). The fibrillization degree of Aβ peptide was measured by ThT fluorescence assay, which probes Aβ for the presence of β-sheet-rich structures. Aliquots (6 μL) of each fibrillization mixture were collected after incubation and added to 1.2 mL of ThT solution (5 μM ThT in 50 mM glycine–NaOH buffer, pH 8.5). ThT fluorescence was observed on a Jasco FP-6200 spectrofluorometer (Jasco). Excitation and emission wavelength were 446 and 490 nm, respectively.
2.6. Hydrophobicity analysis Hydrophobicity of HCAs and their lipophilic derivatives were assessed by calculated partition coefficient (clog P). Clog P values were calculated by ChemBioDraw Ultra 12.0 (CambridgeSoft, Cambridge, MA, USA), which calculates clog P based on the algorithm developed by Medicinal Chemistry Project and Biobyte (BioByte Corporation, Claremont, CA, USA), using the fragment-based method.
2.7. Electron microscopy Transmission electron microscopy was used to characterize the structural morphology of Aβ fibrils both in the presence and in the absence of CA16. Aβ (1–42) peptide (25 μM) was incubated with or without 5 μg/mL CFA at 37 ◦ C. Aliquots (5 μL) of the sample were spotted onto collodion-coated 400 mesh copper grids. The grids were negatively stained with 1% phosphotungstic acid and visualized with a JEM-1400 microscope (JEOL, Tokyo, Japan) operating with an accelerating voltage of 80 kV.
2.8. Statistical analysis
All results are indicated as mean ± SD. Statistical evaluation was carried out by an unpaired Student’s t-test (analysis of variance followed by Scheffe’s multiple range test) with Statcel software version 2.0 (OMS-Publishing, Saitama, Japan).
Antiamyloid Effect of Esterified Phenolic Acids
FIG. 1
(A) Chemical structures of CA, FA, SA, and their lipophilized derivatives. (B) Chemical structure of ChA and its lipophilized derivatives.
3. Results 3.1. Enzymatic lipophilization of HCAs The lipophilic derivatives of HCAs with different alkyl chain lengths were synthesized by lipase treatment. Lipophilized HCAs are named based on their alkyl chain length, for instance, lipophilized CA with a butyl chain is named CA4. Figure 1 shows the chemical structures of HCAs used in this study and their presumed lipophilic derivatives. Table 1 illustrates the behavior of HCAs and the lipophilic derivatives of HCAs on normal-phase TLC, their behaviors in reverse-phase HPLC, and the efficiency of the lipophilizing reaction catalyzed by lipase. In TLC analysis using a silica gel plate, lipophilized HCAs produced by lipase treatment were observed as new single spots exhibited higher Rf values than each HCA. The attachment of a longer alkyl chain tended to lead to higher Rf values. As shown in Fig. 2, lipophilized ChAs were observed as a new single HPLC peak eluting later than ChA, and the attachment of longer alkyl chain resulted in later elution. Similar shifts in HPLC retention time were observed in other lypophilized HCAs (Table 1). Increases in Rf values in TLC, and elongation of retention times in HPLC, imply that lipase treatment of HCAs escalated their lipophilicity, and esterification with alcohols having longer alkyl chains led to higher lipophilicity in HCAs. The clog P value was potentially used as an indicator of lipophilicity. As shown in Table 1, all HCA derivatives exhibited a high clog P value, and longer alkyl
Biotechnology and Applied Biochemistry
chains led to higher clog P. This tendency was consistent with the behavior of HCA derivatives in both normal-phase TLC and reverse-phase HPLC. The reaction efficiency of lipophilization to ChA was in the range of 23.4%–35.3% and was higher among HCA derivatives. In contrast, the reaction efficiency of CA, FA, and SA was far lower than that of ChA lipophilization, and their yield was in the range of 1.7%–10.3%.
3.2. Inhibitory activity of HCAs without lipophilization on Aβ (1–42) fibrillization To investigate whether HCAs have potent antiamyloidal activity, Aβ (1–42) peptide, which is implicated in the onset of Alzheimer’s disease, was employed as the model for amyloidogenic peptide to measure the degree of Aβ fibrillization with or without HCAs. Fibrillization degree of Aβ (1–42) was indicated as ThT fluorescence intensity. As shown in Fig. 3, CA, FA, and ChA significantly decreased Aβ fibrillization both at concentrations of 5 and 50 μM in a dose-dependent manner. In contrast, SA showed no significant inhibition at any concentration. Inhibitory activity of HCAs on Aβ (1–42) fibrillization was compared with Cur, which is known as a potent Aβ fibrilizing inhibitor [21]. Cur exhibited higher activity than FA, SA, and ChA at both 5 and 50 μM, and decreased Aβ fibrillization to 66.3 ± 3.0% and 38.0 ± 5.3%, respectively. In contrast, CA showed almost the same level of inhibition as Cur.
