Characterization of the Extracellular Polymeric Substances Produced

686

Biomacromolecules 2008, 9, 686–695

Characterization of the Extracellular Polymeric Substances Produced by Escherichia coli Using Infrared Spectroscopic, Proteomic, and Aggregation Studies Kevin E. Eboigbodin and Catherine A. Biggs* Department of Chemical and Process Engineering, The University of Sheffield, Mappin Street, Sheffield S1 3JD, United Kingdom Received September 18, 2007; Revised Manuscript Received November 19, 2007

The aim of this study was twofold: first, to characterize the free extracelluar polymeric substances (EPS) and bound EPS produced by Escherichia coli during different growth phases in different media, and then to investigate the role of the free EPS in promoting aggregation. EPS was extracted from a population of E. coli MG1655 cells grown in different media composition (Luria–Bertani (LB) and Luria–Bertani with the addition of 0.5 w/v% glucose at the beginning of the growth phase (LBG)) and at different growth phases (6 and 24 h). The extracted EPS was characterized using Fourier transform infrared spectroscopy and further identified using one-dimensional gel-based electrophoresis and tandem mass spectrometry. E. coli MG1655 was found to produce significantly lower amounts of bound EPS compared to free EPS under all conditions. The protein content of free EPS increased as the cells progressed from the exponential to stationary phase when grown in LB or LBG, while the carbohydrate content only increased across the growth phases for cells grown in LBG. FTIR revealed a variation in the different functional groups such as amines, carboxyl, and phosphoryl groups for free EPS extracted at the different growth conditions. Over 500 proteins were identified in the free EPS, with 40 proteins common in all growth conditions. Proteins with functionality related to amino acid and carbohydrate metabolism, as well as cell wall and membrane biogenesis were among the highest proteins identified in the free EPS extracted from E. coli MG1655 under all growth and media conditions. The role of bound and free EPS was investigated using a standardized aggregation assay. Bound EPS did not contribute to aggregation of E. coli MG1655. The readdition of free EPS to E. coli MG1655 resulted in aggregation of the cells in all growth conditions. Free EPS extracted from the 24 h E. coli MG1655 cultures grown in LB had the greatest effect on aggregation of cells grow in LBG, with a 30% increase in aggregation observed.

Introduction A major player in microbial aggregation or biofilm formation is the natural ability of cells during their growth to produce biopolymers, often called extracellular polymeric substances (EPS).1 EPS were previously thought to be composed mainly of polysaccharides, and therefore traditionally polysaccharides were the most well studied component. This is why the abbreviation “EPS” is mostly used to describe the extracellular polysaccharides or exopolysaccharides.2,3 However, current studies have shown that significant levels of proteins and nucleic acids can be found in the EPS extracted from biofilms,4 pure cultures,5 and activated sludge.6,7 In this paper, the term EPS refers to polysaccharides, proteins, nucleic acid, and other biopolymers situated outside the cell. EPS are complex heterogeneous substances, and their composition and location is due to several metabolic processes such as changes in growth phase, cell breakage due to cell death, active secretion, release of cell surface macromolecules (outer membrane proteins and lipopolysaccharides), and interaction with the environment.2 EPS biosynthesis and composition vary from one bacteria spp. to another and have been shown to be controlled by several environmental factors such as: growth phase,3 growth media,8 temperature,9 limitation of oxygen,10,11 nitrogen,12 and cation (e.g., magnesium, calcium, and phosphate) deficiency.13 Hence * Corresponding author. E-mail: [email protected]. Telephone: (+44) 01142227510. Fax: (+44) 01142227501.

it is not surprising that conflicting reports regarding the compositions of EPS have been given in the literature. Furthermore, the method used to extract bacterial EPS, before analysis, can have an effect on the composition reported. Several methods for extracting EPS have been reported; however, it is difficult to compare results due to the lack of a standard protocol. This is a major drawback in the general analysis of EPS. Zhang et al.14 compared five different methods for extraction of EPS for aerobic/sulfate-reducing biofilm; these were regular centrifugation, EDTA extraction, ultracentrifugation, steaming extraction, and regular centrifugation with formaldehyde (RCF). RCF and the steaming method gave the highest level of carbohydrate and protein content. Liu and Fang,6 also compared EDTA, cation exchange resin, and formaldehyde extraction methods under various conditions with the EPS content from aerobic, acidogenic and methanogenic sludge. The authors observed that formaldehyde plus NaOH was the most appropriate method for extracting bound EPS because it released only relatively low levels of nucleic acid.6 Sheng et al.5 recently extracted EPS from Rhodopseudomonas acidophila using four different extraction methods (EDTA, NaOH, H2SO4, heating/ centrifugation). Here, the authors observed that the EDTA method was the most effective for extracting EPS, again because it released only relatively low levels of nucleic acids (15% for the other three methods) and no cell lysis.5 Proteins were the dominant component that made up the composition of the EPS of Rhodopseudomonas acidophila extracted with EDTA (83.1% of total EPS) compared

10.1021/bm701043c CCC: $40.75  2008 American Chemical Society Published on Web 01/11/2008

Extracellular Polymeric Substances from E. coli

Biomacromolecules, Vol. 9, No. 2, 2008 687

Experimental Procedures

Figure 1. Distribution of bound and free EPS surrounding bacteria.

with NaOH (79.5% of total EPS) and heating/centrifugation (52.7% of total EPS).5 However, for the H2SO4 extraction method, the dominant component was carbohydrates (49.5% of the total EPS with proteins making up only 29.0% of the total EPS). EPS can be classified by its relative proximity to the cell surface. EPS tightly linked via a covalent or noncovalent association are known as capsular EPS (or cell-bound EPS), while EPS which are not directly attached to the cells surface are known as slime (or free EPS).2 A representation of the distribution of bound and free EPS around bacteria can be seen in Figure 1. The type of EPS can also be distinguished based on the extraction method used; free EPS can be separated from the medium by centrifugation with the bound EPS still attached to the cells or aggregates. A further extraction step such as EDTA extraction or ultracentrifugation is required to separate the bound EPS from the cells or aggregates.5 Omoike et al.15 extracted free and bound-EPS from Bacillus subtilis at different growth phases using regular centrifugation and NaCl, respectively. The authors observed that Bacillus subtilis produced significantly higher amounts of free EPS than bound EPS. The growth phase of Bacillus subtilis also influenced the composition of the EPS produced. The protein content of the free EPS were found to significantly increased relative to the carbohydrate content as cells progresses from exponential to stationary phase. EPS are mainly responsible for the structural integrity of aggregates and biofilms as well as playing a key role in the physicochemical properties of these multicellular structures.16 EPS provides an attractive force,16–19 which keeps cells together and attaches aggregates and cells to biotic or aboitic surfaces (in the case of biofilms). The interaction between EPS and the solid surface can be attractive or repulsive depending on the constituents of the EPS or a variation in the affinity of the EPS for the solid surface or aqueous phase.20,21 Previous findings have also revealed that EPS contribute to the surface properties of microoorgansims.22 The role of EPS in aggregation may follow a similar manner as biofilm formation, where the interaction between the free EPS and cells is likely to be influenced by the composition of both the free and bound EPS and the cell surface chemistry of the bacteria.23 In this work, therefore, bound and the free EPS from Escherichia coli harvested from different media, at different growth phases, is first characterized in terms of its carbohydrate and protein content and then more specifically in terms of chemical functional groups using Fourier transform infrared (FTIR). The protein content of the free EPS is further characterized using proteomic analysis tools to identify key metabolic and biological processes that contribute to the proteins found within the free EPS. The contribution of both bound and free EPS to the process of biological aggregation is then investigated using aggregation assays previously reported by Eboigbodin et al.24

