Journal of Adhesion Science and Technology

ISSN: 0169-4243 (Print) 1568-5616 (Online) Journal homepage: www.tandfonline.com/journals/tast20

Protein/Material Interfaces: Investigation on

Model Surfaces

Arnaud Ponche, Lydie Ploux & Karine Anselme

To cite this article: Arnaud Ponche, Lydie Ploux & Karine Anselme (2010) Protein/Material
Interfaces: Investigation on Model Surfaces, Journal of Adhesion Science and Technology,
24:13-14, 2141-2164, DOI: 10.1163/016942410X507966

To link to this article: https://doi.org/10.1163/016942410X507966

Published online: 02 Apr 2012.

Submit your article to this journal

Article views: 111

View related articles

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=tast20
 Journal of Adhesion Science and Technology 24 (2010) 2141–2164
brill.nl/jast

Protein/Material Interfaces: Investigation on Model Surfaces

Arnaud Ponche ∗ , Lydie Ploux and Karine Anselme

Institut de Science des Matériaux de Mulhouse (IS2M), LRC 7228, Université de Haute Alsace,
Mulhouse, France

Abstract
Adhesion of mammalian cells is mediated by a protein layer adsorbed from biological fluids or the extracel-
lular matrix. For bacteria, this ‘conditioning film’ also exerts an influence even if the membrane receptors
are not the same. In fact, in the early stage of adhesion, whatever the surface considered, cells and bacteria
are somehow in contact with proteins. These assumptions shed light on the role of protein/surface interface
in the early stage of cell or bacteria adhesion. In this article, we will first review some of the techniques
currently used for the analysis of such a bio-interface from a surface science point of view. Then, we will
focus on the possibility to chemically model such a bio-interface. To achieve this goal and simplify the
numerous interactions between a protein and cells, the biomolecule can be probed at different scales: amino
acid, peptide and protein as a whole. We will examine how the surface chemistry can help to graft these
fragments, what are the strategies to graft a huge molecule on a surface and what are the relevant parameters
to realize a biomimetic surface, taking into account the modifications undergone by the surface during the
sterilization step.
© Koninklijke Brill NV, Leiden, 2010

Keywords
Surface science, protein, peptide, amino-acid, grafting, cell adhesion, bacteria adhesion

List of abbreviations

APTES Aminopropyltriethoxysilane;
ATR-IR Attenuated Total Reflection/InfraRed Spectroscopy;
ATRP Atom Transfer Radical Polymerization;
BSA Bovine Serum Albumin;
DCC 1,3-Dicyclohexylcarbodiimide;
DFT Density Functional Theory;

*To whom correspondence should be addressed. Institut de Science des Matériaux de Mulhouse (IS2M),
CNRS LRC7228, 15 rue Jean Starcky, BP 2488, 68057 Mulhouse Cedex, France. Tel.: +33 389608800;
Fax: +33 389608799; e-mail: [email protected]

© Koninklijke Brill NV, Leiden, 2010 DOI:10.1163/016942410X507966

2142 A. Ponche et al. / Journal of Adhesion Science and Technology 24 (2010) 2141–2164

DNA DeoxyriboNucleic Acid;

EDC 1-Ethyl-3-(3-dimethylaminopropyl)carbodiimide;
ELISA Enzyme Linked Immuno-Sorbent Assay;
FT-IR Fourier Transform Infrared Spectroscopy;
HSA Human Serum Albumin;
L-DOPA L -3-(3,4-dihydroxyphenyl)alanine;

MALDI-MS Matrix-Assisted Laser Desorption/Ionisation Mass Spectrometry;

NHS N-HydroxySuccinimide;
OEG Oligo(Ethylene Glycol);
OWLS Optical Waveguide Light Spectroscopy;
PBS Phosphate Buffer Solution;
PEG Poly(Ethylene Glycol);
PLL Poly(L-Lysine);
PM-RAIRS Polarization Modulated/Reflection Absorption InfraRed
Spectroscopy;
RGD Peptidic Sequence: Arginine–Glycine–Aspartic Acid;
SAMDI-MS SAMs Assisted Matrix Desorption/Ionisation Mass Spectrometry;
SELDI-MS Surface Enhanced Laser Desorption/Ionisation Mass Spectrometry;
SDS-PAGE Sodium Dodecyl Sulfate/PolyAcrylamide Gel Electrophoresis;
SPR Surface Plasmon Resonance;
STM Scanning Tunneling Microscopy;
ToF-SIMS Time of Flight Secondary Ion Mass Spectrometry;
UHV Ultra High Vacuum;
XPS X-Ray Photoelectron Spectroscopy.

1. Introduction
In 1998, B. Kasemo pointed out that biological surface science will be an impor-
tant part of the future of surface science, as historic fields like semiconductors,
catalysis and materials science were in the past. Ten years after, with the need for
tailoring specific activities, for example in biosensors, biological surface science
has become an outstanding example of multidisciplinary science, at the frontier of
 A. Ponche et al. / Journal of Adhesion Science and Technology 24 (2010) 2141–2164 2143

physics for understanding of surfaces and interactions, chemistry to assemble mole-

cules of interest on a surface, and biology to understand cascade of chemical signals
transducing in cells after adhesion [1].
Adhesion or adsorption of proteins are relevant for many biological processes
and the understanding of such phenomena has become crucial, in the last years, for
the development of high-throughput screening and bio-sensing devices [2]. Indeed,
proteins are among the first molecules (after water) to reach the surface and adsorb
on it. The entire process of mammalian cell adhesion can be divided into several
major phases: cell attachment, cell spreading, actin filaments production (cytoskele-
tal organization) and formation of focal points. At the end of the fourth phase, the
link between cell and surface is secured and cells are able to resist high stripping
forces. The adhesion is mediated and regulated by membrane proteins with a main
familly being integrins. Integrins are heterodimers made of two extracellular chains
(α and β) and two tails which cross the membrane and allow signal transducing in
the intracellular area. For mammalian cells, 18 α-subunits and 8 β-subunits have
been registered giving 24 combinations and so 24 different receptors [3]. An ex-
tra domain called αA or αI divides the integrin familly into two groups depending
on its presence or not (Fig. 1). Such complex domain is responsible for the adhe-
sion of collagen or laminin or even cell–cell adhesion. As adhesive biomolecules,
integrins present also a very high selectivity toward structure of extracellular ma-
trix (ECM). α1 β1 integrins will bind to membrane type IV collagen, whereas α2 β1
family has higher affinity for fibrillar type I collagen [4, 5]. Up to now, subunits
such as α1 to α6 , αv , β1 , β3 , β5 are known to be involved in osteoblast adhesion [6].
Subunits of membrane proteins like integrins or proteoglycans will interact with a
peptide sequence present in adhesive proteins adsorbed at the surface of the sample
(RGD for fibronectin and vitronectin, YIGSR for laminin B1 or KRSR for heparin
binding domains). Each letter of the peptidic sequence stands for one of the twenty
amino acids and some sequences are summarized with their known function in Ta-
ble 1. The expression of each sequence on membrane protein will further trigger
cascade of biochemical events, controlling the future developments of cells [7]. As
for cells, when microorganisms attach to a surface, an interface between a complex
liquid medium and a solid surface is involved. At the early stage of the adhesion
mechanism, membrane receptors will be sensitive to specific fragments of protein
adsorbed. The substrate properties exert a strong influence on the kinetics of adsorp-
tion (polarity, hydrophilicity) and conformation will play a major role in adhesion.
A conditioning film is formed within seconds of exposure in a complex medium.
In liquid phase, this conditioning film has to be considered as the real surface seen
by bacteria [8]. Proteins at an interface can be seen as translators, they convert an
unknown surface chemistry in a layer understandable by the receptors of the cells.
Understanding of this language and initiation of appropriate interactions can induce
specific activities of cells [9]. The effect of the comprehension of such a chemical
language can be to inhibit interactions with the D-fragment of fibrinogen known to
favor inflammatory response [10, 11].
2144 A. Ponche et al. / Journal of Adhesion Science and Technology 24 (2010) 2141–2164

Figure 1. Conformation of adhesive domain αI of integrins showing center part with six β sheets
and on each side three helical structures. (Redrawn from 1QCY.pdb files (Protein Data Bank,
www.rcsb.org) with PyMol Molecular Graphics System (DeLano Scientific, San Carlos, CA USA,
2002 — www.pymol.org).)

