the influence of electrostatic forces on protein adsorption

Colloids and Surfaces B: Biointerfaces 44 (2005) 56–63

The influence of electrostatic forces on protein adsorption

G.V. Lubarsky, M.M. Browne, S.A. Mitchell, M.R. Davidson, R.H. Bradley∗

Advanced Materials and Biomaterials Research Centre, School of Engineering, The Robert Gordon University, Aberdeen, AB10 1FR, UK

Received 12 February 2005; accepted 17 May 2005

Abstract

In this paper we investigate the importance of electrostatic double layer forces on the adsorption of human serum albumin by UV–ozonemodified polystyrene. Electrostatic forces were measured between oxidized polystyrene surfaces and gold-coated atomic force microscope(AFM) probes in phosphate buffered saline (PBS) solutions. The variation in surface potential with surface oxygen concentration was measured.The observed force characteristics were found to agree with the theory of electrical double layer interaction under the assumption of constantpotential. Chemically patterned polystyrene surfaces with adjacent 5�m × 5�m polar and non-polar domains have been studied by AFMbefore and after human serum albumin adsorption. A topographically flat surface is observed before protein adsorption indicating that thepatterning process does not physically modify the surface. Friction force imaging clearly reveals the oxidation pattern with the polar domainsb med on thep yer forcesf ned surfacesi©

K

1

aiescpftapcai[

nc-ins,cell

ting

aryandrents of

er-rbed

e ofn of

oteinemptoteinsaper,key

ions

0d

eing characterised by a higher relative friction compared to the non-polar, untreated domains. Far-field force imaging was perforatterned surface using the interleave AFM mode to produce two-dimensional plots of the distribution of electrostatic double-la

ormed when the patterned polystyrene surfaces is immersed in PBS. Imaging of protein layers adsorbed onto the chemically patterndicates that the electrostatic double-layer force was a significant driving force in the interaction of protein with the surface.

2005 Elsevier B.V. All rights reserved.

eywords:Electrostatic double-layer force; Protein attachment; Polystyrene; AFM; Human serum albumin

. Introduction

The interaction between proteins and the surface ofmaterial is a fundamental phenomenon with important

mplications in a number of biotechnological processes. Forxample the irreversible binding of albumin to hydrophobicurfaces is thought to be the reason for the poor attachment ofells to many polymers. Polymers such as polyethylene andolystyrene, which are essentially non-polar in their native

orm, do not support cell attachment unless treated to increasehe surface energy by the incorporation of heteroatoms suchs oxygen or nitrogen or by the pre-adhesion of attachmentroteins. Surface oxidative treatments are known to promoteell attachment on polymer surfaces with rates of attachmentnd subsequent proliferation being positively influenced by

ncreasing levels of functionalisation and surface energy1–5].

∗ Corresponding author. Tel.: +44 1224 262822; fax: +44 1224 262837.E-mail address:[email protected] (R.H. Bradley).

Chemically micropatterned surfaces, which have futionalised (polar) and unfunctionalised (non-polar) domaand which are consequently adherent or abherent toattachment, provide a potentially useful means of direccell growth to specific regions of device surfaces[6,7].Indeed, for in vitro culture of a range of human primand other cell types, distinctly different attachmentproliferation behaviour are observed on domains of diffechemistry. These differential cell responses to surfacedifferent chemistry are likely, in part, to result from diffences in the composition and orientation of the adsoprotein structure[8].

In a previous study, we investigated the importancsurface chemistry on the orientation and concentratioalbumin adsorbed to polystyrene surfaces[9]. However, thephysical size of proteins such as albumin mean that pradsorption is essentially a colloidal process and any attto understand the interaction between surfaces and prmust take surface forces into account. In the present pwe look at the importance of surface forces as one of thedriving forces behind protein adsorption. Since interact

927-7765/$ – see front matter © 2005 Elsevier B.V. All rights reserved.oi:10.1016/j.colsurfb.2005.05.010

G.V. Lubarsky et al. / Colloids and Surfaces B: Biointerfaces 44 (2005) 56–63 57

between surfaces and proteins occur in aqueous solutions anelectrostatic double layer will exist either through the dis-sociation of surface functional groups and/or through theadsorption of ions from solution. The surface charge is bal-anced by the accumulation of an equal number of oppositelycharged counterions that are either bound to the surface toa form the Stern layer or present in an atmosphere abovethe surface to form an electrostatic double layer. Since pro-teins are usually charged molecules in aqueous solution thepresence of this electrostatic double layer with its associatedelectric field will have an important influence on the interac-tion between polymer surfaces and proteins.