3.3. Inhibitory activity of HCAs lipophilic derivatives on Aβ fibrillization Figure 4 shows the effect of lipophilized derivatives of HCAs on Aβ fibrillization. Lipophilization of HCAs resulted in increased inhibition of Aβ fibrillization, although the inhibitory effect
403
Biotechnology and Applied Biochemistry
TABLE 1
Summary of TLC and HPLC analysis with lipophilicity parameters
The inhibitory effect on Aβ fibril formation and morphology were further studied by transmission electron microscopy. When Aβ alone was incubated for 24 H, clear amyloid fibrils were observed (Fig. 5A). In contrast, incubation of Aβ with 5 μg/mL CA16 for 24 H resulted in a significant decrease in fibril formation (Fig. 5B).
TLC Rf valuea
HPLC retention time (Min)b
CA
0.06
20.8
0.94
CA4
0.19
37.8
3.00
1.7%
4. Discussion
CA8
0.21
46.7
5.02
3.2%
CA16
0.22
52.5
8.72
4.4%
FA
0.36
26.8
1.25
FA4
0.65
42.7
3.30
5.6%
FA8
0.69
48.4
5.32
10.3%
FA16
0.74
54.2
8.86
7.6%
SA
0.36
24.5
1.27
SA4
0.68
41.9
3.32
2.2%
SA8
0.69
48.9
5.34
4.7%
SA16
0.71
52.6
8.87
8.1%
ChA
0.00
18.9
− 0.45
ChA4
0.04
32.4
1.60
35.3%
ChA8
0.05
40.8
3.62
27.3%
ChA16
0.06
49.5
7.77
23.4%
Previous studies have examined the relationship between structures and inhibitory activity on amyloid fibril formation of phenolic compounds. Many researchers have demonstrated that the number or position of aromatic ring and substituents on aromatic rings should act as a critical factor of the inhibitory activity toward amyloid fibril formation [9, 22]. They have suggested that these factors contribute to noncovalent interaction between amyloidgenic peptides and inhibitors of amyloid fibril formation through amyloidgenic aromatic residues. However, little is known about the contribution of lipophilization to antiamyloidal activity of phenolic compounds. In the present study, we have demonstrated that the inhibitory activity of HCAs toward Aβ fibrillization was enhanced by lipophilization. In addition, as a longer alkyl chain was attached, lipophilized HCAs, except for SA, tended to exhibit higher inhibition of Aβ fibrillization. Sawaya et al. [23] have reported that a hydrophobic segment of Aβ peptide plays important roles in Aβ fibril extension. They assume that the hydrophobic segment intermediates binding between an incoming Aβ monomer and Aβ fibril. Therefore, it is possible that lipophilized HCAs can interact with the hydrophobic segment in Aβ via hydrophobic interaction, resulting in lipophilized HCAs, possibly preventing incoming Aβ monomer contact with fibril. From these findings, it was considered that there is a close correlation between lipophilicity and Aβ fibrillization inhibitory activity of HCAs and that the antiamyloidal activity of HCAs increases along with its lipophilicity. Among HCAs, CA and ChA possessing catechol structure showed high antiamyloidal activity (Fig. 3), and the activity of CA was comparable to Cur, which is known as a potent antiamyloidal compound. CA, ChA, and Cur have two hydroxylic groups on the aromatic ring, whereas FA and SA possess only a single hydroxylic group on their aromatic ring. Consistent with this result, previous reports also have demonstrated that phenolic compounds having more hydroxylic groups, such as myricetin, morin, and quercetin, on their aromatic rings exhibit higher antiamyloidal activity [7, 22]. Thus, they assumed that the critical factor for antiamyloidal activity of phenolic compounds is the presence of a hydroxylic group on the aromatic ring, which enhances interactions between amyloidgenic peptide and phenolic compounds by forming hydrogen bonds. ChA, the ester of CA with quinic acid, showed lower activity than CA. In addition, antiamyloidal activities of ChA esters were lower than those of CA esters having the same alkyl chain length. The hydrophilicity of quinic acid might affect the antiamyloid activity of ChA. On the other hand, a methoxy group of phenolic
Compound
clog Pc
Reaction efficiencyd –e
–e
–e
–e
a Rf values in normal phase TLC with chloroform–methanol–acetic acid (63:2:1). b Observed
on ODS column with H3 PO4 aqueous solution–methanol as mobile phase.