Many strategies have been used to immobilize proteins on a solid surface. These
range from adhesion by using adhesive properties of specific proteins (like adhe-
sive proteins secreted by mussels), through development of adhesive layers known
to strongly interact with proteins, to the covalent grafting using an organic divalent
linker. The role of this last molecule is to initiate a covalent bond with the surface
and, at the other end, to react with free chemical groups of protein such as thiols or
primary amine. The last challenge to overcome for inducing a biological reaction
with a biomimetic surface is to keep the reactivity of the molecule of interest after
grafting or even adsorption. Many studies have shown evidence of protein denat-
uration upon adsorption and one can ask whether the biomolecule is still reactive
when its conformation is blocked by the covalent bonds with the surface or if its
mobility is reduced by the vicinity of the surface [12, 13]. In particular, structure
of the hydration layer impacts the conformation of the protein in the adsorbed state
and governs the activity of the protein after adsorption [14].
The purpose of this review is to summarize the major possibilities to realize
model surfaces at different levels for protein interactions studies. By focusing on
experimental characterization, it becomes obvious that bio-interfaces are at the fron-
 A. Ponche et al. / Journal of Adhesion Science and Technology 24 (2010) 2141–2164 2145

Table 1.
Peptide sequences known to promote adhesion

Sequence Amino acids Function References

RGD Arginine–Glycine–Aspartic Cell adhesion [93]

acid
YIGSR Tyrosine–Isoleukin–Glycine– Cell adhesion [7]
Serine–Arginine
KRSR Lysine–Arginine–Serine– Osteoblast adhesion [94]
Arginine
GRGDY Glycine–Arginine–Glycine– Fibroblast adhesion [88]
Aspartic acid–Tyrosine
GRGDSP Glycine–Arginine–Glycine– Cell adhesion (specific [88]
Aspartic acid–Serine–Proline interaction with fibronectin
receptors)
PHSRN Proline–Histidine–Serine– Human fibrosarcoma cells [48]
Arginine–Asparagine adhesion (synergetic with RGD)
Cyclo-DfKRG Cyclic–Aspartic acid–Lysine– Human osteoblast adhesion [98, 99]
Arginine–Glycine and differentiation
FHRRIKA Phenylalanine–Histidine– Osteoblastic mineralization [7]
Arginine–Arginine–Isoleucine–
Lysine–Alanine

tier between physical chemistry and biology. The technique able to probe quantita-
tively the chemistry of bio-interfaces in liquid conditions is still to be developed. For
the moment, informations are gathered by a compromise between a well-controlled
physical adhesion (often in ultra high vacuum conditions) and indirect informations
by techniques operated in liquid phase.
This review does not intend to be exhaustive but to show the most commonly
used spectroscopic techniques for biomaterials characterization as well as classical
chemical routes to graft a cell-binding peptide onto a metallic surface. We will
basically reduce our study to silicon and gold surfaces, the former being the most
interesting for future biosensor devices and the latter a historic model surface for
chemical modifications, self-assembly systems and adsorption studies.

2. Experimental Characterization of Model Surfaces

In his review on biological surface science, B. Kasemo has detailed the succes-
sive phases of adsorption when a native surface is implanted in the human body
[13]. Each of the different layers relevant for cell adhesion implicates biomolecules
and their effects have to be characterized in order to understand the complex phe-
nomenon of adhesion. Biomolecules of importance can be placed on a timescale
as follows: the first molecules reaching the surface are water molecules. Depend-
2146 A. Ponche et al. / Journal of Adhesion Science and Technology 24 (2010) 2141–2164

ing on the substrate, this water layer can create hydroxyl groups at the surface and
bind strongly resulting in a hydrophilic and dense water layer. The more hydropho-
bic substrates will result in a heterogeneous water layer and weaker binding of
water molecules. In a second stage, proteins will adsorb at the surface and the
conformation will depend on the previous hydration layer (denaturation or con-
formational changes). Finally, cells will sense the proteinaceous layer and adhere
with protein-like receptors. As we see, relevant parameters for biological adhesion
are concentration, orientation and conformation of proteins at the surface.
2.1. Spectroscopic Techniques
Among all physical techniques, spectroscopic techniques such as Fourier Transform
Infrared Spectroscopy (FT-IR), X-Ray Photoelectron Spectroscopy (XPS) or Time
of Flight Secondary Ions Mass Spectrometry (ToF-SIMS) are the most common
techniques usable to obtain chemical informations about biomolecules at a surface.
In the case of proteins, many of the «biological» techniques used for characteri-
zation operate in a liquid phase or with crystalline products for X-Ray Diffraction
structural determination. Due to the very low amount of protein in the vicinity or
in direct contact with the surface, compared to the concentration in liquid phase,
the surface signal falls in many cases below the detection limit and one has to
turn towards more sensitive techniques operating in Ultra High Vacuum like XPS
or ToF-SIMS. Such techniques have very low detection limit and different probe
depths. XPS signal emanates from the first 8–9 nm [15] and ToF-SIMS, the most
surface sensitive technique, can give information over a depth of only 1 nm depend-
ing on ion source used.
XPS has been widely used for the determination of the amount of proteins ad-
sorbed on surfaces [16–18]. It is a surface sensitive technique where intensities
of photoelectron peaks give quantitative information and binding energy is related
to chemical environment. This technique is very sensitive but suffers from being
operated in ultra high vacuum and is hardly sensitive to protein conformation or
organization at the surface. Nevertheless, it can give information about surface
coverage as detailed by Jedlicka et al. [19]. The authors decided to modify pep-
tides based on RGD groups of fibronectin or YIGSR peptide from laminin with an
alkoxysilane group. The modified peptide was then combined with a sol–gel solu-
tion and the thin film allowed to dry on a glass coverslip. XPS can give information
on the surface coverage of the silane modified peptides by looking at the attenuation
of the Si2p signal of silica substrate. On a silica substrate, peptide amount is limited
to only 10% of a monolayer (probably due to steric hindrance) and the thickness of
the peptide films is in the range 10–30 Å [19]. XPS can serve as a forensic analysis
technique for each step of successive chemical grafting due to its sensitivity and
is particularly suitable for oligonucleotides grafted on silica surfaces [20, 21] or
extracellular matrix remaining after cell detachment [22].
Techniques like Reflection Absorption Infrared Spectroscopy (RAIRS) comple-
ment very efficiently the XPS analysis in addressing molecular orientation, chem-
 A. Ponche et al. / Journal of Adhesion Science and Technology 24 (2010) 2141–2164 2147

ical binding and electronic structure but are more or less limited to gold surfaces
[23]. Gold surfaces have been extensively studied for self-assembled monolayers,
adsorption phenomena and grafting of biomolecules. A main reason can be the
possibility to use thiol group to create a covalent Au–S–R bond and also the possi-
bility to analyze it very easily using a conventional RAIRS system or Polarization
Modulated-RAIRS. The gold surface is a reference for many surface science studies
and Fourier Transform Infrared Spectroscopy (FT-IR) is a key technique for the sur-
face characterization of molecules whether they are adsorbed or grafted [24]. The
RAIRS characterization of gold surfaces is straightforward, but probing a mono-
layer on silicon is more difficult. Silicon surfaces are transparent to infrared incident
light at the grazing angle giving a very low reflection signal, in contrast to a gold
surface. To overcome these limitations, a multiple transmission/reflection method
has been shown to give very high sensitivity for N-hydroxysuccinimide (NHS)
functionalized monolayers on silicon [25]. Attenuated Total Reflection (ATR) and
RAIRS are sensitive to the ordering of SAMs. The peak positions of symmetric (νa )
and asymmetric (νas ) CH2 stretching modes are shifted by 5 cm−1 from crystalline
monolayers (νa : 2851 ± 1 cm−1 and νas : 2818 ± 1 cm−1 ) to disordered monolayers
(νa : 2855 ± 1 cm−1 and νas : 2822 ± 1 cm−1 ) [26].
Time of Flight Secondary Ions Mass Spectrometry is a surface science technique
operated in Ultra High Vacuum. It analyzes the mass of secondary ions emitted
from the surface under the effect of a focused primary ions gun. This technique
is only quantitative with caution but for protein layer analysis, the signal recorded
can give indirect information about protein conformation or orientation. It is in-
deed possible to follow the abundance of main ions arising from aminoacid such as
C2 NH6 + (m/z = 30) for arginine and alanine or C4 NH8 + for proline (m/z = 70).
These aminoacids are related to different domains of the protein molecule and the
ratio of abundances of these markers can give information on the sub-domain of
the fibronectin retained at the surface [27]. ToF-SIMS technique is very sensitive,
the detection limit is around 1013 species/cm2 (similar to XPS): for compari-
son, a full monolayer represents, depending on the size of the molecule, around
1015 species/cm2 (i.e., ∼1 nmole/cm2 ) [28]. The coverage for peptides or proteins
adsorption is often limited to a sub-monolayer. Such a low concentration limits the
number of techniques usable for quantitative analysis. Statistical analysis of signals
arising from amino acid composition of protein leads ToF-SIMS analysis to dis-
criminate between mixtures of proteins adsorbed on the same surface and can lead
to significant information about the competition among different biomolecules ad-
sorbed on a metallic surface [29, 30]. The structure of the film also has an influence
on the ToF-SIMS signature of the surface. If bridging occurs at the two ends of a
grafted peptide, some characteristic peaks of the free peptide can disappear or vary
in intensity [31]. ToF-SIMS technique has the best lateral resolution (a few tenths
of a nanometer on latest generation apparatus) and thus permit to obtain chemical
images that can unravel the influence of salts on adsorption of the Bovine Serum Al-
bumin on stainless steel [32]. Spectroscopic techniques such as XPS and ToF-SIMS
2148 A. Ponche et al. / Journal of Adhesion Science and Technology 24 (2010) 2141–2164

