- Email: [email protected]
Heme binding site in apomyoglobin may be effectively targeted with small molecules to control aggregation
The International Journal of Biochemistry & Cell Biology 45 (2013) 299–307
Contents lists available at SciVerse ScienceDirect
The International Journal of Biochemistry & Cell Biology journal homepage: www.elsevier.com/locate/biocel
Heme binding site in apomyoglobin may be effectively targeted with small molecules to control aggregation Mehrnaz Azami-Movahed a , Sajad Shariatizi a , Marjan Sabbaghian b , Atiyeh Ghasemi a , Azadeh Ebrahim-Habibi c,∗∗ , Mohsen Nemat-Gorgani a,d,∗ a
Institute of Biochemistry and Biophysics, University of Tehran, P.O. Box 13145-1384, 1417614411 Tehran, Iran Department of Andrology, Reproductive Biomedicine Research Center, Royan Institute, ACECR, Tehran, Iran c Endocrinology and Metabolism Research Center, Tehran University of Medical Sciences, 1411413137 Tehran, Iran d Stanford Genome Technology Center, Stanford University, Palo Alto, CA, USA b
a r t i c l e
i n f o
Article history: Received 3 June 2012 Received in revised form 8 October 2012 Accepted 11 October 2012 Available online 23 October 2012 Keywords: Apomyoglobin Amyloid Aggregation Chrysin Nile red
a b s t r a c t A number of ligands with affinities for the heme binding site of apomyoglobin were tested to control amorphous and fibrillar aggregation in the protein. Several techniques, including fluorescence, dynamic light scattering, transmission electron microscopy, dot blot analysis combined with viability studies were employed for structural characterization and cytotoxicity assessment of the intermediate and final protein structures formed during the aggregation process. Of the small molecules investigated, chrysin and Nile red with high structural similarities to heme were chosen for further studies. Only fibril formation was found to be prevented by Nile red, while chrysin, with a greater structural flexibility, was able to prevent both types of aggregate formation. The two ligands were found to influence aggregation at different stages of intermediate structure formation, an ability determined by their degrees of similarities with heme. Based on structural characterization and toxicity studies, it is concluded that ligands similar in structure to heme may be effective in influencing various stages of aggregate formation and toxicity potencies of the protein structures. Since metalloproteins constitute more than thirty percent of all known proteins, it is concluded that the present strategy may be of general significance. © 2012 Elsevier Ltd. All rights reserved.
1. Introduction Myoglobin is a globular, all-alpha metalloprotein, whose “apo” form (i.e. devoid of heme) is well-characterized relative to its folding dynamics (Jamin, 2005; Fink et al., 1998). Native, molten globule and unfolded states have been obtained for the apo form of horse myoglobin at pH 6.5, pH 4.2 and pH 2, respectively (Fink et al., 1998; Tcherkasskaya and Ptitsyn, 1999). Amyloid formation of myoglobin has been reported for the apo structure at alkaline pH, while the heme-containing form (holomyoglobin) could not be driven toward this state (Fandrich et al., 2003). On the other hand, a mutant form (W7FW14F) of sperm whale apomyoglobin showed fibrillation at physiologic pH and room temperature (Sirangelo et al., 2004, 2009; Infusini et al., 2012). In the present study, different features of
∗ Corresponding author at: Stanford Genome Technology Center, 855 S. California Avenue, Palo Alto, CA 94304, USA. Tel.: +1 650 812 1972. ∗∗ Corresponding author at: Endocrinology and Metabolism Research Center, Tehran University of Medical Sciences, Shariati Hospital, North Kargar Avenue, 1411413137 Tehran, Iran. Tel.: +98 21 88220038. E-mail addresses: [email protected], [email protected] (A. Ebrahim-Habibi), [email protected] (M. Nemat-Gorgani). 1357-2725/$ – see front matter © 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.biocel.2012.10.004
aggregation in wild-type horse apomyoglobin have been investigated, including determination of the roles of pH and electrostatic repulsion on aggregation pathways, which may result in amorphous aggregation or fibril formation. Small molecules were tested as inhibitors of both events and toxicities of intermediate structures were determined. It is suggested that stability of apomyoglobin could affect formation of “on pathway” or “off pathway” intermediates. One common event that is thought to be of importance in aggregate formation of apo-metalloproteins is destabilization of the native protein structure as a result of metal removal. For example, amyloid fibrils of bovine ␣-lactalbumin form either at low pH or by disulfide reduction. These conditions cause release of Ca2+ , leading to reduction of thermal stability and formation of partially folded conformations (Goers et al., 2002; Veprintsev et al., 1997; Ebrahim-Habibi et al., 2010a). Similarly, loss of metal ions, disulfide reduction, and pathologic mutations may drive superoxide dismutase toward amyloid fibrillation (Oztug Durer et al., 2009; Fee and Phillips, 1975; Mei et al., 1992). Fibril formation is also observed in apo-carbonic anhydrase, which may take up a pre-molten globular structure after removal of its zinc ion (Es-haghi et al., 2012). More specifically, conditions leading to loss of heme group of bovine cytochrome c (i.e. mutations in the heme binding site, or incubation
300
M. Azami-Movahed et al. / The International Journal of Biochemistry & Cell Biology 45 (2013) 299–307
at 75 ◦ C and under mild alkaline conditions) results in formation of unfolded structures and consequently, amyloid fibrils (Pertinhez et al., 2001; de Groot and Ventura, 2005). The structure stabilizing role of heme has been demonstrated for the heme-binding section of myoglobin (microglobin) (Ji et al., 2008), and further suggested to prevent apomyoglobin amyloid formation (Iannuzzi et al., 2007).
Ten microliters of well-mixed incubation sample were added to 190 l of the Congo red solution and incubated for 30 min. Absorbance spectra were recorded (400–600 nm) using a Shimadzu UV–visible spectrophotometer (Kyoto, Japan).
2. Materials and methods
2.7. Circular dichroism (CD) measurement
Horse myoglobin, Thioflavin T, Congo red, borate, 2-butanone, dialysis tubing, MTT, Nile red, and chrysin were purchased from Sigma (St Louis, MO, USA). ANS was obtained from Fluka. Glycine, DMSO and all salts and organic solvents were obtained from Merck (Darmstadt, Germany). RPMI medium were purchased from Gibco. FBS, horse serum, streptomycin and penicillin were obtained from Biosera (England). Ultra centrifuge filters were purchased from Millipore and A11 antibody was obtained from Chemicon.
CD spectra in the far-UV region (190–260 nm) were obtained on an AVIV 215 spectropolarimeter (Aviv Associates, Lakewood, NJ, USA), using a 1 mm path cell at room temperature. The protein concentration was 0.4 mg/ml.
2.1. Preparation of apomyoglobin (Teale, 1959) Apomyoglobin was prepared by phase extraction process. Holomyoglobin was dissolved in 0.01 M HCl and mixed with 2butanone on ice. Upon phase separation, the organic layer was removed and the procedure was repeated until clear solution appeared. Then, the sample was dialysed overnight in distilled water to remove extra butanone. Ultra centrifugation was employed to concentrate the apoprotein solutions. 2.2. Turbidity measurements Different buffers (phosphate, borate, sodium acetate), pH (4.2, 6.5, and 9), protein concentrations (0.2–0.4 mg/ml) and temperatures (60–70 ◦ C) were used when screening for the best conditions. Finally, solutions containing 0.3 mg/ml apomyoglobin in sodium acetate (10 mM), sodium phosphate (10 mM) and pH 6.5 and different ligands were prepared and incubated at 65 ◦ C. To monitor protein aggregation, turbidity measurements were made at 350 nm on a Shimadzu UV–visible spectrophotometer (Kyoto, Japan) (Rezaei-Ghaleh et al., 2007a). Temperatures were controlled to within ±0.1 ◦ C. 2.3. Thioflavin T (ThT) binding assay All fluorescence experiments were carried out on a CaryEclipse VARIAN fluorescence spectrophotometer. To investigate fibrillation, apomyoglobin samples were added to ThT solution in a molar ratio of 1:2.5, then mixed thoroughly and incubated for 5 min. Fluorescence excitation and emission were set at 440 and 482 nm and slit widths were set at 5 nm and 10 nm, respectively. 2.4. 8-Anilino-1-naphthalene sulfonate (ANS) fluorescence assays The excitation wavelength was 350 nm and emission spectra were recorded between 400 and 600 nm. Excitation and emission slit widths were set at 5 nm. The final concentrations of ANS and protein were 88 M and 0.03 mg/ml respectively.
2.6. Congo red absorbance assays
2.8. Transmission electron microscopy (TEM) Copper 400 mesh grid was covered with carbon-coated formvar film followed by adsorption of 10 l of apomyoglobin samples. After 2 min, excess fluid was removed with a paper filter, and 1% uranyl acetate added. After another minute, excess dye was removed. Finally, the grids were monitored with a CEM 902A Zeiss microscope (Oberkochen, Germany) and Philips (Japan). 2.9. Dynamic light scattering Dynamic light scattering (DLS) studies were performed on a zeta potential and particle size analyzer (Brookhaven Instrument, Holtsville, NY 11742-1896, USA). The size distribution/abundance of particles was studied in the absence and presence of desired ligands. The final protein concentration used was 58 M. A laser of 657 nm with a fixed detector angle of 90◦ was used. DLS studies were carried out at least in triplicates. 2.10. Dot blot analysis Dot immunoblot analysis was performed to investigate reactivity of anti-oligomeric A11 antibody against apomyoglobin oligomers. Aliquots of protein samples were spotted onto nitrocellulose membranes and dried. Membranes were blocked with 10% dry milk and incubated with A11 antibody (1:1000 dilution) overnight, then incubated with goat-anti rabbit HRP as the secondary antibody (1:2000 dilution) for 1 h. Dot visualization was carried out in a dark room. 2.11. Docking studies Docking was performed with Autodock vina (Trott and Olson, 2010). The 1DWR.pdb file (protein devoid of heme group) was used as receptor. Grid box of 48 × 48 × 44 points was used with ˚ and the grid box center was placed on x = 13.583, a spacing of 1.0 A, y = 21.646, and z = 8.308. The box encompassed the whole protein molecule. Ligands were prepared using Molecular Operating Environment (MOE), 2010.10 (Chemical Computing Group Inc., 1010 Sherbooke St. West, Suite #910, Montreal, QC, Canada, H3A 2R7). Gasteiger charges were assigned to protein and ligand molecules. Exhaustiveness was set on 20 and 100 poses were generated for the ligands. Preparation of the image representing the best pose of each ligand was performed with MOE 2010.10.
2.5. Nile red fluorescence assay 2.12. MTT assay Apomyoglobin (85 M) was added to Nile red (0.3 M) at pH 7 and the increase and shift occurring in the fluorescence emission was monitored. The excitation wavelength was 580 nm and emission spectra were recorded between 600 and 700 nm. Excitation and emission slit widths were both set at 5 nm.
Cell viability assay was assessed using rat pheochromocytoma (PC12) cell line. After 24 h of incubation with samples pretreated under amyloidogenic conditions in the absence and presence of ligands, cells were washed with PBS. 100 l of MTT stock solution
M. Azami-Movahed et al. / The International Journal of Biochemistry & Cell Biology 45 (2013) 299–307
(5 mg/ml in PBS) was added to 900 l of DMEM (without phenol red) containing 10% bovine calf serum followed by incubation for 3 h. Solutions were aspirated, and cells were treated with DMSO for 20 min. MTT absorbance was determined at 570 and 690 nm. All experiments were run in triplicates.
301
3.1. Apomyoglobin aggregation
et al., 2002; Lin et al., 2001; Nilsberth et al., 2001; Weise et al., 2010; Meratan and Nemat-Gorgani, 2012; Meratan et al., 2011). Exposure of hydrophobic sites was monitored using ANS fluorescence. Maximum ANS binding occurred at 20 h after incubation (corresponding to oligomeric species), followed by a decrease to reach a minimum after 48 h, when fibrils are formed, and a new fold is reached (results not shown). The ability of ANS in binding to the heme binding site of native apomyoglobin (Bismuto et al., 1996; Stryer, 1965), suggests an important role for this hydrophobic site in the amyloid structure formation process.
