Synergy on Surfaces: Anti-Biofouling Interfaces Using Surface

Subscriber access provided by Kaohsiung Medical University

Biological and Environmental Phenomena at the Interface

Synergy on Surfaces: Anti-Biofouling Interfaces using SurfaceAttached Antimicrobial Peptides PGLa and Magainin-2 Nitzan Shtreimer Kandiyote, Gunasekaran Mohanraj, Canwei Mao, Roni Kasher, and Christopher J. Arnusch Langmuir, Just Accepted Manuscript • DOI: 10.1021/acs.langmuir.8b01617 • Publication Date (Web): 20 Aug 2018 Downloaded from http://pubs.acs.org on August 21, 2018

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 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 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.

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 31 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

Synergy on Surfaces: Anti-Biofouling Interfaces using Surface-Attached Antimicrobial Peptides PGLa and Magainin-2 Nitzan Shtreimer Kandiyote, Gunasekaran Mohanraj, Canwei Mao, Roni Kasher* and Christopher J. Arnusch* Department of Desalination and Water Treatment, Zuckerberg Institute for Water Research, The Jacob Blaustein Institutes for Desert Research, Ben-Gurion University of the Negev, Sede-Boqer Campus 84990, Israel *Corresponding

authors:

Tel.

972-8-656-3532,

Fax.

[email protected], [email protected]

1 ACS Paragon Plus Environment

972-8-656-3503,

e-mail:

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

Abstract The synergistic effect of antimicrobial compounds is an important phenomenon that can increase the potency of the treatment and might be useful against the formation of biofilms on surfaces. A strong inhibition of microbial viability on surfaces can potentially delay the development of biofilms on treated surfaces, thereby enhancing the performance of water purification technologies and medical devices, e.g., to prevent hospital-acquired infections. However, the synergistic effects of surface-immobilized antimicrobial peptides (AMPs) have not yet been reported. Here, we demonstrate the synergistic antimicrobial effects of the AMPs PGLa and magainin-2 on modified reverse osmosis (RO) membranes. These AMPs are known to act synergistically in the free state, but their antimicrobial synergistic effects have not yet been reported in a surface-immobilized state. The AMPs were functionalized with alkyne linkers and covalently attached to RO membranes modified with azides, using a click chemistry reaction. The resulting RO membranes showed reduced contact angles, indicating increased wettability. XPS spectroscopy confirmed the presence of the two peptides on the membranes via changes in the amounts of carbon, oxygen, and sulfur, which led to an increased S/C ratio, probably due to the sulfur present in the methionine residue of the peptides. The synergistic activity was measured with the free peptides in solution and covalently bound on RO membrane surfaces by observing increased leakage of 5(6)-carboxyfluorescein from large unilamellar vesicles. The synergistic antimicrobial activity against P. aeruginosa was observed using surface-activity assays, where the AMP-modified RO membranes showed an effective inhibition of P. aeruginosa biofilm growth, as compared with unmodified membranes. An enhanced activity of antimicrobials on surfaces might lead to potent antimicrobial surfaces, which could result in more fouling-resistant water treatment membranes. 2 ACS Paragon Plus Environment

Page 2 of 31

Page 3 of 31 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

Keywords: Antimicrobial peptides; PGLa; magainin-2; synergy; reverse osmosis membranes; biofouling. 1. Introduction Desalination of seawater using reverse-osmosis (RO) membranes is an efficient and widely used technology,1-2 in which a semipermeable thin-film composite (TFC) membrane is used to separate solutes and dissolved salts from water.3-4 A severe problem in RO desalination is biofouling due to the formation of biofilms on the top polyamide layer of the TFC membrane, which is susceptible to bacterial attachment and biofilm formation. Importantly, such biofilms cannot be eliminated by oxidizing chemical reagents (e.g., active chlorine species) because the polyamide film of the membrane is sensitive to oxidizing chemicals, while other disinfection methods, such as ultra-violet irradiation, are impractical in RO membranes that are installed in self-contained membrane modules.5 Conversely, imparting antimicrobial activity to the RO membrane may help reduce the number of viable bacteria attached to the surface and, thereby, delay biofilm growth6-8 and require less frequent cleaning of the membrane, thus prolonging its lifetime and increasing process efficiency.9 Several strategies have been demonstrated to reduce or delay biofilm development on the membrane surface. Select examples include membrane surfaces modified with other polymer materials using polymer grafting techniques,10-12 membrane formulations to increase chlorine resistance, which might protect the membrane during chemical cleaning cycles,13 imparting antimicrobial activity using toxic materials such as copper or silver,14-15 or carbon based materials such as graphene derivatives.16-19 One strategy for imparting antimicrobial activity to RO membranes is to modify the membranes with antimicrobial peptides (AMPs)20-22 – short


