Multienzyme Amperometric Gluconic Acid Biosensor Based On Nanocomposite Planar Electrodes For Analysis in Musts and Wines
Multienzyme Amperometric Gluconic Acid Biosensor Based On Nanocomposite Planar Electrodes For Analysis in Musts and Wines
Multienzyme Amperometric Gluconic Acid Biosensor Based On Nanocomposite Planar Electrodes For Analysis in Musts and Wines
31
International Journal of
ELECTROCHEMICAL
SCIENCE
www.electrochemsci.org
1. INTRODUCTION
Gluconic acid, an oxidation product of glucose, occurs widely in nature. It is present in various
foodstuffs naturally or as a permitted additive [1]. In wine industry, gluconic acid represents the most
important indicator of infection of grapes by Botrytis Cinerea [2]. This fungal infection should be
avoided since it affects negatively the quality of grapes and, consequently, wines resulting in serious
Int. J. Electrochem. Sci., Vol. 12, 2017 1184
economic losses of wineries on the worldwide basis. Thus a rapid, specific, simple and cost effective
determination of gluconic acid in musts and wines, which could be performed directly in wineries, is
highly required [3]. A variety of analytical techniques such as chromatography [4-6], electrophoresis
[7], infrared spectroscopy [8], spectrophotometry [9], chemiluminescence [10] and fluorescence [11]
have been developed for this purpose. Enzyme based electrochemical biosensors seem to have a
potential to fulfil well all of the requirements mentioned above [12-14].
Ikeda et al. [15] constructed amperometric biosensor based on gluconate dehydrogenase
(GADH) and p-benzoquinone as an electron-transfer mediator. This concept has been improved by
GADH coimmobilization with tetrathiafulvalene (TTF) using self-assembled monolayers on the
surface of gold disc [16] and screen-printed carbon electrode (SPCE) [3]. As GADH has been not
longer commercially available, alternative systems have been described employing gluconate kinase
(GK) and 6-phospho-d-gluconate dehydrogenase (6PGDH) coimmobilized by an entrapment within
the polysulfone membrane on a graphite-epoxy composite [13] and SPCE [14]. Both of these
biosensors operated at the relatively high working potential of 800 mV with coenzymes ATP and
NADP in measuring media. Recently, the detection of NADPH formed by 6PGDH at suitable low
potential of 100 mV was described using film of polyaniline/poly(2-acrylamido-2-methyl-1-
propanesulfonic acid) electrodeposited on SPCE [17]. Although these biosensors showed a sufficient
analytical performance, they exhibited relatively short lifetimes from several days to 3 months. Storage
and operational stabilities are crucial factors for the potential biosensor commercialization, together
with production costs and the availability of components. To fulfil such a requirement, a glycerol
biosensor based on a novel multi-enzyme cascade with an extended lifetime (at least to 15 months) has
been reported recently creating a new platform for the biosensor preparation with diverse kinases [18].
Here we present the gluconic acid biosensor utilizing a multi-enzyme cascade consisting of
gluconate kinase, creatine kinase, creatinase, sarcosine oxidase and peroxidase immobilized onto
nanocomposite planar electrodes containing multi-walled carbon nanotubes. All these enzymes are
available, cheap, very stable and cooperating effectively at suitable working conditions allowing thus
fabrication of sensitive, robust and stable biosensors. In addition, the use of expensive NADP cofactor
is avoided. The aim of this work was the construction of a gluconate biosensor for analyses of real
oenological samples and compatible with a commercially available portable analytical device like
Omnilab W [19].