In this study, a UV–ozone treatment has been used toproduce surfaces with adjacent regions of different chemistrythat allow the study of the electrostatic forces existing abovenative and oxidised polystyrene surfaces under otherwiseidentical conditions. Atomic force microscopy will be usedto measure the electrostatic forces of these chemicallypatterned surfaces in phosphate buffered saline—a mediumwidely used in cell culture and protein adsorption work.The AFM will also be used to map the topography, fric-tional forces and adhesion forces before and after proteinadsorption.

2. Experimental

2

renec ter-n ablef teinl .5 nmb ilstp is oft er ofc ctrumr witht mem hichp f them d nota teinl

hep mer-c nd2 ve-l onea

wasa ppert thistt pes.

However, the mask used in the present study has bar andspace dimensions of 5�m and thus produces a pattern ofsub-cellular dimensions. Intimate contact between grid andsample was provided using home-built mask aligner, whicheffectively pressed the mask into contact with the surface.The UV irradiation was performed in an ambient air envi-ronment with an exposure time of 60 s. XPS analysis showsthat this treatment incorporates approximately 7 at.% surfaceoxygen into the polystyrene surface. The precise chemistryof the oxidation has been discussed elsewhere[4,5,11,12].

Protein solutions of 0.1 mg/mL were prepared with humanalbumin, essentially fatty acid and globulin free lyophilisedpowder (Cat. No. A 3782, Sigma-Aldrich Company, Dorset,UK) in Dulbecco’s phosphate buffered saline at pH 7.2 (PBS,Cat. No. D8537, Sigma-Aldrich). A 50�L drop of solutionwas placed on chemically patterned surfaces and allowed toincubate at 37◦C. After 1 h samples were removed fromthe incubator and rinsed twice in Milli-Q water. Sampleswere then re-immersed in Milli-Q and placed on a shakerfor 4 h, rinsed twice again in Milli-Q and then allowed to dryovernight at room temperature.

2.2. Electrostatic double-layer force measurement

Force–distance experiments were performed using aDigital Instruments Multimode SPM system (Nanoscope IIIa hee theb em-i rsedi . No.D ofc si lytew

-u asw ntere . The

.1. Surface preparation

The material used throughout this study was polystyut from Nunc untreated cell culture dishes (VWR Inational Ltd). In order to produce flatter surfaces suit

or the study of electrostatic forces and adsorbed proayers we reduced the roughness to between 0.4 and 0y heating the polystyrene to approximately 413 K whressed between two clean silicon wafers. XPS analys

he flattened polystyrene reveals no oxidation or transfontaminants during this process and the XPS C1s speecorded from the processed surfaces is consistenthat of pure PS[10]. The resulting surfaces contain soacroscopic features in the form of periodic steps, wresumably result from thermal stress, and viscous flow oelted PS. These features were relatively large and diffect the subsequent AFM imaging of the adsorbed pro

ayers.The UV source for chemical modification of t

olystyrene surface was a high-intensity low-pressureury grid lamp, which generates UV light at 184.9 a53.7 nm (Jelight Company Inc. Irvine, USA). These wa

engths are known to excite molecular oxygen to form oznd to photosensitise polymer surfaces[11,12].

The production of chemically patterned surfaceschieved by irradiating the polystyrene through a co

ransmission electron microscope grid. We have usedechnique in previously reported tissue culture work[6,7]o control the spatial attachment of a range of cell ty

) in liquid media using an electrochemical fluid cell. Tlectrochemical hardware of the SPM was operated inipotentiostat regime to monitor and control the electroch

cal potentials in the cell. The tip and surface were immen 1% Dulbecco’s phosphate buffered saline (PBS, Cat

8537, Sigma-Aldrich Company, Dorset, UK) solutiononcentration 3.93× 10−3 M at 25◦C. The solution wanjected into the cell, using a syringe, and the electroas hydrodynamically static during measurements.A three-electrode system (Fig. 1) was set up in the liq

id cell of the AFM unit with the gold-coated cantileverorking electrode (WE), a ring-shaped gold wire as coulectrode (CE) and Ag/AgCl as the reference electrode

Fig. 1. Schematic of the experimental three-electrode set up.

