These authors contributed equally to this work.
Heme binding inhibits the fibrillization of amyloidogenic apomyoglobin and determines lack of aggregate cytotoxicity
Article first published online: 2 JAN 2009
Copyright © 2007 The Protein Society
Volume 16, Issue 3, pages 507–516, March 2007
How to Cite
Iannuzzi, C., Vilasi, S., Portaccio, M., Irace, G. and Sirangelo, I. (2007), Heme binding inhibits the fibrillization of amyloidogenic apomyoglobin and determines lack of aggregate cytotoxicity. Protein Science, 16: 507–516. doi: 10.1110/ps.062471107
- Issue published online: 2 JAN 2009
- Article first published online: 2 JAN 2009
- Manuscript Accepted: 19 NOV 2006
- Manuscript Revised: 23 OCT 2006
- Manuscript Received: 31 JUL 2006
- amyloid fibril;
- apomyoglobin aggregation;
- fibrillization inhibition;
- amyloid cytotoxicity
Myoglobin is an α-helical globular protein containing two highly conserved tryptophanyl residues at positions 7 and 14 in the N-terminal region. The double W/F replacement renders apomyoglobin highly susceptible to aggregation and amyloid-like fibril formation under physiological conditions. In this work we analyze the early stage of W7FW14F apomyoglobin aggregation following the time dependence of the process by far-UV CD, Fourier-transform infrared (FTIR) spectroscopy, and heme-binding properties. The results show that the aggregation of W7FW14F apomyoglobin starts from a native-like globin state able to bind the prosthetic group with spectroscopic properties similar to those observed for wild-type apoprotein. Nevertheless, it rapidly aggregates, forming amyloid fibrils. However, when the prosthetic group is added before the beginning of aggregation, amyloid fibrillization is inhibited, although the aggregation process is not prevented. Moreover, the apomyoglobin aggregates formed in these conditions are not cytotoxic differently from what is observed for all amyloidogenic proteins. These results open new insights into the relationship between the structure adopted by the protein into the aggregates and their ability to trigger the impairment of cell viability.
A growing number of diseases appear to be caused by aggregation of misfolded protein that is deposited in the extra- and intracellular space (Kelly 1998; Sigurdsson et al. 2002; Westermark et al. 2002; Dobson 2003; Uversky and Fink 2004). These deposits can be amorphous (disordered) or fibrillar (ordered) (Geddes et al.1968; Sunde and Blake 1997, 1998; Wetzel 2002). Inclusion bodies are an example of amorphous aggregates, and amyloid fibrils are an example of fibrillar or ordered aggregates. The characteristics of different amyloid fibrils, namely structure and morphology, observed by electron microscopy and X-ray fiber diffraction appear to be quite similar. In fact, they display a core cross-β-sheet structure in which continuous β-sheets run perpendicular to the long axis of the fibrils (Sunde et al. 1997). Mature fibrils generally consist of two to six unbranched protofilaments, 2–5 nm in diameter, associated laterally or twisted together to form fibrils with 4–13 nm diameter (Shirahama and Cohen 1967; Shirahama et al. 1973; Jimenez et al. 1999). Amyloid formation is modulated by some critical, structural, and/or environmental factors, the understanding of which may certainly help in finding strategies to reverse fibril formation.
The propensity to form amyloid fibrils is now believed to be a generic property of the polypeptide chain and varies from protein to protein depending on the experimental conditions used to perturb the native structure (Chiti et al. 1999; Ramirez-Alvarado et al. 2000; Fandrich et al. 2001; Dobson 2003). It has been postulated that the generic amyloid conformation, the cross-β-structure, may be a universal, energetic minimum for aggregated proteins (Dobson 2001; Jahn and Radford 2005). Once formed, amyloid fibrils are extremely stable and difficult to solubilize (Smith et al. 2003). The insoluble amyloid structures originate from small water-soluble oligomeric species containing few monomeric units, each not bonded covalently to its neighboring units. It is widely accepted that these oligomeric soluble species are implicated in the toxic process (Kayed et al. 2003).
Whether all non-native, fully or partially folded states of proteins can form amyloid fibrils is still not clear, although it has been indicated that specific denatured states are critical in amyloid formation (Uversky and Fink 2001, 2004; Smith et al. 2003; Plakoutsi et al. 2004; Marcon et al. 2005). In this study, we used the amyloid-forming W7FW14F apomyoglobin to investigate the conformational state that precedes the formation of highly organized amyloid fibrils under conditions in which the wild-type protein maintains a folded structural state in an attempt to understand the molecular events underlying protein aggregation. Myoglobin is a highly soluble globular protein whose native state properties, i.e., the overall α-helical structure, do not lead us to hypothesize any predisposition to forming amyloid fibrils. In fact, most of its amino acid sequence is organized in well-defined α-helices with few or no elements of β-sheets (Evans and Brayer 1988). Recently, we have shown that the apomyoglobin mutant W7FW14F at physiological pH and room temperature forms amyloid fibrils by a population of prefibrillar granular aggregates in about 15 d (Sirangelo et al. 2002, 2004). These fibrils share several morphological, structural, and tinctorial features typical of amyloids, including increased thioflavin T fluorescence and yellow-green birefringence under cross-polarized light upon Congo Red binding. Atomic force microscopy (AFM) of the fibrils indicated an average fibril diameter of about 3–5 nm and a length of >1000 nm (Malmo et al. 2006). Insight into their secondary structural organization was obtained from CD analysis. The spectrum was consistent with that observed for proteins having a β-sheet-rich structure. The estimated percentages of secondary structure were: α 10%, β 37%, turn 22%, and unordered 31%. Similarly to what was observed with amyloidogenic proteins, mature fibrils formed by W7FW14F are not cytotoxic, whereas granular aggregates are (Sirangelo et al. 2004; Malmo et al. 2006). Two factors appear crucial for fibril formation in the case of W7FW14F apomyoglobin: noncovalent interactions and charge effects. In particular, the increased hydrophobicity of the N-terminal segment caused by the double WF substitution and the increased tendency to form β-strands make apomyoglobin more susceptible to aggregation (Eisenberg et al. 1984; Sirangelo et al. 2004); however, this is prevented at low pH because of the electrostatic repulsion. On raising the pH toward neutrality, the net charge of the polypeptide chain decreases and this favors the aggregation process.
