Single ether group in a glycol-based ultra-thin layer prevents surface fouling from undiluted serum

Article abstract of DOI:10.1039/c2cc15692j

Through systematic structural modification, it is shown that the internal, single oxygen atom of simple monoethylene glycol-based organic films is essential for radically altering the fouling behaviour of quartz against undiluted serum, as characterized by the electromagnetic piezoelectric acoustic sensor. The synergy is strongest with distal hydroxyls. The Royal Society of Chemistry 2012.

Full text of DOI:10.1039/c2cc15692j

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2012, 48, 1305–1307
www.rsc.org/chemcomm
COMMUNICATION
Single ether group in a glycol-based ultra-thin layer prevents surface
fouling from undiluted serumw
Sonia Sheikh, David Yi Yang, Christophe Blaszykowski and Michael Thompson*
Received 14th September 2011, Accepted 30th November 2011
DOI: 10.1039/c2cc15692j
Through systematic structural modification, it is shown that the
internal, single oxygen atom of simple monoethylene glycol-
based organic films is essential for radically altering the fouling
behaviour of quartz against undiluted serum, as characterized by
the electromagnetic piezoelectric acoustic sensor. The synergy is
strongest with distal hydroxyls.
plasma and serum.3,4 Unfortunately, few coatings retain such
level of performance when it comes to completely undiluted
fluids,3,4 and are, accordingly, not applicable in either biomedical
or bioanalytical technology.
In order to examine the interactions of surfaces (e.g. SiO2)
with proteins, cells and various biomolecules, we have developed
the highly sensitive electromagnetic piezoelectric acoustic sensor
(EMPAS).6,7 This device, which is operated in a flow-injection
configuration, is capable of the detection of such chemistry in a
real-time and label-free manner. This is demonstrated through
use of the device as an immunosensor dedicated to the serological
detection of anti-HIV antibodies with simultaneous assessment
of cross-reactivity and serum fouling.8 The system operates
through the instigation of acoustic resonance at frequencies up
to 1 GHz in a quartz substrate by an electromagnetic field
associated with a flat spiral coil.6–8 Adsorption of species from
solution, whether from target analytes or fouling moieties, causes
a change in the resonant frequency of the sensor. The response of
the sensor in liquids is governed by interfacial effects such as
alteration of viscoelasticity and slip phenomena rather than
gravimetric effects.6,7,9 Accordingly, the EMPAS system is ideal
for the study of the antifouling behaviour of any coating imposed
onto quartz.
Prevention of the fouling of surfaces by biological species
circulating in the bloodstream and interstitial fluids is of
utmost importance for the development of biocompatible
and fully functional biomedical implants and devices with
extended life expectancy. This phenomenon constitutes a
challenging problem since it can result in implant malfunction,
complete failure or cause adverse outcomes such as restenosis
with respect to stent technology.1 In the worst-case scenario,
the host body triggers an immunological response, which
results in post-operative complications such as peri-implant
tissue inflammation with eventual rejection of the foreign
object.1 In biosensor technology, fouling is more often referred
to as ‘‘non-specific adsorption’’ (NSA) and represents a
serious prevailing concern for many detection platforms that
could be potentially employed for the analysis of bodily fluids
such as blood and serum. These biological matrices are
composed of complex mixtures of interfering biomolecules,
in particular proteins at high concentration (the reference
interval for total serum protein is 60–80 g LÀ1),2 that have
the propensity to adsorb non-specifically to the surface of the
device thereby preventing the accurate measurement of target
analytes present in considerably lower concentrations (down
to ng LÀ1). Tremendous efforts have been devoted over many
years towards the development of protein-resistant surfaces
and numerous types of organic coatings have been reported in
the literature for such purpose, oligo- (OEG) and polyethylene
glycol (PEG) polymers, historically, being the most prevalent.3–5
Ultralow fouling (o5 ng cmÀ2) has now been achieved for
single-protein buffered aqueous solutions as well as diluted blood
The focus of this paper is a comparison of the anti-
fouling behaviour, on quartz, of several organic monolayers
(Scheme 1) constructed from structurally simple surface
modifiers, characterized against undiluted serum with the
EMPAS system.
