Äîêóìåíò âçÿò èç êýøà ïîèñêîâîé ìàøèíû. Àäðåñ îðèãèíàëüíîãî äîêóìåíòà : http://analyt.chem.msu.ru/kinetics/papers/Bagirova%20Effect%20of%20Mercury(II)%20Traces%20Anal%20Lett.%202006.pdf Äàòà èçìåíåíèÿ: Thu Sep 25 14:13:54 2008 Äàòà èíäåêñèðîâàíèÿ: Mon Oct 1 20:17:53 2012 Êîäèðîâêà:

Analytical Letters, 39: 521 ­ 541, 2006 Copyright # Taylor & Francis Group, LLC ISSN 0003-2719 print/1532-236X online DOI: 10.1080/00032710500536137

BIOANALYTICAL

Effect of Mercury(II) Traces on Catalytic Activity of Peanut and Horseradish Peroxidases
Nailya A. Bagirova, Svetlana V. Muginova, Tatyana N. Shekhovtsova, and Irina G. Gazaryan
Chemistry Department, Lomonosov Moscow State University, Moscow, Russia

Robert B. van Huystee
Department of Plant Sciences, University of Western Ontario, London, Ontario, Canada

Abstract: Mercury(II) in the range of 0.1 ­ 1 mg L21 concentrations was found to be a much more efficient inhibitor of native peanut peroxidase (PNP) than of horseradish peroxidase (HRP) in the reaction of o-dianisidine oxidation with hydrogen peroxide. The possible reason for the different degree of mercury(II) effects on the catalytic activity of both enzymes was studied. It was shown that the different number of glycans in PNP and HRP molecules (three and eight, respectively), or their absence in the molecule of wild-type recombinant horseradish peroxidase refolded from E. coli inclusion bodies (recHRP), does not play a significant role in the effects caused by mercury(II). The efficient inhibition of PNP by mercury(II) in the absence of any other additives (for example, thiourea) originates from a greater mobility of the distal calcium ion in the enzyme molecule. A model scheme for the interaction of the studied plant peroxidases with mercury(II) was proposed. The PNP-based enzymatic method for mercury(II) determination with cmin ¼ 0.04 mgL21

Received 25 October 2005; accepted 26 October 2005 The authors thank Prof. Dr. E.S. Chukhray (Physical Chemistry Division of MSU) for fruitful discussions and the Russian Foundation for Basic Research for financial support (project No 04-03-33116). Address correspondence to Tatyana N. Shekhovtsova, Chemistry Department, Lomonosov Moscow State University, Leninskie Gory, 119992 Moscow, Russia. E-mail: shekhov@analyt.chem.msu.ru 521


522

N. A. Bagirova et al.

(0.2 nmol L21) was developed and the possibility of PNP application for analysis of different samples was demonstrated. Keywords: Peanut peroxidase, native and wild-type horseradish peroxidases, inhibition, mercury(II) determination, real samples analysis

INTRODUCTION Peroxidases are hemoproteins that contain glycan chains and calcium ions and catalyze the oxidation of organic and inorganic compounds with hydrogen peroxide (Welinder, 1992). The X-ray structural analysis of plant peroxidases (Schuller et al. 1996; Gajhede et al. 1997; Henriksen et al. 1998) reveals the similar arrangement of key conservative catalytic amino acid residues in the active sites of peroxidases from diverse sources. However, the accessibility of the enzyme active site, the conformation of surrounding peptidic chains, and number of glycans are specific for a particular enzyme. This results in different profiles of peroxidase substrate specificity, different stability, and different sensitivity towards the action of effector compounds. Peroxidases are of great bioanalytical potential, and the recently published data on structures and properties of novel plant peroxidases isolated from peanut (Welinder, 1992), horseradish (Gajhede et al. 1997), barley grain (Hendriksen et al. 1998), and tobacco (Gazaryan, et al. 1998) are of great interest. Horseradish peroxidase (HRP) among other plant peroxidases, is the most studied enzyme, and mercury(II) was found to be the most efficient inhibitor of this enzyme (Dolmanova et al. 1979). A highly sensitive and selective enzymatic method for mercury(II) determination (0.01 ­ 0.5 mg L21) based on HRP inhibition in the reaction of o-dianisidine oxidation in the presence of thiourea has been developed (Shekhovtsova et al. 1996). The analytical application of PNP has just begun. So far it has been applied in biosensors for the determination of aromatic amines and phenols (Munteanu et al. 1998; Bagirova et al. 1998; Bagirova et al. 2000). The minor structural changes between HRP and PNP revealed by structural analysis (Schuller et al. 1996; Gajhede et al. 1997) may explain the differences in their catalytic behavior. The present paper addresses the issues connected with the peroxidase response to the action of mercury(II), which is one of the most toxic metals with a tendency to accumulate in brain, kidneys, and testicles (Zigel 1989). The possible difference in mercury(II) effects on peroxidases of different origin, e.g., HRP and PNP, is both of fundamental and practical importance. The goal of the present work is to study the effect of mercury(II) on PNP catalytic activity, compare the results obtained with the previous ones on HRP inhibition, and develop a novel method for mercury(II) determination using PNP.


Effect of Trace Amounts of Mercury(II)

