Interfacial re-arrangement in initial microbial adhesion to surfaces

This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and education use, including for instruction at the authors institution and sharing with colleagues. Other uses, including reproduction and distribution, or selling or licensing copies, or posting to personal, institutional or third party websites are prohibited. In most cases authors are permitted to post their version of the article (e.g. in Word or Tex form) to their personal website or institutional repository. Authors requiring further information regarding Elsevier’s archiving and manuscript policies are encouraged to visit: http://www.elsevier.com/copyright

Author's personal copy Current Opinion in Colloid & Interface Science 15 (2010) 510–517

Contents lists available at ScienceDirect

Current Opinion in Colloid & Interface Science j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / c o c i s

Interfacial re-arrangement in initial microbial adhesion to surfaces Henk J. Busscher a, Willem Norde a,b, Prashant K. Sharma a, Henny C. van der Mei a,⁎ a b

University of Groningen, University Medical Center Groningen, Department of Biomedical Engineering, Antonius Deusinglaan 1, 9713 AV Groningen, The Netherlands Laboratory of Physical Chemistry and Colloid Science, Wageningen University, Dreijenplein 6, 6703 HB Wageningen, The Netherlands

a r t i c l e

i n f o

Article history: Received 17 March 2010 Received in revised form 17 May 2010 Accepted 17 May 2010 Available online 24 May 2010

a b s t r a c t Upon initial microbial adhesion to a surface, multiple events occur that include interfacial re-arrangements in the region between an adhering organism and a surface. Application of physico-chemical mechanisms to explain microbial adhesion to surfaces requires better knowledge of the interfacial re-arrangement occurring immediately after adhesion than hitherto available. © 2010 Elsevier Ltd. All rights reserved.

Keywords: Microbial adhesion Bond strengthening Biofilm Atomic force microscopy Flow displacement systems Quartz crystal microbalance DLVO theory Surface thermodynamics

1. Introduction Biofilms form on virtually all surfaces exposed to natural and industrial environments and their formation commences with adhesion of microorganisms. Surfaces to which microorganisms adhere can be in a pristine state or covered with a conditioning film consisting of adsorbed macromolecular components [1,2]. The presence of a conditioning film greatly complicates the mechanism of microbial adhesion, as microscopic specific ligand–receptor bonds may be involved in microbial adhesion to e.g., protein-coated surfaces [3–5], whereas on a pristine surface adhesion is mainly governed by the macroscopic properties of the interacting surfaces [6•,7••,8]. Microbial adhesion, as the initial step in biofilm formation, excludes metabolic processes such as excretion of extrapolymeric substances [9] and growth, which justifies to treat initial microbial adhesion according to the mechanisms proposed to be valid for inert particle adhesion [10••]. Neglecting the structural complexity and chemical heterogeneity of microbial cell surfaces [11], indeed a search has been initiated many years ago into microbial zeta potentials [12•], contact angles [13•], cell surface hydrophobicities [14–16], surface free energies [17], and other physico-chemical properties of microbial cell surfaces with the aim of applying surface thermodynamics [17,18•] or

⁎ Corresponding author. Tel.: + 31 50 3633140; fax: +31 50 3633159. E-mail address: [email protected] (H.C. van der Mei). 1359-0294/$ – see front matter © 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.cocis.2010.05.014

DLVO theories [6,19••,20] to explain initial microbial adhesion to surfaces. More recently, various groups have been involved in the direct measurement of microbial interaction forces with substratum surfaces, either pristine or conditioning film coated, employing atomic force microscopy (AFM) [21,22] or optical tweezers [23,24]. Microbial adhesion to surfaces is initially reversible, but even in the absence of metabolic processes becomes essentially irreversible shortly after first contact. Experiments in flow displacement systems have indicated that desorption probabilities of both microorganisms as well as of inert polystyrene particles decrease by several orders of magnitude within 1 to 2 min after contact with a substratum surface [25,26]. AFM has confirmed that microbial adhesion forces indeed strengthen exponentially over time by progressively invoking acid– base interaction forces [27••]. Yet, detailed knowledge of the interfacial re-arrangements responsible for bond strengthening is lacking. Few studies have been conducted over the past decades to address these interfacial rearrangement, but physico-chemical understanding of microbial adhesion within and beyond the initial 1 or 2 min after first contact is essential in order to advance our understanding to a level relevant to practical applications. Quartz crystal microbalance with the ability to measure dissipation (QCM-D) has been applied by several groups [9,28] to assess the physico-chemical nature of the coupling of microorganisms to a surface and the changes occurring over time. In this review, we briefly summarize the mechanisms of microbial adhesion to surfaces as revealed by application of surface thermodynamics and DLVO theories, AFM and QCM-D. Rather than providing a

Author's personal copy H.J. Busscher et al. / Current Opinion in Colloid & Interface Science 15 (2010) 510–517

511

summary of what has been achieved by applying these methods, focus will be on remaining challenges that need to be solved in order to provide a better basis for understanding the mechanism of microbial adhesion to surfaces. 2. Microbial adhesion Strange as it may seem, the simple expression “microbial adhesion” is used in literature for widely different phenomena. Table 1 summarizes essentially different parameters that are generally used to characterize “microbial adhesion”. Yet, these parameters refer to different aspects of the adhesion process. Mass transport conditions are most important in microbial adhesion, as they not only determine the rate at which organisms arrive at a substratum surface [29,30], but they also dictate the shear off and lift forces during adhesion [31]. Under controlled mass transport conditions, experimental deposition rates can be compared with theoretical upper limits, such as resulting from the Smoluchowski–Levich approach [29]. The ratio between experimental deposition rates and theoretical upper limits is called the “deposition efficiency”, and expresses the fraction of organisms arriving at a surface that actually manages to adhere. Deposition rates depend on the microbial concentration in suspension and for that reason a dimensionless deposition rate, the so-called Sherwood number, has been introduced according to [32]

Sh =

jl D0 c

ð1Þ

where j is the deposition rate, l denotes a characteristic length, like e.g. the depth of the flow channel, D0 the microbial diffusion coefficient and c the microbial concentration in suspension. Flow displacement systems offer the best control over experimental mass transport conditions and when equipped with real-time, in situ analysis options also enable to directly measure desorption rate coefficients (β) in the various stages of an experiment as a function of the residence time of adhering particles or microorganisms (see example in Fig. 1). In Fig. 1 it can be observed that bacterial desorption rate coefficients decrease exponentially with increasing residence time (t − τ) of the adhering bacteria on the surface from an initial value, βo to a final value, β∞ and once adhering for longer than 20 to 25 s desorption becomes highly unlikely. The desorption kinetics can be described assuming one relaxation time constant, 1 / δ, according to [33••] −δðt−τÞ

βðt−τÞ = β∞ −ðβ∞ −βo Þe

:

ð2Þ

Fig. 1. Residence-time dependent desorption rate coefficients (β(t − τ)) for S. epidermidis HBH2 3 from hydrophilic glass (black dots) and hydrophobic, DDS-coated glass (white dots). Figure taken from Boks et al. [26].

