Microchemical Journal 93 (2009) 82–86
Contents lists available at ScienceDirect
Microchemical Journal
j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / m i c r o c
Microfluidic networks for surface plasmon resonance imaging real-time
kinetics experiments
Giuseppe Grasso a, Roberta D'Agata a, Laura Zanoli b, Giuseppe Spoto a,c,⁎
a
b
c
Dipartimento di Scienze Chimiche, Università di Catania, Viale Andrea Doria 6, 95125, Catania, Italy
Scuola Superiore di Catania, Università di Catania, Via S. Paolo 73, Catania, Italy
Istituto Biostrutture e Bioimmagini, CNR, Viale A. Doria 6, Catania, Italy
a r t i c l e
i n f o
Article history:
Received 20 February 2009
Accepted 5 May 2009
Available online 9 May 2009
Keywords:
Surface plasmon resonance imaging
Real-time kinetics
Microfluidics
Datura stramonium agglutinin
Asialofetuin
a b s t r a c t
The coupling of microfluidic devices with surface plasmon resonance imaging (SPRI) has emerged in recent
years as a novel approach for the simultaneous monitoring of interactions of biomolecules arrayed onto gold
substrates. In order to minimize a variety of effects which affect the final determination of kinetic parameters
(non-specific interactions above all), difficult choices of appropriate references are often encountered in
carrying out SPRI investigations. A common solution to these problems consists of laborious experimental
setup involving the use of specially designed microchannels and tedious manipulation of the gold substrate
that often produces surface degradation.
In this work, a discussion about appropriate choice of references in SPRI measurements is opened and the use
of alternative microfluidic patterns coupled to the SPRI system is proposed as a solution to the above
mentioned problems. Specifically, a Y-shaped SPRI flow cell has been constructed from masters in polyvinyl
chloride and it has been identified as one of the most suitable experimental approach for obtaining
appropriate referencing during SPRI experiments. The experimental set up has been tested in a real time
study of the interaction between the Datura Stramonium Agglutinin and the asialofetuin and the obtained
results demonstrate the suitability of such microfluidic network in SPRI investigations.
© 2009 Elsevier B.V. All rights reserved.
1. Introduction
The manipulation of fluids in the micro-environment has offered in
recent years the possibility of solving outstanding system integration
issues that are critical for biology and chemistry. In particular,
microfluidic devices, which can be identified by the presence of one
or more channels with at least one dimension less than 1 mm, have
been used for many different applications [1,2] in the field of
chemistry and biology [3] and the area of micro total analysis systems
(μ-TAS), also called “lab-on-a-chip”, or miniaturized analysis systems,
is growing rapidly [4]. The main reason behind such rapid development is that there are many advantages over conventional instrumentation, as this miniaturized working environment is very suitable
for lots of different applications such as cell patterning and
separations [5], DNA analysis [6], enzyme reaction kinetics [7,8] and
many others [9,10]. Microfluidic systems have also a very important
application in the field of biosensor devices and, in recent years, they
have often been coupled with techniques such as surface plasmon
resonance imaging (SPRI) [11,12]. The latter expands the label-free
capability of the standard SPR technique to rapidly evaluate the
⁎ Corresponding author. Dipartimento di Scienze Chimiche, Università di Catania,
Viale Andrea Doria 6, 95125, Catania, Italy.
E-mail address: [email protected] (G. Spoto).
0026-265X/$ – see front matter © 2009 Elsevier B.V. All rights reserved.
doi:10.1016/j.microc.2009.05.001
interaction between an analyte and its biospecific partner immobilized on the sensor surface [13,14] by allowing a multiplexed approach
to the evaluation of biomolecular interactions and to the sensing of
chemical and biological analytes [15–17]. To take fully advantage of
the SPRI approach a precise control of both the patterning of
biomolecules onto the sensor surface as well as the fluidic of the
analyte solution is imposed. In this perspective, the use of microfluidic
devices provides SPRI compatible convenient means for manipulating
very small amounts of sample and for controlling the patterning of a
variety of different biomolecules (i.e. DNA, RNA, peptides, proteins,
carbohydrates) [18,19].