All chemicals were purchased from Sigma-Aldrich (Gillingham, Dorset, UK) unless otherwise stated. All experiments were conducted in triplicate (at least), and the average of the results was reported. Variation in the experimental results is presented as the average ( standard deviation. Bacterial Strains and Growth Studies. Escherichia coli MG1655 was used in this study and cultivated with aeration at 30 °C in Luria–Bertani (LB) medium supplemented with or without 0.5 w/v (%) of glucose at the beginning of the growth experiments as previously reported by Eboigbodin et al.24 LB media supplemented with 0.5 w/v (%) glucose is referred to as LBG throughout this paper. Overnight culture (∼16 h) of E. coli MG1655 was used to inoculate fresh LB or LBG at a 1:100 dilution, and the cells were grown at 30 °C with aeration. The optical density at 595 nm was measured using a spectrophotometer (ThermoSpectronic, UK). Viability tests for the E. coli cells, under same conditions used for the aggregation assay and surface characterization techniques, were performed using a Live/Dead BacLight kit (Molecular Probes, UK) following the manufacturer’s protocol. Cells were viewed with an Axioplan II imaging fluorescence microscope equipped with DF10 filter (Carl Zeiss Ltd., UK). Extracellular Polymeric Substances (EPS) Extraction. Cells were harvested by centrifugation at 5000 rpm for 15 min at 4 °C. The cell pellets were used for extraction of the bound EPS, and the supernatant was used for extraction of the free EPS. Samples were collected from cells grown in LB and LBG after 6 and 24 h. Bound EPS was extracted from the cell pellet using the EDTA method as described by Sheng et al.5 with some modifications, as this method was reported to give reproducible results, releasing low levels of nucleic acids with no cell lysis, compared with other methods.5 Walker et al.41 also used EDTA to extract cell bound EPS from E. coli cells (E. coli D21g) with minimum change in cell viability. Cells were washed twice with 0.9% NaCl to remove any traces of the media. The washed cells were resuspended in 1:1 volume of solution 0.9% NaCl and 2% EDTA then incubated for 60 min at 4 °C. The supernatant (containing bound EPS) was then harvested by centrifugation at 10 000g at 4 °C for 60 min and then filtered through a nitrocellulose membrane (Fisher Scientific, UK). Free EPS was extracted from the supernatant collected after the initial harvesting of the cells by centrifugation at 5000 rpm for 15 min at 4 °C, using the method described by Omoike and Chorover.15 The supernatant was recentrifuged at 10 000g for 30 min at 4 °C to remove residual cells, and then the supernatant containing free EPS was precipitated with 1:3 volume ethanol and stored at -20 °C for 18 h. Free EPS were then removed by centrifugation at 10 000g for 15 min at 4 °C. The extract was resuspended in ultra pure water and dialyzed against ultrapure water to removed ethanol using 2000 MWCO Spectrum DispoDialyzer (Medicell, UK). Both bound and free EPS were stored at -20 °C until needed for further analysis. The EPS was then analyzed for total protein content using the Lowry method with bovine serum albumin (BSA) as a standard.25 The total carbohydrate content was also analyzed using the Anthrone method26 with glucose as the standard. A Student paired t test was used to determine any statistical differences (at a 95% confidence level (P < 0.05)) between the protein or carbohydrate composition of the bound and free EPS extracted from cells at different growth phases, which had been grown in the same media. Fourier Transformation Infrared Spectroscopy (FTIR). Fourier transform infrared (FTIR) spectroscopy is a rapid nondestructive method that has been applied to many biological systems.27–29 The technique is based on the principle that atoms in molecules are not held rigidly apart and, when subjected to infrared radiation (between 300 and 4000 cm-1), the molecule will absorb the energy and the bond will be subjected to a number of different vibrations. Hence the absorption spectrum contains information regarding the molecular structure of the sample. Both bound and free EPS extracted from E. coli MG1655 at different growth phases (6 and 24 h) and growth media (LB and LBG)

688 Biomacromolecules, Vol. 9, No. 2, 2008 were characterized using a Perkin-Elmer Spectrum One Fourier transformation infrared spectrophotometer (Perkin-Elmer, UK). Then 20 µL of EPS, collected from cells grown in LB and LBG after 6 and 24 h, were allowed to dry at room temperature for 45 min on a CaF2 slide. At least 100 scans, with a resolution of 4 cm-1, were collected for all samples between 4000 and 900 cm-1. Absorption band assignments corresponding to functional groups of macromolecules, in the region between 1800 and 900 cm-1, was based on observations of the vibration patterns, previously reported for bacteria24,27,30–32 (see Table S1 in the Supporting Information for a summary of the banding assignments). Protein Extraction from the Free EPS. Proteins were extracted from the free EPS extracted from cells grown in LB and LBG after 6 and 24 h by precipitating with 1:5 ice-cold acetone and stored in -20 °C for 18 h. The solution was then centrifuged at 15 000g for 10 min and the supernatant discarded. Acetone precipitation was then repeated for the pellets containing proteins and then stored for further analysis. The concentration of proteins was then analyzed using the Lowry method with bovine serum albumin (BSA) as a standard.25 Sodium Dodecyl Sulfate Polyacrylamide Gel Electrophoresis (SDS-PAGE). SDS-PAGE was performed on proteins in the free EPS extracted from E. coli MG1655 using a 10% (w/v) polyacrylamide gel, following the method previously described by Laemmli.33 Protein extract (100 µg/mL protein) was dissolved in ultrapure water and then electrophoresed at constant voltage (120 V) until the bromophenol blue tracking dye front reached the bottom of the gel. Low-molecular-weight protein markers (New England Biolabs, UK) were used as protein standards, and the protein bands were stained with Bio-Safe Coomassie blue stain (Bio-Rad, UK). In Gel Digestion. Gels obtained from the SDS-PAGE were rinsed with deionized water, cut into smaller pieces, and then subjected to tryptic digestion as previously described by Gan et al.34 The gel pieces were then destained twice with 200 µL of 200 mM ammonium bicarbonate in 40% acetonitrile (ACN) and then vaccuum-dried for approximately 15-30 min. The gel pieces were then subjected to reduction and alkylation using 200 µL 10 mM dithiothreitol (DTT) and 200 µL 55 mM iodoacetamide (IAA), respectively. Reduction and alkylation buffers were discarded and gel pieces were washed twice with 200 µL 50 mM ammonium bicarbonate and then once with 200 µL 50 mM ammonium bicarbonate in 50% acetonitrile. All liquid from the gel pieces were discarded after centrifugation at 13 000g for 10 s, and the gel pieces were then vaccuum-dried for approximately 20 min. Then 20 µL of trypsin solution (0.4 µg) and 50 µL of 40 mM ammonium bicarbonate in 9% acetonitrile (ACN) were added to the gel pieces and then incubated overnight (16 h) at 37 °C. Peptides were then extracted separately with 50 mM ammonium bicarbonate, acetonitrile, 5% formic acid (FA), and 5% formic acid in 50% acetonitrile. The solution from each extraction stage was combined and vaccuumdried for 20 min and stored at -20 °C prior to mass spectrometric analysis. Nanoliquid Chromatography Electrospray Ionization-Tandem Mass Spectrometry. The dried peptide samples from the in gel digestion were resuspended in 7.5 µL of buffer (0.1% formic acid and 3% ACN) and sonicated for 5 min. Samples were analyzed using a QStar XL Hybrid ESI Quadrupole TOF tandem mass spectrometer, ESI qQ TOF-MS/MS (Applied BioSystems, Framingham, MA; MDSSciex, Concord, Ontario, Canada), coupled with a nanoLC system comprising a combination of an LC Packings Ultimate Pump, Switchos pump, and Famos autosampler (Dionex/LC Packings) as previously described.34 The peptide mixture was separated on a 75 µm Pepmap capillary column. Data acquisition on the mass spectrometer was set to perform data acquisition in the positive ion mode, with a selected mass range of 300–2000m/z. Peptide with 2+ and 3+ charge states were selected for fragmentation. Database Searching. Mass spectra obtained were searched against the mass spectra of protein sequences in the NCBI Database using Mascot 1.6b.16 (www.matrixscience.com). Search parameters were set

