Evidence for the Phospholipid Sponge Effect as the Biocidal

Research Article www.acsami.org

Evidence for the Phospholipid Sponge Effect as the Biocidal Mechanism in Surface-Bound Polyquaternary Ammonium Coatings with Variable Cross-Linking Density Jing Gao,† Evan M. White,‡ Qiaohong Liu,† and Jason Locklin*,†,‡ †

Department of Chemistry and College of Engineering and ‡New Materials Institute, University of Georgia, Athens, Georgia 30602, United States S Supporting Information *

ABSTRACT: Poly quaternary “-oniums” derived from polyethylenimine (PEI), poly(vinyl-N-alkylpyridinium), or chitosan belong to a class of cationic polymers that are efficient antimicrobial agents. When dissolved in solution, the positively charged polycations are able to displace the divalent cations of the cellular phospholipid bilayer and disrupt the ionic cross-links and structural integrity of the membrane. However, when immobilized to a surface where confinement limits diffusion, poly -oniums still show excellent antimicrobial activity, which implies a different biocidal mode of action. Recently, a proposed mechanism, named phospholipid sponge effect, suggested that surface-bound polycationic networks are capable of recruiting negatively charged phospholipids out of the bacterial cell membrane and sequestering them within the polymer matrix.1 However, there has been insufficient evidence to support this hypothesis. In this study, a surface-bound N,N-dodecyl methyl-co-N,Nmethylbenzophenone methyl quaternary PEI (DMBQPEI) was prepared to verify the phospholipid sponge effect. By tuning the irradiation time, the cross-linking densities of surface-bound DMBQPEI films were mediated. The modulus of films was measured by PeakForce Quantitative Nanomechanical Mapping (QNM) to indicate the cross-linking density variation with increasing irradiation time. A negative correlation between the film cross-linking density and the absorption of a negatively charged phospholipid (DPhPG) was observed, but no such correlations were observed with a neutral phospholipid (DPhPC), which strongly supported the action of anionic phospholipid suction proposed in the lipid sponge effect. Moreover, the killing efficiency toward S. aureus and E. coli was inversely affected by the cross-linking density of the films, providing evidence for the phospholipid sponge effect. The relationship between killing efficiency and film cross-linking density is discussed. KEYWORDS: surface-bound antimicrobials, phospholipid absorption, modulus, benzophenone photo-cross-linking

■

INTRODUCTION The membrane-disrupting poly “-onium” (ammonium and pyridinium) cations have been extensively investigated as a category of antimicrobial compounds due to their widespread use, ranging from disinfectants in solution to biocidal coatings immobilized on surfaces.2−5 These -onium polycations have been reported to have high antimicrobial activity toward a broad spectrum of microbes including Gram-positive and Gram-negative bacteria, fungi, and even some viruses.6−8 Additionally, bacteria have shown little antibiotic resistance to poly -oniums due to their nonselective, damaging antimicrobial action of the cellular membrane, which occurs immediately on contact.9,10 The biocidal mechanism of leaching (or soluble) © 2017 American Chemical Society

poly -onium derivatives involves disruption of the ionic integrity of the cell membrane by cationic displacement of the divalent cations (Mg2+, Ca2+) by the quaternary -oniums via electrostatic interaction.11 On the other hand, when covalently immobilized on a surface where there is limited diffusion, poly -oniums are still potent biocides. This suggests that surfacebound poly -oniums kill bacteria through different modes of action. The first biocidal mechanism, known as the polymeric spacer effect, demonstrated that surface-grafted poly -onium Received: November 21, 2016 Accepted: February 1, 2017 Published: February 1, 2017 7745

DOI: 10.1021/acsami.6b14940 ACS Appl. Mater. Interfaces 2017, 9, 7745−7751

Research Article

ACS Applied Materials & Interfaces

estimated using a quantitative live/dead staining assay. The fact that PQA with a high cross-linking density led to diminished antimicrobial activity, along with correlated effects in surface charge density and hydrophobicity, strongly supports the phospholipid sponge effect.

