ia

, Ja

a r t i c l e i n f o

Article history:Received 28 June 2013Accepted 8 September 2013Available online 30 September 2013

Keywords:Block copolymersPolycarbonatePoly(ethylene glycol)Hydrogel

effective antimicrobial formulations that can be applied broadly ina variety of biomedical-associated contexts. Antimicrobial hydro-gels, which are heavily hydrated networks of polymers possessing

being exploited inmedication, disin-[1]. Ideally, thesetively, eliminatingc as well as biolming the emergence

of resistant subpopulation since antimicrobial resistance hasbecome a major global healthcare concern.

The development of an effective antimicrobial hydrogel requiresthe understanding of the different growth behavior and treatmentsusceptibility between planktonic cells and those embeddedwithinbiolms. The formation of biolms is initiated by the deposition ofplanktonic cells onto living tissue or an inanimate surface, such asthose of medical devices. These cells adhere and anchor themselvesto the surfaces via the production of exopolymers. As proliferation

* Corresponding author. Tel.: 65 6824 7106; fax: 65 6478 9084.** Corresponding author. Tel.: 1 408 927 1632; fax: 1 408 927 3310.

E-mail addresses: hedrick@almaden.ibm.com (J.L. Hedrick), yyyang@ibn.a-

Contents lists availab

Biomat

journal homepage: www.elsev

Biomaterials 34 (2013) 10278e10286star.edu.sg (Y.Y. Yang).1. Introduction

In the on-going battle against microbial infections, the risks ofacquiring drug resistance has not ceased despite a considerablenumber of strategies that have been devised for new cellular tar-gets. Bearing this in mind, there is an increasing need for more

the ability to eliminate infectious microbes, aremany pharmaceutical applications, includingfectants, sanitizers and personal care productsmaterials should exert antimicrobial actions effecand preventing the recurrence of both planktoniorganisms. This is essentially important for limit 2013 Elsevier Ltd. All rights reserved.BiolmAntimicrobial0142-9612/$ e see front matter 2013 Elsevier Ltd.http://dx.doi.org/10.1016/j.biomaterials.2013.09.029a b s t r a c t

Current antimicrobial strategies have mostly been developed to manage infections due to planktoniccells. However, microbes in their nature state will tend to exist by attaching to and growing on living andinanimate surfaces that result in the formation of biolms. Conventional therapies for treating biolm-related infections are likely to be insufcient due to the lower susceptibility of microbes that areembedded in the biolm matrix. In this study, we report the development of biodegradable hydrogelsfrom vitamin E-functionalized polycarbonates for antimicrobial applications. These hydrogels wereformed by incorporating positively-charged polycarbonates containing propyl and benzyl side chainswith vitamin E moiety into physically cross-linked networks of ABA-type polycarbonate and poly(-ethylene glycol) triblock copolymers. Investigations of the mechanical properties of the hydrogelsshowed that the G0 values ranged from 1400 to 1600 Pa and the presence of cationic polycarbonate didnot affect the stiffness of the hydrogels. Shear-thinning behavior was observed as the hydrogels dis-played high viscosity at low shear rates that dramatically decreased as the shear rate increased. In vitroantimicrobial studies revealed that the more hydrophobic VE/BnCl(1:30)-loaded hydrogels generallyexhibited better antimicrobial/antifungal effects compared to the VE/PrBr(1:30) counterpart as lowerminimum biocidal concentrations (MBC) were observed in Staphylococcus aureus (Gram-positive),Escherichia coli (Gram-negative) and Candida albicans (fungus) (156.2, 312.5, 312.5 mg/L for VE/BnCl(1:30) and 312.5, 2500 and 625 mg/L for VE/PrBr(1:30) respectively). Similar trends were observedfor the treatment of biolms where VE/BnCl(1:30)-loaded hydrogels displayed better efciency withregards to eradication of biomass and reduction of microbe viability of the biolms. Furthermore, a highdegree of synergistic antimicrobial effects was also observed through the co-delivery of antimicrobialpolycarbonates with a conventionally-used antifungal agent, uconazole. These hydrogels also displayedexcellent compatibility with human dermal broblasts with cell viability >80% after treatment withhydrogels loaded with cationic polymers and/or uconazole at minimum biocidal concentrations (MBC).b IBM Almaden Research Center, 650 Harry Roaa Institute of Bioengineering and Nanotechnology, 31 Biopolis Way, The Nanos, Singapore 138669, Singapored, San Jose, CA 95120, USABlock copolymer mixtures as antimicroberadication

Ashlynn L.Z. Lee a, Victor W.L. Ng a, Weixin Wang aAll rights reserved.l hydrogels for biolm

mes L. Hedrick b, **, Yi Yan Yang a, *

le at ScienceDirect

erials

ier .com/locate/biomateria ls

rialsoccurs, microcolonies appear and the thickening of polymer matrixaround the microcolonies results in the growth of the biolm [2,3].Biolm formation has been recognized as one of the leading causesof a signicant amount of human infections [4,5]. Pathogens thatare commonly associated with biolm-induced chronic infectionsinclude Staphylococcus aureus in chronic rhinosinusitis [6],enteropathogenic Escherichia coli in recurrent urethritis [7,8] andCandida albicans in candiasis [9].

Microbes that grow in biolms have been known to exhibitdramatically higher resistance against antimicrobial agentscompared to free-oating cells [10e12]. Several factors contributeto the intrinsically lower susceptibility to antimicrobials, includingrestricted penetration of antimicrobials into a biolm due tolimited diffusion within the cell polymer matrix, decreased growthrate which minimizes the uptake of biocides into the cells, as wellas the expression of possible resistance genes [13,14]. The minimalinhibitory concentration (MIC) and minimal bactericidal concen-tration (MBC) of antimicrobials to biolm-associated microbes varywidely depending on the strains and drugs used, and in some case,it may be up to 1000-fold higher compared with planktonic bac-teria [2]. For instance, Stoodley et al. reported that biolm-associated S. aureus required 3.5 times higher than the MBC ofoxacillin to provide a 3.0-log reduction in bacterial counts [15], andCeri et al. reported that E. coli required >500 times the MIC ofampicillin for a 3.0-log reduction in bacteria growth [16].

