Enzyme-Immobilized Nanofiltration Membrane To Mitigate Biofouling

Jan 4, 2011 - Recently, enzymatic quorum quenching (in the form of a free enzyme or an immobilized form on a bead) was successfully applied to a subme...
3 downloads 10 Views 3MB Size
ARTICLE pubs.acs.org/est

Enzyme-Immobilized Nanofiltration Membrane To Mitigate Biofouling Based on Quorum Quenching Jae-Hyuk Kim,* Dong-Chan Choi, Kyung-Min Yeon, Sang-Ryong Kim, and Chung-Hak Lee* School of Chemical and Biological Engineering, Seoul National University, Seoul 151-744, Korea

bS Supporting Information ABSTRACT: Recently, enzymatic quorum quenching (in the form of a free enzyme or an immobilized form on a bead) was successfully applied to a submerged membrane bioreactor with a microfiltration membrane for wastewater treatment as a novel approach to control membrane biofouling. In this study, a quorum quenching enzyme (acylase) was directly immobilized onto a nanofiltration membrane to mitigate biofouling in a nanofiltration process. In a flow cell experiment, the acylase-immobilized membrane with quorum quenching activity prohibited the formation of mushroom-shaped mature biofilm due to the reduced secretion of extracellular polymeric substances (EPS). The acylase-immobilized membrane maintained more than 90% of its initial enzyme activity for more than 20 iterative cycles of reaction and washing procedure. In the lab-scale continuous crossflow nanofiltration system operated at a constant pressure of 2 bar, the flux with the acylase-immobilized nanofiltration (NF) membrane was maintained at more than 90% of its initial flux after a 38-h operation, whereas that with the raw NF membrane decreased to 60% accompanied with severe biofouling. The quorum quenching activity of the acylase-immobilized membrane was also confirmed by visualizing the spatial distribution of cells and polysaccharides on the surface of each membrane using confocal laser scanning microscopy (CLSM) image analysis technique.

’ INTRODUCTION Biocake formation on a membrane, which can be caused by both the deposition of bulk biomass and biofilm growth, has been conceived as one of the main obstacles in membrane processes such as membrane bioreactor (MBR)1 and the reverse osmosis/nanofiltration process (RO/NF).2 Many researchers have tried to mitigate biocake formation in various ways, for example, through the addition of additives or media,3,4 changes in system design,5 or modification of the membrane surface.6 However, such physicochemical approaches to the biofouling problem might not be fundamental solutions because the biocake is a “complex living community”, not just a deposited mass. Recently, novel molecular biological approaches were attempted to control biofouling in a membrane process by applying quorum quenching (QQ),7,8 e.g., the disruption of quorum sensing (QS) between microorganisms, which regulates their group behaviors such as biofilm formation, secretion of extracellular polymeric substances (EPS), and revelation of virulence.9 In those studies, a quorum quenching enzyme, acylase I, which can degrade the N-acetyl homoserine lactone (AHL) autoinducer of Gram-negative bacteria, was added to the MBR in the form of a free enzyme or an immobilized form on a magnetic carrier.7,8 Those approaches revealed that biofouling in MBR could be effectively reduced by disrupting the quorum sensing mechanism of microorganisms. Although the addition of free or immobilized acylase into the bulk phase in a membrane system obviously showed the mitigation of membrane biofouling in MBR, more attention should be focused on the membrane where the biocake is formed. Accordingly, in this study, we attempted to immobilize the acylase directly onto the NF membrane surface, aiming at the inhibition of quorum sensing between microorganisms in the biocake on the membrane and thus r 2011 American Chemical Society

the reduction of biofouling. This prepared membrane was characterized by various physicochemical analysis techniques and its enzyme activity and stability were also monitored. The structure and architecture of biocakes formed on the raw and acylase-immobilized NF membranes in a crossflow NF system and in a flow cell unit were analyzed using confocal laser scanning microscopy (CLSM) and a 3D-image analysis program.

