| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Laboratory of Physical Chemistry and Colloid Science, Wageningen University, Wageningen, The Netherlands
Correspondence: Address reprint requests to Renko de Vries, Laboratory of Physical Chemistry and Colloid Science, Wageningen University, PO Box 8038, Wageningen 6700 EK, The Netherlands. Tel.: 31-317-484561; E-mail: renko.devries{at}wur.nl.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
48 h. Nucleation presumably involves a change in the conformation of individual lysozyme molecules. Indeed, long-term circular dichroism measurements at pH 2.0, T = 57°C show a marked change of the secondary structure of lysozyme molecules after 48 h of heating. From atomic force microscopy we find that most of the fibrils have a thickness of 4 nm. These fibrils have a coiled structure with a periodicity of 30 nm and show characteristic defects after every four or five turns. |
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
). For most of these proteins conditions have also been found under which they form fibrillar aggregates in vitro (Chamberlain et al., 2000; Jimenez et al., 2001, 2002; Zurdo et al., 2001).
A very different example of fibrillar protein aggregates is the formation of transparent heat-set gels by various food proteins (Langton and Hermansson, 1992; Kavanagh et al., 2000; Koike et al., 1996). Such gels often form under conditions of strong electrostatic repulsion (pH far away form the protein isoelectric point and low ionic strength). Electron microscopy (Kavanagh et al., 2000; Aymard et al., 1999) and atomic force microscopy (Ikeda and Morris, 2002) have revealed that these transparent gels also consist of fibrillar protein aggregates. Heat-set fibrillar gels have been reported for ovalbumin (Nemoto et al., 1993), ß-lactoglobulin (Kavanagh et al., 2000), and bovine serum albumin (Tani et al., 1995), to name but a few food proteins.
In a previous work (Arnaudov et al., 2003), mainly motivated by food applications, we investigated fibril formation of ß-lactoglobulin (ß-lg) at a temperature of 80°C, pH 2, and low ionic strength. On-line proton NMR spectroscopy and static and dynamic light-scattering were used to characterize the aggregation process. We found that fibrils start forming immediately after the heating has started, without any detectable lag time. Fibril formation is a multistep process: freshly formed fibrils still disintegrate upon slow cooling, but aged fibrils do not. Below an initial protein concentration of 2.5 wt % fibril formation is no longer the dominant process. Instead, at low concentrations, most protein molecules go to an unfolded or partially folded state from which fibrils can no longer be formed.
It is not exactly clear to what extent fibril formation is mechanistically similar for different proteins under similar circumstances. Although some authors have called the fibrils such as those formed by ß-lactoglobulin "amyloid fibrils" (Gosal et al., 2002; Sagis et al., 2004), no detailed comparison has been made, yet. Therefore, we here continue our studies of heat-induced fibril formation by studying fibrillar aggregation of hen egg white lysozyme (HEWL) under conditions (temperature of 57°C, pH 2–4, and low ionic strength) similar to those used in our previous study of ß-lactoglobulin.
HEWL has been recently found to form amyloid fibrils in vitro (Krebs et al., 2000; Goda et al., 2000; Cao et al., 2004). It is also one of the best characterized and most studied of all proteins. Its folding-unfolding has been studied in detail (Redfield and Dobson, 1988; Radford et al., 1992; Itzhaki et al., 1994; Dobson et al., 1994) and unfolding intermediates have been found. It is also an important food protein.
Furthermore, HEWL is homologous to human lysozyme, which is one of the proteins that cause an amyloid disease upon mutation (Booth et al., 1997; Pepys et al., 1993) and for which aggregation is also observed in vitro (Morozova-Roche et al., 2000). The formation of amyloid aggregates in vitro by wild-type and point-mutated human lysozyme has been studied by Morozova-Roche et al. (2000). Amyloid fibril formation was observed upon incubation of the wild-type and the variant human lysozymes at conditions in which partially folded intermediates are highly populated. Seeding experiments proved that the fibril formation was greatly influenced by the presence of the seeds. Amyloid aggregates have also been found to form from equine lysozyme (Malisauskas et al., 2003). Short fibrils and rings were observed by atomic force microscopy (AFM), which was used as a basic monitoring technique.