With the emergence of antibiotic-resistant superbugs, thedevelopment of newer, more potent antibiotics with antimicrobialactions that are different from the conventional small moleculedrugs is crucial [17]. Polymeric biocides are a class of materials thatare engineered as synthetic mimics of naturally occurring host-defense antimicrobial peptides. These amphiphilic, cationic poly-mers are able to selectively target and bind to microbes by elec-trostatic interactions and disintegrate the bacterial membranes byinsertion into the membrane lipid bilayer [18,19]. Such membranedisruption mechanism causes rupture and lysis of the microbes,thereby decreasing the potential of resistance development.

When used in vivo, selectively killing of microbes generallycomes from enhanced long-range electrostatic interaction betweenthe polymeric biocides and microbes in comparison to mammaliancells [19,20]. The design of these polymers is dependent of severalfactors that can greatly affect the antimicrobial activity and selec-tivity, such as the molecular weight, hydrophobic/hydrophilic bal-ance, cationic chemical functionality [21,22]. Chan-Park et al.recently reported the antimicrobial effects of UV-crosslinked chi-tosan-based hydrogels and found that the degree of quaternizationof the polymers played a signicant role in inuencing theirbiocidal effects [23]. At a high cationic chitosan concentration(10wt.%), the hydrogels gave rise to>2.0-log reduction inmicrobialcounts (i.e. >99.0% killing efciency) for Pseudomonas aeruginosa,E. coli, S. aureus and Fusarium solani after 1 h exposure, but>3.0-logreduction (i.e. >99.9% killing efciency) was seen only in S. aureusand F. solani. More recently, the same group reported a photo-polymerized hydrogel system based on the antimicrobial peptide-poly-L-lysine (EPL) grafted to methacrylic acid (MA) [24]. Anti-microbial activities of EPL-MA were retained after polymerizationand the hydrogels were able to reduce the counts of S. aureus andP. aeruginosa by more than 99.9% (3-log reduction) at high EPL-MAconcentrations (21e25 wt.%) after 2 h exposure, while the hydro-gels were less effective against C. albicans and F. solani (fungi) (

ials(Wyatt Technology Corporation, U.S.A.) and Waters HR-4E as well as HR 1 columns(Waters Corporation, U.S.A.). The system was equilibrated at 30 C in THF, whichserved as the polymer solvent and eluent with a ow rate of 1.0 mL/min. Polymersolutions were prepared at a known concentration (ca. 3 mg/mL) and an injectionvolume of 100 mL was used. Data collection and analysis were performed using theAstra software (Wyatt Technology Corporation, U.S.A.; version 5.3.4.14). The col-umns were calibrated with series of polystyrene standards ranging fromMp 360 Da to Mp 778 kDa (Polymer Standard Service, U.S.A.).

2.4. General procedure for polymer synthesis

Triblock copolymers of vitamin E-functionalized polycarbonates and poly(-ethylene glycol), and vitamin E-containing cationic polycarbonates were synthe-sized for preparation of antimicrobial hydrogels.

Triblock copolymer ((MTC-VE)1.25-PEG(20k)-(MTC-VE)1.25): In a 20-mL vial con-taining a magnetic stir bar in the glove box, MTC-VE (58.9 mg, 100 mmol, 4.0 equiv.),HO-PEG-OH (20 kDa, 500 mg, 25 mmol, 1.0 equiv.) and TU (9.3 mg, 25 mmol,1.0 equiv.) were dissolved in dichloromethane (4 mL). To this solution, DBU (3.7 mL,25 mmol, 1.0 equiv.) was added to initiate polymerization. The reaction mixture wasallowed to stir at room temperature and aliquots of samples were taken to monitorthe monomer conversion and evolution of molecular weight by 1H NMR spectros-copy and SEC. After 120 min, the reaction mixture was quenched by the addition ofexcess (w20 mg) of benzoic acid and was precipitated into ice-cold diethyl ether(2 50 mL). The resultant polymer was dried in a vial for about 1e2 days until aconstant sample mass was obtained, as white powder. Selected 1H NMR (400 MHz,CDCl3): d 3.40e4.00 (s, 1815H, OCH2CH2 PEG), 1.00e2.00 (overlapping peaks, VitE),0.75e0.95 (m, 30H, overlapping CH3 on VitE). PDI (GPC): 1.07.

Vitamin E-containing cationic polycarbonates (VE/BnCl(1:30)): In a 20-mL vialcontaining a magnetic stir bar in the glove box, MTC-BnCl (608.8 mg, 2.04 mmol, 30equiv.), MTC-VE (40.0 mg, 68 mmol, 1.0 equiv.) and TU (25.2 mg, 68 mmol, 1.0 equiv.)were dissolved in dichloromethane (3 mL). To this solution, BnOH (7.0 mL, 68 mmol,1.0 equiv.) followed by DBU (10.2 mL, 68 mmol, 1.0 equiv.) were added to initiatepolymerization. The reaction mixture was allowed to stir at room temperature for20 min and quenched by the addition of excess (w20 mg) of benzoic acid. Themixture was then precipitated into ice-cold methanol (50 mL) and centrifugedat5 C for 30 min. The resultant semi-transparent oil was dried under vacuo until afoamy white solid was obtained. GPC analysis of the intermediate was carried outand the polymer was used without further purication. The polymer was subse-quently dissolved in acetonitrile, transferred to a Teon-plug sealable tube andchilled to 0 C. Trimethylamine was added to start the quarternization process. Thereaction mixture was stirred at room temperature for 18 h in the sealed tube. Pre-cipitation of an oilymaterial was observed during the course of reaction. Themixturewas evacuated to dryness under vacuo and freeze-dried to nally yield a white crisp-foamy solid. The desired polymer was characterized by 1H NMR (some of the integralvalues are approximated to the nearest DP values). VE/PrBr(1:30) was obtained in asimilar manner using MTC-PrBr as the monomer. The 1H NMR spectra for all thecationic polycarbonates are provided in the Supplementary Information (Fig. S1).