2. EXPERIMENTAL
Sarcosine oxidase, creatinase, horseradish peroxidase were obtained from Sorachim (Lausanne,
Switzerland), gluconate kinase from CPC Biotech (Burago di Molgora, Italy) and creatine kinase from
USB (Cleveland, OH USA). Potassium hexacyanoferrate(II), N-eicosane, potassium D-gluconate, D-
glucose, D-fructose, sodium-L-lactate, L-ascorbic acid, L-malate disodium salt, sodium acetate,
anhydrous ethanol, magnesium sulphate heptahydrate, glycerol, adenosine 5’-triphosphate (ATP),
creatine phosphate (CP) and chitosan from shrimp shells (85% deacetylated) were supplied by Sigma-
Int. J. Electrochem. Sci., Vol. 12, 2017 1185
Aldrich (St. Louis, USA). Potassium phosphate monobasic and potassium phosphate dibasic were
purchased from Riedel-de Haen (Seelze, Germany). All chemicals used were of analytical grade.
Nanopure water (Millipore Milli-Q, 18 MΩ cm) was used throughout.
Multi-walled carbon nanotubes (MWCNT) (d = 60-100 nm, l = 5-15 μm, 95 % purity) were
purchased from NanoAmor (Houston, USA). Basic planar circular electrodes with the diameter of 1.6
mm and equipped with a Ag/AgCl reference electrode of diameter 2 mm screen-printed on the planar
glass-epoxy-laminate substrate were obtained from Biorealis (Bratislava, Slovakia).
Slightly modified procedure described previously [18, 20] was used for the nanocomposite
preparation. First, 100 mg of N-eicosane were melted at the temperature 45°C. Then, 12 mg of
MWCNT were added and the mixture was stirred vigorously with a mini-stirrer until the homogenous
mixture was obtained. The suspension was subsequently deposited on the surface of cleaned (nanopure
water and ethanol) metal contacts to create thick layer circles. Then, the nanocomposite layer was left
to solidify [21].
The immobilization of individual enzymes with optimum amounts of 0.25 U of GK, 0.15 U of
CK, 0.25 U of SAOX, 0.6 U of CRE and 1 U for HRP dissolved in nanopure water on the electrode
surface was carried out by their sandwiching between two chitosan layers formed by applying its 1 %
(w/w) solution. Individual layers were deposited after the previous one was dried. The prepared
biosensor was stored at room temperature in a desiccator until use.
Polymer IEX in H+ form, 250 mm x 8 mm, 8 μm in diameter (Watrex, Bratislava, Slovakia), guard
column Polymer IEX in H+ form, 10 mm x 4 mm, 8 μm (Watrex, Bratislava, Slovakia); column
temperature 50 °C; mobile phase with isocratic elution - 9 mM H2SO4 in water for HPLC; flow rate 1.0
mL min-1. Data were collected and processed by the software Clarity chromatography station
DataApex (Prague, Czech Republic). Prior to analysis, samples were diluted by a mobile phase and
filtered through 0.22 μm Chromafil AO filters (Macherey-Nagel, Dűren, Germany). Organic acids
were identified by a comparison with retention times of the authentic standard solutions.
Figure 1. Scheme of multi-enzyme cascade immobilized between chitosan layers at the gluconic acid
biosensor with hexacyanoferrate(II) as the redox mediator in solution. The Medox reduction
current is measured at the applied constant potential of -50 mV vs. Ag/AgCl.
The quantities of immobilized enzymes were optimized in the range 0.1 to 0 .8 U for gluconate
kinase (GK), 0.1 to 0.5 U for creatine kinase (CK) and sarcosine oxidase (SAOX), 0.3 to 1.5 U for
creatinase (CRE) and 0.5 to 4 U for horseradish peroxidase (HRP). High enzyme loadings led to the
decrease in current which is likely due to the steric hindrance of protein immobilized on the electrode
surface. On the other hand, low enzyme quantities caused a decrease of the biosensor response and
narrow linear concentration ranges. The optimum amounts found were 0.25 U of GK, 0.15 U of CK,
0.25 U of SAOX, 0.6 U of CRE and 1 U for HRP and, consequently, used in further experiments.
Int. J. Electrochem. Sci., Vol. 12, 2017 1187
Concentration of the mediator in the working medium was studied in the range from 1 to 10 mM
hexacyanoferrate(II) where 2.5 mM was found as the optimum. The concentrations of magnesium ions
(typical kinase activator) and cofactors CP and ATP were optimal at 7.5 mM, 1 mM and 0.5 mM,
respectively. Recently similar concentrations (Mg2+ 7.0 and 2.4 mM, ATP 0.36 and 0.39 mM) were
published in two reports on the bienzymatic gluconic acid biosensors [14, 17].
