Article pubs.acs.org/Biomac
Nanoparticle (Star Polymer) Delivery of Nitric Oxide Effectively Negates Pseudomonas aeruginosa Biofilm Formation Hien T. T. Duong,†,▽ Kenward Jung,‡,▽ Samuel K. Kutty,§ Sri Agustina,† Nik Nik M. Adnan,‡ Johan S. Basuki,† Naresh Kumar,§ Thomas P. Davis,*,⊥,# Nicolas Barraud,*,∥ and Cyrille Boyer*,†,‡ †
Australian Centre for Nanomedicine and ‡Centre for Advanced Macromolecular Design (CAMD), School of Chemical Engineering, School of Chemistry, and ∥Centre for Marine Bio-Innovation, School of Biotechnology and Biomolecular Sciences, University of New South Wales, Sydney, New South Wales, Australia 2052 ⊥ ARC Centre of Excellence in Convergent Bio-Nano Science & Technology, Monash Institute of Pharmaceutical Sciences, Monash University, Parkville, Melbourne Victoria, Australia, 3052 # Department of Chemistry, University of Warwick, Coventry CV4 7AL, United Kingdom §
S Supporting Information *
ABSTRACT: Biofilms are increasingly recognized as playing a major role in human infectious diseases, as they can form on both living tissues and abiotic surfaces, with serious implications for applications that rely on prolonged exposure to the body such as implantable biomedical devices or catheters. Therefore, there is an urgent need to develop improved therapeutics to effectively eradicate unwanted biofilms. Recently, the biological signaling molecule nitric oxide (NO) was identified as a key regulator of dispersal events in biofilms. In this paper, we report a new class of core cross-linked star polymers designed to store and release nitric oxide, in a controlled way, for the dispersion of biofilms. First, core cross-linked star polymers were prepared by reversible addition−fragmentation chain transfer polymerization (RAFT) via an arm first approach. Poly(oligoethylene methoxy acrylate) chains were synthesized by RAFT polymerization, and then chain extended in the presence of 2-vinyl-4,4-dimethyl-5-oxazolone monomer (VDM) with N,N-methylenebis(acrylamide) employed as a cross-linker to yield functional core cross-linked star polymers. Spermine was successfully attached to the star core by reaction with VDM. Finally, the secondary amine groups were reacted with NO gas to yield NO-core cross-linked star polymers. The core cross-linked star polymers were found to release NO in a controlled, slow delivery in bacterial cultures showing great efficacy in preventing both cell attachment and biofilm formation in Pseudomonas aeruginosa over time via a nontoxic mechanism, confining bacterial growth to the suspended liquid.
■
INTRODUCTION
outcomes. Biofilms can also be highly detrimental in industrial settings such as fouled immersed marine surfaces, clogged filtration membranes, or corroded pipes which act as reservoirs for pathogens in food and water processing.4,5 Therefore, novel and efficient measures specifically designed to inhibit or prevent biofilm formation are urgently needed across a wide range of applications. One key strategy to negate biofilm formation is to target the biofilm developmental processes. Biofilm dispersal events are of particular interest for their potential to revert biofilm formation and promote cell detachment.6 Recently, the biological signaling molecule nitric oxide (NO), a diatomic free radical, was identified as a key regulator of biofilm dispersal. NO was found to be produced endogenously in the late developmental stages of mature
Despite intense efforts to develop antimicrobial agents, infectious diseases still have a major detrimental impact on human health and the global economy. At present, infectious diseases rank as the second leading cause of premature death worldwide, accounting for about 15 million deaths each year.1 The failure of antibiotic and antimicrobial agents to affect cures is often attributed to the development of drug resistance by pathogens and the adverse side effects associated with harsh treatment regimes.2 One of the adaptive behaviors adopted by bacteria is to form biofilms, which are highly structured, usually surface-attached, multicellular communities of cells enclosed in a self-produced extracellular polymeric matrix. Bacteria embedded within biofilms exhibit upward of 10−1000-fold higher resistance to biocides and traditional antimicrobials than their planktonic counterparts, and are less susceptible to host immune defenses.3 The inability to fully eradicate biofilms often forms the basis for chronic infections that can then lead to fatal © 2014 American Chemical Society
