Ion channel activity of brain abundant protein BASP1 in planar lipid bilayers Olga S. Ostroumova 1 , Ludmila V. Schagina 1 , Mark I. Mosevitsky 2 and Vladislav V. Zakharov 2 1 Laboratory of Ionic Channels of Cell Membranes, Institute of Cytology of RAS, St Petersburg, Russia 2 Division of Molecular and Radiation Biophysics, Petersburg Nuclear Physics Institute of RAS, Gatchina, Leningrad District, Russia Introduction Under the current concept, many neurodegenerative diseases (e.g. Alzheimer’s, Parkinson’s, Huntington’s diseases, etc.) are associated with the abnormal aggre- gation of amyloid proteins. Soluble oligomers of these proteins are now considered to be the main neurotoxic species that contribute to disease-associated neurode- generation. The precise mechanism of amyloid protein toxicity is not well understood. A large body of evi- dence suggests that amyloid protein oligomers can exert their toxicity by indirect mechanisms, through binding to various receptors, existing ion channels or other cellular proteins (as described for amyloid b-protein [1,2]). Alternatively, the amyloid oligomers can cause membrane permeabilization, which directly Keywords amyloid proteins; BASP1; ion channels; lipid bilayer; protein oligomers Correspondence V. V. Zakharov, Division of Molecular and Radiation Biophysics, Petersburg Nuclear Physics Institute of Russian Academy of Sciences, 188300 Gatchina, Leningrad District, Russia Fax: +7 813 7132303 Tel: +7 812 3282802 E-mail: vlad.v.zakharov@mail.ru (Received 23 September 2010, revised 14 November 2010, accepted 18 November 2010) doi:10.1111/j.1742-4658.2010.07967.x BASP1 (also known as CAP-23 and NAP-22) is a brain abundant myri- stoylated protein localized at the inner surface of the presynaptic plasma membrane. Emerging evidence suggests that BASP1 is critically involved in various cellular processes, in particular, in the accumulation of phosphati- dylinositol-4,5-diphosphate (PIP 2 ) in lipid raft microdomains. We have recently shown that BASP1 forms heterogeneously-sized oligomers and higher aggregates with an outward similarity to oligomers and protofibrils of amyloid proteins. However, BASP1 is not known to be related to any amyloid disease. In the present study, we show that BASP1 induces single channel currents across negatively-charged planar lipid bilayers (containing phosphatidylserine or PIP 2 ) bathed in 0.1–0.2 M KCl (pH 7.5). By their characteristics, BASP1 channels are similar to amyloid protein channels. BASP1 channels exhibit multiple conductance levels, in the range 10–3000 pS, with the most frequently observed conductance state of approximately 50 pS. The channels demonstrate a linear current–voltage relationship and voltage-independent kinetics of opening and closing. Their K + to Cl ) permeability ratio is approximately 14, indicating that BASP1 channels are cation-selective. The ion channel activity of BASP1 is in accordance with the pore-like structure of BASP1 oligomers observed by electron microscopy on a lipid monolayer. Neuronal protein GAP-43, which is functionally related to BASP1 and also forms oligomers, elicited no ion channel currents under the conditions used in the present study. Elucidation of the physiological or pathological roles of ion channel activ- ity of membrane-bound BASP1 oligomers will help to define the precise mechanism of amyloid protein toxicity. Abbreviations BASP1, brain acid-soluble protein-1; GAP-43, axonal growth-associated protein-43; PC, phosphatidylcholine; PE, phosphatidylethanolamine; PIP 2 , phosphatidylinositol-4,5-diphosphate; PS, phosphatidylserine; TEM, transmission electron microscopy. FEBS Journal 278 (2011) 461–469 ª 2010 The Authors Journal compilation ª 2010 FEBS 461 perturbs cellular calcium ion homeostasis. This fact was demonstrated both in lipid bilayer and cell mem- brane experiments [3,4]. Many amyloid proteins, such as amyloid b-protein, islet amyloid polypeptide, poly- glutamine and a-synuclein, have been reported to form ion channels in lipid membranes [5–9]. It has been also suggested that amyloid oligomers can increase mem- brane permeability by channel-independent perturba- tion of lipid bilayers, allowing nonspecific membrane leakage of larger molecules [10–13]. In support of the channel hypothesis, atomic force microscopy of amy- loid oligomers on supported lipid bilayers demon- strates the presence of pore-like (annular) structures [8,14–16]. In addition, electron microscopy of amyloid protein preparations has also revealed the presence of heterogeneous annular oligomers [17–20]. The size diversity of amyloid oligomers probably contributes to the observed heterogeneity of amyloid protein channels with respect to their kinetic behaviour, conductance and ion selectivity [21,22]. BASP1 (brain acid-soluble protein-1; also known as cortical cytoskeleton-associated protein CAP-23 and neuronal tissue-enriched acidic protein NAP-22) is an abundant 22–25 kDa protein in neuronal axon termi- nals, where it resides at the inner surface of the plasma membrane, predominantly in lipid rafts [23,24]. With respect to certain biochemical and func- tional properties, it is similar to axonal growth- associated protein GAP-43 [25,26]. Although GAP-43 is almost exclusively neuron-specific, BASP1 was also detected in considerable amounts in other cell types and tissues [23,27]. Knocking out the BASP1 gene in mice results in severe abnormalities in the nervous system and a high incidence (approximately 90%) of early postnatal lethality, indicating that the protein is critically involved in normal physiological processes [25]. Like GAP-43, BASP1 participates in neurite growth, axon guidance and regeneration [25,28,29]. The molecular mechanism of BASP1 activity presum- ably involves the regulation of actin cytoskeleton dynamics through either direct interaction with actin- binding proteins [30] or modulation of the plasmalemmal distribution of phosphatidylinositol-4,5-diphosphate (PIP 2 ) [31,32]. Recently obtained data indicate that BASP1 may also act as a potential tumor suppressor in non-neuronal cells, where it inhibits transcriptional activity of WT1 and Myc proteins [33,34]. BASP1 and GAP-43 are not related to any amyloid disease. However, recently, we have shown that GAP- 43 and particularly BASP1 form oligomers resembling the oligomers of amyloid proteins by their annular appearance in electron micrographs [35]. In addition, BASP1 can also form linear rod-like aggregates reminiscent of protofibrils of amyloid proteins. In the present study, taking into account an outward similar- ity of BASP1 to amyloid oligomers, we explored whether BASP1 oligomers can induce ionic currents across planar lipid bilayers. Results BASP1 exhibits single channel-like activity on planar lipid bilayers The activity of BASP1 and GAP-43 proteins in nega- tively-charged lipid bilayers under near-physiological conditions (0.1–0.2 m KCl, pH 7.5) was examined using the electrophysiological recording technique. The addi- tion of BASP1 to the solution in the cis-compartment resulted in the spontaneous incorporation of BASP1 molecules into the planar bilayer, which was accom- panied by the appearance of single channel-like currents (Fig. 1A). Figure 2 shows representative exam- ples of current fluctuations of the single BASP1 chan- nels in phosphatidylcholine (PC) ⁄ phosphatidylserine (PS) (50 : 50 mol %), PC ⁄ PIP 2 (90 : 10 mol %) and phosphatidylethanolamine (PE) ⁄ PS (50 : 50 mol %) bilayers at the most frequently observed conductance level. The burst activity is a prominent characteristic of this conductance level, regardless of the bilayer compo- sition, whereas the dwell time distributions of BASP1 channels were slightly different in the bilayers used (Fig. 2). The channels demonstrate slower kinetics in PC ⁄ PIP 2 (90 : 10 mol %) membranes (mean life time is approximately 2.5-fold larger) compared to PC ⁄ PS (50 : 50 mol %) and PE ⁄ PS (50 : 50 mol %) membranes. A B Fig. 1. Study of ion channel activity of BASP1 and GAP-43 in the PC ⁄ PS (50 : 50 mol %) bilayer. (A) BASP1 induces discrete single- channel current fluctuations. (B) GAP-43 has no channel-like activ- ity. Membrane-bathing solutions were 0.2 M KCl, 10 mM Hepes- KOH (pH 7.5) in both compartments. The current records are obtained at 50 mV transmembrane potential. The start of the records corresponds to 50 min after the addition of protein to the cis-compartment solution. Ion channel activity of BASP1 O. S. Ostroumova et al. 462 FEBS Journal 278 (2011) 461–469 ª 2010 The Authors Journal compilation ª 2010 FEBS At the most frequently observed conductance level (shown in Fig. 2), BASP1 channels generate a linear current–voltage relationship with a slope of 50 ± 3 pS, irrespective of the bilayer composition (Fig. 3). The frequency of channel appearance was signifi- cantly dependent on the bilayer composition. BASP1 ion channel activity was only rarely observed in PE ⁄ PS bilayers (two of eight trials), whereas 100% of trials with PC ⁄ PS and PC⁄ PIP 2 bilayers (27 and 13 trials, respectively) resulted in appearance of ion channel cur- rents. A trial was considered negative if no discrete membrane conductance changes were observed within 180 min (starting from the protein addition). The latency time before the appearance of ion channel activity was highly variable (in the range 30–120 min) and decreased with increasing protein concentration. Less latency time (approximately 20 min) was observed when BASP1 was pre-incubated (for 30 min) with PC ⁄ PS (50 : 50 