Residence time, loading force, pH, and ionic strength affect adhesion forces between colloids and...

Residence Time, Loading Force, pH, and Ionic StrengthAffect Adhesion Forces between Colloids and

Biopolymer-Coated Surfaces

Li-Chong Xu, Virginia Vadillo-Rodriguez, and Bruce E. Logan*

Department of Civil and Environmental Engineering, The Pennsylvania State University,University Park, Pennsylvania 16802

Received April 6, 2005

Exopolymers are thought to influence bacterial adhesion to surfaces, but the time-dependent nature ofmolecular-scale interactions of biopolymers with a surface are poorly understood. In this study, the adhesionforces between two proteins and a polysaccharide [Bovine serum albumin (BSA), lysozyme, or dextran]and colloids (uncoated or BSA-coated carboxylated latex microspheres) were analyzed using colloid probeatomic force microscopy (AFM). Increasing the residence time of an uncoated or BSA-coated microsphereon a surface consistently increased the adhesion force measured during retraction of the colloid from thesurface, demonstrating the important contribution of polymer rearrangement to increased adhesion force.Increasing the force applied on the colloid (loading force) also increased the adhesion force. For example,at a lower loading force of ∼0.6 nN there was little adhesion (less than -0.47 nN) measured between amicrosphere and the BSA surface for an exposure time up to 10 s. Increasing the loading force to 5.4 nNincreased the adhesion force to -4.1 nN for an uncoated microsphere to a BSA surface and to as muchas -7.5 nN for a BSA-coated microsphere to a BSA-coated glass surface for a residence time of 10 s.Adhesion forces between colloids and biopolymer surfaces decreased inversely with pH over a pH rangeof 4.5-10.6, suggesting that hydrogen bonding and a reduction of electrostatic repulsion were dominantmechanisms of adhesion in lower pH solutions. Larger adhesion forces were observed at low (1 mM) versushigh ionic strength (100 mM), consistent with previous AFM findings. These results show the importanceof polymers for colloid adhesion to surfaces by demonstrating that adhesion forces increase with appliedforce and detention time, and that changes in the adhesion forces reflect changes in solution chemistry.

Introduction

An understanding of the chemical and physical factorsinfluencing bacterial adhesion is important to bioreme-diation of contaminated soil,1 biofilm formation on or-thopaedic implants,2,3 and biofouling of ship hulls inmarine systems.4 The initial bacterial attachment (re-versible adhesion) to a surface depends on the physico-chemical characteristics of the bacterium, the fluidinterface, and the substratum characteristics.5,6 Thesefactors affect one or more of the many forces that governthe approach and adhesion of a particle to the surface,such as electrostatics,7,8 van der Waals,9,10 hydrophobic,11-13

hydration,14,15 and specific chemicals forces.3

Several investigators have observed that bacterialadhesion and desorption are residence-time-depend-ent.7,16,17 The time dependence of this process can beregarded as the result of interaction and attraction ofbacterial exopolymers (e.g., proteins, lipopolysaccharides,and other biopolymers) with the surface and the formationof weak and strong chemical bonds of the polymers withthe surface. Such interactions can depend on specificprotein acceptor-donor interactions as well as on specific

carbohydrate polymers.5,18,19 The bond between biopoly-mers and a surface may strengthen or weaken over time,leading to adsorption and desorption rates that varybetween bacterial strains.16

Solution ionic strength (IS) and pH are well-known toinfluence the extent of bacterial adhesion to a surface.Increasing the IS compresses the thickness of the elec-trostatic double layer surrounding a bacterium and resultsin an increase in bacterial adhesion.20-22 This phenomenon

* Corresponding author. Phone: 814-863-7908. Fax: 814-863-7304. E-mail: [email protected].

(1) Li, Q.; Logan, B. E. Water Res. 1999, 33, 1090-1100.(2) Sheehan, E.; McKenna, J.; Mulhall, K. J.; Marks, P.; McCormack,

D. J. Orthop. Res. 2004, 22, 39-43.(3) Speranza, G.; Gottardi, G.; Pederzolli, C.; Lunelli, L.; Canteri, R.;

Pasquardini, L.; Carli, E.; Lui, A.; Maniglio, D.; Brugnara, M.; Anderle,M. Biomaterials 2004, 25, 2029-2037.

(4) Schmidt, D. L.; Brady, R. F.; Lam, K.; Schmidt, D. C.; Chaudhury,M. K. Langmuir 2004, 2830-2836.

(5) Gristina, A. G. Science 1987, 4822, 1588-1595.(6) Iwabuchi, N.; Sunairi, M.; Anzai, H.; Morisaki, H.; Nakajima, M.

Colloids Surf., B 2003, 30, 51-60.

(7) Bouttier, S.; Han, K. G.; Ntsama, C.; Bellon-Fontaine, M. N.;Fourmat, J. Colloids Surf., B 1994, 2, 57-65.

(8) Mosley, L. M.; Hunter, K. A.; Ducker, W. A. Environ. Sci. Technol.2003, 37, 3303-3308.

(9) Hayashi, H.; Tsuneda, S.: Hirata, A.; Sasaki, H. Colloids Surf.,B 2001, 22, 149-157.

(10) Poortinga, A. T.; Bos, R.; Norde, W.; Busscher, H. J. Surf. Sci.Rep. 2002, 47, 1-32.

(11) Rijnaarts, H. H. M.; Norde, W.; Bouwer, E.; Lyklema, J.; Zehnder,A. J. B. Environ. Sci. Technol. 1996, 30, 2877-2883.

(12) Vadillo-Rodriguez, V.; Busscher, H. J.; Norde, W.; de Vries, J.;van der Mei, H. C. J. Bacteriol. 2004, 186, 6647-6650.

