Inhibition of Amyloid Formation by Ionic Liquids: Ionic Liquids Affecting

Aug 14, 2009 - Departments of Biochemistry and Chemistry, Tarbiat Modares University, Tehran, Iran. Biomacromolecules , 2009, 10 (9), pp 2468–2475...
0 downloads 0 Views 849KB Size
2468

Biomacromolecules 2009, 10, 2468–2475

Inhibition of Amyloid Formation by Ionic Liquids: Ionic Liquids Affecting Intermediate Oligomers Hamid Reza Kalhor,*,† Mostafa Kamizi,† Jafar Akbari,‡ and Akbar Heydari‡ Departments of Biochemistry and Chemistry, Tarbiat Modares University, Tehran, Iran Received April 15, 2009; Revised Manuscript Received July 24, 2009

In searching for alternative ways to reduce or inhibit amyloid formation, we have studied this process using hen egg white lysozyme (HEWL) in the presence of a low concentration of protic ionic liquids. The ionic liquids were synthesized in a combinatorial fashion maintaining the cationic part (tetramethylguanidinium) with alteration of the anionic component of each compound tested. It was observed that one of these compounds (tetramethylguanidinium acetate) inhibited amyloid formation of HEWL in vitro by nearly 50%. Examination under transmission electron microscopy confirmed the fibril inhibition, and fibrils were observed to be morphologically thinner. To investigate the mechanism of inhibition, intrinsic fluorescence, ANS binding, and circular dichroism analyses were performed. These analyses indicated that the native structure of HEWL was maintained in the presence of the ionic liquid. Performing native PAGE and nondenaturing agarose electrophoresis, it became evident that some of the intermediate oligomers were not converted to protofibrils and that the oligomers were trapped in more stable conformations. Additionally, it was observed that this inhibitory effect was related to the ionic liquid itself and not the solvated ions. It also became evident that the carboxyl functional group was important in the inhibition. The size of the anions and kosmotropicity did not play significant roles in the fibril inhibition.

Introduction The self-assembly of proteins into their native conformations has been one of the seminal topics in biochemistry for more than half a century.1-3 Proper native protein conformations play an essential role in normal cell functioning. If the correct folding of proteins does not occur and appropriate cellular repair machinery fail to function, protein aggregation and amyloid formation may occur, a process evident in several human diseases.3-5 The formation of amyloid has been identified in vitro to be associated with both disease and nondisease related proteins.6,7 In both scenarios, the fibril structures have demonstrated a strikingly similar structure and morphology including enriched β-rich structures, Congo red affinity, Thioflavin T (ThT) binding, and a fibril length of 10-50 nm.8,9 Hen egg white lysozyme (HEWL) has been used as a model protein to study fibril formation in vitro.8,10,11 HEWL can be converted to amyloid fibrils under environmental conditions of low pH and elevated temperatures.8,12 The fibrils formed from HEWL share similar characteristics to other amyloids and are toxic to cultured cells. Although no definitive intermediate in lysozyme fibril formation has been identified, an R-domain intermediate in the folding of HEWL has been recognized.13 In recent years the amyloidogenic region of human lysozyme has been mapped by proteolysis to two small regions, illustrating that amyloid formation does not require the entire lysozyme chain.14 Ionic liquids (ILs) have become important solvents that are used with biological molecules. The advantages of these compounds include their nonvolatility, good solvating properties, variable polarity range, and recyclability, which render these compounds environmentally “green”.15-17 Ionic liquids have been shown to stabilize and solubilize proteins, and to increase * To whom correspondence should be addressed. Tel.: +98-21-82883419. Fax: +98-21-82883463. E-mail: [email protected]. † Department of Biochemistry. ‡ Department of Chemistry.

enzymatic activities.15,18,19 We have been searching for synthetic and biological molecules that can inhibit amyloid formation in vitro. In this quest and in continuation of our interest in IL mediated organic reactions, we have explored tetramethylguanidinium based ionic liquids using a combinatorial fashion. In this study, the cation of the ionic liquids was preserved but the anionic counterpart was altered, generating a variety of functional groups. Utilizing these compounds, we examined their effects on lysozyme fibril formation in vitro. Our results indicated that a few protic ionic liquids could substantially affect HEWL fibril formation in vitro. The most effective of these compounds was ionic liquid 2 (IL-2), which suppressed fibril formation and altered the structural morphology of fibrils. In addition, IL-2 exerted inhibitory effects at relatively low concentrations (µM). The inhibitory effect was most likely manifested by the compound in ionic liquid forms because in nonionic liquid form the similar compound did not significantly inhibit fibrillation. We undertook structural studies to examine the impact IL-2 would have on fibril formation. These studies demonstrated that the ionic liquid compound stabilized native protein structure, resulting in inhibition of β-sheet enriched protofibrils that are required for fibril formation. When a number of ionic liquids that only differ in their anionic components were used, it became evident that the carboxyl functional group from the anionic part of these compounds played a crucial role in this inhibition. Conversely, the size of the anions and kosmotropicity did not play a significant role in the inhibition. By examining the intermediates formed during fibrillation using native PAGE and nondenaturing agarose electrophoresis, it was observed that intermediate oligomers were accumulated in the presence of the ionic liquid, well into the late stage of fibril formation.