58 G.V. Lubarsky et al. / Colloids and Surfaces B: Biointerfaces 44 (2005) 56–63

gold-coated cantilever was electrically linked to the potentio-stat via a copper wire. As the sample was electrically insulatedin the liquid cell, its surface potential was kept constantduring force measurement. To find the optimum workingpotential on the tip the applied voltage was varied incremen-tally between−300 mV and+300 mV using a polystyrenesubstrate with a low oxygen concentration. At each tipvoltage, a force–distance curve was acquired and in thisexperiment, optimum contrast was found with a tip voltageof −120 mV.

All force distance measurements were performed using200�m long V-shaped gold-coated cantilevers with a nom-inal spring constant of 0.06 Nm−1 and tip radius of approx-imately 60 nm. Each force–distance curve was obtainedusing theJ scanner and was an average of 15 approach andretraction cycles. The data was obtained with thexandyscandirections disabled and cycling thez piezo over a specificdistance of 200 nm at a frequency of 0.3 Hz. The deflectionof the cantilever (∆C) was recorded as a function of piezomovement (Z) during both approach and retraction of thetip. Force–distance curves were subsequently obtained byconverting the deflection data and accounting for the relativetip-sample distance, as described by Ducker et al.[13].

2.3. Electrostatic field mapping

staticfi nlyd aves k isu ablet per-a feed-b ighta per-f sure-m inals ly2 easeds rcem d. Bya rfaces apso roge-n

2

mentw ual-i entsw onn tanto ol-l ont both

scanning velocity and applied load. In friction force imag-ing contrast originates from variations in surface chemistryand viscoelastic properties. In a liquid medium and with con-stant experimental conditions the friction force between thecontacting pair is a function of only the adhesion force. Asdescribed above, the adhesion force is a measure of the sur-face energy of the sample, i.e. the friction force image willreveal localised variations in surface energy and chemistry.In order to optimise image contrast the friction force loopwas analysed as a function of the scanning velocity in range0–300�m s−1 with a fixed applied load of 2–3 nN. The slowscan axis was disabled to obtain the friction force as a functionof scanning velocity. Friction force mapping in the AFM isbased on lateral cantilever deflection. The friction force val-ues are recorded by the AFM in volts, which were convertedinto cantilever deflection values and then to force values byemploying the formula proposed by Noy et al.[16] for thecalculation of the lateral spring constant for V-shaped can-tilevers.

2.5. Analysis of the protein layer after adsorption ontopatterned surface

The sample was studied using contact mode topographicalimaging in order to quantify the amount of albumin adsorbedonto the oxidized and non-oxidized domains. The friction-f hy tos in-c n theo facewi cientt thatr nessw ntactm

3

3g

PBSs oxi-d nta -e rodeu ter-a thet tive,t ctivet plei e oftw

The same experimental set-up was used for electroeld imaging of chemically patterned surfaces with the oifference being that the AFM was operated in interlecanning mode. During normal AFM operation, feedbacsed to maintain a constant cantilever deflection. To en

he measurement of cantilever deflections the AFM is oted in interleave mode. During the interleave scan, theack is turned off and the tip is lifted to a pre-selected hebove the chemically patterned polystyrene surface to

orm far-field measurements. The tips used in this meaent were V-shaped gold-coated cantilevers with a nom

pring constant of 0.58 Nm−1 and tip radius of approximate0–60 nm. These cantilevers were used since their incrtiffness prevented the tip jumping to contact during foeasurements with the small separation distances usessembling the scan-lines acquired using different tip-sueparations it is possible to produce two-dimensional mf the electrostatic interaction between the tip and heteeous sample.

.4. Characterisation of patterned surface

The patterned surfaces produced by UV–ozone treatere also characterized by friction force imaging to vis

ze the chemical patterns. The friction force measuremere performed under doubly distilled water with silicitride cantilevers and tips with a nominal spring consf 0.06 Nm−1. A load of 30–35 nN was used for the c

ection of each image. The friction coefficient dependshe nature of the bodies in contact and is a function of

orce image was acquired simultaneously with topograptudy any variations in chemical composition of the albumoated surface. The thickness of the adsorbed protein oxidized and unoxidized regions of the polystyrene suras determined by the AFM tip scratch method[17]. A load-

ng force of 3–5 nN was used as this was found to be suffio scratch through the soft protein layer but was less thanequired to damage the polystyrene substrate. Film thickas then measured by imaging the resulting scratch in coode.