might penetrate the bacterial cell wall, reach the cell membrane, disrupt the phospholipid bilayer structural organization, and thus kill the cell.12 This hypothesis was supported by the dependence of the anitimicrobial activity of surface-grafted quarternized polyethyleneimine (PEI) on the molecular weight of the polymer observed by Lin and Haldar.7,13 However, a series of comb-like polyquaternanry ammonium (PQA) cations with various spacers was reported with similar absorption of a negatively charged phospholipid, which implied a similar biocidal efficacy produced by the PQAs upon contact with bacteria.14 This observation suggested the mechanism deviated from the action of hydrophobic spacer intercalation. Bieser postulated an alternative hypothesis, the so-called “phospholipid sponge effect”, which demonstrated that surface-immobilized poly -oniums could pull anionic phospholipids out of the bacterial membrane and sequester them within the polymer matrix, causing hole formation in the cell membrane.1 In their study, surface-bound cellulose functionalized with PQA was deactivated after treatment with negatively charged phospholipids but not with net-neutral zwitterionic phospholipids. This hypothesis was also supported by the work of Li, in which a surface-coated quaternized chitosan-based hydrogel showed increasing antimicrobial activity with increasing gel porosity.15 However, there has been a lack of direct experimental evidence to prove that the phospholipids are adsorbed and retained in the polymer matrix in order to identify the phospholipid sponge effect as the primary biocidal route of surfaceimmobilized -onium polycations. The benzophenone (BP) chromophore has been well studied and extensively used as a photoactive species for many reasons including its tolerance to a wide range of chemical environments, high photochemical reactivity, and thermal and chemical stability.16−19 Upon absorption of UV light, promotion of one electron from an n orbital to the π* orbital of the carbonyl of BP yields a biradicaloid triplet state where the electron-deficient oxygen n orbital interacts with the surrounding C−H δ bonds, leading to hydrogen abstraction to complete the half-filled n orbital. The two resulting carbon radicals then combine to form a new C−C bond.20 Addition of BP as dopants or comonomers has been used to achieve photo-cross-linking of a variety of polymers in the solid and melt states.21−23 Copolymers containing covalently bound BP cross-linkers have been described as a powerful route to prepare cross-linked polymer networks and hydrogels for various applications including surface-attached films/coatings, biosensors, and patterned sheets.24−27 The photoinduced cross-linking mechanism, kinetics of polymer network formation, and dependence on comonomer chemistry have been well studied in the materials chemistry field.28−30 In this study, we provide evidence for the phospholipid sponge effect in contact-active PQAs using a series of BPsubstituted quaternized PEIs with a range of cross-linking densities. The cross-linking density of the polymer network was varied by tuning irradiation time, and the mechanical properties of the polymer thin films were measured to discriminate different cross-linking densities. The correlation between the negatively charged and the neutral phospholipids absorption and cross-linking density of PQA films was investigated to simulate the contact-active biocidal action. The observation that the swollen film fraction reduced as cross-linking density increased is in line with the concept of a phospholipid sponge effect. More importantly, the dependence of killing efficiency on the cross-linking density, i.e., porosity, of the PQA films was

