Ariadna Grau-Campistany

A bioinspired peptide scaffold with high antibiotic activity and low in vivo toxicity

Bacterial resistance to almost all available antibiotics is an important public health issue. A major goal in antimicrobial drug discovery is the generation of new chemicals capable of killing pathogens with high selectivity, particularly multi-drug-resistant ones. Here we report the design, preparation and activity of new compounds based on a tunable, chemically accessible and upscalable lipopeptide scaffold amenable to suitable hit-to-lead development. Such compounds could become therapeutic candidates and future antibiotics available on the market. The compounds are cyclic, contain two D-amino acids for in vivo stability and their structures are reminiscent of other cyclic disulfide-containing peptides available on the market. The optimized compounds prove to be highly active against clinically relevant Gram-negative and Gram-positive bacteria. In vitro and in vivo tests show the low toxicity of the compounds. Their antimicrobial activity against resistant and multidrug-resistant bacteria is at the membrane level, although other targets may also be involved depending on the bacterial strain.

Tryptophan-containing lipopeptide antibiotics derived from polymyxin B with activity against Gram positive and Gram negative bacteria.

Resistance to all known antibiotics is a growing concern worldwide, and has renewed the interest in antimicrobial peptides, a structurally diverse class of amphipathic molecules that essentially act on the bacterial membrane. Propelled by the antimicrobial potential of this compound class, we have designed three new lipopeptides derived from polymyxin B, sp-34, sp-96 and sp-100, with potent antimicrobial activity against both Gram positive and Gram negative bacteria. The three peptides bind with high affinity to lipopolysaccharide as demonstrated by monolayer penetration and dansyl-displacement. The interaction with the cytoplasmic membrane has been elucidated by biophysical experiments with model membranes of POPG or POPE/POPG (6:4), mimicking the Gram positive and Gram negative bacterial membrane. Trp-based fluorescence experiments including steady-state, quenching, anisotropy and FRET, reveal selectivity for anionic phospholipids and deep insertion into the membrane. All three lipopeptides induce membrane fusion and leakage from anionic vesicles, a process that is favored by the presence of POPE. The molecules bind to zwitterionic POPC vesicles, a model of the eukaryotic membrane, but in a different way, with lower affinity, less penetration into the bilayer and no fusion or permeabilization of the membrane. Results in model membranes are consistent with flow cytometry experiments in Escherichia coli and Staphylococcus aureus using a membrane potential sensitive dye (bis-oxonol) and a nucleic acid dye (propidium iodide), suggesting that the mechanism of action is based on membrane binding and collapse of membrane integrity by depolarization and permeabilization.

Hydrophobic mismatch demonstrated for membranolytic peptides, and their use as molecular rulers to measure bilayer thickness in native cells

Hydrophobic mismatch is a well-recognized principle in the interaction of transmembrane proteins with lipid bilayers. This concept was extended here to amphipathic membranolytic α-helices. Nine peptides with lengths between 14 and 28 amino acids were designed from repeated KIAGKIA motifs, and their helical nature was confirmed by circular dichroism spectroscopy. Biological assays for antimicrobial activity and hemolysis, as well as fluorescence vesicle leakage and solid-state NMR spectroscopy, were used to correlate peptide length with membranolytic activity. These data show that the formation of transmembrane pores is only possible under the condition of hydrophobic matching: the peptides have to be long enough to span the hydrophobic bilayer core to be able to induce vesicle leakage, kill bacteria, and cause hemolysis. By correlating the threshold lengths for biological activity with the biophysical results on model vesicles, the peptides could be utilized as molecular rulers to measure the membrane thickness in different cells.

