Intact and permeabilized cells of the yeast Hansenula polymorpha as bioselective elements for amperometric assay of formaldehyde

Talanta 71 (2007) 934–940

Intact and permeabilized cells of the yeast Hansenula polymorpha as bioselective elements for amperometric assay of formaldehyde Maria Khlupova a , Boris Kuznetsov a , Olha Demkiv b , Mykhailo Gonchar b , Elisabeth Cs¨oregi c , Sergey Shleev a,c,∗ b

a Laboratory of Chemical Enzymology, A.N. Bach Institute of Biochemistry, 119071 Moscow, Russia Department of Cell Regulatory Systems, Institute of Cell Biology, Drahomanov Street 14/16, 79005 Lviv, Ukraine c Department of Analytical Chemistry, Lund University, P.O. Box 124, SE-221 00 Lund, Sweden

Received 2 March 2006; received in revised form 23 May 2006; accepted 26 May 2006 Available online 10 July 2006

Abstract Intact and permeabilized yeast cells were tested as the biorecognition elements for amperometric assay of formaldehyde (FA). For this aim, the mutant C-105 (gcr1 catX) of the methylotrophic yeast Hansenula polymorpha with a high activity of AOX was chosen. Different approaches were used for monitoring FA-dependent cell response including analysis of their oxygen consumption rate by the use of a Clark electrode, as well as assay of oxidation of redox mediator at a screen-printed platinum electrode covered by cells entrapped in Ca-alginate gel. It was shown that oxygen consumption rate of permeabilized cells reached its saturation at 4 mM of FA (23 ◦ C). The detection limit was found to be 0.27 mM. In the presence of redox mediator 2,6-dichlorophenolindophenol (DCIP), the screen-printed platinum band electrode covered by permeabilized cells did not show any current output to FA. In contrast, well-pronounced amperometric response to FA was observed in the case of intact yeast cells in the presence of DCIP. It was shown that current output reached its maximum at 7 mM concentration of FA. The detection limit was found to be 0.74 mM. Obviously, it is necessary to perform a directed genetic engineering of the yeast cells to improve their bioanalytical characteristics in the corresponding biosensors. © 2006 Elsevier B.V. All rights reserved. Keywords: Bioselective analysis; Formaldehyde; Biosensor; Amperometry; Redox mediator; Intact and permeabilized cells; Yeast Hansenula polymorpha

1. Introduction Formaldehyde (FA) is one of the most important chemicals widely used in industry. However, it is classified as a mutagen and human carcinogen [1], as well as one of the chemical mediators of apoptosis [2]. These polluting substance’s properties are sufficient to demonstrate the necessity for FA control in consumer goods, food, and environment. Such control requires the development of simple, cheap, sensitive, and selective methods for the analysis of this extremely toxic agent. A number of approaches for the detection of FA concentration has been used [3,4]. The drawbacks of the methods are the requirement

∗

Corresponding author. Tel.: +7 495 954 44 77; fax: +7 495 954 27 32. E-mail addresses: [email protected], [email protected] (S. Shleev). 0039-9140/$ – see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.talanta.2006.05.069

of derivatization of FA and, in some cases, the necessity of fully anaerobic conditions. FA-sensitive biosensors have been constructed also by the use of highly purified alcohol oxidase (AOX) or bacterial FA dehydrogenase. Previously, the amperometric [5,6], potentiometric [7], and optical [8] biosensors based on these enzymes were designed for the determination of FA concentration. However, due to a non-sufficient stability of FA dehydrogenase and to a broad selectivity of AOX, they did not succeed in the practical applications. Besides, the commercial application of FA dehydrogenase-based biosensors is restricted due to a need in exogenous low molecular cofactors [5]. It is expedient to use cells-based biosensors that could provide high selectivity and sufficient sensitivity. These biosensors are usually simple and much cheaper. For this aim, intact and permeabilized cells of the mutant yeast Hansenula polymorpha C-105 (gcr1 catX) with a high activity of AOX were chosen.

