- Email: [email protected]
An initial-rate potentiometric method for the determination of uric acid using a fluoride ion-selective electrode
Talanta 61 (2003) 493 /499 www.elsevier.com/locate/talanta
An initial-rate potentiometric method for the determination of uric acid using a fluoride ion-selective electrode Farzad Deyhimi *, Rahman Salamat-Ahangari Department of Chemistry, Shahid Beheshti University, Evin-Tehran 19839, Iran Received 24 October 2002; received in revised form 15 April 2003; accepted 9 May 2003
Abstract The principle of a general potentiometric method based on Emerson /Trinder reaction for the assay of various hydrogen peroxide generating systems is reported. Emerson /Trinder reaction, habitually employed as a spectrophotometric indicator reaction, is exploited in this method as a potentiometric indicator reaction. This method is based on the detection of F ions, liberated from the oxidation of a fluorophenol compound used as hydrogen-donor in Emerson /Trinder reaction, by a fluoride ion-selective electrode. The ability and usefulness of this method are illustrated by an initial-rate potentiometric determination of uric acid in aqueous and human serum samples, for which, initial-rate reaction progress curves, linear calibration curve, within-day precisions, upper and lower detection limits, and also its analytical recovery were reported. # 2003 Elsevier B.V. All rights reserved. Keywords: Uric acid; Fluoride ion-selective electrode; Potentiometric method; Initial-rate
1. Introduction In recent years, a variety of a new electrochemical, optical sensors and biosensors have been developed for the assay of different substrates and enzymes. An important example of such development concern uric acid, that because of its role in the metabolism of the purines and related physiological disorders (gout, hyperuricemia, or LeschNyhan syndrome), is one of the most important
* Corresponding author. Tel.: /98-21-240-1765; fax: /9821-240-3041. E-mail address: [email protected] (F. Deyhimi).
parameters monitored in urine and blood samples in clinical chemistry e.g. [1 /11]. Although, ion-selective electrodes were frequently utilized (indirectly) in clinical chemistry as enzyme electrodes for the assay of different substrates and enzymes [12,13], nevertheless, they were, rather rarely, used directly for such purposes. Illustrative examples of such applications are those concerning pH, ammonium, iodide and fluoride ion-selective electrodes [14 /18]. Specifically, fluoride ion-selective electrode was used for the first time by Siddiqi [17,18] for the assay of horseradish peroxidase enzyme (HRP), glucose and cholesterol. This potentiometric method was based on the reaction reported by Hughes and Sanders [19] concerning HRP-catalyzed oxidation
0039-9140/03/$ - see front matter # 2003 Elsevier B.V. All rights reserved. doi:10.1016/S0039-9140(03)00309-6
494
F. Deyhimi, R. Salamat-Ahangari / Talanta 61 (2003) 493 /499
of 4-fluoroaniline (used as hydrogen-donor) by hydrogen peroxide. In this reaction the resulted catalytic cleavage of the covalent carbon /fluorine bond of 4-fluoroaniline generate the release of fluoride ions in the above reaction. The enzymelinked immunoassay of IgG (immunoglobulin G) in human serum, with the fluoride ion-selective electrode and HRP labeled anti-IgG, was then reported by Alexander and Maltra [20]. Siddiqi’s approach was further extended to the determination of other substrates of clinical diagnostic interest (e.g. hemoglobin, amino-acid. . .) [21,22]. Another reaction, routinely exploited as a spectrophotometric indicator reaction in clinical chemistry, is the so called Emerson /Trinder reaction. This reaction was first reported by Emerson in 1943 as a new color test reaction for the determination of phenolic compounds. In this reaction, a quinoneimine dye product is produced by oxidative condensation of a phenol with 4aminoantipyrine (4-AAP) [23]. This spectrophotometric method is still in common use for the determination of phenolic compounds (e.g. [24,25]). Emerson reaction was later employed by Trinder for the determination of blood glucose, where, hydrogen peroxide produced in the glucose/ glucose oxidase reaction, acted as an oxidizing agent in the presence of HRP enzyme in Emerson reaction [26,27]. For this reason this reaction is also known as Trinder reaction. This indicator reaction was afterward used for the spectrophotometric assay of a large number of substrates or enzymes such as uric acid [28], cholesterol [29], free hemoglobin [30] or triglycerides [31] and also by using different organic hydrogen-donor compounds such as different substituted (ortho, meta and para) chloro or bromophenols, 4-hydroxybenzene-sulfonic acid [32], 2,4-dichlorophenol [33], 3,5-dichloro-2-hydroxybenzensulfonic acid [31,34] or different aniline derivatives [35]. However, its exploitation as a potentiometric indicator reaction was not yet reported. Accordingly, this work illustrates the ability of this reaction to be used also as a potentiometric indicator reaction for the assay of various hydrogen peroxide generating systems. In the resulting potentiometric method, the fluoride ion-selective electrode is used to monitor the rate of fluoride ion production from
the HRP-catalyzed oxidative condensation of an organo-fluorine compound with 4-AAP by hydrogen peroxide. The application results of this potentiometric method and its analytical capability for the assay of uric acid, in aqueous and serum samples, are presented.
