| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Department of Biochemistry and Molecular Biology, Indiana University School of Medicine, Indianapolis, Indiana 46202
Correspondence: Address reprint requests to Lili Guo, Dept. of Biochemistry and Molecular Biology, Indiana University School of Medicine, 635 Barnhill Dr., Indianapolis, IN 46202. Tel.: 317-274-7507; Fax: 317-274-4686; E-mail: lguo{at}iupui.edu.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
–3) will have far-reaching implications in developing therapeutics for different protein conformational diseases and also in the de novo design of ß-barrel proteins (4–8). In this context, recently we observed that a short ß- sheet peptide, (xSxG)6, (VSLGLSIGFSVGVSIGWSFGRSRG) forms voltage-gated, high-conductance multichannels closely resembling ß-barrel porins in lipid bilayers (9). A detailed study of the aggregation, channel activity, and peptide-lipid interactions indicated the insertion of the peptide as oligomers into the lipid bilayer followed by rearrangement into a complex ß- barrel-like pore structure comprising several individual ß-sheets or smaller ß-sheet aggregates (10). A dyad repeat sequence and the presence of glycine, serine, and hydrophobic residues similar to that found in ß-barrel proteins are considered as the driving force for porin-like channel formation by this peptide (10).
The unprecedented observation of the porin-like activity by a short ß-sheet peptide with a repeating sequence pattern could provide a useful platform for elucidating the effect of specific amino acid residues on folding and functional properties of ß-sheet peptides in membranes and structural engineering through amino acid replacements. In this work, we substituted four out of six glycine residues of (xSxG)6 with D-alanines (VSLGLSIAFSVAVSIAWSFARSRG, where A is D-alanine; abbreviated as (xSxA/G)6) to examine its effect on conformational and ion channel properties. Peptides having a regularly repeating pattern of L and D amino acids are known to adopt unique structures, especially in membranes (e.g., the single stranded head-to-head ß6-helical dimer structure of gramicidin A (gA)) (11–15). Here, we hypothesized that substitutions of small, flexible glycine residues with D-alanines in (xSxG)6 would lead to a dramatic change in its structural and functional properties. Thus the N-formyl analog of (xSxA/G)6 is likely to form a ß12-helical dimeric structure similar to that of gA as proposed by Kennedy et al. for a N-formyl-(LLLD)6 sequence (16,17). In a ß12-helical model, it is possible to have polar side chains (e.g., serine) inside the helical pore and hydrophobic side chains pointing outside, thereby satisfying the condition for an ideal ion channel inside a lipid bilayer (Fig. 1).
|
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Peptide synthesis
The (xSxG)6 and (xSxA/G)6 peptides were synthesized manually using standard solid phase Fmoc chemistry as explained earlier (9,10). The N-formylation was carried out with p-nitrophenyl formate in N-methyl pyrrolidone. The N-formyl and desformyl analogs of the above peptides are indicated as f- and df-, respectively. The peptides were purified by reverse phase HPLC (a Varian Prostar 210 system fitted with a Prostar 320 ultraviolet/visible detector) on a C4 column (Vydac; 5 µm particle size, 10 mm i.d., and 250 mm length). A methanol-water gradient (solution A: 100% H2O, 0.1% TFA; solution B: 100% CH3OH, 0.09% TFA; Program: B—70% (0 min)-70% (20 min)-90% (39 min)-90% (60 min)-95% (61 min)-100% (85 min)) was used for elution. The collected methanol-water fractions were lyophilized after removing methanol by rotary evaporation. The peptides were characterized by analytical HPLC and molecular weight analysis using a MALDI-TOF (matrix assisted laser desorption-time of flight) mass spectrometer (Micromass, Manchester, UK). The peptide purity was >95%.
