Streptococcus mutans and Streptococcus intermedius Adhesion to

Aug 27, 2008 - Telephone: +31 50 3633140. ... to fibronectin films in a parallel plate flow chamber was completely in line with this modeling, while i...
0 downloads 0 Views 227KB Size
Download PDF
10968

Langmuir 2008, 24, 10968-10973

Streptococcus mutans and Streptococcus intermedius Adhesion to Fibronectin Films Are Oppositely Influenced by Ionic Strength Henk J. Busscher,† Betsy van de Belt-Gritter,† Rene J. B. Dijkstra,† Willem Norde,†,‡ and Henny C. van der Mei*,† Department of Biomedical Engineering, UniVersity Medical Center Groningen, and UniVersity of Groningen, Antonius Deusinglaan 1, 9713 AV Groningen, The Netherlands, and Laboratory of Physical Chemistry and Colloid Science, Wageningen UniVersity, P.O. Box 8038, 6700 EK Wageningen, The Netherlands ReceiVed March 5, 2008. ReVised Manuscript ReceiVed July 15, 2008 Bacterial adhesion to protein-coated surfaces is mediated by an interplay of specific and nonspecific interactions. Although nonspecific interactions are ubiquitously present, little is known about the physicochemical mechanisms of specific interactions. The aim of this paper is to determine the influence of ionic strength on the adhesion of two streptococcal strains to fibronectin films. Streptococcus mutans LT11 and Streptococcus intermedius NCTC11324 both possess antigen I/II with the ability to bind fibronectin from solution, but S. intermedius binds approximately 20× less fibronectin than does the S. mutans strain under identical conditions. Both strains as well as fibronectin films are negatively charged in low ionic strength phosphate buffered saline (PBS, 10× diluted), but bacteria appear uncharged in high ionic strength PBS. Physicochemical modeling on the basis of overall cell surface properties (cell surface hydrophobicity and zeta potentials) demonstrates that both strains should favor adhesion to fibronectin films in a high ionic strength environment as compared to in a low ionic strength environment, where electrostatic repulsion between equally charged surfaces is dominant. Adhesion of S. intermedius to fibronectin films in a parallel plate flow chamber was completely in line with this modeling, while in addition atomic force microscopy (AFM) indicated stronger adhesion forces upon retraction between fibronectin-coated tips and the cell surfaces in high ionic strength PBS than in low ionic strength PBS. Thus, the dependence of the interaction on ionic strength is dominated by the overall negative charge on the interacting surfaces. Adhesion of S. mutans to fibronectin films, however, was completely at odds with theoretical modeling, and the strain adhered best in low ionic strength PBS. Moreover, AFM indicated weaker repulsive forces upon approach between fibronectin-coated tips and the cell surfaces in low ionic strength PBS than in high ionic strength PBS. This indicated that the dependence of the interaction on ionic strength is dominated by electrostatic attraction between oppositely charged, localized domains on the interacting surfaces, despite their overall negative charge. In summary, this study shows that physicochemical modeling of bacterial adhesion to proteincoated surfaces is only valid provided the number of specific interaction sites on the cell surfaces is low, such as on S. intermedius NCTC11324. Nonspecific interactions are dominated by specific interactions if the number of specific interaction sites is large, such as on S. mutans LT11. Its ionic strength dependence indicates that the specific interaction is electrostatic in nature and operative between oppositely charged domains on the interacting surfaces, despite the generally overall negatively charged character of the surfaces.

Introduction Viridans streptococci, although part of the commensal microflora of the human oral cavity, may cause a variety of diseases, ranging from dental caries to endocarditis. Dental caries initiate with the adhesion of these streptococci to salivary films on intraoral surfaces, while endocarditis results from adhesion of viridans streptococci to extracellular matrix proteins, such as fibronectin.1 It is unclear what mechanism(s) these strains utilize in their adhesion to protein-coated surfaces. On the one hand, their adhesion is controlled by the overall physicochemical cell surface properties,2-4 and their energies of interaction with a substratum surface can be calculated from contact angles measured with liquids on the interacting surfaces and zeta potentials using a thermodynamic approach or the (extended) * To whom correspondence should be addressed. Telephone: +31 50 3633140. Fax: +31 50 3633159. E-mail: [email protected]. † University of Groningen. ‡ Wageningen University.