3.1.1. Fibrillation Apomyogloin fibrillation has been studied as an interesting example of a classical all-alpha protein, with the capacity to undergo transition to amyloid beta structure formation (Fandrich et al., 2001). In one of the reports describing amyloid formation in horse muscle apomyoglobin, role of unfolded segments was suggested to be more important than partially unfolded structures (Fandrich et al., 2001), and involvement of the Nterminal segment has been highlighted, especially the section corresponding to residues 1–29 (Picotti et al., 2007; Infusini et al., 2012). In the present study, after obtaining amyloid structures from horse muscle apomyoglobin, intermediate forms were characterized, and cytotoxicity of the various structures was assessed. In order to induce amyloid structures, initially various conditions were tested including use of a wide range of pH (2.5–9), and protein concentrations of 1–4 mg/ml at temperatures of 37 ◦ C and 65 ◦ C. Fibrillation was observed at pH 9 and 65 ◦ C, similar to conditions reported previously (Fandrich et al., 2003). An incubation time of 14 days was used in order to obtain a suitable population of mature fibrils. Although fibril formation could be observed after 24 h, amorphous species were also present at that time point. It should be mentioned that the incubated samples were not subjected to agitation, making the process slower but more reproducible. Accordingly, the selected conditions for further studies consisted of incubation of a 1 mg/ml protein solution made in 50 mM, pH 9 glycine buffer at 65 ◦ C. Amyloid formation under these conditions was tested using common procedures used for this type of studies, i.e. Congo red absorbance, Thioflavin T (ThT) fluorescence, and CD-monitored changes in secondary structure. CD spectra taken at different times are indicative of a gradual transition from alpha to beta structure, which could possibly pass from partially unfolded states of apomyoglobin that are known to contain a native-like core (Fink et al., 1998). Direct evidence of fibril presence was assessed by TEM. Kinetics of fibril formation was studied by following the changes occurring in ThT fluorescence intensity over the incubation period. The observed lag-phase, an exponential growth phase, and a stationary phase corresponded to the typical pattern of a nucleation-dependent behavior (Bhak et al., 2009), in accord with a random nucleation process previously observed for this protein (Fandrich et al., 2006) (results not shown). Cytotoxicity of apomyoglobin amyloid structures was assessed by the MTT assay. PC12 cells were exposed to various species, i.e. holo and apo monomers and oligomers (obtained after 20 h of incubation), structures formed after 4 and 7 days and mature fibrils (14 days). Toxic species were found to be oligomeric structures which possess reactivity to A11 antibody and the TEM image was also indicative of these structures. On the other hand, mature fibrils (and structures formed after 4 and 7 days) revealed no toxicity (results not shown). Experimental data indicate that toxicity of amyloidogenic proteins is caused by prefibrillar intermediates which are formed in the early stages of fibrillation. This toxicity has been associated with their exposed hydrophobic surfaces, which are able to interact with cell membranes (Bucciantini et al., 2002; Lashuel
3.1.2. Amorphous aggregation Self-assembly of polypeptide chains leading to formation of high molecular weight non-ordered aggregates is related to the protein environment conditions such as temperature, salts type and concentration, and protein concentration (Dumetz et al., 2008; Ebrahim-Habibi et al., 2010b). In order to investigate amorphous aggregation of horse muscle apomyoglobin, acidic and alkaline pH values (4.2, 6.5 and 9) close to the protein pI of 7.3 (Banks and Paquette, 1995), were used in the 60–70 ◦ C temperature range [(Tm is 61 ± 2 ◦ C (Fandrich et al., 2003)]. The process was assessed by the increase in the optical density of insoluble particles (Guagliardi et al., 1995; Paulikova et al., 1998; Li et al., 2001; Rafikova et al., 2003; Rezaei-Ghaleh et al., 2007b; Qiu and Macrae, 2008). Amorphous aggregation at considerable extent was found to occur at pH 6.5, where, interestingly, apomyoglobin has a native conformation (Tcherkasskaya and Ptitsyn, 1999). Indeed, formation of amorphous aggregates has been reported to be favored under conditions where proteins adopt a conformation close to their native state (Qin et al., 2007; Morshedi et al., 2010). It should be mentioned that at acidic pH (i.e. pH = 4.2), where the protein is in a partially folded state, we did not observe any significant amount of aggregation in the course of 10 min (Suppl. Fig. 1). In a previous report (Fandrich et al., 2003), in order to study and analyze the amorphous aggregate form of horse apomyoglobin, acidic pH of 4.1 was used to generate these forms. However, an incubation time period of more than 4 days (at 50 ◦ C) was found necessary to reach the “flake-like” aggregates, while shorter incubation times did not bring such results (Fandrich et al., 2003). As expected, the extent of aggregation was found to increase with increasing temperature (Fig. 1A), and protein concentration (Fig. 1B). Based on these results, a 0.3 mg/ml concentration of the protein was used at 65 ◦ C in subsequent experiments, in order to obtain an appreciable aggregation. Congo red and TEM were used to further investigate insoluble aggregate formation (Fig. 1C and D). The TEM image (Fig. 1D) is indicative of a non-fibrillar aggregate morphology (Nilsson, 2004), and clearly different from TEM images obtained from apomyoglobin fibrils (Fandrich et al., 2003). The Congo red spectrum of the amorphous species presents a red shift in comparison with the spectrum obtained from the dye alone, but lacks the increase in absorbance which is seen in the fibrillar species. This may be due to the presence of beta-sheet structures in the amorphous species which are believed to share these structures with amyloid species, though probably occurring to a lesser extent (Tyedmers et al., 2010); furthermore, amorphous and fibrillar species have been suggested to stem from a common partially folded intermediate containing beta-sheet structures (Jiang et al., 2012; Ortore et al., 2011). Concerning insoluble amorphous aggregates obtained from apomyoglobin (under other conditions), the presence of beta-sheet structures has been reported in these species, which are nonetheless devoid of the “well defined orientations” that are characteristics of amyloid fibrils (Fandrich et al., 2003). As summarized in the scheme below, pH seems to have an important role with regard to the type of aggregates formed at a high temperature.
3. Results and discussion
302
M. Azami-Movahed et al. / The International Journal of Biochemistry & Cell Biology 45 (2013) 299–307
Fig. 1. Monitoring absorbance at 350 nm over 720 s as an indicator of amorphous aggregation in apomyoglobin. (A) Use of apomyoglobin concentration of 0.3 mg/ml at pH 6.5 incubated at various temperatures: 60 ◦ C (), 62 ◦ C (), 65 ◦ C (), 67 ◦ C (♦), 70 ◦ C (). (B) Use of various concentrations of apomyoglobin at pH 6.5 and 65 ◦ C: used concentrations were 0.2 mg/ml (), 0.25 mg/ml (), 0.3 mg/ml (), 0.35 mg/ml (♦), 0.4 mg/ml (). (C) Congo red absorbance spectra of apomyoglobin aggregates formed after 10 min incubation at pH 6.5 and 65 ◦ C (- - -), Congo red dye (–) and apomyoglobin amyloid structure obtained after incubation of apomyoglobin at pH 9 and 65 ◦ C for 2 weeks (. . .). (D) TEM image of apomyoglobin aggregates formed after 10 min incubation at pH 6.5 and 65 ◦ C. (E) Apomyoglobin (0.3 mg/ml) at pH 6.5 and 65 ◦ C was incubated with various concentrations of NaCl, of 0 (), 60 mM (♦), 80 mM (), 100 mM (), 400 mM (䊉), and 750 mM ().
In addition to protein stability, protein solubility and net charge may play critical roles (Calloni et al., 2005). Mutations have been reported to change protein net charge and destabilize the native conformation. Consequently, misfolded intermediates appear which are prone to aggregation and result into pathogenic deposits (Chiti et al., 2002; Chiti and Dobson, 2006). ␣-Synuclein is an unstructured protein possessing an excess of acidic residues and is prone to aggregation when its strong negative charge decreases at a low pH (Uversky et al., 2001). In contrast, many
globular proteins form amyloid aggregates at acidic pH, where they have high positive charge (Guijarro et al., 1998; Kad et al., 2003; Frare et al., 2004). An electrostatic component is then mostly recognized as having a role in various types of aggregate formation. Wild-type apomyoglobin could rapidly form amorphous aggregates upon heating at a high temperature and a pH close to neutral where it is uncharged (Banks and Paquette, 1995), while amyloid structures are generated when its net negative charge is increased. Adding NaCl to the protein environment may result in decreasing amounts of amorphous aggregates, and the effect is concentration-dependent (Fig. 1E). In studies on other proteins, aimed at differentiating conditions leading to amorphous or amyloid structures, it has been suggested that amorphous forms could be subjected to less entropic penalties (Rezaei-Ghaleh et al., 2011), and are mainly induced by a prevalence of hydrophobic interactions (Vetri et al., 2007b). On the other hand, formation of amyloid structures, requiring extensive conformational changes in proteins (Vetri et al., 2007a) would be favored by conditions
M. Azami-Movahed et al. / The International Journal of Biochemistry & Cell Biology 45 (2013) 299–307
303
Fig. 2. Dynamic light scattering of apomyoglobin alone (A), and in the presence of Nile red (B). Aliquots were taken after 3 weeks of incubation. Figures show size distribution and average diameter of particles. (C) Apomyoglobin fibrillation in the presence of Nile red monitored by TEM. Amorphous aggregates are observed after 2 weeks of apomyoglobin incubation with Nile red (0.3 M). (D) Cell viability assays using apomyoglobin aggregates formed after 2 weeks in the presence of Nile red, compared with a control cell sample.
where inter-molecular hydrogen bonds occur preferentially over hydrogen bonds that form with the solvent (e.g. water), indicating that electrostatic and hydrophobic forces could be decisive factors in defining aggregate morphology (Vetri et al., 2011). 3.2. Effect of ligands on apomyoglobin aggregation Inhibition of pathologic aggregation by small molecules is now seriously considered as a promising therapeutic approach (Amijee and Scopes, 2009; Re et al., 2010). Both amyloid and amorphous aggregates have been successfully targeted by these compounds, which could also be used as probes to further elucidate the aggregation mechanism itself (Riviere et al., 2009; Sirangelo and Irace, 2010; Sood et al., 2011; Sabbaghian et al., 2011; Rezaei-Ghaleh et al., 2007a; Vilasi et al., 2008, 2011). In the present study, the effects of Nile red and chrysin have been investigated on apomyoglobin aggregation. Nile red is a fluorescent probe which has been used to study presence of hydrophobic surfaces in proteins (Sackett and Wolff, 1987). More specifically, it has been shown to be able to interact with the heme binding pocket of horse apomyoglobin (Polverini et al., 2006). On the other hand, presence of heme may stabilize horse myoglobin structure, increasing its Tm by about 20 ◦ C, and preventing fibrillation (Fandrich et al., 2003). When used as an additive to the amyloidogenic environment, it has been shown to prevent amyloid formation and the resulted cytotoxicity of the double mutant form of whale sperm apomyoglobin (Iannuzzi et al., 2007). Addition of micromolar concentrations of Nile red (0.3 M) to apomyoglobin (85 M) at neutral pH resulted in an increase in the fluorescence emission intensity and a blue shift from 665 nm to 628 nm, indicative of binding of the probe to a hydrophobic pocket in the protein molecule, similar to a previous report (Polverini et al., 2006). To induce fibril formation, pH was adjusted to 9, and the
protein was incubated at 65 ◦ C in the presence of Nile red. As monitored by dynamic light-scattering, after 3 weeks of incubation, control samples formed amyloid-like fibrils, showing diameters of 400 and 8200 nm (Fig. 2A), indicative of formation of intermediates and mature fibrils, respectively. In contrast, samples incubated with Nile red showed two populations of particles, 225 nm and 868 nm (Fig. 2B). TEM images taken after 2 weeks of incubation in the presence of the ligand showed presence of amorphous aggregates (Fig. 2C). Interestingly, these structures were found to be non-toxic by the MTT test (Fig. 2D). This result is similar to that reported using heme to prevent amyloid formation of the double mutant form of sperm whale apomyoglobin (Iannuzzi et al., 2007). The “remodeling” ability of some ligands that results into formation of non-toxic unstructured aggregates instead of the toxic species formed during the amyloidogenic process, has recently been reported for resveratrol (Ladiwala et al., 2010), biflavonoids (Thapa et al., 2011) and rifamycin (Woods et al., 2011). More generally, depending on their structures, aromatic small molecules are believed to act on the amyloidogenic process through different mechanisms, resulting in formation of various non-fibrillar structures (Ladiwala et al., 2011). In the present study, targeting the heme pocket seems to be an efficient strategy to prevent both fibrillation and cytotoxicity of the products generated from horse apomyoglobin. Interaction of Nile red with the heme binding pocket is suggested to occur via a binding mode that would partially overlap the heme position (Polverini et al., 2006). A blind docking experiment was performed with Nile red in apomyoglobin, and the best observed pose is shown in Fig. 3. It seems that this docking tool would give reasonable results, since the method was able to find the real binding site for heme (results not shown). Further to hydrophobic interactions that could occur between Nile red and the aliphatic residues of heme binding pocket, pi–pi interactions were also detected, occurring between His 97 and Phe 43 and one of Nile red aromatic components. Verification
304
M. Azami-Movahed et al. / The International Journal of Biochemistry & Cell Biology 45 (2013) 299–307
Fig. 5. Cell viability of PC-12 cells. Cells were exposed to apomyoglobin (25 M) aggregates formed after 2 weeks in the presence of chrysin (0, 50, 100 M), compared with a control sample.