2. Experimental
2.1. Chemicals All experiments were performed using solutions prepared from analytical grade chemicals and doubly distilled water. The enzymes horseradish peroxidase (HRP, hydrogen peroxide oxidoreductase; E.C. 1.11.1.7), uricase (urate oxidase; E.C. 1.7.3.3) and ascorbate oxidase (L-ascorbate oxidase; EC 1.10.3.3) were purchased from Boehringer-Manheim (Germany). Triton-X100 (polyethylene-glycol tert -octylphenyl ether) and potassium hexacyanoferrate were from Fluka AG (Switzerland) and all other compounds were obtained from Merck (Germany). A stock solution of 1.71 U mg 1 uricase enzyme was obtained by dissolving 4 mg of this enzyme in 1 ml of 150 mM tris(hydroxymethyl-aminomethane) buffer (trisbuffer), at pH 8. 250 mM stock solution of 4fluorophenol (4-FP) and 200 mM solution of 4AAP were both prepared also in tris-buffer (at pH 8). Uric acid standard solution (3.57/104 M) and its spectrophotometric determination kit were obtained from Pars Azmoun (Iran). This spectrophotmetric (Emerson/Trinder based reagent) commercial kit was first used for the determination of uric acid in the laboratory prepared aqueous sample solutions and then in the serum and pool serum samples. The laboratory made aqueous sample solutions of uric acid were prepared by appropriate dilution of an initial stock solution obtained by dissolving (at 60 8C) a mixture of 0.1 g uric acid and 0.08 g lithium carbonate in 100 ml distilled water [36]. Stock solution of NaF (100 mM) in distilled water was also prepared directly by appropriate weighing and dilution.
F. Deyhimi, R. Salamat-Ahangari / Talanta 61 (2003) 493 /499
2.2. Instrumentation The experimental cell potentials were recorded using a high input impedance (/1 GV) Topward multimeter (model 1304, Taiwan, Korea) equipped with a GPIB interface Bus option (1304G) which was connected to a personal computer (Samsung, 386/32 MHz processor) via the GPIB interface card (IEEE488, Keithley, USA) for data acquisition and processing. A laboratory written Basic program combined with Microsoft Excel (XPOffice 2002) software were used for data acquisition and initial-rate calculation. The fluoride ionselective and reference Ag/AgCl electrodes used in this work were from Fluka (Switzerland). The pH of all prepared solutions were adjusted using a combined pH electrode and a pH-meter (model 691), both from Metrohm (Switzerland). All measurements were performed under stirring conditions at room temperature (259/2 8C), except when otherwise indicated.
3. Principle of method 3.1. Principle As mentioned above, in Emerson /Trinder reaction, a red (purple) quinoneimine dye is produced by oxidative condensation of a phenolic compound with 4-AAP in the presence of hydrogen peroxide. The reaction is catalyzed by HRP enzyme (see reaction (2)). The experimental conditions for the determination of uric acid, based on Emerson /Trinder reaction, are available for different hydrogen-donor organic compounds [28,33/35] and are in current use as indicated in commercial kits. Therefore, starting with the available experimental compositions used in the Boehringer-Manheim commercial kit [37], and by substituting just the hydrogen-donor compound by p-fluorophenol, the indicator reaction was first optimized, using both initial-rate and endpoint measurements, by one at a time parameter method. The collection of the experimental absorption data was performed at the maximum of the absorption band of the resulted quinoneimine complex (lmax /505 nm). Subsequently, using a
495
fluoride ion-selective electrode, this reaction was first exploited for the potentiometric determination of H2O2 [38], and then coupled to the uric acid/uricase/allantoin reaction. The determination of uric acid was performed by adjusting the available experimental conditions used in uric acid Boehringer-Manheim commercial kit with those obtained for the indicator reaction. The pH of this mixture was further varied in order to obtain the best performances by monitoring the production of F ions, using potentiometric technique and a F ion-selective electrode. Nevertheless, further work based on chemometrics methods is needed to optimize suitably the experimental conditions utilized in this work and those in current use which were reported in the related published works. However, the detail concerning the experimental facts that Emerson /Trinder reaction can be used directly both in spectrophotometric or potentiometric methods for the determination of phenol or hydrogen peroxide was obtained. Accordingly, this work illustrates the application of the resulting procedure for the potentiometric assay of uric acid in aqueous and serum samples, by using fluoride ion-selective electrode and initial-rate potentiometric method in the following coupled reactions. In the above coupled reactions, the quantity of uric acid is determined via the amount of hydrogen peroxide, first produced in reaction (1) and subsequently consumed proportionally in reaction (2), generating an equal amount of F ions which is detected by a fluoride ion-selective electrode. Using a galvanic cell, containing both fluorideselective and reference electrodes, the production of fluoride can be monitored during the course of reaction by the decrease of cell potential according to the Nernst equation E E?
2:303RT F
log[F ]
(3)
where E is the experimental cell potential, E ? is the cell constant potential, F and R are Faraday and gas constants, respectively, T is Kelvin temperature, S /2.303RT /F is the slope of the fluorideselective electrode and [F ] is the concentration of
496
F. Deyhimi, R. Salamat-Ahangari / Talanta 61 (2003) 493 /499
(1)
(2)
fluoride ion. Clearly, as it turns out well in highly diluted solution, concentration is used instead of activity for the calculation of fluoride ion concentrations in the above equation.
1.43 /106, 2.86 /106, 4.28 /106, 5.71 / 106, 7.14 /106 M, prepared by appropriate dilution of the commercial uric acid standard solution (Pars Azmoun, Iran) and also a labora-
3.2. Procedure
Table 1 Final optimized parameters and the other analytical characteristics of the reported initial-rate potentiometric method for the determination of uric acid.
