Published Ahead of Print 28 October 2011. 10.1128/AEM.06780-11.

2012, 78(1):120. DOI:Appl. Environ. Microbiol. der Mei and Henk J. BusscherBrandon W. Peterson, Prashant K. Sharma, Henny C. van Centrifugal CompactionBacterial Cell Surface Damage Due to

http://aem.asm.org/content/78/1/120Updated information and services can be found at:

These include:

REFERENCEShttp://aem.asm.org/content/78/1/120#ref-list-1This article cites 12 articles, 6 of which can be accessed free at:

CONTENT ALERTS more»articles cite this article),

Receive: RSS Feeds, eTOCs, free email alerts (when new

http://journals.asm.org/site/misc/reprints.xhtmlInformation about commercial reprint orders: http://journals.asm.org/site/subscriptions/To subscribe to to another ASM Journal go to:

on January 28, 2013 by guesthttp://aem

.asm.org/

Dow

nloaded from

Bacterial Cell Surface Damage Due to Centrifugal Compaction

Brandon W. Peterson, Prashant K. Sharma, Henny C. van der Mei, and Henk J. Busscher

Department of Biomedical Engineering, W. J. Kolff Institute, University Medical Center Groningen and University of Groningen, Groningen, The Netherlands

Centrifugal damage has been known to alter bacterial cell surface properties and interior structures, including DNA. Very fewstudies exist on bacterial damage caused by centrifugation because of the difficulty in relating centrifugation speed and con-tainer geometry to the damage caused. Here, we provide a simple, versatile method of analysis for describing the compaction ofbacteria during centrifugation based on a proposed centrifugation coefficient, C. Values of C can be related to different bacterialcell surface properties. Changing the geometry of the centrifugation container or centrifugation speeds changed the value of Csignificantly. Initial deposition rates of Staphylococcus aureus ATCC 12600 to a glass surface decayed exponentially from 4,217to 1,478 cm�2 s�1 with increasing C, while the proportion of staphylococci with a zeta potential of around �15 mV decreasedfrom 97 to 58%. These surface-sensitive parameters were used independently to derive a critical centrifugation coefficient(0.040), above which centrifugation was considered to impact the outcome of surface-sensitive experiments due to cell surfacedamage. The critical centrifugation coefficient could successfully predict staphylococcal cell surface damage, i.e., a significantchange in initial deposition rate or zeta potential distribution, in 84% of all cases included here, whereas the centrifugationspeed could predict damage in only 58% of all cases. Moreover, controlling the centrifugation coefficient within narrow limitsover a series of experiments yielded 43% smaller standard deviations in initial staphylococcal deposition rates than with centrif-ugation at fixed speeds for replicate experiments.

Centrifugation is a common laboratory practice used for har-vesting planktonic bacteria. For harvesting, a wide variety of

forces (roughly ranging from 1,000 to 12,000 � g) (1, 10–12) isused without mentioning a reason for a particular choice. Centrif-ugation in essence involves compacting bacteria into a pellet,causing collisions against each other that result in shear forces onthe bacterial cell surface, which may easily lead to cell surfacedamage with a potential effect on the outcome of surface-sensitiveexperiments. Many experimental protocols choose high centrifu-gation speeds to collect as many bacteria as possible while assum-ing that it does not cause any bacterial cell damage or cell death (4,10). However, cell surface damage during bacterial centrifugationat 15,000 � g was demonstrated to cause significant reductions inEscherichia coli viability compared to centrifugation at 5,000 � g,while little effect was detected on the viability of Psychrobacter sp.strain SW8 or Staphylococcus epidermidis (9). Moreover, bacterialcell surface damage can be expected to affect surface-sensitivephenomena, such as their adhesion to surfaces. Chemical andphysical surface damage to Pseudomonas aeruginosa, for instance,yielded lower initial deposition rates to substratum surfaces thanwere observed for undamaged bacteria (2). In general, harvestingby the high-speed centrifugation (15,000 � g) of bacterial cultureswas found to reduce the surface charge of organisms not protectedby a dense layer of extracellular polymeric substances (9). Kineticexperiments demonstrated differences in surface characteristics inChlamydia psittaci after centrifugation at 1,580 � g, and furtherinvestigation revealed a 10-fold drop in virulence, suggesting thatinterior structures also were modified (1). Possible effects of cen-trifugation often are assumed to be absent or nonexistent, which ispartly because there is no simple method available to predict orevaluate bacterial cell surface damage due to centrifugation. Thelarge range of applied centrifugation speeds likely yields unknownbacterial cell surface damage, causing variations between reportedresults.

