Contrasting effects of extracellular polymeric substances on the surface characteristics of bacterial pathogens and cell attachment to soil particles

Chemical Geology 410 (2015) 79–88

Contents lists available at ScienceDirect

Chemical Geology journal homepage: www.elsevier.com/locate/chemgeo

Contrasting effects of extracellular polymeric substances on the surface characteristics of bacterial pathogens and cell attachment to soil particles Wenqiang Zhao a,c, Sharon L. Walker b, Qiaoyun Huang a, Peng Cai a,⁎ a

State Key Laboratory of Agricultural Microbiology, College of Resources and Environment, Huazhong Agricultural University, Wuhan 430070, China Department of Chemical and Environmental Engineering, University of California, Riverside, CA 92521, USA Key Laboratory of Mountain Ecological Restoration and Bioresource Utilization & Ecological Restoration Biodiversity Conservation Key Laboratory of Sichuan Province, Chengdu Institute of Biology, Chinese Academy of Sciences, Chengdu 610041, China b c

a r t i c l e

i n f o

Article history: Received 27 December 2014 Received in revised form 10 June 2015 Accepted 11 June 2015 Available online 14 June 2015 Editor: J. Fein Keywords: Bacterial pathogen EPS Attachment Soil Zeta potential ATR-FTIR

a b s t r a c t Extracellular polymeric substances (EPSs) have been confirmed to affect bacterial surface properties and cell attachment to minerals. However, no systematic work has been done to clarify the contrasting roles of EPS in cell attachment to natural soil between different pathogenic strains. This study compared the different surface properties and attachment behaviors of two bacterial pathogens (with full or partial EPS) using attenuated total reflection Fourier transform infrared (ATR-FTIR) spectroscopy, potentiometric titration, zeta potential, hydrophobicity analysis, DLVO theory, and attachment tests. Cation exchange resin (CER) was employed to remove the EPS on Streptococcus suis and Escherichia coli such that the contribution of EPS to cell attachment to soil could be determined. ATR-FTIR confirmed the binding sites differed between S. suis and E. coli EPS. Notably, after partial EPS removal the absorption bands of S. suis between 1800 cm−1 and 800 cm−1 shifted or disappeared, whereas the lack of EPS did not affect the infrared absorption peaks for E. coli. This result suggests the overall surface site types within the E. coli EPS were similar to the residual EPS fractions or cell wall. The partial removal of EPS also changed the proton-active site concentrations of both cell types, and reduced the bacterial surface charge densities by 7%–17%. The negative charges on bacterial surfaces followed the order of full EPSS. suis b partial EPS-S. suis b partial EPS-E. coli b full EPS-E. coli (ionic strength 1–100 mM; pH 5.6–5.8). With the removal of EPS, the average hydrophobicities of S. suis increased by 5% while those of E. coli decreased by 11%. EPS removal inhibited the attachment of S. suis to soil particles (b 2 mm) but enhanced E. coli attachment across the IS range of 1–100 mM, which was attributed to the alteration in electrostatic repulsion. At IS 60–100 mM, a sudden reduction in the attachment was observed only for full EPS-S. suis, which could be ascribed to the steric hindrance derived from EPS. However, full EPS-E. coli and partial EPS-E. coli showed similar increasing attachment trends at IS 1–100 mM. This study clearly showed the distinct contribution of EPS to pathogen attachment to soil as a function of cell type and EPS present. © 2015 Elsevier B.V. All rights reserved.

1. Introduction Livestock manure and wastes from animal feeding operations are increasingly applied to lands as agricultural fertilizers for silage, grazing, or crop production (Guber et al., 2007). These biosolids may serve as a source of pathogens that contaminate soil, fresh products, surface/ ground water and water supply systems (Venglovsky et al., 2009). Pathogens can survive for extended periods after they are spread to agricultural land with manure (Nyberg et al., 2010; Toth et al., 2013). The retention of pathogens in upper soil layers could influence their metabolic activities (Rong et al., 2007), survival time (Franz et al., 2008), as well as transport process in surface runoff or saturated soil (Keller and Auset, 2007). A wide variety of bacterial pathogen groups have been ⁎ Corresponding author. E-mail address: [email protected] (P. Cai).

http://dx.doi.org/10.1016/j.chemgeo.2015.06.013 0009-2541/© 2015 Elsevier B.V. All rights reserved.

found in animal feces, such as fecal coliforms, streptococci, Salmonella spp., Campylobacter spp. and Listeria monocytogenes (Unc and Goss, 2004). Over the past decades, the emergence of serious disease outbreaks has been associated with consumption of fecal contaminated food or water, and in many cases animal manures that released into soil environments were identified as the likely source of these outbreaks (Gerba and Smith, 2005; Abit et al., 2012). Therefore, an understanding of the fate of bacterial pathogens in soils – notably the degree to which they attach to the soil – is needed to assess their availability and potential risk to public health. A number of physical and chemical factors have been proposed to govern bacteria–solid surface interactions (Stevik et al., 2004), including particle size (Soupir et al., 2010), surface coating (Li and Logan, 2004), charge property (Jacobs et al., 2007), hydrophobic effect (Shephard et al., 2010), and electrolyte composition (Hong and Brown, 2008). Additionally, biological factors such as bacterial cell

80

W. Zhao et al. / Chemical Geology 410 (2015) 79–88

type (Foppen et al., 2010), extracellular polymeric substances (EPSs) (Kim et al., 2009b), and lipopolysaccharides (Walker et al., 2004) have also been shown to have considerable influence on bacterial attachment. Among the various components of bacterial cell surfaces, EPS is the major heterogeneous component composed dominantly of polysaccharides and proteins, with nucleic acids and lipids as minor constituents (Eboigbodin and Biggs, 2008; Long et al., 2009; Cao et al., 2011). EPS contains various acidic functionalities (carboxyl, phosphoryl, amide, amino, hydroxyl) that ionize in response to changes in solution chemistry, and is of particular importance which affects cell surface characteristics and attachment to solid substrates (Gong et al., 2009; Karunakaran and Biggs, 2011; Mukherjee et al., 2012). Interactions between bacteria and solid surfaces are typically considered as an abiotic physicochemical process that can be approximated by the balance of electrostatic, van der Waals and hydrophobic forces (Hermansson, 1999). It has been shown that bacteria–mineral interactions were well predicted by the classic Derjaguin–Landau–Verwey–Overbeek (DLVO) theory (Hori and Matsumoto, 2010; Zhao et al., 2014). In some cases, the DLVO theory is not always valid to predict cell-surface interactions due to the presence of complex polymer layers extending into the liquid medium (Tsuneda et al., 2003; Kim et al., 2009a). Depending on solution ionic strength, the presence of EPS may alter the extent of microbial attachment due to steric interactions which are not incorporated in DLVO theory (Chen and Walker, 2007; Liu et al., 2010). The influence of EPS on cell interactions with solid surfaces has been drawing increasing attention (Tong et al., 2010). Several studies have investigated the attachment of cells with and without EPS coating to wellcharacterized minerals under controlled conditions via batch attachment tests, parallel plate flow chamber, radial stagnation point flow (RSPF) system, and packed-bed columns. For instance, by comparing the attachment efficiencies of untreated and proteinase K treated Escherichia coli O157:H7 on quartz sand in a batch system, Kim et al. (2009b) reported greater cell attachment occurred for cells with EPSpartially removed when IS ≥ 10 mM, suggesting the presence of EPS hindered cell attachment. Hong et al. (2013) found that via the treatment of cation exchange resin (CER), the reduction in EPS had no apparent effects on Bacillus subtilis adhesion on montmorillonite, kaolinite and goethite when the wet bacteria/mineral mass ratio was less than 0.4. However, at higher cell concentrations, the removal of EPS reduced bacterial adhesion to kaolinite and montmorillonite, and enhanced its adhesion to goethite. Taylor et al. (2014) examined the deposition of E. coli O157:H7 (full or partial coating EPS) on bare and hematite coated quartz in a parallel plate chamber. Their results indicated the ability of EPS to facilitate interactions between cells and surfaces. Kuznar and Elimelech (2006) investigated the attachment kinetics of untreated and digestive enzyme-treated Cryptosporidium parvum oocyst to quartz surfaces using a RSPF system. It showed that the removal of surface macromolecules increased attachment efficiencies. Gargiulo et al. (2007) observed that the proteolitic enzymatic treatment of Rhodococcus rhodochrous significantly decreased the amount of attached cells in the silica sand column, while straining behavior for treated and untreated bacteria was quite similar. Despite the efforts discussed above, it should be noted that the literature mainly focused on homogeneous mineral surfaces; whereas the contribution of EPS in bacterial attachment to more heterogeneous natural soil particle has not been investigated. Furthermore, the majority of studies focused on a single microbe. Overall, very little work has been conducted to comprehensively interpret the EPS-induced dissimilar attachment phenomena that may exist in different pathogens. Hence, this study was developed to address this gap in knowledge. The objective of this paper is to explore the role of EPS in surface characteristics of different pathogens and their subsequent attachment to natural, heterogeneous soil. Two model bacterial pathogens – Streptococcus suis (Gram-positive) and E. coli (Gram-negative) – were selected based on their differing outer membrane properties. The CER extraction method was employed to remove EPS associated with the outer surfaces

