Permeabilization of Lipid Bilayers Is a Common Conformation-dependent Activity of Soluble Amyloid Oligomers in Protein Misfolding Diseases*

Amyloid fibrillization is multistep process involving soluble oligomeric intermediates, including spherical oligomers and protofibrils. Amyloid oligomers have a common, generic structure, and they are intrinsically toxic to cells, even when formed from non-disease related proteins, which implies they also share a common mechanism of pathogenesis and toxicity. Here we report that soluble oligomers from several types of amyloids specifically increase lipid bilayer conductance regardless of the sequence, while fibrils and soluble low molecular weight species have no effect. The increase in membrane conductance occurs without any evidence of discrete channel or pore formation or ion selectivity. The conductance is dependent on the concentration of oligomers and can be reversed by anti-oligomer antibody. These results indicate that soluble oligomers from many types of amyloidogenic proteins and peptides increase membrane conductance in a conformation-spe-cific fashion and suggest that this may represent the common primary mechanism of pathogenesis in amy-loid-related degenerative diseases.

Soluble amyloid oligomers are a common intermediate in the pathway for amyloid fibril formation and have been implicated as the primary toxic species of amyloids related to neurodegenerative disease (1)(2)(3)(4)(5)(6). More recent reports indicate that soluble amyloid oligomers are intrinsically toxic even when they are formed from proteins that are not normally related to degenerative disease (3), and the toxic activity of soluble oligomers may be related to a common generic structure that they share (6). Although the primary mechanism of amyloid toxicity is not clear, the fact that different amyloids reside in either the cytosolic or extracellular compartments and the observation that cytosolic amyloid aggregates are toxic when applied externally to cells (6,7) points to the cell plasma membrane as a potential primary target of amyloid pathogenesis. Indeed, there are many reports of membrane perturbations caused by amyloids like A␤ (8), but it isn't clear whether these effects are specific to soluble oligomers nor whether they are common to other types of amyloids. Here we report that homogeneous populations of spherical amyloid oligomers and protofibrils increase the conductivity of membranes by a non-channel mechanism. This effect is observed for all soluble oligomers tested regardless of protein sequence and is not observed for amyloid fibrils or soluble low molecular weight species, suggesting that the increase in membrane conductivity may be a primary common mechanism of amyloid oligomer pathogenesis.

MATERIALS AND METHODS
Peptide Synthesis-Peptide synthesis: A␤ peptides, prion 106 -126, and IAPP 1 were synthesized by fluoren-9-ylmethoxy carbonyl chemistry using a continuous flow semiautomatic instrument as described previously (9). The purity was checked by analytical reverse phase-high performance liquid chromatography and by electrospray mass spectrometry. Polyglutamine KKQ40KK was a gift from Dr. Ronald Wetzel, and ␣-synuclein was a gift from Dr. Ralf Langen.
Preparation of Oligomers and Fibrils-Lyophilized peptides and proteins were resuspended in 50% acetonitrile in water and re-lyophilized. Soluble oligomers were prepared by dissolving 1.0 mg of peptide or protein in 400 l of hexafluoroisopropanol for 10 -20 min at room temperature. 100 l of the resulting seedless solution was added to 900 l of double distilled H 2 O in a siliconized Eppendorf tube. After 10 -20min incubation at room temperature, the samples were centrifuged for 15 min at 14,000 ϫ g, and the supernatant fraction (pH 2.8 -3.5) was transferred to a new siliconized tube and subjected to a gentle stream of N 2 for 5-10 min to evaporate the hexafluoroisopropanol. The samples were then stirred at 500 r.p.m. using a Teflon-coated micro stir bar for 24 -48 h at 22°C. Aliquots (10 l) were taken at 6 -12-h intervals for observation by atomic force microscopy, electron microscopy (EM), and size exclusion chromatography (SEC). The time at which the oligomers are most populated depends on several factors, such as speed of stirring and the concentration, so it is important to check their size and homogeneity with SEC. The maximum occurs between 6 h to 4 days. For prion 106 -126 the samples were heated at 65°C for 30 min, and poly(Q) was heated for 2 h at 37°C before stirring. These conditions were empirically determined to provide homogeneous populations of oligomers. In the case of IAPP, oligomer formation should be monitored as early as possible. The purity of IAPP is important, because the presence of reduced IAPP promotes fibril formation leading to short-lived oligomers.
Fibrils were prepared under three different conditions, water (pH 3.8 -4.2), 10 mM Tris (pH 7.4), and 50 mM Tris, 100 mM NaCl (pH 7.4), each containing 0.02% sodium azide. The final peptide or protein concentration was 0.3-0.5 mg/ml. The samples were prepared as described above for oligomers but stirred at room temperature for 6 -9 days. Fibril formation was monitored by thioflavin T fluorescence and UV light scattering. Once fibril formation was complete, the solutions were centrifuged at 14,000 x g for 20 min, the fibril pellet was washed three times with the doubly distilled water and then resuspended in the desired buffer. The morphology was verified by negative stain EM.
Bilayers-Bilayers were formed at room temperature by the union of two monolayers formed from a mixture (1:1 by weight) of phosphatidylcholine and phosphatidylserine, phosphatidylethanolamine, and cho-* This work was supported by National Institutes of Health Grants NS31230, AG00538, and AG16573 and by a grant from the Larry L. Hillblom Foundation. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. § These authors contributed equally to the work. ** To whom correspondence should be addressed. E-mail: cglabe@ uci.edu. lesterol as described elsewhere (10,11). The lipid acyl chains are all oleoyl. Briefly lipid monolayers were opposed over a hole ϳ200 m in diameter in a 15-micron-thick Teflon partition dividing the two aqueous phases. The hole, punched by electric spark, was precoated with a 2.5% solution of squalane in n-pentane. Lipids were purchased from Avanti Polar Lipids, Birmingham, AL. Salt solutions contained various concentrations NaCl buffered by 10 mM HEPES-Tris to pH 7.4. Solutions were stirred continuously with magnetic stirring bars, except that stirring was occasionally interrupted to record an I-V curve if stirringinduced noise was too objectionable. Bilayer formation was monitored by measuring capacitance. Silver/silver chloride wires were used as electrodes to apply voltages and record currents across the bilayer. The rear-chamber potential was taken as ground, and the additions were made to the front chamber. For measurements of membrane conductance a ramp protocol (Ϫ150 to ϩ150 mV, 60 mV/s) was used. Voltages were generated and currents digitized at resolution of 12 bits by an AD Laboratory ADC/DAC board running software written in the laboratory. Currents were transduced by an Axopatch 200A amplifier (Axon Instruments, Foster City, CA.) connected to the AD Laboratory board as described above. The ion selectivity was determined by measuring the reversal potential in asymmetrical ionic conditions.
Electron Microscopy-For electron microscopy, 2 l of the sample were adsorbed onto 200-mesh carbon and formvar-coated grids, airdried, and washed for 1 min in distilled water. The samples were negatively stained with 2% uranyl acetate (Ted Pella Inc., Redding, CA) for 2 min and viewed with a Zeiss 10CR microscope (80 kV).
Thioflavin T Fluorescence-Thioflavin T fluorescence was determined as described (12). The fluorescence emission spectrum was measured using a Spex Fluorolog-2 spectrofluorometer Size Exclusion Chromatography-SEC was performed using a Hewlett Packard 1050 liquid chromatograph and a 0.78 ϫ 30 cm Toso Haas G3000SWxl column in 100 mM NaCl, 50 mM Tris, pH 7.4, at a flow of 0.4 ml/min. The eluate was monitored by UV absorbance at 230 nm.