Planar lipid bilayer studies
Planar lipid bilayers were formed by applying DPhPC solution (15 mg/mL in n-decane) over a 200 µm diameter aperture in the wall of a Delrin cup (18). Typically, 
3 µL of a methanolic solution of f-(xSxA/G)6 peptide (0.05 mg/mL) was added to both sides of the preformed bilayer membranes containing 100–1000 mM CsCl solution (unbuffered) to a final concentration of 10–8 M. A custom current/voltage conversion amplifier was used to optimize single-channel recording. The Ag/AgCl electrodes were connected to the cis and trans chambers through 1 M KCl-agar bridges (the side which was connected to the signal ground of the amplifier is defined as trans, and the opposite side to which lipid solution was smeared is defined as cis). Transmembrane currents through voltage-clamped bilayers were low-pass filtered at 1 kHz using an 8-pole Bessel filter (model BC-525C, Warner Instrument Co., Hamden, CA) followed by digitization at 0.5 kHz with a 16 bit data acquisition system (Digidata 1320 A, Axon Instruments, Foster City, CA). The data were acquired and analyzed using a computer and Clampex 8.2 software (Axon Instruments). Curve fits were done using the Marquardt least-squares algorithm.
Peptide-unilamellar lipid vesicles
The peptide-liposomes were prepared by a modified method of Wallace and Blout (19); 0.5 mg of peptide and 5 mg of DMPC (peptide/lipid molar ratio 1:36) were codissolved in 2 mL trifluoroethanol followed by removal of the solvent to get a peptide-lipid film. The lipid film was hydrated with 500 µL of D2O (or H2O) and incubated at 40°C for 2 h followed by sonication for 5 min using a Branson sonifier 450 (output control 4, duty cycle 30%, 35 watts) fitted with a contact tip accessory. Samples were centrifuged at 14,000 rpm for 10 min at room temperature. After these treatments a clear solution was obtained. Liposomes without peptides were also prepared as a control.
Circular dichroism spectroscopy
Circular dichroism (CD) measurements were carried out at 20°C on a JASCO (Easton, MD) J-720 spectrometer. The ellipticity was calibrated with (+)-10-camphorsulfonic acid. Each spectrum was an average of 6–30 scans using quartz cells having path lengths of 0.01 or 0.05 cm. Spectra were recorded in the range 260–180 nm at a speed of 50 nm/min with a resolution of 0.2 nm and are presented on a per amide basis (deg cm2 decimol–1). The mean residue molar ellipticity [
]MRW was calculated using the formula, []MRW = /10 x Cr x l, where MRW is the mean residue weight (molecular weight/number of residues) of the peptide, Cr is the mean residue molar concentration (Cr = n x Cp, where n is the number of peptide bonds and Cp is the molar concentration in mol/liter), l is the path length in cm, and is the ellipticity in mdeg.
IR spectroscopy
Attenuated total reflectance-Fourier transform infrared spectroscopy (ATR-FTIR) spectra were recorded on a Perkin Elmer 2000 spectrometer. Typically, aliquots of 50 µL peptide-liposome samples were loaded onto a ZnSe internal reflection element (IRE) and a hydrated thin film was formed by using a stream of nitrogen gas. The IRE was then placed in an ATR accessory, and FTIR spectra were collected. The background spectrum was collected using the same IRE before the sample measurements. Fourier deconvoluted and second derivative spectra were obtained using Grams analyst 2000 software. Resulting spectra were smoothed by the 7 point Savitsky-Golay function. Band fittings of the deconvoluted spectra were done using a Gaussian function, and starting band parameters were taken from deconvoluted spectra. The initial line widths were set to 12 cm–1 and allowed to vary freely. The relative areas of the components were expressed as a percentage of the area of each fitted region, which was equivalent to an area normalization of the amide bands.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
3 µL of a methanolic solution (0.05 mg/mL) of the peptide on both sides of the bilayer resulted in rapid transitions between the open and closed states and stable current levels typical of single channels. Similar current transitions were difficult to observe when the peptide was added only on one side of the bilayer. Typical channel traces obtained at different applied potentials are shown in Fig. 2 A. A linear current-voltage relationship was obtained under symmetrical ionic conditions (Fig. 2 B). The single-channel conductance as determined from the slope of the I/V curve was 49.6 pS with a 1 M CsCl (unbuffered) solution. The desformyl analog of (xSxA/G)6 failed to show any ion channel activity, whereas both the formyl and desformyl analogs of (xSxG)6 peptide formed high conductance channels under similar experimental conditions (9). The channel trace of f-(xSxG)6 obtained at an applied potential of 10 mV after the addition of 3 µL of a methanolic solution (0.05 mg/mL) of the peptide on one side of the bilayer (channels appeared immediately after the addition of the peptide on the cis side of the bilayer) is shown for comparison (Fig. 2 C). The single-channel conductance of f-(xSxG)6 (1370 pS) was 28 times the conductance of f-(xSxA/G)6 channels (compare the scale in Fig. 2, A and C). The f-(xSxG)6 channel was in the high conductance open state most of the time, and complete closures were rarely seen at low applied potentials. Moreover, several fast flickering type events along with short-lived subconductance states were seen in f-(xSxG)6 channels. On the contrary, the most important characteristics of f-(xSxA/G)6 channel were the absence of multichannels or subconductance states and the occurrence of complete closing events at different applied potentials.