■

EXPERIMENTAL SECTION

Materials. Bromododecane, silver nitrate, iodomethane, and isobutyltrichlorosilane were purchased from Alfa Aesar. Potassium carbonate and tert-amyl alcohol were purchased from JT Baker. Poly(2-ethyl-2-oxazoline) (Mw = 50 kDa)) (Aldrich), hydrochloric acid (BDH), 4-methylbenzophenone (Oxchem), N-bromosuccinimide (Sigma-Aldrich), fluorescein sodium salt (Sigma-Aldrich), and cetyltrimethylammonium bromide (Acros Organic) were used without further purification. 1,2-Diphytanoyl-sn-glycero-3-[phospho-rac-(1glycerol)] sodium salt (DPhPG) and 1,2-diphytanol-sn-glycero-3phosphocholine (DPhPC) were obtained from Avanti. Acetone (BDH) was dried over molecular sieves (EMD Chemical) before usage. The two-color fluorescent LIVE/DEAD BacLight bacterial viability kit L7012 (Molecular Probes, Life Technologies) which contains SYTO 9 green-fluorescent nucleic acid stain and a propidium iodide red-fluorescent nucleic acid stain was utilized to evaluate the bacterial viability. Silicon wafer with a native oxide, glass slides, and quartz slides were used as substrates. All substrates were silylated with an isobutyltrichlorosilane self-assembled monolayer prior to deposition of the polymer coating to provide C−H bonds for cross-linking. Synthesis of Copolymer of N,N-Dodecyl Methyl and N,NMethylbenzophenone Methyl Quaternary PEI (DMBQPEI). The tertiary N-dodecyl-co-N-methylbenzophenone PEI (DMBPEI) was prepared using procedures reported in our previous study.31 The quaternary copolymer, DMBQPEI, was then prepared by quaternization with iodomethane and silver nitrate by the assistance of silver iodide precipitation. The final pure product is a red solid. Experimental details and 1H NMR spectroscopy of DMBPEI and DMBQPEI are provided in the Supporting Information, Figures S6 and S7, respectively. Preparation of Cross-Linked DMBQPEI Immobilized on Surface. DMBQPEI films were deposited on an alkylated silicon wafer, glass, or quartz slides (2.5 × 1 cm) by spin casting with 250 μL of polymer solution (10 mg/mL, in chloroform) at 1000 rpm. Then polymer films were irradiated under UV light for various times in order to obtain the desired cross-linking density. The UV light source was a Compact UV lamp (UVP) with bulbs of wavelength at 254 nm (power 6.5 mW cm−1). DMBQPEI Bulk Polymer and Thin Film Characterizations. The glass transition temperature (Tg) of DMBQPEI was measured by using a differential scanning calorimeter DSC 823e (Mettler Toledo). Thermogravimetric analysis (TGA) was performed with a Discovery TGA (TA Instruments). Water contact angles were measured using a DSA 100 drop shape analysis system (KRŰ SS) with a computercontrolled liquid dispensing system. The thickness of polymer films was measured using an M-2000 spectroscopic ellipsometer (J.A. Woollam Co., Inc.). Surface charge density of DMBQPEI-coated substrates was analyzed by an indirect method using fluorescein sodium dye by UV−vis spectroscopy.32 The morphologies and moduli of the polymer films were collected using a PeakForce QNM (Bruker Multimode AFM, Scanasyst-AIR) DMBQPEI Photo-Cross-linking Kinetics Study. The photocross-linking kinetics of DMBQPEI was investigated by UV−vis spectroscopy on a Cary Bio Spectrophotometer (Varian). The ex-situ photochemical cross-linking study of DMBQPEI was also conducted by infrared spectroscopy using a Thermo-Nicolet model 6700 spectrometer equipped with a variable-angle grazing angle attenuated total reflection (GART-ATR) accessory (Harrick Scientific). Phospholipid Absorption Study. DMBQPEI films were immersed in DPhPG or DPhPC in PBS solution (1 mg/mL) for 45 min, followed by rinsing with PBS and drying under a stream of nitrogen. The thickness changes were measured by spectroscopic ellipsometry. The elemental composition of the phospholipid 7746


Research Article


Scheme 1. Illustration of How the Cross-Linking Density of DMBQPEI Networks Influences the Antimicrobial Efficiency Based on the Phospholipid Sponge Effect

Figure 1. (A) DMBQPEI photo-cross-linking kinetic study by UV−vis spectroscopy. (B) Thickness change and contact angle of DMBQPEImodified surfaces with various irradiation times. adsorbed polymer films was studied using scanning electron microscopy along with energy-dispersive X-ray spectroscopy (FEI Teneo with Oxford EDS, FEI Co., Hillsboro, OR). Live/Dead Bacterial Viability Assay. The antimicrobial activity of DMBQPEI with different cross-linking density was estimated by examining the bacterial cell viability after direct contact with polymer films using the two-color fluorescent live/dead assay. SYTO 9 dye, which yields green fluorescence, labels all bacteria in the population with an intact membrane. In contrast, propidium iodide, which yields red fluorescence, penetrates only bacteria with damaged membranes and displaces the SYTO 9 stain, causing a reduction in green fluorescence and the appearance of red fluorescence. Consequently, bacteria with damaged cell membranes can be distinguished from live ones. S. auresus (ATCC 5538) and E. coli (ATCC 11303) was tested in the assay. The staining method was based on our previous study.33 Quantitative counting of the bacterial cells for the live/dead assay was estimated by using CellProfiler software as recommended by published reports.34,35 PeakForce Quantitative Nanomechanical Mapping (PFQNM). PeakForce QNM is a scanning probe technique which provides topography correlation to quantitative mechanical analysis mapping in real time. To obtain the quantitative elastic modulus results with QNM, the linear portion of the retract curve is fit using the Derjaguin−Müller−Toporov (DMT) model36

Ftip =

4 * E Rd3 + Fadh 3

(1)

where Ftip is the force on the cantilever, Fadh is the adhesion force, R is the tip end radius, d is the tip−sample separation, and E* is the reduced elastic modulus. The relationship between E* and the crosslinking density may be extrapolated for sufficiently thick films and small values of d. For a cross-linked polymer, the cross-linking density influences the modulus of the film, as described in eq 2

3ρRT E* = Mc

(2)

where Mc is the number-average molar mass of the chain lengths between cross-links, ρ is the bulk density of polymer, T is temperature, and R is the gas constant.37,38 As Mc decreases, E* increases, which indicates that the polymer becomes stiffer as the cross-link density increases. The DMT moduli of DMBQPEI thin films were estimated using PeakForce quantitative nanomechanical mapping (QNM) to characterize the cross-linking densities. By fitting the linear portion of the retraction force curve, the Young’s modulus of the film, Es, may be determined by eq 3 7747