Eboigbodin and Biggs at a peptide tolerance of up to 0.2 Da and an MS/MS tolerance of up to 0.2 Da; oxidation of methionine and carbamidomethyl modification of cysteine as modifications and one missed cleavage of trypsin was allowed. Molecular Weight Search (MOWSE) scores greater than 50 were considered as significant.35 The hydrophobicity of proteins identified were determined by the grand average hydropathy (GRAVY) index36 using ProtParam tool37 at http://us.expasy.org/tools/protparam.html. The GRAVY value for a protein is derived from the sum of hydropathy values for each amino acid residue, divided by the length of residues in the sequence. Proteins with a negative GRAVY index are considered hydrophilic, and proteins with a positive GRAVY index are considered hydrophobic, with a GRAVY index value greater than +0.3 used as an indicator of hydrophobic proteins.38 The original cellular location of the proteins identified was also predicted using the program PSORTb v.2.039 at http://psort.org/psortb/ index.html. However, the PSORTb v.2.0 program can not predict lipoproteins, and hence LipoP v1.0 software40 was also used, which predicts lipoproteins by differentiating between lipoprotein signal peptides from other signal peptides.40 Aggregation Studies. To determine the role of EPS in the process of biological aggregation two separate studies were conducted on the bound EPS and free EPS, respectively, as described below. Role of Bound EPS. The role of bound EPS in aggregation of E. coli MG1655 was observed by carrying out an aggregation assay on E. coli MG1655 cells before and after extraction of the bound EPS. The assay was conducted on cells grown in LB and LBG and harvested after 6 and 24 h, respectively. The aggregation assay used has been previously described by Eboigbodin et al.24 Cells were harvested and washed twice with 0.9% NaCl (pH 7.0). The cell pellet was resuspended in 0.9% NaCl, and the optical density (OD) was adjusted to ∼0.6 prior to measurement to ensure standardization across the tests. Then 1 mL of cells was transferred into the cuvettes. As the cells aggregate, they settle to the bottom of the cuvette, and the OD corresponding to cells at the upper part of the cuvette was measured. The percentage aggregation was calculated as the difference in OD600 readings taken at time 0-5 h according to eq 1

% aggregation )

OD0 - ODt × 100 OD0

(1)

Here OD0 is the OD at time 0, at the beginning of the test, and ODt is the OD measured at the end of the test at 5 h.24 The buffer solution of 0.9% NaCl was used in the aggregation studies, rather than distilled water, to prevent any pH effect on E. coli aggregation. Role of Free EPS. To observe the role of free EPS in E. coli MG1655 aggregation, an aggregation assay similar to the assay mentioned above was used, apart from the fact that extracted free EPS is added to the cuvettes containing the washed cells. The process is similar to aggregation studies conducted by Eboigbodin et al.23 for E. coli AB1157 aggregated using sodium polystyrene sulfonate. Free EPS was extracted from cells harvested at 6 and 24 h in both LB and LBG (i.e., four different free EPS samples) and added to cells harvested at 6 and 24 h in both LB and LBG (i.e., four different cells samples). Supporting Information Table S2 outlines the combination of free EPS extracts and cell samples conducted which resulted in 10 different experiments. In each case, 0.1 mL of free EPS extract (0.05 ( 0.005 mg) was transferred into cuvettes containing 0.9 mL of cells (0.5 ( 0.05 mg), and the OD was measured at 600 nm both initially (OD0) and again after 5 h (ODt). The percentage of aggregation was then calculated using eq 1. For each condition, the aggregation studies were conducted in at least triplicate with the variation in the experimental results is presented as the average ( standard deviation.

Results and Discussion Characterization of EPS. Bulk Composition of EPS. The growth studies of E. coli MG1655 cultivated in LB and LBG



Table 1. Carbohydrate and Protein Content of Bound and Free EPS of E. coli MG1655 Extracted at Different Growth Phases, Cultivated in Luria–Bertani (LB) and LB with 0.5 w/v (%) glucose (LBG)a

bound EPS E. coli LB 6 h E. coli LBG 6 h E. coli LB 24 h E. coli LBG 24 h free EPS E. coli LB 6 h E. coli LBG 6 h E. coli LB 24 h E. coli LBG 24 h

carbohydrate

protein

carbohydrate/ protein ratio

3.53 ( 1.91b 1.89 ( 0.07 1.34 ( 0.03b 2.27 ( 1.07

1.43 ( 0.43b 0.88 ( 0.07 0.42 ( 0.25b 0.54 ( 0.25

2.47 2.15 3.19 4.20

14.26 ( 0.02b 10.61 ( 0.21b 7.00 ( 0.44b 20.51 ( 0.79b

9.79 ( 1.03b 7.98 ( 1.44b 25.66 ( 2.26b 17.17 ( 1.38b

1.46 1.33 0.27 1.19

a Variation in the replicates are reported as the average ( SD. b A Student paired t test was used to assess whether the composition of EPS extracted from cells grown in the same media but harvested at the exponential (6 h) and stationary (24 h) phase were statistically different at 95% confidence level (P < 0.05).