The only study of the formation of amyloid fibrils by wild-type HEWL induced by heat treatment alone is that of Krebs et al. (2000). Fibrils were formed from aqueous solutions at pH 2.0 by incubation at 37°C after a brief heat shock at 100°C followed by freezing in liquid nitrogen, or by simple incubation at 65°C. Fibrils were also formed at pH 7.4 in solution containing 30% 2,2,2-trifluorethanol by incubation either at 37°C or at 65°C. Other studies of HEWL amyloid fibril formation invariably involve denaturing agents such as ethanol (Goda et al., 2000; Yonezawa et al., 2002; Tanaka et al., 2001) or even full reduction of the protein (Cao et al., 2004). No study has yet been done on fibril formation by HEWL using the simplest procedure of heating an aqueous solution at a constant temperature. Also there are practically no data about the kinetics of the fibril formation from HEWL. For human lysozyme, Morozova-Roche et al. (2000) studied the fibrillar aggregation induced by heating at 57°C at pH 2. Following these authors, we here study the fibrillar aggregation of HEWL induced by heating at 57°C at pH 2. We also explore the effects of increasing the pH (up to 4) and temperature (up to 80°C). The kinetics of aggregation is studied by on-line static light scattering (SLS). The intensity of the scattered light is particularly sensitive to the size of the scattering objects, which makes the method very useful in detecting the onset of the aggregation. The type of the aggregates and their morphology are studied in detail by AFM. The effect of the temperature and the heating time on the secondary structure of HEWL at different pH values is studied by circular dichroism (CD). Prolonged heating at pH 2 may cause protein hydrolysis, which may or may not influence the course of the aggregation process. Therefore we also perform tris-tricine sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) on the heat-treated protein solutions.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
= 280 nm, using an extinction coefficient of 2.65 L g–1 cm–1.
Methods
Static light scattering
SLS data were obtained at a scattering angle of 90° using an ALV/SLS/DLS-5000 light-scattering apparatus (ALV, Langen, Germany), equipped with an argon ion laser (LEXEL, Palo Alto, CA) operating at a wavelength of 514.5 nm. Two types of experiments were carried out. In the first one the sample holder of the instrument was preheated to the desired temperature (57°C, 60°C, 65°C, or 80°C), the sample was introduced in the sample holder, and the intensity of the scattered light was recorded continuously for as long as 2 days. In the second type of experiment samples of each protein concentration were introduced in a thermostatic bath preheated at the desired temperature, the sample holder of the light scattering apparatus was preheated to the same temperature, and at regular time intervals the samples were quickly replaced from the bath to the light-scattering equipment and the intensity of the scattered light was measured. The intensity of the scattered light was calibrated by the intensity of the scattered light from pure toluene measured before each series of experiments at 25°C. Before heating, the presence of only monomeric protein in the samples was established by dynamic light scattering (scattering angle 90°, temperature 25°C) for each sample.
Atomic force microscopy
Tapping mode AFM was carried out using a Nanoscope III, multimode scanning force microscope (Digital Instruments, Santa Barbara, CA). Observations were performed on dry samples in air prepared as follows. Clean silicon plates were used as substrates. The plates were cleaned first in pure ethanol by ultrasound, then rinsed with pure ethanol and dried with pure and dry nitrogen. The dry silicon plates were subsequently subjected to plasma cleaning. After the plasma cleaning the silicon plates were dipped into test protein solutions for 1 h, taken out and dried immediately using pure and dry nitrogen. The test protein solutions were prepared by taking aliquots at regular time intervals from HEWL solutions kept in tightly closed glass tubes in a preheated bath at the desired temperature, quenching the aliquots in ice-cold water and diluting them to a final protein concentration of 0.1 wt %.
Circular dichroism
Circular dichroism spectra were recorded by using a Jasco J-715 spectropolarimeter (Jasco, Tokyo, Japan). Far-ultraviolet (UV) (185–260 nm) spectra were recorded in 0.01 cm path length cell (Hellma, Muellheim/Baden, Germany) using a step size of 0.5 nm, bandwidth of 1 nm, and scan rate of 50 nm/min. Each protein spectrum was obtained by averaging 10 scans and corrected by subtracting the solvent spectrum. The temperature was varied between 25°C and 90°C by means of a computer-controlled Peltier device (PTC-348WI, Jasco). The temperature scan was carried out on fresh protein solutions with concentration of 0.2 wt %. At every temperature the sample was allowed to equilibrate for 5 min before the wavelength scan series was started. Spectra were also recorded at 25°C from quenched samples of already heated protein with concentrations of 0.1 wt %. The quenched samples were obtained by the procedure described in the previous section.
SDS-PAGE
Gel electrophoresis was performed under nonreducing conditions using commercial SDS/polyacrylamide gels (16.5% Tris-Tricine precast gels, BioRad, Hercules, CA). Reference polypeptide mixture was purchased from BioRad. Coomassie brilliant blue R250 (Sigma, St. Louis, MO) was used for staining.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
|
Atomic force microscopy
AFM is used both to confirm the presence of fibrillar aggregates in the HEWL samples at pH between 2.0 and 4.0, heated for different times at temperatures between 57°C and 80°C, and to study the fibril morphology in as much detail as possible.