Surface engineering using trichlorosilane self-assembling
monolayer (SAM) chemistry for the development of selective
biosensing platforms is currently the object of research in
Department of Chemistry, University of Toronto,
80 St. George Street, Toronto, ON, Canada M5S 3H6.
E-mail: mikethom@chem.utoronto.ca
w Electronic supplementary information (ESI) available: Procedures
for surface modifier synthesis, quartz cleaning, organic film formation
and EMPAS measurements as well as film characterization by angle-
resolved XPS and contact angle goniometry. See DOI: 10.1039/
c2cc15692j
Scheme 1 Surface modification of quartz.
c
This journal is The Royal Society of Chemistry 2012
Chem. Commun., 2012, 48, 1305–1307 1305

View Article Online
The antifouling behaviour of bare and derivatized quartz
surfaces against undiluted goat serum (45–75 mg protein/mL)
was studied with the EMPAS at the ultra-high frequency of
0.94 GHz. From Fig. 2, it is immediately obvious that all
organic adlayers yield a decrease in the shift in resonant
frequency compared to the value observed for unmodified
quartz (À31 kHz). Fully alkylated OTS films perform poorly
(À22 kHz). Substitution for distal hydroxyl groups in OTS-
OH provides no improvement (À23 kHz). Remarkably, the
incorporation of an internal, single atom of oxygen in an
MEG-type film results in a highly significant reduction in the
resonant frequency shift associated with surface adsorption
(À2 kHz for MEG-OH). This represents a >15-fold decrease
relative to the result for bare quartz. In a number of experi-
ments, no shift is observed. Other MEGylated films (MEG-
TFA and MEG-OMe) also exhibit such an effect, although to a
lesser extent. This result confirms the observation that the
internal atom of oxygen in the b position is not only essential,
but also acts synergistically in tandem with the distal hydroxyl
group. These results may reveal why EGylated surfaces are so
effective for the prevention of fouling. EMPAS measurements
generally display reproducible behaviour with the exception of
the MEG-TFA films. Random removal of the labile TFA
groups during the EMPAS experiment, by the running PBS
buffer or/and most certainly serum, is likely responsible for such
an observation.
Scheme 2 Structure of MEG-TFA surface modifier.
our laboratory. In a recent study, that described combination
of the EMPAS with such SAM chemistry, we reported
low-fouling OEGylated mixed layers able to substantially
decrease the adsorption of avidin from relatively concentrated
(0.1 mg mLÀ1) PBS-buffered solutions.10,11 Distributed within
these assemblies were shorter monoethylene glycolated (MEG)
surface modifying building blocks (renamed MEG-TFA
herein—Scheme 2), where the intended role was to space out
the biorecognition element linking residues, causing these to
protrude thereby offering enhanced overall binding ability.10,11
This work led directly to the consideration that pure MEG-TFA
films could possess important antifouling properties when
exposed to complex, real-world bodily fluids such as undiluted
serum. This feature of the work, in turn, led to a study of the
related organic ultra-thin layers (Scheme 1), as well as bare
quartz, in an attempt to gain better understanding of the
structural parameters that influence antifouling.