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

amphipathic cationic peptides, which are generally classified as broadly toxic to microorganisms23 and are considered less susceptible than conventional antibiotics to generate bacterial resistance.24-26 AMPs that can act synergistically to inhibit biofilm formation are of special applicative interest, but, to the best of our knowledge, the synergism of AMPs covalently attached to a surface has never before been demonstrated. For instance, PGLa and magainin-2— two AMPs isolated from the skin of the Xenopus frog—were shown to act synergistically to inhibit bacterial growth,27-28 but their synergistic activity was only demonstrated in their dissolved form.29-31 Moreover, previous reports have indicated that the activity of immobilized AMPs do not necessarily correspond to the activity of the free soluble AMPs counterparts.32 In the present study, we tested whether covalently attaching PGLa and magainin-2 to the surface of RO membranes can enhance the surface antimicrobial activity of the membranes and, thereby, attenuate biofilm formation. To this end, we employed the copper-catalyzed azidealkyne cycloaddition reaction (CuAAC, “click chemistry”)21, 33-34 to functionalize the peptides with an alkyne linker and the RO membranes with an azide. We found a synergistic action of the two surface-attached AMPs in antibacterial surface assays, and all peptide-modified surfaces demonstrated a reduction in biofilm growth, as compared with unmodified surfaces. These findings indicate that synergistic AMPs can be attached to surfaces to increase the effectiveness of AMP-modified membranes and further delay biofilm growth, and that they may be applicable for membrane-based water treatment technologies and medical applications. 35-38


Page 4 of 31

Page 5 of 31 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

2. Materials and Methods 2.1. Materials SW30HR flat-sheet RO membranes were purchased from Dow Water & Process Solutions (Midland, MI). N-Fluorenyl methoxy carbonyl (Fmoc)-protected amino acids, Fmocrink amide (MBHA) resin (100–200 mesh, 0.7 mmol/g), and the coupling reagents Obenzotriazole-N,N,N’,N’-tetramethyl-uronium-hexafluoro-phosphate

(HBTU)

and

N,N-

diisopropylethylamine (DIEA) were purchased from NovaBiochem (Darmstadt, Germany) and N-hydroxysulfosuccinimide

sodium

salt

(sulfo-NHS),

N-(3-dimethylaminopropyl)-N'-

ethylcarbodiimide hydrochloride (EDC), and sodium ascorbate (C6H7NaO6) from Chem-Impex International (Wood Dale, IL). Protected peptides on a rink-amide resin (protected PGLa-resin and protected magainin-2-resin) were purchased from peptide 2.0 (Chantilly, VA). N-Methyl-2pyrrolidone (NMP) peptide synthesis grade, dichloromethane (DCM) peptide synthesis grade, dimethylformamide (DMF) peptide synthesis grade, acetonitrile (ACN) HPLC-grade, and H2O HPLC-grade were purchased from J.T. Baker (Philipsburg, NJ). Tert-Butyl methyl ether (MTBE) HPLC-grade was purchased from Bio-Lab (Jerusalem, Israel). 4-Pentyonic acid (98%) was purchased from Acros Organics (Morris Plains, NJ). Cupric sulfate, 11-Azido-3,6,9trioxaundecan-1-amine (azide linker), 5(6)-carboxyfluorescein, and Triton X-100 10 % (w/v) were purchased from Sigma-Aldrich (St. Louis, MO). 1,2-dioleoyl-sn-glycero-3-phosphocholine (DOPC) and 1,2-dioleoyl-sn-glycero-3-phospho-(1'-rac-glycerol) (DOPG) lipids of >99% purity were purchased from Avanti Polar Lipids (Alabaster, AL). 2.2. Synthesis and Purification of Alkyne-PGLa and Alkyne-magainin-2 Peptide sequence assembly: PGLa and magainin-2 protected peptides (see sequences in Table 1) were synthesized by the CS336S automated peptide synthesizer (CS Bio, Menlo Park, 5 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

