Huey W. Huang

Barrel-Stave Model or Toroidal Model? A Case Study on Melittin Pores

Transmembrane pores induced by amphiphilic peptides, including melittin, are often modeled with the barrel-stave model after the alamethicin pore. We examine this assumption on melittin by using two methods, oriented circular dichroism (OCD) for detecting the orientation of melittin helix and neutron scattering for detecting transmembrane pores. OCD spectra of melittin were systematically measured. Melittin can orient either perpendicularly or parallel to a lipid bilayer, depending on the physical condition and the composition of the bilayer. Transmembrane pores were detected when the helices oriented perpendicularly to the plane of the bilayers, not when the helices oriented parallel to the bilayers. The evidence that led to the barrel-stave model for alamethicin and that to the toroidal model for magainin were reviewed. The properties of melittin pores are closely similar to that of magainin but unlike that of alamethicin. We conclude that, among naturally produced peptides that we have investigated, only alamethicin conforms to the barrel-stave model. Other peptides, including magainins, melittin and protegrins, all appear to induce transmembrane pores that conform to the toroidal model in which the lipid monolayer bends continuously through the pore so that the water core is lined by both the peptides and the lipid headgroups.

Crystallization of Antimicrobial Pores in Membranes: Magainin and Protegrin

Membrane pores spontaneously formed by antimicrobial peptides in membranes were crystallized for the first time by manipulating the sample hydration and temperature. Neutron diffraction shows that magainins and protegrins form stable pores in fully hydrated fluid membranes. At lower hydration levels or low temperature, the membrane multilayers crystallize. In one crystalline phase, the pores in each bilayer arrange in a regular hexagonal array and the bilayers are stacked into a hexagonal ABC lattice, corresponding to the cubic close-packed structure of spheres. In another crystalline phase, the bilayers are modulated into the rippled multilamellae, corresponding to a 2D monoclinic lattice. The phase diagrams are described. Crystallization of the membrane pores provides possibilities for diffraction studies that might provide useful information on the pore structures.

Deformation free energy of bilayer membrane and its effect on gramicidin channel lifetime

The deformation free energy of a lipid bilayer is presented based on the principle of a continuum theory. For small deformations, the free energy consists of a layer-compression term, a splay-distortion term, and a surface-tension term, equivalent to the elastic free energy of a two-layer smectic liquid crystal with surface tension. Minimization of the free energy leads to a differential equation that, with boundary conditions, determines the elastic deformation of a bilayer membrane. When a dimeric gramicidin channel is formed in a membrane of thickness greater than the length of the channel, the membrane deformation reduces the stability of the channel. Previously this effect was studied by comparing the variation of channel lifetime with the surface tension of bilayers (Elliott, J. R., D. Needham, J. P. Dilger, and D. A. Hayden, 1983, Biochim. Biophys. Acta, 735:95-103). The tension was assumed to pull a dimer for a distance z before the channel loses ion conductivity. To account for the data, z was found to be 18 A. With the deformation free energy, the data can be accounted for with z less than or approximately to 1 A, which is consistent with the breaking of hydrogen bonds in a dimer dissociation. Increasing the strength of lipid-protein interactions is not the only consequence of the complete free energy compared with the previous discussions. It also changes the shape of membrane deformation around an embedded channel from convex to concave, and increases the range of deformation from less than 10 A to greater than 20 A. Clearly these will be important factors in the general considerations of lipid-protein interactions and membrane-mediated interactions between proteins. In addition, thermal fluctuations of a membrane are calculated; in particular, we calculate the relations between the intrinsic thickness and the experimentally measured values. The experimental parameters of monoolein-squalene membranes are used for quantitative analyses.