Nevertheless, SPR data are often affected by various artifacts and
the choice of appropriate reference cells in the analysis of kinetic data
represents a frequently addressed problem [20].
In SPRI, in order to overcome to some of the above mentioned
inconveniences, experiments are usually conducted by following a
multistep process where the arrayed surface is first obtained by
microspotting [17,21] or by flowing into poly(dimethylsiloxane)
(PDMS) made microchannels in contact to the functionalized sensor
surface [22,23] a solution of the biomolecule to be arrayed. The
interaction with the selected analytes requires the removal of the
previous PDMS device and the use of a second set of microchannels
that are attached to the surface perpendicular to the previously
arrayed areas of the sensor surface. The intersections of the arrays
G. Grasso et al. / Microchemical Journal 93 (2009) 82–86
with the microchannels define the regions of the sensor surface where
the interactions take place, while the surrounding areas are used as
reference, so to take into account non-specific interactions and/or
changes in bulk refractive index.
A limitation of a similar double step approach is represented by the
need to further manipulate the gold substrate after the anchoring of
the biomolecular array.
We present here results on SPRI experiments aimed at optimizing
the coupling to microfluidic devices. In particular, we discuss a
microfluidic approach based on the use of a device carrying a Yshaped microchannel, which tries to minimize the above mentioned
limitation associated to the current use of PDMS made microfluidic
devices in SPRI experiments. The study of the interaction between the
lectin from Datura Stramonium Agglutinin (DSA) and the asialofetuin
(AF) is reported to demonstrate the use of such microfluidic network
in SPRI.
2. Experimental
2.1. Materials
Reagents were obtained from commercial suppliers and used
without further purification. Triethanolamine (TEA), DSA and AF were
purchased from Sigma-Aldrich; methoxy-polyethyleneglycol amine
(mPEG, MW = 5000) was purchased from Nektar Therapeutics (USA).
Phosphate buffered saline (PBS) solutions at pH = 7.4, (NaCl 137 mM,
KCl 2.7 mM, phosphate buffered 10 mM, Amresco) were used in all the
SPRI experiments.
Gold substrates (GWC Instruments, USA) for SPRI measurements
were obtained by thermally evaporating a gold layer (450 Å) on to SF10 glass slides. Chromium (50 Å) was used as the adhesion layer.
0.0097 g of dithiobis succinimidylpropionate (Lomant's reagent) were
dissolved in 5 ml of dimethyl sulfoxide (DMSO) (Sigma-Aldrich) and
the gold substrates were immersed in the obtained solution for 48 h.
After two rinsing steps with ultra-pure water (Milli-Q element,
Millipore) and absolute ethanol (Sigma-Aldrich), the modified gold
substrates were then used for SPRI studies.
2.2. Microfluidics background and PDMS channels fabrication
Fluids behaviour in the microfluidic environment is different from
the one observed in the macroscopic world [2], resulting very suitable
and advantageous for SPRI investigations. In particular, behaviour of
fluids in the microfluidic regime can be rationalized by quantifying the
relative contribution of a variety of competing physical phenomena by
using dimensionless numbers. While the Péclet dimensionless
number (Pe) expresses the relative importance of convection to
diffusion, the Reynolds number (Re) helps in quantifying the relative
contribution of inertial (fi) and viscous (fv) forces. In the microfluidic
environment viscous forces typically overwhelm inertial forces, and
the resulting flows are therefore laminar. In this scenario, the T-Sensor
is a μ-TAS component that combines microfluidic separation and
detection functions and it has been applied also for quantitative
analysis of molecular interactions [24]. Fluids coming from two
separate microchannels are put in contact in a single and wider
microchannel and interact during parallel flow until they exit the
microstructure. Interdiffusion occurs during the time that they are in
contact and the occurred diffusion at a certain point down the single
channel where the two solutions are put in contact can be obtained
from:
0:5
l = ðDt Þ
83
function of the diameter and so, ultimately, of the molecular weight
(MW) of the particles.