(13) Salerno, M. B.; Logan, B. E.; Velegol, D. Langmuir 2004, 20,10625-10629.

(14) Busalmen, J. P.; de Sanchez, S. R. Int. Biodeterior. Biodegrad.2003, 52, 13-19.

(15) Valle-Delgado, J. J.; Molina-Bolıvar, J. A.; Galisteo-Gonzalez,F.; Galvez-Ruiz, M. J.; Feiler, A.; Rutland, M. W. J. Phys. Chem. 2004,108, 5365-5371.

(16) Meinders, J. M.; van der Mei, H. C.; Busscher, H. J. J. ColloidInterface Sci. 1995, 176, 329-341.

(17) Switalski, L.; Butcher, W. G. Arch. Oral Biol. 1994, 39, 155-161.

(18) Stuart, J. K.; Hlady, V. Langmuir 1995, 11, 1368-1374.(19) Funfstuck, R.; Franke, S.; Hellberg, M.; Ott, U.; Knofel, B.;

Straube, E.; Sommer, M.; Hacker, J. Int. J. Antimicrob. Agents 2001,253-258.

(20) Zita, A.; Hermansson, M. Appl. Environ. Microbiol. 1994, 60,3041-3048.

(21) Deshpande, P. A.; Shonnard, D. R. Water Resour. Res. 1999, 35(5), 1619-1627.

7491Langmuir 2005, 21, 7491-7500

10.1021/la0509091 CCC: $30.25 © 2005 American Chemical SocietyPublished on Web 06/16/2005

is qualitatively consistent with predictions based on DLVOtheory as an increase in the IS will reduce the range ofelectrostatic repulsion between two negatively chargedsurfaces.9,23,24 Changes in solution pH are thought toinfluence bacterial adhesion by altering hydrogen bondingforces between polymers and the surface, as well aschanging electrostatic interactions. For example, Ahimouet al.25 found that lowering the pH from 7 to 4 increasedbacterial adhesion. They attributed the increased adhesionat the lower pH to increased formation of hydrogen bondsbetween the bacteria and surface, while at higher pH theyattributed the reduction in adhesion to the increasedelectrostatic repulsion between the bacteria and thesurface. A consistent change in bacterial adhesion withsolution pH, however, is not always observed. For example,some investigators have found that pH has no effect onadhesion when the pH range is limited to a smaller, andnear neutral, range of 5.5-7.26,27 The solution IS can alsobe a factor in the effect of pH. Busalmen and de Sanchez14

observed that bacterial adhesion increased with pH in alow-IS solution (0.1 M), but adhesion was consistentlylow over a pH range of 2-8 in a high-IS solution (0.6 M).

Several macroscopic techniques have been used to studythe kinetics of microbial adhesion to surfaces, rangingfrom well-defined systems such as parallel plate flowchambers28,29 and packed columns30,31 to simple adhesiontests based on exposing a surface to a bacterial suspensionfor a specific amount of time.32 While these techniquesdistinguish adhesion characteristics of bacteria or sur-faces, they do not provide more detailed information onadhesion forces at nanoscale to molecular scale that influ-ence bacterial adhesion and allow us to better understandthe factors affecting adhesion. By using an atomic forcemicroscope (AFM), it has been possible to directly measurethe physicochemical properties of biopolymers on bacterialsurfaces at molecular scales, including the interactionforces of these surfaces with the AFM tip or other surfacesdue to electrostatic and steric interactions.33-38

It has been shown that the contact time of the AFM tipwith a surface of either proteins or bacteria increases theadhesion force measured during tip retraction.18,39-41 It is

reasonable to assume that this increase in adhesion resultsfrom how the biopolymers on the bacterial surface interactwith the surface over time. However, the time-dependentnature of biopolymer interactions with surfaces has notbeen sufficiently examined, especially as a function ofsolution chemistry and the force applied on a biopolymer.In this study, we used colloid probe AFM to measureinteraction forces between a colloid mounted on an AFMtip with a surface, in the presence or absence of variouspolymers coating these surfaces. These interactions wereexamined both as a function of time and loading force (theforce applied by the colloid probe), as well as solutionchemistry (pH and IS). By using this approach of examin-ing polymer interactions with surfaces, we hope to betterunderstand the molecular-scale factors that affect biopoly-mer interactions with surfaces and, therefore, microbialadhesion to surfaces.

MethodsColloid and Surfaces. Carboxylated polystyrene latex mi-

crospheres (4.5 µm, Polysciences, Warrington, PA) or glass beads(∼4.5 µm, Polysciences) were used as the colloid probe in allAFM experiments. Latex microspheres and glass beads werewashed and rinsed three times with Milli-Q water by centrifuga-tion prior to use. Glass cover slips (25 mm × 60 mm, CorningNo.1 cover glass, Corning, NY) were rigorously cleaned,42 rinsedwith copious amounts of ultrapure water (Milli-Q), and storedin ultrapure water at 4 °C before use.

Covalent Bonding of Biopolymers to Glass and LatexMicrosphere Surfaces. Bovine serum albumin (BSA, 69 kDa),lysozyme (14 kDa) from chicken egg white, and dextran (144kDa; Sigma, St. Louis, MO), were used as model protein orpolysaccharide biopolymers having different charges at pH 7.0.BSA is negatively charged, while lysozyme is positively charged.Dextran is a hydrophilic and neutrally charged polysaccharideat pH 7.0. Glass cover slips were modified by bonding an aminofunctional group to the glass cover slip and then covalentlybonding a protein (BSA or lysozyme) to this amino group aspreviously described.43 BSA was covalently bonded to a car-boxylated latex microsphere using a standard carbodiimide kit(Polysciences).