have the disadvantage that these operate under Ultra High Vacuum conditions, lead-
ing to a complete dehydration of protein films, which is far from being relevant for
biological ‘wet’ conditions. In particular, the real interface or hydration layer (the
first in contact with biomolecules when immersed in a complex medium) is difficult
to probe. Some attempts have been made with neutron reflectivity [33] showing that
a thiolated SAM with PEG end groups undergoes transition from helical structure
to an amorphous state depending on solvation of PEG molecules.
Mass spectrometry (MS) and all derivative techniques are very promising for
applications in biomedical field. MALDI-MS (Matrix Assisted Laser Desorp-
tion/Ionisation) and SELDI-MS (Surface Enhanced Laser Desorption/Ionisation)
are directly usable on solid surfaces, even if mass spectrometry and in particu-
lar MALDI-MS has been developed for liquid samples. With increasing sensitivity
and mass range, all these techniques have the main advantage to detect and iden-
tify proteins. The conventional technique used for MALDI is the so-called ‘dry
droplet method’ where an analyte and a matrix are dissolved in a solvent and al-
lowed to dry on a polished stainless steel plate. During evaporation of the solvent,
a co-crystallization occurs, and analyte molecules remain encapsulated within ma-
trix crystals. After laser desorption, the relative intensities of molecular ions as
a function of the mass/charge ratio differentiate lysozyme (M+ : m/z = 28 585),
HSA (M+ : m/z = 68 000), lactoferrin (M+ : m/z = 85 500) and even IgG (M+ :
m/z = 152 000) adsorbed on a carboxymethylated dextran surface [34]. Reviews
of all innovative techniques surrounding MALDI-MS have been written for bio-
medical applications [35] and characterization of SAMs by MALDI-MS [36, 37].
Recent developments make Laser Desorption Mass Spectroscopy a very promis-
ing technique for bio-interfaces characterization. Surface Enhanced Laser Desorp-
tion/Ionization is based on affinity technology. The principle is to characterize the
analyte present in solution by retaining it on a surface derivatized with subsequent
chemical probe, adding the matrix, drying and analyzing after laser desorption. The
types of probes are as large as SAMs, dextran, antibody- and protein affinity probes
and metal affinity or even hydrophobic interaction bio-probes [38]. When all these
probes are arranged in surface arrays, it reduces considerably the time of analysis
and contributes to the discovery of biomarkers of infection.
2.2. Biological Techniques
Biological characterization of materials in contact with cells or bacteria relies on
fluorescence microscopy. It consists mainly in reacting a fluorescence tag with the
functional group of interest. The illumination of the samples with an appropriate
radiation causes the tag to fluoresce and the microscope acquires a map of localiza-
tion of fluorescence tag, making the technique extremely suitable for micro-arrays
[39, 40]. The main drawback of this technique arises from the limited yield of reac-
tion between the fluorescence tag and the functional group located near the surface
or very close to other groups. Fluorescence markers are usually delocalized mole-
cules, often with aromatic rings. The steric hindrance is important and even if the
 A. Ponche et al. / Journal of Adhesion Science and Technology 24 (2010) 2141–2164 2149

yield is close to 100% in solution when the goal is to tag a protein or a peptide close
to a surface, it can dramatically affect the reaction when the surface concentration
is high. For this reason, a quantitative analysis of functional group of interest by
fluorescence techniques is difficult to use for biological surface engineering. Flu-
orescence spectroscopy can follow the natural fluorescence of tryptophan residue
(295 nm) and thus can be a probe for DNA hybridization, for example. The only
other amino acids presenting natural fluorescence are phenylalanine and tyrosine.
A fully quantitative method used in biology is radiolabelling: a radioactive iso-
tope is used as a tracer and integration of the nuclear signal is linked to the concen-
tration of tagged molecules. This technique requires specific equipments and has
often been used in parallel with other techniques to calibrate or verify the concen-
tration obtained [41, 42]. More common biological characterizations of proteins
involve enzyme-linked immunosorbent assay (ELISA) [43] and electrophoresis
(SDS-PAGE) [44] followed by Western blot immunoassay [45]. Such techniques
work in solution (as many biological techniques) but can be adapted to follow des-
orption of protein, or consumption of protein in the liquid phase when a surface is
immersed. Unfortunately, as the amounts of proteins adsorbed at a metallic surface
are fairly low compared to polymer surfaces, detachment with surfactant and direct
analysis in solution fall in many cases below the detection limit.

3. Biomimetic Materials
As seen in the previous section, analyzing a biological interface is not straightfor-
ward. Due to the chemical and structural complexity of such interface, mimicking
the real interface with molecules immobilized on a surface is the only way to de-
couple the various parameters involved in protein adhesion. The realization of such
a model system is based on the different levels at which a protein can be seen: from
the whole molecule to single constitutive parts of protein (amino-acid), including
peptide. These model interfaces can be fabricated either by grafting or adsorption
and depend strongly on the substrate used. The choice of the peptidic sequence to
be grafted on a surface depends on knowledge of adhesive protein. For example,
domains of fibronectin interacting with surface receptors or other proteins (heparin,
fibrinogen or collagen) are summarized in Fig. 2. Eukaryotic cell binding domain
(including the RGD fragment) is located at the central part of the molecule [46, 47],
whereas bacteria are known to interact with NH2 terminated heparin/fibrin binding
fragment [48].
3.1. Surface Modification for Covalent Grafting
For biotechnology applications, the modified surfaces need to establish specific
binding with biomolecules and reduce non-specific interactions. The establishment
of a covalent bond between a metallic substrate and an organic biomolecule needs
a coupling agent. This coupling agent, when dealing with bio-sensors or implants,
requires to be grafted with a high density to maximize reactivity with molecules
2150 A. Ponche et al. / Journal of Adhesion Science and Technology 24 (2010) 2141–2164

Figure 2. Structure of fibronectin showing the three repeat motifs (types I, II and III) and known
binding domain. RGD and PHSRN binding sequences are located at the center of the cell binding
domain.

of interest. In the case of chemical modification of a metallic surface with an or-

ganic linker, two strategies meet these requirements: self-assembled monolayers
and polymer brushes [49].
3.1.1. Self-Assembled Monolayers for Biological Applications
Even if SAMs have been known for a long time, the technology employing self-
assembled surfaces for biological applications has only been applied since the
beginning of 1990’s via the use of the model system involving thiols on gold. Via
the reaction between an organic molecule and the metallic substrate, it gives a high
density of alkyl groups as binding sites for further reactions [50]. Self-Assembled
Monolayers (SAMs) are usually formed by immersion of the substrate into an active
solution. The active molecule bears a head group which forms bond with substrate
atoms (silane for silicon substrate and thiol for gold), an alkyl chain –(CH2 )n – for
van der Waals interactions with neighboring chains and a surface functional group
–R (usually NH2 or COOH for bio-surfaces) [26]. The molecular assembly of thiol
molecules on a gold surface provides the opportunity to realize hybrid interfaces
on various metallic surfaces. Each type of metallic surface has found a reacting
chemical group to bind covalently with the surface. Preponderant reactive groups,
found in the literature, are thiols, silanes and phosphonates [51] for various metallic
surfaces. For biological applications, SAMs can be considered as a synthetic model
for membrane and thus can provide information on interaction between protein and
cells. It is worth noting that the property of thiols to form strong covalent bonds
with a gold surface has been used to graft peptides on gold without the use of a
linker. Indeed, cysteine is the only natural amino acid to bear a thiol terminated side
chain. As a consequence, a peptide with cysteine residue can be directly grafted on
a gold surface via cysteine side chain [23, 52].
The choice of the functional group –R of the molecule governs further properties
of the surface. For long, oligo-ethyleneglycol (OEG) terminated SAMs have proven
to be protein repellent in certain conditions. Love et al. have expressed the following
three characteristics of SAMs to be a model biological surface [53]:
 A. Ponche et al. / Journal of Adhesion Science and Technology 24 (2010) 2141–2164 2151

(1) Limit non-specific interactions;

