Báo cáo khoa học: a-Synuclein–synaptosomal membrane interactions Implications for fibrillogenesis pot

a-Synuclein–synaptosomal membrane interactions Implications for fibrillogenesis Euijung Jo 1 , Audrey A. Darabie 1 , Kyung Han 1 , Anurag Tandon 1,3 , Paul E. Fraser 1,2 and JoAnne McLaurin 1,4 1 Centre for Research in Neurodegenerative Diseases; Departments of 2 Medical Biophysics, 3 Medicine and 4 Laboratory Medicine and Pathobiology, University of Toronto, Ontario, Canada a-Synucle in exists in two different compartments in vivo – correspondingly existing as two different forms: a mem- brane-bound form that is predominantly a-helical and a cytosolic form that is randomly structured. It h as been sug- gested that these environmental and structural differences may play a role in aggregation propensity and d evelopment of pathological lesions observed in P arkinson’s disease (PD). Such effects may be accentuated by mutations observed in familial PD k indreds. In order to test this hypothesis, wild- type and A53 T mutant a-synuclein interactions with rat brain synaptosomal membranes were examined. Previous data has demonstrated that the A30P mutant has defective lipid binding and therefore was not examined in this study. Electron m icroscopy demonstrated that wild-type a-synuclein fibrillogenesis is accelerated in the presence of synaptosomal membranes whereas the A53T a-synuclein fibrillogenesis i s inhibited under the same conditions. These results suggested that subtle sequence changes in a-synuc- lein could s ignificantly a lter interaction with membrane bilayers. Fluorescence and absorption spectroscopy using environment s ensitive probes demonstrated variations in the inherent lipid properties in the presence and absence of a-synuclein . A ddition of wild-type a-synuclein to synapto- somes did not significantly alter the membrane fluidity at either the fatty acyl chains or headgroup space, suggesting that synaptosomes have a high capacity for a-synuclein binding. I n contrast, synaptosomal membrane fluidity was decreased by A53T a-synuclein binding with concomitant packing of the lipid headgroups. These results suggest that alterations in a-synuclein–lipid interactions may contribute to physiological changes detected in early onset PD. Keywords: anisotropy; electron microscopy; fibrillogenesis; fluorescence s pectroscopy; m embrane. The link between a-synuclein and P arkinson’s disease (PD) is unclear; yet a-synuclein is the major component of the primary neuropathological feature, Lewy bodies [1–5]. The association of a-synuclein with familial Parkinson’s disease was established in several PD kindreds with t he disc overy of two missense mutations A53T and A30P, which suggested an etiological significance rather than a secondary patho- logical e vent [6,7]. More recently, a-synuclein gene triplica- tion has been identified in a large family of early onset autosomal dominant PD [8]. These studies suggest that multiple alterations in a-synuclein protein s equence, expres- sion level or f unction may lead to the downstream clinical manifestation of PD. Immunocytochemistry has revealed a-synuclein positive inclusions within astrocytes and oligo- dendrocytes of PD-patients and glial and neuronal inclu- sions of multiple system atrophy patients [9–14]. These studies suggested that a-synuclein and its abnormal p rotein aggregation might play an active part in these n eurodegen- erative diseases. The physiological function of a-synuclein remains largely unknown, but it has been suggested that it may play a r ole in synaptogenesis and lipid trafficking [15,16]. a-Synuclein protein structure contains seven imperfect repeats of 11 amino acids which form the amino terminal amphipathic a-h elices, a central hydrophobic domain followed by an acidic C-terminal, r ich in g lutamate [15,17]. These primary and secondary structure characteristics su ggested a role fo r a-synuclein in protein–membrane interactions. Initial in vitro experiments showed that the helical structure of wild-type (WT) a-synuclein was in duced and s tabilized by binding to synthetic m embranes [18]. Subsequently, it was found that a-synuclein forms a dimer or trimer when bound to lipid vesicles [19–21]. Furthermore, WT a-synuclein isolated from human SH-SY5Y cells is monomeric in soluble or cytosolic form and oligomeric when associated with lipids [ 22]. Subsequent studies demonstrated that the N-terminal region of a-synuclein bound to lipids while the C-terminus remained soluble and randomly structured [23]. As both familial PD mutations are located in the N -terminal lipid-binding region, it is possible that these mutations may alter the normal equilibrium between a membrane-bound dimeric/oligomeric form and a free cytosolic form of the W T a-synuclein. Previous investigations have shown that A30P and A53T mutations had no effect on the lipid-induced a-helical structure of a-synuclein [24], yet these mutations increase a-synuclein oligomerization, in vitro [25]. M ore d etailed analyses of the A30P mutation demonstrated t hat although the secondary structure could not be distinguished from Correspondence to J. McLaurin, Centre for Research in Neuro- degenerative Diseases, Tanz Neuroscience Building 6 Queen’s Park Crescent West, Toronto, Ontario, M5S 3H2, Canada. Fax: +1 416 978 1878; Tel.: +1 416 978 1035; E-mail: j.mclaurin@utoronto.ca Abbreviations: DPH, 1,6-diphenyl-1,3,5-hexatriene; MC540, Merocyanine 540; PD, Parkinson’s disease; WT, wild-type. (Received 1 0 May 20 04, accepted 8 June 2004) Eur. J. Biochem. 