Introduction

Wastewater treatment is the grand challenge of the 21st century. Many technologies have been developed to meet this challenge, which includes membrane-based treatment processes. One such process is the membrane bioreactor (MBR), which combines an activated sludge process with a solid–liquid separation by ultra or microfiltration membranes replacing the usual sedimentation step [43]. MBRs have many advantages over conventional activated sludge treatments, including small footprint and reactor requirements, good disinfection capability, higher volumetric loading, and less sludge production [18,46]. As a result, MBRs have been applied in numerous fields such as water/wastewater treatment and seawater desalination due to their capability to remove organic matters from wastewaters and salts from seawater, which make them highly suitable for industrial and agricultural purposes [1,9]. Additionally, with the continuous reduction of membrane cost and easy availability, several pilot-scale MBR plants have been installed for industrial and municipal wastewater treatment and are also currently under development [24].

Despite all these great promises, MBRs also face some major problems in operation, and membrane fouling is the issue of concern that limits the development of membrane systems. Recent studies have demonstrated that bacterial biofilms that form on membrane surfaces are the main reason behind membrane biofouling [17]. This leads to a decline in permeate flux, increase in the trans-membrane pressure, and finally reduction in the treatment performance [13]. Moreover, this shortens the life of membranes and leads to higher operation costs of MBRs owing to membrane replacement [6]. Some physicochemical strategies for membrane fouling mitigation such as physical cleaning (i.e., relaxation) and chemical maintenance cleaning are known [24]. More recently, some antimicrobial compounds such as nitrofurazone, chlorhexidine, silver salts, and antibacterial peptides have also been tried for inhibition of membrane biofouling [15]. However, physicochemical strategies cause damage to membranes and thus require high costs for their replacement. On the other hand, antimicrobial compounds are toxic to target as well as non-target microorganisms and also lead to multidrug resistance among them. Thus, all these strategies fail with the fact that the control of membrane biofouling in MBRs is not to kill the bacteria or limit their growth but to inhibit the biofilm formation.

Since the N-acyl homoserine lactone-mediated quorum sensing mechanism of gram-negative bacteria was reported to regulate biofilm formation [3], its inhibition thus suggests a fundamental approach for membrane biofouling control in MBRs without limiting bacterial growth [11]. The MBR treating wastewater has been known to contain various gram-negative bacteria that produce multiple types of AHLs to regulate a single phenotype biofilm formation [22]. Certain natural compounds that function as quorum sensing inhibitors vs. vanillin, furanones, curcumin, (–)-epigallocatechin gallate, etc. have been reported to demonstrate anti biofouling activity without affecting bacterial growth [25,35,36,41]. Thus, the use of such natural quorum sensing inhibition (QSI) compound suggests an alternative approach for inhibition of membrane biofouling, which is important given that the likelihood of developing multidrug resistance among bacteria in wastewater is low. Additionally, exploiting the potential of natural QSI in combinations that can interfere with different AHLs of varying acyl chain length indicates an attractive option for combating multispecies gram-negative bacteria biofouling. The combined effect of natural compounds has been evaluated previously for antibacterial activity against multidrug-resistant pathogenic bacteria [2]. However, no report is available on the combined use of natural compounds for inhibition of AHL-mediated biofilm formation in gram-negative bacteria. Therefore, as a model, two natural compounds with different chemical structure (curcumin (CCM) and (–)-epigallocatechin gallate (EGCG)) were selected in this study and evaluated in combination for inhibition of AHL-mediated biofilm formation in gram-negative bacteria. CCM is a diphenolic compound generally used in turmeric form throughout central and eastern Asia as a spice in food preparations [2]. It has traditionally been used as an antimicrobial and anti-inflammatory agent; however, a few reports revealed that CCM could be used for inhibition of biofilm formation in gram-negative bacteria [31,33,39]. Additionally, a recent study reported the inhibition of QS-mediated biofilm development in Vibrio spp. by CCM [34]. The other compound EGCG, a polyphenol constituent of green tea, has great antimicrobial potential against various gram-negative as well as gram-positive microorganisms [2,12,32,42]. Moreover, it is known to inhibit AHLs-mediated phenotype biofilm formation in Escherichia coli O157:H7 [25]. All these previous reports provide a suitable basis for exploring their combined QSI potential for control of biofilm formation.

