Journal of

Electroanalytical
Chemistry
Journal of Electroanalytical Chemistry 575 (2005) 17
www.elsevier.com/locate/jelechem

Simultaneous monitoring of glucose, lactate, L -glutamate and

hypoxanthine levels in rat striatum by a ow-injection
enzyme electrode array system with in vivo microdialysis sampling
Fen-Fen Zhang a, Qiao Wan a, Chen-Xin Li a, Xiao-Li Wang a, Zi-Qiang Zhu b,
Yue-Zhong Xian a, Li-Tong Jin a,*, Katsunobu Yamamoto c
b

a
Department of Chemistry, East China Normal University, Shanghai 200062, PR China
Department of Electronic Engineering, East China Normal University, Shanghai 200062, PR China
c
BAS Co., Ltd. No. 36-4, 1-Chome, Oshiage, Sumida-Ku, Tokyo 131, Japan

Received 4 June 2004; received in revised form 23 July 2004; accepted 27 July 2004
Available online 18 October 2004

Abstract
A ow-injection enzyme electrode array system with in vivo microdialysis sampling is proposed for the simultaneous measurement of cerebral glucose, lactate, L -glutamate and hypoxanthine concentrations. The enzyme electrode array system was based on
neutral red-doped silica (NRDS) nanoparticles as the electrocatalyst. These uniform NRDS nanoparticles (about 50 3 nm) were
prepared by a water-in-oil microemulsion method, and characterized by the transmission electron microscopy technique. The inside
neutral red dopant maintained its high electron-activity, while the outside nano silica surface prevented neutral red from leaching
out into the aqueous solutions and showed high biocompatibility. These nanoparticles were then mixed with the glucose oxidase,
lactate oxidase, L -glutamate oxidase or xanthine oxidase, and immobilized on the four carbon electrode array, respectively. A thin
Naon
lm was coated on the enzyme layer to prevent interference such as from ascorbic acid and uric acid in the dialysate. The
proposed ow-injection analysis with the NRDS-enzyme electrode array system enables simultaneous monitoring of trace levels of
glucose, L -glutamate, lactate and hypoxanthine in rat striatum.
 2004 Elsevier B.V. All rights reserved.
Keywords: Neutral red-doped silica nanoparticle; In vivo microdialysis sampling; Rat striatum; Flow-injection analysis; Enzyme electrode array

1. Introduction
Rapid measurement of glucose, L -glutamate, and lactate is important in understanding the dynamics of the
energy balance in brain tissue [1,2]. L -glutamate is also
the main excitatory neurotransmitter [1,3], while hypoxanthine is a major metabolite in the degradation of
adenine nucleotide [4,5], and the accumulation of adenosine production is central to its function in the nervous
system and can therefore be expected to dier consider*

Corresponding author. Fax: +86 21 6223 2627.

E-mail address: fenfenzhang@yahoo.com.cn (L.-T. Jin).

0022-0728/$ - see front matter  2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.jelechem.2004.07.039

ably from those of conventional neurotransmitters [6].

Therefore, the simultaneous monitoring of glucose, lactate, L -glutamate and hypoxanthine would be of great
benet for studies on the energy metabolites and neurons communicating in the brain. However, their concentrations in the extracellular brain environment are
still poorly documented [7]. Microdialysis is a powerful
tool for minimally invasive probing of such metabolisms
[8]. In addition, the use of microdialysis probes permits
the release of active agents to the monitoring site and the
calculation of local metabolic turnover rates of the tissue
via clearance methods [9]. In many cases, this sampling
method is generally coupled to high-performance liquid

F.-F. Zhang et al. / Journal of Electroanalytical Chemistry 575 (2005) 17

chromatography (HPLC) or capillary electrophoresis

(CE) [1012]. Nevertheless, microdialysis combined with
an enzymatic electrode possesses simplicity of operation
and substrate selectivity of the enzyme [13].
Neutral red has been reported to act as the electrocatalyst for NADt/NADH regeneration [1416]. NR also
shows catalytic activity toward DNA [17]. Compared
to the traditional methods for dye immobilization onto
the electrode surface (electropolymerization [14], adsorption [17]), dye doped silica nanoparticles synthesized by
using a water-in-oil (W/O) microemulsion method are
advantageous, providing both prolonged long-term stability and showing high biocompatibility [18].
In the present work, we have rst developed novel
electro-active neutral red-doped silica (NRDS) nanoparticles [1921] in a four enzyme electrode array with high
sensitivity and long-term stability. The experiments indicated that the inside neutral red dopant maintained its
high electron-activity as an electrocatalyst, while the
outside nano silica surface prevented NR from leaching
out into the aqueous solutions and showed high biocompatibility. For in vitro monitoring of brain dialysate
with electrochemical biosensors, L -ascorbic acid (L AA) causes major interference. Naon
lm was
dripped on the four sensor array to exclude such electrooxidizable interferants [2123]. The ow-injection performance of this newly prepared NRDS-enzyme
electrode array based on these uniform nanoparticles
and the application to simultaneous determination of
glucose, lactate, L -glutamate and hypoxanthine in in vitro rat brain dialysate were investigated. The FIA system with the NRDS-enzyme electrode array showed
high sensitivity, wide ranges of response, and was without electroactive interferences.