Experimental Evidence for Hydrophobic Matching and Membrane- Mediated Interactions in Lipid Bilayers Containing Gramicidin

Hydrophobic matching, in which transmembrane proteins cause the surrounding lipid bilayer to adjust its hydrocarbon thickness to match the length of the hydrophobic surface of the protein, is a commonly accepted idea in membrane biophysics. To test this idea, gramicidin (gD) was embedded in 1, 2-dilauroyl-sn-glycero-3-phosphocholine (DLPC) and 1, 2-myristoyl-sn-glycero-3-phosphocholine (DMPC) bilayers at the peptide/lipid molar ratio of 1:10. Circular dichroism (CD) was measured to ensure that the gramicidin was in the beta6.3 helix form. The bilayer thickness (the phosphate-to-phosphate distance, or PtP) was measured by x-ray lamellar diffraction. In the Lalpha phase near full hydration, PtP is 30.8 A for pure DLPC, 32.1 A for the DLPC/gD mixture, 35.3 A for pure DMPC, and 32.7 A for the DMPC/gD mixture. Gramicidin apparently stretches DLPC and thins DMPC toward a common thickness as expected by hydrophobic matching. Concurrently, gramicidin-gramicidin correlations were measured by x-ray in-plane scattering. In the fluid phase, the gramicidin-gramicidin nearest-neighbor separation is 26.8 A in DLPC, but shortens to 23.3 A in DMPC. These experiments confirm the conjecture that when proteins are embedded in a membrane, hydrophobic matching creates a strain field in the lipid bilayer that in turn gives rise to a membrane-mediated attractive potential between proteins.

Evidence for Membrane Thinning Effect as the Mechanism for Peptide-Induced Pore Formation

Antimicrobial peptides have two binding states in a lipid bilayer, a surface state S and a pore-forming state I. The transition from the S state to the I state has a sigmoidal peptide-concentration dependence indicating cooperativity in the peptide-membrane interactions. In a previous paper, we reported the transition of alamethicin measured in three bilayer conditions. The data were explained by a free energy that took into account the membrane thinning effect induced by the peptides. In this paper, the full implications of the free energy were tested by including another type of peptide, melittin, that forms toroidal pores, instead of barrel-stave pores as in the case of alamethicin. The S-to-I transitions were measured by oriented circular dichroism. The membrane thinning effect was measured by x-ray diffraction. All data were in good agreement with the theory, indicating that the membrane thinning effect is a plausible mechanism for the peptide-induced pore formations.

Neutron scattering in the plane of membranes: structure of alamethicin pores

A technique of neutron in-plane scattering for studying the structures of peptide pores in membranes is described. Alamethicin in the inserted state was prepared and undeuterated and deuterated dilauroyl phosphatidylcholine (DLPC) hydrated with D2O or H2O. Neutron in-plane scattering showed a strong dependence on deuteration, clearly indicating that water is a part of the high-order structure of inserted alamethicin. The data are consistent with the simple barrel-stave model originally proposed by Baumann and Mueller. The theoretical curves computed with this model at four different deuteration conditions agree with the data in all cases. Both the diameter of the water pore and the effective outside diameter of the channel are determined accurately. Alamethicin forms pores in a narrow range of size. In a given sample condition, > 70% of the peptide forms pores of n and n +/- 1 monomers. The pore size varies with hydration and with lipid. In DLPC, the pores are made of n = 8-9 monomers, with a water pore approximately 18 A in diameter and with an effective outside diameter of approximately 40 A. In diphytanoyl phosphatidylcholine, the pores are made of n approximately 11 monomers, with a water pore approximately 26 A in diameter, with an effective outside diameter of approximately 50 A.

Method of oriented circular dichroism.

We present a new method for determining the orientation of alpha-helical sections of proteins or peptides in membrane. To apply this method, membranes containing proteins must be prepared in a multilayer array. Circular dichroism (CD) spectra of the multilayer sample are then measured at the normal as well as oblique incident angles with respect to the bilayer planes; we call such spectra oriented circular dichroism (OCD). The procedure of OCD measurement, particularly the ways to avoid the spectral artifacts due to the effects of dielectric interfaces, linear dichroism and birefringence, and the method of data analysis are described in detail. To illustrate the method, we analyze the OCD of alamethicin in diphytanoylphosphatidylcholine multilayers. We conclude unambiguously that the helical section of alamethicin is parallel to the membrane normal when the sample is in the full-hydration state, but the helical section rotates to the plane of membrane when the sample is in a low-hydration state. We also obtained the parallel and perpendicular CD spectra of alpha-helix, and found them to be in agreement with previous theoretical calculations based on the exciton theory. These spectra are useful for analyzing protein orientations in future experiments.