In these experiments, in order to fulfill the initial-rate linear kinetic conditions, due to the slowness of the first reaction and its subsequent low generation of hydrogen peroxide molecules in the first step, reaction (2) as left sufficiently to progress and then in the second step, reaction (2) was subsequently started after 10 min by addition of HRP enzyme. The used final concentration of reagents, as explained in the above section, was reported in Table 1. The potentiometric initial-rate was each time calculated from the potential data recorded from the start of reaction by addition of HRP (with a sampling time of 1 s and during about 60 s) and also using Nernst equation.
4. Results and discussion 4.1. Aqueous samples Fig. 1 shows typical reaction progress curves (produced F ions versus time) obtained using standard solutions with final concentrations of
pH value (tris-buffer) Tris-buffer concentration Uricase 4-FP 4-AAP Horseradish peroxidase Ascorbate oxidase Potassium hexacyanoferrate Triton X-100 Linear range (M)
8 150 mM 81 U l 1 30 mM 3 mM 110 U l 1 4.1 kU l 1 0.05 mM 2 g l 1 3.34 /10 8 /4.85/10 5 y /2474.4x/0.0024 R/0.9966
Lower detection limit (M) Upper detection limit (M) Within-day precisions (%RSD) (a) Aqueous samples, N/10 3.28/10 7 M (Low) 6.55/10 6 M (High) (b) Pool serum, N/9 7.54/10 6 M
1 /10 8 4.85 /10 5
Pool serum recovery Added Recovered
1.48% 2.57% 2.37% 20, 40, 60, 80, 100% 101.7, 100.7, 96.8, 101.3, 98.8%
F. Deyhimi, R. Salamat-Ahangari / Talanta 61 (2003) 493 /499
497
sented in Fig. 2. Within-day precisions of this method were also determined for 10 replica by initial-rate potentiometric method for aqueous uric acid standard solutions at low (3.28 /107 M) and high (6.55 /106 M) levels, and presented as the relative standard deviation (%RSD) in Table 1. By graphical method, using the intersection of the linear portions of the uric acid calibration curve, the upper and lower detection limits of this method were also evaluated and reported in Table 1.
4.2. Application to real sample assays
Fig. 1. Typical reaction progress curves (produced F versus time) for the determination of uric acid (T /25 8C). The final concentrations of uric acid solutions were: 1.43 /10 6, 2.86/ 10 6, 4.28/10 6, 5.71/10 6, 7.14/10 6 M (standard solutions), and the concentration of the laboratory prepared aqueous sample solution was: 3.93/10 6 M.
tory prepared aqueous sample solution (3.93 / 106 M). The corresponding linear calibration curve (initial-rate versus concentration) is pre-
In order to establish the usefulness of this method, the determination of uric acid was performed in human serum and pool serum samples obtained from hospital (Taleghani Medical University Hospital, Tehran). These samples were stored at /20 8C until required for analysis. Fig. 3 illustrates the satisfactory correlation (y / 0.9901x/0.7727, r/0.9971) between the results obtained for the determination of uric acid in individual serum samples, determined by the presented method versus those obtained by using the commercial spectrophotmetric kit at 30 8C. Analytical recovery and precision of this initialrate potentiometric method were also determined using pool serum samples prepared by mixing
Fig. 2. Calibration curve (initial-rate of F production versus concentration) obtained from the data presented in Fig. 1 (T /25 8C).
498
F. Deyhimi, R. Salamat-Ahangari / Talanta 61 (2003) 493 /499
individual serum samples. As shown in Table 1, the amount of uric acid recovered in pool serum after addition of 20, 40, 60, 80 and 100% uric acid were: 101.72, 100.74, 96.85, 101.33 and 98.84%, respectively. On the other hand, the precision obtained with pool serum was as found to be: Mean /10.959/0.35 mM, RSD% /3.18, N /9, as summarized in Table 1. The effect of interfering substances in the serum samples for determination of uric acid in the uricase/phenolic compound/4-AAP/peroxidase systems was investigated by many authors [28,31,33,34] and was minimized by the presence of several additive substances. The effect of interference of bilirubin was reduced by the addition of potassium hexacyanoferrate (no interference up to 0.3 mM); ascorbate oxidase was used to eliminate the interference of ascorbate in the therapeutic range, and the incorporation of non-ionic detergent (Triton X-100) permitted the determination of uric acid in lipemic samples (up to 19 mM).
5. Conclusion This work presented the principle of a general potentiometric method based on Emerson /Trinder reaction and the use of a fluoride ion-selective electrode for the assay of different hydrogen peroxide generating systems. The ability of this potentiometric method was illustrated by the assay of uric acid, in view of its importance as mentioned above. The determination of uric acid was performed by coupling Emerson /Trinder reaction, used as a potentiometric indicator reaction, to the reaction of uricase catalyzed the transformation of uric acid to allantoin. The amount of uric acid was determined via the initial-rate potentiometric detection of fluoride ions liberated by the oxidative condensation of 4-FP (used as hydrogen-donor) with 4-AAP in Emerson /Trinder reaction. Using uric acid aqueous samples, the corresponding reaction progress curves, the resulting linear calibration curve, within-day precisions at low and at high levels, and the upper and lower detection limits were determined for this method. The values of uric acid in individual human serum samples obtained with the commercial kit (spectrophotometric method) were also compared with those obtained with the F ion-selective electrode (potentiometric method). The precision and analytical recoveries of the presented method were as well determined in human pool serum samples. Briefly, as summarized in Table 1, the reported method offers, at once, the satisfactory analytical advantages of the spectrophotometric method (precision, recovery, selectivity) and those associated with the electrochemical method (rapidity and insensitivity to color or turbid samples).
