| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
-Methylalanines with Leucines: A FTIR Spectroscopic Analysis and Correlation with Channel Kinetics
* School of Molecular Sciences, De Montfort University, Leicester LE1 9BH, United Kingdom; Unite Mixte de Recherche 5048 Centre National de la Recherche Scientifique-Universite de Montpellier 1, 34090 Montpellier, France; and Unite Mixte de Recherche 6026 Centre National de la Recherche Scientifique-Universite de Rennes I, Campus de Beaulieu, 35042 Rennes, France
Correspondence: Address reprint requests to Dr. Herve Duclohier, E-mail: herve.duclohier{at}univ-rennes1.fr.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
-methylalanine or Aib) for its ion channel activity remains unknown, although a previous study showed that replacement of all Aib residues with leucines preserved the essential channel behavior except for much faster single-channel events. To correlate these functional properties with structural data, here we compare the secondary structures of an alamethicin derivative where all the eight Aibs were replaced by leucines and the native alamethicin. Fourier transform infrared (FTIR) spectra of these peptides were recorded in methanol and in aqueous phospholipid membranes. Results obtained show a significant conformational change in alamethicin upon substitution of its Aib residues with Leu. The amide I band occurs at a lower frequency for the Leu-derivative indicating that its -helices are involved in stronger hydrogen-bonding. In addition, the structure of the Leu-derivative is quite sensitive to membrane fluidity changes. The amide I band shifts to higher frequencies when the lipids are in the fluid phase. This indicates either a decreased solvation due to a more complete peptide insertion or a peptide stretching to match the full thickness of the bilayer. These results contribute to explain the fast single-channel kinetics displayed by the Leu-derivative. |
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
; Sansom, 1993; Cafiso, 1994; Bechinger, 1997; Duclohier and Wroblewski, 2001). Alamethicin is a 20 residue-long natural polypeptide endowed with antibiotic activity, extracted from the fungus Trichoderma viride as a mixture of related compounds (Martin and Williams, 1976) of which the main component has the sequence shown in Fig. 1.
|
; Eisenberg et al., 1973) and in cell membranes (Sackmann and Boheim, 1979), as well as inducing membrane excitability (Mueller and Rudin, 1968; Duclohier and Spach, 2001). Alamethicin has recently been used in a tethered bilayer sensor with the potential application in drug discovery processes (Yin et al., 2003).
In addition to voltage-driven transmembrane incorporation of preformed interfacial aggregates (Schwarz et al., 1987; Vodyanoy et al., 1988), two main kinds of mechanisms have been proposed for the way in which this channel-forming peptide operates: 1), through conformational changes, and 2), the helix dipole models. For instance, Hall et al. (1984) proposed a conformational transition between a bent structure around Pro14 with a C-terminal ß-structure and a N-terminal -helix, and an all linear helix as the basic gating event, broadly in line with Fox and Richards' (1982) hypothesis based on crystal structure. In models stressing the role of the helix dipole moment, some of the inserted helical monomers flip across the membrane dielectric so that transmembrane antiparallel bundles turn parallel (Menestrina et al., 1986). Despite existing evidence from the literature that alamethicin monomers adopts a predominantly -helical secondary structure in membranes (Esposito et al., 1987; Cascio and Wallace, 1988; Haris and Chapman, 1988; Bak et al., 2001), the structure of the helical bundle in the lipid bilayer is not known in detail, especially as regards its open state. However, some modeling and simulations studies offer interesting clues regarding membrane insertion and bundle assembly (e.g., Tieleman et al., 1999).
It has been argued that alamethicin may not constitute an appropriate model peptide for membrane proteins because of its high content of Aibs, uncoded amino acids. However the fact that it is a hydrophobic peptide that is predominantly -helical and whose crystal structure and solution structure are known makes it a useful peptide for understanding membrane protein folding and aspects of ion channel function. As it would be useful to deal with an analog of alamethicin that did not contain any of the Aib residues, thus removing the ambiguity of a partial 310-helical character in alamethicin, a synthetic analog in which all the eight Aibs were replaced by leucine (designated alm-dUL, hereafter; see the amino-acid sequence in Fig. 1) was prepared. Its secondary structure is characterized here and compared to the native alamethicin using Fourier transform infrared (FTIR) spectroscopy to explain the functional modulation brought about by the Aib-Leu substitution.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
-DMPC (dimyristoyl-phosphatidylcholine) used in the FTIR study was purchased from Sigma Chemical. Lipids used for the single-channel conductance assay were palmitoyl-oleoyl phosphatidylcholine (POPC) and dioleoyl-phosphatidtylethanolamine (DOPE), both from Avanti Polar Lipids (Birmingham, AL). Purification of alamethicin and synthesis/purification of the alm-dUL analog.
