Impact of Chemical Heterogeneities of Surfaces on Colonization by

Article pubs.acs.org/journal/abseba

Impact of Chemical Heterogeneities of Surfaces on Colonization by Bacteria Judith Böhmler, Hamidou Haidara, Arnaud Ponche, and Lydie Ploux* Institut de Science des Matériaux de Mulhouse, CNRS-UMR7361, University of Strasbourg/University of Haute-Alsace, UMR7361, Mulhouse, France S Supporting Information *

ABSTRACT: An essential yet never addressed parameter for the control of bacteria on functionalized biomaterial is surely the accessibility and heterogeneity of the functional groups immobilized on the surface. In this context, we investigated the colonization (Escherichia coli K12, Staphylococcus epidermidis RP62A) of precisely engineered surfaces revealing various densities of NH2 and CH3 functional groups. We demonstrated for the first time nonlinear relationships between the NH2/CH3 surface fraction and the quantity of adhered, adhering or detaching bacteria. Plateaus and transition zones were related to the range of NH2/CH3 surface fraction offering stability or sharp variation in bacterium/surface interactions. The nonlinear behavior was attributed to the discrete distribution of positive charges revealed by the bacterial membrane in the continuum of negative charges resulting from the phospholipids, which may correlate with one single specific distribution of positive NH3+ charges on the material surface, because of electrostatic, repulsive interactions occurring at the local, molecular scale. KEYWORDS: bacterial adhesion, surface properties, functional group density, chemical heterogeneities, real-time confocal microscopy

1. INTRODUCTION Despite many research studies over three decades, biofilm formation on surfaces still remains a major source of problems and limitations in the biomedical field. In particular, infections developing on medical devices have severe impact on patient health.1−3 They are a frequent cause of implant removal because of the difficulty, if not impossibility, of treating biofilmrelated infections by systemic antibiotic treatments. Biofilm formation on surfaces is a complex process that involves a large number of factors, including features of the bacterial species and strains, chemical and physical characteristics of the surroundings, and material surface properties and characteristics.4−6 Material properties like surface topography, mechanical properties, and chemistry are known to affect bacterial adhesion. So far, however, the complex relationships between bacterial adhesion and surface properties remain largely unknown. Regarding material surface chemistry, in particular, many studies exist that report influences of the hydrophobic/hydrophilic and charge characters of the material surface7−10 or impact of various chemical functionalities11−13 on bacterial adhesion. However, investigations so far have failed to provide a general, integrative view of the already available results. The large variety of properties that are revealed by bacteria and the large diversity of experimental approaches used to realize the studies are obvious causes of some apparent contradictions. Nevertheless, an additional, important hurdle might arise from insufficient consideration of the material surface properties at the molecular scale. In other words, although effects are usually attributed to macroscopic, hydrophobic/hydrophilic, or charge, characters of the surface, © 2015 American Chemical Society

chemical heterogeneities of the surface, and accessibility of chemical functional groups may also intervene and control the interface between bacteria and the surface. Only a few publications have addressed the influence of chemical heterogeneities of surfaces on bacterial adhesion.14−18 In these publications, the impact of various mixtures of functional groups was investigated by using self-assembledmonolayers (SAMs)-based model surfaces that allowed the control of the relative amounts of each functional group at the surface. Despite the indisputable interest of these studies, their scope is limited by an incomplete description of the SAMscoated surfaces. Moreover, the weak yet real, and potentially crucial, difference in length of the various molecules that constituted a mixed molecular layer may have contributed to the results, which is prone to interpretation bias. As evoked by Wiencek and Fletcher,17 the resulting molecular topography may make some chemical functionalities of the side-chain accessible to bacteria, which are different from the terminal functional groups. Side-chain effects may also be induced by additional molecules sometimes used as a linker between substrate and molecules of interest, and that may be accessible to bacteria as reported by Barth et al.14 Chemical surface properties and therefore molecular accessibility are incompletely controlled. Consequently, effects due to terminal functional groups, side-chain groups, and molecular topographical properties cannot be clearly decoupled. Therefore, dedicated Received: March 26, 2015 Accepted: June 11, 2015 Published: June 11, 2015 693

DOI: 10.1021/acsbiomaterials.5b00151 ACS Biomater. Sci. Eng. 2015, 1, 693−704

Article

ACS Biomaterials Science & Engineering

Figure 1. Description and characteristics of the chemically mixed model surfaces. (a) Simplified representation of a Br/CH3 75% mixed monolayer. Three molecules of 11-Br-undecyltrichlorosilane and one of trichloroundecylsilane are grafted on silicon wafer. (b) Simplified top view of five mixed monolayers revealing various densities of Br or NH2 backfilled with CH3. Initial Br/CH3 surfaces are converted in NH2/CH3 surfaces via N3/CH3 surfaces. (c) Characteristics of the five final NH2/CH3 surfaces (with theoretical surface fractions of 0, 25, 50, 75, and 100%, respectively) in terms of NH2 content, water contact angle, and ellipsometric thickness.

research efforts are still required to both develop adequate molecular model-surfaces and investigate the role of material surface heterogeneities in bacterial adhesion and further surface-associated proliferation. Recently, we have developed highly controlled NH2/CH3 mixed SAM-based platforms, reported a comprehensive description of their properties, and demonstrated their relevance for bacterial adhesion studies.19 Made from mixed SAMs of two silanes with methyl and bromine end groups, respectively (Figure 1a), which are further transformed to amino/methyl groups (Figure 1b), these surfaces provide smooth (topography-free) platforms for further use. Surfaces were thoroughly analyzed to achieve a comprehensive description of their chemical composition, layer structure and molecular organization. Extensions of short-term bacterial colonization on 100%/0% NH2/CH3 terminated surfaces obtained by this Brorganosilane approach and on 100%/0% NH2/CH3 terminated surfaces obtained by the direct grafting of 3-Aminopropyltriethoxysilane (APTES) or N-(6-Aminohexyl)-3-aminopropyltrimethoxysilane (AHAPS) were compared, showing that the quantity of adhered bacteria was the least dispersed, i.e., the most reproducible, on Br-organosilane-based SAMs. This demonstrated for the first time that the presence, even rare and discrete, of secondary chemical groups on the top of one layer composed of another chemical group, predominant in terms of quantity, can significantly act on the bacterial adhesion onto the layer. Thus, the potential impact of the conformation of the grafted or deposited molecules, their organization at the surface, and the potential accessibility to side-chain groups, all leading to chemical heterogeneity at the extreme surface, was suggested. In the present work, we investigated the impact of chemical heterogeneities of a surface on its colonization by bacteria.

Aware of the crucial necessity to prevent potential unexpected side-effects of surface properties, the study was conducted on the previously developed NH2/CH3 model-substrates based on Br-organosilane-based mixed monolayers. Behaviors of bacteria were analyzed during and following their adhesion to the surfaces, and two different bacteria species (Escherichia coli and Staphylococcus epidermidis) were considered aimed at addressing the variety of chemical composition of the bacterial outer membrane, especially owing to the Gram-negative or Grampositive class. Variation in the liquid medium (selective growth medium versus NaCl 9 g/L solution) and the hydrodynamic flow (dynamic versus static condition) used during the adhesion experiment enabled the potential impact resulting from specific, surrounding culture conditions that are likely to interfere with the physical and chemical interactions between surface and bacteria to be addressed. Microbiological investigations were conducted under fluorescence confocal microscopy in batch and real-time observations.