’ EXPERIMENTAL SECTION Materials. Unless otherwise stated, all chemicals were purchased from Sigma-Aldrich and were used without further purification. The stock solutions of N-acetyl-L-methionine (Tokyo Chemical Industry, Japan) and porcine kidney acylase I (EC-Number 3.5.1.14) were prepared by dissolving the appropriate amount of solute in 20 mM phosphate buffer (PB, pH 7.2) solution and stored at 4 °C for further use. The acylase solution was used after filtering with a 0.45-μm GHP membrane filter (Smartpor, PALL Life Sciences, U.S.). Chitosan stock solution (high molecular weight, 1 wt % in 1 wt % acetic acid solution) was also used after filtering with GF/C glass filter. Flat, sheet-type polyamide nanofiltration membrane (NE404090, WOONGJIN Corp., Korea) was used as the raw NF membrane and the subject of enzyme immobilization. According to the manufacturer, the molecular weight cutoff (MWCO) and the NaCl rejection of the NF membrane at 75 psig are approximately 200 Da and 60-70%, respectively, for a 2000 mg/L NaCl solution. Received: October 14, 2010 Accepted: December 15, 2010 Published: January 4, 2011 1601

dx.doi.org/10.1021/es103483j | Environ. Sci. Technol. 2011, 45, 1601–1607

Environmental Science & Technology

ARTICLE

Figure 1. Schematic diagram of acylase immobilization onto the nanofiltration membrane surface by forming a chitosan-acylase matrix.

Bacterial Species and Culture Conditions. Pseudomonas aeruginosa PAO1 tagged with green fluorescent protein (GFP) was used as a model strain of Gram-negative bacteria and this strain is carbenicillin resistant. PAO1 tagged with GFP was precultured on a tryptic soy agar (TSA, Difco, U.S.) plate supplemented with 200 μg/mL of carbenicillin (Gold Biotechnology, U.S.) for 18 h at 37 °C in an incubator. A fresh single colony from the agar plate was inoculated in 3 g/L of tropic soy broth (TSB, Bacto, U.S.) and cultured for 18 h at 200 rpm at 37 °C to reach a midexponential growth phase with a final optical density at 600 nm (OD600) of 0.66 (cell concentration: 2.9 ((0.4)  109 CFU/mL). One mL of cultured solution was centrifuged for 10 min at 9000 rpm to obtain a pure bacterial cell pellet. After discarding the supernatant, the pellet was resuspended in 1 mL of PB solution (20 mM) by vortexing. After repeating the washing processes 3 times (centrifugation and resuspension), the cells were diluted in deionized water to an appropriate ratio and used as inoculums in the flow-cell test or the NF process. Enzyme Immobilization Procedure. A schematic diagram of enzyme immobilization is depicted in Figure 1. Enzyme immobilization onto the NF membrane surface was carried out in a specially designed holder, such that only the active layer of the equipped membrane was exposed to the reagents and was able to be pressurized and filtered by nitrogen gas up to 5 bar. The effective membrane area that was in contact with the reagents was 59.2 cm2 (11.6 cm 5.1 cm). The enzyme immobilization procedure was performed as follows: 400 μL of chitosan stock solution (1 wt % in 1% acetic acid solution) was mixed with 30 mL of deionized water in a glass beaker. Then, NaOH (0.1 M) was added dropwise to the solution until the pH of the solution reached ∼10. After 2 mL of acylase solution (1000 mg/L) was added to the solution, the solution was mixed for 5 min to form aggregation between chitosan and acylase. Then, the solution was poured into the holder containing the raw NF membrane. The holder was pressurized and filtered by nitrogen gas at 4 bar so that the chitosan-acylase aggregates were deposited on the membrane

