ARTICLE pubs.acs.org/Biomac
Chemical and Physical Factors in Design of Antibiofouling Polymer Coatings Inbal Eshet,†,‡ Viatcheslav Freger,*,†,‡ Roni Kasher,*,† Moshe Herzberg,† Jing Lei,§ and Mathias Ulbricht§ †
Zuckerberg Institute for Water Research, Ben-Gurion University of the Negev, Sede Boqer Campus 84990, Israel Unit of Environmental Engineering, Ben-Gurion University of the Negev, Beer-Sheva 84105, Israel § Lehrstuhl f€ur Technische Chemie II, Universit€at Duisburg-Essen, 45117 Essen, Germany ‡
ABSTRACT: Because most “low fouling” polymers resisting bacterial attachment are hydrophilic, they are usually also significantly swollen. Swelling leads to purely physical dilution of interaction and weakens attachment; however, these nonspecific contributions are usually not separated from the specific effect of polymer chemistry. Taking advantage of the fact that chemistry and swelling of hydrogels may be independently varied through the fraction of a cross-linker, the roles of chemistry and physical dilution (swelling) in bacterial attachment are analyzed for selected hydrogels. Using as a quantitative indicator the rate of bacterial deposition in a parallel plate setup under defined flow conditions, the observed correlation of deposition rate with swelling provides a straightforward comparison of gels with different chemistries that can factor out the effect of swelling. In particular, it is found that chemistry appears to contribute similarly to bacterial deposition on hydrogels prepared from acrylamide and a zwitterioninic monomer 2-(methacryloyloxy)ethyl) dimethyl-(3-sulfopropyl) ammonium hydroxide so that the observed differences may be related to swelling only. In contrast, these gels were inferior to PEG-based hydrogels, even when swelling of the latter was lower, indicating a greater contribution of PEG chemistry to reduced bacterial deposition. This demonstrates that swelling must be accounted for when comparing different biofoulingresistant materials. Chemical and physical principles may be combined in hydrogel coatings to develop efficient antibiofouling surfaces.
’ INTRODUCTION Bacterial biofilms are ubiquitous and readily form on surfaces exposed to aqueous environments.1 Biofilm formation may inflict damage in various ways, for example, by causing persistent infections, contaminating biomedical devices, damaging underwater structures and ship hulls, and impairing membrane performance in membrane-based water treatment.24 Design of antibiofouling surfaces, for example, via surface modification, must consider the diversity and persistence of bacteria.57 Nature provides remarkable examples of efficiently preventing attachment of bacteria and biofilms to living surfaces and emphasizes the benefits of combining different principles for this purpose. For instance, the skin of whales and dolphins keeps clean of biofilms by using both passive (a gel-like surface) and active mechanisms (ablation and enzymatic digestion of undesired biological materials).8 Synthetic surfaces combining active and passive approaches have been reported,9,10 but, unfortunately, components actively preventing bacterial growth are likely to lose activity in long-term uses. Nevertheless, there is much room for combining even simple “passive” principles to engineer efficient and sustainable biofouling-resistant surfaces. In this work, we analyze the combination of two known principles, referred to here as “chemical” and “physical”, in one surface engineering strategy. Toward this goal, we implement here a thorough quantitative separation of chemical and physical effects, which has not been common. The “chemical” principle refers to the use of r 2011 American Chemical Society