2. Experimental
2.1. Reagents
Glucose oxidase (GOD, from Aspergillus niger, EC
1.1.3.4. 150 000 U g 1), L -glutamate oxidase (L -GLOD,
EC 1.4.3.11, from Streptomyces sp., 10.8 U mg 1), lactate oxidase (LOD, from Pediococcus species, 37
U mg 1), xanthine oxidase (XOD, EC 1.1.3.22, Grade
I from Buttermilk, 1 U (mg protein) 1, 34.7 mg protein ml 1), b-D -glucose, L -glutamate, lactate, hypoxanthine, ascorbic acid, uric acid and Naon
(1% methyl
alcohol) were purchased from Sigma Chemical Co.
Tetrathyl orthosilicate (TEOS, 98%) was purchased
from United Chemical Technologies (Bristol, PA). Bovine serum album (BSA) was obtained from Huamei
Biochemical (Shanghai, China). Glucose stock solution
was allowed to mutarotate for 24 h before use. Other
reagents were of at least analytical-reagent grade. All
solutions were prepared using twice distilled water.
2.2. Apparatus
The ow-injection system consisted of a LC-10 AS
eluent delivery pump and an SIL -6B injector equipped
with a 20 ll sample loop (Shimadzu, Tokyo, Japan).
Flow-injection amperometric data were collected using
a CHI 1030 workstation (CH instruments, Inc.).
The homemade thin-layer cell consisted of an SCE as
the reference electrode, a gold ake as the counter electrode and four NRDS-enzyme modied carbon-disk
electrodes, 300 lm in diameter, as the working electrode
array. The structure of the homemade thin-layer cell is
shown in Fig. 1. Parts A, B and C were made of Teon,

Fig. 1. Schematic illustration of the homemade thin-layer cell.

F.-F. Zhang et al. / Journal of Electroanalytical Chemistry 575 (2005) 17

and they were xed with four screws to prevent weeping.

Before use, the carbon disc electrode array was successively polished with emery paper and 0.5 lm alumina
powder, and sonicated in twice distilled water.
The microdialysis system consisted of a CMA/101
microdialysis pump (Sweden) and a CMA/11 microdialysis probe (Sweden) with a membrane diameter of 0.24 mm
and a length of 3.0 mm. Ringers solution was used as the
perfusion solution at the rate of 1.0 ll min 1. The components were 140 mmol l 1 NaCl, 1.0 mmol l 1 MgCl2, 1.2
mmol l 1 CaCl2 and 5.0 mmol l 1 NaHCO3, pH 7.4.
Transmission electron microscope (TEM) images
were recorded by a JEOL JEM-100CX-II Electron
Microscope (Japan).

taxic frame. A microdialysis probe was implanted into

the left striatum (coordinates with the skull leveled between bregma and lambda, were x = +3.0, y = +0.6,
z = 7 mm) [26]. Dialysate samples were discarded over
the rst 90 min to allow recovery from the acute eects
of the implantation procedure. Samples were then collected continually in 20 ll sample receivers, and injected
into the sample loop by switching the value. In this manner, four analytes in the dialysates were detected simultaneously at a radial stream NRDS-enzyme electrode
array. The standard addition method was also used by
injecting increasing concentrations of the four analyte
mixed solution.

2.3. Synthesis NRDS nanoparticles

3. Results and discussion

Silica nanoparticles were prepared according to the literature [24]. The W/O microemulsion was prepared rst
by mixing 7.5 ml of cyclohexane, 1.8 ml 1-hexanol and
1.77 ml of Triton X-100 completely. Then 400 ll neutral
red solution (1.0 10 2 mol l 1) was added slowly to the
above mixed solution in an ice cooled ultrasonicator
bath. In the presence of 100 ll TEOS, a polymerization
reaction [25] was initiated by adding 60 ll NH4OH.
The reaction was allowed to continue for 24 h. After
the reaction was completed, the NRDS nanoparticles
were isolated from the microemulsion with acetone,
and washed thoroughly (56 times) with both ethanol
and water to remove any surfactant molecules or any
physically adsorbed NR from the particle surfaces.