CA) on a rink-amide MBHA resin (0.7 mmol/mg loading) using the solid-phase peptide synthesis with the Fmoc chemistry strategy.39 Coupling of 4-pentynoic acid: Dry protected PGLa-resin or protected magainin-2-resin (either commercial or synthesized in our laboratory; 0.1 mmol) was added to DMF (4 mL) for 1 h. The DMF was filtered, and a solution containing 0.5 mmol (5 eq.) 4-pentyonic acid, 0.5 mmol (5 eq.) HBTU, and 1 mmol (10 eq.) DIEA, all dissolved in DMF and agitated for 2h, was added to the swollen resin. Afterward, the peptide-resin was washed thoroughly, first with DMF (3 x 1 min) and then with methylene chloride (3 x 1 min). It was then dried and subsequently cleaved from the resin by adding a solution containing trifluoroacetic acid (TFA; 94%), 1,2ethanedithiol (EDT; 2.5%), water (2.5%), and triisopropylsilane (1%). The solution was separated from the resin and the volume reduced with a stream of N2(g), and cold MTBE was added. The precipitate was washed with cold MTBE and collected by centrifugation. The peptides were purified (>92% purification) by reverse-phase HPLC (Thermo Finnigan, San Jose, CA) using a C18 column (Gemini, 5 microns, 110 Å, 4.6 × 250 mm, Phenomenex, Torrance, CA) and a 40 min binary gradient of 0.1% TFA in water and 0.085% TFA in 75% acetonitrile in water (5%-95%) (Figure S-1). The identity of peptides was confirmed by MALDI-TOF-MS using a Reflex 4 mass spectrometer (Bruker Daltonics, Bremen, Germany), with α-cyano-4hydroxycinnamic acid or 2,5-dihydroxybenzoic acid as the matrix. Alkyne-PGLa, calculated mass = 2047.29 Da; observed mass 2048.26 Da [M+H]+; alkyne-magainin-2, calculated mass = 2544.43 Da; observed mass 2545.35 Da [M+H]+. 2.3. Membrane Modification


Page 6 of 31

Page 7 of 31 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

Azide-modified membranes: Round RO membranes (16 mm in diameter) were first dipped in 70% (v/v) ethanol and immediately washed with double distilled water (DDW). The membranes were then washed in a cooled sonication bath in a 1:1 isopropanol:water solution (3×10 min) and then in DDW (3×10 min). Next, in a closed vial, the membranes were immersed in a sodium phosphate buffer (0.1 M, pH 7.4) containing sulfo-NHS (20 mM), EDC (20 mM), and azide linker (60 mM), and the solution was constantly agitated overnight on a Unimax 1010 orbital platform shaker (Heidolph, Kelheim, Germany) at room temperature. Finally, the membranes were washed with sonication, first in a sodium phosphate buffer (0.1 M, pH 7.4, 3×10 min) and then in DDW (3×10 min), and were stored in DDW at 4 °C until used. AMP-modified membranes: Alkyne-PGLa, alkyne-magainin-2, or a 1:1 (mol/mol) mixture of the two were dissolved in 1 mL of DDW (total concentration: 1 mM) added with cupric sulfate (CuSO4, 1 mM) and sodium ascorbate (2 mM). Then, the peptides were added to the azide-modified membrane and gently agitated for 2 h at 40 °C. The membranes were washed with an HCl aqueous solution (1 mM, 2×10 min) and then with DDW (3×10 min) under sonication, and stored in DDW at 4 °C. Attaching various amounts of peptides to RO membranes: The process was similar to the AMP modification process, except that 4-pentynoic acid was added to the peptides at ratios (peptides:4-pentynoic acid) of 1:0 (denoted R1), 1:1 (denoted R2), 1:1.2, 1:2, 1:4, 1:10, 1:20, or 1:100. The total amount of each alkyne-containing compound was 1 mM. 2.4. Membrane Surface Analyses Attenuated total reflection Fourier transform infrared spectroscopy (ATR-FTIR): The modified membrane surfaces were characterized by a VERTEX 70/80 spectrophotometer 7 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

(Bruker Optiks GmbH, Ettlingen, Germany) using a Miracle ATR attachment with a onereflection diamond-coated KRS-5 element. For each modified membrane, an average spectrum was calculated by the OPUS software (version 6.5) from six spectra obtained from random locations on each membrane surface. Prior to measurements, the sampled membranes were dried overnight in a vacuum at room temperature. Each measurement included, on average, 40 scans at a 4 cm-1 resolution, at the range of 4000-400 cm-1. X-ray photoelectron spectroscopy (XPS): XPS is a highly sensitive technique for analyzing element composition, and chemical binding information can be obtained on surfaces to a depth of ~1-10 nm. The XPS spectrometer ESCALAB 250 (Thermo Fisher Scientific, Waltham, MA) with an ultrahigh vacuum (10-9 bar), installed with an AlKα X-ray source (beam size: 500 µm) and a monochromator, was used. The signals from C 1s, N 1s, O 1s, and S 2p were detected by fixing different separated elements to the experimental data by means of nonlinear least-squares curve fitting. The survey spectra with a pass energy (PE) of 150 eV and the highenergy resolution spectra with a PE of 20 eV were recorded. Spectra were corrected for charging effects by calibrating the carbon 1s peak at ~285.0 eV. Prior to measurements, the sampled membranes were completely dried overnight in a vacuum at room temperature. Surface contact angle measurement: The sessile drop-contact angles of the sampled membranes were measured on the horizontal surfaces with DDW water by an OCA-20 contact angle analyzer (DataPhysics Instruments, Filderstadt, Germany). A water droplet (5 µL) was applied on the surface of the dried membrane. The SCA-20 software (DataPhysics Instruments) was used to calculate the contact angles. For each modified membrane, an average of six drops was calculated.


