Creating gradients of two proteins by differential passive adsorption onto a PEG-density gradient.

Abstract

Many fundamental biological processes, including early embryo development, immune responses and the progression of pathogens, are mediated by gradients of biological molecules. Understanding these vital physiological processes requires the development of biomaterial platforms that can mimic them in-vitro. Such platforms include laboratory generated surface gradients of biological molecules. In this work, we report a method for the generation of surface gradients of two proteins. We used a surface grafting density gradient of polyethylene glycol (PEG) to control protein adsorption. In addition, we used protein size as a tool to control the position and the adsorbed amount of both proteins. To demonstrate our concept, we used fibrinogen as an example of a large protein and lysozyme as an example of a small protein. However, we speculate that the same strategy could be extended to any other pair of large and small proteins. We used X-ray photoelectron spectroscopy and sessile drop contact angle measurements to determine the chemical composition and wettability of the gradients. Protein adsorption was studied by surface plasmon resonance imaging.

Full text

Page 1

Creating gradients of two proteins by differential passive adsorption
onto a PEG-density gradient
Krasimir Vasileva,*, Agnieszka Mierczynskab, Andrew L. Hookc,1, Joseph Chanb,
Nicolas H. Voelckerc, Rob D. Shorta
aMawson Institute, University of South Australia, Mawson Lakes, SA 5095, Adelaide, Australia
bIan Wark Research Institute, University of South Australia, Mawson Lakes, SA 5095, Adelaide, Australia
cSchool of Chemistry, Physics and Earth Sciences, Flinders University, Adelaide 5001, South Australia, Australia
a r t i c l e i n f o
Article history:
Received 12 August 2009
Accepted 15 September 2009
Available online 6 October 2009
Keywords:
PEG gradient
Protein gradients
Protein adsorption
Plasma deposition
Surface grafting
a b s t r a c t
Many fundamental biological processes, including early embryo development, immune responses and
the progression of pathogens, are mediated by gradients of biological molecules. Understanding these
vital physiological processes requires the development of biomaterial platforms that can mimic them in-
vitro. Such platforms include laboratory generated surface gradients of biological molecules. In this work,
we report a method for the generation of surface gradients of two proteins. We used a surface grafting
density gradient of polyethylene glycol (PEG) to control protein adsorption. In addition, we used protein
size as a tool to control the position and the adsorbed amount of both proteins. To demonstrate our
concept, we used fibrinogen as an example of a large protein and lysozyme as an example of a small
protein. However, we speculate that the same strategy could be extended to any other pair of large and
small proteins. We used X-ray photoelectron spectroscopy and sessile drop contact angle measurements
to determine the chemical composition and wettability of the gradients. Protein adsorption was studied
by surface plasmon resonance imaging.
? 2009 Elsevier Ltd. All rights reserved.
1. Introduction
Manyessential biological processes are mediated bygradients of
biological molecules. For example, chemotaxis plays a role in
diverse physiological processes, such as the recruitment of leuko-
cytes to sites of infection, trafficking of lymphocytes throughout the
human body and patterning of neuronal cells in the developing
nervous system [1–7]. Gradients of cytokines play a role in the
ability of the body to resist invading pathogenic microorganisms by
providing immune cells with the directional cues they need to
rapidly migrate to the infection site [2,8]. In embryonic develop-
ment, gradients induce phenomena such as proliferation [9],
differentiation [10], or migration [1,11,12]. In cancer metastasis,
migrating tumor cells utilize protein gradients to escape the orig-
inal tumor, invade new tissues and recruit endothelial cells to
create blood vessels to feed the new tumor site [8,13,14]. Gradients
of proteins such as vascular endothelial growth factor (VEGF) and
colony stimulating factor 1 have been implicated in controlling cell
motility and angiogenesis in metastasising cancer [8,13,15]. These
examples demonstrate that gradients abound in biological systems
and are at the heart of physiological and pathological processes,
many of which remain poorly understood. There is an acute need to
develop biomaterial platforms that simplify and facilitate the study
of biological gradients. This has encouraged an increased research
tempo towards material platforms displaying surface-bound
chemical gradients as a means of providing surfaces on which to
study cell adhesion, proliferation, differentiation, and motility
[16–18], or on which to fabricate gradients of biomolecules and
their synthetic mimics [19–24]. Initial studies on wettability
gradients, showing the effect of conformational changes in cell
adhesion mediating proteins [25–27] have been extended to the
fabrication of surface-bound gradients presenting a variety of
surface functional groups such as carboxyl [28] or amine groups
[29] to explore cell adhesion and proliferation. A recent example by
Wells et al. [30] demonstrates that the adhesion and shape of
embryonic stem cells can be controlled on gradients of carboxyl
groups. However, in real biological systems the situation is much
more complex and gradients of a single protein, which have been
the subject of studies to date, are too simplistic to mimic natural
physiological phenomena. It is likely that gradients of two (or
more) biomolecules synergistically control such processes. In this
* Corresponding author.
E-mail address: krasimir.vasilev@unisa.edu.au (K. Vasilev).
1Present address: Laboratory of Biophysics and Surface Analysis, School of
Pharmacy, University of Nottingham, NG7 2RD, UK.
Contents lists available at ScienceDirect
Biomaterials
journal homepage: www.elsevier.com/locate/biomaterials
0142-9612/$ – see front matter ? 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.biomaterials.2009.09.056
Biomaterials 31 (2010) 392–397