MEG-TFA, MEG-OMe and OTS films were prepared via
immersion of cleaned quartz discs into 1/1000 (v/v) solutions
of the appropriate surface modifier in anhydrous toluene,
for 60 min at room temperature.w All films were next soaked
over-night in a 1/1 (v/v) solution of methanol and Milli-Q
water, generating in the process the desired MEG-OH and
OTS-OH coatings by cleaving the labile trifluoroacetyl (TFA)
terminal groups of respectively MEG-TFA and OTS-TFA
(not shown)w surfaces. The formation of the MEG-TFA
films and subsequent complete removal of the TFA groups
were successfully determined using X-ray photoelectron
spectroscopy (XPS) following the appearance (respectively
the loss) of a peak for fluorine at 689 eV, the one element
unambiguously attributable to MEG-TFA (Fig. 1). Unlike
fluorine however, the signal for carbon (at 285 eV) was not
affected by the following mild aqueous treatment, clearly
demonstrating that the latter had effectively cleaved the
TFA groups without etching the organic backbone from the
quartz substrate. The intensity of the substrate Si signal, given
the escape depth of these electrons, strongly implies the
formation of ultra-thin films. This result is consistent with
ellipsometric measurements, which gave a thickness of B5 A
for MEG-TFA film.w
The shape of the EMPAS response profiles is also notable.
For bare quartz, the resonant frequency first drops sharply
(BÀ25 kHz) then gradually decreases (BÀ7 kHz in B600 s)
before final stabilization (Fig. 3A). While the former phase is
undoubtedly due to serum reaching the quartz surface, pro-
ducing a high level of fouling, the second phase likely reflects
the passage of the viscous sample over the surface, in the flow-
injection configuration. No final rinse-off by the uninter-
rupted buffer flow is observed, clearly indicating that species
responsible for fouling accumulate irreversibly. With respect
to MEG-OH films, the EMPAS profiles display radically
different behaviour, with a progressive and comparatively
limited initial drop in resonant frequency (>À5 kHz in
B250 s) typically followed by a gradual and extensive (up to
1500 s and +2 kHz) rinse-off (Fig. 3B). This observation
seems to indicate that, in this case, the majority of serum
species are adsorbed in a reversible fashion involving transient
interaction with the substrate surface. Interestingly, this could
represent the signature of a sequential ‘‘Vroman-like’’ process,
Fig. 2 EMPAS frequency shift upon injection of undiluted serum onto
Fig. 1 XPS surveys for bare quartz, MEG-TFA and MEG-OH films.
1306 Chem. Commun., 2012, 48, 1305–1307
bare and derivatized quartz surfaces (4 to 8 replicates per data set).
c
This journal is The Royal Society of Chemistry 2012

View Article Online
of surfaces and shown to be physically distinct, solute-free
exclusion zones projecting up to several hundred microns into
the contiguous aqueous phase.17 Quartz also has long-range
water structuring properties but only the molecules within a
few monolayers are rigidly bound to the surface.18 This
explains why quartz presents a surface that is fouled with
facility by serum.
In conclusion, we have presented evidence that short
MEGylated organic monolayers, constructed from structurally
simple surface modifiers in a straightforward two-step sequence,
are able to radically alter the fouling behaviour of quartz
against undiluted serum. A key feature is the observation that
the internal ether oxygen acts in tandem with the distal –OH
moiety, likely through a mechanism involving the instigation of
a special intramolecular zone of hydration, which compromises
the ability of the latter to engage in H-bonding. From a
practical point of view, we believe that this work may prove
to be applicable to other medically and bioanalytically relevant
hydroxylated (bio)materials such as medical grade stainless steel
and ceramics for implantable devices, as well as silicon and
indium-tin oxides for biosensor applications.
Fig. 3 EMPAS profiles for (a) bare quartz and (b) MEG-OH film.
We thank the Natural Sciences and Engineering Research
Council of Canada for support, Dr Peter Brodersen from
Surface Interface Ontario for XPS analysis as well as Prof.
Kenneth S. Burch and Yao Tian (Department of Physics –
University of Toronto) for ellipsometry analysis.
wherein higher mobility, more abundant proteins first adsorb
before being ultimately displaced by less motile, higher affinity
entities.12 Protein adsorption is
a complex, multi-step
process12 and, accordingly, the increase in resonant frequency
may reflect the occurrence of rigidifying viscoelastic pheno-
mena within the fouling adlayer, such as those arising
from protein structural rearrangments. Such structural and
conformational reorganizations have been reported for serum
albumin,13 calcium-binding calmodulin14 and HIV envelop
glycoprotein gp12015 using the conventional bulk acoustic
wave sensor employed for measurement of both series resonant
frequency and energy dissipation or motional resistance.