271, 3180–3189 (2004) Ó FEBS 2004 doi:10.1111/j.1432-1033.2004.04250.x WT a-synuclein, the 3D conformation was d ifferent [26]. F urthermore, t he differences in 3D structure results in defective membrane b inding of the A30P m utant a-synuclein [27,28]. On the other hand, the A53T mutant could not be distinguished from WT a-synu clein with respect to lipid binding or protein structure, yet A53T mutant a-synuclein was less effective than WT at destabili- zing membrane bilayers [19]. In order to i nvestigate this apparent discrepancy, we undertook the examination of A53T mutant a-synuclein in the presence of lipid bilayers formed from rat brain synaptosomes. We chose to evaluate physiologically rele- vant membranes in o rder to determine whether properties associated with a-synuclein-lipid binding in homogeneous lipid environments are replicated in environments consisting of relevant ratios of phospholipids, ganglios ides and sphingomyelin. Furthermore, these studies may h elp to distinguish subtle differences between m utant a nd WT a-synuclein interactions and to further elucidate potential roles for a-synuclein–lipid interactions in vivo.Inorderto examine differences and explain pathological findings, we examined the ability of these membranes to facilitate a-synuclein assembly by electron microscopy. Changes in the m embrane physical characteristics as a result of a-synuclein interactions were examined by fluorescence spectroscopy using environment s ensitive probes. These parameters define the extent to which a-synuclein penetrates the lipid bilayer or disrupts lipid headgroup p acking. Materials and methods Expression and purification of recombinant a-synuclein Human a-synuclein cDNAs, wild-type and A53T mutant were subcloned i nto the plasmid pET-28a (Novagen), u sing NcoIandHindIII restriction s ites. a-Synuclein was over- expressed in Eschericia coli BL21 ( DE3) and i solated over a Q-Sepharose column as described p reviously [19]. Aliquots from all purification steps were analyzed by SDS/PAGE to confirm purity. Protein concentration was determined by Lowry assay. Circular dichroism (CD) spectra were re cor- ded on a Jasco Circular Dichroism Spectrometer (Tokyo, Japan) at 25 °C. Spectra were obtained from 195 to 250 nm, with a 0.5 nm step, 1 nm bandwidth and 10 s collection time per step. The p eptide conformation was determined by adding an aliquot of stock peptide solution into NaCl/P i (pH 7 .4) at a final peptide concentration of 10 l M . Synaptosome isolation Rat g rey matter was dissected a fter cervical dislocation (according to CACC guidelines) a nd homogenize d in 10 volumes of 320 m M sucrose, 5 m M Hepes, pH 7.4 (homo- genizing buffer) using a glass homogenizer. The homogenate was s pun at 1050 g for 10 min. The supernatant was re-spun at 13 30 0 g for 15 min, 4 °C. The p ellet was resuspended in homogenizing buffer and loaded onto a discontinuous ficoll gradient, consisting of 13, 9 and 5% ficoll. The gradient was spun for 35 min at 60 000 g at 4 °C. The synaptosomes were isolated from the 9–13% layer and diluted into H epes buffer. Final synaptosome isolation w as achieved after centrif uga- tion at 13 300 gfor5 min, 4 °C [29,30]. Lipids were extracted from the s ynaptosomes using Folch partition [chloroform/ methanol/water (v/v/v); 2 : 1 : 0.6] and subsequently con- centrated under a stream of N 2 .Thesampleswerestoredat )20 °C until use. Phospholipid concentration was deter- mined using the Bartlett assay [31]. Electron microscopy WT and A53T a-synucleins were i ncubated i n the presence and absence of synaptosomal vesicles at a final peptide concentration of 5.8 l M .Thea-synuclein to lipid ratio was maintained at 1 : 20 (by mass). For negative stain electron microscopy, c arbon-coated pioloform g rids were floated on aqueous solutions of peptides. After the grids were blotted and air-dried, the samples were stained with 1% ( w/v) phosphotungstic acid and examined on a Hitachi 7000 electron microscope operated at 75 kV [32]. Tyrosine fluorescence spectroscopy Steady-state fluo rescence was measured at 20 °Cusing a Photon Technology International (PTI, London, ON, Canada) QM-1 Fluorescence spectrophotometer equipped with excitation intensity correction a nd a m agnetic stirrer. Tyrosine emission spectra from 290 to 340 nm were collected (excitation wavelength 281 nm, 0.5 sÆnm )1 ,band pass 4 nm). A cuvette with a 1 cm path-length was used. For aggregation s tudies, 10 l M of WT or A53T a-synuclein or hen e gg-white lysozyme was incubated in the presence or absence of synaptosomal membrane vesicles a t a 1 : 20 molar ratio for up to 96 h with stirring [33]. The samples were measured for fluorescence as a measure of total tyrosine fluorescence and then centrifuged for 30 min at 15 600 g in order to sediment fibres. The relative amount of tyrosine fluorescence in the supernatant was then deter- mined as a measure of soluble protein fraction. Steady-state fluorescence anisotropy Anisotropy experiments were performed on a PTI fluori- meter e quipped with manual polarizers as described previ- ously [34]. Excitation a nd emission wavelengths w ere s et at 360 nm and 425 nm with a slit width of 1 and 4 nm, respectively. Our system was calibrated initially using 1,6- diphenyl-1,3,5-hexatriene (DPH; Molecular Probes) in mineral oil, which should give an anisotropy equal to 1. The g-factor was calculated using horizontally polarized excitation and subsequent comparison of the horizontal and vertical emissions, which for our instrument is 0.883. Lipid vesicles were diluted to 250 lgÆmL )1 in NaCl/P i , incubated for 20–30 min in t he presence and absence of a-synuclein or lysozyme, and then incubated subsequently for a further 30 min with DPH at a 1 : 500 probe/lipid ratio. Fluores- cence intensity was measured with the excitation polarizer in the vertical position and the analyzing emission polarizer in the vertical ( I VV ) and horizontal (I VH ) positions and anisotropy, r, was calculated using Eqn (1); r ¼ I VV À gI VH I VV þ 2gI VH ð1Þ Lipid vesicles in the absence of DPH were measured in order to evaluate the effect of light scattering on our measurements. Ó FEBS 2004 Lipid-induced a-synuclein fibrillogenesis (Eur. J. Biochem. 