The relaxation time constant is the characteristic time of bond maturation between an adhering bacterium and a substratum surface and essentially describes the rate at which the bond matures. As not only bacteria but also inert polystyrene particles show similar bond-maturation characteristics [34], it is emphasized that this type of bond maturation is purely physico-chemical in nature and has little or nothing to do with metabolic processes inherent to the use of microorganisms. Physico-chemically, bond maturation has been associated with the progressive removal of interfacial water, unfolding of surface structures or rotation of an entire particle to have its most favourable site opposing a substratum surface. Interestingly, increasing the residence time of a BSA-coated microsphere on a surface consistently increased the adhesion force measured during retraction of the sphere from the surface [35,36], demonstrating the important contribution of protein unfolding to bond maturation. In the absence of in situ observation methods of microbial adhesion, substrata with adhering organisms have to be taken out of a microbial suspension, and sometimes rinsed to remove so-called “loosely adhering” organisms prior to actual enumeration. These actions, and particularly traversing of a substratum with adhering micron-sized particles through a liquid–air interface, exert high detachment forces upon adhering particles [37•], causing the removal of an unknown number of adhering organisms from the substratum surface [38,39]. Moreover, under sufficiently vigorous rinsing all adhering organisms may appear as “loosely adhering”. Thus the outcome of experiments involving application of detachment forces higher than those prevailing under the conditions of attachment do

Table 1 Various parameters used in the literature referring to different aspects of what is generally called “microbial adhesion”. Parameter

Description

Units

Initial deposition rate

Number of organisms initially arriving and adhering at the surface per unit time and area under the experimental mass transport conditions Number of organisms arriving and adhering at the surface per unit time and area under the experimental mass transport conditions Ratio between experimental deposition rate and a theoretically calculated upper limit Number of organisms initially leaving the surface per unit time and area under the experimental mass transport conditions Number of organisms leaving the surface per unit time and area under the experimental mass transport Number of organisms adhering to the surface per unit area at time under the experimental mass transport conditions Number of organisms adhering to the surface per unit area after application of a detachment force at time t exceeding the one of the experimental mass transport conditions

cm− 2 s− 1

Deposition rate Deposition efficiency Initial desorption rate Final desorption rate Adhesion number Retention number

cm− 2 s− 1

cm− 2 s− 1 cm− 2 s− 1 cm− 2 cm− 2

Author's personal copy 512

H.J. Busscher et al. / Current Opinion in Colloid & Interface Science 15 (2010) 510–517

not reflect “adhesion” but rather “retention”, and it is strongly advocated that a distinction needs to be made between microbial adhesion and retention. In case data are referred to as microbial retention, it is imperative that the detachment forces exerted are estimated. Another interesting distinction between different modes of microbial adhesion is between mobile and immobile adhesion [40••,41]. Depending on the strain and substratum involved, it has been observed that microorganisms can reside in the close vicinity of a substratum surface, while still being able to slide over the surface. This is named “mobile” adhesion. Alternatively, it has also been observed that organisms adhere and immediately immobilize at the position of initial contact (“immobile” adhesion). The distinction between mobile and immobile adhesion is of utmost importance as it yields information about the distribution of localized high affinity sites over a substratum surface [42]. This distinction can only be made with real-time, in situ observation possibilities. 3. Surface free energy approach and (extended)-DLVO theories Surface thermodynamic analyses of microbial adhesion to surfaces are always based on measured contact angles with liquids on microbial lawns prepared on membrane filters as well as on the accompanying substratum surfaces. Measured contact angles are then converted into microbial and substratum surface tensions γ which, at constant pressure and temperature, equal the surface Gibbs free energies. Conversion of contact angle data in surface free energies can be done by various theoretical concepts [43,44]. Currently the concept of Lifshitz–Van der Waals/acid–base components [19••] appears to have become the most widely used. In this concept, surface free energies are separated in a Lifshitz–Van der Waals (γ LW) and an acid– base component (γ AB). The acid–base component can be further separated into an electron-donating (γ−) and accepting parameter (γ+), according to γ=γ

LW

+γ

AB

ð3Þ

where γ

AB

pffiffiffiffiffiffiffiffiffiffiffiffiffi = 2 γþ γ− :

ð4Þ

Contact angles are usually measured with two polar liquids (preferably water and formamide) and one or two apolar liquids (preferably methyleneiodide, and sometimes also α-bromonaphthalene). Using the known surface energy parameters of the liquids and the measured contact angles, three or four (depending on the number of liquids used) equations of the form Eq. (5) can be derived for the different liquids [45••] qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiffiffiffiffi γl ð1 + cosθl Þ þ − γ− γmLW γlLW + γþ m γl + m γl = 2

ð5Þ

where the subscript m denotes the microbial cell and l the liquid used for measuring the contact angle, θl. By solving these equations, the microbial cell surface free energy components and parameters can be calculated. Similar equations have to be solved for substratum surface, s, in order to determine its surface free energy. As a next step, assuming adhesion takes place in an aqueous environment (w), the interfacial free energy of the ms interface, on the one hand, is compared with the interfacial free energies of the mw and the sw interfaces on the other, to yield the interfacial free energy of adhesion ΔG and its Lifshitz–Van der Waals, ΔG LW, and acid–base, ΔG AB, components LW

AB

ΔGmws = ΔGmws + ΔGmws = γms −γmw −γsw

ð6Þ

qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi LW LW LW γms = γm + γs −2 γmLW γsLW

ð7Þ

qffiffiffiffiffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiffiffiffiffi AB AB AB − − þ γms = γm + γs −2 γþ m γs −2 γm γs :