The aim of this study is to elucidate the mechanism enabling the W7FW14F apomyoglobin mutant to form amyloid fibrils by the identification of the conformational state underlying the early events of the aggregation process. Preliminary attempts indicated that the aggregation process might involve a conformational state different both from the native state and the well-characterized partially folded state of this protein (Goto et al. 1990; Hughson et al. 1990; Bismuto et al. 1992; Jennings and Wright 1993; Sirangelo et al. 1994; Kataoka et al. 1995). However, the time dependence of protein aggregation and the early events of the process were not investigated in detail. The results presented in this paper indicate that, in the early stage of the aggregation process, the W7FW14F apomyoglobin mutant adopts a native-like globin fold able to bind the prosthetic group. Thus, the amyloid aggregation detected at physiological pH and room temperature cannot be ascribed to the inability to correctly fold, introduced by the double tryptophanyl substitution as previously suggested (Sirangelo et al. 2000, 2002, 2004). The effect of heme on the aggregation process was also investigated. The heme binding does not prevent apomyoglobin aggregation, but the aggregates formed in these conditions do not exhibit fibrillar morphology and are not cytotoxic. This opens new insights into the correlation between structure and cytotoxicity of protein aggregates.
It is generally assumed that the conformation of apomyoglobin depends on both pH and salt concentration (Hsu and Woody 1971; Goto and Fink 1990; Bismuto et al. 1992; Sirangelo et al. 1994). Specifically, it is native at neutral pH, partially folded near pH 4.0, and fully unfolded at pH 2.0 in the presence of very low amounts of salt. The partially folded state contains the A, G, and H helices folded as in the native conformation, whereas the rest of the molecule is essentially unfolded (Goto et al. 1990; Hughson et al. 1990; Bismuto et al. 1992; Jennings and Wright 1993; Sirangelo et al. 1994; Kataoka et al. 1995). In mildly acidic condition, i.e., pH 4.0, the amyloid-forming W7FW14F apomyoglobin mutant adopts a soluble, partially folded conformation similar to that of the wild-type protein, but with a slightly lower helical content. The pH increase from 4.0 to 7.0 causes protein aggregation and amyloid formation instead of resulting in the formation of the soluble native state (Sirangelo et al. 2002).
In order to understand the molecular mechanism underlying W7FW14F amyloid fibril formation, we focused our attention on the early events of the aggregation process. The time dependence of apomyoglobin aggregation at pH 7.0 was investigated by far-UV circular dichroism. Registration of CD spectra started immediately after pH neutralization of the partially folded conformation, and spectra were then recorded in time (Fig. 1). During the first 6 h of observation, the intensity of the CD signal decreased gradually, whereas the shape of the CD spectra did not change, all spectra showing the two minima at 208 and 222 nm, characteristic of an α-helical conformation. The inset of Figure 1 shows the time dependence of the molar ellipticity at 208 nm in the first 6 h of the aggregation process. The data were analyzed by fitting the experimental values to logarithmic as well as exponential functions. The best fit was obtained using a logarithmic function: [θ](t) = [θ]0 + A · ln t. Extrapolation of the ellipticity values to t = 1 sec allowed us to build the CD spectrum of the conformational state attained immediately after pH neutralization (Fig. 2); also shown in Fig. 2 is the spectrum recorded 15 d after the beginning of aggregation process. The spectrum is largely modified with respect to that detected at early times and, as expected, typical of polypeptide, mostly organized in the β-structure. The analysis of the extrapolated CD spectrum by CDPro algorithm (Sreerama and Woody 2000) is shown in Table 1 in comparison with that of wild-type apomyoglobin at neutral pH. It is interesting to observe that the content of the secondary structure of the amyloid-forming apomyoglobin approaches that of wild-type protein. Therefore, it is evident that in the early events of the aggregation process, W7FW14F apomyoglobin adopts a conformation similar to native wild-type protein and, only later, a transition from α-helical to β-cross-structure occurs.
Fourier transform infrared (FTIR) spectroscopy further corroborated our conclusion. FTIR spectroscopy is a sensitive and commonly used technique of monitoring the α to β transition underlying amyloid formation (Dzwolak et al. 2002, 2003; Fandrich et al. 2003). Figure 3 shows the FTIR spectra in the amide I′ region recorded immediately after pH adjustment and after 15 d from the beginning of the aggregation process. At the beginning of the aggregation process, the position of the amide I′ is at 1647 cm−1, a wavenumber typical for proteins mostly organized in α-helical conformation (Nevskaya and Chirgadze 1976). The spectrum recorded after 15 d shows an amide I′ maximum close to 1617 cm−1. This value typically belongs to amyloid conformation (Zandomeneghi et al. 2004) and is identical to that found for the amide I′ of amyloid fibrils formed by wild-type apomyoglobin (Fandrich et al. 2003). It is noteworthy to mention that the shift of the amide I′ maximum is already evident after 24 h from the beginning of the aggregation process.