271) 3181 Laurdan generalized polarization Steady-state excitation and e mission spectra were collected on the PTI fluorimeter. Laurdan (Molecular Probes) was added to preformed lipid vesicles in the presence and absence of a-synuclein at a 500 : 1 lipid/probe ratio. The laurdan generalized polarization (GP) parameter as devel- oped by Parasassi and colleagues [35] was calculated as follows. The emission GP parameter is given by Eqn (2): GP em ¼ I 400 À I 340 I 400 þ I 340 ð2Þ where, I 400 and I 340 are the flu orescence intensities measured at all emission wavelengths between 420 and 520 nm. Using the fixed excitation wavelength of 400 and 340 nm, respectively. The excitation GP is given by Eqn (3): GP ex ¼ I 440 À I 490 I 440 þ I 490 ð3Þ where, I 440 and I 490 are t he fluorescence intensities at each excitation wavelength from 320 to 420 nm, measured at fixed emission wavelengths o f 440 nm and 4 90 nm, r espect- ively. Merocyanine 540 absorption spectroscopy Merocyanine 540 (MC540, Molecular Probes) absorption spectra were obtained at room temperature on a Beckman spectra DU530. The dye was added to preformed vesicles at a p robe/lipid ratio o f 1 : 500 [36]. Final MC540 molar concentration in the cuvette was 21.3 l M . Absorption spectra were obtained between 400 and 600 nm with 1 n m steps. The lipid alone baseline in t he absence of MC540 was subtracted from all spectra, a nd then corrected by referring the absorbances at 600 nm to zer o (Eqns 4 and 5). [monomer] ¼ A À½e D  C=2 e m À e D =2 ð4Þ [dimer] ¼ ðC À [monomer]Þ 2 ð5Þ Where, A is the absorbance at 569 nm, e is the constant for MC540 dimer or monomer a t t he given wavelength, e m ¼ 1.511 · 10 5 and e D ¼ 5400, while C is the final MC540 concentration. After this correction, the absorbance values at 569 nm were used to calculate the dimerization constant (K dapp )as described by Bernik & Disalvo [37] (Eqn 6). K dapp ¼ [dimer] [monomer] 2 ð6Þ Results a-Synuclein morphological characteristics Recombinant a-synucleins, both wild-type (WT) and A53T mutant, were over-expressed and purified from Eschericia coli BL21 as described previously [19]. After purification, both WT a nd A53T mutant a-synuclein ran as s ingle bands on SDS/PAGE (Fig. 1A) and displayed typical random secondary structure when diluted into N aCl/P i ,pH7.3 (Fig. 1 B). These results confirm both the purity and the random structure o f t hese monomeric a-synuclein prepara- tions. Previous studies have suggested that WT a-synuclein becomes oligomeric upon binding to acidic phospholipids [28,38] as well as when bound to membranes from r at brain or neuroblastoma cells [22,39]. Furthermore, a-synuclein has been located to and associated with the presynaptic terminals and synaptosomal membrane surfaces b y immu- nogold localization s tudies [16]. T herefore, bilayers com- posed of lipids isolated f rom r at brain synaptosomes represent a physiologically relevant system in which to examine a-synuclein–lipid interactions. In order to distin- guish a-synuclein–lipid interactions from a-synuclein–pro- tein interactions, synaptosomal lipids were isolated by Folch Fig. 1. Physical properties of W T and A53T mutant a-synuclein. Both WT (lan e 1) and A53T mutant (lane 2) a-synuclein demonstrated pre- dominantly single band s w hen elec trophores ed b y SDS /PAGE (A). Th e second ary structu re of th e a-synucleins was determined using c ircular dichroism spectroscopy (B). Both WT (h)andA53T(s)mutanta-synuclein were randomly structured when diluted into NaCl/P i , as illustrated by a single m inimum below 200 nm. 3182 E. Jo et al.(Eur. J. Biochem. 271) Ó FEBS 2004 partition and subsequently sonicated to form small uni- lamellar vesicles. To determine if synaptosomal membranes promote assembly of WT and A53T mutant a-synuclein into amyloid fibres, we examined a-synuclein structural characteristics by negative stain electron microscopy (Fig. 2). In the absence of lipid and at low l M concentrations, WT a-synuclein did not form detectable amyloid fibres after a 3-day incubation (Fig. 2 A) but upon extensive incubation formed fibres (data not shown). In contrast, abundant a-synuclein fibres were detected in the presence of synaptosomal lipid vesicles (Fig. 2 B). The fibres appeared to be associated with both the surface and the edge of vesicles suggesting a direct interaction between fibre and the vesicle bilayer. These results are consistent with our previous studies in which WT a-synuclein assembled into aggregates and protofibrils in the presence of phosphatidylcholine/phosphatidylserine bilayers [19]. Similarly, Lee and coworkers d emonstrated that membrane-bound WT a-syn uclein could seed the aggregation of cytosolic WT a-synuclein as determined using SDS/PAGE analyses [39]. These data are consistent with partial insertion of a-synuclein into th e bilayer, which acts as an anchor for site-directed fibril assembly. Preced- ence for this mechanism of amyloid formation has been Fig. 2. Negative stain electron microscopy of a-synuclein in the presence of synaptosomal membranes. WT a-synuclein incubated in buffer alone (A) did not form fib rils. When incubated in the presenc e of synaptosomal membranes (B) abundant a-synuclein fibrils could be detected with organization along the v esicle surface. A53T mu tant a-synuclein formed abundant fib rils of varyin g length and l ateral aggregation (C). In the presence o f syn aptosomal membranes only a few protofibrils of A53T mutant a-synuclein were detected (D). W hen in cubated alone, lysozyme formed a few fibres of varying lengths (E) and were indistinguishable from fibres found in the presence of synaptosomal membranes (F). Scale bar is 50 nm for all. Ó FEBS 2004 Lipid-induced a-synuclein fibrillogenesis (Eur. J. Biochem. 