of cells such that the EPS-extracted cells could be compared with those with full EPS. Surface properties of full EPS- and partial EPS-bacteria were extensively characterized via attenuated total reflection Fourier transform infrared (ATR-FTIR) spectroscopy, potentiometric titration, zeta potential, and hydrophobicity analysis. In addition to attachment tests of full EPS/partial EPS-pathogens, DLVO theory was applied to interpret results and evaluate the contribution of EPS to cell attachment across the ionic strength (IS) range tested (1–100 mM). 2. Materials and methods 2.1. Bacterial pathogens and soil particles The two model organisms used in this study were Gram-positive pathogenic bacteria S. suis SC05 and Gram-negative E. coli WH09, both obtained from the State Key Laboratory of Agricultural Microbiology. These cell types were isolated from soils around a pig farm in Wuhan, Hubei Province, China (Zhao et al., 2014). The bacteria were grown and harvested according to the protocols described in the Supplementary data. After harvest, the cells were washed two additional times with sterilized distilled-deionized water (ddH2O) to remove all traces of the growth medium (Kim et al., 2010). Washed cells were subsequently resuspended in ddH2O at a concentration of 6.0 × 109 cells per mL and then divided into two portions. One portion of cell suspension was used as untreated bacteria (full EPS). The other portion of cell suspension was used to prepare partial EPS-bacteria via the employment of cation exchange resin (CER) technique (Long et al., 2009; Tong et al., 2010). The detailed CER treatment procedure is presented in the Supplementary data. This method has been reported as the most effective way to remove EPS from cell surfaces and most of the cells were intact (Aguilera et al., 2008; Tourney et al., 2008). Preliminary experiments were also conducted to test the viability of cells. Bacterial suspension (~108 cells mL−1) was stained with a dye solution consisting of 40 μL Live/Dead BacLight stain (L-7012, Molecular Probes, Eugene, OR) in 2 mL of 1–100 mM KNO3 for 15 min. Most of the cells (94.8%–98.0%) were confirmed to be viable (green color) by using fluorescence microscopy (IX-70, Olympus). A Yellow-Brown soil was collected from the top 20 cm of farmland in Wuhan, Hubei Province, China. After removing the organic residues and stones, the soil was air-dried, sieved to pass through 2-mm sieves, and autoclaved at 121 °C and 0.105 MPa for 30 min (repeated for 3 times) (Guber et al., 2005). No microorganisms were detected in the sterilized soil through plate counting procedures on tryptone soy agar (TSA). The soil particles were then oven-dried at 60 °C prior to the attachment experiments. 2.2. ATR-FTIR spectroscopy To compare the functional groups on full EPS- and partial EPS-cell surfaces and ensure the effectiveness of the EPS removal, infrared spectra of bacteria were determined by ATR-FTIR spectroscopy. This technique permits in situ investigation of functional groups on bacterial surfaces in water (Parikh and Chorover, 2006; Ojeda et al., 2008). The washed bacterial suspensions in ddH2O were centrifuged at 8000 ×g for 10 min, and the cell pellets were spread on the ZnSe internal reflecting element (IRE) crystal to obtain their spectra. The ATR-FTIR spectra for the full EPS/partial EPS-bacteria were collected over the scan range of 3500 cm−1 to 800 cm−1 (Vertex 70, Bruker Optics, Germany) (Parikh and Chorover, 2006). Each sample was scanned 256 times with a resolution of 4 cm−1, and the obtained spectra were baseline corrected. The FTIR spectra of both full EPS- and partial EPS-bacteria were normalized to the height of the peak at 1539 cm−1 for S. suis and 1545 cm−1 for E. coli (amide II), respectively. Following the same protocol, the supernatants of bacterial suspensions were also scanned as the background spectra (Ueshima et al., 2008).

W. Zhao et al. / Chemical Geology 410 (2015) 79–88

2.3. Potentiometric titration Potentiometric titration was performed by an automatic potentiometric titrator (Metrohm titrator 836, Metrohm, Switzerland) to analyze the acid–base properties of full EPS- and partial EPS-pathogen surfaces (Fein et al., 2005; Ueshima et al., 2008; Kenny and Fein, 2011). Bacterial suspensions with concentrations of 6.0 × 109 cells per mL in 0.01 M KCl were titrated using 0.09 M HCl and 0.09 M KOH solutions under a N2 atmosphere at 25 °C. Each suspension was continuously stirred with a small magnetic stir bar during the titration. A known amount of HCl was added at the beginning of the experiment to lower the pH to 2.5. The cell suspension was equilibrated for 40 min and titrated to pH 10.0 with KOH (Fang et al., 2010). The surface charge density was calculated from the amount of base consumed by suspended cells during the titration between pH 2.5 and pH 10.0, accounting for the cell surface area (Chen and Walker, 2007). At each titration step, a pH electrode stability of 0.1 mV S−1 was attained before the addition of the next aliquot of titrant. Blank titration in the same electrolyte solution (0.01 M KCl) was performed and each experiment was conducted in triplicate. The bacterial surface site types were acidic and discrete. The deprotonation process can be generally expressed by the following reaction (Borrok and Fein, 2004): R–AH
→R–A– þ H þ

ð1Þ

where R is the bacterium to which the site type A is attached. The equilibrium constant Ka for reaction (1) can be given by: K a ¼ ½R–A– aHþ = ½R–AH


ð2Þ

where [R − A−] and [R − AH°] stand for the concentrations of deprotonated and protonated sites, respectively, and aH+ represents the activity of protons in the bulk solution. Titration data were modeled using the software ProtoFit to obtain dissociation constants (pKa) and their corresponding site concentrations for proton-active chemical moieties on bacterial surfaces (Turner and Fein, 2006).

81

microbial growth and maintain bacterial viability (Gantzer et al., 2001; Guber et al., 2007). Initial experiments showed that 1 h was sufficient for the cell–soil interaction to reach a plateau. The mixtures were centrifuged at 100 ×g for 15 min, and the soil particles with any attached bacteria settled to the bottom of the tubes (Guber et al., 2007; Pachepsky et al., 2008). The calibration curves and the amount of un-attached cells in the supernatant were measured by UV–visible spectrophotometer (UV1102, Shanghai Tianmei Scientific Instrument Co., China) at a wavelength of 600 nm. Control tubes with only bacterial suspensions (or only soil suspensions) were processed in the same manner as the experimental tubes to correct for bacterial attachment to tube walls (or optical density of the soil supernatant) (Dai and Boll, 2003). The amount of attached bacteria was calculated from the difference between the amount applied and that recovered in the supernatant (Guber et al., 2005, 2007). 2.6. Application of DLVO theory DLVO theory was applied to calculate the interaction energy profiles existing between the pathogens and soil particles. Total interaction energies were quantified as the sum of van der Waals and electrostatic interactions, which were calculated by considering the system as sphereplate model (Haznedaroglu et al., 2009; Kim et al., 2010; Cai et al., 2013). Calculation details are given in the Supplementary data. 2.7. Statistical analysis All the potentiometric titration, zeta potential, hydrophobicity and attachment experiments were performed in triplicate. The data are presented as the mean ± standard deviation of the mean (x ± SDM). Statistical differences between mean values were analyzed using a Student's t test. Linear regression analysis (Origin 8.0, OriginLab Co., Northampton, USA) were applied to evaluate the statistical correlations among bacterial surface property, DLVO energy barrier and attachment data. P values greater than or equal to 0.05 suggest differences are statistically insignificant within a 95% confidence interval.

2.4. Zeta potential and hydrophobicity of bacteria and soil 3. Results and discussion Zeta potential and hydrophobicity of bacteria and soil across a range of IS (1–100 mM KCl) were measured and calculated according to the protocols described in the Supplementary data. The hydrodynamic diameters of bacteria were determined by dynamic light scattering in a Zetasizer (Nano ZS90, Malvern Instruments Ltd., UK). The specific surface areas (SSA) of bacteria were measured via methylene blue adsorption method (He and Tebo, 1998). Total EPS, sugar, and protein contents were assayed by ethanol precipitation, phenol–sulfuric acid, and bicinchoninic acid (Biosynthesis Co., Ltd., Beijing) method, respectively (Chen and Walker, 2007; Cao et al., 2011). The above cell properties are listed in Table S1. Basic soil properties, including soil pH (5.8), texture (clay 14.2%; silt 64.1%; sand 21.7%), SSA (26.8 m2 g−1), organic matter content (14.7 g kg−1) and cation exchange capacity (10.0 cmol kg−1), were obtained by a pH meter (FE20, Mettler Toledo, Switzerland), pipette method, nitrogen adsorption measurement (Autosorb-1, Quantachrome Instruments, USA), K2Cr2O7 digestion and NH4C2H3O2 displacement, respectively (Gee and Bauder, 1986; Evangelou and Blevins, 1988). 2.5. Bacterial attachment to soil particles Bacterial suspensions (contained 1.05 × 1010 cells) were mixed with 200 mg of soil particles in 20-mL centrifuge tubes. The bacteria-to-soil surface area ratios were 1:1.5 for S. suis and 1:3.6 for E. coli, respectively. KCl solution and ddH2O were added to reach a final volume of 15 mL (1, 5, 10, 20, 40, 60, 80 and 100 mM KCl). Cell and soil mixtures were stirred on a rotary shaker for 1 h at 150 rev min−1 and 8 °C to minimize

3.1. ATR-FTIR spectra of bacterial cells Fig. 1a displays the infrared spectra of full EPS- and partial EPS-S. suis from 1800 cm−1 to 800 cm−1. Based on previously reported spectral features for bacteria (Jiang et al., 2004; Yee et al., 2004; Parikh and Chorover, 2006; Ojeda et al., 2008; Ueshima et al., 2008; Gao and Chorover, 2009), the absorption bands of S. suis are summarized in Table S2. The peak at 1740 cm−1 corresponds to C_O stretching associated with lipids and fatty acids; 1645, 1539, and 1460 cm−1, for the C_O stretching of amide Ι, N\\H bending/C\\N stretching of amide II, and \\CH3/\\CH2 bending of proteins, respectively; 1397 cm−1, for the symmetric stretching C_O; 1234 cm−1, for the stretching of PO− \COOH/ 2 in phosphodiesters and phosphates, or vibrations of \ C\\O\\C in esters; 1067 cm−1, for the C\\OH/C\\O\\C/C\\C vibrations in polysaccharides, or P_O stretching of phosphodiester moieties; 887/ 843 cm−1, for the C\\O\\P/P\\O\\P ring vibrations in polysaccharides, or O\\P\\O vibration. The infrared bands of full EPS-S. suis supported previous work that bacterial surface macromolecules were composed of a combination of protein, lipid, and carbohydrate (Eboigbodin and Biggs, 2008). After the partial removal of the EPS, some absorption peaks (1740 and 843 cm−1) disappeared. The band at 1645 cm−1 shifted to 1640 cm−1. The disappearance and shift of some bands implied that some EPS constituents were removed, or the residual EPS molecules exposed after the cleaving of EPS. Ha et al. (2010) also reported that the intensities of peaks were significantly lower in the spectrum of Shewanella oneidensis without EPS.

