Comparative analysis of the orientation of transmembrane peptides using solid-state 2H- and 15N-NMR: mobility matters

Membranolytic Mechanism of Amphiphilic Antimicrobial β-Stranded [KL]n Peptides

Amphipathic peptides can act as antibiotics due to membrane permeabilization. KL peptides with the repetitive sequence [Lys-Leu]_n-NH₂ form amphipathic β-strands in the presence of lipid bilayers. As they are known to kill bacteria in a peculiar length-dependent manner, we suggest here several different functional models, all of which seem plausible, including a carpet mechanism, a β-barrel pore, a toroidal wormhole, and a β-helix. To resolve their genuine mechanism, the activity of KL peptides with lengths from 6–26 amino acids (plus some inverted LK analogues) was systematically tested against bacteria and erythrocytes. Vesicle leakage assays served to correlate bilayer thickness and peptide length and to examine the role of membrane curvature and putative pore diameter. KL peptides with 10–12 amino acids showed the best therapeutic potential, i.e., high antimicrobial activity and low hemolytic side effects. Mechanistically, this particular window of an optimum β-strand length around 4 nm (11 amino acids × 3.7 Å) would match the typical thickness of a lipid bilayer, implying the formation of a transmembrane pore. Solid-state ¹⁵N- and ¹⁹F-NMR structure analysis, however, showed that the KL backbone lies flat on the membrane surface under all conditions. We can thus refute any of the pore models and conclude that the KL peptides rather disrupt membranes by a carpet mechanism. The intriguing length-dependent optimum in activity can be fully explained by two counteracting effects, i.e., membrane binding versus amyloid formation. Very short KL peptides are inactive, because they are unable to bind to the lipid bilayer as flexible β-strands, whereas very long peptides are inactive due to vigorous pre-aggregation into β-sheets in solution.

Membrane Interactions of Latarcins: Antimicrobial Peptides from Spider Venom

A group of seven peptides from spider venom with diverse sequences constitute the latarcin family. They have been described as membrane-active antibiotics, but their lipid interactions have not yet been addressed. Using circular dichroism and solid-state ¹⁵N-NMR, we systematically characterized and compared the conformation and helix alignment of all seven peptides in their membrane-bound state. These structural results could be correlated with activity assays (antimicrobial, hemolysis, fluorescence vesicle leakage). Functional synergy was not observed amongst any of the latarcins. In the presence of lipids, all peptides fold into amphiphilic α-helices as expected, the helices being either surface-bound or tilted in the bilayer. The most tilted peptide, Ltc2a, possesses a novel kind of amphiphilic profile with a coiled-coil-like hydrophobic strip and is the most aggressive of all. It indiscriminately permeabilizes natural membranes (antimicrobial, hemolysis) as well as artificial lipid bilayers through the segregation of anionic lipids and possibly enhanced motional averaging. Ltc1, Ltc3a, Ltc4a, and Ltc5a are efficient and selective in killing bacteria but without causing significant bilayer disturbance. They act rather slowly or may even translocate towards intracellular targets, suggesting more subtle lipid interactions. Ltc6a and Ltc7, finally, do not show much antimicrobial action but can nonetheless perturb model bilayers.

Examination of pH dependency and orientation differences of membrane spanning alpha helices carrying a single or pair of buried histidine residues

We have employed the peptide framework of GWALP23 (acetyl-GGALWLALALALALALALWLAGA-amide) to examine the orientation, dynamics and pH dependence of peptides having buried single or pairs of histidine residues. When residue L8 is substituted to yield GWALP23-H⁸, acetyl-GGALWLAH⁸ALALALALALWLAGA-amide, the deuterium NMR spectra of ²H-labeled core alanine residues reveal a helix that occupies a single transmembrane orientation in DLPC, or in DMPC at low pH, yet shows multiple states at higher pH or in bilayers of DOPC. Moreover, a single histidine at position 8 or 16 in the GWALP23 framework is sensitive to pH. Titration points are observed near pH 3.5 for the deprotonation of H8 in lipid bilayers of DLPC or DMPC, and for H16 in DOPC. When residues L8 and L16 both are substituted to yield GWALP23-H^8,16, the ²H NMR spectra show, interestingly, no titration dependence from pH 2-8, yet bilayer thickness-dependent orientation differences. The helix with H8 and H16 is found to adopt a transmembrane orientation in thin bilayers of DLPC, a combination of transmembrane and surface orientations in DMPC, and then a complete transition to a surface bound orientation in the thicker DPoPC and DOPC lipid bilayers. In the surface orientations, alanine A7 no longer fits within the core helix. These results along with previous studies with different locations of histidine residues suggest that lipid hydrophobic thickness is a first determinant and pH a second determinant for the helical orientation, along with possible side-chain snorkeling, when the His residues are incorporated into the hydrophobic region of a lipid membrane-associated helix.