ð8Þ

Note that application of Eq. (6) requires the development of expressions similar to Eqs. (7) and (8) for the mw- and sw-interface. Application of surface thermodynamics a priori requires that microbial adhesion is in equilibrium, with equal “on” and “off” rates. Whether or not the reversibility of microbial adhesion as hitherto measured [46–49••], actually allows application of surface thermodynamics has remained an open question that most researchers prefer to ignore. Moreover, it can be doubted whether the sm interface, as intended in Eqs. (7) and (8), has really formed and whether such an interface, if formed, remains the same over time. In this respect it is likely, that the microbial cell surface during contact angle measurements is in a different state than when the organism is adhering. Numerous papers have tried to establish relationships between microbial adhesion numbers and interfacial free energies of adhesion. Although sometimes relationships have been established for small collections of different strains [17,50], generally microbial adhesion was found not to relate with the interfacial free energies of adhesion and to occur despite thermodynamically unfavourable (ΔG N 0) conditions. In this respect it is interesting to note that the general observation has been that the long-range Lifshitz–Van der Waals interactions are nearly always favourable (ΔGLW b 0) and these must thus be responsible for adhesion to occur [18•]. Acid–base interactions require closer approach due to their short-range character [19•] and eventually dominate the interaction as they can become much stronger than the Lifshitz–Van der Waals ones (see Fig. 2) until Born repulsion sets in. Once attractive acid–base interactions have become operative, they may dictate more complicated features of the adhesion process, such as reversibility [26] or local immobilization [40••] of adhering microorganisms. Unlike the thermodynamic approach, the extended DLVO (X-DLVO) theory predicts the free energy of adhesion ΔG as a function of separation distance H between a microbial cell and substratum surface. In the classical DLVO theory, ΔG(H) is made up of contributions from Lifshitz–Van der Waals interactions ΔG LW(H) and electrical double layer interactions ΔG EL(H), whereas in the X-DLVO theory the contribution of Lewis acid–base interaction ΔG AB(H) is added [51]. Accordingly, ΔGðH Þ = ΔG

LW

EL

ðH Þ + ΔG ðH Þ + ΔG

AB

ðH Þ:

ð9Þ

The Lifshitz–Van der Waals interaction is nearly always attractive (ΔG LW(H) b 0) and can be calculated using Eq. (10) ΔG

LW

ðH Þ = −

   A mws 2aðH + aÞ H + 2a − ln 12 HðH + 2aÞ H

ð10Þ

where Amws is the Hamaker constant for interaction between m and s across w, and a is the radius of the microbial cell. Hamaker constants Amws can either be taken from the literature [52–55], or from ΔGLW mws occurring in Eq. (6) according to 2

LW

Amws = −12πdo ΔGmws

ð11Þ

where do denotes the minimum separation distance (1.57 Å). The electrical double layer interaction (ΔG EL) depends on the surface potentials of the microbial cells (ψ1) and substratum surface (ψ2) and can be quantified by Eq. (12) " # −κH   2ψ ψ   1+e EL 2 2 −2κH 1 2 ln + ln 1−e ΔG ðH Þ = πεa ψ1 + ψ2 1−e−κH ψ21 + ψ22 ð12Þ

Author's personal copy H.J. Busscher et al. / Current Opinion in Colloid & Interface Science 15 (2010) 510–517

513

Fig. 2. Gibbs free energy of interaction as a function of distance between microorganisms and substratum surfaces, according to the X-DLVO theory. Total (solid line), Lifshitz–Van der Waals (□), acid–base (○) and electrostatic (Δ) interaction energies as a function of distance. (A) Repulsive electrostatic and acid–base interactions, only allowing microbial cells to come as close as 240 Å. (B) Repulsive acid–base interactions and reduced electrostatic repulsion due to high ionic strength conditions, allowing microbial cells to come closer (approximately 70 Å). (C) Repulsive electrostatic and attractive acid–base interactions, allowing microorganisms to approach the surface up to the secondary minimum, that is separated from the deep primary minimum by an energy barrier (about 45 kT in the present example). (D) Attractive acid–base interactions complete overrule electrostatic repulsion, reduced due to high ionic strength conditions, yielding attractive conditions over the entire distance range.

where ε is the permittivity of the aqueous phase and κ− 1 is the double layer thickness. In practice, the surface potentials are often replaced by the zeta potentials (ζ). This is a reasonable approach in case of hard particles, but for soft particles like microbial cell surfaces, the surface coat may be penetrable to ions, and zeta potentials may yield an overestimation of the surface potential [56, 57••], if a hard-sphere model like the Smoluchowski one is used. Since most naturally occurring surfaces including microbial cell surfaces are negatively charged under physiological conditions [12•], electrical double layer interactions are usually repulsive. Increasing the ionic strength of the aqueous phase reduces the thickness of the electrical double layer (1 / κ), thereby decreasing the magnitude of electrical double layer interactions at a given particle substratum separation distance. The acid–base interactions, quantified by Eq. (13), could be both repulsive or attractive depending on the hydrophobicity of the two surfaces: AB

AB

½ðdo −H Þ = λ

ΔG ðHÞ = 2πaλΔG adh e

ð13Þ

where λ is the correlation length of molecules in water (6 Å) and ΔGAB adh the acid–base component of the Gibbs free energy of adhesion as occurring in Eq. (6) [51]. Examples of X-DLVO interaction energies as a function of separation distance are presented in Fig. 2A–D. Based on available literature, it must be concluded however, that the XDLVO theory at best describes some aspects of microbial adhesion qualitatively [50], since firm, generally valid relationships between XDLVO parameters and adhesion numbers are lacking [18•]. The reasons for searching relationships between surface thermodynamics or DLVO-parameters, like the depth of the interaction minimum, and microbial adhesion numbers have never become quite clear. It can even be argued that the depth of an interaction minimum may have nothing to do with the number of adhering organisms. Once a microorganism is captured in an interaction minimum, desorption is only likely if its depth is less than several kT's, unless adhering microorganisms are challenged to detach from their position in the interaction minimum by prevailing detachment forces. Xia et al. [58•] proposed an equation describing a relation between the depth of the interaction minimum and microbial desorption. Since interaction minima up to 20 kT deep have been described, calculation of

