Original Article A novel method for measuring serum ornithine carbamoyltransferase Hiroaki Ishikawa 1, Takeo Matsuzawa 2, Koji Ohashi 2 and Yoichi Nagamura 2 Abstract Addresses 1 Department of Medical Technology Fujita Health University College 2 Department of Clinical Chemistry School of Health Sciences Fujita Health University Toyoake, Aichi 470-1192, Japan Correspondence Hiroaki Ishikawa E-mail: hishikaw@fujita-hu.ac.jp Background Serum ornithine carbamoyltransferase is a diagnostic marker of hepatic disorders due to its localization in periportal mitochondria. Methods We have developed a new method for the determination of serum ornithine carbamoyltransferase. It is based on the reverse reaction of ornithine carbamoyltransferase, using ornithine-ketoacid aminotransferase, 1 -pyrroline-5- carboxylate dehydrogenase and glutamate dehydrogenase, which together convert citrulline through ornithine to glutamate. The glutamate is then quantitatively measured using glutamate oxidase and Trinder s reagent. Results The results obtained by this method agreed well with those obtained using the diacetylmonoxime method as a gold standard [correlation coef cient (r ) ˆ 0 973 P50 001]. The endogenous amino acids sensitive to this method in serum (glutamate, ornithine and 1 -pyrroline-5-carboxylate) were eliminated by the initial futile reaction. The new method appears to be more accurate at low levels of ornithine carbamoyltransferase activity than the diacetylmonoxime method. Conclusions Here we report a new method for serum ornithine carbamoyltransferase assay which might be useful for clinical diagnosis of hepatic disorders, including hepatic cancer. Introduction Ornithine carbamoyltransferase (OCT; EC 2.1.3.3) is the second of the ve enzymes that constitute the urea cycle. It catalyses the transfer of carbamoyl phosphate to the d-amino residue of ornithine to produce citrulline and inorganic phosphate. This enzyme is a mitochondrial matrix enzyme which occurs mainly in the liver, where it is exclusively localized in periportal hepatocytes and the small intestine. The OCT activity in the liver is so high that the activity released into serum has been used as a diagnostic marker of hepatic disorders, especially those involving damage to periportal mitochondria. Serum OCT activity was assayed originally by Archibald s method 1 using diacetylmonoxime, which reacts non-speci cally with citrulline, one of the reaction products of the OCT reaction. Diacetylmonoxime also reacts with urea and several other guanidine compounds. This method, however, is not suitable for automated analysis because of the necessity of urease pre-treatment of the serum, the use of strong acids (a mixture of concentrated sulphuric and phosphoric acids) and a boiling step. 2 For automated analysis, a speci c enzyme-based assay would be ideal. Here we describe such an assay, which determines OCT through the reverse reaction catalysed by the enzyme and which, following automation, is suitable for high-throughput screening. Materials and methods Principle Citrulline is converted to ornithine by the reverse reaction of OCT, then ornithine is quantitatively transaminated 3 and oxidized producing two molecules of glutamate in the presence of ornithine ^ ketoacid aminotransferase (OKT; EC 2.6.1.13) and 1 -pyrroline-5-carboxylate dehydrogenase (P5CDH; EC 1.5.1.12). 4 Glutamate is quantitated using glutamate oxidase (GOD; EC 1.4.3.11) in the presence of Trinder s reagent, 4-aminoantipyrine and TOOS (N-ethyl-N-(2- hydroxy-3-sulphopropyl)-m-toluidine sodium salt dihydrate), which form a coloured quinoneimine derivative (absorption maximum ˆ 555 nm) following 264 &2003 The Association of Clinical Biochemists
Measuring serum ornithine carbamoyltransferase 265 Reagents for Archibald s diacetylmonoxime method were OCT assay reagent 2 (OCT-test, Wako) obtained from Wako Chemical Company (Tokyo, Japan). OKT, P5CDH and POD were stored at 7208C; other reagents were kept at 48C. Recombinant rat OCT was prepared as described previously. 