W. Zhao et al. / Chemical Geology 410 (2015) 79–88

1063

885

887

843

1234

800

Full EPS-E. coli Partial EPS-E. coli

860

The titration curves of the model pathogens before and after CER treatment are presented in Fig. 2. The amount of OH− consumed during the titration from pH 2.5 to pH 10.0 reflects the bacterial buffering capacity originated from the protons dissociated from surface acidic chemical moieties. The inflection points on the titration curves indicate the deprotonation of these different site types (Ams et al., 2004; Fang et al., 2010). Table S4 shows that the overall OH− amount consumed by E. coli was larger than S. suis regardless of the EPS presence. The buffering capacities of full EPS-E. coli were higher as compared to those of partial EPS-E. coli, which suggested that the extraction of EPS removed some of the macromolecules and their dissociable chemical moieties. Dissimilarly, no obvious difference was found between the buffering capacities of full EPS-S. suis and partial EPS-S. suis. The relative order of surface charge densities as calculated from the titration curves (Chen and Walker, 2007) was full EPS-E. coli (541.1 μC cm−2) N partial EPSE. coli (447.5 μC cm−2) N full EPS-S. suis (213.0 μC cm−2) N partial EPSS. suis (198.9 μC cm−2). After EPS removal, the surface charge density on E. coli surface decreased by 17%, which was greater than that on S. suis (7%). The potentiometric titration data were further evaluated by fitting by the non-electrostatic surface complexation model to quantify site concentrations and acid dissociation constants (pKa) for bacteria (Turner and Fein, 2006). The four-site model yielded the best fit to the titration curves (R2 N 0.99), which implied the presence of four types of chemical moieties on the cell surfaces. As shown in Table 1, the pKa

857

1085

1232

(a)

Full EPS-S. suis 1 Full EPS-S. suis 2 Full EPS-S. suis 3 Partial EPS-S. suis 1 Partial EPS-S. suis 2 Partial EPS-S. suis 3

3

-

OH consumed (10

As shown in Fig. 1b, E. coli exhibited fewer infrared absorption peaks than S. suis. E. coli cell clearly had a different combination of binding sites exposed on the outer surface such that the absorption band pattern differed from S. suis. The main absorption bands are as follows (Table S3): 1641 cm−1 (amide Ι band); 1545 cm−1 (amide II band); 1457 cm−1 (\\CH3/\\CH2 bending); 1399 cm−1 (symmetric stretching C_O); 1232 cm−1 (\\COOH/C\\O\\C vibrations in esters or symmetric −1 stretching of PO− (C\\OH/ 2 from phosphate moieties); 1085 cm −1 C\\O\\C/C\\C vibrations in polysaccharides); and 860 cm (asymmetric stretching vibration of O\\P\\O). These results revealed the presence of hydroxyl, carboxyl, amino, phosphoryl, and phosphate binding sites on E. coli surface. The fewer absorption peaks indicated that the site types on E. coli surface were less heterogeneous than S. suis. The overall bands were similar for the two E. coli types. No significant differences in peak positions (less than FTIR resolution of 4 cm−1) were observed between the bands of the full EPS-E. coli and partial EPS-E. coli from 1800 cm−1 to 800 cm−1. The partial removal of EPS from E. coli surface had no effect on the binding site types, which suggested that the overall composition of sites within the E. coli EPS was similar to that of the residual EPS fractions or cell wall in general (Ueshima et al., 2008). This finding agreed with previous reports that the FTIR spectra of untreated and CER-treated B. subtilis or Pseudomonas putida cells were similar (Fang et al., 2011; Hong et al., 2013). The peak positions of the four bacterial IR spectra in this study were comparable to those of other bacterial strains except the peak at 1740 cm−1 of full EPS-S. suis (Parikh and Chorover, 2006; Ueshima et al., 2008; Ha et al., 2010).

6

0 2

4

6

8

10

pH

(b)

-1

Fig. 1. ATR-FTIR spectra of full EPS/partial EPS- (a) S. suis and (b) E. coli biomass samples on ZnSe. The IR spectra were collected over the scan range of 1800 cm−1 to 800 cm−1 with a resolution of 4 cm−1. Band positions are indicated in the figure.

mol cell )

800

12

-16

1400 1200 1000 -1 Wavenumber (cm )

9

8 Full EPS-E. coli 1 Full EPS-E. coli 2 Full EPS-E. coli 3 Partial EPS-E. coli 1 Partial EPS-E. coli 2 Partial EPS-E. coli 3

4

-

1600

OH consumed (10

1800

-16

-1

1399 1400

1231

1457 1455

1545

1640

1085

Absorbance (a.u.)

1231

1538

1460

1395

1400 1200 1000 -1 Wavenumber (cm )

1545

(b)

1600

3.2. Potentiometric titration

mol cell )

1800

1640

1460

1397

1645

1539

Full EPS-S. suis Partial EPS-S. suis

1641

1740

Absorbance (a.u.)

(a)

1067

82

0 2

4

6

8

10

pH Fig. 2. Potentiometric titration data for full EPS/partial EPS- (a) S. suis and (b) E. coli (6.0 × 109 cells per mL) biomass samples in 0.01 M KCl. Each experiment was conducted in triplicate.

W. Zhao et al. / Chemical Geology 410 (2015) 79–88 Table 1 Dissociation constants (pKa) and site concentrations of surface functional groups for full EPS- and partial EPS-pathogens.

Full EPS-S. suis

Partial EPS-S. suis

Full EPS-E. coli

Partial EPS-E. coli

pKa 1.64 ± 0.17 5.55 ± 0.16 7.33 ± 0.34 9.56 ± 0.17 2.28 ± 0.16 5.08 ± 0.25 6.38 ± 0.17 9.39 ± 0.45 3.31 ± 0.30 5.26 ± 0.53 6.82 ± 0.15 9.46 ± 0.07 2.44 ± 0.30 4.88 ± 0.41 7.13 ± 0.38 10.09 ± 0.50

Site concentration (10−16 mol cell−1)

Total site concentration (10−16 mol cell−1)

0.29 ± 0.12 2.11 ± 0.14 0.89 ± 0.07 1.67 ± 0.06 1.91 ± 0.33 1.50 ± 0.21 0.48 ± 0.10 0.92 ± 0.25 1.37 ± 0.33 0.98 ± 0.09 1.49 ± 0.51 3.09 ± 0.39 0.92 ± 0.16 0.78 ± 0.20 1.67 ± 0.15 3.07 ± 0.37

4.96 ± 0.13

concentrations of carboxyl/phosphoryl or reduced the phosphodiester/ amine/hydroxyl moieties on B. licheniformis S-86 (Tourney et al., 2008). Such deviation among various studies likely resulted from the differences in cell strains and growth phase, which lead to the differential expression of EPS molecules on bacterial surfaces (Eboigbodin et al., 2007). 3.3. Charge properties of bacteria and soil

4.81 ± 0.69

6.93 ± 0.89

6.44 ± 0.55

values for full EPS- and partial EPS-cells were determined to be 1.6–3.3 (pKa1), 4.9–5.6 (pKa2), 6.4–7.3 (pKa3) and 9.4–10.1 (pKa4). According to the typical deprotonation constants for short-chained organic site types (Fein et al., 1997), the proton-active sites on bacterial surfaces likely corresponded to the phosphodiester (pKa b 4), carboxylic (4 b pKa b 6), phosphoric (pKa ≈ 7), and hydroxyl/amine (9 b pKa b 11) moieties. ATR-FTIR data showed PO− \P\\O 2 /C_O/O\ stretching and N\\H bending (Tables S2 and S3), further confirming the presence of these chemical moieties on bacteria. Previous studies also found four-site/three-site models with pKa values of 3.2–4.2 (pKa1), 4.2–5.9 (pKa2), 6.4–7.7 (pKa3) and 9.0–10.6 (pKa4) for E. coli, S. oneidensis, P. putida, Rhizobium tropici, and Agrobacterium sp. with or without EPS (Castro and Tufenkji, 2007; Ueshima et al., 2008; Ha et al., 2010; Kenny and Fein, 2011). The deprotonation constants (pKa2/pKa3/pKa4) in this study compared well to the above values in the literatures. However, the pKa1 values (1.6–3.3) of E. coli and S. suis were obviously lower than those reported on other bacteria (3.2–4.2). It indicated that the bacterial strains used in this work possessed smaller proton binding capacities of phosphodiester moieties. Furthermore, the pKa values significantly decreased or increased (P b 0.05) after EPS removal (e.g., pKa3 for S. suis and pKa4 for E. coli), suggesting that the presence of EPS molecules could alter the protonation behaviors of the biomass. The total site concentrations on full EPS-S. suis and full EPSE. coli surfaces were 4.96 × 10−16 and 6.93 × 10−16 mol cell−1, respectively; while those of partial EPS-S. suis (4.81 × 10−16 mol cell−1) and partial EPS-E. coli (6.44 × 10−16 mol cell−1) were very close to the full EPS-bacteria. No significant differences were found between the total site concentrations on full EPS-cell and on partial EPS-cell surfaces (P N 0.05). The two strains used in this study had more surface site concentrations than another pathogenic E. coli O157:H7 (2.74 × 10−16 mol cell−1) reported by Kim et al. (2009b). For each surface site, the decreasing trend for site concentrations was not always observed for the four site types after EPS removal (Table 1). It is interesting to note that the site concentration of phosphodiester moiety for partial EPS-S. suis (1.91 × 10−16 mol cell−1) was larger than that for full EPS-S. suis (0.29 × 10−16 mol cell−1 ). Partial EPS-E. coli also had more phosphoric moieties (increased by 0.18 × 10−16 mol cell−1) than full EPS-E. coli. This phenomenon revealed that more phosphodiester and phosphoric moieties on the outer cell wall may be exposed with the removal of EPS, although the overall site concentrations decreased. It has been reported that EPS had different effects on cell surface site concentrations. For example, EPS removal increased the concentrations of all four site types on B. subtilis (Hong et al., 2013); whereas this treatment did not significantly alter the site