|
4 pA in the x axis) representative of mainly two levels, a low conductance state and a high conductance state, were seen (Fig. 3 C). The ion channel data were further analyzed by generating open times distributions corresponding to the open conductance state using dwell-time histograms. The binned data were fitted exponentially using the Marquardt least-squares algorithm. The open time events in f-(xSxA/G)6 channel events could be best fitted with a single time constant (Fig. 3, D and E). However, in f-(xSxG)6 channels the openings to the large conductance state are best characterized with two time constants (
1 and 2) indicating the complex nature of the channel (Fig. 3 F).
|
40% of the total events represent the open state of the channel. The proportion of the high conductance channel component in f-(xSxG)6 channel was maximum (>80%) at low applied potentials (<±40 mV) and a sharp decrease in the high conductance open channel events was observed at higher potentials, showing a voltage-dependent behavior. A two state voltage gating of f-(xSxG)6 channels with a tendency to close both at positive and negative potentials at applied potentials >40 mV has already been reported (9).
|
).
|
NEt3Bz+, and glucosammonium, suggesting a pore diameter >10.5 Å corresponding to the diameter of the largest permeating cation (NEt3Bz+) obtained from Corey, Pauling, and Koltun (space-filling) models (9). A comparative account of single-channel conductance of f-(xSxG)6 and f-(xSxA/G)6 channels in the presence of different electrolytes under similar experimental conditions are given in Table 1. Among the organic cations tested, only the ammonium ion (4 Å) was permeable in the case of the f-(xSxA/G)6 channel, and measurable channel activity was not observed in the presence of electrolytes containing larger cations such as 
glucosammonium, and NEt3Bz+ with a diameter of 6, 9, 10, and 10.5 Å, respectively, indicating a relatively smaller pore compared to the f-(xSxG)6 channel. Based on these observations, we believe that the diameter of the narrowest portion of the f-(xSxA/G)6 channel should be 4 Å similar to gramicidin channel. The selectivity sequence in terms of conductances or conductance ratios of the f-(xSxA/G)6 channel for monovalent cations followed the order Cs > K> Na >Li. The conductance in CsCl is 3 times that in LiCl solution, which suggests that the channel favors the passage of Cs over Li cation. Although the radius of bare Cs ion (1.69 Å) is much larger than that of lithium (0.6 Å), the inner hydration shell of the former is small and is easily dehydrated. Therefore the process of cation passage through the peptide channel might be controlled predominantly by dehydration energies (20). Moreover, the overall ion selectivity of the channel could be influenced by other factors like the effective size of the pore and the electrostatic interaction between the ions and the pore interior.