M. Khlupova et al. / Talanta 71 (2007) 934–940

AOX inside the permealized cells remains in active state for a long time [9]. Previously, it was found that cells of the methylotrophic yeast H. polymorpha, capable of metabolizing FA, can be employed as FA-selective biorecognition elements in the pHFET-based potentiometric biosensors [9–11]. Microbial sensors for alcohol assays based on the employment of permeabilized yeast cells H. polymorpha were also constructed [12]. Therefore, the goal of the present work was to test several approaches for amperometric determination of FA by the use of genetically modified intact and permeabilized cells of the yeast H. polymorpha with a high level of AOX activity. 2. Experimental 2.1. Materials and cells The solution of FA (1 M) was prepared by hydrolysis of the corresponding amount of paraformaldehyde (300 mg; 10 ml water) in sealed ampoule at 105 ◦ C during 6 h. As a bioselective element, the cells of the mutant strain of the thermotolerant methylotrophic yeast H. polymorpha C105 (gcr1 catX) constructed in Institute of Cell Biology (Lviv, Ukraine) was used [13,14]. It has impairment in glucose catabolite repression of AOX synthesis, is catalase-defective and has ability to over-produce AOX in a glucose-containing growth medium, in contrast to the wild type strain unable to synthesize this enzyme in the presence of glucose. 2.2. Chemicals K2 HPO4 was from “Reachim” (Moscow, Russia); digitonin, CaCl2 , KCl, MgSO4 , and KH2 PO4 were from “Merck” (Darmstadt, Germany). Tris, paraformaldehyde, succinate, and 2,6-dichlorophenolindophenol (DCIP) were from “Sigma” (St. Louis, MO, USA); buffers were prepared using double-distilled water. 2.3. Cultivation of the yeast H. polymorpha and preparation of permeabilized cells as a source of endogenous alcohol oxidase Cells of the mutant H. polymorpha C-105 were cultivated in flasks on shaker (200 rpm) at 30 ◦ C to the middle of the exponential growth phase (∼24 h) in the medium containing (g/L): glucose, 10.0; (NH4 )2 SO4 , 3.5; KH2 PO4 , 1.0; MgSO4 , 0.5; CaCl2 , 0.1; yeast extract, 3.0. The pH of the medium was 5.5. Permeabilized cells were obtained using the yeast cells H. polymorpha C-105 as described in [12]. Yeast cells at a concentration of 4–5 mg/ml were treated with 0.1% digitonin in 50 mM K-phosphate buffer (pH 7.0) for 15 min at 30 ◦ C periodically shaking. The cells were washed twice with the initial buffer, harvested at 4 ◦ C by centrifugation at 1000 rpm, lyophilized, and kept at −15 ◦ C. Before each experiment, lyophilized yeast cells were re-suspended in 0.1 M K-phosphate buffer, pH 7.0 at concentration 0.8 mg/ml and allowed to swell during 30 min at 37 ◦ C. Just after the swelling, the cell suspension was used in the experiments.

935

2.4. Measurements of oxygen consumption rate by the use of the electrode of Clark type Potentiostat BAS CV-50W (BAS CV-50W Electrochemical Analyzer with BAS CV-50W software v. 2.1, Bioanalytical Systems, West Lafayette, IN, USA) was used for the measurement of respiratory activity of cells in the solution. The suspension (1.2 ml) of cells (permeabilized and swollen in buffer or intact in growth medium) was placed into the electrochemical cell of the Clark type, which was thermostatically controlled at 37 ◦ C or at room temperature (23 ◦ C). Formaldehyde in various concentrations was added to the constantly stirring suspension. The total volume of the electrochemical cell was 1.5 ml. The potential of −600 mV was applied to the electrode in the open Clark cell. After that the steady-state current was established during 300 s. Then the current record began and continued for 400 s while the cell was still opened for the first 100 s. The cell was hermetically closed then and the course of oxygen discharge current with time was recorded for the following 300 s. The rate of oxygen consumption (y) in ␮M/s was obtained from the following equation: y=

