Differences between Chemisorbed and Physisorbed Biomolecules on Particle Deposition to Hydrophobic Surfaces † MICHAEL B. SALERNO, ‡ SAM ROTHSTEIN, ‡ CHISOMAGA NWACHUKWU, ‡ HAITHEM SHELBI, ‡ DARRELL VELEGOL, ‡ AND BRUCE E. LOGAN* ,§ Department of Chemical Engineering and Department of Civil and Environmental Engineering, The Pennsylvania State University, University Park, Pennsylvania 16802 This study examines differences between chemisorbed and physisorbed biomolecules on bacterial adhesion to both hydrophobic and hydrophilic surfaces that are biologically nonspecific. Bacteria-sized latex microspheres were used as a simplified model in order to study these factors that affect microbial adhesion. Two biomolecules (protein A, poly-D-lysine) were covalently bound to microspheres in order to study the effect of proteins on particle filtration rates in columns packed with glass beads. When poly-D- lysine or protein A was covalently bonded to the microspheres, sticking coefficients (R) for the microspheres increased by up to an order of magnitude as compared with uncoated latex microspheres. The glass packing beads were then made hydrophobic by covalently attaching silane groups with different carbon-chain lengths (0.2, 1.2, and 2.8 nm). Sticking coefficients for the uncoated microspheres on these silanized packing beads (R) 0.15 at 1 mM ionic strength; 0.76 at 100 mM) were larger than those on uncoated glass packing beads (0.02 at 1 mM; 0.15 at 100 mM). In addition, adhesion increased with ionic strength on both hydrophobic and hydrophilic surfaces. Physical adsorption gave different results. When either dextran or protein A was physically adsorbed to both the microspheres and the column, no appreciable change in adhesion was observed. Covalently attaching protein A to the microspheres increased their hydrophobicity, but sticking coefficients were large regardless of the substrate hydrophobicity as a result of biomolecule-surface interactions. This study demonstrates that, at high ionic strength, covalently attached hydrophobic species give much higher sticking coefficients for particles than do physically adsorbed species. Introduction Controlling and understanding bacterial adhesion is im- portant in many fields, including prevention of biofilms on medical implants, reduction of drag on ship hulls (1, 2), increased efficiency of water filtration systems (3-5), and improved efficacy of bioaugmentation methods for treating polluted landfills (6-9). A number of surface forces (10) (e.g., electrostatic and van der Waals interactions, receptor-ligand interactions) (11) determine the adhesion of the bacteria. These forces have been shown to depend on solution ionic strength (12-14), bacterial growth conditions (15), substrate physical properties (16), and bacterial surface molecular structure (17). While discerning the importance of the various forces is difficult when a bacterium is nearly touching the substratesespecially in light of nonuniformly distributed biomolecules on the bacterial surface (18)swe hypothesize in this work that bacterial surface proteins have a large impact on increasing adhesion. The attractive forces caused by proteins arise from their having hydrophobic regions and their electrostatic charge. Traditionally, bacterium-substrate interactions have been described by classical Derjaguin-Landau-Verwey-Over- beek (DLVO) theory. This model accounts for van der Waals (VDW) and electrostatic forces, and it neglects complexities such as surface roughness (19-21) and charge nonuniformity (22-24). At separations greater than 50 nm, the attractive VDW forces are dominant but still relatively weak (25). As the bacterium approaches closer to the substrate, the VDW attraction becomes larger. However, when both the bacterium and the substrate are negatively charged (the most common case), the VDW attraction is counteracted by repulsive electrostatic interactions (26, 27). Electrostatic interactions become less important at higher ionic strength, and so bacterial adhesion has usually been shown to increase with ionic strength (28-31). Nevertheless, there are cases where the DLVO prediction concerning ionic strength fails (12), and so extensions to DLVO theory have been developed to account for hydro- phobicity (16, 32) and surface free energy (33-35). Hydro- phobic groups on bacterial surfaces have been shown to increase the adhesion of many strains because the groups remove water adsorbed to surfaces (36, 37). Thus, highly hydrated surfaces have actually been found to resist bacterial adhesion (37). Some bacteria have been found to be hydrophobic (35), while most have been found to be hydrophilic (14). One study even showed that adhesion of bacteria was similar to both hydrophobic and hydrophilic surfaces (38). While in some studies the extensions explain adhesion data (34, 39, 40), this is not always the case (41-43). One explanation for the mixed success is that nanoscale (18, 44- 46) or molecular arguments do not appear, even at very small bacterium-substrate separations (47). For example, X-ray photoelectron spectroscopy (XPS) experiments suggest that protein molecules are involved in hydrophobic forces (48, 49). Studies have shown that proteins produced by bacteria (e.g., fimbriae, flagella, or enzymes) are important in the initial adhesion stages (50). Although the evidence suggests that proteins are important to adhesion, it is not clear whether this results primarily from nonspecific hydrophobic interactions or more specific receptor-ligand interactions (44). Our study aims to clarify the role of biomolecules on bacterial adhesion to both hydrophobic and hydrophilic surfaces that are biologically nonspecific. Our approach is to use colloidal microspheres that have a size similar to many bacteria (i.e., micron size). The microspheres are coated with biomolecules, and thus the colloids have a well-defined surface chemistry and a well-defined geometry. † This paper is part of the Charles O’Melia tribute issue. * Corresponding author telephone: (814)863-7908; fax: (814)863- 7304; e-mail: blogan@psu.edu. ‡ Department of Chemical Engineering. § Department of Civil and Environmental Engineering. Environ. Sci. Technol. 2005, 39, 6371-6377 10.1021/es050204l CCC: $30.25  2005 American Chemical Society VOL. 39, NO. 17, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 9 6371 Published on Web 05/28/2005