have previously been reported by Eboigbodin et al.24 with no significant difference in growth rate between cells cultivated in LB or LBG. Under the conditions of this study, after 6 h, E. coli MG1655 cells were still in the exponential phase (between mid and late exponential phase), and at 24 h, the cells were in the stationary growth phase. The uptake of 0.5 w/v (%) glucose during batch cultivation of cells was also described with over 97% of glucose taken up by the cells after 6 h.24 Viability of cells was found to be approximately 83 and 81% for cells at exponential (6 h) and stationary (24 h) phase, respectively, in both types of media.24 Table 1 shows the protein and carbohydrate content of both the bound and free EPS produced at different growth phases in LB and LBG. The carbohydrate content of bound EPS extracted from cells cultivated in LB significantly decreased as cell progressed from the exponential phase (6 h) to the stationary phase (24 h). Conversely, for cells cultivated in LBG, the carbohydrate content of the bound EPS did not change significantly (at the 95% confidence level) during both the exponential phase (6 h) and stationary growth phase (24 h). The protein content of bound EPS extracted from the cells cultivated in LB also significantly decreased as the cells proceeded from the exponential (6 h) to the stationary (24 h) growth phase. However, there was no significant change in protein content for cells grown in LBG between the exponential and stationary growth phases. The low values observed in Table 1 for bound EPS (protein and carbohydrate) suggest that E. coli MG1655 produces very little EPS in the growth phases for which the experiment was conducted. The amount of EPS (protein and carbohydrate) produced by this strain is very low when compared to previously reported work (e.g., the protein and carbohydrate content of EPS from Pseudomonas putida in batch cultures was 214 and 54 mg/g, respectively4), but is consistent with previous reports of E. coli strains that show little or no EPS.41,42 When comparing the concentration of bound versus free EPS in Table 1, it can be seen that E. coli MG1655 produces higher amounts of free EPS compared to bound EPS under the same growth conditions. The carbohydrate content of free EPS decreased as the cells progressed from the exponential to stationary phase for cells cultivated in LB. However, for cells cultivated in LBG, the carbohydrate concentration increased as the cells progressed from exponential to stationary phase. This increase in carbohydrate content is not simply due to the presence of glucose remaining in the supernatant as the glucose

Figure 2. FTIR spectra of bound EPS extracted from E. coli MG1655 cultivated with Luria–Bertani (LB) and extracted during exponential (6 h) and stationary phase (24 h).

concentration in the supernatant after 24 h has been measured as negligible.24 The protein content of the free EPS from cells cultivated in both LB and LBG increased as the cells progressed from the exponential to stationary phase (Table 1). The increase in the protein content is consistent with previous findings suggesting that E. coli cells shed their outer membrane or periplasmic proteins during the growth phase.43 The most observable change in carbohydrate/protein (C/P) ratio for free EPS was for cells cultivated in LB as they proceed from the exponential to stationary phase (Table 1). This finding is consistent with Omokie et al.,15 who showed that C/P ratio of free EPS extracted from Bacillus subtilis cultivated in LB decreased from 2.29 to 0.85 as cells proceeded from exponential to stationary phase. As the amount of carbohydrate and protein concentration varied in the free EPS extracted from E. coli cells, due to a change in media (LB and LBG) and growth phase, it stands to reason that their physiochemical characteristics may also differ. While measurement of the bulk carbohydrate and protein content gives a general overview of the EPS composition and allows comparison with other microorgansims, FTIR analysis and largescale study of proteins (i.e., proteomics), as describe below, enables a more detailed picture of the EPS to be observed. FTIR Characterization Studies of EPS. The spectra of bound EPS extracted from E. coli MG1655 cultivated in LB at exponential (6 h) and stationary (24 h) phases are shown in Figure 2. The effect of a change in media (LBG) on the bound EPS content of E. coli MG1655 harvested at exponential (6 h) and stationary phase (24 h) is also analyzed using FTIR analysis spectroscopy (Figure 3). Each spectrum contains information about the functional groups arising from predominantly carbohydrate, proteins, and nucleic acids. In Figures 2 and 3, the band at approximately the 1645 cm-1 region corresponds to the amine I, which are characteristic of functional groups CdO stretching vibrations of proteins. The peak at 1450 cm-1 corresponding to bending of CH3 and CH2 groups from proteins (δCH2, δCH3) may also have a contribution from an amine III band. The peak at ∼1400 cm-1 arises from to the stretching C-O (υC-O) of carboxylic groups which overlap with the amide III band making it difficult to distinguish. The bands between 1200 and 950 cm-1 are attributed to the vibrations of C-O-P and C-O-C stretching

690 Biomacromolecules, Vol. 9, No. 2, 2008

Figure 3. FTIR spectra of bound EPS extracted from E. coli MG1655 cultivated with Luria–Bertani with 0.5 w/v (%) glucose (LBG) and extracted during exponential (6 h) and stationary phase (24 h).

of diverse polysaccharide groups, and the bands at 1260 and 1080 cm-1 exhibit the stretching of PdO (υPdO) of phosphoryl and phoshodiester groups from phosphorylated proteins, polyphosphate products, and nucleic acids. [Note: the absorption band assignments corresponding to functional groups of these macromolecules, in the region between 1800 and 900 cm-1, was based on observations of the vibration patterns, previously reported for bacteria24,27,30–32 (also see Table S1 in Supporting Information for a summary of the banding assignments).] Differences in spectra were observed for bound EPS extracted from cells harvested at different growth phases and media (Figures 2 and 3). Amide I (1645 cm-1) and II bands (1450 cm-1) arising mainly from proteins were relatively low for bound EPS extracted from cells cultivated in LB and harvested at the exponential phase, when compared to its stationary phase counterpart. The bands between 1200 and 950 cm-1 corresponding to polysaccharide groups for bound EPS extracted from cells cultivated in LBG at 6 h is lower compared to the bound EPS extracted from cells cultivated in LBG at 24 h (Figure 3). Interestingly, the spectra of bound EPS from cells cultivated with LBG (Figure 3) at both growth phases, displayed similar patterns to bound EPS extracted from cells cultivated in LB but harvested at just the stationary growth phase (24 h) (Figure 2). The shape of the spectra of bound EPS extracted from cells cultivated in LBG at the exponential and stationary phase did not change significantly, which is in contrast with bound EPS extracted from cells cultivated in LB at the exponential and stationary phase. At this stage, it is difficult to draw any correlation between the FTIR spectrum of the bound EPS and the total protein and carbohydrate concentrations found in Table 1, as the FTIR spectrum provides a relative comparison of functional groups rather than absolute quantification. Figure 4 shows the spectra of free EPS extracted from E. coli MG1655 cultivated in LB at exponential (6 h) and stationary (24 h) phase. The relative absorbance of the bands between 1650 and 900 cm-1 increased as cells progress from the exponential to stationary phase. Figure 5 shows the spectra of free EPS extracted from E. coli MG1655 cultivated in LBG at different growth phases. The relative intensity of the peaks in the proteins and polysaccharides region also increased with increasing growth phases. The ratio of amine I and II also varied between free EPS extracted from E. coli cultivated in LBG at different growth phases, suggesting that the quantity and types of proteins differs. This is investigated further using proteomic analysis. One major difference between bound and free EPS is the presence or absence of a peak at ∼1545 cm-1 corresponding

Eboigbodin and Biggs

Figure 4. FTIR spectra of free EPS extracted from E. coli MG1655 cultivated with Luria–Bertani (LB) and extracted during exponential (6 h) and stationary phase (24 h).