Zeta potential is defined as the potential at the shear plane of electrical double layer, which reflects the net surface charge of bacteria or soil particle (Kirby and Hasselbrink, 2004). The zeta potentials determined under varied IS conditions are shown in Fig. 3. Within the IS range tested, the natural soil sample and pathogens were all negatively charged. Full EPSS. suis was the least negatively charged with its zeta potential increasing slightly from −11.9 mV to −6.5 mV with the increase of IS from 1 mM to 100 mM. The zeta potentials of partial EPS-S. suis (from − 33.0 mV to −9.4 mV), full EPS-E. coli (from −37.7 mV to −13.4 mV) and partial EPS-E. coli (from −35.1 mV to −11.9 mV) also became less negative with increasing IS from 1 to 60 mM. This reduction of zeta potential was attributed to the double layer compression and charge screening by counter ions outside the cell surfaces (Walker et al., 2004; Shephard et al., 2010). Above 60 mM, the zeta potentials of partial EPS-S. suis/E. coli and full EPS-S. suis remained constant (P N 0.05), while only the full EPSE. coli increased slightly by 3.2 mV. This result suggests that the compression of the double layer is effectively achieved around 60 mM (Mills et al., 1994; Zhao et al., 2012). Similar results were also observed in literature that the zeta potentials of E. coli O157:H7 and Salmonella pullorum SA1685 were unchanged (P N 0.05) at high IS conditions (Castro and Tufenkji, 2007; Haznedaroglu et al., 2009). The surface charges of the model soil showed less variation than the bacteria, with the zeta potentials changing only from −21.7 mV to −15.1 mV (1–100 mM). Over the IS range examined, the relative zeta potentials of the cell types were as follows (Fig. 3): full EPS-E. coli N partial EPS-E. coli N partial EPS-S. suis N full EPS-S. suis. The charges of bacterial cells are related to the nature of macromolecules (e.g., EPS and lipids) in the outer cell wall (Wilson et al., 2001; Hong and Brown, 2006). Ionizable sites (e.g., phosphodiester, carboxylic, phosphoric, and amine/hydroxyl moieties) that make up these macromolecules are responsible for the surface charge exhibited under different IS (Castro and Tufenkji, 2007). Hence, the different site concentrations of cell walls for the four bacteria could explain the observed variability in the zeta potentials. At the tested pH range (5.6–5.8), the deprotonation of phosphodiester (pKa b 4.0) and carboxylic moieties (pKa b 5.6) primarily contributed to the overall net negative charges of the cells (Kim et al., 2009b). Titration results in Table 1 indicated that the sum of phosphodiester and carboxylic site

-10 Zeta potential (mV)

Pathogen type

83

-20 Full EPS-S. suis Partial EPS-S. suis Full EPS-E. coli Partial EPS-E. coli Yellow-Brown soil

-30

-40 0

20

40 60 80 Ionic strength (mM)

100

Fig. 3. Surface zeta potentials of full EPS/partial EPS-pathogens and Yellow-Brown soil at IS 1–100 mM (unadjusted pH 5.6–5.8). Error bars are the standard deviation of three replicates.

84

W. Zhao et al. / Chemical Geology 410 (2015) 79–88

concentrations for partial EPS-S. suis increased by 1.01 × 10−16 mol cell−1 compared with full EPS-S. suis, resulting in more negative charges on partial EPS-S. suis. In addition, the negative surface charges of partial EPSS. suis (i.e., absolute zeta potential values) declined rapidly by 71.5% from 1 mM to 100 mM, while the surface charge trend of full EPS-S. suis was less sensitive to IS and decreased only by 45.4%. It suggests that EPS components on S. suis are able to reduce the effect of solution IS on the variation of S. suis surface charges. In contrast, the observed reduction in the total concentrations of these two chemical moieties (decreased from 2.35 × 10−16 mol cell−1 to 1.70 × 10−16 mol cell−1) on E. coli after CER treatment may explain the reduced amounts of surface negative charges. Furthermore, both E. coli types exhibited the similar trends in zeta potential, which suggests that EPS presence did not change the general trends of E. coli surface charges at various IS. It is interesting to mention here that EPS removal also did not affect E. coli surface site types and the band positions from 1800 cm−1 to 800 cm−1 (Fig. 1b), likely due to the limited coverage of EPS on E. coli cell. Meanwhile, before and after the EPS removal via CER treatment, S. suis showed a dissimilar change in the surface charge trends over the tested IS range, and the resulting absorption peaks shifted or disappeared (Fig. 1a). Therefore, it is proposed that the change in observed cell surface charge is consistent with the loss of chemical moieties with the partial removal of EPS molecules. 3.4. Surface hydrophobicity of bacteria and soil Hydrophobicity values were quantified based on the percentage of bacterial or soil partitioning in a hydrocarbon phase (n-dodecane) and the results are presented in Fig. 4. Across the IS range tested, the hydrophobicities of S. suis and E. coli ranged from 21.4% to 42.5% and from 3.4% to 32.4%, respectively. These hydrophobicity values were comparable to those of other bacterial strains (S. pullorum SA1685 — 26.5%; E. coli O2:H7 — 2.8%; E. coli O157:H7 — 25.9%; E. coli O157:H16 — 33.1%) previously reported (Li and McLandsborough, 1999; Haznedaroglu et al., 2009; Kim et al., 2010). In the presence of 1–10 mM KCl, the hydrophobicities of both cell types with full EPS were approximately 30%, whereas the values gradually decreased to 21.4% and 8.9% at 100 mM for E. coli and S. suis, respectively. A decreasing trend for E. coli O157:H7 hydrophobicity as IS increased (26.7%–16.4% across range of IS from 1 mM to 100 mM) was also reported in the literature (Haznedaroglu et al., 2009). Bacterial surface polymers (e.g., EPS) are compressed at high IS (Abu-Lail and Camesano, 2003), and probably led to the decrease in the exposure of hydrophobic sites outside the cell surface (Alizadeh-Pasdar and Li-Chan, 2000). For cells with partial EPS, their hydrophobicities remained variable and

were around 35% for S. suis and 10% for E. coli, respectively, exhibiting irregular trends. No significant differences (P N 0.05) in hydrophobicity were observed for soil particle (70.0% ± 1.6%) over the entire IS range. Notably, the overall hydrophobicity values of the four cell types fall in the following order: partial EPS-S. suis ≥ full EPS-S. suis ≥ full EPSE. coli ≥ partial EPS-E. coli. The only exception was full EPS-E. coli at 100 mM, which may be due to the conformational changes of hydrophobic moieties on EPS molecules at high IS condition (Shephard et al., 2010). The hydrophobicities of S. suis were higher than E. coli, indicating more nonpolar sites on S. suis surface — regardless of EPS presence. Probably, a greater amount of acidic and polar sites are exposed on the surface of E. coli (Table 1, 6.44/6.93 × 10−16 mol cell−1), resulting in a lower percentage of cells partitioning into the hydrocarbon (Walker et al., 2005b). A greater hydrophobicity value was found for partial EPS-S. suis (30.6%–42.5%) than full EPS-S. suis (21.4%–34.0%). This difference may be ascribed to the removal of large amount of proteins (28.70 μg/108 cells) relative to sugar content (12.78 μg/108 cells) in S. suis EPS (Table S1). The outer proteins of bacteria that existed in EPS are mostly hydrophilic, which decreases cell surface hydrophobicity (Nikaido, 2003; Walker et al., 2005b). Further evidence of the sensitivity of bacterial hydrophobicity to proteins in extracellular polymers can be found in previous studies (Bruinsma et al., 2001; Chavant et al., 2002). The FTIR spectra in Fig. 1a also confirmed that the absorption peak for protein at 1645 cm−1 (amide Ι) shifted after treatment with CER. Interestingly, the average hydrophobicity values of partial EPS-E. coli decreased by 11% as compared with those of full EPS-E. coli at IS 1–100 mM. E. coli EPS had higher sugar content (11.73 μg/108 cells) than protein (4.80 μg/108 cells). Thus, the major parts of the molecular composition removed by CER were likely to be the uncharged hydrocarbon compounds C\\(C,H) in polysaccharides, which contained a relatively larger number of nonpolar sites than proteins (Kim et al., 2009b). It has been reported that the C\\(C,H) constituents could contribute to the increase in hydrophobicity (Hamadi et al., 2008). Therefore, a reduction of E. coli hydrophobicity after CER treatment can be attributed to the removal of C\\(C,H) constituents (polysaccharides) in EPS. 3.5. Influence of ionic strength on cell attachment Fig. 5 shows the attachment of bacteria to soil particles as a function of IS. The cell attachment percentages followed the sequence of full EPSS. suis (51.8%–95.6%) N partial EPS-S. suis (21.5%–77.5%) N partial EPSE. coli (19.9%–55.6%) N full EPS-E. coli (10.4%–37.6%). These obtained different attachment trends suggested that the pathogen type and EPS

120 Attachment percentage (%)

Hydrophobicity (%)

75 Full EPS-S. suis Partial EPS-S. suis Full EPS-E. coli Partial EPS-E. coli Yellow-Brown soil

60 45 30 15

Full EPS-S. suis Partial EPS-S. suis Full EPS-E. coli Partial EPS-E. coli

100 80 60 40 20

0 0

20

40 60 80 Ionic strength (mM)

100

Fig. 4. Hydrophobicities (%) of full EPS/partial EPS-pathogens and Yellow-Brown soil as a function of IS (1–100 mM). The hydrophobicity was quantified as the percentage of cells or soil particles partitioned into hydrocarbon (n-dodecane) phase. Error bars are the standard deviation of three replicates.

0 0

20

40 60 80 Ionic strength (mM)

100

Fig. 5. Effect of IS on the attachment percentages of pathogens to Yellow-Brown soil (unadjusted pH 5.6–5.8, IS 1–100 mM). Error bars are the standard deviation of three replicates.