Page 2

work, we present an approach to generate a surface density
gradient of two proteins via differential passive adsorption onto
a polyethylene glycol (PEG) density gradient. For this ‘‘proof of
concept’’, we have chosen fibrinogen and lysozyme as model
proteins, but the guiding principle behind this work is generic and
works on the basis of protein size discrimination. Our rational for
selecting these model proteins is that fibrinogen is a relativelylarge
protein (Mw340,000 g/mol) and is of a similar molecular weight to
cell adhesion mediators such as fibronectin (Mw440,000 g/mol),
whilst lysozyme, a much smaller protein (Mw 14,600 g/mol), is
similar in molecular weight to typical chemokines [8]. To control
the surface density of these two proteins, we have employed
a surface-bound gradient of grafted PEG density that is used to
control the adsorption profiles of these two proteins.
2. Materials and methods
2.1. Materials
Octadiene, allylamine, fibrinogen, lysozyme, potassium sulphate, sodium cya-
noborohydride and phosphate buffered saline were purchased from Sigma–Aldrich.
PEG-dialdehyde (Mw3000 g/mol) was purchased from Rapp Polymere (Germany).
Surface plasmon resonance chips (SPRchips) were from GWC Technologies Inc., USA.
Polished silicon wafers (p-type boron doped, orientation <100>) were obtained
from Virginia Semiconductors (USA). High-purity ultra-pure Milli-Q grade
(18.2 MUcm) with a final filtering step through a 0.22 mm filter was used for all
washing procedures and solution preparations.
2.2. Fabrication of gradients of amine functional groups
Gradients of octadiene and allylamine were plasma-deposited onto SPRchips or
polished siliconwafers using method and apparatus as described previously [24,31].
The shape of the gradient was controlled by the rate at which the octadiene:allyl-
amine ratio was changed. The initial flow rate of octadiene was 10 sccm, which was
linearly reduced to 0 sccm between 3 and 10 mm positions. The flow rate of allyl-
amine was linearly increased from 0 sccm at 3 mm to 10 sccm at 10 mm. The plasma
was excited using a 13.56 MHz radiofrequency generator. The power of deposition
was 20 W and remained constant during the whole process of deposition.
2.3. PEG grafting
PEG-dialdehyde (Mw3000 g/mol) was grafted onto plasma treated surfaces as
previously described [32]. Briefly, PEG-dialdehyde (2 mg/ml) was dissolved in
phosphate buffer, pH 7.4 containing 0.6 M K2SO4. Freshly prepared gradients of
amine functional groups were immersed in the solution of PEG-dialdehyde. The
grafting reaction was carried out at 60?C overnight, followed by thorough washing
with Milli-Q water.
2.4. X-ray photoelectron spectroscopy (XPS)
The XPS spectra were recorded on a Kratos AXIS Ultra DLD spectrometer using
a monochromated Al Ka radiation source (hn¼1486.7 eV) operating at 15 kV and
10 mA. The elements present in the sample surface were identified from a survey
spectrum recorded over the energy range 0–1100 eV at a pass energy of 160 eV and
a resolution of 1.0 eV. The areas under the selected photoelectron peaks in the