Figure 5. FTIR spectra of free EPS extracted from E. coli MG1655 cultivated with Luria–Bertani with 0.5 w/v (%) glucose (LBG) and extracted during exponential (6 h) and stationary phase (24 h).

to the bending of N-H amides. This peak is predominately low or absent in bound EPS (Figure 2 and 3) in contrast to free EPS (Figures 4 and 5). The band at ∼1240 cm-1, corresponding to the asymmetric stretching PdO of phosphodiester backbone of nucleic acids is present in all spectra of free EPS (Figures 4 and 5) but is absent in spectra from bound EPS (Figures 2 and 3). These findings suggest that the content of bound EPS is predominantly dominated by proteins and carbohydrates. However, free EPS, while also dominated by proteins and carbohydrates, contains nucleic acids. The presence of nucleic acid in EPS has also been seen in mixed and pure culture systems.4–7 Identification of Proteins in the Free EPS. Figure 6 shows the SDS-PAGE profiles of proteins present in the free EPS from E. coli MG1655 cultivated in both LB and LBG and harvested during the exponential and stationary growth phases. Proteins were found to be distributed over a wide range of molecular weights (10–70 kDa). SDS-PAGE revealed differences in the protein profiles from free EPS harvested at different growth phases and from the two types of media. This techniques correlates well with the major differences in protein patterns already observed using the FTIR. More distinct protein bands were observed in free EPS extracted from stationary phase cells, cultivated in LB compared with free EPS from exponential phase in LB, and from free EPS from both growth phases in LBG. On the basis of the mass spectrometry data and analysis, a total of 503 proteins were identified across all the free EPS extracts from cells cultivated in LB and LBG at both exponential and stationary growth phases. A total of 79 and 175 proteins were identified from the free EPS extracted from cells harvested


Figure 6. One-dimensional SDS-PAGE gel of proteins from the free EPS extracted from E. coli MG1655 cells harvested at different growth phases in Luria–Bertani (LB) and Luria–Bertani with 0.5% (w/v) glucose (LBG). Lane 1: proteins from free EPS extract from cells grown in LB, harvested at 6 h. Lane 2: proteins from free EPS extract from cells grown in LBG, harvested at 6 h. Lane 3: proteins from free EPS extract from cells grown in LB, harvested at 24 h. Lane 4: proteins from free EPS extract from cells grown in LBG, harvested at 24 h. Lane 5: MW protein marker.

during the exponential (6 h) and stationary phase (24 h), cultivated in LB, respectively. For E. coli cells cultivated in LBG, a total of 117 and 132 proteins were identified from free EPS extracted from cells harvested at exponential (6 h) and stationary phase (24 h), respectively. Within these 503 proteins, 40 proteins were found in all growth conditions (The master list of proteins identified in free EPS extracted from E. coli MG1655 under all conditions is available in the Supporting Information). E. coli proteins identified in the free EPS in all growth conditions were categorized according to their biological function using the Encyclopedia of Escherichia coli K-12 Genes and Metabolism (http://ecocyc.org/) and the InterPro (http:// www.ebi.ac.uk/interpro/index.html) databases (Figure 7). Proteins with functionality related to amino acid metabolism, carbohydrate metabolism, cell wall and membrane, cellular processes, and translation, were among the highest proteins identified in the free EPS extracted from E. coli MG1655 under all growth and media conditions. Other proteins identified include proteins involved in energy production, hypothetical proteins, lipid metabolism, nucleotide and nucleoside metabolism, transport and secretion, and transcription.44,45 Several enzymes involved in glycolytic, pentose phosphate, and gluconeogenesis pathways, which are involved in carbohydrate metabolism were also found in the free EPS proteome of E. coli MG1655 cultivated in all growth phases and media. Enzymes involved in glycolytic pathway such as glyceraldehyde-3-phosphate dehydrogenase, phosphoglycerate kinase, triosephosphate isomerase, phosphofructokinase, phosphoglyceromutase, and phosphopyruvate hydratase (enolase) were found. Enzymes involved in amino acid metabolism such as aspartate aminotransferase and L-threonine 3-dehydrogenase were also identified. Ribosomal proteins involved in protein synthesis such as 30S and 50S ribosomal protein were also found in the free EPS of E. coli under all conditions.


Figure 7. Distribution of proteins identified in free EPS from E. coli MG1655 extracted at different growth phases, cultivated in Luria–Bertani (LB) and LB with 0.5 w/v (%) glucose (LBG) according to their functionality groups.

As anticipated, the outer membrane and related proteins were also found in abundance in the free EPS of E. coli MG1655. These include outer membrane porin protein A (OmpA), outer membrane protein X (OmpX), outer membrane protein W (OmpW), and outer membrane porin protein C (OmpC), which were all detected in all growth phases and media. The presence of outer membrane proteins in the extracellular proteome of E. coli have been suggested to be either due to cell lysis43,46 or shedding of membrane macromolecules during cell growth.43 Outer membrane porin proteins are involved in transport of sugars and other solutes and ions in and out of the membrane, and as such, changes in environmental condition may lead to the expression of different outer membrane proteins.47 Hence the varied number of outer membrane proteins identified in all growth conditions used in this study is not surprising. The number of outer membrane proteins and related macromolecules were higher for cells cultivated in LBG than LB regardless of the growth phase. This may be due to the expression of additional outer membrane porins required for glucose transport.48,49 Proteins involved in various cellular processes such as adaptation of specific conditions, cell protection, motility, and cell-to-cell communication were also identified in the free EPS under all growth conditions. The DNA protection protein (DPS), which binds to DNA to protect it from oxidative damage, was also identified under all conditions of study. DPS have been reported to be hyperexpressed during biofilm formation, prolonged starvation, and stationary phase.50 Proteins involved in maintaining protein folding before they are transported across the cytoplasm were also identified; these include heat shock proteins Hsp90, chaperone Hsp60, molecular chaperone Hsp70, and trigger factor. The proteins, chaperone Hsp60, molecular chaperone Hsp70, and trigger factor were identified from all conditions of study, while heat shock proteins Hsp90 was only identified in stationary growth phase (both in 24hrLB and 24hrLBG). Flagella hook-associated protein (flgK) was only found in the free EPS for stationary phase E. coli cells cultivated in LB. In E. coli, one of the mechanisms of translocation of proteins produced in cytoplasm to the outer membrane or extracellular environment is a via the general secretory (SecB) pathway. SecB, which also acts to prevent misfolding of proteins, binds



Table 2. Characterization of the Proteins Identified from the Free EPS Extracted from E. coli MG1655 at different growth phases, cultivated in Luria–Bertani (LB) and LB with 0.5 w/v (%) Glucose (LBG) 6 h LB, 6 h LBG, 24 h LB, 24 h LBG, % of totalb % of totalb % of totalb % of totalb locationa cytoplasmic cytoplasmic membrane extracellular outer membrane periplasmic unknown

50.6 1.3

59.0 0.9

65.7 1.7

52.3 0.8

0.0 8.9 3.8 35.4

0.0 12.8 1.7 25.6

0.0 5.7 3.4 23.4

0.0 10.6 6.1 30.3

GRAVYcg+0.3 lipoproteinsd pIe MW (kDa)f

1 4d 4.6–11.5 57–100

1 4d 4.2–11.5 57–104

0 2d 4.3–11.5 57–151

0 3d 4.3–11.5 57–151

a Cellular location was predicted using PSORTb v.2.0 software (http://www.psort.org/psortb/). b Percentage of total proteins observed. c Grand average hydropathy (GRAVY) values were calculated using the ProtParam tool at http://us.expasy.org/tools/protparam.html. d Lipoprotein signals were obtained using the LipoP tool at www.cbs.dtu.dk/services/LipoP/. e The isoelectric point (pI) range, i.e., the pH at which the proteins have no net electrical charge. f The molecular weight (MW) range.