Therefore, the aim of this study is to demonstrate how centrif-

ugal compaction affects the outcome of surface-sensitive experi-ments due to cell surface damage and to propose a predictivemethod from which possible cell surface damage can be inferred,using Staphylococcus aureus ATCC 12600 as an example strain. Acentrifugation coefficient, C, is proposed that can be easily calcu-lated from the measured volume of a single bacterium, the result-ing bacterial pellet volume, and the number of cells within thepellet. The coefficient C is suggested to be predictive for cell sur-face damage to occur or not during centrifugation, and its predic-tive nature will be validated by studying initial deposition to glassand zeta potentials of staphylococci centrifuged at differentspeeds.

MATERIALS AND METHODSTheoretical background: coefficient of centrifugation. During centrifu-gation, bacteria in a suspension are compacted into a pellet. This creates ashear force that is active on the contact points between bacteria duringcompaction, which depends on the orientation of the bacteria with re-spect to each other and the direction of the centrifugal force. Conse-quently, a coefficient of centrifugation, C, describing the compactionwithin a pellet can be defined according to the equation

C � 0.63�Vbacterium nbacteria

Vpellet� (1)

in which the influence of the centrifugation speed is expressed by theratio between the total bacterial volume occupied (i.e., the volume of asingle bacterium [Vbacterium]) multiplied by the number of bacteria in thepellet (nbacteria) and the volume of the pellet (Vpellet). The factor 0.63

Received 2 September 2011 Accepted 19 October 2011

Published ahead of print 28 October 2011

Address correspondence to Prashant K. Sharma, p.k.sharma@umcg.nl.

Copyright © 2012, American Society for Microbiology. All Rights Reserved.

doi:10.1128/AEM.06780-11

120 aem.asm.org 0099-2240/12/12.00 Applied and Environmental Microbiology p. 120–125

on January 28, 2013 by guesthttp://aem

.asm.org/

Dow

nloaded from

represents the random angle orientation of the bacteria with respect toeach other and the direction of the centrifugal force (Fig. 1a) (8). Withincreasing centrifugation speed, more bacterial cells leave the suspensionand enter into the pellet, while increased compaction leads to a smallerpellet, which yields a higher coefficient of centrifugation.

The determination of the coefficient of centrifugation according toequation 1 requires the measurement of the volume of a single bacterium,the number of bacteria in a pellet, and the pellet volume. For sphericallyshaped bacteria, the radius of a single bacterium can be determined bymicroscopic techniques, such as scanning electron, atomic force, orphase-contrast microscopy, or by more indirect techniques, such as dy-namic light scattering. The measured radius then can be propagated intothe bacterial volume (Vbacterium). For rod-shaped bacteria, other measuresand formulas should be applied. The enumeration of the total number ofbacteria in a pellet can be done after the dispersal of the pellet into sus-pension, followed by direct counting under a phase-contrast microscope.Pellet geometry strictly coincides with the shape of the container used forcentrifugation. Typically, pellets form according to three specific geomet-ric shapes at the base/side of the container and possess a cone (equation 2),cylindrical wedge (equation 3), or spherical cap (equation 4) shape, withvolumes given by

Vpellet � cone��R2d'

3 (2)

Vpellet � wedge�h3b�a�3R2 � a2�� 3R2�b � R��� (3)

Vpellet � cap��d'6 �3a2 � d'2� (4)

where R is the radius of the container, d= is the depth of the pellet of aspherical cone or cap, h is p � cos� (� being the fixed angle of the rotor andp the measured diameter across the vertical plane), b is the radius acrossthe curved edge of the cylinder (equal to p � sin�), a is the measuredradius across the straight edge of the cylinder, and � is the angle of inclineof the pellet segment, which is set equal to cos�1[1 � (b/R)] (Fig. 1b).