3.1. Characterization and electrochemical behavior of

nanoparticles

2.4. Construction of a Naon
/NRDS-enzyme electrode

array
A typical NRDS-modied GODLODL -GLOD
XOD four enzyme sensor array was prepared as follows.
A solution of 50 ll PBS with NRDS nanoparticles in 50
ll PBS containing 10 mg BSA and 1 mg GOD for the
GOD electrode (0.4 mg LOD for the LOD electrode,
0.2 mg L -GLOD for the L -GLOD electrode, or 10 ll
XOD for the XOD electrode), was mixed with 20 ll
5% glutaraldehyde solution. About 1 ll of the resulting
enzyme solution with NRDS nanoparticles was coated
onto the CE surface and air-dried at room temperature.
The enzyme and NRDS nanoparticle coated carbon
electrode array was further modied with a thin layer
of Naon
by dripping 1 ll of 1% (w/v) Naon
/methanol solution and allowing the solvent to dry in air. When
not in use, it was stored in PBS in the dark at 4 C.

Neutral red-doped silica nanoparticles prepared by

the microemulsion method were extremely uniform in
size, 50 3 nm in diameter, and were characterized by
TEM (Fig. 2). Based on a calculation done on 60 individual nanoparticles, the relative standard deviation
(RSD) of their size distribution was less than 2.8%.
Fig. 3 depicts cyclic voltammograms (CV) of the bare
GCE (a) and the Naon
/XOD-NRDS sensor (b) in
PBS (pH 6.9). At a scan rate of 100 mV s 1, almost symmetric waves and 70 mV peak-to-peak separations between the potentials of the anodic (Epa) and the
cathodic peaks (Epc) exhibited the features of rapid
charge transfer at surface-bound species [21]. The results
showed that NRDS nanoparticles kept their high electron-transfer eciency. In addition, no obvious of the
peak currents was observed after 50 cycles, indicating
that NR is not leached out from the SiO2 network under
these conditions. Furthermore, in the 5.0 10 6 mol l 1

2.5. FIA with a Naon
/NRDS-enzyme electrode array

system for simultaneous detection of biological samples
A male SD rat weighing about 250 g was anesthetized
with urethane (1.5 g kg 1, i.p.) and placed in a stereo-

Fig. 2. TEM image of neutral red-doped silica nanoparticles (NRDS).

F.-F. Zhang et al. / Journal of Electroanalytical Chemistry 575 (2005) 17

12
6

-I / nA

-6

b
-12
-18
0.5

c
0.4

0.3

0.2

0.1

0.0

-0.1

-0.2

-0.3

E / V vs. SCE
Fig. 3. Cyclic voltammograms for: (a) bare GCE in PBS; (b) Naon
/
XOD-NRDS/GCE in PBS; (c) Naon
/XOD-NRDS/GCE in
5.0 10 6 mol l 1 hypoxanthine solution. Scan rate: 100 mV s 1.

hypoxanthine solution, a striking change in the voltammogram occurs, Fig. 3(c). Both anodic and cathodic
currents were increased. The increases of the peak currents were dependent on the hypoxanthine concentration. The other three sensors showed CV responses
similar to that of Naon
/XOD-NRDS. This is the
main characteristic of the electrocatalytic reaction by
mediators [27].

tively; despite the use of the same operating potential,

there was no apparent cross reactivity between the four
sensing parts, which gave four reproducible peaks simultaneously. In the absence of the NRDS-modied four
enzyme sensor array, the amperometric response for
5.0 10 3 mol l 1 glucose decreased to 23%, 2.0 10 3
mol l 1 lactate decreased to 66%, 5.0 10 3 mol l 1 L glutamate lactate decreased to 12% and 1.0 10 3
mol l 1 hypoxanthine lactate decreased to 37%
(Fig. 4). This indicated that the NRDS nanoparticles
present electrocatalytic activity toward the four analytes. The catalytic eect might be attributed to a high
eciency of the electron-transfer mediator NRDS nanoparticles. On the other hand, it might be due to the
hydrophobic silica nanoparticles providing a biocompatible environment and improving the enzyme activity
[18,28,29].
3.4. Linearity, detection limits of Naon
/NRDS-enzyme
electrode array by FIA system
Typical ow-injection responses for the Naon
/
NRDS-enzyme electrode array with an applied potential