VE/BnCl(1:30): 1H NMR (400 MHz, CDCl3) d 7.20e7.70 (m, 109H, overlappingpeaks of initiator Ph and Bn), 5.00e5.40 (m, 54H, overlapping peaks of initiatorCH2Ph and Bn), 4.67 (m, 52H), 4.30 (m, 104H), 3.07 (m, 234H, N(CH3)), 1.00e2.00(overlapping peaks, VitE), 0.70e0.90 (m, 12H, overlapping CH3 on VitE); PDI of in-termediate (GPC): 1.21; Actual DP of VE/BnCl(1:30) (from 1H NMR)w 1:26.

VE/PrBr(1:30): 1H NMR (400 MHz, CDCl3) d 7.10e7.60 (m, 5H, initiator Ph), 5.15(m, 2H, initiator CH2Ph), 4.00e4.50 (m, 144H), 3.46 (m, 48H), 3.30e3.20 (m, 216H,N(CH3)), 2.10 (m, 48H), 1.00e2.00 (overlapping peaks, VitE), 0.70e0.90 (m, 12H,overlapping CH3 on VitE). PDI of intermediate (GPC): 1.13. Actual DP of VE/PrBr(1:30)(from 1H NMR)w 1:24.

2.5. Rheological experiments

Blank hydrogel and antimicrobial polymer-loaded hydrogels were prepared bydissolving the triblock copolymer or a mixture of the triblock copolymer and avitamin E-containing cationic polycarbonate in HPLC gradewater at 25 C and 4wt%.The rheological analysis of the hydrogels was performed on an ARES-G2 rheometer(TA Instruments, U.S.A.) equipped with a plateeplate geometry of 8 mm diameter.Measurements were taken by equilibrating the gels at 25 C between the plates at agap of 1.0 mm. The data were collected under a controlled strain of 0.2% and afrequency scan of 1.0e100 rad/s. Gelation properties of the polymer solutions wasmonitored bymeasuring the shear storage modulus (G0), as well as the loss modulus(G00), at each point. For shear-thinning studies, the viscosity of the hydrogels wasmonitored as function of shear rate from 0.1 to 10 s1.

2.6. In vitro uconazole release

The release of uconazole from the 4 wt.% (MTC-VE)1.25-PEG(20k)-(MTC-VE)1.25hydrogel was studied. Fluconazole-containing hydrogel was prepared by rst dis-solving the antifungal drug in HPLC grade water at room temperature. This solution

A.L.Z. Lee et al. / Biomater10280was then used to dissolve (MTC-VE)1.25-PEG(20k)-(MTC-VE)1.25 and the mixture wasleft to stand overnight for the formation of hydrogel. The release study wasperformed by placing the hydrogel (1 mL) containing 4 mg/L of uconazole in adialysis membrane tube with MWCO of 1000 Da (Spectrum Laboratories, U.S.A.),which was then immersed in 40 mL of the release medium phosphate-bufferedsaline (PBS, pH 7.4). To study if the presence of the antimicrobial polycarbonatehas any effect on drug release, 4 mg/L of VE/BnCl(1:30) was added to the hydrogelduring the gelation process. The samples were kept shaking on an orbital shaker at100 rpm at 37 C. At designated time intervals, 0.5 mL of the release medium wasremoved and replaced with fresh medium. The removed medium was analyzed forits drug content. To do this, the deposited uconazole was dissolved in 1.5 mL ofmobile phase (A e 10 mM sodium acetate buffer adjusted to pH 5.0, B e Methanol;65%A/35%B) and analyzed using HPLC at 210 nm.

2.7. Killing efciency tests

E. coli and S. aureuswere reconstituted from its lyophilized form according to themanufacturers protocol, and cultured in TSB at 37 C under constant shaking of300 rpm, while C. albicanswas cultured in YMB at room temperature under constantshaking of 50 rpm. Prior to treatment, the microbes were rst inoculated overnightto enter into log growth phase. Cationic polycarbonate (VE/BnCl(1:30) or VE/PrBr(1:30))-containing hydrogels were prepared using 4 wt.% of the triblockcopolymer (MTC-VE)1.25-PEG(20k)-(MTC-VE)1.25 and varying contents of VE/BnCl(1:30), VE/PrBr(1:30) and/or uconazole. Hydrogels (50 mL) were placed intoeachwell of a 96-well microplate containing an equal volume of microbe suspension(3 105 CFU/mL). Prior to this, the concentration of microbe solution was adjustedto give an initial optical density (O.D.) reading of approximately 0.07 at 600 nmwavelength on a microplate reader (TECAN, Switzerland), which corresponds to theconcentration of McFarland 1 solution (3 108 CFU/mL). The culture plate was kepteither at 37 C for bacterial samples or room temperature for C. albicans underconstant shaking of 300 or 50 rpm respectively for 24 h. After treatment, the sam-ples were taken for a series of tenfold dilution, and plated onto agar plates. Theplates were incubated for 24 h at 37 C and the number of colony-forming units(CFU) was counted. Microbes treated with hydrogel without cationic polycarbonateswere used as negative control, and each test was carried out in 3 replicates. Mini-mum bactericidal concentration, MBC, is dened as the concentration of the cationicpolycarbonate that eliminates >99.9% of the microbes.

2.8. Analysis of synergism between antimicrobial polycarbonate and uconazole

To assess the antifungal effects of the polymereuconazole combination, thecheckerboard and isobologram methods of analyzing drug interactions were used.For the checkerboard method, the fractional inhibitory concentration (FBC) wascalculated for each component in each combination dose [30,31]. The types andextent of interactionwas determined by calculating the FBC index, which is the ratioof the MBC of a drug in combination and MBC of the drug alone. For two interactingdrugs, A and B, the sum of the FBCs indicates the extent of the interaction. Synergy isdened as SFIC index 0.5. Indifference was dened as SFIC index of >0.5 but 4and antagonism as a SFIC index of >4.0 [32]. As for the isobologram method, eval-uation of drug interactionwas performed at the MBC level. Using graphical analysis,the concentrations required to produce the effect of >99.9% killing efciency weredetermined for each component and plotted on the x and y axes of a two-coordinateplot. A line is drawn to connect these two points and this is dened as the line ofadditivity. After that, treatment was then performed with the drugs in combinationat varying concentrations. The concentrations of uconazole and polycation in thecombination that provided the same effect were placed in the same plot. Effect of thedrug interactionwas determined according to the position of the points with respectto the line of additivity. Synergy, additivity, and antagonism are represented whenthe point is located below, on, and above the line, respectively [33].

