Anal. Chem. 2000, 72, 2211-2215
Amperometric Detection of Histamine with a Methylamine Dehydrogenase Polypyrrole-Based Sensor Kui Zeng,† Hiroyasu Tachikawa,*,† Zhenyu Zhu,‡ and Victor L. Davidson*,‡
Department of Chemistry, Jackson State University, P.O. Box 17910, 1400 Lynch Street, Jackson, Mississippi 39217, and Department of Biochemistry, The University of Mississippi Medical Center, Jackson, Mississippi 39216-4505
Methylamine dehydrogenase (MADH) has been immobilized in a polypyrrole (PPy) film on an electrode surface and used as an amine sensor for the determination of primary amines. Its response to histamine has been characterized in detail. The PPy film containing MADH was formed electrochemically on a gold minielectrode (1mm diameter) in the presence of ferricyanide. The film was then coated with Nafion. This enzyme electrode did not require any additional cofactors and was not sensitive to oxygen. It exhibited a maximum response current to histamine at applied potentials of 0.24-0.33 V and at pH 7.5-8.5. This MADH-PPy sensor exhibited a response time of less than 3 s. The immobilized MADH on the electrode exhibited Michaelis-Menten behavior similar to that of the free enzyme in solution with a Km value of 1.3 mM. This sensor could be used to reliably detect histamine over a concentration range from approximately 25 µM to 4 mM. This is the first example of a biosensor that uses an immobilized enzyme that possesses the tryptophan tryptophylquinone prosthetic group. Quinoproteins are a recently characterized class of enzymes that use novel quinones as redox cofactors. Quinoprotein dehydrogenases do not require dissociable soluble cofactors, such as NAD+ or NADP+, and do not react directly with oxygen. These dehydrogenases normally donate electrons to other protein-bound redox centers, such as cytochromes and copper proteins.1-3 Quinoproteins have been characterized that oxidize a wide range of substrates including sugars, alcohols, aldehydes, and amines.1,2 Methylamine dehydrogenase (MADH) possesses tryptophan tryptophylquinone (TTQ, Chart 1) as a covalently bound prosthetic group.4 MADH, from Paracoccus denitrificans, is an H2L2 heterotetramer with subunit molecular masses of ∼47 and 15 kDa †
Jackson State University. The University of Mississippi Medical Center. (1) Davidson, V. L. In Biosensor Design and Application; Mathewson, P., Finley, J. W., Eds.; ACS Symposium Series 511; American Chemical Society: Washington, DC, 1992; pp 1-9. (2) Davidson, V. L.; Jones, L. H. Anal. Chim. Acta 1991, 249, 235-240. (3) Principles and Applications of Quinoproteins; Davidson, V. L., Ed.; Marcel Dekker: New York, 1993 and references therein. (4) McIntire, W. S.; Wemmer, D. E.; Christoserdov, A. Y.; Lindstrom, M. E. Science 1991, 252, 817-824. ‡
10.1021/ac9911138 CCC: $19.00 Published on Web 04/12/2000
© 2000 American Chemical Society
Chart 1. Structure of Tryptophan Tryptophylquinone (TTQ)a
a
The C6 and C7 positions are labeled.
that oxidizes primary amines to their corresponding aldehydes plus ammonia.5 To complete its catalytic cycle in vivo, it transfers electrons to a type I copper protein, amicyanin.6 The chemical conversion of substrate to product occurs in the enzyme active site at the C6 carbonyl of TTQ, but the subsequent electron transfer to amicyanin occurs from the edge of the other indole ring of TTQ that is exposed at the enzyme surface.7 This allows the TTQ prosthetic group to act as a bridge between the active site chemistry and surface-mediated redox reactions. This feature of these TTQ enzymes suggests that they may be useful components of enzyme-based electrodes. MADH has its highest affinity for short-chain aliphatic amines such as methylamine. However, it will also recognize other larger primary amines, including histamine.8 While the detection of primary amines such as methylamine has limited commercial application, analysis of the primary amine histamine is an important tool in the study of allergic responses and elevated histamine levels, which have been associated with a variety of pathological conditions such as gastric disorders, mastocytosis, (5) Husain, M.; Davidson, V. L. J. Bacteriol. 1987, 169, 1712-1717. (6) Husain, M.; Davidson, V. L. J. Biol. Chem. 1985, 260, 14625-14629. (7) Brooks, H. B.; Davidson, V. L. Biochemistry 1994, 33, 5696-5701. (8) Davidson, V. L. Biochem. J. 1989, 261, 107-111.