Bacterial strain, culturing, and harvesting by centrifugation. S. au-reus ATCC 12600 was precultured from a blood agar plate in 10 ml oftryptone soya broth (TSB; Oxoid Ltd., Basingstoke, England) for 24 h andsubsequently subcultured into 200 ml fresh TSB under static conditions at37°C for 18 h. Sixty-ml aliquots were taken from the culture and centri-fuged (rotor JA14, 200 ml at a 25° incline; Beckman J2-MC; BeckmanCoulter, Palo Alto, CA; and rotor JA18, 70 ml at 23° incline; Beckman

FIG 1 Schematic of applied forces and geometries of bacterial pellets. (a) The contact point between two bacteria during centrifugation. The shear force actingon the two cells due to compaction during centrifugation can be expressed as the sinus component (�) of the centrifugal force. In a pellet, the angle � takes randomvalues between 0 and 90 degrees. (b) Pellet shapes and dimensions used in the calculation of the pellet volume (Vpellet) (equations 2 to 4). Measured dimensionsare 2a, p, and d= for both the cylindrical wedge and spherical cap geometries. The cone geometry was not used in the manuscript.

Centrifugal Cell Surface Damage

January 2012 Volume 78 Number 1 aem.asm.org 121

on January 28, 2013 by guesthttp://aem

.asm.org/

Dow

nloaded from

Coulter, Galway, Ireland) at 1,330, 6,250, 6,480, 14,800, or 15,300 � g for5 min at 10°C. Note that the two rotors differed in rotor axis angles,container volumes, and degrees of filling.

Experimental determination of centrifugation coefficients. To de-termine the volume of a single bacterium, a 60-ml aliquot was centrifugedat medium speed (6,480 � g) for 5 min at 10°C (JA14 rotor), and thesupernatant was discarded. The pellet was washed twice with 10 mlphosphate-buffered saline (PBS; 5 mM K2HPO4, 5 mM KH2PO4, 150 mMNaCl, pH 7.0) and was resuspended by flushing PBS against the side wallof the container using a pipette. The final pellet was resuspended in 10 mlPBS to a density of 108 per ml using a Bürker-Türk counting chamber forenumeration. The hydrodynamic radius of the staphylococci was mea-sured using dynamic light scattering (Zetasizer nano series nano-ZS;Malvern Instruments, Worcestershire, United Kingdom). Three mea-surements were combined from three subcultures to obtain the averageradius for S. aureus ATCC 12600.

Pellet volumes were measured after the different centrifugations inJA14 and JA18 rotor containers. After discarding the supernatant, thepellet dimensions 2a, p, and d= were measured using a caliper. Subse-quently, equation 2, 3, or 4 was chosen to calculate the pellet volumedepending on the geometry of the container. Pellets then were resus-pended in 10 ml PBS and diluted, and the number of bacteria in the pelletwas determined using a Bürker-Türk counting chamber. In all cases, S.aureus ATCC 12600 appeared as single cells during enumeration.