Materials and Methods Materials. HEWL, ThT, ANS (8-anilino-1-naphthalene sulfonic acid), and agarose were purchased from Sigma. Glycine, Congo red,

10.1021/bm900428q CCC: $40.75  2009 American Chemical Society Published on Web 08/14/2009

Inhibition of Amyloid Formation by Ionic Liquids polyacrylamide, Coomassie Brilliant Blue, and bovine serum albumin (BSA) were purchased from Merck, Germany. Bisacrylamide was purchased from Scharlau, Spain, and β-alanine from Applichem, Germany. All materials used for the preparation of the ionic liquids were purchased from Merck, Germany. Formation of HEWL Fibril. To calculate the exact concentration of the protein, the UV absorbance of the lysozyme solution was measured at 280 nm (ε ) 37970 M-1 cm-1). HEWL solutions of 2 mg/mL (139 µM) were prepared by dissolving HEWL in a glycine buffer (0.1 M, pH ) 2.5). The protein samples were incubated at 57 °C for several days. For each time point, a specific amount was taken out for measurements. For preparation of preformed fibrils, 1 mM lysozyme in H2O/HCl, pH 2.0, was incubated for 4 days at 65 °C. For seeded reactions, the amount of preformed fibril was 5% of the total concentration of the lysozyme protein. Thioflavin T Binding Assay. Stock of ThT was prepared by dissolving ThT in sodium phosphate (NaH2PO4 50 mM, Na2HPO4 50 mM, pH 6) and the solution was filtered using a 0.2 µm filter. A total of 10 µL of the sample to be tested was mixed with ThT (total concentration was 13 µM). ThT fluorescence intensity measurements were subsequently performed by exciting wavelength at 440 nm and emission intensities were recorded at 485 nm using a Shimadzu RF5000 spectrofluorophotometer. Both excitation and emission slit widths were kept within 5 nm. ANS Fluorescence Assay. Stocks of ANS fluorescence assay were prepared by dissolving ANS in absolute ethanol and filtered using a 0.2 µm filter. A total of 10 µL of the samples was mixed with ANS (total concentration of samples being 50 µM). Excitation wavelength was regulated at 365 nm and emission spectra subsequently recorded from 430 to 550 nm. Excitation and emission slit widths were kept in 5 and 3 nm, respectively. Circular Dichroism (CD). CD spectra of HEWL samples were recorded from 190 to 260 nm (far-UV) using a Jasco J810 spectropolarimeter. Concentration samples were 0.1 mg/mL, and a 0.5 cm quartz cell was utilized. Samples were then either centrifuged at 13000 g for 10 min to remove insoluble fibrils or were used directly without centrifugation. Intrinsic Fluorescence. Diluted protein samples (10 µL at a concentration of 1.39 µM) were used for reading Trp intrinsic fluorescence. Excitation wavelength was regulated in 295 nm and emission wavelengths were recorded from 310 to 400 nm. Transmission Electronic Microscopy (TEM). Lysozyme samples (10 µL) were placed on Formvar coated grids stained with 1% (w/v) uranyl acetate. The grids were subsequently washed with deionized water, examined, and photographed at an accelerating voltage of 80 kV. Preparation of Ionic Liquids. All experiments were undertaken under dry nitrogen containing environments free of oxygen. All solvents were used as commercially obtained, and the ionic liquids were prepared using the previously reported methods without modification.20 Tetramethylguanidine ionic liquids were synthesized by neutralizing tetramethylguanidine with selected acids. As an example, IL-2 was prepared by adding stoichiometric amounts of 1,1,3,3 tetramethylguanidine and acetic acid. In a 250 mL flask, 50.0 mmol of tetramethylguanidine was added to 40.0 mL of dichloromethane. A 50.0 mmol solution of acetic acid in 30.0 mL of dichloromethane was progressively added to the mixture until the reaction was complete (within 2 h). The solvent was then removed by evaporation under reduced pressure. The residue was washed with hexane and evaporated in a vacuum until IL-2 (a white solid) was attained. NMR spectra were performed to verify the syntheses.

Results Combinatorial Approach To Search for Inhibitors of Amyloid Formation In Vitro Using Ionic Liquids. We wanted to take advantage of the ease of synthesizing ionic liquids in a

Biomacromolecules, Vol. 10, No. 9, 2009

2469

Figure 1. Fibrillation kinetics of HEWL. ThT binding assay of HEWL in the presence of 400 µM of IL-2 (b) and in the absence of IL-2 (O). HEWL (2 mg/mL) was incubated at pH 2.5 and 57 °C for several days. At specific time points, aliquots were taken out for measurements. Each time point is an average of three independent experiments.