Mechanism of alamethicin insertion into lipid bilayers

Alamethicin adsorbs on the membrane surface at low peptide concentrations. However, above a critical peptide-to-lipid ratio (P/L), a fraction of the peptide molecules insert in the membrane. This critical ratio is lipid dependent. For diphytanoyl phosphatidylcholine it is about 1/40. At even higher concentrations P/L > or = 1/15, all of the alamethicin inserts into the membrane and forms well-defined pores as detected by neutron in-plane scattering. A previous x-ray diffraction measurement showed that alamethicin adsorbed on the surface has the effect of thinning the bilayer in proportion to the peptide concentration. A theoretical study showed that the energy cost of membrane thinning can indeed lead to peptide insertion. This paper extends the previous studies to the high-concentration region P/L > 1/40. X-ray diffraction shows that the bilayer thickness increases with the peptide concentration for P/L > 1/23 as the insertion approaches 100%. The thickness change with the percentage of insertion is consistent with the assumption that the hydrocarbon region of the bilayer matches the hydrophobic region of the inserted peptide. The elastic energy of a lipid bilayer including both adsorption and insertion of peptide is discussed. The Gibbs free energy is calculated as a function of P/L and the percentage of insertion phi in a simplified one-dimensional model. The model exhibits an insertion phase transition in qualitative agreement with the data. We conclude that the membrane deformation energy is the major driving force for the alamethicin insertion transition.

Structure of transmembrane pore induced by Bax-derived peptide: Evidence for lipidic pores

The structures of transmembrane pores formed by a large family of pore-forming proteins and peptides are unknown. These proteins, whose secondary structures are predominantly α-helical segments, and many peptides form pores in membranes without a crystallizable protein assembly, contrary to the family of β-pore-forming proteins, which form crystallizable β-barrel pores. Nevertheless, a protein-induced pore in membranes is commonly assumed to be a protein channel. Here, we show a type of peptide-induced pore that is not framed by a peptide structure. Peptide-induced pores in multiple bilayers were long-range correlated into a periodically ordered lattice and analyzed by X-ray diffraction. We found the pores induced by Bax-derived helical peptides were at least partially framed by a lipid monolayer. Evidence suggests that the formation of such lipidic pores is a major mechanism for α-pore-forming proteins, including apoptosis-regulator Bax.

Lipid-alamethicin interactions influence alamethicin orientation.

Whereas the barrel-stave configuration is accepted by most investigators as a good description of the conducting state of alamethicin, there are conflicting interpretations on its nonconducting state; in the absence of an applied field, some found alamethicin molecules on the membrane surface, but others found them incorporated in the hydrophobic core of the membrane. This problem is resolved by the discovery of a phase-transitionlike behavior of alamethicin in the membrane. As a function of lipid/peptide ratio L/P and the chemical potential of water mu, alamethicin molecules were observed to switch between two states: in one, the majority of the peptide molecules bind parallel to the membrane surface; in another, the majority of the peptide molecules insert perpendicularly into the membrane. The state of alamethicin was monitored by the method of oriented circular dichroism (OCD; Wu, Y., H. W. Huang, and G. A. Olah, 1990, Biophys. J. 57:797-806) using aligned multilayer samples in the liquid crystalline L(alpha) phase. If L/P exceeds a critical value, most of the peptide molecules are on the membrane surface. If L/P is below the critical value, most of the peptide molecules are incorporated in the membrane when mu is high; when mu is low, most of them are again on the membrane surface. In a typical conduction experiment of voltage dependence, alamethicin molecules are in a partition equilibrium between the aqueous phase and the lipid phase before the application of voltage; in the lipid phase, the lipid/peptide ratio is such that most of alamethicin molecules are on the membrane surface. This is the nonconducting state of alamethicin. The OCD analysis showed that there is essentially no change in the secondary structure when alamethicin changes between the surface state and the inserted state. The voltage-gating mechanism can be explained if we assume that these surface peptide molecules probabilistically turn into the membrane core to form channels due to the dipole-electric field interactions. We speculate that the phase-transitionlike behavior is a manifestation of membrane-mediated intermolecular interactions between peptide molecules.