Notes and references
1 P. Thevenot, W. Hu and L. Tang, Curr. Top. Med. Chem., 2008,
8, 270.
2 J. N. Adkins, S. M. Varnum, K. J. Auberry, R. J. Moore,
N. H. Angell, R. D. Smith, D. L. Springer and J. G. Pounds,
Mol. Cell. Proteomics, 2002, 1, 947.
3 C. Zhao, L. Li, Q. Wang, Q. Yu and G. Zheng, Langmuir, 2011,
27, 4906.
´
4 H. Vaisocherova, Z. Zhang, W. Yang, Z. Cao, G. Cheng,
A. D. Taylor, M. Piliarik, J. Homola and S. Jiang, Biosens.
Bioelectron., 2009, 24, 1924.
5 E. Ostuni, R. G. Chapman, R. E. Holmlin, S. Takayama and
G. M. Whitesides, Langmuir, 2001, 17, 5605.
6 M. Thompson, S. M. Ballantyne, L.-E. Cheran, A. C. Stevenson
and C. R. Lowe, Analyst, 2003, 128, 1048.
7 S. M. Ballantyne and M. Thompson, Analyst, 2004, 129, 219.
8 S. Sheikh, C. Blaszykowski and M. Thompson, Talanta, 2011,
85, 816.
9 J. S. Ellis and M. Thompson, Langmuir, 2010, 26, 11558.
10 S. Sheikh, J. C.-C. Sheng, C. Blaszykowski and M. Thompson,
Chem. Sci., 2010, 1, 271.
Finally, when MEG-OMe films are not submitted to
the subsequent aqueous treatment, the EMPAS profiles show
different response upon serum injection (À13 vs. À7 kHz,
Fig. 4). These results appear to corroborate the widely
proposed hypothesis that surface hydration is intimately
involved in protein repellency. This argument proposes the
concept of a ‘‘water barrier’’, wherein embedded and inter-
facial water molecules are tightly bound into permeated
structures that have an energy cost in terms of disturbance.16
Such barriers have been elegantly highlighted for several types
11 M. Thompson, S. Sheikh, J. C.-C. Sheng and C. Blaszykowski,
US patent application No. 13/154,067.
12 M. Wahlgren and T. Arnebrant, Trends Biotechnol., 1991, 9,
201.
´
13 S. Belegrinou, I. Mannelli, P. Lisboa, F. Bretagnol, A. Valsesia,
G. Ceccone, P. Colpo, H. Rauscher and F. Rossi, Langmuir, 2008,
24, 7251.
14 X. Wang, J. S. Ellis, E.-L. Lyle, P. Sundaram and M. Thompson,
Mol. BioSyst., 2006, 2, 184.
15 H.-S. Lee, M. Contarino, M. Umashankara, A. Schon, E. Freire,
¨
A. B. Smith III, I. M. Chaiken and L. S. Penn, Anal. Bioanal.
Chem., 2010, 396, 1143.
16 S. Chen, L. Li, C. Zhao and J. Zheng, Polymer, 2010, 51, 5283.
17 J.-M. Zheng, W.-C. Chin, E. Khijniak, E. Khijniak Jr. and
G. H. Pollack, Adv. Colloid Interface Sci., 2006, 127, 19.
18 N. K. Roberts and G. Zundel, J. Phys. Chem., 1980, 84, 3655.
Fig. 4 EMPAS profiles for MEG-OMe coatings (a) soaked
(15% RSD, n = 6) or (b) not soaked (8% RSD, n = 5) into
MeOH/H2O, overnight.
c
This journal is The Royal Society of Chemistry 2012
Chem. Commun., 2012, 48, 1305–1307 1307