“non-fouling” building blocks that exhibit low affinity to cells and biomacromolecules such as polyethylene glycol (PEG),11,12 zwitterionic moieties,13,14 and overall neutral complexes of oppositely charged polyelectrolytes or polyampholytes built of units bearing alternating charges.15,16 Such materials are typically incorporated to surfaces through procedures based on self-assembly or graft polymerization.57 The “physical” principle analyzed here is nonspecific “dilution” of molecular interactions using highly hydrated materials. In general, one expects that for a given chemistry fewer interactions per contact area should always reduce the thermodynamic driving force for bacterial adhesion to the watergel interface, provided that it is impermeable to bacteria. Both principles are apparently interrelated because the majority of “non-fouling” substances are inherently hydrophilic and the importance of hydration has been recognized.17,18 However, it is emphasized here that chemistry and hydration (swelling) may be varied independently. This is most easily realized in surface coatings based on hydrogels.19,20 Apart from independent control of immobilized water fraction, hydrogels offer several benefits: Their mesh-like structure acts as a steric barrier for particles, bacteria, and large molecules, and no inherent limit is imposed on thickness, as is the Received: April 7, 2011 Revised: May 21, 2011 Published: May 26, 2011 2681
dx.doi.org/10.1021/bm200476g | Biomacromolecules 2011, 12, 2681–2685
Biomacromolecules
ARTICLE
Scheme 1. Monomers Used in This Work for Hydrogel Preparation
case for self-assembled layers or grafted brushes. Then, even if highly swollen, the hydrogel layer may efficiently screen interactions with the underlying surface. In addition, the hydrogel configuration may better suit certain low-fouling materials, for example, neutral polyampholyte complexes.15 Finally, hydrogels may offer additional advantages in specific applications; for example, in membrane filtration, the low transport resistance of swollen hydrogels ensures minimal impact on performance.20,21 Despite many studies of hydrogel surfaces in various contexts of bioadhesion and biofouling,19,20,2227 the roles of chemistry and hydration were usually not clearly separated. While examining the effect of swelling, non-“low-fouling” building blocks or moderately hydrated gels were used, and, whereas the focus was on chemistry, degrees of swelling or hydration were seldom reported. Hence, in this work, we specifically focused on bacterial attachment to hydrogels that (a) are composed of low-fouling moieties and (b) have a high and known swelling. It will be shown that the correlation of bacterial deposition rates with hydrogel swelling provides a comparison of gels with different chemistries and can factor out the nonspecific effect of swelling and that this enables one to estimate the specific contributions of chemical and physical factors to the antibiofouling function of hydrogel-based coatings.
’ MATERIALS AND METHODS Materials. Poly(ethylene glycol) methacrylate (PEGMA; 360 g/mol), 2-(methacryloyloxy)ethyl) dimethyl-(3-sulfopropyl) ammonium hydroxide (SPE), cross-linker N,N0 -methylenebisacrylamide (MBA), initiators N, N,N0 ,N0 -tetramethylethylenediamine (TEMED), ammonium persulfate (APS), 3-(trimethoxysilyl)propyl methacrylate (MPS), and LuriaBertani (LB) broth (50% tryptone, 25% yeast extract, 25% NaCl) were purchased from Sigma-Aldrich. Acrylamide (AAm)-MBA mixtures containing different fractions of cross-linker (2.5, 3.33, and 5%) were purchased from Biological Industries. Hydrogel Synthesis and Swelling Measurements. Highly swollen hydrogels were prepared using three water-soluble monomers, PEGMA, SPE, and AAm, and the degree of swelling was varied through the fraction of MBA. The structure of these building blocks is shown in Scheme 1. A monomer solution (0.5 M AAm, 0.5 M SPE, or 0.4 M PEGMA) containing 0 to 10 mol % MBA, 0.005 M APS, and 0.01 M TEMED was polymerized overnight in a Petri dish. Slabs of a hydrogel were immersed in DI water for 1 week. Monomer conversion (g90%) was verified for polyPEGMA gels by analyzing the water phase using total organic carbon analysis. Thereafter, the gel was taken out of solution and quickly blotted dry with a tissue paper, and its weight Ws was measured. Hereafter, the dry weight Wd was determined after drying under ambient conditions for 5 days and then in vacuum for 12 h; swelling was calculated as S ¼ ðWs Wd Þ=Wd
ð1Þ
For known S the characteristic mesh size of the gel was estimated using the following relation28 ξ ¼ ðCn nÞ1=2 lv1=3
Figure 1. Swelling and estimated average mesh size of the examined hydrogels.