(1) Chia, J. S.; Yeh, C. Y.; Chen, J. Y. Infect. Immun. 2000, 68, 1864–1870. (2) Bos, R.; van der Mei, H. C.; Busscher, H. J. FEMS Microbiol. ReV. 1999, 23, 179–229. (3) Hermansson, M. Colloids Surf., B 1999, 14., 105–119. (4) Sharma, P. K.; Rao, K. H. Colloids Surf., B 2003, 29, 21–38.

Derjaguin-Landau-Verwey-Overbeek (DLVO) theory.5 More often than not, however, these physicochemical approaches based on overall surface properties fail, especially when applied to understand bacterial adhesion to protein-coated surfaces. On the other hand, viridans streptococci, including Streptococcus mutans and Streptococcus intermedius, express one or more members of a family of structurally and antigenically related surface proteins, termed antigen I/II, that are essential for their adhesion to salivary films and fibronectin-coated surfaces.6 In reality, nonspecific contributions originating from the overall cell surface properties work jointly with specific contributions from localized, high affinity domains on the interacting surfaces in establishing adhesion, and this is what makes a proper understanding of bacterial adhesion to surfaces in general very difficult.7 S. mutans LT11 and S. intermedius NCTC11324 both possess antigen I/II surface proteins and bind fibronectin, albeit in phosphate buffered saline (PBS) S. mutans binds about 20× more fibronectin from solution than does S. intermedius.6 Interestingly, this differential ability to bind fibronectin cannot (5) Van Oss, C. J. Colloids Surf., B 2007, 54, 2–9. (6) Petersen, F. C.; Assev, S.; van der Mei, H. C.; Busscher, H. J.; Scheie, A. A. Infect. Immun. 2002, 70, 249–256. (7) Busscher, H. J.; Cowan, M. M.; van der Mei, H. C. FEMS Microbiol. ReV. 1992, 8, 199–209.

10.1021/la8016968 CCC: $40.75  2008 American Chemical Society Published on Web 08/27/2008

Streptococcal Adhesion to Fibronectin Films

Langmuir, Vol. 24, No. 19, 2008 10969

be deduced from the overall cell surface characteristics of the two strains, as both are hydrophilic, as judged from water contact angles, and negatively charged. This paper is aimed at determining up to what extent nonspecific and specific interactions (i.e., molecular recognition phenomena) contribute to the adhesion of these two strains to fibronectin films and how their adhesion is influenced by ionic strength. To this end, bacterial adhesion to fibronectin, adsorbed on glass slides in a parallel plate flow chamber, is studied, while the interaction forces between the cell surfaces and fibronectin are determined using atomic force microscopy (AFM). All experiments are carried out in high and low ionic strength PBS.