spectrum were used to calculate the atomic percentages. High-resolution (0.1 eV)
spectra were then recorded for pertinent photoelectron peaks at a pass energy of
20 eV to identify the chemical state of each element. All the binding energies (BEs)
were referenced to the C1s neutral carbon peak at 285 eV, to compensate for the
effect of surface charging. The analysis area was 700 ?300 mm2. The processing and
curve-fitting of the high-resolution spectra was performed using CasaXPSTM
software.
2.5. Sessile drop contact angle
Static contact angle measurements were carried out using the sessile drop
technique on gradients formed on flat siliconwafers. A small droplet (approximately
100 mL) of high-purity Milli-Q water, (supplied by an Elga UHQ water system,
conductivity less than 1?108Scm?1, surface tension 72.8 mNm?125?C), was
deposited on a given position of the surface using a motorised syringe. The
silhouette of the droplet was captured and imaged with a progressive scan CCD
camera (JAI CV-M10BX, Japan). The contact angle was determined by drawing
a tangent close to the edge of the droplet – at the three-phase point (using Image J
software – Drop-Snake). Experiments were conducted at 22?C in a class-100
cleanroom.
2.6. Surface plasmon resonance imaging
Surface Plasmon Resonance Imaging (SPRi) was done on a SPRImagerII from
GWC Technologies Inc., USA. Imaging was done over a circular area of 12 mm
diameter, which was exposed to 1 mg per ml protein solution in PBS buffer in a flow
cell. Two spots of approximately 1 mm2corresponding to each position of the
gradient were used to determine the change in the angle of plasmon resonance as
a result of a biomolecular adsorption event. The change in angle of resonance was
converted to layer thickness using Winspal software utilising a four layer model
composed of prism, gold layer, protein layerand buffer. A refractive indexof 1.45 was
used for the protein layer.
3. Results and discussion
PEG grafted with a sufficient surface graft-density is known to
resist protein adsorption [32,33]. However, if the surface grafting
density of PEG is not sufficient then proteins can penetrate the PEG
and adsorb to the underlying surface. In order to achieve this we
have adopted an approach that is schematically depicted in Fig. 1.
Initially, PEG was grafted to a gradient of amine functional groups
created via plasma co-polymerisation. This PEG gradient spanned
12 mm and featured a high PEG densityat the high ‘‘amine’’ density
end of the gradient, decreasing across the surface in a controlled
gradient manner, governed by the decreased availability of amine
binding sites. The PEG gradient was then used to control the
adsorption of the two proteins. Specifically, our strategy was first to
incubate the larger protein (fibrinogen) with the PEG gradient,
Fig. 1. Schematic representation of the experimental concept. In step 1 a density gradient of surface grafted PEG is created on a chemical gradient of amine functional groups. The
large protein is adsorbed in step 2 followed by adsorption of a small protein (step 3).
K. Vasilev et al. / Biomaterials 31 (2010) 392–397 393