All the above mentioned details have to be considered if a
diffusion-controlled mixing of miscible liquids in the microfluidic
regime has be obtained by putting in contact two different fluids in a
Y-shaped microchannel [25,26]. As long as the latter has appropriate
dimensions and a right choice of flow rate is made [27], the fluids flow
in a laminar mode and the mass transport is diffusion rather than
convection limited. As a consequence, the boundary between the two
miscible fluids can be regarded as a dynamic interface that can be
manipulated and put to practical use [2].
A Y-shaped master in polyvinyl chloride (PVC) was specially
designed for this work and poly(dimethylsiloxane) (PDMS) flow cells
were fabricated from such master as described elsewhere [28]. The
overall length of the Y-shaped channel was 1 cm and the two branches
had identical dimensions (80 μm depth, 0.4 cm length, 0.5 mm width)
that differ from the one of the single channel (80 μm depth, 0.7 cm
length, 1 mm width). At the ends of each channel circular reservoirs
(diameter 400 μm) were pierced into the PDMS by using a piercing
tool of appropriate size. C-Flex tubes (Upchurch Scientific) were
inserted in such reservoirs in order to connect the PDMS microfluidic
cell to a Masterflex L/S (Cole-Parmer, USA) peristaltic pump,
operating at 500 μl/min in order to minimize any mass transport
effect. An image of the device carrying the Y-shaped flow cell is shown
in Fig. 1. An Re value of about 100 can be calculated for the
experimental condition maintained during the SPRI experiments.
Such value indicates that our Y-shaped microchannel ensures laminar
flow conditions during SPRI experiments.
2.3. SPRI experiments and data analysis
All the SPRI experiments were carried out by using an SPR imager
apparatus (GWC Technologies, USA) elsewhere described in detail
[29]. SPR images were analyzed by using the V++ software (version
4.0, Digital Optics Limited, New Zealand) and the software package
Image J 1.32j (National Institutes of Health, USA). SPRI provides data as
pixel intensity units on a 0–255 scale. SPRI curves used for real-time
kinetic experiments were obtained by plotting versus time the
difference between the integrated areas of preselected regions of
interest (ROIs) of successive SPR images of the chip. As specified by
the manufacturer, raw data were converted in percentage of
ð1Þ
where l is the distance (perpendicular to the flow direction) that
spherical particles will diffuse in time t, while D is the diffusion
coefficient which is, for a given temperature and solvent viscosity, a
Fig. 1. The image shows the Y-shaped flow cell built as described in the text. The PDMSmade flow cell is put in contact to the SPRI gold sensor surface. Solutions are then
flowed inside the pipes by a peristaltic pump.
84
G. Grasso et al. / Microchemical Journal 93 (2009) 82–86
reflectivity (%R), or Δ%R in the case of difference images, by using the
formula:
kR = 1004 0:85 Ip = Is
where Ip and Is refer to the reflected light intensity detected using pand s-polarized light, respectively. Kinetic experiments were carried
out by sequentially acquiring 15 frames averaged SPR images with 3 s
time delay between them.
Briefly, in order to study the DSA/AF interaction two different
solutions were eluted in the two branches of a Y-shaped microchannel
in contact with a functionalized gold substrate:
i. 2.08 μM AF in PBS (immobilization of AF);
ii. PBS (reference).
In order to minimize the non-specific interaction of the flowing
biomolecules and the SPRI chip surface, a solution 4 mM mPEG in
0.1 M TEA is then flowed in both channels. Afterwards, a DSA solution
is eluted in both channels and its interaction with the immobilized AF
is monitored in one half of the channel, while the other half is used for
signal referencing. In particular, couples of ROIs (for signal and
reference) were selected from each half of the channel. Each ROI for
signal was no more than 200 μm far from the reference ROI.
Advantages of such approach are discussed in the Results and
discussion section, while the typical Δ%R change over time for the
described elution sequence is reported in Fig. 2.