Fig. 3. Best pose of docking results obtained for Nile red in horse apomyoglobin. The binding occurs in the hydrophobic heme-binding pocket. Pi–pi interactions could be seen, with participation of His 97 and Phe 43 and the aromatic components of Nile red.
of the first five docking poses revealed that at least one of these interactions would occur in all of these situations. As a polyphenolic compound, the flavonoid chrysin appeared to have the ability of influencing apomyoglobin aggregation. The effectiveness of flavonoids as anti-amyloid compounds has been reported in multiple studies, either on the process itself (Gasiorowski et al., 2011) or on its cytotoxicity (Lebeau et al., 2001; Zhu et al., 2007). Consequently, apomyoglobin was incubated with
50 and 100 M chrysin and samples were taken for ThT binding assay at various times. Our results indicated that the ligand, when used at 50 M concentration, was ineffective, while the presence of 100 M chrysin prolonged the lag phase without suppressing amyloid fibril formation. As shown in Fig. 4A, when chrysin (100 M) was present, after the lag time, fluorescence enhancement was observed, but reached only 50% of the maximal value recorded at the plateau phase in the absence of the ligand. Average diameter of control samples was found to be about 5000 nm as suggested by DLS runs, whereas the treated samples showed two populations of particles with average diameters of 170 and 1500 nm (results not shown). After 3 weeks, control fibrils extend to 8200 nm (Fig. 2A), but in samples incubated with chrysin, three populations of aggregates were observed with average diameters of 200, 1600, and 6800 nm (Fig. 4B). TEM images indicated formation of oligomeric structures in the presence of chrysin (Fig. 4C). Since chrysin was found to increase the lag phase, and based on the TEM results, apomyoglobin should have stayed in a cytotoxic oligomeric state in its presence. This was proven by incubating PC12 cells with
Fig. 4. (A) ThT fluorescence kinetics of apomyoglobin incubated at pH 9 and 65 ◦ C for 23 days in the absence () and presence () of chrysin (100 M). The lag phase is observed to increase (reaching 2 weeks) in the presence of chrysin. (B) Dynamic light scattering of apomyoglobin incubated at pH 9 and 65 ◦ C in the presence of chrysin. Aliquots of the protein sample were taken after 3 weeks of incubation. Figure shows size distribution and average diameter of particles. (C) Apomyoglobin fibrillation in the presence of chrysin monitored by TEM. The sample was incubated at pH 9 and 65 ◦ C with chrysin (100 M) for 2 weeks. Oligomeric intermediates are observed.
M. Azami-Movahed et al. / The International Journal of Biochemistry & Cell Biology 45 (2013) 299–307
305
Fig. 6. Surface hydrophobicity of apomyoglobin monitored by ANS fluorescence. Spectra were taken after 10 h (A), 2 days (B), and 3 days (C), from apomyoglobin in the absence of ligands (. . .), and in the presence of Nile red (—) or chrysin (- -). Control spectra of ANS alone are also shown (− ··).
the apomyoglobin structures formed in the presence of the effective concentration (100 M) of chrysin, which caused a decrease in reduced MTT levels (Fig. 5). Chrysin itself is devoid of toxicity: when the cells were incubated with different amounts of chrysin (25–100 M) or DMSO (0.1%), no toxicity was observed (results not shown). A comparison of the best docking pose of chrysin with the best pose of Nile red suggested the possibility for involvement of a common binding site (results not shown). However, pi–pi interactions are more favored with the fluorescence probe, in comparison with chrysin. Nile red has also a larger structure and one more aromatic component, which may help it to make more extensive interactions, while its more rigid structure would make it more similar to the natural ligand (heme). The fact that both ligands are binding to the heme binding site was further confirmed by an ANS fluorescence experiment, where samples of incubated protein with the ligands were taken at 10 h, 2 days and 3 days, and probed with ANS fluorescence (Fig. 6A–C, respectively). Results indicate a lower ANS fluorescence intensity when the ligands are present at day 2. The relatively low difference of intensity on 10 h and three days may be due to the competitive character of ligand and ANS binding for the same site. A possible mechanism of action of these two ligands on amyloid formation is presented in the below scheme:
Interestingly, chrysin is able to diminish amorphous aggregation of apomyoglobin. Upon incubation with 50–500 M chrysin at 65 ◦ C and pH 6.5, turbidity as indicative of aggregation was clearly diminished, in a dose-dependent manner (Fig. 7A). Moreover, DLS
measurements showed that the effective diameter of the aggregates decreased considerably in the presence of chrysin (Fig. 7B). A remarkable feature of this ligand is the effective inhibitory dosage (EC50% = 250 M): this value is lower than the usual concentrations used to inhibit amorphous aggregation (e.g. Sabbaghian et al., 2011). Amorphous aggregates formation is mostly related to increasing hydrophobic surfaces becoming exposed (Vetri et al., 2007b), thus chrysin is probably causing lowering of that exposure to some extent, but is not able to completely preserve the native fold. It could also be suggested that this small molecule is acting in a similar manner in both processes. At any rate, some stabilization of the structure occurs, and this experiment is suggestive of a possible use for aromatic ligands as preventive aids in deleterious amorphous aggregation. On the other hand, Nile red (used from 0.3 M to mM range) was not able to prevent amorphous aggregation of apomyoglobin. This is suggestive of the fact that Nile red is acting more similarly as the native ligand (heme), since holomyoglobin was also found to form amorphous aggregates when incubated at high temperatures (results not shown). Thus, depending on the process that is targeted (i.e. amorphous aggregates or amyloid formation), structures possessing different similarities with the native ligand could be effectively used.
306
M. Azami-Movahed et al. / The International Journal of Biochemistry & Cell Biology 45 (2013) 299–307
Fig. 7. (A) Amorphous aggregation of apomyoglobin (0.3 mg/ml) at pH 6.5 and 65 ◦ C incubated with various concentrations of chrysin: 0 M (), 50 M (♦), 100 M (), 200 M (), 300 M (䊉), 400 M () and 500 M (). (B) Effective diameter of amorphous aggregates alone () and in the presence of 250 M chrysin (), monitored by DLS (incubation was done at pH 6.5 and 65 ◦ C).
4. Conclusion In the present study, the inhibitory effects of two small molecules on amyloid fibrillation of wild type horse apomyoglobin are described. It seems that chrysin keeps apomyoglobin in a very toxic oligomeric state, while Nile red causes formation of small nontoxic aggregates, in a similar manner to heme. Results presented are indicative of involvement of the heme binding pocket in amorphous and amyloid aggregation of apomyoglobin. Since heme binding is related to stabilization of myoglobin, it could be suggested that stability of apomyoglobin structure may affect morphology and toxicity of intermediate structures. With metalloproteins representing over 30% of all proteins, the described strategy may provide a useful approach for control of protein aggregation and related cytotoxicity. Conflict of interest All authors declare not to have any conflict of interest. Acknowledgment This work was supported by grants from the research council of the University of Tehran and the Iranian National Science Foundation (INSF). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/ j.biocel.2012.10.004. References Amijee H, Scopes DI. The quest for small molecules as amyloid inhibiting therapies for Alzheimer’s disease. Journal of Alzheimer’s Disease 2009;17:33–47. Banks PR, Paquette DM. Monitoring of a conjugation reaction between fluorescein isothiocyanate and myoglobin by capillary zone electrophoresis. Journal of Chromatography A 1995;693:145–54. Bhak G, Choe YJ, Paik SR. Mechanism of amyloidogenesis: nucleation-dependent fibrillation versus double-concerted fibrillation. BMB Reports 2009;42:541–51. Bismuto E, Sirangelo I, Irace G, Gratton E. Pressure-induced perturbation of apomyoglobin structure: fluorescence studies on native and acidic compact forms. Biochemistry 1996;35:1173–8. Bucciantini M, Giannoni E, Chiti F, Baroni F, Formigli L, Zurdo J, et al. Inherent toxicity of aggregates implies a common mechanism for protein misfolding diseases. Nature 2002;416:507–11. Calloni G, Zoffoli S, Stefani M, Dobson CM, Chiti F. Investigating the effects of mutations on protein aggregation in the cell. Journal of Biological Chemistry 2005;280:10607–13. Chiti F, Calamai M, Taddei N, Stefani M, Ramponi G, Dobson CM. Studies of the aggregation of mutant proteins in vitro provide insights into the genetics of amyloid diseases. Proceedings of the National Academy of Sciences of the United States of America 2002;99(Suppl. 4):16419–26.
Chiti F, Dobson CM. Protein misfolding, functional amyloid, and human disease. Annual Review of Biochemistry 2006;75:333–66. de Groot NS, Ventura S. Amyloid fibril formation by bovine cytochrome c. SPECTROSCOPY – OTTAWA THEN AMSTERDAM 2005;19:199–205. Dumetz AC, Chockla AM, Kaler EW, Lenhoff AM. Effects of pH on protein–protein interactions and implications for protein phase behavior. Biochimica et Biophysica Acta 2008;1784:600–10. Ebrahim-Habibi MB, Amininasab M, Ebrahim-Habibi A, Sabbaghian M, NematGorgani M. Fibrillation of alpha-lactalbumin: effect of crocin and safranal, two natural small molecules from Crocus sativus. Biopolymers 2010a;93:854–65. Ebrahim-Habibi A, Morshedi D, Rezaei-Ghaleh N, Sabbaghian M, Nemat-Gorgani M. Protein–protein interactions leading to aggregation: perspectives on mechanism, significance and control. Journal of the Iranian Chemical Society 2010b;7:521–44. Es-haghi A, Shariatizi S, Ebrahim-Habibi A, Nemat-Gorgani M. Amyloid fibrillation in native and chemically-modified forms of carbonic anhydrase. II: role of surface hydrophobicity. Biochimica et Biophysica Acta (BBA) – Proteins & Proteomics 2012;1824:468–77. Fandrich M, Fletcher MA, Dobson CM. Amyloid fibrils from muscle myoglobin. Nature 2001;410:165–6. Fandrich M, Forge V, Buder K, Kittler M, Dobson CM, Diekmann S. Myoglobin forms amyloid fibrils by association of unfolded polypeptide segments. Proceedings of the National Academy of Sciences of the United States of America 2003;100:15463–8. Fandrich M, Zandomeneghi G, Krebs MR, Kittler M, Buder K, Rossner A, et al. Apomyoglobin reveals a random-nucleation mechanism in amyloid protofibril formation. Acta Histochemica 2006;108:215–9. Fee JA, Phillips WD. The behavior of holo- and apo-forms of bovine superoxide dismutase at low pH. Biochimica et Biophysica Acta 1975;412:26–38. Fink AL, Oberg KA, Seshadri S. Discrete intermediates versus molten globule models for protein folding: characterization of partially folded intermediates of apomyoglobin. Folding and Design 1998;3:19–25. Frare E, Polverino De Laureto P, Zurdo J, Dobson CM, Fontana A. A highly amyloidogenic region of hen lysozyme. Journal of Molecular Biology 2004;340:1153–65. Gasiorowski K, Lamer-Zarawska E, Leszek J, Parvathaneni K, Yendluri BB, BlachOlszewska Z, et al. Flavones from root of Scutellaria baicalensis Georgi: drugs of the future in neurodegeneration? CNS & Neurological Disorders – Drug Targets 2011;10:184–91. Goers J, Permyakov SE, Permyakov EA, Uversky VN, Fink AL. Conformational prerequisites for alpha-lactalbumin fibrillation. Biochemistry 2002;41:12546–51. Guagliardi A, Cerchia L, Rossi M. Prevention of in vitro protein thermal aggregation by the Sulfolobus solfataricus chaperonin. Evidence for nonequivalent binding surfaces on the chaperonin molecule. Journal of Biological Chemistry 1995;270:28126–32. Guijarro JI, Sunde M, Jones JA, Campbell ID, Dobson CM. Amyloid fibril formation by an SH3 domain. Proceedings of the National Academy of Sciences of the United States of America 1998;95:4224–8. Iannuzzi C, Vilasi S, Portaccio M, Irace G, Sirangelo I. Heme binding inhibits the fibrillization of amyloidogenic apomyoglobin and determines lack of aggregate cytotoxicity. Protein Science 2007;16:507–16. Infusini G, Iannuzzi C, Vilasi S, Birolo L, Pagnozzi D, Pucci P, et al. Resolution of the effects induced by W→F substitutions on the conformation and dynamics of the amyloid-forming apomyoglobin mutant W7FW14F. European Biophysics Journal 2012;41:615–27. Jamin M. The folding process of apomyoglobin. Protein and Peptide Letters 2005;12:229–34. Ji HF, Shen L, Grandori R, Muller N. The effect of heme on the conformational stability of micro-myoglobin. The FEBS Journal 2008;275:89–96. Jiang D, Rauda I, Han S, Chen S, Zhou F. Aggregation pathways of the amyloidbeta(1–42) peptide depend on its colloidal stability and ordered beta-sheet stacking. Langmuir 2012., http://dx.doi.org/10.1021/la3021436. Kad NM, Myers SL, Smith DP, Smith DA, Radford SE, Thomson NH. Hierarchical assembly of beta2-microglobulin amyloid in vitro revealed by atomic force microscopy. Journal of Molecular Biology 2003;330:785–97.