mol %) liposomes, which were used as 0.2 0.4 0.6 Relative frequency Time (s) Time (s) Time (s) t t = 0.29 ± 0.03 s t t = 0.82 ± 0.24 s t t = 0.33 ± 0.03 s 0.2 0.4 0.6 Relative frequency 0.2 0.4 0.6 Relative frequency 0.0 0.1 0.2 Relative frequency Current (pA) Current (pA) Current (pA) I = –4.6 ± 0.5 pA I = –1.9 ± 0.3 pA I = –5.2 ± 0.6 pA 0.0 0.2 0.4 Relative frequency 5 pA 5 s –100 mV 4 pA A B C 5 s –75 mV 2 pA 5 s –50 mV 0.0 0.5 1.0 1.5 0246 0123 –6 –5 –4 –3 –7 –6 –5 –4 –3 –3 –2 –1 0.0 0.1 0.2 Relative frequency Fig. 2. Single ion channel activity of BASP1 in planar lipid bilayers of different composition: (A) PC ⁄ PS (50 : 50 mol %); (B) PC ⁄ PIP 2 (90 : 10 mol %); (C) PE ⁄ PS (50 : 50 mol %). Experimental conditions were the same as described in Fig. 1. The applied transmembrane potential is indicated above the current records. The dotted line corresponds to the closed state of the channel (0 pA). Current amplitude his- tograms (central panels) and channel dwell time histograms (right panels) for the corresponding bilayer compositions are shown. The mean ± SD current, I, and dwell time, s, were evaluated by Gaussian and single-exponential fit, respectively (with significance at P < 0.05 by the chi-square test). The total numbers of events used for the analysis were 50–150 and 50–400, respectively. –200 –100 100 200 –10 –5 5 10 V (mV) I (pA) Fig. 3. Amplitude of the BASP1-induced single-channel currents, I, plotted as a function of transmembrane potential, V. Bilayer compositions were: squares, PC ⁄ PS (50 : 50 mol %); triangles, PC ⁄ PIP 2 (90 : 10 mol %); circles, PE ⁄ PS (50 : 50 mol %). Experi- mental conditions were the same as described in Fig. 1. Each point of the I–V plot represents the mean ± SD of at least ten current readings. The line shows a linear regression fit of the data points. O. S. Ostroumova et al. Ion channel activity of BASP1 FEBS Journal 278 (2011) 461–469 ª 2010 The Authors Journal compilation ª 2010 FEBS 463 carriers to incorporate the protein into PC ⁄ PS (50 : 50 mol %) bilayer. Under these conditions, a ser- ies of stepwise current changes was observed, which probably corresponds to successive fusions of lipo- somes (containing active BASP1 channels) with the bilayer (Fig. 4A). In three of three trials, the ion channel activity of BASP1 in PC ⁄ cholesterol ⁄ PIP 2 (63 : 32 : 5 mol %) bilayers was similar to that observed in the absence of cholesterol. The inclusion of cholesterol into the bilayer did not significantly affect either the conductance of BASP1 channels or the latency time of their appearance (30–50 min). By contrast to BASP1, GAP-43 did not affected bilayer conductance under the con ditions used (Fig. 1B). Channel-like current fluctuations were not observed in either PC ⁄ PS (50 : 50 mol %), PE ⁄ PS (50 : 50 mol %), PC ⁄ PIP 2 (90 : 10 mol %) or PC ⁄ cholesterol⁄ PIP 2 (63 : 32 : 5 mol %) bilayers, regardless of the protein concentration and method of incorporation (addition of soluble GAP-43 or GAP-43-containing liposomes). Multiple conductance states of BASP1 channels Besides the most frequently observed conductance level of approximately 50 pS, BASP1 often induced the formation of ion channels with higher conductance. Figure 5 shows the spontaneous transitions of BASP1 channels between multiple conductance levels, when the protein was incorporated into the bilayer directly from the membrane-bathing solution. The conductance transitions were observed within the picosiemens range. As a rule, the higher conductance states were more long-lived than the 50 pS conductance state (Fig. 5). Their usual life time was from several seconds up to several minutes (Fig. 1A). Rarely, conductance levels of less than 50 pS or more than 1000 pS were also observed. Multichannel activity of BASP1 was studied by liposome-mediated protein delivery to the bilayer. Multichannel measurements produced linear current– A B 100 pA 10 s 0 pA 0 pA 50 pA 20 s –50 mV –25 mV –100 –50 50 100 –900 –600 –300 300 600 900 I (pA) V (mV) Fig. 4. Multichannel activity induced by BASP1 incorporated into PC ⁄ PS (50 : 50 mol %) lipid bilayer by fusion of BASP1-containing liposomes. Membrane-bathing solutions were 0.1 M KCl, 1 mM CaCl 2 ,5mM Hepes-KOH (pH 7.5). (A) Current records obtained at the transmembrane potential of )50 mV (upper) and )25 mV (lower). (B) I–V plots corresponded to different time intervals after liposome addition: squares, 30 min; circles, 45 min; triangles, 145 min. The lines show linear regression fits of three data series with a slope (conductance) of approximately 200 pS, 1 nS and 10 nS, respectively. The larger conductance corresponds to a larger number of BASP1 channels incorporated by liposome fusion. The data in each series were measured within 20 s. 100 mV 5 pA ABC 