combinatorial fashion and use them to regulate protein fibril formation in vitro. To achieve this goal a series of ionic liquids were synthesized wherein the cationic counterpart was composed of tetramethylguanidinium and the anionic counterpart was modified to generate various functional groups (Table S1).20,21 Each compound was then assayed separately for in vitro fibril formation. HEWL was converted to amyloid in vitro using elevated temperatures and a high concentration of hydrogen (pH 2.5) as is commonly performed.6,22 Ionic liquids in various concentrations were added to the in vitro reaction at the commencement of the assay. To monitor fibrillation, we used several methods including SDS-PAGE, Congo red binding assay, and turbidity (data not shown). However, the most sensitive assay to monitor amyloid formation turned out to be ThT binding assay (Figure 1). As shown in Figure 1, the kinetics of fibril formation commenced with a lag phase, followed by an exponential elongation period, until it finally plateaued. It became evident that IL-2 had a significant inhibitory effect on fibril formation. To shorten the time of fibrillation, the in vitro reaction was seeded with a minute amount of preformed lysozyme fibril as described in Materials and Methods. As shown in Figure 2a, the same inhibitory role of IL-2 on the fibril formation was observed. The effects of other ionic liquids on HEWL fibril formation were also tested, and it became evident that IL-2 was the best at impeding fibrillation (Figure 2b). Calculation of Polymerization Rate and Fibril Characterization in the Presence of Ionic Liquid. To obtain a numerical value for the effect that IL-2 has on the rate of polymerization, the lag time and polymerization rate were calculated based on the sigmoid-shaped kinetics, as described in the Materials and Methods (Supporting Information).23 We observed that the lag time for fibril formation in the presence of IL-2 and its absence is 94 and 87 h, respectively, and that the aggregation constant (k) in the presence of IL-2 K IL-2 ) 6.7 × 10-4 min-1, and in its absence K ) 9.06 × 10-4 min-1. Our calculation supports the notion that the inhibitory effect of IL-2 was mostly observed in the elongation period. To be certain that the ionic liquid compound affected amyloid formation, the fibrils formed in the presence and absence of IL-2 was visualized by TEM. As shown in Figure 3, the ionic liquid clearly demonstrated an inhibitory effect on fibril formation. In the absence of IL-2, the amyloid fibrils were more abundant, and their diameters ranged from 10-15 nm (Figure 3a). However, in the presence of IL-2, a greater number of immature fibrils of a smaller diameter formed (ranging from 5 to 6 nm, Figure 3b).

2470

Biomacromolecules, Vol. 10, No. 9, 2009

Kalhor et al.

We also examined ANS binding to lysozyme at different times during the progression toward amyloid formation. As it has been well established, ANS binds to the hydrophobic reagent of macromolecules.27 The level of ANS binding increased as the incubation time of lysozyme in the amyloidogenic condition increased, as one would have expected (Figure 4b). However, the ANS binding was much lower when the IL-2 was present in the reaction mixture. This indicated that in the presence of the ionic liquid, the unfolding of lysozyme was retarded. This effect was especially pronounced in the late stage of the fibril formation.

Figure 2. Fibrillation kinetics of seeded HEWL. ThT binding assay of HEWL (a) in the presence of 400 µM (b) and in the absence of IL-2 (O); (b) in the presence of 400 µM IL-2 (b), 400 µM IL-12 (9), 400 µM IL-7 (4), 400 µM IL-10 (O), and in the absence of any ionic liquid ([). The fibrillation reaction was performed exactly as in Figure 1, with the exception that samples were seeded (5% final) with preformed fibrils of lysozyme at the beginning of the assay.

To fully investigate the mechanism of fibril formation in the presence of the ionic liquid, TEM was performed on samples from the early stage of the elongation period (96 h, Figure 1). In the absence of IL-2, there existed some amorphous nanoparticles that may represent some intermediate oligomers or perhaps small protofibrils (Figure 3c). However, to our surprise, relatively few of these particles were observed when the ionic liquid was present in the assay (Figure 3d). This may suggest that, in the presence of IL-2, lesser amounts of intermediate oligomers were detected by TEM. Intrinsic Fluorescence, ANS Binding, and Circular Dichroism Studies in the Presence of Ionic Liquid. One approach to study the unfolding of a protein during fibril formation is to use its intrinsic Trp fluorescence.24 We measured the intrinsic florescence of lysozyme in the presence or absence of the ionic liquid at various time points leading to its aggregation. In the absence of ionic liquid in the late stage of lysozyme amyloid formation (132 h), there was a noticeable increase in the intensity of the emitted fluorescence, which was not observed when the ionic liquid had been present in the sample (Figure 4a). Indeed, HEWL has six Trp residues, five of which are near (angstrom distance away) sulfur atoms when the protein is in the folded state, and consequently, these residues would be quenched. However, when the protein unfolds, as in our assay, the Trp residues could emit more fluorescence as they move away from the sulfur atoms.25 There was also a red shift observed when no ionic liquid was present. This shift is most likely indicative of the denaturation state of the protein as previously shown with other proteins during aggregation.26 However, it must be noted that Trp fluorescence could be sensitive to a number of factors including quenching by ionic liquid itself.