Page 3

which we expected to adsorb at the low PEG density end of the
gradient. Importantly, the adsorbed fibrinogen passivates the
surface to the subsequent binding of lysozyme. Then the smaller
protein (lysozyme) was incubated with the surface; the binding of
this protein would occur only where there was a ‘‘window’’ for it to
adsorb, between the fibrinogen-coatedend of the surface (low PEG)
and the high PEG density end of the surface.
To create a gradient of PEG density, we first fabricated a gradient
of ‘‘amine’’ functionality by the plasma co-polymerisation of
octadiene and allylamine through a moving mask, as described
previously [24,31,34]. Polymer was deposited through a slot and
with the slot at the 0 mm position on the surface, we started a feed
of ‘‘neat’’ hydrocarbon, octadiene. As the slot moved, we gradually
exchanged the hydrocarbon monomer for the amine-containing
monomer, allylamine. From detailed XPS analysis of plasma poly-
mers of this monomer (Fig. 2), we believe these deposits contain
a portion of primary amines [35,36]. The gradients were formed
over a 12 mm distance on SPRchips or silicon wafers. Fig. 2 shows
the nitrogen to carbon (N/C) atomic ratio measured across the
surface. As is seen from this figure, the surface is hydrocarbon-rich
at the 1 mm position. The concentration of nitrogen gradually
increases and reaches a maximum at about 9 mm. The slight drop
in N/C on the 11 mm position may be related to a maximum ratio
between 0.10 and 0.12, above which further incorporation of
nitrogen cannot result in a stable film. We cannot exclude that the
XPS is probing an area of the sample beyond the plasma treated
zone, which would also result in a decreased nitrogen signal.
Although the X-ray spot size at FWHM is 300?700 mm2the actual
surface measured is larger.
Dialdehyde-terminated PEG chains (Mw¼3000 g/mol) were
immobilised to the plasma-polymerised allylamine gradient by
means of cloud-point grafting [32]. Successful fabrication of a PEG
gradientwasconfirmedbyhigh-resolutionXPS,usingtheintensityof
thecarbonC1ssignalat286.5 eV,whichcorrespondstotheC–Obond
of the PEG. Fig. 3a shows the growth in this signal across the surface
with the component at 286.5 eV increasing in the same direction as
the N/C ratio, across the surface from the 1 to 9 mm positions, at
the expense of the aliphatic component at 285 eV. This result is
consistent with an increasing density of surface-immobilised PEG.
Successful immobilisation validates the hypothesis that some of the
nitrogen seen in the plasma-polymerised allylamine corresponds to
primary amine [35]. Despite the low nitrogen signal (N/C) at the
1 mm position, we can still detect a significant C–O signal at this
position, corresponding to grafted PEG. The low nitrogen signal at
1 mm is indicative of allylamine diffusion under the mask. The
intensity of the C–O component arises from an amplification effect,
because if we assume that one PEG chain binds to one amine group,
the original nitrogen signal would be expected to be amplified 136
times,giventheaveragechainlengthofthePEG(andinXPSeachC–O
Fig. 2. N/C atomic ratio across the allylamine–octadiene plasma polymer gradient as
determined by XPS.
Fig. 3. a) Evolution of the XPS C1s peak along the allylamine–octadiene plasma
polymer gradient surface. b) C–O/C–C atomic ratios as determined from fitting of the
C1s XPS spectra.
K. Vasilev et al. / Biomaterials 31 (2010) 392–397 394