2. MATERIALS & METHODS 2.1. Mixed Monolayer Preparation and Characterization. Preparation of the model surfaces was done as previously described.19 Briefly, silicon wafers (100), purchased from Mat Technology (France), polished on one side, were cut into 1 cm × 1 cm pieces and cleaned by 10 min immersion in CHCl3 (Sigma-Aldrich) and ultrasound treatment (frequency 45 kHz). Such cleaned, nonfunctionalized silicon wafers were used as an internal control. For further mixed monolayer preparation, various ratios of 11-Brundecyltrichlorosilane and trichloroundecylsilane (ABCR, Germany) were used to create SAMs-coated surfaces of five different Br/CH3 surface fractions (theoretical values of 0, 25, 50, 75, and 100%). The terminal bromine groups were commuted into azide by nucleophilic substitution reaction and azide-to-amine reduction was performed subsequently as described elsewhere.19 Samples were finally rinsed 694


Article

ACS Biomaterials Science & Engineering Table 1. Bacteria Strains Used in the Present Studya bacteria strain

ATCC no/ references

description

Escherichia coli SCC1

MG1655 with chromosomal insertion of PA1/04/03/_gfpmut3b CmR

20

Staphylococcus epidermidis RP62A

slime producing coagulase-negative- staphylococci MetR, ClinR, EryR, GenR CMS, TetS

ATCC 35984 23 24

estimated outer membrane characters

refs

from high to moderate hydrophilic high electron donor from no to moderate electron acceptor weak hydrophilic high electron donor from no to weak electron acceptor

22b 12b 25c 26d 27e

a

Physical-chemical properties of their outer membrane (hydrophilicity/hydrophobicity, electron donor and acceptor characters) were estimated on a bibliographic basis (see references in table). bEstimation based on the microbial adhesion to solvents (MATS) method.28 cEstimation based on salt agglutination.29 dEstimation based on the microbial adhesion to hydrocarbons (MATH) method.30 eEstimation based on contact angle measurements.31 observed in the last rinsing solution by using a fluorescence/reflection mode upright confocal microscope (LSM700, Carl ZEISS) equipped with a long working distance objective (9.1 mm). On each surface, micrographs were randomly taken on 10 different locations. Micrographs were analyzed by CellC33 and ImageJ V.1.44d software with LSMtoolbox V4.0 g plugins34 for determining the quantity of adhered bacteria. Each experiment was conducted with two surfaces of each NH2/CH3 surface fraction and experiments were repeated three times. Average and standard deviation of adherent bacteria numbers were normalized to the average of data obtained for internal control surfaces in the same experiment. Results depicted in Figures 2 and 4 are

with Milli-Q water and ethanol before drying under nitrogen stream. Before microbiological experiments, we sterilized grafted surfaces by immersing for 30 min in 70% ethanol in Milli-Q water solution, before rinsing in sterile Milli-Q water. The mixed monolayers were thoroughly characterized and analyzed elsewhere.19 The characterization aimed at providing the comprehensive description and knowledge of the mixed monolayers. Briefly, surface fractions of both functional groups grafted on the surface, layer structures and molecular organization in the grafted layer were evaluated at each grafting and chemical conversion step. For that purpose, XPS (survey and high-resolution spectra) measurements and analysis using CasaXPS 2.3.12 software (Casa Software Ltd., Teignmouth, UK, www.casaxps.com) were made to determine atomic concentrations and chemical environments. A DSA100 goniometer was used for static and dynamic water contact angle measurements. Ellipsometry (M-O33K001) at 532 nm provided thickness measurements of the grafted layers in both air and liquid conditions. 2.2. Bacteria Species and Culture. Bacterial experiments were conducted with two bacteria species: the fluorescent, slime- and GFPproducing Escherichia coli (E. coli) SCC120 and the slime-producing Staphylococcus epidermidis (S. epidermidis) ATCC35984 (RP62A; purchased by Prof I. Spilioupoulo, Patras, Greece) (Table 1).

average and standard deviation of the so-calculated colonization compared to internal control (n = 60). Significant two-by-two differences between the average of colonization on the diverse NH2/CH3 surface fractions (μX%, X = {0;25;50;75;100}) were evaluated for normalized numbers of adherent bacteria, by bilateral Student’s t tests (application conditions: independent data and equal variances assessed by F-test) with significance thresholds (α) of 0.01 and 0.05. According to,35 the alternative hypothesis (H1: μX% ≠ μY%) was assumed to be true when the main hypothesis (H0: μX% = μY%) was rejected.

Here, −80 °C frozen bacteria were cultured overnight at 30 °C on Luria−Bertani (LB purchased from Sigma-Aldrich) agar plate and at 37 °C on Brain-Heart-Infusion (BHI purchased from Sigma-Aldrich) agar plate for E. coli and S. epidermidis, respectively. First, precultures were prepared with one colony of E. coli and S. epidermidis in LB or BHI, respectively, and incubated overnight (about 18 h) at 30 and 37 °C, respectively. Second, cultures were prepared with 10% of the overnight cultures in LB or BHI for E. coli and S. epidermidis, respectively. Bacteria were harvested by centrifugation. According to the experiment (see the next session), bacteria pellets were resuspended either in NaCl (150 mM; pH 6.8) solution or in the so-called M63G-B1 E. coli-selective medium (pH 6.8).21 Bacterial suspensions were adjusted to an absorbance at 600 nm (A600 nm) of 0.01 (5 × 106 CFU mL−1) and 0.1 (5 × 107 CFU mL−1) for static and dynamic experiments, respectively.

2.4. Evaluation of Bacterial Adhesion and Detachment under Live Imaging. A specific flow cell was designed for experiments conducted under the Zeiss LSM700 confocal microscope (Figure S1). Flow cells were made with the support of the Technical University of Denmark, Lyngby, Denmark36 and were displayed elsewhere.37 Sterilized flow cell and accessories were connected together in sterile conditions to form a closed flow setup (Figure S2). The complete setup was placed under the confocal microscope in a thermoregulated chamber (Okolab, Germany) allowing a constant temperature of 30 °C for E. coli and 37 °C for S. epidermidis. A 0.47 s−1 shear rate at sample surface was used throughout the experiment corresponding to a 2.6 mL/min flow. The pump was run at least 1 h to fill the system with NaCl 9 g/L solution and ensure flow and temperature stability. Then, 10 mL of bacteria suspension (A600 nm of 0.1) was injected for 4 min before closing the system for 5 min. The system was then open and fresh NaCl 9g/L solution was injected for 9 min for the elimination of planktonic bacteria. At the sample surface, these experimental procedures allowed the adhesion of injected bacteria for around 15 min. The system was then closed and pictures were taken every 2 min over 150 min in the same location. The fluorescence and reflection microscopic modes were used to visualize E. coli and S. epidermidis,, respectively (Figure S3). Resulting images were analyzed by the Particle Tracker of ImageJ V.1.44d software with LSMtoolbox V4.0 g34 and particle tracking38 plugins, leading to the calculation of number of adhered bacteria per micrograph versus time. Because a study in dynamic conditions needs long-lived experiments for each individual material surface sample, statistical reproduction has been fixed at two replicates per condition (n = 2). Due to less sensitivity to statistical variations, the bacterial adhesion rate was preferred to absolute value of adhered bacteria as indicators for comparing bacterial behavior on surfaces of different NH2/CH3

2.3. Evaluation of Bacterial Adhesion by Microbiological Batch Experiments. Surfaces with various NH2/CH3 surface fractions (theoretical values of 0, 25, 50, 75, and 100%) were immersed in 0.1 A600 nm E. coli suspensions prepared in NaCl 9 g/L or M63G-B1 medium, or a 0.1 A600 nm S. epidermidis suspension, prepared in NaCl 9 g/L. As an internal control, which aims to normalize experimental data and further compare experimental replicates, a cleaned, nonfunctionalized silicon wafer was used. Surfaces were incubated for 2 h at 30 and 37 °C for E. coli and S. epidermidis, respectively. Surfaces were thoroughly and carefully rinsed six times with NaCl 9 g/L solution to remove nonattached bacteria without creating air-surface interface.32 S. epidermidis adhered on substrates were fluorescently stained by adding 1 μL mL−1 of a 5.10−3 M Syto9 (Invitrogen) stock solution to the last NaCl rinsing solution. Surfaces were directly 695