To better examine the native-like structural properties adopted by W7FW14F apomyoglobin in the early events of the aggregation process, we followed the holoprotein reconstitution by examining the Soret region spectral properties. Hemin was added to the amyloid-forming apomyoglobin at pH 4.0, and then the pH was raised to neutrality. Figure 4 shows the appearance of the characteristic Soret band near 410 nm when the pH was gradually raised from 4.0 to 7.0. This suggests the presence of a heme-binding region similar to that of the native wild-type protein. Figure 5 shows the CD activity in the Soret spectral region of the reconstituted amyloid-forming holoprotein in comparison with the naturally expressed and reconstituted wild-type myoglobins. It is known that the optically inactive heme group acquires activity upon binding to apomyoglobin. The induced Cotton effect arises from a coupled oscillator interaction between the electronic transitions of the heme and the allowed π-π* transitions of nearby histidine, phenylalanine, and tyrosine side chains (Hartley et al. 1999). The sign and the magnitude of the acquired CD activity are related to the nature of the chromophore binding site. The spectra shown in Figure 5 indicate that no difference exists between naturally expressed and reconstituted wild-type holoprotein, both spectra presenting a strong negative dichroic activity band with emission minimum centered at 409 nm, a value coincident with the wavelength maximum of the Soret absorption spectrum. The reconstituted W7FW14F holoprotein displayed a similar negative band with a slightly red-shifted minimum. Differences in the position of the Soret absorption are generally caused by polarity changes of the protein environment around the heme group. At present it is arduous to assess the origin of this shift: It could be due to an increased solvent accessibility to heme or to a different interaction between the heme moiety and the side chains of surrounding amino acids.
The presence of heme does not affect the protein propensity to aggregate. We followed the time dependence of apomyoglobin aggregation at neutral pH in the presence of heme by measuring the absorbance at 280 nm of supernatant solution after centrifugation. The results are shown in Figure 6 in comparison with those obtained in the absence of the prosthetic group. It is evident that the rate of protein aggregation increases significantly when heme is added to the protein solution. The aggregation rates obtained fitting the experimental data to an exponential function as y(t) = y0 + Ae−kt were: k+eme = (0.64 ± 0.18) h−1; k−eme = (0.39 ± 0.11) h−1.
Additional confirmation was obtained by SDS-PAGE. Protein samples were centrifuged immediately after neutralization, and the pellet and the supernatant fractions were analyzed. The results are shown in Figure 7. In the presence of heme, the band relative to the pellet fraction, collected at the beginning of the aggregation reaction, was more intense than that observed in the absence of heme. On the contrary, the protein band relative to the pellet, collected in the absence of heme, was less intense than that of the supernatant solution.
Direct structural information on the organization of the polypeptide backbone in the aggregates formed in the presence of heme was obtained from FTIR spectroscopy. The spectrum recorded 24 h after the beginning of the aggregation reaction performed in the presence of the prosthetic group is shown in Figure 8. When heme is present, the spectrum maintains the typical β characteristics (Zandomeneghi et al. 2004), although with a shift of the amide I′ position from 1617 to 1627 cm−1. However, both values are within the wavenumber range associated with the presence of the amyloid structure. The FTIR spectrum did not change in time even after 15 d of observation.
AFM was used to study the morphological features of the aggregates formed in the presence and in the absence of the prosthetic group. The images were recorded over a period of 15 d. Spherical aggregates predominated in the early aggregation times (0–1 h) both in the absence (Fig. 9A) and in the presence of heme (Fig. 9E). The only detectable difference at this time is that the average height (i.e., diameter) of the aggregates formed in the absence of heme is lower than that of the samples incubated in the presence of the prosthetic group. After 2 d, pore-like (or annular) structures with a lateral diameter ranging between 100 and 500 nm appear only in protein aggregates grown without heme (Fig. 9B), whereas the samples incubated in the presence of heme retain the same spherical structure observed at early times of aggregation. At later times (7–15 d), the spherical and annular structures disappeared almost completely (Fig. 9C) and mature fibrils were formed only in the absence of heme (Fig. 9D), while granular structures were persistent in the sample aggregated in the presence of the prosthetic group (Fig. 9F). These data indicate that the presence of heme inhibits the formation of amyloid fibrils.
The overall results indicate that the amyloidogenic apomyoglobin adopts, at the beginning of the aggregation process, a helical conformation able to bind the prosthetic group. Probably, the early stages of the aggregation consist of the association of helical monomers into oligomers with a subsequent structural transition resulting in the formation of β-strands, which later associate into a regular fibrillar structure. Such association does not occur when hemin is bound with a consequent inhibition of the amyloid fibril formation.