Research Article

ACS Applied Materials & Interfaces Table 1. DMBQPEI Thin Film Thickness and Charge Density with Various Irradiation Times irradiation time (min) thickness (nm) charge density (nm−2)

before irradiation after irradiation

0 75.44 N/A 10.83 ± 0.31

1 75.63 71.09 10.44 ± 0.62

3 77.03 68.15 10.82 ± 0.90

5 78.54 64.82 10.83 ± 0.71

Figure 2. (A) DMT modulus measured by AFM Peak-Force QNM and antimicrobial efficiency of DMBQPEI-modified surfaces with various irradiation time; live−dead staining fluorescent images of S. aureus on polymer-modified surfaces with various irradiation times: no irradiation (B), 1 min (C), 3 min (D), and 5 min (E) (scale bar 10 μm). 2 ⎤−1 ⎡ 1 − v2 1 − vtip s ⎥ E* = ⎢ + ⎢⎣ Es Etip ⎥⎦

cross-linked with an irradiation time of 0, 1, 3, and 5 min were chosen for the following studies due to their considerable difference in cross-linking density. FTIR spectra of films irradiated with these time periods were also recorded to monitor the photochemical conversion (Figure S3). The gradual decrease of the BP carbonyl peak at 1665 cm−1 is indicative of the photo-cross-linking reaction. Table 1 shows the dry thickness of spin-casted DMBQPEI films before and after various irradiation times. A thickness decrease in the polymer films indicates an increase in bulk density, which is due to ascending cross-linking density with prolonged irradiation. The moduli of DMBQPEI films with different irradiation times are plotted in Figure 2A and found to increase with longer UV exposure. As more of the BP reacts with surrounding alkyl groups, side chain movements are restricted and the polymer network becomes denser, resulting in a higher modulus. The AFM topographic and modulus images of DMBQPEI films are shown in Figure S4. Thus, a series of surface-bound PQA with a range of cross-linking densities was prepared to investigate the biocidal action of these films. In the earlier discussion on the phospholipid sponge effect, surface-bound polycations are presumed to draw anionic lipids from the bacterial cell membrane and sequester these ions within the polymer matrix when bacteria are in close proximity. Conversely, neutral lipids are not attracted to the ionic network. Furthermore, the polycations with high porosity should have a high likelihood in receiving anionic lipids given the same charge density (Scheme 1). To confirm this assumption, DMBQPEI films with different cross-linking densities were exposed to both neutral and charged phospholipid solutions. The film swelling upon anionic lipid absorption is directly proportional to the polymer network porosity, which is inversely related to its cross-linking density. DMBQPEI films were exposed to blank (control), DPhPC (neutral), and DPhPG (anionic)/PBS solutions, and their interactions with phospholipids were evaluated by measuring the film thickness changes (Figure 1B). When immersed in

(3)

where vs and vtip are the Poisson’s ratio of the sample and tip, respectively. The software assumes that the tip modulus,Etip, is infinite, which effectively removes the term from the calculation.39 The sample Poisson’s ratio of 0.3 was used for calculating Es.

■

RESULTS AND DISCUSSION The DMBQPEI polymer composition was verified using NMR spectroscopy, which revealed that the molar ratio of dodecyl to benzophenone pendant groups was 2:1 (structure shown in Scheme 1). The hydrophobic quaternary PEI with long alkyl side chains has been demonstrated as one of the most potent surface-bound antimicrobial polycations.9,13,40,41 The glass transition temperature, Tg, of DMBQPEI was broad and measured to be ∼39 °C at the onset (Figure S1). Due to the local thermal heating caused by UV lamp irradiation, the measured temperature (using an IR thermometer) at the irradiation spot was 49 °C, well above the glass transition temperature, which encourages segmental motion and facilitates hydrogen abstraction and carbon recombination between the benzophenone chromophore and other C−H bonds on the silylated surface and along the polymer backbone and pendant groups. The TGA thermogram (Figure S1) shows the onset of polymer degradation near 150 °C, verifying film thermal stability near irradiation temperatures. In the presence of the long alkyl side chains, a robust covalent polymeric network is formed on alkylated surfaces triggered by UV irradiation, and the cross-linking density of the polymer films can be tuned by varying irradiation time. The photo-cross-linking kinetics of DMBQPEI was investigated by UV−vis spectroscopy (Figure 1A). The conversion of the cross-linking reaction was determined by plotting the decay of absorbance peaks of the BP moiety (λmax at 255 nm in Figure S2) as a function of irradiation dosage. On the basis of the kinetics of the photoreaction, DMBQPEI films that were 7748