|
) at various peptide concentrations (C). The data were plotted on a log-log scale (Fig. 6). From the obtained relationship log () = 1.7 log (C), the value of slope (1.7) gives an approximate estimate of the number of molecules per channel (21). This method would give only an approximate value of peptide units comprising the channel due to the difficulty in knowing the exact number of molecules moved into lipid bilayer at various peptide concentrations. Nevertheless, the value of 1.7 could be considered as a lower limit of peptide units required for channel formation.
|
,22). In methanol, the spectrum of f-(xSxA/G)6 displayed a maximum around 198 nm and a minimum around 219 nm, suggesting predominantly a ß-structure in solution.
|
). The arginine absorptions near 1635 cm–1 and 1673 cm–1 usually exhibit large shifts of –50 and –70 cm–1, respectively, upon hydrogen/deuterium exchange (23). This was confirmed by a comparative analysis of the secondary derivative spectra in water and D2O solutions (data not shown). The main feature of the amide 1 region of the spectra of f-(xSxA/G)6- and df-(xSxA/G)6-incorporated liposome samples was the occurrence of a strong band around 1627 cm–1 and another in the high frequency region around 1690 cm–1 (Fig. 8). The spectral bands could be resolved into at least five individual absorptions that are centered around 1692, 1684, 1669, 1652, and 1627 cm–1 through deconvolution (Table 2). The occurrence of a 1627 cm–1 band with a high percentage relative area (60%, see Table 2) suggests predominantly a ß-conformation in membranes (24). The high frequency band found in antiparallel ß-sheet peptides due to transition dipole coupling is usually very weak compared to the low frequency band. So, one of the weaker components (1692 cm–1) in the high frequency region (1660–1700 cm–1) could be ascribed to the antiparallel hydrogen bonding and the rest to turn structures (24). A broad band around 1530 cm–1 in the amide 2 region (peptide-liposome samples in H2O) further supported a ß-structure (data not shown).
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
). The formation of highly conductive channels (mean single-channel conductance = 1.37 nS), the two state gating of the channels at potentials >±40 mV, the observation of fast and slow kinetic events, and the existence of subconductance states demonstrate such a similarity (9). The permeation of a wide variety of ions of different sizes suggests the formation of nonselective channels with a pore diameter >10.5 Å (corresponding to the calculated diameter of the largest permeating cation, NEt3Bz+). Based on these results we concluded that resemblance of (xSxG)6 channels to that of porins could be due to the formation of a ß-barrel-like structure possibly through peptide aggregation (9). This was further supported by CD and infrared (IR) studies indicating a ß-sheet conformation in membranes stabilized by antiparallel hydrogen bonding and possibly some ß-turns (9). Furthermore, inhibition of (xSxG)6 channel by Congo red, oligomeric structures in association with lipid bilayers detected by electron microscopy, and gel electrophoresis of peptide-incorporated-liposomes indicating multimeric species suggests that (xSxG)6 channels are formed by peptide aggregation (10).
In this work we hypothesized that an (LLLD)6 pattern of sequence having a formyl group at the N-terminus would form a ß12-helical dimeric structure in membranes through hydrogen bonding between two ß-helical monomers via N-formyl groups as in gramicidin (11,17). Substantial variation in the ion channel properties of (xSxG)6 upon D-alanine substitutions strongly suggests the formation of a different type of structural species in membranes. Here, we tried to validate the proposed ß12-helical model for f-(xSxA/G)6 by combining several lines of independent structural and functional information. Most importantly, both the formyl and desformyl analogs of (xSxG)6 were membrane active, whereas the desformyl analog of (xSxA/G)6 did not form channels. Second, the formation of single channels without multiple states or subconductance states in the case of f-(xSxA/G)6 suggests that the channels do not have a complex structure. Usually multiple conductance states of complex channels result from the cooperation of several individual pore-forming subunits or the presence of distinct conformational species of the