Fig. 3 shows tapping mode AFM height images of fibrils formed from HEWL solutions at pH 2.0 heated at 57°C, 60°C, and 80°C. Fibrils are formed in all cases, though at 80°C a considerable amount of spherical aggregates are formed, too. The fibrils formed are very long, thin, and straight.
|
|
10%) of thin fibrils can be observed in Fig. 5 a with an average thickness of 2.5 ± 0.3 nm. Most fibrils are thick and have an average diameter of 4.0 ± 0.7 nm.
|
30 nm and the difference between the peaks is also 30 nm one may suspect the presence of a length scale in the fibrils smaller than the subunit length and equal to 30 nm.
To validate the latter assumption we carried out a set of experiments with super sharp tapping mode AFM tips with a curvature at the tip of 2 nm (Nanosensors, Wetzlar-Blankenfeld, Germany). The standard tapping mode AFM tips have a curvature at the tip of 10 nm, which makes the observation of lateral details of the order of 20 nm practically impossible because of the sample-tip convolution effects. Fig. 6 shows four consecutive tapping mode AFM height images of a fibril formed upon heating of HEWL, pH 2.0, at 57°C. Indeed one can identify a coiled substructure of the fibril with a periodicity of 30 nm (Fig. 6, c and d). The substructure observed by us in this preliminary experiment will be studied in more detail in the near future.
|
|
Fig. 8 a shows CD spectra of HEWL at pH 2.0 measured at temperatures from 25 to 85°C at 5°C intervals. The spectra from 25°C to 45°C practically coincide with the native one. A major change in the spectrum occurs between 50°C and 60°C as the ellipticity between 208 and 230 nm increases and the minimum at 208 nm moves to a lower wavelength with increasing temperature, as the final value reached at 85°C is 202 nm. Above 60°C all spectra practically coincide. After the T ramp the sample was cooled back to 25°C. The spectrum of that sample did not fully coincide with the initial spectrum measured at 25°C (data not shown), i.e., the unfolding transition induced by heating is not fully reversible.
|
In Fig. 8 c spectra of HEWL at pH 4.0 measured at temperatures from 25°C to 90°C at 5°C intervals are plotted. In this case the spectra resemble the native one up to 70°C. The change in the spectrum occurs between 70°C and 85°C. Above 85°C the spectra coincide.
For all pH values studied, an isodichroic point is observed at 204 nm, suggesting that unfolding is a two-state cooperative process (Fändrich et al., 2003). To determine midpoint temperatures for the unfolding transition, we use the ellipticity at 222 nm (Yang et al., 1994; Sasahara et al., 2000). A plot of the molecular residue ellipticity at 222 nm of HEWL solutions at pH 2.0, 3.0, and 4.0 as a function of temperature is given in Fig. 9. The data are fitted by sigmoid curves. Midpoint temperatures of the transition obtained in this way are 54.8 ± 0.2°C for pH 2.0, 72.7 ± 0.7°C for pH 3.0, and 78.0 ± 0.5°C for pH 4.0. Transition temperatures decrease very rapidly at low pH, indicating a much less stable protein molecule at pH 2.0.
|
-helix in the secondary structure of the protein. Also, a jump in the ellipticity is observed between 44 and 50 h of heating. This is shown in Fig. 10 b, where the molecular residue ellipticity at 222 nm during the continuous heating of the HEWL sample is plotted. From a sigmoid fit of the experimental data we found that the jump occurred at a heating time of 48 h (see Fig. 10 b). This time coincides with the lag time after which the aggregation is detected by light scattering. One possible scenario is a process in which the partly destabilized lysozyme, which is still not capable of forming fibrils, undergoes a structural change after which it is susceptible to amyloid aggregation.
|
202 nm, which is typical for a random-coil protein. The spectrum of the fibrils (OLA, only large aggregates) has a similar minimum around 200 nm, but a small minimum at 218 nm can be observed that may be due to the presence of ß-sheet.
|
|
Gradual fragmentation is indeed observed, but only at pH 2.0 and for very long times of heating at 57°C (lanes 4, 5, 7, and 8). The protein that is not heated (lanes 1–3) is intact as only one bright line can be seen for each of the three samples. Although the line of the monomeric protein is thick and bright even for the sample heated for 11 days (lane 8), one can see one or more lines corresponding to lower-molecular weight species appearing below the native one. The process is slow: for the sample heated for 24 h (lane 4) a line corresponding to 10-kDa fragments is barely visible. The brightness of this line increases with heating time (lanes 5, 7, and 8). One can also see additional lines for the samples heated for 7 and 11 days, corresponding to fragments with molecular weights <6,500.