SPRI kinetic data were firstly fitted by assuming a simple 1:1
interaction model between the immobilized AF and the flowing DSA
by using the following equation resulting from the integration of the
rate equation [30]:
ΔkR =
Cka ΔRmax
f1 − exp½−ðka C + kd Þt g
ka C + kd
where C is the analyte concentration, ΔRmax is the SPRI response at the
surface saturation, ka and kd are the adsorption and desorption rate
constants, respectively. (see Fig. 3).
Kinetic data were also fitted by assuming a surface heterogeneity
model [31,32]. In this case the ClampXP data analysis package [33] was
used. Data fittings that produced the average values reported in
Fig. 3. Time-dependent SPRI curves obtained after the adsorption of differently
concentrated (0.7 μM, 1.0 μM and 1.3 μM) DSA solutions on the surface immobilized AF
and in-line reference correction. Kinetics parameters were obtained as explained in the
text. Dashed lines show results obtained when a pseudo first-order kinetic model was
considered. Gray lines show the fits obtained when a surface heterogeneity model was
chosen. Residuals for the non-linear fits obtained in the latter case are also shown.
Table 1 were obtained with a residual randomly scattered around the
baseline with amplitude lower than ±0.25 Δ%R (Fig. 3) roughly
equivalent to the typical instrumental noise of the SPRI apparatus.
3. Results and discussion
One of the major challenges encountered in SPRI investigations is
represented by the search for appropriate signal referencing. In-line
referencing is the best solution to the minimal manipulation of the
substrate and the use of parallel microchannels with different flowing
solution (one of the microchannels is adopted as reference cell)
represents an appealing approach [34]. Nevertheless, this configuration can give rise to misleading referencing, mainly due to two
combining factors: the spatial limitation in the microchannels
proximity and the spatial dependent surface inhomogeneity [23]. In
order to prove this point, the SPRI signal difference from two parallel
microchannels was recorded as a function of the distance between the
two microchannels where the same solution (ultra-pure water in this
case) was flowed and the results are reported in Fig. 4. In this case the
rather large error bars are the result of the inevitable difference in SPRI
signal registered within the same microchannel due to the above
mentioned factors. Nevertheless, it is evident from the graph that
differences in the SPRI signal between different microchannels tend to
increase proportionally with the distance between them. A partial
solution to this problem is to build microfluidic platforms where fluids
in different microchannels are flowed very close in space. Unfortunately, there is a practical impossibility to build platforms having
Table 1
Kinetics parameters for DSA/AF interaction obtained from the SPRI experiments when
an interaction model assuming heterogeneity in the surface sites was taken into
account.
Ka1 (M− 1 s− 1)
Fig. 2. Change in percent reflectivity over time for the interaction between the
immobilized AF and the eluted DSA (1.0 μM). All the experimental steps are indicated in
the graph and explained in the text. The two lines refer to two different 200 μm far ROIs
chosen on the Y-shaped microchannel for in-line referencing the SPRI raw data.
DSA/AF
Literature⁎
1.7(± 0.8) × 10
5.7 × 105
5
Kd1 (s− 1)
Ka2 (M− 1s− 1)
−4
7.0(± 0.5) × 10
1.3 × 10− 3
1.2(± 0.5) × 10
Kd2 (s− 1)
4
6.9(± 0.4) × 10− 4
A comparison with values reported in the literature is also reported.
⁎Ref. [37]. Data refer to the interaction of DSA with glycopeptides that possess a single
N-linked sugar chain and that were obtained by digesting AF with lysyl endopeptidase.
G. Grasso et al. / Microchemical Journal 93 (2009) 82–86
Fig. 4. Average difference in the SPRI response obtained by selecting differently distant
ROIs from image areas corresponding to two microchannels within which the same
fluid was flowing (water, flow rate of 0.5 ml min− 1). Error bars represent the dispersion
(standard deviation) of the pixel values calculated from replicated measurements
(different ROIs) from the same image. The point on the far left refers to detection carried
out within the largest microchannel of the Y-shaped flow cell. The shown results refer to
the specific optical system used for the study.
microchannels as close as appropriate referencing would impose. On
the contrary, spatial limitation in the microchannels proximity and
spatial dependent surface inhomogeneity are almost eliminated by
the use of a Y-shaped microchannel (point on the far left of the graph
of Fig. 4). In this case the referencing is carried out in the same channel
so that surface inhomogeneities are dramatically reduced as sample
and reference solutions flow very close in space (sample and reference
areas are typically a few hundred microns away).