Page 8 of 31

Page 9 of 31 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

2.5. Antibacterial Activity in Solution The minimal inhibitory concentration (MIC) of alkyne-PGLa and alkyne-magainin-2 in solution was tested with a broth dilution assay in a 96-well microtiter plate (Costar, Corning, NY),40-41 using the Gram-negative bacteria Pseudomonas aeruginosa (PAO1 strain) and Escherichia coli in Luria-Bertani (LB) broth media. In addition, Klebsiella oxytoca bacteria were tested in Tryptic Soy Broth (TSB) media. For all bacterial tests, 100 µL of LB or TSB was added to each well, and then 100 µL of the AMP (2 mg/mL) was added to the first row and serially diluted in each column. Next, the bacteria were incubated overnight in 1 mL of their respective growth medium (P. aeruginosa at 30 °C, E. coli at 37 °C, and K. oxytoca at 25 °C), which was then used to seed a fresh culture in 100 mL of fresh media.29, 42-43 The fresh culture was shaken and incubated at the above-mentioned temperatures until turbidity reached an optical density (OD) of ~0.8, as measured at the wavelength of 600 nm by a Genesys 8 spectrophotometer (Thermo Scientific, Loughborough, UK). The bacteria concentration was adjusted to approximately 2×107 CFU/mL by dilution (verified by the plate-count method) and was further diluted ×100 with growth media (to a concentration of 2×105 CFU/mL); then, 100 µL of the final solution was added to each well. The plates were incubated overnight at the above-mentioned temperatures, and then the MICs were determined by locating the lowest peptide concentration that did not show turbidity, verified by measuring absorption at 600 nm. Sterilized water served as the positive control and the medium served as the negative control. Results from five independent experiments, performed in duplicates, were averaged. The checkerboard assay was performed according to previously published methods.44


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

2.6. Permeabilization Measurements using Large Unilamellar Vesicles Preparation of large unilamellar vesicles (LUVs): Lipid vesicles were prepared as previously described.45 Briefly, lipids were dissolved in chloroform and a DOPC/DOPG mixture46 (1/1 molar ratio) was made and added to a test tube. The chloroform was evaporated using N2(g) and further dried under vacuum for 15 min. Subsequently, the mixture was hydrated with 50 mM of 5(6)-carboxyfluorescein in Tris buffer (10 mM Tris, 150 mM NaCl, pH 7.4) and vortex-mixed to produce multilamellar vesicles (MLVs). After hydration, the MLVs were subjected to eight freeze-thaw cycles (freezing in liquid nitrogen, thawing in a warm water bath). The LUV suspensions were generated by multiple extrusions through a polycarbonate filter (0.2 µm pore size, 10 times) mounted in mini-extruder (Avanti Polar Lipids) at room temperature. Untrapped 5(6)-carboxyfluorescein was removed by subsequent gel filtration (Sephadex G-50, Sigma-Aldrich, St. Louis, MO) at room temperature, with Tris buffer (10 mM Tris, 150 mM NaCl, pH 7.4) as the eluent. Fluorescence leakage assay: Tris buffer (100 µL, 10 mM, pH 7.4) was added to the wells of a 96-wells plate. A 100 µL solution of peptides (10 µM) in Tris buffer was added to the first well and the peptides were serially diluted 1:1 across the plate. Then LUVs (100 µL, 20 µM or 50 µM lipid in 10 mM Tris, pH 7.4) were added to each well giving a final concentration of 10 µM or 25 µM. 5(6)-carboxyfluorescein release from the LUVs was monitored with a fluorescence spectrophotometer (Cary Eclipse, Varian, Palo Alto, CA; excitation wavelength 492 nm, emission wavelength 515 nm). A 100% leakage was defined by the fluorescence obtained after the addition of 100 µL Triton X-100 to a final concentration of 0.1% (w/v) and a 0% leakage was defined by the fluorescence obtained after the addition of 100 µL Tris buffer.47 The normalized leakage was calculated according to equation 1:28 10 ACS Paragon Plus Environment

Page 10 of 31

Page 11 of 31 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

Equation 1. Normalized leakage (%) = 100 ∗ (F − F଴ )/(F௧ − F଴ ) where F and Ft denote the fluorescence intensity before and after the addition of 0.1% X-100 triton, respectively, and F0 denotes the fluorescence of intact vesicles. Six measurements were performed in triplicate at room temperature. Synergy between the two AMPs was concluded if the sum of fractional inhibitory concentrations [ƩFIC = FIC(A) + FIC(B)] was below 0.5, as calculated according to equations 2 and 3:

