Analysis of dopamine and tyrosinase activity on Ion-Sensitive Field-Effect Transistor (ISFET) devices

Article abstract of DOI:10.1002/chem.200700734

Dopamine (1) and tyrosinase (TR) activities were analyzed by using chemically modified ion-sensitive fieldeffect transistor (ISFET) devices. In one configuration, a phenylboronic acid functionalized ISFET was used to analyze 1 or TR. The formation of the boronate-1 complex on the surface of the gate altered the electrical potential associated with the gate, and thus enabled 1 to be analyzed with a detection limit of 7×10^-5 M. Similarly, the TR-induced formation of 1, and its association with the boronic acid ligand allowed a quantitative assay of TR to be performed. In another configuration, the surface of the ISFET gate was modified with tyramine or 1 to form functional surfaces for analyzing TRactivities. The TR-induced oxidation of the tyramine- or 1-functionalized ISFETs resulted in the formation of the redox-active dopaquinone units. The control of the gate potential by the redox-active dopaquinone units allowed a quantitative assay of TR to be performed. The dopaquinone-functionalized ISFETs could be regenerated to give the 1-modified sensing devices by treatment with ascorbic acid.

Full text of DOI:10.1002/chem.200700734

FULL PAPER
DOI: 10.1002/chem.200700734
7288
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 7288 – 7293

FULL PAPER
Analysis of Dopamine and Tyrosinase Activity on Ion-Sensitive Field-Effect
Transistor (ISFET) Devices
[
a]
Ronit Freeman, Johann Elbaz, Ron Gill, Maya Zayats, and Itamar Willner*
Abstract: Dopamine (1) and tyrosinase
TR) activities were analyzed by using
chemically modified ion-sensitive field-
effect transistor (ISFET) devices. In
duced formation of 1, and its associa-
tion with the boronic acid ligand al-
lowed a quantitative assay of TR to be
performed. In another configuration,
the surface of the ISFETgate was
modified with tyramine or 1 to form
functional surfaces for analyzing TR
activities. The TR-induced oxidation of
the tyramine- or 1-functionalized
ISFETs resulted in the formation of
the redox-active dopaquinone units.
The control of the gate potential by the
redox-active dopaquinone units al-
lowed a quantitative assay of TR to be
performed. The dopaquinone-function-
alized ISFETs could be regenerated to
give the 1-modified sensing devices by
treatment with ascorbic acid.
(
one configuration,
a phenylboronic
acid functionalized ISFETwas used to
analyze 1 or TR. The formation of the
boronate–1 complex on the surface of
the gate altered the electrical potential
associated with the gate, and thus ena-
bled 1 to be analyzed with a detection
Keywords: biosensors · enzymes ·
ion-sensitive field-effect transistors ·
monolayers · tyrosinase
À5
limit of 7”10 m. Similarly, the TR-in-
Introduction
3-(3,4-dihydroxyphenyl)alanine (l-dopa) or 1, which are fur-
ther oxidized by the enzyme to form their respective qui-
none products (Scheme 1).
There has been growing interest in ion-sensitive field-effect
transistors (ISFETs) as label-free transducers for biorecogni-
tion events. Different ISFET-based biosensors have been
developed, which include devices that monitor biocatalytic
[1]
[2]
transformations, that probe antigen–antibody binding pro-
[3]
[4]
Scheme 1. Oxidation of 3 to 1, and subsequently to dopaquinone by the
enzyme tyrosinase.
cesses and DNA hybridization, and very recently, that
follow the formation of complexes between aptamers and
[5]
low molecular weight substrates. In all of these biosensor
systems the biocatalytic processes or the biorecognition
events alter the gate potential of the ISFETdevices, thus al-
lowing the electronic transduction of the sensing processes.