Analytical Chemistry, Vol. 72, No. 10, May 15, 2000 2211
and chronic myelogenous leukemia.9 Histamine is also produced in several types of fish and fish products as a result of the microbial decarboxylation of the amino acid histidine. This occurs after harvest, and if the fish are not properly handled histamine may accumulate to toxic levels. Histamine has been identified as the causative toxic agent in scombroid food poisoning, one of the most common causes of illness that is associated with consumption of fish.10 The ingestion of high levels of histamine causes a potentially fatal allergic-like reaction.11 It has also been suggested that histamine levels in fish and fish products may be a general indicator of the extent of deterioration. An enzyme-based electrode using MADH in conjunction with the mediator tetrocyanoquinodimethane (TCNQ) was previously constructed for the determination of histamine.12 With that electrode, a solution of MADH was kept in contact with a modified electrode by containing it in a semipermeable membrane (i.e., dialysis membrane). That study demonstrated the potential value of MADH as a component of an amine sensor, but clearly a more efficient and stable electrode design is necessary for any potential practical applications. In this work, we prepared an amine sensor by incorporating MADH into a polypyrrole (PPy) matrix on a gold minielectrode in the presence of ferricyanide anions by electrochemical polymerization. PPy is a widely used matrix for immobilizing enzymes on the electrode surface.13 The PPy film can be easily prepared by oxidative electropolymerization from pyrrole monomers in aqueous or nonaqueous solvent in the presence of both small and large anions.14-16 This is the first report of an enzyme electrode using an immobilized TTQ-containing enzyme. The properties of this MADH-based sensor are described with particular emphasis on its ability to detect histamine.
EXPERIMENTAL SECTION Chemicals and Materials. MADH was purified from P. denitrificans (ATCC 13543) as described previously.17 Nafion perfluorinated ion-exchange resin (5 wt % solution in a mixture of alcohol and water) was obtained from Aldrich. All other chemicals were reagent grade and were used without further purification. Deionized, doubly distilled water was used to prepare all aqueous solution. The gold minielectrodes used in this study were obtained from Cypress Systems, Inc., Lawrence, KS. Preparation of Electrodes. Gold minielectrodes (1-mm diameter) were polished successively with 30-, 15-, 9-, 6-, 3-, and 1-µm diamond polishing pastes followed by polishing with 0.05(9) Beaven, M. A.; Robinson-White, A.; Roderick, N. V.; Kauffman, G. L. Klin. Wochenschr. 1982, 60, 873-881. (10) Morrow, J. D.; Margolies, G. R.; Rowland, J.; Roberts, L. J. N. Engl. J. Med. 1991, 324, 716-720. (11) Sanchez-Guerrero, I. M.; Vidal, J. B.; Escudero, A. I. J. Allergy Clin. Immunol. 1997, 100, 433-434. (12) Loughran, M. G.; Hall, J. M.; Turner, A. P. R.; Davidson, V. L. Biosens. Bioelectron. 1995, 10, 569-576. (13) Sun, Z.; Tachikawa, H. In Biosensors and Chemical Sensors; Edelman, P. G., Wang, J., Eds; ACS Symposium Series 487; American Chemical Society: Washington, DC,, 1992; pp 134-149 and references therein. (14) Diaz, A. F.; Bargon, J. In Handbook of Conducting Polymers; Skotheim, T. A., Ed.; Marcel Dekker: New York, 1986; p 81 and references therein. (15) Street, C. B. In Handbook of Conducting Polymers; Skotheim, T. A., Ed.; Marcel Dekker: New York, 1986; p 265 and references therein. (16) Choi, C. S.; Tachikawa, H. J. Am. Chem. Soc. 1990, 112, 1757-1768. (17) Davidson, V. L. Methods Enzymol. 1990, 188, 241-246.