Equation 2. FIC(A) =

Equation 3. FIC (B) =

୑୍େ (୅୆) ୑୍େ (୅) ୟ୪୭୬ୣ

୑୍େ (୅୆) ୑୍େ (୆) ୟ୪୭୬ୣ

=

=

஼௢௡௖௘௡௧௥௔௧௜௢௡ ௜௡ ௖௢௠௕௜௡௔௧௜௢௡ ୍େఱబ (୅) ୟ୪୭୬ୣ

஼௢௡௖௘௡௧௥௔௧௜௢௡ ௜௡ ௖௢௠௕௜௡௔௧௜௢௡ ୍େఱబ (୆) ୟ୪୭୬ୣ

=

஼௢௡௖௘௡௧௥௔௧௜௢௡ ௜௡ ௖௢௠௕௜௡௔௧௜௢௡ ୉େఱబ (୅) ୟ୪୭୬ୣ

=

஼௢௡௖௘௡௧௥௔௧௜௢௡ ௜௡ ௖௢௠௕௜௡௔௧௜௢௡ ୉େఱబ (୆) ୟ୪୭୬ୣ

where EC50 represents the concentration of a peptide that gives a half-maximal response and IC50 represents the concentration of a peptide that was required for a 50% inhibition.44, 48 Leakage assay for AMP-modified RO membranes: The membranes modified with alkynePGLa, alkyne-magainin-2, or a 1:1 mixture of the two AMPs were cut with a hole-puncher to circles with a diameter of 6.3 mm. Then, each circle was placed in one well of a 96-wells plate and was mixed with 200 µL LUVs (final concentration: 10 µM or 25 µM lipid in 10 mM Tris, pH 7.4) and agitated on a shaker. After 24 h, the solution was transferred to a new 96-wells plate and the 5(6)-carboxyfluorescein release from the LUVs was monitored by the fluorimeter (excitation wavelength = 492 nm, emission wavelength = 515 nm). 2.7. Contact Killing Assay A fresh single colony of P. aeruginosa was inoculated in 10 mL of LB broth and incubated overnight at 30 °C on a shaker at 150 rpm. The resulting culture was re-diluted in


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

5 mL LB broth (a 1:100 ratio) and was allowed to grow for 6 h to the exponential growth phase (OD600 ~0.5). The cells were diluted with a sterile PBS buffer solution (pH 7.4) to 104 CFUs. Circular membranes (16 mm in diameter) were taped with a double-sided adhesive tape to a sterile glass slide and placed inside a sterile Petri dish. To these membranes, 100 µL of bacterial culture (~4000 cells) from the dilution tube was transferred to the AMP modified surface and evenly spread on it by covering it with a microscope coverslip, as previously described.49 The membranes were then incubated for 3 h at room temperature, the coverslips were gently removed, and the membranes, together with the coverslips, were washed with 0.5 mL of sterile PBS by continuous aspiration. The plate-count method was used to quantify the viable CFU with sterile LB agar incubated for 24 h at 30 °C. 2.8. Biofilm Growth Assay The flow cell (25 mm × 49 mm × 32 mm) was washed with 70% ethanol and sterile DDW. The modified and unmodified membranes were cut in squares (1 cm2), attached to a microscope coverslip (24 mm × 40 mm; Menzel-Glaser, Australia) using a double-sided adhesive tape, and placed vertically inside the flow cell.16-17 A culture of P. aeruginosa was prepared at 30 °C, as described in section 2.6 above, to the exponential growth phase and then diluted to OD 0.1 at 600 nm. The diluted P. aeruginosa culture (50 mL) was flowed across the surfaces at 2.5 mL/min using a peristaltic pump (LabM1-2 YZ1515x, Shenchen, China). The inlet line of the cell was then connected to an LB liquid medium (5 L) and ran for 36 h at 2 mL/min at room temperature. The membrane-attached coverslips were then removed and the bacteria were stained. The staining solution was prepared as follows: propidium iodide (1.5 µL, 20 mM) and SYTO 9 (1.5 µL, 3.34 mM) were added to 0.997 mL of a 0.1 M sodium chloride solution to stain dead and live bacteria, respectively. Then, 100 µL of the staining solution was 12 ACS Paragon Plus Environment