M. Azami-Movahed et al. / The International Journal of Biochemistry & Cell Biology 45 (2013) 299–307 Ladiwala AR, Lin JC, Bale SS, Marcelino-Cruz AM, Bhattacharya M, Dordick JS, et al. Resveratrol selectively remodels soluble oligomers and fibrils of amyloid Abeta into off-pathway conformers. Journal of Biological Chemistry 2010;285:24228–37. Ladiwala AR, Dordick JS, Tessier PM. Aromatic small molecules remodel toxic soluble oligomers of amyloid beta through three independent pathways. Journal of Biological Chemistry 2011;286:3209–18. Lashuel HA, Hartley D, Petre BM, Walz T, Lansbury Jr PT. Neurodegenerative disease: amyloid pores from pathogenic mutations. Nature 2002;418:291. Lebeau A, Esclaire F, Rostene W, Pelaprat D. Baicalein protects cortical neurons from beta-amyloid (25–35) induced toxicity. Neuroreport 2001;12:2199–202. Li J, Zhang S, Wang C. Only the reduced conformer of alpha-lactalbumin is inducible to aggregation by protein aggregates. Journal of Biochemistry 2001;129:821–6. Lin H, Bhatia R, Lal R. Amyloid beta protein forms ion channels: implications for Alzheimer’s disease pathophysiology. FASEB Journal 2001;15:2433–44. Mei G, Rosato N, Silva Jr N, Rusch R, Gratton E, Savini I, et al. Denaturation of human Cu/Zn superoxide dismutase by guanidine hydrochloride: a dynamic fluorescence study. Biochemistry 1992;31:7224–30. Meratan AA, Ghasemi A, Nemat-Gorgani M. Membrane integrity and amyloid cytotoxicity: a model study involving mitochondria and lysozyme fibrillation products. Journal of Molecular Biology 2011;409:826–38. Meratan AA, Nemat-Gorgani M. Mitochondrial membrane permeabilization upon interaction with lysozyme fibrillation products: role of mitochondrial heterogeneity. Biochimica et Biophysica Acta 2012;1818:2149–57. Morshedi D, Ebrahim-Habibi A, Moosavi-Movahedi AA, Nemat-Gorgani M. Chemical modification of lysine residues in lysozyme may dramatically influence its amyloid fibrillation. Biochimica et Biophysica Acta 2010;1804:714–22. Nilsberth C, Westlind-Danielsson A, Eckman CB, Condron MM, Axelman K, Forsell C, et al. The ‘Arctic’ APP mutation (E693G) causes Alzheimer’s disease by enhanced Abeta protofibril formation. Nature Neuroscience 2001;4:887–93. Nilsson MR. Techniques to study amyloid fibril formation in vitro. Methods 2004;34:151–60. Oztug Durer ZA, Cohlberg JA, Dinh P, Padua S, Ehrenclou K, Downes S, et al. Loss of metal ions, disulfide reduction and mutations related to familial ALS promote formation of amyloid-like aggregates from superoxide dismutase. PLoS ONE 2009;4:e5004. Ortore MG, Spinozzi F, Vilasi S, Sirangelo I, Irace G, Shukla A, et al. Time-resolved small-angle X-ray scattering study of the early stage of amyloid formation of an apomyoglobin mutant. Physical Review E: Statistical, Nonlinear, and Soft Matter Physics 2011;84:061904. Paulikova H, Molnarova M, Podhradsky D. The effect of heparin and pentosan polysulfate on the thermal stability of yeast alcohol dehydrogenase. Biochemistry and Molecular Biology International 1998;46:887–94. Pertinhez TA, Bouchard M, Tomlinson EJ, Wain R, Ferguson SJ, Dobson CM, et al. Amyloid fibril formation by a helical cytochrome. FEBS Letters 2001;495:184–6. Picotti P, De Franceschi G, Frare E, Spolaore B, Zambonin M, Chiti F, et al. Amyloid fibril formation and disaggregation of fragment 1–29 of apomyoglobin: insights into the effect of pH on protein fibrillogenesis. Journal of Molecular Biology 2007;367:1237–45. Polverini E, Cugini G, Annoni F, Abbruzzetti S, Viappiani C, Gensch T. Molten globule formation in apomyoglobin monitored by the fluorescent probe Nile Red. Biochemistry 2006;45:5111–21. Qin Z, Hu D, Zhu M, Fink AL. Structural characterization of the partially folded intermediates of an immunoglobulin light chain leading to amyloid fibrillation and amorphous aggregation. Biochemistry 2007;46:3521–31. Qiu Z, Macrae TH. ArHsp21, a developmentally regulated small heat-shock protein synthesized in diapausing embryos of Artemia franciscana. Biochemical Journal 2008;411:605–11. Rafikova ER, Kurganov BI, Arutyunyan AM, Kust SV, Drachev VA, Dobrov EN. A mechanism of macroscopic (amorphous) aggregation of the tobacco mosaic virus coat protein. International Journal of Biochemistry and Cell Biology 2003;35:1452–60. Re F, Airoldi C, Zona C, Masserini M, La Ferla B, Quattrocchi N, et al. Beta amyloid aggregation inhibitors: small molecules as candidate drugs for therapy of Alzheimer’s disease. Current Medicinal Chemistry 2010;17:2990–3006. Rezaei-Ghaleh N, Ebrahim-Habibi A, Moosavi-Movahedi AA, Nemat-Gorgani M. Effect of polyamines on the structure, thermal stability and 2,2,2trifluoroethanol-induced aggregation of alpha-chymotrypsin. International Journal of Biological Macromolecules 2007a;41:597–604. Rezaei-Ghaleh N, Ebrahim-Habibi A, Moosavi-Movahedi AA, Nemat-Gorgani M. Role of electrostatic interactions in 2,2,2-trifluoroethanol-induced structural changes and aggregation of alpha-chymotrypsin. Archives of Biochemistry and Biophysics 2007b;457:160–9.
307
Rezaei-Ghaleh N, Giller K, Becker S, Zweckstetter M. Effect of zinc binding on beta-amyloid structure and dynamics: implications for Abeta aggregation. Biophysical Journal 2011;101:1202–11. Riviere C, Delaunay JC, Immel F, Cullin C, Monti JP. The polyphenol piceid destabilizes preformed amyloid fibrils and oligomers in vitro: hypothesis on possible molecular mechanisms. Neurochemical Research 2009;34:1120–8. Sabbaghian M, Ebrahim-Habibi A, Hosseinkhani S, Ghasemi A, Nemat-Gorgani M. Prevention of thermal aggregation of an allosteric protein by small molecules: some mechanistic insights. International Journal of Biological Macromolecules 2011;49:806–13. Sackett DL, Wolff J. Nile red as a polarity-sensitive fluorescent probe of hydrophobic protein surfaces. Analytical Biochemistry 1987;167:228–34. Sirangelo I, Malmo C, Iannuzzi C, Mezzogiorno A, Bianco MR, Papa M, et al. Fibrillogenesis and cytotoxic activity of the amyloid-forming apomyoglobin mutant W7FW14F. Journal of Biological Chemistry 2004;279:13183–9. Sirangelo I, Iannuzzi C, Vilasi S, Irace G, Giuberti G, Misso G, et al. W7FW14F apomyoglobin amyloid aggregates-mediated apoptosis is due to oxidative stress and AKT inactivation caused by Ras and Rac. Journal of Cellular Physiology 2009;221:412–23. Sirangelo I, Irace G. Inhibition of aggregate formation as therapeutic target in protein misfolding diseases: effect of tetracycline and trehalose. Expert Opinion on Therapeutic Targets 2010;14:1311–21. Sood A, Abid M, Sauer C, Hailemichael S, Foster M, Torok B, et al. Disassembly of preformed amyloid beta fibrils by small organofluorine molecules. Bioorganic and Medicinal Chemistry Letters 2011;21:2044–7. Stryer L. The interaction of a naphthalene dye with apomyoglobin and apohemoglobin. A fluorescent probe of non-polar binding sites. Journal of Molecular Biology 1965;13:482–95. Tcherkasskaya O, Ptitsyn OB. Molten globule versus variety of intermediates: influence of anions on pH-denatured apomyoglobin. FEBS Letters 1999;455:325–31. Teale F. Cleavage of the haem-protein link by acid methylethylketone. Biochimica et Biophysica Acta 1959;35:543. Thapa A, Woo ER, Chi EY, Sharoar MG, Jin HG, Shin SY, et al. Biflavonoids are superior to monoflavonoids in inhibiting amyloid-beta toxicity and fibrillogenesis via accumulation of nontoxic oligomer-like structures. Biochemistry 2011;50:2445–55. Trott O, Olson AJ. AutoDock Vina: improving the speed and accuracy of docking with a new scoring function, efficient optimization, and multithreading. Journal of Computational Chemistry 2010;31:455–61. Tyedmers J, Mogk A, Bukau B. Cellular strategies for controlling protein aggregation. Nature Reviews Molecular Cell Biology 2010;11:777–88. Uversky VN, Li J, Fink AL. Evidence for a partially folded intermediate in alpha-synuclein fibril formation. Journal of Biological Chemistry 2001;276: 10737–44. Veprintsev DB, Permyakov SE, Permyakov EA, Rogov VV, Cawthern KM, Berliner LJ. Cooperative thermal transitions of bovine and human apo-alpha-lactalbumins: evidence for a new intermediate state. FEBS Letters 1997;412:625–8. Vetri V, Canale C, Relini A, Librizzi F, Militello V, Gliozzi A, et al. Amyloid fibrils formation and amorphous aggregation in concanavalin A. Biophysical Chemistry 2007a;125:184–90. Vetri V, Librizzi F, Leone M, Militello V. Thermal aggregation of bovine serum albumin at different pH: comparison with human serum albumin. European Biophysics Journal 2007b;36:717–25. Vetri V, D’Amico M, Fodera V, Leone M, Ponzoni A, Sberveglieri G, et al. Bovine serum albumin protofibril-like aggregates formation: solo but not simple mechanism. Archives of Biochemistry and Biophysics 2011;508:13–24. Vilasi S, Iannuzzi C, Portaccio M, Irace G, Sirangelo I. Effect of trehalose on W7FW14F apomyoglobin and insulin fibrillization: new insight into inhibition activity. Biochemistry 2008;47:1789–96. Vilasi S, Sarcina R, Maritato R, De Simone A, Irace G, Sirangelo I. Heparin induces harmless fibril formation in amyloidogenic W7FW14F apomyoglobin and amyloid aggregation in wild-type protein in vitro. PLoS ONE 2011;6:e22076. Weise K, Radovan D, Gohlke A, Opitz N, Winter R. Interaction of hIAPP with model raft membranes and pancreatic beta-cells: cytotoxicity of hIAPP oligomers. ChemBioChem 2010;11:1280–90. Woods LA, Platt GW, Hellewell AL, Hewitt EW, Homans SW, Ashcroft AE, et al. Ligand binding to distinct states diverts aggregation of an amyloid-forming protein. Nature Chemical Biology 2011;7:730–9. Zhu JT, Choi RC, Chu GK, Cheung AW, Gao QT, Li J, et al. Flavonoids possess neuroprotective effects on cultured pheochromocytoma PC12 cells: a comparison of different flavonoids in activating estrogenic effect and in preventing beta-amyloid-induced cell death. Journal of Agricultural and Food Chemistry 2007;55:2438–45.