Erratum to: Biosurfactant production by AL 1.1, a Bacillus licheniformis strain isolated from Antarctica: production, chemical characterization and properties

Biosurfactants are of great interest due to the demand for natural products with low toxicity. Nevertheless, their production is not competitive when cost is a limiting factor. Strain AL 1.1, isolated on Deception Island (Antarctica), identified as Bacillus licheniformis, produced lipopeptides when grown using a variety of carbohydrates. Biosurfactant production, but not growth, was optimal at 30 °C. The culture conditions and medium composition dictated biosurfactant production. Basic optimization of culture and extraction parameters gave a production yiels of 860 mg/L purified extract in 24 h. The purified biosurfactant yielded a mixture of lipopeptide homologues, with molecular weights between 1006 and 1034. The peptide moiety consists of glutamine as the N-terminal amino acid, two leucines, valine, aspartic, leucine and isoleucine as the C-terminal amino acid. The lipid moiety contains a mixture of β-hydroxy fatty acids ranging in size from C14 to C16. These results indicate a similarity with lichenysin groups A, D or G. The organic extract reduced surface tension to 28.5 mN/m and achieved a critical micelle concentration of 15 mg/L. This highly effective and efficient behavior characterized the product as a powerful surfactant. Its stability under a wide pH range, high temperatures and variable concentrations of salt, as well as its emulsifying properties, suggest potential application in cosmetic industrial processes.

Extending the Hydrophobic Mismatch Concept to Amphiphilic Membranolytic Peptides

A series of nine amphiphilic, pore-forming α-helical KIA peptides (KIAGKIA repeats) with lengths between 14 and 28 residues were studied by solid-state (15)N NMR to determine their alignment in oriented lipid bilayers. In a 2:1 mixture of 1,2-dimyristoyl-sn-glycero-3-phosphatidylcholine (DMPC) with its corresponding 1-myristoyl-2-hydroxy-sn-glycero-3-phosphocholine (lyso-MPC), which has a highly positive spontaneous curvature, the helix tilt angle was found to vary steadily with peptide length. The shortest peptide was aligned transmembrane and upright, while the longer ones successively became tilted away from the membrane normal. This behavior is in agreement with the hydrophobic matching concept, conceived so far only for hydrophobic helices. In 1,2-dioleoyl-sn-glycero-3-phosphatidylcholine, with a negative spontaneous curvature, all KIA peptides remained flat on the bilayer surface, while the cylindrical DMPC lipids permitted a slight tilt. Peptide insertion thus depends critically on the intrinsic lipid curvature, and helix orientation is then fine-tuned by membrane thickness. A refined toroidal pore model is proposed.

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.

Helix Fraying and Lipid-Dependent Structure of a Short Amphipathic Membrane-Bound Peptide Revealed by Solid-State NMR

The amphipathic α-helical peptide KIA14 [(KIAGKIA)2-NH2] was studied in membranes using circular dichroism and solid-state NMR spectroscopy to obtain global as well as local structural information. By analyzing 2H NMR data from 10 analogues of KIA14 that were selectively labeled with Ala- d3, those positions that are properly folded into a helix could be determined within the membrane-bound peptide. The N-terminus was found to be unraveled, whereas positions 4-14 formed an ideal helix all the way to the C-terminus. The helicity did not change when Gly residues were replaced by Ala- d3 but was reduced when Ile was replaced, indicating that large hydrophobic residues are required for membrane binding and helix formation. The reduced helicity was strongly correlated with a decrease in peptide-induced leakage from lipid vesicles. The orientation of the short KIA14 peptide was assessed in several lipid systems and compared with that of the longer KIA21 sequence [(KIAGKIA)3-NH2]. In 1,2-dioleoyl- sn-glycero-3-phosphatidylcholine, both peptides are aligned flat on the membrane surface, whereas in 1,2-dimyristoyl- sn-glycero-3-phosphatidylcholine (DMPC)/1-myristoyl-2-hydroxy- sn-glycero-3-phosphatidylcholine (lyso-MPC) both are inserted into the membrane in an upright orientation. These two types of lipid systems had been selected for their strongly negative and positive spontaneous curvature, respectively. We propose that in these cases, the peptide orientation is largely determined by the lipid properties. On the other hand, in plain DMPC and 1,2-dilauroyl- sn-glycero-3-phosphatidylcholine, which have only a slight positive curvature, a marked difference in orientation is evident: the short KIA14 lies almost flat on the membrane surface, whereas the longer KIA21 is more tilted. We thus propose that out of the lipid systems tested here, DMPC (with hardly any curvature) is the least biased lipid system in which peptide orientation and realignment can be studied, allowing to compare and discriminate the intrinsic effects of the properties of the peptides as such.