Samples heated for 11 days at 57°C, pH 2.0, were again separated into monomers and fibrils using filtration. Lanes 10 and 11 correspond to low-molecular weight species and long aggregates, respectively, after being separated from a sample of HEWL at pH 2.0 heated for 11 days. The lines in lane 11 practically coincide with the lines in lane 8—an identical sample not subjected to separation of the low-molecular weight species and the fibrils. The lines in lane 10 are similar to those in lanes 8 and 11 but less pronounced due to the smaller amount of protein applied to the gel in that lane. Clearly the fibrils do not exclusively consist of either full-length protein or protein fragments.
Finally, for pH 3.0 quenched after 38 days of heating at 57°C (lane 9), no hydrolysis is observed in SDS-PAGE, which indicates that hydrolysis is not a prerequisite for fibril formation.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
; Malisauskas et al., 2003; Krebs et al., 2000). Obviously the protein molecule has to be at least partially destabilized to form fibrils (Dobson, 2001). Another common observation is that, at the same time, the protein molecule must be at least partially structured to form fibrils. In the work of Krebs et al. (2000) this is achieved by first heating the proteins at 100°C, freezing them in liquid nitrogen, followed by incubation at 37°C. We observe similar behavior using the much simpler procedure of incubation at a fixed temperature that was also used by Morozova-Roche et al. (2000) in their study of fibril formation of both native and mutated human lysozyme.
At pH 2.0, HEWL forms fibrils after incubation at temperatures between 57 and 80°C. At 57°C, close to the midpoint temperature of the unfolding transition, fibrils are formed after only 2 days of incubation. At 80°C, we observe fibrils but also a lot of spherical aggregates. This is probably due to the fact that at this temperature, the protein molecule is already almost completely unfolded. Although the spherical aggregates are not the focus of this study, we should mention that sometimes proteins that form amyloid aggregates in vitro also form granular aggregates at specific conditions. For the case of the SH3 domain from bovine phosphatidyl-inositol-3'-kinase (PI-SH3) and the N-terminal domain of the Escherichia coli HypF protein (HypF-N) Bucciantini et al. (2002) have shown that not only fibrillar aggregates but also granular aggregates can be cytotoxic.
At pH 3.0 the protein is much more stable, and fibril formation only occurs after heating at 57°C for 11 days. Finally, at pH 4.0 we could not observe fibrils even after 42 days of incubation at 57°C because the protein is practically native at these conditions. Incubation at 80°C, pH 4.0, produced only small spherical aggregates. Spherical or granular aggregates are in some cases considered as precursors or nuclei for a fibril formation as in the case of amyloid ß-protein studied by Lomakin et al. (1997). This is obviously not the case for HEWL aggregation in the pH range studied by us. At pH 2.0 we observe spherical aggregates together with fibrils mostly in samples heated at 80°C (Fig. 3 a). In samples heated at 57°C and at pH 2.0 we rarely observe spherical aggregates. Upon heating of HEWL solutions at pH 3.0 at 57°C we do not observe any spherical aggregates. If spherical aggregates were precursors or nuclei for fibril formation one would expect to observe them before fibril formation or together with the fibrils. As one can see in Fig. 3, a–d, the spherical aggregates observed by us are larger in diameter than the fibrils so they cannot be a precursor to the fibril formation.
Unfolding properties of the more unstable lysozyme at pH 2.0 seem to be unique in promoting the fibrillar aggregation of HEWL: simply heating to a temperature close to the midpoint of the unfolding transition, as we did for pH 4.0, is not enough. It is not just the distances to the midpoint of the unfolding transition that determines whether fibrils are formed or not. Evidently, HEWL unfolding pathways are different at pH 2.0 and pH 4.0, and fibril formation sensitively depends on these pathways. Unfolding pathways will also be different for unfolding induced by other means, such as by adding denaturants. The mechanism of fibril formation may therefore also be different for those cases. On the other hand, since there is such a wide range of solution conditions that give rise to fibril formation, for such a wide range of proteins, it is not unreasonable to expect that there are at least some mechanistic similarities, and that ultimately studies on model systems such as ours may contribute to a better understanding of fibril formation, e.g., in mutated human lysozyme.
In our SLS experiments, we observe that there is always a very distinct lag time before fibril formation starts. We have found that even when heating at 80°C, fibrils are observed by AFM only after 2 days, whereas spherical aggregates start forming immediately, as detected by light scattering and confirmed by AFM. If we assume that the lag time is due to some nucleation event, the concentration dependence of the lag time should reflect the number of protein monomers involved in nucleation. Although we are somewhat hampered by the fact that there is such a large spread in the lag time for individual samples, the averaged data suggest that there is in fact no significant dependence of lag time on concentration. This would imply that the nucleation event is in fact a monomolecular event, e.g., a slow transition of individual lysozyme molecules to a partially folded state that initiates fibril formation.