Materials and Methods Bacterial Strains and Growth Media. The streptococcal strains used in this study included S. mutans LT11 and S. intermedius NCTC11324, both in possession of antigen I/II and able to bind to fibronectin films, albeit S. mutans LT11 binds fibronectin much more extensively than S. intermedius NCTC11324.6 The strains were stored at -80 °C in brain heart infusion broth (BHI; OXOID, Basingstoke, U.K.) supplemented with 7% (v/v) dimethyl sulfoxide (DMSO). Strains were grown for 24 h in BHI at 37 °C in 5% CO2. This culture was used to inoculate a second overnight culture yielding early stationary phase bacteria,6 which were harvested by centrifugation, washed twice, and resuspended in high (10 mM potassium phosphate and 150 mM NaCl, pH 7.0) or low ionic strength PBS (10× diluted PBS), to a final concentration of 3 × 108 mL-1. The presence of antigen I/II in exponential and early stationary phase bacteria has been demonstrated by immunoblotting.6 Bacterial Adhesion to Fibronectin Films in a Parallel Plate Flow Chamber. Streptococcal adhesion to the bottom glass plate of a parallel plate flow chamber was determined with a phase-contrast microscope coupled to a CCD-MXR camera and an image analyzer. Fibronectin (from human plasma, Sigma Aldrich, Zwijndrecht, The Netherlands) was dissolved in high and low ionic strength PBS to a final concentration of 20 µg mL-1. The bottom glass plate was coated with fibronectin for 2 h at room temperature. The experiments were performed with bacteria suspended in high and low ionic strength PBS to reveal the influence of electrostatic interactions, which are generally shielded at high ionic strength. The bacterial flow rate was adjusted to 1.4 mL min-1 under the influence of hydrostatic pressure, yielding a shear rate of 15 s-1. Live images were taken every 0.5-2 min during the first 30 min and thereafter at 10-30 min intervals up to 4 h, when flow was stopped. The number of adhering bacteria was determined from the stored images. The initial deposition rate was calculated as the number of bacteria that adhered during the first 20 min per unit of time and area, while the number of streptococci adhering after 4 h was recorded as a number of weakly and more strongly adhering organisms. Weakly adhering organisms were considered those organisms that were able to adhere under the prevailing shear forces of the experiment but that could not withstand the passage of a liquid-air interface through the flow chamber. Analysis of the forces accompanying the passage of a liquid-air interface over adhering micrometer-sized particles has indicated that the detachment force acting perpendicular to the surface amounts to approximately 10-7 N. Atomic Force Microscopy. For AFM, bacteria were immobilized in an isopore polycarbonate membrane,8 while the AFM tips (Veeco Inc., New York; NP probes, nominal tip radius 20 nm) were coated with a fibronectin film by immersion in a fibronectin solution (20 µg mL-1 in high or low ionic strength PBS) for 30 min with the aid of a micromanipulator. All membranes with immobilized bacteria (see Figure 1 for an example) and fibronectin-coated AFM tips were immediately used for measurements. AFM measurements were done at room temperature in high and low ionic strength PBS using an optical level microscope (Nanoscope III Digital Instruments, Santa Barbara, CA). An array of 32 × 32 (8) Kasas, S.; Ikai, A. Biophys. J. 1995, 68, 1678–1680.

Figure 1. S. mutans LT11 trapped in a polycarbonate membrane filter as imaged by AFM using an uncoated silicon-nitride AFM tip.

force-distances curves with z-displacements of 100-200 nm at z-scan rates = 10 Hz were collected over the entire field of view when a bacterium was imaged. The slopes of the retraction force curves in the region where the probe and sample are in contact were used to convert the voltage into cantilever deflection. The conversion of deflection into force was carried out using a nominal spring constant for the fibronectin-coated tips of 0.06 N m-1, as determined by the Cleveland method.9 Force-distance curves taken over the top of each bacterium studied were analyzed in order to determine various characteristic parameters. Approach curves were fitted to an exponential function, where the interaction force F is described as

F ) F0 exp(-d ⁄ λ)

(1)

in which F0 is the force at zero separation between the interacting surfaces, d is the separation distance, and λ is the decay length of the interaction force F. The retracting curves were used to generate adhesion maps. Adhesion maps were produced by taking the strongest adhesive force detected during retraction at each position as the value for adhesion and by plotting that value against the x-y position of each force-distance curve. From the adhesion maps, an area of ∼800 × 800 nm2 over the top of each bacterium was selected to generate an adhesion distribution histogram from which median, mode, and range values for the adhesion force Fadh were calculated between functionalized AFM tips and the bacterial cell surface for each experimental condition studied. Five different organisms were studied for each particular case, and a new tip was used for each bacterium. However, contamination of the tip by bacterial cell surface components or desorption of fibronectin from the tip was seldom observed, as judged from aberrant force-distance curves measured on glass before and after each measurement on a bacterial cell surface. Physicochemical Characterization of the Interacting Surfaces. Zeta potentials of the two strains were determined by particulate microelectrophoresis in high and low ionic strength PBS at pH 7.0 (Lazer Zee Meter 501, PenKem, Bedford Hills, NY), equipped with image analysis options. Briefly, the microelectrophoresis chamber was filled with a bacterial suspension, and a voltage difference of 150 V was applied over the chamber. The velocity of each individual bacterium was determined by image sequence analysis, and from this their zeta potential was derived, assuming the HelmholtzSmoluchowski equation holds. In order to ascertain that the bacterial cell surface did change during, for instance, flow chamber adhesion experiments or AFM measurements, zeta potentials were also (9) Cleveland, J. P.; Manne, S.; Bocek, D.; Hansma, P. K. ReV. Sci. Instrum. 1993, 64, 403–405.