Page 12 of 31

Page 13 of 31 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

added to cover the biofilm surface and stored, protected from light, for 20 min. The surfaces were gently washed (3 times) by adding 0.25 mL of the sodium chloride solution (0.1 M) and then the excess electrolyte was carefully removed by touching the edges with absorbing paper. The biofilm of P. aeruginosa (PAO1 strain) was visualized using the Zeiss LSM 510 META confocal microscope, with a dry objective Plan-NeoFluar (20× magnification and 0.5 numerical aperture). Data were analyzed using the Imaris 3D imaging software (Bitplane, Zurich, Switzerland), and quantitative analysis (biofilm volume and average thickness) was conducted using COMSTAT on Matlab 2015b.50 The analysis was performed on RO membranes modified using reaction conditions R1 and R2 (see section 2.3). Average results and standard deviation of five representative images are reported. 2.9. Scanning Electron Microscopy (SEM) P. aeruginosa were cultured in LB broth at 30 °C to an exponential phase as above. The cells were harvested by centrifugation at 3000 rpm for 10 min, washed and suspended with PBS buffer to an OD of 0.1. The membranes (10 mm x 10 mm) fabricated using R1 conditions were attached to a glass slide, and 50 µL of the bacterial culture was placed on each membrane sample. A sterile cover slip was gently placed on top to spread the inoculum across the membrane sample and then the samples were incubated at room temperature for 3 hours. The cover slip was removed and 2.5% (w/v) glutaraldehyde (100 µL) was added and kept undisturbed for overnight at 4 °C. The samples were washed twice with PBS and 1 % osmium tetraoxide (100 µL) was added on the surface and incubated for 1 h, and then washed twice with PBS. The samples were dehydrated by immersion in aqueous ethanol solutions (50, 70, 90 and 100%) for 15 min each, and then tertiary butanol for 30 min. The dehydrated sample was air dried and coated with gold and observed by SEM (JSM-7400F, JEOL, Tokyo, Japan). 13 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

3. Results and Discussion 3.1. Peptide Synthesis and Antimicrobial Activity We previously developed click chemistry procedures for the attachment of AMPs to RO membranes, which were also compatible with the presently used PGLa and magainin-2.21 The two AMPs were functionalized with an alkyne group at their N-terminus51 and reacted with RO membranes modified with azide groups. Alkyne-PGLa and alkyne-magainin-2 either were synthesized in-house using standard solid-phase peptide synthesis procedures or acquired commercially. The peptides were purified by RP-HPLC and their identity confirmed with mass spectrometry MALDI-TOF-MS analysis (see Materials and Methods, and Figure S-1). As previous reports showed a synergistic activity of PGLa and magainin-2 in a 1:1 peptide ratio,29, 52-53

we tested the MIC of the alkyne-functionalized AMPs against P. aeruginosa, E. coli, and K.

oxytoca, and compared these with the AMPs at a 1:1 ratio (Table 1). To test whether the alkynefunctionalized AMPs caused permeabilization of lipid membranes, the alkyne-peptides were tested using model membranes, large unilamellar vesicles (LUVs) composed of DOPC:DOPG at a 1:1 ratio. In both the MIC and model membrane assays, we found that the alkynefunctionalized AMPs, at a 1:1 ratio, demonstrated synergistic activity (ΣFIC ≤ 0.5 in all cases; see equations 2-4 in section 2.6 above) and a mode of action that was similar to PGLa and magainin-2.29, 52 Other peptide ratios tested did not show effective synergy (Table S-1 and Table S-2). Table 1. The antimicrobial and lipid membrane activity of alkyne-modified AMPs in solution. MICs against P. aeruginosa, E. coli, and K. oxytoca are shown for the alkyne-peptides alone and


Page 14 of 31

Page 15 of 31 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

for their 1:1 mixture. The EC50 is reported for the membrane permeabilization using an assay with LUVs, and the ΣFIC value was calculated using equations 2-4.

EC50 MIC (µM) (µM) P.

E.

K.

aeruginosa

coli

oxytoca

Alkyne-PGLa-NH2

121

50

>200

0.89

Alkyne-magainin-2-NH2

98

50

49.1

0.84

54

11

11.4

0.15

0.5

0.22

0.14

0.17

a

Peptide

LUVs

Alkyne-PGLa-NH2:Alkyne-magainin-2-NH2 (1:1) ΣFIC a

Sequences:

Alkyne-PGLa, Alkyne-GMASKAGAIAGKIAKVALKAL-NH2 Alkyne-magainin-2, Alkyne-GIGKFLHSAKKFGKAFVGEIMNS-NH2 3.2. RO Membrane Modification The surface modification procedure was performed in a similar manner to a previously developed protocol, in which short peptides were attached to RO membranes via the free carboxyl group of the aromatic polyamide of the membrane (Figure 1).21 As detailed below, both XPS and FTIR measurements indicated that the two AMPs were covalently attached to the surface and these surfaces were compared with the surface of un-modified RO membranes. The 15 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