With this background, the present study investigated the combined effect of CCM and EGCG on inhibition of AHL-mediated biofilm formation in bacteria from wastewater. Initially, the minimum inhibitory concentrations (MICs) of the individual compounds and their combination were determined against individual bacteria as well as their multispecies culture. Furthermore, the inhibition of biofilm formation in test bacteria was studied at one-half MICs. The potential of CCM and EGCG to interfere with a wide-range of AHL-mediated phenotype expression in known biosensor strains was evaluated. The effect of CCM and EGCG alone and their combination on inhibition of a multispecies bacterial culture biofilm formation on PVDF membrane surface was studied with a Center for Disease Control (CDC) biofilm reactor. Formation of biofilm on the membrane surface was estimated by extracellular polymeric substances (EPS) measurement. Finally, the quantification of biofilm was carried out by confocal laser scanning microscopy (CLSM) observations while the biofilm morphology was analyzed by scanning electron microscopy.

Materials and Methods

Chemicals and Microbiological Media

Curcumin powder (≥90% purity) extracted from Curcuma longa, green tea constituent (–)-epigallocatechin gallate (≥95% purity), phosphate-buffered saline, osmium tetroxide, 5-bromo-4-chloro-3-indolyl-β-D-galactopyranoside (X-gal), N-butanoyl-ʟ-homoserine lactone (C4-HSL), N-hexanoyl-ʟ-homoserine lactone (C6-HSL), and N-decanoyl-ʟ-homoserine (C10-HSL) were purchased from Sigma-Aldrich (St. Louis, MO, USA). The chemical structures and known QSI activities of the plant-derived compounds CCM and EGCG are given in Table 1. The LIVE/DEAD BacLight bacterial viability kit containing solutions of 3.34 mM SYTO9 nucleic acid stain in dimethyl sulfoxide (DMSO) and 20 mM propidium iodide in DMSO was purchased from Invitrogen, Molecular Probes (Eugene, OR, USA).

Table 1.Chemical structure and quorum sensing inhibition activity of CCM and EGCG used in the present study.

Biofilm-Forming Bacteria, Biosensor Strains, and Culture Conditions

Nine gram-negative biofilm-forming bacteria isolated from membrane bioreactor activated sludge in our previous study were used in this work [22]. These included Aeromonas hydrophila subsp. hydrophila strain NA1 (Acc. No. KF938658), Aeromonas hydrophila subsp. dhakensis strain LBA2 (Acc. No. KF938660), Enterobacter ludwigi strain SWA1 (Acc. No. KF938661), Pseudomonas japonica strain TSA3 (Acc. No. KF938663), Enterobacter cancerogenus strain LBA4 (Acc. No. KF938664), Klebsiella variicola strain SWA2 (Acc. No. KF938665), Citrobacter freundii strain R2A5 (Acc. No. KF938666), Serratia marcescens strain SWA6 (Acc. No. KF938667), and Raoultella ornithinolytica strain TSA7 (Acc. No. KF938668). At the start of each experiment, all the biofilm-sforming bacteria were routinely grown in Luria-Bertani (LB) broth (BD-Difco, Franklin Lakes, NJ, USA) for 24 h at 28 ± 0.2℃ under shaking conditions (120 rpm). Then, the cultured cells were harvested by washing with sterilized phosphate-buffered saline (PBS, 0.01M, pH 7.4) and pelleted by low-speed centrifugation (1,000 ×g for 10 min, 4 ± 0.2℃) for three times [47]. The resulted cells were resuspended in sterile PBS and named the starter culture. The cell concentration of starter culture was estimated by measuring the optical density (OD595nm) using a UV-Visible spectrophotometer (Genesys 10 UV; Thermo Fisher Scientific Inc., Waltham, MA, USA). Additionally, countable 10-fold dilutions were prepared in PBS and aliquots of 10 μl were spread on LB agar plates. After overnight incubation at 28 ± 0.2℃, colonies on the plates were counted to determine the number of colony-forming units (CFU). The linear relationship between OD595nm and number of CFU was established for each bacterium starter culture to get desired cell concentrations.