The two main forms of alamethicin differ by the identity of residue at position 18: Glu or Gln for analogs referred as Rf30 and Rf50, respectively, the charged form being more abundant. In addition, minor fractions have an alanine in place of -aminoisobutyric acid (differing by a CH2) at position 6. The two main alamethicin forms used here were separated using reversed phase HPLC with a Brownlee Aquapore (Santa Clara, CA) RP-300 C18 column, and the analysis was performed in a similar manner to that described by Gisin et al. (1977). As the FTIR spectra of the two forms of alamethicin were found to be identical, most of the measurements were made on the mixture as purchased from Sigma. As for the Leu-derivative, Alm-dUL, it was synthesized on solid-phase resin and HPLC-purified as previously described (Molle et al., 1988).
Single-channel conductance assay in planar lipid bilayers
For comparing the single-channel activity displayed by native alamethicin and Alm-dUL, bilayers were formed at the tip of patch-clamp pipettes as already described (Coronado and Latorre, 1983). The same standard conditions (e.g., 1 M KCl) and lipid mixture (POPC/DOPE; 7:3) already used before in most of our comparative conductance assays of a set of alamethicin derivatives (e.g., Molle et al., 1996) were used here. With other workers in the field, we found that this lipid mixture represents a good compromise for bilayer stability, ease of peptide insertion, and channel behavior. Indeed, correlated with an increased tendency for nonlamellar lipid structure or curvature strain, the incorporation of DOPE in planar bilayers favors more open substates and longer open times, i.e., some channel stabilization as shown by Keller et al. (1993).
Samples preparation and Fourier transform infrared spectroscopy
Spectra of the native alamethicin and of alm-dUL in methanol were obtained with 1–2 mg of the polypeptide dissolved in 100 µl of the solvent. Spectra were recorded on a 1750 Perkin-Elmer (Wellesley, MA) FTIR spectrometer at 15°C and 25°C by signal averaging 64 scans with a resolution of 4 cm-1. Native and synthetic alamethicin in lipid dispersions were prepared as follows. Required amounts of alamethicin and DMPC were dissolved in chloroform to achieve a peptide/lipid molar ratio of 0.05–0.15. After solvent evaporation under nitrogen, the samples were dried under vacuum. The required volume of H20 or 2H20 was added to each of the dried samples, before mixing with a vortex and then incubated above the transition temperature of DMPC (Tt = 23°C) for 3–4 h. This incubation step was omitted for some samples in 2H20 so that the hydrogen-deuterium exchange of the amide protons in the first several hours can be monitored. The concentration of alamethicin used in these measurements was 
10 mg/ml. Infrared spectra were recorded at 1°C, 15°C, 20°C, and 30°C by signal averaging 400 scans with a resolution of 4 cm-1. Further details about the methods of recording and analyzing FTIR spectra are described in our earlier publications (Haris et al., 1986, 1994; Haris and Chapman, 1988).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
, 1996), all macroscopic and single-channel conductance data displayed by alm-dUL in planar lipid bilayers, as compared to native alamethicin, point to a very similar behavior as regards voltage- and concentration-dependences, and single-channel amplitudes. The only parameter that is significantly altered is the kinetics of single-channel substates as confirmed by the experiment shown in Fig. 2, performed for the purpose of this study: alm-dUL induces channels whose open lifetimes are about one order of magnitude faster than with native alamethicin (Fig. 2).
|
3 cm-1) than for alamethicin. On the other hand, the amide II band occurs at a lower frequency by 1 cm-1 for the native alamethicin. In addition, absorbances in the 1675 cm-1, 1639 cm-1, and 1621 cm-1 regions are significantly lower for alm-dUL.
|
) is similar to FTIR spectra of alamethicin recorded in methanol (Fig. 3). In that earlier study, FTIR spectra of native alamethicin in aqueous dispersions of DMPC, recorded above and below the lipid transition temperature, were shown to be virtually identical (Haris and Chapman, 1988).