3.2. Optimization of ow-injection analysis

In the FIA system, the ow rate used for measurement the four analytes is an important parameter since
the process involves the enzymatic reaction kinetics
and the diusion of the glucose, lactate, L -glutamate
and hypoxanthine and their products through the
NRDS-modied enzyme electrode array. An optimal
ow rate of 1.0 ml min 1 was obtained by evaluating
the analytical performance of the sensor array, peak
width and the measurement sensitivity.
The eect of the detection potential was assessed
from the ow injection hydrodynamic voltammogram.
The glucose anodic response started at +0.10 V, rose
sharply to +0.70 V, and leveled o at higher values
(not shown). All subsequent work, thus, employed a
detection potential of +0.70 V. A similar potential (on
the current plateau) was used for the detection of lactate, L -glutamate and hypoxanthine too.
3.3. Amperometric response on Naon
/NRDS-enzyme
electrode array and Naon
/enzyme electrode array
When a mixed solution (5.0 10 3 mol l 1 glucose,
2.0 10 3 mol l 1 lactate, 5.0 10 3 mol l 1 L -glutamate and 1.0 10 3 mol l 1 hypoxanthine) was injected
into the sample loop, the sensing parts of the Naon
/
NRDS-enzyme electrode array responded selectively to
glucose, lactate, L -glutamate and hypoxanthine, respec-

Fig. 4. Amperometric response on the Naon
/NRDS-enzyme electrode array (a) and Naon
/enzyme electrode array (b) by the FIA
system with 20 ll injections of: (A) 5.0 mM glucose, (B) 2.0 mM
lactate, (C) 5.0 mM L -glutamate and (D) 1.0 mM hypoxanthine.
Applied potentials (for four sensor array), +0.70 V vs SCE. The mobile
phase was 0.1 M phosphate buer pH 6.9. The ow rate was 1
ml min 1.

F.-F. Zhang et al. / Journal of Electroanalytical Chemistry 575 (2005) 17

of 0.7 V is shown in Fig. 5 The four sensor array responded rapidly to injections of the corresponding target analytes, with a nearly instantaneous rise in the
current. The linear calibration curves, correlation coecients and detection limits of the four analytes are summarized in Table 1. The sensitivities for the four analytes
are superior to FIA-UV, for which the analytical data
are not shown here.

3.5. Reproducibility and stability of Naon
/NRDSenzyme electrode array

The reproducibility is ascertained by monitoring the
current response for ten replicate injections of the four
analyte mixture consisting of 1.0 10 3 mol l 1 glucose,
2.0 10 3 mol l 1 lactate, 5.0 10 3 mol l 1 L -glutamate and 1.0 10 3 mol l 1 hypoxanthine. The relative
standard deviations (RSD) of the peak currents are 2.5%
for glucose, 3.1% for lactate, 2.4% for L -glutamate and
4.6% for hypoxanthine, indicating a good reproducibility and stability of Naon
/NRDS-enzyme electrode array for FIA.
In addition, the sensitivity of the Naon
/NRDS-enzyme electrode array shows no observable change after
two weeks of storage in PBS at 4 C or after successive
potential cycles, indicating that this array is very stable
and shows long shelf-life.
3.6. Study on the interference

Fig. 5. Simultaneous ow-injection analysis of glucose + lactate + L glutamate + hypoxanthine mixed solutions. (A) Response to increasing
levels of glucose: 0.5 mM (a), 2.0 mM (b), 4.0 mM (c), 6.0 mM (d), 8.0
mM (e), and 10.0 mM (f). (B) Response to increasing levels of lactate:
0.5 mM (a), 2.0 mM (b), 4.0 mM (c), 6.0 mM (d), 8.0 mM (e), and 10.0
mM (f). (C) Response to increasing levels of L -glutamate: 1.0 mM (a),
2.0 mM (b), 4.0 mM (c), 6.0 mM (d), 8.0 mM (e), and 10.0 mM (f).
(D) Response to increasing levels of hypoxanthine: 0.1 mM (a), 0.2
mM (b), 0.4 mM (c), 0.6 mM (d), 0.8 mM (e), and 1.2 mM (f). Other
conditions as in Fig. 4.