same protein (25,26). Hence, these two possibilities can be eliminated in the case of the f-(xSxA/G)6 channel. Moreover, low single-channel conductance (49.6 pS) and open channel events represented by a single time constant () also rule out the possibility of a complex channel. So, the possibility of a channel formed from a single molecule or a dimer could be envisaged. Another interesting difference between the f-(xSxG)6 and f-(xSxA/G)6 channels is the absence of a voltage gating behavior in the latter. The fraction of open channels of f-(xSxA/G)6 constitutes 30–40% of total channel events over a wide range of applied potentials. The presence of distinct open and closing events might represent an equilibrium between conducting and nonconducting forms of the channel. Moreover, if a channel is formed from the f-(xSxA/G)6 monomer one would expect that the channel would be open for a long period of time without several closing events and the fraction of open channel events would be high. However this assumption will be good only if no conformational changes of the peptide take place within the membrane. Based on the above discussion it is reasonable to believe that the f-(xSxA/G)6 channel is formed by the association of two monomeric units via N-formyl groups. The effect of peptide concentration on channel conductance also supports the idea of a peptide dimer as the possible channel species. We could not observe current transitions by adding f-(xSxA/G)6 just on one side of the membrane, suggesting that the f-(xSxA/G)6 molecule cannot readily cross the membrane and the channel is probably formed by dimerization, one molecule approaching from each side of the bilayer. The possibility of a channel comprising 
-helical or ß-sheet structure is also ruled out based on the following considerations. First, the -helical conformation would be less favored in the bilayer because serine side chains would be exposed to hydrocarbon tails in the interior of the membrane. Second, the pore formation by ß- sheet peptides is known to proceed via aggregation, resulting in complex multichannels having several subconductance states (9,21,25,27).
The examination of the pore size of the f-(xSxA/G)6 channel with a set of organic ammonium cations showed that the largest cation that is permeable through the channel is the ammonium cation, indicating that the diameter of the narrowest portion of the channel is 4 Å (28). In the proposed ß-helical model, the f-(xSxA/G)6 channel comprises three peripheral pores each about the same internal diameter of the gramicidin channel (4.5 Å). Or in other words, a ß12-helical structure is composed of three gramicidin-like pores. Space-filling molecular models showed that each pore has an internal diameter of 4.5 Å. Though the maximum diameter of the ß12 helix structure (from one end to other end of the ß12-helix) is expected to be three times that of gramicidin, the effective internal diameter for the passage of ions will be close to that of gramicidin as the channel interior is partly blocked by side chains of serine. This could explain the impermeability of membranes to larger cations such as 
and 
As explained earlier a salient feature of the proposed structure is that the serine hydroxyl groups lining the interior of the channel may solvate mobile ions, contributing to ion selectivity. The ionic selectivity of f-(xSxA/G)6 channels (pK+/pCl– = 6.9) is consistent with the prediction that ions traversing the channel would be coordinated by serine hydroxyl groups. The hydroxyl group is an amphivalent ligand and may coordinate either anions or cations. The majority of the current is carried by cations because first, the oxygen atom of the OH dipole is more negative than the hydrogen atom is positive (thus the situation HO–M+–OH is more favorable, energetically, than OH–X––HO) and second, because geometric constraints within the channel would not allow a favorable OH–X– bond when the anion is in one of the three peripheral pores of the channel. In addition, the walls of the channel are composed of peptide groups which can supply a part of the coordination sphere for the ions in transit, and the same argument about the relative strengths of the oxygen and hydrogen atom in the peptide group applies (29). For these reasons cations would be accommodated much more readily than anions in the three peripheral chambers of the channel.