i2 − i1 202 × t2 − t 1 i0

where i1 , i2 are the currents measured on the electrode of Clark type in the course of the reaction of FA oxidation by the yeast cells, in nA; t1 , t2 are the corresponding times of measured currents, in s; 202 ␮M is the solubility of oxygen in water at 37 ◦ C by saturation in air (265 ␮M is the solubility of oxygen in water at 23 ◦ C); i0 is the current measured after the stationary regime has been established in the absence of yeast cells, in nA. The devices with intact and permeabilized yeast cells were used for the measurement of oxygen consumption rate (Fig. 1A). The suspension of cells (2 ml) was filtered through Schleicher & Schull filter (Germany) with pore diameter 0.45 ␮m and washed with 2 ml of K-phosphate buffer, pH 7.0. The filter was cut out according to the size of biosensor’s electrode, 10 mm in diameter, and pressed to biosensor’s membrane. The filter side covered by yeast cells was oriented to the membrane as shown in Fig. 1A. Each filter was used for a series of measurements. The recording of current started in 50 s after the potential of −600 mV (versus Ag/AgCl/0.2 M KCl) was applied to electrode and the record continued for 300 s. FA was injected by 100 s just after the stationary regime has been established. Oxygen consumption was calculated by the slope of the corresponding kinetic curves. Both devices were also used for the measurement of oxygen consumption rate at 1 mM FA of the intact and permeabilized yeast cells pre-treated by different concentrations of FA (from 2 to 18 mM). These concentrations are plotted on the abscissa axis. The time of pre-treatment was approximately 10 min, after each pre-treatment the electrode was washed with of K-phosphate buffer, pH 7.0 for 10 min. 2.5. Amperometric studies by using a redox mediator Principal scheme of monitoring of intracellular redox reactions using 2,6-dichlorophenolindophenol (DCIP) as a mediator

936

M. Khlupova et al. / Talanta 71 (2007) 934–940

Fig. 1. (A) The construction of biosensor based on Clark electrode with immobilized yeast cells: 1, electrode (Pt, Ag/AgCl); 2, electrolyte; 3, Teflon membrane (20 ␮m); 4, the layer of yeast cells; 5, bacterial filter (Schleicher & Schull filter (Germany) with pore diameter 0.45 ␮m); 6, fixing gauze; 7, analysed solution of FA in K-phosphate buffer; 8, stir. (B) Scheme of intracellular redox reactions coupled with electrochemical oxidation of the mediator DCIP, which is given according to [15] with some additions and changes.

as well as a putative mechanism of DCIP reaction cycle in yeast cells are shown in Fig. 1B. In contrast to the double mediator system (menadione/menadiol plus ferricyanide/ferrocyanide) only one mediator was used in our electrochemical studies [15]. Intact and permeabilized yeast cells were immobilized on the top of platinum band screen-printed electrode in a Ca2+ -alginate gel as described in [15]. For this aim, 2 mg of permeabilized yeast cells were re-suspended in 2.5 ml of 10 mM Tris–succinate buffer, pH 7.0, then 15 ␮l of 2% sodium alginate was added to the 15 ␮l of the suspension of permeabilized yeast cells for further formation of cell-alginate suspension. As for intact yeast cells 1 ml of suspension of cells in growth medium (108 cells/ml; 1.61 mg abs. dry weight per 1 ml) was centrifuged, then the precipitation was washed by 10 mM Tris–succinate buffer, pH 7.0 and then this washed precipitation was re-suspended in 15 ␮l of buffer. After that, 15 ␮l of 2% sodium alginate was added to the 15 ␮l of the suspension of intact yeast cells for further formation of cell-alginate suspension. The electrode was coated with 15 ␮l of cell-alginate suspension by the sampler; the excess of cell-alginate was removed by a tip. The calcium-alginate gel was formed after immersing the coated

electrode in solution of 100 mM CaCl2 for 2 min. Measurements were carried out immediately after the cell immobilization was completed. To maintain the integrity of the gel measurements were performed in the buffer containing 10 mM CaCl2 . One layer of immobilized cells was used for one series of experiments. Electrochemical measurements were carried out at a fixed potential of +200 mV (versus Ag/AgCl) in an electrochemical cell under aerobic conditions with vigorous stirring. Eight-channel screen-printed electrode from “BVT Technologies” (Brno, Czech Republic) was used for this aim. It contained in itself working Pt electrode and Ag as a combined reference and a counter electrode. Yeast cells were immobilized on this electrode. The electrode was placed then into the electrochemical cell containing buffer solution (10 mM Tris–succinate, pH 7.0; 100 mM KCl, and 10 mM CaCl2 ). Measurements started after the steady-state current became established. By 200 s 2 ml of buffer containing mediator DCIP was added to a final concentration of 0.1 mM. After the resulting response had reached a steady state, the measurements were continued by addition of FA up to 10 mM.