A selection of the suitable working potential plays an important role in a suppression of
possible interferences such as wine polyphenols. A direct detection of hydrogen peroxide (Figure 1)
could be also utilized at the potential over 600 mV which is accompanied by a risk of false interfering
signals. For this reason, peroxidase with the hexacyanoferrate(III/II) redox couple were employed
allowing the detection at a significantly lower potential. A working potential value was tested within
the range -200 mV to +150 mV vs Ag/AgCl and that of -50 mV has been selected as possessing
satisfactory biosensor sensitivity and avoiding from undesired interferences. This was proved in next
experiments.
100
relative response (%)
80
60
40
20
5 6 7 8 9 10
pH
Figure 2. Effect of pH on the gluconic acid biosensor relative response (pH 7.5 correspond to 100%).
Experimental conditions: 5 mM hexacyanoferrate(II), 50 μM gluconic acid, applied constant
potential -50 mV vs. Ag/AgCl.
As optimum pH for single used enzymes varies, selection of suitable pH value for multienzyme
biosensor was crucial for its good functionality and stability. The dependence of the amperometric
response of the biosensor after addition of 10 mM gluconic acid (pKa 3.86) on pH of the phosphate
buffer solution (PB) is shown in Fig. 2. The current value increased with increasing pH until and
reached the maximum at the pH 7.5. Stability of the biosensor was optimal at this pH as well.
Int. J. Electrochem. Sci., Vol. 12, 2017 1188
800
700
600
500
current (nA)
400
300
200
100
0
0 200 400 600 800 1000 1200
gluconic acid concentration (µM)
Figure 3. Calibration curve obtained for the gluconic acid biosensor. Experimental conditions: 5 mM
hexacyanoferrate(II), 0.1 M phosphate buffer, pH 7.5, applied constant potential -50 mV vs.
Ag/AgCl.
The biosensor exhibited a good linearity within the range 4-620 μM with the correlation
coefficient of 0.990 (n = 21). The detection limit obtained was 2.6 μM (based on signal/noise = 3) and
the sensitivity 45.3 nA μM-1 cm-2. These characteristics are better or comparable to those reported
previously for diverse gluconate biosensors with exception of the significantly lower detection limit
0.19 μM [16]. The response of the presented biosensor was relatively fast. The current started to
increase approximately 4 s after the injection of standard or sample solutions reaching a 90% steady-
state current within 45 s. The final response time was found to be about 70 s. Reproducibility of the
measurement with the same biosensor was tested as a current response on 50 μM gluconic acid and at
the average response of 42.7 nA is represented by RSD of 4.17% (n=6). In Table 1, analytical
parameters of the biosensor developed in this work are compared with those reported previously.
Operation and storage stabilities of the prepared biosensor were tested. No loss of the biosensor
sensitivity after 30 successive additions of 50 μM gluconic acid or wine samples was observed. The
biosensor exhibited a response above 90 % or above 55 % of the original one after 3 days or 7 days,
respectively, at storage in 0.1 M PB of pH 7.5 at 4 °C after a periodic use. Humidity and high
Int. J. Electrochem. Sci., Vol. 12, 2017 1189
temperature are typically considered as the most negative factors which can affect the storage stability
of enzymatic biosensors. To perform the long term stability study, a batch of 80 biosensors was
prepared. The sensitivities of 5 biosensors from this batch were evaluated in regular intervals. When
the biosensors were held in a desiccator at room temperature without any previous use for 12 months,
their sensitivity remained about 90 % of the initial value. We suppose that the long storage and
operation stability was achieved due to good stability of the immobilization of employed enzymes in
the chitosan matrix. Such excellent long-term storage stability was observed in our laboratory also at
other biosensors based on the chitosan sandwich as it was discussed in the previous work [22]. With
respect to facts that the stability contributes to a reliability of biosensors and could significantly reduce
the operational costs of potential users, this is of great importance for the expected use of the biosensor
in control processes and routine monitoring. The stability of the present biosensor is significantly
higher when compared to those reported in the literature (Table 1).