W. Zhao et al. / Chemical Geology 410 (2015) 79–88

content had great influences on subsequent cell attachment to soil at varied IS. Under the experimental conditions of pH 5.6–5.8, the pathogens and soil particle were all negatively charged (Fig. 3), resulting in the electrostatic repulsive forces existing between cells and soil (Haznedaroglu et al., 2008). Full EPS-S. suis was the least negatively charged (from − 11.9 mV to − 6.5 mV), which yielded the weakest repulsive forces between full EPS-S. suis and soil, and hence the greatest attachment to soil was observed. Similarly, the greatest absolute zeta potentials of full EPS-E. coli (from − 37.7 mV to − 13.4 mV) generated the largest repulsive force, and subsequently the least attachment was found. The attachment behavior could be quantitatively explained by the zeta potential values. Moreover, the sphere-plate DLVO theory was applied to gain insight into the relative interaction energies between the model bacteria and the soil as a function of separation distance (Hogg et al., 1966; Gregory, 1981; Torkzaban et al., 2007). A total positive interaction energy represents a repulsive force, whereas, a total negative interaction energy indicates an attraction (Vigeant and Ford, 1997). Fig. S1 illustrates the repulsive interactions existed at separation distances between bacteria and soil up to approximately 50 nm (1–60 mM), while some favorable conditions for cell attachment were predicted across the range of 60–100 mM, due to the total attraction regardless of separation distance (Haznedaroglu et al., 2009; Zhao et al., 2014). Energy barriers that inhibited pathogen interaction in the primary minima were found to exist at separation distances of ~3 nm. As shown in Table 2, the repulsive energy barriers of cell–soil interactions were in the order of full EPSE. coli (0–377.1 kT) N partial EPS-E. coli (0–362.1 kT) N partial EPS-S. suis (0–268.5 kT) N full EPS-S. suis (0–76.2 kT). These heights of energy barriers were inversely related with the sequence of cell attachment percentages. It indicated that the sphere-plate DLVO model could quantitatively elucidate the different interaction energy trends and attachment behaviors of full EPS/partial EPS-cells. The attachment percentages of full EPS-S. suis and partial EPS-S. suis remarkably increased by 85.2%–158.4% with IS increasing from 1 mM to 60 mM (P b 0.05). This attachment trend was in agreement with the increasing zeta potentials displayed in Fig. 3, which caused a reduction of electrostatic repulsion with increasing IS. DLVO prediction in Table 2 also revealed that the energy barriers of both S. suis types decreased (268.5–1.9 kT) with increasing IS, and the interactions became chemically favorable at 60 mM (i.e., no energy barrier) (Haznedaroglu et al., 2009). While IS was above 60 mM, the zeta potentials of full EPSS. suis (Fig. 3) were statistically unchanged (P N 0.05), and the trends of calculated DLVO energy curves at 60–100 mM got very close to each other (Fig. S1a). Thus, the extent of electrostatic repulsions were similar and the percentages of attached full EPS-S. suis was expected to reach a plateau. However, a decreasing trend was observed for full EPS-S. suis with increasing IS from 60 mM to 100 mM, which deviated from the prediction of DLVO theory. It suggested that additional mechanisms may play a role in the full EPS-S. suis attachment to soil at 80 and

Table 2 Energy barrier heights between pathogens and Yellow-Brown soil as a function of IS (unadjusted pH 5.6–5.8). IS (mM)

1 5 10 20 40 60 80 100

Energy barrier of pathogen–soil particle (kT) Full EPS-S. suis

Partial EPS-S. suis

Full EPS-E. coli

Partial EPS-E. coli

76.2 49.2 16.3 1.9 NB NB NB NB

268.5 159.7 101.4 58.7 15.7 NB NB NB

377.1 284.4 206.0 132.7 53.3 13.6 NB NB

362.1 230.8 137.3 111.6 52.8 NB NB NB

Note: NB denotes no energy barrier.

85

100 mM, which are not incorporated in the traditional DLVO theory. The possible mechanism involved will be discussed in the next section. The attachment of full EPS/partial EPS-E. coli to soil particles also showed increasing trends in the range of 1–100 mM (Fig. 5). This was an expected phenomenon due to double-layer compression occurring with increasing IS. Zeta potentials of E. coli cells confirmed this with decreased absolute values at higher IS (Fig. 3). As also indicated in Table 2, E. coli experienced smaller energy barriers (377.1–13.6 kT) at higher IS, and achieved its largest attachment capacity at 100 mM (37.6% and 55.6% for full EPS-E. coli and partial EPS-E. coli, respectively) where no barrier existed. Additionally, the comparable curves of full EPS-E. coli and partial EPS-E. coli at varied IS implied that EPS did not affect the attachment tendency of E. coli to soil particles. This result is generally in agreement with the zeta potential trends of full EPS/partial EPS-E. coli over the IS range, which may be attributed to the similar surface site types between full EPS-E. coli and partial EPS-E. coli (Fig. 1b). Cell attachment data at varied IS were plotted against their surface properties to statistically investigate the main factors influencing the overall attachment phenomena. As seen in Fig. S2, a significant correlation (Y = − 0.12 × X + 56.9, R2 = 0.602, P b 0.01) was observed between bacterial attachment percentage (Y) and corresponding energy barrier (X) at 1–60 mM. This linear regression could be used to predict the extent of pathogen attachment in soil on the basis of existed energy barriers. Moreover, in the range of 60–100 mM where no repulsive energies appeared, the total interaction energies of partial EPS-S. suis and full EPS/partial EPS-E. coli at short distances of 0–5 nm became more attractive with increased IS (Figs. S1 b–d). The result verified that the attachment behaviors of the three bacterial types (partial EPS-S. suis and full EPS/partial EPS-E. coli) to natural soil particle at high IS were also in accordance with the sphere-plate DLVO calculations. 3.6. Other proposed mechanisms of EPS-induced cell attachment In the current study, non-DLVO type forces may contribute to the attachment behaviors. The first possibility is hydrophobic effect. Previous studies showed that larger cell surface hydrophobicity was found to enhance its attachment affinity for soil minerals (Stenström, 1989; Kristian Stevik et al., 2004; Meyer et al., 2006). Attractive hydrophobic interaction can occur at separation distance of b10 nm and decay exponentially with distance (Israelachvili and Pashley, 1984; Rochex et al., 2004; Meyer et al., 2006). However, since DLVO calculations (Fig. S1) suggest that electrostatic repulsion dominates interactions as far as 50 nm, the cells will not approach the soil closely enough (in the order of 10 nm) to feel the hydrophobic force. This is further supported by the lack of correlation between attachment and hydrophobicity values. As shown in Fig. 6, the decrease in bacterial attachment to soil was generally observed with the increase of hydrophobicity. Although EPS had obvious impacts on cell hydrophobicities (Fig. 4), no significant positive correlation (P N 0.05) could be established (Fig. 6), indicating that the hydrophobic interaction had negligible effect on cell attachment. Since the attractive secondary energy minima of cell–soil interactions were very small (around − 1 kT, calculated by DLVO theory) in this study, it should not be sufficient to retain large amounts of bacteria on the soil surfaces. The average thermal energy of a bacterium is reported to be 0.5–1 kT (Hahn and O'Melia, 2004; Tufenkji and Elimelech, 2004). Some weakly attached cells may be knocked out of the secondary minima if only van der Waals and electrostatic forces were involved. Under this condition, other attractive forces possibly contribute to the bacterial attachment (e.g., surface roughness, charge heterogeneity and hydrodynamic force) (Li and Logan, 2004; Walker et al., 2005b; Chen et al., 2010). The inability of DLVO theory to capture the interaction behavior entirely is partially due to the fact that the soil used in this study was not as homogeneous and smooth as DLVO theory assumes. Soil particles were irregularly shaped, and consisted of different particle sizes (clay 14.2%;

86

W. Zhao et al. / Chemical Geology 410 (2015) 79–88

Attachment percentage (%)

100 Full EPS-S. suis Partial EPS-S. suis Full EPS-E. coli Partial EPS-E. coli

80 60

Negative correlation

40 20 0 0

10

20 30 Hydrophobicity (%)

40

50

Fig. 6. Relationships between bacterial hydrophobicities and attachment percentages of bacteria to Yellow-Brown soil (IS 1–100 mM).

silt 64.1%; sand 21.7%). The rough soil surfaces were likely to promote pathogen attachment. It was reported that roughness can lower surface energy, so that the electrostatic repulsive interactions are considerably less between the cell and rough surfaces than they are for more idealized smooth surfaces (Holm et al., 1992; Shellenberger and Logan, 2002; Li and Logan, 2004). The next possible mechanism involved is surface charge heterogeneity. In this work, uniform distribution of surface charge is assumed to calculate electrostatic repulsion energy (Hogg et al., 1966; Ohshima et al., 1983; Walker et al., 2005b). This assumption, however, is not valid since real soil and cell surfaces are heterogeneous in nature (Song et al., 1994). It is well known that soil particles consist of multiple mineral and organic matter types (Brady and Weil, 1999). Soil grains have a tendency to aggregate and form heterogeneous surfaces exhibiting different charge distribution (Kemper and Rosenau, 1986; Tombácz and Szekeres, 2006). Local attachment of bacteria may occur to various mineral faces where the electrostatic repulsions between cells and some of these sites may have regions of lesser or different charge such that greater interaction is feasible (Song et al., 1994; Cai et al., 2013). The charge heterogeneity on cell surface could also result in the reduction of electrostatic energy barrier and more stable cell attachment (Redman et al., 2004). It has been reported that a variety of charge heterogeneities can be found on top of the peptidoglycan layer in Gram-positive (S. suis) or on the outer membrane in Gram-negative bacteria (E. coli), which arise from the differences in the arrangement of charged functional groups (Poortinga et al., 2002; Walker et al., 2005a). Hence, the local surface charge heterogeneity on soil/cell surfaces are important factors strengthening attachment. Another possible interaction mechanism involved is hydrodynamic force. The cells and soil particles in attachment assay were suspended and mixed at 150 rev min−1. It is possible that some level of shear force caused by this mixing may provide additional kinetic energy to help cells attach to the solid surfaces (Mohamed et al., 2000; Thomas et al., 2002; Chen et al., 2010; Hong et al., 2012). There is also a possibility of polymer bridging as facilitated by EPS. It is well accepted that cell surface polymers may form covalent or hydrogen bonds that contribute to the attachment if the polymers are long enough to extend across the separation distance (Grasso et al., 2002; Chen and Walker, 2007). In this study, polymer bridging was found to play negligible roles in pathogen attachment. This is supported by data wherein the attachment of partial EPS-E. coli (19.9%–55.6%) was significantly greater (P b 0.05) than those of full EPS-E. coli (10.4%– 37.6%). EPS on E. coli surface did not have substantial polymer bridging forces to improve its attachment. Besides, the attachment percentages