Article


content of the solution used for the surface elaboration, and the Br concentration detected on the surface. Similar XPS analysis conducted on the azide and amine surfaces after SN2 bromineto-azide and azide-to-amine conversions, respectively, showed a similar linear relationship between Br-organosilane solution content and azide or amine concentration detected on the surface, showing that conversions did not affect the monolayer properties. This finally demonstrates the possibility to control the NH2/CH3 surface fraction by the initial Br-organosilane solution content, as shown in Figure 1c. The thickness of the grafted layers was measured by ellipsometry in a range of 1.3−1.8 nm in liquid (Figure 1c) and 1.3−1.7 nm in air. Very good agreement with the expected thickness (1.6 nm) demonstrates that the coating is structured as a monolayer whatever the NH2/CH3 surface fraction. This was confirmed by the low Br 3d/SiOx surface fraction that was determined by high-resolution XPS analysis (0.15 ± 0.02 compared to 0.42 ± 0.05 for a typical multilayer19) with the SiOx signal coming from the native silicon oxide layer present under ambient conditions at the surface of all silicon substrates. Moreover, the good agreement between wettability measurements (59 ± 2° and 108 ± 1° for NH2/CH3 100% and NH2/CH3 0%, respectively) (Figure 1c) and expected values39 demonstrated that silane molecules were densely packed and organized as expected for SAMs, which was also indicated by the similarity of thickness measured in air and liquid. Finally, this comprehensive description of layer structure, molecular organization and chemical composition of the extreme surface of chemically mixed coating allows us to ensure, first, that NH2 and CH3 groups were the only, accessible functional groups at the extreme surface, and second, that NH2 and CH3 groups were present at the surface in the quantity expected because of the Br-silane solution used for surface elaboration. 3.2. Relationship between Bacterial Adhesion in Static Conditions of Culture and the Surface Content in Chemical Groups. E. coli and S. epidermidis bacteria strains, both showing hydrophilic and electron donor characters on their outer surface (Table 1), were cultured in static conditions on mixed monolayer model surfaces of five different NH2/CH3 surface fractions. When experiments were carried out in NaCl 9 g/L (ionic strength, IS = 0.15 M), E. coli and S. epidermidis bacteria strains both significantly and preferentially adhered on NH2/CH3 0% (i.e., CH3 100%) than on NH2/CH3 100% surfaces. Furthermore, CH3-rich surfaces were in general more colonized than surfaces revealing a lower quantity of CH3 groups (Figure 2a, b). Such general, direct decreasing trend of the number of adherent bacteria (NAdhBact) with CH3 content is in agreement with results already reported,40 including for bacteria strains with hydrophilic outer membrane properties.41 In particular, Wiencek and Fletcher16 and Burton et al.15 observed higher adhesion on CH3-rich surfaces than on more hydrophilic surfaces (OH-rich surfaces in both studies) by using mixed OH/CH3 monolayers. However, these studies were conducted with other bacteria strains (Pseudomonas sp. and E. coli RP437 respectively) and medium (Instant Ocean Marine salts, IS = 0.48 M, and LB medium, IS = 0.18 M, respectively) than the present study. Using identical S. epidermidis bacteria as in the present study but chemically homogeneous yet non model surfaces, Katzikogianni et al.27 also reported similar general trend of surface colonization: surfaces that mainly revealed CH3-, NH2- and OH-groups were respectively colonized with decreasing bacteria amounts, in phosphatebuffered saline (PBS) of two various ionic strengths (0.01 M,

Figure 2. Number of (a) E. coli and (b) S. epidermidis bacteria adhered on mixed monolayers in NaCl 9g/L medium and static culture conditions, related to internal control. Colonization of internal control with E. coli and S. epidermidis was 2.2 ± 0.1 × 106 bacteria/cm2 and 3.5 ± 0.2 × 106 bacteria/cm2, respectively. §: Significant difference (α = 0.01) compared to NH2/CH3 0%; #: Significant difference (α = 0.05) compared to NH2/CH3 100%. Values demarcated with dotted lines have similar values and form plateau-like zones, separated by transition zones. contents. Adhesion rate was calculated for the first 20 min of experiment as the slope of the linear regression of the 10 first non-null values for each experiment (Supporting Information, Figure S3), which is a time range with quasi-linear evolution of the number of adhered bacteria. Adhesion rate is proportionally related to the number of adhered bacteria at a given time, which finally allows the comparison of trends obtained in dynamic and static conditions. Mean frequencies of attachment or detachment events, and total numbers of bacteria that were mobile between surface and suspension (i.e., that either attached or detached) throughout the experiment were also determined.

3. RESULTS & DISCUSSION 3.1. Mixed Monolayer Model Surfaces. The mixed monolayers were obtained by mixing two silanes with methyl and bromine end groups, respectively (Figure 1a), and by further in situ transformation including SN2 reaction to convert bromine to amino nucleophilic groups (Figure 1b).19 Their thorough analysis was mainly based on the determination of Br/SiOx, Br/CH3 and NH2/CH3 surface fractions, layer thickness, layer compactness and wettability. As previously reported,19 High-Resolution XPS analysis of the elaborated surfaces revealed a good correlation between the Br-organosilane 696


Article


heterogeneity of physiological state51−54 in the bacterial population, and were compared to those obtained for the inoculation bacterial suspension (Figure S4). No significant differences were observed between populations adhered on the different NH2/CH3 mixed surfaces displaying similar length and width characteristics. Therefore, populations on the different NH2/CH3 mixed surfaces were hypothesized to not be significantly different. The second hypothesis is that nonlinear bacterial adhesion vs NH2/CH3 surface fraction relationship rather results from the nonproportional variation in the relative contribution of the diverse interactions (hydrophobic, van der Waals, electrostatic, etc.) occurring between bacterium and NH2/CH3 surface when the NH2/CH3 mixed surface content varies. In part, the nonlinear decrease in the attractive electrostatic interactions between bacterium and NH2/CH3 surface due to the increase of NH2/CH3 mixed surface content may be involved. Indeed, as evoked above, only hydrophobic and van der Waals attractive interactions exist on NH2/CH3 0% (Figure 3b), between bacterium membrane and hydrophobic, apolar and uncharged material surface. Nonetheless, competition for the surface between bacterium membrane and aqueous medium is favorable to close bacterium/surface contact, although the compound layers of adsorbed water and counterions surrounding the charged, hydrophilic membranes of both bacteria strains used in this study12,27,50 may induce steric hindrance and, consequently, slightly limit this contact. Close bacterium/surface contact leads to stronger van der Waals interactions, and therefore to stable adhesion of bacterium to CH3 100% (i.e., NH2/CH3 0%) surfaces in static culture conditions, i.e., in the absence of any external constraints on the bacterium/surface contact other than Brownian forces.55 On NH2/CH3 X% (X ≠ 0) surfaces, electrostatic interactions between the negatively charged bacterial membrane and positively charged NH3+-revealing surfaces at the working pH (∼6.8) led to additional attractive interactions between bacterium and the surface. Adhesion of bacterium to NH2/CH3 X% (X ≠ 0) surfaces is therefore expected to be favored in comparison to NH2/CH3 0% surfaces. However, layers of counterions at the bacterial and the NH2/CH3 surfaces (due to electrostatic interactions between the charged surfaces and ions present in the surrounding liquid)42,50 may reduce the electrostatic attraction length (Debye length) by screening both NH3+ and negative bacterium charges. Furthermore, close bacterium/surface contact is limited by the presence at the surface of a layer of adsorbed water molecules due to the hydrophilic and polar characters of the NH2-rich surfaces. The increased density of NH2 groups on NH2/CH3 mixed surfaces primarily and proportionally affects the lateral expansion of the screening layer as a 2D continuum, which initially starts from discrete, closely packed “multimolecular” hydration clusters that form at low concentrations of isolated NH2 groups and/or clusters of NH2 groups. With increasing NH2 group surface fraction, these discrete multimolecular hydration clusters expand and connect laterally to form a 2D continuum of hydration layer. Although an effective growth of the overall thickness of this hydration layer is then expected and favored by the coherence of this 2D close-packing, there is a priori no theoretical basis to support that the thickness of the hydration layer should linearly increase with surface fraction of polar/charged NH2−NH3+ groups. Therefore, a nonlinear decrease in attractive interactions and, subsequently, a nonlinear reduction in NAdhBact might be expected when NH2/CH3 content increases. In addition, the density of the positive charge at the bacterial outer membrane may also impact bacterial adhesion on