Since the lysozyme stability plays a crucial role in the fibril formation, we studied it with respect to temperature at different pH values, as well as the effect of the heating time on the protein structure at pH 2.0 and at 57°C. The stability of HEWL has been extensively studied (Redfield and Dobson, 1988; Radford et al., 1992; Itzhaki et al., 1994; Dobson et al., 1994; Yang et al., 1994; Hoshino et al., 1997; Kamatari et al., 1998; Yonezawa et al., 2002). Most of the studies, however, are concerned with the effect of alcohols on the structure and stability of the HEWL molecule, and not with the effect of temperature on the protein structure in water solutions. According to Kamatari et al. (1998), methanol induces an expanded helical conformation in HEWL. Similar effects were observed by Hoshino et al. (1997) in trifluoroethanol, where a highly cooperative transition to an open helical conformation was observed. Yang et al. (1994) studied a synthetic peptide corresponding to the antiparallel triple-stranded ß-sheet in HEWL by CD and size-exclusion chromatography. They found that upon heating the polypeptide at pH 2.0 all the structure was lost at 65°C. The addition of trifluoroethanol, however, led to the formation of -helical structure. Judging from CD data, Yonezawa et al. (2002) argue that the secondary structure of HEWL depends strongly on the protein concentration and on the presence of ethanol. For low protein concentration (2 mg/ml), the structure of HEWL is predominantly helical up to 80% ethanol, whereas for higher protein concentration (5 mg/ml), the structure is predominantly ß-sheet above 75% ethanol.
Our studies of the structure of HEWL as a function of temperature and pH showed the following. The presence of single isodichroic points in Fig. 8, a–c, suggests that at all pH values studied by us the change in the secondary structure of the molecule, with respect to increasing temperature, is a two-state cooperative process (Fändrich et al. 2003). Our CD data also show (see Fig. 10 a) that when the temperature is kept close to the midpoint of the unfolding transition, after a rapid initial change, the spectrum keeps changing more slowly, toward increasing -helix content, suggesting slow and continuous changes in the secondary structure of the partially folded HEWL molecules.
The jump observed in the molecular residue ellipticity at 222 nm (Fig. 10 b) shows that the change in the lysozyme structure as a function of the heating time occurs simultaneously for a large part of the protein molecules. The jump occurs within the time at which the fibril formation starts (see Light scattering). Another interpretation could be that the change in the secondary structure is due not to a slow collective transition of the protein molecules but to a very slowly increasing number of unfolded molecules due to the lower probability of the unfolding at a lower temperature.
Though some fibrils may inevitably form in the sample studied by CD upon prolonged heating, their concentration is not high enough to significantly influence the CD spectra. Results obtained from dilute solutions during continuous heating and results from concentrated solutions quenched and diluted after prolonged heating may differ, since the latter contain many more fibrils. That is why we also carried out experiments with HEWL samples quenched and diluted after certain heating times. Our results in Fig. 11 show that after 11 days of heating of HEWL at pH 2.0 at 57°C, the structure of the free monomers, as well as that of the monomers incorporated in the fibrils, differs a lot from the structure observed in Fig. 10 a and is predominantly a random coil. The minimum at 218 nm observed in the spectrum of the separated fibrils (OLA) could be caused by formation of intermolecular ß-sheets by the protein molecules in the fibrils.
The kinetic data from the light scattering suggest that the fibril formation is not a cooperative multimolecular process, but rather a monomolecular one. The rate of fibril elongation should hold further mechanistic information. Fitting a kinetic model to the experimental data could help to resolve the type of the nucleation event, if present, and the type of the kinetic reaction (Ferrone, 1999; Hofrichter, 1986). Unfortunately, the rate of increase of the light scattering intensity (after the lag time) shows such a large spread that at present we cannot extract any reliable information on the rate of fibril elongation. Further analysis of the light scattering data is also complicated because the fibrils are strongly interacting and start overlapping already when very small amounts of fibrils have been formed.