2212
Analytical Chemistry, Vol. 72, No. 10, May 15, 2000
µm alumina suspension. Then they were sonicated for 5 min in deionized, doubly distilled water, followed by extensive rinsing with deionized, doubly distilled water and then methanol. The MADH-PPy film was formed by applying a constant potential at 750 mV vs Ag/AgCl (saturated KCl) in an aqueous solution containing 30 mg/mL MADH, 0.35 M pyrrole, and 0.05 M potassium ferricyanide. The thickness of the MADH-PPy film was controlled by the charge passed during the electropolymerization and was estimated to be 1000 Å by passing 48 mC/cm2. To prepare a Nafion-coated surface, the MADH-PPy electrode was immersed in a Nafion solution. Then the electrode was removed from the solution and air-dried for several minutes. The procedure was repeated if necessary.18 Electrochemical Measurement. Electrochemical experiments were carried out with a CH Instruments model CHI 832 electrochemical detector. The electrochemical cell that was used consisted of a (1-mm diameter) gold minielectrode as the working electrode, a platinum wire as the auxiliary electrode, and an Ag/ AgCl (saturated KCl) as the reference electrode. All the electrochemical experiments were performed at room temperature (22 ( 2 °C). RESULTS AND DISCUSSION Immobilization of MADH in the PPy Film. The MADHPPy film was formed by applying a constant potential of 750 mV in an aqueous solution containing MADH, pyrrole, and ferricyanide. It was found that MADH was reduced within 2 h after pyrrole monomer was added to the MADH solution. Since the oxidized form of MADH in the MADH-PPy film must be maintained in order to detect amines in the solution, potassium ferricyanide was added to keep the MADH in the oxidized form and mediate electron transfer from reduced MADH to the electrode surface. There was no amperometric response to histamine (or other amines), if Fe(CN)63- was absent in the electrolyte during the electropolymerization of PPy. Similarly, no amperometric response current was seen at the PPy-coated gold electrode if no MADH was in the electrolyte for the electrochemical formation of PPy. Cyclic Voltammetric Behavior of the MADH-PPy Film. Figure 1 shows the cyclic voltammogram (CV) of a MADH-PPy film on a gold electrode in 0.1 M potassium phosphate, pH 7.5, containing 0.2 M KCl. The redox pair near +0.2 V is due to the redox reaction of ferricyanide/ferrocyanide. The broad cathodic peak at -410 mV and anodic peak at -290 mV is most likely due to the redox reaction of PPy. The E1/2 value (-350 mV) of this redox pair is 300-400 mV more negative than that of PPy doped with small anions but is characteristic of PPy doped with large anions.16 Since MADH carries the negative charge at pH 7.5,5 the negatively charged MADH and ferricyanide anions were entrapped in the PPy matrix to balance the charge during the polymerization. The possible contribution of MADH to this redox current was examined by recording the CV at different values of pH between 6.1 and 7.5. The E1/2 value of MADH in solution is known to vary with pH over this range.19,20 However, no pH effect on the E1/2 of this peak was observed, confirming that this current is due to the redox reaction of PPy. (18) Gerhart, G. A.; Oke, A. F.; Nagy, G.; Maghaddam, B.; Adams, R. N. Brain Res. 1984, 290, 390-395.
Figure 1. Cyclic voltammogram of MADH-PPy/gold (1-mm diameter) electrode in 0.1 M potassium phosphate, pH 7.5, containing 0.2 M KCl. Scan rate, 100 mV/s.
Scheme 1. TTQ-mediated Electron Transfer from Substrate Amine to the Electrode Surface on the MADH-PPy/Gold Electrode
Catalytic Response of the MADH-PPy Film Electrode to Histamine. The amperometric response of histamine was recorded at a MADH-PPy/gold electrode in 0.1 M pH 7.5 potassium phosphate solution containing 0.2 M KCl at 245 mV. MADH entrapped in the PPy film catalyzes the oxidation of histamine as illustrated in Scheme 1. As shown in this scheme, ferricyanide in the PPy film is likely to mediate the oxidation of TTQH- to TTQ to complete each catalytic cycle. Figure 2 shows the relationship between the amperometric response current and the applied potential to the electrode. The amperometric response current to histamine at the MADH-PPy/gold electrode increased as the applied potential was increased from +0.10 to +0.24 V. The maximum response currents were observed at potentials ranging from +0.24 to +0.33 V, and then current decreased as the applied potential moved to more positive potentials. Since the highest amperometric response was observed at approximately +0.24 V, which is slightly more positive than the formal potential of the Fe(CN)64-/Fe(CN)63- pair (see Figure 1), the observed potential dependency results support the mediated oxidation of MADH enzymes as shown in Scheme 1. There may be a couple of factors that contribute the decreased current at more positive potentials. (19) Davidson, V. L.; Brooks, H. B.; Graichen, M. E.; Jones, L. H.; Hyun, Y.-L. Methods Enzymol. 1995, 258, 176-190. (20) Zhu, Z.; Davidson, V. L. J. Biol. Chem. 1998, 273, 14254-14260.