Two biosensor strains Chromobacterium violaceum 026 and Agrobacterium tumefaciens A136, which are deficient in AHL production but respond to a wide range of exogenous AHLs by color production, were used for QSI bioassay. Both the biosensor strains were routinely grown in LB medium supplemented with appropriate antibiotics at 28 ± 0.2℃ for 24 h under shaking condition (120 rpm) [22].

Determination of MICs

The MICs of CCM, EGCG, and their combination against all the individual biofilm-forming bacteria were determined in 96-well flat-bottom polystyrene cell culture plates (SPL Life Sciences Co. Ltd., Pocheon-Si, Korea). For this, 190 μl of LB medium containing different concentrations of CCM (50-500 μg/ml), EGCG (100-1,000 μg/ml), and equal amount of each compound as their combination (50-500 μg/ml) were added into the horizontal wells of microtiter plates. Wells without compounds served as a positive growth control, while antibiotic streptomycin (50-500 μg/ml) was used as the standard antibacterial. All the wells were inoculated with 10 μl of active culture from each of the test bacteria to get a final concentration of 105 CFU. After incubation at 28 ± 0.2℃ for 24 h, the OD595nm was measured with a plate reader (iMark Microplate Absorbance Reader 168-1135; Bio-Rad, Hercules, CA, USA) and the MIC recorded as the lowest concentration where complete inhibition of visible growth was observed. The MICs of CCM, EGCG, and their combination against a multispecies culture of nine bacteria were also determined with a similar protocol. All the MIC determinations were made in triplicate and mean values are presented.

Effects of CCM, EGCG, and their Combination on Biofilm Inhibition

The effects of CCM, EGCG, and their combination on biofilm formation by AHL-producing bacteria on solid surface (1.9 cm2) were investigated in 24-well flat-bottom polystyrene cell culture plates (SPL Life Sciences Co. Ltd., Korea) according to an earlier report [35]. The one-half MICs of CCM, EGCG, and their combination were prepared in LB medium and 900 μl of each was placed into the horizontal wells of individual microtiter plates. Then, 100 μl of test bacteria culture was added to each well to get the final concentration of 105 CFU/ml. The plates were statically (without agitation) incubated at 28 ± 0.2℃ for 24 h to allow attachment of bacteria on the solid surface. After attachment, the wells were rinsed with sterile distilled water and the plate was dried at 60 ± 1.0℃ for 30 min. Subsequently, the wells were stained with 0.2 ml of 0.1% (w/v) crystal violet solution for 20 min, washed again with sterile distilled water, and then 1 ml of ethanol was added into each well to dissolve the remaining crystal violet. Aliquots of 200 μl were withdrawn and added to a 96-well cell culture plate, and the OD595nm was measured with a plate reader. The OD values were considered as an index of bacterial adherence to solid surface and biofilm formation [48]. The absorbance values of the controls were subtracted from the experimental values, and the percent biofilm inhibition was determined. The biofilm inhibition assay was performed in triplicate and the results are represented as the mean ± standard deviation.