Page 4

is counted twice). As the PEG chains are bis-functionalised, it is
possiblethatbothendstethertothesurface,whichwouldchangethe
amplification effect.
In Fig.3b,wehaveplottedtheC–O/C–Cpeakarearatioacrossthe
PEG gradient. As anticipated, this ratio reaches maximum at 9 mm.
The slight decrease at 11 mm can be attributed to the effects dis-
cussed above. The C–O/C–H and XPS data from the ‘‘amine’’-end tell
us that, in comparison with Kingshott et al. [32], we have not
immobilised as thick a layer of PEG as can be achieved by use of
a dialdehyde-PEG on a plasma-polymerised n-heptylamine surface;
however, this is a comparison with an optimised procedure on
a surface of homogeneous chemistry. Our data at the ‘‘amine’’-end
compares better with the dialdehyde-PEG immobilised to an allyl-
amine plasma polymer to form a PEG brush layer, which has been
shown to be highly resistant to proteins including lysozyme [32].
We imaged the surface morphology of the PEG gradient at the 1,
3, 5, 7, 9, and 11 mm positions by atomic force microscopy (AFM).
Imaging was done in the dry state using tapping mode. Represen-
tative images are shown in Fig. 4. An ‘‘island’’ surface morphology
was observed on the low PEG density side of the gradient. These
islands have a typical valley-to-peak height of 11–12 nm (z direc-
tion), as shown in the Supplementary Information (SI) Fig. S1, and
vary in width (x–y) and the spacing between islands, as can be seen
from the scale bar in the top left hand side image (1 mm position).
As the PEG density increased, the valley-to-peak height of these
islands gradually decreased (Fig. S1), and concurrently the footprint
of these islands and the distances between them also decreased
(Fig. 4). In the region of highest PEG grafting density (9 mm and
11 mm positions), an almost smooth morphology is observed, with
the valley–peak height of less than 2 nm. This morphology is
consistent with a PEG brush layer. The root mean square roughness
(RMS) calculated from these imaged is shown in Fig. S2 (SI). RMS
decreases with increasing PEG density.
To the best of our knowledge, there have been no prior reports
on the surface morphology of dialdehyde-PEG immobilisation to
aminated surfaces. However, the dry-state images at the ‘‘amine’’-
end are consistent with the results of Cole et al. [37] who reported
fairly featureless monofunctional PEG (Mw5000) coated surfaces
with an average RMS of 0.5 nm, whilst the underlying allylamine
was reported to have an RMS of 0.3 nm.
In order to confirm that the observed surface morphologies at
lower ‘‘amine’’ density arise from the grafted PEG, we imaged the
‘‘naked’’ surface of an octadiene–allylamine chemical gradient,
prior to PEG grafting. Fig. S3 in the SI shows the topography and
phase images at 1, 5, and 10 mm positions, respectively. These
images reveal the surface of the octadiene–allylamine chemical
gradient to be very smooth. The phase images do not show any
distinguishable features, which suggest a smooth transition from
plasma-polymerisedhydrocarbon to allylamine without the
formation of islands or patches.
The AFM results should be discussed in the context of the cloud
point immobilisation. By definition, under the conditions of cloud
point, micelles are formed in the PEG solution. These micelles
comprise dialdehyde-functionalised PEG that is reactive towards
surface-bound amines. On the lower surface energy (hydrocarbon)
end of the gradient, once immobilised, we speculate that micelles
do not spread and tether through only a limited number of contact
points. We speculate that at this end we see aggregates of micelles,
whole micelles, or more likely, significant fragments of these
aggregates/micelles that are retained after the washing steps,
which would ordinarily (with monofunctional aldehydes) break up
the micelles leaving just immobilised individual chains. At higher
amine group density, the greater surface energy and density of
binding sites forces the micelles to spread on the surface. Micelles
break up or, conversely, smaller micelles fill the gaps between
already adsorbed micelles; the overall effect after washing is that
we see a more uniform and smoother surface morphology on the
higher ‘‘amine’’-end of the gradients. At the mid-point along the
gradient (5–7 mm positions) we do not see a variation in feature
sizes like those seen at the 1 mm position, for example.
We do not believe that there is any phase separation occurring
in the plasma polymer layer, to give regions that are, for example,
‘‘amine’’-rich, as the AFM images of the naked plasma-polymerised
gradient do not support surface phase separation. This suggests
there is sufficient mixing of the monomers at the molecular level
within the plasma [38].
Sessile drop water contact angle measurements across the
gradient are shown in Fig. 5. These confirm the suspected change in
wettability from one side of the gradient to the other prior to PEG
grafting (triangles). In Fig. 5, we also show the effective change in
Fig. 4. Atomic force microscopy images measured in the dry state at different positions along the allylamine–octadiene plasma polymer gradient. For all images x- and y-scales are
2 mm, z-scale is 25 nm.
K. Vasilev et al. / Biomaterials 31 (2010) 392–397 395