Even though conformational studies using IR and CD spectral methods cannot ascertain the exact structure of a peptide in membranes, they can give valuable clues regarding the nature of hydrogen bonding and backbone structure. Distinct spectral features of f-(xSxA/G)6 in membranes compared to that of f-(xSxG)6 suggest a structural change upon incorporating D-alanines instead of glycine residues. The CD spectrum of f-(xSxG)6 had the characteristic feature of a ß-sheet protein with a maximum and minimum around 205 and 220 nm, respectively, in membranes (22,30). However, the CD spectrum of f-(xSxA/G)6 is different from that of typical ß-sheet proteins in having an additional negative band around 226 nm along with negative and positive bands around 215 and 200 nm, respectively. The presence of two minima is usually ascribed to -helical structures but at different positions (209 and 222 nm) than in f-(xSxA/G)6 peptide (30). Moreover, the IR study did not show any evidence of an -helical structure in membranes and indicated a ß-sheet-type hydrogen-bonding pattern (31). Considering the influence of one tryptophan residue present in f-(xSxA/G)6, the 226 nm band could also arise from changes in the tryptophan environment. For example, membrane-bound tryptophan-containing short peptides are reported to show a similar peak around 230 nm (1,32). A comparison of the CD spectra of f-(xSxA/G)6 in membrane and methanol solution seems to support the idea of an environment-sensitive dichroism of tryptophan residue due to the absence of a 226 nm band in the latter. However, the absence of a similar band in the spectrum of f-(xSxG)6 (which also has one tryptophan) in a membrane environment suggests that the 226 nm band arises more likely from a change in backbone conformation of the f-(xSxA/G)6 peptide and subsequent variation in the microenvironment of tryptophan residue upon D-alanine substitution. Taken together, the data indicate that f-(xSxA/G)6 has a different backbone structure compared to -helical and ß-sheet proteins and is stabilized by ß-sheet-type hydrogen bonding.
A comparison of conductance values of f-(xSxA/G)6 in this study and that of gA reported in the literature revealed that, interestingly, both peptides have almost the same single-channel conductance in DPhPC lipid bilayers (49.6 pS for f-(xSxA/G)6 and 47.2 for gA) in the presence of 1 M CsCl (33). However, it is known that the conductance of the gA channel arises in part from the presence of four tryptophan residues which alter the energy profile for ion permeation through long-range electrostatic interactions (34). For example, an 5-fold decrease in single-channel conductance of gA was observed upon substituting three tryptophan residues with phenylalanines (35). So it would be more appropriate to compare the single-channel conductance of f-(xSxA/G)6 (has only one tryptophan) with that of a gramicidin analog having only one tryptophan residue; an 9-fold difference in the single-channel conductance of f-(xSxA/G)6 (30 pS in 1 M NaCl) compared to the gA analog (3.4 pS in 1 M NaCl) having one tryptophan residue (35) implies that the presence of three peripheral pores with an overall diameter three times that of the gramicidin channel and the presence of serine hydroxyl groups which can coordinate permeating ions may be contributing to the higher conductivity of f-(xSxA/G)6 channel.
In summary, this work demonstrates a dramatic change in ion channel and conformational properties of a (xSxG)6 peptide upon replacing glycine residues with D-alanines. Significant changes in the structural and functional properties as a result of glycine to D-alanine substitution of (xSxG)6 suggest that D-amino acid recurrence in a peptide sequence overwhelmingly affect the backbone structure of the peptide in membranes. Moreover, glycines in (xSxG)6 seem to have a structural role in aggregation and in the formation of ß-barrel-like pore formation due to their flexibility, lack of side chains, and the ability to facilitate sheet-to-sheet packing interactions. Using a simple ß12-helical model, we proposed that the ion channel properties of D-alanine repeat (f-(xSxA/G)6) peptide could be easily explained and correlated with the structure. We hope that the success in correlating the functional properties of the channel with the proposed ß-helical model will stimulate the design of gramicidin-like ß-helical motifs with a variety of applications. Moreover, the transformation of a complex porin-like channel into a simple gramicidin-like channel through glycine to D-alanine substitutions provides an interesting possibility for the structural engineering of different pore-forming ß-sheet proteins in the design of novel macromolecular devices. With the ability to predict structural models of smaller proteins and to correlate with their functional properties, we are moving toward the ultimate goal of understanding the dependence of protein sequence on conformation, folding, and function.
|
ACKNOWLEDGEMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Submitted on August 16, 2005; accepted for publication October 19, 2005.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
2. Bishop, C. M., W. F. Walkenhorst, and W. C. Wimley. 2001. Folding of ß-sheets in membranes: specificity and promiscuity in peptide model systems. J. Mol. Biol. 309:975–988.[CrossRef][Medline]
3. Rausch, J. M., J. R. Marks, and W. C. Wimley. 2005. Rational combinatorial design of pore-forming ß-sheet peptides. Proc. Natl. Acad. Sci. USA. 102:10511–10515.
