Direct Measurement of Pore Dynamics and Leakage Induced by a

Subscriber access provided by UNIV OF NEBRASKA - LINCOLN

Article

Direct measurement of pore dynamics and leakage induced by a model antimicrobial peptide in single vesicles and cells Matthew G. Burton, Qi M Huang, Mohammed Akhter Hossain, John D. Wade, Enzo A. Palombo, Michelle L Gee, and Andrew Harry A. Clayton Langmuir, Just Accepted Manuscript • DOI: 10.1021/acs.langmuir.6b00596 • Publication Date (Web): 09 Jun 2016 Downloaded from http://pubs.acs.org on June 10, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Langmuir is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Direct measurement of pore dynamics and leakage induced by a model antimicrobial peptide in single vesicles and cells Matthew G. Burton,† Qi M. Huang,† Mohammed A. Hossain,†‡ John D. Wade,†‡ Enzo A. Palombo§, Michelle L. Gee† and Andrew H. A. Clayton,§* †

School of Chemistry, University of Melbourne, Parkville, Victoria 3010, Australia

‡

Florey Department of Neuroscience and Mental Health, Centre for Neuroscience Research

Chemistry, University of Melbourne, Parkville, Victoria 3010, Australia §

Faculty of Science, Engineering and Technology, Swinburne University of Technology,

Hawthorn, Victoria 3122, Australia E-mail addresses of co-authors: [email protected] (Matthew G. Burton); [email protected] (Qi M. Huang); [email protected] (Mohammed A. Hossain); [email protected] (John D. Wade); [email protected]; (Enzo A. Palombo); [email protected] (Michelle L. Gee); [email protected] (Andrew H. A. Clayton) Corresponding Author

ACS Paragon Plus Environment

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 38

*E-mail address: [email protected] (Andrew H. A. Clayton); Tel.: (+61)9214-5719. Fax: (+61)9214-5435. Keywords: peptide-membrane interaction, antimicrobial peptides, cell leakage, pore formation, fluorescence lifetime imaging microscopy

ABSTRACT

Antimicrobial peptides are promising therapeutic alternatives to counter growing antimicrobial resistance. Their precise mechanism of action remains elusive, however, particularly with respect to live bacterial cells. We investigated the interaction of a fluorescent melittin analogue with single Giant Unilamellar Vesicles, Giant Multilamellar Vesicles and bilamellar gram-negative Escherichia coli (E. coli) bacteria. Time-lapse fluorescence lifetime imaging microscopy was employed to determine the population distribution of the fluorescent melittin analogue between pore state and membrane surface state, and simultaneously measure the leakage of entrapped fluorescent species from the vesicle (or bacterium) interior. In Giant Unilamellar Vesicles, leakage from vesicle interior was correlated with an increase in level of pore states, consistent with a stable pore formation mechanism. In Giant Multilamellar Vesicles, vesicle leakage occurred more gradually and did not appear to correlate with increased pore states. Instead pore levels remained at a low steady-state level, which is more in line with coupled equilibria. Finally, in single bacterial cells, significant increases in pore levels were observed over time, which were correlated with only partial loss of cytosolic contents. These observations suggested that pore formation, as opposed to complete dissolution of membrane, was responsible for the leakage of contents in these systems, and that the bacterial membrane has an adaptive capacity that resists


2

Page 3 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

peptide attack. We interpret the three distinct pore dynamics regimes in the context of the increasing physical and biological complexity of the membranes.

INTRODUCTION Antimicrobial resistance remains a serious looming threat for humanity, with numerous strains of resistant pathogens emerging in recent decades. Infection by resistant strains is correlated with increased morbidity and mortality rates, along with increases in both direct and indirect economic costs due to the treatment process, as well as through losses in human productivity.1-4 Research into new therapeutic agents that act via different mechanisms to conventional antibiotics continues to be highly appealing. Antimicrobial peptides (AMPs), also known as Host Defense Peptides, have been found across a vast large array of living organisms, utilized directly to combat infection and disease, or as immuno-modulators tied to various bodily functions.5-9 Due to their ability to disrupt and permeate the cell wall of gram-positive and gram-negative bacteria, they have become promising templates for the design of future therapeutics. Despite enormous efforts dedicated to study these promising compounds, a complete understanding of their mechanistic behavior remains elusive. Even more puzzling is identifying which key steps in the interactions of peptides with cell membranes are responsible for bacterial inhibition and cell death. A variety of different interaction mechanisms have been proposed, the outcome of which appears to be heavily influenced by the experimental conditions used, such as lipid composition, membrane fluidity, curvature of lipids, electrolyte concentration, pH, membrane surface charge, and osmotic pressure.7, 10-15 Some studies suggest that peptidemediated membrane disruption can lead to either leakage of key intracellular machinery critical for survival, or disruption of essential metabolic pathways.16-17 Evidence for this stems from


3

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 38

observations that many AMPs permeate the membrane, resulting in the formation of pore-like structures, such as barrel stave and toroidal pores, or from the disruption of the membrane structure in a detergent-like manner.18-19 Alternatively, intracellular components and key proteins may be attacked directly by permeating AMPs, disrupting their function without major damage to the cellular membrane.14, 18, 20 AMPs of radically different sequence and structure can display similar net behavior,21-24 but it is not well known whether one or several simultaneous modes of peptide action occur. Melittin, a 26-residue peptide isolated from the European honeybee Apis mellifera, has been extensively studied since its discovery as a model of AMP behavior.12, 25 It displays amphiphilic character, and adopts an α-helical configuration when bound to artificial lipid membranes.10, 26-27 Studies have demonstrated that melittin initially adopts a parallel orientation to the membrane surface, but at sufficiently high peptide concentrations, orients perpendicularly and bends the membrane as a toroidal pore; a model well described by the work of Huang et al. and Matsuzaki et al..28-32 Peptide-induced reduction of membrane thickness, correlated with an increase in population of peptide in a trans-membrane orientation is observed as the system approaches equilibrium. Dynamics measurements on single vesicles have timed changes in membrane thickness to peptide binding and eventual release of internal contents consistent with a pore formation mechanism.33 Similar investigations have also been performed for other antimicrobial peptides.22, 34 However, to the best of our knowledge, direct simultaneous measurements on pore formation kinetics and contents release have not yet been realized. In the present work, we have measured the dynamics of pore formation of a fluorescent melittin analogue, and internal contents release from single giant vesicles and single bacterial cells. The melittin analogue has a Pro-14 to Lys-14 substitution (P14K), which is further grafted to the