QSI Bioassay

The potential of CCM and EGCG to interfere with short-chain (C4-HSL and C6-HSL) and long-chain (C10-HSL) AHL-mediated phenotype expression in AHL-deficient biosensor strains C. violaceum CV026 and A. tumefaciens A136 was investigated by agar well diffusion assay [21]. For violacein pigmentation assay, warm molten LB medium (0.8% agar) maintained at 40 ± 1.0℃ was individually incorporated with 10 μM of C4-HSL and C6-HSL as an exogenous AHL and inoculated with 1% of overnight grown culture of C. violaceum CV026 having OD595nm of 1.0. Additionally, for β-galactosidase activity, warm molten LB medium (0.8% agar) was incorporated with 80 μg/ml of X-gal as a visualizing agent, 10 μM of C10-HSL as exogenous AHLs, and inoculated with 1% of overnight grown culture of A. tumefaciens A136 having OD595nm of 1.0. The contents were gently mixed and 15 ml was immediately poured per Petri dish. The agar plates were allowed to solidify and wells of 6.5 mm diameter were made with a cork borer. Fifty microliters of a filter-sterilized solution of CCM (150 μg/ml) and EGCG (275 μg/ml) made in DMSO was loaded per well. DMSO was used as the negative control and 2(5H)-furanone (100 μg/ml) as the positive control for QSI [35]. The plates were incubated at 28 ± 0.2℃ for 24 to 48 h and observed for halo zones of violacein inhibition in C. violaceum CV026 and inhibition of blue coloration from β-galactosidase activity in A. tumefaciens A136.

Batch Studies with CDC Biofilm Reactor

The effects of CCM, EGCG, and their combination on inhibition of biofilm formation by a multispecies culture of nine bacteria were investigated by CDC biofilm reactor studies (BioSurface Technologies Corp, Bozeman, MT, USA) as described earlier [19]. The CDC reactor, medium storage tank, polypropylene coupon holder, magnetic stirrer bar, and tubing were autoclaved at 121℃ for 20 min and connected under a clean bench. The PVDF microfiltration flat membrane of pore size 0.22 (Merck Millipore, Darmstadt, Germany) was cut into 1.5 cm × 1.5 cm pieces, sterilized by immersing in 40% ethanol for 1 h followed by UV light exposure for 4 h (2 h each side), and stuck onto one side of removable coupon holders using sterile double-sided cellophane tape. Then, the rods were placed in the CDC reactor in such a way that the fixed membrane faced feed side out, rendering a total surface area of 2.25 cm2. All the steps were carried out under a clean bench.

At the beginning, the CDC biofilm reactors were filled with 349 ml of 1/10th strength LB medium containing one-half MIC of CCM (150 μg/ml), EGCG (275 μg/ml), and their combination (CCM 100 μg/ml and EGCG 100 μg/ml). Then, the reactors were inoculated with 0.1 ml each (105 CFU/ml) of starter culture of nine bacteria as a multispecies culture to initiate biofilm formation. The starter culture of each bacterium was prepared as described in the above section, with only the exception of dissolving in 1/10th strength LB medium instead of PBS. The reactor was run at 28 ± 0.2℃ and 150 rpm for 24 h. The control reactor was run in parallel under similar conditions, with the exception of natural compounds in 1/10th strength LB medium.

Extraction and Measurement of EPS

The EPS formed on PVDF membrane surface were extracted by a thermal method [28]. Briefly, after 24 h of incubation, the membrane specimens were carefully detached from coupon holders and gently washed with 30 ml of 0.9% NaCl solution. Next, the washed membrane specimens were moved into conical tubes containing 15 ml of 0.9% NaCl solution, vortexed for 5 min, and then sonicated for 60 min (B5510, Branson Ultrasonics, Danbury, CT, USA). Thereafter, the membrane specimens were removed from solution and the content was centrifuged (5,000 ×g for 20 min, 4 ± 0.2°C) to collect released biofilms. The supernatant was collected, filtered through a 0.45 μm cellulose acetate filter (Sterlitech Corporation, Kent, WA, USA), and used to measure dissolved EPS. The remaining pellets were further dissolved in 15 ml of 0.9% NaCl solution, heated at 80 ± 1.0°C for 60 min in a water bath, cooled to room temperature, and then centrifuged (5,000 ×g for 20 min, 4 ± 0.2°C). Finally, the resultant supernatant was filtered through a 0.45 μm cellulose acetate filter and the filtrate was collected as bound EPS. The EPS were defined as the sum of proteins and polysaccharides that were quantified by using the Lowry method with bovine serum albumin as standard [27] and anthrone method with glucose as standard [8].