Transmembrane Helix Integrity versus Fraying To Expose Hydrogen Bonds at a Membrane-Water Interface

Transmembrane helices dominate the landscape for many membrane proteins. Often flanked by interfacial aromatic residues, these transmembrane helices also contain loops and interhelix segments, which could help in stabilizing a transmembrane orientation. Using ²H nuclear magnetic resonance spectroscopy to monitor bilayer-incorporated model GWALP23 family peptides, we address systematically the issue of helix fraying in relation to the dynamics and orientation of highly similar individual transmembrane helices. We inserted aromatic (Phe, Trp, Tyr, and His) or non-aromatic residues (Ala and Gly) into positions 4 and 5 adjacent to a core transmembrane helix to examine the side-chain dependency of the transmembrane orientation, dynamics, and helix integrity (extent and location of unraveling). Incorporation of [²H]alanine labels enables one to assess the helicity of the core sequence and the peptide termini. For most of the helices, we observed substantial unwinding involving at least three residues at both ends. For the unique case of histidine at positions 4 and 5, an extended N-terminal unwinding was observed up to residue 7. For further investigation of the onset of fraying, we employed A^4,5GWALP23 with ²H labels at residues 4 and 5 and found that the number of terminal residues involved in the unwinding depends on bilayer thicknesses and helps to govern the helix dynamics. The combined results enable us to compare and contrast the extent of fraying for each related helix, as reflected by the deviation of experimental ²H quadrupolar splitting magnitudes of juxta-terminal alanines A3 and A21 from those represented by an ideal helix geometry.

Best of Two Worlds? How MD Simulations of Amphiphilic Helical Peptides in Membranes Can Complement Data from Oriented Solid-State NMR

The membrane alignment of helical amphiphilic peptides in oriented phospholipid bilayers can be obtained as ensemble and time averages from solid state ²H NMR by fitting the quadrupolar splittings to ideal α-helices. At the same time, molecular dynamics (MD) simulations can provide atomistic insight into peptide-membrane systems. Here, we evaluate the potential of MD simulations to complement the experimental NMR data that is available on three exemplary systems: the natural antimicrobial peptide PGLa and the two designer-made peptides MSI-103 and KIA14, whose sequences were derived from PGLa. Each peptide was simulated for 1 μs in a DMPC lipid bilayer. We calculated from the MD simulations the local angles which define the side chain geometry with respect to the peptide helix. The peptide orientation was then calculated (i) directly from the simulation, (ii) from back-calculated MD-derived NMR splittings, and (iii) from experimental ²H NMR splittings. Our findings are that (1) the membrane orientation and secondary structure of the peptides found in the NMR analysis are generally well reproduced by the simulations; (2) the geometry of the side chains with respect to the helix backbone can deviate significantly from the ideal structure depending on the specific residue, but on average all side chains have the same orientation; and (3) for all of our peptides, the azimuthal rotation angle found from the MD-derived splittings is about 15° smaller than the experimental value.

Helix formation and stability in membranes

In this article we review current understanding of basic principles for the folding of membrane proteins, focusing on the more abundant alpha-helical class. Membrane proteins, vital to many biological functions and implicated in numerous diseases, fold into their active conformations in the complex environment of the cell bilayer membrane. While many membrane proteins rely on the translocon and chaperone proteins to fold correctly, others can achieve their functional form in the absence of any translation apparatus or other aides. Nevertheless, the spontaneous folding process is not well understood at the molecular level. Recent findings suggest that helix fraying and loop formation may be important for overall structure, dynamics and regulation of function. Several types of membrane helices with ionizable amino acids change their topology with pH. Additionally we note that some peptides, including many that are rich in arginine, and a particular analogue of gramicidin, are able passively to translocate across cell membranes. The findings indicate that a final protein structure in a lipid-bilayer membrane is sequence-based, with lipids contributing to stability and regulation. While much progress has been made toward understanding the folding process for alpha-helical membrane proteins, it remains a work in progress. This article is part of a Special Issue entitled: Emergence of Complex Behavior in Biomembranes edited by Marjorie Longo.