2.9. Biolm formation and treatment

S. aureus and E. coli were grown overnight in TSB at 37 C and diluted in TSB to3 106 and 3 108 CFU/mL, respectively, before use. C. albicanswas grown overnightin YMB at room temperature and diluted to 3 105 CFU/ml before use. The diluted cellsuspension (100 mL) was then inoculated into each well of 96-well plate and culturedfor 7e10 days depending on their growth rates. Due to differences in the rate of biolmformation, S. aureus and C. albicans were kept shaking at 100 rpm, 37 C and 50 rpm,25 C respectively, while E. Coli was incubated without shaking at 37 C. The culturemediumwas changed everyday with PBS being added to wash off the planktonic andloosely adhered cells before it was replacedwith freshmedium. Treatmentwas carriedout by rst removing the spent medium and the biolm was washed gently with PBSto remove the planktonic and loosely adhered cells. After that, the biolm was incu-bated with 50 mL of polycation-loaded hydrogel at MBCs for 24 h.

2.10. Biomass assay

The biomass left after treatment was analyzed using crystal violet (CV) stainingassay. The spent medium and hydrogel was gently removed and the biolm was

34 (2013) 10278e10286gently washed with PBS to remove the planktonic cells. Fixation was carried out byadding 100 mL of methanol to the biolm and removing it after 15 min. Crystal violet

A.L.Z. Lee et al. / Biomaterials 34 (2013) 10278e10286 10281staining (0.1 w/v %, 100 mL) was added to the xed biolm and incubated for 10 min.Excess crystal violet was washed off thoroughly using water of HPLC grade. Theremaining crystal violet bound to the biolm was extracted using 200 mL of 33%glacial acetic acid. An aliquot of 150 mL was then taken from each well and trans-ferred to a fresh 96-well plate. The absorbance was thenmeasured at 570 nm using amicroplate reader (Tecan, Switzerland) and the biomass of the remaining biolmwas expressed as a percentage of control group.

Scheme 1. Syntheses of (MTC-VE)n-PEG-(MTC-VE)n and vitamin E-containing polycationicsystem (inset).

A B

0

500

1000

1500

Blank Gel VE/BnCl(1:30)Gel

VE/PrBr(1:30)Gel

G' (P

a)

Fig. 1. Mechanical properties of hydrogels. (A) G0 values of blank and cationic polycarbonpolycation-loaded hydrogels were prepared by dissolving (MTC-VE)1.25-PEG(20k)-(MTC-VE)2.11. XTT reduction assay

XTT assay was used for quantifying viable cells in the biolms after hydrogeltreatment bymeasuring themitochondrial enzyme activity of the cells. It is based onthe reduction of 2,3-bis (2-methoxy-4-nitro-5-sulfophenyl)-5-[(phenylamino)carbonyl]-2H-tetrazolium hydroxide (XTT) in themetabolically activemicrobial cellsto a water soluble formazan. Briey, XTT solution (1 mg/mL) and menadione

polymers. Schematic illustration of incorporating polycationic polymers into hydrogel

0

200

400

600

800

1000

0 2 4 6 8 10Shear rate (s

-1)

Vis

co

sity

(P

a.s

)

VE/BnCl(1:30) GelVE/PrBr(1:30) Gel

ate-containing hydrogels. (B) Viscosity of hydrogels as a function of shear rate. The1.25 (4wt.%) and cationic polycarbonates (0.1 wt.%) in HPLC grade water.

cross-links between the polymer chains with the application ofshear stress. Thereby, this indicates that the hydrogels can be well-spread over the skin for topical treatment of dermal infections.

3.3. Antimicrobial activities of cationic polycarbonate-containinghydrogels

Antimicrobial activities of two different polycation-loadedhydrogels were evaluated against S. aureus, E. coli and C. albicansas representative models of Gram-positive and Gram-negativebacteria, and fungus respectively. These microbes are commonpathogens that often manifest on dermal wounds [35e37], and aretypically treated via topical delivery of antibiotics to infected areas[38,39].

99.87100.00 100.00 100.00

99.34

99.80100.00

98.8

99.0

99.2

99.4

99.6

99.8

100.0

78.1 156.2 312.5 625Conc. of Polymer (mg/L)

Killin

g Ef

ficie

ncy

(%)

A

100.00 100.00

99.63

100.00 100.00

99.88

99.6

99.7

99.8

99.9

100.0

156.2 312.5 625 1250 2500 5000Conc. of Polymer (mg/L)

Killin

g Ef

ficie

ncy

(%)

B

105.0 BnCl PrBrC

ials 34 (2013) 10278e10286solution (0.4mM) were individually prepared by dissolution in de-ionized (DI) water.Right before the assay, the two components were mixed together at a volume ratioof XTT:menadione 5:1. During the assay, the mediumwas rst removed and biolmwas carefully washed with PBS to remove planktonic cells. PBS (120 mL) and the XTT-menadionemixture (14.4 mL) were then added to eachwell and incubated for 3 h. Analiquot of 100 mL was then taken from each well and transferred to a fresh 96-wellplate. The absorbance was then measured at 490 nm using the microplate reader(Tecan, Switzerland) and cell viability of the remaining biolm was expressed as apercentage of the control group.

2.12. Field emission-scanning electron microscopy (FE-SEM)

After treatment with the hydrogels, the biolmwas gently washed with PBS andxed with 4% formaldehyde for 30 min. The xed biolmwas washed with DI waterto remove formaldehyde and a series of ethanol washes (35, 50, 75, 90, 95 and 100%)was carried for dehydration of the sample. After two days of air-drying, the sampleswere mounted onto carbon tape and coated with platinum for SEM analysis under aeld emission-scanning electron microscope (JEOL JSM-7400F, Japan).