CLSM Observations

After incubation of 24 h, the membrane specimens were detached from the coupon holders and immediately stained with a LIVE/DEAD Backlight kit (Life Technologies, Grand Island, NY, USA) according to the manufacturer’s instructions. Briefly, a total of 200 μl of fluorescent staining solution containing propidium iodide (6 μM) and SYTO9 (5 μM) was added onto the membrane specimens for 30 min at room temperature under the dark. The excess stain was carefully washed with deionized sterile water and the membrane specimens were mounted on glass slides (covered with a coverslip). Microscopic observation and image acquisition were performed on the stained membrane specimens using confocal laser scanning microscopy (LSM 710, ZEISS, Oberkochen, Germany). The microscope was equipped with argonion and diode-pumped solid-state lasers sets for monitoring SYTO9 (excitation wavelength = 488 nm, emission wavelength = 505/30 nm) and propidium iodide (excitation wavelength = 543 nm; emission wavelength = 566/585 nm), respectively. Biofilm images were observed with a 40× water immersion objective (numerical aperture of 1.2) and a series of z-stacks were acquired through optical sectioning at 1 μm of slice thickness. Each membrane specimen with adhered biofilm was scanned randomly (4-5 positons) with the same setting of excited laser intensity, background level, contrast, and electronic zoom size. The generated images covered an area of 424.68 μm × 424.68 μm with a resolution of 1,024 × 1,024 pixels of 8-bit. The three-dimensional reconstruction of the CLSM images was carried out using Zen 2009 software. Finally, the images were analyzed, and the total biomass (μm3/μm2), surface coverage (%), and average thickness (μm) of biofilm layer were determined by the MATLAB 5.1 (The MathWorks, Inc., Natick, MA, USA) software COMSTAT ver. 1.1 [14].

Scanning Electron Microscopy (SEM)

After CDC reactor study, all the membrane specimens were detached from coupon holders and subjected to biofilm morphology analysis by a field emission scanning electron microscope (FESEM S-4200; Hitachi, Tokyo, Japan). For this, the membrane specimens were air dried for 2 h at room temperature, fixed with 2.5% (v/v) glutaraldehyde in phosphate buffer (PBS; 0.01 M, pH 7.4) for 1 h, and washed three times for 10 min with PBS. The second fixation step was performed by keeping the membrane specimens in PBS supplemented with 1% (v/v) osmium tetroxide for 1 h. Then, the excess amount of osmium tetroxide was removed by washing three times with PBS for 10 min, dehydrated for 20 min with ascending ethanol concentrations (30%, 50%, 70%, 90%, and 100%), and dried for 5 h in desiccators at room temperature. Finally, the membrane specimens were coated with a platinum layer and examined at 15 kV voltage.

Analysis of EPS by FTIR

In order to characterize the organic matter in the form of EPS, Fourier transform infrared spectroscopy (FTIR) analysis was carried out. The membrane specimens detached from coupon holders were carefully washed with deionized sterile water for three times to remove all the medium and residual compounds. After being dried at 25 ± 1.0℃ in the dark for 5 h, the membrane surface characterization was carried out with a Nicolet iS 10 FTIR Spectrometer (Thermo Fisher Scientific Inc., Waltham, MA, USA) [29]. The membrane specimens were pressed in the sample holder and spectra were collected in air, in the mid-infrared region ranging from 400 to 4,000 cm-1, with the average of 16 scans at a resolution of 4. All spectra were base-line corrected and the background spectrum was subtracted from that of the membrane spectra to remove any atmospheric absorbance peaks.