4

Page 5 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

fluorescent probe Alexa Fluor 430. The analogue retains strong α-helical character and lytic behavior, and further displays polarity-sensitive fluorescent lifetime decays, allowing it to report on the local microenvironment experienced by the centre of the peptide helix.35 Based upon our prior work, we have shown that the excited-state lifetime of the Alexa Fluor 430 fluorescence is ~4.8 ns on membrane surfaces and is quenched to ~1.5 ns upon formation of toroidal pores.36 Accordingly, our experimental strategy involved using time-lapse fluorescence lifetime imaging microscopy to determine the population distribution of fluorescent-melittin between pore state and membrane surface state, and simultaneously measure the leakage of an entrapped fluorescent species (Lucifer Yellow from the vesicle; eGFP from bacteria).35 Both Lucifer Yellow and eGFP can be excited and emit at the same wavelength range as the modified peptide analogue, but have distinctly different fluorescence lifetimes, allowing for simultaneous monitoring of peptide and entrapped fluorescent species. The paper is organized as follows. First, we measure the dynamics of pore formation and contents leakage from Giant Unilamellar Vesicles (GUVs) composed of 1,2-dipalmitoyl-sn-glycero-3phosphatidylchloine (DPPC) in the gel phase, which encapsulate the fluorescent dye Lucifer Yellow. Next, measurements were performed on Giant Multilamellar Vesicles (GMVs) composed of DPPC, also encapsulating Lucifer Yellow dye. Finally, the dynamics of pore formation and contents release were measured on single Escherichia coli bacteria that expressed cytosolic enhanced Green Fluorescent Protein (eGFP). These systems, which consistent of membranes with increasing physical (unilamellar and multilamellar) and biological complexity (synthetic versus living), display varying pore formation and leakage dynamics.


5

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 38

MATERIALS AND METHODS Materials 1,2-dipalmitoyl-sn-glycero-3-phosphatidylchloine (DPPC, powder form) was purchased from Avanti Polar Lipids (Alabaster, USA). UV/Vis spectroscopy grade chloroform and methanol, as well as 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES, 99% purity) were purchased from Merck (Darmstadt, Germany). Lucifer Yellow (CH dilithium salt) was purchased from Sigma-Aldrich (Castle Hill, Australia). The melittin analogue, melittin P14K Alexa Fluor 430, was synthesized on an Fmoc-PAL-EG-PS resin, as previously detailed.35 The Alexa Fluor 430 fluorescent label was purchased from Molecular Probes (Carlsbad, USA). All materials were used without further purification. High purity water with a resistivity of 18.2 MΩ (Total Organic Content 1-4 ppb) was generated by an ELGA PureFlex system (ELGA LabWater, United Kingdom), and was used in the preparation of all aqueous solutions. HEPES buffer solutions (10 mM HEPES, pH 7.40 ± 0.05) filtered through 0.22 µm hydrophilic poly(1,1,2,2-tetrafluoroethylene) membranes (Millipore Corp., Bedford, USA) were used in the preparation of all peptide and dye solutions. pH was adjusted using small volumes of concentrated sodium hydroxide (5 M). Lucifer Yellow fluorescent dye solutions (0.5 mM) were covered with aluminum foil to avoid exposure to light sources, and were used immediately after preparation. The melittin analogue solutions were prepared at 5–300 µM concentrations, and were stored at 4 °C in the dark for no longer than two weeks.

Preparation of giant vesicles Giant vesicles composed of pure or doped DPPC were prepared following the solvent evaporation method.37 Briefly, a concentrated solution of DPPC powder dissolved in


6

Page 7 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

spectroscopy grade chloroform (0.1 M) was added to a chloroform:methanol mixture (9.8:1 ratio). Lucifer Yellow fluorescent dye (0.5 mM) in HEPES buffer was then added to the organic phase, with the organic phase subsequently removed via rotary evaporation for 2-3 minutes at 40 rpm, 40°C (±1°C), and 150 mBar minimum pressure. A small volume (1-2 mL) of the resulting vesicle suspension was taken into syringe, with any non-encapsulated dye removed via progressive dilution with 10 mM HEPES, using a 0.45 µm poly(1,1,2,2-tetrafluoroethylene) syringe filter to prevent loss of the vesicles (Millipore Corp., Bedford, USA). Upon successful removal of unencapsulated dye, the vesicle suspension was transferred to glass vials, and stored sideways in the dark at room temperature (23°C ± 3 °C) for no longer than two weeks. Both GUVs and GMVs could be obtained from the same sample. We distinguished between GMVs and GUVS with the following criteria. Both GMVs and GUVs exhibited fluorescence from trapped Lucifer Yellow dye. GMVs were visible under bright-field illumination whilst GUVs were not visible under bright-field illumination.

Preparation of Escherichia coli bacteria Escherichia coli (strain JM109) transfected to self-express cytosolic eGFP were prepared first by inoculating 3 mL D2O Luria Broth with a single colony taken off culture plates and incubated for 3-4 hrs at 37 °C and 120 rpm. 800 µL of this solution was diluted into 200 µL of 80% glycerol solution and stored at -80 °C. Activation of the frozen bacteria was achieved by inoculating a fresh 3 mL solution of D2O Luria Broth using colonies taken from the frozen bacterial pellet, followed by overnight incubation at 37 °C and 120 rpm. Luria agar plates were inoculated with the activated bacterial solution and incubated overnight at 37 °C. Plate grown bacteria were then used to inoculate 3 mL Luria Broth containing


7

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 38

ampicillin (100 µg/mL), followed by further incubation at 37 °C at 120 rpm for 12 hours. The resultant solution was transferred into 100 mL of ampicillin-containing Luria Broth and incubated for 3 hrs. 1 mL of isopropryl-β-D-1-thiogalactopryanoside (0.1 M) was then added, followed by additional incubation for 4 hrs. Finally, centrifugation at 10,000 x g for 1-3 min formed a bacterial pellet, and was stored at -20 °C after removal of the supernatant. Prior to experiments, frozen pellets were resuspended in 5 mL Luria Broth and allowing to thaw for at least one hour. To determine the total lipid content for a given experiment with E. coli cells, the average cell size length and width was determined via optical bright-field microscopy observations using a 100x magnification. Approximating the surface area (SA) of the rod-shaped cell was achieved by combining the surface area of two hemispheres with radius r, and a cylinder with length l and width w (SAE. coli = 2[1/2(4πr2) + (πlw)]). The average length and width of E. coli cells was determined to be 2 µm and 0.5 µm respectively, yielding a total surface area of ~3.14 µm2. This value was doubled owning the presence of inner and outer membrane bilayers in gram-negative E. coli cells. Using an averaged lipid headgroup surface area size of 2x10-8 µm2, the total lipid molecules per bacterial cell could then be determined. In a 0.5 mL sample of E. coli, it was found there were approximately 108 viable bacterial cells. Multiplying the number of viable cells by the approximate number of lipid molecules per cell enabled the determination of the total lipid content within a typical sample. This then enabled the determination of the global L:P ratio, which could be achieved by altering the amount of peptide delivered to the system.