271) 3183 proven for Alzheimer’s amyloid-b peptide fibrillogenesis at the surface of bilayers [40,41]. Under identical conditions and in the absence of syna- ptosomal lipid vesicles, A53T mutant a-synuclein formed abundant fibres after a 3-day i ncubation period (Fig. 2C). The fibrils were of varying lengths with a characteristic 10–12 nm diameter [42,43]. These short fibres demonstra- ted varying degrees of lateral aggregation into larger bundles up to 130 nm in diameter. In t he presence of synaptosomal lipid vesicles, very few A53T a-synuclein fibrils could be d etected as c ompared to when i ncubated in the absence of lipid (Fig. 2D). The few fibrils that were detected had the morphological characteristics of protofi- brils but were always intimately associated with synapto- somal vesicle edges. In contrast to WT a-synuclein, synaptosomal membranes inhibit the formation of A53T mutant a-synuclein amyloid fibres, suggesting that the A53T mutation affects the mode of interaction with lipid bilayers. To demonstrate the specificity of a-synuclein–synapto- somal membrane interactions, hen egg w hite lysozyme was used as a c ontrol amyloid-forming protein that is n ot found in the nervous system. Under identical conditions, lysozyme incubated alone for 3 days demonstrated ver y few fibr es of varying lengths (Fig. 2E) and was not d istinguishable from fibres formed in the presence of synaptosomal membranes (Fig. 2 F). Th ese results suggest that synaptosomal lipid vesicles do not alter lysozyme fibril formation. In order to correlate the morphological studies of a-synuclein fibre formation with quantitative fibril growth, the intrinsic tyrosine fluorescence of a-synuclein was used to monitor the amou nt of soluble protein after incubating in the p resence of synaptosomal membrane v esicles [33]. After 2, 48 and 96 h of incubation , soluble a-synuclein or lysozyme was separated from aggregated and fibrillar protein by low-speed centrifugation (Fig. 3). These condi- tions are s ufficient to pellet protein aggregates and fibrils but not unilamellar lipid vesicles. In agreement with the electron microscopy studies, the amount of soluble WT a-synuclein decreased s ignificantly over t ime w hereas A53T a-synuclein remained soluble. An aliquot from each sample was examined by electron microscopy; WT but not A53T a-synuclein fibres were detected as described above. Furthermore, lysozyme in the presence of synaptosomal membrane vesicles aggregated over time but to a l esser extent than WT a-synuclein (Fig. 3). Fatty acyl chain mobility To further characterize the differences in WT and A53T mutant a-synuclein binding to lipid bilayers an d to deter- mine the most influential lipid prop erties that govern a-synuclein fibrillogenesis, we examined the effect of a-synuclein on synaptosomal membrane fluidity. The availability of fluorescent dyes that penetrate to varying levels within the bilayer and exhibit fluorescent properties characteristic of their local environment allow u s to address the extent t o which a-synuclein inserts into t he lipid bilayer. Specifically, the effects of a-synuclein on the mobility of t he fatty a cyl chains within the bilayer can be determined using the steady-state fluorescence anisotropy of the dye, DPH [34]. The relative motion of the D PH molecule within the bilayer is determined by polarized fluorescence and expressed as r, the anisotropy constant, that is inversely proportional to the degree of membrane fluidity. The relative fluidity of synaptosomal membranes was considered gel-like as indicated by an r-value close to 0.2 (Fig. 4 ). Addition of WT a-synuclein to the synaptosomal membranes had little effect on the m embrane fluidity and was not dependent on the a-synuclein/lipid ratio. These results suggest that although WT a-synuclein binds lipid vesicles, it may not insert into the bilaye r or alternatively that synaptosomal membranes have a high capacity for a-synuclein binding. These results appear to be in contrast with the r eport by Sharon a nd colleagues, who s howed that WT a-synuclein decreased the fluidity of whole cell mem- brane preparations [44]. The differences between the two studies may be accounted for by either the presence of proteins or the combination of plasma, nuclear, endosomal, lysosomal and Golgi membranes in this preparation. Our experiments address only synaptosomal membrane inter- actions and therefore represent a small population within the whole cell membrane preparation. In contrast, A53T mutant a-synucle in significantly decreased internal bilayer fluidity of synaptosomal vesicles as demonstrated by the elevated anisotropy constant (Fig. 4 ). Increasing A53T concentration further decreased synaptosomal membrane fluidity. These results suggest that the A 53T muta nt a-synuclein either inserts directly into the fatty acyl chains or that synaptosomes h ave a low capacity for A53T mutant binding due to subtle changes in structure. The substitution of Ala53 fi Thr of a-synuclein is predicted to partially disrupt the N-terminal a-helix and extend the Fig. 3. Tyrosine fluorescence was used to determine the extent of a-synuclein aggregation in the presence of synaptosomal membranes. WT and A53T a-synuclein or lysozyme were incubated in the presence of synaptosomal membranes f or 2 (solid bars), 48 (hatched bars) and 96 h ( open ba rs). The e xtent o f aggr egat ion was det er mined using a ratio of tyrosine fl uorescence before and after centrifugation. The ratio of tyrosine fluorescence after immediate mixing was set at 100% an d all other conditions were normalized to this value. The re su lts are the mean ± SEM for t hree independent e xperiments. 3184 E. Jo et al.(Eur. J. Biochem. 271) Ó FEBS 2004 b-sheet structure of the central hydrophobic domain [15,45]. The decreased membrane fluidity as a result of the A53T substitution m ay be explained by the fact that lipid bilaye rs can readily accommodate a-helices but are disrupted by b-structured transmembrane features. The specificity of a-synuclein-synaptosomal membrane preparation was further probed by examining the effects of lysozyme on synaptosomal membrane fluidity (Fig. 4). This amyloid- forming peptide increases the fluidity of these v esicles i n a concentration dependent manner suggesting that peptide sequence i s important for specific membrane perturbations. Dynamics of lipid headgroups and interface In order to further characterize the differences in WT and A53T-synpatosomal membrane binding, the dynamics of the polar headgroups and the polarity of the lipid interface were analyzed as a measure of protein–membrane inter- actions. To obtain f urther insight into t he mechanism of a-synuclein–bilayer interactions, laurdan fluorescence spectroscopy was utilized. L aurdan’s naphthalene ring is located at the glycerol backbone and is a nchored in the bilayer by t he lauroyl moiety, thereby imparting fluorescent characteristics that are dependent o n the polarity of its environment [35,46]. The advantage of laurdan over other fluorescent probes is that it is nonfluorescent