Influence of the Length and Charge on the Activity of α-Helical Amphipathic Antimicrobial Peptides

Hydrophobic mismatch is important for pore-forming amphipathic antimicrobial peptides, as demonstrated recently [Grau-Campistany, A., et al. (2015) Sci. Rep. 5, 9388]. A series of different length peptides have been generated with the heptameric repeat sequence KIAGKIA, called KIA peptides, and it was found that only those helices sufficiently long to span the hydrophobic thickness of the membrane could induce leakage in lipid vesicles; there was also a clear length dependence of the antimicrobial and hemolytic activities. For the original KIA sequences, the cationic charge increased with peptide length. The goal of this work is to examine whether the charge also has an effect on activity; hence, we constructed two further series of peptides with a sequence similar to those of the KIA peptides, but with a constant charge of +7 for all lengths from 14 to 28 amino acids. For both of these new series, a clear length dependence similar to that of KIA peptides was observed, indicating that charge has only a minor influence. Both series also showed a distinct threshold length for peptides to be active, which correlates directly with the thickness of the membrane. Among the longer peptides, the new series showed activities only slightly lower than those of the original KIA peptides of the same length that had a higher charge. Shorter peptides, in which Gly was replaced with Lys, showed activities similar to those of KIA peptides of the same length, but peptides in which Ile was replaced with Lys lost their helicity and were less active.

Influence of glutamic acid residues and pH on the properties of transmembrane helices

Negatively charged side chains are important for the function of particular ion channels and certain other membrane proteins. To investigate the influence of single glutamic acid side chains on helices that span lipid-bilayer membranes, we have employed GWALP23 (acetyl-GGALW⁵LALALALALALALW¹⁹LAGA-amide) as a favorable host peptide framework. We substituted individual Leu residues with Glu residues (L12E or L14E or L16E) and incorporated specific ²H-labeled alanine residues within the core helical region or near the ends of the sequence. Solid-state ²H NMR spectra reveal little change for the core labels in GWALP23-E12, -E14 and -E16 over a pH range of 4 to 12.5, with the spectra being broader for samples in DOPC compared to DLPC bilayers. The spectra for samples with deuterium labels near the helix ends on alanines 3 and 21 show modest pH-dependent changes in the extent of unwinding of the helix terminals in DLPC and DOPC bilayers. The combined results indicate minor overall responses of these transmembrane helices to changes in pH, with the most buried residue E12 showing no pH dependence. While the Glu residues E14 and E16 may have high pK_a values in the lipid bilayer environment, it is also possible that a paucity of helix response is masking the pK_a values. Interestingly, when E16 is present, spectral changes at high pH report significant local unwinding of the core helix. Our results are consistent with the expectation that buried carboxyl groups aggressively hold their protons and/or waters of hydration.

Membrane Thinning and Thickening Induced by Membrane-Active Amphipathic Peptides

Membrane thinning has been discussed as a fundamental mechanism by which antimicrobial peptides can perturb cellular membranes. To understand which factors play a role in this process, we compared several amphipathic peptides with different structures, sizes and functions in their influence on the lipid bilayer thickness. PGLa and magainin 2 from X. laevis were studied as typical representatives of antimicrobial cationic amphipathic α-helices. A 1:1 mixture of these peptides, which is known to possess synergistically enhanced activity, allowed us to evaluate whether and how this synergistic interaction correlates with changes in membrane thickness. Other systems investigated here include the α-helical stress-response peptide TisB from E. coli (which forms membrane-spanning dimers), as well as gramicidin S from A. migulanus (a natural antibiotic), and BP100 (designer-made antimicrobial and cell penetrating peptide). The latter two are very short, with a circular β-pleated and a compact α-helical structure, respectively. Solid-state (2)H-NMR and grazing incidence small angle X-ray scattering (GISAXS) on oriented phospholipid bilayers were used as complementary techniques to access the hydrophobic thickness as well as the bilayer-bilayer repeat distance including the water layer in between. This way, we found that magainin 2, gramicidin S, and BP100 induced membrane thinning, as expected for amphiphilic peptides residing in the polar/apolar interface of the bilayer. PGLa, on the other hand, decreased the hydrophobic thickness only at very high peptide:lipid ratios, and did not change the bilayer-bilayer repeat distance. TisB even caused an increase in the hydrophobic thickness and repeat distance. When reconstituted as a mixture, PGLa and magainin 2 showed a moderate thinning effect which was less than that of magainin 2 alone, hence their synergistically enhanced activity does not seem to correlate with a modulation of membrane thickness. Overall, the absence of a typical thinning response in the case of PGLa, and the increase in the repeat distance and membrane thickening observed for TisB, demonstrate that the concept of peptide-induced membrane thinning cannot be generalized. Instead, these results suggest that different factors contribute to the resulting changes in membrane thickness, such as the peptide orientation in the bilayer, and/or bilayer adaptation to hydrophobic mismatch.