523

EXPERIMENTAL Reagents Potassium biphthalate, KOH, H3BO3, Na2B4O7 (Reakhim, Russia), and Tris (Serva, Germany) were used to prepare 0.1 mol L21 phthalate (pH 4.0 ­ 6.5), 0.15 mol L21 borate (pH 7.0) and 0.05 mol L21 Tris (pH 7.5) buffer solutions. Hydrogen peroxide solutions were purchased from Merck, Germany. The concentration of H2O2 solutions was determined by permanganometry. The solid preparations of histidine (Sigma, USA), o-dianisidine, and thiourea (Reakhim, Russia) were used throughout. The solutions of o-dianisidine were prepared daily by dissolving weighed amounts in ethanol; histidine and thiourea were dissolved in water. Mercury(II) nitrate solutions (1 g L21) were prepared by dissolving weighed amounts of the purified metal in a minimum quantity of distilled concentrated HNO3. The stock solution of Fe(III) (1 g L21) was prepared by dissolving accurately weighted amounts of iron (purified by carbonyl method) in 3 ­ 4 drops of concentrated HNO3 (very pure) and subsequent dilution with water to the required volume. The solutions of other metals with 1 g L21 concentration were prepared by dissolving the exact portions of their salts in HNO3-acidified water (1.00 mL concentrated HNO3 per 100.0 mL water). The solutions with lower metal concentrations were prepared by successive dilution of stock solutions with HNO3-acidified water. The diluted solutions of metal salts had pH 3.0. To prepare solutions of samarium(III) chloride, samarium(III) oxide (Reakhim, Russia) was dissolved in 5 mL of high-purity concentrated HCl by heating the solution till the oxide was completely dissolved. The solution was evaporated to wet salts, which were then dissolved in 50 mL water under slight heating. The concentration of samarium(III) stock solution was determined by absorbance [1403 ¼ 3.3 mol21 L cm21 (Peshkova and Gromova 1985). The solutions with lower concentrations of samarium(III) were prepared by dilution of its stock solutions with water. The solution of CH3HgI was prepared daily by dissolving an accurately weighed amount in ethanol; potassium ethylenediaminetetraacetate (EDTA) was dissolved in water. MilliQ water was used thoughout the experiments to prepare all the aqueous solutions. Three preparations of plant peroxidases (EC 1.11.1.7) were used. The HRP isozyme C was purchased from Sigma, USA (RZ ¼ 2.2). Cationic PNP was isolated from the cultural medium of Arachis Hypogea cells (Sesto and van Huystee 1989) (RZ ¼ 2.0). Wild-type recHRP (RZ ¼ 2.1) was expressed and refolded from E. coli inclusion bodies as described earlier (Gazaryan et al. 1994). Stock solutions of PNP and HRP at concentrations of 1 ­ 10 mmol L21 were made by dissolving the solid preparations of the enzymes in 0.05 mol L21 Tris-HCl (pH 7.5) and 0.15 mol L21 borate (pH 7.0) buffer solutions, respectively. Wild-type recHRP was stored in


524

N. A. Bagirova et al.

5 mmol L21 solution in 0.05 mol L21 Tris-HCl buffer, pH 7.5. The concentrations of PNP, native HRP, and wild-type recombinant HRP stock solutions were determined by the Soret band absorbance [1405 ¼ 112 (Lambeir et al. 1985), 1403 ¼ 94 (Shannon et al. 1996), 1403 ¼ 102 mmol21 L cm21 (Gazaryan et al. 1994), respectively]. Solutions with lower enzyme concentrations were prepared daily by dilution of their stock solutions with 0.05 mol L21 Tris-HCl (pH 7.5) in the case of PNP and recHRP, and 0.15 mol L21 borate buffer solution (pH 7.0) in the case of native HRP. The specific activities of PNP, native and recHRP towards o-dianisidine determined in accordance with the protocol (Lebedeva et al. 1977) were equal to 700, 700, and 750 units mg21, respectively. The solid enzyme preparations and solutions were refrigerated at × 48C. Apparatus The absorbance of peroxidases and samarium (III) solutions was measured with a Shimadzu UV-2201 spectrophotometer (Japan) (l ¼ 1 cm). A KFK-2 photoelectro-colorimeter (Russia, Zagorsk) (l ¼ 460 nm; l ¼ 2 cm; deionized water as a reference) was used for measuring the absorbance of the reaction solutions. The pH of aqueous solutions was measured with an accuracy of + 0.005 using a potentiometer in the pH-meter mode (Econics-Expert-001, Russia). Procedure The following components were placed sequentially into a glass test-tube with a ground-glass stopper: phthalate buffer solution (pH 5.0), enzyme solution, inhibitor solution (mercury(II) or thiourea or their mixture), o-dianisidine solution, and water up to 10 mL total volume of the reaction mixture. Finally, hydrogen peroxide solution was introduced. At the moment when hydrogen peroxide was added and the reaction solution was mixed, a stop-watch was started and the absorbance at 460 nm was measured at 15 s intervals for 2 min. Blank experiments were carried out without adding inhibitors. Kinetic curves were plotted as absorbance vs. time and the absolute value of the initial velocity of the indicator reaction (y 0, mmol21 L min21 was calculated as y o ¼ DC/Dt ¼ DA/Dt . 1/l1 ¼ tan a/l1 where Dc is the increment of the product concentration at the reaction time (Dt), Aisa light absorption, l is the cell length (2 cm), 1 is the molar absorbance coefficient of the product of o-dianisidine oxidation at 460 nm (30 mmol L21cm21) (Lebedeva et al. 1977); tan a is the average value of a slope of a kinetic curves plotted as absorbance (A) at 460 nm vs. time (t, s) (n ¼ 5). In experiments with preincubation, a mixture consisting of buffer, enzyme solution, and mercury(II) (or thiourea or their mixtures) was standed for a certain time; then the other components were introduced in the reaction mixture. The inhibition degree (I, %) was calculated as I (%) ¼ 100% . (y o ­ y )/y o, where y o and y are the initial velocities of o-dianisidine oxidation without and


Effect of Trace Amounts of Mercury(II)

525

in the presence of an inhibitor [mercury(II) or thiourea or their mixture], respectively. The relative activity (y /y o) of plant peroxidases was determined using the initial velocities of the indicator reaction--y o and y --mentioned earlier. All experiments were carried out at room temperature. Note that the velocity of o-dianisidine oxidation changes up to 2% with a change in temperature of 18C (Ugarova et al. 1979).

Samples Preparation Aquatic samples containing Hg(II) were collected from Moscow river. The collected samples were acidified with concentrated HNO3 to pH 4.0, then were transferred to plastic bulbs and stored at 48C until analysis. For the study, samples of podzolic soil from the Moscow region were selected and stored in PE flasks. The soil samples preparation for analysis was conducted in accordance with the procedures described in (Veselova and Shekhovtsova (2000) and Zhou et al. (1996)). The pretreatment of biological samples (urine and blood serum) included the dilution of a urine sample with water acidified with 2 mol L21 HCl (1.00 mL HCl per 9.00 mL of the sample) and precipitation of blood proteins with a 20% trichloroacetic acid and following centrifugation.