To monitor the global-secondary structural changes during fibrillation, we performed CD analyses (Figure 5). We were first concerned that fibril formation might interfere with the CD spectra. To clear this point, we compared the CD spectra of a sample that was centrifuged at high speed to remove fibrils to the same sample without the extra step. We subsequently found no difference in the CD spectra between the two samples (data not shown). Therefore, we decided to use samples directly for CD measurements. To see if IL-2 had any effect on the native form of lysozyme, we measured far-UV CD spectra on lysozyme in the native form in the presence and absence of IL-2. The results obtained from CD 190-260 nm revealed no effect on HEWL (native form) in the presence of IL-2 (data not shown). However, the far-UV CD spectra at different time points, when HELW was subjected to amyloidogenic conditions (high temperature and low pH), showed clearly different ellipticity when its structure was compared to proteins in the presence of IL-2 (Figure 5a,b). There was a clear shift to the β-sheet enriched structure (216-218 nm) as the time of incubation in the denaturing condition increased (Figure 5a). However, in the presence of IL-2, there was a decreased shift toward the β-sheet dominated structure (Figure 5b). As shown in Figure 5c, the difference in the ellipticity of HEWL at late stages of the fibril formation clearly depicted the differences in the secondary structural feature discussed. To better observe the secondary structural changes that were taking place, we plotted the absolute value of ellipticity at 218 nm for two comparing samples with increased incubation time. It became evident that, in the absence of IL-2, there was a significant increase in the absolute value of ellipticity at 218 nm as the time of the incubation increased (Figure 6). However, in the presence of IL-2, there was not a significant increase in the value of ellipticity as the fibrillation of HEWL continued. The positive correlation between the ellipticity at 215-218 nm (β-sheet rich structure) and incubation time has been shown to be indicative of oligomer polymerization in the fibrillation pathway of proteins, corresponding to the elongation phase of the ThT binding assay.28 Mechanism of Ionic Liquid Inhibition. To further elucidate the mechanisms of ionic liquid inhibition on amyloid fibril formation, we compared the cationic and anionic components of IL-2 separately (as tetramethylguanidine and acetic acid), a mixture of these components together, or as an ionic liquid compound as in previous experiments (IL-2). As shown in Figure 7, none of the compounds inhibited fibrillation, with the exception of IL-2. This indicated that fibril inhibition only occurred when the compound was in the “ionic liquid” form. In other words, neither the cationic nor the anionic counterparts alone or when combined (as two ions) could inhibit fibrillation. This illustrated to us that the inhibitory role of IL-2 was likely related to the ionic liquid features of this compound. Interestingly enough, performing the same experiment, as demonstrated in Figure 7, we used sodium acetate (400 µM and 1 mM), and

Inhibition of Amyloid Formation by Ionic Liquids

Biomacromolecules, Vol. 10, No. 9, 2009

2471

Figure 3. TEM images of HEWL fibril formation (a, c) in the absence of IL-2 and (b, d) in the presence of 400 µM of IL-2. The reaction conditions were exactly as in Figure 1. For (a) and (b), the samples were taken out from the late stage of fibril formation (244 h), and for (c) and (d), the samples were taken out at the early stage of the elongation phase (96 h). The bar below each panel represents 100 nm.

we did not detect any fibril inhibition at either concentration (data not shown). To more clearly study the mechanism of IL-2 inhibitory effects on fibril formation, we performed native PAGE and nondenaturing agarose electrophoresis. We examined protein fibrillation at various time points in both the presence and the absence of the ionic liquid (Figures S1 and S2). In the early stage of amyloid formation there existed both monomers and intermediate oligomers in the sample containing IL-2, as well as in the control (Figure S1, lanes 2-5). However, in the late stage (240 h) there existed fewer intermediate oligomers and monomers than when IL-2 was present in the sample reaction (Figure S1, lanes 6 and 7). The same phenomenon was observed when nondenaturing agarose gel electrophoresis was performed; that is, the intermediate oligomers were further shown to be largely absent in the late stage of fibril formation when IL-2 was not present in the sample reaction (Figure S2). These results suggest a possible mechanism for fibril inhibition by IL-2; the ionic liquid suppresses conversion of oligomeric intermediates into protofibrils, which could ultimately lead to decreased fibril formation. To investigate further the effect of the ionic liquid on fibril formation, a range of different concentrations of IL-2 was used

in the fibril formation assay. The optimal concentration that inhibited fibril formation was 400 µM (Figure 8). To our surprise at very low and high concentrations, the ionic liquid had no effect on fibril formation. In the high concentration range (above 50 mM) there was inhibition of fibrillation, but it was discovered that at this concentration IL-2 increased the alkalinity of the solution, which would automatically affect the fibril formation (data not shown). To see if IL-2 also inhibits fibril formation in other proteins, we tested fibrillation in BSA. We chose BSA because as opposed to HEWL it is an acidic protein (pI 4.5) and can form fibrils at pH 7.0.29 IL-2 was able to retard the fibrillation of BSA, but the inhibition appeared to be less effective when compared to the HEWL fibril’s inhibition at similar concentrations (Figure S3).

Discussion In this work we investigated the inhibitory effects of ionic liquids on HEWL amyloid formation. The lysozyme protein was chosen because it is frequently used as a model protein in many studies.11,12,30 HEWL is composed of mainly R-helical and two short β-strands; the structure of the lysozyme contains four

2472

Biomacromolecules, Vol. 10, No. 9, 2009

Kalhor et al.

Figure 4. Intrinsic fluorescence spectra and ANS binding assay. (a) Trp intrinsic fluorescence spectra of HEWL fibril at different time points; native HEWL (9), HEWL at 24 (O), 96 (0), and 132 h (s), and HEWL in the presence of 400 µM of IL-2 at 24 ( · · · ), 96 (]), and 132 h (4). (b) ANS binding assay of HEWL at different time points; native HEWL (4) and HEWL at 24 (]), 96 (+), and 175 h (b), and HEWL in the presence of 400 µM IL-2 at 24 (×), 96 (9), and 175 h (O), with the spectrum of ANS alone (0). The conditions for the assay were exactly as in Figure 2.