References
Fig. 3. Plot of the results obtained by the reported initial-rate potentiometric method versus those obtained by using the spectrophotmetric method, for the determination of uric acid in individual serum samples (T /30 8C). The normal range of uric acid [39] are delimited by the vertical dashed lines.
[1] J. Zen, P. Chen, Anal. Chem. 69 (1997) 5087. [2] T. Nakaminami, S. Ito, S. Kuwabata, H. Yoneyama, Anal. Chem. 71 (1999) 4278. [3] E. Popa, Y. Kubota, D.A. Tryk, A. Fujishima, Anal. Chem. 72 (2000) 1724. [4] J. Wang, M. Chatrathi, B. Tian, R. Polsky, Anal. Chem. 72 (2000) 2514. [5] A. Brajter-Toth, K. Abou El-Nour, E.T. Cavalheiro, R. Bravo, Anal. Chem. 72 (2000) 1576.
F. Deyhimi, R. Salamat-Ahangari / Talanta 61 (2003) 493 /499 [6] J. Galban, Y. Andreu, M.J. Almenara, S. de Marcos, J.R. Castillo, Talanta 54 (2001) 847. [7] J.M. Zen, Y.Y. Lai, H.H. Yang, A.S. Kumar, Sensors and Actuators B 84 (2002) 237. [8] S. Kasai, Y. Hirano, N. Motochi, H. Shiku, M. Nishizawa, T. Matsue, Analyt. Chim. Acta 458 (2002) 263. [9] M.M. Castillo-Ortega, D.E. Rodriguez, J.C. Encinas, M. Plascencia, F.A. Mendez-Velarde, R. Olayao, Sensors Actuators B 85 (2002) 19. [10] S. Kasai, Y. Hirano, N. Motochi, H. Shiku, M. Nishizawa, Analyt. Chim. Acta 458 (2002) 263. [11] T. Nakaminami, S. Ito, S. Kuwabata, H. Yoneyama, Anal. Chem. 71 (1999) 1928. [12] L.J. Blum, P.R. Coulet (Eds.), Biosensor Principles and Applications, M. Dekker Inc, New York, 1991. [13] B.R. Eggins, Biosensors: An Introduction, Wiley, New York, 1997. [14] J. Ruzicka, E.H. Hansen, A.K. Ghose, Anal. Chem. 51 (1979) 199 /203. [15] G.G. Guibault, R.K. Smith, J.G. Montalvo, Jr., Anal. Chem. 41 (1969) 600 /605. [16] R.A. Lienado, G.A. Rechnitz, Anal. Chem. 45 (1973) 826 / 833. [17] I.W. Siddiqi, J.A. Brochot, J. Clin. Chem. Clin. Biochem. 54 (1981) 83. [18] I.W. Siddiqi, Clin. Chem. 28 (1982) 1962 /1967. [19] G.M.K. Hughes, B.C. Sanders, J. Chem. Soc. (1954) 4630 /4634. [20] P.W. Alexander, C. Maltra, Anal. Chem. 54 (1982) 68 /71. [21] R.K. Kobos, S.D. Abbot, H.W. Levin, H. Kilkson, D.R. Peterson, J.W. Dickinson, Clin. Chem. 33 (1987) 153 /158.
499
[22] E. Athanasiou-Malaki, M.A. Koupparis, Analyst 112 (1987) 757 /761. [23] E. Emerson, J. Org. Chem. 8 (1943) 417 /428. [24] M.B. Ettinger, C.C. Ruchhoft, R.J. Lishka, Anal. Chem. 23 (1951) 1783 /1788. [25] Y. Fiamegos, C. Stalikas, G. Pilidis, Anal. Chim. Acta 467 (2002) 105 /114. [26] P. Trinder, Ann. Clin. Biochem. 6 (1969) 24 /37. [27] D. Barham, P. Trinder, Analyst 97 (1972) 142 /145. [28] P. Kabasakalian, S. Kalliney, A. Westcott, Clin. Chem. 19 (1973) 522 /524. [29] C.C. Allain, L.S. Poon, C.S.G. Chan, W. Richmond, P.C. Fu, Clin. Chem. 20 (1974) 470 /475. [30] K. Bauer, J. Clin. Chem. Clin. Biochem. 19 (1981) 971 / 976. [31] P. Fossati, L. Prencipe, Clin. Chem. 28 (1982) 2077 /2080. [32] F. Wang, Y. Wu, X. Wu, S. Shong, Y. Ci, Fresenius J. Anal. Chem. 344 (1992) 556 /558. [33] S. Klose, M. Stoltz, E. Munz, R. Portenhauser, Clin. Chem. 24 (1978) 250 /255. [34] P. Fossati, L. Prencipe, G. Berti, Clin. Chem. 26 (1980) 227 /231. [35] K. Tamaoku, Y. Murao, K. Akiura, Anal. Chim. Acta 136 (1982) 121 /127. [36] R. Richterich, J.P. Colombo (Eds.), Clinical Chemistry, Theory Practice and Interpretation, Wiley, New York, 1981, p. 454. [37] M.-H. Town, J. Clin. Chem. Clin. Biochem. 23 (1985) 591. [38] L. Parvin, MS Chemistry Thesis, Shahid Beheshti University, Tehran, Iran, 1999. [39] C.A. Burtis, E.R. Ashwood (Eds.), Tietz Textbook of Clinical Chemistry, second ed., W.B. Saunders Co, Philadelphia, 1994.