(Y) of full EPS/partial EPS-pathogens were significantly governed by zeta potentials (X) of bacteria (Y = 2.1 × X + 86.1, R2 = 0.768, P b 0.01), rather than EPS content. Data and the above analysis suggest that EPS influenced cell attachment via electrostatic interactions stemming from the charged sites on the exposed surface polymers (full or partial EPS). Any contribution of polymer bridging or hydrophobicity was insignificant. Finally, the short-range steric effect was likely to be responsible for the abnormal decreasing attachment trend of full EPS-S. suis between 60 and 100 mM (Kuznar and Elimelech, 2006). Table 2 illustrates that the energy barriers of full EPS-S. suis did not exist at this particular IS (60–100 mM). Total interaction energies were all attractive and S. suis cell was tightly attached at smaller separation distances. Under high IS conditions, the variation of electrostatic force was reduced. The strong steric repulsion could overcome the electrostatic force and contribute to the attachment at short distances (Abu-Lail and Camesano, 2003). It has been documented that the presence of excess ions suspended among the polymers may increase the intramolecular electrostatic interactions between individual polymer units, which lead to the polymers being more rigid. The rigid polymers restrain themselves from altering steric conformation, and interact directly with the soil surface to produce a steric repulsion (Chen and Walker, 2007; Kim et al., 2010). This hypothesis was corroborated by comparing with the attachment behavior of partial EPS-S. suis. Specifically, the attachment of partial EPS-S. suis increased with IS from 60 mM to 100 mM (i.e., attachment percentages of 55.7%, 67.0% and 77.5% at 60, 80 and 100 mM, respectively), verifying that the steric hindrance directly originated from the EPS components on S. suis. Such phenomenon has also been reported by previous studies. For instance, Kim et al. (2009b) found the adhesion of E. coli O157:H7 to quartz sand decreased at IS (KCl) ≥ 10 mM. Chen and Walker (2007) reported that the steric force was repulsive when the IS (CaCl2) was greater than 300 mM, as confirmed by the attachment efficiency of Halomonas pacifica to quartz decreasing at high IS. Otto et al. (1999) observed the number of attached E. coli MS7 to quartz crystal microbalance surfaces increased up to an IS (phosphate-buffered saline) of 122.5 mM or 185 mM, above which adhesion either leveled off or decreased. It should be mentioned that the solution IS value at which steric repulsion appeared in this study (≥80 mM) was distinctly different from the above investigations. This difference was possibly ascribed to the different natures of EPS on cell surfaces and dissimilar solution chemistry (Rijnaarts et al., 1999). On the contrary, E. coli had lower total EPS content than S. suis (Table S1). The small amount of EPS on E. coli (26.25 μg/108 cells) was probably less rigid so as not to be sensitive to steric repulsion. 4. Conclusions This work directly demonstrates that the EPS constituents could play contrasting roles in bacterial attachment to natural soil particles, as a function of cell type. Detailed insights into the mechanisms regarding the variation of EPS-induced pathogen surface properties were also provided. The obtained results showed that bacterial EPS had distinct influences on the surface charges and hydrophobicities of S. suis and E. coli, due to their different acidic site concentrations and EPS components. EPS removal impacted pathogen attachment to soil via the modification of cell surface charges. Steric repulsion from the EPS on S. suis surface was possibly involved in its attachment under high IS conditions. In contrast to previous reports using low-heterogeneous mineral surfaces (Li and Logan, 2004; Vadillo-Rodríguez et al., 2005; Chen and Walker, 2007), the effects of “EPS bridging” or hydrophobic interaction on bacterial attachment to soil particles were insignificant. Other physical (e.g., soil surface roughness and hydrodynamic shear force) and chemical factors (e.g., charge heterogeneity) may play more important roles and contribute to the attachment processes of full EPS/partial EPSbacteria. Results of cell attachment to pure mineral experiments may

W. Zhao et al. / Chemical Geology 410 (2015) 79–88

have limited implications for predicting the real functions of EPS in bacterial attachment phenomena in natural soil. Classic DLVO theory, including electrostatic and van der Waals forces, was able to explain the attachment trends of pathogens regardless of EPS content under low soil IS conditions. Findings from the current study highlight the importance of understanding the diverse functions of EPS that exist in bacteria–soil interactions, which is vital to better assess the fates of different pathogens in soil–aquatic environments. Acknowledgments This work was supported by the National Natural Science Foundation of China (41171196), the National Basic Research Program of China (2015CB150504), Foundation for the Author of National Excellent Doctoral Dissertation of China (201066), and the Fundamental Research Funds for the Central Universities (2011PY005). The authors would like to thank Professor Junlong Zhao from the State Key Laboratory of Agricultural Microbiology for providing the S. suis SC05 and E. coli WH09 isolates. Appendix A. Supplementary material Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.chemgeo.2015.06.013. References Abit, S.M., Bolster, C.H., Cai, P., Walker, S.L., 2012. Influence of feedstock and pyrolysis temperature of biochar amendments on transport of Escherichia coli in saturated and unsaturated soil. Environ. Sci. Technol. 46, 8097–8105. Abu-Lail, N.I., Camesano, T.A., 2003. Role of ionic strength on the relationship of biopolymer conformation, DLVO contributions, and steric interactions to bioadhesion of Pseudomonas putida KT2442. Biomacromolecules 4, 1000–1012. Aguilera, A., Souza-Egipsy, V., Martín-Úriz, P.S., Amils, R., 2008. Extraction of extracellular polymeric substances from extreme acidic microbial biofilms. Appl. Microbiol. Biotechnol. 78, 1079–1088. Alizadeh-Pasdar, N., Li-Chan, E.C., 2000. Comparison of protein surface hydrophobicity measured at various pH values using three different fluorescent probes. J. Agric. Food Chem. 48, 328–334. Ams, D.A., Fein, J.B., Dong, H., Maurice, P.A., 2004. Experimental measurements of the adsorption of Bacillus subtilis and Pseudomonas mendocina onto Fe-oxyhydroxidecoated and uncoated quartz grains. Geomicrobiol J. 21, 511–519. Borrok, D., Fein, J.B., 2004. Distribution of protons and Cd between bacterial surfaces and dissolved humic substances determined through chemical equilibrium modeling. Geochim. Cosmochim. Acta 68, 3043–3052. Brady, N.C., Weil, R.R., 1999. The Nature and Properties of Soil. 12th ed. Prentice-Hall Inc., Upper Saddle River, New Jersey. Bruinsma, G.M., Rustema-Abbing, M., van der Mei, H.C., Busscher, H.J., 2001. Effects of cell surface damage on surface properties and adhesion of Pseudomonas aeruginosa. J. Microbiol. Methods 45, 95–101. Cai, P., Huang, Q., Walker, S.L., 2013. Deposition and survival of Escherichia coli O157:H7 on clay minerals in a parallel plate flow system. Environ. Sci. Technol. 47, 1896–1903. Cao, Y., Wei, X., Cai, P., Huang, Q., Rong, X., Liang, W., 2011. Preferential adsorption of extracellular polymeric substances from bacteria on clay minerals and iron oxide. Colloids Surf. B: Biointerfaces 83, 122–127. Castro, F.D., Tufenkji, N., 2007. Relevance of nontoxigenic strains as surrogates for Escherichia coli O157:H7 in groundwater contamination potential: role of temperature and cell acclimation time. Environ. Sci. Technol. 41, 4332–4338. Chavant, P., Martinie, B., Meylheuc, T., Bellon-Fontaine, M.N., Hebraud, M., 2002. Listeria monocytogenes LO28: surface physicochemical properties and ability to form biofilms at different temperatures and growth phases. Appl. Environ. Microbiol. 68, 728–737. Chen, G.X., Walker, S.L., 2007. Role of solution chemistry and ion valence on the adhesion kinetics of groundwater and marine bacteria. Langmuir 23, 7162–7169. Chen, G.X., Hong, Y., Walker, S.L., 2010. Colloidal and bacterial deposition: role of gravity. Langmuir 26, 314–319. Dai, X., Boll, J., 2003. Evaluation of attachment of Cryptosporidium parvum and Giardia lamblia to soil particles. J. Environ. Qual. 32, 296–304. Eboigbodin, K.E., Biggs, C.A., 2008. Characterization of the extracellular polymeric substances produced by Escherichia coli using infrared spectroscopic, proteomic, and aggregation studies. Biomacromolecules 9, 686–695. Eboigbodin, K.E., Ojeda, J.J., Biggs, C.A., 2007. Investigating the surface properties of Escherichia coli under glucose controlled conditions and its effect on aggregation. Langmuir 23, 6691–6697. Evangelou, V., Blevins, R., 1988. Effect of long-term tillage systems and nitrogen addition on potassium quantity–intensity relationships. Soil Sci. Soc. Am. J. 52, 1047–1054. Fang, L., Huang, Q., Wei, X., Liang, W., Rong, X., Chen, W., Cai, P., 2010. Microcalorimetric and potentiometric titration studies on the adsorption of copper by extracellular