. Results and discussion

.1. Electrostatic double-layer force betweenold-coated tip and oxidised PS

Fig. 2 shows the force–distance curves, acquired inolution between a gold-coated AFM tip and a PS surfaceised for 10 s. Each set of data inFig. 2represents a differepplied tip voltage in the range−300 to+300 mV. In the presnce of an electrolyte solution with the gold-coated electnder electrochemical control, the polystyrene–gold inction is a strong function of the applied potential. As

ip voltage polarity was changed from positive to negahe force between the tip and surface went from attrao repulsive respectively. The ability to control tip-samnteraction by varying the tip voltage is strong evidenche electrostatic origin of this interaction sincevdW forcesould experience no such voltage dependence.


Fig. 2. Normalised force vs. distance curves obtained using various tip volt-ages on polystyrene surface modified by a 10 s UV–ozone treatment.

The surface forces existing between two bodies in aqueoussolutions can be ascribed to a combination of van der Waals,electrostatic, hydrophobic, and hydration forces[18]. In thiswork the hydrophobic forces are negligible since the surfacesof the gold-coated tip and oxidised PS have polar functionalgroups on their surfaces and are therefore largely hydrophilic.We have previously made a thorough study of Young’s mod-ulus[19] variation and of the van der Waals interactions fordifferent systems involving oxidised polystyrene[20]. Wefound that the oxygen incorporation in the surface and subsur-face regions of the polystyrene due the UV–ozone treatmentleads to a reduction of the Young’s modulus as well as chang-ing the magnitude ofvdW interactions. Furthermore, thedistance scaling for thevdW force was found to be lim-ited to between 8 and 12 nm with the magnitude of the forceat this distance of about 0.03–0.7 mNm−1. This comparesto a force measured in the present study of 2.5 mNm−1 at adistance of 10 nm using oxidised PS with an oxygen concen-tration of 15 at.%. If surface forces of an origin other thanelectrostatic had a dominating influence these systems, it isunlikely that they would display a similar distance scaling asthat calculated and/or measured for the electrostatic double-layer force. With this in mind and with clear evidence of thecontrollability of the system by the tip voltage, we believethat in this study the portion of force measured in the range10–50 nm is due to the electrostatic nature of the interactiono fort

PSs fixedt noa weref sionf e–d f thea tancec in

Fig. 3. Normalised force vs. distance curves obtained in phosphate bufferedsaline on polystyrene surface modified by UV–ozone with different treat-ment times (�-0, ♦-20,©-40,�-60 and�-180 s). Tip voltage was fixed at−120 mV.

tip-sample distance was better then 1 nm and the resolutionin force measurements was about 0.3 nN. To performquantitative analysis of curves obtained we modelled theelectrostatic force between a sphere and a flat surfaceseparated by a distanceD with simple continuum theory.

According to the Poisson–Boltzmann theory the pressurebetween two infinite flat plates due to an electrostatic poten-tial and the presence of ions is given by:

p = nkT (eqψ/kT + e−qψ/kT − 2) − εε0

2

(dψ

dx

)2

(1)

whereψ is the potential which depends on the positionx, qis the unit charge,ε andε0 are the dielectric permittivitiesof water and of free space,k is the Boltzmann constant,T isthe absolute temperature,ndenotes the number of moleculesper cubic metre. The first and the second terms are due tothe osmotic pressure and Maxwell stress, respectively. Thepressurep does not depend onx and is equal for all pointsbetween interacted bodies. In the case of surfaces with similarand constant potentialψ0, the condition in the mid-plane isdψ/dx = 0. The theoretical analysis of the force due the inter-action between two electrostatic double layers has been madeon two different assumptions: i.e. constant potential[21] andconstant charge[22]. In the present study we chose a constantp n ino : thes theg es n, thee ivena

F

nly. The force curves obtained were purely repulsivereated samples and did not jump into primary contact.

Force distance curves were also obtained fromurfaces, UV–ozone treated for up to 180 s, using aip potential of−120 mV. In the present work there wasttempt made to acquire retraction curves, as these

ound to contain a large jump-away step due to the adheorce. Acquisition of the retraction portion of the forcistance plot significantly decreased the resolution opproach curves. The resultant normalised force–disurves are shown inFig. 3. The resolution, obtained

otential boundary condition as the experimental situatiour study resembled the constant potential assumptionurface potentialVS is not equal to the applied voltage onold-coated tipVT, furthermore,VT � VS. In the case of thphere–plane model under constant potential assumptiolectrical double layer force as a function of distance is gs[21,23].