(2) Tune density of biomolecules or composition of ligand;
(3) React with the ligands while maintaining their biological activities.
The first characteristic can be easily achieved by tailoring the end groups of the
SAMs to be oligo-ethylene glycol [54] or poly(ethylene glycol) (PEG) [55]. This
aspect will be further developed in the next section about ‘polymer brushes’. For
oligo-ethylene glycol ended SAMs, a minimum of two units of ethylene glycol
are necessary to obtain repellent property but longer size (n = 6) shows robust re-
pellent properties towards fibrinogen and lysozyme, irrespective of the solvent [54].
Ethylene glycol is not the only molecule giving repellent surface. In particular, N,N-
dimethyl amide functional group presents reduced affinity for protein. Alkanethiols
terminated with permethylated sorbitol and acetylpiperazine were even more re-
sistant to fibrinogen adsorption than (EG)n group [56]. A detailed, recent XPS
characterization of an ethylene glycol repellent layer can be found in [57]. It shows
that the layer is no more stable after 24 h incubation in PBS buffer at 37◦ C and
loses its repellent properties. The instability is due to the hydrolysis of the siloxane
bond which is only observed at a temperature of 37◦ C.
The second characteristic can be realized with mixed SAMs. For a gold surface,
when two thiol molecules with different end groups are co-reacted from solution,
the ratio of molecules immobilized onto the surface is related to the molar compo-
sition of the solution. It is then possible to tune the density of active ligands and
repellent end groups to limit non-specific adsorption in the mean time. Finally, the
use of mixed SAMs to realize oligonucleotide microarrays with biotinylated- and
oligoethylene-alkanethiols on a gold surface has been shown to immobilize high
amount of isolated streptavidin. The immobilization protocol keeps the biological
activity intact and allows to realize high selectivity sensors [24].
Self-assembled monolayers can as well be modified to bear a phosphorylated
analog of amino acids such as serine, threonine and tyrosine leading, after assem-
bly, to biomimetic surfaces [58]. Nevertheless, an important point is the general
tendency to idealize the structure of a so-called self-assembled monolayer. The lat-
eral dimension of the end groups can interfere with the auto-assembly processes
[53] resulting in a less ordered adlayer. For quantitative applications like biosen-
sors, manufacturers have to ensure that the accessibility of the reacting groups is
not affected by the process. The experimental conditions necessary to obtain a true
monolayer are very drastic and some silanes (usually short silanes) like amino-
propyltriethoxysilane (APTES or γ -APS) are known to give easily a mulitlayer
structure [59]. APTES is often used as a coupling agent for biomaterials appli-
cations as it yields a high density multilayer when adsorbed from liquid phase.
A better control of the structure can be obtained in vapor phase and, in this case,
molecularly thick films are observed [60].
2152 A. Ponche et al. / Journal of Adhesion Science and Technology 24 (2010) 2141–2164

3.1.2. Polymer Brushes

Polymer brushes architecture can be used as SAMs for maximizing and controlling
density of grafting. One of the main applications, for biomaterials, is the biofoul-
ing aspect of OEG or PEG. The difference between the two is related to the size
of the molecule. To adsorb on a hydrophilic surface, a protein needs to remove
water molecules from the immobilized hydration layer. Through conformational
changes, the protein gains enough energy to overcome the energy barrier of the hy-
dration layer. As removed water molecules turn to a liquid phase, entropy increases
and this phenomenon is thermodynamically favorable. PEG macromolecules have
strong interactions with water molecules, resulting in a thermodynamically stable
layer. The energy barrier is too high and conformational changes of proteins are not
sufficient to displace the hydration layer [61]. The biomolecule adsorption is turned
unfavorable and such surfaces are called ‘protein repellent’ or ‘protein resistant’.
Coating a surface by PEG or OEG layer is then a very efficient way to remove
all non-specific adhesion of proteins on a surface. Tuning the repellent properties of
OEG by mixing with hydrophobic SAMs can also tune the conformation of proteins
from helical to amorphous layers [62].
Experimentally, such polymer surfaces can be realized either by chemisorption
on the basis of reactive end-groups of SAMs in the most conventional strategy
or by physisorption involving block copolymers. Physisorption currently involves
copolymers such as poly(L-lysine)-graft-poly(ethylene glycol). PLL-g-PEGs inter-
act electrostatically (via the PLL positively charged block) with various negatively
charged oxide layers including TiO2 , SiO2 [63] and Nb2 O5 [64]. On such surfaces,
the resistance of HSA (Human Serum Albumin) and fibrinogen has proven to be ef-
fective over a time span of 30 h. In optimal conditions, a minimum of only 5 ng/cm2
of HSA and fibrinogen can be attained depending on the size of the PEG and the
graft ratio (typically 1–5 lysine residues for 1 PEG chain). The quantity on an
uncoated oxide surface was around 600 ng/cm2 (measured by Optical Waveguide
Lightmode Spectroscopy, OWLS) [63].
Nature-inspired coatings can also meet all the requirements for immobiliza-
tion of biomolecules. The composition of adhesive proteins secreted by mussels
is particularly rich in catechol L-DOPA (3,4-dihydroxy-L-phenylalanine) and ly-
sine. Researchers have anticipated the binding affinities of L-DOPA to realize PEG
modified analog and generate, by simple ‘dip and rinse’ protocols, protein-repellent
surfaces [65]. Lee et al. [66] realized an adherent polydopamine film by dip coat-
ing from a dopamine solution onto a wide range of organic and inorganic surfaces.
Polymerization of dopamine in controlled basic conditions (Tris Buffer, pH 8.5)
leads to the growth of an adhesive film on various substrates (including metals, ox-
ides, polymers and ceramics), the structure of which is closely related to melanin.
As for melanin, the exact chemical structure is not known but such a film presents
the opportunity to react with nucleophiles like primary amine groups of proteins to
realize a covalent immobilization [66]. A more recent article has shown that thick-
ness of dopamin films is proportional to the time of immersion if fresh dopamine
 A. Ponche et al. / Journal of Adhesion Science and Technology 24 (2010) 2141–2164 2153

solution is regularly provided [67]. The polydopamine thin film is a good candi-
date for further conjugation of biomolecules (peptide via NH2 groups, sugars via
SH groups) or for surface engineering like photolithography or electroless deposi-
tion of metals [68]. It should be mentioned, as well, the possibility to immobilize
an ATRP (Atom Transfer Radical Polymerization) precursor (usually brominated
molecules) on the surface and, on the basis of the precursor, initiate an atom trans-
fer radical polymerization. This could lead to protein-repellent polymer brushes
of poly(carboxybetaine) [69] or poly(sulfobetaine) [70]. Adsorption of human fib-
rinogen, lysozyme and Human Chorionic Gonadotropin is reduced to lower than
0.3 ng/cm2 (Surface Plasmon Resonance determination) with carboxybetaine. With
sulfobetaine polymer brushes, the low fouling behavior is evidenced by a very low
amount of fibrinogen of 0.3 ng/cm2 and amounts of lysozyme and bovine serum
albumin in the range 5–10 ng/cm2 .
Polymer brushes or SAMs are usually the initial step of a biomolecule grafting
strategy on silicon or more generally metallic surfaces. As such surfaces will be
in contact with cells and bacteria, it is important to estimate contamination and
evaluate the effect of sterilization step on SAMs or polymer layers.
3.1.3. Chemical Modification Induced by Sterilization or Cleaning Steps
XPS and other surface-specific spectroscopic techniques can give a strong insight
into carbonaceous species present at a surface. This layer is usually called organic
contamination and is formed in less than a second when a high surface energy
surface (metals) is in contact with the atmosphere. It is very difficult to remove
all contaminants from a surface even with strong detergents, solvents or even us-
ing UV and plasma-based techniques [71]. Cleaning processes usually modify the
layer by incorporating oxygenaceous species and tend to homogenize the conta-
mination rather than really removing it [72]. The sterilization step is particularly
critical when organic layer or polymer surfaces are used, leading to strong mod-
ification of the outermost surface [73, 74]. When a contaminated surface will be
in contact with a complex medium like serum, the equilibrium between initially
adsorbed contaminants and surface will be displaced, leading preferentially to ad-
sorption of biomolecules. The same competition occurs with pre-adsorbed proteins
like HSA or fibrinogen, when placed in contact with a solution of another protein.
Displacement of equilibrium is modulated by the affinity of each protein for the
surface. HSA or fibrinogen layers are easily displaced on glass whereas they are
more resistant when first adsorbed on polystyrene.
3.2. Amino Acids
Amino acids are the building blocks or monomeric units of proteins. On their own,
they can be energy metabolites, essential nutriments in animals, or play a bio-
chemical role as neurotransmitters for glycine [75]. For lightest amino acids and
depending on sublimation behavior, the study of adsorption in Ultra High Vac-
uum (UHV) leads to highly controlled deposition rate and conformation of isolated
molecules.
2154 A. Ponche et al. / Journal of Adhesion Science and Technology 24 (2010) 2141–2164