To couple dextran to a glass surface, it was first modified withcarboxyl functional groups using succinic anhydride and dimethylsulfoxide (DMSO) as described elsewhere.44 Amino-functionalizedglass surfaces were incubated in the succinoylated dextransolution (10 mg/mL) containing 1-[3-(dimethylamino)propy]-3-ethylcarbodiimide hydrochloride (EDC, 400 mM) and sulfo-N-hydroxysulfosuccinimide (115 mM, Pierce Biotechnology Inc.Rockford, IL) in sodium phosphate buffer (10 mM, pH 7.5) andallowed to react overnight with gentle shaking at 50 rpm. Afterbeing coated with biopolymers, the slides were rinsed with Milli-Qwater before being used in AFM experiments.

The ú potentials of uncoated colloids and colloids coated withBSA were measured 5 times using 25 cycles/analysis (ZetaPALSanalyzer, Brookhaven Instruments Corp., Holtzville, NY).

AFM Experiments. All AFM experiments were performedusing a BioScope AFM (Digital Instrument, Santa Barbara, CA)with a Nanoscope IIIa control system (Version 4.32r3). Biopoly-mer coatings were imaged using tapping mode in liquid at adrive frequency of ∼8 kHz and a drive amplitude of 100 mV.Silicon nitride AFM tips (Veeco NanoProbe NP-20; narrow longcantilever) were used for imaging.

Adhesion force measurements were performed in contact modeusing a colloid probe. Colloid probes were prepared by gluing

(22) Rijnaarts, H. H. M.; Norde, W.; Lyklema, J.; Zehnder, A. J. B.Colloids Surf., B 1999, 14, 179-195.

(23) Azeredo, J.; Visser, J.; Oliveira, R. Colloids Surf., B 1999, 14,141-148.

(24) Brant, J. A.; Childress, A. E. Environ. Eng. Sci. 2002, 19, 413-427.

(25) Ahimou, F.; Denis, F. A.; Touhami, A.; Dufrene, Y. Langmuir2002, 18, 9937-9941.

(26) Sadr Ghayeni, S. B.; Beatson, P. J.; Schneider, R. P.; Fane, A.G. J. Membr. Sci. 1998, 138, 29-42.

(27) Jewett, D. G.; Hilbert, T. A.; Logan, B. E.; Arnold, R. G.; Bales,R. C. Water Res. 1995, 29, 1673-1680.

(28) Meinders, J. M.; Noordmans, J.; Busscher, H. J. J. ColloidInterface Sci. 1992, 152, 265-280.

(29) McClaine, J. W.; Ford, R. M. Biotechnol. Bioeng. 2002, 78, 179-189.

(30) Gross, M. J.; Albinger, O.; Jewett, D. G.; Logan, B. E.; Bales, R.C.; Arnold, R. G. Water Res. 1995, 29, 1151-1158.

(31) Burks, G. A.; Velegol, S. B.; Paramonova, E.; Lindenmuth, B.E.; Feick, J. D.; Logan, B. E. Langmuir 2003, 19, 2366-2371.

(32) Li, B. K.; Logan, B. E. Colloids Surf., B 2004, 36, 81-90.(33) Razatos, A.; Ong, Y. L.; Boulay, F.; Elbert, D.; Hubbell, J. A.;

Sharma, M. M.; Georgiou, G. Langmuir 2000, 16, 9155-9158.(34) Fang, H. H. P.; Chan, K. Y.; Xu L.-C. J. Microbiol. Methods

2000, 40, 89-97.(35) Ducker, W. A.; Senden, T. J.; Pashley, R. M. Nature 1991, 353

(6341), 239-241.(36) Camesano, T. A.; Abu-Lail, N. I. Biomacromolecules 2002, 3,

661-667.(37) Velegol, S.; Logan, B. E. Langumir 2002, 18, 5256-5262.(38) Emerson, R. J.; Camesano, T. A. Appl. Environ. Microbiol. 2004,

70, 6012-6022.(39) Hemmerle, J.; Altmann, S. M.; Maaloum, M.; Horbert, J. K. H.;

Heinrich, L.; Voegel, J. C. Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 6705-6710.

(40) Mondon, M.; Berger, S.; Ziegler, C. Anal. Bioanal. Chem. 2003,375, 849-855.

(41) Vadillo-Rodriguez, V.; Busscher, H. J.; Norde, W.; de Vries, J.;van der Mei, H. C. J. Colloid Interface Sci. 2004, 278, 251-254.

(42) Camesano, T. A.; Natan, M. J.; Logan, B. E. Langmuir 2000, 16,4563-4572.

(43) Xu, L.-C.; Logan, B. E. Environ. Sci. Technol. 2005, 39, 3592-3600.

(44) Stevens, M. M.; Allen, S.; Davies, M.; Roberts, C. J.; Schacht,E.; Tendler, S. J.; VanSteenkiste, S.; Williams, P. M. Langmuir 2002,18, 6659-6665.

7492 Langmuir, Vol. 21, No. 16, 2005 Xu et al.

(DEVCON epoxy S-31/31345) the latex microsphere to a tiplesssilicon nitride cantilever (Veeco NanoProbe NP-OW) as previouslydescribed.45 The spring constant of the cantilevers (all takenfrom the same wafer) was determined using the thermal tuningmethod (Nanoscope V6.12r2) as 0.0613 ( 0.0030 N/m. Afterpreparation, the colloid probe was stored in the refrigerator (4°C) prior to use. Each probe was inspected before an experimentusing a microscope (Olympus IX70) to ensure probe integrity.