disulfide bonds.31 There is a very high homology in both sequence and structure between HEWL and human lysozyme, and each protein has been shown to produce fibrils.8,31 There have been a number of reports that have used HEWL to look for inhibitors of fibril formation in vitro.32-35 In this work, we took the advantage of the combinatorial properties of ionic liquids to investigate the mechanism of fibril inhibition. Ionic liquids were used because of the recent rise in their applications in biochemical research.15,18,36 Ionic liquids can also be synthesized with relative ease and in a combinatorial manner, which enables them to be used in investigating the mechanisms behind biological processes.20,15,37 In our studies, we used a series of ionic liquids (Table S1) that were composed of an identical cationic component but differed according to their anionic counterparts. These compounds were used mainly in low concentrations (µM) to observe their effects on amyloid formation of lysozyme in vitro. Our results clearly indicated that only specific types of the ionic liquids had an inhibitory effect on the process of amyloid formation in vitro. One of the best inhibitors, tetramethylguanidinium acetate (IL-2), showed a significant (around 50%) inhibition of amyloid formation. When the ThT binding assay was used, it was demonstrated that the inhibitory effect occurred predominantly in the elongation phase of fibril formation, with little or no effect on the lag (initiation) phase of fibril formation (Figure 1). When TEM was used, we observed a reduction in fibril formation when IL-2 was used in the reaction and fibrils also demonstrated a thinner morphology (Figure 3b). To investigate the structural changes that HEWL undergoes in the presence of IL-2, we conducted intrinsic fluorescence,

Figure 5. Far-UV CD spectra HEWL fibrillation in the presence of IL-2. (a) HEWL at 24 (0), 48 (4), 96 (]), 132 (O), and 0 h (9). (b) HEWL in the presence of 400 µM of IL-2 at 24 (4), 48 (×), 96 (O), 132 (0), and 0 h (]). (c) HEWL in the presence of 400 µM (0) and in the absence (O) of IL-2 at 132 h, and HEWL at 0 h (s). Fibrillation assay was performed as in Figure 2.

Figure 6. CD spectra at 218 nm. CD ellipticity of absolute value at 218 nm for HEWL (O) and in the presence of 400 µM of IL-2 (2).

ANS binding, and far-UV CD analyses. The intrinsic fluorescence and ANS binding studies suggested that in the presence of the ionic liquid HEWL underwent far less unfolding. The CD analyses confirmed that there was less β sheet content in the presence of ionic liquid (Figure 5c). Additionally, there was a clear correlation between the increase ellipticity and the

Inhibition of Amyloid Formation by Ionic Liquids

Figure 7. Fibrillation kinetics of HEWL in the presence of 400 µM of tetramethylguanidine (0), 400 µM of acetic acid (b), equal mixture of tetramethylguanidine (200 µM) and acetic acid (200 µM; 4), 400 µM of IL-2 (9), and HEWL alone (]). The reaction conditions for fibrillation were exactly as in Figure 2. Each time point is an average of three independent experiments.

Figure 8. Fibrillation kinetics of HEWL in various concentrations of IL-2. The reaction condition for fibrillation was as exactly as in Figure 2.

increased incubation time at 218 nm when ion-containing liquid was not present. However, this correlation was not as evident when ionic liquid was present (Figure 6). It has previously been suggested that the increase ellipticity at 218 with time corresponds to the polymerization of oligomers into protofibrils in the process of fibril formation.28,38 We were curious to know if the inhibitory effect was due to the ionic liquid itself or as a consequence of the solvated ions. To test this, we used each component independently, in a mixture together (acetic acid/tetramethylguanidine), or as an ionic liquid (IL-2). To our surprise, we observed that inhibition of fibrillation only occurred when the ionic liquid compound was used (Figure 7). This result demonstrates that the ionic liquid compound exerts its effects when both the cationic and anionic components are held together as an ionic liquid compound. This would be a novel application of ionic liquids functioning as solutes. In many of the applications of ionic liquids in chemistry and biochemistry, ionic liquids are either used as solvents, cosolvents, or within biphasic systems. To better investigate the intermediate oligomers that were formed in the presence of IL-2, we performed native electrophoresis at various stages of fibril formation (Figure S1 and S2). Our results indicated that in the presence of IL-2, there was an accumulation of late stage oligomers with a reduced conversion to mature fibrils. It may be reasoned that the more folded monomers that were detected in the presence of ionic liquid might be due to accumulation of the intermediate oligomers in the late stage of fibril formation which that would shift the equilibrium in favor of more native monomers. Nevertheless, one cannot completely rule out a direct binding of IL-2 to the partially unfolded monomer.