6 Serum aspartate and alanine aminotransferase activities were determined with Iatrozyme TA-Lq reagent 7 obtained from Iatron Laboratories, Inc. (Tokyo, Japan). Figure 1. Reaction scheme for the conversion of citrulline to glutamate during the quantitative determination of OCT activity. Cit ˆ citrulline; Pi ˆ inorganic phosphate; CarbP ˆ carbamoyl phosphate; Orn ˆ ornithine; akg ˆ 2-oxoglutarate; Glu ˆ glutamate; P5C ˆ pyrroline-5-carboxylate; NADP + ˆ nicotine adenine dinucleotide phosphate, oxidized; NADPH ˆ nicotine adenine dinucleotide phosphate, reduced; 4AA ˆ 4-aminoantipyrine; TOOS ˆ N-ethyl-N-(2-hydroxy-3-sulphopropyl)-mtoluidine sodium salt dihydrate; OCT ˆ ornithine carbamoyltransferase; OKT ˆ ornithine-ketoacid aminotransferase; P5CDH ˆ pyrroline-5-carboxylate dehydrogenase; GDH ˆ glutamate dehydrogenase; GOD ˆ glutamate oxidase; POD ˆ peroxidase. oxidation by peroxidase (POD; EC 1.11.1.7). In the course of the reaction sequence, three molecules of glutamate arise per molecule of citrulline: one is produced by OKT from 2-oxoglutarate, another from pyrroline-5-carboxylate by P5CDH and a third by glutamate dehydrogenase (GDH; EC 1.4.1.3) used for the regeneration of NADP + from NADPH. This leads to signal ampli cation and contributes to the sensitivity of the assay (see Fig. 1). Furthermore, endogenous glutamate plus ornithine and 1 -pyrroline-5-carboxylate in the serum can be eliminated by an initial futile reaction without addition of 4-aminoantipyrine of Trinder s reagent; therefore only glutamate produced in the present method is quantitated by glutamate oxidase and Trinder s reagent. Enzymes and reagents OKT and P5CDH were puri ed from Bacillus sphaericus. 5 GDH and POD were purchased from Boehringer Mannheim GmbH (Mannheim, Germany). GOD was obtained from Seikagaku Corporation (Tokyo, Japan). 4-Aminoantipyrine and TOOS were from Dojindo Laboratories (Kumamoto, Japan). L-Canaline was from Sigma (St Louis MO, USA). Serum samples Venous blood samples were obtained from 150 healthy persons and 55 patients, including those with hepatic disorders, at the clinical laboratory of Fujita Health University Hospital. The separated sera were stored at 7858C until use. The research protocol was approved by the Research Ethics Committee of Fujita Health University. Procedures Enzyme reaction mixtures A, B, C and D were prepared as follows: Mixture A consisted of 160 mmol of potassium phosphate bu er (ph 8 0), 7 6 U of OKT, 6 5 U of P5CDH, 32 mmol of NADP +, 50 mmol of 2-oxoglutarate, 300 mmol of ammonium chloride and 48 U of GDH in a nal volume of 1mL. Mixture B contained 17 mmol of TOOS, 400 U of POD and 6 8 U of GOD in a nal volume of 5 ml of water. Mixture C was 10 mmol of 4-aminoantipyrine and 570 mmol of L-citrulline in a nal volume of 2 ml of water. Mixture D contained 10 mmol of 4-aminoantipyrine in a nal volume of 2 ml of water. Mixture A was prepared before use and was stable for 1 day if kept on ice; the other mixtures could be used for 4^5 days if kept at 48C. Mixture A+B was made before starting the reaction by mixing equal volumes of mixtures A and B. For each serum sample a separate blank reaction was carried out. To two 1 5-mL plastic centrifuge tubes, 20 ml of serum and 100 ml of mixture A+B were added. Both tubes were incubated at 378C for 30 min (to eliminate endogenous glutamate, ornithine and 1 -pyrroline-5-carboxylate) (the futile reaction); then 20 ml of mixture D was added to the blank tube and 20 ml of mixture C to the OCTassay tube. After 60 min of incubation at 378C, the reaction was terminated by the addition of 500 ml of 2610 74 mol/l L-canaline, and the mixture was centrifuged at 7000 g for 10 min to remove any turbidity. The absorbance of the supernatant was measured spectrophotometrically at 555 nm within 1h. The enzyme activity (U/L) was calculated from the di erence between the absorbances of the blank tube and the OCT assay tube using the molar extinction coe cient of the end-product glutamate produced in the reaction mixture (e ˆ 7800 cm/mol at 555 nm) which was obtained from the calibration curve as shown in Fig. 2. Ann Clin Biochem 2003: 40: 264 268
266 Ishikawa et al. an almost linear response of the assay to the very small amount of OCT added. Due to the higher speci c activity of the recombinant OCT, the incubation time in this experiment could be shortened to 15 min with the other conditions being very similar to those described in`materials and methods (see Fig. 3). Figure 2. Linearity of the assay following the addition of glutamate. Varying amounts (5 80 nmol) of glutamate were added to the reaction mixture. The reaction was carried out as described in Materials and methods. Statistical calculations Statistical calculations for this study were carried out using the software package StatView purchased from Abacus Concepts, Inc. (Berkeley, CA, USA). Results Linearity of the assay Initially we checked the linearity of this method by measuring the absorbance of the reaction mixture after the addition of 5^80 nmol of glutamate to the reaction mixture. A linear dependence of the absorbance at 555 nm in the range 0^1 0 on the amount of glutamate added (5^80 nmol) was con rmed (Fig. 2). The two preparations of recombinant rat OCT showed Termination of the assay The enzyme reaction was