surface until no liquid was left in the holder. The membrane was completely filtered again with 30 mL of PB solution (20 mM) to wash out the possible remaining solution on the surface and support layer of the membrane. Then, the deposited aggregates were cross-linked with 5 mL of glutaraldehyde (GA, 0.2%) for 30 min under gentle shaking. The cross-linked membrane (referred to as the Acy-NF membrane) was washed again by filtering PB solution as described above to remove any remaining GA and stored in PB solution (20 mM) at 4 °C. Measurement of Enzyme Activity and Stability. The enzyme activity of the prepared membrane was evaluated by measuring the removal rate of N-acetyl-L-methionine chosen as a surrogate molecule of AHL. To measure the enzyme activity of prepared membranes, the membrane coupon (2 cm  6 cm) was immersed in a glass vial containing 10 mL of N-acetyl-L-methionine solution (0.1 mM, in 20 mM PB) and shaken at 200 rpm at 30 °C. At a designated time interval, an aliquot was withdrawn from the vial and the concentration of N-acetyl-L-methionine was quantified using high-performance liquid chromatography (HPLC). Enzyme stability of prepared membranes was measured under iterative uses by following similar methods described in the literature.10,11 To measure the enzyme stability of prepared membranes under iterative uses, the membrane sample was recovered from a previous cycle and washed with plenty of fresh PB solution. Then, the sample was stored in the same conditions as described above only without the substrate for the next cycle. The relative activity was calculated by the ratio of residual activity to initial activity. Flow Cell Operation. A dual channel flow cell reactor (FC 281, BioSurface Technologies, U.S.) was used to investigate the effect of acylase immobilization onto a membrane surface on the structure and/or composition of biofilm that was formed on it under no driving force conditions.12,13 All of the flow cell devices and feed media (20 L, TSB 60 mg/L in 2 mM PB solution) were autoclaved before the flow cell operation at 121 °C for 20 min and 5 h, respectively. 1602

dx.doi.org/10.1021/es103483j |Environ. Sci. Technol. 2011, 45, 1601–1607

Environmental Science & Technology

ARTICLE

Figure 2. Schematic diagram of the laboratory-scale nanofiltration system.

The raw and Acy-NF membranes (74 mm 24 mm for each) were placed into each channel after a 15-min ultraviolet disinfection. Then, each flow cell channel was inoculated with the 3 mL of washed bacterial suspension diluted in deionized water (∼7  105 CFU/mL). After 1 h of cell attachment onto the membrane surface, the feed media started flowing through the flow cell channel at a constant flow rate of 1.5 mL/min using a peristaltic pump. During the whole 5 days of the flow cell operation, bacterial growth on each membrane surface was monitored with time, and biofilm images at the entrance region of the flow cell unit were captured at the designated times using a digital camera (DXM1200, Nikon, Japan) connected to a fluorescence microscope (E600, Nikon, Japan). After the 5-day operation, the flow cell unit was disassembled and a membrane specimen (2 cm 2 cm) at the entrance region of the flow cell was cut from each membrane sample. The biofilm formed on each membrane specimen was observed with confocal laser scanning microscopy (CLSM, C1 plus, Nikon, Japan) after staining with fluorescent dyes. The detailed methods of biofilm staining and image analysis are described in the Supporting Information. Nanofiltration System. Figure 2 depicts the lab-scale crossflow NF system. The system consists of a feed tank containing media (60 mg/L TSB, 1 mM trisodium citrate in 2 mM PB), a reservoir with a working volume of 1.2 L, two piston pumps, and NF modules operated in parallel at a constant pressure of 2 bar. Carbenicillin (6 mg/L) was added to the feed media before the NF operation to prevent possible contamination. The effective membrane area of the NF membrane was 14.2 cm2. The permeate was continuously discarded after measuring water flux and the retentate was recycled to the reservoir. The media in the feed tank was semicontinuously added to the reservoir to maintain the total working volume. Other operating conditions of the NF system are summarized in the Supporting Information. Before the NF operation, the NF system was sterilized and thoroughly cleaned by applying the method described in the

literature.2 After the cleaning procedure, the NF system was stabilized only with the feed media for 1 h at 2 bar. Then, the washed cell culture was inoculated into the reservoir to achieve an initial cell concentration of ∼102 CFU/mL. After the 38-h NF operation, the membranes were carefully taken out from each membrane module and the biocake on each membrane was investigated by the same method as used in flow cell experiments. Analytical Methods. N-acetyl-L-methionine was quantified using HPLC on a 150 mm 4.6 mm Luna C8 column (Phenomenex, U.S.) connected to a UV detector (2487 dual λ absorbance detector; Waters, USA). The mobile phase was a mixture of water/ acetonitrile (85/15 v/v, 0.1 vol% phosphoric acid) pumped at a flow rate of 1 mL/min. The detection wavelength was set at 210 nm. Limit of detection (LOD) of N-acetyl-L-methionine was calculated to be 0.9 μM. The zeta potential value of the membrane was determined using a commercially available electrophoresis measurement apparatus (ELS-8000, Otsuka Electronics, Japan) with a plate sample-cell. Polystyrene latex particles (diameter 200 nm; Otsuka Electronics, Japan) coated with hydroxy propyl cellulose were dispersed in 0.01 M NaCl and were used as mobility-monitoring particles. Attenuated total reflective Fourier transform infrared (ATRFTIR) spectra were obtained using a Nicolet spectrophotometer 5700 (Thermo Electron Corp., U.S.) with a ZnSe crystal at an incident angle of 45°. All spectra (32 scans at a resolution of 4.0 cm-1) were recorded at 25 °C and were corrected for the atmospheric background spectra.