to the N-terminal signal sequence of newly synthesized precursor polypeptides(preproteins) to be translocated and targets these proteins to the Sec apparatus.51,52 A major protein involved in the general secretory pathway, molecular chaperone SecB, was identified in the proteome of free EPS from E. coli MG1655 in all growth conditions. Oligopeptide transporter protein (oppA), which facilitates the transport of oligopeptide into and out of E. coli cells or between E. coli cells, was also identified in the free EPS proteome under all conditions of study. It is well-known that cell-to-cell communication via a quorum sensing in bacteria regulates several genes during cell growth and plays an important role in biofilm formation.53,54 In this study, the quorum sensing protein (autoinducer-2, LuxS) was identified in only the proteome of the exponential phase of free EPS from E. coli MG1655 cultivated in LBG (6hrLBG). E. coli secretes quorum sensing molecules (AI-2) into the media during exponential phase growth, and it is internalized by the cells immediately after the exponential phase when the cells are cultivated in LB. However, in E. coli cells cultivated in LB supplemented with glucose, AI-2 remains in the media even after the exponential phase.55 This is due to the fact that glucose inhibits the uptake of AI-2 quorum sensing autoinducer (AI-2) via a decrease of cyclic adenosine monophosphate (cAMP) concentration.56 Once the glucose in the media diminishes, AI-2 can then be internalized by the cell. Hence this may account for why AI-2 was only identified in the proteome of free EPS from E. coli cells harvested at 6 h in LBG. Other key proteins in E. coli were also identified, most of which were not unique to all growth conditions used in this study (see Supporting Information). For example, the protein involved in flagella biosynthesis, hook-filament junction protein 1, was only identified from the proteome of free EPS of E. coli MG1655 cultivated in LB, which was harvested at the stationary phase (24 h LBG). Proteins identified from free EPS extracted from E. coli MG1655 were further characterized based on their location and hydrophobicities. Table 2 shows the location of proteins identified determined by PSORTb as well the indication of the protein hydrophobicities determined from the GRAVY index. Only one protein in the free EPS, oxaloacetate decarboxylase,

was found to have a GRAVY index above or equal to +0.3 (0.322) and regarded as hydrophobic. The protein, oxaloacetate decarboxylase, was only identified in the free EPS proteome of E. coli MG1655 harvested at the exponential phase (6 h), cultivated in either LB or LBG. All other proteins identified were found to be below the +0.3 GRAVY index value. The PSORTb program was used to predict the cellular location of identified proteins under the different growth conditions used in this study. The proteins identified in the free EPS were predominantly cytoplasmic proteins for all conditions (Table 2). A higher percentage of outer membrane proteins was seen in the free EPS proteome of E. coli MG1655 cultivated in LBG than in the free EPS proteome of E. coli MG1655 cultivated in LB. The increase in outer membrane proteins observed for cells cultivated in LBG may be due to the presence of outer membrane proteins required for glucose transport. Furthermore, the percentage of periplasmic proteins was found to be higher in the free EPS proteome of E. coli MG1655 cultivated in LBG, harvested at the stationary phase (6.1%), than in cells harvested at the exponential phase (1.7%) in the same growth medium. Conversely, the percentage of periplasmic proteins was found to be relatively constant in the free EPS proteome of E. coli MG1655 cultivated in LB in both the exponential and stationary phases. The percentage of proteins located in the cytoplasmic membrane was found to be slightly higher in the free EPS of E. coli MG1655, cultivated in LB, than in LBG. However, the PSORTb program is unable to predict the locations of some proteins, hence a significant percentage of the proteins in the free EPS of E. coli MG1655 were designated as unknown. Another drawback of the PSORTb program is that it is unable to detect membrane lipoproteins. Therefore the program LipoP v1.0 was used to address this limitation. However, only small numbers of lipoproteins were identified in the free EPS proteome of E. coli MG1655. Role of EPS in Aggregation of E. coli MG1655. Effect of Bound EPS on the Aggregation Capability of E. coli MG1655. The role of bound EPS on the aggregation ability of E. coli MG1655 was investigated by carrying out the aggregation assay on E. coli MG1655 before and after the extraction of bound EPS. The percentage aggregation of E. coli MG1655 cultivated in both LB and LBG harvested at the exponential (6 h) and stationary (24 h) phase before and after the extraction of bound EPS is shown in Figure 8. The results revealed that, under the same conditions, i.e., the same growth phase or the same media, the presence or absence of bound EPS did not affect the % aggregation of E. coli MG1655. This may be due to the very low amount of bound EPS produced during the conditions of the experiments. Furthermore, the primary role of bound EPS may not be in the initial cell-to-cell adhesion but in holding cells together after adhesion.57 It is also important to note here that no free EPS was present and hence the aggregation is largely governed by the cell’s themselves. The difference in aggregation capability of E. coli MG1655 harvested at different growth phase and media is therefore controlled by its surface chemistry.24,58 In a previous study, it was observed that stationary phase cells cultivated in LB had a less negative charge than cells at the exponential phase and as such are more likely to aggregate.58 However, in Figure 8, it appears that for cells cultivated on LBG, the effect of glucose on aggregation was more dominant than a growth phase surface charge effect, and as such, cells cultivated in LBG harvested at the exponential displayed a higher aggregation ability than cells cultivated in LBG, which were harvested at



Figure 8. Percentage aggregation of E. coli MG1655 cultivated in Luria–Bertani (LB) and LB with 0.5 w/v (%) glucose (LBG) harvested at exponential (6 h) and stationary phase (24 h) before and after bound EPS extraction.

the stationary phase. A similar reduction in aggregation ability over the growth phase when glucose is initially present in the media was also seen in Eboigbodin et al.24 Aggregation in E. coli MG1655 by Free EPS. From a physiochemical point of view, bacteria can be considered to be negatively charged colloidal “particles”59,60 surrounded by bound and free polyelectrolyte, i.e., EPS. A suspension of nonbiological colloidal particles in the presence of nonadsorbing polymers have been observed to aggregate via a process known as depletion.61,62 The interaction is due to an imbalance in osmotic pressure when the nonadsorbing polymers are excluded from the region between particles, which sets up a net attractive potential between the particles enhancing the potential for aggregation. This approach can be used to describe biological aggregation, assuming that the EPS produced by the bacteria are nonadsorbing, e.g., free EPS. Eboigbodin et al.23 previously reported similar depletion interactions can occur in the aggregation of E. coli AB1157 by using a nonadsorbing polymer sodium polystyrene sulfonate (SPS). In this case, the level of interaction was dependent on the both the electrokinetic properties of the cell surface as dictated by the growth phase and the concentration of polymer added.55 Hence, we investigated this work further to show if the free EPS produced by E. coli MG1655 can also induce interaction between the cells via depletion, which will further explain the function of free EPS in bacteria aggregation. The aggregation ability of free EPS extracted from E. coli MG1655 cultivated in LB and harvested at exponential phase (6 h) and stationary phase (24 h) on different samples of E. coli MG1655 cells is shown in Figure 9. The results reveal that free EPS triggers the aggregation capability of E. coli MG1655 harvested at different growth phases. The addition of free EPS extracted from cells at exponential growth phase (6 h) or the stationary phase (24 h) to washed E. coli MG1655 harvested at the stationary phase (24 h) displayed a similar aggregation ability (compare Figure 9c and d). However, the addition of free EPS extracted from cells at exponential growth phase (6 h) or stationary phase (24 h) to cells harvested at the exponential phase (6 h) displayed different aggregation ability (compare Figure 9a and b). Cells harvested at exponential phase were found to display an increase in aggregation ability when free EPS, harvested at stationary phase, was added (Figure 9b) relative to the addition of free EPS extracted at 6 h (Figure 9a). However, it should be noted that