In the present study we demonstrate the detection of dop-
amine (1) by a functionalized ISFETdevice, and also dem-
onstrate the use of three different ISFETsystems to follow
the activity of tyrosinase (TR). TR oxidizes phenol deriva-
tives, such as tyrosine or tyramine (3), in the presence of O2
to form their respective catechol derivatives, for example, l-
Elevated amounts of TR were detected in melanoma
cancer cells, and the enzyme is considered to be a marker
for this type of malignant cells. It was also reported that
[6]
the loss of 1 in neurons might cause diseases such as Parkin-
[7]
sonꢁs disease. Compound 1 is a central neurotransmitter,
and its sensitive detection, particularly by using integrated
miniaturized devices, could be valuable for the invasive
monitoring of neural responses. Different methods to ana-
[8]
lyze 1 were reported, which included electrochemical
means or optical methods that involved the 1-induced for-
[9]
mation of Au nanoparticles. TR was probed electrochemi-
cally by the enzyme-stimulated formation of 1 on electrodes
[
a] R. Freeman, J. Elbaz, R. Gill, Dr. M. Zayats, Prof. I. Willner
Institute of Chemistry
The Hebrew University of Jerusalem
Jerusalem 91904 (Israel)
Fax : (+972)2-652-7715
E-mail: willnea@vms.huji.ac.il
and the voltammetric detection of the product by labeling it
[10]
with the redox-active ferrocene–boronic acid.
Similarly,
TR activity was monitored through the biocatalyzed oxida-
tion of tyrosine to form l-dopa and the product-mediated
formation of Au nanoparticles, which were optically detect-
Chem. Eur. J. 2007, 13, 7288 – 7293
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7289

I. Willner et al.
[
9]
ed. Alternatively, CdSe quantum dots (QDs) were func-
tionalized with methyl ester of tyrosine, and the enzyme-cat-
alyzed oxidation of the substrate to dopaquinone resulted in
the fluorescence quenching of the QDs, thus enabling the
[11]
optical biosensing of TR activity.
Results and Discussion
Scheme 2 shows one configuration for detecting 1 and ana-
lyzing TR activity. The Al O gate of the ISFETdevice was
2
3
functionalized with an aminopropylsiloxane film that was
further treated with 4-carboxyphenyl boronic acid (2). The
resulting 2-functionalized gate was then employed as the
active surface for the association of 1 (Scheme 2, route a).
The formation of the charged boronate complex on the sur-
face of the gate alters the gate potential, thus enabling the
quantitative analysis of 1 by the ISFETdevice. Figure 1A
shows the gate-to-source potential changes of the 2-func-
tionalized device after interaction with different concentra-
Figure 1. A) Calibration curve that corresponds to the analysis of 1 by
the 2-functionalized ISFETdevice. B) Calibration curve that corresponds
to the analysis of different concentrations of TR.
resulting mixture that consisted of 1 and dopaquinone was
then reduced with citric acid to give 1 as the sole product,
and the resulting solution was analyzed by using the ISFET
device (Scheme 2, route b). The association of 1 with the
boronic acid ligand resulted in the changes in potential on
the surface of the gate (Figure 1B). As the concentration of
TR increased, the concentration of 1 generated also in-
creased and the potential changes were enhanced. The gate
potential leveled off at a concentration of around 1.2 U of
TR, and the detection limit for the analysis of TR was
around 0.25 U. It should be noted that treatment of the 2-
functionalized gate with ther-
tions of 1 for 20 min. The detection limit for analyzing 1 is
À5
7
”10 m, which is very similar to the detection limits for
[8]
[9]
electrochemical or optical methods previously reported.
The 2-functionalized ISFETdevice was also employed to
analyze TR activity. To achieve this aim, 3 was treated with
different concentrations of TR for a fixed time interval of
20 min before the enzyme was thermally deactivated. The
mally deactivated TR within
the same concentration range
resulted in minute potential
changes (DV=2 mV), which
implies that nonspecific adsorp-
tion of TR to the gate is not oc-
curring, and thus the potential
changes are as a result of the
formation of the 1–boronate
complex.