The different morphological properties shown by W7FW14F apomyoglobin aggregates formed in the presence and in the absence of heme prompted us to investigate the relationship between molecular features and toxicity of protein assemblies to cultured cells. The cytoxicity of W7FW14F apomyoglobin aggregates was evaluated by the MTT assay. Soluble prefibrillar oligomers have been implicated as the primary toxic species of amyloids; a decrease in reduced MTT levels is generally observed in the presence of these species, whereas fibrils have little or no effect (Booth et al. 1997; Kayed et al. 2003; Stefani and Dobson 2003; Sirangelo et al. 2004; Malmo et al. 2006). In particular, we tested the toxicity associated with aggregates formed in the absence and in the presence of heme at the beginning of the aggregation process and after 15 d. The results are shown in Figure 10. The MTT reduction level decreased significantly to 50% (p < 0.001) compared with the control level by exposing the cells for 24 h to W7FW14F prefibrillar aggregates, whereas mature fibrils displayed no significant effect. Interestingly, the aggregates formed in the presence of heme were not significantly toxic for cells even in the early stage of the aggregation process. This was further confirmed by examining the cytotoxic effect of pellet and supernatant fractions separated immediately after the beginning of the aggregation process. Both fractions did not decrease the level of MTT reduction. Moreover, no effect was detected increasing the protein concentration in the supernatant fraction. This led us to conclude that the absence of toxicity could be related to the presence of a lower amount of soluble oligomers resulting from the faster aggregation kinetics. In conclusion, the presented data indicate that the amyloid-mediated cell toxicity is strictly related to the appearance of a basic toxic fold on the aggregation pathway. Probably, the heme inhibits the formation of the structural determinants that make the aggregates able to interact with cell components, such as membranes, and trigger the impairment of cell viability.
Amyloid fibril formation is known to be driven both by hydrophobic and electrostatic interactions. “In vitro,” the process is generally initiated when the native state of a protein is slightly destabilized, originating partially folded intermediates that are more prone to aggregation. Fibril formation occurs after a defined period of time, called the lag time, during which protein monomers associate forming soluble oligomers. These oligomers serve as nuclei for fibril elongation (Booth et al. 1997; Bucciantini et al. 2002). High-denaturant concentration does not favor the formation of protein aggregates, because denaturant–protein interactions prevent contacts among protein molecules. Similarly, the establishment of intermolecular interactions in strongly acidic conditions is avoided by the electrostatic repulsion among positively charged groups. Therefore, the in vitro formation of amyloid fibrils is favored in conditions in which partially folded conformations are populated (Uversky and Fink 2004). Similarly, peptides or proteins that are natively unfolded need to adopt a partially folded conformation either before or during fibril assembly. This seems to happen in the formation of fibrils formed by the β amyloid peptide associated with Alzheimer's disease (Petkova et al. 2002; Torok et al. 2002). More recently, a growing interest has been turned to the study of aggregation under experimental conditions not drastically different from the natural setting, in which proteins are initially in a native-like conformation (Plakoutsi et al. 2004; Marcon et al. 2005; Sambashivan et al. 2005). The amyloid-forming W7FW14F apomyoglobin lends itself to structural studies for exploring the amyloid formation in a context similar to the physiological medium. In our early studies, we proposed that the pathway of apomyoglobin amyloid formation at neutral pH starts from a conformational state different both from the native and the well-characterized acidic partially folded state. In particular, we suggested that the apomyoglobin amyloid precursor could be a misfolded form possessing a high level of β structure (Sirangelo et al. 2002). In the present study, we analyze the apomyoglobin aggregation in the earliest stages of the process. The results show that the conformational state of the amyloid-forming apomyoglobin at physiological pH presents, at the beginning of the aggregation process, only minor structural modifications compared with the native form of the wild-type protein. This was probed by examining the time dependence of far UV CD activity, FTIR spectroscopy, Soret absorption, and Soret rotational strength. The following similarities at pH 7.0 between amyloid-forming apomyoglobin and wild-type protein were detected: (1) the content of helical structure is quite similar, i.e., 56% and 65%, respectively; (2) the amide I′ maximum in FTIR spectra is positioned at 1647 cm−1, a wavenumber typical of helical structures, (3) both proteins bind the prosthetic group with similar spectroscopic properties. Despite these similarities, the W7FW14F apomyoglobin rapidly aggregates forming amyloid fibrils.
The question arises whether aggregation of the amyloid-forming apomyoglobin under physiological conditions starts from a native-like conformation or via partially unfolded states in rapid equilibrium with the native-like structure. It has been reported by Dobson and coworkers (Fandrich et al. 2003) that amyloid fibril formation of wild-type apomyoglobin does not correlate with the presence of partially folded species. On the other hand, there are examples in the recent literature that natively folded states are a precursor for amyloid formation (Bousset et al. 2002, 2003; Laurine et al. 2003; Plakoutsi et al. 2004). Fibril maturation has been shown to involve the early formation of ring-shaped protofilaments, from which mature fibrils emanate. Annular structures have been detected during fibril maturation of both amyloid disease-associated and unassociated proteins (Lashuel et al. 2002b, 2003; Relini et al. 2004). The formation of ring-shaped structures has been reported to be an indicator of persistence of native-like regions during fibril formation (Malisauskas et al. 2003). In this respect, the observation that W7FW14F apomyoglobin forms ring-shaped structures during its fibrillization suggests that the protein could aggregate through a native-like conformation.