Results and Discussion

Inhibitory Effects of CCM, EGCG, and their Combination

Initially, to gain an understanding of the inhibitory effects of CCM, EGCG, and their combination on bacterial growth, the MICs were determined. The result of preliminary study on the MICs of CCM and EGCG alone and their combination against biofilm-forming bacteria in wastewater are listed in Table 2. Both compounds CCM and EGCG alone were capable of inhibiting the growth of all tested bacteria in the concentration range 200-350 μg/ml and 350-650 μg/ml, respectively. However, the combination of CCM and EGCG in equal amount significantly lowered the MICs values up to 100-250 μg/ml. These results suggested that the activity of both compounds in combination greatly inhibits the growth of tested bacteria as compared with that of each alone. The results of the present study falls in line with the findings of Betts and Wareham [2], in which the combined activity of CCM with EGCG reduced the MIC of CCM by 3- to 7-fold against multidrug-resistant strains of Acinetobacter baumannii. Additionally, Sasidharan et al. [40] also observed that CCM in combination with third-generation antibiotic cephalosporins significantly reduced the antibiotic MICs against pathogenic bacteria associated with infectious diarrhea. Different mechanisms for the antibacterial effect of CCM have been known, which includes inhibition of core metabolic pathways associated with folic acid metabolism (shikimate dehydrogenase) [4] and bacterial cell division protein FtsZ [37]. On the other hand, EGCG is known to possess antibacterial activity by the inhibition of bacterial type II fatty acid synthesis [50] and dihydrofolate reductase, which results in blocking of folic acid metabolism [44].

Table 2.Values are the mean of three independent experiments. aAn equal amount of CCM and EGCG was used as combination for determination of MIC.

Effects of CCM, EGCG, and Their Combination on Biofilm Inhibition

Bacterial biofilm formation is known as one of the most crucial factors to cause membrane biofouling in MBRs treating wastewaters, and use of natural compounds suggests an eco-friendly approach for the inhibition of biofilm formation without affecting microbial growth [23]. Thus, the effects of CCM, EGCG, and their combination on the inhibition of biofilm formation in bacteria from wastewater were investigated. Results of the study at one-half MICs indicate that CCM and EGCG alone inhibited 52-68% and 58-78% of biofilm formation, respectively, whereas their combination inhibited almost complete (95-99%) biofilm formation (Table 3). These findings suggest that the combination of CCM and EGCG was more effective in the inhibition of biofilm formation than that of the individual compounds. The enhanced biofilm inhibition by the combination of compounds might be due to the interference of a wide range of AHLs by the two compounds. It was documented that the combination of natural compounds xylitol with ursolic acid significantly inhibited the biofilm formation in oral pathogenic bacteria S. mutans 159 and S. sobrinus 33478 [51].

Table 3.Values are given as the mean ± SD, obtained from three independent experiments.

Quorum Sensing Inhibition

The results of the QSI bioassay suggest that CCM (150 μg/ml) inhibited only C4 and C6 HSL-mediated phenotype expression of violacein pigmentation in C. violaceum 026 (Fig. 1). On the other hand, EGCG (275 μg/ml) inhibited C4 and C6 HSL-mediated violacein pigmentation as well as the C10 HSL-mediated phenotype expression of blue coloration by β-galactosidase activity in A. tumefaciens A136. It is known that the C. violaceum 026 strain bears a LuXR homolog, CviR, and regulates the violacein pigmentation when induced by C4 and C6 exogenous HSLs [30]. Another strain, A. tumefaciens A136, bears the traI-lacZ fusion (pCF218) (pCF372) plasmids and displays blue coloration from the hydrolysis of X-gal by β-galactosidase activity in response to a variety of exogenous AHLs (C6 to C14 HSLs) [20,38]. These results suggest that CCM possesses only short-chain AHL-mediated phenotype expression inhibitory activity, whereas EGCG inhibits both short-chain long-chain AHL-mediated phenotype expression. As a prerequisite, it was confirmed that both the compounds at tested concentration did not show bactericidal effects against biosensor strains as determined by CFU count (data not shown). The QSI activity of CCM has been well reported for the inhibition of C6 HSL-mediated violacein pigmentation in C. violaceum 026 [33]. However, we report for the first time the QSI mechanism of EGCG, which acts through down-regulating the expression of C4, C6, and C10 HSL-mediated phenotypes violacein pigmentation in C. violaceum 026 and β-galactosidase activity in A. tumefaciens A136.

