4720
Langmuir 2006, 22, 4720-4727
Interaction Forces Measured Using AFM between Colloids and Surfaces Coated with Both Dextran and Protein Li-Chong Xu and Bruce E. Logan* Department of CiVil and EnVironmental Engineering, The PennsylVania State UniVersity, UniVersity Park, PennsylVania 16802 ReceiVed December 20, 2005. In Final Form: March 2, 2006 Both proteins and polysaccharides are biopolymers present on a bacterial surface that can simultaneously affect bacterial adhesion. To better understand how the combined presence of proteins and polysaccharides might influence bacterial attachment, adhesion forces were examined using atomic force microscopy (AFM) between colloids (COOHor protein-coated) and polymer-coated surfaces (BSA, lysozyme, dextran, BSA+dextran and lysozyme+dextran) as a function of residence time and ionic strength. Protein and dextran were competitively covalently bonded onto glass surfaces, forming a coating that was 22-33% protein and 68-77% dextran. Topographic and phase images of polymercoated surfaces obtained with tapping mode AFM indicated that proteins at short residence times (3.3°) is 97.3% for the BSA+dextran coating and 85.8% for the lysozyme+dextran coating. This observation of a very high percentage of dextran relative to that of the protein suggests that the outermost layer of coating surface was effectively covered with dextran, with the protein (BSA or lysozyme) relatively “hidden” beneath that of the dextran. This relatively apparent abundance of the dextran versus that of the protein on the outer layer of coating is probably the result of competitive adsorption processes between protein and dextran onto the silane glass surface, which is a process likely dependent on the surface charge of the biopolymers and their sizes. Adhesion of Colloids to BSA, Dextran, or BSA+Dextran Coated Surfaces. The maximum adhesion forces measured from retraction force curves between a COOH latex colloid and polymer coated glass surfaces at short residence times (1 ms) increased in the order BSA e BSA+dextran < dextran for the 1 mM IS solution (Figure 4). Saw-tooth shaped retraction force curves were observed, indicating multiple bonds were broken between the colloid and polymer surfaces when the probe was retracted from the surface. The larger values obtained for dextran than for the two proteins were expected based on previous findings.26 The greater values for the BSA+dextran than for the surface
coated only with BSA (at short residence times) are primarily due to the colloid interacting with the larger dextran molecules. Adhesion forces consistently increased with residence time for all cases (Figure 5). However, the largest interaction forces at longer retention times (5-100 s) were obtained for the BSA+dextran complex, compared to those for the BSA or dextran alone (Figure 5). For example, the adhesion forces for BSA+dextran increased from -0.3 ( 0.26 nN at 1 ms to -6.6 ( 1.5 nN at 100 s, while the forces for BSA increased from -0.25 ( 0.19 nN to -3.2 ( 1.5 nN and dextran from -1.6 ( 0.4 to -4.7 ( 0.7 nN only. The change of adhesion force order indicates that the presence of both molecules on the surface resulted in overall greater adhesion forces than that expected from measurements of the forces for the individual molecules. This greater adhesion force must be due to the rearrangement of molecules in the BSA+dextran complex relative to that of the surfaces over long exposure times, producing more bonds between the colloid and polymers. In 100 mM IS solution, similar trends were observed between the colloid and the polymer coated surfaces at short residence times, although the overall forces were greatly reduced. The BSA+dextran surface produced lower adhesion forces than those
4724 Langmuir, Vol. 22, No. 10, 2006
Xu and Logan
changes in the polymers as a result of the high versus low IS.25 The conformational change of biopolymers in high IS solution was evidenced by the reduction in the dimension of biopolymer molecules and thickness of surface versus that in low IS. For example, the average dimension of BSA was 12.9 ( 3.6 nm in 100 mM IS solution versus 18.6 ( 5.3 nm in 1 mM IS solution, and the thickness of coating was reduced from 5.2 nm (1 mM) to 2.0 nm (100 mM). AFM topographic images have previously been shown for these molecules by Xu et al.26 When the colloid was coated with a protein (BSA), adhesion forces were generally larger than those examined under the same conditions with the uncoated colloid, and adhesion forces always increased with time (Figure 6). In the low IS solution, the BSA+dextran surfaces again demonstrated the largest adhesion force over time. In the high IS solution, however, the greatest adhesion force was consistently obtained between the proteincoated colloid and the dextran. Adhesion of Colloids to Lysozyme, Dextran, and Lysozyme+ Dextran Coated Surfaces. In