Contents lists available at SciVerse ScienceDirect
The International Journal of Biochemistry & Cell Biology journal homepage: www.elsevier.com/locate/biocel
Heme binding site in apomyoglobin may be effectively targeted with small molecules to control aggregation Mehrnaz Azami-Movahed a , Sajad Shariatizi a , Marjan Sabbaghian b , Atiyeh Ghasemi a , Azadeh Ebrahim-Habibi c,∗∗ , Mohsen Nemat-Gorgani a,d,∗ a
Institute of Biochemistry and Biophysics, University of Tehran, P.O. Box 13145-1384, 1417614411 Tehran, Iran Department of Andrology, Reproductive Biomedicine Research Center, Royan Institute, ACECR, Tehran, Iran c Endocrinology and Metabolism Research Center, Tehran University of Medical Sciences, 1411413137 Tehran, Iran d Stanford Genome Technology Center, Stanford University, Palo Alto, CA, USA b
a r t i c l e
i n f o
Article history: Received 3 June 2012 Received in revised form 8 October 2012 Accepted 11 October 2012 Available online 23 October 2012 Keywords: Apomyoglobin Amyloid Aggregation Chrysin Nile red
a b s t r a c t A number of ligands with affinities for the heme binding site of apomyoglobin were tested to control amorphous and fibrillar aggregation in the protein. Several techniques, including fluorescence, dynamic light scattering, transmission electron microscopy, dot blot analysis combined with viability studies were employed for structural characterization and cytotoxicity assessment of the intermediate and final protein structures formed during the aggregation process. Of the small molecules investigated, chrysin and Nile red with high structural similarities to heme were chosen for further studies. Only fibril formation was found to be prevented by Nile red, while chrysin, with a greater structural flexibility, was able to prevent both types of aggregate formation. The two ligands were found to influence aggregation at different stages of intermediate structure formation, an ability determined by their degrees of similarities with heme. Based on structural characterization and toxicity studies, it is concluded that ligands similar in structure to heme may be effective in influencing various stages of aggregate formation and toxicity potencies of the protein structures. Since metalloproteins constitute more than thirty percent of all known proteins, it is concluded that the present strategy may be of general significance. © 2012 Elsevier Ltd. All rights reserved.
1. Introduction Myoglobin is a globular, all-alpha metalloprotein, whose “apo” form (i.e. devoid of heme) is well-characterized relative to its folding dynamics (Jamin, 2005; Fink et al., 1998). Native, molten globule and unfolded states have been obtained for the apo form of horse myoglobin at pH 6.5, pH 4.2 and pH 2, respectively (Fink et al., 1998; Tcherkasskaya and Ptitsyn, 1999). Amyloid formation of myoglobin has been reported for the apo structure at alkaline pH, while the heme-containing form (holomyoglobin) could not be driven toward this state (Fandrich et al., 2003). On the other hand, a mutant form (W7FW14F) of sperm whale apomyoglobin showed fibrillation at physiologic pH and room temperature (Sirangelo et al., 2004, 2009; Infusini et al., 2012). In the present study, different features of
∗ Corresponding author at: Stanford Genome Technology Center, 855 S. California Avenue, Palo Alto, CA 94304, USA. Tel.: +1 650 812 1972. ∗∗ Corresponding author at: Endocrinology and Metabolism Research Center, Tehran University of Medical Sciences, Shariati Hospital, North Kargar Avenue, 1411413137 Tehran, Iran. Tel.: +98 21 88220038. E-mail addresses: [email protected], [email protected] (A. Ebrahim-Habibi), [email protected] (M. Nemat-Gorgani). 1357-2725/$ – see front matter © 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.biocel.2012.10.004
aggregation in wild-type horse apomyoglobin have been investigated, including determination of the roles of pH and electrostatic repulsion on aggregation pathways, which may result in amorphous aggregation or fibril formation. Small molecules were tested as inhibitors of both events and toxicities of intermediate structures were determined. It is suggested that stability of apomyoglobin could affect formation of “on pathway” or “off pathway” intermediates. One common event that is thought to be of importance in aggregate formation of apo-metalloproteins is destabilization of the native protein structure as a result of metal removal. For example, amyloid fibrils of bovine ␣-lactalbumin form either at low pH or by disulfide reduction. These conditions cause release of Ca2+ , leading to reduction of thermal stability and formation of partially folded conformations (Goers et al., 2002; Veprintsev et al., 1997; Ebrahim-Habibi et al., 2010a). Similarly, loss of metal ions, disulfide reduction, and pathologic mutations may drive superoxide dismutase toward amyloid fibrillation (Oztug Durer et al., 2009; Fee and Phillips, 1975; Mei et al., 1992). Fibril formation is also observed in apo-carbonic anhydrase, which may take up a pre-molten globular structure after removal of its zinc ion (Es-haghi et al., 2012). More specifically, conditions leading to loss of heme group of bovine cytochrome c (i.e. mutations in the heme binding site, or incubation
300
M. Azami-Movahed et al. / The International Journal of Biochemistry & Cell Biology 45 (2013) 299–307
at 75 ◦ C and under mild alkaline conditions) results in formation of unfolded structures and consequently, amyloid fibrils (Pertinhez et al., 2001; de Groot and Ventura, 2005). The structure stabilizing role of heme has been demonstrated for the heme-binding section of myoglobin (microglobin) (Ji et al., 2008), and further suggested to prevent apomyoglobin amyloid formation (Iannuzzi et al., 2007).
Ten microliters of well-mixed incubation sample were added to 190 l of the Congo red solution and incubated for 30 min. Absorbance spectra were recorded (400–600 nm) using a Shimadzu UV–visible spectrophotometer (Kyoto, Japan).
2. Materials and methods
2.7. Circular dichroism (CD) measurement
Horse myoglobin, Thioflavin T, Congo red, borate, 2-butanone, dialysis tubing, MTT, Nile red, and chrysin were purchased from Sigma (St Louis, MO, USA). ANS was obtained from Fluka. Glycine, DMSO and all salts and organic solvents were obtained from Merck (Darmstadt, Germany). RPMI medium were purchased from Gibco. FBS, horse serum, streptomycin and penicillin were obtained from Biosera (England). Ultra centrifuge filters were purchased from Millipore and A11 antibody was obtained from Chemicon.
CD spectra in the far-UV region (190–260 nm) were obtained on an AVIV 215 spectropolarimeter (Aviv Associates, Lakewood, NJ, USA), using a 1 mm path cell at room temperature. The protein concentration was 0.4 mg/ml.
2.1. Preparation of apomyoglobin (Teale, 1959) Apomyoglobin was prepared by phase extraction process. Holomyoglobin was dissolved in 0.01 M HCl and mixed with 2butanone on ice. Upon phase separation, the organic layer was removed and the procedure was repeated until clear solution appeared. Then, the sample was dialysed overnight in distilled water to remove extra butanone. Ultra centrifugation was employed to concentrate the apoprotein solutions. 2.2. Turbidity measurements Different buffers (phosphate, borate, sodium acetate), pH (4.2, 6.5, and 9), protein concentrations (0.2–0.4 mg/ml) and temperatures (60–70 ◦ C) were used when screening for the best conditions. Finally, solutions containing 0.3 mg/ml apomyoglobin in sodium acetate (10 mM), sodium phosphate (10 mM) and pH 6.5 and different ligands were prepared and incubated at 65 ◦ C. To monitor protein aggregation, turbidity measurements were made at 350 nm on a Shimadzu UV–visible spectrophotometer (Kyoto, Japan) (Rezaei-Ghaleh et al., 2007a). Temperatures were controlled to within ±0.1 ◦ C. 2.3. Thioflavin T (ThT) binding assay All fluorescence experiments were carried out on a CaryEclipse VARIAN fluorescence spectrophotometer. To investigate fibrillation, apomyoglobin samples were added to ThT solution in a molar ratio of 1:2.5, then mixed thoroughly and incubated for 5 min. Fluorescence excitation and emission were set at 440 and 482 nm and slit widths were set at 5 nm and 10 nm, respectively. 2.4. 8-Anilino-1-naphthalene sulfonate (ANS) fluorescence assays The excitation wavelength was 350 nm and emission spectra were recorded between 400 and 600 nm. Excitation and emission slit widths were set at 5 nm. The final concentrations of ANS and protein were 88 M and 0.03 mg/ml respectively.
2.6. Congo red absorbance assays
2.8. Transmission electron microscopy (TEM) Copper 400 mesh grid was covered with carbon-coated formvar film followed by adsorption of 10 l of apomyoglobin samples. After 2 min, excess fluid was removed with a paper filter, and 1% uranyl acetate added. After another minute, excess dye was removed. Finally, the grids were monitored with a CEM 902A Zeiss microscope (Oberkochen, Germany) and Philips (Japan). 2.9. Dynamic light scattering Dynamic light scattering (DLS) studies were performed on a zeta potential and particle size analyzer (Brookhaven Instrument, Holtsville, NY 11742-1896, USA). The size distribution/abundance of particles was studied in the absence and presence of desired ligands. The final protein concentration used was 58 M. A laser of 657 nm with a fixed detector angle of 90◦ was used. DLS studies were carried out at least in triplicates. 2.10. Dot blot analysis Dot immunoblot analysis was performed to investigate reactivity of anti-oligomeric A11 antibody against apomyoglobin oligomers. Aliquots of protein samples were spotted onto nitrocellulose membranes and dried. Membranes were blocked with 10% dry milk and incubated with A11 antibody (1:1000 dilution) overnight, then incubated with goat-anti rabbit HRP as the secondary antibody (1:2000 dilution) for 1 h. Dot visualization was carried out in a dark room. 2.11. Docking studies Docking was performed with Autodock vina (Trott and Olson, 2010). The 1DWR.pdb file (protein devoid of heme group) was used as receptor. Grid box of 48 × 48 × 44 points was used with ˚ and the grid box center was placed on x = 13.583, a spacing of 1.0 A, y = 21.646, and z = 8.308. The box encompassed the whole protein molecule. Ligands were prepared using Molecular Operating Environment (MOE), 2010.10 (Chemical Computing Group Inc., 1010 Sherbooke St. West, Suite #910, Montreal, QC, Canada, H3A 2R7). Gasteiger charges were assigned to protein and ligand molecules. Exhaustiveness was set on 20 and 100 poses were generated for the ligands. Preparation of the image representing the best pose of each ligand was performed with MOE 2010.10.
2.5. Nile red fluorescence assay 2.12. MTT assay Apomyoglobin (85 M) was added to Nile red (0.3 M) at pH 7 and the increase and shift occurring in the fluorescence emission was monitored. The excitation wavelength was 580 nm and emission spectra were recorded between 600 and 700 nm. Excitation and emission slit widths were both set at 5 nm.
Cell viability assay was assessed using rat pheochromocytoma (PC12) cell line. After 24 h of incubation with samples pretreated under amyloidogenic conditions in the absence and presence of ligands, cells were washed with PBS. 100 l of MTT stock solution
M. Azami-Movahed et al. / The International Journal of Biochemistry & Cell Biology 45 (2013) 299–307
(5 mg/ml in PBS) was added to 900 l of DMEM (without phenol red) containing 10% bovine calf serum followed by incubation for 3 h. Solutions were aspirated, and cells were treated with DMSO for 20 min. MTT absorbance was determined at 570 and 690 nm. All experiments were run in triplicates.
301
3.1. Apomyoglobin aggregation
et al., 2002; Lin et al., 2001; Nilsberth et al., 2001; Weise et al., 2010; Meratan and Nemat-Gorgani, 2012; Meratan et al., 2011). Exposure of hydrophobic sites was monitored using ANS fluorescence. Maximum ANS binding occurred at 20 h after incubation (corresponding to oligomeric species), followed by a decrease to reach a minimum after 48 h, when fibrils are formed, and a new fold is reached (results not shown). The ability of ANS in binding to the heme binding site of native apomyoglobin (Bismuto et al., 1996; Stryer, 1965), suggests an important role for this hydrophobic site in the amyloid structure formation process.