RESULTS AND DISCUSSION
Investigating the conformational basis of soluble amyloid oligomer toxicity requires the preparation homogenous and relatively stable populations of conformationally distinct soluble monomers, oligomeric intermediates and fibrils. Homogeneous samples of low molecular weight species (monomer or dimer), spherical oligomers, and fibrils were prepared from A␤40 and A␤42 (Alzheimer disease) ␣-synuclein (Parkinson disease), IAPP (Type II diabetes), polyglutamine (KKQ40KK) (Huntington disease), and prion (106 -126) H1 (Prion diseases), using the procedures we have described previously (6). Although the conditions that promote the formation of spherical oligomers are remarkably similar for the different proteins and

Membrane Permeabilization by Amyloid Oligomers 46364
peptides, the conditions were optimized for each peptide or protein to obtain the most homogenous samples possible (see "Materials and Methods"). All preparations were visualized by electron microscopy (Fig. 1A). The soluble oligomer samples contain a homogeneous population of spherical particles with an average diameter of 3-5 nm and are free of detectable mature amyloid fibrils. The fibrillar samples contain predominately long mature amyloid fibrils with very few, if any, spherical oligomers. The homogeneity was analyzed by size exclusion chromatography (Fig. 1B). Oligomeric A␤42 elutes as a symmetrical peak of ϳ90 -110 kDa and contains less than 10% low molecular weight monomer or dimeric species. The soluble low molecular weight samples contain a peak that elutes at the position expected for monomer or dimer and contains less than 1% of high molecular weight species (13).
We examined the effect of these relatively homogeneous samples on membrane conductivity using planar lipid bilayers (10). Spherical A␤42 oligomers specifically increase the conductance of the bilayer (Fig. 2A). The increase in conductivity is approximately proportional to the concentration of oligomers. No increase in conductance was observed for low molecular weight A␤ species (monomer or dimer) (Fig. 2B) or fibrils (Fig. 2C). Moreover, the prior addition of soluble low molecular weight A␤ or A␤ fibrils does not prevent the conductivity increase caused by the subsequent addition of spherical A␤ oligomers (Fig. 2, B and C). A␤40 spherical oligomers also exhibited the same conductivity increase when added to lipid bilayers (supplementary Fig. 1).
The increase in membrane conductance in response to oligomer addition occurs in the absence of any evidence of discrete ion channel or pore formation (Fig. 3A). The high sensitivity recording indicates that there is little change in the noise level in the current trace as the current increases from 0 to ϳ100 pA after oligomer addition. Previous studies of the interaction of A␤ with membranes suggested that A␤ forms pores or channels in membranes (8,14,15). The increase in conductivity reported here is unique and distinct from previous reports of pore or channel formation by A␤. In particular, we do not find discrete unitary conductance changes or evidence of open and closed states that are characteristic of ion channels. The conductance induced by soluble oligomers is not inhibited by Tris ions or Congo red that have been reported to inhibit A␤ channels (15,16). We examined the ion selectivity by measuring the bionic reversal potential. The observed reversal potentials are zero, indicating that the conductance is not ion selective (Fig. 3B). The conductance we observe for soluble A␤ oligomers is ϳ100fold greater than previously reported for similar concentrations of A␤ that were not conformationally characterized (14,15). Rather than forming pores or channels, soluble oligomers appear to enhance the ability of ions to move through the lipid bilayer on their own. Although the explanation for these differences is not entirely clear, the conditions used in the previous reports of amyloid A␤ ion channels seem unlikely to include significant amounts of soluble amyloid oligomers.
Since preincubation of soluble oligomers with oligomer-specific antibody inhibits the toxicity of all types of soluble oligomers in cell culture (6), we tested the effect of this antioligomer antibody on membrane conductance. The increase in conductivity caused by A␤ spherical oligmers is rapidly reversed (Fig. 3C) upon the addition of a conformation-dependent antibody that is specific for amyloid oligomers (6) but is not reversed by nonspecific control antibodies (data not shown). The fact that anti-oligomer antibody can reverse the conductivity increase suggests the oligomeric A␤ does not undergo a