Fig. 1.QSI activity of CCM and EGCG against C4, C6 and C10 HSL-mediated phenotype expression in AHL-deficient biosensor strains C. violaceum CV026 and A. tumefaciens A136.

CDC Biofilm Reactor Study

To further analyze the AHL-mediated inhibition of biofilm formation in a multispecies culture of nine bacteria by CCM, EGCG, and their combination, the CDC biofilm reactor study was carried out. In a number of gram-negative bacteria, the formation of biofilm is known to be quorum-sensing regulated and positively correlated with the EPS production [3,33]. The EPS are one of the major components of biofilms and have a decisive role in building and keeping the structural integrity of microorganisms [26,49]. Considering the fact that EPS are the main contributors to membrane biofouling in MBR technology [7], the development of EPS formation on PVDF membrane surface was investigated. The proteins and carbohydrates have been identified as the major components of EPS [45]. Thus, the concentration of EPS in terms of protein and carbohydrate was monitored after 24 h of reactor run. Results of the study revealed that the concentrations of protein and carbohydrate in EPS were 5 and 12 μg/cm2 in the CCM incorporated reactor, 4 and 10 μg/cm2 in the EGCG incorporated reactor, and 1 and 2 μg/cm2 in the combined compounds incorporated reactor (Table 4). The PVDF membranes from all the three natural compounds-containing reactors possessed significantly lower protein/carbohydrate content than that of control reactor, which had 14 and 36 μg/cm2 of protein/carbohydrate ratio. These observations indicate that carbohydrates constituted a major fraction of biofilm in the control reactor, whereas the natural compounds incorporated reactors showed a much less amount, suggesting inhibition of biofilm formation. The calculated EPS concentrations from the CDC biofilm reactor exposed PVDF membranes were 50 μg/cm2 in control, 17 μg/cm2 in CCM, 14 μg/cm2 in EGCG, and 3 μg/cm2 in combined compounds.

Table 4.Values are given as the mean ± SD, obtained from three independent experiments.

The EPS analysis data suggest that EPS production in the CCM + EGCG incorporated reactor was much lower than that of the CDC biofilm reactor containing CCM or EGCG alone, which means the biofilm in the combined compounds reactor develops much slower or not over the 24 h run time than that in other two reactors. The low EPS content in the combined compounds incorporated CDC biofilm reactor may be due to the down-regulation of biofilm phenotype associated with a wide range of AHLs from various bacteria by the combined action of both the compounds, as CCM and EGCG alone were found to inhibit only particular AHLs (Fig. 1).

CLSM Observations

The LIVE/DEAD staining of PVDF membrane specimens exposed to a multispecies culture of nine bacteria in the presence of CCM or EGCG alone and their combination after 24 h of run time was carried out. This time period was selected because it was sufficient for the tested bacteria to form a biofilm and therefore provided a considerable measure of biofilm inhibition by the test compounds. The biofilms formed in terms of live and dead cells were viewed three-dimensionally using the image analysis software Zee 2009 and results are shown in Fig. 2. For the control reactor, the CLSM image was dominated by green fluorescence, indicating the heavy amount of biofilm formation by the multispecies culture of nine bacteria on the PVDF membrane surface (Fig. 2A). However, CCM and EGCG alone significantly inhibited the biofilm formation on membrane surfaces, as evidenced by the decreased intensity of green as well as red fluorescence (Figs. 2B and 2C). Consequently, the combination of CCM and EGCG almost completely inhibited the biofilm formation, as there was only negligible green fluorescent color observed (Fig. 2D). These CLSM observations clearly suggest that combination of CCM with EGCG results in the strong inhibition of biofilm formation compared with that by the individual compounds.

Fig. 2.Three-dimensional reconstructions of CLSM images of PVDF membrane surfaces.