In order to visualize this phenomenon, in Fig. 5 an SPR image of two
different solutions (pure water and a 10% ethanol in water solution)
flowing in the two inlets of a Y-shaped microchannel is shown. The two
solutions occupy similar areas of the channel as their viscosity coefficients
do not differ considerably. The absence of appreciable mixing occurs also
in the case of solutions typically used for the study of biomolecular
interactions. As an example it is possible to take into account a DSA
solution (DSA in PBS buffer) and the pure PBS buffer flowing in the two
inlets of a Y-shaped microchannel. The contact time between the
solutions (volume of the mixing microchannel = 0.16 μl, flow rate
0.5 ml min− 1) is only t≈0.02 s while a value of D=9.5×10− 7 cm2/s
can be considered for DSA (value obtained from ClampXP software. This
value is in the range of values reported for similar proteins. As an example
Fig. 5. SPR image of a Y-shaped microchannel where two different solutions (pure water
and a solution of ethanol in water 10%) are eluted in the microfluidic device. The two
solutions occupy similar areas of the channel as their viscosity coefficients do not differ
considerably and no significant mixing is observed.
85
for the bovine serum albumin (BSA), having a MW of 66,000 Da, it is
D=6.3×10− 7 cm2/s) [35]. By considering Eq. (1), it is possible to
estimate a distance of 45 μm perpendicular to the flow direction to be
affected by the diffusion at the end of the Y-shaped microchannel. DSA
diffusion can be therefore considered negligible in the Y-shaped
microchannel whose lateral dimension is, in our case, of 1 mm so that
adjacent solutions (the DSA solution and the pure buffer solution) can be
treated as if they were flowing in two different microchannels. Clearly, a
similar conclusion can be drawn for all the other biomolecules that are
normally investigated by SPRI technique as the value of D is usually in the
range of 10− 6 ÷10− 8 cm2/s.
SPRI investigation with in-line referencing can then be carried out
by choosing as reference ROIs of the Y-shaped microchannel where
non-specific interactions are expected to occur. Reference ROIs can be
selected very close to the signal ROIs (about 200 μm far in our case) as
to optimize the signal referencing. Lacks in the homogeneity of the
local SPRI chip illumination can be also easily minimized. It is
important to highlight that, if the reference choice is wrong, there
could be a difference in the final resonance shifts, so that the kinetic
curve observed for a particular biomolecular interaction is affected by
severe errors.
It is obvious from the above explained reasons that in-line
referencing within the same microchannel where two fluids flow in
laminar mode has to be considered as the best reference choice for
SPRI investigations.
In order to evaluate the applicability of the new experimental
setup, DSA/AF interacting system was chosen as model.
In order to correct raw data for non-specific adsorption and
changes in bulk refractive index, Δ%R for each ROI was normalized to
the average of the Δ%R measured for the background ROIs obtained
from adjacent regions in the Y-shaped microchannel where nonspecific interactions were expected to occur. In Fig. 6 the SPR
difference image acquired after DSA transit in the Y-shaped microchannel is shown and the brighter area is the result of the specific
DSA/AF interaction. Correspondently, changes in percent reflectivity
over time for the two different areas of the Y-shaped microchannel are
reported in Fig. 2. The power of the SPRI approach is clearly visible by
comparing Figs. 2 and 6. The operator can easily visualize the
interaction processes under investigation, choose the most appropriate ROIs for his purposes and obtain the correspondent kinetic data.
The method is less vulnerable to random factors such as imperfections
of the gold substrate, variations in the illumination etc. that are often
the cause of errors in the calculation of kinetics parameters by SPR as
well as less time consuming.