Frequency-domain Fluorescence Lifetime Imaging Microscopy measurements


8

Page 9 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

All fluorescence lifetime imaging experiments were performed in duplicate using frequencydomain Lambert Instruments Fluorescence Lifetime Attachment (Lambert Instruments, Leutingwolde, The Netherlands) interfaced with an inverted microscope (Ti Eclipse, Nikon Inc., Japan). 200-300 µL of giant vesicle suspension was aliquoted into an 8-well sample chamber at room temperature (~23-26 °C, i.e. vesicles in the gel phase), and allowed to settle for ~1 hr. Giant vesicles were then observed in situ through a 100x NA 0.75 oil objective (Nikon Plan-Fluor, Nikon Inc, Japan). The fluorescence excitation source was a 448 nm 1 W LED (Lambert Instruments, Leutingwolde, The Netherlands) set at a modulation frequency of 35 MHz, for excitation of Alexa Fluor 430 and Lucifer Yellow. The emission signal was first passed through a HSi-400 HyperSpectral Imaging unit (Gooch & Housego, Ilminster, United Kingdom) before reaching the detector, and was set to a collection wavelength of 530 nm with a 18.7 nm bandwidth for Alexa Fluor 430/Lucifer Yellow fluorescence. Peptide was dispensed to the vesicle solution via micro-syringe (dead-time 30-60 seconds) prior to FLIM acquisition. Phase and modulation lifetimes, τφ and τM, respectively, were determined by acquisition of a phase image stack (12 phase images per recording) taken periodically. Photobleaching was corrected by using acquiring the phase stack using a random permutation of recording order procedure.38 The reference used for all lifetime determinations was Rhodamine 6G (lifetime τ: 4.1 ns).39 Fluorescence lifetime image stacks were recorded every 1-2 minutes, for the first 15 minutes after the introduction of peptide, followed by 2-3 minute intervals from the 15 minute mark for up to 2 hours. All experiments on giant vesicles were carried out at global lipid-to-peptide concentration ratios of 50:1, 25:1 and 5:1. For fluorescence lifetime studies on E. coli cells, samples were prepared in a similar fashion to the giant vesicles. Thawed E. coli samples (0.5 mL) were spun down at 10,000 x g for 5 minutes.


9

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 38

The Luria Broth supernatant was removed and replaced with HEPES buffer (0.5 mL), followed by gentle shaking to re-suspend the bacteria. 200 µL of the re-suspended solution was aliquoted into individual wells of an 8-well chamber, and left undisturbed for 2 hrs to allow adhesion of the cells to the well surface. HEPES buffer was then siphoned from the sample, followed by addition of 150 µL melittin analogue solution prepared at sufficient concentration to yield L:P ratios from 2.5:1 to 60:1. FLIM images for the E. coli experiments were acquired using a 474 nm 1 W LED, with the HSi-400 HyperSpectral Imaging unit set to collect at 540 nm, with a 23.30 nm bandwidth. A single sample was monitored for a minimum of two hours, and up to four.

FLIM data analysis Areas representing individual cells or vesicles were analyzed using the region of interest selection tool provided by Lambert instruments LIFA software. Pixel-by-pixel data were averaged spatially over the specified region of interest to obtain cell or vesicle averaged information. Data exported from the software consisted of lists of phase lifetimes (τφ, in ns) and modulation lifetimes (τM, in ns) as a function of time. These lifetimes were converted to modulation (m) and phase (φ) values with the formulae: φ = arctan(ωτφ)

(1)

m = [(1+(ωτM)2)-1]0.5 (2) where ω (fundamental circular frequency) = 0.2514 ns-1. The FLIM data were further transformed into phasor space, where x-axis = B = mcosφ, and y-axis = A = msinφ. The phasor space is convenient because complex fluorescence can be presented by a simple point on the phasor plot, with changes in fluorescence over time represented by a trajectory in phasor space.


10

Page 11 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Moreover decomposition of complex fluorescence into biologically or physically relevant states can be performed. To further decompose the fluorescence from fluorescent peptide and fluorescent cells (or vesicles) into fractional states, the fractional fluorescence from surface peptide and peptide pore states were computed as a function of time. This analysis assumes that the phasors of the surface peptide state (Bsurf,Asurf), pore peptide state (Bpore,Apore) and entrapped dye molecules (Bdye,Adye) are time-invariant but the relative fluorescence contributions from these states is time dependent.

For a given measured phasor at a specific time t, with components, B(t)=mtcosφt and A(t)= mtsinφt (where mt and φt are the values of m and φ at time t respectively), the fractional fluorescence from the peptide pore state, fpore(t), the fraction fluorescence from the peptide surface state, fsurf(t) and the fractional fluorescence from intracellular (or intra-vesicular) dye, fdye(t) is given by:40

fsurf(t)=[(A(t)-Adye)(Bpore-Bdye)-(B(t)-Bdye)(Apore-Adye)]/den (3)

fpore(t)=[(B(t)-Bdye)(Asurf-Adye)-(A(t)-Adye)(Bsurf-Bdye)]/den (4) den=(Asurf-Adye)(Bpore-Bdye)-(Bsurf-Bdye)(Apore-Adye) (5)

fdye(t)=1- fsurf(t)- fpore(t)

(6)

The values of the dye or GFP phasors were obtained by direct measurement of a solution of Lucifer Yellow (Adye = 0.486, Bdye = 0.381) and bacteria expressing GFP (Adye = 0.439, Bdye = 0.739). The phasor corresponding to the pore peptide state (Apore = 0.298, Bpore = 0.902) was obtained from several sources, (i) associative anisotropy decay analysis of peptide in supported


11

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 38

bilayers which assigned a short lifetime to restricted motion as expected for a pore state, (ii) observation of increase in quenched lifetime state upon interaction of peptide with membranes,36, 41

(iii) observation of peptide aggregates with short lifetime in super-concentrated solution of

peptide.42 The surface peptide phasor (Asurface = 0.504, Bsurface = 0.478) was deduced from previous frequency-domain measurements of peptide in lipid vesicles,43 and time-resolved measurements of peptide in methanol (dieletric constant similar to the headgroup region of lipid bilayer).35 This model assumes that the Alexa Fluor 430 lifetimes for pore state and surface state are invariant to membrane-type (i.e. synthetic lipid versus bacterial cell wall). We think this approximation is reasonable. The dielectric constant of membrane interfaces has been reported to be in the range of 10-30. Our previous photo-physical characterization of Alexa Fluor 430 revealed that the lifetime of the probe was 4.5-5.0 ns for moderately polar solvents pyridine, acetonitrile, ethanol and methanol. Thus the expected variation in lifetime is about 10%. The Alexa Fluor 430 pore state has a short fluorescence lifetime that is not expected to depend on environment because it is caused by close proximity of probes, not by differences in dielectric constant of the surrounding environment.