Homo- and heteromeric interaction strengths of the synergistic antimicrobial peptides PGLa and magainin 2 in membranes

PGLa and magainin 2 (MAG2) are amphiphilic α-helical frog peptides with synergistic antimicrobial activity. In vesicle leakage assays we observed the strongest synergy for equimolar mixtures of PGLa and MAG2. This result was consistent with solid-state (15)N-NMR data on the helix alignment in model membranes. The Hill coefficients determined from the vesicle leakage data showed that the heterodimeric (PGLa-MAG2) interactions were stronger than the homodimeric (PGLa-PGLa and MAG2-MAG2) interactions. This result was also reflected in the free energy of dimerization determined from oriented circular dichroism and quantitative solid-state (19)F-NMR analysis.

Alanine scan and (2)H NMR analysis of the membrane-active peptide BP100 point to a distinct carpet mechanism of action

The short membrane-active peptide BP100 [KKLFKKILKYL-NH2] is known as an effective antimicrobial and cell penetrating agent. For a functional alanine scan each of the 11 amino acids was replaced with deuterated Ala-d3, one at a time. MIC assays showed that a substitution of Lys did not affect the antimicrobial activity, but it decreased when a hydrophobic residue was replaced. In most cases, a reduction in hydrophobicity led to a decrease in hemolysis, and some peptide analogues had an improved therapeutic index. Circular dichroism showed that BP100 folds as an amphiphilic α-helix in a bilayer. Its alignment was determined from (2)H NMR in oriented membranes of different composition. The azimuthal rotation angle was the same under all conditions, but the average helix tilt angle and the dynamical behavior of the peptide varied in a systematic manner. In POPC/POPG bilayers, with a negative spontaneous curvature, the peptide was found to lie flat on the bilayer surface, and with little wobble. In DMPC/DMPG, with a positive spontaneous curvature, BP100 at higher concentrations became tilted obliquely into the membrane, with the uncharged C-terminus inserted more deeply into the lipid bilayer, experiencing significant fluctuations in tilt angle. In DMPC/DMPG/lyso-MPC, with a pronounced positive spontaneous curvature, the helix tilted even further and became even more mobile. The 11-mer BP100 is obviously too short to form transmembrane pores. We conclude that BP100 operates via a carpet mechanism, whereby the C-terminus gets inserted into the hydrophobic core of the bilayer, which leads to membrane perturbation and induces transient permeability.

3D hydrophobic moment vectors as a tool to characterize the surface polarity of amphiphilic peptides

The interaction of membranes with peptides and proteins is largely determined by their amphiphilic character. Hydrophobic moments of helical segments are commonly derived from their two-dimensional helical wheel projections, and the same is true for β-sheets. However, to the best of our knowledge, there exists no method to describe structures in three dimensions or molecules with irregular shape. Here, we define the hydrophobic moment of a molecule as a vector in three dimensions by evaluating the surface distribution of all hydrophilic and lipophilic regions over any given shape. The electrostatic potential on the molecular surface is calculated based on the atomic point charges. The resulting hydrophobic moment vector is specific for the instantaneous conformation, and it takes into account all structural characteristics of the molecule, e.g., partial unfolding, bending, and side-chain torsion angles. Extended all-atom molecular dynamics simulations are then used to calculate the equilibrium hydrophobic moments for two antimicrobial peptides, gramicidin S and PGLa, under different conditions. We show that their effective hydrophobic moment vectors reflect the distribution of polar and nonpolar patches on the molecular surface and the calculated electrostatic surface potential. A comparison of simulations in solution and in lipid membranes shows how the peptides undergo internal conformational rearrangement upon binding to the bilayer surface. A good correlation with solid-state NMR data indicates that the hydrophobic moment vector can be used to predict the membrane binding geometry of peptides. This method is available as a web application on http://www.ibg.kit.edu/HM/.