The morphology of the fibrils formed from HEWL at different conditions varies from short and flexible (Goda et al., 2000) to long and stiff (Krebs et al., 2000; Cao et al., 2004). Cao et al. (2004) report 2–5 nm thickness and 1- to 2-µm length. The fibrils observed by Goda et al. (2000) are much shorter, 100–200 nm, and a bit thicker, 7 nm. A thickness of 7 nm is also reported by Krebs et al. (2000) with fibrils longer than 1 µm. We studied the morphology of the fibrils formed upon heating of HEWL water solutions at pH 2.0 and 3.0 at 57°C in great detail by AFM. For samples at pH 2.0 heated at 57°C we observe two populations of fibrils—thin, 2.5 ± 0.3 nm, and thick, 4.0 ± 0.7 nm (see Fig. 3, c and d, and Fig. 5 a). Similar sizes have been observed by Cao et al. (2004). The experimental data about the fine structure of the fibrils observed with the help of super-sharp AFM tips (Fig. 6) allows us to speculate about the origin of the fibril morphology. The thick fibrils seem to have a coiled structure with a period of 30 nm. The two populations of the rodlike subunit (Fig. 4, a and b, and Fig. 5 b) with lengths of 124 ± 9 nm and 157 ± 11 could be due to a tension developing in the coil resulting in a defect every four or five turns. The thin fibrils most probably result from a straight association of protein molecules without any twist because the rodlike subunit is not observed in that case.
The fibrils obtained upon heating of HEWL at pH 3.0 at 57°C are also very long (of the order of 5 µm), with a similar thickness to those obtained at pH 2.0. They seem slightly more flexible and the rodlike unit is not observed.
Although protein hydrolysis does take place at pH 2.0, we do not think that it is crucial in the aggregation process. Malisauskas et al. (2003) arrived at a similar conclusion for the case of equine lysozyme. The arguments are that, first of all, at pH 3.0 there is no hydrolysis, but judging from light scattering and AFM experimental data, fibril formation is observed to be very similar to that at pH 2.0, albeit much slower, since HEWL is much more stable at pH 3.0. Second, if fragments play a special role in the aggregation process, one might expect that they would be preferentially incorporated into the fibrils. However, separated fibrils and (partly hydrolyzed) monomers are nearly identical in SDS-PAGE, suggesting an even distribution of fragments.
Now we can draw some parallels between fibrils formed from ß-lg and HEWL. In both cases fibrils are formed at low ionic strength far from the isoelectric point of the protein. On the other hand the mechanism of fibril formation seems to depend strongly on the unfolding pathway of the protein at the particular conditions. In the case of ß-lg the aggregation starts immediately when the heating is started (Arnaudov et al., 2003). This means that the structure of the ß-lg molecule at 80°C, pH 2.0, is appropriate for fibril formation. Whether that is due to the large ß-sheet content in the secondary structure of ß-lg is a matter for further study. In the case of HEWL, simply increasing the temperature above the melting transition of the protein is not sufficient for fibril formation. The difference in the behavior of HEWL at pH 2.0, 3.0, and 4.0 can hardly be due to electrostatics because at all these pH values the protein molecule is strongly charged (between 11 and 17 positive charges for the pH range between 4.0 and 2.0, respectively; Kuehner et al., 1999). The difference must be sought in the different unfolding pathways at different conditions. Even at pH 2.0, 2 days are needed for the HEWL molecules to adopt a configuration suitable for forming fibrils. Therefore, despite the similarities between the fibrillar structures formed from different proteins and the similarities between the conditions at which they form, a detailed approach is needed to establish the mechanism of fibril formation for every individual protein.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
HEWL fibrils are formed after a lag time that is practically concentration-independent. This means that the governing process for the fibril formation is the change in the structure of single protein molecules caused by a prolonged exposure to a temperature close to the midpoint of the unfolding transition.
The fibril morphology is complex. The fibrils formed at pH 2.0 are long and straight with a length of the order of 5 µm and predominant thickness of 4 nm and consist of stiff rodlike subunits with length either 124 or 157 nm. On a smaller scale the fibrils consist of a coiled structure with a period of 30 nm that gives the appearance of the rodlike subunits probably because of defects occurring every four or five turns.
The fibrils consist mostly of full-length HEWL, although, some fragments due to hydrolysis at pH 2.0 and 57°C are probably incorporated into the fibrils. At any rate the hydrolysis of the protein is not the cause of the aggregation since at pH 3.0 no hydrolysis is detected but fibrils do form.
Submitted on July 1, 2004; accepted for publication October 4, 2004.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Aymard, P., T. Nicolai, and D. Durand. 1999. Static and dynamic scattering of ß-lactoglobulin aggregates formed after heat-induced denaturation at pH 2. Macromolecules. 32:2542–2552.[CrossRef]
Booth, D. R., M. Sunde, V. Bellotti, C. V. Robinson, W. L. Hutchinson, P. E. Fraser, P. N. Hawkins, C. M. Dobson, S. E. Radford, C. C. F. Blake, and M. B. Pepys. 1997. Instability, unfolding and aggregation of human lysozyme variants underlying amyloid fibrillogenesis. Nature. 385:787–793.[CrossRef][Medline]
Bucciantini, M., E. Giannoni, F. Chiti, F. Baroni, L. Formigli, J. Zurdo, N. Taddel, G. Ramponi, C. M. Dobson, and M. Stefani. 2002. Inherent toxicity of aggregates implies a common mechanism for protein misfolding diseases. Nature. 416:507–511.[CrossRef][Medline]
Cao, A., D. Hu, and L. Lai. 2004. Formation of amyloid fibrils from fully reduced hen egg white lysozyme. Protein Sci. 13:319–324.