An initial-rate potentiometric method for the determination of uric acid using a fluoride ion-selective electrode Farzad Deyhimi *, Rahman Salamat-Ahangari Department of Chemistry, Shahid Beheshti University, Evin-Tehran 19839, Iran Received 24 October 2002; received in revised form 15 April 2003; accepted 9 May 2003
Abstract The principle of a general potentiometric method based on Emerson /Trinder reaction for the assay of various hydrogen peroxide generating systems is reported. Emerson /Trinder reaction, habitually employed as a spectrophotometric indicator reaction, is exploited in this method as a potentiometric indicator reaction. This method is based on the detection of F ions, liberated from the oxidation of a fluorophenol compound used as hydrogen-donor in Emerson /Trinder reaction, by a fluoride ion-selective electrode. The ability and usefulness of this method are illustrated by an initial-rate potentiometric determination of uric acid in aqueous and human serum samples, for which, initial-rate reaction progress curves, linear calibration curve, within-day precisions, upper and lower detection limits, and also its analytical recovery were reported. # 2003 Elsevier B.V. All rights reserved. Keywords: Uric acid; Fluoride ion-selective electrode; Potentiometric method; Initial-rate
1. Introduction In recent years, a variety of a new electrochemical, optical sensors and biosensors have been developed for the assay of different substrates and enzymes. An important example of such development concern uric acid, that because of its role in the metabolism of the purines and related physiological disorders (gout, hyperuricemia, or LeschNyhan syndrome), is one of the most important
* Corresponding author. Tel.: /98-21-240-1765; fax: /9821-240-3041. E-mail address: [email protected] (F. Deyhimi).
parameters monitored in urine and blood samples in clinical chemistry e.g. [1 /11]. Although, ion-selective electrodes were frequently utilized (indirectly) in clinical chemistry as enzyme electrodes for the assay of different substrates and enzymes [12,13], nevertheless, they were, rather rarely, used directly for such purposes. Illustrative examples of such applications are those concerning pH, ammonium, iodide and fluoride ion-selective electrodes [14 /18]. Specifically, fluoride ion-selective electrode was used for the first time by Siddiqi [17,18] for the assay of horseradish peroxidase enzyme (HRP), glucose and cholesterol. This potentiometric method was based on the reaction reported by Hughes and Sanders [19] concerning HRP-catalyzed oxidation
0039-9140/03/$ - see front matter # 2003 Elsevier B.V. All rights reserved. doi:10.1016/S0039-9140(03)00309-6
494
F. Deyhimi, R. Salamat-Ahangari / Talanta 61 (2003) 493 /499
of 4-fluoroaniline (used as hydrogen-donor) by hydrogen peroxide. In this reaction the resulted catalytic cleavage of the covalent carbon /fluorine bond of 4-fluoroaniline generate the release of fluoride ions in the above reaction. The enzymelinked immunoassay of IgG (immunoglobulin G) in human serum, with the fluoride ion-selective electrode and HRP labeled anti-IgG, was then reported by Alexander and Maltra [20]. Siddiqi’s approach was further extended to the determination of other substrates of clinical diagnostic interest (e.g. hemoglobin, amino-acid. . .) [21,22]. Another reaction, routinely exploited as a spectrophotometric indicator reaction in clinical chemistry, is the so called Emerson /Trinder reaction. This reaction was first reported by Emerson in 1943 as a new color test reaction for the determination of phenolic compounds. In this reaction, a quinoneimine dye product is produced by oxidative condensation of a phenol with 4aminoantipyrine (4-AAP) [23]. This spectrophotometric method is still in common use for the determination of phenolic compounds (e.g. [24,25]). Emerson reaction was later employed by Trinder for the determination of blood glucose, where, hydrogen peroxide produced in the glucose/ glucose oxidase reaction, acted as an oxidizing agent in the presence of HRP enzyme in Emerson reaction [26,27]. For this reason this reaction is also known as Trinder reaction. This indicator reaction was afterward used for the spectrophotometric assay of a large number of substrates or enzymes such as uric acid [28], cholesterol [29], free hemoglobin [30] or triglycerides [31] and also by using different organic hydrogen-donor compounds such as different substituted (ortho, meta and para) chloro or bromophenols, 4-hydroxybenzene-sulfonic acid [32], 2,4-dichlorophenol [33], 3,5-dichloro-2-hydroxybenzensulfonic acid [31,34] or different aniline derivatives [35]. However, its exploitation as a potentiometric indicator reaction was not yet reported. Accordingly, this work illustrates the ability of this reaction to be used also as a potentiometric indicator reaction for the assay of various hydrogen peroxide generating systems. In the resulting potentiometric method, the fluoride ion-selective electrode is used to monitor the rate of fluoride ion production from
the HRP-catalyzed oxidative condensation of an organo-fluorine compound with 4-AAP by hydrogen peroxide. The application results of this potentiometric method and its analytical capability for the assay of uric acid, in aqueous and serum samples, are presented.