’ RESULTS AND DISCUSSION Characterization of the Acylase-Immobilized NF Membrane. FE-SEM images of the raw and Acy-NF membranes

are presented in Figure S2. The polyamide skin layer of the NF 1603

dx.doi.org/10.1021/es103483j |Environ. Sci. Technol. 2011, 45, 1601–1607

Environmental Science & Technology

ARTICLE

Figure 3. ATR-FTIR spectra of the raw and acylase-immobilized NF membranes: (a) wavenumber range between 800 and 1800 cm-1 and (b) wavenumber range between 2500 and 4000 cm-1.

Figure 4. (a) Enzyme activity and (b) stability of the Acy-NF membrane under iterative cycles.

membrane was almost entirely covered with chitosan-acylase matrix. Because the chitosan-acylase matrix formed on the membrane acts as an additional barrier to the transportation of water molecules, the water permeability of the Acy-NF membrane was reduced to ∼70% of that of the raw NF membrane. The surface charge of the raw NF membrane was highly negative (zeta potential -70 mV) due to the carboxylic acid (-COOH) functional group, which can be deprotonated in water. After forming the chitosan-acylase matrix on the NF membrane, however, the absolute value of negative charge decreased to -20 mV. The ATR-FTIR spectra of the raw NF and Acy-NF membranes were analyzed to identify the changes in surface functional groups before and after acylase immobilization (Figure 3). The peaks at 1489 and 1585 cm-1 are characteristic of the polysulfone support layer of a NF membrane, and two peaks at 1666 and 1543 cm-1 are attributed to amide I (CdO stretching) and amide II (N;H in-plane bending) in the polyamide skin layer, respectively (Figure 3a, black line).14 After forming the acylase-chitosan matrix on the NF membrane, the ATR-IR spectrum was greatly changed. The most intense and broad band at 950-1200 cm-1 was due to the nature of the

chitosan ring (Figure S1). It was clearly observed that the intensity of the amide I and amide II peaks were noticeably increased, which would originate from the peptide bond in the acylase.15 In the wavenumber ranging from 2500 to 4000 cm-1 (Figure 3b), the broad band corresponds to the hydrogen bonding of the hydroxyl group originating from the chitosan as well as the acylase. The weak primary amine group (-NH2) comes from the chitosan and acylase, while the alkane (-CH3) is attributed to the acetyl group of chitosan. Enzyme Activity and Stability of the Acylase-Immobilized Membrane. The quorum quenching activity and stability of the Acy-NF membrane were determined by measuring the removal rate of N-acety-L-methionine and relative activity during 21 iterative cycles of reuse, respectively. As shown in Figure 4a, the Acy-NF membrane readily degraded the substrate (initial activity 0.39 μM/min per cm2 of membrane) whereas the raw NF membrane hardly removed the substrate despite its potential physicochemical adsorption. Although the initial activity of the prepared membrane is quite high, it can also lose its enzyme activity by the leaching and/or denaturation of the enzyme under iterative reuse. However, the Acy-NF membrane was found to be 1604

dx.doi.org/10.1021/es103483j |Environ. Sci. Technol. 2011, 45, 1601–1607

Environmental Science & Technology

ARTICLE

Figure 5. Reconstructed CLSM images of P. aeruginosa biofilms cultivated on the (a) raw NF and (b) Acy-NF membranes in the flow cell unit after a 5-day operation and then stained with SYTO 9 and ConA. Merged images represent overlays of green (cell) and red (EPS) signal collected by CLSM. Magnification 400. Image size 315 μm  315 μm.