3.2.1. Physisorption
Smallest amino acids are easily evaporated in vacuum conditions using a Knudsen
cell operated at low temperature. This possibility induces a high control on the ad-
sorption of glycine or alanine on a polycrystalline surface. Such analysis provides a
further understanding of self-assembly processes and orientation of adsorbed mole-
cules using Scanning Tunneling Microscopy (STM) and DFT calculations [76].
Working in Ultra High Vacuum provides a good control on the adsorption (flow
rate, pressure and even orientation) and is an extremely clean process (absence of
solvents, salts or other biomolecules). But when the evaporation of the molecule
is not possible, the adsorption step must be realized in ‘wet’ conditions. In a way,
it is a step closer to biological conditions but it also suffers from possible conta-
mination of the surface for spectroscopic characterization and possible changes in
conformation during dehydration when exposed to UHV conditions.
Among all amino acids, glycine is the smallest and has been the most studied in
UHV over the last fifteen year. It has been deposited on a variety of well-controlled
crystallographic surfaces like Cu(1 0 0) [77] or Pt(1 1 1) [78]. The adsorption be-
havior of glycine in UHV conditions is controlled by the temperature. RAIRS study
shows that glycine mainly adsorbs in anionic form at room temperature on Cu(1 1 0)
and the plane of carboxylate ion is oriented perpendicular to the surface. S-Alanine
shows the same adsorption trends in anionic forms at room temperature but in con-
trast to glycine, the surface is not fully covered (only a third of surface atoms are
covered, i.e., 0.33 monolayer). A high coverage of alanine can be obtained at a tem-
perature of 420 K, with a more complex crystal phase and a saturated monolayer is
present in a (3 × 2) phase at 470 K as revealed by STM [79].
Physical adsorption and desorption behaviors of cysteine or cystine on Au(1 1 1)
have also been analyzed in light of the possibility of thiol groups to interact with a
gold surface. In this case, voltammetry experiments can give a strong insight into
the surface structure [80]. Cysteine interacts with divalent cations like Cu2+ and
can be used to realize multilayers by electrostatic interactions [81]. Organization
of the cysteine molecules on the surface can only be unraveled using synchrotron
radiation for XPS measurements because of the low deposition of L-cysteine in
UHV [82]. When adsorption occurs from the liquid phase, the structure of the final
film can be compared to alkanethiol SAMs structure [83].

3.2.2. Chemisorption
Reaction of thiols with gold atoms can be used to realize a direct interface between
cysteine (or all cysteine containing peptides) and gold surface. These L-cysteine
layers present interesting electronic properties [84] and are used to modify gold
electrode to induce specific complexation of Cu2+ ions [85]. Such chiral films,
grown with amino acid chiral enantiomers can induce a discriminative crystalliza-
tion of enantiomers from racemic solution [86].
 A. Ponche et al. / Journal of Adhesion Science and Technology 24 (2010) 2141–2164 2155

3.3. Peptides
A sequence of amino acids is called a peptide and grafting a peptide (in place of
the whole protein) can induce specific interaction with membrane receptors of cells
and bacteria. RGD sequence, recognized by eukaryotic cell surface receptors, has
been found in numerous adhesive proteins of extracellular matrices (fibronectin,
vitronectin, osteopontin, collagen, fibrinogen) [87]. Biomimicking surfaces with
RGD peptides is less effective in cell attachment assay than fibronectin itself. But,
use of short peptides has the advantage to increase selectivity. Thus, GRGDSP short
peptide becomes more specific towards fibronectin receptor [88].
3.3.1. Adsorption
As seen earlier, interfaces between metals, insulator or semiconductors and bio-
molecules are of importance for further development of biosensors. A systematic
study [89] on nine surfaces, three solvents and homopeptides based on the twenty
amino acids has shown that adsorption was maximized for polar and charged ho-
mopeptides on Si3 N4 or SiO2 surfaces. This adsorption behavior is strongly de-
pendent on the pH of the peptide solution and roughly, maximum peptide adhesion
density is obtained for pH < pKa of the free amino acid. Concentration of the pep-
tide plays also a major role and adsorption is very limited when the concentration
is lower than 0.01 mM [89].
In the biosensor field, semiconductor surfaces are of high interest and the adsorp-
tion of biomolecules on such surfaces is a very recent field of research. Some studies
have analyzed the adsorption behavior of peptides and have compared to the highly
ordered SAMs obtained with alkanethiol. It is found that the amounts adsorbed are
fairly low and cationic peptides interact more strongly with the surface of InP(1 0 0)
[90].
We have seen that peptides can be considered as part of a protein. In order to
find active adhesive site of a protein toward a surface, this protein can be digested
with a trypsin into peptides fragments. The adhesion behavior of isolated fragments
can then be probed toward the surface. By this method, it is shown that regions of
the β-lactoglobulin rich in glutamine and aspartic acid residues are involved in the
adhesion process. By varying the pH, the authors show that electrostatic interactions
alone cannot explain the adhesive behavior of the peptides considered. As a matter
of fact, adhesion of the peptides of interest is irreversible at pH 3.3 where carboxylic
groups of glutamine and aspartic acid are non-ionized [91].
3.3.2. Top-Down Grafting
Top-down grafting is an approach where isolated biomolecules are synthesized and
derivatized with suitable head groups which can further react with a surface. These
can be a thiol group of a cysteine residue to bind to a gold surface or a peptide
modified by a silane to bind to a silicon surface [92]. The complete grafting protocol
is as follows: oxidation/cleaning of the surface, APTES immobilization, peptide
coupling to amine groups of the APTES layer with carbodiimide chemistry (EDC
or DCC coupling agents activate carboxylic groups of peptide). Via this method,
2156 A. Ponche et al. / Journal of Adhesion Science and Technology 24 (2010) 2141–2164

Figure 3. Principal chemical routes for peptide immobilization on SAMs and polymer brushes.

one can immobilize approximately 80 pmol of short peptide RGDS or KRSR per
square centimeter on glass surfaces [93, 94]. The usual chemical routes to graft
a peptide on a self-assembled layer are summarized in Fig. 3 with examples on
NH2 - and COOH-terminated SAMs and use of linkers like glutaraldehyde and N-
succinimidyl-3-maleimidylpropionate.
Even if all chemical steps are well controlled, one needs to take care of the side
reactions which can take place in a complex structure. For example, if we consider
the use of a bifunctional linker with a maleimide at one end and succinimidyl es-
ter at the other (Fig. 3), the latter group reacts with primary amine of the SAMs
and the maleimide group is then free to react with a thiolated peptide. This leads
to covalent grafting of a peptide usually by a cysteine residue (containing a –SH
group). In an ideal view and without taking side reactions into account, all succin-
imidyl groups will react with amine groups of the substrate and maleimide groups
are oriented towards the air. But if we consider that maleimide groups can react
with amine with a limited yield, this can lead to succinimidyl oriented away from
the surface. Free amine of the peptide grafted by the first way reacts easily with
 A. Ponche et al. / Journal of Adhesion Science and Technology 24 (2010) 2141–2164 2157

succinimidyl groups and create a bridging of the peptide along the surface as evi-
denced by RAIRS [31]. This kind of immobilization of peptide along the surface
will annihilate all bioactivity and block recognition sites. This emphasizes the im-
portance of a careful control of the chemistry as well as the structure for biological
applications [95]. A beautiful example of monitoring of all these reaction steps by
XPS and evaluation of yield of surface reaction can be found in [96]. The results
obtained by XPS and radiolabelling converge to give a value of maleimide modified
APTES surface of 2.2 ± 0.5 groups/nm2 whereas it falls to 0.2–0.4 group/nm2 when
the maleimide is reacted with peptides. Density of peptides in top-down grafting is
limited by the steric hindrance of the peptide. As an example, the yield of reaction
(estimated by XPS) of cysteine on maleimide surface is around 75–80% depending
on the type of maleimide reactant. For larger peptide, it is constrained within the
range of 15–20%.
Some specifically designed peptides can be used for supramolecular architecture.
In the precursor AMBBA-Cys-Glu-Cys-Glu, the two cysteine residues are expected
to attach and orient the peptide horizontally on the gold surface. Such implementa-
tion is important to obtain a specific organization and orient the glutamine so that
it can give strong interaction with BSA or Chymotrypsin [97]. Up to now, most
of the short peptides investigated had a linear geometry. Cyclopeptides, based on
RGD motifs, have also been immobilized on hydroxyapatite and titanium surfaces.
Cyclo-DfKRG tends to favor short time attachment compared to RGD linear se-
quence on hydroxyapatite surfaces but also induces modulated cell differentiation
according to the anchor used (thiol or phosphonate) [98, 99].
3.3.3. Bottom-up Grafting or Solid Phase Synthesis
As mentioned earlier, when the peptide to be grafted has a large cross section, it lim-
its intrachain interactions between molecules constituting the SAMs and induces a
disorder. To achieve a better coverage and a good ordering, there exists another way
of grafting called ‘Bottom-up grafting’ or in situ biomolecule synthesis. Usually,
peptides are synthesized following the Merrifield reaction which combines each
amino acid step by step by creating a peptidic bond between the carboxylic acid of
the first amino acid and the amino group of the second. Protected amino acids are
used to avoid side reactions. The solid support is generally made of polymer beads
or hydrogels. The first step is to introduce a cleavable linker grafted on the poly-
mer. The peptide is assembled on this linker by a succession of coupling reactions
and de-protections according to the primary structure. Finally, by a cleaving step,
the peptide can be released in the liquid phase. Even if recent progress in synthe-
sis has succeeded to reach an yield of 98% for each coupling step, introduction of
20 amino acids in the structure leads to an overall yield of 0.9820 , i.e., 67%. The
synthesis of a complete protein with thousands of amino acids is, via this method,
hardly conceivable.
Up to now, this synthesis has been limited to solid surfaces and small peptides.
The use of peptide synthesizer can, therefore, increase the productivity. The proto-
col maintains the same steps except that there is no need for insertion of a cleavable
2158 A. Ponche et al. / Journal of Adhesion Science and Technology 24 (2010) 2141–2164