Recently, Chow et al. (2003) reported that the 1–36 N-terminal fragment of wild-type apomyoglobin displays a high level of β-structure and forms macroscopic aggregates when the pH gets closer to neutrality. The WF substitutions at positions 7 and 14 certainly contribute to increase the hydrophobicity and the β-propensity of the N-terminal region (Sirangelo et al. 2004), thus making the protein more prone to aggregate under physiological conditions. Moreover, a recent report showed that the mutations that play a critical role in the rate-determining step of apomyoglobin aggregation are those located within the N-terminal region of the molecule (Vilasi et al. 2006). These observations strongly suggest that aggregation of W7FW14F apomyoglobin could occur through the association of the N-terminal regions of natively folded molecules.
In the W7FW14F apomyoglobin amyloid aggregation, the starting point is the protein folded in a native-like conformation and the end point is the same protein aggregated and forming a cross-β-structure. However, it may not be excluded that the starting point is a partially non-native state as it could be suspected by the observed differences in the Soret spectral properties and far UV CD. The first step could involve an amyloidogenic form constituted by a native-like or partially non-native structure with a high tendency to form intermolecular β-sheets. Small oligomers are formed during the lag phase, which can be detected with thioflavin T assay (Sirangelo et al. 2004); then they convert into protofilaments and, finally, into mature fibrils. If heme is present, fibril maturation does not occur, although apomyoglobin aggregates show an FTIR spectrum consistent with the presence of β structure. Amide I′ components occur in the wavenumber range from 1600 to 1700 cm−1 and arise primarily from stretching vibration of main-chain carbonyl groups. Thus, this technique may provide direct structural information about the spatial orientation of the polypeptide backbone. It has been reported that the wavenumber range of amide I′ components of amyloid fibrils extends from 1611 to 1630 cm−1, whereas that of native β-sheet globular protein is comprised from 1630 to 1643 cm−1 (Zandomeneghi et al. 2004). These differences have been ascribed to a different distribution of Φ/φ dihedral angles, the average number of strands per sheet, and the β-sheet twist. Surprisingly, thermally induced nonfibrillar aggregates as well as inclusion bodies display infrared spectra with amide I′ maximum in the amyloid region (Jackson and Mantsch 1991; Fink 1998). However, the differences detected for apomyoglobin aggregates formed in the presence and in the absence of heme, i.e., 1617 versus 1627 cm−1, although in the same range, could be related to the occurrence of structural modifications between the two aggregated states.
The general picture emerging from these data indicates that the double mutation WF changes the physico-chemical characteristics of the N-terminal region of the molecule, making the native state of the protein prone to aggregate. The binding of the prosthetic group inhibits the formation of mature fibrils, probably avoiding that the specific region making up the core of the fibrils could form intermolecular interactions. It is well known that the interaction of heme with apomyoglobin strongly stabilizes the native structure, particularly the E and F helices (Sirangelo et al. 2000). We have recently identified, by H/D exchange followed by limited proteolysis (G. Infusini, C. Iannuzzi, S. Vilasi, L. Birolo, D. Pagnozzi, P. Pucci, G. Irace, and I. Sirangelo, unpubl.), that A, B, E, and part of D and G participate in the formation of the amyloid fibril core. This led us to hypothesize that the mutations introduced in the N-terminal region are responsible for the increased propensity to aggregate and that other molecular regions, especially the E helix, are involved in fibril elongation.
An accumulation of evidence has led to the hypothesis that the oligomeric precursors mediate cell dysfunction and cell death in degenerative diseases associated with protein misfolding, whereas the fibrillar forms of misfolded proteins are much less toxic or inert (Lambert et al.1998; Hartley et al. 1999; Bucciantini et al. 2002; Lashuel et al. 2002a; Stefani and Dobson 2003; Sirangelo et al. 2004; Malmo et al. 2006). The common pathogenic mechanism, shared by the oligomeric forms of both amyloid disease-associated and nonassociated proteins, has been suggested to be intimately related to their similar molecular organization. A demonstration came from the finding that an oligomer-specific antibody, which recognizes micellar Aβ and not Aβ fibrils, also recognizes soluble oligomers of other amyloidogenic proteins and peptides (Kayed et al. 2003). The aggregates formed in the presence of heme are not toxic for cells, even in the early stage of the aggregation process, different from that detected in the absence of heme and in the case of other amyloidogenic proteins. At present, it is hard to find an explanation for this result. Recently, Singer and Dewji (2006) proposed that the common molecular structure responsible for the toxicity of the early oligomeric aggregates of amyloidogenic proteins and peptides is the cylindrically wound double-β-stranded subunit that Perutz et al. (2002) derived from the X-ray diffraction studies of polyglutamine fibers. This structure interacts with cell membrane, forming a water-filled channel of about 15 nm in diameter, large enough to permit the diffusion of small ions. On the other hand, native-like assemblies of yeast prion Ure2p retaining the α-helix content of the soluble monomer have been recently reported to be highly toxic to cultured cells (Pieri et al. 2006).
The observations reported in this paper add new insights into the complex relationship between structure and pathogenic activity, showing that apomyoglobin aggregates with rather similar secondary organization can have different biological effects. Thus, the cytotoxic effect is not due to nonspecific reactions caused by aggregated protein, but it is strictly related to the formation of a basic toxic fold able to interact with cell membrane.