In order to obtain the kinetics parameters for our model system,
we carried out the same SPRI experiment with three different
concentrations of DSA (0.7 μM, 1.0 μM and 1.3 μM respectively. See
Fig. 3). The SPRI kinetic data were fitted by assuming an interaction
model that differs from the simple 1:1 interaction. A deviation from
Fig. 6. SPR difference image acquired after DSA elution in the Y-shaped microchannel.
The brighter area is due to the specific interaction between the previously immobilized
AF and the eluted DSA. Concentration of the latter was 1.0 μM in this case. The presence
of two small air bubbles is visible as black spot in the image, but they do not interfere
with the kinetics parameters calculations as long as proper ROIs are selected.
86
G. Grasso et al. / Microchemical Journal 93 (2009) 82–86
the simple first-order kinetics was expected on the basis of the well
known general multivalency of glycoprotein–lectin interaction [31,32]
and was verified by fitting SPRI kinetic data with a first-order kinetic
rate equation [36] (Fig. 3, dashed lines). When kinetic data were fitted
by using an interaction model (see Experimental section) that
assumes heterogeneity in the active surface sites (Fig. 3, gray lines)
better results were obtained. In particular, adsorption and desorption
rate constants in the range expected for the selected biomolecular
system [37] were obtained (Table 1), thus supporting the validity of
the new approach that allows in-line referencing without timeconsuming gold substrate/microfluidic interface manipulation that
could disrupt the system.
4. Conclusions
An SPRI experimental setup based on the use of a Y-shaped
microchannel for in-line referencing is discussed. The advantages of
the method have been discussed and a comparison with other SPRI
experimental approaches has been outlined. In particular, the
methods improves the possibilities for in-line referencing offered by
devices using parallel microchannels. The latter can give rise to
misleading referencing, mainly due to two combining factors: the
spatial limitation in the microchannels proximity and the spatial
dependent surface inhomogeneity. Moreover, the multistep process
typically required as far as SPRI experiments using parallel microchannel devices are going to be concerned is now avoided and the
potential introduction of artefacts in SPRI data minimized.
In order to show the applicability of the newly designed SPRI
procedure, an example of a real biomolecular interaction has been
investigated and the DSA/AF interaction has been used as a model
system. Kinetic parameters have been determined and the results
confirmed the suitability of the experimental approach for real-time
kinetic SPRI experiments.
This work shows some of the results that can be achieved by
coupling the SPRI system with microfluidic devices. The microfluidic
approach represents one of the most suitable ways for making the
most of the SPRI technique as long as a proper patterning of the
surface can be achieved. The described procedure can be expanded to
much more sophisticated microfluidic patterns and this will widen the
applicability of the SPRI technique, giving to the latter the prominent
role that deserves for the investigation of biomolecular interactions.
Acknowledgments
We thank MIUR (PRIN 2007 n 2007F9TWKE) for financial support.
The authors wish to acknowledge the helpful comments and suggestions
made by Prof. F. Lisdat and Dr. S. Weibel.
References
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]
[33]
[34]
[35]
[36]
[37]
G.M. Whitesides, Nature 442 (2006) 368–373.
T.M. Squires, S.R. Quake, Rev. Mod. Phys. 77 (2005) 977–1026.
K. Ohno, K. Tachikawa, A. Manz, Electrophoresis 29 (2008) 4443–4453.
J. West, M. Becker, S. Tombrink, A. Manz, Anal. Chem. 80 (2008) 4403–4419.
B. Wolf, M. Brischwein, H. Grothe, C. Stepper, J. Ressler, T. Weyh, Microsystems 16 (2007)
269–307.
X. Liang, S.Y. Chou, Nano Lett. 8 (2008) 1472–1476.
S.Y. Jung, Y. Liu, C.P. Collier, Langmuir 24 (2008) 4439–4442.
G. Grasso, M. Fragai, E. Rizzarelli, G. Spoto, K.J. Yeo, J. Am. Soc. Mass Spectrom. 18 (5)
(2007) 961–969 18.
C.J. Kastrup, J.Q. Boedicker, A.P. Pomerantsev, M. Moayeri, Y. Bian, R.R. Pompano, T.R.
Kline, P. Sylvestre, F. Shen, S.H. Leppla, Nat. Chem. Biol. 4 (2008) 742–750.