tests where lysozyme was used instead of BSA, the maximum adhesion forces between the latex colloid and the polymer coated glass surfaces generally produced the same trends as observed with BSA and the different surfaces (Figure 7). The adhesion forces between the lysozyme and a surface were smaller than that between dextran and the colloid (IS ) 1 mM), but at residence times of 5 s or longer, the adhesion forces for lysozyme+dextran were larger than those for the individual protein or dextran (Figure 8a). With lysozyme, a similar trend was observed in the 100 mM IS solution, as adhesion forces with the lysozyme+dextran were also consistently the largest after a colloid retention time of 5 s (Figure 8b). Adhesion forces were also examined between a BSA-coated colloid and a lysozyme+dextran coated surface. Results were similar to those obtained with the same BSA-coated colloid and the individual polymers versus combined polymers, although at the longest residence times the adhesion force for the lysozymecoated surface was approximately equal to that of the lysozyme+dextran coated surface (Figure 9a). Adhesion forces increased with residence time in all cases. In 100 mM IS solution, the adhesion forces for lysozyme and lysozyme+dextran were similar and slightly less than that those obtained for dextran (Figure 9b), consistent with the analogous measurements using BSA+dextran (Figure 6b).
Discussion
Figure 3. AFM 2-D topographic (a, c, e, g, i) and phase (b, d, f, h, j) images of proteins and dextran. BSA (a, b); lysozyme (c, d); dextran (e, f). BSA+ dextran (g, h), and lyszoyme+dextran (i, j) (color scale: 10 nm height or 10° for phase).
with BSA or dextran at short residence times (1.0). Under excess conditions, there is partial (44) Dickinson, E. In Biopolymer mixtures; Harding, S. E., Hill, S. E., Mitchell, J. R., Eds.; Nortingham University Press: Nortingham, 1995; pp 349-372. (45) Antonov, Y. J.; Wolf, B. A. Biomacromolecules 2005, 6, 2980-2989.
Colloids Dextran- and Protein-Coated Surfaces
Langmuir, Vol. 22, No. 10, 2006 4725
Figure 4. Representative retraction force curve of COOH latex colloid and BSA, dextran, BSA + dextran surfaces in 1 mM IS solution, showing the trend of adhesion force with time and polymers.
Figure 5. Adhesion forces of carboxylated latex colloid with biopolymer coated surfaces. (a) 1 mM IS, (b) 100 mM IS.
destabilization of the secondary and tertiary structure of the protein. In our study, we first produced a stable intermediate with each individual polymer through the addition of NHS before the polymers were poured onto the glass surfaces. This step reduced the likelihood of the formation of interpolymer complexes, although we cannot completely rule out the possibility of such interpolymer complexes in the final coating. The presence of the protein did not apparently alter the relative structure of the dextran as AFM phase analysis indicated dextran interactions distant from the surface. This suggests that, in the absence of the colloid probe pushing into the two polymers, the protein molecules remained hidden beneath dextran molecules and did not greatly affect their structure via inter-complex formation. The combined presence of both a protein and dextran on a surface consistently produced the same trend in the adhesion forces measured during colloid retraction from the surface at low IS, and also in many cases at high IS. We observed that the
Figure 6. Adhesion forces of BSA-coated latex colloid with biopolymer coated surfaces. (a) 1 mM IS, (b) 100 mM IS.
adhesion force was smaller for protein+dextran at short residence times compared to those obtained for the individual dextran molecules, but the adhesion forces were largest at the longest residence time (100 s). Similar results were obtained at high IS (100 mM) except for the two tests where the colloid was coated with BSA (Figures 6b and 9b). This trend in increased adhesion with the two molecules was observed using two different proteins (BSA and lysozyme) and two types of colloids (uncoated COOH and coated with BSA). Thus, we conclude that the observed increase in adhesion with the protein+dextran was less of a consequence of the type of polymer, and more a result of the interactions between these different types of polymers in the presence of the colloid. The adhesion force measured between a colloid and a surface mostly depends on the chemical bonds and electrostatic forces as well as the strength of these interactions.31,46 Protein and dextran were held on the surface through covalent bonding of specific functional groups to the surface. When both the protein and the
4726 Langmuir, Vol. 22, No. 10, 2006
Xu and Logan
Figure 7. Representative retraction force curve of COOH latex colloid and lysozyme, dextran, and lysozyme+dextran surfaces in 1 mM IS solution, showing the trend of adhesion force with time and polymers.