2. Experimental
2.1. Chemicals All experiments were performed using solutions prepared from analytical grade chemicals and doubly distilled water. The enzymes horseradish peroxidase (HRP, hydrogen peroxide oxidoreductase; E.C. 1.11.1.7), uricase (urate oxidase; E.C. 1.7.3.3) and ascorbate oxidase (L-ascorbate oxidase; EC 1.10.3.3) were purchased from Boehringer-Manheim (Germany). Triton-X100 (polyethylene-glycol tert -octylphenyl ether) and potassium hexacyanoferrate were from Fluka AG (Switzerland) and all other compounds were obtained from Merck (Germany). A stock solution of 1.71 U mg 1 uricase enzyme was obtained by dissolving 4 mg of this enzyme in 1 ml of 150 mM tris(hydroxymethyl-aminomethane) buffer (trisbuffer), at pH 8. 250 mM stock solution of 4fluorophenol (4-FP) and 200 mM solution of 4AAP were both prepared also in tris-buffer (at pH 8). Uric acid standard solution (3.57/104 M) and its spectrophotometric determination kit were obtained from Pars Azmoun (Iran). This spectrophotmetric (Emerson/Trinder based reagent) commercial kit was first used for the determination of uric acid in the laboratory prepared aqueous sample solutions and then in the serum and pool serum samples. The laboratory made aqueous sample solutions of uric acid were prepared by appropriate dilution of an initial stock solution obtained by dissolving (at 60 8C) a mixture of 0.1 g uric acid and 0.08 g lithium carbonate in 100 ml distilled water [36]. Stock solution of NaF (100 mM) in distilled water was also prepared directly by appropriate weighing and dilution.
F. Deyhimi, R. Salamat-Ahangari / Talanta 61 (2003) 493 /499
2.2. Instrumentation The experimental cell potentials were recorded using a high input impedance (/1 GV) Topward multimeter (model 1304, Taiwan, Korea) equipped with a GPIB interface Bus option (1304G) which was connected to a personal computer (Samsung, 386/32 MHz processor) via the GPIB interface card (IEEE488, Keithley, USA) for data acquisition and processing. A laboratory written Basic program combined with Microsoft Excel (XPOffice 2002) software were used for data acquisition and initial-rate calculation. The fluoride ionselective and reference Ag/AgCl electrodes used in this work were from Fluka (Switzerland). The pH of all prepared solutions were adjusted using a combined pH electrode and a pH-meter (model 691), both from Metrohm (Switzerland). All measurements were performed under stirring conditions at room temperature (259/2 8C), except when otherwise indicated.
3. Principle of method 3.1. Principle As mentioned above, in Emerson /Trinder reaction, a red (purple) quinoneimine dye is produced by oxidative condensation of a phenolic compound with 4-AAP in the presence of hydrogen peroxide. The reaction is catalyzed by HRP enzyme (see reaction (2)). The experimental conditions for the determination of uric acid, based on Emerson /Trinder reaction, are available for different hydrogen-donor organic compounds [28,33/35] and are in current use as indicated in commercial kits. Therefore, starting with the available experimental compositions used in the Boehringer-Manheim commercial kit [37], and by substituting just the hydrogen-donor compound by p-fluorophenol, the indicator reaction was first optimized, using both initial-rate and endpoint measurements, by one at a time parameter method. The collection of the experimental absorption data was performed at the maximum of the absorption band of the resulted quinoneimine complex (lmax /505 nm). Subsequently, using a
495
fluoride ion-selective electrode, this reaction was first exploited for the potentiometric determination of H2O2 [38], and then coupled to the uric acid/uricase/allantoin reaction. The determination of uric acid was performed by adjusting the available experimental conditions used in uric acid Boehringer-Manheim commercial kit with those obtained for the indicator reaction. The pH of this mixture was further varied in order to obtain the best performances by monitoring the production of F ions, using potentiometric technique and a F ion-selective electrode. Nevertheless, further work based on chemometrics methods is needed to optimize suitably the experimental conditions utilized in this work and those in current use which were reported in the related published works. However, the detail concerning the experimental facts that Emerson /Trinder reaction can be used directly both in spectrophotometric or potentiometric methods for the determination of phenol or hydrogen peroxide was obtained. Accordingly, this work illustrates the application of the resulting procedure for the potentiometric assay of uric acid in aqueous and serum samples, by using fluoride ion-selective electrode and initial-rate potentiometric method in the following coupled reactions. In the above coupled reactions, the quantity of uric acid is determined via the amount of hydrogen peroxide, first produced in reaction (1) and subsequently consumed proportionally in reaction (2), generating an equal amount of F ions which is detected by a fluoride ion-selective electrode. Using a galvanic cell, containing both fluorideselective and reference electrodes, the production of fluoride can be monitored during the course of reaction by the decrease of cell potential according to the Nernst equation E E?
2:303RT F
log[F ]
(3)
where E is the experimental cell potential, E ? is the cell constant potential, F and R are Faraday and gas constants, respectively, T is Kelvin temperature, S /2.303RT /F is the slope of the fluorideselective electrode and [F ] is the concentration of
496
F. Deyhimi, R. Salamat-Ahangari / Talanta 61 (2003) 493 /499
(1)
(2)
fluoride ion. Clearly, as it turns out well in highly diluted solution, concentration is used instead of activity for the calculation of fluoride ion concentrations in the above equation.
1.43 /106, 2.86 /106, 4.28 /106, 5.71 / 106, 7.14 /106 M, prepared by appropriate dilution of the commercial uric acid standard solution (Pars Azmoun, Iran) and also a labora-
3.2. Procedure
Table 1 Final optimized parameters and the other analytical characteristics of the reported initial-rate potentiometric method for the determination of uric acid.