The FTIR spectra of alm-dUL in lipid dispersion are similar to the spectra of native alamethicin also recorded in aqueous lipid dispersion. However, as for the samples in methanol (see Fig. 3), the main amide I component for alm-dUL occurs at a lower (2–4 cm-1) frequency than for native alamethicin. This clearly indicates a change in the hydrogen-bonding pattern within the amide groups (Alvarez et al., 1987) of alm-dUL upon substitution of Aib residues with Leu.
Fig. 4 shows the second-derivative spectra of alm-dUL in aqueous lipid dispersion above (30°C) and below (15°C) the lipid phase transition. It can be seen that the main amide I component is at 1657 cm-1 below the lipid phase transition and this shifts to 1661 cm-1 when the temperature is raised above Tt. As the spectrum of this sample recorded at 1°C was found to be virtually identical to that recorded at 15°C, the observed shift above the lipid phase transition is not due to a direct effect of temperature on the polypeptide structure. It is also important to note that the amide II peak shifts from 1548 cm-1 to l546 cm-1 on raising the temperature above the lipid phase transition. The simultaneous shift of the amide I and amide II band toward higher and lower frequency, respectively, suggests an increased hydrogen-bonding within the peptide groups (Alvarez et al., 1987).
|
1°C and 30°C, but the second-derivative spectra were similar to that recorded at 15°C (see Fig. 5). There was no significant shift of the amide I components to suggest any change in the secondary structure of native alamethicin as a function of lipid fluidity, as was previously reported for samples in H2O (Haris and Chapman, 1988).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
-helical structure, the aim of studying alm-dUL was also to use it as a model channel-forming peptide, more akin to normal protein channels (that is without the noncoded Aib). Given the significantly increased hydropathy index of Leu versus Aib and the greater bulkiness of its side chain, stronger hydrophobic interactions were expected both with the lipid acyl chains and between monomers. An alternative synthetic analog was also designed where all Aibs were replaced by Ala, but both its channel-forming ability and its conformational stability were rather poor (our unpublished results).
FTIR spectra of alm-dUL were obtained so that comparisons could be made with native alamethicin. Spectra obtained in methanol are similar in the amide I and amide II regions for both peptides although some small, but significant, differences were apparent. Most importantly, the amide I maximum occurs at a lower frequency for alm-dUL indicating an increase in amide hydrogen-bonding, as is also supported by a small positive shift of the amide II band toward higher frequency. As already mentioned, it is well known that an increase in amide hydrogen-bonding causes the amide I and amide II band frequencies to decrease and increase, respectively (Alvarez et al., 1987; Haris and Chapman, 1988). The fact that the amide I maximum for alm-dUL occurs at a lower frequency, and hence more closer to the absorbance expected from normal helices, suggests that it has a smaller amount of 310-helical structure compared with the native alamethicin. It is also noteworthy that in the FTIR spectrum of alm-dUL, the amide band shape is much more symmetrical, and absorbance near 1670-cm-1 and 1640-cm-1 intensities are lower as compared with native alamethicin. These observations are also in favor of some conformational change and support a contribution from 310-helical structure to the amide I band in alamethicin.
In this work, the secondary structure of both native alamethicin and alm-dUL in aqueous lipid dispersion was studied above and below the lipid phase transition. Contrasting with alamethicin, the frequency of the amide I maximum for alm-dUL was found to increase as the temperature was raised above the lipid phase transition temperature Tt. The increase in amide I band frequency suggests a weakening of hydrogen-bonding between the peptide groups above Tt . The fact that the amide I band is now located at frequency of 1661 cm-1 possibly indicates an increase in 310-helical structure that has weaker hydrogen-bonding compared to normal helical structure. However, the changes observed could be due to other factors. For instance, the polypeptide structure might be more flexible in a fluid lipid bilayer compared with a rigid one. Increased flexibility can be related to the weakening of hydrogen-bonds, which in turn may explain the shift of the amide I band toward higher frequency. In addition, this latter effect could be due to further penetration of the peptide into the hydrophobic core and the resulting reduced hydration of the peptide groups. Reciprocally, it was recently reported that alamethicin induces different changes in lipid mobility above and below a critical peptide concentration (Kikukawa and Araiso, 2002) corresponding to a drastic change in alamethicin orientation (Huang, 2001).