c Calculated
by the algorithm developed by Medicinal Chemistry Project and Biobyte.
d Percentage e Without
of HCAs converted to ester.
lipase treatment.
was different depending on the HCA compounds. The relative fluorescence intensity of CA4 was 41.6 ± 6.1%, which shows a similar level to that of CA; however, the inhibitory activity of CA8 and CA16 increased with the attachment of longer chains and decreased Aβ fibrillization to 19.3 ± 2.3% and 8.8 ± 2.3%, respectively (Fig. 4A). Similarly, all lipophilic derivatives of FA and ChA exhibited higher activity than intact FA and ChA (Figs. 4B and 4D). Lipophilization with a hexadecyl chain showed the most inhibition of Aβ fibrillization, followed by octyl and butyl chains. Although SA itself showed no significant activity without lipophilization, lipophilized SA inhibited Aβ fibrillization significantly (Fig. 4C). However, the decrease in the inhibitory effect was not observed in lipophilization of SA with a longer alkyl chain.
404
Antiamyloid Effect of Esterified Phenolic Acids
FIG. 2
FIG. 3
UV chromatograms of ChA and purified ChA lipophilization. (A) ChA only, (B) purified ChA4, (C) purified ChA8, and (D) purified ChA16.
Effect of HCAs and curcumin on Aβ (1–42) fibrillization. CA, FA, SA, ChA, or Cur were incubated with 25 μM Aβ at a concentration of 5 or 50 μM for 24 H. Aβ fibrillization was assessed by ThT fluorescence intensity. Values with different superscripts are significantly different (P < 0.05).
compounds might attenuate the antiamyloidal activity because the inhibitory activity of SA containing two methoxy groups was lowest among HCAs. Also, lipophilization of SA with short alkyl chains resulted in an increase in antiamyloidal activity, whereas the introduction of longer alkyl chain decreased the antiamyloidal activity. It seems that the introduction of the alkyl chain contributes to the enhancement of interaction with amyloidogenic peptide. However, hydrophobicity of the methoxy
Biotechnology and Applied Biochemistry
group on their aromatic ring of SA would interfere when longer alkyl chain was introduced to phenolic compounds. Further studies are needed to reveal the relationship between hydrophilic and hydrophobic groups on lipophilized HCAs and their antiamyloidal activity. To fully understand the antiamyloidal activity of lipophilized HCAs, in vivo investigation is obviously necessary. Murakami et al. [24] reported that esterification of FA with 2-ethylhexanol remarkably improved cellular intake at human promyelocytes cell line HL-60. Generally, the cell permeability depends on lipophilicity, and lipophilization of HCAs might be thus expected as the way to improve cell permeability. It has been reported that the optimal clog P range for cell permeability is 1–3 in terms of pharmacokinetics [25]. Clog P of CA4 and ChA4 reached this range, whereas that of CA and ChA was less than 1 (Table 1). On the other hand, CA16, FA16, and SA16 exhibited clog P higher than 8. Even if CA16 showed the highest value of the antiamyloidal activity, the esterified phenolic compounds with too high clog P might not show potent activity in vivo because of low cell permeability. The stability of newly synthesized lipophilic derivatives of HCAs is an important parameter for further in vivo investigation. The lipophilized HCAs collected from column chromatography was evaporated under reduced pressure; however, no hydrolytic reaction was observed in HPLC chromatograms (data not shown), indicating that the lipophilic derivatives were stable during this process. In conclusion, we demonstrated that CA and ChA possessing catechol structure inhibited Aβ fibrillization to a greater extent. Furthermore, higher antiamyloidal activities of HCA derivatives except for SA are associated with having a longer alkyl chain. These results suggest the importance of lipophilicity and catecholic structure on antiamyloidal activity of HCAs and provide understanding about the relationship between
405
Biotechnology and Applied Biochemistry
FIG. 4
Effect of lipophilized HCAs on Aβ (1–42) fibrillization. (A) Lipophilic derivatives of CA. (B) Lipophilic derivatives of FA. (C) Lipophilic derivatives of SA. (D) Lipophilic derivatives of ChA. HCAs, or their lipophilized derivatives (50 μM) were incubated for 24 H with 25 μM Aβ. Aβ fibrillization was assessed by ThT fluorescence intensity. Values with different superscripts are significantly different (P < 0.05).
antiamyloidal activity of phenolic compounds and their lipophilicity. Our results will contribute to the development of antiamyloidal agents derived from naturally occurring phenolic compounds.
5. Acknowledgements This study was supported by a grant-in-aid from the Ministry of Education, Culture, Sports, Science and Technology of Japan (23580161). We also appreciate Mr. Takahiro Ohno for his assistance with electron microscopy.