In these experiments, in order to fulfill the initial-rate linear kinetic conditions, due to the slowness of the first reaction and its subsequent low generation of hydrogen peroxide molecules in the first step, reaction (2) as left sufficiently to progress and then in the second step, reaction (2) was subsequently started after 10 min by addition of HRP enzyme. The used final concentration of reagents, as explained in the above section, was reported in Table 1. The potentiometric initial-rate was each time calculated from the potential data recorded from the start of reaction by addition of HRP (with a sampling time of 1 s and during about 60 s) and also using Nernst equation.
4. Results and discussion 4.1. Aqueous samples Fig. 1 shows typical reaction progress curves (produced F ions versus time) obtained using standard solutions with final concentrations of
pH value (tris-buffer) Tris-buffer concentration Uricase 4-FP 4-AAP Horseradish peroxidase Ascorbate oxidase Potassium hexacyanoferrate Triton X-100 Linear range (M)
8 150 mM 81 U l 1 30 mM 3 mM 110 U l 1 4.1 kU l 1 0.05 mM 2 g l 1 3.34 /10 8 /4.85/10 5 y /2474.4x/0.0024 R/0.9966
Lower detection limit (M) Upper detection limit (M) Within-day precisions (%RSD) (a) Aqueous samples, N/10 3.28/10 7 M (Low) 6.55/10 6 M (High) (b) Pool serum, N/9 7.54/10 6 M
1 /10 8 4.85 /10 5
Pool serum recovery Added Recovered
1.48% 2.57% 2.37% 20, 40, 60, 80, 100% 101.7, 100.7, 96.8, 101.3, 98.8%
F. Deyhimi, R. Salamat-Ahangari / Talanta 61 (2003) 493 /499
497
sented in Fig. 2. Within-day precisions of this method were also determined for 10 replica by initial-rate potentiometric method for aqueous uric acid standard solutions at low (3.28 /107 M) and high (6.55 /106 M) levels, and presented as the relative standard deviation (%RSD) in Table 1. By graphical method, using the intersection of the linear portions of the uric acid calibration curve, the upper and lower detection limits of this method were also evaluated and reported in Table 1.
4.2. Application to real sample assays
Fig. 1. Typical reaction progress curves (produced F versus time) for the determination of uric acid (T /25 8C). The final concentrations of uric acid solutions were: 1.43 /10 6, 2.86/ 10 6, 4.28/10 6, 5.71/10 6, 7.14/10 6 M (standard solutions), and the concentration of the laboratory prepared aqueous sample solution was: 3.93/10 6 M.
tory prepared aqueous sample solution (3.93 / 106 M). The corresponding linear calibration curve (initial-rate versus concentration) is pre-
In order to establish the usefulness of this method, the determination of uric acid was performed in human serum and pool serum samples obtained from hospital (Taleghani Medical University Hospital, Tehran). These samples were stored at /20 8C until required for analysis. Fig. 3 illustrates the satisfactory correlation (y / 0.9901x/0.7727, r/0.9971) between the results obtained for the determination of uric acid in individual serum samples, determined by the presented method versus those obtained by using the commercial spectrophotmetric kit at 30 8C. Analytical recovery and precision of this initialrate potentiometric method were also determined using pool serum samples prepared by mixing
Fig. 2. Calibration curve (initial-rate of F production versus concentration) obtained from the data presented in Fig. 1 (T /25 8C).
498
F. Deyhimi, R. Salamat-Ahangari / Talanta 61 (2003) 493 /499
individual serum samples. As shown in Table 1, the amount of uric acid recovered in pool serum after addition of 20, 40, 60, 80 and 100% uric acid were: 101.72, 100.74, 96.85, 101.33 and 98.84%, respectively. On the other hand, the precision obtained with pool serum was as found to be: Mean /10.959/0.35 mM, RSD% /3.18, N /9, as summarized in Table 1. The effect of interfering substances in the serum samples for determination of uric acid in the uricase/phenolic compound/4-AAP/peroxidase systems was investigated by many authors [28,31,33,34] and was minimized by the presence of several additive substances. The effect of interference of bilirubin was reduced by the addition of potassium hexacyanoferrate (no interference up to 0.3 mM); ascorbate oxidase was used to eliminate the interference of ascorbate in the therapeutic range, and the incorporation of non-ionic detergent (Triton X-100) permitted the determination of uric acid in lipemic samples (up to 19 mM).
5. Conclusion This work presented the principle of a general potentiometric method based on Emerson /Trinder reaction and the use of a fluoride ion-selective electrode for the assay of different hydrogen peroxide generating systems. The ability of this potentiometric method was illustrated by the assay of uric acid, in view of its importance as mentioned above. The determination of uric acid was performed by coupling Emerson /Trinder reaction, used as a potentiometric indicator reaction, to the reaction of uricase catalyzed the transformation of uric acid to allantoin. The amount of uric acid was determined via the initial-rate potentiometric detection of fluoride ions liberated by the oxidative condensation of 4-FP (used as hydrogen-donor) with 4-AAP in Emerson /Trinder reaction. Using uric acid aqueous samples, the corresponding reaction progress curves, the resulting linear calibration curve, within-day precisions at low and at high levels, and the upper and lower detection limits were determined for this method. The values of uric acid in individual human serum samples obtained with the commercial kit (spectrophotometric method) were also compared with those obtained with the F ion-selective electrode (potentiometric method). The precision and analytical recoveries of the presented method were as well determined in human pool serum samples. Briefly, as summarized in Table 1, the reported method offers, at once, the satisfactory analytical advantages of the spectrophotometric method (precision, recovery, selectivity) and those associated with the electrochemical method (rapidity and insensitivity to color or turbid samples).