Figure 9. Percentage aggregation of E. coli MG1655 cultivated in Luria–Bertani (LB) at different growth phase by the addition of free EPS extracted from E. coli MG1655 cells cultivated in LB at different growth phases.

cells harvested at stationary phase (24 h) in LB displays a higher aggregation capability without the addition of free EPS than cells harvested after the exponential phase (6 h) (Figure 8). This may account for the lesser effect of free EPS on cells harvested at stationary phase. The aggregation ability of free EPS extracted from E. coli MG1655 cultivated in LBG harvested at exponential phase (6 h) and stationary phase (24 h) is shown in Figure 10. The addition of free EPS harvested in the exponential phase (6 h) from cells grown in LBG had a marked effect on aggregation ability of cells at exponential phase (Figure 10a) compared to stationary phase (Figure 10c). A similar effect was seen for free EPS harvested at 24 h when added to cells at the exponential and stationary phase (Figure 10b and d). However, it should also be noted that cultivating in LBG harvested at exponential phase (6 h) displays a higher aggregation capability, without addition of free EPS, at all times than cells harvested after the stationary phase (24 h) (Figure 10) and as such may also account for the lesser effect of free EPS on cells harvested at stationary phase.24 The results thus far show that free EPS extracted from E. coli MG1655 cultivated in LB and harvested at stationary growth phase displayed the highest ability to induce aggregation. This



Cells harvested at exponential phase were found to display an increase in aggregation ability when free EPS, harvested at stationary phase, was added. These findings suggest that free EPS can induce cell aggregation and their ability to induce aggregation may be due to its protein content. The composition of free EPS determines its ability to induce aggregation in E. coli, which in itself is controlled by the growth phase and media. EPS extracted from cultures at different growth phases, have a varying capacity to induce aggregation. Hence it is possible to force aggregation of E. coli cells with both inert and biologically produced polymers.

Figure 10. Percentage aggregation of E. coli MG1655 cultivated in Luria–Bertani with 0.5 w/v (%) glucose (LBG) by the addition of free EPS extracted from E. coli MG1655 cells cultured in LBG and harvested at different growth phases.

Acknowledgment. We thank the UK’s Biotechnology and Biological Sciences Research Council (BBSRC) for funding (BB/C505391/1), and the Engineering and Physical Sciences Research Council (EPSRC) for a studentship, an Advanced Research Fellowship (EP/E053556/1), and further funding (GR/ S72467/01(P)) that enabled access to the FTIR. We also acknowledge J. J. Ojeda for his expertise and technical support in the FTIR experiments, Professor P. Williams, University of Nottingham, for the E. coli MG1655 strain, and Professor P. C. Wright and Dr. A. P. L. Snijders for technical support and access to the mass spectrometry facilities Supporting Information Available. Summary of the band assignments from the literature (Table S1), a summary of the combination of the different samples of free EPS and E. coli cells harvested at different times and grown in different media used in the aggregation studies (Table S2), summary of the major proteins identified from the free EPS of E. coli MG1655 that are common to all growth conditions (Table S3), and the master list of proteins identified. This material is available free of charge via the Internet at http://pubs.acs.org.

References and Notes

Figure 11. Percentage aggregation of E. coli MG1655 cultivated in Luria–Bertani with 0.5 w/v (%) glucose (LBG) at 6 and 24 h by the addition of free EPS extracted from E. coli MG1655 cells harvested at 24 h from Luria–Bertani (LB) media.

was investigated further by the addition of free EPS extracted from cells at the stationary growth phase (24 h) cultivated in LB to cells cultivated in LBG at both exponential and stationary phase (Figure 11). Interestingly, the aggregation ability of cells cultivated in LBG for both exponential and stationary growth phase increased by about 30% (compared with Figure 10) when free EPS extracted from E. coli cells cultivated in LB at stationary growth phase (24 h LB) was added.

Conclusions E. coli MG1655 was found to produce a small amount of bound EPS and a relatively large amount of free EPS under the conditions used for this study. The protein content of free EPS was found to significantly increase as E. coli MG1655 cells progress from exponential to stationary phase. A total of 503 proteins were identified from the free EPS extracted from cells harvested during the exponential (6 h) and stationary phase (24 h), cultivated in LB and LBG, respectively.

(1) Burdman, S.; Jurkevitch, E.; Soria-Diaz, M. E.; Serrano, A. M. G.; Okon, Y. FEMS Microbiol. Lett. 2000, 189, 259–264. (2) Wingender, J.; Neu, T. R.; Flemming, H. C. Microbial Extracellular Polymeric Substances; Springer: Berlin, 1999. (3) Bramhachari, P. V.; Dubey, S. K. Lett. Appl. Microbiol. 2006, 43, 571–577. (4) John, A.; Nielsen, P. H. Water Sci. Technol. 1995, 32, 157–164. (5) Sheng, G.-P.; Yu, H.-Q.; Yu, Z. Appl. Microbiol. Biotechnol. 2005, 67, 125–130. (6) Liu, H.; Fang, H. H. P. J. Biotechnol. 2002, 95, 249–256. (7) Frolund, B.; Palmgren, R.; Keiding, K.; Nielsen, P. H. Water Res. 1996, 30, 1749–1758. (8) Kristina, D.; Rinker, R. M. K Biotechnol. Bioeng. 2000, 69, 537– 547. (9) Macedo, M. G.; Lacroix, C.; Champagne, C. P. Biotechnol. Prog. 2002, 18, 167–173. (10) Bayer, A. S.; Eftekhar, F.; Tu, J.; Nast, C. C.; Speert, D. P. Infect. Immun. 1990, 58, 1344–1349. (11) Bayer, A. S.; O’Brien, T.; Norman, D. C.; Nast, C. C. J. Antimicrob. Chemother. 1989, 23, 21–35. (12) Degeest, B.; De Vuyst, L. Appl. EnViron. Microbiol. 1999, 65, 2863– 2870. (13) Dunne, W. M., Jr.; Buckmire, F. L Microbios 1985, 43, 193–216. (14) Zhang, X.; Bishop, P. L.; Kinkle, B. K. Water Sci. Technol. 1999, 39, 211–218. (15) Omoike, A.; Chorover, J. Biomacromolecules 2004, 5, 1219–1230. (16) Flemming, H. C.; Wingender, J. Water Sci. Technol. 2001, 43, 1–8. (17) Wingender, J.; Flemming, H.-C. In Biotechnology; Winter, J., Ed.; Wiley-VCH: New York, 1999; Vol. 8, pp 63–86. (18) Kreft, J. U.; Wimpenny, J. W. Water Sci. Technol. 2001, 43, 135– 141. (19) Sutherland, I. W. Water Sci. Technol. 2001, 43, 77–86. (20) Jucker, B. A.; Harms, H.; Hug, S. J.; Zehnder, A. J. B. Colloids Surf., B 1997, 9, 331–343. (21) Jucker, B. A.; Harms, H.; Zehnder, A. J. B. Colloids Surf., B 1998, 11, 33–45.