Figure 6. Spatial distribution of areal porosity of P. aeruginosa biofilms cultivated on the raw NF (black circle) and Acy-NF membranes (red triangle) in the flow cell unit: (a) based on cells and (b) based on polysaccharides.

highly stable and retained more than 90% of its initial activity even after 21 iterative cycles of reaction and washing procedure (Figure 4b). Structure and Composition of Biofilm Formed in the Flow Cell Unit. P. aeruginosa PAO1 is a representative microorganism of Gram-negative bacterium that uses AHL as a signal molecule and forms biofilm by its quorum sensing mechanism.16 Therefore, the differences in structure and composition of biofilm formed on the raw NF and Acy-NF membranes in the flow cell unit were compared using PAO1 as a model strain. After 5 days of flow cell operation, the biovolume of PAO1 biofilm cultivated on the Acy-NF membrane was measured to be ∼0.9 μm3 per μm2 membrane area, which is only 24% of that cultivated on the raw NF membrane (∼3.7 μm3 per μm2 membrane area). Moreover, the mushroom-like structure of the PAO1 biofilm, which could be induced by quorum sensing, was observed on the raw NF membrane (Figure 5a) whereas it was hardly observed on the Acy-NF membrane (Figure 5b). The PAO1 biofilm formed on the Acy-NF membrane appeared relatively thin and flat compared to that on the raw NF membrane. To further make quantitative comparison between the two biofilms, the architectures of both biofilms were analyzed using an image analysis technique. The architecture of the biofilm on each membrane was also compared in terms of spatial distribution of areal porosities based

Figure 7. Comparison of flux profiles between the two nanofiltration systems equipped with the raw and Acy-NF membranes under a continuous mode.

on cells and polysaccharides (Figure 6). The thickness of biofilm on the raw NF membrane reached around 45 μm, whereas that on the Acy-NF membrane was only 20 μm. The porosity of the biofilm on the raw NF was much smaller than that on the Acy-NF membrane along the entire depth of the biofilm in terms of 1605

dx.doi.org/10.1021/es103483j |Environ. Sci. Technol. 2011, 45, 1601–1607

Environmental Science & Technology

ARTICLE

Figure 8. Reconstructed CLSM images of biofilm formed on the (a) raw and (b) Acy-NF membranes after a 38-hour operation of the continuous NF of P. aeruginosa and then stained with SYTO 9 and ConA. Magnification 400. Image size 315 μm  315 μm.

polysaccharides as well as cells. This result is similar to those of others who reported that the PAO1 strain lacking a functional quorum sensing system for some reason (mutation in gene16 or addition of quorum sensing inhibitor17) forms flat and undifferentiated biofilm. In our case, the quorum quenching acylase on the surface of the Acy-NF membrane could effectively degrade AHL produced by PAO1 at the substratum level and thus prevent the maturation of the PAO1 biofilm. The areal porosity distributions (Figure 6) coincided well with the visual shapes of the biofilms (Figure 5). The spatial distribution and amount of EPS present on the membrane surface is very important not only because EPS itself is a biofoulant but also because it can make the biofilm matrix denser and harder, which in turn make it difficult to remove. Quantitatively, the volume of polysaccharides on the Acy-NF membrane was 0.7 μm3 per μm2 membrane area, only 30% of that on the raw NF membrane (2.1 μm3 per μm2 membrane area). This is expected since EPS secretion is strongly dependent on the quorum sensing mechanism.8,18 Application of the Acy-NF Membrane to a Continuous NF System. A crossflow nanofiltration with each membrane was operated in parallel to compare the flux between the two membranes (Figure 7). For the raw NF membrane, the flux started to decline after 12 h of the NF operation and continuously decreased to 60% of its initial flux. It is worth noting that the starting point of such a flux decline corresponds to the beginning point of the exponential growth of PAO1. In contrast, for the Acy-NF membrane, any remarkable decrease in flux was not observed all the time during the NF operation, e.g., it still maintained more than 90% of its initial flux even at the end of the NF operation. This indicates that the biofilm, e.g., the water permeation barrier, did not form substantially on the Acy-NF membrane. To further investigate such a big difference in flux between the two membranes, each membrane was taken out after the 38-h NF operation and the structure of biocake formed on each membrane surface was visualized using CLSM. First, the amount of biomass on the Acy-NF membrane formed in the crossflow NF system (Figure 8b) was even much smaller than that formed in the flow cell (Figure 5b), despite the forced deposition of microorganisms from the bulk phase onto the membrane surface due to convective flow. This is because the shear rate in the NF system (cross-flow velocity 0.12 m/s) would certainly be much higher than that in the flow cell (cross-flow velocity 0.00072 m/s). Second, Figure 8 shows a sharp difference in the biocake structure between the two membranes. In the biocake formed on the raw NF membrane where no quorum quenching was involved, the mature biofilm and dense EPS (polysaccharides) indicating severe biofouling were clearly observed (Figure 8a).