Although the analysis of TR
by the 2-functionalized ISFET
device is feasible, the activation
of the biocatalytic oxidation of
3
in solution, the need to ther-
mally deactivate the enzyme,
and the subsequent association
of the reaction product onto
the gate surface, turn the ana-
lytical protocol into a complex
procedure. To facilitate the
analysis of TR, we looked to
design a surface-confined pro-
cess that would enable the sep-
aration of the reaction product
from the enzyme, and would
eliminate the time interval re-
quired to bind the reaction
product to the functionalized
Scheme 2. Schematic representation of the detection of 1 (route a) and the analysis of TR activity (route b) by
using a 2-functionalized ISFETdevice.
7290
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 7288 – 7293

Ion-Sensitive Field-Effect Transistor
FULL PAPER
Scheme 3. Analysis of TR activity by using a 3-functionalized ISFETdevice (route a) and a 1-functionalized ISFETdevice (route b).
gate. Scheme 3 shows the configuration of the ISFETdevice
for the analysis of TR on the surface of the gate. The amino-
propylsiloxane-functionalized gate was treated with glutaric
dialdehyde, and subsequently with 3 (Scheme 3, route a).
The modified surface was then treated with TR (under air)
to yield the 1–dopaquinone mixed film. The subsequent re-
duction of the dopaquinone units with ascorbic acid yielded
the 1-modified gate, in which the gate potential was control-
led by the quantity of redox-active 1 present. Note that the
biocatalytic transformation on the surface of the gate, and
the subsequent reduction of the surface-confined reaction
product by ascorbic acid allowed the stepwise separation of
Figure 2. Time-dependent gate-to-source potential changes (DVgs) upon
analyzing the activity of TR (2 U). Measurements were performed in
phosphate buffer (0.01m, pH 6.3).
the 1-modified gate from the enzyme and the reducing
agent, and the rapid generation of clean 1-modified ISFET
devices.
Figure 2 shows the changes in the gate potentials upon
treating the 3-modified ISFETwith TR (2 U) for different
time intervals, according to route (a) in Scheme 3. As the re-
action time increased, the changes in the gate potential were
enhanced, and they leveled off to a saturation value of
with 1 (Scheme 3, route b). This device was fabricated by
the covalent attachment of 1 to the glutaric dialdehyde
modified gate (instead of 3).
Curve a in Figure 3A shows the changes in the potentials
of the gate modified with 1 upon treatment with different
concentrations of TR for a fixed time interval of 20 min.
The changes in the gate potentials increase as the concentra-
tion of TR increases, and they level off at a TR concentra-
tion of 1.6 U. Each of the 1-modified gates treated with TR
(results shown in Figure 3A, curve a) was subsequently
treated with ascorbate. The base potential of the 1-modified
gates was observed for all of the devices (Figure 3A, curve
DV =70 mV after about 15 min. These results are consistent
gs
with the increase in the concentration of 1 on the surface of
the gate as the reaction with TR is prolonged. When 3 is
fully oxidized, the gate potential reaches its saturation
value. To prove that the TR-induced oxidation of 3 leads to
the formation of dopaquinone and that the reduction of the
latter by ascorbate is essential to yield the 1-functionalized
gate, the gate of an ISFETdevice was directly modified
Chem. Eur. J. 2007, 13, 7288 – 7293
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7291