RESULTS Dynamics of peptide-induced pore formation and leakage of Giant Unilamellar Vesicles Results of time-lapse fluorescence lifetime imaging microscopy experiments on GUVs entrapped with Lucifer Yellow are collected in Figure 1. At the 50:1 L:P concentration ratio, changes in the fluorescence proportion of peptide surface, peptide pore and entrapped fluorescent species were observed over a time range of two hours. As shown in Figure 1d, after peptide addition at time zero (t = 0min), a gradual increase in peptide surface state was observed (from 0 to 0.75 over a 120 min period). After a lag phase of about 40 minutes, a sudden increase in pore


12

Page 13 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

state proportion from 0 to 0.13 was observed. At this moment, the entrapped fluorescent dye decreased dramatically (from 0.75 to 0.25) consistent with a sudden loss of vesicle contents. The intensity data shown is the fluorescence intensity of both the peptide and the entrapped dye averaged over the vesicle. Variations in the initial intensity values can occur between experiments due to differing concentrations of entrapped dye, as well as due to the relative rates of peptide binding, pore formation and leakage of entrapped contents. Regardless, the behavior of the intensity as shown in Figure 1e confirms that at the point of initial pore formation, a decrease in total intensity is observed. Inspection of the phasor trajectory on the phasor plot (Figure 1f) reveals peptide binding to membrane surface and pore formation as distinct, resolved steps on our experimental time-scale. The net vector on the phasor diagram reveals an overall increase in pore states at the expense of entrapped dye. However, after vesicle release there appeared to be a further increase in pore states over time. This suggests that vesicle release occurs once a critical concentration (or relative concentration of pores is formed). At the 5:1 L:P ratio (i.e. ten-fold more peptide), it appears that the same interaction process is occurring as the 50:1 L:P system but at a dramatically increased rate. This can be readily seen by observing the fractional fluorescence population distribution over time (Figure 1a). Within the first few minutes after peptide addition, there is a detectable level of peptide pore states (0.13), which is again correlated with a sudden, rapid decrease in the entrapped fluorescence dye, coinciding with a distinct decrease in the fluorescence intensity (Figure 1b). Images of the GUV fluorescence signals (Figure 2) show that rapid loss of internal fluorescence contents occurs between 5 and 6 minutes after image acquisition leaving a characteristic pattern of fluorescence at the membrane due to the peptide. The system appears to have already reached equilibrium within ~35 minutes, as indicated by the fraction of pore peptide and surface peptide fluorescence


13

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 38

populations approaching a steady state (Figure 1b). Analysis of the fluorescence lifetime trajectory plotted in phasor space (Figure 1c) reveals that at time t = 0 min, there is already a significant proportion of pore and surface peptide states but with a further increase in the pore state over time. As above, the net vector on the phasor diagram reveals an overall increase in pore states at the expense of trapped dye. To summarize, for the 50:1 L:P and 5:1 L:P systems, sudden vesicle contents release is triggered at the point when the fluorescence fraction pore states is 0.12-0.13 and the surface fraction is 0.4-0.5. Taking into account the differences in fluorescence lifetimes of the two-states, this corresponds to a molar ratio of between 0.8:1 to 1:1.

Dynamics of peptide induced pore formation and Giant Multilamellar Vesicle leakage At 50:1 L:P concentration ratio, the fractional fluorescence population distribution for GMVs (Figure 3d) is markedly different to that of the corresponding GUV data. Membrane peptide pores remain at a low level steady-state population over the two-hour observation period (< 0.1). Consequently, there is slow, gradual leakage of the entrapped fluorescent dye (Figure 3d & e) indicated by the change in fluorescence fraction from 0.9 to 0.5 over 2 hours, which is correlated with increased surface binding of peptide. The lifetime trajectory on the phasor plot (Figure 3f) shows only surface binding with little pore formation and incomplete vesicle contents leakage on the observation time-scale. At the 5:1 L:P ratio, a small fraction of peptide pores are rapidly formed upon addition of the peptide to the vesicle (Figure 3a). Shortly afterward, slow leakage of the entrapped fluorescent dye occurs, but there is no correlative increase in the population of pore states. The relatively slow dynamics of the fluorescence changes is also evident in the images of the GMVs at selected


14

Page 15 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

time points (Figure 4). The peptide pore fraction remains at a constant level of 0.12 over the twohour observation period (Figure 3a). However, there is a measurable increase in the peptide surface state population (from 0.25 to 0.75), which does appear to be correlated with the loss of entrapped fluorescent dye (from 0.5 to 0.1). The lifetime trajectory on the phasor plot clearly shows a complete loss of all the entrapped dye without any change to the population of the peptide pore state. The net vector in phasor space indicates loss of entrapped dye is correlated with an increase in peptide surface state (at constant pore level). This can also be seen from the decrease in fluorescence lifetime of the GMV fluorescence image from 7 ns for the entrapped dye state (at time=0 min) to about 4.5 ns characteristic of the peptide surface state (at time=120min) (Figure 4b). In summary, release of vesicle contents is gradual and apparently correlated with an increase in peptide surface states with the pore states remaining at a low time-invariant level.

Dynamics of pore formation and peptide-induced leakage in E. coli bacteria Initial deconvolution of the fractional fluorescence from E. coli yielded scattered data that made it difficult to determine any behavioral trends. This is likely due to the relatively low intensity values of the E. coli data, as seen in Figure 5c,f,j. In order to smooth out the experimental data, a simple exponential function of the form “y = Ze(-t/s) + offset” was fitted to the phasor trajectory data values (Figure 5 a,d,i), where Z and s are arbitrary fitting values, using the ProFit curve fitting function (Levenberg-Marquardt algorithm, software version 6.2.16). The generated AB values from the fit were then used in the fractional fluorescence deconvolution as normal, in order to reveal the behavior of the peptide interaction.