Surface-sensitive experiments. (i) Bacterial deposition in a parallelplate flow chamber. Bacterial deposition to surfaces is highly sensitive tochanges in bacterial cell and substratum surface properties. Therefore,staphylococcal deposition experiments were carried out in a parallel plateflow chamber. First, staphylococci were resuspended in PBS at a density of3 � 108 per ml after the different centrifugations. Initial bacterial deposi-tion rates (j0) were determined on the bottom glass plate of the flowchamber with dimensions (3) of 17.5 by 1.6 by 0.075 cm. Briefly, thebacterial suspension flowed through the flow chamber at a wall shear rateof 15 s�1 for 60 min at 21°C. Images were taken every minute, and thenumber of bacteria adhering per cm2 was enumerated using in-house-developed software based on MATLAB. From a plot of the number ofadhering bacteria versus time, the initial deposition rate (j0, in cm�2 s�1)was calculated by linear regression analysis. All experiments were carriedout five times with different bacterial cultures.

(ii) Zeta potential distribution. To reveal possible cell surface damagefrom differences in zeta potentials, bacteria were suspended after the dif-ferent centrifugations in low-ionic-strength PBS (5 mM K2HPO4, 5 mMKH2PO4, 30 mM NaCl, pH 7.0). Zeta potential distributions were mea-sured with a Lazer Zee meter 501 (PenKem, Bedford Hills, NY) equippedwith image analysis options for zeta sizing. Briefly, the microelectropho-resis chamber was filled with a bacterial suspension at a density of 107 to108 bacteria per ml, and a voltage difference of 150 V was applied to thechamber. The velocity of each individual bacterium was determined byimage sequence analysis and expressed as a zeta potential, assuming thatthe Helmholtz-Smoluchowski equation holds. Each zeta potential distri-bution presented contains a minimum of 100 individual bacteria. Zetapotentials were determined six times from six different bacterial cultures.

Data analysis. Centrifugation coefficients were grouped according tothe centrifugal force applied and compiled into box-plot format withmedian and quartiles, with whiskers representing minimum and maxi-mum values (GraphPad Prism v. 5.00) (Fig. 2). Number of bacteria har-vested, pellet volume, and calculated centrifugation coefficients as afunction of centrifugation force then were compared using unpaired two-tailed Student t tests with Welch’s correction for unequal variances. Toeliminate increased alpha errors (0.05) from multiple Student t tests, thereported P values were multiplied by the number of tests performed: twofor comparison of container geometry (at 6,480 and 6,250 � g and at15,300 and 14,800 � g) and three for speed comparisons (6,480 and15,300 � g in a JA14 rotor and 1,330, 6,250, and 14,800 � g in a JA18rotor).

RESULTSCentrifugation coefficients. The hydrodynamic radius of S. au-reus ATCC 12600 in PBS was found to be 480 � 86 nm, which,combined with numbers of staphylococci in the pellets (4.5 � 1010

to 5.8 � 1010 staphylococci; Fig. 2a) and the pellet volumes (378 to405 mm3 in the JA14 rotor and 182 to 268 mm3 in the JA18 rotor;Fig. 2b), yielded the centrifugation coefficients presented in Fig. 2cas a function of the centrifugation speed. Incidentally, atomicforce and phase-contrast microscopy yielded an average radius ofS. aureus of 480 and 500 nm, respectively. Changing the centrifu-gation speed or geometry of the container did not significantlyalter the number of bacteria harvested in a pellet. Pellet volume,however, decreased with increasing speed and changed signifi-cantly (P � 0.0002) with the geometry of the centrifugation con-tainers.

FIG 2 Effects of centrifugation speed and container geometry on (a) the num-ber of S. aureus ATCC 12600 cells harvested in a pellet, (b) the pellet volumeafter centrifugation, and (c) the centrifugation coefficient (C) for S. aureusATCC 12600. Shaded quartiles indicate those determined using the JA14 rotor,while unfilled quartiles designate those determined using the JA18 rotor.Whiskers indicate the minimum and maximum values for five measurements.Statistically significant differences for centrifugal force are indicated with anasterisk (P � 0.05) or double asterisks (P � 0.01), and those for rotor geometryare indicated as ### (P � 0.001). The horizontal line in panel c indicates thecritical C value, as derived from initial deposition rates and zeta potentialdistributions (Fig. 3 and 4).

Peterson et al.