desorption rates based on this equation confirmed the generally low desorption of microorganisms from substratum surfaces [26, 49••] and suggested equal numbers of microorganisms captured on different substratum surfaces. This raises the question of the actual significance of surface thermodynamics or DLVO theories for microbial adhesion. Most importantly, both theories confirm that microorganisms are able to adhere to almost any substratum surface and do so through ubiquitously present, attractive Lifshitz–Van der Waals forces. Acid– base interactions determine much more the conditions prevailing at the interface between an adhering microorganism and a substratum surface such as the degree and extent of water ordering near a surface [59] and when attractive they will allow a more rapid close approach through the viscous layer of interfacial water in between the interacting surfaces than when repulsive. Therewith, attractive acid–base interactions become pivotal for specific ligand–receptor and other localized interactions to occur [7••, 8] and determine the rate at which bond maturation and the transition from reversible to irreversible adhesion takes place. The absence of attractive acid–base interactions, however, is not necessarily the only condition that impedes the development of close approach ligand–receptor interactions, as steric-hindrance by polymer-brush like structures on the microbial cell surface may do so as well [11].

4. Microbial interaction forces using AFM Microbial interaction forces can be measured with AFM using different methods. The simplest one is the measurement of the interaction force between the native AFM tip and organisms immobilized on a surface [60]. Single molecule ligand–receptor interaction forces have been measured by attaching ligands to the AFM tip [21,61]. To mimic the situation in practical systems more closely, interaction forces between microbial functionalized AFM probes and a substratum surface are measured. Interactions between microorganisms may be determined also using a microbial functionalized AFM probe and a surface covered by microorganisms [62–66]. In either case, regular checks of the AFM tip need to be carried out on a clean surface in order to verify that the integrity of the tip or microbial probe has not occurred (contamination of the tip by detached cell

Author's personal copy 514

H.J. Busscher et al. / Current Opinion in Colloid & Interface Science 15 (2010) 510–517

surface components, detachment of microorganisms from a microbial probe, etc.), which could greatly influence these measurements. For a number of strain collections, a direct relationship between AFM interaction forces and microbial cell surface properties [35,67] and/or microbial adhesion to surfaces [64] has been found. For instance, probing bacterial strains that possess specific antigens on their surface by an AFM cantilever coated with the corresponding antigen-binding protein showed stronger interaction forces than strains lacking those specific antigens. This finding is corroborated by calorimetric data revealing that interaction of the corresponding protein with the antigen-containing bacterial cell surface is enthalpically far more favourable than its interaction with antigen-deficient mutants [68–70]. These observations are not generic however [64]. There are a number of reasons for this that require further investigation. Firstly, during AFM the contact is forced upon the system, while during adhesion experiments approach of an organism toward a substratum surface is mediated by convective-diffusion mostly, after which the interaction forces take over and accelerate approach toward the surface. In all cases, this yields a less forced and penetrative contact than during approach in AFM measurements. In AFM, approach is generally continued until a contact force of around 8 nN is invoked, which is much larger than calculated DLVO-interaction forces [27••]. There are a few experiments reported in the literature in which the maximum force causing reversal of a microbial AFM tip from approach to retract is varied [71]. Increasing even the reversal force for an inert microsphere attached to an AFM cantilever toward a simple BSAcoated surface from a low loading force of 0.6 nN to 5.4 nN is accompanied by an increase in adhesion force from less than −0.5 nN to −7.5 nN [69]. This observation clearly indicates that different cell surface groups may be involved in microbial adhesion under flow than during AFM measurements, explaining the lack of relationships between adhesion to substratum surfaces and AFM interaction force measurements. Bond maturation may well be another reason for the lack of generic relations between AFM interaction forces and actual adhesion. Residence-time dependent desorption under flow demonstrates that microbial adhesion essentially becomes irreversible within seconds to minutes after arrival of an individual organism on a substratum surface (see Fig. 1). Therewith, adhesion numbers, established after a finite contact time, mostly represent organisms that have already undergone extensive bond maturation, while AFM interaction forces are most frequently measured with 0 s surface delay, i.e. the time during which contact is allowed before retracting the AFM probe. Measurements of AFM force distance curves allowing a surface delay of up to 120 s have clearly indicated that the interaction forces become stronger over time, as can be seen in the example given in Fig. 3. AFM interaction force strengthening can be described by an exponential function similar to Eq. (2)   − t= F ðt Þ = F0 + ðF∞ −F0 Þ 1−e τ

ð14Þ

where F0 is the adhesion force at 0 s and F∞ is the adhesion force after bond maturation with a characteristic time constant of τ. Time constants derived from AFM interaction force strengthening according to Eq. (14), do not always correspond, however, with time constants according to which desorption rate coefficients decrease [27••,63,70]. Detailed analysis of the retract force distance curves enables to define the nature of the interaction forces. Poisson analysis of minor peaks appearing in the downward flanks of retract curves has been employed to identify the acid–base versus non-specific Lifshitz–Van der Waals interaction forces in microbial adhesion [73••]. In addition, acid–base interactions have seldom been revealed in retract force

Fig. 3. The adhesion force upon retract of an AFM cantilever with attached S. sanguinis ATCC10556 from a salivary conditioning film (closed symbols) and a BSA-coating (open symbols) as a function of the surface delay time [72].

distance curves measured with a 0 s surface delay, and only appear when the bond has been allowed to mature (see also Fig. 3). Recently a new AFM is introduced, the PeakForce QNM (VEECO Instruments Inc.), which makes it possible to investigate the role of mechanical properties in biological structures and processes. Binding and recognition events and therewith re-arrangements at the interface can potentially be followed by fluorescent-labelling, since this AFM is build on a fluorescent microscope. 5. Use of the quartz crystal microbalance in microbial adhesion One of the challenges in our current state of understanding the physico-chemical mechanisms of microbial adhesion to surfaces is to find out what happens in the interfacial region between an adhering organism and a substratum surface during bond maturation after initial contact. Experiments in flow displacement systems as well as AFM interaction force measurements have clearly demonstrated that even in the absence of metabolic processes, events happen in this region that facilitate bond maturation. QCM-D was essentially developed to determine rigid and homogeneous adsorption of molecules with an accuracy of ng/cm2. The adsorbed mass Δm is derived from a decrease in resonance frequency Δf, according to Sauerbrey's equation [56] Δm = −

CQCM Δf n

ð15Þ

where CQCM is an instrumental sensitivity constant (17.7 ng/cm2) and n is the overtone number of the oscillating frequency (1, 3, 5, 7, 9, 11 or 13). When equipped with the possibility to measure energy dissipation, QCM-D can be used to study the visco-elastic properties of soft polymer films and complex biomolecular systems at the solution– surface interface by fitting a Voigt model to the frequency and dissipation shifts [74,75]. QCM-D is able to do so because the evanescent wave caused by the lateral oscillation of the sensor surface, penetrates the adjacent fluid up to 250 nm depth [76,77] according to Penetration Depth ðmÞ =

rffiffiffiffiffiffiffiffi η πf ρ

ð16Þ

where η is the viscosity, ρ is the specific gravity of the adjacent fluid and f is the sensor oscillation frequency. Oscillations at higher overtones make it possible to gather the signal from a region even closer to the sensor surface.