I. Willner et al.
could be useful sensing devices to monitor TR activity for
the rapid detection of melanoma cancer cells. In the present
studies we used two different principles to control the gate
potential. The first approach involved the alteration of the
change associated with the gate by linking the product of
the biocatalytic transformation with the boronic acid ligand
that was attached to the ISFETdevice. Th e second method
to control the gate potential included the generation of a
redox-active product (dopaquinone) that gave an electro-
chemical potential on the surface of the gate. All previous
methods for the analysis of TR activity involved metallic
[
9]
[11]
nanoparticles, semiconductor QDs,
or functionalized
Figure 3. A) Gate-to-source potential changes upon analyzing different
concentrations of TR by using the same ISFET device functionalized
with 1. Curve a) shows the potential changes as a result of oxidation of 1
by different concentrations of TR that yield the dopaquinone. Curve b)
shows the changes in the gate-to-source potential after treatment of the
dopaquinone-modified ISFETwith ascorbic acid (1 m m) in phosphate
buffer (0.01m, pH 6.3). The results indicate the regeneration of the base
potential of the ISFETdevice prior to analysis of any of the TR concen-
trations. B) Gate-to-source potential changes upon analyzing different
concentrations of TR by using the 3-modified ISFET. Measurements
were performed in phosphate buffer (0.01m, pH 6.3).
[10]
redox molecules, as labels for the optical or electrochemi-
cal detection of TR activity. The present study demonstrates
a label-free procedure for analyzing TR activities with com-
parable sensitivities to previous methods.
Experimental Section
Materials: Compounds 1, 2, 3, TR, 3-aminopropyltriethoxysilane, 1-[3-
(
dimethylamino)propyl]-3-ethylcarbodiimide hydrochloride (EDC), as-
corbic acid, citric acid, and glutaric dialdehyde were purchased from
Sigma or Aldrich and used as supplied. Ultrapure water obtained from
Barnstead NANOpure DIamond was used in all experiments.
b), which implies that all of the enzyme-generated dopaqui-
none was reduced to form the 1-functionalized surfaces. In
fact, the TR-stimulated oxidation of 3 to dopaquinone and
the back reduction of dopaquinone to 1 are fully reversible
Preparation of the modified ISFET devices: The primary modification of
the Al O gate of an ISFETdevice was achieved by treating the ISFET
2 3
with a solution of 3-aminopropyltriethoxysilane (0.2 mL, 10% (v/v) in tol-
uene) at room temperature for 12 h.
(
Figure 3A). These results confirm that the oxidation of 3
Preparation of the 2-functionalized ISFET: The silylated chips were thor-
oughly rinsed with toluene followed by a HEPES buffer solution (10 mm,
pH 6.3), and were subsequently dried in air at room temperature for
30 min. Compound 2 was then covalently linked to the aminosiloxane-
functionalized gate interface by treating the gate with HEPES buffer
on the surface lead to the formation of dopaquinone, which
is reduced by ascorbate to form the 1-functionalized surfa-
ces. Furthermore, the results imply that the 1-functionalized
ISFETdevices may be used as biosensors to follow TR ac-
tivity, and that the reduction of the dopaquinone product
may be used as a means of regenerating the interface.
(
0.2 mL) solution that contained 2 (1 mm) and EDC (10 mm) for 2 h. The
resulting modified ISFETdevices were thoroughly rinsed with the
HEPES buffer solution and dried in air for 30 min.
The development of the 3-modified ISFETdevice as a
bioelectronic device that transduces the biocatalyzed forma-
tion of 1 on the surface of the gate enabled us to construct a
new biosensor configuration for analyzing TR activity. Fig-
ure 3B shows the changes in the potential of the gate modi-
fied with 3 upon the analysis of different concentrations of
TR. In these experiments, the 3-modified ISFETs were
treated with different concentrations of TR (for a fixed time
interval of 20 min), and the resulting dopaquinone-function-
alized gates were reduced with ascorbate (Scheme 3, route
a). As the concentration of TR was increased, the change in
Preparation of the 1- or 3-functionalized ISFETs: Compound 1 was cova-
lently linked to the aminosiloxane-functionalized gate interface by treat-
ing of the gate with glutaric dialdehyde (0.2 mL, 10% (v/v) solution in
water) at room temperature for 20 min. The chips were rinsed with water
and then with a HEPES buffer solution (10 mm, pH 6.3) and then treated
with 1 (0.2 mL, 1 mm in HEPES buffer solution) for 20 min. The 3-modi-
fied gates were prepared by the same procedure as described for 1.