S.Y. The, R. Lin, L.H. Hung, A.P. Lee, Lab Chip 8 (2008) 198–220.
J.D. Taylor, M.J. Linman, T. Wilkop, Q. Cheng, Anal. Chem. 81 (2009) 1146–1153.
R. D'Agata, G. Grasso, G. Spoto, Open Spectr. J. 1 (2008) 1–9.
J. Homola, Chem. Rev. 108 (2008) 462–493.
J.S. Shumaker-Parry, C.T. Campbell, Anal. Chem. 76 (2004) 907–917.
G. Arena, A. Contino, E. Longo, C. Sgarlata, G. Spoto, V. Zito, Chem. Commun.16 (2004)
1812–1813.
A.W. Wark, H.J. Lee, R.M. Corn, in: R.B.M. Schasfoort, A.J. Tudos (Eds.), Handbook of
Surface Plasmon Resonance, The Royal Society of Chemistry, 2008, pp. 246–274.
R. D'Agata, G. Grasso, G. Iacono, G. Spoto, G. Vecchio, Org. Biomol. Chem. 4 (2006)
610–612.
H.J. Lee, A.W. Wark, R.M. Corn, Langmuir 22 (2006) 5241–5250.
R. D'Agata, R. Corradini, G. Grasso, R. Marchelli, G. Spoto, ChemBioChem 9 (2008)
2067–2070.
R.J. Ober, E.S. Ward, Anal. Biochem. 271 (1999) 70–80.
J.S. Shumaker-Parry, M.H. Zareie, R. Aebersold, C.T. Campbell, Anal. Chem. 76 (2004)
918–929.
A.R. Wheeler, S. Chah, R.J. Whelan, R.N. Zare, Sens. Actuators, B 98 (2004) 208–214.
H.J. Lee, T.T. Goodrich, R.M. Corn, Anal. Chem. 73 (2001) 5525–5531.
E. Fu, T. Chinowsky, K. Nelson, K. Johnston, T. Edwards, K. Helton, J. Grow, J.W.
Miller, P. Yager, Ann. N.Y. Acad. Sci. 1098 (2007) 335–344.
A. Hatch, A.E. Kamholz, K.R. Hawkins, M.S. Munson, E.A. Schilling, B.H. Weigl, P.
Yager, Nat. Biotechnol. 19 (2001) 461–465.
A.E. Kamholz, E.A. Schilling, P. Yager, Biophys. J. 80 (2001) 1967–1972.
J. Chun-Ping, W. Chung-Yi, L. Yu-Cheng, W. Ching-Yi, Lab Chip 3 (2003) 77–81.
G. Grasso, M. Fragai, E. Rizzarelli, G. Spoto, K.J. Yeo, J. Mass Spectrom. 41 (2006)
1561–1569.
B.P. Nelson, A.G. Frutos, J.M. Brockman, R.M. Corn, Anal. Chem. 71 (1999)
3928–3934.
D.J. Oshannessy, M. Brighamburke, K.K. Soneson, P. Hensley, I. Brooks, Anal.
Biochem. 212 (1993) 457–468.
S.S. Komath, M. Kavithab, M.J. Swamy, Org. Biomol. Chem. 4 (2006) 973–988.
M. Ambrosi, N.R. Cameron, B.G. Davis, Org. Biomol. Chem. 3 (2005) 1593–1608.
D.G. Myszka, T.A. Morton, Trends Biochem. Sci. 23 (1998) 149–150.
H.-B. Pyo, Y.-B. Shin, M.-G. Kim, H.C. Yoon, Langmuir 21 (2005) 166–171.
C.T. Culbertson, S.C. Jacobson, J.M. Ramsey, Talanta 56 (2002) 365–373.
T.A. Morton, D.G. Myszka, I.M. Chaiken, Anal. Biochem. 227 (1995) 176–185.
Y. Shinohara, F. Kim, M. Shimizu, M. Goto, M. Tosu, Y. Hasegawa, Eur. J. Biochem.
223 (1994) 189–194.