e ciently terminated by the addition of L-canaline, which is a potent inhibitor of OKT. 9 We tested how long this inhibitor could suppress the increase of absorbance at 555 nm after termination of 20 serum samples, including low and high activities in the normal range. Our results showed that the inhibitor was active for 1h; the absorbance then gradually increased, indicating that the absorbance should be measured within 60 min after termination of the reaction with L-canaline (see Fig. 4). Comparison of the present method and the diacetylmonoxime method The OCT activities of serum samples from 55 patients with hepatic disorders and 13 healthy individuals were determined by the present method and the diacetylmonoxime method. A linear relationship was obtained between the two methods and the correlation coe cient (r) was calculated to be 0 973 (P50 001) (see Fig. 5a). The Bland ^ Altman di erence plot comparing the present method with the diacetylmonoxime method showed a concentration-dependent positive bias (Fig.5b). 10 At an activity of less than 11 IU/L, the diacetylmonoxime method sometimes yielded a value of zero (see also Fig. 6), whereas the new method consistently gave non-zero results. The initial futile reaction was useful, because the absorbance at 555 nm of the blank tube was usually less than 0 100. Figure 3. Linearity of the assay following the addition of recombinant OCT activity. Two different preparations of recombinant rat ornithine carbamoyltransferase (1 4 ml) were added to the reaction mixture and assayed. The initial futile reaction was omitted, and the enzyme reaction was carried out only for 15 min. The values were obtained with two recombinant OCT preparations, 452 IU/L (open circles) and 155 IU/L (closed circles), respectively. Figure 4. Effect of termination of the reaction by L-canaline. Ornithine carbamoyltransferase activities of 20 serum samples were assayed and terminated by the addition of L-canaline. The increase of the absorbance was then monitored. The values indicate the mean+standard deviation.
Measuring serum ornithine carbamoyltransferase 267 Figure 6. Correlation of the two methods for the determination of ornithine carbamoyltransferase (OCT) activities within the normal range. OCT activities of 150 normal serum samples were assayed by the present method and by the diacetylmonoxime method; the values were plotted using StatView. whereas the present method produced an almost symmetric single peak in the histogram (Fig. 6). Figure 5. Relationship between ornithine carbamoyltransferase (OCT) activity measured by the present method and the diacetylmonoxime method (a) and Bland Altman difference plot (b). OCT activities of 55 patients serum samples and 13 normal serum samples were assayed by the present method and by the diacetylmonoxime method. The values were plotted and the Pearson correlation coef cient was calculated using StatView. (a) The solid line represents a linear regression equation of y ˆ 3 986x+32 1 (r ˆ 0 973, P50 001). (b) Differences [(OCT activity by the diacetylmonoxime method)7(oct activity by the present method)] are plotted against the average values by the two methods. SD ˆ standard deviation. Range of normal values The OCT activities of serum samples from 150 healthy persons were measured by the present method and by the diacetylmonoxime method, and their normal ranges and histograms were compared. The mean value (2 standard deviations) was calculated to be 6 301 (5 59) for the present method and 7 392 (10 96) for the diacetylmonoxime method. The normal range was estimated as 0 71^11 89 IU/L for the new method and 0 00^18 35 IU/L for the diacetylmonoxime method. The histograms and scatter graphs of the values obtained by the two methods are shown in Fig. 6. As pointed out in the previous section (see Fig. 5), the diacetylmonoxime method gave two peaks consisting of a high null and a broader lower distribution, Reproducibility Measurement of OCT activity ten times in two serum samples, one with an activity of 56 0 IU/L and another with an activity of 9 8 IU/L, gave a percentage coe cient of variation (%CV) of 3 4 and 6 5, respectively, using the new method. Discussion OCT is an enzyme located in the mitochondrial matrix of the liver and small intestine. There is no activity of this enzyme in the kidney, heart, skeletal muscle, brain or other tissues. In particular, the activity of this enzyme is rich in the liver where it is localized exclusively in periportal mitochondria. Its release into the circulation should therefore be a useful liver-speci c diagnostic marker of damage to the periportal mitochondria in liver diseases, such as various types of hepatitis and liver carcinomas. On the other hand, in the case of intestinal disorders, the intestinal enzyme may be released into the alimentary canal and not into the circulation. This new method utilizes the reverse reaction of OCT; 4 1mol of ornithine that is produced from the substrate citrulline is nally converted into 3 mol of glutamate through the actions of OKT, P5CDH and GDH. OCT itself favours the forward reaction, but is forced to catalyse the reverse reaction by a successive conversion of ornithine into glutamate and Ann Clin Biochem 2003: 40: 264 268