M. Khlupova et al. / Talanta 71 (2007) 934–940

937

For permeabilized cells, oxygen can be consumed only in AOX-catalyzed reaction, because FDH is not functioning due to a leakage of its cofactors, glutathione and NAD, from the treated cells. Taking into account these considerations, it can be expected a much more expressed sensitivity of the intact cells to the deteriorate effect of FA excess on the cells which really was observed in experiments. 3.2. Biosensors based on the Clark electrode

Fig. 2. Dependence of oxygen-consuming activity of intact (doted lines) and permeabilized cells (solid lines) of the yeast Hansenula polymorpha C-105 in the solution on FA concentration. The experiments were performed by using the Clark electrochemical cell at 37 and 23 ◦ C. Conditions: concentration of the permeabilized cells in a buffer solution (0.64 mg of dried cells/ml); concentration of the intact cells in a growth medium (108 cells/ml; 1.61 mg/ml); 0.1 M Kphosphate buffer, pH 7.0.

The biosensor based on intact and permeabilized yeast cells (Fig. 3A) was also used for the study of the dependence of the rate of oxygen consumption on the concentration of FA. The current as a function of FA concentration is shown in Fig. 3A. The device with immobilized permeabilized yeast cells allowed determination of FA concentration up to 4 mM at 23 ◦ C, e.g. similar to the characteristics obtained for the cells suspended in the solution. The device with immobilized intact cells did not allow determi-

3. Results and discussion 3.1. Monitoring of FA-dependent oxygen-consuming activity of the intact and permeabilized cells by the Clark electrode The dependence of oxygen consumption rate on the FA concentration at 37 and 23 ◦ C (room temperature) in the presence of intact or permeabilized yeast cells in the solution was studied using the Clark electrode. Dependence of cells’ oxygen consumption rate on FA concentration is presented in Fig. 2. In the case of intact cells, the curves reached their saturation values at 1 mM of FA at 37 ◦ C and 4 mM at 23 ◦ C (Fig. 2, doted lines). But the drop in the curves at higher FA concentrations, sharply expressed at 37 ◦ C, makes the use of intact cells as a bioselective element in biosensors of the Clark type rather complicated. As for the permeabilized cells, the curves reached their limitation value at 4 and 12 mM of FA at 37 and 23 ◦ C, respectively. A difference in profiles of the dynamic range for permeabilized and intact cells can be explained by a different nature of the enzymes involved in FA oxidation. For intact cells, both enzymes, formaldehyde dehydrogenase (FDH) and AOX, could be responsible for biosensor’s response (AOX, consuming oxygen directly, and FDH, via re-oxidation of NADH in mitochondria coupled with oxygen consumption) AOX

CH2 O + H2 O + O2 = HCOOH + H2 O2 , CH2 O + H2 O + NAD+ = HCOOH + NADH(H+ ), FDH

2NADH(H+ ) + O2

2NAD+ + 2H2 O

mitochondria

=

Fig. 3. (A) Dependence of oxygen-consuming activity of immobilized intact (doted line) and permeabilized cells (solid line) of the yeast H. polymorpha C105 on FA concentration. The experiments were carried out at 23 ◦ C in 0.1 M K-phosphate buffer, pH 7.0. (B) The changes of response for Clark electrode functionalized by immobilized intact (doted line) and permeabilized yeast cells (solid line), pre-treated initially by FA at different concentrations for 10 min. All signals were measured for 1 mM FA.

938

M. Khlupova et al. / Talanta 71 (2007) 934–940

Table 1 Bioanalytical characteristics of the mutant yeast-based amperometric biosensors selective to formaldehyde Bioelement/transduction

Linear dynamic range (mM)

Detection limit (mM)

Sensitivity

Immobilized intact cells/Clark electrode Immobilized permeabilized cells/Clark electrode Immobilized intact cells/redox mediator DCIP Immobilized permeabilized cells/redox mediator DCIP

0.05–3.0a

0.76a

1.15a ␮M s−1

mM−1

Correlation coefficient 0.978

0.3–4.0

0.27

0.44 ␮M s−1 mM−1

0.998

1.0–7.0

0.74

8.62 nA mM−1

0.991

No signal

Notes: All parameters in the table were calculated as described in [18]; measurements were carried out at 23 ◦ C. a At correlation coefficient lower than 0.99.