2.13. Cytotoxicity test

Human dermal broblasts were seeded onto a 96 well plate at a density of20,000 cells per well and incubated overnight at 37 C. The medium was removedand 50 mL of colorless DMEM was added to each well, followed by 50 mL of thehydrogels containing different concentrations of cationic polycarbonates and/oruconazole. The plate was then incubated for 24 h at 37 C. CellTitre-blue (Promega,USA) and DMEM were then mixed at a ratio of 2:5 by volume. After 24 h of treat-ment, 100 mL of this mixture was then added to each well and the cells were left toincubate in the dark at 37 C for 4 h. Cells that were untreated were used as control.Subsequently, the absorbance at 570 nm was measured. The readings were thenexpressed as a percentage of cell viability of the control group.

3. Results and discussion

3.1. Polymer synthesis

The syntheses of (MTC-VE)n-PEG-(MTC-VE)n triblock co-polymers and vitamin E-containing cationic polycarbonates aresummarized in Scheme 1. The organocatalytic ring opening poly-merizations (ROP) of MTC-VE, initiated by HO-PEG-OH and benzylalcohol, respectively, were achieved using DBU/thiourea as cata-lysts [34]. The polycationic polymers were obtained subsequentlyby quaternization with trimethylamine. The nal compositions ofthe polymers were analyzed and conrmed by GPC and/or protonNMR. The triblock copolymer had an average number of 2.5 VEmolecules in each chain, i.e. (MTC-VE)1.25-PEG(20k)-(MTC-VE)1.25.The nal cationic polycarbonates, VE/BnCl(1:30) and VE/BnCl(1:30), contain one MTC-VE molecule each, 24 units of MTC-PrBr and 26 units of MTC-BnCl, respectively. There was a slightdecrease in the degree of polymerization (DP) for both polymerswhen compared to the pre-quaternized precursors (For VE/BnCl(1:30), DP 1:30 (before), 1:26 (after); For VE/PrBr(1:30),DP 1:27 (before), 1:24 (after)). While trimethylamine is basic, it isnot strongly nucleophilic. Under the mild reaction condition,degradation is minimal. From NMR analysis, all quaternization re-actions were complete (Fig. S1).

3.2. Mechanical properties of hydrogel

The triblock copolymer formed hydrogels at 4 wt% throughphysical cross-links between ower-like micellar networks(Scheme 1). The G0 value of the hydrogel was around 1500 Pa(Fig. 1A). The addition of the cationic polycarbonates VE/PrBr(1:30)and VE/BnCl(1:30) at 0.1 wt.% (equivalent to 1000 mg/L) to thehydrogel did not cause signicant difference in stiffness. In addi-tion, as shown in Fig. 1B, the polycation-loaded hydrogels displayedhigh viscosity at low shear rates, indicating a rm and well-bodiedstructure. As the shear rate increased, the viscosity of the gels felldrastically, and eventually became a thin liquid. This shear-thinning

A.L.Z. Lee et al. / Biomater10282behaviour of the hydrogels resulted from the disruption of physical99.94

82.22

100.0099.92

85.98

98.97

80.0

85.0

90.0

95.0

100.0

156.2 312.5 625.0Conc. of Polymer (mg/L)

Killin

g Ef

ficie

ncy

(%)

Fig. 2. Antimicrobial activity of polycation-loaded (MTC-VE)1.25-PEG(20k)-(MTC-

VE)1.25 (4wt.%) hydrogels against (A) S. aureus, (B) E. coli and (C) C. albicans. VE/BnCl(1:30) and VE/PrBr(1:30).

99.2299.99

99.5799.96

98.0

98.5

99.0

99.5

100.0

100.5

10 10 40 40

78 156 39 78

Fluc (mg/L)

VE/BnCl (mg/L)

Killin

g Ef

ficie

ncy

(%)

0

50

100

150

200

250

300

0 100 200 300 400 500 600Fluconazole (mg/L)

VE

/B

nC

l (m

g/L

)

\

BA

Fig. 3. Synergism between uconazole (Fluc) and VE/BnCl(1:30) in (MTC-VE)1.25-PEG(20k)-(MTC-VE)1.25 (4wt.%) hydrogel against C. albicans. (A) Killing efcacy of uconazole andVE/BnCl(1:30) combination. (B) An isobologram analysis representing the synergy between uconazole (10 mg/L) and VE/BnCl(1:30) (156 mg/L, MBC) or between uconazole(40 mg/L) and VE/BnCl(1:30) (78 mg/L, MBC). Synergy between the two compounds is demonstrated as the drug combination dose falls to the left of the line of additivity.

A

20

40

60

80

100

Ctrl Blank Gel VE/BnCl(1:30)

Gel

VE/PrBr(1:30)

Gel

Ctrl Blank Gel VE/BnCl(1:30)

Gel

VE/PrBr(1:30)

Gel

Ctrl Blank Gel VE/BnCl(1:30)

Gel

VE/PrBr(1:30)

Gel

Ctrl Blank Gel VE/BnCl(1:30)

Gel

VE/PrBr(1:30)

Gel

Cell

Via

bili

ty (%

)

C

20

40

60

80

100

Cell V

ia

bilit

y (

%)

Ctrl Blank Gel VE/BnCl(1:30)

Gel

VE/PrBr(1:30)

Gel

E

Ctrl Blank Gel VE/BnCl(1:30)

Gel

VE/PrBr(1:30)

Gel

0

20

40

60

80

100

Cell

Via

bili

ty

(%

)

B

20

40

60

80

100B

io

mas

s (%

)

D

20

40

60

80

100

Bio

mass

(%

)

F

20

40

60

80

100

Bio

ma

ss

(%

)

Fig. 4. Biolm eradication by polycation-loaded (MTC-VE)1.25-PEG(20k)-(MTC-VE)1.25 (4wt.%) hydrogels against (A and B) S. aureus, (C and D) E. coli and (E and F) C. albicans.Reduction in metabolic activities and biomass of the various biolms is shown in (A, C and E) and (B, D and F) respectively.