in aqueous media and is independent of pH and lipid headgroups; therefore fl uorescence only r eflects the polarity of the probe associated with the bilayer. The spectral p roperties of laurdan are described by the general polarization equation and render information about th e lipid phase, polarity and coexistence of multiple lipid phases within a membrane [35,46]. Laurdan excitation a nd emission spectra in the presence of synaptosomal membrane bilayers demonstrate t he char- acteristic red excitation at 340 nm and blue excitation at 380 n m, whereas the emission spectra indicate a single maximum at 430 nm indicative of the blue emission (Fig. 5 A). The intensity of the red excitation b and correlates with a polar environment o r strong hydrogen bon ding Fig. 5. Laurdan emission and excitation spectra of synaptosomal membranes in the presence of a-synuclein. Vesicles alone (ÆÆÆÆ) and after addition of WT (––) and A53T ( ) a-synuclein show similar o verall spectral characteristics ( A). T he dec reased intensity as a result of the presence of a-synuclein demonstrates the decrease in polarity of the headgroup–fatty acyl chain interface. The excitation and emission generalized polarization of laurdan in the presence of both WT and A53T a-synuclein are not depe ndent on wavelength (B). Generalized polarization values were calculated from excitation and emission scans before (ÆÆÆÆ) a nd after addition o f WT (––) a nd A53T ( ). Fig. 4. The effect of a-synuclein on membrane fluidity o f synaptosomal lipid bilayers as determined by DPH anisotropy. The addition of 5 and 10 lgofWTa-synuclein to synaptosomal membrane vesicles did n o t affect membrane flu idity. Addition o f A53T m utant a-synuclein significantly decreased t he membrane fluidity in a concentration dependent manner. In contrast, lysozym e increased membrane fluidity in a concentratio n dependent manner. D ata represent the mea n ± SD of three independent experiments. Student t-test indicates *P <0.01, P < 0.001 compared to lipid alone. Ó FEBS 2004 Lipid-induced a-synuclein fibrillogenesis (Eur. J. Biochem. 271) 3185 which occur in gel phase lipid bilayers where little relaxation occurs. The ratio of the blue to red components of the excitation scan generates d ata o n the polarity of the probe. The addition of both WT a nd A53T a-synuclein to laurdan containing synaptosomal membranes g enerates a b lue-shift in the excitation curve and decreases the intensity of laurdan fluorescence (Fig. 5A). Furthermore, the blue : red excita- tion ratio changes from 1.00 for synaptosomal membranes to 1 .02 and 1.04 after addition of A53T and WT a-synuc- lein, respectively. The blue-shift in t he excitation scan represents a less polar environment, suggestive of d ecreased H 2 O molecules and increased packing of lipid molecules. Furthermore, the general polarization emission (GP em )is unaffected by addition of A53T a-synuclein confirming that the micropolarity or hydration of the interfacial region of the lipid bilayer is unchanged, whereas WT binding increases the GP em indicating an altered packing of the interfacial region. Finally, the wavelength dependence of the general polarization lends eviden ce for the phase behaviour of bilayers in the presence and absence of proteins. The GP ex and GP em of synaptosomal membranes are independent of wavelength in the p resence of b oth WT and A53T a-synuclein indicating that the lipids do not undergo protein-induced phase change (Fig. 5B). These results suggest that binding of WT and A53T a-synuclein to synaptosomal membranes disrupts the inter facial region of the lipid molecule to varying extents but both can be easily accommodated within the lipid structure. Lipid headgroup packing and surface properties To distinguish a-synuc lein surface binding from insertion into the bilayer, we examined the lipid headgroup spacing using merocyanine 540 (MC540) absorption spectral prop- erties. The unique spectral properties result from binding of monomeric MC540 and subsequent dimerization, which are both dependen t on lipid headgroup packing and fluidity [36]. MC540 spectra in the presence of synaptosomal lipid vesicles are characteristic of a fluid headgroup packing w ith characteristic maxima at 530 and 570 nm (Fig. 6). The fluid headgroup space allows for extensive monomeric MC540 insertion as i ndicated by the predominance of the 5 70 nm maxima. Addition of WT a-synuclein increased the intensity of both maxima, indicating a slightly more fluid environ- ment and an increased headgroup space or the presence of packing defects (Fig. 6). The M C540 spectra are consistent with WT a-synuclein–synaptosomal interactions occurring predominantly at t he headgroup space. These data correlate well with our anisotropy studies, which demonstrate no change in the fatty acyl chain mobility as a result of WT a-synuclein binding and our electron microscopy data, which showed enhanced fibre f ormation after lipid binding of WT a-synuclein. Predominant headgroup b inding would position WT a-synuclein in an ideal location to act as a seed for fibril formation. In contrast, a ddition of A53T mutant a-synuclein significantly decreased the intensity o f the MC540 spectra indicating increased packing of the lipid headgroups and d ecreased accessibility f or MC540 binding (Fig. 6). These results are consistent with varying levels of A53T insertion into the synapt osomal bilayer, which a ffect both the headgroup and fatty acyl chain mobility. One interpretation of these results is that A53T mutant a-synuclein may insert into the bilayer to a greater degree than WT a-synuclein such that the hydrophobic domain is buried within the bilayer. Masking of the hydrophobic, b- sheet promoting region would effectively inhibit a-synuclein self-assembly into fibrils. The MC540 monomer-dimer equilibrium is relevant to the packing properties of the bilayer a nd can be used as a n indication of lipid headgroup spacing [36]. W e have calculated the apparent dimerization constant, K dapp ,for MC540 in synaptosomal lipid bilayers in the presence and absence o f W T and A53T mutant a-synuclein (Table 1). The K dapp of