10970 Langmuir, Vol. 24, No. 19, 2008

Busscher et al.

Table 1. Zeta Potentials (mV) in High (PBS) and Low (PBS/10) Ionic Strength PBS, Contact Angles (deg), and Surface Free Energies (mJ/m2) of Fibronectin Films As Well As of S. mutans and S. intermedius Cell Surfacesa fibronectin film parameter

PBS

zeta potential θwater θformamide θmethyleneiodide θR-bromonaphthalene γLW γAB γγ+

-26

S. mutans LT11

PBS/10

PBS

-42

0(1)

53 37 44 29 38.3 9.6 23.4 1.0

S. intermedius NCTC11324

PBS/10

PBS

-22(-20)

0(1)

26 23 52 37 34.5 18.8 47.4 1.9

PBs/10 -15(-13) 39 43 51 36 35.0 6.2 47.4 0.2

a Zeta potentials in parenthesis were measured after 4 h exposure to the corresponding buffer. SD in bacterial zeta potentials and contact angles over three separate cultures amount to 4 mV and 4°, respectively. SD in fibronectin film zeta potentials and contact angles over three separately prepared substrata amount to 2.5 mV and 10°, respectively.

measured on bacteria after 4 h of exposure to high and low ionic strength PBS. Zeta potentials of the fibronectin films were derived from the pressure dependence of the streaming potentials measured in a parallel plate flow chamber according to

∆Estr εε0 ζ ) ∆p ηκsp

(2)

where εε0 is the dielectric permittivity, η is the viscosity, and κsp is the specific conductivity of the liquid. The walls of the flow chamber were fibronectin-coated microscope slides separated by a 0.1 mm Teflon gasket, while two rectangular platinum electrodes (5 × 25 mm2) were located at both ends of the flow chamber. Streaming potentials were measured in high and low ionic strength PBS at 10 different pressures ranging from 5 to 40 kPa. Each pressure was applied for 10 s in both directions. The surface free energies of the streptococcal strains and fibronectin films were assessed by contact angle measurements employing the sessile drop technique on bacteria deposited onto membrane filters. To this end, bacteria, resuspended in demineralized water, were deposited onto a 0.45 µm pore size filter (Millipore, Billerica, MA) using negative pressure. A lawn of approximately 50 stacked layers of bacteria was produced on the filter. The filters, as well as the fibronectin films on glass plates, were dried for 30 min in order to measure plateau contact angles with water. Contact angles were subsequently determined with water, formamide, 1-bromonaphthalene, and methyleneiodide and also with high and low ionic strength PBS. Surface free energies were calculated according to Van Oss5 and expressed in terms of a Lifshitz-van der Waals (γLW) and acid-base (γAB) component, with the latter involving an electrondonating (γ-) and accepting (γ+) parameter. These surface free energies were subsequently employed in AB ∆Gadh ) ∆GLW adh + ∆Gadh

in which the

∆GLW

parameter can be calculated from

LW LW ∆GLW adh ) - 2 √γmv - √γlv

)(√γ

(

while the

∆GAB

(3)

LW sv -

√γlvLW)

(4)

parameter can be calculated from

+ + + ∆GAB adh ) 2 √γlv √γsv + √γmv - √γlv + √γlv √γsv + √γmv -

[ (

) ( √ ) - √γ γ + γlv

+ sv mv -

√γ-svγ+mv] (5)

where γlv, γsv, and γmv are the liquid-vapor, solid-vapor, and microorganism-vapor interfacial free energies, respectively.