Overall, the results presented here can be summarized as follows:
) that showed by circular dichroism and NMR experiments that the substitution of Aib by Leu increased the -helical content and also decreased the length of alm-dUL. ).
We conclude that substitution of Aib residues with Leu produces a more stable structure for the monomers that display strong interaction with phospholipid membranes. A NMR study in SDS micelles (Brachais et al., 1998) concluded also that Alm-dUL helices showed regular structure along their axes that are lightly bent near the Pro14 although a higher flexibility was observed in MeOH. These structural findings are correlated with functional data comparing channel-forming activity of both peptides in planar lipid bilayers (Fig. 2). It thus appears that some structural flexibility in addition to an adequate peptide length/bilayer thickness match favors long-lived channels. In a previous article, Brachais et al. (1995) considered that alamethicin was largely helical with several 310 parts in the C-terminal, whereas Alm-dUL seemed to be exclusively in -helical conformation according to NMR experiments. Thus the authors claimed that Alm-dUL would be certainly shorter than alamethicin and consequently less able to cross completely the membrane under applied voltage. This parameter would be responsible of faster fluctuations of ion channels induced by Alm-dUL in planar lipid bilayers. Further molecular modeling study (Brachais et al., 1998) confirmed the shortening of Alm-dUL compared to the one of Alm (26.5 Å against 29 Å).
|
ACKNOWLEDGEMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
The support of Groupement de Recherche Centre National de la Recherche Scientifique No. 964 "Canaux Peptidiques" is also acknowledged.
Submitted on March 19, 2003; accepted for publication October 21, 2003.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
-methylmannose binding studied by Fourier transform infrared spectroscopy. Biochim. Biophys. Acta. 916:5–12.[Medline]
Bak, M., R. P. Bywater, M. Hohwy, J. K. Thomsen, K. Adelhorst, H. J. Jakobsen, O. W. Sorensen, and N. C. Nielsen. 2001. Conformation of alamethicin in oriented phospholipid bilayers determined by 15N solid-state nuclear magnetic resonance. Biophys. J. 81:1684–1698.
Bechinger, B. 1997. Structure and functions of channel-forming peptides: magainins, cecropins, melittin and alamethicin. J. Membr. Biol. 156:197–211.[Medline]
Boheim, G. 1974. Statistical analysis of alamethicin channels in black lipid membranes. J. Membr. Biol. 19:277–303.[Medline]
Brachais, L., D. Davoust, and G. Molle. 1995. Conformational study of a synthetic analogue of alamethicin. Influence of the conformation on ion-channel lifetimes. Int. J. Pept. Protein Res. 45:164–172.[Medline]
Brachais, L., C. Mayer, D. Davoust, and G. Molle. 1998. Influence of the secondary structure on the pore forming properties of synthetic alamethicin analogs: NMR and molecular modelling studies. J. Pept. Sci. 4:344–354.[Medline]
Cafiso, D. S. 1994. Alamethicin: a peptide model for voltage gating and protein-membrane interactions. Annu. Rev. Biophys. Biomol. Struct. 23:141–165.[Medline]
Cascio, M., and B. A. Wallace. 1988. Conformation of alamethicin in phospholipid vesicles: implication for insertion models. Proteins. 4:89–98.[Medline]
Coronado, R., and R. Latorre. 1983. Phospholipid bilayers made from monolayers on patch-clamp pipettes. Biophys. J. 43:231–236.
Duclohier, H., and G. Spach. 2001. Artificial membrane excitability revisited and implications for the gating of voltage-dependent ion channels. Gen. Physiol. Biophys. 20:1–15.
Duclohier, H., and H. Wroblewski. 2001. Voltage-dependent pore formation and antimicrobial activity by alamethicin and analogues. J. Membr. Biol. 184:1–12.[Medline]
Eisenberg, M., J. E. Hall, and C. A. Mead. 1973. The nature of the voltage-dependent conductance induced by alamethicin in black lipid membranes. J. Membr. Biol. 14:143–176.[Medline]
Esposito, G., J. A. Carver, J. Boyd, and I. D. Campbell. 1987. High-resolution 1H NMR study of the solution structure of alamethicin. Biochemistry. 26:1043–1050.[Medline]
Fox, R. O., and F. M. Richards. 1982. A voltage-gated ion channel model inferred from the crystal structure of alamethicin at 1.5 Å resolution. Nature. 300:325–330.[Medline]
Gisin, B. F., S. Kobayashi, and J. E. Hall. 1977. Synthesis of a 19-residue peptide with alamethicin-like activity. Proc. Natl. Acad. Sci. USA. 74:115–119.