5 s ~ 500 pS ~ 100 pS 15 pA 3 s 50 mV ~ 50 pS ~ 300 pS 200 mV ~ 50 pS ~ 200 pS 10 pA 5 s Fig. 5. Multiple conductance levels induced by direct incorporation of BASP1 into planar lipid bilayers of different composition: (A) PC ⁄ PS (50 : 50 mol %); (B) PC ⁄ PIP 2 (90 : 10 mol %); (C) PE ⁄ PS (50 : 50 mol %). Experimental conditions were the same as described in Fig. 1. Applied transmembrane potential is indicated above the current records. The dotted line corresponds to the closed state of the channel (0 pA). Conductance values are indicated adjacent to the corresponding current jumps. Ion channel activity of BASP1 O. S. Ostroumova et al. 464 FEBS Journal 278 (2011) 461–469 ª 2010 The Authors Journal compilation ª 2010 FEBS voltage relationships, indicating that all conductance levels exhibit ohmic behaviour (Fig. 4B). Cation-selective behaviour of BASP1 channels To determine the charge selectivity of BASP1 channels, we studied the protein-induced currents in the asym- metrical 0.5 ⁄ 0.05 m KCl solution system (Fig. 6A). At zero membrane potential, a net negative current (approximately ) 1.3 pA) was measured (Fig. 6B), which corresponded to a positive charge moving from the trans- to the cis-compartment. This result demonstrates that the bulk of the current is carried by potassium ions. The reversal potential (V-intercept in Fig. 6B) was estimated to be approximately 50 mV. Similar values of the reversal potential (45 ± 5 mV) were obtained for seven bilayers, with the overall con- ductance in the range 10–100 pS. Using the Goldman– Hodgkin–Katz equation [36], we calculated that the permeability ratio P K ⁄ P Cl is approximately 14, whereas the transport number for K + ions (i.e. the fraction of current carried by K + ions) is equal to approximately 0.93. These results indicate that BASP1 channels are cation-selective. BASP1 channels are voltage-independent Using multichannel measurements, we investigated the effect of transmembrane voltage on gating of BASP1 channels. Ion channel voltage gating may be character- ized by an effective gating charge [37]. Switching the applied voltage polarity (± 50 mV) did not signifi- cantly change the absolute value of BASP1-induced steady-state transmembrane current. A two-fold reduc- tion of the applied voltage (from )50 to )25 mV) decreased the current in the same manner (Fig. 7A). Using the steady-state current measurements, we deter- mined the effective gating charge of BASP1 channels as being approximately 0 (Fig. 7B). Consequently, the channels are not voltage-gated. Taking into account the linear current–voltage relationships of BASP1 channels (Fig. 4B), these data clearly indicate that the channels are voltage-independent. Electron microscopy of BASP1 oligomers on a lipid monolayer Using transmission electron microscopy (TEM), we pre- viously demonstrated that BASP1 oligomers adsorbed on the nitrocellulose support film predominantly have an annular structure with a central pore (Fig. 8A) [35]. The size of annular BASP1 oligomers was confirmed as being very heterogeneous, with a diameter in the range 10–25 nm. To examine the structure of BASP1 oligomers under conditions close to those employed in the present study, we used the lipid monolayer tech- nique. Figure 8B shows that BASP1 oligomers formed on PC ⁄ PS (50 : 50 mol %) monolayer patches have a similar pore-like appearance. The mean ± SD diame- ter of the oligomers was 17.2 ± 3.8 nm (n = 52). Discussion In the present study, we have shown for the first time that the brain abundant protein BASP1 forms ion channels in a manner similar to that observed for amy- loid proteins. In the present experiments, we used neg- atively-charged lipid bilayers, which were previously found to induce the formation of BASP1 and GAP-43 oligomers [35] and stimulate the activity of amyloid protein channels [5–7,9,38]. The BASP1-induced A B –300 –200 –100 100 200 –9 –6 –3 3 6 V (mV) I (pA) –55 mV 50 mV 170 mV –280 mV 45 mV 55 mV 20 mV 95 mV 0 pA 3 pA 10 s Fig. 6. Cation-selectivity of BASP1 channels. (A) BASP1-induced current recorded in asymmetric system at different transmembrane potentials. Lipid bilayer composition was PC ⁄ cholesterol ⁄ PIP 2 (63 : 32 : 5 mol %). Membrane-bathing solutions: in the cis- compartment, 0.5 M KCl, 5 mM Hepes-KOH (pH 7.5); in the trans- compartment, 0.05 M KCl, 5 mM Hepes-KOH (pH 7.5). BASP1 was added to the solution in the cis-compartment. Values of transmem- brane potential are indicated. (B) I–V plot corresponded to the current record shown in (A). O. S. Ostroumova et al. Ion channel activity of BASP1 FEBS Journal 278 (2011) 461–469 ª 2010 The Authors Journal compilation ª 2010 FEBS 465 increase in membrane conductance occurred in the form of discrete unitary conductance changes with evi- dence