Results Theoretical Modeling of Adhesion of S. mutans and S. intermedius to Fibronectin Films. Table 1 summarizes the contact angles and zeta potentials of the streptococcal cell surfaces

Table 2. Adhesion Energies (mJ/m2) between S. mutans and S. intermedius Cell Surfaces and Adsorbed Fibronectin Films in High (PBS) and Low (PBS/10) Ionic Strength PBS S. mutans LT11 parameter

PBS

PBS/10

LW

-3.7 13.3 absent 9.6

-4.0 7.8 repulsive .3.8

∆G ∆GAB ∆GEL ∆GTOTAL

S. intermedius NCTC11324 PBS -3.8 12.9 absent 9.1

PBS/10 -4.2 6.0 repulsive .1.9

and fibronectin films in high and low ionic strength PBS. Both bacterial strains are hydrophilic as judged by their water contact angles. Interestingly, contact angles measured with high or low ionic strength PBS are not significantly different from those measured with water (data not shown). When the measured contact angles are converted into surface free energies, this yields strong electron-donating surface free energy parameters, γ-, and small electron-accepting parameters, γ+. Fibronectin films are more hydrophobic than the bacterial cell surfaces, which translates to a smaller electron-donating surface free energy parameters γ-. Whereas ionic strength has little or no influence on the contact angles measured, it strongly influences the zeta potentials of the interacting surfaces. The bacterial cell surfaces are negatively charged in low ionic strength PBS, but due to counterion adsorption they appear uncharged in high ionic strength PBS. Importantly, zeta potentials did not change significantly after 4 h of exposure to high or low ionic strength buffers, indicating that the cell surfaces did not undergo major changes during flow chamber adhesion experiments or AFM measurements. Also, the fibronectin films are more negatively charged in low ionic strength PBS than in high ionic strength PBS. Table 2 summarizes the interaction free energies that can be calculated from the above overall physicochemical surface properties. Thermodynamic analysis shows that, for both strains and regardless of ionic strength, the Lifshitz-van der Waals interaction is attractive at contact, whereas the acid-base interaction is unfavorable. The electrostatic interaction, resulting from electrical double layer overlap, is difficult to calculate quantitatively, because the interaction distance between the cell surface and the fibronectin film, the value of which is required to calculate the interaction in the DLVO theory, is unknown. Due to the absence of charge on the bacterial cell surfaces, however, there is no such electrostatic interaction operating in high ionic strength PBS, but in low ionic strength PBS there must be a sizable electrostatic repulsion. In conclusion, from the evaluation of the overall physicochemical surface properties of the streptococci and fibronectin films, it is inferred that it is unfavorable for both strains to adhere to fibronectin films, regardless of ionic strength. In low ionic

Streptococcal Adhesion to Fibronectin Films

Figure 2. Phase contrast images of S. mutans LT11 (left panel) and S. intermedius NCTC11324 (right panel) after 4 h of adhesion to a fibronectin film in a parallel plate flow chamber. The top series is pertinent to low ionic strength PBS, while the bottom series refers to high ionic strength PBS. Bar markers represent 25 µm.