Research Article

ACS Applied Materials & Interfaces PBS, the cross-linked films were observed to shrink to some extent (red columns in Figure 1B). Slides treated with control, DPhPC, and DPhPG solutions were further examined using SEM. A clear morphological change in the polymer film was observed for the film exposed to the anionic DPhPG surfactant, while the control and DPhPC solutions showed little to no morphological difference (Figure S9(A−C). Energy-dispersive X-ray spectroscopy (EDS) was used to determine the qualitative elemental composition of the film (Figure S9(DE)). A clear phosphorus signal was observed in the DPhPGcontaining EDS spectrum, which confirmed the adsorption of the anionic phospholipid. No phosphorus signal was observed from either the DPhPC exposed film or the control EDS spectra. Furthermore, the corresponding EDS mapping to the DPhPG-reated film (Figure S10) shows the relative intensities of the identified elements, with a uniform distribution of P signal intensity throughout the film. The cross-linking reaction produces a hydroxyl group for each molecule of photoconverted BP. The conversion of carbonyl to alcohol may be probed with contact angle measurements; concordantly, as photo-cross-linking proceeded to 5 min, the water contact angle reduced from 92° to 82° (black squares in Figure 1B). Therefore, the hydrophilicity of the DMBQPEI film increases with irradiation time. The charge densities of all films regardless of irradiation time remained statistically constant (Table 1) as the overall amount of polyelectrolyte bound to the surface did not change with irradiation. Irradiation time also influences the final film thickness of the covalent networks along with the swelling behavior toward phospholipids. When comparing the irradiated dry films after treatment with solutions of zwitterionic (DPhPC) versus anionic (DPhPG) lipids, the films treated with anions swell while immersion in both the neutral zwitterions and the control PBS solutions lead to a decrease in film thickness. It is likely that photochemical irradiation not only leads to cross-linking but also can lead to β-scission as a potential pathway for the carbon radical after hydrogen abstraction. 42 Hydrogen abstraction along the polymer backbone can lead to a chain scission event, which reduces the molecular weight of the polymer and can result in a reduction in the film thickness if soluble oligomers are present. We observed this behavior previously when using BP to cross-link polyacrylates.43 The overall reduction in film thickness that occurs in the films treated with PBS and zwitterionic lipids suggests that the films do not sequester ions, while the films treated with DPhPG solutions are capable of sequestering anions, resulting in film swelling. Moreover, the film thickness increase with anionic lipids was negatively correlated to the cross-linking density. This trend is in agreement with the concept of the phospholipid sponge effect: the lightly cross-linked film possesses high interior volume for anionic lipids to diffuse into, so the polymer networks swell to a great extent with lipids “stored” within it. Densely cross-linked films have a lower overall pore volume and stiffer polymeric matrix so that few lipids can be adsorbed. In order to confirm that the surface-bound poly -oniums are lethal to bacteria on contact due to the phospholipid sponge effect, DMBQPEI with various cross-linking density were challenged against S. aureus and E. coli and the antimicrobial activity of the films was determined using live−dead stain assay. Figure 2B, 2C, 2D, and 2E shows the representative fluorescent images of S. aureus cells in contact with DMBQPEI-modified

surfaces irradiated for 0, 1, 3, and 5 min, respectively. Using films irradiated under identical conditions, Figure S8 (B−E) shows representative fluorescent images of E. coli cells in contact with DMBQPEI-modified surfaces. Figure S5 shows the fluorescent images of both Gram-positive and Gram-negative bacteria on bare glass as a control. A quantitative evaluation of the killing efficiency was achieved by analyzing the red fluorescent (membrane disrupted) and green fluorescent (membrane intact) intensities using CellProfiler software, and the results are plotted in Figure 2A (red circles) for S. aureus and Figure S8A for E. coli. It was found that increased crosslinking of DMBQPEI films also negatively impacted their killing efficiency. Relative to the non-cross-linked film, the film with the highest modulus, ergo the highest cross-linking density, lost 32% and 47% of antimicrobial activity for S. aureus and E. coli, respectively. To exclude the possibility that charge density variation affects the antimicrobial activity, the number of surface-bound ammoniums of four films was measured. The films were found to have similar charge density (Table 1); therefore, it is revealed that the void volume of the polycation networks plays a role in the contact-active biocidal action. The negative correlation between the modulus and the killing efficiency of the immobilized DMBQEI further supports a phospholipid sponge killing mechanism (Figure 2A). The change in the volume of voids within the polycations networks is able to influence the antimicrobial activity only by this mechanism, considering the same molecular composition and comparable charge densities.