Biomacromolecules, Vol. 10, No. 9, 2009

2473

The inhibitory effect of IL-2 was only observed at low concentrations of ionic liquid (maximum effect at 400 µM). At higher concentrations a decreased effect was identified (Figure 8). A reduced effect at higher concentrations might be consequent to micelle formation that can occur with ionic liquids in aqueous solutions.39 Examining BSA fibrillation, IL-2 successfully inhibited fibril formation of at pH 7 although the effect was not as significant compared to that observed in HEWL. This indicates the mechanism through which IL-2 may exert its effects could be working with other proteins such as BSA, and perhaps it did not require a restricted electrostatic requirement. However, the inhibition is much more pronounced in acidic environments as many proteins have demonstrated fibril formation when the electrostatic feature of proteins or the environment is altered.40-42 In examining the results presented in Table S1, the question still remains as to what the underlying mechanisms of ionic liquid mediated fibril inhibition are. However, ionic liquids have some interesting characteristics that might shed some light into their behavior. One of the dominating characteristics of ionic liquids is their polarity.15,43 This polarity may influence the interaction of the ionic liquid with the solvent and the protein as well. Indeed, there have been reports that ionic liquids in aqueous solutions (very high mole fraction of water) are not dissociated completely into ions and these molecules form clusters in aqueous solutions.44 Solvents of high cohesive energy like water create a “solvophobic” phenomenon, and ILs have shown not to dissociate ion pairs but rather water as solvents may disturb the ionic liquid cohesion.45-47 This behavior of ionic liquids in water may explain why when ionic liquids containing both cationic and anionic parts were present, there was an inhibition of fibril formation (Figure 7). Solvated ions (those that are completely dissociated into ions) did not demonstrate the inhibition. A common feature that can be derived from these behaviors of ILs in water may be the importance of the hydrophobic moiety in the cationic part of ILs. Interestingly, the structure of guanidinium, in aqueous solution, revealed that the cation was not well hydrated, and it was labeled as one of the least hydrated cations.48 Guanidinium is similar to the cationic part of our ionic liquids; in addition, the cationic group of the ILs used in our experiments is even more hydrophobic with four methyl groups and could most likely form hydrophobic subgroups in water. It can be deduced from the data provided (Table S1) that there might be a specific chemical functional group that is responsible for the inhibitory effect of these compounds. This functional group could most likely be the carboxylic acid moiety that is present in almost all of the compounds that have shown some effect on the retardation of amyloid formation. The carboxylic moiety could easily participate in hydrogen bonding interaction with the protein’s backbone and perhaps stabilize the protein or its intermediate oligomers. Indeed, the carboxylic acid moiety has been identified as an essential structural feature in several tau inhibitors.49,50 However, IL-9 (tetramethylguanidinium lactate), which has not demonstrated any significant effect on fibrillation, also carries a carboxyl moiety. This alludes to the fact that ionic liquids could display complex properties, which are not readily explained. For example, in this particular type of ionic liquid (which bears a hydroxyl group next to its carboxylic moiety), the hydroxyl group has demonstrated unique properties that may change the physicochemical properties of ionic liquids leading to a higher density due to potential hydrogen bonding of the hydroxyl group.15

2474

Biomacromolecules, Vol. 10, No. 9, 2009

Kalhor et al.

Figure 9. Schematic diagram of a proposed mechanism of fibril inhibition by IL-2. The protein fibrillation pathway begins with the destabilization and partial unfolding of the native protein (in elevated temperature and low pH). Partial unfolded proteins are converted to the intermediates oligomers and subsequently are turned into protofibrils and amyloid at the end. When the protein is exposed to IL-2, the oligomers are formed, but some of the oligomers are specifically trapped by IL, leading to fewer protofibrils and amyloid at the end.

The possibility that the ionic liquids used in this work may act as Hoffmiester series was examined by arranging some of the anions used in our work based on their size as formate (IL1), acetate (IL-2), trifluoroacetate (IL-3), trifluoromethanesulfonate (IL-4), tetrafluoroborate (IL-5), nitrate (IL-6), chloride (IL-7), and perchlorate (IL-8). The size of these anionic ions increases from formate to perchlorate. Additionally, the anions are arranged in order of increasing kosmotropicity H2PO4- > CH3CO2- > HSO4- > HCO2- > Cl- > NO3- > ClO4-. Nevertheless, we do not see a trend of fibril inhibition as either the size of the anionic part increases or the kosmotropicity. This may be due to the fact that because these compounds were used in low concentrations, the Hoffmiester effect would not play a significant role.51 A proposed mechanism of ionic liquid mediated fibril inhibition is shown in Figure 9. It could be deduced from our results that the ionic liquid could directly bind to the partially unfolded protein. The binding of the IL-2 to proteins can assist in stabilizing the protein, as its functional groups are more exposed to the environment and it can inhibit fibril formation by interfering with protofibril and subsequently fibril formation. One type of noncovalent binding through which the IL may be affecting the protein stability would be through the formation of cation-π interactions.52 Lysozyme has a high percentage of aromatic amino acids with six Trp residues. As the biomolecule undergoes partial unfolding, the aromatic residues are more exposed and, therefore, cation-π interaction between the cation part of the IL and those exposed aromatic residues could take place ultimately resulting in protein stabilization. Additionally, if these aromatic residues are involved in oligomer-oligomer interactions, presence of cation-π interactions could also slow down the rate of fibril formation.53 The reason why IL-2 only partially inhibits fibril formation may be related to the fact that IL-2 is able to recognize specific conformations of intermediate oligomers.54 Indeed, as native PAGE illustrated, particular oligomers seemed to be stabilized in the presence of IL-2 (labeled as an arrow in Figure S1). In recent years there have been significant efforts made to inhibit fibril formation in disease-related amyloidosis, as it has

been identified that oligomers are even more toxic to cells.55,56 It has been suggested that efforts to stop amyloidosis must be directed toward those intermediates present during the process of protein fibrillation. Consequently, it may be useful to employ various functionalized ionic liquids with their interesting (but incompletely understood) features to inhibit fibrillation, particularly during the early stages involving intermediate oligomers. As we were preparing to submit our work, a short report came to print that demonstrated the use of ionic liquids in dissolving and solvating amyloid fibril.57 However, in this paper the ionic liquids were used as solvents and not as solutes, and the type of ionic liquids utilized were entirely different from the ones we applied in our study. One of our ongoing projects is to examine the impact of the various functional groups of the cationic part of ionic liquids in amyloid formation.