122 aem.asm.org Applied and Environmental Microbiology

on January 28, 2013 by guesthttp://aem

.asm.org/

Dow

nloaded from

Effects on the outcome of surface-sensitive experiments. Thenumbers of staphylococci adhering to the glass substratum in-creased linearly with time for at least 15 min under the presentexperimental conditions (Fig. 3). The slope of the linear deposi-tion during the first 15 min varied with the value of C (Fig. 3a) andresulted in an exponential decay (R2 � 0.75) with increasing cen-trifugation coefficients (Fig. 3b). The median of the initial depo-sition rates observed amounted to 2,703 cm�2 s�1 and wasadapted to derive a critical centrifugation coefficient (0.040),above which centrifugation was considered to severely affect theinitial staphylococcal deposition rate.

The zeta potential distribution of S. aureus ATCC 12600showed multiple subpopulations with different zeta potentialsand full widths at half maximum (Fig. 4), ranging from �32 to �4mV. The major subpopulation had a zeta potential of around �15mV (Fig. 4a) regardless of the centrifugation coefficient, albeit indifferent proportions of the total population. The occurrence ofbacteria with a zeta potential of around �15 mV decreased from97 to 58% with increasing centrifugation coefficients, which wasconcurrent with the development of other subpopulations ofstaphylococci with less negative zeta potentials. The percentage ofthe staphylococcal subpopulation with a zeta potential of around

�15 mV is given as a function of the centrifugation coefficient inFig. 4b and also can be used to determine a critical centrifugationcoefficient, defined similarly to the method used for the initialdeposition rates (Fig. 3b) but based on the median percentage sizeof the �15-mV subpopulation. Note that the critical centrifuga-tion coefficient from zeta potential distributions above which cen-trifugation is considered to severely impact the cell surface prop-erties (0.037) coincides well with the one derived from initialdeposition rates (0.040).

Predictive capability of the critical centrifugation coeffi-cient. If the predictive nature of the critical centrifugation coeffi-cient was 100%, then the outcome of any surface-sensitive exper-iment at a centrifugation coefficient higher than the critical onewould yield data below the median outcome of that experimentalset and vice versa. To quantify the predictive nature of the criticalcoefficient of centrifugation, we determined the percentage of ex-periments fulfilling this condition. This yielded a predictive capa-bility with respect to the initial staphylococcal deposition rate andzeta potential distribution of 95 and 74%, respectively. Therefore,the average predictive capability of the critical centrifugation co-efficient amounts to 84%, which is higher than that from a similaranalysis on the basis of centrifugation speed that yielded a predic-

FIG 3 Bacterial deposition in a parallel plate flow chamber. (a) The mean number of S. aureus ATCC 12600 cells deposited on glass as a function of time fordifferent ranges of the centrifugation coefficient. (b) The initial deposition rate of S. aureus ATCC 12600 to glass as a function of the centrifugation coefficientapplied during the harvesting of the staphylococci. The curves obey an exponential decay with a correlation coefficient (R2) equal to 0.75. The dotted horizontalline indicates the median deposition rate from which a critical centrifugation coefficient of 0.040 is derived.

FIG 4 Bacterial velocities and derivation of zeta potentials. (a) Example of the effect of centrifugation on the zeta potential distribution in a low-ionic-strengthsuspension of S. aureus ATCC 12600 for two different centrifugal forces. Each distribution is based on the measurement of the zeta potentials of between 100 and125 bacteria. (b) The percentage of staphylococci contained within the subpopulation with a zeta potential of around �15 mV as a function of the centrifugationcoefficient. The dotted horizontal line indicates the median percentage size of this subpopulation, from which a critical centrifugation coefficient of 0.037 isderived.

Centrifugal Cell Surface Damage

January 2012 Volume 78 Number 1 aem.asm.org 123

on January 28, 2013 by guesthttp://aem

.asm.org/

Dow

nloaded from

tive capability of only 58%. Importantly, controlling the centrifu-gation coefficient within narrow limits of 0.02 units over a series ofexperiments yielded 43% smaller standard deviations in initialstaphylococcal deposition rates than when the centrifugationspeed was fixed.