of open and closed states that are characteristic of ion channels. It is notable that cholesterol, which is a main component of lipid rafts, had no noticeable effect on the BASP1 channel activity. Similarly, BASP1 oligomerization induced by negatively-charged liposomes was also unaffected by the addition of cholesterol [35]. BASP1 channels show cation selectiv- ity, generate linear voltage–current relationships and appear to be voltage-independent. Similar characteris- tics were reported for ion channels formed by amyloid b-protein and prion protein fragment 106–126 [5,21,39,40]. In addition, similar to most amyloid pro- tein channels, BASP1 channels exhibit spontaneous transitions between multiple conductance states (chan- nel subtypes), including those with sufficiently high conductance (up to several thousands of picosiemens) [7,9,21,39–41]. The conductance levels of BASP1 channels are highly heterogeneous. Some levels are characterized by burst activity (such as the 50 pS conductance level), whereas others are long-lived. Similar to amyloid protein channels, BASP1 channels have irregular latency of appearance [21]. The latency may be explained by the time-consuming step of BASP1 oligomerization into an active channel structure. In sup- port of this suggestion, liposome-mediated incorpora- tion of preformed BASP1 oligomers into the lipid bilayer resulted in a smaller latency time. We have shown previously that BASP1 is predominantly mono- meric in solution but forms oligomers in the presence of liposomes composed of anionic lipids [35]. The forma- tion of an ion-conducting channel probably proceeds through the association of several BASP1 monomers on the surface of the anionic lipid bilayer. The BASP1 molecule has a net negative charge, while its N-terminus is positively charged. Electrostatic binding of the N-terminal domains to anionic lipids probably provides the proper conformation and orientation of the protein molecules allowing them to self-associate. Using electron microscopy, we demonstrated that BASP1 oligomers predominantly have an annular structure with a central pore. This was confirmed for oligomers formed on the nitrocellulose support film, as well as on the PC ⁄ PS lipid monolayer. We suggest that the pore-like structure of BASP1 oligomers is responsi- ble for the generation of ion-permeable channels in lipid bilayers. The pore-like BASP1 oligomers are very heterogeneous in diameter (10–25 nm), which can account for the multiple conductance states of BASP1 channels. Different conductance levels are probably related to different numbers of the protein monomers in the oligomeric channel structure. A similar explana- tion was proposed for multiple single-channel conduc- tances of amyloid protein channels [14,16,21,42]. In the presence of GAP-43, no ion channels were observed under the conditions used. It should be noted that, in contrast to BASP1, GAP-43 preferably forms discoid rather than annular oligomers [35]. The most intriguing question concerns the physio- logical or pathological role of BASP1 channels. First, –25 mV –50 mV 0 pA 100 pA 5 min 50 mV A B –2 0 2 4 8 12 VF/RT ln (G, pS) Fig. 7. Voltage-independent behavior of BASP1 channels. (A) Records of BASP1-induced multichannel activity at different trans- membrane potentials. Experimental conditions were the same as described in Fig. 4. Arrows indicate voltage steps. (B) A semiloga- rithmic plot of BASP1-induced membrane conductance, G,asa function of the applied voltage, V, at steady-state conditions for two independent experiments. Taking into account the linear cur- rent–voltage relationships of BASP1 channels (Figs 3 and 4B), the average number of open channels under steady-state conditions, N ch , is proportional to steady-state membrane conductance, G. The effective gating charge, Q, that characterizes the effect of electric field on the conformational equilibrium of the channels can be cal- culated by the formula: Q = d(lnN ch ) ⁄ d(VF ⁄ RT ) [46], where R, T and F represent the universal gas constant, the absolute tempera- ture and the Faraday constant, respectively. The linear regression fit of the plotted data gives Q = 0.081 ± 0.037 (squares) and Q = 0.004 ± 0.046 (circles). Ion channel activity of BASP1 O. S. Ostroumova et al. 466 FEBS Journal 278 (2011) 461–469 ª 2010 The Authors Journal compilation ª 2010 FEBS BASP1 oligomer formation may be related to a certain pathology at which the oligomers exhibit toxicity as a result of their ion channel activity. However, BASP1 is not known to be associated with any amyloid disease. Recent data have indicated that, in neurons, BASP1 forms large oligomeric complexes, which can be revealed by glutaraldehyde cross-linking of proteins in plasma membrane rafts (V. V. Zakharov & O. S. Vity- uk, unpublished results). This suggests that the mem- brane-bound oligomers probably represent a functional form of BASP1. Recently, the nonpathological protein stefin B (in amyloid-like aggregates and in the native form) was shown to also form ion channels in planar lipid bilayers [43]. It has been suggested that the ion channel activity of stefin B may be a physiological function of the protein. Whether or not raft-associated BASP1 oligomers act as ion channels in vivo (e.g. as a kind of leak channels) remains unclear. Further experiments are needed to address this possibility. The studies of ion channel activity of nonpathological amyloid-like proteins probably will help to define the precise mechanism of amyloid protein toxicity. Materials and methods BASP1 and GAP-43 proteins were isolated from bovine brain as described previously [44]. The proteins were purified to homogeneity using a combination of preparative polyacrylamide gel electrophoresis and reversed-phase HPLC [35]. Protein stock solutions (0.5–3.5 mgÆmL )1 ) in water were stored at )20 °C in aliquots. Water was distilled twice and deionized. PC (1,2-dioleoyl-sn-glycero- 3-phosphocholine), PS (1,2-dioleoyl-sn-glycero-3-phospho- serine), PE (1,2-dioleoyl-sn-glycero-3-phosphoethanolamine) and porcine brain PIP 2 were purchased from Avanti Polar Lipids (Alabaster, AL, USA). Cholesterol was obtained from Sigma-Aldrich (St Louis, MO, USA). Solvent-free planar lipid bilayers were formed according to the lipid monolayer opposition technique [45] on a 50 lm diameter aperture in the 10 lm thick Teflon film separating two compartments (cis and trans) of the Teflon chamber. The aperture was pre-treated with hexadecane. Under symmetric conditions, an electric potential was applied to the bilayer with a pair of Ag ⁄ AgCl electrodes via 2 m KCl, 1.5% agarose bridges. Under asymmetric con- ditions (0.5 ⁄ 0.05 m KCl), Ag ⁄ AgCl electrodes were used. In the latter case, the difference in the two half-cell potentials (measured to be approximately 53 mV) was subtracted from the measured potential to obtain the transmembrane potential. The trans-compartment was held at virtual ground. Negative current represents the flow of cations from trans-tocis-side. Proteins were added to the mem- brane-bathing solution in the cis-compartment to a final concentration of 2–10 lm. In some experiments, BASP1 and GAP-43 were preliminary incorporated into PC ⁄ PS (50 : 50 mol %) liposomes (protein ⁄ lipid molar ratio of 1 : 50, lipid concentration of 2 mm) by a method described previously [35]. 5 lL of the protein ⁄ liposome preparation was added to the membrane-bathing solution in the cis- compartment (1.5 mL). To facilitate fusion of the liposomes with the bilayer, CaCl 2 was added to the solutions in both compartments up to 1 mm. Transmembrane currents were recorded with a custom made amplifier and digitized with a pclamp-compatible board. Data collection was performed with a 2 kHz sampling frequency and low-pass filtering at 200 Hz. Data were analyzed using pclamp, version 9.0 (Axon Instruments, Foster City, CA, USA) and Origin, version 7.0 (OriginLab Corp., Northampton, MA, USA). All experiments were carried out at room temperature (21 ± 2 °C). TEM specimens were prepared using the lipid monolayer technique. 1 lL of 0.1 mgÆmL )1 lipid solution (PC ⁄ PS, 50 : 50 mol %) in chloroform ⁄ methanol (19 : 1) was spread on 60 lL of the buffer (10 mm sodium phosphate, pH 7.3, 0.1 m NaCl) in a Teflon well (4 mm in diameter) and allowed to incubate in a humid chamber for 1 h at room temperature. Formvar ⁄ carbon-coated copper grid (200 mesh) (SPI Supplies, West Chester, PA, USA) was placed onto the top of the buffer in the well for 1 min. Then the grid with adsorbed lipid monolayer was transferred onto a 10 lL drop of BASP1 solution (0.1 mgÆmL )1 ) in the same AB Fig. 8. TEM micrographs of BASP1 oligomers stained with uranyl acetate. (A) BASP1 oligomers adsorbed on the nitrocellulose support film. (B) BASP1 oligomers adsorbed on the lipid monolayer composed of PC and PS (50 : 50 mol %). Dark areas correspond to the lipid mono- layer patches on the Formvar ⁄ carbon support film. Black objects represent electron-dense material. Scale bars = 50 nm. O. S. Ostroumova et al. Ion channel activity of BASP1 FEBS Journal 278 (2011) 461–469 ª 2010 The Authors Journal compilation ª 2010 FEBS 467 buffer and exposed for 20 min. Subsequently, the grid was successively rinsed on three drops of water (for 30 s on each) and stained with 1% uranyl acetate for 30 s. The excess fluid was drained off with filter paper, and the grid was air dried. The specimens of BASP1 oligomers on the nitrocellulose support film were prepared as described pre- viously [35]. Acknowledgements This work was supported by the Russian Foundation for Basic Research (grants 08-04-00432, 09-04-00883), the Molecular and Cell Biology Program of the Presid- ium of the Russian Academy of Sciences, the President of Russian Federation (grant SS-3796.2010.4) and the Russian State Contract P1372 (MES, FTP ‘‘SSEPIR’’). References 1 Cappai R & Barnham KJ (2008) Delineating the mech- anism of Alzheimer’s disease A beta peptide neurotoxic- ity. Neurochem Res 33, 526–532. 