Author's personal copy H.J. Busscher et al. / Current Opinion in Colloid & Interface Science 15 (2010) 510–517

Common use of QCM toward microbial adhesion is aimed at relating the shift in resonance frequency with the number of organisms adhering to the sensor surface [28,78,81]. Fibronectincoated gold electrodes for instance, have been exposed to suspensions with different concentrations of Staphylococcus epidermidis, yielding a linear relationship between changes in resonant frequency and the logarithm of the bacterial concentration [82,83]. Unfortunately, like many QCM studies, this study does not include simultaneous observation of microbial adhesion to the sensor surface and measurement of frequency shifts in the QCM chamber, which made validation difficult. Moreover, attachment of a rigid mass to the crystal surface should, according to Eq. (15), be accompanied by proportional, negative frequency shifts. However, several experiments yielded non-linear relationships between the number of adhering bacteria and the QCM frequency shift [79,84] and even positive frequency shifts have been observed [85], making the frequency shift an unreliable indicator for the number of adhering microbial cells. This indicates that a non-rigid coupling exists between the adhering microorganisms and the sensor surface, which seems to depend on the microbial surface morphology [28,85], as illustrated in Fig. 4 and confirming that no real, stable ms interface is formed during adhesion, as required for the application of surface thermodynamics. QCM-D has the ability to measure the dissipative nature of nonrigid masses adhering to the crystal surface in the form of oscillation amplitude decay. Full interpretation of the frequency change and dissipation signal either by Sauerbrey's equation [56] or the Voigt model [74,75] is impossible because microbial adhesion causes heterogeneous distribution of large masses connected to the sensor surface via a point contact, while the equations assume a homogeneous distribution of the adhering mass. Hence other analyses need to be developed to interpret the QCM signal for microbial adhesion. Comparison of the QCM-D response upon adhesion of a fimbriated and non-fimbriated Escherichia coli strain to sensor surfaces showed that the frequency and dissipation shifts per attached bacterium increased with increasing ionic strengths (particularly on hydrophobic surfaces) for non-fimbriated bacteria, whereas the adhesion of a fimbriated strain caused only low-level frequency and dissipation shifts irrespective of ionic strength up to 500 mM [28]. It was concluded that non-fimbriated bacteria may get a better, more rigid contact with increasing ionic strengths due to an increased area of contact between the organism and the surface, whereas fimbriated bacteria seem to have a flexible, non-rigid contact with the surface at

515

all ionic strengths included. This was confirmed by the use of a series of Streptococcus salivarius mutants, possessing various surface appendages of known lengths, demonstrating that only adhesion of a “bald” bacterium, completely devoid of surface appendages, was registered as a non-rigid coupling with a frequency decrease (Fig. 4) [85]. Adhesion of bacteria possessing surface appendages either yield a much smaller decrease or increase in frequency, despite the fact they adhered in higher numbers. Furthermore, the magnitude of frequency and dissipation shifts was found to be influenced by the distance at which the cell body was held from the sensor surface, i.e. by the lengths of the different surface appendages. Analogously, also Oantigen mediated adhesion of E. coli to concanavalin A on sensor surfaces indicated a rigid coupling with little dissipation [80]. A novel analysis of the dissipation signal has been recently proposed to determine the time dependent bacterium–substratum bond maturation [86]. Independent of the surface morphology of a series of S. salivarius mutants it was demonstrated that the bacterium–substratum interface changes within the first 55 s after arrival on the surface, whereas a rigid and abiotic silica particle did not show this effect. Note, that also this time period roughly coincides with the characteristic time constants for resident-time dependent desorption under flow and bond maturation observed in AFM. QCM-D is a very effective tool to study the microorganism– substratum interface in detail, in particular changes in the interaction with time. However, the QCM-D signal needs to be normalized against the number of adhering microorganisms in real-time to avoid misinterpretations due to detachment and presence of aggregates on the sensor surface. It is unlikely that QCM-D will develop into a new technique to monitor the number of microbial cells adhering on a surface or to identify the species of microbial cells in a sensor type application, for use e.g. in a water pipeline. This is due to its high sensitivity for any adhering mass, i.e. any biopolymers adhering outside the microorganism–substratum interface could influence the signal leading to its misinterpretation. Nevertheless, one of the challenges in advancing our understanding of microbial adhesion phenomena that could be achieved by QCM-D is to determine the exact details of the interfacial re-arrangements during bond maturation. 6. Conclusions This review makes clear how little we understand of microbial adhesion mechanisms to surfaces. Application of surface thermodynamics and DLVO theories may appear tempting and sometimes explain experimental observations, but it is not a priori obvious that the conditions required for application these approaches are always met. The dynamics occurring in the interfacial region between an adhering microorganism and a substratum surface causes bond strengthening within seconds to minutes after initial contact, therewith strongly decreasing the reversibility of adhesion. Once interfacial re-arrangements have increased the depth of the interaction minimum to less than several kT's, desorption becomes highly unlikely. Direct evidence of the interfacial re-arrangements occurring in the first minutes after initial contact between microorganisms and substratum surfaces can be obtained by QCM-D and will allow better definition of the interface. It can be anticipated that a better definition of the ms interface will enable more successful application of surface thermodynamics and DLVO theories at the sub-micron level than hitherto achieved. References and recomended readings

Fig. 4. Frequency shift, Δf, as a function of the number of adhering streptococci, determined using QCM-D equipped with image analysis options for direct enumeration of the numbers of bacteria adhering to the crystal surface for two strains of S. salivarius HBC12 (■) and HB7 (○). Data taken from [85].