■

CONCLUSIONS In conclusion, a quaternary ammonium polymer, DMBQPEI, with pendent benzophenone as a photoreactive cross-linker was prepared, and the systematic variation of cross-linking density was performed to validate the phospholipid sponge effect. Variable cross-linking density was achieved by tuning the irradiation time, and the photo-cross-linking process was examined via UV−vis spectroscopy, static contact angle measurements, and PeakForce QNM. A series DMBQPEI films with different cross-linking densities was exposed to neutral and anionic phospholipids to simulate the action of lipid suction proposed in the phospholipid sponge effect. The thickness changes of these films after lipid exposure were monitored to characterize the absorption amount. A negative correlation between the cross-linking density and the absorption of negatively charged phospholipid was found, which supports the hypothesis that anionic phospholipids are sequestered in the polycationic networks, whereas neutral lipids were not adsorbed by DMBQPEI films. More convincing evidence to prove that bacterial death is caused by the loss of the membrane lipids was obtained by studying the relationship between cross-linking density of DMBQPEI films and killing efficacy on contact with S. aureus and E. coli. Using the live/dead stain assay, we quantitatively evaluated the antimicrobial activity of DMBQPEI films with increasing cross-linking densities. It was found that killing efficiency is inversely dependent on the DMT modulus, given similar surface charge density and hydrophobicity of films. To our knowledge, this the first report demonstrating that the cross-linking density of surface-bound poly -onium networks is a defining parameter in antimicrobial activity. This observation can be readily explained by the phospholipid sponge effect: lipids are pulled from the cell membrane when bacteria are on 7749


Research Article


(8) Botequim, D.; Maia, J.; Lino, M. M. F.; Lopes, L. M. F.; Simões, P. N.; Ilharco, L. M.; Ferreira, L. Nanoparticles and Surfaces Presenting Antifungal, Antibacterial and Antiviral Properties. Langmuir 2012, 28, 7646−7656. (9) Milović, N. M.; Wang, J.; Lewis, K.; Klibanov, A. M. Immobilized N-alkylated polyethylenimine avidly kills bacteria by rupturing cell membranes with no resistance developed. Biotechnol. Bioeng. 2005, 90, 715−722. (10) Lin, J.; Tiller, J. C.; Lee, S. B.; Lewis, K.; Klibanov, A. M. Insights into bactericidal action of surface-attached poly(vinyl-Nhexylpyridinium) chains. Biotechnol. Lett. 2002, 24, 801−805. (11) Kügler, R.; Bouloussa, O.; Rondelez, F. Evidence of A ChargeDensity Threshold for Optimum Efficiency of Biocidal Cationic Surfaces. Microbiology 2005, 151, 1341−1348. (12) Tiller, J. C.; Liao, C.-J.; Lewis, K.; Klibanov, A. M. Designing Surfaces that Kill Bacteria on Contact. Proc. Natl. Acad. Sci. U. S. A. 2001, 98, 5981−5985. (13) Lin, J.; Qiu, S.; Lewis, K.; Klibanov, A. M. Bactericidal Properties of Flat Surfaces and Nanoparticles Derivatized with Alkylated Polyethylenimines. Biotechnol. Prog. 2002, 18, 1082−1086. (14) Bieser, A. M.; Thomann, Y.; Tiller, J. C. Contact-Active Antimicrobial and Potentially Self-Polishing Coatings Based on Cellulose. Macromol. Biosci. 2011, 11, 111−121. (15) Li, P.; Poon, Y. F.; Li, W. F.; Zhu, H. Y.; Yeap, S. H.; Cao, Y.; Qi, X. B.; Zhou, C. C.; Lamrani, M.; Beuerman, R. W.; Kang, E. T.; Mu, Y. G.; Li, C. M.; Chang, M. W.; Leong, S. S. J.; Chan-Park, M. B. A polycationic antimicrobial and biocompatible hydrogel with microbe membrane suctioning ability. Nat. Mater. 2011, 10, 149−156. (16) Lin, A. A.; Sastri, V. R.; Tesoro, G.; Reiser, A.; Eachus, R. On the Crosslinking Mechanism of Benzophenone-Containing Polyimides. Macromolecules 1988, 21, 1165−1169. (17) Turro, N. J. Modern Molecular Photochemistry; Benjamin/ Cummings Pub. Co.: Menlo Park, CA, 1978. (18) Park, M.-K.; Deng, S.; Advincula, R. C. pH-Sensitive Bipolar Ion-Permselective Ultrathin Films. J. Am. Chem. Soc. 2004, 126, 13723−13731. (19) Pahnke, J.; Rühe, J. Attachment of Polymer Films to Aluminium Surfaces by Photochemically Active Monolayers of Phosphonic Acids. Macromol. Rapid Commun. 2004, 25, 1396−1401. (20) Dorman, G.; Prestwich, G. D. Benzophenone photophores in biochemistry. Biochemistry 1994, 33, 5661−5673. (21) Qu, B. J.; Rånby, B. Photocross-linking of low-density polyethylene. I. Kinetics and reaction parameters. J. Appl. Polym. Sci. 1993, 48, 701−709. (22) Doytcheva, M.; Dotcheva, D.; Stamenova, R.; Orahovats, A.; Tsvetanov, C.; Leder, J. Ultraviolet-induced crosslinking of solid poly(ethylene oxide). J. Appl. Polym. Sci. 1997, 64, 2299−2307. (23) Graves, R.; Pintauro, P. N. Polyphosphazene membranes. II. Solid-state photocrosslinking of poly[(alkylphenoxy) (phenoxy)phosphazene] films. J. Appl. Polym. Sci. 1998, 68, 827−836. (24) Toomey, R.; Freidank, D.; Rühe, J. Swelling Behavior of Thin, Surface-Attached Polymer Networks. Macromolecules 2004, 37, 882− 887. (25) Beines, P. W.; Klosterkamp, I.; Menges, B.; Jonas, U.; Knoll, W. Responsive Thin Hydrogel Layers from Photo-Cross-Linkable Poly(N-isopropylacrylamide) Terpolymers. Langmuir 2007, 23, 2231− 2238. (26) Mateescu, A.; Wang, Y.; Dostalek, J.; Jonas, U. Thin Hydrogel Films for Optical Biosensor Applications. Membranes 2012, 2, 40. (27) Kim, J.; Hanna, J. A.; Byun, M.; Santangelo, C. D.; Hayward, R. C. Designing Responsive Buckled Surfaces by Halftone Gel Lithography. Science 2012, 335, 1201−1205. (28) Christensen, S. K.; Chiappelli, M. C.; Hayward, R. C. Gelation of Copolymers with Pendent Benzophenone Photo-Cross-Linkers. Macromolecules 2012, 45, 5237−5246. (29) Suyama, K.; Miyamoto, Y.; Matsuoka, T.; Wada, S.; Tsunooka, M. Photo-initiated thermal crosslinking of copolymers bearing pendant base generating groups. Polym. Adv. Technol. 2000, 11, 589−596.