A.L.Z. Lee et al. / Biomaterials 34 (2013) 10278e10286 10283

Two vitamin E-containing cationic polycarbonates, VE/BnCl(1:30)and VE/PrBr(1:30), were loaded into 4 wt.% (MTC-VE)1.25-PEG(20k)-(MTC-VE)1.25 hydrogel in different concentrations. As the concen-tration of the polycations increased, a greater density of cationicchargewas expected to be displayed on the hydrogel surface and thiswould give rise to better inhibition effects. The hydrogels werechallenged with an inoculum of 3 105 CFU/ml and proliferationcapacity of the survived cells was assessed 24 h later via the spreadplate technique. This method of examining the antimicrobial activityis akin to measuring the minimum inhibitory concentration (MIC) ofthe polycations in solution [40]. Our study showed that the poly-cations were broadly antimicrobial towards the tested bacteria andfungi. As seen from Fig. 2(AeC), the inhibition efcacy of VE/BnCl(1:30)- and VE/PrBr(1:30)-loaded hydrogels on the proliferationof the differentmicrobe species varied. Amongst the testedmicrobes,the hydrogels are most effective in preventing Gram-positiveS. aureus proliferation as it required the lowest MBCs for bothpolycation-loaded gels (156.2 and 625 mg/L for VE/BnCl(1:30) andVE/PrBr(1:30), respectively). On the other hand, the hydrogelsinhibited the proliferation of E. coli (Gram-negative) less effectively(MBCs: 312.5 and 2500 mg/L for VE/BnCl(1:30)and VE/PrBr(1:30)respectively), and this was likely due to the presence of additionallipopolysaccharide-containing outer membrane that could impede

relevant antimicrobial agents. The benets of combination ther-apy have been long-established with appealing attributes such asthe evasion of drug resistance and minimized drug concentrationsfor desired therapeutic actions [32,42]. However, it should be notedthat augmentation in drug interaction is both system- and target-dependent and it is vital to utilize appropriate therapeutic part-ners for favorable effects. In this study, antifungal drug uconazolewas added together with polycation during the preparation of thehydrogels for combination treatment against C. albicans. Thepresence of cationic polycarbonate did not affect the drug releaseprole signicantly, and most drug molecules were released fromthe hydrogels at 2 h (Fig. S2). Fluconazole is a member of the azolefamily of antifungal agents that possesses good activity againstC. albicans and exhibits low toxicity. While azole-based drugs areconsidered safe for clinical applications, the main disadvantage isthat they are only fungistatic and resistance can develop easily withextended usage [43]. Our results showed that even at a high con-centration of 500 mg/L, uconazole only exerted fungistatic prop-erties with less than 3.0log reduction in total microbial count(Fig. S3). By combining this fungistatic drug and polycation in ahydrogel matrix, signicant improvement in the therapeutic ef-cacy was observed at two combination doses (10 mg/L uconazoleand MBC of VE/BnCl(1:30) (156 mg/L); 40 mg/L uconazole and

A.L.Z. Lee et al. / Biomaterials 34 (2013) 10278e1028610284the penetration of the polycations. A similar phenomenon was alsoobserved in other antimicrobial systems [23,40]. The large differenceinMBC values of the two polycations against E. coli indicates that theGram-negative bacteria possess greater sensitivity towards theamphiphilic balance [41] of these vitamin E-functionalized poly-cations compared to the Gram-positive counterpart. Antimicrobialeffect on C. albicans was intermediary to that of the two bacteriaspecies with MBC values of 312.5 and 625 mg/L for VE-BnCl(1:30)and VE/PrBr(1:30), respectively. This further illustrates that themore hydrophobic polycation VE/BnCl(1:30) generally exhibitsgreater antimicrobial/antifungal effects compared to VE/PrBr(1:30).

3.4. Synergism analysis of co-delivery of uconazole and cationicpolycarbonate using hydrogels

Another strategy to enhance the antimicrobial efcacy of thepolycation-loaded hydrogels is the co-delivery with clinically-Fig. 5. SEM images of biolms treated with polycation-loaded MBC of VE/BnCl(1:30) (78 mg/L)), and the hydrogels werefungicidal (> 99.9% eradication) (Fig. 3A). In addition, FBC indexof the combination doses were w0.5 and w0.25 respectively,indicating synergistic interaction between uconazole andVE/BnCl(1:30). Furthermore, the isobologram method of analyzingdrug interactions further illustrated the strong synergism betweenuconazole and VE-BnCl(1:30) (Fig. 3B).

3.5. Biolm eradication

To explore efcacy of the antimicrobial hydrogels in biolmelimination, various microbes (S. aureus, E. coli and C. albicans) werecultured for several days to develop biolms prior to the treatment.Polycations were loaded at MBC concentrations into (MTC-VE)1.25-PEG(20k)-(MTC-VE)1.25 (4wt.%) hydrogels and placed onto thebiolms. After 24 h, the biomass and cell viability of these biolmswere analyzed. From Fig. 4A, C and E, hydrogels loaded with(MTC-VE)1.25-PEG(20k)-(MTC-VE)1.25 (4wt.%) hydrogels.

VE/BnCl(1:30) were as efcient as those loaded with VE/PrBr(1:30)in reducing the proliferation and viability of S. aureus (Gram-posi-tive) and C. albicans (fungus). The major difference was observed inE. coli (Gram-negative) where cells treated with hydrogels loadedwith the more hydrophobic VE/BnCl(1:30) had signicantly lowerviability compared to those treated with VE/PrBr(1:30)-loadedhydrogels. In addition, the ability of the polycation-loaded hydro-gels to eradicate the biomass displayed a similar trend to reductionof cell viability residing in the biolms where VE/BnCl(1:30) hadstronger activity in removing E. coli biomass and similar efciencyin the treatment of S. aureus and C. albicans biolms compared toVE/PrBr(1:30) (Fig. 4B, D and F). Furthermore, SEM imagesdemonstrated that biolms treated with VE/BnCl(1:30)-loadedhydrogels showed extensive cell destruction and clearance

A.L.Z. Lee et al. / BiomaterialsFig. 6. Viability of human dermal broblasts after 24 h treatment with (MTC-VE)1.25-

PEG(20k)-(MTC-VE)1.25 (4wt.%) hydrogels loaded with (A) VE/BnCl(1:30), (B) VE/PrBr(1:30) and (C) VE/BnCl(1:30) and uconazole at different concentrations.(Fig. 5). Only ruptured cell fragments remained for S. aureus andC. albicans. The images also showed good correlation with thequantication assays of cell viability and biomass where VE/BnCl(1:30) hydrogel was signicantly more effective in eradicatingE. coli biolm compared to VE/PrBr(1:30) hydrogel.