where Cn is the polymer characteristic ratio (the value of methyl methacrylate 14.6 was used), l = 1.54 Å is the sp3 CC bond length, n is the molar ratio of monomer and cross-linker (assuming full conversion), and υ is the polymer volume fraction equal to (S þ 1)1, assuming the same density for polymer and water. Glass-Anchored Hydrogels. Glass slides were cleaned in Piranha solution, thoroughly rinsed, and immersed in a 0.4% solution of MPS and 0.005% acetic acid in 50% ethanol for 30 min. Thereafter, the slides were dried in a flow of nitrogen, cured at 105 °C for 1 h, ultrasonicated in water for 1 min, and dried. A surface-modified slide was pressed against a regular slide keeping a gap between the two by means of a 300 μm thick spacer. The gap was filled with a monomer/MBA/initiator solution, and the assembly was left to polymerize overnight. The upper slide was detached after the assembly was immersed in water. Bacterial Deposition. Pseudomonas fluorescens F113 bacteria tagged with green fluorescence protein29 (kindly received from Prof. Ulrich Karlson) were precultured for 12 h in 20 g/L LB medium, and a 1 mL aliquot was transferred to 100 mL of fresh LB medium and cultured for another 12 h. The cultured bacteria in the stationary stage of growth were centrifuged out, washed by resuspending/centrifuging in water three times, and finally resuspended in a clean 0.1 M NaCl solution to optical density 0.05 (bulk concentration Cb = 1.28 107 bacteria/cm3, standard deviation 6 105 bacteria/cm3). The suspension was fed to a FC 81 parallel plate flow chamber having a channel of cross-section 20 2 mm2 (BioSurface Technologies, Bozeman, MT) at a flow rate 2 mL/min corresponding to a wall shear rate of 6.15 cm/s. A glass slide with an anchored hydrogel layer formed the bottom plate. The chamber was mounted on the stage of an Axioscope fluorescence microscope equipped with a digital camera (Zeiss). Deposited cells were imaged and counted at 100 magnification through the top plate every 5 min for 30 min. Deposition coefficient Kd was determined as follows Kd ¼
ð2Þ 2682
1 dN ACb dt
ð3Þ
dx.doi.org/10.1021/bm200476g |Biomacromolecules 2011, 12, 2681–2685
Biomacromolecules
ARTICLE
where dN/dt is the average slope of number of bacteria versus time and A is the area of view.
’ RESULTS AND DISCUSSION Hydrogel Characteristics. Monomers used in this work to prepare highly swollen hydrogels are often employed for making nonfouling surfaces. PEG-based polymers have been used commonly for this purpose;14,27,30 however, the oxidative and hydrolytic stability of the polymer is relatively poor.31 SPE represents the family of zwitterionic building blocks including phosphorylcholine, sulfobetaine, and carboxybetaine functional groups that are being considered as more stable alternatives to PEGs.6,14,32,33 Note that PEGMA (disregarding the hydroxyl terminus of the side group) and SPE meet the well-known criteria for protein adsorption-resistant surfaces (hydrophilicity, zero net charge and presence of hydrogenbond accepting groups only),12 whereas AAm does not because it contains a hydrogen-bond-donating amide group. (See Scheme 1.) It needs to be emphasized that these criteria for protein resistance, especially, the absence of hydrogen-bond-donating groups, have been the subject of some controversy34 because they seem to contradict some findings, and somewhat different criteria have been proposed as well.11 Nevertheless, they have been supported by many data and are still widely accepted; therefore, they will be used in the present discussion. Equilibrium swelling of prepared hydrogels was in the range 6 to 33 (85 to 97% water) and was inversely related to the crosslinker content (Figure 1). It varied with the type of monomer, and, because some compositions failed to produce stable and homogeneous gels, swelling ranges for monomers used in this study did not fully overlap. Curiously, PEGMA formed a stable gel even without added cross-linker (0% MBA). A possible reason is that PEGMA could contain some PEG-dimethacrylate as an impurity that could act as a cross-linker, yet it is not unlikely that gelation could be a result of interactions and entanglements of PEG side chains playing the role of noncovalent physical crosslinks.35 Swelling of all gels was weakly affected by pH and ionic strength, except for the lower swelling of SPE-based gels at acidic pH, apparently, due to protonation of the sulfonic group. The mesh sizes of the gels could be estimated from swelling using eq 2 and assuming full conversion. For all gels used for bacterial deposition experiments (see the next section), it was between 3.5 and 12.6 nm (Figure 1), far below 1 μm, the typical size of bacteria. It was then expected that deposited bacteria would only be able to adhere to the hydrogel surface and would not settle partially within