Materials and methods
Protein purification and assembly into oligomers and fibrils
Wild-type and W7FW14F mutant myoglobin were expressed and purified essentially as described elsewhere (Sirangelo et al. 2002). Briefly, proteins were expressed in the Escherichia coli M15[pREP4] strain as amino-terminal His-tagged forms and purified via affinity chromatography on Ni2+-nitrilotriacetic acid resin (Qiagen). The wild-type myoglobin was expressed in soluble form and purified under native conditions. The heme was removed by the 2-butanone extraction procedure (Teale 1959). The W7FW14F mutant was sequestered into insoluble inclusion bodies and purified under denaturing conditions. Refolding was achieved by removing denaturant by dialysis against 10 mM NaH2PO4, 10 mM C2H3O2Na (pH 2.0), containing decreasing concentrations of urea. The pH of the protein solution was then adjusted at the desired value, depending on different experimental conditions. Protein concentration was determined under denaturing conditions and absorption was measured at 280 and 275 nm for wild-type and mutant apomyoglobin, respectively. The molar extinction coefficients, calculated from tryptophan and tyrosine content (Edelhoch 1967), were ε280 = 13,500 M−1 cm−1 and ε275 = 3750 M−1 cm−1. The assembly into fibrils was achieved by raising the pH of a 40-μM protein solution to 7.0.
CD spectra were recorded at 25°C on a Jasco J-715 spectropolarimeter using thermostated quartz cells of 0.1 cm and 1-cm path length for far and near-UV, respectively. Spectral acquisition was taken at 0.2-nm intervals with a 4-sec integration time and a bandwidth of 1.0 nm. An average of three scans was obtained for all of the spectra. Photomultiplier absorbance did not exceed 600 V in the spectral region analyzed. Data were corrected for buffer contributions and smoothed using the software provided by the manufacturer (System Software version 1.00). All measurements were performed under nitrogen flow. Protein concentration was 20 × 10−6 M for far-UV and 4 × 10−6 M with a hemin/protein molar ratio of 1:1 for near-UV measurements. The results were expressed as mean residue ellipticity [Θ]MRW in units of degree cm2 dmol−1. A mean residue weight of 115 was used for the peptide chromophore. Protein secondary structure estimation was performed using CDPro software, which contains three software packages, i.e., CDSSTR, CONTIN/LL, and SELCON3 (Sreerama and Woody 2000). The secondary structure percent contents reported are the arithmetic means of the estimates obtained from each program.
Soret-region absorption spectra were recorded at 25°C on a Jasco V-550 double-beam spectrophotometer. Protein concentration was 8 × 10−6 M with a hemin/protein molar ratio of 1:1.
FTIR spectra were recorded on a Multiscope FT-IR Microscope coupled with a Spectrum One spectrometer (Perkin Elmer). The FTIR spectra in transmission mode were collected (4000 cm−1–600 cm−1 range) at a resolution of 4 cm−1 with 16 accumulations per run. For each spectrum, signals corresponding to the water and CO2 vapors were automatically subtracted and the baseline corrected.
Spectra were recorded with dry samples of W7FW14F apomyoglobin obtained through repeated cycles of lyophilization and dissolution in D2O at a concentration of 40 μM.
Atomic force microscopy
Tapping mode AFM in air was performed by using a Solver Pro Scanning Probe Microscope (NT-MDT, Moscow, Russia). Rectangular silicon cantilevers 100-μm long with resonant frequencies in the range of from 190 to 325 kHz and a nominal force constant between 5.5 and 22.5 N/m (typical 11.5 N/m) were used. AFM images with a size between 5 and 15 μm were recorded at a scan rate below 1 Hz with 256 × 256 pixels per image. The cantilevers had integrated tips with a curvature radius of 10 nm and a tip cone angle <220.
The W7FW14F apomyoglobin samples were diluted to a concentration of 50 nM in 2 mM MgCl2. Immediately afterward, 10–20 μL of protein solution were deposited onto freshly cleaved ruby mica. The sample was incubated for 2 min, rinsed with water, and blown dry with nitrogen at 0.5 bars.
Cell culture and incubation with protein aggregates
NIH-3T3 cells (mouse fibroblasts, American Type Culture Collection) were cultured in Dulbecco's modified eagle's medium high glucose supplemented with 10% bovine calf serum and 3.0 mM glutamine in a 5.0% CO2 humidified environment at 37°C. A total of 50 units/mL penicillin and 50 μg/mL streptomycin were added to the medium. The cells were plated at a density of 100, 000 cells/well on 12-well plates in 1 mL of medium. After 24 h, protein samples were added to the cell medium at a final concentration of 20 μM in culture medium. Cells incubated with culture medium without protein served as control. The morphological characteristics of aggregates were checked by AFM before the addition to the cell medium.
Cell viability was assessed as the inhibition of the ability of cells to reduce the metabolic dye MTT to a blue formazan product (Hansen et al. 1989). After 24 h of incubation with protein samples, cells were rinsed with PBS. A total of 100 μL of a stock MTT solution (5 mg/mL in PBS) was then added to 900 μL of DMEM without phenol red, containing 10% bovine calf serum/well, and incubation was continued for an additional 3 h. The medium was aspirated, and cells were treated with isopropyl alcohol-0.1 M HCl for 20 min. Levels of reduced MTT were determined by measuring the difference in absorbance at 570 and 690 nm. Data are expressed as average percent reduction of MTT with respect to the control ± S.D. from three independent experiments carried out in triplicate. For statistical analysis, a two-tailed Student's t-test with unequal variance at a significance level of 5%, unless otherwise indicated, was used.