Extracellular Polymeric Substances from E. coli (22) Tsuneda, S.; Jung, J.; Hayashi, H.; Aikawa, H.; Hirata, A.; Sasaki, H. Colloids Surf., B 2003, 29, 181–188. (23) Eboigbodin, K. E.; Newton, J. R.; Routh, A. F.; Biggs, C. A. Langmuir 2005, 21, 12315–12319. (24) Eboigbodin, K. E.; Ojeda, J. J.; Biggs, C. A. Langmuir 2007, 23, 6691– 6697. (25) Lowry, O. H.; Rosebrough, N. J.; Farr, A. L.; Randall, R. J. J. Biol. Chem. 1951, 193, 265–275. (26) Gaudy, A. F. Ind. Water Wastes 1962, 7, 17–22. (27) Schmitt, J.; Flemming, H.-C. Int. Biodeterior. Biodegrad. 1998, 41, 1–11. (28) Gennis, R. B. FEBS Lett. 2003, 555, 2–7. (29) Schuster, K. C.; Mertens, F.; Gapes, J. R. Vib. Spectrosc. 1999, 19, 467–477. (30) Dittrich, M.; Sibler, S. J. Colloid Interface Sci. 2005, 286, 487–495. (31) Lin, M.; Al-Holy, M.; Chang, S.-S.; Huang, Y.; Cavinato, A. G.; Kang, D.-H.; Rasco, B. A. Int. J. Food Microbiol. 2005, 105, 369–376. (32) Maquelin, K.; Kirschner, C.; Choo-Smith, L.-P.; van den Braak, N.; Endtz, H. P.; Naumann, D.; Puppels, G. J. J. Microbiol. Methods 2002, 51, 255–271. (33) Laemmli, U. K. Nature 1970, 227, 680–685. (34) Chee Sian Gan, K. F.; Phillip, R.; Wright, C. Proteomics 2005, 5, 2468–2478. (35) David, N.; Perkins, D. J.; David, C. P.; Creasy, John, M.; Cottrell, S. Electrophoresis 1999, 20, 3551–3567. (36) Kyte, J.; Doolittle, R. F. J. Mol. Biol. 1982, 157, 105–132. (37) Gasteiger E.; Hoogland C.; Gattiker A.; Duvaud S.; Wilkins M. R.; Appel, R. D.; BairochA. In The Proteomics Protocols Handbook; Walker, J. M., Ed.; Humana Press.: Totowa, NJ, 2005, pp 571–607. (38) Mitra, S. K.; Gantt, J. A.; Ruby, J. F.; Clouse, S. D.; Goshe, M. B. J. Proteome Res. 2007, 6, 1933–1950. (39) Gardy, J. L.; Laird, M. R.; Chen, F.; Rey, S.; Walsh, C. J.; Ester, M.; Brinkman, F. S. L. Bioinformatics 2005, 21, 617–623. (40) Juncker, A. S.; Willenbrock, H.; von Heijne, G.; Brunak, S.; Nielsen, H.; Krogh, A. Protein Sci. 2003, 12, 1652–1662. (41) Walker, S. L.; Hill, J. E.; Redman, J. A.; Elimelech, M. Appl. EnViron. Microbiol. 2005, 71, 3093–3099. (42) Razatos, A.; Ong, Y.-L.; Sharma, M. M.; Georgiou, G. Proc. Natl. Acad. Sci. U.S.A. 1998, 95, 11059–11064.

Biomacromolecules, Vol. 9, No. 2, 2008 695 (43) Han, L.; Enfors, S.-O.; Ha¨ggstro¨m, L. Bioprocess Biosyst. Eng. 2003, 25, 205–212. (44) Oda, K.; Kakizono, D.; Yamada, O.; Iefuji, H.; Akita, O.; Iwashita, K. Appl. EnViron. Microbiol. 2006, 72, 3448–3457. (45) Nandakumar, M. P.; Cheung, A.; Marten, M. R. J. Proteome Res. 2006, 5, 1155–1161. (46) Li, M.; Rosenshine, I.; Tung, S. L.; Wang, X. H.; Friedberg, D.; Hew, C. L.; Leung, K. Y. Appl. EnViron. Microbiol. 2004, 70, 5274–5282. (47) Nikaido, H. Microbiol. Mol. Biol. ReV. 2003, 67, 593–656. (48) Thomas, A. D.; Booth, I. R. J. Gen. Microbiol. 1992, 138, 1829– 1835. (49) Sato, M.; Machida, K.; Arikado, E.; Saito, H.; Kakegawa, T.; Kobayashi, H. Appl. EnViron. Microbiol. 2000, 66, 943–947. (50) Martinez, A.; Kolter, R. J. Bacteriol. 1997, 179, 5188–5194. (51) Scott, J. R.; Barnett, T. C. Annu. ReV. Microbiol. 2006, 60, 397–423. (52) Fekkes, P.; de Wit, J. G.; van der Wolk, J. P. W.; Kimsey, H. H.; Kumamoto, C. A.; Driessen, A. J. M. Mol. Microbiol. 1998, 29, 1179– 1190. (53) Xu, L.; Li, H.; Vuong, C.; Vadyvaloo, V.; Wang, J.; Yao, Y.; Otto, M.; Gao, Q. Infect. Immun. 2006, 74, 488–496. (54) Waters, C. M.; Bassler, B. L. Annu. ReV. Cell DeV. Biol. 2005, 21, 319–346. (55) Eboigbodin, K. E.; Newton, J. R. A.; Routh, A. F.; Biggs, C. A. Appl. Microbiol. Biotechnol. 2006, 73, 669–675. (56) Wang, L.; Hashimoto, Y.; Tsao, C.-Y.; Valdes, J. J.; Bentley, W. E. J. Bacteriol. 2005, 187, 2066–2076. (57) Danese, P. N.; Pratt, L. A.; Kolter, R. J. Bacteriol. 2000, 182, 3593– 3596. (58) Eboigbodin, K. E.; Newton, J.; Routh, A.; Biggs, C. Appl. Microbiol. Biotechnol. 2006, 73, 669–675. (59) Sonohara, R.; Muramatsu, N.; Ohshima, H.; Kondo, T. Biophys. Chem. 1995, 55, 273–277. (60) van Loosdrecht, M. C.; Lyklema, J.; Norde, W.; Schraa, G.; Zehnder, A. J. Appl. EnViron. Microbiol. 1987, 53, 1898–1901. (61) Asakura, S.; Oosawa, F. J. Chem. Phys. 1954, 22, 1255–1256. (62) Asakura, S.; Oosawa, F. I. J. Polym. Sci. 1958, 183–192.

BM701043C

Characterization of the Extracellular Polymeric Substances Produced

Recommend Documents