Single tryptophan and tyrosine comparisons in the N-terminal and C-terminal interface regions of transmembrane GWALP peptides

Hydrophobic membrane-spanning helices often are flanked by interfacial aromatic or charged residues. In this paper, we compare the consequences of single Trp → Tyr substitutions at each interface for the properties of a defined transmembrane helix in the absence of charged residues. The choice of molecular framework is critical for these single-residue experiments because the presence of "too many" aromatic residues (more than one at either membrane-water interface) introduces excess dynamic averaging of solid state NMR observables. To this end, we compare the outcomes when changing W(5) or W(19), or both of them, to tyrosine in the well-characterized transmembrane peptide acetyl-GGALW(5)(LA)6LW(19)LAGA-amide ("GWALP23"). By means of solid-state (2)H and (15)N NMR experiments, we find that Y(19)GW(5)ALP23 displays similar magnitudes of peptide helix tilt as Y(5)GW(19)ALP23 and responds similarly to changes in bilayer thickness, from DLPC to DMPC to DOPC. The presence of Y(19) changes the azimuthal rotation angle ρ (about the helix axis) to a similar extent as Y(5), but in the opposite direction. When tyrosines are substituted for both tryptophans to yield GY(5,19)ALP23, the helix tilt angle is again of comparable magnitude, and furthermore, the preferred azimuthal rotation angle ρ is relatively unchanged from that of GW(5,19)ALP23. The extent of dynamic averaging increases marginally when Tyr replaces Trp. Yet, importantly, all members of the peptide family having single Tyr or Trp residues near each interface exhibit only moderate and not highly extensive dynamic averaging. The results provide important benchmarks for evaluating conformational and dynamic control of membrane protein function.

Canonical azimuthal rotations and flanking residues constrain the orientation of transmembrane helices

In biological membranes the alignment of embedded proteins provides crucial structural information. The transmembrane (TM) parts have well-defined secondary structures, in most cases α-helices and their orientation is given by a tilt angle and an azimuthal rotation angle around the main axis. The tilt angle is readily visualized and has been found to be functionally relevant. However, there exist no general concepts on the corresponding azimuthal rotation. Here, we show that TM helices prefer discrete rotation angles. They arise from a combination of intrinsic properties of the helix geometry plus the influence of the position and type of flanking residues at both ends of the hydrophobic core. The helical geometry gives rise to canonical azimuthal angles for which the side chains of residues from the two ends of the TM helix tend to have maximum or minimum immersion within the membrane. This affects the preferential position of residues that fall near hydrophobic/polar interfaces of the membrane, depending on their hydrophobicity and capacity to form specific anchoring interactions. On this basis, we can explain the orientation and dynamics of TM helices and make accurate predictions, which correspond well to the experimental values of several model peptides (including dimers), and TM segments of polytopic membrane proteins.

The Transmembrane Helix Tilt May Be Determined by the Balance between Precession Entropy and Lipid Perturbation

Hydrophobic helical peptides interact with lipid bilayers in various modes, determined by the match between the length of the helix’s hydrophobic core and the thickness of the hydrocarbon region of the bilayer. For example, long helices may tilt with respect to the membrane normal to bury their hydrophobic cores in the membrane, and the lipid bilayer may stretch to match the helix length. Recent molecular dynamics simulations and potential of mean force calculations have shown that some TM helices whose lengths are equal to, or even shorter than, the bilayer thickness may also tilt. The tilt is driven by a gain in the helix precession entropy, which compensates for the free energy penalty resulting from membrane deformation. Using this free energy balance, we derived theoretically an equation of state, describing the dependence of the tilt on the helix length and membrane thickness. To this end, we conducted coarse-grained Monte Carlo simulations of the interaction of helices of various lengths with lipid bilayers of various thicknesses, reproducing and expanding the previous molecular dynamics simulations. Insight from the simulations facilitated the derivation of the theoretical model. The tilt angles calculated using the theoretical model agree well with our simulations and with previous calculations and measurements.