6. References
FIG. 5
406
Electron micrographs after 24 H of incubation of Aβ (1–42) in the absence (A) and presence (B) of CA16. CA16 (50 μM) were incubated for 24 H with 25 μM Aβ. The samples were negatively stained with 1% phosphotungstic acid. The scale bar indicates 200 nm.
[1] Williamson, G., and Manach, C. (2005) Am. J. Clin. Nutr. 81, 243S–255S. [2] Aggarwal, B. B., Kumar, A., and Bharti, A. C. (2003) Anticancer Res. 23, 363–398. [3] Baur, J. A., Pearson, K. J., Price, N. L., Jamieson, H. A., Lerin, C., Kalra, A., Prabhu, V. V., Allard, J. S., Lopez-Lluch, G., and Lewis, K. (2006) Nature 444, 337–342. [4] Korkina, L. (2007) Cell Mol. Biol. 53, 15–25. [5] Blennow, K., Leon, M. d., and Zetterberg, H. (2006) Lancet 368, 387–403. [6] Ono, K., Yoshiike, Y., Takashima, A., Hasegawa, K., Naiki, H., and Yamada, M. (2003) J. Neurochem. 87, 172–181. [7] Ono, K., Hamaguchi, T., Naiki, H., and Yamada, M. (2006) Biochim. Biophys. Acta 1762, 575–586. [8] Ono, K., Hirohata, M., and Yamada, M. (2005) Biochem. Biophys. Res. Commun. 336, 444–449. [9] Porat, Y., Abramowitz, A., and Gazit, E. (2006) Chem. Biol. Drug Des. 67, 27–37.
Antiamyloid Effect of Esterified Phenolic Acids
[10] Garrido, J., Gaspar, A., Garrido, E. M., Miri, R., Tavakkoli, M., Pourali, S., Saso, L., Borges, F., and Firuzi, O. (2012) Biochimie 94, 961–967. [11] Salem, J. H., Humeau, C., Chevalot, I., Harscoat-Schiavo, C., Vanderesse, R., Blanchard, F., and Fick, M. (2010) Proc. Biochem. 45, 382–389. [12] Serafim, T. L., Carvalho, F. S., Marques, M. P., Calheiros, R., Silva, T., Garrido, J., Milhazes, N., Borges, F., Roleira, F., Silva, E. T., Holy, J., and Oliveira, P. J. (2011) Chem. Res. Toxicol. 24, 763–774. [13] Chebil, L., Humeau, C., Falcimaigne, A., Engasser, J. M., and Ghoul, M. (2006) Proc. Biochem. 41, 2237–2251. [14] Jayaprakasam, B., Vanisree, M., Zhang, Y., Dewitt, D. L., and Nair, M. G. (2006) J. Agric. Food Chem. 54, 5375–5381. [15] Walle, T. (2007) Mol. Pharm. 4, 826–832. [16] Figueroa-Espinoza, M. C., and Villeneuve, P. (2005) J. Agric. Food Chem. 53, 2779–2787. [17] Clifford, M. N. (1999) J. Sci. Food Agr. 79, 362–372.
Biotechnology and Applied Biochemistry
[18] Compton, D. L., Laszlo, J. A., and Berhow, M. A. (2000) J. Am. Oil Chem. Soc. 77, 513–519. [19] Stevenson, D. E., Parkar, S. G., Zhang, J., Stanley, R. A., Jensen, D. J., and Cooney, J. M. (2007) Enzyme Microb. Tech. 40, 1078–1086. [20] Katayama, S., Ogawa, H., and Nakamura, S. (2011) J. Agric. Food Chem. 59, 12691–12696. [21] Ono, K., Hasegawa, K., Naiki, H., and Yamada, M. (2004) J. Neurosci. Res. 75, 742–750. [22] Reinke, A. A., and Gestwicki, J. E. (2007) Chem. Biol. Drug Des. 70, 206–215. [23] Sawaya, M. R., Sambashivan, S., Nelson, R., Ivanova, M. I., Sievers, S. A., Apostol, M. I., Thompson, M. J., Balbirnie, M., Wiltzius, J. J. W., McFarlane, H. T., Madsen, A. O., Riekel, C., and Eisenberg, D. (2007) Nature 447, 453–457. [24] Murakami, A., Kadota, M., Takahashi, D., Taniguchi, H., Nomura, E., Hosoda, A., Tsuno, T., Maruta, Y., Ohigashi, H., and Koshimizu, K. (2000) Cancer Lett. 157, 77–85. [25] Waring, M. J. (2010) Expert Opin. Drug Dis. 5, 235–248.
407