the hydrogel. Bacterial Deposition Results. The antibiofouling properties of the gels were quantified by initial deposition of P. fluorescens F113 bacteria onto glass-anchored gel films in a parallel plate flow setup.26,36 This setup ensures well-defined and reproducible hydrodynamic conditions that are known to have a strong affect on deposition kinetics.37 It also offers a proper balance between attachment and detachment hydrodynamic forces as well as gravity and adequately represents the situation encountered in many biomedical devices, marine applications, and membrane filtration.3739 Microscopic observation and bacterial counts versus elapsed time showed linear deposition kinetics, indicating that the surface coverage within 30 min was far from saturation (Figure 2A,B). The deposition results for the different hydrogels expressed as the deposition coefficient (eq 3) versus measured swelling are summarized in Figure 2C. An inverse relation between swelling of the gels and deposition coefficient is well-observed, yet each
Figure 2. Initial deposition of P. fluorescens F113 on different gels: (A) a typical fluorescence image of gel surface after 30 min of deposition, deposited bacteria appear as bright dots, scale bar 100 μm; (B) a representative plot showing a linear deposition kinetics; (C) summary of measured deposition coefficients on gels for different monomers and degrees of cross-linking at wall shear rate 6.15 cm1. The percentage near each symbol indicates MBA fraction; the result for bare glass is shown for comparison.
chemistry produced a distinct trend. We suggest that the location of such a trend curve on the plot rather than a particular value of deposition coefficient can be considered as an inherent (chemical) antifouling characteristic of the specific monomer. In this way, the physical effect of swelling is factored out. Therefore, PEGMA-based gels show systematically lower deposition coefficients than SPE- and AAm-based gels, indicating better inherent antifouling properties of PEGMA. In fact, the least swollen polySPE hydrogel did not show advantage over bare glass surface. Interestingly, the polySPE and polyAAm gels fall on about the same trend curve, meaning that their inherent propensity to bacterial adhesion under flow condition is quite similar. In fact, polyAAm gels perform better than polySPE and nearly as good as polyPEGMA, but this appears to be entirely due to their higher swelling. Discussion of Deposition Results. Considering the popular criteria of protein resistance12 with reservations previously pointed out, the apparently similar inherent chemical characteristics of AAm and SPE could be a result of an unfavorable hydrogen-donating group in AAm being offset by the fairly large fraction of hydrophobic groups in SPE. Regardless of the specific reason, Figure 2C clearly indicates that despite the fact that zwitterionic polymers are often considered to be viable alternatives to PEGs, SPE appears not to be as inherently low-fouling as PEGMA, which is in agreement with previously reported data for protein fouling of hydrogel-modified membranes.14 These results emphasize the importance of accounting for swelling as an independent physical factor, superimposed on the chemical 2683
dx.doi.org/10.1021/bm200476g |Biomacromolecules 2011, 12, 2681–2685
Biomacromolecules nature of the gel, when comparing chemically different materials. The results point to the possibility of improving the antifouling performance of hydrogels materials, even nonoptimal by chemical criteria, through combination with high swelling. The trend of deposition monotonically decreasing with swelling observed in Figure 2C is in agreement with most reports on adhesion of blood platelets,23,25 bacteria,22,24,26 aquatic microorganisms,22,38 and marine barnacles40 to hydrogels, even though the degree of swelling in most cases was lower than that used here. At least two groups, however, reported a nonmonotonic relation and reversal of the trend for deposition of blood25 and diatom cells38 on gels at S ≈ 10 (water content ∼90%), that is, in the range of swelling similar to the present study. The cells in these studies were allowed to settle on the surface for extended time rather than be carried along the surface by the flow. Both reports also pointed to irregularities of the gel structure, such as heterogeneity or incomplete gelation, and thus the cells could settle partially within the gel, which could strengthen adhesion. Here we thoroughly avoided the use of such irregular gels; for example, chemically favorable but microphase-separated poly(N-vinyl pyrrolidone) gels were discarded.41 Also, the duration and hydrodynamics in the present experiments precluded settling within the gel, which could explain the monotonic dependency on swelling observed here throughout the whole examined range.