15

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 38

Remarkable consistency in the behavior of the peptide interacting with the E. coli cells was observed across all L:P regimes studied. There was an initial rise in the fraction of pore states within the E. coli cells, coinciding with a decrease in the peptide surface states (Figure 5 b,e,h). Once the pore states were detected, there was an initiation of some eGFP leakage, as indicated by the sudden decrease in fraction of eGFP emission. Complete leakage was not observed. The movement of the fluorescence lifetime data within the phasor space (Figure 6) proceeded towards a stronger contribution of a quenched fluorescent pore state over time at the expense of eGFP fluorescence. This also occurred at all L:P ratios studied. Figure 7 shows typical images of cells at selected time points after peptide addition. These images confirm that only partial leakage of contents occurs from the bacterial cells. An interesting observation in the fluorescence intensity data for the 20:1 L:P ratio data (Figure 5j) is that there is an apparent increase in the intensity value, which appears to be at odds with the rest of the experimental data. We attribute this due to the very low initial intensity values of the cytosolic eGFP within the E. coli cells in this particular set of samples. In this case addition of peptide to the cells will result in an initial increase in total fluorescence reflecting the combination of peptide and GFP fluorescence. However, we are still able to deconvolve the contributions of each fluorescent state using the FLIM. To ensure the data was representative, several individual cells were subject to the same analysis. Figure 8 shows three individual E. coli cells all exposed to an L:P ratio of 20:1, again revealing considerable consistency in phasor component values and fractional fluorescence. Analysis for the 10:1 and 2.5:1 L:P ratio data performed in the same manner revealed similar insights (data not shown).


16

Page 17 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

DISCUSSION We have measured the dynamics of the interaction of a model peptide with single vesicles and cells. Significantly, we have been able to determine the dynamics of peptide surface-state to peptide pore-state transitions and correlate the populations of these states with vesicle or cell contents release in parallel. It is important to stress that the results of the experiments with the vesicles were obtained with lipids in the gel phase to be consistent with our prior work. Thus the results in terms of dynamics may not necessarily be representative of peptide-membrane behavior in fluid phase systems. The kinetics data on the GUVs are consistent with a model where pore formation is correlated with vesicle contents release. As peptide binds to the membrane surface there is a critical density at which the peptides begin to form pores. For both 50:1 L:P and 5:1 L:P systems, dye release is triggered at the point when the pore-to-surface mole concentration ratio of 0.8:1 to 1:1 i.e. when the concentration of pore states and surface states are approximately equal. After dye leakage, the pore-to-surface mole concentration ratio increases to beyond 1:1, indicating that dye release occurs at a point prior to the system reaching steady-state equilibrium. These observations are in accordance with the toroidal pore model for melittin as first proposed by Huang et al.,28 and recently explored in more detail by Lee et al..44 In this model, the surface state of the peptide initially has a lower free energy than the trans-membrane state, owing to the energetic cost of forming a pore in the membrane. As more peptide binds in the surface state, the membrane experiences a distortion that causes the membrane to thin, decreasing the free energy difference between the trans-membrane pore state and the surface state. Eventually as enough surface peptide binds, the energies of the two states are equal and the trans-membrane pore state forms, akin to a second-order phase transition. It is at this point of coexistence of the surface and pore


17

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 38

states (i.e. at pore to surface concentration ratio close to 1:1) that we observe the sudden release of contents from the GUVs. For the GMVs, the observed kinetics indicate different behavior of the peptide. The pore levels remained at a low steady-state level, with a very slow increase in surface state over time, correlated with long time-scale biphasic vesicle contents release. These dynamics are suggestive of a more complex process or series of processes leading to graded release of vesicle contents. From the measurements on the unilamellar system we know that in order for pore formation to occur and dye to be released, a critical surface density of peptides must first be achieved at the outermost bilayer. In a multilamellar system, however, every membrane bilayer would have to have its own surface peptide concentration exceeded before pore formation can occur; bilayers deeper in the stack are effectively protected until the upper layers have been permeated (Figure 9). Overall, this creates a series of equilibria between several bilayers that does not exist in the unilamellar case; peptide participating in pore formation in one bilayer can translocate across and become readily accessible to the next bilayer, which then goes through the same series of interaction steps. In effect, up to the point of equilibrium the pore state serves as a kinetic intermediate being continuously formed and broken up. This can help explain the pseudo-steady state behavior of the pore state, the correlation between surface binding and vesicle release and the slow kinetics of the later. Another possible consequence of the higher ordered multilamellar structure is that when the peptide is able to permeate the membrane structure, disruptions to the overall structure would incur a significantly higher energetic cost, as more phospholipid molecules are affected by deformations in the packing structure. A rough measure of this change in effective stability can be made if we assume that the 5:1 L:P system reaches equilibrium at 2


18

Page 19 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

hours. From the fraction pore of 0.125 and the fraction of surface state of 0.75, and the calculations above, we obtain a pore to surface concentration ratio of 0.5:1, indicating somewhat less stable pore formation than the unilamellar vesicles in this instance. The dynamics of pore formation and eGFP release in the individual E. coli bacteria contains some elements in common with both the GUV and GMV data, but neither synthetic system replicates the observed dynamics entirely. For the live bacterial cells, a decrease in surface states, an increase in pore formation and partial leakage occurs within 20 minutes of peptide addition. As for the GUVs, the eGFP release is triggered from the bacterial cells once the fractional fluorescence from pores reaches about 0.06-0.13, the surface state proportions are 0.3-0.5 and the pore to surface concentration ratio at eGFP release is in the range 0.5:1 to 1:1. Unlike the GUVs, however, the bacteria display only partial release of contents over the time scale of examination. Pores are formed but their ability to release eGFP appears to be time-dependent. This may be accounted for by the complex bi-lamellar structure of the gram-negative E. coli cell. The outer membrane is composed largely of lipopolysaccharide, which is known to hamper dye leakage in model membranes.45 It is further interesting to note that Rangarajan et al.46 observed slow diffusion of Cecropin A in E. coli cells past the LPS layer, accounting for the observed lag phase in outer membrane permeabilization. The formation of smaller, stable pores comparable to the size of ~100 nm located in the outer membrane once a critical threshold concentration was reached was also observed, but this stability was not explicitly stated to also exist in the cytoplasmic membrane. It is tempting to speculate that stable pores are being formed in the outer membrane in our E. coli system, owing to the presence of LPS, but that only limited transient pores are formed in the inner membrane, perhaps due to outer membrane protection and interference from the periplasmic space.