IS = 0.17 M, and 0.10 M, IS = 1.70 M). This general trend is usually interpreted as the result of hydrophobic and apolar characters of CH3-rich surfaces, which leads to predominant van der Waals interactions between CH3-rich surface and bacterium. On the contrary, hydrophilic and polar characters of NH2-rich surfaces result in a layer of adsorbed water, hindering short-range interactions between NH2-rich surface and bacterium.16,42,43 Nevertheless, hydrophilic and polar NH2-rich surfaces also exhibit positive charges (NH3+) at pH inferior to the Iso-Electric Point (IEP) of immobilized NH2 groups (IEP = 7.3−7.6 (44−48)), resulting in attractive electrostatic interactions between negatively charged bacteria27,49,50 and positively charged NH2-rich mixed surfaces. This may favor the adhesion of bacterium to the NH2-rich surface, in relation to the quantity of NH2-groups present at the surface, and consequently reduce the differences in colonization between CH3-rich and NH2-rich surfaces resulting from hydrophobic interactions, or may even reverse the trend. Eventually, the observed trend is the complex result of several opposite effects, i.e., attractive hydrophobic interactions on CH3-rich surfaces, prevention of bacteria/surface close contact due to the hydration layers and electrostatic interactions on positively charged NH2-rich surfaces. These are mainly attractive at the microscopic level of the bacterium, due to the global negative charge of the bacterial surface, but may also become repulsive at the molecular level when bacterium and the NH2-rich surface are close to contact allowing the dilute positive charges of the bacterial membrane and of the NH2-rich surface to interact. Compared to on CH3-rich surfaces where they are absent, these repulsive electrostatic interactions, even if low, may impact the global interaction between bacterium and NH2-rich surfaces, reducing the relative contribution of the negative charges of the bacterial membrane in the bacterium/NH2-rich surface interaction, and finally participating in lower colonization of NH2-rich surfaces compared to CH3-rich surfaces. Nevertheless, in contrast to the most commonly accepted opinion in the bacterial adhesion scientist community, the present study shows a nonlinear relationship between NAdhBact and NH2/CH3 surface content (Figure 2) in the range from 0% to 100% NH2/CH3 for both bacteria species. Such nonlinear relationship, with plateaus and transition zones, is reported for the first time, in contrast to the linear relationships reported by Burton et al.15 and Wiencek and Fletcher16 in static and dynamic conditions, respectively. Zones are related to different ranges of NH2/CH3 surface fraction and are differently marked for E. coli (Figure 2a) and S. epidermidis (Figure 2b). As shown in Figure 2, experimental curves for both bacteria strains can be fitted by linear functions highlighting zones of stability besides transition zones (ca. 50% of NH2/CH3 for E. coli, ca. 10% for S. epidermidis). For both bacteria species, adhered bacteria amounts were reduced to about 20% for NH2/CH3 surface content higher than these thresholds. One hypothesis is that such differences in the amount of bacteria adhered on different groups of NH2/CH3 mixed surfaces may arise from the selection, during the adhesion process, of different bacteria populations coming from the inoculation bacterial suspension. According to the global, physical-chemical properties of both bacterial outer membrane and NH2/CH3 mixed surface, bacterial population adhered on 0 and 25% NH2/CH3 mixed surfaces may differ from population adhered on 50, 75, and 100% NH2/CH3 mixed surfaces. To test this hypothesis, we measured width and length of bacteria adhered on the diverse NH2/CH3 mixed surfaces as an indicator of 697


Article


Figure 3. Illustration of the main hypothesis leading to plateaus and transition zones in NAdhBact versus X% (i.e., NH2/CH3 surface fraction) relationships. (a) Scheme depicting interactions between the bacterial membrane and the material surface placed in static conditions in NaCl 9 g/L medium. (b−e) Membrane of Gram-negative bacterium is schematized by only depicting the phospholipid bilayer (negative charge in green) and two membrane proteins (positive charge in red). Four different situations of interactions between bacterial membrane and NH2/CH3 mixed surface are depicted, corresponding to four specific density or range of density of positive charge on NH2/CH3 surfaces: X = 0%, 0% < X < X0, X = X0, X > X0, where X0 is a critical density related to the density of positive charge of the bacterial membrane. The red dashed area is depicts the solid angle in which the positive charge of the bacterial surface may interact with positive charge of the material surface.

the largest amount (highest density) of accessible groups on bacteria membranes. In contrast, positive charges, originated from the sparse, peptidic structures in both Gram-negative and Gram-positive bacteria membranes, result in a discrete distribution of positive charges on the membrane at physiological pH less than IEP of NH2 group. We therefore assume that specific conditions of bacterium/surface contact exist in which these positive (amino) membrane charges and the ammonium (NH3+) of the supporting surface can create, beyond a certain surface fraction of NH2/CH3, repulsive electrostatic interactions, which increase in intensity and start to balance the attractive and stable bacterium/surface contact. This hypothesis is depicted in Figure 3c−e. Electrostatic interactions at the bacterium/material surface are the result of the competition between (i) the overall, attractive interaction between the continuous distribution of negative charges of the bacterial membrane and the positively charged amino groups of the material surface, and (ii) local and discrete, repulsive interactions between positively charged chemical groups, sparsely distributed at the bacterium surface, and the positively charged amino groups of the material surface. Because of the decreasing intensity of the repulsive Coulomb force with the inverse square of the distance between the interacting charges, only positively charged groups of the material surface that are within a finite lateral size r around the normal distance (d0) to the positive charge of the bacterial surface (d ≤ d0, r = r (φ), Figure 3b) may interact with the positive charge of bacterium, and contribute to the repulsive electrostatic interaction. The distance r defines a critical area on the material surface facing a

NH2/CH3 mixed surfaces in a nonlinear way, leading to transition zones at specific, threshold-like values of surface content and delimitating two different behaviors of bacteria in term of their adhesion to the surface (reduction of about 20% of adhered bacteria amount when NH2/CH3 surface fraction is higher than the critical threshold mentioned above). This process certainly involves the intrinsic chemical heterogeneity of the bacteria outer membrane (Figure S5): Whatever bacteria species, the main components of outer wall of Gram-negative bacteria like E. coli are phospholipids that provide, through phosphate groups, the most important contribution to the total negative charge of the bacterial outer membrane at pH close to the physiological value (∼7).50,56 Aside from other negatively charged groups like carboxylic acids, amino groups resulting from the proteins of the cell membrane are also present at the bacterial surface, and are responsible for positive charges at the membrane surface at pH less than IEP of NH2 groups.44−48 The cell wall of Gram-positive bacteria like S. epidermidis is composed of a peptidoglycan macromolecule containing a large quantity of attached molecules like teichoic acids, teichuronic acids, and polyphosphates.57 The overall, negative charge of the Gram-positive bacteria membrane is originated at physiological pH from these teichoic and teichuronic acids and polyphosphates, rich in phosphate groups. Besides, peptides are linked to the glycan chains showing free amino residues and, therefore, positive charges at pH less than IEP of NH2 group. Therefore, whatever the Gram-class of the bacteria, phosphate groups of the membrane provide at physiological pH a quasicontinuous distribution of negative charges since they represent 698