contact with surface-bound DMBQPEI and sequestered by the polymer network. Moreover, the larger the void volume of the network, the more lipids the polymer network can uptake, resulting in a higher killing efficacy of the film. These results should aid in the design of antimicrobial surfaces in future studies.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsami.6b14940. DSC thermogram of DMBQPEI, TGA thermogram of DMBQPEI, 1H NMR of DMBPEI and DMBQPEI, UV− vis spectroscopy of DMBQPEI film, FTIR spectroscopy of DMBQPEI thin film, AFM topographic and modulus maps of DMBQPEI films, live−dead staining fluorescent images of S. aureus and E. coli on bare glass, SEM images of DMBQPEI films exposed to phospholipids, and EDS mapping of DPhPG treated polymer film (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. ORCID

Jason Locklin: 0000-0001-9272-2403 Author Contributions

The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The authors gratefully acknowledge financial support by the Centers for Disease Control and Prevention (contract 2002016-91933). Special thanks to Dr. Sean Hopkins for quantitative analysis of live/dead stain assay and Yutian Ke for SEM analysis.

■

REFERENCES

(1) Bieser, A. M.; Tiller, J. C. Mechanistic Considerations on Contact-Active Antimicrobial Surfaces with Controlled Functional Group Densities. Macromol. Biosci. 2011, 11, 526−534. (2) Ferreira, L.; Zumbuehl, A. Non-leaching Surfaces Capable of Killing Microorganisms on Contact. J. Mater. Chem. 2009, 19, 7796− 7806. (3) Yatvin, J.; Gao, J.; Locklin, J. Durable Defense: Robust and Varied Attachment of Non-leaching Poly″-Onium″ Bactericidal Coatings to Reactive and Inert Surfaces. Chem. Commun. 2014, 50, 9433−9442. (4) Yudovin-Farber, I.; Beyth, N.; Nyska, A.; Weiss, E. I.; Golenser, J.; Domb, A. J. Surface Characterization and Biocompatibility of Restorative Resin Containing Nanoparticles. Biomacromolecules 2008, 9, 3044−3050. (5) Koplin, S. A.; Lin, S.; Domanski, T. Evaluation of the Antimicrobial Activity of Cationic Polyethylenimines on Dry surfaces. Biotechnol. Prog. 2008, 24, 1160−1165. (6) Ravikumar, T.; Murata, H.; Koepsel, R. R.; Russell, A. J. SurfaceActive Antifungal Polyquaternary Amine. Biomacromolecules 2006, 7, 2762−2769. (7) Haldar, J.; An, D.; Á lvarez de Cienfuegos, L.; Chen, J.; Klibanov, A. M. Polymeric coatings that inactivate both influenza virus and pathogenic bacteria. Proc. Natl. Acad. Sci. U. S. A. 2006, 103, 17667− 17671. 7750