Multiple probes were made at the same time to obtain consistentresponses of the probes between experiments, and each probewas tested for a consistent response to a clean glass surface priorto an experiment. Any probe that did not display behaviorconsistent with other probes from that batch was discarded.

All force measurements were made in phosphate buffersolutions (PBS) with an IS of 1 or 100 mM at a scan rate of 1 Hzand z scan size of 1 µm. A series of PBS (IS, 1 and 100 mM) wereprepared that had pH values of 4.5, 7.3, 8.8, and 10.6. To set theloading force applied by the colloid to a surface at values of 5.4,3.6, 2.4, 1.2, and 0.6 nN, the trigger mode setting was used inthe software with the threshold of the cantilever deflection setat the appropriate value. The ramp delay in the software was setat different times, producing different colloid residence times.The time needed for the tip to reach the desired loading forcefrom the point of contacting sample is defined as the ramp time,while the residence time is defined as the time after the loadingforce has reached the set value until the tip begins to be retracted(Figure 1). Force measurements were performed at residencetimes of 0.001-50 s at four to five randomly selected locationsover every sample studied. Ten force curves were obtained ateach location and at a fixed residence time. To avoid changes insalt concentration due to evaporation, solution levels wereexamined and adjusted as needed with PBS during an experi-ment. Interaction force curves were shown to be reproducible forthe same operating conditions by examining different locationson the surface. Nonreproducible curves were assumed to indicatetip contamination, and therefore the probe was discarded. Forceswere reported as positive when the cantilever was deflected

Figure 1. Colloid probe force curve drawn to show the timescale of the ramp time needed to achieve a loading force of 5.4nN.

Figure 2. AFM 3-D images of (a) bare glass, (b) BSA, (c) lysozyme, and (d) dextran on glass surfaces in 1 mM IS. (Scale: x, y,100 nm/div; z, 10 nm/div).

Colloid and Biopolymer-Coated Surface Adhesion Langmuir, Vol. 21, No. 16, 2005 7493

upward from the sample and negative when the cantilever wasdrawn down to the sample. The adhesion force is the valuemeasured at the point of maximum deflection during the colloidprobe retraction from the surface.

Model. To quantitatively relate the adhesion forces measuredusing the different biopolymer-coated surfaces and colloids withtime, adhesion forces were modeled using a simple power-lawrelationship of

where F is the adhesion force, T is the residence time, and A andn are empirical coefficients determined from log-log plots.

ResultsCharacteristics of Biopolymer Coatings and Col-

loids. In general, biopolymers were covalently bondedand uniformly adsorbed on the glass surfaces (Figure 2).Uniform coatings were also confirmed from force mea-surements with colloid probes by the observation ofconsistent interaction forces in force curves from differentlocations of each sample. The presence of the biopolymercoating on a surface is indicated by surface roughness.For bare glass, the roughness (Rq) was only 0.060 nm,while surfaces coated with BSA, lysozyme, and dextranhad average roughness values of 0.50, 0.30, and 1.50 nm(IS ) 1 mM). On the basis of the surface roughness, thethickness of the biopolymer coatings was calculated asfollows: BSA, 5.2 nm; lysozyme, 3.0 nm; and dextran,11.1 nm. Solution IS changed the conformation of biopoly-mers. In high-IS solution (100 mM) polymers appeared tobecome more folded and compressed, resulting in the

reduction of the effective diameter of the polymers asevidenced by a reduction in the thickness and roughnessof the surface. This phenomenon was consistently observedfor each biopolymer. For example, in 100 mM solution thethickness of the BSA coating was 2.0 nm and the roughnesswas 0.23 nm. As a result, the average diameter of the BSAwas calculated to be 12.9 ( 3.6 nm in the 100 mM ISsolution versus 18.6 ( 5.3 nm in the 1 mM IS solution(Figure 3).

A carboxylated latex microsphere is a “soft” micropar-ticle with the elastic modulus of 3 × 109 N/m2 and Poissonratio of 0.35, while the glass surface is rigid and had aelastic modulus of 64.7 × 109 N/m2 and a Poisson ratio of0.19.43 When a microsphere was coated with BSA, therewas an average of (1.61 ( 0.22) × 10-4 ng of BSA on asurface. The ú potential of uncoated carboxylated latexmicrospheres was -107 ( 1.7 mV in 1 mM solution and-44.6 ( 1.8 mV in 100 mM IS solution (pH ) 7.3). Coatingthe colloid with BSA decreased the ú potential of the latexmicrospheres to -85.6 ( 0.8 and -24.6 ( 1.4 mV in 1 and100 mM IS solutions, respectively. This decrease in úpotential likely results from coverage of the colloid surfacewith BSA that has a lower negative charge than the surfaceat neutral pH

Loading Force. Interaction distances and adhesionforces between an uncoated microsphere colloid probe anda glass surface consistently increased with loading forcein the presence of a biopolymer regardless of biopolymertype. Examples of retraction curves measured at threedifferent loading forces for a BSA-coated microsphereinteracting with a BSA-coated glass, shown in Figure 4,

Figure 3. Section analysis of AFM topographic image of BSA in (a) 1 mM and (b) 100 mM IS solutions.