In the Naon
/NRDS-enzyme electrode array, the

outside Naon
lm could prevent both anionic electroactive interferences from reaching the electrode surface
and fouling of the array [2123]. The presence of 0.2
mM ascorbic acid and 0.2 mM uric acid in the buer
containing 1.0 mM glucose, 1.0 mM lactate, 5.0 lM L glutamate and 5.0 lM hypoxanthine did not aect the
FIA response current, suggesting that, especially at
low glucose, lactate, L -glutamate and hypoxanthine concentrations, there was little interference from AA, etc.,
which increased the sensitivity in measuring the concentrations of the four analytes in in vitro dialysate
samples.
3.7. Determination of glucose, lactate,
hypoxanthine level in living samples

L -glutamate

and

The microdialysis probe was implanted into the brain

of an anesthetized S.D. rat of about 250 g. The dialysate
was collected at a perfusion rate of 2.0 ml min 1 after

Table 1
Analytical data of the four analytes in FIA with the Naon
/NRDS-enzyme electrode array systemA
Analytes
Glucose
Lactate
L -glutamate

Hypoxanthine
A
B

Regression equation (ax + ba)B

Correlation coecient (r)

Linear range (mol l 1)

6

y = 6.5536x + 0.1781
y = 1.7299x + 0.4555

0.9988
0.9967

1.0 10 1.0 10
5.0 10 41.0 10

y = 1.1502x + 0.2158
y = 0.6132x + 0.2906

0.9949
0.9970

1.0 10 68.0 10
1.0 10 41.0 10

y = 17.843x + 1.4845
y = 1.5557x + 1.8773

0.9902
0.9981

5.0 10 72.0 10
5.0 10 51.0 10

y = 0.0861x + 0.2488
y = 0.5314x + 0.2042

0.9923
0.9974

5.0 10 75.0 10
1.0 10 41.2 10

FIA conditions as in Fig. 5.

Where y and x represent the peak current and the concentration of the analytes, respectively.

10
5.0

5.0

2.0

2.0

Detection limit/mol l

(r = 3)

F.-F. Zhang et al. / Journal of Electroanalytical Chemistry 575 (2005) 17

is very sensitive and useful for the simultaneous monitoring of glucose, lactate, L -glutamate and hypoxanthine levels in in vitro dialysate samples. Uniform
NR-doped silica nanoparticles retained a high electron-transfer eciency and showed electrocatalytic
activity toward the four analytes. There was negligible
interference from oxidizable species (such as ascorbate,
and urate) in the extracellular space of rat brain, and
the system was useful to the study of brain metabolism
and neuron communication. The system has the potential to be applied to monitor the four analytes in other
parts of living cells, such as hepatic tissue and the blood
stream.

Acknowledgments
Financial support is acknowledged from the National
Natural Science Foundation of China (No. 20175006,
20305007) and the specialized Research Fund for Nanotechnology from Shanghai (No. 0214nm078 and No.
0359nm002).

Fig. 6. Typical FIA response of Naon
/NRDS-enzyme electrode

array in the brain dialysate collected from rat striatum (a), and
addition of (A) 2.0 mM glucose, (B) 2.80 mM lactate, (C) 0.70 mM L glutamate and (D) 0.10 mM hypoxanthine mixture standard solutions
(b). Other conditions as in Fig. 4.

Table 2
Content of the four analytes in rat striatum dialysate sample and
recoveriesa
Analytes
Glucose
Lactate
L -glutamate
Hypoxanthine

Detected/
lmol l 1

Added/
mmol l

412
685
1.06
5.91

2.00
2.80
0.70
0.10

Found/
mmol l
2.48
3.41
0.68
0.11

Recovery/
%
102.8
97.8
97.1
103.8

The values shown are calculated from the calibration curves and
are the mean of n = 3 in each case.

1.5 h. Fig. 6. shows the FIA responses to glucose, lactate, L -glutamate and hypoxanthine of the dialysate
sample. Recovery studies were also carried out by adding known amounts of glucose, lactate, L -glutamate
and hypoxanthine mixture to the dialysate sample. The
results showed good recoveries, ranging from 97.1% to
103.8% for the four analytes (see Table 2), which correlate well with those in the literature [1,3033].