Research Article

ACS Applied Materials & Interfaces (30) Higuchi, H.; Yamashita, T.; Horie, K.; Mita, I. Photo-crosslinking reaction of benzophenone-containing polyimide and its model compounds. Chem. Mater. 1991, 3, 188−194. (31) Dhende, V. P.; Samanta, S.; Jones, D. M.; Hardin, I. R.; Locklin, J. One-Step Photochemical Synthesis of Permanent, Nonleaching, Ultrathin Antimicrobial Coatings for Textiles and Plastics. ACS Appl. Mater. Interfaces 2011, 3, 2830−2837. (32) Murata, H.; Koepsel, R. R.; Matyjaszewski, K.; Russell, A. J. Permanent, Non-Leaching Antibacterial Surfaces2: How High Density Cationic Surfaces Kill Bacterial Cells. Biomaterials 2007, 28, 4870−4879. (33) Gao, J.; Huddleston, N. E.; White, E. M.; Pant, J.; Handa, H.; Locklin, J. Surface Grafted Antimicrobial Polymer Networks with High Abrasion Resistance. ACS Biomater. Sci. Eng. 2016, 2, 1169−1179. (34) Carpenter, A. E.; Jones, T. R.; Lamprecht, M. R.; Clarke, C.; Kang, I. H.; Friman, O.; Guertin, D. A.; Chang, J. H.; Lindquist, R. A.; Moffat, J.; Golland, P.; Sabatini, D. M. CellProfiler: image analysis software for identifying and quantifying cell phenotypes. Genome Biology 2006, 7, R100. (35) Kamentsky, L.; Jones, T. R.; Fraser, A.; Bray, M.-A.; Logan, D. J.; Madden, K. L.; Ljosa, V.; Rueden, C.; Eliceiri, K. W.; Carpenter, A. E. Improved structure, function and compatibility for CellProfiler: modular high-throughput image analysis software. Bioinformatics 2011, 27, 1179−1180. (36) Derjaguin, B. V.; Muller, V. M.; Toporov, Y. P. Effect of contact deformations on the adhesion of particles. J. Colloid Interface Sci. 1975, 53, 314−326. (37) Charlesby, A. Elastic modulus formulae for a crosslinked network. International Journal of Radiation Applications and Instrumentation. Part C. Radiation Physics and Chemistry 1992, 40, 117−120. (38) Larché, J. F.; Seynaeve, J. M.; Voyard, G.; Bussière, P. O.; Gardette, J. L. Characterizing Mesh Size Distributions (MSDs) in Thermosetting Materials Using a High-Pressure System. J. Phys. Chem. B 2011, 115, 4273−4278. (39) Pittenger, B. B.; Erina, N.; Su, C. Quantitative mechanical property mapping at the nanoscale with PeakForce QNM. Veeco Application Note #128; Bruker Nano Surfaces Division: Santa Barbara, CA, 2010. (40) Park, D.; Wang, J.; Klibanov, A. M. One-Step, Painting-Like Coating Procedures To Make Surfaces Highly and Permanently Bactericidal. Biotechnol. Prog. 2006, 22, 584−589. (41) Lin, J.; Qiu, S.; Lewis, K.; Klibanov, A. M. Mechanism of bactericidal and fungicidal activities of textiles covalently modified with alkylated polyethylenimine. Biotechnol. Bioeng. 2003, 83, 168−172. (42) Torikai, A.; Sekigawa, Y.; Fueki, K. Photo-degradation of blends of polystyrene and poly(methyl methacrylate). Polym. Degrad. Stab. 1988, 21, 43−54. (43) Gao, J.; Martin, A.; Yatvin, J.; White, E.; Locklin, J. Permanently grafted icephobic nanocomposites with high abrasion resistance. J. Mater. Chem. A 2016, 4, 11719−11728.

7751


Evidence for the Phospholipid Sponge Effect as the Biocidal

Recommend Documents