Potentiometric ISFET measurements: ISFETdevices (IM T, Neuchâtel,
Switzerland) with Al O gates (20 mm”700 mm) were used in all experi-
2
3
ments. A conventional Ag/AgCl electrode was used as a reference elec-
trode. The 2-functionalized gate was immersed in a working cell com-
posed of phosphate buffer solution (0.8 mL, 0.01m, pH 8.8), which con-
tained 1 (1 mm). To follow the time-dependent potential changes on the
surface of the gate, the measurement of the gate-to-source potentials
were monitored for a time interval of 30 min. To extract the calibration
curve for 1, the 2-functionalized ISFETs were exposed to various concen-
trations of 1 for 20 min, and the resulting devices were immersed in a
working cell composed of phosphate buffer solution (0.8 mL, 0.01m,
pH 8.8).
the gate-to-source potential (DV ) was enhanced, which is
gs
consistent with the fact that the coverage of 1 on the surface
of the gate was increased.
Conclusion
To follow the activity of TR, a solution of3 (1 mm) was treated with vari-
able concentrations of TR in phosphate buffer (200 mL, 10 mm, pH 6.3)
for a fixed time interval of 20 min. Subsequently, the enzyme was ther-
mally deactivated by heating the sample at 808 for 2 min. The resulting
mixture consisted of 1 and dopaquinone. The mixture was treated with
citric acid (1 mm) for 5 min to reduce to 1 any dopaquinone that had
been generated. The 2-functionalized ISFETdevices were then treated
with the mixtures (0.2 mL) for a time interval of 20 min. The resulting de-
In conclusion, the present study has demonstrated the novel
use of ISFETdevices for the analysis of 1 or for the assay of
TR. The 2-functionalized ISFETcould provide a useful min-
iaturized implantable electronic system for monitoring the
neurotransmitter 1. Similarly, 3- or 1-functionalized ISFETs
7292
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 7288 – 7293

Ion-Sensitive Field-Effect Transistor
FULL PAPER
vices were immersed in a working cell composed of phosphate buffer
[
[
1] a) P. Bergveld, Sens. Actuators B 2003, 88, 1–20; b) S. V. Dzyade-
vych, A. P. Soldtakin, A. V. Elꢁskaya, C. Martelet, N. Jaffrezic-Re-
nault, Anal. Chim. Acta 2006, 568, 248–258; c) M. J. SchÅning, A.
Poghossian, Electroanalysis 2006, 18, 1893–1900.
2] a) S. V. Pogorelova, M. Zayats, A. B. Kharitonov, E. Katz, I. Willner,
Sens. Actuators B 2003, 89, 40–47; b) J. J. Xu, W. Zhao, X. L. Luo,
H. Y. Chen, Chem. Commun. 2005, 792–794; c) O. A. Biloivan, S. V
Dzyadevych, O. A. Boubriak, A. P. Soldtakin, A. V. Elꢁskaya, Elec-
troanalysis 2004, 16, 1883–1889.
(
0.8 mL, 10 mm, pH 6.3).
To follow the TR activity by using the 3-functionalized ISFET, the modi-
fied device was immersed in a working cell composed of phosphate
buffer (0.8 mL, 10 mm, pH 6.3) that contained TR (2 U), and the re-
sponse of the device was monitored for a time interval of 20 min. The
calibration curve for analyzing TR activity by using the 3-modified
ISFETdevices was extracted by interacting the modified sensing devices
with different concentrations of TR in phosphate buffer (0.8 mL, 10 mm,
pH 6.3) for 20 min. The resulting devices were then treated with ascorbic
acid (1 mm) for 5 min to transform all of the dopaquinone that was gener-
ated to 1. The resulting gate was then immersed in a working cell com-
posed of phosphate buffer (0.8 mL, 10 mm, pH 6.3), and the potentiomet-
ric responses of the devices were monitored.