2 Sakono M & Zako T (2010) Amyloid oligomers: forma- tion and toxicity of Abeta oligomers. FEBS J 277, 1348–1358. 3 Kagan BL, Azimov R & Azimova R (2004) Amyloid peptide channels. J Membr Biol 202, 1–10. 4 Glabe CG (2006) Common mechanisms of amyloid oli- gomer pathogenesis in degenerative disease. Neurobiol Aging 27, 570–575. 5 Arispe N, Rojas E & Pollard HB (1993) Alzheimer disease amyloid beta protein forms calcium channels in bilayer membranes: blockade by tromethamine and aluminum. Proc Natl Acad Sci USA 90, 567–571. 6 Mirzabekov TA, Lin MC & Kagan BL (1996) Pore formation by the cytotoxic islet amyloid peptide amylin. J Biol Chem 271, 1988–1992. 7 Hirakura Y, Azimov R, Azimova R & Kagan BL (2000) Polyglutamine-induced ion channels: a possible mechanism for the neurotoxicity of Huntington and other CAG repeat diseases. J Neurosci Res 60, 490–494. 8 Quist A, Doudevski I, Lin H, Azimova R, Ng D, Frangione B, Kagan B, Ghiso J & Lal R (2005) Amyloid ion channels: a common structural link for protein-misfolding disease. Proc Natl Acad Sci USA 102, 10427–10432. 9 Zakharov SD, Hulleman JD, Dutseva EA, Antonenko YN, Rochet JC & Cramer WA (2007) Helical alpha-synuclein forms highly conductive ion channels. Biochemistry 46, 14369–14379. 10 Kayed R, Sokolov Y, Edmonds B, McIntire TM, Milton SC, Hall JE & Glabe CG (2004) Permeabilization of lipid bilayers is a common conformation-dependent activity of soluble amyloid oligomers in protein misfolding diseases. J Biol Chem 279, 46363–46366. 11 Green JD, Kreplak L, Goldsbury C, Li Blatter X, Stolz M, Cooper GS, Seelig A, Kistler J & Aebi U (2004) Atomic force microscopy reveals defects within mica supported lipid bilayers induced by the amyloidogenic human amylin peptide. J Mol Biol 342, 877–887. 12 Sokolov Y, Kozak JA, Kayed R, Chanturiya A, Glabe C & Hall JE (2006) Soluble amyloid oligomers increase bilayer conductance by altering dielectric structure. J Gen Physiol 128, 637–647. 13 Valincius G, Heinrich F, Budvytyte R, Vanderah DJ, McGillivray DJ, Sokolov Y, Hall JE & Losche M (2008) Soluble amyloid beta-oligomers affect dielectric membrane properties by bilayer insertion and domain formation: implications for cell toxicity. Biophys J 95, 4845–4861. 14 Lin H, Bhatia R & Lal R (2001) Amyloid beta protein forms ion channels: implications for Alzheimer’s disease pathophysiology. FASEB J 15, 2433–2444. 15 Lal R, Lin H & Quist AP (2007) Amyloid beta ion channel: 3D structure and relevance to amyloid channel paradigm. Biochim Biophys Acta 1768, 1966–1975. 16 Jang H, Arce FT, Capone R, Ramachandran S, Lal R & Nussinov R (2009) Misfolded amyloid ion channels present mobile beta-sheet subunits in contrast to con- ventional ion channels. Biophys J 97, 3029–3037. 17 Lashuel HA, Hartley D, Petre BM, Walz T & Lansbury PT Jr (2002) Neurodegenerative disease: amyloid pores from pathogenic mutations. Nature 418, 291. 18 Lashuel HA, Hartley DM, Petre BM, Wall JS, Simon MN, Walz T & Lansbury PT Jr (2003) Mixtures of wild-type and a pathogenic (E22G) form of Abeta40 in vitro accumulate protofibrils, including amyloid pores. J Mol Biol 332, 795–808. 19 Diociaiuti M, Polzi LZ, Valvo L, Malchiodi-Albedi F, Bombelli C & Gaudiano MC (2006) Calcitonin forms oligomeric pore-like structures in lipid membranes. Biophys J 91, 2275–2281. 20 Kayed R, Pensalfini A, Margol L, Sokolov Y, Sarsoza F, Head E, Hall J & Glabe C (2009) Annular protofibrils are a structurally and functionally distinct type of amyloid oligomer. J Biol Chem 284, 4230–4237. 21 Hirakura Y, Lin MC & Kagan BL (1999) Alzheimer amyloid abeta1-42 channels: effects of solvent, pH, and Congo Red. J Neurosci Res 57, 458–466. 22 Kourie JI, Culverson AL, Farrelly PV, Henry CL & Laohachai KN (2002) Heterogeneous amyloid-formed ion channels as a common cytotoxic mechanism: implications for therapeutic strategies against amyloido- sis. Cell Biochem Biophys 36, 191–207. 23 Mosevitsky MI, Capony JP, Skladchikova G, Novitskaya VA, Plekhanov A & Zakharov VV (1997) The BASP1 family of myristoylated proteins abundant Ion channel activity of BASP1 O. S. Ostroumova et al. 468 FEBS Journal 278 (2011) 461–469 ª 2010 The Authors Journal compilation ª 2010 FEBS in axonal termini. Primary structure analysis and physico-chemical properties. Biochimie 79, 373–384. 