19

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 38

Unlike artificial vesicles, bacterial cells are living, dynamic systems, capable of adapting to changes within the local environment. Bacteria in particular have a vast array of additional defenses in case of attack, such as porins and efflux pumps to control the flow of molecules and proteins in and out of the cell, enzymes to cleave attacking peptides, or, even the capacity to produce outer membrane vesicles as membrane decoys.47-50 Modification of the lipid membrane composition can also be employed under stressful stimuli, or to evade attack by host immune responses, which includes antimicrobial peptides.51 All of these mechanisms act to protect the living cell by either preventing the peptides from reaching the membrane target, or by removing peptide interacting with the membrane before it can permeate the critically important cytoplasm.

CONCLUSIONS Our work here has investigated the peptide-membrane interaction of a melittin analogue with two vesicle model membranes, and gram-negative E. coli bacteria, with simultaneous monitoring of both the fluorescent peptide and entrapped fluorescent dyes. Analysis of the fluorescence lifetime imaging microscopy data reveals a common pore formation mechanism to all systems studied, but that the stability of the membrane pores formed varies depending on the permeability of the membrane structure. Giant Multilamellar Vesicles in the gel phase could only facilitate the formation of pores at low population relative to surface peptide states, whilst Giant Unilamellar Vesicles facilitated more stable pores with a reasonable population fraction. Although gramnegative E. coli are expected to behave similarly to multilamellar systems, we find that the peptide-membrane interaction is more akin to that seen in GUVs with the formation of highly stable pores, but with significantly reduced leakage of internal contents. We propose that the complex dynamics of the living cell prevent attack of the inner membrane, which is required for


20

Page 21 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

leakage of the eGFP and intracellular contents from the cytoplasm. The specific mechanisms through which the E. coli accomplishes this for this particular system is not known, but may well be one of several potential adaptive mechanisms displayed by other bacteria in response to antimicrobial peptide attack.

AUTHOR INFORMATION Corresponding Author *E-mail address: [email protected] (A.H.A.C.); Tel.: (+61)9214-5719. Fax: (+61)92145435. Notes ABBREVIATIONS GUV, Giant Unilamellar Vesicle; GMV, Giant Multilamellar Vesicle; FLIM, Fluorescence Lifetime Imaging Microscopy; L:P, lipid-to-peptide (concentration ratio); L:P* Critical lipid-topeptide (concentration ratio); DPPC, 1,2-dipalmitoyl-sn-glycero-3-phosphochloine; HEPES, 4(2-hydroxyethyl)-1-piperazineethanesulfonic acid; eGFP enhanced Green Fluorescent Protein ACKNOWLEDGEMENTS AHAC and MLG gratefully thank the Australian Research Council for their generous financial support of this project in the form of Discovery Grant DP110100164.


21

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 38

REFERENCES 1.

Maragakis, L. L.; Perencevich, E. N.; Cosgrove, S. E. Clinical and Economic Burden of

Antimicrobial Resistance. Expert Rev. Anti Infect. Ther. 2008, 6, 751-763. 2.

Smith, R. D.; Coast, J. Antimicrobial Resistance: A Global Response. Bull. World Health

Organ. 2002, 80, 126-133. 3.

Cosgrove, S. E.; Carmeli, Y. The Impact of Antimicrobial Resistance on Health and

Economic Outcomes. Clin. Infect. Dis. 2003, 36, 1433-1437. 4.

Hirsch, E. B.; Tam, V. H. Impact of Multidrug-Resistant Pseudomonas Aeruginosa

Infection on Patient Outcomes. Expert Rev. Pharmacoecon. Outcomes Res. 2010, 10, 441-451. 5.

Kruse, T.; Kristensen, H. H. Using Antimicrobial Host Defense Peptides as Anti-Infective

and Immunomodulatory Agents. Expert Rev. Anti Infect. Ther. 2008, 6, 887-895. 6.

Jenssen, H.; Hamill, P.; Hancock, R. E. W. Peptide Antimicrobial Agents. Clin.

Microbiol. Rev. 2006, 19, 491-511. 7.

Pasupuleti, M.; Schmidtchen, A.; Malmsten, M. Antimicrobial Peptides: Key

Components of the Innate Immune System. Crit. Rev. Biotechnol. 2012, 32, 143-171. 8.

Oppenheim, J. J.; Biragyn, A.; Kwak, L. W.; Yang, D. Roles of Antimicrobial Peptides

Such as Defensins in Innate and Adaptive Immunity. Ann. Rheum. Dis. 2003, 62, 17-21. 9.

Zaiou, M. Multifunctional Antimicrobial Peptides: Therapeutic Targets in Several Human

Diseases. Journal of Molecular Medicine-Jmm 2007, 85, 317-329.


22

Page 23 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

10. Shai, Y. Mechanism of the Binding, Insertion and Destabilization of Phospholipid Bilayer Membranes by α -Helical Antimicrobial and Cell Non-Selective Membrane-Lytic Peptides. Biochim. Biophys. Acta 1999, 1462, 55-70. 11. Terra, R. M. S.; Guimaraes, J. A.; Verli, H. Structural and Functional Behavior of Biologically Active Monomeric Melittin. J. Mol. Graph. Model. 2007, 25, 767-772. 12. Dempsey, C. E. The Actions of Melittin on Membranes. Biochim. Biophys. Acta 1990, 1031, 143-161. 13. Benachir, T.; Lafleur, M. Osmotic and pH Transmembrane Gradients Control the Lytic Power of Melittin. Biophys. J. 1996, 70, 831-840. 14. Teixeira, V.; Feio, M. J.; Bastos, M. Role of Lipids in the Interaction of Antimicrobial Peptides with Membranes. Prog. Lipid Res. 2012, 51, 149-177. 15. Allende, D.; Simon, S. A.; McIntosh, T. J. Melittin-Induced Bilayer Leakage Depends on Lipid Material Properties: Evidence for Toroidal Pores. Biophys. J. 2005, 88, 1828-1837. 16. Hancock, R. E. W.; Sahl, H.-G. Antimicrobial and Host-Defense Peptides as New AntiInfective Therapeutic Strategies. Nat Biotech 2006, 24, 1551-1557. 17. Diamond, G.; Beckloff, N.; Weinberg, A.; Kisich, K. O. The Roles of Antimicrobial Peptides in Innate Host Defense. Curr. Pharm. Des. 2009, 15, 2377-2392. 18. Brogden, K. A. Antimicrobial Peptides: Pore Formers or Metabolic Inhibitors in Bacteria? Nature Reviews Microbiology 2005, 3, 238-250.