synaptosomal membranes was not altered significantly after binding WT a-synuclein suggesting that the membrane can easily accommodate WT a-synuclein. Addition of A53T a-syn uclein inc reased the Kdapp by 10-fold indicating that dimerization of MC540 within the headgroup space was d ecreased. These results support the notion that A53T mutant a-synuclein interactions with synaptosomal membranes organize the headgroup packing and ultimately the bilayer fluidity, thereby increasing membrane rigidity. Discussion Overall, our data suggest that WT a-synuclein binds predominantly to the headgroups of physiologically Fig. 6. The interaction of a-synuclein with the lipid headgroups of syn- aptosomal membranes was examined using MC540 absorption spectro- scopy. MC540 spectra demonstrate the fluid packing of the synaptosomal membrane h eadgroups (––). Addition o f W T a-synuc- lein ( ) resulted i n an increase i n the intensity of t h e MC540 spectra indicative of a-synuclein–headgroup interactions. In c ontrast, A53T a-synucle in decreased the in tensity of the sp ectra indicating increased packing o f the headgroups (ÆÆÆÆ). Table 1. Effect of a-synuclein mutations on the apparent dimerization constant (K dapp ) o f merocyanine 540 in synaptosomal m embranes. Peptides were added to lipid vesicles at a 1 : 20 ratio with a final peptide concentration of 6.9 lm. Sample Apparent dimerization constant Synaptosomes 1.67 · 10 5 WT a-synuclein 2.17 · 10 5 A53T mutant 1.54 · 10 6 3186 E. Jo et al.(Eur. J. Biochem. 271) Ó FEBS 2004 relevant lipid mixtures causing p acking defects in the headgroup space. The molecule penetrates into the interfa- cial lipid space as evidenced by the decrease in H 2 0 content and increased order of this region, but does not penetrate to the fatty acyl chains as no change in fluidity was detected. Predominant s urface binding of WT to synaptosomal membranes m ay help to explain the reversible lipid binding function of a-synuclein as this would create the least amount of disturbance in overall membrane structure [19,47]. In contrast, A53T binding causes increased lipid head- group packing, sub tle changes in the lipid interfacial space and a d ecrease in t he fluidity o f the fatty a cyl chains. These results are consistent with insertion of A53T into the bilayer. Our data suggest that a mutation linked t o familial early onset PD af fects no t on ly the s elf-asse mbly of a-synuclein but also the interaction with lipid bilayers. These results have implications for development of PD pathology, such as Lewy bodies, and extend our understanding of the effect of mutations in a-synuclein that may result in early onset forms of the disease. Lewy bodies are composed of a-synuclein, lipids and ubiquitin [48]. Immunohistochemical analyses have demon strated that lipids are distributed diffusely in homogenous Lewy bodies or are highly localized to the periphery of concentric Lewy bodies. It h as been proposed that lipids m ay either facilitate the incorporation of a-synuclein or influence a-synuclein fibril elo ngation [48]. Our r esults demonstrate that WT a-synuclein binding to synaptosomal membranes not only enhances fibril forma- tion but also propagates fibril growth along the bilayer surface. These results suggest that heterogenous seeding of a-synuclein fibrillogenesis m ay be one mechanism by which Lewy body formation progresses in PD. These results are consistent with previous reports that have shown seeding of WT cytosolic a-synuclein with me mbrane-bound a-synuc- lein [39]. Furthermore, synaptosomal membranes have a high capacity to bind WT a-synuclein, raising the possibility that control of membrane to cytosol distribution of a-synuclein may be important in nerve terminals. L ipid loading of primary neuronal cells demonstrated the redis- tribution of WT a-synuclein from cytosol t o the surface of lipid droplets, resulted in the prevention o f triglyercides hydrolysis [21]. Our results are in contrast to single lipid environments of acidic phospholipids, which have demon- strated decreased fibre formation in the presence of phosphatic acid and phosphatidylglycerol [49]. The differ- ence may be explained by the presence of a full repertoire of phospholipids, gangliosides, cholesterol and sphingomyelin, affecting not only overall membrane fluidity but also the surface charge of the lipid bilayer. In contrast to WT a-synuclein, A53T mutant a-synuclein binding to synaptosomal membranes decreases fibril for- mation, which seems to contradict in vitro self assembly models that have shown e nhanced fibrillogenesis of t he A53T mutant in comparison to WT a-synuclein [25,50–52]. However, these p revious in v itro fibrillogenesis studies were performed in t he absence of membranes. Our data suggest that the effects of the A53T mut ant a-synuclein may be elicited by altering the ratio of membrane-bound to cytosolic a-synuc lein. If decreased synaptosomal membrane fluidity resulting from A53T mutant a-synuclein binding further i nhibits a-synuclein–membrane interactions, t hen the relative a-synuclein concentration in the cytosol would be elevated and self-assembly may dominate. These results are consistent with the hypothesis t hat molecular crowding within the cytoplasm may contribute to amyloid-related disorders [53,54]. Furthermore, d ecreased m embrane fluidity will also affect normal cellular function and specifically synaptic signalling. In conclusio n, the equilib- rium between membrane-bound and cytosolic a-synuclein may be crucial for physiological function such that any significant shift in the equilibrium due to missense mutations or changes in membrane fluidity may c ause abnormal protein aggregation and Lewy body formation. Acknowledgements The authors would l ike to thank the Electron M icroscopy Suite at the University of Toronto for use of Hitachi 7000 electron microscope (CIHR M aintenance Grant). This work was supported by the Canadian Institutes of Health Research (J. M., P. E. F., P. H.), the Natural Sciences and Engineering Research C ouncil of Canada (J. M.), Ontario M ental Health Foundation (P.E.F.)andtheScottishRite Charitable Foundation (P. E. F., J. M.). The authors acknowledge support from the Ontario Alzheimer’s Asso ciation. J. M. was the Year 2001 Young Investigator Fund Scholarship recipient. E. J. was the recipient of a Postdoctoral Fellowship from the P arkin son’s Founda- tion of Canada. References 1. Goedert,M.,Jakes,R.,Crowther,R.A.,Hasegawa,M.,Smith, M.J. & Spillantini, M.G. (1998) Intraneuronal filamentous tau protein and alpha-synuclein deposits in neurodegenerative dis- eases. Biochem. Soc. Trans. 