=πRεκ(

(ψT2 +ψS

2)(1 − coth(κD)) + 2ψTψS

cosh(κD)

)(2)


κ =√√√√∑

i

2e2nizi2

εkT(3)

whereψT andψS are the surface potentials of the sphereand the plane respectively,R is the radius of sphere,D is theshortest distance between the sphere and the plane surface,κ

the inverse of Debye length of the electrolyte,ni is the con-centration of ions of typei in bulk solution,zi its charge ande is the electron charge. For aqueous solutions at 25◦C andconverting ion concentrations into molar terms, the Debyeparameter yieldsκ = 2.32× 109(

∑ciz

2i )

1/2, whereci is inM andκ is in m−1.

An atomic force microscope-based experiment consists ofa spherical tip mounted on the end of a force-sensing leverinteracting with a planar surface. The force-response of thelever is described by Hooke’s LawF = −K∆C, whereK isthe spring constant corresponding to normal deflection∆C,of the lever. Analysis of plots, shown inFig. 3, revealed twodistinct regions that can be defined as follows: the cantileverin its rest position (∆C = 0); the cantilever deflection is linearwith respect to sample displacement (∆C = Z) representing

points where the tip is in contact with the sample. The portionof the force distance curve that deviates from the cantileverrest fitted line to where it meets the zero tip-surface separationposition was used to fit experimental data with Eq.(2). Fig. 4shows the force curves derived from Eq.(2) in the case of 1%Dulbecco’s phosphate buffered saline solution. The Debyelength κ−1 for this solution, calculated from Eq.(3), wasfound to be 6.9 nm.Fig. 4clearly shows evidence of double-layer interaction, which is attractive on untreated polystyrenesurface and repulsive on all surfaces modified with UV–ozonetreatment.

A summary of the surface potentials best-fitted to theforce–distance data obtained from all UV–ozone mod-ified surfaces is shown inFig. 5. The magnitude ofthe fitted potential increases with increasing UV–ozonetreatment time and, hence surface oxygen concentration.For the tip voltage of−120 mV, the experimental datais described reasonably well by the sphere–plane modelunder the assumption of constant potential. Note, the mea-sured surface roughness of the samples used through-out these experiments is small compared to the Debyelength.

F(

ig. 4. Experimental force vs. distance data (©) fitted with theoretical curves cac) after 60 s and (d) after 180 s. (See text for discussions.)

lculated from Eq.(2): (a) untreated PS surface, (b) after 20 s UVO treatment,


Fig. 5. Surface potentials used to fit interaction forces between gold-coatedtip (VT = ψT = −120 mV) and UV–ozone treated polystyrene sampleswith different surface oxygen concentrations in 1%Dulbecco’s phosphatebuffered saline at pH 7.2.

3.2. Electrostatic field mapping of chemically patternedPS surface

Fig. 6 shows a stacked plot of four interleaved measure-ments representing different tip heights obtained from chem-ically patterned PS.

As can be seen in the plot, the cantilever deflections arepurely repulsive when the tip is over treated (polar) regions ofthe PS surface. Minimal deflection of the tip is observed overuntreated regions. As discussed previously, these deflectionsare believed to be due solely to electrostatic forces and thusthis plot presents a powerful means to visualise the distribu-tion of the electrostatic double layer forces on a chemicallypatterned surface.

These long-range (∼30–40 nm) electrostatic forces occur-ring in electrolytic solution will obviously have a majorimpact on colloidal interactions, such as protein adsorption,involving that surface. All proteins contain at least two ion-isable groups i.e. the amino and carboxyl groups. At a pH

Fig. 6. Plots of cantilever deflection over chemically patterned PS as a func-tion of tip-surface separation. Cantilever repulsions are clearly seen overtreated regions of the sample, however, there is no deflection detected overuntreated zones.

above the isoelectric point (pI) a protein in solution willpossess a net negative charge due to the deprotonation ofcarboxylic acid (COOH) groups. As mentioned above, thepH used in these experiments, i.e. 7.2, is above the iso-electric point of human serum albumin at 4.6[24]. Thiswould lead to repulsion between the similarly charged proteinmolecule and the UV–ozone treated areas of the polystyrenesurface.