Figure 8. Adhesion forces of carboxylated latex colloid with lysozyme, dextran, and lysozyme+dextran coated surfaces. (a) 1 mM IS, (b) 100 mM IS.
dextran are on the surface, the unattached portions of the molecules can bond to each other, adsorb onto the glass surface, or extend into solution. For short contact times (0.001 to 1 s) we observed lower adhesion forces and phase images that suggested dextrans extending into the solution past the proteins for the protein+ dextran coated surface (Figure 10). With an increase in contact time, however, adhesion forces increased with time, with the largest forces obtained for the protein+dextran case compared to the individual polymers. This suggests molecule rearrangement that led to a different overall interaction between the combined molecules than that observed with the individual molecules. The increase in adhesion force with time has previously been observed and attributed to water exclusion and biopolymer rearrangement.26 This rearrangement of the molecules in their binding to the (46) Mondon, M.; Berger, S.; Ziegler, C. Anal. Bioanal. Chem. 2003, 375, 849-855.
Figure 9. Adhesion forces of BSA-coated latex colloid with lysozyme, dextran, and lysozyme+dextran coated surfaces. (a) 1 mM IS, (b) 100 mM IS.
different surfaces results in an overall reduced energy state. Withdrawing the colloid from the surface following molecule rearrangement consequently required greater forces relative to that measured at shorter times with little molecule rearrangement. For the case of the protein+dextran, we surmise that reorientation at longer residence times led to more bonds forming between the molecules anchored to the surface and the colloid, producing a larger adhesion force than observed for the single molecule cases (Figure 10). This trend of increased adhesion in the presence of both protein and dextran molecules can be used to infer molecular-scale mechanisms of bacterial adhesion to a surface. When bacteria collide with a surface, the probability of detachment decreases with time as demonstrated by sharply tailing breakthough curves in tests with packed columns.47 Vadillo-Rodriguez et al.31 have shown that the force needed to retract an AFM tip from the surface of a bacterium increases over time, consistent with this
Colloids Dextran- and Protein-Coated Surfaces
Langmuir, Vol. 22, No. 10, 2006 4727
Figure 10. Proposed the structure of protein + dextran and interaction models for colloid and protein + dextran surface.
observation of increased adhesion over time. The surfaces of Gram negative bacteria are covered with, phospholipids, proteins, and LPS molecules. The LPS molecule contains a large O-antigen that extends out into the bulk liquid, whereas the smaller proteins imbedded in the bacterial outer membrane remain relatively shielded by the LPS.21,36 This situation is analogous to that observed here, with the long dextran molecules covering the proteins (either BSA or lysozyme) when both were bonded to the surface. This was shown in phase images where it was observed that dextran molecules dominated the overall interaction between the colloid and the protein+dextran mixture, and adhesion forces measured at short colloid residence times reflected a surface with a mixture of dextran and the protein. We conclude that proteins on the bacterial surface eventually dominate in the interaction of the cell at the surface, resulting in greater adhesion than expected even on the basis of the mixture of the two types of molecules. These observations are also consistent with polymer-surface interactions observed with FTIR.22 Bacterial cell surfaces consist of a much more complex mixture (47) Camesano, T. A.; Unice, K. M.; Logan, B. E., Colloids Surf. A: Physicochem. Eng. Aspects 1999, 160, 291-307.
of proteins, polysaccharides, and other biopolymers than the relatively simple cases examined here. However, the current work with protein+dextran coated surfaces demonstrates how the combined presence of both a protein and a polysaccharide can affect the adhesion of a particle to a surface. Because the polysaccharides on the surface are expected to extend out into solution, their interactions with a surface will dominate at short residence times. However, proteins hidden behind the dextran will subsequently interact with the surface and further increase the adhesion force between the cell and the surface. Measurements that characterize only the average surface properties of the bacterium, such as zeta potentials, do not capture the more complex behavior of molecules at a surface. Only through direct measurements of the adhesion forces over time can we better understand and unravel the complex factors that affect bacterial adhesion. Acknowledgment. This research was supported by the National Science Foundation (NSF) Grants CHE-0089156 and CHE-0431328. LA053443V