Article


So far, no published work has reported, or even addressed, evidence of a critical density of (or critical distance between) surface functional groups that may be involved in nonspecific bacterium/surface interactions. More generally, very little is known regarding the critical density of (or distance between) grafted molecules that may govern the specific or unspecific adhesion of bacteria at the molecular scale. On the contrary, the distance between ligand molecules grafted on surfaces has been highlighted as controlling binding force of adherent eukaryotic cells.59 The distance between arginine-glycine-aspartic acid peptides (RGD peptides), well-known to be involved in cell adhesion to surfaces, was shown to determine the separation of individual integrins and thus the assembly of the integrin clusters that were only stable when ligands were spaced below a critical value.60 Finally, the development of stable focal adhesions, their number, and size as well as the cellular adhesion strength was proved to be influenced by the local more than the global ligand density. The ability of bacteria to bind single molecules at the surface of biological tissues through ligand/receptor interaction61 leads us to expect similar involvement of the individual functional groups of a material surface: Variations at the molecular scale in the surface distribution of grafted molecules like carbohydrate residues may influence bacterial adhesion. The multivalence character of bacteria/ mannose specific interactions, involving FimH adhesins located on E. coli pili,62 is an example of bacteria/host interactions in which the binding strength is probably highly impacted by the distance between mannose residues. This is also suggested by the higher affinity of bacteria for branched oligomannosides of specific size.14 3.3. Effect of the Culture Medium on the Relationship between Bacteria Adhesion and the Surface Content in Chemical Groups. The number of adhered E. coli bacteria on mixed monolayer model surfaces of five different NH2/CH3 surface fractions were compared in static conditions in two different liquid media (NaCl 9 g/L, IS = 0.15 M; M63G, IS = 0.20 M). A significant difference in the NAdhBact vs NH2/CH3

positive charge of the bacterial surface, and a critical interaction distance d ≥ d0, through the solid angle φ, r = r (φ). We propose therefore that, due to the discrete distribution of the positive charge in a quasi-negatively charged continuum at the bacterium surface, a critical density of positive charges on the material surface (X0) exists, above which at least one positive charge of the material surface is located within the interacting area of each positive charge of the bacterial membrane. This specific situation where the density of interacting positive charges on material and bacterium surfaces (X ≈ X0, Figure 3d) increases and becomes important can lead to a repulsive contribution that is able to compete with the overall attractive, electrostatic interaction between bacterium and material surface. Up to X0 (Figure 3c), only very rare positive charges of the material surface may be located close to positive charges of the bacterium surface. Therefore, the overall, resulting, repulsive interaction at the bacterium scale does not significantly contribute to the overall bacterium/material surface interaction. As soon as the surface density of positive charges on material surface allows at least one or more of these charges to be within the interacting area with the positive charges of the bacterial membrane (X ≈ X0, Figure 3d) on bacteria/material contact, the overall, attractive interaction between bacterium and material surface is significantly reduced. For X > X0 (Figure 3e), the repulsive interaction intensity that involves one bacterial positive charge at X ≈ X0 increases proportionally to the number of positive charge of the material surface involved in the interaction. However, because of the probably small dimension of the critical area matching with the interaction range of the positive charge of bacterial surface, the quantity of positive charges of the material surface that contributes an increase of the electrostatic repulsive force between bacterium and material surface is therefore finite. The quantity of adhered bacteria observed on surfaces with different NH2/CH3 surface fractions, all higher than X0, is consequently found to be roughly constant for both bacteria (Figure 3e). Obviously, additional effects should be expected from the layers of counterions formed on positively charged material surfaces. The thickness of these layers increases with the density of positive surface charges, reducing the strength of the repulsive interactions and therefore leading to the weak yet insignificant decrease of NAdhBact in the NH2/CH3 surface fraction ranges of the plateau zones. Eventually, the density of positive charges on the bacterial membrane is a decisive factor in the definition of X0. Any increase or decrease of this density reduces or rises the strength of the repulsive interactions, shifting the range of the transition zone as observed for the two bacterial species Indeed, although plateaus and transition zones were observed for both E. coli and S. epidermidis bacteria, some differences were observed. In particular, the transition zones correspond to different ranges of NH2/CH3 surface fraction (Figure 2a, b). This might be related to the differences in distribution and quantity of positive charges that exist on the bacterial outer membranes of both bacterial species. In particular, higher N/C ratios have been reported for Gram-positive microorganisms, especially for Staphylococci compared to E. coli.56 Although nitrogen of Staphylococci also results from amide bonds of peptidoglycan,58 the presence of a high quantity of amino residues, i.e., peptides linked to the glycan chains,57 in probable higher quantity than proteins at the E. coli surface, may explain that X0 (∼10%) of S. epidermidis is significantly smaller than of X0 (∼50%) of E. coli.

Figure 4. Number of E. coli bacteria adhered on NH2/CH3 mixed monolayers of five different NH2/CH3 contents after inoculation and culture in NaCl 9g/L and M63G culture medium,21 and static culture conditions. Colonization of internal control in NaCl 9g/L and M63G were 2.2 ± 0.1 × 106 bacteria/cm2 and 2.1 ± 0.1 × 106 bacteria/cm2, respectively.*: Significant difference (α = 0.01) between M63G and NaCl 9 g/L conditions. 699


Article


Figure 5. Illustration of the modifications in close contact and subsequent interactions between bacterial membrane and NH2/CH3 surface by changing the surrounding medium. Scheme depicts interactions between the bacterial membrane and the material surface in static conditions, in(a, c) NaCl 9 g/L and (b, d) M63G medium.21 The membrane of Gram-negative bacterium is schematized by only depicting the phospholipid bilayer (negative charge in green) and two membrane proteins (positive charge in red). Two different situations of interactions between bacterial membrane and NH2/CH3 mixed surface are depicted, corresponding to mixed surfaces with X = 0% and X > 0% NH2/CH3 surface fractions. ta, tb and t′a, t′b are thickness of the counterion layers in NaCl 9g/L and M63G media, on cell membranes and material surfaces, respectively.

reduced in M63G compared to NaCl 9 g/L. However, this should a priori have affected both attractive and repulsive interactions in a similar way, leading to insignificant modification of adhered bacteria number. 3.4. Relationship between Bacteria Adhesion in Dynamic Conditions of Culture and the Surface Content in Chemical Groups. Hydrodynamic conditions used in published studies mentioned above15,16,27 highly differed from one study to another (Table S1). When hydrodynamic flow was applied,16,27 the shear stress was high enough to significantly affect the bacterial metabolism through direct flow-related stress.63 In the present study, the flow chamber was designed to produce low, yet laminar and well-controlled shear stress on the surface. This prevents from flow-related modifications of the bacterial metabolism and therefore allows the study of the single effect of shear stress on the adhesion and proliferation of bacteria to surfaces. This methodological precaution is rendered essential by the potentially strong sensitivity of bacterial behavior to very weak changes that are expected to arise from slight variations of the material surface chemical composition. E. coli and S. epidermidis bacteria strains were cultured in NaCl 9 g/L under a shear rate of 0.47 s−1 on the mixed monolayer model surfaces of the five different NH2/CH3 surface fractions. Figures 6a and 6b show that adhesion rate versus NH2/CH3 surface content relationships do not have identical transition zones as adhered bacteria number versus NH2/CH3 surface content relationships determined in static conditions, both for E. coli and S. epidermidis bacterial species. Nevertheless, one particularity of these relationships deserves being underscored: Although these relationships are far from linear, the general