Page 5

surface wettability after PEG grafting (filled circles). To rule out any
possibility that this change arose from the influence of the grafting
buffer on the plasma polymer (rather than the immobilised PEG),
both gradients were treated identically prior to measuring the
contact angle, only omitting the PEG for one gradient (triangles).
Prior to PEG grafting, the water contact angle gradually decreased
from 80?on the ‘‘octadiene’’-end of the gradient (position 1 mm) to
68?on the ‘‘amine’’-rich part (position 11 mm). After PEG grafting,
the contact angle on the ‘‘octadiene’’-end decreased only by 4?,
whilst the contact angle decreased by almost 15?on the ‘‘amine’’-
end of the gradient. This increasing hysteresis is consistent with
a transition from a low PEG coverage zone, where the underlying
plasma polymer controls the wettability of the surface, to a PEG
brush regime, where the PEG layer controls the contact angle.
Finally, we have used the relative PEG density across the
gradient to control the adsorption of two different proteins, which
are almost an order of magnitude different in terms of their phys-
ical dimensions; fibrinogen has dimensions of 50–70? 450–500 Å
[39] and lysozyme can be considered as a 26? 45 Å ellipsoid with
an axial ratio of 1.73 [40].
Our rationale for the two-protein adsorption experiment was
based upon the premise that it would be significantly more difficult
to block the adsorption of lysozyme on the grafted PEG gradient
than fibrinogen. At low grafting density, we anticipated that
fibrinogenwouldadsorb, but as the grafteddialdehyde-PEG density
increased, adsorption would be resisted. This is borne out by the
data in Fig. 6, where SPR imaging (SPRi) was used to determine
the thickness of the adsorbed fibrinogen layer (rhomboids) along
the gradient, which decreases by almost 3-fold. SPRi enabled
measurements to be taken across the entire gradient simulta-
neously. The varied SPR response due to the variation in surface
chemistry across the gradient was considered for all measure-
ments. Adsorbed protein is presented as an adsorbed layer thick-
ness in Fig. 6.
Subsequent to the adsorption of fibrinogen and washing with
buffer, we challenged the gradient surface with lysozyme. Minimal
amounts of adsorbed lysozyme were measured on the low-density
PEG side of the gradients. The adsorbed lysozyme gradually
increased towards the central part of the gradient, reaching
a maximum at a position of 4 mm, but decreased where the grafted
PEG density was highest. The error bars support the view that this
is a reproducible phenomenon, and, without further data, we
explain this result in terms of the different effectiveness of
a particular PEG density to prevent different proteins from
adsorbing to the underlying plasma polymer layer.
We assume that the underlying plasma polymer gradient is
‘‘sticky’’ to both proteins. The PEG provides effective screening
for fibrinogen, even at low density, hence, the rapid reduction
in fibrinogen adsorption seen along the gradient in Fig. 6. Adsorbed
fibrinogen (in turn) screens the plasma-polymerised surface at the
low-density PEG end of the gradient, preventing subsequent lyso-
zyme adsorption, and at the high-density PEG end of the gradient,
the dense PEG brush effectively prevents protein adsorption. The
screening of lysozyme by the PEG brush grafted onto allylamine
plasma polymers is supported by ref. 32. In the central zone of
the gradient, we observe maximum lysozyme binding. This is the
only region of the plasma-polymerised gradient not effectively
screened; i.e. the lysozyme adsorption data are the composite
result of screening of the plasma polymer layer by both, the
fibrinogen (most effective at the low ‘‘amine’’-end of the gradient)
and PEG (most effective at the high ‘‘amine’’-end), which leaves
a ‘‘window’’ in the centre of the gradient.
An important control is to demonstrate that the adsorbed
fibrinogen can passivate the underlying surface against subsequent
lysozyme adsorption. This substantiates the claim that the lyso-
zyme is indeed only binding where a ‘‘window’’ of opportunity
exists, whereupon, the underlying plasma-polymerised surface is
not being effectively screened by the adsorbed fibrinogen layer or
where the PEG density is not yet sufficient to resist lysozyme
binding. To show that fibrinogen binding passivates a surface to the
binding of lysozyme, we conducted SPR measurements after
fibrinogen adsorption to a pure octadiene plasma polymer-coated
surface, which, after rinsing, was challenged by lysozyme. Kinetic
data as well as SPR angular scans are shown in Figs. S4–S6 in the
supporting information. These data show that when the surface is
first passivated with a layer of fibrinogen there is only a small
change in reflectivity (film thickness) when exposed to lysozyme,
as also seen on the low PEG side of the gradient.
4. Conclusions
In summary, we demonstrate a facile method for creating
surface density gradients of PEG by first generating a gradient of
amine functionality that is subsequently utilised for creating
a surface density gradient of grafted PEG-dialdehyde. We also
demonstrate that we can establish two different protein gradients,
Fig. 5. Sessile drop water contact angle measurements along the gradient surfaces.
Triangles show contact angles along the allylamine–octadiene gradient. Filled circles
show contact angles along the PEG gradient.
Fig. 6. The thickness of the adsorbed protein layer measured by SPRi for first fibrin-
ogen adsorption (rhomboids) and followed by lysozyme (triangles) as a function of the
position across the PEG gradient.
K. Vasilev et al. / Biomaterials 31 (2010) 392–397 396