[
3] a) M. Zayats, O. A. Raitman, V. I. Chegel, A. B. Kharitonov, I. Will-
ner, Anal. Chem. 2002, 74, 4763–4773; b) G. A. J. Besselink,
R. B. M. Schasfoort, P. Bergveld, Biosens. Bioelectron. 2003, 18,
1
109–1114.
4] a) J.-K. Shin, D.-S. Kim, H.-J. Park, G. Lim, Electroanalysis 2004, 16,
912–1918; b) F. Uslu, S. Ingebrandt, D. Mayer, S. Bçcker-Meffert,
M. Odenthal, A. Offenhꢂusser, Biosens. Bioelectron. 2004, 19, 1723–
731; c) J. Fritz, E. B. Cooper, S. Gaudet, P. K. Sorger, S. R. Manalis,
[
The analysis of TR activity by the 1-functionalized ISFETwas achieved
by exposing the modified device to samples that contained different con-
centrations of TR outside of the working cell for 20 min. The resulting
device was immersed in a working cell composed of phosphate buffer
1
1
Proc. Natl. Acad. Sci. USA 2002, 99, 14142–14146.
5] M. Zayats, Y. Huang, R. Gill, C. A. Ma, I. Willner, J. Am. Chem.
Soc. 2006, 128, 13666.
6] C. Angeletti, V. Khomitch, R. Halaban, D. L. Rimm, Diagnostic Cy-
topathology 2004, 31, 33–37.
(
0.8 mL, 10 mm, pH 6.3) and its potentiometric response was monitored.
[
[
To regenerate the 1-functionalized sensing ISFETdevice, the TR -treated
ISFETwas then treated with ascorbic acid (1 m m) for 5 min. In all of the
ISFETdevices, the potentiometric signal between the source of the
ISFETand the reference electrode was recorded by using a semiconduc-
tor parameter analyzer (HP 4155B). The parameter that was monitored
was the change in the gate-to-source voltage (Vgs), whereas the drain cur-
[
7] C. W. Olanow, Neurology 1990, 40, 32–37.
[
8] a) M. Lahav, A. N. Shipway, I. Willner J. Chem. Soc. Perkin Trans. 2
1
999, 9, 1925–1931; b) S. M. Strawbridge, S. J. Green, J. H. R.
rent (I
d
) or the drain-to-source voltage (Vds) remained at a constant value
Tucker, Chem. Commun. 2000, 2393–2394; c) S. Tembe, M. Karve,
S. Inamdar, S. Haram, J. Melo, S. F. DꢁSouza, Anal. Biochem. 2006,
3
(I
d
=100 mA or Vds =1.5 V). All of the measurements were carried out
for 5 min at ambient temperatures without stirring to simulate the condi-
tions for possible future in vivo applications. Reproducibility of the meas-
urements was Æ2 mV in a number of experiments (n=3). The detection
49, 72–77.
9] a) R. Baron, M. Zayats, I. Willner, Anal. Chem. 2005, 77, 1566–
571; b) X. M. Feng, C. J. Mao, G. Yang, W. H. Hou, J. J. Zhu, Lang-
muir 2006, 22, 4384–4389.
10] D. Li, R. Gill, R. Freeman, I. Willner, Chem. Commun. 2006, 5027–
029.
[
1
[
12]
limits were estimated according to IUPAC recommendations as the in-
tercept of the two linear segments in the calibration curves.
[
[
5
11] R. Gill, R. Freeman, J. P. Xu, I. Willner, S. Winograd, I. Shweky, U.
Banin, J. Am. Chem. Soc. 2006, 128, 15376–15377.
Acknowledgements
[12] R. P. Buck, E. Lindner, Pure Appl. Chem. 1994, 66, 2527–2536.
This study is supported by the Israel Ministry of Science as an Infrastruc-
ture Project and by the Intel Cooperation.
Received: May 15, 2007
Published online: August 8, 2007
Chem. Eur. J. 2007, 13, 7288 – 7293
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7293