F ) - ATn (1)


demonstrate that both the maximum retraction forces andmaximum pull-off distances increased with the forceapplied by the colloid. Changing the applied loading forcedid not affect approach curves. The maximum pull-offdistance was 105 ( 45 nm at loading force 0.6 nN, whileit reached 145 ( 30 and 184 ( 49 nm when the loadingforces were increased to 2.4 and 5.4 nN, respectively. Theincrease in adhesion with a range of applied forces is shownin Figure 5a for both uncoated and BSA-coated micro-spheres interacting with a BSA-coated surface in solutionswith a low (1 mM) and high (100 mM) IS. The maximumadhesion force between a BSA-coated microsphere and aBSA-coated glass surface was -2.1 ( 0.5 nN in a 1 mMIS solution, with a slightly lower maximum adhesion forceof -1.4 ( 0.3 nN in a 100 mM solution (5.4 nN appliedforce). Similar trends between the measured adhesionforce and the loading force were observed when lysozyme

or dextran were investigated at IS ) 100 mM (Figure 5b).The increase in the adhesion force is thought to resultfrom an increase in the number of chemical bonds formedbetween the biopolymers when the surfaces are forcedinto greater contact using higher loading forces.

The solution IS affected the measured adhesion forces,but only at the higher loading forces. No significant effectof IS was found for the adhesion forces between twosurfaces at loading forces < 3.6 nN, when both surfaceswere coated with BSA (for example, see Figure 5a).However, at a higher loading force of 5.4 nN, the adhesionforce was larger at the low-IS (1 mM) solution for BSA-coated surfaces and the colloid (uncoated or BSA-coated)than at high IS (100 mM). This finding of adhesion beinginversely related to IS has been previously observed,43

but only a single high loading force of ∼6.0 nN waspreviously examined.

Residence Time. The adhesion force measured be-tween a colloid and biopolymer-coated surfaces increasedwith residence time over a range of 0.001-50 s (fixed pH) 7.3 and IS ) 100 mM) (Figure 6). The largest changesin adhesion force were generally observed when theresidence time was in the range of 1-20 s. This increaseinadhesionwithresidence time in thepresenceofpolymershas been noted by others.39-41

Increasing the loading force produced greater adhesionforces over most of the range of residence times examined(Figure 6). At the lower loading force of 0.6 nN, there wasonly a slight increase in the adhesion forces for residencetimes up to 5 s. For example, the adhesion force of anuncoated microsphere to a BSA-coated surface was only-0.15 nN within a residence time of 5 s, but the forceincreased to -0.47 and -0.84 nN when residence timewas increased to 10 and 20 s, respectively. At a higherloading force of 5.4 nN, the adhesion forces were -3.3,-4.1, and -4.8 nN at residence times of 5, 10, and 20 s(Figure 6a). For the BSA-coated microsphere and BSAsystem, the adhesion forces increased to as much as -7.5nN at a residence time of 10 s, while it is only -1.36 nNat a residence time 0.1 s. These results reinforce the abovefinding that the loading force significantly influences theadhesion between biopolymers and surfaces and showsthat this phenomenon can be generalized to differentresidence times of the colloid at biopolymer-coated sur-faces.

Adhesion forces increased when both a colloid and theflat surface were coated with biopolymers compared toforces measured between an uncoated microsphere anda biopolymer-coated surface for the range of residencetimes examined here (Figure 6). At the shortest residencetime (0.001 s), the adhesion forces measured decreased inthe order BSA-lysozyme > BSA-dextran > BSA-BSA(loading force of 5.4 nN). However, the maximum adhesionforces measured at the longest residence times were largerfor BSA-coated microspheres interacting with BSA-coatedsurfaces, than those for BSA-coated microspheres andlysozyme- or dextran-coated surfaces (loading force of 5.4nN). Recall that BSA, lysozyme, and dextran are nega-tively, positively, and neutrally charged polymers atneutral pH. This suggests that electrostatic interactionswere dominant during short contact times, while thechemical bonding between polymer molecules, polymerrearrangement, and water exclusion became more im-portant to the measured adhesion force with increasingresidence times. The adhesion force between uncoatedmicrospheres and a bare glass surface was also largerthan that observed for an uncoated glass bead and a glass

(45) Li, X.; Logan, B. E. Langmuir 2004, 20, 8817-8822.

Figure 4. Representative approach and retraction force curvesof BSA-coated microspheres to BSA-coated glass surfaces atdifferent loading forces (IS ) 100 mM; pH ) 7.3).

Figure 5. Effect of loading force on the adhesion of biopolymersand surfaces: (a) interactions between BSA and a microsphereor BSA-coated microsphere (residence time ) 1 ms; pH ) 7.3);(b) interactions between lysozyme or dextran and uncoated orBSA-coated microsphere (IS ) 100 mM; residence time ) 1ms).


surface (Figure 7) but smaller than that observed forbiopolymer coated surfaces. For example, the adhesionforce of an uncoated microsphere and a glass surface wasonly -2.1 ( 0.7 nN at a residence time of 10 s (IS ) 100mM), compared to adhesion forces to BSA, lysozyme, anddextran of -4.1 ( 0.7, -4.7 ( 1.2, and -4.0 ( 0.4 nN,respectively. The adhesion force between a microsphereand an uncoated glass surface also reached a maximum

value sooner (within ∼10 s; Figure 7) than observed forbiopolymer-coated surfaces (>20-50 s). The adhesion forcebetween the uncoated latex microsphere and glass waslarger for an IS of 1 mM than an IS of 100 mM, at thelongest residence time, consistent with previous findingson the effect of IS on adhesion.43

There was no change in the adhesion force measuredbetween an uncoated glass bead and an uncoated glasssurface up to a residence time of 50 s (Figure 7). Weattribute the lack of a change in the adhesion force of theglass-glass system to the absence of polymers on the glasssurface. The observation that adhesion force increasedwith the uncoated latex microsphere with an increase inresidence time suggests that there are polymers on thelatex microsphere itself. This indicates the latex micro-sphere is not a truly “hard surface” in comparison to theglass bead.