Received: March 19, 2014 Revised: May 27, 2014 Published: June 10, 2014 2583
dx.doi.org/10.1021/bm500422v | Biomacromolecules 2014, 15, 2583−2589
Biomacromolecules
Article
Scheme 1. Synthesis of P(OEGA)-b-P(VDM) Core Cross-Linked Star Polymers, Followed by Spermine and NO Donor Conjugation
biofilms to induce dispersal events.6,7 Molecular analyses revealed that NO triggers a signaling pathway involving the conserved intracellular second messenger cyclic di-GMP, which in turn activates a range of effectors leading to dispersal.8−10 Further, the addition of low, nontoxic doses of NO (in the picomolar to nanomolar range) to established biofilms, using NO donors, has been found to induce dispersal in a broad range of microbial species, thereby restoring sensitivity to a range of antimicrobials and antibiotics.11−14 Thus, the use of NO represents a promising strategy for the control of biofilms in medical and industrial applications. However, the direct delivery of gaseous NO is fraught with problems as NO has high reactivity and a consequential short half-life in biological systems, making it challenging to administer at an appropriate dose rate and for a sustained time period. To overcome the short half-life of NO, a range of smallmolecules capable of decomposition, releasing NO, under specific conditions have been developed; these include nitrates, nitroprussides, S-nitrosothiols (RSNOs), and N-diazeniumdiolates (NONOates). NO donors have the potential to be used as biofilm dispersal agents especially when treating existing biofilms, where a short burst of NO may be able to disperse the majority of the biofilm.15 However, common NO donors lack both stability and specificity and are unlikely to be suitable as preventative agents when a slow and sustained release of NO is needed to negate the bacterial attachment processes confining the bacteria to a planktonic mode of growth. One promising avenue to overcome the challenge of controlled NO delivery is the development of nanoparticles for the storage and subsequent release of NO. Nanoparticle based delivery has many advantages over the direct administration of small molecule antibiotics by improving pharmacokinetics and accumulation, reducing side effects, and more importantly overcoming drug resistance.16,17 In recent work, NONOate has been successfully conjugated to a range of polymeric systems, including a matrix of ethylene/vinyl acetate,18 star polymers,19
or micelles.19−23 Nanoparticle delivery of NO has the potential to address many of the shortcomings of small molecule NO donors, by encapsulating NO donors in a hydrophobic microenvironment (e.g., in the core of micelles), protecting the NO donors from multiple release triggers (e.g., light, heat, copper, enzymes), enhancing NO donor stability, and thus controlling the release of NO for extended (and preprogrammed) times. Another advantage to using nanoparticle carriers is the potential to accommodate multiple NO donors in each nanoparticle, resulting in a significant increase in the local concentration and, as a consequence, enhanced release and response. Nanoparticle delivery of NO can also be combined with a concomitant delivery with other therapeutic agents to potentially exploit positive synergistic effects.24−27 Herein, we report a novel NONOate conjugated star polymer that is able to maintain a slow release of NO in the presence of the model biofilm-forming organism Pseudomonas aeruginosa, completely inhibiting bacterial attachment and biofilm formation over time in a non-growth-inhibitory fashion. This paper describes for the first time the use of core cross-linked star polymer (star polymer) for the encapsulation of NONOates for the prevention of biofilms. In contrast to small organic NO donors, star polymer provides a slow release of NO for a long period (several days), which could be employed for the prevention and the dispersion of biofilm. We show that the NO star polymer confines growth of the bacterial population to a suspended liquid, while keeping surfaces completely free of biofilm.