surface fraction relationships was observed. Contrary to what was observed in NaCl 9 g/L, NAdhBact increases regularly, though only slightly, with NH2/CH3 surface fraction in M63G medium (Figure 4). Interestingly, NAdhBact measured on surfaces with a high NH2/CH3 surface fraction (X ≥ 60%) were similar in NaCl 9 g/L and M63G media, whereas NAdhBact related to low NH2/CH3 surface fraction (X < 60%) was significantly lower in M63G medium than in NaCl 9 g/L. This suggests that the difference between the two situations mainly arises from differences in interactions involving CH3-groups. Indeed, as depicted in Figure 5, counterion layers at the bacterium and charged material surfaces should have various thicknesses in NaCl 9 g/L and M63G mainly because of the size of the ions present in the media (100 mM of K+ and NH4+, and 10 mM of PO4− and SO42− in M63G), since valence of the main ions and IS were similar. On CH3 100% surfaces (NH2/CH3 0%), the increase in bacterium/material surface distance due to counterion layers at the bacterium surface should have reduced close contact, reducing both attractive hydrophobic interaction and van der Waals forces between bacterium and material surface (Figure 5a, b). It therefore appears that a bacterium approaching a surface is highly sensitive to tiny external perturbations, resulting here in less stability of the bacterial adhesion, and finally in less immobilization of bacteria in M63G than in NaCl 9 g/L. On surfaces containing NH2 groups (NH2/CH3 X %, X > 0), adhesion of bacteria was probably affected by the difference in thickness of counterion layers occurring both at the bacterium and charged material surfaces, which reduced the close contact between bacterium and material surface in M63G compared to NaCl 9 g/L (Figure 5c, d). Electrostatic interactions were therefore probably slightly 700


Article


resists the hydrodynamic shear, but only its tangential component, the adhesive frictional Ff, which is related to the normal adhesion force Fnadh through the apparent frictional coefficient μ* of the interface, Ff ≈ μ*(Fnadh. + Lbact.), Lbact. being the negligible load of the bacterium in the culture medium,64 corrected here by the gravity difference. Dimensionally, when reported to the square of the bacterium/surface contact area (CA), this is equivalent to a tangential resisting adhesive pressure (stress), Padh≈ Ff/(CA), which directly compares with the driving hydrodynamic shear.64 One can reasonably imagine that the adhesion arising from these attractive but low strength physical forces may be insufficient to resist the driving (detaching) shear force produced by the hydrodynamic flow. Therefore, the probability of the bacterium being detached and forced to slide (or roll) off the surface is high, consequently leading to a low probability of immobilization and low adhesion rate of the bacterium on the material surface (Figure 7a). On NH2/CH3 X% (X ≠ 0) surfaces, attractive electrostatic interactions additionally contribute to increase the attractive and resisting forces, achieving a level that can balance the driving/detaching shear-force. The overall result in this case is a decreased probability for bacterium detachment and its sliding (or rolling) entrainment in the flow (Figure 7b). Therefore, one expects bacterium to be more easily immobilized on such surfaces, and their adhesion to be more stable against flow. This is confirmed by the frequency of attachment or detachment events during the growth phase of sessile populations after the arrival of new bacteria in a flow cell was stopped (i.e., time >20 min) (Figures 8). Since experiments were conducted in non-nutritive medium (NaCl 9 g/L), cell proliferation in this phase was negligible (as confirmed elsewhere; data not published). Therefore, attachment and detachment events were likely the result of rearrangements, i.e., mobility of bacteria between surface and suspension. Figure 8 demonstrates that this mobility was higher, and consequently that stability of adhesion was much less on CH3-rich surfaces, compared to NH2-rich surfaces. Finally, because of dynamic conditions that hamper the formation of counterion layers contrary to static culture conditions, electrostatic interactions may be stronger here than in static culture conditions. Dynamic conditions can therefore be considered more (respectively less) favorable to bacterial adhesion on NH2-rich (respectively CH3-rich) surfaces than static conditions, which may explain the slightly opposite trend of adhesion rate vs NH2/CH3 surface content observed under hydrodynamic flow compared to static conditions. Surfaces with attractive, electrostatic properties for globally,

Figure 6. Adhesion rates of (a) E. coli bacteria and (b) S. epidermidis bacteria in NaCl 9g/L medium. Dotted line depicts the linear fitting. Linear equation and associated regression coefficient is noted on the graphic.

trend of the adhesion rates versus NH2/CH3 surface content, determined by linear regression, is a positive attachment which globally slightly increases with NH2/CH3 surface content. These gentle, even nearly zero slopes, display an overall behavior opposite to those observed under static culture conditions. These two opposite trends can be explained as follows. The behavior of an adhering or already attached particle, and in particular, of a bacterium under shear-flow can be accounted for essentially by two competing forces: (i) an attaching of resisting force arising from the (bacterium/surface) attractive forces, or the established adhesion of the bacterium to the underlying substrate and, (ii) a shear force arising from the hydrodynamic flow.64 On pure CH3 surface (NH2/CH3 0%), the resisting adhesive force is provided only by the van der Waals and hydrophobic interactions between the bacterium and the support. Moreover, it is not the whole adhesion force that

Figure 7. Scheme depicting interactions between the bacterial membrane and the material surface in dynamic conditions, i.e., under hydrodynamic flow. The membrane of Gram-negative bacterium is schematized by only depicting the phospholipid bilayer (negative charge in green) and two membrane proteins (positive charge in red). 701


Article


the results reported here is that surface concentrations in chemical groups should be considered as a crucial parameter to provide efficient antibacterial adhesion properties to a (bio)material surface. The appropriate and optimal surface content of functional groups should be searched for in weak as in large variations of surface concentrations.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsbiomaterials.5b00151. Table S1 and Figures S1−S5 (PDF)

■

AUTHOR INFORMATION

Corresponding Author

Figure 8. Frequency of attachment and detachment events of E. coli bacteria on five mixed surfaces with different NH2/CH3 contents, calculated for micrographs that were taken every 2 min from 20 to 150 min after the experiment began (injection of bacteria in the flow experimental system and washing procedure of unattached bacteria were performed in the first 20 minutes). Circles depict total events (i.e., sum of attachment and detachment events), whereas bin categories specify the relative proportions of attachment and detachment events.

*E-mail: [email protected]. Phone: +33 (0)389608798. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The article has been edited and proofread by Proof-ReadingServices.org. All works done at Institut de Science des Matériaux de Mulhouse were funded by the French “Centre National de la Recherche Scientifique” (CNRS). The present project was also especially funded by Region Alsace (France). In addition, the authors sincerely thank Jules Valentin and Adeline Marguier for their technical help and Prof Jean-Claude Block for stimulating discussions.

negatively charged objects may be therefore more colonized by bacteria compared to CH3 and CH3-rich surfaces, in contrast to that which happens in static culture conditions.

■

4. CONCLUSIONS Using well-organized and well-controlled model surfaces providing various surface contents of NH2 in CH3 end groups, we demonstrated that E. coli and S. epidermidis behaviors on surfaces with chemical heterogeneities at the molecular scale do not follow a linear evolution with the density of functional groups. Rather, our results show that the ability of bacteria to sense chemical heterogeneities may vary according to ranges of surface concentrations in chemical groups, depending on bacteria species. Transition zones are highlighted, which may be related to the ability of bacteria to distinguish differences in the nature and surface concentrations of functional groups through their own membrane chemical composition. An optimal density of the positive charges on the material surface exists that matches the density of the positively charged molecular groups on the bacterial membrane, leading to a threshold of NH2/CH3 content up to which bacterial adhesion is maximal. Surfaces with NH2/CH3 content in a range in which significant variations of positive charges on the material surface can only slightly modify the strength of bacterium/material surface interaction may allow more constant adhesion whatever NH2/CH3 surface content, leading to plateaus in the relationship between bacterial adhesion and NH2/CH3 surface content. In addition, bacterial adhesion was shown to be more stable on surfaces with NH2/CH3 high contents, because attachment and detachment events under low shear-flow were much less on NH2-rich surfaces than CH3-rich surfaces, probably because of higher attractive electrostatic interactions playing as resisting forces on NH2-rich surfaces. In dynamic culture conditions, the abrupt variations of bacterial behavior observed for slight variations of the chemical composition of the surface also demonstrates that chemical surfaces heterogeneities at the molecular scale can strongly affect adhesion and colonization of bacteria on surfaces. Eventually, an important consequence of