3.1.1. Fibrillation Apomyogloin fibrillation has been studied as an interesting example of a classical all-alpha protein, with the capacity to undergo transition to amyloid beta structure formation (Fandrich et al., 2001). In one of the reports describing amyloid formation in horse muscle apomyoglobin, role of unfolded segments was suggested to be more important than partially unfolded structures (Fandrich et al., 2001), and involvement of the Nterminal segment has been highlighted, especially the section corresponding to residues 1–29 (Picotti et al., 2007; Infusini et al., 2012). In the present study, after obtaining amyloid structures from horse muscle apomyoglobin, intermediate forms were characterized, and cytotoxicity of the various structures was assessed. In order to induce amyloid structures, initially various conditions were tested including use of a wide range of pH (2.5–9), and protein concentrations of 1–4 mg/ml at temperatures of 37 ◦ C and 65 ◦ C. Fibrillation was observed at pH 9 and 65 ◦ C, similar to conditions reported previously (Fandrich et al., 2003). An incubation time of 14 days was used in order to obtain a suitable population of mature fibrils. Although fibril formation could be observed after 24 h, amorphous species were also present at that time point. It should be mentioned that the incubated samples were not subjected to agitation, making the process slower but more reproducible. Accordingly, the selected conditions for further studies consisted of incubation of a 1 mg/ml protein solution made in 50 mM, pH 9 glycine buffer at 65 ◦ C. Amyloid formation under these conditions was tested using common procedures used for this type of studies, i.e. Congo red absorbance, Thioflavin T (ThT) fluorescence, and CD-monitored changes in secondary structure. CD spectra taken at different times are indicative of a gradual transition from alpha to beta structure, which could possibly pass from partially unfolded states of apomyoglobin that are known to contain a native-like core (Fink et al., 1998). Direct evidence of fibril presence was assessed by TEM. Kinetics of fibril formation was studied by following the changes occurring in ThT fluorescence intensity over the incubation period. The observed lag-phase, an exponential growth phase, and a stationary phase corresponded to the typical pattern of a nucleation-dependent behavior (Bhak et al., 2009), in accord with a random nucleation process previously observed for this protein (Fandrich et al., 2006) (results not shown). Cytotoxicity of apomyoglobin amyloid structures was assessed by the MTT assay. PC12 cells were exposed to various species, i.e. holo and apo monomers and oligomers (obtained after 20 h of incubation), structures formed after 4 and 7 days and mature fibrils (14 days). Toxic species were found to be oligomeric structures which possess reactivity to A11 antibody and the TEM image was also indicative of these structures. On the other hand, mature fibrils (and structures formed after 4 and 7 days) revealed no toxicity (results not shown). Experimental data indicate that toxicity of amyloidogenic proteins is caused by prefibrillar intermediates which are formed in the early stages of fibrillation. This toxicity has been associated with their exposed hydrophobic surfaces, which are able to interact with cell membranes (Bucciantini et al., 2002; Lashuel
3.1.2. Amorphous aggregation Self-assembly of polypeptide chains leading to formation of high molecular weight non-ordered aggregates is related to the protein environment conditions such as temperature, salts type and concentration, and protein concentration (Dumetz et al., 2008; Ebrahim-Habibi et al., 2010b). In order to investigate amorphous aggregation of horse muscle apomyoglobin, acidic and alkaline pH values (4.2, 6.5 and 9) close to the protein pI of 7.3 (Banks and Paquette, 1995), were used in the 60–70 ◦ C temperature range [(Tm is 61 ± 2 ◦ C (Fandrich et al., 2003)]. The process was assessed by the increase in the optical density of insoluble particles (Guagliardi et al., 1995; Paulikova et al., 1998; Li et al., 2001; Rafikova et al., 2003; Rezaei-Ghaleh et al., 2007b; Qiu and Macrae, 2008). Amorphous aggregation at considerable extent was found to occur at pH 6.5, where, interestingly, apomyoglobin has a native conformation (Tcherkasskaya and Ptitsyn, 1999). Indeed, formation of amorphous aggregates has been reported to be favored under conditions where proteins adopt a conformation close to their native state (Qin et al., 2007; Morshedi et al., 2010). It should be mentioned that at acidic pH (i.e. pH = 4.2), where the protein is in a partially folded state, we did not observe any significant amount of aggregation in the course of 10 min (Suppl. Fig. 1). In a previous report (Fandrich et al., 2003), in order to study and analyze the amorphous aggregate form of horse apomyoglobin, acidic pH of 4.1 was used to generate these forms. However, an incubation time period of more than 4 days (at 50 ◦ C) was found necessary to reach the “flake-like” aggregates, while shorter incubation times did not bring such results (Fandrich et al., 2003). As expected, the extent of aggregation was found to increase with increasing temperature (Fig. 1A), and protein concentration (Fig. 1B). Based on these results, a 0.3 mg/ml concentration of the protein was used at 65 ◦ C in subsequent experiments, in order to obtain an appreciable aggregation. Congo red and TEM were used to further investigate insoluble aggregate formation (Fig. 1C and D). The TEM image (Fig. 1D) is indicative of a non-fibrillar aggregate morphology (Nilsson, 2004), and clearly different from TEM images obtained from apomyoglobin fibrils (Fandrich et al., 2003). The Congo red spectrum of the amorphous species presents a red shift in comparison with the spectrum obtained from the dye alone, but lacks the increase in absorbance which is seen in the fibrillar species. This may be due to the presence of beta-sheet structures in the amorphous species which are believed to share these structures with amyloid species, though probably occurring to a lesser extent (Tyedmers et al., 2010); furthermore, amorphous and fibrillar species have been suggested to stem from a common partially folded intermediate containing beta-sheet structures (Jiang et al., 2012; Ortore et al., 2011). Concerning insoluble amorphous aggregates obtained from apomyoglobin (under other conditions), the presence of beta-sheet structures has been reported in these species, which are nonetheless devoid of the “well defined orientations” that are characteristics of amyloid fibrils (Fandrich et al., 2003). As summarized in the scheme below, pH seems to have an important role with regard to the type of aggregates formed at a high temperature.
3. Results and discussion
302
M. Azami-Movahed et al. / The International Journal of Biochemistry & Cell Biology 45 (2013) 299–307
Fig. 1. Monitoring absorbance at 350 nm over 720 s as an indicator of amorphous aggregation in apomyoglobin. (A) Use of apomyoglobin concentration of 0.3 mg/ml at pH 6.5 incubated at various temperatures: 60 ◦ C (), 62 ◦ C (), 65 ◦ C (), 67 ◦ C (♦), 70 ◦ C (). (B) Use of various concentrations of apomyoglobin at pH 6.5 and 65 ◦ C: used concentrations were 0.2 mg/ml (), 0.25 mg/ml (), 0.3 mg/ml (), 0.35 mg/ml (♦), 0.4 mg/ml (). (C) Congo red absorbance spectra of apomyoglobin aggregates formed after 10 min incubation at pH 6.5 and 65 ◦ C (- - -), Congo red dye (–) and apomyoglobin amyloid structure obtained after incubation of apomyoglobin at pH 9 and 65 ◦ C for 2 weeks (. . .). (D) TEM image of apomyoglobin aggregates formed after 10 min incubation at pH 6.5 and 65 ◦ C. (E) Apomyoglobin (0.3 mg/ml) at pH 6.5 and 65 ◦ C was incubated with various concentrations of NaCl, of 0 (), 60 mM (♦), 80 mM (), 100 mM (), 400 mM (䊉), and 750 mM ().
In addition to protein stability, protein solubility and net charge may play critical roles (Calloni et al., 2005). Mutations have been reported to change protein net charge and destabilize the native conformation. Consequently, misfolded intermediates appear which are prone to aggregation and result into pathogenic deposits (Chiti et al., 2002; Chiti and Dobson, 2006). ␣-Synuclein is an unstructured protein possessing an excess of acidic residues and is prone to aggregation when its strong negative charge decreases at a low pH (Uversky et al., 2001). In contrast, many
globular proteins form amyloid aggregates at acidic pH, where they have high positive charge (Guijarro et al., 1998; Kad et al., 2003; Frare et al., 2004). An electrostatic component is then mostly recognized as having a role in various types of aggregate formation. Wild-type apomyoglobin could rapidly form amorphous aggregates upon heating at a high temperature and a pH close to neutral where it is uncharged (Banks and Paquette, 1995), while amyloid structures are generated when its net negative charge is increased. Adding NaCl to the protein environment may result in decreasing amounts of amorphous aggregates, and the effect is concentration-dependent (Fig. 1E). In studies on other proteins, aimed at differentiating conditions leading to amorphous or amyloid structures, it has been suggested that amorphous forms could be subjected to less entropic penalties (Rezaei-Ghaleh et al., 2011), and are mainly induced by a prevalence of hydrophobic interactions (Vetri et al., 2007b). On the other hand, formation of amyloid structures, requiring extensive conformational changes in proteins (Vetri et al., 2007a) would be favored by conditions
M. Azami-Movahed et al. / The International Journal of Biochemistry & Cell Biology 45 (2013) 299–307
303
Fig. 2. Dynamic light scattering of apomyoglobin alone (A), and in the presence of Nile red (B). Aliquots were taken after 3 weeks of incubation. Figures show size distribution and average diameter of particles. (C) Apomyoglobin fibrillation in the presence of Nile red monitored by TEM. Amorphous aggregates are observed after 2 weeks of apomyoglobin incubation with Nile red (0.3 M). (D) Cell viability assays using apomyoglobin aggregates formed after 2 weeks in the presence of Nile red, compared with a control cell sample.
where inter-molecular hydrogen bonds occur preferentially over hydrogen bonds that form with the solvent (e.g. water), indicating that electrostatic and hydrophobic forces could be decisive factors in defining aggregate morphology (Vetri et al., 2011). 3.2. Effect of ligands on apomyoglobin aggregation Inhibition of pathologic aggregation by small molecules is now seriously considered as a promising therapeutic approach (Amijee and Scopes, 2009; Re et al., 2010). Both amyloid and amorphous aggregates have been successfully targeted by these compounds, which could also be used as probes to further elucidate the aggregation mechanism itself (Riviere et al., 2009; Sirangelo and Irace, 2010; Sood et al., 2011; Sabbaghian et al., 2011; Rezaei-Ghaleh et al., 2007a; Vilasi et al., 2008, 2011). In the present study, the effects of Nile red and chrysin have been investigated on apomyoglobin aggregation. Nile red is a fluorescent probe which has been used to study presence of hydrophobic surfaces in proteins (Sackett and Wolff, 1987). More specifically, it has been shown to be able to interact with the heme binding pocket of horse apomyoglobin (Polverini et al., 2006). On the other hand, presence of heme may stabilize horse myoglobin structure, increasing its Tm by about 20 ◦ C, and preventing fibrillation (Fandrich et al., 2003). When used as an additive to the amyloidogenic environment, it has been shown to prevent amyloid formation and the resulted cytotoxicity of the double mutant form of whale sperm apomyoglobin (Iannuzzi et al., 2007). Addition of micromolar concentrations of Nile red (0.3 M) to apomyoglobin (85 M) at neutral pH resulted in an increase in the fluorescence emission intensity and a blue shift from 665 nm to 628 nm, indicative of binding of the probe to a hydrophobic pocket in the protein molecule, similar to a previous report (Polverini et al., 2006). To induce fibril formation, pH was adjusted to 9, and the
protein was incubated at 65 ◦ C in the presence of Nile red. As monitored by dynamic light-scattering, after 3 weeks of incubation, control samples formed amyloid-like fibrils, showing diameters of 400 and 8200 nm (Fig. 2A), indicative of formation of intermediates and mature fibrils, respectively. In contrast, samples incubated with Nile red showed two populations of particles, 225 nm and 868 nm (Fig. 2B). TEM images taken after 2 weeks of incubation in the presence of the ligand showed presence of amorphous aggregates (Fig. 2C). Interestingly, these structures were found to be non-toxic by the MTT test (Fig. 2D). This result is similar to that reported using heme to prevent amyloid formation of the double mutant form of sperm whale apomyoglobin (Iannuzzi et al., 2007). The “remodeling” ability of some ligands that results into formation of non-toxic unstructured aggregates instead of the toxic species formed during the amyloidogenic process, has recently been reported for resveratrol (Ladiwala et al., 2010), biflavonoids (Thapa et al., 2011) and rifamycin (Woods et al., 2011). More generally, depending on their structures, aromatic small molecules are believed to act on the amyloidogenic process through different mechanisms, resulting in formation of various non-fibrillar structures (Ladiwala et al., 2011). In the present study, targeting the heme pocket seems to be an efficient strategy to prevent both fibrillation and cytotoxicity of the products generated from horse apomyoglobin. Interaction of Nile red with the heme binding pocket is suggested to occur via a binding mode that would partially overlap the heme position (Polverini et al., 2006). A blind docking experiment was performed with Nile red in apomyoglobin, and the best observed pose is shown in Fig. 3. It seems that this docking tool would give reasonable results, since the method was able to find the real binding site for heme (results not shown). Further to hydrophobic interactions that could occur between Nile red and the aliphatic residues of heme binding pocket, pi–pi interactions were also detected, occurring between His 97 and Phe 43 and one of Nile red aromatic components. Verification
304
M. Azami-Movahed et al. / The International Journal of Biochemistry & Cell Biology 45 (2013) 299–307
Fig. 5. Cell viability of PC-12 cells. Cells were exposed to apomyoglobin (25 M) aggregates formed after 2 weeks in the presence of chrysin (0, 50, 100 M), compared with a control sample.