4. Conclusions
It was found from the results that the present FIA
with a Naon
/NRDS-enzyme electrode array system

References
[1] M.G. Boutelle, L.K. Fellows, C. Cook, Anal. Chem. 64 (1992)
1790.
[2] N.F. Shram, L.I. Netchiporouk, C. Martelet, N. Jarezic-Renault,
C. Bonnet, R. Cespuglio, Anal. Chem. 70 (1998) 2618.
[3] A. Hashimoto, J. Kanda, T. Oka, Brain Res. Bull. 53 (2000) 347.
[4] L.Q. Mao, F. Xu, Q. Xu, L.T. Jin, Anal. Biochem. 292 (2001) 94.
[5] L.Q. Mao, Katsunobu Yamamoto, Anal. Chim. Acta 415 (2000)
143.
[6] E. Llaudet, N.P. Botting, J.A. Crayston, N. Dale, Biosens.
Bioelectron. 18 (2003) 43.
[7] L.I. Netchiporouk, N.F. Shram, N.J. Renault, C. Martelet, R.
Cespuglio, Anal. Chem. 68 (1996) 4358.
[8] M. Ellmerer, L. Schaupp, Z. Trajanoski, G. Jobst, I. Moser, G.
Urban, F. Skrabal, P. Wash, Biosens. Bioelectron. 13 (1998) 1007.
[9] P.S. Petrou, I. Moser, G. Jobst, Biosens. Bioelectron. 18 (2003)
613.
[10] J. Woitzik, N. Abromeit, F. Schaefer, Anal. Biochem. 289 (2001)
10.
[11] J.X. Zhou, D.M. Heckert, H. Zuo, C.E. Lunte, S.M. Lunte,
Anal. Chim. Acta 379 (1999) 307.
[12] M.A. Malone, H. Zuo, S.M. Lunte, M.R. Smyth, J. Chromatogr.
A 700 (1995) 73.
[13] F. Xu, L. Wang, M.N. Gao, L.T. Jin, J.Y. Jin, Talanta 57 (2002)
365.
[14] A.A. Karyakin, Y.N. Ivanova, E.E. Karyakina, Electrochem.
Commun. 5 (2003) 677.
[15] B. Grundig, G. Wittstock, U. Rudel, B. Strehlitz, J. Electroanal.
Chem. 395 (1995) 143.
[16] S.M. Chen, K.C. Lin, J. Electroanal. Chem. 511 (2001) 101.
lvarez, P. Rodrguez-Granda, M.J. Lobo-Cast[17] P. de-los-SantosA
anon, A.J. Miranda-Ordieres, P. Tunon-Blanco, Electrochem.
Commun. 5 (2003) 267.
[18] M. Qhobosheane, S. Santra, P. Zhang, W.H. Tan, Analyst 126
(2001) 1274.
[19] B. Strehlitz, B. Grundig, W. Schumacher, P.M.H. Kroneck, K.D. Vorlop, H. Kotte, Anal. Chem. 68 (1996) 807.

F.-F. Zhang et al. / Journal of Electroanalytical Chemistry 575 (2005) 17

[20] M. Abo, Y. Ogasawara, Y. Tanaka, A. Okubo, S. Yamazaki,
Biosens. Bioelectron. 18 (2003) 735.
[21] M. Zhu, S.B. Han, Z.B. Yuan, J. Electroanal. Chem. 480 (2000)
255.
[22] D.J. Harrison, R.F.B. Turner, H.P. Balters, Anal.Chem. 60
(1988) 2002.
[23] J.H. Yu, S.Q. Liu, H.X. Ju, Biosens. Bioelectron. 19 (2003) 401.
[24] F.J. Arriagada, K. Osseo-Asare, J. Colloid Interf. Sci. 211 (1999)
210.
[25] W. Stober, A. Fink, E. Bohn, J. Colloid Interf. Sci. 26 (1968) 62.
[26] X.M. Bao, Y. Su, The Stereotaxic Atlas of the Rat Brain, Peoples
Healthy Press, Beijing, China, 1991.

[27] Y.X. Sun, B.X. Ye, W.M. Zhang, X.Y. Zhou, Anal. Chim. Acta
363 (1998) 75.
[28] X.L. Luo, J.J. Xu, W. Zhao, H.Y. Chen, Sensor Actuat. B 97
(2004) 249.
[29] Z.J. Chen, X.M. Ou, F.Q. Tang, L. Jiang, Colloids Surf., B:
Biointerf. 7 (1996) 173.
[30] T. Yao, T. Yano, H. Nishino, Anal. Chim. Acta 510 (2004) 53.
[31] T. Yao, T. Yano, Talanta 63 (2004) 771.
[32] J.F. Stover, K. Lowitzsch, O.S. Kempski, Neurosci. Lett. 238
(1997) 25.
[33] M.W. Lada, T.W. Vickroy, R.T. Kennedy, Anal. Chem. 69 (1997)
4560.