Flow-injection determination of 3-hydroxybutyrate in serum with an immobilized 3-hydroxybutyrate dehydrogenase reactor and fluorescence detection

Flow-injection determination of 3-hydroxybutyrate in serum with an immobilized 3-hydroxybutyrate dehydrogenase reactor and fluorescence detection

Talanra. Vol. 41. No. 9. WI. I583-1586. 1994 G&right ‘cj IWb’&evier Scicn& Ltd Printed in Great Britain. All rights reserved 0039-9140/94$7.00+ 0.00 ...

319KB Sizes 0 Downloads 11 Views

Talanra. Vol. 41. No. 9. WI. I583-1586. 1994 G&right ‘cj IWb’&evier Scicn& Ltd

Printed in Great Britain. All rights reserved 0039-9140/94$7.00+ 0.00

SHORT

COMMUNICATION

FLOW-INJECTION DETERMINATION OF 3-HYDROXYBUTYRATE IN SERUM WITH AN IMMOBILIZED 3-HYDROXYBUTYRATE DEHYDROGENASE REACTOR AND FLUORESCENCE DETECTION NOBUTOSHIKIBA, HIDEKAZU KOEMADOand MOTOH~SAFURUSAWA Department of Applied Chemistry and Biotechnology, Faculty of Engineering, Yamanashi U.niversity, Kofu 400, Japan (Received

23 December

1993. Revised

14 February

1994. Accepted

11 February

1994)

Summary-A flow-injection system with an immobilized enzyme reactor is proposed for the determination of 3-hydroxybutyrate. 3-Hydroxybutyrate dehydrogenase is immobilized on aminated poly(viny1 alcohol) beads and packed into a stainless-steel column (4 cm x 4 mm I.D.). Serum is diluted and filtered. Sample solution (20 ~1) is injected into the carrier stream [4mM NAD+ in glycine buffer (pH 9.3)]. The NADH formed is detected at 465 nm (excitation at 340 nm). The calibration graph is linear for 0.7-5OOpM 3-hydroxybutyrate; the detection limit is OS@.

The concentration of 3-hydroxybutyrate (HB) in blood is significantly affected by various physiological and pathological conditions in mammals. Since under starvation and diabetic conditions HB production is markedly increased, it is desirable to have a reliable clinical assay method. The most commonly used assay methods for HB are enzymatic,‘-’ which involve 3-hydroxybutyrate dehydrogenase (EC 1.1.1.30, HBDH) (enzymatic reaction; HB + nicotinamide adenine dinucleotide (NAD+) = acetoacetate + reduced nicotinamide adenine dinucleotide (NADH) + H+). Among them manual methods’*3.6,7 based on the spectrophotometric measurement of NADH are timeconsuming because an end-point technique is used to obtain the sufficient sensitivity. While centrifugal methods for autoanalyser have been standardized,2 a small diagnostic laboratory does not always have access to such sophisticated instruments for the determination of HB. In this situation an immobilized HBDH electrode in batch system has been developed for the rapid determination of HB in undiluted whole blood and plasma.’ If an immobilized HBDH is used in a continuous-flow system, manual

handling is minimized and the reproducibility is enhanced. The immobilized HBDH has not been used as a reactor in continuous-flow system. Since the equilibrium constant6 for the reaction catalyzed by HBDH is 1 x 10e9, for sensitive assay a higher pH (pH 9.5) and a higher concentration of NAD+ (15mM) were used to shift the equilibrium of the reaction in the direction of NADH.’ In most previous studies, the immobilized enzymes in continuousflow system were all in packed-bed type reactors and glass beads were used as support for the covalent attachment of enzymes.8 However, in this case glass beads were not used as the support for HBDH since they are unstable in alkaline solution.9 In this paper, a flow-injection fluorimetric method for the determination of HB with immobilized HBDH reactor is reported. The enzyme was covalently immobilized with glutaraldehyde on poly(viny1 alcohol) beads which are relatively stable in alkaline solution. A fluorimetric procedure was used to monitor the appearance of NADH. The flow-injection method was applied to the determination of HB in serum.