Conclusion In this paper, we demonstrated that certain tetramethylguanidinium-based protic ionic liquids were able to inhibit protein fibril formation. The effect of the most potent of these ionic liquids was examined in detail using various techniques. When TEM was used, it was observed that IL-2 affected both the morphology and the amount of fibril formed. Native gel analyses indicated that some of the oligomeric intermediates were not converted to subsequent steps of polymerization in the presence of the ionic liquid. A more folded conformation of the exposed protein was detected in the presence of the ionic liquid during fibril formation, and there was a significant reduction in the β-sheet content of exposed proteins in the presence of IL-2 as justified by CD analyses. The inhibitory effect was best observed in µM range, and the effect was consequent to the presence of the ionic liquid and not the solvated ions. When the effects of various ionic liquids that differed in their ionic functional groups with maintenance of similar cationic components were compared, it was deduced that the presence of the carboxyl functional group played a significant role in fibril inhibition. The size of the anion, however, did not play a significant role

Inhibition of Amyloid Formation by Ionic Liquids

in this inhibitory process. The strategies and compounds identified could ultimately be examined for their effects on other amyloidogenic proteins that cause a wide range of pathological diseases in humans. Acknowledgment. We would like to thank Mr. Beiranvand for TEM technical assistance and Dr. S. Z. Bathaie for permission to use her CD instrument. We are also grateful to Dr. S. Azizian Kohan for critical reading of the manuscript and his valuable suggestions. This work was supported by Tarbiat Modares University. Supporting Information Available. Table S1, calculation of polymerization rate, native PAGE, nondenaturing agarose electrophoresis, and inhibition of BSA fibrillation. This material is available free of charge via the Internet at http://pubs.acs.org.

References and Notes (1) Thornton, J. M.; Orengo, C. A.; Todd, A. E.; Pearl, F. M. J. Mol. Biol. 1999, 2, 333–342. (2) Ohnishi, S.; Takano, K. Cell. Mol. Life. Sci. 2004, 61, 511–524. (3) Dobson, C. M. Semin. Cell. DeV. Biol. 2004, 1, 3–16. (4) Wetzel, R. Cell 1996, 65, 699–702. (5) Kelly, J. W. Curr. Opin. Struct. Biol. 1998, 8, 101–6. (6) Chiti, F.; Webster, P.; Taddei, N.; Clark, A.; Stefani, M.; Ramponi, G.; Dobson, C. M. Proc. Natl. Acad. Sci. U.S.A. 1999, 7, 3590–3594. (7) Lansbury, P. T., Jr. Proc. Natl. Acad. Sci. U.S.A. 1999, 7, 3342–3344. (8) Morozova-Roche, L. A.; Zurdo, J.; Spencer, A.; Noppe, W.; Receveur, V.; Archer, D. B.; Joniau, M.; Dobson, C. M. J. Struct. Biol. 2000, 2-3, 339–351. (9) Makin, O. S.; Serpell, L. C. FEBS J. 2005, 23, 5950–5961. (10) Krebs, M. R.; Wilkins, D. K.; Chung, E. W.; Pitkeathly, M. C.; Chamberlain, A. K.; Zurdo, J.; Robinson, C. V.; Dobson, C. M. J. Mol. Biol. 2000, 3, 541–549. (11) Vernaglia, B. A.; Huang, J.; Clark, E. D. Biomacromoclecules 2004, 5, 1362–70. (12) Cao, A.; Hu, D.; Lai, L. Protein Sci. 2004, 2, 319–324. (13) Matagne, A.; Chung, E. W.; Ball, L. J.; Radford, S. E.; Robinson, C. V.; Dobson, C. M. J. Mol. Biol. 1998, 5, 997–1005. (14) Frare, E.; Mossuto, M. F.; Polverino de Laureto, P.; Dumoulin, M.; Dobson, C. M.; Fontana, A. J. Mol. Biol. 2006, 3, 551–561. (15) Greaves, T. L.; Drummond, C. J. Chem. ReV. 2008, 1, 206–37. (16) Rogers, R. D.; Seddon, K. R. Science 2008, 302, 792–793. (17) Wasserscheid, P.; Welton T. Ionic Liquids in Synthesis, 2nd ed.; WileyVCH: New York, 2007. (18) Fujita, K.; MacFarlane, D. R.; Forsyth, M. Chem. Commun. (Cambridge, U.K.) 2005, 38, 4804–4806. (19) Fujita, K.; MacFarlane, D. R.; Forsyth, M.; Yoshizawa-Fujita, M.; Murata, K.; Nakamura, N.; Ohno, H. Biomacromoclecules 2007, 8, 2080. (20) Wasserscheid, P.; Driessen-Hoelscher, B.; Hal, R.; Steffens, H.; Zimmerman, J. Chem Commun. 2003, 16, 2038–2039. (21) Zhu, A.; Jiang, T.; Wang, D.; Han, B.; Liu, L.; Huang, J.; Zhang, J.; Sun, D. Green Chem. 2005, 7, 514–517. (22) Nilsson, M. R. Methods 2004, 1, 151–160. (23) Nielsen, L.; Khurana, R.; Coats, A.; Frokjaer, S.; Brange, J.; Vyas, S.; Uversky, V. N.; Fink, A. L. Biochemistry 2001, 20, 6036–6046. (24) Dusa, A.; Kaylor, J.; Edridge, S.; Bodner, N.; Hong, D. P.; Fink, A. L. Biochemistry 2006, 8, 2752–2760. (25) Geddes, C. ReViews in fluorescence; Springer: New York, 2006. (26) Marchal, S.; Shehi, E.; Harricane, M. C.; Fusi, P.; Heitz, F.; Tortora, P.; Lange, R. J. Biol. Chem. 2003, 34, 31554–31563.