RESULTS AND DISCUSSION Individual Mercury(II) Effect on the Catalytic Activity of PNP The reaction of o-dianisidine oxidation with hydrogen peroxide was used as an indicator reaction for studying the influence of Hg(II) on the catalytic activity of PNP. The choice of this indicator reaction was determined by the fact that its mechanism, kinetics, and optimum conditions were established previously (Bagirova et al. 1999): concentration of PNP, 0.1 nmol L21; o-dianisidine, 0.12 mmol L21; hydrogen peroxide, 0.15 mmol L21; a phtalate buffer solution, pH ¼ 4.8 ­ 5.7. Moreover, the effect of Hg(II) on the catalytic activity of HRP was studied in detail in the reaction of o-dianisidine oxidation (Dolmanova et al. 1979; Shekhovtsova et al. 1996). It was stated that PNP was inhibited by Hg(II) at significantly lower concentrations (0.001 ­ 1 mg L21) than those for HRP (25 ­ 100 mg L21) without preincubation with the enzyme. The inhibitory effect from Hg(II) decreased with a decrease in the metal concentration, however, a linear dependence of the velocity of the indicator reaction on Hg(II) concentration was observed at 0.1 ­ 1 mg L21 Hg(II) only. As an example of 0.5 mg L21 Hg(II) concentration, the working conditions of metal determination was optimized. So, it was shown that the maximum difference in velocities of the indicator reaction in the absence


526

N. A. Bagirova et al.

and presence of Hg(II) was observed with 60 mmol L21 o-dianisidine and 0.15 mmol L21 hydrogen peroxide concentrations, and almost at the same pH range (5.0 ­ 5.7) as the optimum for the indicator reaction in the absence of the inhibitor (Fig. 1, curves 1, 2). This result agrees well with the data obtained previously (Shekhovtsova et al. 1996) for the mercury(II) effect on the activity of native HRP. Hence, the introduction of Hg(II) to the indicator system within the indicated pH range does not result in a significant change in the intrinsic pK values of both enzymes. The nature of dissociating groups in HRP was descibed in

Figure 1. The pH-dependence of the indicator reaction velocity in a phthalate buffer solution in the absence 1), and presence of Hg(II) 2), thiourea (tinc ¼ 20 min) 3); thiourea (tinc ¼ 20 min) and Hg(II) 4). Concentrations used: PNP, 0.1 nmol L21; o-dianisidine, 60 mmol L21; hydrogen peroxide, 0.15 mmol L21; thiourea, 1 mmol L21; Hg(II), 0.5 mgL21.


Effect of Trace Amounts of Mercury(II)

527

detail by Phelps et al (1971). They were found to be carboxyl groups of Asp and Glu residues (pKa 4.3) and His imidazole groups with a pKa of 6.65. In the case of PNP, a remarkable feature of pH-dependence is a sharp maximum at pH 4.9 ­ 5.7. A similar pH-dependence for native HRP has a plateau in a wider range of pH 4.5 ­ 6.8 (Shekhovtsova et al. 1993). The difference in pH-dependence for the enzymes may be evidence of either a different environment of the same catalytically active ionogenic groups or the participation of different groups in catalysis. The Dixon method (Varfolomeev and Gurevich 1999) provides an estimation of the pKa value of the first catalytically active ionogenic group of PNP as pKa,1 ¼ 5.0. As to pKa,2, we propose that its pKa is 6.0. The location of the maximum with pKa 5.2 was more expedient from our point of view for the obtained pH-curves. We made an attempt to increase the inhibitory effect of mercury(II) and the dependence of PNP inhibition degree on the time of its preincubation with metal was studied. As shown in Fig. 2 (curve 1), the inhibitory action

Figure 2. The dependence of the PNP inhibitory degree on the preincubation time with Hg(II) 1), and thiourea 2), Concentrations used: PNP, 0.1 nmol L21; o-dianisidine, 60 mmol L21; hydrogen peroxide, 0.15 mmol L21; thiourea, 1 mmol L21; Hg(II), 0.5 mg L21.


528

N. A. Bagirova et al.

of Hg(II) was enhanced up to I ¼ 35% for relatively short (1 ­ 20 min) preincubation times and did not change with longer times (. 20 min). A decrease in the PNP activity with time indicates irreversible binding of the inhibitor to the enzyme. An irreversible character of Hg(II)-induced changes in the enzyme molecule was confirmed by the methods of dilution and dialysis carried out as described in (Ugarova and coworkers (1981)). So, it was stated that the inhibition degree of PNP by 0.5 mg L21 Hg(II) increases not more than 4% with increasing of dialysis time from 2 up 24 h. On the contrary, Hg(II) was found (Shekhovtsova et al. 1996) to be a reversible competitive inhibitor of HRP in the reaction of o-dianisidine oxidation. Thus, a considerable difference in the mechanism of mercury(II) influence on the both studied plant peroxidases exists.

Influence of Mercury(II) on the Catalytic Activity of PNP in the Presence of Thiourea The inhibitory action of Hg(II) on HRP increases significantly in the presence of a number of sulfur-containing organic compounds and, to the highest extent thiourea (Dolmanova et al. 1979; Shekhovtsova et al. 1979). The possibility of making HRP ready for the attack of another inhibitor--Hg(III)--with the help of thiourea, which reduces disulfide bridges formed by cysteine residues in the enzyme molecule, was pointed out as a probable reason for such action of thiourea (Shekhovtsova et al. 1979; Torchinskyi et al. 1977). Thus, the tertiary enzyme structure partially decomposes, the protein globule unfolds, and the interaction of Hg (II) with both carboxyl and thiol groups, formed as a result of S-S bonds reduction, becomes easier. It is well known that the affinity of Hg (II) ions to the latter is high (Dolmanova et al. 1987). It should be noted that the simultaneous addition of thiourea and Hg(II) to the indicator reaction catalyzed by HRP did not increase the inhibitory effect of the HRP. Preincubation of a mixture of the enzyme with thiourea for not less than 30 min, which resulted in a 15 ­ 20% decrease in the enzyme activity, was necessary. The Hg(II) had to be added to the indicator reaction only after this stage. The maximum inhibitory effect (I . 30%) of mercury(II) was achieved after successive incubation of the mixture [HRP × thiourea × Hg(II)] for 30 min. Under these conditions, native HRP is inhibited effectively by 0.001 ­ 1 mg L21 Hg(II) at the concentration of thiourea 0.5 mmol L21. Thus, the addition of thiourea enhances the sensitivity of Hg(II) determination using HRP by a factor of 20.000 (Dolmanova et al. 1979; Shekhovtsova et al. 1979). In the case of PNP, thiourea at 1 mmol L21 concentrations exerts an inhibitory action on PNP, which increases with time (Fig. 2, curve 2). As can be seen from Fig. 2, preincubation of PNP with a 1 mmol L21 thiourea solution for 20 min leads to a increase in the inhibition degree by 24% in comparison with that observed without preincubation. It was shown, additionally,


Effect of Trace Amounts of Mercury(II)