[1] Palmer Jr RJ, Stoodley P. Biofilms 2007: broadened horizons and new emphases. J Bacteriol 2007;189:7948–60. • ••

of special interest. of outstanding interest.

Author's personal copy 516

H.J. Busscher et al. / Current Opinion in Colloid & Interface Science 15 (2010) 510–517

[2] Schneider RP, Marshall KC. Retention of the Gram negative marine bacterium SW8 on surfaces — effects of microbial physiology, substratum nature and conditioning films. Colloids Surf B Biointerfaces 1994;2:387–96. [3] Greene C, McDevitt D, Francois P, Vaudaux PE, Lew DP, Foster TJ. Adhesion properties of mutants of Staphylococcus aureus defective in fibronectin-binding proteins and studies on the expression of fnb genes. Mol Microbiol 1995;17: 1143–52. [4] Vaudaux P, Suzuki R, Waldvogel FA, Morgenthaler JJ, Nydegger UE. Foreign body infection: role of fibronectin as a ligand for the adherence of Staphylococcus aureus. J Infect Dis 1984;150:546–53. [5] Peacock SJ, Foster TJ, Cameron BJ, Berendt AR. Bacterial fibronectin-binding proteins and endothelial cell surface fibronectin mediate adherence of Staphylococcus aureus to resting human endothelial cells. Microbiology 1999;145:3477–86. [6] Hermansson M. The DLVO theory in microbial adhesion. Colloids Surf B • Biointerfaces 1999;14:105–19. [7] Busscher HJ, Cowan MM, Van der Mei HC. On the relative importance of specific •• and non-specific approaches to oral microbial adhesion. FEMS Microbiol Rev 1992;8:199–209. [8] Busscher HJ, Weerkamp AH. Specific and nonspecific interactions in bacterial adhesion to solid substrata. FEMS Microbiol Rev 1987;46:165–73. [9] Long G, Zhu P, Shen Y, Tong M. Influence of extracellular polymeric substances (EPS) on deposition kinetics of bacteria. Environ Sci Technol 2009;43:2308–14. [10] Mozes N, Marchal F, Hermesse MP, Van Haecht JL, Reuliaux L, Leonard AJ, et al. •• Immobilization of microorganisms by adhesion: interplay of electrostatic and non electrostatic interactions. Biotechnol Bioeng 1987;30:439–50. [11] Dorobantu LS, Bhattacharjee S, Foght JM, Gray MR. Analysis of force interactions between AFM tips and hydrophobic bacteria using DLVO theory. Langmuir 2009;25:6968–76. [12] Jucker BA, Harms H, Zehnder AJ. Adhesion of the positively charged bacterium • Stenotrophomonas (Xanthomonas) maltophilia 70401 to glass and Teflon. J Bacteriol 1996;178:5472–9. [13] Van der Mei HC, Bos R, Busscher HJ. A reference guide to microbial cell surface • hydrophobicity based on contact angles. Colloids Surf B Biointerfaces 1998;11:213–21. [14] Rosenberg M, Bar-Ness Greenstein R, Barki M, Goldberg S. Hydrophobic interactions as a basis for interfering with microbial adhesion. Adv Exp Med Biol 1996;408:241–8. [15] Lichtenberg D, Rosenberg M, Sharfman N, Ofek I. A kinetic approach to bacterial adherence to hydrocarbon. J Microbiol Meth 1985;4:141–6. [16] Van Loosdrecht MC, Lyklema J, Norde W, Schraa G, Zehnder AJ. The role of bacterial cell wall hydrophobicity in adhesion. Appl Environ Microbiol 1987;53:1893–7. [17] Absolom DR, Lamberti FV, Policova Z, Zingg W, Van Oss CJ, Neumann AW. Surface thermodynamics of bacterial adhesion. Appl Environ Microbiol 1983;46:90–7. [18] Bos R, Van der Mei HC, Busscher HJ. Physico-chemistry of initial microbial adhesive • interactions — its mechanisms and methods for study. FEMS Microbiol Rev 1999;23:179–230. [19] Van Oss CJ. Hydrophobicity of biosurfaces — origin, quantitative-determination •• and interaction energies. Colloids Surf B Biointerfaces 1995;5:91–110. [20] Rutter PR, Vincent B. Attachment mechanisms in the surface growth of microorganisms. In: Bazin MJ, Prosser JI, editors. Physiological models in microbiology, Volume IIFlorida, USA: CRC Press; 1988. p. 87–107. [21] Gaboriaud F, Dufrene YF. Atomic force microscopy of microbial cells: application to nanomechanical properties, surface forces and molecular recognition forces. Colloids Surf B Biointerfaces 2007;54:10–9. [22] Liu Y, Strauss J, Camesano TA. Adhesion forces between Staphylococcus epidermidis and surfaces bearing self-assembled monolayers in the presence of model proteins. Biomaterials 2008;29:4374–82. [23] Bjornham O, Fallman E, Axner O, Ohisson J, Nilsson UJ, Boren T, et al. Measurements of the binding force between the Helicobacter pylori adhesin BabA and the Lewis b blood group antigen using optical tweezers. J Biomed Opt 2005;10:44024. [24] Salomo M, Keyser UF, Struhalla M, Kremer F. Optical tweezers to study single Protein A/Immunoglobulin G interactions at varying conditions. Eur Biophys J Biophys Lett 2008;37:927–34. [25] Meinders JM, Busscher HJ. Adsorption and desorption of colloidal particles on glass in a parallel-plate flow chamber — influence of ionic-strength and shear rate. Coll Pol Sci 1994;272:478–86. [26] Boks NP, Kaper HJ, Norde W, Busscher HJ, Van der Mei HC. Residence time dependent desorption of Staphylococcus epidermidis from hydrophobic and hydrophilic substrata. Colloids Surf B Biointerfaces 2008;67:276–8. [27] Boks NP, Busscher HJ, Van der Mei HC, Norde W. Bond-strengthening in •• Staphylococcal adhesion to hydrophilic and hydrophobic surfaces using atomic force microscopy. Langmuir 2008;24:12990–4. [28] Otto K, Elwing H, Hermansson M. Effect of ionic strength on initial interactions of Escherichia coli with surfaces, studied on-line by a novel quartz crystal microbalance technique. J Bacteriol 1999;181:5210–8. [29] Elimelech M. Particle deposition on ideal collectors from dilute flowing suspensions — mathematical formulation, numerical-solution, and simulations. Sep Technol 1994;4:186–212. [30] Adamczyk Z, Van de Ven TGM. Deposition of particles under external forces in laminar flow through parallel-plate and cylindrical channels. J Colloids Interface Sci 1981;80:340–56. [31] Busscher HJ, Van der Mei HC. Microbial adhesion in flow displacement systems. Clin Microbiol Rev 2006;19:127–41. [32] Elimelech M, Gregory J, Jia X, Williams RA. Particle deposition and aggregation — measurement, modelling and simulation. Woburn, MA: Butterworth-Heinemann; 1995.