DISCUSSION

Centrifugation is a necessary evil in harvesting microorganismsfrom liquid cultures, and the potential damage arising from cen-trifugal compaction usually is ignored due to a lack of an easyassessment method. Here, we have introduced a new method forpredicting cell surface damage from effects on the outcome of twosurface-sensitive experiments, based on a so-called centrifugationcoefficient. The centrifugation coefficient can be easily calculatedfrom the pellet volume after centrifugation, the number of bacte-ria in the pellet, and the volume of a single bacterium, and it isindicative of the degree of compaction of bacteria in a pellet andthus of the shear forces acting on the cell surface during compac-tion. The centrifugation of S. aureus ATCC 12600 at coefficientsabove 0.040 resulted in a significant reduction in initial staphylo-coccal deposition rates to glass and a change in their zeta potentialdistributions. The centrifugation coefficient describes the averagecompaction observed throughout the pellet rather than the com-paction at one point due to differential compaction observed atdifferent depths within a pellet. It is interesting that the criticalcentrifugation coefficient appeared independent of whether it wasderived from the deposition behavior of the strain or its negativecell surface charge density (zeta potential distribution), suggestingthat other experiments involving the bacterial cell surface alsocould have been used to arrive at this critical value.

Hydrostatic pressures within a 15-ml centrifuge container at10,000 � g can be as high as 10 atm, and it is likely that cell surfacematerial is stripped off the cell surface by centrifugation (5). Cellsurface molecules are anchored to the membrane, and Grandboiset al. (6) have demonstrated that most anchored organic bondsrupture at 4.5 nN. The upper threshold of 4.5 nN for anchoredorganic bonds therefore was used to estimate whether centrifugalforces could cause cell surface damage. The forces that bacteria ina pellet experience depend not only on their position in the pelletbut also on the centrifugation speed and the degree of compactionaccording to the equation

F(h) � m(h)fC (5)

in which m(h) is the total mass of the bacteria above height h, f isthe centrifugal acceleration, and C is the centrifugation coeffi-cient. As an example, we have plotted the pellet volume as a func-tion of its height in Fig. 5a. It can be seen that only a fraction of allbacteria in a pellet actually experience a shear force above 4.5 nN.The mass at the threshold height m(hthres), below which bacteriaare affected by centrifugation, can be calculated from equation 5,assuming the forces experienced [F(h)] to be equal to the molec-ular bond rupture force of 4.5 nN and inserting the known cen-trifugal acceleration f and the measured coefficient of centrifuga-tion C. From the calculated mass affected, one can use equation 2,3, or 4 depending on the container geometry to calculate the vol-ume affected. Figure 5b shows that the volume fraction of bacteriaaffected by centrifugation increases with the coefficient of centrif-ugation, and even when centrifugation is performed below thecritical centrifugation coefficient a sizeable fraction of the bacteriacan be expected to experience a shear force of 4.5 nN or higher.

However, since these shear forces occur below the critical coeffi-cient value, they did not severely affect the surface-sensitive prop-erties measured here. In the proposed analysis of cell surface dam-age, we assumed that possible effects of bacterial resuspensionduring harvesting could be neglected. We believe this assumptionto be valid, since we observed no difference in the zeta potentialdistribution of staphylococci resuspended by spontaneous disso-lution over time and by pipetting up to six times (data not shown).