Hall, J. E., I. Vodyanoy, T. M. Balasubramanian, and G. R. Marshall. 1984. Alamethicin, a rich model for channel behavior. Biophys. J. 45:233–247.
Haris, P. I., D. C. Lee, and D. Chapman. 1986. A Fourier transform infrared investigation of the structural differences between ribonuclease-A and ribonuclease-S. Biochim. Biophys. Acta. 874:255–265.[Medline]
Haris, P. I., and D. Chapman. 1988. Fourier transform infrared spectra of the polypeptide alamethicin and a possible structural similarity with bacteriorhodopsin. Biochim. Biophys. Acta. 943:375–380.[Medline]
Haris, P. I., B. Ramesh, S. Brazier, and D. Chapman. 1994. The conformational analysis of a synthetic S4 peptide corresponding to a voltage-gated potassium channel protein. FEBS Lett. 349:371–374.[Medline]
Helluin, O., J.-Y. Dugast, G. Molle, A. R. Mackie, S. Ladha, and H. Duclohier. 1997. Lateral diffusion and conductance properties of a fluorescein-labelled alamethicin in planar lipid bilayers. Biophys. J. 1330:284–292.
Huang, H. W. 2001. Action of antimicrobial peptides: two-state model. Biochemistry. 39:8347–8352.
Keller, S. L., S. M. Bezrukov, S. M. Gruner, M. W. Tate, I. Vodyanoy, and V. A. Parsegian. 1993. Probability of alamethicin conductance states varies with nonlamellar tendency of bilayer phospholipids. Biophys. J. 65:23–27.
Kikukawa, T., and T. Araiso. 2002. Changes in lipid mobility associated with alamethicin incorporation into membranes. Arch. Biochem. Biophys. 405:214–222.[Medline]
Martin, D. R., and R. J. Williams. 1976. Chemical nature and sequence of alamethicin. Biochem. J. 153:181–190.[Medline]
Menestrina, G., K. P. Voges, G. Jung, and G. Boheim. 1986. Voltage-dependent channel formation by rods of helical polypeptides. J. Membr. Biol. 93:111–132.[Medline]
Molle, G., J.-Y. Dugast, H. Duclohier, and G. Spach. 1988. Conductance properties of des-Aib-Leu-des-Pheol-Phe-alamethicin in planar lipid bilayers. Biochim. Biophys. Acta. 938:310–314.
Molle, G., J.-Y. Dugast, G. Spach, and H. Duclohier. 1996. Ion channel stabilization of synthetic alamethicin analogs by rings of inter-helix H-bonds. Biophys. J. 70:1669–1675.
Mueller, P., and D. O. Rudin. 1968. Action potentials induced in bimolecular membranes. Nature. 217:713–719.[Medline]
Sackmann, B., and G. Boheim. 1979. Alamethicin-induced single channel conductance fluctuations in biological membranes. Nature. 282:336–339.[Medline]
Sansom, M. S. P. 1993. Structure and function of channel-forming peptaibols. Q. Rev. Biophys. 26:365–421.[Medline]
Schwarz, G., H. Gerke, V. Rizzo, and S. Stankowski. 1987. Incorporation kinetics in a membrane, studied with the pore-forming peptide alamethicin. Biophys. J. 52:685–692.
Tieleman, D. P., H. J. C. Berendsen, and M. S. P. Sansom. 1999. An alamethicin channel in a lipid bilayer: dynamic simulations. Biophys. J. 76:1757–1769.
Vodyanoy, I., J. E. Hall, and V. Vodyanoy. 1988. Alamethicin adsorption to a planar lipid bilayer. Biophys. J. 53:649–658.
Woolley, G. A., and B. A. Wallace. 1992. Model ion channels: gramicidin and alamethicin. J. Membr. Biol. 129:109–136.[Medline]
Yin, P., C. J. Burns, P. D. J. Osman, and B. A. Cornell. 2003. A tethered bilayer sensor containing alamethicin channels and its detection of amiloride based inhibitors. Biosens. Bioelectron. 18:389–397.[Medline]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