[33] Dabros T, Van de Ven TGM. Kinetics of coating by colloidal particles. J Colloids •• Interface Sci 1982;89:232–44. [34] Meinders JM, Busscher HJ. Influence of ionic-strength and shear rate on the desorption of polystyrene particles from a glass collector as studied in a parallelplate flow chamber. Coll Surf A-Physicochemical Eng Aspects 1993;80:279–85. [35] Xu LC, Vadillo-Rodriguez V, Logan BE. Residence time, loading force, pH, and ionic strength affect adhesion forces between colloids and biopolymer-coated surfaces. Langmuir 2005;21:7491–500. [36] Xu LC, Logan BE. Interaction forces measured using AFM between colloids and surfaces coated with both dextran and protein. Langmuir 2006;22:4720–7. [37] Leenaars AFM. A new approach to the removal of sub-micron particles from solid • (silicon) substrates. In: Mittal KL, editor. Particles on surfaces 1: detection, adhesion and removal. New York, USA: Plenum Press; 1988. p. 361–72. [38] Gomez-Suarez C, Busscher HJ, Van der Mei HC. Analysis of bacterial detachment from substratum surfaces by the passage of air–liquid interfaces. Appl Environ Microbiol 2001;67:2531–7. [39] Sharma PK, Gibcus MJ, Van der Mei HC, Busscher HJ. Influence of fluid shear and microbubbles on bacterial detachment from a surface. Appl Environ Microbiol 2005;71:3668–73. [40] Boks NP, Kaper HJ, Norde W, Van der Mei HC, Busscher HJ. Mobile and immobile •• adhesion of staphylococcal strains to hydrophilic and hydrophobic surfaces. J Colloids Interface Sci 2009;331:60–4. [41] Vigeant MAS, Ford RM, Wagner M, Tamm LK. Reversible and irreversible adhesion of motile Escherichia coli cells analyzed by total internal reflection aqueous fluorescence microscopy. Appl Environ Microbiol 2002;68:2794–801. [42] Anderson BN, Ding AM, Nilsson LM, Kusuma K, Tchesnokova V, Vogel V, et al. Weak rolling adhesion enhances bacterial surface colonization. J Bacteriol 2007;189:1794–802. [43] Sharma PK, Rao KH. Analysis of different approaches for evaluation of surface energy of microbial cells by contact angle goniometry. Adv Colloids Interface Sci 2002;98:341–463. [44] Bellon-Fontaine MN, Mozes N, Van der Mei HC, Sjollema J, Cerf O, Rouxhet PG, et al. A comparison of thermodynamic approaches to predict the adhesion of dairy microorganisms to solid substrata. Cell Biophys 1990;17:93–106. [45] Van Oss CJ. Development and applications of the interfacial tension between water •• and organic or biological surfaces. Colloids Surf B Biointerfaces 2007;54:2–9. [46] Busscher HJ, Uyen MHMJ, Weerkamp AH, Postma WJ, Arends J. Reversibility of adhesion of oral Streptococci to solids. FEMS Microbiol Lett 1986;35:303–6. [47] Rijnaarts HHM, Norde W, Lyklema J, Zehnder AJB. The isoelectric point of bacteria as an indicator for the presence of cell-surface polymers that inhibit adhesion. Colloids Surf B Biointerfaces 1995;4:191–7. [48] Xu CP, Boks NP, De Vries J, Kaper HJ, Norde W, Busscher HJ, et al. Staphylococcus aureus-fibronectin interactions with and without fibronectin-binding proteins and their role in adhesion and desorption. Appl Environ Microbiol 2008;74:7522–8. [49] Meinders JM, Van der Mei HC, Busscher HJ. Deposition efficiency and reversibility •• of bacterial adhesion under flow. J Colloids Interface Sci 1995;176:329–41. [50] Sharma PK, Rao KH. Adhesion of Paenibacillus polymyxa on chalcopyrite and pyrite: surface thermodynamics and extended DLVO theory. Colloids Surf B Biointerfaces 2003;29:21–38. [51] Van Oss CJ, Good RJ, Chaudhury MK. The role of Van der Waals forces and hydrogen-bonds in hydrophobic interactions between bio-polymers and lowenergy surfaces. J Colloids Interface Sci 1986;111:378–90. [52] Nir S, Andersen M. Van der Waals interactions between cell-surfaces. J Membr Biol 1977;31:1–18. [53] Nir S. Van der Waals interactions between surfaces of biological interest. Prog Surf Sci 1977;8:1–58. [54] Norde W. Colloids and interfaces in life sciences. New York: Marcel Dekker; 2003. [55] Lyklema J. Appendix 9. Fundamentals of interface and colloid science, Volume 1: fundamentals. Academic Press; 1991. [56] Duval JFL, Busscher HJ, Van de Belt-Gritter B, Van der Mei HC, Norde W. Analysis of the interfacial properties of fibrillated and nonfibrillated oral streptococcal strains from electrophoretic mobility and titration measurements: evidence for the shortcomings of the ‘classical soft-particle approach’. Langmuir 2005;21: 11268–82. [57] Ohshima H. Electrophoresis of soft particles. Adv Colloids Interface Sci 1995;62: •• 189–235. [58] Xia Z, Woo L, Van de Ven TGM. Microrheological aspects of adhesion of Escherichia • coli on glass. Biorheology 1989;26:359–75. [59] Van Oss CJ, Giese RF, Docoslis A. Water, treated as the continuous liquid in and around cells. Cell Mol Biol 2001;47:721–33. [60] Mendez-Vilas A, Gallardo-Moreno AM, Calzado-Montero R, Gonzalez-Martin ML. AFM probing in aqueous environment of Staphylococcus epidermidis cells naturally immobilised on glass: physico-chemistry behind the successful immobilisation. Colloids Surf B Biointerfaces 2008;63:101–9. [61] Hinterdorfer P, Dufrene YF. Detection and localization of single molecular recognition events using atomic force microscopy. Nat Meth 2006;3:347–55. [62] Vadillo-Rodriguez V, Busscher HJ, Norde W, De Vries J, Dijkstra RJ, Stokroos I, et al. Comparison of atomic force microscopy interaction forces between bacteria and silicon nitride substrata for three commonly used immobilization methods. Appl Environ Microbiol 2004;70:5441–6. [63] Vadillo-Rodriguez V, Busscher HJ, Norde W, De Vries J, Van der Mei HC. Atomic force microscopic corroboration of bond aging for adhesion of Streptococcus thermophilus to solid substrata. J Colloids Interface Sci 2004;278:251–4. [64] Vadillo-Rodriguez V, Busscher HJ, Van der Mei HC, De Vries J, Norde W. Role of lactobacillus cell surface hydrophobicity as probed by AFM in adhesion to surfaces at low and high ionic strength. Colloids Surf B Biointerfaces 2005;41:33–41.