3.3. Imaging of chemically patterned polystyrene surface

To study albumin attachment onto chemically patternedsurfaces we patterned the PS to produce an array of 5�m× 5�m oxidised squares within an unoxidized grid.Fig.7 shows a topographical image obtained from this surfacealongside a friction force image, obtained simultaneously.The mean roughness across the whole pattern is still0.4–0.5 nm with no grid pattern visible. This clearly showsthat the resulting surface is topographically homogeneous

F S surfa rii cale ba

ig. 7. (a) Homogeneous topography image of oxidatively patterned Pmage showing 5�m × 5�m oxidized regions separated by 5�m bars. (S

ce (features produced during pressing visible as diagonal lines). (b) Fction forcer is 10�m.)


Fig. 8. (a) Topography image showing differing amounts of albumin adsorbed by polar and non-polar domains of the pattern. (b) A typical friction forceimageof the albumin adsorbed surface. (c) Line-scans from pre-adsorbed surface showing homogeneous topography across the chemical pattern (dot-dashedline)and indicating that the albumin layer on the non-polar areas is 8 nm thicker than that on the polar for the adsorbed surface (solid line).

but that the chemical pattern is clearly resolvable using AFMfriction-force imaging.

Friction-force images of this surface clearly reveal thechemical pattern with the 5�m × 5�m oxidized domains,which have an oxygen level of 7 at.% showing as lighter-coloured, higher friction, regions compared to the 5�m gridbars which have no oxygen.

3.4. Imaging of protein layer on chemically patternedsurface

Adsorption of protein onto the chemically patterned sur-face leads to the AFM images shown inFig. 8(a) and (b).In contrast to the images produced before protein adsorptionthe topographical image (Fig. 8(a)) shows that the unoxi-dized domains now have greater relief than the oxidized dueto the adsorption of a thicker layer of albumin by the for-mer. AFM line-scans across this patterned surface, (Fig. 8(c)),indicate this difference in thickness is approximately 8 nm.The friction force image from this surface (Fig. 8(b)) indicatesthis surface to be effectively homogeneous with no contrastobserved.

The thickness of the adsorbed protein on the oxidized andunoxidized regions of the polystyrene surface was determinedby the AFM tip scratch method[17]. This revealed the thick-ness on the unoxidized polystyrene to be in the range 9–10 nma ncesi thod

may also reflect differences in molecule orientation and filmstructure on the regions of different surface chemistry.

Not surprisingly, the electrostatic forces between the simi-larly charged protein molecules and polystyrene surface leadsto a reduction in the amount of adsorbed protein. Soltys-Robitaille et al. investigated protein fouling of contact lensesand found that, at the physiological pH, positively chargedlysozyme was detected on anionic contact lens surfaces butwas not detected on uncharged and positively charged sur-faces. Conversely, HSA, which possessed a negative charge,was only detected on the surface of positively charged contactlenses and not on non-ionic and negatively charged surfaces[25]. Ladam et al. found that negatively charged proteinmolecules do adsorb onto surfaces with the same sign and canform dense layers up to a monolayer coverage[24]. The thick-ness of the adsorbed protein layer was found to be stronglydependent on the ionic concentration of the solution withadsorbed thickness increasing through a maximum with NaClconcentration before decreasing.

4. Conclusions

In this work, we investigated the influence of electrostaticforces on protein adsorption onto polystyrene surfacesoxidized by UV–ozone. Electrical double-layer interactionsb izedt PBS

nd 1–2 nm on the UV–ozone treated areas. The differen thickness of albumin measured by the tip scratch me

etween an AFM tip and polystyrene surfaces, oxido different oxygen concentrations, were measured in


solution using atomic force microscopy, with the surfacepotential of gold-coated tip controlled by bipotentiostat.Experimental force–distance curves have been fitted usingthe sphere–plane model based on the Poisson-Boltzmanntheory. The constant potential assumption has been madein this model. Two-dimensional mapping of the electricaldouble-layer in PBS solution over a sample with non-uniform surface charge distribution was achieved using theAFM interleave mode with a biased, gold-coated tip.