3.6. Cytotoxicity studies

The effects of the hydrogels on mammalian cells were evaluatedby treating human dermal broblasts with hydrogels containingMBC concentrations of the cationic polycarbonates and/or ucon-azole. As shown in Fig. 6AeC, the blank hydrogels resulted in slightretardation of cell proliferation as the cell viability was w88%compared to the untreated group. This is possibly due to theslowing down of nutrient andmetabolite diffusion as a result of thehydrogels present on the cell surface. Importantly, no signicantcytotoxicity was observed from the treatment as the cell viabilityremained > 80% for all treatment groups.

4. Conclusion

Biodegradable antimicrobial hydrogels have been successfullyprepared by incorporating vitamin E-functionalized cationic poly-carbonates into physically cross-linked hydrogels fabricated fromABA-type triblock copolymer (MTC-VE)1.25-PEG(20k)-(MTC-VE)1.25. These hydrogels exhibit thixotropic property, which is idealfor topical applications. They are effective against both Gram-positive and Gram-negative bacteria as well as fungus with morethan 99.9% killing efciency upon contact. In addition, thesephysically cross-linked hydrogels are able to eradicate the biomassand greatly reduce viability of microbes residing as S. aureus, E. coliand C. albicans biolms. The co-delivery of antimicrobial poly-carbonates with the conventionally-used antifungal drug ucona-zole using the hydrogel provides a high degree of synergisticantifungal effect on C. albicans. Importantly, the hydrogels con-taining the cationic polycarbonates and/or uconazole at theirminimum biocidal concentrations do not induce signicant cyto-toxicity towards human dermal broblasts. Therefore, these anti-microbial polycarbonate-loaded hydrogels may be used toeliminate both planktonic microbes and their biolms, and repre-sent a promising approach for the prevention and treatment of skininfections.

Acknowledgments

This work was funded by the Institute of Bioengineering andNanotechnology (Biomedical Research Council, Agency for Science,Technology and Research, Singapore) and IBM Almaden ResearchCenter, USA.

Appendix A. Supplementary data

Supplementary data related to this article can be found online athttp://dx.doi.org/10.1016/j.biomaterials.2013.09.029.

References

[1] Malmsten M. Antimicrobial and antiviral hydrogels. Soft Matter 2011;7:8725e36.

[2] Hiby N, Bjarnsholt T, Givskov M, Molin S, Ciofu O. Antibiotic resistance ofbacterial biolms. Int J Antimicrob Agents 2010;35:322e32.

[3] Tenke P, Riedl CR, Jones GL, Williams GJ, Stickler D, Nagy E. Bacterial biolmformation on urologic devices and heparin coating as preventive strategy. Int JAntimicrob Agents 2004;23:67e74.

[4] Bryers JD. Medical biolms. Biotechnol Bioeng 2008;100:1e18.

34 (2013) 10278e10286 10285[5] Hall-Stoodley L, Stoodley P. Evolving concepts in biolm infections. CellMicrobiol 2009;11:1034e43.

[6] Bendouah Z, Barbeau J, Hamad WA, Desrosiers M. Biolm formation byStaphylococcus aureus and Pseudomonas aeruginosa is associated with an un-favorable evolution after surgery for chronic sinusitis and nasal polyposis.Otolaryngol Head Neck Surg 2006;134:991e6.

[7] Fihn SD, Johnson C, Stamm WE. Escherichia coli urethritis in women withsymptoms of acute urinary tract infection. J Infect Dis 1988;157:196e9.

[8] Johnson JR. Virulence factors in Escherichia coli urinary tract infection. ClinMicrobiol Rev 1991;4:80e128.

[9] Pfaller MA, Diekema DJ. Epidemiology of invasive candidiasis: a persistentpublic health problem. Clin Microbiol Rev 2007;20:133e63.

[10] Mah TFC, OToole GA. Mechanisms of biolm resistance to antimicrobialagents. Trends Microbiol 2001;9:34e9.

[11] Davies D. Understanding biolm resistance to antibacterial agents. Nat RevDrug Discov 2003;2:114e22.

[12] Xu KD, McFeters GA, Stewart PS. Biolm resistance to antimicrobial agents.Microbiology 2000;146:547e9.

[13] Donlan RM. Biolm formation: a clinically relevant microbiological process.Clin Infect Dis 2001;33:1387e92.

[14] Lewis K. Riddle of biolm resistance. Antimicrob Agents Chemother 2001;45:999e1007.

[15] Fux CA, Wilson S, Stoodley P. Detachment characteristics and oxacillin resis-tance of Staphyloccocus aureus biolm emboli in an in vitro catheter infectionmodel. J Bacteriol 2004;186:4486e91.

[16] Ceri H, Olson ME, Stremick C, Read RR, Morck D, Buret A. The calgary biolmdevice: new technology for rapid determination of antibiotic susceptibilitiesof bacterial biolms. J Clin Microbiol 1999;37:1771e6.

[17] Kuroda K, Caputo GA. Antimicrobial polymers as synthetic mimics of host-defense peptides. Wiley Interdiscip Rev Nanomed Nanobiotechnol 2013;5:49e66.

[18] Tew GN, Liu D, Chen B, Doerksen RJ, Kaplan J, Carroll PJ, et al. De novo designof biomimetic antimicrobial polymers. Proc Natl Acad Sci U S A 2002;99:5110e4.