529

that the maximum difference between the velocities of the indicator reaction in the absence and presence of 0.5 mg L21 Hg(II) was observed at the concentration of thiourea of 1 mmol L21. This concentration was chosen as optimum for further investigations. Study of pH influence on the velocity of o-dianisidine oxidation in the presence of thiourea showed that the inhibitory action of thiourea and its mixture with Hg(II) on PNP activity is maximum at pH 5.0 ­ 5.5 (Fig. 1, curves 3 and 4), which is considered to be optimum for the investigation of the combined effects of these inhibitors. The catalytic activity of PNP in the presence of thiourea only, as well as in the mutual presence of both inhibitors, reached the maximum at pH values higher than those without thiourea (Fig. 1, curves 3 and 4). A change of pH dependence in the presence of thiourea showed a change in the environment of the first catalytically active ionogenic group of PNP with the attributed pK1 of 5.0. The same situation was observed for HRP (Dolmanova et al. 1979). Such character of pH-dependence of the velocity of o-dianisidine oxidation catalyzed by PNP and HRP in the presence of thiourea may probably result from the following tautomerization (Barton and Ollis 1980) (Scheme 1). The maximum decrease in the catalytic activity of PNP (I ¼ 38%) is achieved for the enzyme preincubation with thiourea in buffer solution during 20 min followed by Hg(II) addition and incubation of the obtained four-component mixture for 20 min. However, the degree of the individual inhibitory action of 0.5 mg L21 Hg(II) for its 20 min preincubation with PNP (I ¼ 35%) (Fig. 2, curve 1) is comparable (I ¼ 38%) to that of the combined action of mercury(II) with thiourea for 40 min preincubation. Thus, the data show that the degree of Hg(II) influence on the catalytic activity of PNP and HRP in the same indicator process in the presence of thiourea is also different. Evidently, the interaction of Hg(II) with PNP and HRP proceeds according to different mechanisms probably connected with some differences in their structures. We made an attempt to find out possible correlations between the different degree of the mercury(II) influence on PNP and HRP and the structures of these enzymes. Comparative Study of the Mercury(II) Effect on the Catalytic Activity of PNP, and Native and Wild-Type HRP Comparative analysis of the existing data on the structures of plant peroxidases showed that different numbers of glycans in the molecules of PNP

Scheme 1.


530

N. A. Bagirova et al.

and HRP (3 and 8, respectively) (Wan and van Huystee 1993; Welinder 1979) seems to be one of the main differences in the structures of these enzymes. Therefore, we supposed that functional groups of the protein globule of PNP could be more accessible for the Hg(II) effect. Thus, Hg(II) inhibits PNP at low concentrations without the enzyme pretreatment with thiourea. The following question arises: to what degree can peroxidase with no glycans attached be inhibited by Hg(II) Wild-type HRP expressed in E. coli is a non-glycosylated peroxidase (Smith et al. 1991). The influence of metals on the catalytic activity of this enzyme has not been investigated yet. The investigation of the effect of Hg(II) in a wide range of its concentrations on the catalytic activity of three peroxidases (native PNP, HRP, and recHRP) in the reaction of o-dianisidine oxidation has shown that low Hg(II) concentrations (0.1 ­ 1 mg L21) inhibited PNP only (Fig. 3). In contrast to PNP, which was inhibited immediately after the addition of 0.5 mg L21 Hg(II) and the inhibitory effect of the metal increased with time (Fig. 4, curve 1), native and recHRP were inhibited weakly with 1 mgL21 Hg(II) only after their preincubation for at least 45 min (Fig. 4, curves 2, 3). Thus, these data showed that the inhibition degree of Hg(II) did not depend directly on the number of glycans in the molecule of peroxidase. Otherwise, Hg(II) would have a pronounced inhibitory effect on wildtype recHRP. It was interesting to examine a change in the inhibitory action of mercury(II) for long preincubation times with thiourea for three studied

Figure 3. The dependence of the catalytic activity of PNP 1), native 2), and wild-type recombinant 3) HRP on Hg(II) concentration. Concentrations used: PNP and native HRP, 0.1 nmol L21; wild-type HRP, 0.08 nmol L21; o-dianisidine, 60 mmol L21; hydrogen peroxide, 0.15 mmol L21. A phthalate buffer solution, pH 5.0; tinc ¼ 0.


Effect of Trace Amounts of Mercury(II)

531

Figure 4. The dependence o 3) HRP on the preincubation HRP, 0.1 nmol L21; wild-type gen peroxide, 0.15 mmol L21;

f the catalytic activity of PNP 1), native 2), and wild-type time with Hg(II). Concentrations used: PNP and native HRP, 0.08 nmol L21; o-dianisidine, 60 mmol L21; hydroHg(II), 0.5 mgL21 (phthalate buffer solution, pH 5.0).

peroxidases under the same conditions. Native HRP was inhibited by thiourea after its preincubation for 1.5 h (Fig. 5, curve 1). A difference between the velocities of the indicator reaction in the absence and presence of Hg(II) appeared at preincubation times of more than 2 h (Fig. 5, curves 1 and 10 ). This difference increased with time and reached a maximum after 4 h preincubation of HRP with thiourea, which was consistent with the previous observations (Dolmanova et al. 1979). Wild-type recHRP was inhibited by simultaneous addition of thiourea and Hg(II), although, thiourea itself did not change the indicator reaction velocity after its preincubation with the enzyme for less than 2 h (Fig. 5, curves 2 and 20 ). The inhibitory action of Hg(II) on wild-type recHRP was maximum for 2.5 ­ 3 h preincubation with thiourea (Fig. 5, curve 20 ). In the case of PNP, thiourea decreased the indicator reaction velocity immediately after its introduction to the reaction and continued to reduce it


532

N. A. Bagirova et al.

Figure 5. The dependence of the catalytic activity of native (1 and 10 ), wild-type recombinant (2 and 20 ) HRP and PNP (3 and 30 ) on the preincubation time with thiourea in the absence (1 ­ 3) and presence (10 ­30 ) of Hg(II). Concentrations used: PNP and native HRP, 0.1 nmol L21; wild-type HRP, 0.08 nmol L21; o-dianisidine, 60 mmol L21; hydrogen peroxide, 0.15 mmol L21; thiourea, 2.5 mmol L21; Hg(II), 1 mgL21 (phthalate buffer solution, pH 5.0).