REFERENCES

(1) Anguita-Alonso, P.; Hanssen, A. D.; Patel, R. Prosthetic joint infection. Expert Rev. Anti-Infect. Ther. 2005, 3, 797−804. (2) Bagshaw, S. M.; Laupland, K. B. Epidemiology of intensive care unit-acquired urinary tract infections. Curr. Opin. Infect. Dis. 2006, 19, 67−71. (3) Costerton, J. W.; Stewart, P. S.; Greenberg, E. P. Bacterial biofilms: a common cause of persistent infections. Science 1999, 284, 1318−1322. (4) Donlan, R. M. Biofilms: microbial life on surfaces. Emerging Infect. Dis. 2002, 8, 881−890. (5) Katsikogianni, M.; Missirlis, Y. F. Concise review of mechanisms of bacterial adhesion to biomaterials and of techniques used in estimating bacteria-material interactions. Eur. Cell. Mater. 2004, 8, 37− 57. (6) Ploux, L.; Ponche, A.; Anselme, K. Bacteria/Material interfaces: role of the material and cell wall properties. J. Adhes. Sci. Technol. 2010, 24, 2165−2201. (7) An, Y. H.; Friedman, R. J. Concise review of mechanisms of bacterial adhesion to biomaterial surfaces. J. Biomed. Mater. Res. 1998, 43, 338−348. (8) Bakker, D. P.; Busscher, H. J.; van Zanten, J.; de Vries, J.; Klijnstra, J. W.; van der Mei, H. C. Multiple linear regression analysis of bacterial deposition to polyurethane coatings after conditioning film formation in the marine environment. Microbiology 2004, 150, 1779− 1784. (9) Roberts, J. A. Inhibition and enhancement of microbial surface colonization: the role of silicate composition. Chem. Geol. 2004, 212, 313−327. (10) Terada, A.; Yuasa, A.; Tsuneda, S.; Hirata, A.; Katakai, A.; Tamada, M. Elucidation of dominant effect on initial bacterial adhesion onto polymer surfaces prepared by radiation-induced graft polymerization. Colloids Surf., B 2005, 43, 99−107. (11) Krsko, P.; Kaplan, J. B.; Libera, M. Spatially controlled bacterial adhesion using surface-patterned poly(ethylene glycol) hydrogels. Acta Biomater. 2009, 5, 589−596. 702


Article

ACS Biomaterials Science & Engineering (12) Ploux, L.; Beckendorff, S.; Nardin, M.; Neunlist, S. Quantitative and morphological analysis of biofilm formation on self-assembled monolayers. Colloids Surf., B 2007, 57, 174−181. (13) Katsikogianni, M.; Missirlis, Y. Bacterial adhesion onto materials with specific surface chemistries under flow conditions. J. Mater. Sci.: Mater. Med. 2010, 21, 963−968. (14) Barth, K. A.; Coullerez, G.; Nilsson, L. M.; Castelli, R.; Seeberger, P. H.; Vogel, V.; Textor, M. An engineered mannoside presenting platform: Escherichia coli adhesion under static and dynamic conditions. Adv. Funct. Mater. 2008, 18, 1459−1469. (15) Burton, E. A.; Simon, K. A.; Hou, S.; Ren, D.; Luk, Y.-Y. Molecular Gradients of Bioinertness Reveal a Mechanistic Difference between Mammalian Cell Adhesion and Bacterial Biofilm Formation. Langmuir 2009, 25, 1547−1553. (16) Wiencek, K. M.; Fletcher, M. Bacterial adhesion to hydroxyland methyl-terminated alkanethiol self-assembled monolayers. J. Bacteriol. 1995, 177, 1959−1966. (17) Wiencek, K. M.; Fletcher, M. Effects of substratum wettability and molecular topography on the initial adhesion of bacteria to chemically defined substrata. Biofouling 1997, 11, 293−311. (18) Qian, X.; Metallo, S. J.; Choi, I. S.; Wu, H.; Liang, M. N.; Whitesides, G. M. Arrays of Self-Assembled Monolayers for Studying Inhibition of Bacterial Adhesion. Anal. Chem. 2002, 74, 1805−1810. (19) Böhmler, J.; Ponche, A.; Anselme, K.; Ploux, L. Self-Assembled Molecular Platforms for Bacteria/Material Biointerface Studies: Importance to Control Functional Group Accessibility. ACS Appl. Mater. Interfaces 2013, 5, 10478−10488. (20) Miao, H.; Ratnasingam, S.; Pu, C. S.; Desai, M. M.; Sze, C. C. Dual fluorescence system for flow cytometric analysis of Escherichia coli transcriptional response in multi-species context. J. Microbiol. Methods 2009, 76, 109−119. (21) Vidal, O.; Longin, R.; Prigent-Combaret, C.; Dorel, C.; Hooreman, M.; Lejeune, P. Isolation of an Escherichia coli K-12 Mutant Strain Able To Form Biofilms on Inert Surfaces: Involvement of a New ompR Allele That Increases Curli Expression. J. Bacteriol. 1998, 180, 2442−2449. (22) Mauclaire, L.; Brombacher, E.; Bünger, J. D.; Zinn, M. Factors controlling bacterial attachment and biofilm formation on mediumchain-length polyhydroxyalkanoates (mcl-PHAs). Colloids Surf., B 2010, 76, 104−111. (23) Christensen, G. D.; Simpson, W. A.; Bisno, A. L.; Beachey, E. H. Adherence of slime-producing strains of Staphylococcus epidermidis to smooth surfaces. Infect. Immun. 1982, 37, 318−326. (24) Christensen, G. D.; Parisi, J. T.; Bisno, A. L.; Simpson, W. A.; Beachey, E. H. Characterization of clinically significant strains of coagulase-negative staphylococci. J. Clin. Microbiol. 1983, 18, 258−269. (25) Fonseca, A. P.; Granja, P. L.; Nogueira, J. A.; Oliveira, D. R.; Barbosa, M. A. Staphylococcus epidermidis RP62A adhesion to chemically modified cellulose derivatives. J. Mater. Sci.: Mater. Med. 2001, 12, 543−548. (26) Nicolau Korres, A. M.; de Farias V. Aquije, G. M.; Buss, D. S.; Aires Ventura, J.; Bueno Fernandes, P. M.; Ribeiro Fernandes, A. A. Comparison of Biofilm and Attachment Mechanisms of a Phytopathological and Clinical Isolate of Klebsiella pneumoniae Subsp. pneumoniae. Scientific World Journal 2013, 2013, 6 pages. (27) Katsikogianni, M. G.; Missirlis, Y. F. Interactions of bacteria with specific biomaterial surface chemistries under flow conditions. Acta Biomater. 2010, 6, 1107−1118. (28) Bellon-Fontaine, M.-N.; Rault, J.; van Oss, C. J. Microbial adhesion to solvents: a novel method to determine the electrondonor/electron-acceptor or Lewis acid-base properties of microbial cells. Colloids Surf., B 1996, 7, 47−53. (29) Sherman, P. M.; Houston, W. L.; Boedeker, E. C. Functional heterogeneity of intestinal Escherichia coli strains expressing type 1 somatic pili (fimbriae): assessment of bacterial adherence to intestinal membranes and surface hydrophobicity. Infect. Immun. 1985, 49, 797− 804.