This work has been supported by a grant from Ministero dell'Università e della Ricerca Scientifica e Tecnologica (FIRB 2003, “Folding e aggregazione di proteine, unità INBB”). We are grateful to Silvia Tacchi for her assistance with the AFM measurements.
- 1992. Salt-induced refolding of myoglobin at acidic pH: Molecular properties of a partly folded intermediate. Arch. Biochem. Biophys. 298: 624–629. , , and
- 1997. Instability, unfolding and aggregation of human lysozyme variants underlying amyloid fibrillogenesis. Nature 385: 787–793. , , , , , , , , , and , et al.
- 2002. The yeast prion Ure2p retains its native α-helical conformation upon assembly into protein fibrils in vitro. EMBO J. 21: 2903–2911. , , , and
- 2003. The native-like conformation of Ure2p in fibrils assembled under physiologically relevant conditions switches to an amyloid-like conformation upon heat-treatment of the fibrils. J. Struct. Biol. 141: 132–142. , , , and
- 2002. Inherent toxicity of aggregates implies a common mechanism for protein misfolding disease. Nature 416: 507–511. , , , , , , , , , and
- 1999. Designing conditions for in vitro formation of amyloid protofilaments and fibrils. Proc. Natl. Acad. Sci. 96: 3590–3594. , , , , , , and
- 2003. Chain length dependence of apomyoglobin folding: Structural evolution from misfolded sheets to native helices. Biochemistry 42: 7090–7099. , , , , , and
- 2001. The structural basis of protein folding and its links with human disease. Philos. Trans. R. Soc. Lond. B Biol. Sci. 356: 133–145.
- 2003. Protein folding and misfolding. Nature 426: 884–890.
- 2002. Fourier transform infrared spectroscopy in high-pressure studies on proteins. Biochim. Biophys. Acta 1595: 131–144. , , , and
- 2003. Aggregation of bovine insulin probed by DSC/PPC calorimetry and FTIR spectroscopy. Biochemistry 42: 11347–11355. , , , and
- 1967. Spectroscopic determination of tryptophan and tyrosine in proteins. Biochemistry 6: 1948–1954.
- 1984. Analysis of membrane and surface protein sequences with the hydrophobic moment plot. J. Mol. Biol. 179: 125–142. , , , and
- 1988. Horse heart metmyoglobin. A 2.8-A resolution three-dimensional structure determination. J. Biol. Chem. 263: 4263–4268. and
- 2001. Amyloid fibrils from muscle myoglobin. Nature 410: 165–166. , , and
- 2003. Myoglobin forms amyloid fibrils by association of unfolded polypeptide segments. Proc. Natl. Acad. Sci. 100: 15463–15468. , , , , , and
- 1998. Protein aggregation: Folding aggregates, inclusion bodies and amyloid. Fold. Des. 3: R9–R23.
- 1968. “Cross-β” conformation in proteins. J. Mol. Biol. 32: 343–358. , , , and
- 1990. Phase diagram for acidic conformational states of apomyoglobin. J. Mol. Biol. 214: 803–805. and
- 1990. Acid-induced folding of proteins. Proc. Natl. Acad. Sci. 87: 573–577. , , and
- 1989. Re-examination and further development of a precise and rapid dye method for measuring cell growth/cell kill. J. Immunol. Methods 119: 203–210. , , and
- 1999. Protofibrillar intermediates of amyloid β-protein induce acute electrophysiological changes and progressive neurotoxicity in cortical neurons. J. Neurosci. 19: 8876–8884. , , , , , , , and
- 1971. The origin of the heme Cotton effects in myoglobin and haemoglobin. J. Am. Chem. Soc. 93: 3515–3525. and
- 1990. Structural characterization of a partly folded apomyoglobin intermediate. Science 249: 1544–1548. , , and
- 1991. Protein secondary structure from FT-IR spectroscopy: Correlation with dihedral angles from three-dimensional Ramachandran plots. Can. J. Chem. 69: 1639–1642. and
- 2005. The yin and yang of protein folding. FEBS J. 272: 5962–5970. and
- 1993. Formation of a molten globule intermediate early in the kinetic folding pathway of apomyoglobin. Science 262: 892–896. and
- 1999. Cryo-electron microscopy structure of an SH3 amyloid fibril and model of the molecular packing. EMBO J. 18: 815–821. , , , , , , and
- 1995. Structural characterization of the molten globule and native states of apomyoglobin by solution X-ray scattering. J. Mol. Biol. 249: 215–228. , , , , , and
- 2003. Common structure of soluble amyloid oligomers implies common mechanism of pathogenesis. Science 300: 486–489. , , , , , , and
- 1998. The alternative conformations of amyloidogenic proteins and their multi-step assembly pathways. Curr. Opin. Struct. Biol. 8: 101–106.
- 1998. Diffusible, nonfibrillar ligands derived from Aβ1-42 are potent central nervous system neurotoxins. Proc. Natl. Acad. Sci. 95: 6448–6453. , , , , , , , , , and , et al.
- 2002a. Neurodegenerative disease: Amyloid pores from pathogenic mutations. Nature 418: 291. , , , , and
- 2002b. α-Synuclein, especially the Parkinson's disease-associated mutants, forms pore-like annular and tubular protofibrils. J. Mol. Biol. 322: 1089–1102. , , , , , , and
- 2003. Mixtures of wild-type and a pathogenic (E22G) form of Aβ40 in vitro accumulate protofibrils. including amyloid pores. J. Mol. Biol. 332: 795–808. , , , , , , and
- 2003. Lithostathine quadruple-helical filaments form proteinase K-resistant deposits in Creutzfeldt-Jakob disease. J. Biol. Chem. 278: 51770–51778. , , , , , , , , , and , et al.