strength PBS, however, the overall interaction is much less favorable than that in high ionic strength PBS due to the absence of electrostatic repulsion in high ionic strength PBS, with little or no difference to be expected between the adhesion of S. mutans and S. intermedius. The sole attraction operative for both strains in high and low ionic strength PBS is constituted by Lifshitzvan der Waals forces. Adhesion of S. mutans and S. intermedius to Fibronectin Films. Figure 2 illustrates the impact of ionic strength on the adhesion of both strains from phase contrast images after 4 h of adhesion in high and low ionic strength buffers. Interestingly, the impact of ionic strength on the adhesion of the two strains is completely opposite, as can also be seen from the adhesion kinetics of the two strains under both ionic strength conditions (shown in Figure 3). Consequently, and despite the theoretical similarity in free energy of adhesion for the two strains, experimental differences in initial deposition rates and numbers of adhering bacteria after 4 h of adhesion to fibronectin films between S. mutans LT11 and S. intermedius NCT11324 are enormous and statistically highly significant (p < 0.001) both in high and low ionic strength PBS (see Table 3). In high ionic strength PBS, S. mutans LT11 deposits slower and in lower numbers than it does in low ionic strength PBS. Moreover, 27% of the relatively few bacteria that adhere from high ionic strength PBS adhere weakly (i.e., are detached by a passing air bubble through the flow chamber). Oppositely, S. intermedius NCTC11324 deposits slower and in lower numbers in low ionic strength PBS than in high ionic strength PBS. Here, 65% of the few bacteria that adhere from low ionic strength PBS adhere weakly. Herewith, S. intermedius NCTC11324 behaves in line with the above theoretical modeling, but the behavior of S. mutans LT11 is completely at odds with physicochemical modeling based on overall bacterial cell surface properties, which in fact rules out adhesion. Interaction Forces between S. mutans and S. intermedius and Fibronectin Films. Figure 4 illustrates the influence of ionic strength on the force-distance curves between both strains of streptococci and fibronectin-coated AFM tips in high and low ionic strength buffer, whereas Table 4 summarizes the quantitative characteristics of the approach and retraction curves. S. mutans LT11 experiences a strong repulsion upon approach of the fibronectin-coated tip in high ionic strength PBS that extends

Langmuir, Vol. 24, No. 19, 2008 10971

over approximately 38 nm (median). In low ionic strength PBS, this repulsion decreases by a factor of 2 and also extends over a 2× smaller repulsive force range (significant at p < 0.01, signed rank test for the median). Differences in adhesion forces upon retraction are not statistically different in high and low ionic strength PBS. S. intermedius NCTC11324 exhibits much smaller differences in repulsion upon approach when comparing high and low ionic strength PBS than S. mutans LT11 with operating distances ranging between 35 and 29 nm depending upon ionic strength. Adhesion forces upon retraction, however, are weaker in low ionic strength PBS than in high ionic strength PBS, not only when judged from the median (significant at p < 0.05, signed rank test for the median) but especially when considering the range value. In conclusion, the strong repulsion experienced by S. mutans LT11 upon approach of the fibronectin film in high ionic strength PBS and in the absence of differences in adhesion forces upon retraction at both ionic strengths is in line with its reduced adhesion to a fibronectin film in high ionic strength PBS. In addition, whereas repulsion upon approach is similar at both ionic strengths for S. intermedius NCTC11324, its stronger adhesion experienced upon retraction from a fibronectin film in high ionic strength PBS is in agreement with its increased adhesion in high ionic strength PBS.

Discussion Adhesion of S. intermedius NCTC11324 to fibronectin films is influenced by overall repulsive electrostatic interactions and is highest in a high ionic strength environment, in line with physicochemical modeling on the basis of its overall cell surface properties (cell surface hydrophobicity and zeta potential). Localized, specific antigens, mediating fibronectin binding on S. mutans LT11 through molecular recognition, on the other hand govern adhesion of this strain to fibronectin films, and its adhesion is at odds with physicochemical modeling on the basis of overall surface properties. Since its adhesion is highest in a low ionic strength environment, where electrostatic interactions are strongest, it must be concluded that the electrostatic interaction between the S. mutans antigen I/II fibronectin binding proteins and the fibronectin film is between oppositely charged domains on the interacting surfaces. It is speculated that, unlike for S. mutans, physicochemical modeling of S. intermedius adhesion is valid, because it presumably has a much lower density of antigen I/II fibronectin binding proteins than does S. mutans.6 Evidently, the localized, positively charged domains on neither of the two cell surfaces are capable of creating an overall positive cell surface. Clearly, intervening influences of relatively sparsely distributed, localized positive charges in the sea of negative charges on the S. intermedius cell surface are small and do not invalidate a physicochemical approach on the basis of overall cell surface properties. The higher density of localized positive charges on the S. mutans cell surfaces, however, intervenes strongly with a physicochemical approach of its adhesion mechanism based on overall cell surface properties, and a more microscopic approach is needed in order to understand the physicochemistry of the interaction. AFM provides an ideal tool for such a microscopic approach, although the current status of understanding how the forces measured via AFM determine actual bacterial adhesion to macroscopic substrata is poor. It has been hypothesized that the repulsive forces measured upon approach of an AFM tip to a bacterial cell surface determine whether adhesion occurs upon first contact between an organism and a substratum surface, and

10972 Langmuir, Vol. 24, No. 19, 2008

Busscher et al.