26, 463–471. 2. Mezey, E., Dehejia, A.M., Harta, G., Tresser, N., Suchy, S.F., Nussbaum, R.L., Brownstein, M.J. & Polymeropoulos, M .H. (1998) Alpha synuclein is pres entinLewybodiesinsporadic Parkinson’s d isease. Mol. Psyc hiatry 3, 493–499. 3. Spillantini, M.G., Schmidt, M .L., Lee, V.M., Trojanowski, J.Q., Jakes, R. & Goedert, M. ( 1997) Alpha-synuclein i n Lewy bodies. Nature 388 , 839–840. 4. Trojanowski, J.Q. & Lee, V.M. (1998) Aggregation of n eurofila- ment and alpha-synuclein proteins in Lewy bodies: implications for the pathogenesis of Parkinson disease a nd Lewy body dementia. Arch. N eurol. 55, 151 –152. 5. Wakabayashi, K., Matsumoto, K., Takayama, K., Yoshimoto, M. & Takahashi, H. (1997) N ACP, a presynaptic protein, immunoreactivity in L ewy bodies in Parkinson’s disease. Neurosci. Lett. 239, 45–48. 6. Polymeropoulos, M .H., Lavedan, C., L eroy, E., Ide, S.E., Dehejia, A., Dutra, A., Pike, B., Root, H., Rubenstein, J., Boyer, R., et al. (1997) Mutation in the alpha-synuclein gene identified in families with P arkinson’s disease. Science 27 6, 2045–2047. 7. Kruger, R., Kuhn, W., Muller, T., Woitalla, D., Graeber, M., Kosel, S., Przuntek, H., Epplen, J.T., Schols, L. & Riess, O. (1998) Ala30Pro mutation in the gene e ncoding a lpha-synuclein i n Par- kinson’s disease. Nat. Genet. 18, 106–108. 8. Singelton, A.B., Farrer, M., Johnson, J., Singleton, A., H ague, S., Kachergus, J., Hulihan, M., Peuralinna, T., Dutra, A., Nussbaum, R., et al. (2003) a-Synuclein locus triplication c auses Parkinson’s Disease. Science 302 , 841. 9. Arai, T ., Ueda, K., Ikeda, K., A kiyama, H., Haga, C., Kondo, H., Kuroki, N., Niizato, K., Iritani, S. & Tsuchiya, K. (1999) Argyrophilic glial inclusions in the midbrain of patients with Parkinson’s disease and diffuse Lewy b ody disease are immuno- positive for N ACP/a-synuclein Neurosci. Lett. 259, 83–86. Ó FEBS 2004 Lipid-induced a-synuclein fibrillogenesis (Eur. J. Biochem. 271) 3187 10. Arima, K., Ueda, K., Sunohara, N., Arakawa, K., Hirai, S., Nakamura, M., Tonozuka-Uehara, H. & Kawai, M. (1998) NACP/alpha-synuclein immunoreactivity in fibrillary components of neuronal and oligodendroglial cytoplasmic inclusions in the pontine nuclei in multiple system atrophy. Acta Neuropathol. (Berl) 96, 4 39–444. 11. Spillantini, M.G., Crowther, R.A., Jakes, R., Cairns, N.J., Lan- tos, P.L. & Goedert, M. (1998) Filamentous a-synuclein inclu- sions link multiple s ystem atrophy w ith P arkinson’s disease and dementia with L e wy bodies Neurosci. L ett. 251, 205– 208. 12. Tu, P.H., Galvin, J.E., Baba, M., Giasson, B., Tomita, T., Leight, S.,Nakajo,S.,Iwatsubo,T.,Trojanowski,J.Q.&Lee,V.M. (1998) Glial cytoplasmic inclusions in white matter oligoden- drocytes of multiple system atrophy brains contain insoluble alpha-synuclein. Ann. Neurol. 44, 415–422. 13. Wakabayashi, K., Hayashi, S., Kakita, A., Yamada, M., Toyoshima, Y., Yoshimoto, M . & Takahashi, H. (1998) Accu- mulation of alpha-synuclein/NACP is a cytopathological fe ature common to Lewy body disease a nd multiple system atrophy. Acta Neuropathol. ( Berl) 96, 4 45–452. 14.Wakabayashi,K.,Hayashi,S.,Yoshimoto,M.,Kudo,H.& Takahashi, H. (2000) NACP/alpha-synuclein-positive filamentous inclusions in astrocytes and oligodendrocytes o f Parkinson’s dis- ease brains. Ac ta Neuropathol. (Be rl) 99, 14–20. 15. Clayton, D.F. & George, J.M. (1998) The synucleins: a family of proteins invo lve d in synaptic fu nction, plasticity, neurodegenera- tion and d isease. Trends N eurosci. 21, 2 49–254. 16. Clayton, D.F. & George, J.M. (1999) Synucleins in synaptic plasticity and neurodegenerative disorders. J. Neurosci. Res. 58, 120–129. 17. Goedert, M. (1997) Familial Parkinson’s disease. The awakening of alpha-synuclein. Nature 388, 232–233. 18. Davidson, W .S., Jonas, A., Clayton, D.F. & Ge orge, J.M. (1998) Stabilization of a-synuclein secondary structure upon binding to synthetic membranes. J. Biol. Chem. 273, 9443–9449. 19. Jo,E.,McLaurin,J.,Yip,C.M.,St.GeorgeHyslop,P.&Fraser, P.E. (2000) a-Synuclein membrane interactions and lipid speci- ficity. J. Bio l. Chem. 27 5, 34328–34334. 20. Uversky, V.N., Lee, H J., Li, J., Fink, A.L. & Lee, S J. (2001) Stabilization of partially folded conformation during synuclein oligomerization in both purified and cytosolic preparations. J. Biol. Chem. 276, 43495–43498. 21. Cole, N .B., Murphy, D.D., Grider, T., Rue ter, S., Brasaemle, D. & Nussbaum, R.L. (2002) Lipid droplet binding and oligomer- ization properties o f the Pa rkinson’s d isease protein a-synuclein. J. Biol. Chem. 277, 6344–6352. 22. Leng, Y., Chase, T.N. & B ennett, M.C. (2001) Muscarinic receptor stimulation i nduces t ranslocation o f an a-synuclein oligomer from plasma membrane to a light vesicle fraction in cytoplasm. J. Biol. Chem. 276, 28212–28218. 23. Eliezer, D., Kutluay, E., B ussell, R. & Browne, G. (2001) Con- formational propertie s of a-synuclein in its free and lipid-asso- ciated states. J. Mol. Biol. 307 , 1061–1073. 24. Perrin, R.J., Woods, W .S., Clayton, D.F. & George, J .M. (2000) Interaction of hum an-synuclein an d p arkinson’s disease v ariants with phospholipids. Structural an alysis using s ite-directed muta- genesis. J. Biol. Chem. 275, 34393–34398. 25. Conway, K.A., Lee, S J., Rochet, J C., D ing, T.T., Williamson, R.E. & Lansbury, P.T. J r (2000) Acceleration o f oligomerization, not fibrillization, is a shared property of both alpha-synuclein mutations linke d t o early-onset Parkinson’s d isease: i mplications for pathogenesis and therapy Proc. Natl Acad. Sci. USA 97, 571–576. 26. McLean, P.J., Kawamata, H., Ribich, S. & Hyman, B.T. (2000) Membrane association and protein conformation of a-synuclein in intact neurons. Effect of Parkins on’s disease-linked mutations. J. Biol. Chem. 275, 8812–8816. 