23

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 38

19. Matsuzaki, K. Why and How Are Peptide-Lipid Interactions Utilized for Self-Defense? Magainins and Tachyplesins as Archetypes. Biochim. Biophys. Acta 1999, 1462, 1-10. 20. Sochacki, K. A.; Barns, K. J.; Bucki, R.; Weisshaar, J. C. Real-Time Attack on Single Escherichia Coli Cells by the Human Antimicrobial Peptide LL-37. Proc. Natl. Acad. Sci. USA 2011, 108, E77-E81. 21. Lohner, K.; Staudegger, E.; Prenner, E. J.; Lewis, R.; Kriechbaum, M.; Degovics, G.; McElhaney, R. N. Effect of Staphylococcal δ-Lysin on the Thermotropic Phase Behavior and Vesicle Morphology of Dimyristoylphosphatidylcholine Lipid Bilayer Model Membranes. Differential Scanning Calorimetric, P-31 Nuclear Magnetic Resonance and Fourier Transform Infrared Spectroscopic, and X-Ray Diffraction Studies. Biochemistry 1999, 38, 16514-16528. 22. Tamba, Y.; Yamazaki, M. Single Giant Unilamellar Vesicle Method Reveals Effect of Antimicrobial Peptide Magainin 2 on Membrane Permeability. Biochemistry 2005, 44, 1582315833. 23. Last, N. B.; Miranker, A. D. Common Mechanism Unites Membrane Poration by Amyloid and Antimicrobial Peptides. Proc. Natl. Acad. Sci. USA 2013, 110, 6382-6387. 24. Hristova, K.; Dempsey, C. E.; White, S. H. Structure, Location, and Lipid Perturbations of Melittin at the Membrane Interface. Biophys. J. 2001, 80, 801-811. 25. Raghuraman, H.; Chattopadhyay, A. Melittin: A Membrane-Active Peptide with Diverse Functions. Biosci. Rep. 2007, 27, 189-223.


24

Page 25 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

26. Wessman, P.; Morin, M.; Reijmar, K.; Edwards, K. Effect of α-Helical Peptides on Liposome Structure: A Comparative Study of Melittin and Alamethicin. J. Colloid Interface Sci. 2010, 346, 127-135. 27. Toraya, S.; Nishimura, K.; Naito, A. Dynamic Structure of Vesicle-Bound Melittin in a Variety of Lipid Chain Lengths by Solid-State Nmr. Biophys. J. 2004, 87, 3323-3335. 28. Huang, H. W. Action of Antimicrobial Peptides: Two-State Model. Biochemistry 2000, 39, 8347-8352. 29. Matsuzaki, K.; Yoneyama, S.; Miyajima, K. Pore Formation and Translocation of Melittin. Biophys. J. 1997, 73, 831-838. 30. Lee, M. T.; Hung, W. C.; Chen, F. Y.; Huang, H. W. Mechanism and Kinetics of Pore Formation in Membranes by Water-Soluble Amphipathic Peptides. Proc. Natl. Acad. Sci. USA 2008, 105, 5087-5092. 31. Matsuzaki, K.; Murase, O.; Fujii, N.; Miyajima, K. An Antimicrobial Peptide, Magainin 2, Induced Rapid Flip-Flop of Phospholipids Coupled with Pore Formation and Peptide Translocation. Biochemistry 1996, 35, 11361-11368. 32. Huang, H. W.; Chen, F.-Y.; Lee, M.-T. Molecular Mechanism of Peptide-Induced Pores in Membranes. Phys. Rev. Lett. 2004, 92, (198304) 1-4 33. Mally, M.; Majhenc, J.; Svetina, S.; Zeks, B. The Response of Giant Phospholipid Vesicles to Pore-Forming Peptide Melittin. Biochim. Biophys. Acta 2007, 1768, 1179-1189.


25

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 38

34. Tamba, Y.; Ohba, S.; Kubota, M.; Yoshioka, H.; Yamazaki, M. Single GUV Method Reveals Interaction of Tea Catechin (-)-Epigallocatechin Gallate with Lipid Membranes. Biophys. J. 2007, 92, 3178-3194. 35. Rapson, A. C.; Hossain, M. A.; Wade, J. D.; Nice, E. C.; Smith, T. A.; Clayton, A. H. A.; Gee, M. L. Structural Dynamics of a Lytic Peptide Interacting with a Supported Lipid Bilayer. Biophys. J. 2011, 100, 1353-1361. 36. Burton, M. G.; Huang, Q. M.; Hossain, M. A.; Wade, J. D.; Clayton, A. H. A.; Gee, M. L. Long Time-Scale Interaction Dynamics between a Model Antimicrobial Peptide and Giant Unilamellar Vesicles. Langmuir 2013, 29, 14613-21. 37. Moscho, A.; Orwar, O.; Chiu, D. T.; Modi, B. P.; Zare, R. N. Rapid Preparation of Giant Unilamellar Vesicles. Proc. Natl. Acad. Sci. USA 1996, 93, 11443-11447. 38. Van Munster, E. B.; Gadella, T. W. J. Φflim: A New Method to Avoid Aliasing in Frequency-Domain Fluorescence Lifetime Imaging Microscopy. J. Microsc.-Oxf. 2004, 213, 2938. 39. Magde, D.; Rojas, G. E.; Seybold, P. G. Solvent Dependence of the Fluorescence Lifetimes of Xanthene Dyes. Photochem. Photobiol. 1999, 70, 737-744. 40. Clayton, A. H. A.; Hanley, Q. S.; Verveer, P. J. Graphical Representation and Multicomponent Analysis of Single-Frequency Fluorescence Lifetime Imaging Microscopy Data. J. Microsc.-Oxf. 2004, 213, 1-5.