group in the peptide chain as we want the peptide to be firmly grafted onto the sur-
face. A recent paper has reported the use of microwave assisted peptide synthesizer
for the synthesis of a 10-unit peptide on a silicon surface [95].
3.4. Proteins
Proteins are the most complex biomolecules due to their size (up to 3000 kDa) and
also due to the multiple interactions a protein can establish to adsorb onto a surface.
In liquid, a protein exists in the native state but can adopt multiple conformations
when adsorbed on a surface.
3.4.1. Adsorption and Denaturation
From an entropical point of view, hydrophobic interactions between a surface and
a protein induce a disorganization of the water layer that is thermodynamically fa-
vorable. On the contrary, the displacement of water molecules from a hydrophilic
surface needs more energy and this explains the lower amounts of proteins ad-
sorbed on hydrophilic surfaces compared to hydrophobic surfaces. This is not to
say that adsorption does not occur on a hydrophilic surface, in fact it implies other
mechanisms like charge interactions or protein conformation changes providing the
required amount of energy to displace water molecules [12]. Structural changes or
loss of secondary structure called denaturation will result in a stronger attachment
of protein and will depend on the time of residence. The energy gap between na-
tive and denaturated states has been measured by calorimetry, and it is evaluated
at 10–60 kJ/mol [100], such an amount of energy corresponds to only a few hy-
drogen bonds. This destabilized state can then be easily reached upon adsorption.
Proteins with a high internal stability, non-deformable (wide gap between the two
states), and ‘hard’ proteins (lysozyme or ribonuclease) will be unable to gain suf-
ficient energy to displace water layer and thus will adsorb in very low amounts on
hydrophilic surfaces. For this kind of protein, higher amount adsorbed will imply
electrostatic interactions with a pH dependence. On the contrary, ‘soft’ proteins
(BSA, HSA, lactoalbumin, casein or hemoglobin) will adsorb on most surfaces by
gaining energy through conformational changes. This phenomenon can be probed
by studying adsorption–desorption cycle and will be considerably different between
a flexible protein like BSA and egg lysozyme hard protein. As the denaturation
process is driven, somehow, by the binding sites of the surface accessible to the
adsorbed protein, the competition with protein in the liquid phase will limit confor-
mational changes and thus, prevent a loss of biological activity [12]. As a specific
class of proteins, enzymes tend to be more sensitive to loss of biological activ-
ity after adsorption [101]. In their review of protein adsorption onto solid surface,
Nakanishi et al. [91] listed all the techniques applicable to measure amounts of
adsorbed proteins, and the most used are the depletion method (decrease in solute
concentration), Quartz Crystal Microbalance (QCM), ellipsometry, or radio-isotope
labelling. These techniques are sufficiently sensitive to measure protein concentra-
tion in the range of 0.15 mg/m2 (lactoalbumin on silica gel, depletion) to 6.3 mg/m2
(Ferritin on gold, QCM). Detection of conformational changes upon adsorption
 A. Ponche et al. / Journal of Adhesion Science and Technology 24 (2010) 2141–2164 2159

is generally unraveled by FT-IR spectroscopy, Atomic Force Microscopy, or flu-

orescence of tryptophan residues [91]. Mixed SAMs are also good candidates to
investigate the influence of hydrophobicity on the affinity of proteins for the sur-
face. For example, a mixture of thiol terminated by hydroxyl or methyl groups will
give a surface concentration of each thiol comparable to the solution ratio. By ad-
justing the ratio of the two thiols in solution, one can balance the hydrophilicity of
the surface from 0% OH to 100% OH. When adsorbed from single-protein solution,
fibronectin and vitronectin follow the usual trend for proteins, i.e., the more the sur-
face is hydrophilic, the less proteins are adsorbed. When adsorbed from serum, the
amount of fibronectin is lower, revealing the competition for adsorption with other
proteins. On the contrary, even if vitronectin undergoes competition with other pro-
teins from the serum, the hydrophilicity of the surface no more impacts the amount
of protein and the higher amount is observed for 100% OH samples. The antigen
activity evaluated on different surfaces indicates that hydrophilic surfaces preserve
better cell binding activity [102]. On mono-functional SAMs, adsorption of en-
zyme (human glucocerebrosidase) appears to be more strongly dependent on the
pH. Maximum adsorption is obtained in very acidic conditions (pH 4.5), and sur-
face binding is maximized by electrostatic interaction. But, on the contrary, specific
activity of the enzyme is enhanced near the isoelectric point of the enzyme (pH
7–8). This illustrates the influence of denaturation or conformational changes on
bioactivity: proteins adsorbed at pH near the isoelectric point are less sensitive to
structural rearrangements, keeping the enzyme after adsorption in a more efficient
tertiary structure for bioactivity [102]. Conformation controls biological activity of
the protein but, unfortunately, such information is very difficult to obtain. Some
attempts were made by ToF-SIMS but only partial information could be obtained
[27]. Monte Carlo simulation can help to understand denaturation or stable state of
protein adsorbed [103].
3.4.2. Grafting of Proteins
Mixed SAMs have been used on gold to investigate adsorption behavior of recom-
binant Protein A (PrA) and its biological activity towards rabbit IgG. Mixed SAMs
on gold are made by immersion in a mixture of two thiols. The surface concentra-
tion of each thiol appears to be correlated to the composition of the solution. XPS
and FT-IR studies show that classical activation by NHS is more efficient when the
surface carboxylic group is more accessible, i.e., when the two thiols have different
sizes. Unfortunately, when the lengths of the thiols are different, phase segregation
occurs. For PrA grafting, the best surface ratio was around 0.25. The authors have
also shown that mixed SAMs with equal lengths lead to a much higher amount
of PrA by an adsorption process and not by binding. This is confirmed by mea-
suring binding capacity towards rabbit IgG. This affinity is low on surface where
physisorption occurs even if the concentration is higher, whereas the affinity is
concentration-dependent for PrA grafted on mixed SAMs [104]. We have already
seen that cysteine amino acid was the only one to bear a thiol group and this prop-
erty makes it a good candidate for SAMs on gold surface. The side thiol group can
2160 A. Ponche et al. / Journal of Adhesion Science and Technology 24 (2010) 2141–2164

also be used to realize covalent grafting when it is part of the macromolecular chain
of a protein. If sulfur atoms are not involved in a disulfide bond giving a superstruc-
ture to the protein, they can be exploited to bind azurin [105]. Variable fragments of
antigen molecules can be grafted on gold via SAMs/maleimide strategy to realize
biosensor for virus detection. In that case, the target is a virus, known to interact
with the fusion protein grafted on the gold surface [106]. The protein is grafted on
the maleimide linker via a cysteine tag as shown in Fig. 3.

4. Summary and Conclusions

Interfaces play a major role in most of biological reactions. When dealing with
a surface reaction (like protein impact on osseointegration of implants or bacter-
ial colonization), techniques able to analyze the interface in ‘wet’ conditions are
scarce. For the moment, most of chemical information is obtained from spectro-
scopic techniques operated ex situ in Ultra High Vacuum. The relevance of water–
surface interactions for biological adhesion was pointed out by Kasemo in 1998
[1]. We actually know theoretically the impact of water molecules or hydration
layer on the conformation or ability of proteins to bind to a surface but experimen-
tal techniques able to probe a liquid–solid interface and provide physical–chemical
information at an atomic scale are still to be developed.
Due to the difficulty in probing a real ‘wet’ interface, studies have focussed on
realization of model surfaces with a simplified interface, made of peptide or amino
acids. On the basis of these simplified models, one expects to understand more ef-
ficiently the behavior of proteins in the vicinity of surfaces and its influence on
cell adhesion and bacterial colonization. Some studies have already shown that ad-
sorbed proteins ‘translate’ the surface chemistry into interactions understandable
by cell membrane proteins. Understanding of protein conformation, and interac-
tions with cells and bacteria, will open the way to chemically engineered implants
and specific cell response. Nevertheless, in vivo mechanisms are a step forward in
complexity due to the multiple partners involved and the three-dimensional charac-
teristics of extracellular matrices. We actually know that the rate at which fibroblasts
adhere on cell-derived three-dimensional matrix is considerably higher than onto a
two-dimensional substrate [107–109]. Such studies aim to control the chemistry
and also the topography of the materials to be analyzed. This requires a strong
expertise in surface science, chemistry, cell mechanics and biology: the need for
multidisciplinary approaches in biological surface science has never been so crucial
to address the requirements of bio-engineered materials with controlled biological
response. Designing model peptidic surfaces with a high control over the chem-
istry can help to determine interactions between cells or bacteria and surfaces. For
additional information about influence of chemistry and topography on cells and
bacteria, readers can refer to the article of K. Anselme et al. or the more specific
article focused on interfaces with bacteria written by L. Ploux et al. in this same
Journal of Adhesion Science and Technology issue.
 A. Ponche et al. / Journal of Adhesion Science and Technology 24 (2010) 2141–2164 2161