27. Jenson, P.H., Nielsen, M.S., Jakes, R., Dotti, C.G. & Goedert, M. (1998) Binding of a-synuclein to brain vesicles is abolished by familial Parkinson’s disease m utation. J. Biol. C hem. 273, 26292– 26294. 28. Jo,E.,Fuller,N.,Rand,R.P.,St.George-Hyslop,P.&Fraser, P.E. (2002) D efective membrane interactions of familial Parkin- son’s disease mutant A30P alpha-synuclein. J. Mol. Biol. 315, 799–807. 29. Tandon, A., Bannykh, S., Kowalchyk, J.A., Banerje e, A., Martin, T.F.J. & Balch, W .E. ( 1998) Differential regulation o f exocytosis by calcium and CAPS in semi-intact s yn aptosom es. Neuron 21, 147–154. 30. Tandon, A., Tan, P.K., Bannykh, S., Banerjee , A. & Balch, W.E. (1998) N eurotransmitter release from semi-intact synaptosomes. Methods 16 , 198–203. 31. Bartlett, G.R. (1959) Phospholipid analyses. Bioche mistr y 342, 466–468. 32. Waschuk, S.A., Elton, E.A., Darabie, A.A., F raser, P.E. & McLaurin, J. (2001) Cellular membrane composition defines Ab–lipid interactions. J. Biol. Chem. 276, 33561–33356. 33. McLaurin, J., Franklin, T., Chakrabartty, A. & Fraser, P.E. (1998) Phosphatidylinositol and i nositol in volvement in Alzh ei- mer amyloid-b fibril growth and arrest. J. Mol. Biol. 278, 183– 194. 34. Van Blitterswijk, W.J., Hoeven, R.P. & Van Der Meer, B.W. (1981) Lipid structural order parameters ( reciprocal of fluidity) i n biomembranes derived f rom steady-state fluorescence polarization measurements. Biochim. Biophys. Acta 64 4, 323–332. 35. Parasassi, T., DeStasio, G., Ravagnan, G., Rusch, R.M. & Gratton, E. (1991) Quantitation of lipid phases in phospholipid vesicles by the generalized polarization of laurdan fluorescence. Biophys. J . 60, 179–189. 36. Luzardo, M.D.C., Bernik, D.L., P azos, I.F., Figueroa, S., Lanio, M.E., V erez, V. & Disalvo, E.A. (1999) Phase and surf ace prop- erties of lipid bilayers containing neoglycolipids. Arch. Biochem. Biophys. 363 , 81–90. 37. Bernik, D.L. & Disalvo, E.A. (1996) Gel state surface properties of phosphatidylcholine liposomes as measured with merocyanine 540. Chem. Phys Lipids 82, 1 11–123. 38. Narayanan, V. & Scarlata, S. (2001) Membrane binding and se lf association o f a-synuclein. Biochemistry 40, 9 927–9934. 39. Lee, H J., Choi, C. & Le e, S J. (2002) M em brane-b ound a-syn- uclein has a high aggregat ion propensity and the ability to s eed the a ggregation of the cytosolic form. J. Biol. Chem. 277, 671– 678. 40. Yip, C.M. & McLaurin, J. (2001) Amyloid-b peptide assembly: a critical step in fibrillogenesis a nd membrane disruption. Bio phys. J. 80, 1359–1371. 41. Ji, S .R., Wu, Y . & Sui, S.F. (2002) Cholesterol is an im portant factor affecting the membrane insertion of b-amyloid peptide (Ab1–40), which may potentially inhibit the fibril formation. J. Biol. Chem. 277, 6273–6279. 42. Conway, K.A., Harper, J.D. & Lansbury, P.T. (1998) Fibrils formed in vitro from a-synuclein and two mutant forms linked t o Parkinson’s disease are typical amyl oid. Biochemistry 39, 2552– 2563. 43. Giasson, B.I., Uryu, K., Trojanowski, J.Q. & Lee, V.M Y. (1999) Mutant an d wild type human a-synucleins a ssemble into elonga- ted fi laments w ith distinct morphologies in vitro. J. Biol. Chem. 274, 7619–7622. 44. Sharon, R., Bar-Joseph, I., Mirick, G.E., Serhan, C .N. & Selkoe, D.J. (2003) Altered fatty acid composition of Dopaminergic neurons expressing a-synuclein and human brains with a-syn u- cleinopathies. J. Biol. Chem. 278, 49874–49881. 3188 E. Jo et al.(Eur. J. Biochem. 271) Ó FEBS 2004 45. Bussell, R. & E liezer, D. (20 01) Residual structure and dy namics in P arkinson’s d isease-associated m utants of a-synuclein. J. Biol. Chem. 276, 45996–46003. 46. Bagatolli, L.A., P arasassi, T., Fidelio, G .D. & Gratton, E. (1990) A Model for the interaction of 6-laurolyl-2-(N,N-dimethylamino) naphthalene with lipid environments: implications for sp ectral properties. Photochem. Photobiol. 70, 557 –564. 47. Bussell, R. & Eliezer, D. (2003) A structural and functional role for11-merrepeatsina-synuclein and other exchangeable lipid binding proteins. J. M ol. Biol. 329, 763–778. 48. Gai, W.P., Yuan, H.X., Li, X.Q., Power, J.T.H., Blumbers, P.C. & Jensen, P.H. (2000) In situ and in vitro study of colocalization and segregation of alpha-synuclein, ubiquitin, and lipids in Lewy bodies. Exp. Neurol. 166, 324–333. 49. Zhu, M. & Fink, A.L. (2003) Lipid binding inhibits a-synuclein fibril formation. J. Biol. Chem. 278, 16873–16877. 50. Conway, K.A., Harper, J.D. & Lansbury, P.T. (1998) Accelerated in vitro fibril formation by a mutant alpha-synuclein linked to early-onset Parkinson di sease. Natur e Med. 4, 1 318–1320. 51. Conway, K.A., Harper, J.D. & Lansbury, P.T. Jr (2000) Fibrils formed in vitro from alpha-synuclein and two mutant forms linked to Parkinson’s disease are typical amyloid. B ioche mi stry 39, 2552–2563. 52. Narhi, L., Wood, S.J., Steavenson, S., Jiang, Y., Wu, G.M., Anafi, D. et al. (1999) Both familial Parkinson’s disease mutations accelerate a-synuclein aggregation. J. Biol. Chem. 274, 9843– 9846. 53. Yang,D.S.,Yip,C.M.,Huang,T.H.,Chakrabartty,A.&Fraser, P.E. (1999 ) Manipulating t he amyloid-aggregation p athway wi th chemical chaper one. J. Biol. Chem. 274, 3 2970–32974. 54. Shtilerman, M.D., Ding, T.T. & Lansbury, P.T. (2002) Molecular crowding accelerates fibrillization of a-synuclein: could an increase in the c ytoplasmic p rotein c on centration i nduc e Park inson’s dis- ease? Biochemistry 41, 3855–3860. Ó FEBS 2004 Lipid-induced a-synuclein fibrillogenesis (Eur. J. Biochem. 271) 3189 . a-Synuclein–synaptosomal membrane interactions Implications for fibrillogenesis Euijung Jo 1 , Audrey A. Darabie 1 , Kyung Han 1 ,. to distinguish subtle differences between m utant a nd WT a-synuclein interactions and to further elucidate potential roles for a-synuclein–lipid interactions in vivo.Inorderto examine differences and explain. anchor for site-directed fibril assembly. Preced- ence for this mechanism of amyloid formation has been Fig. 2. Negative stain electron microscopy of a-synuclein in the presence of synaptosomal membranes.