The CLSM images of biofilm formed on membrane surface after 24 h by the multispecies culture of nine bacteria were quantified and measured in terms of total biomass, surface coverage, and mean thickness, using the program COMSTAT ver. 1.1. The biofilm quantification parameters are presented in Table 5. Results of the biofilm measurement suggest that the amount of total biomass formed on the control membrane specimen was much higher (0.508 μm-3 μm-2) than that of individual CCM- and EGCG treated membranes. It was also observed that over 96% less biomass was formed on the combined compounds treated membrane. Additionally, the biofilm formed on the combined compounds treated membrane was much thinner (not detected) and covered only 1% membrane surface compared with that of control. From the CLSM image analysis, it was clearly revealed that the quorum quenching effect of combined compounds results into strong inhibition of biofilm formation.

Table 5.Values are given as the mean ± SD, obtained from three independent experiments. nd = Not detected.

SEM Analysis

The surface morphologies of control and natural compounds treated PVDF membranes are presented in Fig. 3. A typical thick biofilm layer formed by the multispecies culture of nine bacteria was observed on the control membrane with SEM (Fig. 3A). Additionally, the CCM and EGCG treated membrane surfaces also showed the presence of biofilm structure, with less density than that of the control membrane (Figs. 3B and 3C). By contrast, membrane treated with the combination of CCM and EGCG was considerably varied in terms of surface morphology and did not show any biofilm structure on the membrane surface, which indicates the inhibition of biofilm formation (Fig. 3D). Comparison of these images reveals that biofilm formation was significantly interfered with by the combination of CCM and EGCG. These SEM results also support the CLSM observation, where the mean biofilm thickness of CCM + EGCG treated membrane was almost non-existent (i.e., not detected).

Fig. 3.SEM images of PVDF membrane surfaces.

FTIR Analysis of the Membranes

The FTIR spectra of virgin, control as natural compounds-unexposed, and treated PVDF membrane specimens are presented in Fig. 4. The FTIR spectra of all the membrane specimens showed a broad region of peaks between 1,401 and 1,069 cm-1, which are due to C–F stretching of carboxylic acids. Additionally, the C–H band of trisubstituted alkenes was detected in the range of 839.6-795.8 cm-1. However, the control membrane surface showed three additional peaks at 3,024, 2,957, and 2,921 cm-1, which correspond to OC–H stretching of epoxides and C–H stretching of alkanes, respectively. Moreover, CCM and EGCG treated membranes also showed a similar peak pattern as the control but with less intensity. The appearance of a specific peak above wave number 3,000 cm-1 in the control as well as individual CCM and EGCG treated membranes is indicative of a hydroxyl group in polysaccharides, which is a major component of biofilm [29]. Furthermore, there was a peak at 1,160 cm-1, corresponding to H–N–H bonds, indicative of protein. However, the significant difference in the transmittance intensity of IR spectrum indicated that the amount of biofilm formed on the control membrane surface was substantially higher than that of CCM and EGCG treated membrane specimens. On the other hand, the membrane treated with the combination of CCM and EGCG did not show the appearance of specific peaks associated with EPS, which suggests the inhibition of biofilm formation. The biofilm is known to be made up of EPS produced by bacteria that grow on a solid surface [5,10]. Therefore, the results of the IR analysis suggest that the combination of CCM and EGCG is more effective in inhibition of EPS formation and thus biofilm formation on PVDF membrane surfaces. Similar IR results have been reported by Mafirad et al. [29] for PVDF fouled membranes.

Fig. 4.FTIR spectra of PVDF virgin membrane, control as natural compounds-unexposed, and treated membrane specimens.

The overall results of the study indicate that the combination of CCM and EGCG exhibits enhanced activity towards inhibition of AHL-mediated biofilm formation in bacteria from MBR activated sludge. CCM inhibits the expression of short-chain AHL-mediated phenotype, whereas EGCG inhibits both short- and long-chain AHL-mediated phenotype expression in known biosensor bacteria. The CDC biofilm reactor study demonstrated that both compounds alone could reduce EPS production in terms of biofilm formation in a multispecies bacterial culture; however, their combination was more effective. It appears that the use of CCM and EGCG in combination has good potential for the inhibition of AHL-mediated biofilm formation in gram-negative bacteria and thus could be used for mitigation of membrane biofouling in MBRs treating wastewaters.