Biomacromolecules, Vol. 10, No. 9, 2009

2475

(27) Hawe, A.; Sutter, M.; Jiskoot, W. Pharm. Res. 2008, 7, 1487–99. (28) Bartolini, M.; Bertucci, C.; Bolognesi, M.; Cavalli, A.; Melchiorre, C.; Andrisano, V. ChemBioChem 2007, 8, 2152–2161. (29) Holm, N. K.; Jespersen, S. K.; Thomassen, L. V.; Wolff, T. Y.; Sehgal, P.; Thomsen, L. A.; Christiansen, G.; Andersen, C. B.; Knudsen, A. D.; Otzen, D. E. Biochem. Biophys. Acta 2007, 9, 1128–1138. (30) Sasahara, K.; Nitta, K. Proteins 2006, 1, 127–135. (31) De Felice, F. G.; Vieira, M. N.; Meirelles, M. N.; Morozova-Roche, L. A.; Dobson, C. M.; Ferreira, S. T. FASEB J. 2004, 10, 1099–1101. (32) Morshedi, D.; Rezaei-Ghaleh, N.; Ebrahim-Habibi, A.; Ahmadian, S.; Nemat-Gorgani, M. FEBS J. 2007, 24, 6415–6425. (33) Steven, S. S.; Po-Han, C.; Ying-Tz, H. J. Mol. Catal. B: Enzym. 2006, 43, 49–57. (34) Lieu, V. H.; Wu, J. W.; Wang, S. S.; Wu, C. H. Biotechnol. Prog. 2007, 3, 698–706. (35) Gazova, Z.; Bellova, A.; Daxnerova, Z.; Imrich, J.; Kristian, P.; Tomascikova, J.; Bagelova, J.; Fedunova, D.; Antalik, M. Eur. Biophys. J. 2008, 7, 1261–1270. (36) De Diego, T.; Lozano, P.; gmouh, S.; Vaultier, M.; Iborra, J. L. Biomacromolecules 2005, 6, 1457–1464. (37) Hardacre, C.; Holbrey, J. D.; Nieuwenhuyzen, M.; Youngs, T. G. Acc. Chem. Res. 2007, 11, 1146–55. (38) Lin, M. S.; Chen, L. Y.; Tsai, H. T.; Wang, S. S.; Chang, Y.; Higuchi, A.; Chen, W. Y. Langmuir 2008, 11, 5802–5808. (39) Blesic, M.; Marques, M.; Plechkova, N.; Seddon, K.; Rebelo, L.; Lopes, L. Green Chem. 2007, 9, 481–490. (40) Picotti, P.; De Franceschi, G.; Frare, E.; Spolaore, B.; Zambonin, M.; Chiti, F.; de Laureto, P. P.; Fontana, A. J. Mol. Biol. 2007, 5, 1237– 1245. (41) Marcon, G.; Plakoutsi, G.; Canale, C.; Relini, A.; Taddei, N.; Dobson, C. M.; Ramponi, G.; Chiti, F. J. Mol. Biol. 2005, 2, 323–335. (42) Fink, A. L. Acc. Chem. Res. 2006, 9, 628–634. (43) Greaves, T. L.; Weerawardena, A.; Krodkiewska, I.; Drummond, C. J. J. Phys. Chem. B 2008, 3, 896–905. (44) Zhang, L.; Xu, Z.; Wang, Y.; Li, H. J. Phys. Chem. B 2008, 112, 6411–6419. (45) Weingartner, H. Angew. Chem., Int. Ed. 2008, 47, 654–670. (46) Cammarata, L.; Kazarian, S. G.; Salterb., P. A.; Welton, T. Phys. Chem. Chem. Phys. 2001, 3, 5192–5200. (47) Spickermann, C.; Thar, J.; Lehmann, S. B. C.; Zahn, S.; Hunger, J.; Buchner, R.; Hunt, P. A.; Welton, T.; Kirchner, B. J. Chem. Phys. 2008, 129, 104505–104517. (48) Mason, P. E.; Neilson, G. W.; Dempsey, C. E.; Barnes, A. C.; Cruickshenk, M. Proc. Natl. Acad. Sci. U.S.A. 2003, 100, 4557–4561. (49) Bulic, B.; Pickhardt, M.; Schmidt, B.; Mandelkow, E. M.; Waldmann, H.; Mandelkow, E. Angew. Chem., Int. Ed. 2009, 10, 1740–1752. (50) Bulic, B.; Pickhardt, M.; Khlistunova, I.; Biernat, J.; Mandelkow, E. M.; Mandelkow, E.; Waldmann, H. Angew. Chem., Int. Ed. 2007, 48, 9215–9219. (51) Broeing, J. M.; Bommarius, A. J. Phys. Chem. 2005, 109, 20612– 20619. (52) Tsou, l.; Tatko, C. D.; Waters, M. L. J. Am. Chem. Soc. 2002, 124, 14917–14921. (53) Johnson, R. M.; Hecht, K.; Deber, C. M. Biochemistry 2007, 32, 9208– 9214. (54) Moler, A.; Fabian, H.; Sokolowski, H.; Lutsch, G.; Gast, K.; Damaschun, G. Amyloid 2004, 11, 215–231. (55) Kayed, R.; Head, E.; Thompson, J. L.; McIntire, T. M.; Milton, S. C.; Cotman, C. W.; Glabe, C. G. Science 2003, 300, 486–9. (56) Ferreira, S. T.; Vieira, M. N.; De Felice, F. G. IUBMB Life 2007, 4-5, 332–45. (57) Byrne, N.; Angel, C. A. Chem. Commun. 2009, 7, 1046–8.

BM900428Q