Table 1. Comparison of analytical parameters obtained with the electrochemical gluconic acid
biosensors.
3.4 Interferences
Generally the interferences are one of the more important factors affecting a relatively low
dissemination of biosensors in various sectors. The interference studies were performed to verify a
specificity of the proposed biosensor to the gluconate determination. Various components of wine may
affect the accuracy of analysis results by interfering with enzymes activity or by direct oxidation on the
electrode. Particularly, polyphenols, alcohols, sugars and some organic acids should be considered as
possible interferences. The effect of potential interferences present in musts and wines such as ethanol
(10%), glucose (18 g L-1), fructose (18 g L-1), glycerol (10 g L-1), L-lactic (4 g L-1), citric (2 g L-1),
tartaric (4 g L-1), malic (4 g L-1), acetic (2 g L-1) and L-ascorbic acid (50 mg L-1) on the response of the
biosensor was evaluated. The presence of sugars, ethanol and acids did not change the current signal.
The effect of polyphenols was also tested by the addition of red wines and using bare electrodes
covered with chitosan double layer without enzymes and the optimized composition of measuring
Int. J. Electrochem. Sci., Vol. 12, 2017 1190
solution. No significant current changes in the biosensor response were observed, thus, polyphenols
also do not interfere under the proposed experimental conditions.
Finally, the presented multi-enzyme biosensor was applied to the analysis of gluconic acid in
various musts and wines without any pretreatment. The results obtained are summarized in Table 2.
The determination was performed by injecting 10 μL of sample and calibration solution in 1
mL of measuring media. As it can be seen in Table 1 that the results obtained with the biosensor are in
good agreement with those obtained by HPLC as the reference method. The analyzed musts and wines
contained low amounts of gluconic acid apart from Tokaj. So the samples were spiked with a known
concentration of gluconic acid. Recovery values about 100 % demonstrate a good reliability of the
proposed biosensor. Tokaj is a noble Slovak wine naturally containing gluconic acid as a result of
particular cultivation conditions and production process.
Table 2. Results of the determination of gluconic acid in must and wine samples using the biosensor
and HPLC as a standard analytical method. Results are expressed with standard deviation
(n=3).
4. CONCLUSIONS
The novel multienzyme biosensor based on commercially available materials for the selectively
quantification of gluconic acid is reported. 5 enzymes were immobilized between chitosan layers on
the surface of planar nanocomposite electrode. The simple and effective immobilization technique
provided long-term storage stability, low fabrication costs, and good analytical performance. The
biosensor exhibited wide linear range, low detection limit, high sensitivity, short measuring time and
interference-free measurements. It was successfully applied to gluconic acid determination in real must
and wine samples, validating the biosensor performance through comparison with the reference HPLC
method. The performance characteristics of this new analytical tool may satisfy requirements of wine
industry being particularly suitable for small-medium winemakers. The developed biosensor is
compatible with the small device Omnilab W.
ACKNOWLEDGEMENTS
This work was supported by the Scientific Grant Agency VEGA of the Slovak Republic (Project No.
1/0361/14), the Slovak Research and Development Agency under the Contract No. APVV-0797-11,
and the Competence Centre for SMART Technologies for Electronics and Informatics Systems and
Services (Project ITMS 26240220072) funded by the Research & Development Operational
Programme from the ERDF.
References
© 2017 The Authors. Published by ESG (www.electrochemsci.org). This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution license
(http://creativecommons.org/licenses/by/4.0/).