’ CONCLUSIONS We have shown that both hydrogel swelling and chemistry are important in determining bacterial adhesion on hydrogels under tangential flow. Whereas these factors are rarely separated in the design of surfaces resisting bacterial and biofilm adhesion, they may be independently varied, which was accomplished in this work using hydrogel surfaces. Under well-defined hydrodynamic conditions, each hydrogel chemistry shows a distinct trend of bacterial deposition coefficient versus degree of swelling. Such trends are then suggested here as a basis for consistent and straightforward comparison between antifouling coating materials in a way that factors out the physical (“dilution”) effects of swelling. In particular, comparison of the observed trends reveals that the zwitterionic monomer SPE surprisingly appears no better than AAm that does not fulfill the widely used chemical “low-fouling” criteria yet shows a superior performance due to purely physical effect of higher swelling. These two monomers were inferior to PEGMA, for which the effect of low-fouling chemistry was apparently greater. Although the relation between initial bacterial deposition and biofouling in long-term application is not straightforward,42 and, from a practical standpoint, the effectiveness of antibiofouling coatings should include other characteristics as well, such as mechanical strength, chemical stability, uptake of molecular foulants, and so on, the conclusions of this work may be used as guiding principles in the development, analysis, and design of novel antifouling hydrogel coatings. ’ AUTHOR INFORMATION Corresponding Author
*E-mail:
[email protected] (V.F.),
[email protected] (R.K.).
’ ACKNOWLEDGMENT This work was supported by German-Israeli Foundation for Scientific Research and Development (G.I.F.), grant no. 95325.10/2007. We thank one of the reviewers for pointing out the
ARTICLE
presence of dimethacrylated impurities in commercial PEGMA monomers.
’ REFERENCES (1) Flemming, H. C.; Wingender, J. Nat. Rev. Microbiol. 2010, 8, 623. (2) Costerton, J. W.; Stewart, P. S.; Greenberg, E. P. Science 1999, 284, 1318. (3) Chambers, L. D.; Stokes, K. R.; Walsh, F. C.; Wood, R. J. K. Surf. Coat. Technol. 2006, 201, 3642. (4) Mansouri, J.; Harrisson, S.; Chen, V. J. Mater. Chem. 2010, 20, 4567. (5) Banerjee, I.; Pangule, R. C.; Kane, R. S. Adv. Mater. 2011, 23, 690. (6) Krishnan, S.; Weinman, C. J.; Ober, C. K. J. Mater. Chem. 2008, 18, 3405. (7) Rosenhahn, A.; Schilp, S.; Kreuzer, J.; Grunze, M. Phys. Chem. Chem. Phys. 2010, 12, 4275. (8) Baum, C.; Fleischer, L.-G.; Roessner, D.; Meyer, W.; Siebers, D. Biorheology 2002, 39, 703. (9) Cheng, G.; Xue, H.; Li, G. Z.; Jiang, S. Y. Langmuir 2010, 26, 10425. (10) Jiang, S. Y.; Cao, Z. Q. Adv. Mater. 