- 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. , , , , , and
- 2006. Tetracycline inhibits W7FW14F apomyoglobin fibril extension and keeps the amyloid protein in a pre-fibrillar, highly cytotoxic state. FASEB J. 20: 346–347. , , , , , , and
- 2005. Amyloid formation from HypF-N under conditions in which the protein is initially in its native state. J. Mol. Biol. 347: 323–335. , , , , , , , and
- 1976. Infrared spectra and resonance interactions of amide-I and II vibration of α-helix. Biopolymers 15: 637–648. and
- 2002. Amyloid fibers are water-filled nanotubes. Proc. Natl. Acad. Sci. 99: 5591–5595. , , , and
- 2002. A structural model for Alzheimer's β-amyloid fibrils based on experimental constraints from solid state NMR. Proc. Natl. Acad. Sci. 99: 16742–16747. , , , , , , and
- 2006. The yeast prion Ure2p native-like assemblies are toxic to mammalian cells regardless of their aggregation state. J. Biol. Chem. 281: 15337–15344. , , , , , , and
- 2004. Aggregation of the Acylphosphatase from Sulfolobus solfataricus: The folded and partially unfolded states can both be precursors for amyloid formation. J. Biol. Chem. 279: 14111–14119. , , , and
- 2000. A systematic exploration of the influence of the protein stability on amyloid fibril formation in vitro. Proc. Natl. Acad. Sci. 97: 8979–8984. , , and
- 2004. Monitoring the process of HypF fibrillization and liposome permeabilization by protofibrils. J. Mol. Biol. 338: 943–957. , , , , , , , , , and , et al.
- 2005. Amyloid-like fibrils of ribonuclease A with three-dimensional domain-swapped and native-like structure. Nature 437: 266–269. , , , , and
- 1967. High-resolution electron microscopic analysis of the amyloid fibril. J. Cell Biol. 33: 679–708. and
- 1973. Fibrillar assemblage of variable segments of immunoglobulin light chains: An electron microscopic study. J. Immunol. 110: 21–30. , , , and
- 2002. Infectivity of amyloid diseases. Trends Mol. Med. 8: 411–413. , , and
- 2006. Evidence that Perutz's double-β-stranded subunit structure for β-amyloids also applies to their channel-forming structures in membranes. Proc. Natl. Acad. Sci. 103: 1546–1550. and
- 1994. Solvent and thermal denaturation of the acidic compact state of apomyoglobin. FEBS Lett. 338: 11–15. , , and
- 2000. The effect of tryptophanyl substitution on folding and structure of myoglobin. Eur. J. Biochem. 267: 3937–3945. , , , , and
- 2002. Tryptophanyl substitutions in apomyoglobin determine protein aggregation and amyloid-like fibril formation at physiological pH. J. Biol. Chem. 277: 45887–45891. , , , , , and
- 2004. Fibrillogenesis and cytotoxic activity of the amyloid-forming apomyoglobin mutant W7FW14F. J. Biol. Chem. 279: 13183–13189. , , , , , , and
- 2003. A systematic investigation into the effect of protein destabilization on β 2-microglobulin amyloid formation. J. Mol. Biol. 330: 943–954. , , , , and
- 2000. Estimation of protein secondary structure from dichroism circular spectra: Comparison of CONTIN, SELCON, and CDSSTR methods with an expanded reference set. Anal. Biochem. 287: 252–260. and
- 2003. Protein aggregation and aggregate toxicity: New insights into protein folding, misfolding diseases and biological evolution. J. Mol. Med. 81: 678–699. and
- 1997. The structure of amyloid fibrils by electron microscopy and X-ray diffraction. Adv. Protein Chem. 50: 123–159. and
- 1998. From the globular to the fibrous state: Protein structure and structural conversion in amyloid formation. Q. Rev. Biophys. 31: 1–39. and
- 1997. Common core structure of amyloid fibrils by synchrotron X-ray diffraction. J. Mol. Biol. 273: 729–739. , , , , , and
- 1959. Cleavage of the heme-protein link by acid methylethylketone. Biochim. Biophys. Acta 35: 543.
- 2002. Structural and dynamic features of Alzheimer's Aβ peptide in amyloid fibrils studied by site-directed spin labeling. J. Biol. Chem. 277: 40810–40815. , , , , , , and
- 2001. Evidence for a partially folded intermediate in α synuclein fibril formation. J. Biol. Chem. 276: 10737–10744. and
- 2004. Conformational constraints for amyloid fibrillation: The importance of being unfolded. Biochim. Biophys. Acta 1698: 131–153. and
- 2006. Kinetics of amyloid aggregation of mammal apomyoglobins and correlation with their amino acid sequences. FEBS Lett. 580: 1681–1684. , , , , , , and
- 2002. Amyloid fibril protein nomenclature-2002. Amyloid 9: 197–200. , , , , , , , , , and
- 2002. Ideas of order for amyloid fibril structure. Structure 10: 1031–1036.
- 2004. FTIR reveals differences between native β-sheet proteins and amyloid fibrils. Protein Sci. 13: 3314–3321. , , , and