On the other hand, only some microcolonies, planktonic bacteria, and little EPS were found on the Acy-NF membrane (Figure 8b), suggesting that the bacteria on the Acy-NF membrane could not grow into a mature biofilm due to the quorum quenching activity of the Acy-NF membrane. The biovolume of the biocake composed of cells and polysaccharides on the Acy-NF membrane was calculated to be only 0.06 ((0.02) μm3 per cm2 membrane area, which is around 5% of that on the raw NF membrane (1.15 ((0.18) μm3 per cm2 membrane area). All of these results strongly suggest that the newly developed membrane with quorum quenching activity has a great antibiofouling feature by suppressing EPS secretion and thus biofilm maturation.

’ ASSOCIATED CONTENT

bS

Supporting Information. Materials for the chemical structure of chitosan, operating conditions of the lab-scale nanofiltration system, biofilm staining and image analysis, and FE-SEM images of the raw and acylase-immobilized nanofiltration membranes. This material is available free of charge via the Internet at http://pubs.acs.org.

’ AUTHOR INFORMATION Corresponding Author

*J.-H.K. e-mail: [email protected]. C.-H.L. phone: þ 82-2-8807075; fax: þ82-2-874-0896; e-mail: [email protected].

’ ACKNOWLEDGMENT This work was supported by the Korea Science and Engineering Foundation (KOSEF) grant funded by the government (MOST) (2007-0056709) and Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology (MEST) (2010-0029061). We thank WOONGJIN Co. Ltd., Korea, for providing the nanofiltration membranes. ’ REFERENCES (1) Hwang, B. K.; Lee, W. N.; Yeon, K. M.; Park, P. K.; Lee, C. H.; Chang, I. S.; Anja, D.; Matthias, K. Correlating TMP increase with microbial characteristics in the bio-cake on the membrane surface in a membrane bioreactor. Environ. Sci. Technol. 2008, 42 (12), 3963– 3968. (2) Herzberg, M; Elimelech, M. Biofouling of reverse osmosis membranes: role of biofilm-enhanced osmotic pressure. J. Membr. Sci. 2007, 295, 11–20. 1606