Figure 2. Effect of applied potential on the amperometric response current for histamine. The MADH-PPy/gold electrode was in 0.1 M potassium phosphate, pH 7.5, containing 0.2 M KCl with 1.5 mM histamine present.
As the applied potential is moved to more positive potentials, an increase of Fe(CN)63- concentration accompanied by a decrease of Fe(CN)64- concentration is expected to occur in the PPy film. Under the high oxidation potentials, an increase in the rate of electron transfer for the oxidation of Fe(CN)64- to Fe(CN)63- at the gold electrode is also expected. Therefore, when a very positive potential is applied, a large oxidation current must flow as soon as histamine is added to the solution, but it decays quickly to a steady-state level which is diffusion controlled by the limiting Fe(CN)64-. This steady-state current level is what is measured. Another possible factor for the decreased amperometric current at very positive potentials may be that diffusion of the positively charged substrate histamine into the PPy-MADH film is less efficient. This type of potential dependency has been observed for the amperometric determination of H2O2 at the glassy carbon (GC) electrode modified with a catalyst polymer polymetallophthalocyanine (PMePc) film.21 The highest amperometric current response with H2O2 at the PMePc/GC electrode was observed near the formal potential of the PMePc(Red)/PMePc(Ox) pair. The response time of this biosensor with histamine was less than 3 s, which is typical for this type of conducting polymerbased biosensor. This response time is sufficiently fast to meet the requirements for in situ monitoring of dynamic processes in many biosensor applications. The pH effect on the amperometric response current with histamine (1.5 mM) was investigated in the pH range of 6.5-9.5. The highest response current was observed in the pH range of 7.5-8.5. This is similar to what was observed with MADH in solution where the pH optimum for its steady-state reaction with soluble redox dyes as electron acceptors was pH ∼7.5.8 Since the pH of most biological fluids is in this (21) Sun, Z.; Tachikawa, H. Anal. Chem. 1992, 64, 1112-1117.
Analytical Chemistry, Vol. 72, No. 10, May 15, 2000
2213
Table 1. Amperometric Response of Histamine and Potential Interferences at the MADH-PPy/Gold Electrode in 0.1 M PBS (0.2 M KCl, pH 7.5) substrate
physiologic concn in blooda (µM)
test concn (µM)
response (nA)b
histamine histamine (in serum)d glutamine creatine creatinine urea ascorbic acid
0.1-1 (1-240c) 0.1-1 (1-240c) 450-750 8-53 53-124 2500-7000 20-100
100 100 650 350 250 5000 100
75 75 0 0 0 0 0
a Reference 22. b Zero current indicates that the signal-to-noise ratio is less than 3:1. c The range of concentration observed for certain pathological conditions as described in text. d Measurement was made in new-born calf serum.
Figure 3. Calibration curve for the determination of histamine at the MADH-PPy/gold electrode. The electrode was in 0.1 M potassium phosphate, pH 7.5, containing 0.2 M KCl with an applied potential of +245 mV. The solid line represents a fit of the data to the MichaelisMenten equation: I ) Imax[histamine]/(Km + [histamine]). Error bars indicated the deviation of three repetitive determinations.