pH. The solution pH influenced both the approach andretraction force curves. A long-range repulsive forceobserved in approach curves at the two higher pHs (8.8and 10.6) was absent at the lower pH of 4.5, for both theuncoated and BSA-coated microspheres with a BSA-coatedglass surface (Figure 8). This lowest pH is below theisoelectric point of BSA (∼6.1). A long-range repulsiveforce was also observed for lysozyme (isoelectric point ∼9.0) with an uncoated microsphere at pHs of 8.8 and 10.6and with a BSA-coated microsphere at pHs of 7.3-10.6.For dextran, a repulsive force was observed over the pHrange of 7.3-10.6 in interactions with either uncoated- or

Figure 6. Effect of residence time on the adhesion forces of different biopolymers with uncoated (a, c, and e) and BSA-coatedmicrospheres (b, d, and f) for three different loading forces (9, 5.4 nN; b, 2.4 nN; 2, 0.6 nN) in 100 mM IS at pH 7.3.

Figure 7. Time-dependent adhesion forces of glass surfacewith uncoated microspheres or glass beads in 1 or 100 mM IS(pH ) 7.3; loading force ) 5.4 nN).


BSA-coated microspheres. The observation of long-rangerepulsive force at higher pH is consistent with the findingin the interaction of two different biopolymers (BSA andapoferritin) and silica sphere by Valle-Delgado et al.,15,46

who showed that the presence of hydration forces at highpH and a high salt concentration resulted in non-DLVOrepulsive interactions.

Retraction curves in low-pH solutions exhibited long-range, sawlike patterns that are indicative of multiplebonds being broken between the colloid and polymers asthe probe is retracted from a surface (Figure 8). Theadhesion forces are larger when both surfaces were coatedby biopolymers than those when only one surface wascoated. For example, the adhesion forces were as large as-1.9 ( 0.5 nN at a pH 4.5 when both surfaces were BSA-coated and decreased to -0.6 ( 0.2 nN when an uncoatedmicrosphere was used (see Figure 9a). Similar results wereobserved with lysozyme and dextran (Figure 9b, c). Resultsshowed that the adhesion force of a microsphere (uncoatedor BSA-coated) to biopolymer-coated surfaces alwaysdecreased with an increase in pH at IS ) 100 mM. Notethat, for all the cases, the adhesion forces betweenbiopolymers and uncoated microspheres were significantlysmaller than those between biopolymers and BSA-coatedmicrospheres in the more acidic solution (pH of 4.5). Incontrast, there was no significant difference in theadhesion forces in the interactions of microspheres withbiopolymer-coated surfaces in the higher pH solutions (pHof 8.8 or 10.6). This relatively larger adhesion force at thelower pH could be due to either increased hydrogenbonding which results from the lower pH conditions orincreased electrostatic repulsive forces at the higher pHs.25

Ionic Strength. A comparison of the adhesion data onthe basis of IS reveals that adhesion force was inverselyrelated to IS for both uncoated- and BSA-coated micro-spheres and all three different biopolymer coated surfaces(p < 10-5 for all cases, Analysis of Variance test; Figure10). This finding is in contrast to what we would expecton the basis of a DLVO model for two like-charged surfaces.It is expected that as the IS decreases, two surfaces wouldhave less adhesion due to an increase in the repulsivelayer thickness between the two surfaces. Similarly, it iswell-known that bacterial adhesion generally increaseswith ionic strength.20-22 On the basis of the opposite findinghere, that adhesion decreased with IS, we surmise thatconformational changes (Figure 3) of biopolymers due tothe different IS produced unanticipated changes inmolecular-scale interactions between the biopolymers andsurfaces, resulting in lower adhesion forces at high ISthan at lower IS.42

Model. Examples of several colloid-surface adhesionforce interactions are shown in Figures 11 and 12, withmodel results from all experiments shown in Table 1. Theparameter n using this model was nearly constant for thevarious conditions, with an average value of 0.29 ( 0.06((SD), and all values within (2 SD except for those withuncoated microspheres and BSA- or lysozyme-coatedsurfaces at the lowest loading force of 0.6 nN. Themagnitude of A varied for the different experimentalconditions, but in general larger values of A wereassociated with higher loading forces, low pH, or low IS(Table 1).

Discussion

The presence of a biopolymer at a surface consistentlyproduced an increase in the adhesion force between thecolloid and the surface over time. This change in adhesionwith the residence time was observed for all threebiopolymer types, suggesting that the increase in adhesionwas less of a function of the type of polymer and more aconsequence of the physical presence of the biopolymer inthe space between the surfaces. Using an empirical modelto correlate residence timeandadhesiondata, itwasshownthat the model constant, n, was the same for almost all

(46) Valle-Delgado, J. J.; Molina-Bolıvar, J. A.; Galisteo-Gonzalez,F.; Galvez-Ruiz, M. J. M.; Feiler, A.; Rutland, M. W. J. Phys.: Condens.Matter 2004, 16, S2383-S2392.

Figure 8. Representative approach and retraction force curvesof colloids to BSA-coated glass surfaces at different pH (IS )100 mM).

Figure 9. Effect of pH and type of biopolymer on adhesionforces for uncoated and BSA-coated microspheres (IS ) 100mM): (a) BSA, (b) lysozyme, and (c) dextran.


polymers under different conditions of IS and pH. Webelieve this consistent effect of residence time is primarilya result of two factors: water exclusion and biopolymerrearrangement. As the colloid is forced into contact witha surface, water is pushed out from between these surfaces.This change in water content should change the orienta-tion of charged portions of the molecule with itself andthe two surfaces. Over relatively long periods of time(∼20-50 s), the polymer molecules will rearrange, bridge,and then bind to the opposing surface until all bonds reachthe lowest energy state for the imposed condition. With-drawing the colloid from the surface after this biopolymerrearrangement therefore requires a greater force relativeto that initially needed.