dx.doi.org/10.1021/es103483j |Environ. Sci. Technol. 2011, 45, 1601–1607

Environmental Science & Technology

ARTICLE

(3) Hwang, B. K.; Lee, W. N.; Park, P. K.; Lee, C. H.; Chang, I. S. Effect of membrane fouling reducer on cake structure and membrane permeability in membrane bioreactor. J. Membr. Sci. 2007, 288, 149–156. (4) Lee, W. N.; Kang, I. J.; Lee, C. H. Factors affecting filtration characteristics in membrane-coupled moving bed biofilm reactor. Water Res. 2006, 40 (9), 1827–1835. (5) Yeon, K. M.; Park, J. S.; Lee, C. H.; Kim, S. M. Development of a new MBR process: Membrane coupled high-performance compact reactor for wastewater treatment. Water Res. 2005, 39 (10), 1954–1961. (6) Liu, C. X.; Zhang, D. R.; He, Y.; Zhao, X. S.; Bai, R. Modification of membrane surface for anti-biofouling performance: Effect of antiadhesion and anti-bacteria approaches. J. Membr. Sci. 2010, 346, 121– 130. (7) Yeon, K. M.; Cheong, W. S.; Oh, H. S.; Lee, W. N.; Hwang, B. K.; Lee, C. H.; Beyenal, H.; Lewandowski, Z. Quorum Sensing: A new biofouling control paradigm in a membrane bioreactor for advanced wastewater treatment. Environ. Sci. Technol. 2009, 43 (2), 380–385. (8) Yeon, K. M.; Lee, C. H.; Kim, J. B. Magnetic enzyme carrier for effective biofouling control in the membrane bioreactor based on enzymatic quorum quenching. Environ. Sci. Technol. 2009, 43 (19), 7403–7409. (9) Fuqua, C.; Winans, S. C.; Greenberg, E. P. Census and consensus in bacterial ecosystems: The LuxR-LuxI family of quorumsensing transcriptional regulators. Annu. Rev. Microbiol. 1996, 50, 727–751. (10) Kim, J.; Lee, J.; Na, H. B.; Kim, B. C.; Youn, J. K.; Kwak, J. H.; Moon, K.; Lee, E.; Park, J.; Dohnalkova, A.; Park, H. G.; Gu, M. B.; Chang, H. N.; Grate, J. W.; Hyeon, T. A magnetically separable, highly stable enzyme system based on nanocomposites of enzymes and magnetic nanoparticles shipped in hierarchically ordered, mesocellular, mesoporous silica. Small 2005, 1 (12), 1203–1207. (11) Lee, J.; Lee, Y.; Youn, J. K.; Bin Na, H.; Yu, T.; Kim, H.; Lee, S. M.; Koo, Y. M.; Kwak, J. H.; Park, H. G.; Chang, H. N.; Hwang, M.; Park, J. G.; Kim, J.; Hyeon, T. Simple synthesis of functionalized superparamagnetic magnetite/silica core/shell nanoparticles and their application as magnetically separable high-performance biocatalysts. Small 2008, 4 (1), 143–152. (12) Cheng, G.; Li, G.; Xue, H.; Chen, S.; Bryers, J. D.; Jiang, S. Zwitterionic carboxybetaine polymer surfaces and their resistance to long-term biofilm formation. Biomaterials 2008, 30 (28), 5234–5240. (13) Stapper, A. P.; Narasimhan, G.; Ohman, D. E.; Barakat, J.; Hentzer, M.; Molin, S.; Kharazmi, A.; Høiby, N.; Mathee, K. Alginate production affects Pseudomonas aeruginosa biofilm development and architecture, but is not essential for biofilm formation. J. Med. Microbiol. 2004, 53 (7), 679–690. (14) Kim, J. H.; Park, P. K.; Lee, C. H.; Kwon, H. H. Surface modification of nanofiltration membranes to improve removal of organic micro-pollutants (EDCs and PhACs) in drinking water treatment: Graft polymerization and cross-linking followed by functional group substitution. J. Membr. Sci. 2008, 321, 190–198. (15) Belfer, S.; Fainchtain, R.; Purinson, Y.; Kedem, O. Surface characterization by FTIR-ATR spectroscopy of polyethersulfone membranes-unmodified, modified and protein fouled. J. Membr. Sci. 2000, 172, 113–124. (16) Davies, D. G.; Parsek, M. R.; Pearson, J. P.; Iglewski, B. H.; Costerton, J. W.; Greenberg, E. P. The involvement of cell-to-cell signals in the development of a bacterial biofilm. Science 1998, 280, 295–298. (17) Hentzer, M.; Riedel, K.; Rasmussen, T. B.; Heydorn, A.; Andersen, J. B.; Parsek, M. R.; Rice, S. A.; Eberl, L.; Molin, S.; Høiby, N.; Kjelleberg, S.; Givskov, M. Inhibition of quorum sensing in Pseudomonas aeruginosa biofilm bacteria by a halogenated furanone compound. Microbiology 2002, 148 (1), 87–102. (18) Sakuragi, Y.; Kolter, R. Quorum-sensing regulation of the biofilm matrix Genes (pel) of Pseudomonas aeruginosa. J. Bacteriol. 2007, 189 (4), 5383–5386.

1607

dx.doi.org/10.1021/es103483j |Environ. Sci. Technol. 2011, 45, 1601–1607