24 Maekawa S, Sato C, Kitajima K, Funatsu N, Kumanogoh H & Sokawa Y (1999) Cholesterol-depen- dent localization of NAP-22 on a neuronal membrane microdomain (raft). J Biol Chem 274, 21369–21374. 25 Frey D, Laux T, Xu L, Schneider C & Caroni P (2000) Shared and unique roles of CAP23 and GAP43 in actin regulation, neurite outgrowth, and anatomical plasticity. J Cell Biol 149, 1443–1454. 26 Mosevitsky MI (2005) Nerve ending ‘signal’ proteins GAP-43, MARCKS, and BASP1. Int Rev Cytol 245, 245–325. 27 Mosevitsky M & Silicheva I (2011) Subcellular and regional location of ‘brain’ proteins BASP1 and MARCKS in kidney and testis. Acta Histochem 113, 13–18. 28 Caroni P, Aigner L & Schneider C (1997) Intrinsic neuronal determinants locally regulate extrasynaptic and synaptic growth at the adult neuromuscular junction. J Cell Biol 136, 679–692. 29 Bomze HM, Bulsara KR, Iskandar BJ, Caroni P & Skene JH (2001) Spinal axon regeneration evoked by replacing two growth cone proteins in adult neurons. Nat Neurosci 4, 38–43. 30 Odagaki S, Kumanogoh H, Nakamura S & Maekawa S (2009) Biochemical interaction of an actin-capping protein, CapZ, with NAP-22. J Neurosci Res 87 , 1980– 1985. 31 Caroni P (2001) New EMBO members’ review: actin cytoskeleton regulation through modulation of PI(4,5)P(2) rafts. EMBO J 20, 4332–4336. 32 Epand RM, Vuong P, Yip CM, Maekawa S & Epand RF (2004) Cholesterol-dependent partitioning of PtdIns(4,5)P2 into membrane domains by the N-terminal fragment of NAP-22 (neuronal axonal myristoylated membrane protein of 22 kDa). Biochem J 379, 527–532. 33 Carpenter B, Hill KJ, Charalambous M, Wagner KJ, Lahiri D, James DI, Andersen JS, Schumacher V, Royer-Pokora B, Mann M et al. (2004) BASP1 is a transcriptional cosuppressor for the Wilms’ tumor suppressor protein WT1. Mol Cell Biol 24, 537–549. 34 Hartl M, Nist A, Khan MI, Valovka T & Bister K (2009) Inhibition of Myc-induced cell transformation by brain acid-soluble protein 1 (BASP1). Proc Natl Acad Sci USA 106, 5604–5609. 35 Zakharov VV & Mosevitsky MI (2010) Oligomeric structure of brain abundant proteins GAP-43 and BASP1. J Struct Biol 170, 470–483. 36 Hodgkin AL (1951) The ionic basis of electrical activity in nerves and muscle. Biol Rev 26, 339–409. 37 Hille B (1992) Ionic Channels of Excitable Membranes, 2nd edn. Sinauer Associates Inc., Sunderland, MA. 38 Alarcon JM, Brito JA, Hermosilla T, Atwater I, Mears D & Rojas E (2006) Ion channel formation by Alzheimer’s disease amyloid beta-peptide (Abeta40) in unilamellar liposomes is determined by anionic phospholipids. Peptides 27, 95–104. 39 Arispe N, Pollard HB & Rojas E (1993) Giant multilevel cation channels formed by Alzheimer disease amyloid beta-protein [A beta P-(1-40)] in bilayer membranes. Proc Natl Acad Sci USA 90, 10573– 10577. 40 Lin MC, Mirzabekov T & Kagan BL (1997) Channel formation by a neurotoxic prion protein fragment. J Biol Chem 272, 44–47. 41 Kourie JI & Culverson A (2000) Prion peptide fragment PrP[106-126] forms distinct cation channel types. J Neurosci Res 62, 120–133. 42 Mustata M, Capone R, Jang H, Arce FT, Ramachandran S, Lal R & Nussinov R (2009) K3 fragment of amyloidogenic beta(2)-microglobulin forms ion channels: implication for dialysis related amyloidosis. J Am Chem Soc 131, 14938–14945. 43 Rabzelj S, Viero G, Gutierrez-Aguirre I, Turk V, Dalla Serra M, Anderluh G & Zerovnik E (2008) Interaction with model membranes and pore formation by human stefin B: studying the native and prefibrillar states. FEBS J 275, 2455–2466. 44 Zakharov VV & Mosevitsky MI (2007) M-calpain- mediated cleavage of GAP-43 near Ser41 is negatively regulated by protein kinase C, calmodulin and calpain-inhibiting fragment GAP-43-3. J Neurochem 101, 1539–1551. 45 Montal M & Mueller P (1972) Formation of bimolecu- lar membranes from lipid monolayers and a study of their electrical properties. Proc Natl Acad Sci USA 69, 3561–3566. 46 Malev VV, Schagina LV, Gurnev PA, Takemoto JY, Nestorovich EM & Bezrukov SM (2002) Syringomycin E channel: a lipidic pore stabilized by lipopeptide. Biophys J 82, 1985–1994. O. S. Ostroumova et al. Ion channel activity of BASP1 FEBS Journal 278 (2011) 461–469 ª 2010 The Authors Journal compilation ª 2010 FEBS 469 . Ion channel activity of brain abundant protein BASP1 in planar lipid bilayers Olga S. Ostroumova 1 , Ludmila V. Schagina 1 , Mark I. Mosevitsky 2 and Vladislav V. Zakharov 2 1 Laboratory of Ionic. bilayers. Results BASP1 exhibits single channel- like activity on planar lipid bilayers The activity of BASP1 and GAP-43 proteins in nega- tively-charged lipid bilayers under near-physiological conditions. regeneration [25,28,29]. The molecular mechanism of BASP1 activity presum- ably involves the regulation of actin cytoskeleton dynamics through either direct interaction with actin- binding proteins