[19] Engler AC, Wiradharma N, Ong ZY, Coady DJ, Hedrick JL, Yang YY. Emergingtrends in macromolecular antimicrobials to ght multi-drug-resistant in-

[25] Li Y, Fukushima K, Coady DJ, Engler AC, Liu S, Huang Y, et al. Broad-spectrumantimicrobial and biolm-disrupting hydrogels: stereocomplex-driven su-pramolecular assemblies. Angew Chem Int Ed 2013;52:674e8.

[26] Kuroda K, DeGrado WF. Amphiphilic polymethacrylate derivatives as anti-microbial agents. J Am Chem Soc 2005;127:4128e9.

[27] Azzi A, Stocker A. Vitamin E: non-antioxidant roles. Prog Lipid Res 2000;39:231e55.

[28] Zampieri N, Zuin V, Burro R, Ottolenghi A, Camoglio FS. A prospective study inchildren: pre-and post-surgery use of vitamin E in surgical incisions. J PlastReconstr Aesthet Surg 2010;63:1474e8.

[29] Pratt RC, Nederberg F, Waymouth RM, Hedrick JL. Tagging alcohols with cycliccarbonate: a versatile equivalent of (meth)acrylate for ring-opening poly-merization. Chem Commun 2008:114e6.

[30] Lorian V. Antibiotics in laboratory medicine. 5th ed. Philadelphia: LippincottWilliams & Wilkins; 2005.

[31] Mackay ML, Milne K, Gould IM. Comparison of methods for assessing synergicantibiotic interactions. Int J Antimicrob Agents 2000;15:125e9.

[32] White RL, Burgess DS, Manduru M, Bosso JA. Comparison of three differentin vitro methods of detecting synergy: time-kill, checkerboard, and E test.Antimicrob Agents Chemother 1996;40:1914e8.

[33] Keith CT, Borisy AA, Stockwell BR. Multicomponent therapeutics for net-worked systems. Nat Rev Drug Discov 2005;4:71e8.

[34] Kiesewetter MK, Shin EJ, Hedrick JL, Waymouth RM. Organocatalysis: opportu-nitiesandchallenges forpolymersynthesis.Macromolecules2010;43:2093e107.

[35] Howell-Jones RS, Wilson MJ, Hill KE, Howard AJ, Price PE, Thomas DW.A review of the microbiology, antibiotic usage and resistance in chronic skinwounds. J Antimicrob Chemother 2005;55:143e9.

[36] Archer GL. Staphylococcus aureus: a well-armed pathogen. Clin Infect Dis1998;26:1179e81.

[37] Lilic D, Gravenor I, Robson N, Lammas DA, Drysdale P, Calvert JE, et al.Deregulated production of protective cytokines in response to Candida albi-cans infection in patients with chronic mucocutaneous candidiasis. InfectImmun 2003;71:5690e9.

[38] Castellano JJ, Shai SM, Ko F, Donate G, Wright TE, Mannari RJ, et al.

A.L.Z. Lee et al. / Biomaterials 34 (2013) 10278e1028610286fections. Nano Today 2012;7:201e22.[20] Nederberg F, Zhang Y, Tan JPK, Xu K, Wang H, Yang C, et al. Biodegradable

nanostructures with selective lysis of microbial membranes. Nat Chem2011;3:409e14.

[21] Muoz-Bonilla A, Fernndez-Garca M. Polymeric materials with antimicro-bial activity. Prog Polym Sci 2012;37:281e339.

[22] Kenawy ER, Worley SD, Broughton R. The chemistry and applications ofantimicrobial polymers: a state-of-the-art review. Biomacromolecules2007;8:1359e84.

[23] Li P, Poon YF, Li W, Zhu HY, Yeap SH, Cao Y, et al. A polycationic antimicrobialand biocompatible hydrogel with microbe membrane suctioning ability. NatMater 2011;10:149e56.

[24] Zhou C, Li P, Qi X, Sharif ARM, Poon YF, Cao Y, et al. A photopolymerizedantimicrobial hydrogel coating derived from epsilon-poly-L-lysine. Bio-materials 2011;32:2704e12.Comparative evaluation of silver-containing antimicrobial dressings anddrugs. Int Wound J 2007;4:114e22.

[39] Zumbuehl A, Ferreira L, Kuhn D, Astashkina A, Long L, Yeo Y, et al. Antifungalhydrogels. Proc Natl Acad Sci U S A 2007;104:12994e8.

[40] Veiga AS, Sinthuvanich C, Gaspar D, Franquelim HG, Castanho MARB,Schneider JP. Arginine-rich self-assembling peptides as potent antibacterialgels. Biomaterials 2012;33:8907e16.

[41] Palermo EF, Kuroda K. Structural determinants of antimicrobial activity inpolymers which mimic host defense peptides. Appl Microbiol Biotechnol2010;87:1605e15.

[42] Chait R, Craney A, Kishony R. Antibiotic interactions that select againstresistance. Nature 2007;446:668e71.

[43] Klepser ME, Wolfe EJ, Jones RN, Nightingale CH, Pfaller MA. Antifungalpharmacodynamic characteristics of uconazole and amphotericin B testedagainst Candida albicans. Antimicrob Agents Chemother 1997;41:1392e5.

Block copolymer mixtures as antimicrobial hydrogels for biofilm eradication1 Introduction2 Materials and methods2.1 Materials2.2 Nuclear magnetic resonance (NMR) spectroscopy2.3 Molecular weight determination by size exclusion chromatography (SEC)2.4 General procedure for polymer synthesis2.5 Rheological experiments2.6 In vitro fluconazole release2.7 Killing efficiency tests2.8 Analysis of synergism between antimicrobial polycarbonate and fluconazole2.9 Biofilm formation and treatment2.10 Biomass assay2.11 XTT reduction assay2.12 Field emission-scanning electron microscopy (FE-SEM)2.13 Cytotoxicity test

3 Results and discussion3.1 Polymer synthesis3.2 Mechanical properties of hydrogel3.3 Antimicrobial activities of cationic polycarbonate-containing hydrogels3.4 Synergism analysis of co-delivery of fluconazole and cationic polycarbonate using hydrogels3.5 Biofilm eradication3.6 Cytotoxicity studies

4 ConclusionAcknowledgmentsAppendix A Supplementary dataReferences