range, this type of biosensor should potentially be useful in clinical applications. The catalytic response of the MADH-PPy/gold electrode to histamine exhibited classical Michaelis-Menten kinetic behavior (Figure 3), very similar to that of the free enzyme in solution. The apparent Michaelis-Menten constant (Km) for histamine was calculated to be 1.3 mM. This value is in good agreement with the value obtained by steady-state kinetic analysis of free MADH in solution.8 The reaction was first order at low substrate concentrations and showed an approximately linear relationship at histamine concentrations ranging up to 500 µM. At the lower concentration limit for detection of histamine (25 µM), the signalto-noise ratio is ∼3:1. To calculate this, the amplitude of the oscillation (noise) of the baseline was compared with the magnitude of the change in signal from the baseline when 25 µM histamine was added to the solution. As seen in Figure 3, the hyperbolic curve, which describes the Michaelis-Menten behavior of the enzyme electrode, can be used for fairly accurate determination of histamine up to concentrations of ∼4 mM. The response of the MADH-PPy/gold electrode was both reproducible and stable. As shown in Figure 3, this electrode had a reproducible response over the concentration range tested. Effects of Potential Interfering Agents. MADH catalyzes the oxidation of primary aliphatic amines with a substrate specificity that decreases with increasing length of the aliphatic chain. The response of the MADH-PPy/gold electrode to other primary amines was tested. Interestingly, while methylamine is the best substrate for the free enzyme in solution, it did not react with the enzyme electrode. This may be because methylamine is relatively hydrophilic compared to histamine. The more hydrophilic methylamine may not be able to efficiently diffuse through the Nafion coating and PPy matrix to react with MADH. Two other more 2214 Analytical Chemistry, Vol. 72, No. 10, May 15, 2000
hydrophobic amines, propylamine and butylamine, did react well with the enzyme electrode. Each exhibited Michaelis-Menten kinetic behavior similar to that seen with histamine. Butylamine and propylamine exhibited Km values of 410 and 680 µM, respectively. No significant amount of either of these amines is expected to be present in blood or other biological samples. Furthermore, MADH does not react with secondary amines, tertiary amines, or aromatic amines.5 The amperometric response currents of MADH-PPy/gold electrode to other potential interfering agents that contain amino groups were also examined (Table 1). Four amino-containing compounds, which are present in blood and other biological samples, were tested: glutamine, creatine, creatinine, and urea.22 No response currents were observed for any of these compounds at approximately normal physiological concentrations. We have also tested the electrode against serum. The response of histamine in serum was basically identical to that in pH 7.5 phosphate buffer. Ascorbic acid, a common component of blood and biological samples, must always be considered a potential problem when an enzyme electrode is used for analysis of biological samples. The previously described MADH-based enzyme electrode, in which the enzyme was not immobilized, exhibited a large response to ascorbate that was ∼60% of the response to histamine.12 To eliminate the interference from ascorbic acid, the surface of the MADH-PPy/gold electrode was coated with Nafion (see Experimental Section). The MADH-PPy-Nafion/gold electrode did not respond to 100 µM ascorbic acid (Table 1). We also tested two possible interferences that may be present in decaying tissue: putrescine (1,4-diaminobutane) and cadaverine (1,5-diaminopentane). No response currents were observed for either of these amines at a concentration of 1000 µM. Potential Applications. Histamine sensors based on the design described in this paper could have potential applications in medicine and the food industry. Although histamine levels in blood are normally quite low, 0.1-1.0 µM,9 they may be elevated to levels within the detection range of the MADH-PPy electrode during certain pathological conditions. For example, elevated blood histamine levels have been observed in at least two neoplastic marrow diseases, polycythemia vera (1-6 µM) and chronic myelogenous leukemia (4-240 µM),9 the latter is in the (22) Higgins, S. J.; Turner, A. J.; Wood, E. J. Biochemistry for the Medical Sciences; Longman Scientific and Technical: Essex, U.K., 1994.
biosensor concentration detection range (25 µM-4 mM). With further refinement of the sensor design, it should be possible to improve the signal-to-noise ratio at low concentrations and increase sensitivity to below 25 µM. The MADH-PPy electrode could also be a desirable alternative to chromatographic and electrophoretic techniques that have been used for the determination of histamine in fish.23 (23) Mopper, B.; Sciacchitano, C. J. J. Assoc. Off. Anal. Chem. 1994, 77, 881884.
ACKNOWLEDGMENT This work was supported in part by the National Science Foundation (Grant CHE-9718644) and the National Institutes of Health (Grant S06GM08047).
Received for review September 27, 1999. Accepted February 1, 2000. AC9911138
Analytical Chemistry, Vol. 72, No. 10, May 15, 2000
2215