26

Page 27 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

41. Gee, M. L.; Burton, M.; Grevis-James, A.; Hossain, M. A.; McArthur, S.; Palombo, E. A.; Wade, J. D.; Clayton, A. H. A. Imaging the Action of Antimicrobial Peptides on Living Bacterial Cells. Sci. Rep. 2013, 3, 42. Burton, M. Probing Lipid-Peptide Interactions in Model Systems Via Fluorescence Lifetime Imaging Microscopy. Master of Science (Thesis), University of Melbourne, Parkville, Australia 2011. 43. Ningsih, Z.; Hossain, M. A.; Wade, J. D.; Clayton, A. H. A.; Gee, M. L. Slow Insertion Kinetics During Interaction of a Model Antimicrobial Peptide with Unilamellar Phospholipid Vesicles. Langmuir 2012, 28, 2217-2224. 44. Lee, M. T.; Sun, T. L.; Hung, W. C.; Huang, H. W. Process of Inducing Pores in Membranes by Melittin. Proc. Natl. Acad. Sci. USA 2013, 110, 14243-14248. 45. Allende, D.; McIntosh, T. J. Lipopolysaccharides in Bacterial Membranes Act Like Cholesterol in Eukaryotic Plasma Membranes in Providing Protection against Melittin-Induced Bilayer Lysis. Biochemistry 2003, 42, 1101-1108. 46. Rangarajan, N.; Bakshi, S.; Weisshaar, J. C. Localized Permeabilization of E. Coli Membranes by the Antimicrobial Peptide Cecropin A. Biochemistry 2013, 52, 6584-6594. 47. Pages, J. M.; James, C. E.; Winterhalter, M. The Porin and the Permeating Antibiotic: A Selective Diffusion Barrier in Gram-Negative Bacteria. Nature Reviews Microbiology 2008, 6, 893-903. 48. MacDonald, I. A.; Kuehn, M. J. Stress-Induced Outer Membrane Vesicle Production by Pseudomonas Aeruginosa. J. Bacteriol. 2013, 195, 2971-2981.


27

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 28 of 38

49. MacDonald, I. A.; Kuehn, M. J. Offense and Defense: Microbial Membrane Vesicles Play Both Ways. Res. Microbiol. 2012, 163, 607-618. 50. El-Halfawy, O. M.; Valvano, M. A. Non-Genetic Mechanisms Communicating Antibiotic Resistance: Rethinking Strategies for Antimicrobial Drug Design. Expert Opinion on Drug Discovery 2012, 7, 923-933. 51. Zhang, Y. M.; Rock, C. O. Membrane Lipid Homeostasis in Bacteria. Nature Reviews Microbiology 2008, 6, 222-233.


28

Page 29 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

TOC Graphic


29

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 30 of 38

FIGURE 1 - Dynamics of peptide pore formation and GUV leakage by time-lapse FLIM (a,d) Fractional fluorescence of surface peptide (blue), pore peptide (red) and entrapped dye (green) as a function of time. (b,e) fluorescence intensity (arbitrary units) of the whole vesicle as a function of time. (c,f) Phasor trajectory plot, so-called AB plot, for 5:1 and 50:1 L:P respectively. Fitted lines in fractional fluorescence plots are for clarity of trends only. The arrow in the phasor trajectory plots indicates the net direction of the data vector over time (initial to final).


30

Page 31 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

FIGURE 2 – Time-lapse images of the melittin analogue interacting with a DPPC GUV under, a) fluorescence intensity, b) fluorescence lifetime, and c) optical bright-field microscopy at the indicated time intervals t(min). The GUV appears to be invisible with no contrast under brightfield illumination. Intensity for fluorescence intensity and bright-field images are in arbitrary units. Global L:P ratio is 5:1. Scale bar = 5 µm


31

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 32 of 38

FIGURE 3 - Dynamics of peptide pore formation and GMV leakage by time-lapse FLIM (a,d) Fractional fluorescence of surface peptide (blue), pore peptide (red) and entrapped dye (green) as a function of time. (b,e) Fluorescence intensity (arbitrary units) of the whole vesicle as a function of time. (c,f) Phasor trajectory for 5:1 and 50:1 L:P respectively. Fitted lines in fractional fluorescence plots are for clarity of trends only. The arrow in the phasor trajectory plots indicates the net direction of the data vector over time (initial to final).


32

Page 33 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

FIGURE 4 – Time-lapse images of the melittin analogue interacting with a DPPC GMV under, a) fluorescence intensity, b) fluorescence lifetime, and c) optical bright-field microscopy at the time intervals “t” specified. The GMV membrane generates a visible contrast ring. Intensity for fluorescence intensity and bright-field images are in arbitrary units. Global L:P ratio is 5:1. Scale bar = 5 µm


33

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 34 of 38

FIGURE 5 - Dynamics of pore formation and eGFP leakage from single E. coli cells (a,d,h) Sine (red, A = msin(φ)) and Cosine (blue, B = mcos(φ)) components of the phasor as a function of time. Solid line is the fit of an exponential plus offset function. (b,e,i) Fractional fluorescence of eGFP (green), peptide surface state (blue) and peptide pore state (red) as a function of time. (c,f,j) fluorescence intensity (arbitrary units) as a function of time. L:P ratios are 2.5:1, 10:1, and 20:1 respectively.


34

Page 35 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

FIGURE 6 - Phasor plot representation of the fluorescence from the fluorescent melittin analogue interacting with E. coli expressing cytosolic eGFP at the specified L:P concentration ratio. Data corresponds to Figure 5. The arrow in the phasor plots indicates the net direction of the phasor (vector) over time (initial to final).


35

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 36 of 38

Figure 7 – Time-lapse images of a) Fluorescence intensity, and b) fluorescence lifetime of E. coli bacteria containing cytosolic eGFP, exposed to the fluorescent melittin analogue at various time intervals t(min). Global L:P ratio is 2.5:1. Scale bar = 5 µm.


36

Page 37 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

FIGURE 8 - Plot of sine (red, A = msin(φ) and cosine (blue, B = mcos(φ) components of the phasor as a function of time for the fluorescence lifetime data obtained from three individual E. coli cells (a) to c)) at L:P 20:1, along with the corresponding fractional fluorescence plots. There is remarkable consistency and stability in the values of the phasor components within the cell after ~30 minutes. Data points in Fractional Fluorescence plots are: peptide surface state (blue diamonds) peptide pore state (red squares) and eGFP (green inverted triangles).


37

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 38 of 38

FIGURE 9 – Schematic representation of the peptide criticality effect in GMVs. Peptide binds with the outermost bilayer akin to a unilamellar vesicle, until the critical ratio L:P* (L:P* 1) is reached and insertion occurs. Once peptide is able to translocate across the membrane, additional peptide encounters the next bilayer, which in turn must exceed L:P* (L:P* 2) for the same process to occur. Both bilayers exist in equilibrium to each other. This will dramatically reduce the total number of possible pore structures for a fixed amount of available peptide.


38

Direct Measurement of Pore Dynamics and Leakage Induced by a

Recommend Documents