nation of FA concentration due to a very complicated profile of the calibration curve (Fig. 3A, doted line). Probably, two-peak pattern of the curve can be explained by a two-enzyme nature of FA oxidation in intact cells. The first stage in an increase of current may be related with FDH-catalyzed reaction, which has a lower KM value to FA, than AOX (0.25 mM [16] versus 10.5 mM for AOX [17]). The following decrease in the curve could be caused by a damaging effect of FA on cellular functions, first of all, on NAD+ regeneration. The last stage of an increase in current can be explained by involvement of AOX in FA oxidation in intact cells at saturated for this enzyme FA concentrations. In Table 1 the basic parameters of the biosensors based on the Clark electrode are summarized. As can be seen from the table, the device with intact yeast cells H. polymorpha possessed quite lower detection limit compared to the biosensor based on the permeabilized immobilized cells in spite of the more than twice higher sensitivity. It should be also mentioned that statistical analysis of the calibration curve of the device with intact cells did not revealed a well-expressed linear range of the response of the biosensor, e.g. correlation coefficient for linear regression was lower than 0.99 (Table 1). Thus, the immobilized intact cells did not provide a good linear relationship for the determination of FA. These results differ from that ones obtained in a previously reported work by using the pH-FET-based potentiometric transducer coupled with intact yeast cells, where the calibration curve was linear within the range of analyte concentration from 5 to 50 mM [10]. However, both designed amperometric biosensors for FA detection were significantly more sensitive as compared to intact yeast cells-based potentiometric one. To evaluate a deteriorate effect of high concentrations of FA on the characteristics of the biosensors, the change of the current at FA concentration of 1 mM was measured after the preliminary treatment of the permeabilized and intact cells by increasing concentrations of FA (2–18 mM). Obviously, a decrease in current for permeabilized yeast cells was observed after such treatment (Fig. 3B, solid line). As for the intact cells (Fig. 3B, doted line), pretreatment of the cells by increasing FA concentrations caused an initial decrease of the response (probably, due to inhibition

Fig. 4. Typical real time current responses of biosensors based on intact (doted line) and permeabilized (bold solid line) H. polymorpha cells immobilized on platinum electrode (10 mM Tris–succinate buffer, pH 7.0, 10 mM CaCl2 , and 100 mM KCl). The responses were measured after addition of DCIP followed by addition of FA up to 5 mM.

of NAD+ -regeneration in the cells), followed by a temporally increased signal. 3.3. Biosensor based on a screen-printed platinum electrode covered with the cells The current produced by the yeast cells in the presence of mediators was measured using the screen-printed platinum electrode covered with the cells immobilized in a Ca2+ -alginate gel. It was shown that both intact and permeabilized cells produced rather high current after DCIP addition as shown in Fig. 4. The current responses to DCIP were approximately twice higher for intact yeast cells than for permeabilized ones. After FA addition typical signal of the electrode with immobilized intact cells can be clearly seen, whereas no response towards FA was obtained for the permeabilized cells-based device (Fig. 4).

M. Khlupova et al. / Talanta 71 (2007) 934–940

939

inside the cells, in FA-selective bioelement in amperometric biosensors based on the electrode of Clark type. It is of importance that the activity of the enzyme is preserved enough well under the usual conditions of a storage at 4 ◦ C in the dried cells. The dynamic range for FA in the case of the Clark electrode has been found to be 0.3–4.0 mM. The deteriorate effect of FA on the cells/enzyme resulted in a drop of the signal at higher concentrations of FA. These results differ from that ones obtained in a previously reported work by using the pH-FET-based potentiometric transducer coupled with intact yeast cells, where the calibration curve was linear within the range of analyte concentration from 5 to 50 mM [10]. Thus, the designed amperometric biosensor for FA detection is approximately 10 times more sensitive compared to the intact yeast cells-based potentiometric device. As for the amperometric biosensors by using redox mediator DCIP, the permeabilized yeast cells cannot be used as a bioselective element for amperometric assay of FA. The electron transfer from AOX to the mediator is practically absent in permeabilized yeast cells, because of competition in the reaction with oxygen, natural electron acceptor of the enzyme. However, intact cells are perspective bioelements for the construction of amperometric biosensors by using redox mediators. The devices for FA detection designed in this work possessed the broad linear dynamic range (1.0–7.0 mM of FA) along with the high sensitivity and acceptable detection limit. Obviously, it is necessary to perform a further directed genetic engineering of the yeast cells to improve their bioanalytical characteristics in the corresponding biosensors. Acknowledgement The work was financially supported by INTAS Open Call project No. 03-51-6278. Fig. 5. (A) Typical real time current responses of the biosensor based on intact H. polymorpha cells immobilized on platinum electrode (10 mM Tris–succinate buffer, pH 7.0, 10 mM CaCl2 , and 100 mM KCl) at different concentrations of FA. (B) Calibration curve for the biosensor.