Page 6

facilitated by the PEG-density gradient, by sequential adsorption of
first a large protein and second a small protein. In this work, we
used model proteins as a ‘‘proof of concept’’, however, the same
strategy can be easily extended to other proteins. An example
would be a gradient of a cell adhesion protein such as fibrinogen
(large protein, Mw 440,000 g/mol) and a typical chemokine
(Mw11,000–12,000 g/mol). We speculatethat suchgradients of two
or more proteins will provide a platform that mimics more closely
the complexity of real biological gradients, thus providing a tool for
researchers to address in a more meaningful manner fundamental
biological processes such as cell migration and locomotion.
Appendix
The full colour image can be found in the online version, at doi:
10.1016/j.biomaterials.2009.09.056.
Appendix. Supplementary information
Supplementary data associated with this article can be found in
the online version, at doi:10.1016/j.biomaterials.2009.09.056.
References
[1] Bryant DM, Mostov KE. From cells to organs: building polarized tissue. Nat Rev
Mol Cell Biol 2008;9:887–901.
[2] Cattaruzza S, Perris R. Proteoglycan control of cell movement during wound
healing and cancer spreading. Matrix Biol 2005;24:400–17.
[3] Gurdon JB, Bourillot PY. Morphogen gradient interpretation. Nature 2001;
413:797–803.
[4] Lecuit T, Lenne PF. Cell surface mechanics and the control of cell shape, tissue
patterns and morphogenesis. Nat Rev Mol Cell Biol 2007;8:633–44.
[5] Luo LQ. Developmental neuroscience – two gradients are better than one.
Nature 2006;439:23–4.
[6] Rogulja D, Irvine KD. Regulation of cell proliferation by a morphogen gradient.
Cell 2005;123:449–61.
[7] Vasilyev A, Liu Y, Mudumana S, Mangos S, Lam PY, Majumdar A, et al. collective
cell migration drives morphogenesis of the kidney nephron. PLOS Biol
2009;7:101–14.
[8] Jin T, Xu XH, Hereld D. Chemotaxis, chemokine receptors and human disease.
Cytokine 2008;44:1–8.
[9] Raballo R, Rhee J, Lyn-Cook R, Leckman JF, Schwartz ML, Vaccarino FM. Basic
fibroblast growth factor (Fgf2) is necessary for cell proliferation and neuro-
genesis in the developing cerebral cortex. J Neurosci 2000;20:5012–23.
[10] Ashe HL, Briscoe J. The interpretation of morphogen gradients. Development
2006;133:385–94.
[11] Baum B, Settleman J, Quinlan MP. Transitions between epithelial and
mesenchymal states in development and disease. Semin Cell Dev Biol
2008;19:294–308.
[12] Zhu AJ, Scott MP. Incredible journey: how do developmental signals travel
through tissue? Genes Dev 2004;18:2985–97.
[13] Eccles SA. Targeting key steps in metastatic tumour progression. Curr Opin
Genet Dev 2005;15:77–86.
[14] Koizumi K, Hojo S, Akashi T, Yasumoto K, Saiki I. Chemokine receptors in
cancer metastasis and cancer cell-derived chemokines in host immune
response. Cancer Sci 2007;98:1652–8.
[15] Wang WG, Eddy R, Condeelis J. The cofilin pathway in breast cancer invasion
and metastasis. Nat Rev Cancer 2007;7:429–40.
[16] Genzer J,Bhat RR.Surface-bound
2008;24:2294–317.
[17] Keenan TM, Folch A. Biomolecular gradients in cell culture systems. Lab Chip
2008;8:34–57.
soft matter gradients.Langmuir
[18] Kim MS, Khang G, Lee HB. Gradient polymer surfaces for biomedical appli-
cations. Prog Polym Sci 2008;33:138–64.
[19] Arnold M, Hirschfeld-Warneken VC, Lohmuller T, Heil P, Blummel J, Cav-
alcanti-Adam EA, et al. Induction of cell polarization and migration by
a gradient of nanoscale variations in adhesive ligand spacing. Nano Lett
2008;8(7):2063–9.
[20] Dertinger SKW, Jiang XY, Li ZY, Murthy VN, Whitesides GM. Gradients of
substrate-bound laminin orient axonal specification of neurons. Proc Natl
Acad Sci U S A 2002;99:12542–7.