Fig. 3. Best pose of docking results obtained for Nile red in horse apomyoglobin. The binding occurs in the hydrophobic heme-binding pocket. Pi–pi interactions could be seen, with participation of His 97 and Phe 43 and the aromatic components of Nile red.
of the first five docking poses revealed that at least one of these interactions would occur in all of these situations. As a polyphenolic compound, the flavonoid chrysin appeared to have the ability of influencing apomyoglobin aggregation. The effectiveness of flavonoids as anti-amyloid compounds has been reported in multiple studies, either on the process itself (Gasiorowski et al., 2011) or on its cytotoxicity (Lebeau et al., 2001; Zhu et al., 2007). Consequently, apomyoglobin was incubated with
50 and 100 M chrysin and samples were taken for ThT binding assay at various times. Our results indicated that the ligand, when used at 50 M concentration, was ineffective, while the presence of 100 M chrysin prolonged the lag phase without suppressing amyloid fibril formation. As shown in Fig. 4A, when chrysin (100 M) was present, after the lag time, fluorescence enhancement was observed, but reached only 50% of the maximal value recorded at the plateau phase in the absence of the ligand. Average diameter of control samples was found to be about 5000 nm as suggested by DLS runs, whereas the treated samples showed two populations of particles with average diameters of 170 and 1500 nm (results not shown). After 3 weeks, control fibrils extend to 8200 nm (Fig. 2A), but in samples incubated with chrysin, three populations of aggregates were observed with average diameters of 200, 1600, and 6800 nm (Fig. 4B). TEM images indicated formation of oligomeric structures in the presence of chrysin (Fig. 4C). Since chrysin was found to increase the lag phase, and based on the TEM results, apomyoglobin should have stayed in a cytotoxic oligomeric state in its presence. This was proven by incubating PC12 cells with
Fig. 4. (A) ThT fluorescence kinetics of apomyoglobin incubated at pH 9 and 65 ◦ C for 23 days in the absence () and presence () of chrysin (100 M). The lag phase is observed to increase (reaching 2 weeks) in the presence of chrysin. (B) Dynamic light scattering of apomyoglobin incubated at pH 9 and 65 ◦ C in the presence of chrysin. Aliquots of the protein sample were taken after 3 weeks of incubation. Figure shows size distribution and average diameter of particles. (C) Apomyoglobin fibrillation in the presence of chrysin monitored by TEM. The sample was incubated at pH 9 and 65 ◦ C with chrysin (100 M) for 2 weeks. Oligomeric intermediates are observed.
M. Azami-Movahed et al. / The International Journal of Biochemistry & Cell Biology 45 (2013) 299–307
305
Fig. 6. Surface hydrophobicity of apomyoglobin monitored by ANS fluorescence. Spectra were taken after 10 h (A), 2 days (B), and 3 days (C), from apomyoglobin in the absence of ligands (. . .), and in the presence of Nile red (—) or chrysin (- -). Control spectra of ANS alone are also shown (− ··).
the apomyoglobin structures formed in the presence of the effective concentration (100 M) of chrysin, which caused a decrease in reduced MTT levels (Fig. 5). Chrysin itself is devoid of toxicity: when the cells were incubated with different amounts of chrysin (25–100 M) or DMSO (0.1%), no toxicity was observed (results not shown). A comparison of the best docking pose of chrysin with the best pose of Nile red suggested the possibility for involvement of a common binding site (results not shown). However, pi–pi interactions are more favored with the fluorescence probe, in comparison with chrysin. Nile red has also a larger structure and one more aromatic component, which may help it to make more extensive interactions, while its more rigid structure would make it more similar to the natural ligand (heme). The fact that both ligands are binding to the heme binding site was further confirmed by an ANS fluorescence experiment, where samples of incubated protein with the ligands were taken at 10 h, 2 days and 3 days, and probed with ANS fluorescence (Fig. 6A–C, respectively). Results indicate a lower ANS fluorescence intensity when the ligands are present at day 2. The relatively low difference of intensity on 10 h and three days may be due to the competitive character of ligand and ANS binding for the same site. A possible mechanism of action of these two ligands on amyloid formation is presented in the below scheme:
Interestingly, chrysin is able to diminish amorphous aggregation of apomyoglobin. Upon incubation with 50–500 M chrysin at 65 ◦ C and pH 6.5, turbidity as indicative of aggregation was clearly diminished, in a dose-dependent manner (Fig. 7A). Moreover, DLS
measurements showed that the effective diameter of the aggregates decreased considerably in the presence of chrysin (Fig. 7B). A remarkable feature of this ligand is the effective inhibitory dosage (EC50% = 250 M): this value is lower than the usual concentrations used to inhibit amorphous aggregation (e.g. Sabbaghian et al., 2011). Amorphous aggregates formation is mostly related to increasing hydrophobic surfaces becoming exposed (Vetri et al., 2007b), thus chrysin is probably causing lowering of that exposure to some extent, but is not able to completely preserve the native fold. It could also be suggested that this small molecule is acting in a similar manner in both processes. At any rate, some stabilization of the structure occurs, and this experiment is suggestive of a possible use for aromatic ligands as preventive aids in deleterious amorphous aggregation. On the other hand, Nile red (used from 0.3 M to mM range) was not able to prevent amorphous aggregation of apomyoglobin. This is suggestive of the fact that Nile red is acting more similarly as the native ligand (heme), since holomyoglobin was also found to form amorphous aggregates when incubated at high temperatures (results not shown). Thus, depending on the process that is targeted (i.e. amorphous aggregates or amyloid formation), structures possessing different similarities with the native ligand could be effectively used.
306
M. Azami-Movahed et al. / The International Journal of Biochemistry & Cell Biology 45 (2013) 299–307
Fig. 7. (A) Amorphous aggregation of apomyoglobin (0.3 mg/ml) at pH 6.5 and 65 ◦ C incubated with various concentrations of chrysin: 0 M (), 50 M (♦), 100 M (), 200 M (), 300 M (䊉), 400 M () and 500 M (). (B) Effective diameter of amorphous aggregates alone () and in the presence of 250 M chrysin (), monitored by DLS (incubation was done at pH 6.5 and 65 ◦ C).
4. Conclusion In the present study, the inhibitory effects of two small molecules on amyloid fibrillation of wild type horse apomyoglobin are described. It seems that chrysin keeps apomyoglobin in a very toxic oligomeric state, while Nile red causes formation of small nontoxic aggregates, in a similar manner to heme. Results presented are indicative of involvement of the heme binding pocket in amorphous and amyloid aggregation of apomyoglobin. Since heme binding is related to stabilization of myoglobin, it could be suggested that stability of apomyoglobin structure may affect morphology and toxicity of intermediate structures. With metalloproteins representing over 30% of all proteins, the described strategy may provide a useful approach for control of protein aggregation and related cytotoxicity. Conflict of interest All authors declare not to have any conflict of interest. Acknowledgment This work was supported by grants from the research council of the University of Tehran and the Iranian National Science Foundation (INSF). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/ j.biocel.2012.10.004. References Amijee H, Scopes DI. The quest for small molecules as amyloid inhibiting therapies for Alzheimer’s disease. Journal of Alzheimer’s Disease 2009;17:33–47. Banks PR, Paquette DM. Monitoring of a conjugation reaction between fluorescein isothiocyanate and myoglobin by capillary zone electrophoresis. Journal of Chromatography A 1995;693:145–54. Bhak G, Choe YJ, Paik SR. Mechanism of amyloidogenesis: nucleation-dependent fibrillation versus double-concerted fibrillation. BMB Reports 2009;42:541–51. Bismuto E, Sirangelo I, Irace G, Gratton E. Pressure-induced perturbation of apomyoglobin structure: fluorescence studies on native and acidic compact forms. Biochemistry 1996;35:1173–8. Bucciantini M, Giannoni E, Chiti F, Baroni F, Formigli L, Zurdo J, et al. Inherent toxicity of aggregates implies a common mechanism for protein misfolding diseases. Nature 2002;416:507–11. Calloni G, Zoffoli S, Stefani M, Dobson CM, Chiti F. Investigating the effects of mutations on protein aggregation in the cell. Journal of Biological Chemistry 2005;280:10607–13. Chiti F, Calamai M, Taddei N, Stefani M, Ramponi G, Dobson CM. Studies of the aggregation of mutant proteins in vitro provide insights into the genetics of amyloid diseases. Proceedings of the National Academy of Sciences of the United States of America 2002;99(Suppl. 4):16419–26.
Chiti F, Dobson CM. Protein misfolding, functional amyloid, and human disease. Annual Review of Biochemistry 2006;75:333–66. de Groot NS, Ventura S. Amyloid fibril formation by bovine cytochrome c. SPECTROSCOPY – OTTAWA THEN AMSTERDAM 2005;19:199–205. Dumetz AC, Chockla AM, Kaler EW, Lenhoff AM. Effects of pH on protein–protein interactions and implications for protein phase behavior. Biochimica et Biophysica Acta 2008;1784:600–10. Ebrahim-Habibi MB, Amininasab M, Ebrahim-Habibi A, Sabbaghian M, NematGorgani M. Fibrillation of alpha-lactalbumin: effect of crocin and safranal, two natural small molecules from Crocus sativus. Biopolymers 2010a;93:854–65. Ebrahim-Habibi A, Morshedi D, Rezaei-Ghaleh N, Sabbaghian M, Nemat-Gorgani M. Protein–protein interactions leading to aggregation: perspectives on mechanism, significance and control. Journal of the Iranian Chemical Society 2010b;7:521–44. Es-haghi A, Shariatizi S, Ebrahim-Habibi A, Nemat-Gorgani M. Amyloid fibrillation in native and chemically-modified forms of carbonic anhydrase. II: role of surface hydrophobicity. Biochimica et Biophysica Acta (BBA) – Proteins & Proteomics 2012;1824:468–77. Fandrich M, Fletcher MA, Dobson CM. Amyloid fibrils from muscle myoglobin. Nature 2001;410:165–6. Fandrich M, Forge V, Buder K, Kittler M, Dobson CM, Diekmann S. Myoglobin forms amyloid fibrils by association of unfolded polypeptide segments. Proceedings of the National Academy of Sciences of the United States of America 2003;100:15463–8. Fandrich M, Zandomeneghi G, Krebs MR, Kittler M, Buder K, Rossner A, et al. Apomyoglobin reveals a random-nucleation mechanism in amyloid protofibril formation. Acta Histochemica 2006;108:215–9. Fee JA, Phillips WD. The behavior of holo- and apo-forms of bovine superoxide dismutase at low pH. Biochimica et Biophysica Acta 1975;412:26–38. Fink AL, Oberg KA, Seshadri S. Discrete intermediates versus molten globule models for protein folding: characterization of partially folded intermediates of apomyoglobin. Folding and Design 1998;3:19–25. Frare E, Polverino De Laureto P, Zurdo J, Dobson CM, Fontana A. A highly amyloidogenic region of hen lysozyme. Journal of Molecular Biology 2004;340:1153–65. Gasiorowski K, Lamer-Zarawska E, Leszek J, Parvathaneni K, Yendluri BB, BlachOlszewska Z, et al. Flavones from root of Scutellaria baicalensis Georgi: drugs of the future in neurodegeneration? CNS & Neurological Disorders – Drug Targets 2011;10:184–91. Goers J, Permyakov SE, Permyakov EA, Uversky VN, Fink AL. Conformational prerequisites for alpha-lactalbumin fibrillation. Biochemistry 2002;41:12546–51. Guagliardi A, Cerchia L, Rossi M. Prevention of in vitro protein thermal aggregation by the Sulfolobus solfataricus chaperonin. Evidence for nonequivalent binding surfaces on the chaperonin molecule. Journal of Biological Chemistry 1995;270:28126–32. Guijarro JI, Sunde M, Jones JA, Campbell ID, Dobson CM. Amyloid fibril formation by an SH3 domain. Proceedings of the National Academy of Sciences of the United States of America 1998;95:4224–8. Iannuzzi C, Vilasi S, Portaccio M, Irace G, Sirangelo I. Heme binding inhibits the fibrillization of amyloidogenic apomyoglobin and determines lack of aggregate cytotoxicity. Protein Science 2007;16:507–16. Infusini G, Iannuzzi C, Vilasi S, Birolo L, Pagnozzi D, Pucci P, et al. Resolution of the effects induced by W→F substitutions on the conformation and dynamics of the amyloid-forming apomyoglobin mutant W7FW14F. European Biophysics Journal 2012;41:615–27. Jamin M. The folding process of apomyoglobin. Protein and Peptide Letters 2005;12:229–34. Ji HF, Shen L, Grandori R, Muller N. The effect of heme on the conformational stability of micro-myoglobin. The FEBS Journal 2008;275:89–96. Jiang D, Rauda I, Han S, Chen S, Zhou F. Aggregation pathways of the amyloidbeta(1–42) peptide depend on its colloidal stability and ordered beta-sheet stacking. Langmuir 2012., http://dx.doi.org/10.1021/la3021436. Kad NM, Myers SL, Smith DP, Smith DA, Radford SE, Thomson NH. Hierarchical assembly of beta2-microglobulin amyloid in vitro revealed by atomic force microscopy. Journal of Molecular Biology 2003;330:785–97.