XPS analysis indicated that the main elements found in the AMP-modified and unmodified membranes were carbon, nitrogen, oxygen, and sulfur, and these elements were calibrated relative to a carbon 1s peak (~285 eV; Table 2). Notably, the S/C ratio was slightly higher in the AMP-modified surfaces than in the unmodified surfaces (namely, 0.061, 0.074, and 0.065 for membranes modified with PGLa, magainin-2, and their 1:1 mixture, respectively, as compared with 0.053 for the unmodified membrane), possibly due to the presence of sulfur in the methionine residue of PGLa and magainin-2.54-55 Next, we used ATR-FTIR to determine the presence of the AMPs on the RO membrane surfaces. All membranes exhibited absorbance spectra typical of a TFC membrane, which included signals from the polyamide layer and many signals from the underlying polysulfone layer. The strong absorbance peak at 1487 cm-1 was assigned to the polysulfone component and was used to normalize the other peaks. An increase in the amide-I absorption signal at 1664 cm-1 was observed on the AMP-modified membranes (Table 2, Figure S-2). For example, the 1664/1487 cm-1 peak ratio was ~0.29 for the unmodified membranes and ~0.44 for the AMP-modified membranes (Table 2). The wettability of the membrane surface, tested by the sessile water drop contact angle, also showed evidence for peptide modification, as the surfaces were more hydrophilic when the peptides were attached: the water contact angle was lower in modified than in unmodified membranes (38-40° vs. ~52°, respectively). This increase in surface hydrophilicity might lead to decreased adsorption of bacterial cells to the RO membrane surfaces and be beneficial in reducing fouling.56 Peptides are depicted as red or blue lines (Figure 1), however the content or types of secondary structures were not predicted or measured, although previously have been shown to play a role in the mechanism of action of surface attached peptides.46


Page 16 of 31

Page 17 of 31 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 1. Attachment of the AMPs PGLa and magainin-2 to RO membrane surfaces using click chemistry (CuAAC). Reaction conditions: a) 60 mM azide linker, 20 mM sulfo-NHS, and 20 mM EDC in a sodium phosphate buffer (0.1 M, pH 7.4) at room temperature, overnight; b) alkyne-PGLa, alkyne-magainin-2, or a 1:1 (mol:mol) mixture of alkyne-PGLa and alkynemagainin-2 (1 mM), CuSO4, (1 mM), and sodium ascorbate (2 mM), 40 °C for 2 h. Table 2. XPS, ATR-FTIR, and contact-angle data of the unmodified and AMP-modified RO membranes.

XPS (atomic %)

FTIR Contact

Membrane

C

N

O

S

S/C

modification

-1

1664 cm /1487 angle (°) -1

ratio

cm

Unmodified

63.8

2.9

29.9

3.4

0.053

0.29 ± 0.03

52 ± 5

PGLa

60.5

2.4

33.4

3.7

0.061

0.46 ± 0.03

38 ± 4

Magainin-2

60.8

3.1

31.7

4.5

0.074

0.45 ± 0.02

38 ± 4

PGLa:Magainin-2

60.2

3.5

32.3

3.9

0.065

0.46 ± 0.01

40 ± 5.5


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

(1:1)

3.3. Antibacterial Activity of Modified RO Surfaces Our biofilm inhibition strategy was based on the antimicrobial surface activity of the membranes, which is key to delaying bacterial colonization by inactivating the attached bacteria.57 The antimicrobial activity of the membrane surfaces was measured by contact-killing measurements21,

49

using P. aeruginosa, where the surfaces of the unmodified and AMP-

modified membranes were inoculated and incubated with the bacteria. To measure the effect of the amount of peptides bound to the membrane and the synergistic effect of binding a mixture of peptides on surface antibacterial activity, a series of RO membranes was prepared by performing the peptide-coupling step with decreasing amounts of a competing non-peptide alkyne compound, which we expected to generate surfaces with increasing amounts of peptide. As the concentration of peptides during the surface attachment reaction increased, the 1664 cm-1/1487 cm-1 FTIR peak ratio, both of each peptide alone and of the combination of peptides, increased as well, ranging between 0.4 and 0.45 – which was significantly higher than that of an unmodified membrane (~0.3) (Figure 2). In line with these findings, the increase in the amount of peptides also corresponded with an increase in the antimicrobial surface activity of the membranes (Figure 3). The bacterial inhibition values of the AMP-modified membranes were normalized to those of the unmodified membranes. As compared with the surfaces of the AMP-modified membranes, surfaces functionalized only with the azide linker were not active. In addition, as compared with unmodified membranes, membranes functionalized with 4-pentynoic acid without a peptide demonstrated an increased bacterial growth (Figure 3). The IC50 values of the modified membranes were determined using the concentration of peptides that corresponded to 18 ACS Paragon Plus Environment

Page 18 of 31

Page 19 of 31 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

what was used in the reaction conditions. The calculated ΣFIC was 0.29, indicating synergy between the peptides attached on the surface. SEM analysis of the membrane surfaces after bacterial attachment showed less bacteria growth on the modified-AMPs than on the control surfaces (Figure S-3). As well, bacteria with obvious cell damage could be identified on the AMP modified membranes whereas the unmodified membranes showed bacteria in a consistent shape (Figure S-3).