References
1. B. Kasemo, Current Opinion Solid State Mater. Sci. 3, 451–459 (1998).
2. P. J. Walla, Modern Biophysical Chemistry. Wiley VCH, Weinheim (2009).
3. M. A. Arnaout, S. Goodman and J.-P. Xiong, Current Opinion Cell Biology 19, 495–507 (2007).
4. M. Tulla, M. Lahti, J. S. Puranen, A.-M. Brandt, J. Käpylä, A. Domogatskaya, T. A. Salminen,
K. Tryggvason, M. S. Johnson and J. Heino, Expl. Cell Research 314, 1734–1743 (2008).
5. Y. Nymalm, J. S. Puranen, T. K. M. Nyholm, J. Käpylä, H. Kidron, O. T. Pentikäinen,
T. T. Airenne, J. Heino, J. P. Slotte, M. S. Johnson and T. A. Salminen, J. Biol. Chem. 279,
7962–7970 (2004).
6. B. Stevens, Y. Yang, A. Mohandas, B. Stucker and K. T. Nguyen, J. Biomedical Mater. Res. B
85, 573–582 (2007).
7. H. Shin, S. Jo and A. G. Mikos, Biomaterials 24, 4353–4364 (2003).
8. R. M. Donlan, Emerging Infectious Diseases 8, 881–890 (2002).
9. G. L. Bowlin and G. Wnek, Encyclopedia of Biomaterials and Biomedical Engineering. Taylor
& Francis (2005).
10. W.-J. Hu, J. W. Eaton and L. Tang, Blood 98, 1231–1238 (2001).
11. B. G. Keselowsky, A. W. Bridges, K. L. Burns, C. C. Tate, J. E. Babensee, M. C. LaPlaca and
A. J. Garcia, Biomaterials 28, 3626–3631 (2007).
12. C. E. Wilson, R. E. Clegg, D. I. Leavesley and M. J. Pearcy, Tissue Engineering 11, 1–17 (2005).
13. B. Kasemo, Surface Sci. 500, 656–677 (2002).
14. E. A. Vogler, Adv. Colloid Interface Sci. 74, 69–117 (1998).
15. A.-S. Duwez, J. Electron Spectr. Rel. Phenom. 134, 97–138 (2004).
16. M. Collaud Coen, R. Lehmann, P. Groning, M. Bielmann, C. Galli and L. Schlapbach, J. Colloid
Interface Sci. 233, 180–189 (2001).
17. Y. F. Dufrene, T. G. Marchal and P. G. Rouxhet, Appl. Surface Sci. 144–145, 638–643 (1999).
18. G. M. Harbers, L. J. Gamble, E. F. Irwin, D. G. Castner and K. E. Healy, Langmuir 21, 8374–
8384 (2005).
19. S. S. Jedlicka, J. L. Rickus and D. Y. Zemlyanov, J. Phys. Chem. B 111, 11850–11857 (2007).
20. N. Cottenye, F. Teixeira, A. Ponche, G. Reiter, K. Anselme, W. Meier, L. Ploux and C. Vebert-
Nardin, Macromolecular Biosci. 8, 1161–1172 (2008).
21. A. V. Saprigin, C. W. Thomas, C. S. Dulcey, C. H. Patterson and M. S. Spector, Surface Interface
Anal. 36, 24–32 (2004).
22. H. E. Canavan, X. Cheng, D. J. Graham, B. D. Ratner and D. G. Castner, Langmuir 21, 1949–
1955 (2005).
23. R. M. Petoral and K. Uvdal, Colloids Surfaces B 25, 335–346 (2002).
24. M. Riepl, K. Enander, B. Liedberg, M. Schaferling, M. Kruschina and F. Ortigao, Langmuir 18,
7016–7023 (2002).
25. H-B. Liu, N. V. Venkataraman, T. E. Bauert, M. Textor and S-J. Xiao, J. Phys. Chem. A 112,
12373–12377 (2008).
26. D. K. Aswal, S. Lenfant, D. Guerin, J. V. Yakhmi and D. Vuillaume, Analytica Chimica Acta
568, 84–108 (2006).
27. J. B. Lhoest, E. Detrait, P. van den Bosch de Aguilar and P. Bertrand, J. Biomed. Mater. Res. 41,
95–103 (1998).
28. Y. Xing, N. Dementev and E. Borguet, Current Opinion Solid State Mater. Sci. 11, 86–91 (2007).
29. C. Brüning, S. Hellweg, S. Dambach, D. Lipinsky and H. F. Arlinghaus, Surface Interface Anal.
38, 191–193 (2006).
2162 A. Ponche et al. / Journal of Adhesion Science and Technology 24 (2010) 2141–2164

30. R. E. Rawsterne, J. E. Gough, F. J. M. Rutten, N. T. Pham, W. C. K. Poon, S. L. Flitsch, B. Malt-

man, M. R. Alexander and R. V. Ulijn, Surface Interface Anal. 38, 1505–1511 (2006).
31. S.-J. Xiao, S. Brunner and M. Wiedland, J. Phys. Chem. B 108, 16508–16517 (2004).
32. C. Poleunis, C. Rubio, C. Compere and P. Bertrand, Appl. Surface Sci. 203–204, 693–697 (2003).
33. J. Fick, R. Steitz, V. Leiner, S. Tokumitsu, M. Himmelhaus and M. Grunze, Langmuir 20, 3848–
3853 (2004).
34. H. J. Griesser, P. Kingshott, S. L. McArthur, K. M. McLean, G. R. Kinsel and R. B. Timmons,
Biomaterials 25, 4861–4875 (2004).
35. B. J. Houseman and M. Mrksich, Trends Biotechnol. 20, 279–281 (2002).
36. M. Mrksich, ACS Nano 2, 7–18 (2008).
37. J. Su and M. Mrksich, Langmuir 19, 4867–4870 (2003).
38. N. Tang, P. Tornatore and S. R. Weinberger, Mass Spectrometry Reviews 23, 34–44 (2004).
39. S. Nagl, M. Schaeferling and O. S. Wolfbeis, Microchimica Acta 151, 1–21 (2005).
40. P.-H. Chua, K.-G. Neoh, E.-T. Kang and W. Wang, Biomaterials 29, 1412–1421 (2008).
41. C. C. Barrias, C. L. Martins, C. S. Miranda and M. A. Barbosa, Biomaterials 26, 2695–2704
(2005).
42. I. C. Goncalves, M. C. L. Martins, M. A. Barbosa and B. D. Ratner, Biomaterials 26, 3891–3899
(2005).
43. M.-C. Clochard, N. Betz, M. Goncalves, C. Bittencourt, J.-J. Pireaux, K. Gionnet, G. Deleris and
A. Le Moel, Nucl. Instrum. Methods Phys. Res. B 236, 208–215 (2005).
44. J. G. Archambault and J. L. Brash, Colloids Surfaces B 39, 9–16 (2004).
45. I. Dalle-Donne, R. Rossi, D. Giustarini, A. Milzani and R. Colombo, Clinica Chemica Acta 329,
23–38 (2003).
46. K. Ichihara-Tanaka, K. Tikani and K. Sekigushi, J. Cell Sci. 108, 907–915 (1995).
47. W. Kang, S. Park and J.-H. Jang, Biotechnology Letters 30, 55–59 (2008).
48. R.-I. Manabe, N. Oh-e, T. Maeda, T. Fukuda and K. Sekiguchi, J. Cell Biology 139, 295–307
(1997).
49. W. Senaratne, L. Andruzzi and C. K. Ober, Biomacromolecules 6, 2427–2448 (2005).
50. M. Mrksich and G. M. Whitesides, Trends Biotechnol. 13, 228–235 (1995).
51. J. Amalric, P. H. Mutin, G. Guerrero, A. Ponche, A. Sotto and J.-P. Lavigne, J. Mater. Chem. 19,
141–149 (2009).
52. E. Mateo-Marti, C. Briones, E. Roman, E. Briand, C. M. Pradier and J. A. Martin-Gago, Lang-
muir 21, 9510–9517 (2005).
53. J. C. Love, L. A. Estroff, J. K. Kriebel, R. G. Nuzzo and G. M. Whitesides, Chemical Reviews
105, 1103–1169 (2005).
54. L. Li, S. Chen, J. Zheng, B. D. Ratner and S. Jiang, J. Phys. Chem. B 109, 2934–2941 (2005).
55. Z. Yang, J. A. Galloway and H. Yu, Langmuir 15, 8405–8411 (1999).
56. E. Ostuni, R. G. Chapman, R. E. Holmlin, S. Takayama and G. M. Whitesides, Langmuir 17,
5605–5620 (2001).
57. C. M. Dekeyser, C. C. Buron, K. Mc. Evoy, C. C. Dupont-Gillain, J. Marchand-Brynaert and
A. M. Jonas, J. Colloid Interface Sci. 324, 118–126 (2008).
58. A. Borgh, J. Ekeroth, R. M. Petoral Jr, K. Uvdal, P. Konradsson and B. Liedberg, J. Colloid
Interface Sci. 295, 41–49 (2006).
59. H. G. Hong, M. Jiang, S. G. Sligar and P. W. Bohn, Langmuir 10, 153–158 (1994).
60. D. G. Kurth and T. Bein, Langmuir 11, 3061–3067 (1995).
61. P. Harder, M. Grunze, R. Dahint, G. M. Whitesides and P. E. Laibinis, J. Phys. Chem. B 102,
426–436 (1998).
 A. Ponche et al. / Journal of Adhesion Science and Technology 24 (2010) 2141–2164 2163