[21] Kang CE, Gemeinhart EJ, Gemeinhart RA. Cellular alignment by grafted
adhesion peptide surface density gradients. J Biomed Mater Res Part A
2004;71A:403–11.
[22] Maheshwari G, Brown G, Lauffenburger DA, Wells A, Griffith LG. Cell adhesion
and motility depend on nanoscale RGD clustering. J Cell Sci 2000;113:
1677–86.
[23] Plummer ST, Wang Q, Bohn PW, Stockton R, Schwartz MA. Electrochemically
derived gradients of the extracellular matrix protein fibronectin on gold.
Langmuir 2003;19:7528–36.
[24] Whittle JD, Barton D, Alexander MR, Short RD. A method for the deposition of
controllable chemical gradients. Chem Commun 2003;14:1766–7.
[25] Lee JH, Jung HW, Kang IK, Lee HB. Cell behavior on polymer surfaces with
different functional-groups. Biomaterials 1994;15:705–11.
[26] Lee JH, Lee HB. A wettability gradient as a tool to study protein adsorption
and cell-adhesion on polymer surfaces. J Biomat Sci – Polym E 1993;4:
467–81.
[27] Uedayukoshi T, Matsuda T. Cellular-responses on a wettability gradient surface
with continuous variations in surface compositions of carbonate and
hydroxyl-groups. Langmuir 1995;11:4135–40.
[28] Li B, Ma YX, Wang S, Moran PM. Influence of carboxyl group density on neuron
cell attachment anddifferentiation
outgrowth. Biomaterials 2005;26:4956–63.
[29] Zelzer M, Majani R, Bradley JW, Rose F, Davies MC, Alexander MR. Investiga-
tion of cell-surface interactions using chemical gradients formed from plasma
polymers. Biomaterials 2008;29:172–84.
[30] Wells N, Baxter MA, Turnbull JE, Murray P, Edgar D, Parry KL, et al. The
geometric control of E14 and R1 mouse embryonic stem cell pluripotency
by plasma polymer surface chemical gradients. Biomaterials 2009;30:
1066–70.
[31] Parry KL, Shard AG, Short RD, White RG, Whittle JD, Wright A. ARXPS char-
acterisation of plasma polymerised surface chemical gradients. Surf Interface
Anal 2006;38:1497–504.
[32] Kingshott P, McArthur S, Thissen H, Castner DG, Griesser HJ. Ultrasensitive
probing of the protein resistance of PEG surfaces by secondary ion mass
spectrometry. Biomaterials 2002;23:4775–85.
[33] Pasche S, De Paul SM, Voros J, Spencer ND, Textor M. Poly(L-lysine)-graft-
poly(ethylene glycol) assembled monolayers on niobium oxide surfaces:
a quantitative study of the influence of polymer interfacial architecture on
resistance to protein adsorption by ToF-SIMS and in situ OWLS. Langmuir
2003;19:9216–25.
[34] Beck AJ, Jones FR, Short RD. Plasma copolymerization as a route to the fabri-
cation of new surfaces with controlled amounts of specific chemical func-
tionality. Polymer 1996;37:5537–9.
[35] Hook AL, Thissen H, Hayes JP, Voelcker NH. Spatially controlled electro-stim-
ulated DNA adsorption and desorption for biochip applications. Biosens Bio-
electron 2006;21:2137–45.
[36] Whittle JD, Short RD, Douglas CWI, Davies J. Differences in the aging of
allyl alcohol, acrylic acid, allylamine, and octa-1,7-diene plasma polymers
as studied by X-ray photoelectron spectroscopy. Chem Mater 2000;12:
2664–71.
[37] Cole MA, Voelcker NH, Thissen H. Electro-induced protein deposition on low-
fouling surfaces. Smart Mater Struct 2007;16:2222–8.
[38] Beck AJ, Jones FR, Short RD. Mass spectrometric study of the radiofrequency-
induced plasma polymerisation of styrene and propenoic acid. J Chem Soc,
Faraday Trans 1998;94:559–65.
[39] Hall CE, Slayter HS. The fibrinogen molecule: its size, shape, and mode of
polymerization. J Biophys Biochem Cytol 1959;5:11–7.
[40] Cheetham JC, Artymiuk PJ, Phillips DC. Refinement of an enzyme complex
with inhibitor bound at partial occupancy. Hen egg-white lysozyme and Tri-
N-acetylchitotriose at 1.75 Å resolution. J Mol Biol 1992;224:613–28.
behavior: gradient-guidedneurite
K. Vasilev et al. / Biomaterials 31 (2010) 392–397397