Figure 2. ATR-FTIR characterization of RO membranes modified with various peptide concentrations. Peptides (a) PGLa, (b) magainin-2, or (c) PGLa:magainin-2 (1:1 mol:mol), together with 4-pentynoic acid at the indicated concentrations, were dissolved in 1 mL DDW 19 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

with CuSO4 (1 mM) and sodium ascorbate (2 mM), and were added to the azide modified membranes for 2 h at 40 °C. Values indicate means ± SD.

Figure 3. Inhibition of P. aeruginosa bacteria on AMP-modified membranes. First the membranes were modified with the azide linker, then the membranes were prepared by attaching different proportions of peptides and 4-pentynoic acid, as indicated. The IC50 values for membrane surfaces modified with PGLa, magainin-2, or the 1:1 mixture of these peptides were 0.12 mM, 0.29 mM, and 0.05 mM, respectively. *Negative bacterial inhibition indicates bacterial growth relative to the unmodified membrane (stdev. ± 24.7%). 3.4. Biofilm Inhibition by AMP-Modified RO Membranes To investigate the ability of AMP-modified membranes to inhibit biofilm formation, two conditions were alternatively used during the membrane coupling reaction: in the first condition (R1), a 1 mM peptide concentration was used, and in the second condition (R2), a 0.5 mM peptide concentration was used together with a 0.5 mM concentration of 4-pentynoic acid. The inhibition of biofilm growth was determined using a flow cell, in which the modified membranes were exposed first to a bacterial (P. aeruginosa) culture for 20 min to enable bacteria to adhere to the surfaces, and then to a sterile growth medium for 36 h to allow viable adhered bacteria to 20 ACS Paragon Plus Environment

Page 20 of 31

Page 21 of 31 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

form biofilm. Subsequently, the thickness of the biofilm estimated by quantification of both dead and live bacteria was determined using confocal laser scanning microscopy (CLSM). These analyses indicated that the biofilms formed on the surfaces of all AMP-modified membranes were thinner than those formed on the surfaces of the unmodified membranes, including unmodified membranes, membranes with an azide linker or 4-pentynoic acid (Figure 4). Specifically, modifying the membranes using R1 conditions reduced the average biofilm thickness by 70–90%, as compared with the unmodified membranes. Under the R1 condition, although modifying the membranes with a 1:1 mixture of PGLa and magainin-2 showed similar biofilm thickness, as compared with that of membranes modified with only one of the peptides, the synergy was seen in that the mixture of peptides increased the ratio of dead to living cells in the biofilm from 0.24 and 0.06, to 1.49, for PGLa, magainin-2, and the 1:1 mixture, respectively, demonstrating a stronger antimicrobial effect.58 This stronger effect might indicate that such an antimicrobial surface could affect the composition of the biofilm and ultimately the anti-fouling effect. For the surfaces of membranes modified with R2 conditions, PGLa or magainin-2 reduced the biofilm thickness by ca. 70%, as compared with those of the unmodified membranes and modified with only 4-pentyonic acid. However, in the R2 condition, the membranes modified with the 1:1 PGLa:magainin-2 mixture showed a significant increase in biofilm inhibition, reducing the biofilm thickness by ca. 97%, as compared with the unmodified membranes (Figure 4). Taken together, these findings indicate that synergistic effects can be possible even when the AMPs PGLa and magainin-2 are covalently attached to a solid support.


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

Figure 4. The average thickness of a P. aeruginosa biofilm (including both dead and living bacteria) on RO membranes under flow conditions. Control membranes consisted of unmodified RO membranes, and membranes modified with the azide linker or 4-pentynoic acid. Two coupling conditions were used when preparing the membranes: 1 mM of the peptides (denoted R1), and 0.5 mM of the peptides with 0.5 mM of 4-pentynoic acid (denoted R2). A representative image of the R1 condition is shown above each category, where green represents live bacteria and red represents dead bacteria. 3.5. Mode of Action of the Surface Bound AMPs To investigate the antimicrobial mode of action of the membrane-bound AMPs, we contacted the surfaces of the AMP-modified membranes with LUVs, which contained 5(6)-carboxyfluorescein and were composed of lipids that mimic the negatively charged outer membrane leaflet of bacteria (Figure 5a).59-60 We postulated that leakage of the 5(6)-carboxyfluorescein from the LUVs would indicate that the bound AMPs can permeabilize lipid membranes and, thus indicating a lytic mode of action.61-62 The leakage of 5(6)-carboxyfluorescein was 17-26% 22 ACS Paragon Plus Environment

Page 22 of 31

Page 23 of 31 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

greater in all AMP-modified membranes than in unmodified membranes. Membranes modified with a 1:1 PGLa:magainin-2 mixture were 8-9% more active (p

Synergy on Surfaces: Anti-Biofouling Interfaces Using Surface

Recommend Documents