Fig. 5A shows typical real time current responses of the biosensor based on intact H. polymorpha cells immobilized on platinum electrode, and Fig. 5B presents a calibration curve for FA concentrations. The dependence of the electrode response on FA concentration is linear within the range from 1 to 7 mM of FA. The sensitivity of the sensor towards FA was found to be 8.62 nA/mM. It was shown that the detection limit for the device based on the use of mediator for FA detection was much higher compared to the biosensor based on the Clark electrode (Table 1). The reproducibility between five cells modified platinum electrodes, prepared and used different days, was about 25%. 4. Conclusion The permeabilized cells of the mutant yeast H. polymorpha C-105 can be used as a source of AOX, naturally immobilized

References [1] V.J. Feron, H.P. Til, F. De Vrijer, R.A. Woutersen, F.R. Cassee, P.J. Van Bladeren, Mutat. Res. 259 (1991) 363. [2] E. Tyihak, L. Trezl, B. Szende, Ann. NY Acad. Sci. 851 (1998) 259. [3] W.H. Chan, T.Y. Xie, Anal. Chim. Acta 339 (1997) 173. [4] A. Kettrup (Ed.), Anal. Hazard. Subst. Air 4 (1999). [5] B. Winter, K. Cammann, GBF Monogr. 17 (1992) 75. [6] E.A.H. Hall, M. Preuss, J.J. Gooding, M. Hammerle, NATO ASI Ser., Ser. 2: Environ. 38 (1997) 227. [7] Y.I. Korpan, A.P. Soldatkin, M.V. Gonchar, A.A. Sibirny, T.D. Gibson, A.V. El’skaya, J. Chem. Technol. Biotechnol. 68 (1997) 209. [8] K.P. Rindt, S. Scholtissek, GBF Monogr. 13 (1989) 405. [9] Y.I. Korpan, M.V. Gonchar, N.F. Starodub, A.A. Shul’ga, A.A. Sibirny, A.V. El’skaya, Anal. Biochem. 215 (1993) 216. [10] Y.I. Korpan, M.V. Gonchar, A.A. Sibirny, C. Martelet, A.V. El’skaya, T.D. Gibson, A.P. Soldatkin, Biosens. Bioelectron. 15 (2000) 77. [11] M. Gonchar, M. Maidan, Y. Korpan, V. Sibirny, Z. Kotylak, A. Sibirny, FEMS Yeast Res. 2 (2002) 307. [12] M.V. Gonchar, M.M. Maidan, O.M. Moroz, J.R. Woodward, A.A. Sibirny, Biosens. Bioelectron. 13 (1998) 945. [13] M.V. Gonchar, G.P. Ksheminska, N.M. Hladarevska, A.A. Sibirny, in: T.M. Lachowicz (Ed.), Genetics of Respiratory Enzymes in Yeasts, Wroclaw University Press, Wroclaw, Poland, 1980, p. 222.

940

M. Khlupova et al. / Talanta 71 (2007) 934–940

[14] M.V. Gonchar, L.B. Kostryk, A.A. Sibirny, Appl. Microbiol. Biotechnol. 48 (1997) 454. [15] A. Heiskanen, J. Yakovleva, C. Spegel, R. Taboryski, M. Koudelka-Hep, J. Emneus, T. Ruzgas, Electrochem. Commun. 6 (2004) 219. [16] H. Schuette, J. Flossdorf, H. Sahm, M.R. Kula, Eur. J. Biochem. 62 (1976) 151.

[17] S.V. Shleev, G.P. Shumakovich, O.V. Nikitina, O.V. Morozova, H.M. Pavlishko, G.Z. Gayda, M.V. Gonchar, Biochemistry (Moscow) 71 (2006) 245. [18] B. Haghighi, L. Gorton, T. Ruzgas, L.J. Jonsson, Anal. Chim. Acta 487 (2003) 3.