■
RESULTS AND DISCUSSION Core Cross-Linked Star Polymer (Star polymer) Formation. Core cross-linked star polymers (star polymer) were synthesized using an “arm-first” approach,28,29 as developed by our group and previously described for applications in magnetic resonance imaging,30 drug delivery,31 and siRNA delivery.32 Reversible addition−fragmentation transfer (RAFT) polymerization was employed to synthesize 2584
dx.doi.org/10.1021/bm500422v | Biomacromolecules 2014, 15, 2583−2589
Biomacromolecules
Article
Figure 1. (A) ATR-FTIR spectra of P(OEGA) arm, P(OEGA)-b-P(VDM) star, P(OEGA)-b-P(Sper) star, P(OEGA)-b-P(Sper/NO) star, and P(OEGA)-b-P(Sper/NO) after NO release. (B) UV−vis absorption of (red line) P(OEGA)-b-P(VDM) star, (green line) P(OEGA)-b-P(Sper) star, (cyan line) P(OEGA)-b-P(Sper/NO) star, and (blue line) after reaction with Griess assay. Notes: Nitric oxide (NO) of P(OEGA)-b-P(Sper/NO) star polymer (cyan line) in water is confirmed by the presence of absorption band at 252 nm. The presence of an absorption signal at 548 nm (after treatment with Griess reagent) confirmed the presence of NO in solution.
Table 1. Summary of the Molecular Weights, PDI, Size (DLS), and Zeta Potential for the Polymers Prepared in This Study samples POEGMA P(OEGMA)-b-P(VDM) star P(OEGMA)-b-P(Sper) star P(OEGMA)-b-P(Sper) starg P(OEGMA)-b-P(Sper/NO) star P(OEGMA)-b-P(Sper/NO) starh
Mn,theo (g/mol)a 11 600
Mn,NMR (g/mol)b 11 500
Mn,SEC (g/mol)c
PDISECc
14 500 210 000 235 000
1.11 1.40 1.48
240 000 230 000
1.46 1.46
sizeDLS (nm)d 4 21 25 34 32 27
PDIDLSd 0.12 0.18 0.38 0.31 0.21 0.19
sizeTEMe i
nd 18 23 ndi 28 19
ζ (mV)f nd −2 (± 2) +24 (±2) −2 (±3) +3 (±3) +18 (±2)
a
Theoretical molecular weight. bMolecular weight calculated by NMR. cMolecular weight and PDI determined by SEC. dSize (z-number) and PDI of nanoparticles by DLS. eAverage size determined by TEM. fAverage zeta-potential. gThe measurement was performed in basic water (pH = 9.0). h After NO release. iNot determined.
P(OEGA) arm using chain transfer agent 1 (RAFT 1, nbutyltrithiocarbonate isopropionate) and AIBN (2,2′-azobisisobutylonitrile) as radical initiator with the ratio of [OEGA]: [RAFT 1]:[AIBN] = 25:1:0.1 in toluene at 70 °C (Scheme 1). After 4 h of reaction, ∼90% OEGA conversion was achieved and the resultant polymer arms were purified by precipitation in petroleum spirit. After purification, P(OEGA) was then characterized using size exclusion chromatography (SEC) and 1 H NMR. The molecular weight of the POEGMA based on SEC was measured to be 14 500 g mol−1 (PDI = 1.11) by DMAc SEC (Supporting Information (SI), Figure S1), which was slightly higher than the molecular weight given by 1H NMR spectroscopy (Mn,NMR of 11 500 g/mol, calculated using the following equation: Mn,NMR = [(I4.1ppm/2)/(I0.8ppm/3)] × MWOEGA + MWRAFT1). This slight MWT difference can be attributed to the differences in hydrodynamic volume between the P(OEGA) and the polystyrene standards. This assumption was confirmed by the comparison between theoretical and experimental molecular weights. The theoretical molecular weight was also calculated using Mn,theo = [OEGA]0/[RAFT1]0 × MWOEGA + MWRAFT1, with [OEGA]0 and [RAFT1]0 corresponding to OEGA and RAFT1 concentrations. The theoretical value (Mn,theo = 11 600 g/mol) is in good agreement with molecular weight obtained by NMR. The star polymer was then assembled. Briefly, P(OEGA) arms were reacted in the presence of 2-vinyl-4,4-dimethyl-5oxazolone monomer (VDM) and N,N-methylenebis(acrylamide) employed as cross-linker. The molar ratio of P(OEGA) arm, VDM, and the cross-linker was set to 1:16:8, as it resulted in the highest incorporation of P(OEGA) arm into