during its preincubation with the enzyme (Fig. 5, curve 3). At the same time, the inhibitory effect of Hg(II) on PNP after their pre-incubation for 4 ­ 5 h (Dy /y o ¼ 0.10 ­ 0.12) was similar to the combined action of Hg(II) with thiourea (Dy /y o ¼ 0.10), when they were added to the reaction simultaneously (without pre-incubation) (Fig. 5, curves 3 and 30 ). The indicator reaction velocity changed only slightly after Hg(II) addition to the mixture of the enzyme and thiourea incubated for more than 1 h but less than 3 h. The character of the data obtained for native and wild-type recHRP leads us to the conclusion that thiourea makes both enzymes "ready" for


Effect of Trace Amounts of Mercury(II)

533

the interaction with Hg(II) ions by reduction of disulfide bridges. Moreover, the absence of glycans in the molecule of wild-type recHRP makes the access of the inhibitor to the enzyme globule easier and shortens the time needed to observe the maximum effect of Hg(II). Thus, a significant decrease in the velocity of the indicator reaction catalyzed by PNP after the addition of Hg(II) to the enzyme that was not ready for the interaction with it indicates an additional mechanism functioning in the case of PNP.

Possible Reasons for the Mercury(II) Influence on the Catalytic Activity of PNP The character of Hg(II)'s effect on PNP activity may be connected with the incorporation of metal ions in the cavity of the enzyme globule near its active site or with the presence of some functional groups in the enzyme that are capable of interacting with Hg(II). It was shown that samarium(III), which has the closest ionic radius to Hg(II) [0.112 and 0.113 nm (Lurie 1989), respectively], at its concentrations of 0.1 ­ 300 mg L21 did not inhibit PNP after its immediate addition to the indicator system as well as after 1 h preincubation with the enzyme. Thus, the first assumption is not valid. The Hg(II) ions exhibit a high affinity for SH-groups in proteins, which is much higher than that for other functional groups in proteins (Torchinskyi 1977). Mercury(II) inhibits effectively the enzymes of different classes containing free SH-groups like urease (Andrews and Resthel 1970), formate dehydrogenase (Popov and Egorov 1979), and alcohol dehydrogenase (Magonet et al. 1992). On the other hand, Hg(II) exerts weak inhibitory action on cholinesterase, which mainly contains disulfide bridges and an insufficient number of free thiol groups (Schumacher et al. 1986; van Huystee et al. 1992). This data may point to the presence of free thiol groups in PNP. However, the resolved crystal structures of PNP and HRP confirm the presence of eight cysteine residues forming four disulfide bridges in protein globules of both enzymes (Schuller et al. 1996; Gajhede et al. 1997). The existing data show that disulfide bonds do not interact with metal ions at room temperature and pH 4 ­ 8 (Torchinskyi 1977). The splitting of S-S bonds of PNP by Hg(II) ions is barely possible, as the conditions of the indicator reaction are not severe. The data of our study of the influence of a number of metal ions [Pb(II), Cd(II), Zn(II), Bi(III), Cu(II), Co(II), Ni(II), and Mn(II)], on the catalytic activity of PNP in the reaction of o-dianisidine oxidation showed that the enzyme was inhibited to the highest extent by metals with an affinity to sulfur. The correlation between the degree of the inhibition of PNP by metal ions and their affinity to SH-groups decreased as follows: Hg(II) . Pb(II) . Cd(II) . Zn(II) (Torchinskyi et al. 1977).


534

N. A. Bagirova et al.

Study of the Histidine Influence on the Inhibitory Effect of Mercury(II) There is an opinion (Torchinskyi 1977) that heavy metal ions, and mercury(II) particularly, may bind other functional groups like imidazolic, carboxylic, etc. (although less firmly in comparison with SH-groups). One may consider Hg(II) ions bonding with imidazolic groups of His residues in the active site of PNP (His-40, 42, and 170) as a cause for a decrease in the enzyme catalytic activity. If this is true the presence of histidine in solution should relieve the inhibitory effect of Hg(II). To test this assumption, the inhibitory action of Hg(II) on PNP in the presence of histidine was studied. It was shown that 1 mmol L21 histidine did not affect the enzyme catalytic activity, i.e., the degrees of the inhibitory action of Hg(II) in the absence and presence of this amino acid were equal within the experimental error (4% and 3%, respectively). The degree did not change even after the addition of an aliquot of a mixture of histidine and Hg(II) preincubated for 15 ­ 60 min. Thus, this experiment showed that the inhibitory action of Hg(II) was evidently not a result of its interaction with the imidazolic group of active site histidines (Martell and Sillen 1978). The amino acid analysis of cationic PNP (Schuller et al. 1996) and HRP isozyme C (Welinder et al. 1979) showed that these enzymes contain 17 and 20 residues of Asp, and 6 and 7 residues of Glu, respectively. In other words, HRP has four extra COOH groups compared to PNP. However, the accessibility of carboxylic groups in the enzymes for any compounds, and Hg(II) in particular, might differ due to the different location of these residues in the protein globule.

The Role of Calcium Ions in the Inhibitory Action of Mercury(II) ~ It was previously shown (Rodriguez-Maranon et al. 1993; Barber et al. 1995) that one of the differences in the structures of HRP and PNP is the ease of dissociation of the distal calcium ion in the molecule of the latter. The data on the effect of EDTA on the activity of HRP and PNP can also give evidence a higher mobility of the calcium ion in the molecule of the latter. In particular, EDTA at 0.1 mmol L21 concentrations reduced the velocity of o-dianisidine oxidation catalyzed by PNP, however, it inhibited HRP only after its preincubation with the enzyme for 12 h (Shekhovtsova et al. 1993). The comparison of the crystal structures of recombinant HRP and PNP (Schuller et al. 1996; Gajhede et al. 1997) shows that the distal calciumbinding cavity is smaller in HRP and it is screened from solution by Phe 130 residue. The PNP has a serine residue in this position, and thus, the Ca-binding cavity is somehow exposed to the solution. The disulfide bridge


Effect of Trace Amounts of Mercury(II)