Figure 3. Examples of the adhesion kinetics of S. mutans LT11 (left panel) and S. intermedius NCTC11324 (right panel) during 4 h of adhesion to a fibronectin film in a parallel plate flow chamber. The top series is pertinent to low ionic strength PBS, while the bottom series refers to high ionic strength PBS. Table 3. Initial Deposition Rates and Number of Adhering Streptococci after 4 h to a Fibronectin Film under Moderate Flow (Shear Rate 10 s-1) in High (PBS) and Low Ionic Strength PBS (PBS/10), Together with the Percentage of Weakly Bound Streptococci, i.e., Those That Could Not Withstand the Passage of a Liquid-Air Interface S. mutans LT11 parameter initial deposition rate (cm-2 s-1) number after 4 h (106 cm-2) percentage weakly bound (%)

S. intermedius NCTC11324

PBS

PBS/10

PBS

PBS/10

335 ( 82

1573 ( 177

1608 ( 143

230 ( 80

3.3 ( 1.0

22.7 ( 5.3

21.1 ( 2.3

3.2 ( 0.8

27 ( 4

0

0

65 ( 4

evidence for this hypothesis indeed exists.10 Subsequently, once a bacterium is adhering, it has to maintain its position on a surface which requires adequate adhesion forces especially under conditions of flow.11 This rationale led to a second hypothesis that the adhesion forces measured upon retraction of an AFM tip from a bacterial cell surface determine whether bacteria can remain adhering under flow, for which evidence exists as well.12 However, under many conditions, bacterial adhesion is the net result of a dynamic process of attachment and detachment, and neither of the above two hypotheses is supported by observations.13 Also, in view of these hypotheses, the two strains studied (10) Vadillo-Rodriguez, V.; Busscher, H. J.; Norde, W.; De Vries, J.; van der Mei, H. C. Langmuir 2003, 19, 2372–2377. (11) Rutter, P. R.; Vincent, B. In Microbial Adhesion to Surfaces; Berkeley, R. C. W., Lynch, J. M., Melling, J., Rutter, P. R., Vincent, B., Eds.; Ellis Horwood Ltd.: Chicester, England, 1980; pp 71-91. (12) Vadillo-Rodriguez, V.; Busscher, H. J.; Norde, W.; De Vries, J.; van der Mei, H. C. Microbiology 2004, 150, 1015–1022. (13) Nejadnik, M. R.; van der Mei, H. C.; Busscher, H. J.; Norde, W. Appl. EnViron. Microbiol. 2008, 74, 916–919.

here present extremes. S. mutans LT11 adhered to fibronectin films in higher numbers from low ionic strength PBS than from high ionic strength PBS, as it experienced much stronger repulsive forces upon approach of the fibronectin-coated AFM tip to the cell surface in high ionic strength PBS than in low ionic strength PBS. Alternatively, S. intermedius NCTC11324 adhered in higher numbers from high ionic strength PBS. This occurred in the absence of major differences in repulsive forces upon approach between experiments in high and low ionic strength PBS, which makes us conclude that this is due to the stronger adhesion forces measured upon retraction of the fibronectin-coated AFM tip from the cell surface in high ionic strength PBS as compared to in low ionic strength PBS. The adhesion forces measured in the current study for the specific interaction between fibronectin-coated AFM tips and antigen I/II on the streptococcal cell surfaces are much smaller than those measured in a similar way for the specific interaction between fibronectin-coated AFM tips and fibronectin binding proteins on Staphylococcus aureus cell surfaces.14 This suggests that the interaction with fibronectin of the two strains involves different chemical groups and the small adhesion forces for the streptococcal strains are probably a reflection of the lack of attractive acid-base interactions (see Table 2), as existing, for instance, in the interaction between Escherichia coli and silicon nitride.15 The nature of the repulsion generally observed upon approach between two biological surfaces has been attributed to electrosteric (14) Yongsunthon, R.; Fowler, V. G. J.; Lower, B. H.; Vellano, F. P.; Alexander, E.; Reller, L. B.; Corey, G. R.; Lower, S. K. Langmuir 2007, 23, 2289–2292. (15) Abu-Lail, N. I.; Camesano, T. A. Langmuir 2006, 22, 72967301. (16) Razatos, A.; Ong, Y. L.; Sharma, M. M.; Georgiou, G. Proc. Natl. Acad. Sci. U.S.A. 1998, 95, 11059–11064. (17) Ong, Y. L.; Razatos, A.; Georgiou, G.; Sharma, M. M. Langmuir 1999, 15, 2719–2725. (18) Atabek, A.; Camesano, T. A. J. Bacteriol. 2007, 189, 8503–8509.