Author's personal copy H.J. Busscher et al. / Current Opinion in Colloid & Interface Science 15 (2010) 510–517 [65] Postollec F, Norde W, De Vries J, Busscher HJ, Van der Mei HC. Interactive forces between co-aggregating and non-co-aggregating oral bacterial pairs. J Dent Res 2006;85:231–4. [66] Waar K, Van der Mei HC, Harmsen HJ, De Vries J, Atema-Smit J, Degener JE, et al. Atomic force microscopy study on specificity and non-specificity of interaction forces between Enterococcus faecalis cells with and without aggregation substance. Microbiology 2005;151:2459–64. [67] Vadillo-Rodriguez V, Busscher HJ, Norde W, De Vries J, Van der Mei HC. Relations between macroscopic and microscopic adhesion of Streptococcus mitis strains to surfaces. Microbiology 2004;150:1015–22. [68] Busscher HJ, Van de Belt-Gritter B, Dijkstra RJB, Norde W, Petersen FC, Scheie AA, et al. •• Intermolecular forces and enthalpies in the adhesion of Streptococcus mutans and an antigen I/II-deficient mutant to laminin films. J Bacteriol 2007;189:2988–95. [69] Xu CP, Van de Belt-Gritter B, Busscher HJ, Van der Mei HC, Norde W. Calorimetric comparison of the interactions between salivary proteins and Streptococcus mutans with and without antigen I/II. Colloids Surf B Biointerfaces 2007;54:193–9. [70] Xu CP, Van de Belt-Gritter E, Dijkstra RJB, Norde W, Van der Mei HC, Busscher HJ. Interaction forces between salivary proteins and Streptococcus mutans with and without antigen I/II. Langmuir 2007;23:9423–8. [71] Vadillo-Rodriguez V, Schooling SR, Dutcher JR. In situ characterization of differences in the viscoelastic response of individual gram-negative and grampositive bacterial cells. J Bacteriol 2009;191:5518–25. [72] Van der Mei HC, Rustema-Abbing M, De Vries J, Busscher HJ. Bond strengthening in oral bacterial adhesion to salivary conditioning films. Appl Environ Microbiol 2008;74:5511–5. [73] Abu-Lail NI, Camesano TA. Specific and nonspecific interaction forces between •• Escherichia coli and silicon nitride, determined by Poisson statistical analysis. Langmuir 2006;22:7296–301. [74] Marx KA. Quartz crystal microbalance: a useful tool for studying thin polymer films and complex biomolecular systems at the solution–surface interface. Biomacromolecules 2003;4:1099–120. [75] Dutta AK, Nayak A, Belfort G. Viscoelastic properties of adsorbed and cross-linked polypeptide and protein layers at a solid–liquid interface. J Colloids Interface Sci 2008;324:55–60.

517

[76] Kanazawa KK, Gordon JG. Frequency of a quartz microbalance in contact with liquid. Anal Chem 1985;57:1770–1. [77] Kanazawa KK, Gordon JG. The oscillation frequency of a quartz resonator in contact with a liquid. Anal Chim Acta 1985;175:99–105. [78] Otto K, Silhavy TJ. Surface sensing and adhesion of Escherichia coli controlled by the •• Cpx-signaling pathway. Proc Natl Acad Sci USA 2002;99:2287–92. [79] Poitras C, Tufenkji N. A QCM-D-based biosensor for E. coli O157:H7 highlighting the relevance of the dissipation slope as a transduction signal. Biosens Bioelectron 2009;24:2137–42. [80] Shen Z, Huang M, Xiao C, Zhang Y, Zeng X, Wang PG. Nonlabeled quartz crystal microbalance biosensor for bacterial detection using carbohydrate and lectin recognitions. Anal Chem 2007;79:2312–9. [81] Serra B, Gamella M, Reviejo AJ, Pingarron JM. Lectin-modified piezoelectric biosensors for bacteria recognition and quantification. Anal Bioanal Chem 2008;391:1853–60. [82] Pavey KD, Barnes LM, Hanlon GW, Olliff CJ, Ali Z, Paul F. A rapid, non-destructive method for the determination of Staphylococcus epidermidis adhesion to surfaces using quartz crystal resonant sensor technology. Lett Appl Microbiol 2001;33: 344–8. [83] Pavey KD, Ali Z, Olliff CJ, Paul F. Application of the quartz crystal microbalance to the monitoring of Staphylococcus epidermidis antigen–antibody agglutination. J Pharm Biomed Anal 1999;20:241–5. [84] Nivens DE, Chambers JQ, Anderson TR, Tunlid A, Smit J, White DC. Monitoring microbial adhesion and biofilm formation by attenuated total reflection/Fourier transform infrared spectroscopy. J Microbiol Meth 1993;17:199–213. [85] Olsson ALJ, Van der Mei HC, Busscher HJ, Sharma PK. Influence of cell surface appendages on the bacterium–substratum interface measured real-time using QCM-D. Langmuir 2009;25:1627–32. [86] Olsson ALJ, Van der Mei HC, Busscher HJ, Sharma PK. Novel analysis of bacterium– substratum bond-maturation measured using a quartz crystal microbalance. Langmuir 2010;26:11113–7.