M. Azami-Movahed et al. / The International Journal of Biochemistry & Cell Biology 45 (2013) 299–307 Ladiwala AR, Lin JC, Bale SS, Marcelino-Cruz AM, Bhattacharya M, Dordick JS, et al. Resveratrol selectively remodels soluble oligomers and fibrils of amyloid Abeta into off-pathway conformers. Journal of Biological Chemistry 2010;285:24228–37. Ladiwala AR, Dordick JS, Tessier PM. Aromatic small molecules remodel toxic soluble oligomers of amyloid beta through three independent pathways. Journal of Biological Chemistry 2011;286:3209–18. Lashuel HA, Hartley D, Petre BM, Walz T, Lansbury Jr PT. Neurodegenerative disease: amyloid pores from pathogenic mutations. Nature 2002;418:291. Lebeau A, Esclaire F, Rostene W, Pelaprat D. Baicalein protects cortical neurons from beta-amyloid (25–35) induced toxicity. Neuroreport 2001;12:2199–202. Li J, Zhang S, Wang C. Only the reduced conformer of alpha-lactalbumin is inducible to aggregation by protein aggregates. Journal of Biochemistry 2001;129:821–6. Lin H, Bhatia R, Lal R. Amyloid beta protein forms ion channels: implications for Alzheimer’s disease pathophysiology. FASEB Journal 2001;15:2433–44. Mei G, Rosato N, Silva Jr N, Rusch R, Gratton E, Savini I, et al. Denaturation of human Cu/Zn superoxide dismutase by guanidine hydrochloride: a dynamic fluorescence study. Biochemistry 1992;31:7224–30. Meratan AA, Ghasemi A, Nemat-Gorgani M. Membrane integrity and amyloid cytotoxicity: a model study involving mitochondria and lysozyme fibrillation products. Journal of Molecular Biology 2011;409:826–38. Meratan AA, Nemat-Gorgani M. Mitochondrial membrane permeabilization upon interaction with lysozyme fibrillation products: role of mitochondrial heterogeneity. Biochimica et Biophysica Acta 2012;1818:2149–57. Morshedi D, Ebrahim-Habibi A, Moosavi-Movahedi AA, Nemat-Gorgani M. Chemical modification of lysine residues in lysozyme may dramatically influence its amyloid fibrillation. Biochimica et Biophysica Acta 2010;1804:714–22. Nilsberth C, Westlind-Danielsson A, Eckman CB, Condron MM, Axelman K, Forsell C, et al. The ‘Arctic’ APP mutation (E693G) causes Alzheimer’s disease by enhanced Abeta protofibril formation. Nature Neuroscience 2001;4:887–93. Nilsson MR. Techniques to study amyloid fibril formation in vitro. Methods 2004;34:151–60. Oztug Durer ZA, Cohlberg JA, Dinh P, Padua S, Ehrenclou K, Downes S, et al. Loss of metal ions, disulfide reduction and mutations related to familial ALS promote formation of amyloid-like aggregates from superoxide dismutase. PLoS ONE 2009;4:e5004. Ortore MG, Spinozzi F, Vilasi S, Sirangelo I, Irace G, Shukla A, et al. Time-resolved small-angle X-ray scattering study of the early stage of amyloid formation of an apomyoglobin mutant. Physical Review E: Statistical, Nonlinear, and Soft Matter Physics 2011;84:061904. Paulikova H, Molnarova M, Podhradsky D. The effect of heparin and pentosan polysulfate on the thermal stability of yeast alcohol dehydrogenase. Biochemistry and Molecular Biology International 1998;46:887–94. Pertinhez TA, Bouchard M, Tomlinson EJ, Wain R, Ferguson SJ, Dobson CM, et al. Amyloid fibril formation by a helical cytochrome. FEBS Letters 2001;495:184–6. Picotti P, De Franceschi G, Frare E, Spolaore B, Zambonin M, Chiti F, et al. Amyloid fibril formation and disaggregation of fragment 1–29 of apomyoglobin: insights into the effect of pH on protein fibrillogenesis. Journal of Molecular Biology 2007;367:1237–45. Polverini E, Cugini G, Annoni F, Abbruzzetti S, Viappiani C, Gensch T. Molten globule formation in apomyoglobin monitored by the fluorescent probe Nile Red. Biochemistry 2006;45:5111–21. Qin Z, Hu D, Zhu M, Fink AL. Structural characterization of the partially folded intermediates of an immunoglobulin light chain leading to amyloid fibrillation and amorphous aggregation. Biochemistry 2007;46:3521–31. Qiu Z, Macrae TH. ArHsp21, a developmentally regulated small heat-shock protein synthesized in diapausing embryos of Artemia franciscana. Biochemical Journal 2008;411:605–11. Rafikova ER, Kurganov BI, Arutyunyan AM, Kust SV, Drachev VA, Dobrov EN. A mechanism of macroscopic (amorphous) aggregation of the tobacco mosaic virus coat protein. International Journal of Biochemistry and Cell Biology 2003;35:1452–60. Re F, Airoldi C, Zona C, Masserini M, La Ferla B, Quattrocchi N, et al. Beta amyloid aggregation inhibitors: small molecules as candidate drugs for therapy of Alzheimer’s disease. Current Medicinal Chemistry 2010;17:2990–3006. Rezaei-Ghaleh N, Ebrahim-Habibi A, Moosavi-Movahedi AA, Nemat-Gorgani M. Effect of polyamines on the structure, thermal stability and 2,2,2trifluoroethanol-induced aggregation of alpha-chymotrypsin. International Journal of Biological Macromolecules 2007a;41:597–604. Rezaei-Ghaleh N, Ebrahim-Habibi A, Moosavi-Movahedi AA, Nemat-Gorgani M. Role of electrostatic interactions in 2,2,2-trifluoroethanol-induced structural changes and aggregation of alpha-chymotrypsin. Archives of Biochemistry and Biophysics 2007b;457:160–9.
307
Rezaei-Ghaleh N, Giller K, Becker S, Zweckstetter M. Effect of zinc binding on beta-amyloid structure and dynamics: implications for Abeta aggregation. Biophysical Journal 2011;101:1202–11. Riviere C, Delaunay JC, Immel F, Cullin C, Monti JP. The polyphenol piceid destabilizes preformed amyloid fibrils and oligomers in vitro: hypothesis on possible molecular mechanisms. Neurochemical Research 2009;34:1120–8. Sabbaghian M, Ebrahim-Habibi A, Hosseinkhani S, Ghasemi A, Nemat-Gorgani M. Prevention of thermal aggregation of an allosteric protein by small molecules: some mechanistic insights. International Journal of Biological Macromolecules 2011;49:806–13. Sackett DL, Wolff J. Nile red as a polarity-sensitive fluorescent probe of hydrophobic protein surfaces. Analytical Biochemistry 1987;167:228–34. Sirangelo I, Malmo C, Iannuzzi C, Mezzogiorno A, Bianco MR, Papa M, et al. Fibrillogenesis and cytotoxic activity of the amyloid-forming apomyoglobin mutant W7FW14F. Journal of Biological Chemistry 2004;279:13183–9. Sirangelo I, Iannuzzi C, Vilasi S, Irace G, Giuberti G, Misso G, et al. W7FW14F apomyoglobin amyloid aggregates-mediated apoptosis is due to oxidative stress and AKT inactivation caused by Ras and Rac. Journal of Cellular Physiology 2009;221:412–23. Sirangelo I, Irace G. Inhibition of aggregate formation as therapeutic target in protein misfolding diseases: effect of tetracycline and trehalose. Expert Opinion on Therapeutic Targets 2010;14:1311–21. Sood A, Abid M, Sauer C, Hailemichael S, Foster M, Torok B, et al. Disassembly of preformed amyloid beta fibrils by small organofluorine molecules. Bioorganic and Medicinal Chemistry Letters 2011;21:2044–7. Stryer L. The interaction of a naphthalene dye with apomyoglobin and apohemoglobin. A fluorescent probe of non-polar binding sites. Journal of Molecular Biology 1965;13:482–95. Tcherkasskaya O, Ptitsyn OB. Molten globule versus variety of intermediates: influence of anions on pH-denatured apomyoglobin. FEBS Letters 1999;455:325–31. Teale F. Cleavage of the haem-protein link by acid methylethylketone. Biochimica et Biophysica Acta 1959;35:543. Thapa A, Woo ER, Chi EY, Sharoar MG, Jin HG, Shin SY, et al. Biflavonoids are superior to monoflavonoids in inhibiting amyloid-beta toxicity and fibrillogenesis via accumulation of nontoxic oligomer-like structures. Biochemistry 2011;50:2445–55. Trott O, Olson AJ. AutoDock Vina: improving the speed and accuracy of docking with a new scoring function, efficient optimization, and multithreading. Journal of Computational Chemistry 2010;31:455–61. Tyedmers J, Mogk A, Bukau B. Cellular strategies for controlling protein aggregation. Nature Reviews Molecular Cell Biology 2010;11:777–88. Uversky VN, Li J, Fink AL. Evidence for a partially folded intermediate in alpha-synuclein fibril formation. Journal of Biological Chemistry 2001;276: 10737–44. Veprintsev DB, Permyakov SE, Permyakov EA, Rogov VV, Cawthern KM, Berliner LJ. Cooperative thermal transitions of bovine and human apo-alpha-lactalbumins: evidence for a new intermediate state. FEBS Letters 1997;412:625–8. Vetri V, Canale C, Relini A, Librizzi F, Militello V, Gliozzi A, et al. Amyloid fibrils formation and amorphous aggregation in concanavalin A. Biophysical Chemistry 2007a;125:184–90. Vetri V, Librizzi F, Leone M, Militello V. Thermal aggregation of bovine serum albumin at different pH: comparison with human serum albumin. European Biophysics Journal 2007b;36:717–25. Vetri V, D’Amico M, Fodera V, Leone M, Ponzoni A, Sberveglieri G, et al. Bovine serum albumin protofibril-like aggregates formation: solo but not simple mechanism. Archives of Biochemistry and Biophysics 2011;508:13–24. Vilasi S, Iannuzzi C, Portaccio M, Irace G, Sirangelo I. Effect of trehalose on W7FW14F apomyoglobin and insulin fibrillization: new insight into inhibition activity. Biochemistry 2008;47:1789–96. Vilasi S, Sarcina R, Maritato R, De Simone A, Irace G, Sirangelo I. Heparin induces harmless fibril formation in amyloidogenic W7FW14F apomyoglobin and amyloid aggregation in wild-type protein in vitro. PLoS ONE 2011;6:e22076. Weise K, Radovan D, Gohlke A, Opitz N, Winter R. Interaction of hIAPP with model raft membranes and pancreatic beta-cells: cytotoxicity of hIAPP oligomers. ChemBioChem 2010;11:1280–90. Woods LA, Platt GW, Hellewell AL, Hewitt EW, Homans SW, Ashcroft AE, et al. Ligand binding to distinct states diverts aggregation of an amyloid-forming protein. Nature Chemical Biology 2011;7:730–9. Zhu JT, Choi RC, Chu GK, Cheung AW, Gao QT, Li J, et al. Flavonoids possess neuroprotective effects on cultured pheochromocytoma PC12 cells: a comparison of different flavonoids in activating estrogenic effect and in preventing beta-amyloid-induced cell death. Journal of Agricultural and Food Chemistry 2007;55:2438–45.