87

polymeric substances (EPS), minerals and their composites. Bioresour. Technol. 101, 5774–5779. Fang, L., Wei, X., Cai, P., Huang, Q., Chen, H., Liang, W., Rong, X., 2011. Role of extracellular polymeric substances in Cu(II) adsorption on Bacillus subtilis and Pseudomonas putida. Bioresour. Technol. 102, 1137–1141. Fein, J.B., Daughney, C.J., Yee, N., Davis, T.A., 1997. A chemical equilibrium model for metal adsorption onto bacterial surfaces. Geochim. Cosmochim. Acta 61, 3319–3328. Fein, J.B., Boily, J.F., Yee, N., Gorman-Lewis, D., Turner, B.F., 2005. Potentiometric titrations of Bacillus subtilis cells to low pH and a comparison of modeling approaches. Geochim. Cosmochim. Acta 69, 1123–1132. Foppen, J.W., Lutterodt, G., Roling, W.F., Uhlenbrook, S., 2010. Towards understanding inter-strain attachment variations of Escherichia coli during transport in saturated quartz sand. Water Res. 44, 1202–1212. Franz, E., Semenov, A.V., Termorshuizen, A.J., De Vos, O., Bokhorst, J.G., Van Bruggen, A.H., 2008. Manure-amended soil characteristics affecting the survival of E. coli O157: H7 in 36 Dutch soils. Environ. Microbiol. 10, 313–327. Gantzer, C., Gillerman, L., Kuznetsov, M., Oron, G., 2001. Adsorption and survival of faecal coliforms, somatic coliphages and F-specific RNA phages in soil irrigated with wastewater. Water Sci. Technol. 43, 117–124. Gao, X., Chorover, J., 2009. In-situ monitoring of Cryptosporidium parvum oocyst surface adhesion using ATR-FTIR spectroscopy. Colloids Surf. B: Biointerfaces 71, 169–176. Gargiulo, G., Bradford, S., Simunek, J., Ustohal, P., Vereecken, H., Klumpp, E., 2007. Bacteria transport and deposition under unsaturated conditions: the role of the matrix grain size and the bacteria surface protein. J. Contam. Hydrol. 92, 255–273. Gee, G.W., Bauder, J.W., 1986. Particle-size analysis. In: Klute, A. (Ed.), Methods of Soil Analysis, Part 1. Physical and Mineralogical Methods. American Society of Agronomy, Madison, pp. 383–411. Gerba, C.P., Smith, J.E., 2005. Sources of pathogenic microorganisms and their fate during land application of wastes. J. Environ. Qual. 34, 42–48. Gong, A.S., Bolster, C.H., Benavides, M., Walker, S.L., 2009. Extraction and analysis of extracellular polymeric substances: comparison of methods and extracellular polymeric substance levels in Salmonella pullorum SA 1685. Environ. Eng. Sci. 26, 1523–1532. Grasso, D., Subramaniam, K., Butkus, M., Strevett, K., Bergendahl, J., 2002. A review of nonDLVO interactions in environmental colloidal systems. Rev. Environ. Sci. Biotechnol. 1, 17–38. Gregory, J., 1981. Approximate expressions for retarded van der Waals interaction. J. Colloid Interface Sci. 83, 138–145. Guber, A.K., Shelton, D.R., Pachepsky, Y.A., 2005. Effect of manure on Escherichia coli attachment to soil. J. Environ. Qual. 34, 2086–2090. Guber, A.K., Pachepsky, Y.A., Shelton, D.R., Yu, O., 2007. Effect of bovine manure on fecal coliform attachment to soil and soil particles of different sizes. Appl. Environ. Microbiol. 73, 3363–3370. Ha, J., Gélabert, A., Spormann, A.M., Brown, G.E., 2010. Role of extracellular polymeric substances in metal ion complexation on Shewanella oneidensis: batch uptake, thermodynamic modeling, ATR-FTIR, and EXAFS study. Geochim. Cosmochim. Acta 74, 1–15. Hahn, M.W., O'Melia, C.R., 2004. Deposition and reentrainment of Brownian particles in porous media under unfavorable chemical conditions: some concepts and applications. Environ. Sci. Technol. 38, 210–220. Hamadi, F., Latrache, H., Zahir, H., Elghmari, A., Timinouni, M., Ellouali, M., 2008. The relation between Escherichia coli surface functional groups' composition and their physicochemical properties. Braz. J. Microbiol. 39, 10–15. Haznedaroglu, B.Z., Bolster, C.H., Walker, S.L., 2008. The role of starvation on Escherichia coli adhesion and transport in saturated porous media. Water Res. 42, 1547–1554. Haznedaroglu, B.Z., Kim, H.N., Bradford, S.A., Walker, S.L., 2009. Relative transport behavior of Escherichia coli O157:H7 and Salmonella enterica Serovar Pullorum in packed bed column systems: influence of solution chemistry and cell concentration. Environ. Sci. Technol. 43, 1838–1844. He, L.M., Tebo, B.M., 1998. Surface charge properties of and Cu(II) adsorption by spores of the marine Bacillus sp. strain SG-1. Appl. Environ. Microbiol. 64, 1123–1129. Hermansson, M., 1999. The DLVO theory in microbial adhesion. Colloids Surf. B: Biointerfaces 14, 105–119. Hogg, R., Healy, T., Fuerstenau, D., 1966. Mutual coagulation of colloidal dispersions. Trans. Faraday Soc. 62, 1638–1651. Holm, P.E., Nielsen, P.H., Albrechtsen, H.J., Christensen, T.H., 1992. Importance of unattached bacteria and bacteria attached to sediment in determining potentials for degradation of xenobiotic organic contaminants in an aerobic aquifer. Appl. Environ. Microbiol. 58, 3020–3026. Hong, Y., Brown, D.G., 2006. Cell surface acid–base properties of Escherichia coli and Bacillus brevis and variation as a function of growth phase, nitrogen source and C:N ratio. Colloids Surf. B: Biointerfaces 50, 112–119. Hong, Y., Brown, D.G., 2008. Electrostatic behavior of the charge-regulated bacterial cell surface. Langmuir 24, 5003–5009. Hong, Z., Rong, X., Cai, P., Dai, K., Liang, W., Chen, W., Huang, Q., 2012. Initial adhesion of Bacillus subtilis on soil minerals as related to their surface properties. Eur. J. Soil Sci. 63, 457–466. Hong, Z., Chen, W., Rong, X., Cai, P., Dai, K., Huang, Q., 2013. The effect of extracellular polymeric substances on the adhesion of bacteria to clay minerals and goethite. Chem. Geol. 360, 118–125. Hori, K., Matsumoto, S., 2010. Bacterial adhesion: from mechanism to control. Biochem. Eng. J. 48, 424–434. Israelachvili, J., Pashley, R., 1984. Measurement of the hydrophobic interaction between two hydrophobic surfaces in aqueous electrolyte solutions. J. Colloid Interface Sci. 98, 500–514. Jacobs, A., Lafolie, F., Herry, J., Debroux, M., 2007. Kinetic adhesion of bacterial cells to sand: cell surface properties and adhesion rate. Colloids Surf. B: Biointerfaces 59, 35–45.