The effect of residence time on the measured adhesionforce produced by polymers is supported by previousstudies where polymers are present, ranging from biopoly-mer-coated material surfaces to bacteria. For example,Mondon and co-workers40 observed the adhesion forcebetween a protein-modified AFM tip and titanium surfaces

increased with interaction time, reaching a maximumadhesion force within∼2 s. Vadillo-Rodriguez et al.41 foundusing bacteria that the adhesion force between the AFMtip and the cell surface increased with interaction time upto 100 s. They attributed this increase in adhesion forceto the increased attachment of extracellular polymericsubstances to the AFM tip over time. In experiments oninteractions of a fibrinogen-coated AFM tip with a silicasurface, Hemmerle et al.39 observed multiple consecutiveruptures during tip retraction and found that the meannumber and strength of these ruptures increased steadilywith interaction time. They suggested that the extent ofbonding increased with retention time resulting in in-creased adhesion forces. It was found in the current studythat, in the absence of an added biopolymer, the latexmicrosphere exhibited an increase in adhesion force overtime to a glass surface, while the uncoated glass bead didnot. Thus, we conclude that carboxylated functional groupson the microsphere surface function in the same manneras covalently bonded polymers on its surface. This finding

Figure 10. Effect of ionic strength and type of biopolymer on adhesion forces to (a) uncoated microspheres and (b) BSA-coatedmicrospheres (pH ) 7.3; loading force ) 6.0 nN).

Table 1. Fitting Parameters for the Model Used To Relate Adhesion Force (F) to the Residence Time (T)

sample colloid loading force (nN) pH IS (mM) A n R2

loading force effect BSA uncoated microsphere 5.4 7.3 100 2.04 ( 0.14 0.26 ( 0.02 0.982.4 7.3 100 1.39 ( 0.19 0.33 ( 0.04 0.980.6 7.3 100 0.02 ( 0.003 1.31 ( 0.05a 0.91

BSA BSA-coated microsphere 5.4 7.3 100 3.45 ( 0.22 0.29 ( 0.02 0.992.4 7.3 100 1.75 ( 0.12 0.35 ( 0.02 0.990.6 7.3 100 0.34 ( 0.08 0.35 ( 0.08 0.90

lysozyme uncoated microsphere 5.4 7.3 100 2.29 ( 0.27 0.34 ( 0.03 0.942.4 7.3 100 1.20 ( 0.14 0.28 ( 0.04 0.950.6 7.3 100 0.09 ( 0.01 0.48 ( 0.04a 0.98

lysozyme BSA-coated microsphere 5.4 7.3 100 3.88 ( 0.45 0.24 ( 0.04 0.832.4 7.3 100 1.22 ( 0.08 0.35 ( 0.02 0.990.6 7.3 100 0.38 ( 0.06 0.38 ( 0.04 0.97

dextran uncoated microsphere 5.4 7.3 100 2.35 ( 0.16 0.19 ( 0.02 0.952.4 7.3 100 1.81 ( 0.07 0.22 ( 0.01 0.990.6 7.3 100 0.25 ( 0.03 0.26 ( 0.04 0.92

dextran BSA-coated microsphere 5.4 7.3 100 2.96 ( 0.18 0.25 ( 0.02 0.982.4 7.3 100 1.87 ( 0.14 0.29 ( 0.02 0.960.6 7.3 100 1.21 ( 0.12 0.37 ( 0.03 0.98

pH effect lysozyme uncoated microsphere 2.4 7.3 100 1.20 ( 0.14 0.28 ( 0.04 0.952.4 4.5 100 1.30 ( 0.18 0.31 ( 0.04 0.94

BSA BSA-coated microsphere 5.4 8.8 100 0.98 ( 0.12 0.37 ( 0.04 0.985.4 7.3 100 3.45 ( 0.22 0.29 ( 0.02 0.99

ionic strength effect dextran BSA-coated microsphere 5.4 7.3 100 2.96 ( 0.18 0.25 ( 0.02 0.985.4 7.3 1 5.69 ( 0.52 0.21 ( 0.03 0.932.4 7.3 100 1.87 ( 0.14 0.29 ( 0.02 0.962.4 7.3 1 4.86 ( 0.45 0.25 ( 0.03 0.960.6 7.3 100 1.21 ( 0.12 0.37 ( 0.03 0.980.6 7.3 1 2.86 ( 0.14 0.26 ( 0.02 0.99

av (mean ( SD) 0.29 ( 0.06a Not included in the calculation of the average value of n.


of polymers at the microsphere surface is consistent withthe findings of others of a “hairy” surface of the latexmicrospheres, and direct observations using AFM ofsurface heterogeneity of latex microsphere surfaces.47,48

The observation that the value of n in our model wasconstant for so many different conditions, combined withthese other studies, is good evidence of the physical basisof this residence time effect.

The magnitude of the adhesion force was a function ofthe type of the biopolymer, with the adhesion force varyingwith the AFM protocol (loading force) and the solutionchemistry (pH and IS). The measured adhesion forces inthe presence of a biopolymer increased by as much as 15×when the applied load to the colloid was increased from0.6 to 5.4 nN (9×; Figure 5). The difference in the measuredadhesion force with the various polymers and solutionconditions is reflected in the large variations obtained forthe coefficient A in our model. A high A value appears tobe correlated with increased loading force. The largeincrease in the adhesion with loading force suggests thatpushing the colloid into a polymer layer promotes in-creased binding of polymers with the introduced surface,resulting in stronger chemical bonds. The longer time

needed to apply a higher loading force (ramp time) cannotexplain this trend with loading force as the additionaltime needed to achieve the highest loading force is onlyabout 0.05 s, or half the time needed at the lower loadingforce (2.4 nN).