2010, 22, 920. (11) Kane, R. S.; Deschatelets, P.; Whitesides, G. M. Langmuir 2003, 19, 2388. (12) Ostuni, E.; Chapman, R. G.; Liang, M. N.; Meluleni, G.; Pier, G.; Ingber, D. E.; Whitesides, G. M. Langmuir 2001, 17, 6336. (13) Cheng, G.; Li, G.; Xue, H.; Chen, S.; Bryers, J. D.; Jiang, S. Biomaterials 2009, 30, 5234. (14) Susanto, H.; Ulbricht, M. Langmuir 2007, 23, 7818. (15) Chen, S.; Jiang, S. Adv. Mater. 2008, 20, 335. (16) Herzberg, M.; Sweity, A.; Brami, M.; Kaufman, Y.; Freger, V.; Oron, G.; Belfer, S.; Kasher, R. Biomacromolecules 2011, 12, 1169–1177. (17) Chen, S. F.; Li, L. Y.; Zhao, C.; Zheng, J. Polymer 2010, 51, 5283. (18) Mendelsohn, D.; Yang, S. Y.; Hiller, J. A.; Hochbaum, A. I.; Rubner, M. F. Biomacromolecules 2003, 4, 96. (19) Peppas, N. A.; Hilt, J. Z.; Khademhosseini, A.; Langer, R. Adv. Mater. 2006, 18, 1345. (20) Yang, Q.; Adrus, N.; Tomicki, F.; Ulbricht, M. J. Mater. Chem. 2011, 21, 2783. (21) Sagle, A. C.; Van Wagner, E. M.; Ju, H.; McCloskey, B. D.; Freeman, B. D.; Sharma, M. M. J. Membr. Sci. 2009, 340, 92. (22) Ekblad, T.; Bergstroem, G.; Ederth, T.; Conlan, S. L.; Mutton, R.; Clare, A. S.; Wang, S.; Liu, Y. L.; Zhao, Q.; D’Souza, F.; Donnelly, G. T.; Willemsen, P. R.; Pettitt, M. E.; Callow, M. E.; Callow, J. A.; Liedberg, B. Biomacromolecules 2008, 9, 2775. (23) Fax€alv, L.; Ekblad, T.; Liedberg, B.; Lindahl, T. L. Acta Biomater. 2010, 6, 2599. (24) Rasmussen, K.; Østgaard, K. Water Res. 2003, 37, 519. (25) Kulik, E.; Ikada, Y. J. Biomed. Mater. Res. 1996, 30, 295. (26) Cook, A. D.; Sagers, R. D.; Pitt, W. G. J. Biomed. Mater. Res. 1993, 27, 119. (27) Krsko, P.; Libera, M. Mater. Today 2005, 8, 36. (28) Lowman, A. M.; Peppas, N. A. Macromolecules 1997, 30, 4959. (29) Boldt, T. S.; Sørensen, J.; Karlson, U.; Molin, S.; Ramos, C. FEMS Microbiol. Ecol. 2004, 48, 139. (30) Li, L.; Chen, S.; Zheng, J.; Ratner, B. D.; Jiang, S. J. Phys. Chem. B 2005, 109, 2934. (31) Qian, Z. Y.; Li, S.; He, Y.; Liu, X. B. Polym. Degrad. Stab. 2004, 84, 41. (32) Chen, S.; Zheng, J.; Li, L.; Jiang, S. J. Am. Chem. Soc. 2005, 127, 14473. (33) Cho, W. K.; Kong, B.; Choi, I. S. Langmuir 2007, 23, 5678. (34) Yang, Q.; Kaul, C.; Ulbricht, M. Langmuir 2010, 26, 5746. (35) Sagle, A. C.; Ju, H.; Freeman, B. D.; Sharma, M. M. Polymer 2009, 50, 756. (36) Roosjen, A.; Kaper, H. J.; van der Mei, H. C.; Norde, W.; Busscher, H. J. Microbiology 2003, 149, 3239. 2684
dx.doi.org/10.1021/bm200476g |Biomacromolecules 2011, 12, 2681–2685
Biomacromolecules
ARTICLE
(37) Roosjen, A.; Boks, N. P.; van der Mei, H. C.; Busscher, H. J.; Norde, W. Colloids Surf., B 2005, 46, 1. (38) Rasmussen, K.; Østgaard, K. Biofouling 2001, 17, 103. (39) Subramani, A.; Hoek, E. M. V. J. Membr. Sci. 2008, 319, 111. (40) Murosaki, T.; Noguchi, T.; Hashimoto, K.; Kakugo, A.; Kurokawa, T.; Saito, J.; Chen, Y. M.; Furukawa, H.; Gong, J. P. Biofouling 2009, 25, 657. (41) Wu, Y.-D.; Freeman, B. D. J. Membr. Sci. 2009, 344, 182. (42) Lee, W.; Ahn, C. H.; Hong, S.; Kim, S.; Lee, S.; Baek, Y.; Yoon, J. J. Membr. Sci. 2010, 351, 112.
2685
dx.doi.org/10.1021/bm200476g |Biomacromolecules 2011, 12, 2681–2685
