1012—JOURNAL OF FOOD SCIENCE—Volume 61, No. 5, 1996 Amperometric Biosensor for Total Histamine, Putrescine and Cadaverine using Diamine Oxidase KEITH B. MALE, PIERRE BOUVRETTE, JOHN H.T. LUONG, and BERNARD F. GIBBS ABSTRACT Diamine oxidase was purified from porcine kidney to homogeneity and immobilized onto a porous preactivated nylon membrane. The enzymic membrane was attached to an amperometric electrode for determination of the total concentrations of histamine, cadaverine and putrescine ac- cumulated in fish fillets during storage. For detecting hydrogen peroxide released, the working platinum electrode was poised at +400 mV vs Ag/AgCl to prevent electrode fouling as well as minimize electroactive interferences. The biosensor was linear up to 6 mM with a lower detec- tion limit of 25 μM for the three substrates and steady state response was achieved within 1.5 min. The enzyme membranes were stable at 5°C for 2 mo and usable for at least 60 assays. Key Words: amperometric biosensor, diamine oxidase, cadaverine, pu- trescine, fish freshness INTRODUCTION CHEMICAL INDICATORS of fish freshness, including ammonia, amines, diamines, volatile acids, trimethylamine and nucleotide degradation products, have been proposed as methods for de- termination of fish quality (Hughes and Jones, 1966; Chang et al., 1976; Mietz and Karmas, 1977). The nucleotide degradation product hypoxanthine causes product off-taste upon storage (Jones et al., 1964). Its concentration is a good indicator of fish freshness for many species and various procedures have been developed for determination of hypoxanthine in fish (Jones and Murray, 1964; Burt et al., 1968; Watanabe et al., 1983; Luong et al., 1989). A more elaborate freshness index, the K-value, was introduced to take into account three nucleotide degradation products, inosine-5'-monophosphate (IMP), inosine and hypo- xanthine (Saito et al., 1959). The K-value, defined as a ratio between concentrations of hypoxanthine and inosine and the to- tal concentrations of hypoxanthine, inosine and IMP, reflects the early post-mortem enzymatic changes in fish tissue as the ratio increases from 0 to 1. The accumulation of histamine and di- amines then becomes considerable after the levels of hypoxan- thine have reached a maximum and these compounds are responsible for the odor of decaying meat. Therefore, for a more complete test the level of nucleotide degradation products such as hypoxanthine should be integrated with a test indicating spoil- age compounds produced by microbial action (Jahns and Rand, 1977). Histamine and the two diamines, putrescine and cadaverine, were observed to accumulate in fish tissue as spoilage com- menced (Mietz and Karmas, 1977). Cadaverine is a biogenic amine produced enzymatically by decarboxylation of lysine, whereas putrescine and histamine are formed by decarboxyla- tion of ornithine and histidine, respectively. Levels of histamine, putrescine and cadaverine have been suggested as rapid fish spoilage indicators and these compounds have also been impli- cated in human poisoning when present at considerable concen- trations (Kim and Bjeldanes, 1979; Bartholomew et al., 1986; Todd et al., 1992). Current analytical techniques have several deficiencies. The HPLC method was complicated and time-con- The authors are affiliated with the Biotechnology Research Insti- tute, National Research Council Canada, Montreal, Quebec, Can- ada, H4P 2R2. suming (Mietz and Karmas, 1977), so other alternatives were considered. An enzymatic assay using diamine oxidase was ca- pable of selectively detecting histamine, putrescine and cadav- erine (Mondovi et al., 1964). The enzyme is abundant in porcine kidney, plasma, human placenta and pea seedlings. A compre- hensive review of properties, kinetics, and specificity of this enzyme has been published (Zeller, 1963). The reaction can be followed by several methods including substrate disappearance, oxygen consumption or the production of ammonia, hydrogen peroxide and an aldehyde (Zeller, 1963). RCH 2 NH 2 + H 2 O + O 2 → RCHO + NH 3 + H 2 O 2 Although diamine oxidase is commercially available from sev- eral suppliers, the specific activity of this costly enzyme is very low. For instance, the Sigma diamine oxidase purified from hog kidney only has a specific activity of 0.06 unit/mg solid (1 μmol putrescine oxidized/hr at 37°C). Using the more conventional unit definition (μmol/min), the specific activity would be 0.001 unit/mg solid. In general, for development of an amperometric biosensor for histamine and/or diamines with satisfactory re- sponse, sensitivity and detection limit, the enzyme must have a specific activity of about 1 unit/mg solid. The main objective of our work was to develop and evaluate the performance of an amperometric biosensor using immobi- lized diamine oxidase, an enzyme which has been purified to homogeneity in our laboratory. The biosensor was applied for determination of histamine/diamines in fish tissue as spoilage occurred. The applicability of the biosensor in terms of sensi- tivity, reliability, accuracy and stability was also examined and the results were compared with those of a high performance amino acid (HPAA) analyzer. MATERIALS & METHODS Materials Cadaverine (1,5-diaminopentane), putrescine (1,4-diaminobutane), glutaraldehyde, methyl--D-glucopyranoside, horseradish peroxidase (HRP), heparin-agarose, hydroxyapatite (HA-ultrogel) and bovine serum albumin (BSA) were purchased from Sigma Chemical Co. (St. Louis, MO) and histamine and 4-aminoantipyrine were obtained from Aldrich (Milwaukee, WI). Concanavalin A-sepharose was a product of Phar- macia (Uppsala, Sweden). All other chemicals were purchased from An- achemia (Montreal, QC), except hydrogen peroxide (Fisher Scientific Co., Fair Lawn, NJ). The preactivated Immunodyne membrane (pore size 3 μm) was supplied by Pall BioSupport Corporation (East Hills, NY). Porcine kidneys, as well as samples of sole and rainbow trout, were purchased from local markets. Purification of diamine oxidase from porcine kidney Cortexes of 10–12 fresh or frozen porcine kidneys (ca. 1.3 kg) were minced and homogenized for 3–5 min in 4L of 0.1M sodium phosphate buffer, pH 7.4, using a Waring Blendor, until a smooth suspension was obtained. After heating to 66°C for 15–20 min, the resulting homogenate was cooled to 25°C, followed by centrifugation for 30 min at 14,300  g, 10°C. Ammonium sulfate (35% saturation) was then added to the resulting crude extract and the resulting solution was stirred for 1 hr at 4°C. After centrifugation, ammonium sulfate was added to the superna- tant to a final concentration of 60% (saturated level at 4°C). The resulting precipitate (35–60%), containing the DAO activity obtained by centrif- ugation, was then dissolved in 0.1M sodium phosphate buffer, pH 7.4,