4. Trzesniewska, K., M. Brzyska, and D. Elbaum. 2004. Neurodegenerative aspects of protein aggregation. Acta Neurobiol. Exp. (Warsz.). 64:41–52.[Medline]
5. Arispe, N. 2004. Architecture of the Alzheimers AßP ion channel pore. J. Membr. Biol. 197:33–48.[CrossRef][Medline]
6. Lin, H., R. Bhatia, and R. Lal. 2001. Amyloid ß protein forms ion channels: implications for Alzheimer's disease pathophysiology. FASEB J. 15:2433–2444.
7. Quist, A., I. Doudevski, R. Azimova, D. Ng, B. Frangione, B. Kagan, J. Ghiso, and R. Lal. 2005. Amyloid ion channels: a common structural link for protein-misfolding disease. Proc. Natl. Acad. Sci. USA. 102:10427–10432.
8. Wimley, W. C. 2003. The versatile ß-barrel membrane protein. Curr. Opin. Struct. Biol. 13:404–411.[CrossRef][Medline]
9. Thundimadathil, J., R. W. Roeske, and L. Guo. 2005. A synthetic peptide forms voltage-gated porin-like ion channels in lipid bilayer membranes. Biochem. Biophys. Res. Commun. 330:585–590.[CrossRef][Medline]
10. Thundimadathil, J., R. W. Roeske, H.-Y. Jiang, and L. Guo. 2005. Aggregation and porin-like channel activity of a ß-sheet peptide. Biochemistry. 44:10259–10270.[CrossRef][Medline]
11. Ketchem, R. R., W. Hu, and T. A. Cross. 1993. High-resolution conformation of gramicidin A in a lipid bilayer by solid-state NMR. Science. 261:1457–1460.
12. Townsley, L. E., W. A. Tucker, S. Sham, and J. F. Hinton. 2001. Structures of gramicidins A, B, and C incorporated into sodium dodecyl sulfate micelles. Biochemistry. 40:11676–11686.[CrossRef][Medline]
13. Navarro, E., R. Tejero, E. Fenude, and B. Celda. 2001. Solution NMR structure of a D, L alternating oligonorleucine as a model of ß helix. Biopolymers. 59:110–119.[CrossRef][Medline]
14. Sanchez-Quesada, J., H. S. Kim, and M. R. Ghadiri. 2001. A synthetic pore-mediated transmembrane transport of glutamic acid. Angew. Chem. Int. Ed. Engl. 40:2503–2506.[CrossRef]
15. Cornelissen, J. J. L. M., J. J. J. M. Donnoers, R. D. Helder, W. S. Graswinckel, G. A. Metselaar, A. E. Rowan, N. A. J. M. Sommerdijk, and R. J. M. Nolte. 2001. ß-helical polymers from isocyanopeptides. Science. 293:676–678.
16. Kennedy, S. J., H. R. Besch, A. M. Watanabe, A. R. Freeman, and R. W. Roeske. 1977. Properties of beta-helical ion channels. Biophys. J. 17:A87–A87. (Abstr.)
17. Kennedy, S. J. 1978. Structures of membrane proteins. J. Membr. Biol. 42:265–279.[CrossRef][Medline]
18. Benz, R., K. Janko, W. Boos, and P. Lauger. 1978. Formation of large, ion-permeable membrane channels by the matrix protein (porin) of Escherichia coli. Biochim. Biophys. Acta. 511:305–319.[Medline]
19. Wallace, B. A., and E. R. Blout. 1979. Conformation of an oligopeptide in phospholipid vesicles. Proc. Natl. Acad. Sci. USA. 76:1775–1779.