Chamberlain, A. K., C. E. MacPhee, J. Zurdo, L. A. Morozova-Roche, H. A. O. Hill, C. M. Dobson, and J. J. Davis. 2000. Ultrastructural organization of amyloid fibrils by atomic force microscopy. Biophys. J. 79:3282–3293.
Dobson, C. M. 2001. The structural basis of protein folding and its links with human disease. Philos. Trans. R. Soc. Lond. B. 356:133–145.[CrossRef][Medline]
Dobson, C. M., P. A. Evans, and S. E. Radford. 1994. Understanding how proteins fold: the lysozyme story so far. Trends Biochem. Sci. 19:31–37.[CrossRef][Medline]
Fändrich, M., V. Forge, K. Buder, M. Kittler, C. M. Dobson, and S. Diekmann. 2003. Myoglobin forms amyloid fibrils by association of unfolded polypeptide segments. Proc. Natl. Acad. Sci. USA. 100:15463–15468.
Ferrone, F. 1999. Analysis of protein aggregation kinetics. Methods Enzymol. 309:256–274.[Medline]
Goda, S., K. Takano, Y. Yamagata, R. Nagata, H. Akutsu, S. Maki, K. Namba, and K. Yutani. 2000. Amyloid protofilament formation of hen egg lysozyme in highly concentrated ethanol solution. Protein Sci. 9:369–375.[Abstract]
Gosal, W. S., A. H. Clark, P. D. A. Pudney, and S. B. Ross-Murphy. 2002. Novel amyloid fibrillar networks derived from a globular protein: beta-lactoglobulin. Langmuir. 18:7174–7181.[CrossRef]
Hofrichter, J. 1986. Kinetics of sickle hemoglobin polymerization III. Nucleation rates determined from stochastic fluctuations in polymerization progress curves. J. Mol. Biol. 189:553–571.[CrossRef][Medline]
Hoshino, M., Y. Hagohara, D. Hamada, M. Kataoka, and Y. Goto. 1997. Trifluoroethanol-induced conformational transition of hen egg-white lysozyme studied by small-angle X-ray scattering. FEBS Lett. 416:72–76.[CrossRef][Medline]
Ikeda, S., and V. J. Morris. 2002. Fine-stranded and particulate aggregates of heat-denatured whey proteins visualized by atomic force microscopy. Biomacromolecules. 3:382–389.[CrossRef][Medline]
Itzhaki, L. S., P. A. Evans, C. M. Dobson, and S. E. Radford. 1994. Tertiary interactions in the folding pathway of hen lysozyme: kinetic studies using fluorescent probes. Biochemistry. 33:5212–5222.[CrossRef][Medline]
Jimenez, J. L., E. J. Nettleton, M. Bouchard, C. V. Robinson, C. M. Dobson, and H. R. Saibil. 2002. The protofilament structure of insulin amyloid fibrils. Proc. Natl. Acad. Sci. USA. 99:9169–9201.
Jimenez, J. L., G. Tennent, M. Pepys, and H. R. Saibil. 2001. Structural diversity of ex vivo amyloid fibrils studied by cryo-electron microscopy. J. Mol. Biol. 311:241–247.[CrossRef][Medline]
Kamatari, Y. O., T. Konno, M. Kataoka, and K. Akasaka. 1998. The methanol-induceed transition and the expanded helical conformation in hen lysozyme. Protein Sci. 7:681–688.[Abstract]
Kavanagh, G. M., A. H. Clark, and S. B. Ross-Murphy. 2000. Heat-induced gelation of globular proteins: part 3. Molecular studies on low pH ß-lactoglobulin gels. Int. J. Biol. Macromol. 28:41–50.[CrossRef][Medline]
Koike, A., N. Nemoto, and E. Doi. 1996. Structure and dynamics of ovalbumin gels: 1. Gel induced by high-temperature heat treatment. Polymer. 37:587–593.[CrossRef]
Krebs, M. R. H., D. K. Wilkins, E. W. Chung, M. C. Pitkeathly, A. K. Chamberlain, J. Zurdo, C. V. Robinson, and C. M. Dobson. 2000. Formation and seeding of amyloid fibrils from wild-type hen lysozyme and a peptide fragment from the ß-domain. J. Mol. Biol. 300:541–549.[CrossRef][Medline]
Kuehner, D. E., J. Engmann, F. Fergg, M. Wernick, H. W. Blanch, and J. M. Prausnitz. 1999. Lysozyme net charge and ion binding in concentrated aqueous electrolyte solutions. J. Phys. Chem. B. 103:1368–1374.