Membrane Permeabilization by Amyloid Oligomers 46365
conformation change to a form that is not recognized by the antibody upon membrane interaction and that the oligomers are only peripherally associated with the bilayer rather than stably inserted into the hydrocarbon core. Anti-oligomer antibody appears to remove A␤ from its association with the membrane, because the infrared amide I signal at 1627 cm Ϫ1 that is specific for membrane-associated A␤ drops precipitously after anti-oligomer antibody addition. 2 Since other amyloidogenic proteins also form spherical oligomers that have a common conformation-dependent structure and are also cytotoxic (6), we tested whether they have similar effects on membrane conductivity. Spherical oligomers from ␣-synuclein, IAPP, poly(Q), and prion 106 -126 also specifically increase bilayer conductivity in a concentration-dependent fashion (Fig. 4, A-D). As observed for A␤, low molecular weight species and fibrils from these amyloid-forming proteins and peptides have no effect on membrane conductivity when tested at concentrations up to 8 M (supplementary Fig. 2). The increase in conductivity caused by ␣-synuclein, IAPP, poly(Q), and prion 106 -126 oligomers was also reversed by addition of anti-oligomer antibody (supplementary Fig. 3). Preincubation of soluble oligomers with anti-oligomer antibody also prevents the increase in membrane conductance when they are applied to lipid bilayers (data not shown).
Using homogeneous populations of low molecular weight species, spherical oligomers, and fibrils, we found that all types of spherical oligomers examined specifically induce dramatic increases in the conductivity of membranes. Since the monomeric or low molecular weight species and amyloid fibrils do not induce a conductivity change, this effect is specific for the particular conformation associated with spherical oligomers. We have recently shown that the soluble oligomers share a common structure that is associated with cellular toxicity, suggesting that soluble oligomers share a common primary mechanism of pathogenesis (6). This predicts that oligomers would also share a common target, and since some types of amyloid oligomers are cytosolic, while others are extracellular, the common target must be accessible to both the cytosolic and extracellular compartments. The plasma membrane of the cell is one of the few targets that is accessible to both compartments. We demonstrate that the increase in membrane conductivity induced by spherical oligomers is also a common property shared by all oligomers tested. This suggests that the increase in membrane permeability caused by spherical oligomers may represent the common primary mechanism of pathogenesis.
The different effects of low molecular weight A␤ and oligomeric A␤ in increasing membrane conductance is correlated with the different effects on membrane structure that have been reported as determined by small angle x-ray diffraction (17). Monomeric or low molecular weight A␤ produced a broad increase in electron density in the center of the bilayer accom-panied by a reduction in the width of the bilayer from 51 to 46 Å. In contrast, oligomeric, aggregated A␤ induces a marked decrease in electron density throughout the membrane due to an increased molecular volume of the lipid acyl chains. The decreased density of the hydrocarbon core could explain the increased conductivity of the membrane that we observed upon oligomer addition. The increase in membrane conductivity could lead to depolarization of the plasma membrane, which would be detrimental to the function of cells and especially so for neuronal function. The membrane conductance increase we report here can also account for a wide range of effects that have been reported for A␤, such as defects of cytosolic ion homeostasis and signaling as a direct consequence of the membrane conductance increase (18). Our results suggest that spherical amyloid oligomers or protofibrils are responsible for a generalized increase in membrane conductance, and this may represent the common mechanism of pathogenesis for amyloidrelated degenerative diseases.