The UV–ozone/masking technique is shown to be capableof producing 5�m chemical features on smooth PS surfaceswithout introducing topographical heterogeneity. However,chemical heterogeneity is characterized by higher AFM-measured friction at oxidized domains of PS making thechemical pattern resolvable. Adsorption of albumin resultsin thicker layer on unoxidized (non-polar) regions of thispattern in relation to the oxidized regions. This is clear fromthe topography image of this surface and the associated line-scans. The results provide visual evidence that non-polar PSsurfaces adsorb more albumin than polar PS, which is con-sistent with the finding of several other workers[26,27]. Wehave shown that the electrostatic double force, introduced inan electrolyte near a charged surface, is an important driv-ing factor (attractive or repulsive) for the process of proteinadsorption onto a surface. The process of the spatial adsorp-tion of proteins onto surfaces may be precisely controlled byt facea

R

Res.

med.

[3] J.G. Steele, B.A. Dalton, G. Johnson, P.A. Underwood, Biomaterials16 (1995) 1057.

[4] D.O.H. Teare, N. Emmison, C. Ton-That, R.H. Bradley, Langmuir 16(2000) 2818.

[5] D.O.H. Teare, N. Emmison, C. Ton-That, R.H. Bradley, J. Colloid Inter-face Sci. 234 (2001) 84.

[6] S.A. Mitchell, N. Emmison, A.G. Shard, Surf. Interface Anal. 33 (2002)742.

[7] S.A. Mitchell, A.H.C. Poulsonn, M.R. Davidson, N. Emmison, A.G.Shard, R.H. Bradley, Biomaterials 25 (2004) 4086.

[8] S.W. Hamilton, A.J. Johnstone, R.H. Bradley, European Cells andMaterials (December (Suppl. 2)) 2003.

[9] M.M. Browne, G.V. Lubarsky, M.R. Davidson, R.H. Bradley, SurfaceSci. 553 (2004) 155.

[10] G. Beamson, D. Briggs, High Resolution XPS of Organic Polymers,John Wiley and Sons, Chichester UK, 1992.

[11] R.H. Bradley, I. Mathieson, J. Colloid Interface Sci. 194 (1997)338.

[12] D.O.H. Teare, N. Emmison, C. Ton-That, R.H. Bradley, Surf. InterfaceAnal. 29 (2000) 276.

[13] W.A. Ducker, T.J. Senden, R.M. Pashley, Langmuir 8 (1992) 1831–1836.

[16] A. Noy, C.D. Frisbie, L.F. Rozsnyai, M.S. Wringhton, C.M. Lieber, J.Am. Chem. Soc. 117 (1995) 7943.

[17] C. Ton-That, A.G. Shard, R.H. Bradley, Langmuir 16 (5) (2000)2281.

[18] J.N. Israelachvili, Intermolecular and Surface Forces, Academic Press,London, 1985, p. 144.

[19] GV. Lubarsky, MR. Davidson, RH. Bradley, Surface Sci. 558 (2004)135–144.

[20] G.V. Lubarsky, M.R. Davidson, R.H. Bradley, J. Phys. D: Appl. Phys.,

[ 1966)

[[[ inier,

[ be

[ ite-

[

he design of the electrostatic field profile near the surnd by the adjustment of the protein isoelectric point.

eferences

[1] S.D. Johnson, J.M. Anderson, R.E. Marchant, J. Biomed. Mater.26 (1992) 915.

[2] J.G. Steele, B.A. Dalton, G. Johnson, P.A. Underwood, J. BioMater. Res. 27 (1993) 927.

submitted for publication.21] R. Hogg, T.W. Healy, D.W. Fuerstenau, Trans. Faraday Soc. 62 (

1638.22] J. Gregory, J. Colloid. Interface Sci. 51 (1975) 44.23] M. Hull, J.A. Kitchener, Trans. Faraday Soc. 65 (1969) 3093.24] G. Ladam, P. Schaaf, G. Decher, J.-C. Voegel, F.J.G. Cuis

Biomolec. Eng. 19 (2002) 273.25] C.E. Soltys-Robitaille, D.M. Ammon Jr., P.L. Valint Jr., G.L. Gro

III, Biomaterials 22 (2001) 3257.26] E. Ostuni, B.A. Grzybowski, M. Mrksich, C.S. Roberts, G.M. Wh

sides, Langmuir 19 (2003) 1861.27] N.B. Sheller, S. Petrash, M.D. Foster, Langmuir 14 (1998) 4535.

the influence of electrostatic forces on protein adsorption

Documents