the core cross-linked star polymers. At the end of the polymerization, the conversion of VDM and N,N-methylenebis(acrylamide) was determined by NMR to be 85% and 99%, respectively (SI, Figure S2). The resulting star polymer was then purified by precipitation in diethyl ether, removing any unreacted P(OEGA) arms, cross-linker, or VDM monomer. SEC data revealed a significant shift from the individual arm molecular weight with the synthesis of core cross-linked star polymer, with Mn,SEC of 177 000 g/mol and a polydispersity index of 1.39. SEC triple detection was also invoked to determine the absolute molecular weight (MW = 210 000 g/ mol, PDI 1.40). The number of arms was calculated using the following equation: Narm = MW/MnPOEGA. Each star polymer was constituted by 14 arms. This value is consistent with our previous reports determined for similar star polymers, where pentafluorophenyl ester acrylate were employed instead of VDM.33 In addition, we estimated an arm incorporation of ∼71% using the deconvulation of SEC traces. The successful incorporation of VDM monomer into the P(OEGA) arms was confirmed by 1H NMR spectra via a signal at 1.3 ppm originating from the VDM and the ester group from OEGA at 4.1 ppm (SI, Figure S3B). In addition, ATR-FTIR analysis was employed, confirming the presence of the azlactone group at 1820 cm−1 (Figure 1A). The absorptions at 1630, 1720, and 1100 cm−1 were indicative of the cross-linking amide, OEGA ester, and ether groups, respectively, again confirming the incorporation of arms into star polymer structures.34 Spermine was reacted with VDM star polymer, followed by dialysis purification; the formation of star conjugated with spermine was confirmed by the disappearance of the characteristic 2585
dx.doi.org/10.1021/bm500422v | Biomacromolecules 2014, 15, 2583−2589
Biomacromolecules
Article
azlactone absorption at 1820 cm−1 in the FTIR spectrum, confirming azlactone ring-opening via amidation.35 A broad absorption at around 3500 cm−1 was observed consistent with the secondary amines of the conjugated spermine in the FTIR spectrum. NMR analysis revealed the presence of new signals at 3.3−3.4 and 1.5−1.8 ppm attributed to the presence of spermine groups. In addition, X-ray photoelectron spectroscopy (XPS) confirmed the attachment of spermine by the presence of new signal at 401 and 400 eV. Elemental analysis was invoked to quantify the yield of the reaction between spermine and azlactone using the amount of nitrogen present in the star before and after reaction (SI, Table 1). We estimated that ∼42% of VDM was successfully modified with spermine. This value is in accord with NMR results, where we found that ∼40% of VDM was reacted. The formation of the core cross-linked star polymer nanoparticle was confirmed by dynamic light scattering (DLS) showing a number-weighted particle size of ∼25 nm (SI, Figure S4), in accord with TEM results (SI, Figure S5). Zeta potential measurements on the initial star polymer in water revealed a neutral surface charge consistent with the P(OEGA) hydrophilic layers (SI, Figure S6). 36 After conjugation of spermine, the zeta potential of the star polymers shifted from −2 (±1) to +24 (±2) mV, indicative of the protonation of primary and secondary amine (NH3+ or NH2+) in water. This assumption was confirmed by the measure of zeta potential in basic water (pH 9.0). We observed that the zeta potential of the star polymer in basic media is close to −2 (±3) mV, which demonstrates that the positive charge was due to the protonation of the secondary and tertiary amine (SI, Figure S6). UV−vis analysis was also employed to monitor the disappearance of the absorption at 310 nm (Figure 1B) after reaction with spermine, which suggests the aminolysis of the trithiocarbonate RAFT group.37 Nitric Oxide (NO) Conjugation to Spermine Conjugated Star Polymers (NO star). Spermime conjugated star polymers