268 Ishikawa et al. then into 2-oxoglutarate and ammonia in the present method. The present method is an enzymatic method and is sensitive to interference from endogenous glutamate, ornithine and 1 -pyrroline-5-carboxylate present in the serum. The total concentration of these endogenous amino acids is less than 3 nmol in 20 ml of serum. However, the colouration and therefore interference due to them in the serum can be e ectively eliminated by the initial futile reaction without 4- aminoantipyrine. In practice, the blank tube usually gave an absorbance of less than 0 1 absorbance unit at 555 nm. Thus only the glutamate produced by the method is quantitatively measured by glutamate oxidase and Trinder s reagent. The present method gave a narrower reference range (0 71^11 89 IU/L) than the diacetylmonoxime method (0 00^18 35 IU/L). Furthermore, unlike the diacetylmonoxime method, it did not give rise to false zero values at very low activities. Despite this discrepancy, the agreement between the two methods was good, as indicated by the high correlation coe cient (0 973). A %CV of 3 4 was obtained following analysis of patients serum samples, which is satisfactory for the routine analysis of patients samples. In the screening of 55 patients serum samples, the correlation between OCT value and other values such as aspartate aminotransferase (AST) or alanine aminotransferase (ALT) was studied. The correlation coe cient was 0 888 with ALT and 0 951 with AST. The increased ratio of OCT/ALT has been reported as a useful marker for diagnosis of liver carcinoma, 8 which may enhance the utility of highthroughput screening of serum OCT activity in an autoanalyser. The termination of the reaction by the addition of L-canaline at a nal concentration of1 56610 4 mol/l was e ective for only 60 min. The K i value of L- canaline for OKT was reported as 10 77 mol/l, 9 but L-canaline is able to bind several other proteins in a non-speci c manner. 9 If it was supplemented with catalase, the termination would be more e ective. Acknowledgements The authors thank Mrs K Matsunaga and K Isobe, Amano Pharmaceutical Company, for supplying OKT and P5CDH, Dr H Ogawa for preparation of recombinant rat OCT, and Dr M Mori for supplying rat OCT cdna. We also thank Dr A Treumann for reading this manuscript and Ms M Sasaki and Ms AUchida for their technical assistance. The work was supported by a Grant-in-Aid from Fujita Health University. References 1 Archibald RM. Determination of citrulline and allantoin and demonstration of citrulline in blood plasma. J Biol Chem 1944; 156: 121 42 2 Ohshita M, Takeda H, Kamiyama Y, Ozawa K, Honjo I. A direct method for the estimation of ornithine carbamoyltransferase activity in serum. Clin Chim Acta 1976; 67: 145 52 3 Matsuzawa T, Ito M, Ishiguro I. Enzymatic assays of L-ornithine and L- 1 -pyrroline-5-carboxylate in tissues, and ornithine-load test in human subjects. Anal Biochem 1980; 106: 1 6 4 Isobe K, Matsuzawa T, Nagamura Y. A new enzymatic determination of ornithine carbamoyltransferase activity. Anal Lett 1993; 26: 475 86 5 Isobe K, Matsuzawa T, Soda K. Crystallization and characterization of 1-pyrroline-5-carboxylate dehydrogenase from Bacillus sphaericus. Agric Biol Chem 1987; 51: 1947 53 6 Matsuzawa T, Kobayashi T, Tashiro K, Kasahara M. Changes in ornithine metabolic enzyme induced by dietary protein in small intestine and liver: intestine liver relationship in ornithine supply to liver. J Biochem 1994; 116: 721 7 7 Lippi U, Guidi G. A new colorimetric ultramicromethod for serum glutamic oxalacetic and glutamic pyruvic transaminase determination. Clin Chim Acta 1970; 28: 431 7 8 Watanabe Y, Mori S, Fujiyama S, Sato T, Mori M. Clinical evaluation of serum ornithine carbamoyltransferase by enzyme-linked immunosorbent assay in patients with liver disease. Enzyme Protein 1995; 48: 18 26 9 Kito K, Sanada Y, Katunuma N. Mode of inhibition of ornithine aminotransferase by L-canaline. J Biochem 1978; 83: 202 6 10 Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986; i: 307 10 Accepted for publication 8 January 2003