1584

N. KIBA et al.

EXPERIMENTAL Materials

HBDH (D-3-hydroxybutyrate: NAD+ oxidoreductase, from Pseudomonas sp., grade III, 120 U/mg) was obtained from Toyobo (Osaka). Activity for the HBDH was measured spectrophotometrically at 340 nm with o,L-3-hydroxybutyrate as substrate at pH 8.3 at 40°C. NAD+ (free acid, 96%) and NADH (disodium salt, 98%) were from Kohjin (Tokyo). Poly(viny1 alcohol) beads (G-520, 13 pm diameter) were from Shyowa Denko (Tokyo). D,L-3Hydroxybutyrate sodium salt (D,L-HB salt) was from Nacalai Tesque (Kyoto). All other chemicals were of analytical-reagent grade. Glycine buffer (pH 9.3) consisting of O.lM glycine-0.1 M NaCl-O.lM NaOH was prepared. NAD+ solution [8mM NAD+] in O.OlM phosphate buffer (pH 7) was prepared daily. Stock HB solution (10mM) was made by dissolving D,L-HB salt in water. D-Isomer concentration was assumed to be half of D,L-HB salt. The method for the amination of the beads was similar to that described previously.” The aminated beads were packed into a stainlesssteel column (4 cm x 4 mm I.D.) by the slurrypacking method. Glutaraldehyde (2%) in O.lM phosphate buffer (pH 7) was pumped through the column for 3 hr at 0.4 ml/min at 25°C and the column washed with deaerated water for 30 min at 0.5 ml/min. Enzyme solution [HBDH 5 mg (600 U) in 10 ml of 0.05M phosphate buffer (pH 7)] was circulated through the column for 6 hr at 25°C. After the immobilization process, activity in the solution was measured. HBDH was immobilized on the beads with a 72% yield; it was assumed that inactivation of HBDH during the immobilization process did not occur.

Serum (5 ~1) was diluted IO-fold with the glycine buffer (pH 9.3) and filtered on a ultrafiltration membrane (Advantec QOlOO,nominal molecular weight cut-off 10,000). The filtrate (20 ~1) was injected via a sample injector. The time taken to prepare the sample solution was about 1 min. RESULTS

AND DISCUSSION

Optimization

The effect of pH on the activity of the reactor was studied in the pH range 8.8-9.8 using glycine buffer. As shown in Fig. 1, the optimum pH was about 9.3. The temperature dependence of the reactor was investigated over the range 20-55°C. The reactor exhibited maximum activity at 40°C as shown in Fig. 1. Compared with free HBDH (optimum pH and temperature were 8.3 and 55°C respectively), the maximum pH and temperature were ca. 1 pH unit more basic and 15°C lower, respectively. The effect of NAD+ concentration on the activity was studied over the range 0.5-lO.OmM at the 5OOpM HB level. The response increased with increasing concentration, first rapidly and then gradually. Above 5mM, the response was almost constant. A concentration of 8mM NAD+ was chosen to prevent interference of acetoacetate which is one of ketone bodies and a product of the enzymatic reaction; at this concentration, the NAD+ concentration on the reactor is 4mM.

Temperature 20

(‘C) 40

30

50

Apparatus and procedure

The glycine buffer (pH 9.3) and the NAD+ solution were each pumped by a double-plunger pump (KHU-W-52, Kyowa Seimitsu, Japan) at a flow-rate of 0.3 ml/min and mixed before entering an injector @VI-6M2, Sanuki, Japan) with a 20 ~1 loop. The combined solution entered the reactor which was maintained at 40°C with a thermostatted water-bath. Total flow-rate through the reactor was 0.6 ml/min. The NADH produced was monitored at 465 nm (excitation wavelength was 340 nm and entrance and exit slits were 10 nm each) with spectrofluorimeter (FP-210, Jasco, Japan) with flowthrough cell (15 ~1).

9.0

9.4

9.8

PH Fig. 1. Effects of (0) pH and (0) temperature activity of the immobilized HBDH.

on the

3-Hydroxybutyrate in serum

The peak-height decreased linearly as the flow-rate of carrier stream increased from 0.4 to 1.0 ml/min. The peak-height at 0.4 ml/min was about 2.4 times that at 1.0 ml/min. A total flow-rate of 0.6 ml/min was selected, as a compromise between sensitivity and sample at this flow-rate, the sample throughput: throughput was 30 hr-‘; the throughput was identical to that of the centrifugal analyzer.2 Under the conditions of 4mM NAD+ at pH 9.3 and 40°C the conversion efficiency of the reactor was 85% immediately after the preparation of the reactor. The conversion efficiency was measured by using NADH. Under the same conditions, L- and D-lactates, L-malate and 3hydroxypropionate did not give any response at 1mM level. The presence of up to three times molar amounts of acetoacetate did not interfere with the measurements of HB. The operational stability of the reactor was evaluated over 4 weeks. The reactor was used for analysis of 150 samples (lO@f HB) for 5 hr per day and stored at 4°C in O.lM phosphate buffer (pH 7) when not in use. The conversion efficiencies decreased gradually to 82, 78,75 and 72% after 1 (1050), 2 (2100), 3 (3150) and 4 weeks (4200 injections), respectively. Calibration