References
Fig. 3. Plot of the results obtained by the reported initial-rate potentiometric method versus those obtained by using the spectrophotmetric method, for the determination of uric acid in individual serum samples (T /30 8C). The normal range of uric acid [39] are delimited by the vertical dashed lines.
[1] J. Zen, P. Chen, Anal. Chem. 69 (1997) 5087. [2] T. Nakaminami, S. Ito, S. Kuwabata, H. Yoneyama, Anal. Chem. 71 (1999) 4278. [3] E. Popa, Y. Kubota, D.A. Tryk, A. Fujishima, Anal. Chem. 72 (2000) 1724. [4] J. Wang, M. Chatrathi, B. Tian, R. Polsky, Anal. Chem. 72 (2000) 2514. [5] A. Brajter-Toth, K. Abou El-Nour, E.T. Cavalheiro, R. Bravo, Anal. Chem. 72 (2000) 1576.
F. Deyhimi, R. Salamat-Ahangari / Talanta 61 (2003) 493 /499 [6] J. Galban, Y. Andreu, M.J. Almenara, S. de Marcos, J.R. Castillo, Talanta 54 (2001) 847. [7] J.M. Zen, Y.Y. Lai, H.H. Yang, A.S. Kumar, Sensors and Actuators B 84 (2002) 237. [8] S. Kasai, Y. Hirano, N. Motochi, H. Shiku, M. Nishizawa, T. Matsue, Analyt. Chim. Acta 458 (2002) 263. [9] M.M. Castillo-Ortega, D.E. Rodriguez, J.C. Encinas, M. Plascencia, F.A. Mendez-Velarde, R. Olayao, Sensors Actuators B 85 (2002) 19. [10] S. Kasai, Y. Hirano, N. Motochi, H. Shiku, M. Nishizawa, Analyt. Chim. Acta 458 (2002) 263. [11] T. Nakaminami, S. Ito, S. Kuwabata, H. Yoneyama, Anal. Chem. 71 (1999) 1928. [12] L.J. Blum, P.R. Coulet (Eds.), Biosensor Principles and Applications, M. Dekker Inc, New York, 1991. [13] B.R. Eggins, Biosensors: An Introduction, Wiley, New York, 1997. [14] J. Ruzicka, E.H. Hansen, A.K. Ghose, Anal. Chem. 51 (1979) 199 /203. [15] G.G. Guibault, R.K. Smith, J.G. Montalvo, Jr., Anal. Chem. 41 (1969) 600 /605. [16] R.A. Lienado, G.A. Rechnitz, Anal. Chem. 45 (1973) 826 / 833. [17] I.W. Siddiqi, J.A. Brochot, J. Clin. Chem. Clin. Biochem. 54 (1981) 83. [18] I.W. Siddiqi, Clin. Chem. 28 (1982) 1962 /1967. [19] G.M.K. Hughes, B.C. Sanders, J. Chem. Soc. (1954) 4630 /4634. [20] P.W. Alexander, C. Maltra, Anal. Chem. 54 (1982) 68 /71. [21] R.K. Kobos, S.D. Abbot, H.W. Levin, H. Kilkson, D.R. Peterson, J.W. Dickinson, Clin. Chem. 33 (1987) 153 /158.
499
[22] E. Athanasiou-Malaki, M.A. Koupparis, Analyst 112 (1987) 757 /761. [23] E. Emerson, J. Org. Chem. 8 (1943) 417 /428. [24] M.B. Ettinger, C.C. Ruchhoft, R.J. Lishka, Anal. Chem. 23 (1951) 1783 /1788. [25] Y. Fiamegos, C. Stalikas, G. Pilidis, Anal. Chim. Acta 467 (2002) 105 /114. [26] P. Trinder, Ann. Clin. Biochem. 6 (1969) 24 /37. [27] D. Barham, P. Trinder, Analyst 97 (1972) 142 /145. [28] P. Kabasakalian, S. Kalliney, A. Westcott, Clin. Chem. 19 (1973) 522 /524. [29] C.C. Allain, L.S. Poon, C.S.G. Chan, W. Richmond, P.C. Fu, Clin. Chem. 20 (1974) 470 /475. [30] K. Bauer, J. Clin. Chem. Clin. Biochem. 19 (1981) 971 / 976. [31] P. Fossati, L. Prencipe, Clin. Chem. 28 (1982) 2077 /2080. [32] F. Wang, Y. Wu, X. Wu, S. Shong, Y. Ci, Fresenius J. Anal. Chem. 344 (1992) 556 /558. [33] S. Klose, M. Stoltz, E. Munz, R. Portenhauser, Clin. Chem. 24 (1978) 250 /255. [34] P. Fossati, L. Prencipe, G. Berti, Clin. Chem. 26 (1980) 227 /231. [35] K. Tamaoku, Y. Murao, K. Akiura, Anal. Chim. Acta 136 (1982) 121 /127. [36] R. Richterich, J.P. Colombo (Eds.), Clinical Chemistry, Theory Practice and Interpretation, Wiley, New York, 1981, p. 454. [37] M.-H. Town, J. Clin. Chem. Clin. Biochem. 23 (1985) 591. [38] L. Parvin, MS Chemistry Thesis, Shahid Beheshti University, Tehran, Iran, 1999. [39] C.A. Burtis, E.R. Ashwood (Eds.), Tietz Textbook of Clinical Chemistry, second ed., W.B. Saunders Co, Philadelphia, 1994.