88

W. Zhao et al. / Chemical Geology 410 (2015) 79–88

Jiang, W., Saxena, A., Song, B., Ward, B.B., Beveridge, T.J., Myneni, S.C., 2004. Elucidation of functional groups on gram-positive and gram-negative bacterial surfaces using infrared spectroscopy. Langmuir 20, 11433–11442. Karunakaran, E., Biggs, C.A., 2011. Mechanisms of Bacillus cereus biofilm formation: an investigation of the physicochemical characteristics of cell surfaces and extracellular proteins. Appl. Microbiol. Biotechnol. 89, 1161–1175. Keller, A.A., Auset, M., 2007. A review of visualization techniques of biocolloid transport processes at the pore scale under saturated and unsaturated conditions. Adv. Water Resour. 30, 1392–1407. Kemper, W.D., Rosenau, R.C., 1986. Aggregate stability and size distribution. In: Klute, A. (Ed.), Methods of Soil Analysis, Part 1. Physical and Mineralogical Methods. American Society of Agronomy, Madison, pp. 425–444. Kenny, J.P.L., Fein, J.B., 2011. Importance of extracellular polysaccharides on proton and Cd binding to bacterial biomass: a comparative study. Chem. Geol. 286, 109–117. Kim, H.N., Bradford, S.A., Walker, S.L., 2009a. Escherichia coli O157:H7 transport in saturated porous media: role of solution chemistry and surface macromolecules. Environ. Sci. Technol. 43, 4340–4347. Kim, H.N., Hong, Y., Lee, I., Bradford, S.A., Walker, S.L., 2009b. Surface characteristics and adhesion behavior of Escherichia coli O157:H7: role of extracellular macromolecules. Biomacromolecules 10, 2556–2564. Kim, H.N., Walker, S.L., Bradford, S.A., 2010. Macromolecule mediated transport and retention of Escherichia coli O157:H7 in saturated porous media. Water Res. 44, 1082–1093. Kirby, B.J., Hasselbrink, E.F., 2004. Zeta potential of microfluidic substrates: 1. Theory, experimental techniques, and effects on separations. Electrophoresis 25, 187–202. Kuznar, Z.A., Elimelech, M., 2006. Cryptosporidium oocyst surface macromolecules significantly hinder oocyst attachment. Environ. Sci. Technol. 40, 1837–1842. Li, B., Logan, B.E., 2004. Bacterial adhesion to glass and metal-oxide surfaces. Colloids Surf. B: Biointerfaces 36, 81–90. Li, J., McLandsborough, L., 1999. The effects of the surface charge and hydrophobicity of Escherichia coli on its adhesion to beef muscle. Int. J. Food Microbiol. 53, 185–193. Liu, X.M., Sheng, G.P., Luo, H.W., Zhang, F., Yuan, S.J., Xu, J., Zeng, R.J., Wu, J.G., Yu, H.Q., 2010. Contribution of extracellular polymeric substances (EPS) to the sludge aggregation. Environ. Sci. Technol. 44, 4355–4360. Long, G., Zhu, P., Shen, Y., Tong, M., 2009. Influence of extracellular polymeric substances (EPS) on deposition kinetics of bacteria. Environ. Sci. Technol. 43, 2308–2314. Meyer, E.E., Rosenberg, K.J., Israelachvili, J., 2006. Recent progress in understanding hydrophobic interactions. Proc. Natl. Acad. Sci. 103, 15739–15746. Mills, A.L., Herman, J.S., Hornberger, G.M., DeJesús, T.H., 1994. Effect of solution ionic strength and iron coatings on mineral grains on the sorption of bacterial cells to quartz sand. Appl. Environ. Microbiol. 60, 3300–3306. Mohamed, N., Rainier, T.R., Ross, J.M., 2000. Novel experimental study of receptormediated bacterial adhesion under the influence of fluid shear. Biotechnol. Bioeng. 68, 628–636. Mukherjee, J., Karunakaran, E., Biggs, C.A., 2012. Using a multi-faceted approach to determine the changes in bacterial cell surface properties influenced by a biofilm lifestyle. Biofouling 28, 1–14. Nikaido, H., 2003. Molecular basis of bacterial outer membrane permeability revisited. Microbiol. Mol. Biol. Rev. 67, 593–656. Nyberg, K.A., Vinnerås, B., Ottoson, J.R., Aronsson, P., Albihn, A., 2010. Inactivation of Escherichia coli O157:H7 and Salmonella Typhimurium in manure-amended soils studied in outdoor lysimeters. Appl. Soil Ecol. 46, 398–404. Ohshima, H., Chan, D.Y.C., Healy, T.W., White, L.R., 1983. Improvement on the Hogg– Healy–Fuerstenau formulas for the interaction of dissimilar double layers: II. Curvature correction to the formula for the interaction of spheres. J. Colloid Interface Sci. 92, 232–242. Ojeda, J., Romero-Gonzalez, M., Pouran, H., Banwart, S., 2008. In situ monitoring of the biofilm formation of Pseudomonas putida on hematite using flow-cell ATR-FTIR spectroscopy to investigate the formation of inner-sphere bonds between the bacteria and the mineral. Mineral. Mag. 72, 101–106. Otto, K., Elwing, H., Hermansson, M., 1999. Effect of ionic strength on initial interactions of Escherichia coli with surfaces, studied on-line by a novel quartz crystal microbalance technique. J. Bacteriol. 181, 5210–5218. Pachepsky, Y.A., Yu, O., Karns, J.S., Shelton, D.R., Guber, A.K., van Kessel, J.S., 2008. Straindependent variations in attachment of E. coli to soil particles of different sizes. Int. Agrophysics 22, 61–66. Parikh, S.J., Chorover, J., 2006. ATR-FTIR spectroscopy reveals bond formation during bacterial adhesion to iron oxide. Langmuir 22, 8492–8500. Poortinga, A.T., Bos, R., Norde, W., Busscher, H.J., 2002. Electric double layer interactions in bacterial adhesion to surfaces. Surf. Sci. Rep. 47, 1–32. Redman, J.A., Walker, S.L., Elimelech, M., 2004. Bacterial adhesion and transport in porous media: role of the secondary energy minimum. Environ. Sci. Technol. 38, 1777–1785. Rijnaarts, H.H., Norde, W., Lyklema, J., Zehnder, A.J., 1999. DLVO and steric contributions to bacterial deposition in media of different ionic strengths. Colloids Surf. B: Biointerfaces 14, 179–195. Rochex, A., Lecouturier, D., Pezron, I., Lebeault, J.M., 2004. Adhesion of a Pseudomonas putida strain isolated from a paper machine to cellulose fibres. Appl. Microbiol. Biotechnol. 65, 727–733.

Rong, X., Huang, Q., Chen, W., 2007. Microcalorimetric investigation on the metabolic activity of Bacillus thuringiensis as influenced by kaolinite, montmorillonite and goethite. Appl. Clay Sci. 38, 97–103. Shellenberger, K., Logan, B.E., 2002. Effect of molecular scale roughness of glass beads on colloidal and bacterial deposition. Environ. Sci. Technol. 36, 184–189. Shephard, J.J., Savory, D.M., Bremer, P.J., McQuillan, A.J., 2010. Salt modulates bacterial hydrophobicity and charge properties influencing adhesion of Pseudomonas aeruginosa (PA01) in aqueous suspensions. Langmuir 26, 8659–8665. Song, L., Johnson, P.R., Elimelech, M., 1994. Kinetics of colloid deposition onto heterogeneously charged surfaces in porous media. Environ. Sci. Technol. 28, 1164–1171. Soupir, M.L., Mostaghimi, S., Dillaha, T., 2010. Attachment of Escherichia coli and enterococci to particles in runoff. J. Environ. Qual. 39, 1019–1027. Stenström, T., 1989. Bacterial hydrophobicity, an overall parameter for the measurement of adhesion potential to soil particles. Appl. Environ. Microbiol. 55, 142–147. Stevik, Kristian T., Aa, K., Ausland, G., Fredrik Hanssen, J., 2004. Retention and removal of pathogenic bacteria in wastewater percolating through porous media: a review. Water Res. 38, 1355–1367. Taylor, A.A., Chowdhury, I., Gong, A.S., Cwiertny, D.M., Walker, S.L., 2014. Deposition and disinfection of Escherichia coli O157:H7 on naturally occurring photoactive materials in a parallel plate chamber. Environ. Sci.: Processes Impacts 16, 194–202. Thomas, W.E., Trintchina, E., Forero, M., Vogel, V., Sokurenko, E.V., 2002. Bacterial adhesion to target cells enhanced by shear force. Cell 109, 913–923. Tombácz, E., Szekeres, M., 2006. Surface charge heterogeneity of kaolinite in aqueous suspension in comparison with montmorillonite. Appl. Clay Sci. 34, 105–124. Tong, M., Long, G., Jiang, X., Kim, H.N., 2010. Contribution of extracellular polymeric substances on representative gram negative and gram positive bacterial deposition in porous media. Environ. Sci. Technol. 44, 2393–2399. Torkzaban, S., Bradford, S.A., Walker, S.L., 2007. Resolving the coupled effects of hydrodynamics and DLVO forces on colloid attachment in porous media. Langmuir 23, 9652–9660. Toth, J.D., Aceto, H.W., Rankin, S.C., Dou, Z., 2013. Short communication: survey of animalborne pathogens in the farm environment of 13 dairy operations. J. Dairy Sci. 96, 5756–5761. Tourney, J., Ngwenya, B.T., Mosselmans, J., Tetley, L., Cowie, G.L., 2008. The effect of extracellular polymers (EPS) on the proton adsorption characteristics of the thermophile Bacillus licheniformis S-86. Chem. Geol. 247, 1–15. Tsuneda, S., Aikawa, H., Hayashi, H., Yuasa, A., Hirata, A., 2003. Extracellular polymeric substances responsible for bacterial adhesion onto solid surface. FEMS Microbiol. Lett. 223, 287–292. Tufenkji, N., Elimelech, M., 2004. Deviation from the classical colloid filtration theory in the presence of repulsive DLVO interactions. Langmuir 20, 10818–10828. Turner, B.F., Fein, J.B., 2006. Protofit: a program for determining surface protonation constants from titration data. Comput. Geosci. 32, 1344–1356. Ueshima, M., Ginn, B.R., Haack, E.A., Szymanowski, J.E., Fein, J.B., 2008. Cd adsorption onto Pseudomonas putida in the presence and absence of extracellular polymeric substances. Geochim. Cosmochim. Acta 72, 5885–5895. Unc, A., Goss, M.J., 2004. Transport of bacteria from manure and protection of water resources. Appl. Soil Ecol. 25, 1–18. Vadillo-Rodríguez, V., Busscher, H.J., van der Mei, H.C., de Vries, J., Norde, W., 2005. Role of lactobacillus cell surface hydrophobicity as probed by AFM in adhesion to surfaces at low and high ionic strength. Colloids Surf. B: Biointerfaces 41, 33–41. Venglovsky, J., Sasakova, N., Placha, I., 2009. Pathogens and antibiotic residues in animal manures and hygienic and ecological risks related to subsequent land application. Bioresour. Technol. 100, 5386–5391. Vigeant, M., Ford, R.M., 1997. Interactions between motile Escherichia coli and glass in media with various ionic strengths, as observed with a three-dimensional-tracking microscope. Appl. Environ. Microbiol. 63, 3474–3479. Walker, S.L., Redman, J.A., Elimelech, M., 2004. Role of cell surface lipopolysaccharides in Escherichia coli K12 adhesion and transport. Langmuir 20, 7736–7746. Walker, S.L., Hill, J.E., Redman, J.A., Elimelech, M., 2005a. Influence of growth phase on adhesion kinetics of Escherichia coli D21g. Appl. Environ. Microbiol. 71, 3093–3099. Walker, S.L., Redman, J.A., Elimelech, M., 2005b. Influence of growth phase on bacterial deposition: interaction mechanisms in packed-bed column and radial stagnation point flow systems. Environ. Sci. Technol. 39, 6405–6411. Wilson, W.W., Wade, M.M., Holman, S.C., Champlin, F.R., 2001. Status of methods for assessing bacterial cell surface charge properties based on zeta potential measurements. J. Microbiol. Methods 43, 153–164. Yee, N., Benning, L.G., Phoenix, V.R., Ferris, F.G., 2004. Characterization of metal– cyanobacteria sorption reactions: a combined macroscopic and infrared spectroscopic investigation. Environ. Sci. Technol. 38, 775–782. Zhao, W., Liu, X., Huang, Q., Rong, X., Liang, W., Dai, K., Cai, P., 2012. Sorption of Streptococcus suis on various soil particles from an Alfisol and effects on pathogen metabolic activity. Eur. J. Soil Sci. 63, 558–564. Zhao, W., Walker, S.L., Huang, Q., Cai, P., 2014. Adhesion of bacterial pathogens to soil colloidal particles: influences of cell type, natural organic matter, and solution chemistry. Water Res. 53, 35–46.