The calibration graph of peak-height against HB concentration was linear over the range 0.7-500pIi4 with a correlation coefficient of 0.998 (10 data points) under the same conditions as described in Apparatus and Procedure; since the serum sample was diluted lo-fold, this method can be applied to the assay of serum containing 7@4-5mM HB. Below a concentration of 0.7@4 a concave graph was obtained and above a concentration of 5OOpM, the graph curved convexly. The relative standard deviation (RSD) for 15 replicate injections of 5.OpM HB was 0.82% with the reactor having the conversion efficiency of 80%. By the use of the reactor having the conversion efficiency of 70%, the RSD for same runs was 1.25%. To obtain precise results, the reactor must be used within the conversion efficiency of 80%. The limit of detection (signal-to-noise ratio = 3) with the reactor having the conversion coefficiency of 85% was 0.5piU (1 ng in a 20 ~1 injection) with 5.5% RSD.

1585

Table I. Recovery of 3-hydroxybutyrate added to pooled serum* Added (PM)

Recoveredt (PM)

30.0 150 300 1050 2100

31.1 I55 303 1060 2070

4200

4030

Recovery (%) 104 103 101 101 99 96

*Values corrected for HB (29.3~144) already present in serum. tAll values are means (PM) (n = 5).

Pooled human serum was repeatedly analyzed for 4 weeks with the reactor. The reactor was used for analyses of 120 samples in a day and standards were measured at &I-sample intervals, in order to correct the variation of the conversion efficiency. The reactor was renewed every 2 weeks (about 1600 injections) because more precise results were obtained. The method gave satisfactorily precise and reproducible results; for serum containing 52.1pM HB, the withinday RSD was 0.95% and day-to-day RSD 1.3%. A serum of known HB concentration was supplemented with HB to give final concentration of 0.069 to 4.2mM. The recoveries were in the range 96-104%, as shown in Table 1. The results (n = 11, from 35@4 to 3.lmM) were compared with those obtained by manual spectrophotometric method3 with soluble HBDH. The calculated linear regression and correlation coefficient were y = 0.992x + 0.21 and r = 0.993, respectively. The flow-injection system with immobilized HBDH reactor and fluorescence detection is useful for the sensitive and reliable measurement of HB and can easily be used for analysis of serum. The HBDH immobilized on polymer beads is stable enough to permit the measurement of more than 1600 samples.

REFERENCES D. H. Williamson,

J. Mellanby

and H. A. Krebs,

Biochem. J., 1962, 82, 90. N. F. Nuwayhid. G. F. Johnson and R. D. Feld, C/in.

Chem., 1988, 34, 1790.

Application

P. A. Rue11and G. C. Gass. Ann. Clin. Biochem.. 1991,

This method was applied to the determination of HB in serum.

zs, 183. G. Palleschi, H. S. Rathore and M. Mascini. Anal. Chim. Acra, 1988, ur), 223.

1586

N. KIBA er

5. C. J. McNeil, J. A. Spoors, J. M. Cooper, K. G. M. M. Alberti and W. H. Mullen, Anal. Chim. Acta, 1990,237, 99. 6. H. A. Krebs, J. Mellanby and D. H. Williamson, Biochem. J., 1962, 82, 96.

7. A. Brashear and G. A. Cook, 478.

Anal. Biochem., 1983,131,

al.

8. P. W. Carr and L. D. Bowers,

Immobilized

Enzymes in

Wiley, New York, 1980. 9. E. P. Plueddemann, Silane Coupling Agents, pp. 227-230. Plenum Press, New York, 1982. 10. N. Kiba, Y. Inoue and M. Furusawa, Anal. Chim. Acra, Analytical

and Clinical

1991, 243, 183.

Chemistry,

pp. 176-I 80.