were then dissolved in acetonitrile in a Parr hydrogenation apparatus, purged with nitrogen, and then stirred with NO gas for 48 h at 80 psi (5 bar) (Scheme 1). NO reacted with one of the secondary amine of spermine to yield N-diazeniumdiolate (or NONOate) groups,38 which were stabilized inside the core of the polymeric nanoparticles. Each spermine molecule can react with 2 NO molecules.38 The NO star polymer was purified by precipitation in diethyl ether and dried under vacuum at room temperature for 3 h to yield a yellow/orange powder. The NO star polymers can be easily redispersed in water. DLS showed a slight increase of size from 25 to 32 nm after NO conjugation (SI, Figure S4), without a significant change in PDI value in accord with TEM results. In addition, we measured the zeta potential of the resultant NO star polymers (+3, ±3mV), and this indicated a significant reduction in surface charge due to the formation of the Ndiazeniumdiolate (zwitterionic compound) (SI, Figure S6). The successful attachment of NO was confirmed by FTIR, UV−vis, and XPS analyses. First, FTIR showed the absence of amine bond (NH bond) at 3300 cm−1 and the presence of new signal at 1370 cm−1, which confirms the conjugation of NO onto secondary amine (Figure 1A and SI, Figure S7). Second, NO star polymer was analyzed by XPS and elemental analysis (SI, Figures S8 and S9). N(1s) analysis showed the presence of new signal at 401.3 eV attributed to NONOate groups. In addition, we observed an increase of nitrogen amount in the sample in accord with the mechanism proposed (SI, Table S1). Using the
results from elemental analysis, we were able to calculate the yield of conjugation of NO onto spermine to be greater than 95%. The theoretical amount of NO loaded in the star can be calculated using NNO = 2Nspermine, with Nspermine being the number of spermine reacted in the star polymer, assuming that one secondary amine of spermine can react with two NO molecules. Finally, UV−vis spectroscopy was employed to determine the amount of NO conjugated to the star polymers using the characteristic absorbance of NO in water at 252 nm (ε = 8500 M−1 cm−1) (Figure 1B).39,40 Both results obtained by XPS and UV−vis spectroscopy were in good agreement. When the NO star polymer was dispersed in water, NO gas was slowly released, with an accelerated release observed at pH values lower than 7.5.41−43 Qualitative and quantitative analysis of the released NO from star polymers was performed using a Griess assay.44 Briefly, NO star polymers were incubated at different time points in phosphate buffer water at pH 7.0 (Figure 2). The star polymers were separated from the solution
Figure 2. Cumulative release of NO from NO star polymer at pH 7.0 (concentration of star polymer 1 mg/mL, experiments were performed in duplicate) at 37 °C.
using a centrifuge filter with a molecular weight cutoff of 3500 Da. Then, the filtrate solution was incubated with nitrate reductase and its cofactor to detect NO-derived nitrate and nitrite for 3 h, which allows a maximum NO release and subsequent conversion to nitrite in the aqueous medium.45 The addition of a Griess reagent to the nitrite sample formed a diazenium salt that was converted instantaneously to an azo dye (Figure 1B), characterized by the presence of a pink color and a UV−vis absorption at 548 nm. Based on a calibration curve, the NO released profile was determined from the NO star polymers at pH 7.0. The release rate of NO was estimated to be around 355 nM/h for 400 μg/mL of polymer (or 886 nM/ h/mg/mL of NO star). After a rapid release of NO in the first hour, we observed that NO star polymers released continuously NO for 70 h. This is a significant result, as small organocompounds, such as spermine and NONOate, have a short halflife (