62. C. Hoffmann and G. E. M. Tovar, J. Colloid Interface Sci. 295, 427–435 (2006).
63. G. L. Kenausis, J. Voros, D. L. Elbert, N. Huang, R. Hofer, L. Ruiz-taylor, M. Textor,
J. A. Hubbell and N. D. Spencer, J. Phys. Chem. B 104, 3298–3309 (2000).
64. S. Pasche, S. M. De Paul, J. Voros, N. D. Spencer and M. Textor, Langmuir 19, 9216–9225
(2003).
65. J.-Y. Wach, B. Malisova, S. Bonazzi, S. Tosatti, M. Textor, S. Zurcher and K. Gademann, Chem.
Eur. J. 14, 10579–10584 (2008).
66. H. Lee, J. Rho and P. B. Messersmith, Adv. Mater. 20, 1–4 (2008).
67. F. Bernsmann, A. Ponche, C. Ringwald, J. Hemmerle, J. Raya, B. Bechinger, J.-C. Voegel,
P. Schaaf and V. Ball, J. Phys. Chem. C 113, 8234–8242 (2009).
68. H. Lee, S. M. Dellatore, W. M. Miller and P. B. Messersmith, Science 318, 426–430 (2007).
69. Z. Zhang, S. Chen and S. Jiang, Biomacromolecules 7, 3311–3315 (2006).
70. Z. Zhang, S. Chen, Y. Chung and S. Jiang, J. Phys. Chem. B 110, 10799–10804 (2006).
71. F. Bretagnol, H. Rauscher, M. Hasiwa, O. Kylian, G. Ceccone, L. Hazell, A. J. Paul, O. Lefranc
and F. Rossi, Acta Biomaterialia 4, 1745–1751 (2008).
72. S. Caillou, P. A. Gerin, C. J. Nonckreman, S. Fleith, C. C. Dupont-Gillain, J. Landoulsi,
S. M. Pancera, M. J. Genet and P. J. Rouxhet, Electrochimica Acta 54, 116–122 (2008).
73. S. Fleith, A. Ponche, R. Bareille, J. Amedee and M. Nardin, Colloids Surfaces B 44, 15–24
(2005).
74. L. Ploux, K. Anselme, A. Dirani, A. Ponche, O. Soppera and V. Roucoules, Langmuir 25, 8161–
8169 (2009).
75. D. Voet and J. G. Voet, Biochemistry, 3rd edn. Wiley International Edition (2004).
76. F. Höök, B. Kasemo, M. Grünze and S. Zauscher, ACS Nano 2, 2428–2436 (2008).
77. S. M. Barlow, K. J. Kitching, S. Haq and N. V. Richardson, Surface Sci. 401, 322–335 (1998).
78. P. Löfgren, A. Krozer, J. Lausmaa and B. Kasemo, Surface Sci. 370, 277–292 (1997).
79. S. M. Barlow and R. Raval, Surface Sci. Reports 50, 201–341 (2003).
80. G. Hager and A. G. Brolo, J. Electroanal. Chem. 550–551, 291–301 (2003).
81. W.-W. Zhang, C.-S. Lu, Y. Zou, J.-L. Xie, X.-M. Ren, H.-Z. Zhu and Q.-J. Meng, J. Colloid
Interface Sci. 249, 301–306 (2002).
82. G. Gonella, S. Terreni, D. Cvetko, A. Cossaro, L. Mattera, O. Cavalleri, R. Rolandi, A. Morgante,
L. Floreano and M. Canepa, J. Phys. Chem. B 109, 18003–18009 (2005).
83. O. Cavalleri, L. Oliveri, A. Dacca, R. Parodi and R. Rolandi, Appl. Surface Sci. 175–176, 357–
362 (2001).
84. M. M. Beerbom, R. Gargaliano and R. Schlaf, Langmuir 21, 3551–3558 (2005).
85. W. Yang, J. J. Gooding and D. B. Hibbert, J. Electroanal. Chem. 516, 10–16 (2001).
86. D. H. Dressler and Y. Mastai, Chirality 19, 358–365 (2007).
87. E. Ruoslahti and M. Pierschbacher, Science 238, 491–497 (1987).
88. M. Tirrell, E. Kokkoli and M. Biesalski, Surface Sci. 500, 61–83 (2002).
89. R. L. Willet, K. W. Baldwin, K. W. West and L. N. Pfeifer, Proc. Natl Acad. Sci. USA 22, 7817–
7822 (2005).
90. H. H. Park and A. Ivanisevic, J. Phys. Chem. C 111, 3710–3718 (2007).
91. K. Nakanishi, T. Sakiyama and K. Imamura, J. Bioscience Bioeng. 91, 233–244 (2001).
92. J. P. Cloarec, Y. Chevelot, E. Laurenceau, M. Phaner-Goutorbe and E. Souteyrand, ITBM-RBM
29, 105–127 (2008).
93. K. C. Dee, T. T. Andersen and R. Bizios, J. Biomed. Mater. Res. 40, 371–377 (1998).
94. K. C. Dee, T. T. Andersen and R. Bizios, Biomaterials 20, 221–227 (1999).
2164 A. Ponche et al. / Journal of Adhesion Science and Technology 24 (2010) 2141–2164

95. W. K. J. Mosse, M. L. Koppens, T. R. Gengenbach, D. B. Scanlon, S. L. Gras and W. A. Ducker,

Langmuir 25, 1488–1494 (2009).
96. S.-J. Xiao, M. Textor and N. D. Spencer, Langmuir 14, 5507–5516 (1998).
97. T. Baas, L. Gamble, K. D. Hauch, D. G. Castner and T. Sasaki, Langmuir 18, 4898–4902 (2002).
98. S. Pallu, C. Bourgeta, R. Bareille, C. Labrugere, M. Dard, A. Sewing, A. Jonczyk, M. Vernizeau,
M.-C. Durrieu and J. Amédée-Vilamitjan, Biomaterials 26, 6932–6940 (2005).
99. M. C. Durrieu, S. Pallu, F. Guillemot, R. Bareille, J. Amedee, Ch. Baquey, C. Labrugere and
M. Dard, J. Mater. Sci. — Materials in Medicine 15, 779–786 (2004).
100. L. Razumovsky and S. Damodaran, Langmuir 15, 1392–1399 (1999).
101. J. H. Wei, T. Kacar, C. Tamerler, M. Sarikaya and D. S. Ginger, Small 5, 689–693 (2009).
102. C. C. Barrias, M. C. L. Martins, G. Almeida-Porada, M. A. Barbosa and P. L. Granja, Biomate-
rials 30, 307–316 (2009).
103. H.-J. Hsu, S.-Y. Sheu and R.-Y. Tsay, Colloids Surfaces B, 67, 183–191 (2008).
104. E. Briand, M. Salmain, C. Compere and C.-M. Pradier, Colloids Surfaces B, 53, 215–224 (2006).
105. Q. Chi, J. Zhang, J. U. Nielsen, E. P. Friis, I. Chorkendorff, G. W. Canters, J. E. T. Andersen and
J. Ulstrup, J. Amer. Chem. Soc. 122, 4047–4055 (2000).
106. L. Torrance, A. Ziegler, H. Pittman, M. Paterson, R. Tothb and I. Eggleston, J. Virological Meth-
ods 134, 164–170 (2006).
107. B. Geiger, Science 294, 1601–1602 (2001).
108. E. Cukierman, R. Pankov, D. R. Stevens and K. M. Yamada, Science 294, 1708–1712 (2001).
109. M. R. Dusseiller, M. L. Smith, V. Vogel and M. Textor, Biointerphases 1, 1–4 (2006).