Streptococcal Adhesion to Fibronectin Films

Langmuir, Vol. 24, No. 19, 2008 10973

Figure 4. Examples of force-distance curves between S. mutans LT11 (left panel) and S. intermedius NCTC11324 (right panel) and a fibronectincoated AFM tip. The top series is pertinent to low ionic strength PBS, while the bottom series refers to high ionic strength PBS. Table 4. Median, Mode, and Range of the Distributions Measured for the Repulsive Force at Contact (F0), Repulsive Force Range (Λ), and Adhesion Force (Fadh) for the Interaction between S. mutans LT11 and S. intermedius NCTC11324 with Fibronectin-Coated AFM Tips in High (PBS) and Low (PBS/10) Ionic Strength PBSa S. mutans LT11 parameter

S. intermedius NCTC11324

PBS

PBS/10

PBS

PBS/10

F0 (nN)

median mode range

10.8 10.9 24.2

5.5 2.3 18.2

8.2 7.2 15

7.8 9.6 10.7

Λ (nm)

median mode range

38 6 225

19 15 38

35 22 390

29 33 133

Fadh (nN)

median mode range

-0.2 -0.1 -4.7

-0.2 -0.2 -4.3

-0.1 0.0 -3.2

0.0 0.0 -0.9

a All experiments were done in five-fold with separately prepared fibronectin-coated AFM tips and different bacterial cultures, yielding 443 and 274 force-distance curves in high and low ionic strength PBS, respectively.

repulsion.16-18 Therefore, it can be expected that the range of these repulsive forces depends on ionic strength. S. intermedius NCTC11324, with its lower density of localized positive charges, shows a minor decrease in repulsive force range when the ionic strength is decreased, but a decrease in ionic strength for S. mutans LT11 causes a strong increase in electrostatic attraction

between negative and localized positive surface groups that causes collapse of the surface structures present on this strain,6 concurrent with the decrease of the repulsive force range measured by AFM. Apart from the suggestion that the S. intermedius NCTC11324 cell surface contains fewer antigen I/II receptors than the S. mutans LT11 cell surface, it has been suggested that the molar mass of the antigens at the surfaces of S. intermedius and S. mutans antigens is different, that is, amounting to 160 and 185 kDa, respectively. This mass difference has been confirmed by immunoblotting, while amino acid sequencing demonstrated about 250 fewer alanine- and proline-rich sequences in the S. intermedius sequence than in the S. mutans sequence.6 These alanine- and proline-rich sequences may, among others, constitute the positively charged, localized domains on the S. mutans cell surface that overrule the influence of overall cell surface properties on their adhesion to fibronectin films. In summary, this study shows that physicochemical modeling of bacterial adhesion to protein-coated surfaces is only valid provided the number of specific interaction sites on the cell surfaces is low, such as on S. intermedius NCTC11324. Nonspecific interactions are dominated by specific interactions if the number of specific interaction sites is large, such as on S. mutans LT11. Its ionic strength dependence indicates that the specific interaction is electrostatic in nature and operative between oppositely charged domains on the interacting surfaces, despite the generally overall negatively charged character of the surfaces. LA8016968