535

formed by Cys 44 and Cys 49 residues is thought to be responsible for the stability of plant peroxidases and forms a loop encompassing several of distal calcium ligand residues (Schuller et al. 1996). In principle, Hg(II) ion can substitute or more probably change the position of the distal Ca(II) near the disulfide bond in the enzyme molecule, taking into account the resemblance of their ionic radii [0.112 and 0.106 nm (Lurie et al. 1989), respectively]. Such a substitution may cause a dramatic effect on the stability of the distal domain hold only by a hydrogen bond network including Cys 44 ­ Cys 49 and Ca(II) as key elements. The heavy metal ion may lead to a partial collapse of the hydrogen bond network in the enzyme distal domain and the change in the catalytic His position, which leads to the 6-coordinated iron instead of the 5-coordinated. Such a change in the catalytic His position was shown to be responsible for the significant drop in catalytic activity of HRP towards phenolic substrates upon collapse of the hydrogen bond network by mutation of Glu 64 (Tanaka et al. 1998) and Asn 70 residues (Nagano et al. 1996). The character of the dependences of relative activities of plant peroxidases on the time of their preincubation with Hg(II) (Fig. 4) makes it possible to propose the following three-step model of the interaction of the studied enzymes with Hg(II) (Scheme 2). The first reversible step (Scheme 2, step 1) takes place in the absence of Hg(II) and involves [according to (Kutuzova and Ugarova 1981)] the dissociation of the active enzyme molecule--ECa2×, with a rate constant of k1 and a reverse reactivation involving the combination of E with calcium ion with a velocity constant of k21. The second irreversible step (Scheme 2, step 2) characterizes the probable formation of a catalytically inactive form of the enzyme in the presence of Hg(II)--ECa2×Hg2×--with a rate constant of k2. Next, the inactive form fECa2×Hg2×g can be transformed reversibly into EHg2× and calcium ion according to step 3. Thus, we have two

Scheme 2.


536

N. A. Bagirova et al.

possible processes in the indicator system in the presence of Hg(II) that can take place: the first one is decomposition of the coordination environment of the distal calcium ion in enzyme molecules taking place for a short incubation time (1 ­ 30 min) and the second is the enzyme denaturation by Hg(II) after their preincubation for 30 ­ 60 min (Fig. 4). The complex fECa2×Hg2×g for PNP formed according to step 2, Scheme 2, converts readily to the inactive form EHg2× with the first-order effective rate constant of k3eff ¼ 0.01 min21 (Fig. 4, curve 1). It may be a result of a greater mobility of distal calcium ion in the molecule of PNP. Our observations can be explained on the basis of the model presented in Scheme 2. In the short time of preincubation of plant peroxidases with Hg(II), the equilibrium characterized by a constant of K1 (Scheme 2, Eq. (1) is established in the indicator system. The process of calcium dissociation continues until a new equilibrium [with an equilibrium constant of K2 (Scheme 2, Eq. (2)] is attained (Fig. 4). In the cases of native and wild-type recombinant HRP, the irreversible step 2 does not take place at all (due to insufficient mobility of calcium ion in their molecules) or Hg(II) interacts with another (not calcium) site in the enzyme globule. However, the fact that the interaction of native and wild-type recombinant HRP with Hg(II) may be quite slow should not be neglected completely. Thus, the most probable reason for the different character of Hg(II) influence on PNP and HRP is the greater mobility of the distal calcium ion in the molecule of PNP. However, the influence of different number of glycans and the possibility of the interaction of Hg(II) ions with carboxylic groups of the protein globules of the enzymes should not be neglected.

Determination of Mercury(II) Using PNP The revealed inhibitory effect of Hg(II) on the catalytic activity of PNP in the reaction of o-dianisidine oxidation was used for mercury(II) determination. For analysis, a 7 mL portion of a phthalate buffer solution (pH 5.0), 0.1 mL of a 10 nmol L21 PNP solution; 0.1 mL of 0.1 ­ 100 mg L21 standard solution of Hg(II) (or an aliquot of the sample), and 0.3 mL of a 0.1 M EDTA solution were placed sequentially into a glass test tube with a ground-glass stopper [blank experiments were carried out in the absence of Hg(II)]. Next, 0.1 mL of a 6 mmol L21 o-dianisidine solution and water up to 10 mL total volume of the reaction mixture were added into the same test-tube. Finally, 0.1 mL of a 15 mmol L21 hydrogen peroxide solution was introduced. At the moment when hydrogen peroxide was added and the reaction solution was mixed, a stop-watch was started and the absorbance at 460 nm was measured at 15 s intervals for 2 min. Kinetic curves were plotted as absorbance vs. time and the absolute value of the initial velocity


Effect of Trace Amounts of Mercury(II)

537

of the indicator reaction (y 0, mmol L21/min) was calculated as described earlier. A linear dependence of the velocity of o-dianisidine oxidation on Hg(II) concentration made it possible to build a calibration curve for the determination of the latter in the range of 0.1 ­ 1 mg L21 concentrations. The calibration equation for Hg(II) determination is y ¼ 1.70 ­ 0.08 . x, where y is the initial velocity (y , mmol L21/min) of the indicator reaction and x is the concentration of Hg(II) (mg L21). The correlation coefficient is 0.9998,

Table 1. Comparison of Hg(II) influence on the catalytic activity of PNP and HRP in the reaction of o-dianisidine oxidation (0.1 mol/L phthalate buffer solution, pH ¼ 5.0) Enzyme Characteristics Inhibitory type Optimum concentrations of reagents for Hg(II) determination PNP Irreversible Enzyme ­ 0.1 nmol L21 o-Dianisidine-- 60 mmol L21 H2O2--0.15 mmol L21 HRP Reversible competitive Enzyme--0.3 nmol L21 o-Dianisidine ­ 50 mmol L21 H2O2--10-50 mmol L21 Thiourea--0.5 mmol L21 25 ­ 100 (in the absence of thiourea); 0.001 ­ 0.5 (in the presence of thiourea) 8 in the presence of thiourea 60 preincubation of HRP × thiourea (30 min) and HRP × thiourea × Hg(II) (30 min) 5 ­ 7 days

Applicable concentrations 0.1 ­ 1 range, mgL21

Cmin(Hg(II)), ng L2

40 in the absence of thiourea Duration of analysis, min 5 immediate addition of the reagents

1

Stability of working solution of the enzyme (nmol L21) Selectivity

3 months

Determination of 0.5 mg L21 Hg(II) is possible in the presence of 50 excess of Cd(II); 100 excess of Pb(II); 104 excess of Fe(III) (in the presence of 3 mmol L21 EDTA); methylmercury 10 mmol L21 105 excesses of Cd(II) and Bi(III); 103 excess of Pb(II); 104 excess of Fe(III) (in the presence of 0.1 mmol L21 tartaric acid); methylmercury 0.1 mmol L21