20. Hille, B. 1992. Ion Channels of Excitable Membranes. Sinauer, Sunderland, MA.
21. Lin, M.-C. A., and B. L. Kagan. 2002. Electrophysiological properties of channels induced by Aß25–35 in planar lipid bilayers. Peptides. 23:1215–1228.[CrossRef][Medline]
22. Johnson, W. C. Jr. 1988. Secondary structure of proteins through circular-dichroism spectroscopy. Annu. Rev. Biophys. Biophys. Chem. 17:145–166.[CrossRef][Medline]
23. Chirgadze, Y. N., O. V. Fedorov, and N. P. Trushima. 1975. Estimation of amino acid residue side chain absorption in the infrared spectra of protein solutions in heavy water. Biopolymers. 14:679–694.[CrossRef][Medline]
24. Barth, A., and C. Zscherp. 2002. What vibrations tell us about proteins. Q. Rev. Biophys. 35:369–430.[CrossRef][Medline]
25. Hirakura, Y., M.-C. Lin, and B. L. Kagan. 1999. Alzheimer amyloid Aß 1–42 channels: effects of solvent, pH, and Congo red. J. Neurosci. Res. 57:458–466.[CrossRef][Medline]
26. Minn, A. J., P. Velez, S. L. Schendel, H. Liang, S. W. Muchmore, S. W. Fesik, M. Fill, and C. B. Thompson. 1997. Bcl-XL forms an ion channel in synthetic lipid membranes. Nature. 385:353–357.[CrossRef][Medline]
27. Tobkes, N., B. A. Wallace, and H. Bayley. 1985. Secondary structure and assembly mechanism of an oligomeric channel protein. Biochemistry. 24:1915–1920.[CrossRef][Medline]
28. Qi, Z., M. Sokabe, K. Donowaki, and H. Ishida. 1999. Structure-function study on a de novo synthetic hydrophobic ion channel. Biophys. J. 76:631–641.
29. Koeppe II, R. E., J.-L. Mazet, and O. S. Andersen. 1990. Distinction between dipolar and inductive effects in modulating the conductance of gramicidin channels. Biochemistry. 29:512–520.[CrossRef][Medline]
30. Sreerama, N., and R. W. Woody. 2004. Computation and analysis of protein circular dichroism spectra. Methods Enzymol. 383:318–351.[Medline]
31. Krimm, S., and J. Bandekar. 1986. Vibrational spectroscopy and conformation of peptides, polypeptides and proteins. Adv. Protein Chem. 38:181–364.[Medline]
32. Ladokhin, A. S., M. E. Selsted, and S. H. White. 1997. Bilayer interactions of indolicidin, a small antimicrobial peptide rich in tryptophan, proline, and basic amino acids. Biophys. J. 72:794–805.[Medline]
33. Mattice, G. L., and R. E. Koeppe II. 1995. Stabilizing effect of D-alanine in gramicidin channels. Biochemistry. 34:6827–6837.[CrossRef][Medline]
34. Fonseca, V., P. Daumas, L. Ranjalahy-Rasoloarijao, F. Heitz, R. Lazaro, Y. Trudelle, and O. S. Andersen. 1992. Gramicidin channels that have no tryptophan residues. Biochemistry. 31:5340–5350.[CrossRef][Medline]
35. Becker, M. D., D. V. Greathouse, R. E. Koeppe II, and O. S. Andersen. 1991. Amino acid sequence modulation of gramicidin channel function. Effects of tryptophan-to-phenylalanine substitutions on the single channel conductance and duration. Biochemistry. 30:8830–8839.[CrossRef][Medline]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) channels at different applied potentials obtained by Gaussian fitting of the all-point-histograms of channel data. In the case of f-(xSxA/G)6, the fraction of open channel events remained almost the same at different voltages, suggesting little voltage dependence on channel activity. The proportion of the high conductance channel component in f-(xSxG)6 channels is maximum (>80%) at low applied potentials (<±40 mV), and a sharp decrease in high conductance open channel events can be seen at higher potentials, showing a voltage-dependent behavior.