The solution IS affected the adhesion force, as shownby a lower adhesion force with an increase in the IS (Figure10). We believe that this reduction in adhesion is a resultof the change in the initial conformation of the biopolymerin a solution at a fixed IS (Figure 3). In a low-IS solution,biopolymers will be in a more unfolded state and furtherexpanded into solution, providing more potential bindingsites when a surface is pressed onto the polymer.49,50

However, as IS is increased, the polymers will becomemore folded and compressed, forming a denser core andproviding fewer interaction sites. The more condensedstructure of a polymer at higher IS could result in theformation of weaker bonds with a surface, leading to asmaller adhesion force as observed here (Figure 11c). Thisspeculation is supported by the observed changes inmolecule conformation shown in Figure 3. Depletion ofwater within a polymer layer also has an important effectrelative to IS because the concentration of cations at acharged interface, and therefore within the biopolymersand the colloid surface layer, increases with bulk saltconcentration.50 Experiments by Vakarelski et al.51 showedthat potassium ions adsorbed to silica and mica producedlower adhesion forces with increasing IS. Similarly, in aprevious study we showed that adhesion forces betweencolloid and protein layer surface decreased as the saltconcentration increased.43

The solution pH can also affect the adhesion of abiopolymer to a colloid surface. We believe that this resultis primarily a function of the changes in hydrogen bondingand electrostatic interactions as a function of pH. Theadhesion between two surfaces decreased when pH wasincreased over a range of 4.5-10.6, as shown by forcevalues in Figure 9 and by values of A as shown in Table1. At a low pH, hydrogen bonding between carboxyl groupsin polymers should increase adhesion forces, while at highpH we expect long-range repulsive interactions to domi-nate due to the electrostatic double-layer forces betweentwo negatively charged biopolymer and colloid surfaces.Repulsive forces were observed here in approach curvesat high pH (see Figure 8). Our observations on the effectof pH on adhesion are consistent with the finding ofAhimou et al.25 In their study it was shown that there wasa high adhesion force between an AFM tip and a bacterialcell surface at a low pH ) 4, with a repulsive force observedat a higher pH ) 10.

Figure 11. Examples of log-log plots of adhesion force (F)versus time (T) for surfaces coated with different biopolymers:(a) lysozyme and uncoated microsphere (IS ) 100 mM; pH )7.3); (b)BSAandBSA-coatedmicrosphere (IS ) 100mM; loadingforce ) 5.4 nN); and (c) dextran and BSA-coated microsphere(pH 7.3). Lines shown are based on a linear regression.

Figure 12. Example of a best-fit model result for adhesionforces measured for BSA-coated surfaces to BSA-coated mi-crospheres (IS ) 100 mM; pH ) 7.3). Lines shown are best fitsbased on eq 1 using constants listed in Table 1.


Relevance of These Results to Bacterial Adhesion.Bacterial cell surfaces are composed of a complex mixtureof proteins, phospholipids, polysaccharides, and otherbiopolymers. These exopolymers are known to be animportant factor in bacterial adhesion to a surface. Thecurrent work with polymer-coated surfaces here highlightsthe importance of particle residence time on adhesion toa surface. In all cases examined here, increased residencetime increased the adhesion force in the presence of apolymer. However, we found that increasing the ISdecreased adhesion, a result that is opposite to thatobserved in most bacterial adhesion studies. This differentfinding may be a consequence of the way that AFMexperiments are conducted versus how bacteria approach

and contact a surface. In an AFM experiment, the colloidprobe is forced onto the biopolymer-coated surface. Themain mechanism driving bacterial adhesion to a surfaceis normally diffusion, although convective flow and cellsedimentation are also important collision mechanisms.52

By forcing an AFM tip or colloid onto a surface, we maybe missing an important component to the adhesionprocess, thereby underestimating the role of long-rangeinteractions such as steric and electrostatic effects. Thus,while we can simulate the interactions between polymersand surfaces once a particle collides with a surface, wemay currently be missing interactions, particularly at lowIS, that occur on the approach of a particle to a surfacethat may prohibit collisions. The development of AFMprotocols that better mimic the approach and probabilisticcollision of a bacterium with a surface will help us to betterstudy and understand the role of IS in bacterial adhesion.

Acknowledgment. This material is based upon worksupported by the National Science Foundation (NSF)Grants CHE-0089156 and CHE-0431328.

LA0509091

(47) Elimelech, M.; O’Melia, C. R. Colloids Surf. 1990, 44, 165-178.(48) Tan, S.; Sherman, R. L. Jr.; Qin, D.; Ford, W. T. Langmuir 2005,

21, 43-49.(49) Damaschun, G.; Damaschun, H.; Gast, K.; Zirwer, D. J. Mol.

Biol. 1999, 291, 715-725.(50) Abu-Lail, N. I.; Camesano, T. A. Biomacromolecules 2003, 4,

1000-1012.(51) Vakarelski, I. U.; Ishimura, K.; Higashitani, K. J. Colloid

Interface Sci. 2000, 227, 111-118.(52) Logan, B. E. Environmental Transport Processes; John Wiley &

Sons: New York, 1999.


Residence time, loading force, pH, and ionic strength affect adhesion forces between colloids and...

Documents

Transcript of Residence time, loading force, pH, and ionic strength affect adhesion forces between colloids and...