538

N. A. Bagirova et al.

RSD ¼ 8% (n ¼ 5) at the concentration of Hg(II) of 0.4 mg L21. The detection limit of Hg(II) is 0.04 mg L21. Studies of the possibility of Hg(II) determination in the presence of those metal ions [Pb(II), Cd(II), Zn(II), Bi(III)], which cause the greatest inhibitory effect on PNP activity, as well as organic mercury compound (methylmercury), showed that they interfere with the determination of 0.5 mg L21 Hg(II) at concentrations presented in the Table 1. The interfering effect of Fe(III) (at the level of its maximum permissible concentrations-- 0.3 mg L21), which activates PNP and is present in the majority of environmental and biological samples, may be diminished by addition of 3 mmol L21 EDTA. The comparative analytical characteristics of mercury(II) determination using PNP and HRP are presented in the Table 1. The sensitivity and selectivity of the developed procedure are lower in comparison with those for the HRP-based procedure in the presence of thiourea (Shekhovtsova et al. 1996). However, since it is not necessary to add the second inhibitor, thiourea, into the indicator reaction catalyzed by PNP, incubate it with the enzyme, and, finally, incubate the four-component mixture [a buffer solution × PNP × thiourea × Hg(II)], the application of PNP simplifies the procedure of mercury(II) determination and diminishes the time of analysis to 5 min (instead of 30 min in the case of HRP). Moreover, the stability of PNP solution is rather high as opposed to HRP solution (Table 1). The proposed procedure using PNP was used for Hg(II) determination in samples of river waters, soil extracts, blood serum, and urine. Several examples of Hg(II) determination are presented in Tables 2 and 3.

CONCLUSIONS Prospects of PNP application for determination of mercury(II) were demonstrated. The differences in the individual influence of mercury(II) and its combined effect with thiourea on the catalytic activity of two plant
Table 2. Results of mercury(II) determination in aceticammoniac (I) and acidic soil extracts without (II) and using microwave radiation (III) (n ¼ 5, P ¼ 0.95) Pre-treatment procedure I II III Introduced, mg/kg 2.5 2.5 5.0 2.5 5.0 Found, mg/kg 2.5 2.3 5.0 3.0 5.8 + + + + + 0.2 0.3 0.2 0.3 0.6 RSD, % 16 10 14 6 10


Effect of Trace Amounts of Mercury(II) Table 3. Results of mercury(II) determination obtained by enzymatic method (I) and atomic absorption method with cool vapor (II) (n ¼ 5, P ¼ 0.95) Concentration of mercury(II), mgL21 Sample River water Blood serum Urine I 0.15 + 0.03 1.2 + 0.2 4.8 + 0.4 II 0.15 1.2 5.0

539

peroxidases (PNP and HRP) were revealed and summarized. It was shown that high sensitivity of PNP to the mercury(II) action is determined particularly by the greater mobility of calcium ions in the molecule of PNP. The sensitive, selective, simple, and rapid enzymatic procedure for mercury(II) determination based on its inhibitory effect on PNP in the reaction of o-dianisidine oxidation was developed. Samples of river water, soil extracts, blood serum, and urine were analyzed with the use of the proposed technique.

REFERENCES
Andrews, A. and Reithel, F.J. 1970. The thiol groups of jack bean urease. Arch. Biochem. Biophys., 141: 538 ­ 546. Bagirova, N.A. and Shekhovtsova, T.N. 1999. Kinetics and chemism of o-dianisidine and 3,30 ,5,50 -tetramethylbenzidine oxidation reactions catalyzed by peroxidases from peanut and horseradish and hydrogen peroxide. Kinetica and Catalysis, 40 (2): 241 ­ 267. Bagirova, N.A., Tabatchikova, N.V., Shekhovtsova, T.N., and van Huystee, R.B. 2000. Enzymatic method for the determination of phenols with the use of peanut peroxidase. J. Anal. Chem., 55 (1): 82 ­ 89. Bagirova, N.A., Shekhovtsova, T.N., Shopova, E.A., and van Huystee, R.B. 1998. Enzymatic determination of a- and b-naphthols using peanut peroxidase. Mend. Commun., 4): 157 ­ 159. ~ Barber, K.R., Rodriguez-Maranon, M.J., Shaw, G.S., and van Huystee, R.B. 1995. Structural influence of calcium on the heme cavity of cationic peanut peroxidase as determined by 1H-NMR spectroscopy. Eur. J. Biochem., 232: 825 ­ 833. Barton, D. and Ollis, W.D. 1980. Comprehensive Organic Chemistry. The Synthesis and Reactions of Organic Compounds. Pergamon Press.: New York; Vol. 3. Dolmanova, I.F., Shekhovtsova, T.N., and Kutcheryaeva, V.V. 1987. Assay of enzyme effectors. Talanta, 34: 201 ­ 205. Dolmanova, I.F., Ershova, E.V., Shekhovtsova, T.N., and Nad', V.Yu. 1979. Kinetic determination of mercury microamounts using peroxidase. Zhurn. Analit. Khim. (in Russian), 34: 1644 ­ 1657.