(30) Rosenberg, M.; Gutnick, D.; Rosenberg, E. Adherence of bacteria to hydrocarbons: A simple method for measuring cell-surface hydrophobicity. FEMS Microbiol. Lett. 1980, 9, 29−33. (31) Busscher, H. J.; Weerkamp, A. H.; van der Mei, H. C.; van Pelt, A. W.; de Jong, H. P.; Arends, J. Measurement of the surface free energy of bacterial cell surfaces and its relevance for adhesion. Appl. Environ. Microbiol. 1984, 48, 980−983. (32) Anselme, K.; Davidson, P.; Popa, A. M.; Giazzon, M.; Liley, M.; Ploux, L. Response to comment on ″The interaction of cells and bacteria with surfaces structures at the nanoscale″. Acta Biomater. 2011, 7, 1936−1937. (33) Selinummi, J.; Seppälä, J.; Yli-Harja, O.; Puhakka, J. Software for quantification of labeled bacteria from digital microscope images by automated image analysis. BioTechniques 2005, 39, 859−863. (34) Rasband, W. ImageJ; U. S. National Institutes of Health: Bethesda, MD, 1997. (35) Scherrer, B. Biostatistique; Gaëtan Morin Éditeur - Chenelière Éducation: Montreal, 2007; Vol. 1. (36) Coico, R.; Kowalik, T.; Quarles, J.; Stevenson, B.; Taylor, R. Current Protocols in Microbiology. Eds.John Wiley & Sons,: Hoboken, NJ, 2005. (37) Cottenye, N.; Anselme, K.; Ploux, L.; Vebert-Nardin, C. Vesicular structures self-assembled from oligonucleotide-polymer hybrids: mechanical prevention of bacterial colonization upon their surface tethering through hybridization. Adv. Funct. Mater. 2012, 22, 4891−4898. (38) Sbalzarini, I. F.; Koumoutsakos, P. Feature point tracking and trajectory analysis for video imaging in cell biology. J. Struct. Biol. 2005, 151, 182−195. (39) Aswal, D. K.; Lenfant, S.; Guerin, D.; Yakhmi, J. V.; Vuillaume, D. Self assembled monolayers on silicon for molecular electronics. Anal. Chim. Acta 2006, 568, 84−108. (40) Tegoulia, V. A.; Cooper, S. L. Staphylococcus aureus adhesion to self-assembled monolayers: effect of surface chemistry and fibrinogen presence. Colloids Surf., B 2002, 24, 217−228. (41) Cerca, N.; Pier, G. B.; Vilanova, M.; Oliveira, R.; Azeredo, J. Quantitative analysis of adhesion and biofilm formation on hydrophilic and hydrophobic surfaces of clinical isolates of Staphylococcus epidermidis. Res. Microbiol. 2005, 156, 506−514. (42) Hermansson, M. The DLVO theory in microbial adhesion. Colloids Surf., B 1999, 14, 105−119. (43) Hori, K.; Matsumoto, S. Bacterial adhesion: from mechanism to control. Biochem. Eng. J. 2010, 48, 424−434. (44) Lee, S.-H.; Lin, W.-C.; Kuo, C.-H.; Karakachian, M.; Lin, Y.-C.; Yu, B.-Y.; Shyue, J.-J. Photooxidation of amine-terminated selfassembled monolayers on gold. J. Phys. Chem. C 2010, 114, 10512− 10519. (45) Lin, J.-H.; Chang, H.-Y.; Kao, W.-L.; Lin, K.-Y.; Liao, H.-Y.; You, Y.-W.; Kuo, Y.-T.; Kuo, D.-Y.; Chu, K.-J.; Chu, Y.-H.; Shyue, J.-J. Effect of Surface Potential on Extracellular Matrix Protein Adsorption. Langmuir 2014, 30, 10328−10335. (46) Lin, Y.-C.; Yu, B.-Y.; Lin, W.-C.; Lee, S.-H.; Kuo, C.-H.; Shyue, J.-J. Tailoring the surface potential of gold nanoparticles with selfassembled monolayers with mixed functional groups. J. Colloid Interface Sci. 2009, 340, 126−130. (47) Shyue, J.-J.; De Guire, M. R.; Nakanishi, T.; Masuda, Y.; Koumoto, K.; Sukenik, C. N. Acid-base properties and zeta potentials of self- assembled monolayers obtained via in situ transformations. Langmuir 2004, 20, 8693−8698. (48) Zhang, H.; He, H.-X.; Wang, J.; Mu, T.; Liu, Z.-F. Force titration of amino group-terminated self-assembledmonolayers using chemical force microscopy. Appl. Phys. A: Mater. Sci. Process. 1998, 66, S269− S271. (49) Roy, H.; Dare, K.; Ibba, M. Adaptation of the bacterial membrane to changing environments using aminoacylated phospholipids. Mol. Microbiol. 2009, 71, 547−550. (50) Poortinga, A. T.; Bos, R.; Norde, W.; Busscher, H. J. Electric double layer interactions in bacterial adhesion to surfaces. Surf. Sci. Rep. 2002, 47, 1−32. 703


Article

ACS Biomaterials Science & Engineering (51) Gasol, J. M.; del Giorgio, P. A.; Massana, R.; Duarte, C. M. Active versus inactive bacteria: size-dependence in a coastal marine plankton community. Mar. Ecol.: Prog. Ser. 1995, 128, 91−97. (52) Koch, A. L. Distribution of cell size in growing cultures of bacteria and the applicability of the Collins-Richmond principle. J. Gen. Microbiol. 1966, 45, 409−417. (53) Weart, R. B.; Lee, A. H.; Chien, A.-C.; Haeusser, D. P.; Hill, N. S.; Levin, P. A. A metabolic sensor governing cell size in bacteria. Cell 2007, 130, 335−347. (54) Hill, N. S.; Buske, P. J.; Shi, Y.; Levin, P. A. A moonlighting enzyme links Escherichia coli cell size with central metabolism. PLoS Genet. 2013, 9, e1003663. (55) Basu, S.; Nandakumar, K.; Masliyah, J. H. A model for detachment of a partially wetting drop from a solid surface by shear flow. J. Colloid Interface Sci. 1997, 190, 253−257. (56) Harkes, G.; Feijen, J.; Dankert, J. Adhesion of Escherichia coli on to a series of poly(methacrylates) differing in charge and hydrophobicity. Biomaterials 1991, 12, 853−860. (57) Navarre, W. W.; Schneewind, O. Surface proteins of Grampositive bacteria and mechanisms of their targeting to the cell wall envelope. Microbiol. Mol. Biol. Rev. 1999, 63, 174−229. (58) Harkes, G.; Van der Mei, H. C.; Rouxhet, P. G.; Dankert, J.; Busscher, H. J.; Feijen, J. Physicochemical characterization of Escherichia coli. A comparison with Gram-positive bacteria. Cell Biophys. 1992, 20, 17−32. (59) Walter, N.; Selhuber, C.; Kessler, H.; Spatz, J. P. Cellular Unbinding Forces of Initial Adhesion Processes on Nanopatterned Surfaces Probed with Magnetic Tweezers. Nano Lett. 2006, 6, 398− 402. (60) Deeg, J. A.; Louban, I.; Aydin, D.; Selhuber-Unkel, C.; Kessler, H.; Spatz, J. P. Impact of Local versus Global Ligand Density on Cellular Adhesion. Nano Lett. 2011, 11, 1469−1476. (61) Casadevall, A.; Pirofski, L.-a. Host-Pathogen Interactions: Basic Concepts of Microbial Commensalism, Colonization, Infection, and Disease. Infect. Immun. 2000, 68, 6511−6518. (62) Disney, M. D.; Zheng, J.; Swager, T. M.; Seeberger, P. H. Detection of Bacteria with Carbohydrate-Functionalized Fluorescent Polymers. J. Am. Chem. Soc. 2004, 126, 13343−13346. (63) Liu, Y.; Tay, J.-H. Metabolic response of biofilm to shear stress in fixed-film culture. J. Appl. Microbiol. 2001, 90, 337−342. (64) Mougin, K.; Haidara, H.; Schultz, J. Shear-Induced Detachment of Micrometer-Scale Soft Droplets Embedded onto a Rigid Substrate. Relation to Biological Systems. Langmuir 2001, 17, 5952−5957.

704


Impact of Chemical Heterogeneities of Surfaces on Colonization by

Recommend Documents