Langton, M., and A.-M. Hermansson. 1992. Fine-stranded and particulate gels of ß-lactoglobulin and whey protein at varying pH. Food Hydrocolloids. 5:523–539.
Lomakin, A., D. B. Teplov, D. A. Kirschner, and G. B. Benedek. 1997. Kinetic theory of fibrillogenesis of amyloid ß-protein. Proc. Natl. Acad. Sci. USA. 94:7942–7947.
Malisauskas, M., V. Zamotin, J. Jass, W. Noppe, C. M. Dobson, and L. A. Morozova-Roche. 2003. Amyloid protofilaments from the calcium-binding protein equine lysozyme: formation of ring and linear structures depends on pH and metal ion concentration. J. Mol. Biol. 330:879–890.[CrossRef][Medline]
Morozova-Roche, L. A., J. Zurdo, A. Spencer, W. Noppe, V. Receveur, D. B. Archer, M. Joniau, and C. M. Dobson. 2000. Amyloid fibril formation and seeding by wild-type human lysozyme and its disease-related mutational variants. J. Struct. Biol. 130:339–351.[CrossRef][Medline]
Nemoto, N., A. Koike, K. Osaki, T. Koseki, and E. Doi. 1993. Dynamic light scattering of aqueous solutions of linear aggregates induced by thermal denaturation of ovalbumin. Biopolymers. 33:551–559.[CrossRef][Medline]
Pepys, M. B., P. N. Hawkins, D. R. Booth, D. M. Vigushin, G. A. Tennent, A. K. Soutar, N. Totty, G. Nguyen, C. C. F. Blake, C. J. Terry, T. G. Feest, A. M. Zalin, and J. J. Hsuan. 1993. Human lysozyme gene mutations cause hereditary systemic amyloidoses. Nature. 362:553–557.[CrossRef][Medline]
Radford, S. E., C. M. Dobson, and P. A. Evans. 1992. The folding of hen lysozyme involves partially structured intermediates and multiple pathways. Nature. 358:302–307.[CrossRef][Medline]
Redfield, C., and C. M. Dobson. 1988. Sequential 1H NMR assignment and secondary structure of hen egg white lysozyme in solution. Biochemistry. 27:122–136.[CrossRef][Medline]
Sagis, L. M. C., C. Veerman, and E. van der Linden. 2004. Mesoscopic properties of semiflexible amyloid fibrils. Langmuir. 20:924–927.[CrossRef][Medline]
Sasahara, K., M. Demura, and K. Nitta. 2000. Partially unfolded state of hen lysozyme studied by circular dichroism spectroscopy. Biochemistry. 39:6475–6482.[CrossRef][Medline]
Sunde, M., and C. C. F. Blake. 1998. From the globular to the fibrous state: protein structure and structural conversion in amyloid formation. Q. Rev. Biophys. 31:1–39.[CrossRef][Medline]
Tanaka, S., Y. Oda, M. Ataka, K. Onuma, S. Fujiwara, and Y. Yonezawa. 2001. Denaturation and aggregation of hen egg lysozyme in aqueous ethanol solution studied by dynamic light scattering. Biopolymers. 59:370–379.[CrossRef][Medline]
Tani, F., M. Murata, T. Higasa, M. Goto, N. Kitabatake, and E. Doi. 1995. Molten globule state of protein molecules in heat-induced transparent food gels. J. Agric. Food Chem. 43:2325–2331.[CrossRef]
Yang, J. J., M. Pitkeathly, and S. E. Radford. 1994. Far-UV circular dichroism reveals a conformational switch in a peptide fragment from the ß-sheet of hen lysozyme. Biochemistry. 33:7345–7353.[CrossRef][Medline]
Yonezawa, Y., S. Tanaka, T. Kubota, K. Wakabayashi, K. Yutani, and S. Fujiwara. 2002. An insight into the pathway of the amyloid fibril formation of hen egg white lysozyme obtained from a small-angle X-ray and neutron scattering study. J. Mol. Biol. 323:237–251.[CrossRef][Medline]
Zurdo, J., I. Guijarro, and C. M. Dobson. 2001. Preparation and characterization of purified amyloid fibrils. J. Am. Chem. Soc. 123:8141–8142.[CrossRef][Medline]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