540

N. A. Bagirova et al.

Gajhede, M., Schuller, D.J., Henriksen, A., Smith, A.T., and Poulos, T.L. 1997. ° Crystal structure of horseradish peroxidase C at 2.15 A resolution. Nature Struct. Biol., 4 (12): 1032 ­ 1038. Gazaryan, I.G., Ouporov, I.V., Chubar, T.A., Fechina, V.A., Mareeva, E.A., and Lagrimini, L.M. 1998. Effect of pH on tobacco anionic peroxidase stability and its interaction. Biochemisty-US., 63: 600 ­ 606. Gazaryan, I.G., Doseeva, V.V., Galkin, A.G., and Tishkov, V.I. 1994. Effect of single-point mutations Phe41 ! His and Phe143 ! Glu on folding and catalytic properties of recombinant horseradish peroxidase expressed in E.coli FEBS Lett., 354: 248 ­ 250. Henriksen, A., Welinder, K.G., and Gajhede, M. 1998. Structure of barley grain peroxi° dase refined at 1.9 A resolution. J. Biol. Chem., 273: 2241 ­ 2248. Kutuzova, G.D. and Ugarova, N.N. 1981. Stabilization of horseradish peroxidase in acetylatin of enzyme in the presence of calcium ions. Bioorg. Khim., 7 (1): 75 ­ 85. Lambeir, A.M., Dunford, H.B., van Huystee, R.B., and Lobarzewski, J. 1985. Spectral and kinetic properties of a cationic peroxidase secreted by cultured peanut cells. Can. J. Biochem. Cell. Biol., 63: 1086 ­ 1092. Lebedeva, O.V., Ugarova, N.N., and Berezin, I.V. 1977. Kinetic study of oxidation reaction of o-dianisidine with hydrogen peroxide in the presence of horseradish peroxidase. Biokhimiya (in Russian), 42 (8): 1372 ­ 1379. Lurie, Yu.Yu. 1989. Reference Book on Analytical Chemistry; Khimiya: Moscow. Magonet, E., Hayen, P., Delforge, D., Delive, E., and Remacle, J. 1992. Importance of the structural zinc atom for the stability of yeast alcohol dehydrogenase. Biochem. J., 287 (2): 361 ­ 367. Martell, A.E. and Sillen, L.G. 1978. Stability Constants of Metal ­ ions Complexes. The Chemical Society Burlington House.: London. ¨ Munteanu, F.D., Lindgren, A., Emneus, J., Gorton, L., Ruzgas, T., Csoregi, E., Ciucu, A., van Huystee, R.B., Gazaryan, I.G., and Lagrimini, L.M. 1998. Bioelectrochemical monitoring of phenols and aromatic amines in flow injection using novel plant peroxidases. Anal. Chem., 70: 2596 ­ 2600. Nagano, S., Tanaka, M., Ishimori, K., Watanabe, Y., and Morishima, I. 1996. Catalytic roles of the distal site asparagin--histidin couple in peroxidases. Biochemistry, 35: 14251 ­ 14258. Peshkova, V.M. and Gromova, M.I. 1965. Laboratory Works on Spectrophotometry. (in Russian). Moscow: Izd. Mosk. Un-ta. Phelps, Ch., Forlani, L., and Antonini, E. 1971. The hydrogen ion equilibria of horseradish peroxidase and apoperoxidase. Biochem. J., 124 (3): 605 ­ 614. Popov, V.O. and Egorov, A.M. 1979. Investigation of essential SH-groups of bacterial formiate dehydrogenase. Biokhimiya (in Russian), 44 (2): 207 ­ 213. ~ Rodriguez-Maranon, M.J., Stillman, M.J., and van Huystee, R.B. 1993. Co-dependency of calcium and porphyrin for an integrated molecular structure of peanut peroxidase: A circular dichroism analysis. Biochem. Biophys. Res. Commun., 194 (1): 326 ­ 332. Schuller, D.J., Ban, N., van Huystee, R.B., McPherson, A., and Poulos, T.L. 1996. The crystal structure of peanut peroxidase. Structure, 4: 311 ­ 321. Schumacher, M., Camp, S., Maulet, Y., Newton, M., McPhee ­ Quigley, K., and Taylor, P. 1986. Primary structure of Torpedo Californica acetylcholinesterase deduced from cDNA sequences. Nature, 319: 407 ­ 409. Sesto, P.A. and van Huystee, R.B. 1989. Purification and yield of a cationic peanut peroxidase from peanut suspension celll culture. Plant Sci., 61: 163 ­ 165.


Effect of Trace Amounts of Mercury(II)

541

Shannon, L.M., Kay, E., and Lew, J.Y. 1996. Peroxidase isozymes from horseradish roots. Isolation and physical properties. J. Biol. Chem., 241: 2166 ­ 2172. Shekhovtsova, T.N., Muginova (Chernetskaya), S.V., Mizgunova, U.M., and Dolmanova, I.F. 1996. Applications of oxidases in analysis. Quim. Anal., 15: 312 ­ 320. Shekhovtsova, T.N., Chernetskaya, S.V., and Dolmanova, I.F. 1993. Enzymatic determination of trace iron (III) and some peroxidase inhibitors. J. Anal. Chem., 48 (1): 129 ­ 136. Sigel, H. 1986. Metal Ions in Biological Systems. Concepts on Metal Ion Toxicity. Sigel, H. and Sigel, A. (eds.); Moscow, Mir, 1993, 368. (translated from English, based on "Metal Ions in Biological Systems;" Vol. 20, Sigel, H. and Sigel, A. (eds.); Marcel Dekker, New York, 1986). Marcel Dekker Inc.: New York, 20. Smith, A.T., Santama, N., Dacey, S., Edwards, M., Bray, C., Thorneley, R.N.F., and Burke, J.F. 1991. Expression of a synthetic gene for horseradish peroxidase C in E. coli and folding and activation of the recombinant enzyme with Ca2× and heme. J. Biol. Chem., 265 (22): 13335 ­ 13343. Tanaka, M., Ishimori, K., and Morishima, I. 1998. Structural roles of the highly conserved Glu residue in the heme distal site of peroxidase. Biochemistry, 37: 2629 ­ 2630. Torchinskyi, Yu.M. 1977. Sulfur in Proteins; Nauka: Moscow. Ugarova, N.N., Rozhkova, G.D., and Berezin, I.V. 1979. Chemical modificationof the a0 -amino-groups of lysine residues in horseradish peroxidase and its effects on the catalytic properties and thermostability of the enzyme. Biochim. Biophys. Acta, 570: 31 ­ 42. Ugarova, N.N., Popova, I.M., Dolmanova, I.F., and Shekhovtsova, T.N. 1981. Inhibition and inactivation of horseradish peroxidase in the presence of thiourea. Biokhimia (in Russian), 46 (6): 1026 ­ 1034. van Huystee, R.B., Xu, Y., and o'Donnell, J. 1992. Variation in soret band absorption of peroxidase due to calcium. Plant Physiol. Biochem., 30: 293 ­ 297. Varfolomeev, S.D. and Gurevich, K.G. 1999. Biokinetika. Fair-Press.: Moscow. Veselova, I.A. and Shekhovtsova, T.N. 2000. Visual determination of lead(II) by inhibition using alkaline phosphatase immobilized on polyurethane foam. Anal. Chim. Acta, 413: 95 ­ 101. Wan, L. and van Huystee, R.B. 1993. A study on glycosilation of cationic peanut peroxidase. Biochim. Biophys. Res. Commun., 194: 1398 ­ 1405. Welinder, K.G. 1992. Superfamily of plant, fungal and bacterial peroxidases. Curr. Opin. Struct. Biol., 25: 388 ­ 393. Welinder, K.G. 1979. Aminoacid sequence studies of horseradish peroxidase. Eur. J. Biochem., 96: 483 ­ 502. Zhou, C.Y., Wong, M.K., Koh, L.L., and Wee, Y.C. 1996. Comparison of acid mixtures in high-pressure microwave digestion methods for the determination of the total mercury in sediments by CVAAS. Analytical Science, 12: 471 ­ 475.