In summary, shear forces active during the centrifugal com-paction cause surface damage and affect bacterial cell surface-sensitive properties, including their deposition to surfaces andsurface charge. The proposed centrifugation coefficient, C, can beused to predict cell surface damage with an impact on the cellsurface-sensitive properties selected. The centrifugation coeffi-cient should be kept low by using a wide, flat-bottom container tominimize the pellet height and the shear forces during compac-tion. For S. aureus ATCC 12600, a critical value of 0.040 for C wasfound, resulting in an average success rate of 84% in predicting cellsurface damage affecting initial staphylococcal deposition ratesand zeta potentials. Furthermore, controlling the centrifugationcoefficient within narrow limits decreased the standard deviationsbetween replicates more strongly than controlling the centrifuga-tion speed. This is an important conclusion of general applicabil-ity, since it may help to improve the reproducibility within oneand among different laboratories when performing cell surface-sensitive experiments with centrifuged bacteria. The measure-ment of the centrifugation coefficient is simple and solely requires

FIG 5 Volume affected within a bacterial pellet. (a) Example of the relationbetween the height in a bacterial pellet with the volume of the pellet (left axis)and the shear force exerted on the bacteria (right axis). Data represent a 5-mm-thick pellet of S. aureus ATCC 12600 centrifuged at 6,480 � g, yielding acoefficient C value of 0.05. Volume affected is indicated by the green shadedregion. (b) The centrifugation coefficient, C, as a function of fractional volumeaffected by shear force larger than 4.5 nN at two centrifugation speeds. Thehorizontal line indicates the critical centrifugation coefficient value for C es-tablished here.

Peterson et al.

124 aem.asm.org Applied and Environmental Microbiology

on January 28, 2013 by guesthttp://aem

.asm.org/

Dow

nloaded from

an estimate of the volume of the bacterial pellet and a single bac-terium, as demonstrated here for S. aureus ATCC 12600.

REFERENCES1. Allan I, Pearce JH. 1979. Modulation by centrifugation of cell suscepti-

bility to chlamydial infection. J. Gen. Microbiol. 111:87–92.2. Bruinsma GM, Rustema-Abbing M, van der Mei HC, Busscher HJ.

2001. Effects of cell surface damage on surface properties and adhesion ofPseudomonas aeruginosa. J. Microbiol. Methods 45:95–101.

3. Busscher HJ, van der Mei HC. 2006. Microbial adhesion in flow displace-ment systems. Clin. Microbiol. Rev. 19:127–141.

4. Deupree SM, Schoenfisch MH. 2008. Quantitative method for determin-ing the lateral strength of bacterial adhesion and application for charac-terizing adhesion kinetics. Langmuir 24:4700 – 4707.

5. Gilbert P, Coplan F, Brown MR. 1991. Centrifugation injury of gram-negative bacteria. J. Antimicrob. Chemother. 27:550 –551.

6. Grandbois M, Beyer M, Rief M, Clausen-Schaumann H, Gaub HE.1999. How strong is a covalent bond? Science 283:1727–1730.

7. Ionescu M, Belkin S. 2009. Simple quantification of bacterial envelope-associated extracellular materials. J. Microbiol. Methods 78:302–306.

8. Khoei AR. 2002. Numerical simulation of powder compaction processesusing an inelastic finite element analysis. Mater. Design. 23:523–529.

9. Pembrey RS, Marshall KC, Schneider RP. 1999. Cell surface analysistechniques: what do cell preparation protocols do to cell surface proper-ties? Appl. Environ. Microbiol. 65:2877–2894.

10. Romaní AM, et al. 2008. Relevance of polymeric matrix enzymes duringbiofilm formation. Microb. Ecol. 56:427– 436.

11. Van der Mei HC, Rustema-Abbing M, de Vries J, Busscher HJ. 2008.Bond strengthening in oral bacterial adhesion to salivary conditioningfilms. Appl. Environ. Microbiol. 74:5511–5515.

12. Vats N, Lee SF. 2000. Active detachment of Streptococcus mutans cellsadhered to epon-hydroxylapatite surfaces coated with salivary proteins invitro. Arch. Oral Biol. 45:305–314.

Centrifugal Cell Surface Damage

January 2012 Volume 78 Number 1 aem.asm.org 125

on January 28, 2013 by guesthttp://aem

.asm.org/

Dow

nloaded from