Determination of l -phenylalanine in serum by flow-injection analysis using immobilized phenylalanine dehydrogenase and fluorimetric detection

Talanta ELSEVIER

Talanta 44 (1997) 131-134

Short communication

Determination of L-phenylalanine in serum by flow-injection analysis using immobilized phenylalanine dehydrogenase and fluorimetric detection Nobutoshi

Kiba*, Akiko

Itagaki, Motohisa

Furusawa

Department of Applied Chemistry and Biotechnology, Faculty of Engineering, Yamanashi University, KoJu 400, Japan Received 25 March 1996; revised 7 June 1996; accepted 12 June 1996

Abstract

A flow-injection system with an immobilized enzyme reactor is proposed for the determination of L-phenylalanine. Phenylalanine dehydrogenase from Rhodoccus sp. M4 was immobilized on tresylated poly (vinyl alcohol) beads (13 /~m) and packed into a stainless-steel column (5 cm × 4 mm i.d.). Serum sample was deproteinized with tungstic acid and filtered through an ultrafiltration membrane. The sample solution (30 /~l) was injected into the carrier stream (water). The NADH formed was detected at 465 nm (excitation at 340 nrn). The calibration graph was linear for 0.9-600/~m L-phenylalanine; the detection limit was 0.3/zm. The sample throughout was 25 h-1 without carryover. The half-life period of the immobilized enzyme was 23 days.

Keywords: Flow-injection analysis; Fluorescence detection; Immobilized enzyme reactor; L-phenylalanine; Serum

I. Introduction

The measurement of L-phenylalanine (Phe) is of clinical importance in the diagnosis and therapy of phenylketonuria. Enzymatic methods have been developed for the selective determination of Phe based on the conversion of Phe in to phenylpyruvate with concomitant reduction of nicotineamide adenine dinucleotide ( N A D +) to reduced nicotineamide adenine dinucleotide * Corresponding author. Fax: (81)552-208568.

( N A D H ) with phenylalanine dehydrogenase (EC 1.4.1.-., PheDH) [1-8]. P h e D H has been immobilized on a nylon coil and applied in a bioluminescence continuous-flow system for the determination of Phe in serum [9]. The PheDHs used were purified from Thermoactinomyces sp. [1], Brevibacterium sp. [2], Rhodococcus sp. [3-8] and Bacillus badius [9]. A m o n g them, the enzymes from Rhodococcus sp. are more specific for Phe and are stable in the presence of glycerol [10,11]. The P h e D H from Rhodococcus sp. M4 has been immobilized on aminopropyl-substituted controlled-pore glass (CPG) and used as a reactor in

0039-9140/97/$15.00 © 1997 Elsevier Science B.V. All rights reserved PII S0039-9140(96)02016-4

132

N. Kiba et al. / Talanta 44 (1997) 131 134

the fluorimetric flow-injection system for the determination of Phe in serum [12,13]. The method is not very sensitive because the reactor was used in the neutral buffer (pH 7.1) to prolong the lifetime of the reactor; the optimum pH for the PheDH is 10.0 [10,11] and CPG is unstable in alkaline solution [14]. This paper describes a flow-injection system for the enzymatic determination of Phe with PheDH immobilized on poly(vinyl alcohol) beads contained in a packed-bed reactor. The enzyme from Rhodoccus sp. M4 was used. The enzymatic reaction was performed in a basic buffer (pH 10.0). The N A D H produced in the reactor was monitored fluorimetrically. The flow-injection method was applied to the determination of Phe in serum.

2. Experimental 2.1. Materials and reagents

PheDH from Rhodococcus sp. M4 was obtained from Calbiochem (San Diego, CA, USA) with an activity of 12 U ml ~. The activity was measured with Phe as substrate at pH 10.0 at 30°C. N A D + (free acid, 96%) and N A D H (disodium salt, 99.5%) were purchased from Kohjin (Tokyo, Japan) and Boehringer Mannheim (Mannheim, Germany), respectively. L-Amino acids were purchased from Sigma (St. Louis, MO, USA). Poly(vinyl alcohol) beads (GS-520, 13 /tm diameter) were obtained from Showa Denko (Tokyo, Japan). All other chemicals, which were purchased from Nacalai Tesque (Tokyo, Japan), were of analytical-reagent grade. A glycine buffer (pH 10.0) consisting of 0.2 M glycine-0.2 M sodium chloride-0.2 M sodium hydroxide-2% (v/v) glycerol was used. N A D + solution (8 m M N A D + in water) was prepared fresh daily. The carrier solution was water. 2.2. Preparation o f the immobilized enzyme reactor

PheDH was immobilized on the poly(vinyl alcohol) beads. The beads (1 g) were washed with dry acetone (20 ml). The beads were suspended in 20 ml of a mixture of dry acetone and pyridine (1:1 v/v).

With vigorous magnetic stirring, 1 ml of 2,2,2-trifluoroethanesulphonyl chloride (tresyl chloride) was added dropwise to the suspension during 5 min. The reaction was continued for 5 min. The beads were washed with acetone (10 ml) and 1 mM HC1 (20 ml). The beads were packed into a stainless-steel column (5 cm x 4mm i.d.) by the slurry-packing method. Enzyme solution (PheDH 26 U in 10 ml of 2% (v/v) glycerol in 10 mM phosphate buffer) was circulated through the column at 0.2 ml m i n for 6 h at 10°C. During the process the enzyme solution was kept at 2-5°C. The reactor was washed with 10% glycerol in 0.1 M phosphate buffer (pH 7.0) and stored in a refrigerator when not in use. The coupling yield was evaluated by measuring the decrease in the activity of PheDH in the enzyme solution after the process. The PheDH was immobilized with a 95% yield. Also, epoxy-activated beads (epoxy-beads) and glutaraldehyde-activated beads (glutar-beads) were used for the coupling of PheDH; the preparation methods for the activated beads were similar to those described previously [15] and coupling conditions were identical with those described above. The yields for epoxy-beads and glutar-beads were 5% and 24%, respectively. 2.3. Flow system and procedure

A schematic diagram of the flow system is shown in Fig. 1. The system consisted of three piston

ilC

:. . . . . . . . . '

Fig. 1. Schematicdiagram of the flow-injectionsystem for the determination of L-phenylalaninewith an immobilized phenylalanine dehydrogenasereactor. A, 8 mM NAD + solution; B, 2% (v/v) glycerol in 0.2 M glycine buffer; C, carrier solution (water); S, injector with a 30 ~1 loop; MC, mixing coil (100 cm × 0.5 mm i.d.); WB, water-bath thermostated at 20°C; CR, enzyme column reactor (5 cm x 4 mm i.d.); F, spectrofluorimeterwith a flow-throughcell (15/tl); SC, signal cleaner; R, recorder; W, waste. All connectingtubing (0.5 mm i.d.) was made of PTFE.

N. Kiba et al. / Talanta 44 (1997) 131 134

pumps (Hitachi L-6000), an injection valve (Sanuki SVI-6M2) equipped with a 30/~1 loop, a reactor, a spectrofluorimeter (Jasco FP-210) with a flow-through cell (15 /~1) connected to a signal cleaner (SIC SC77) and a recorder (TOA FBR251A). The reactor was maintained at 20°C. Serum (50/ll) was deproteinized by adding 10% (w/v) sodium tungstate solution (100/tl) and 0.06 M sulphuric acid (100 /~1). The mixture was filtered through an ultrafiltration membrane (Advantec Q0100, nominal molecular weight cut-off 10 000) and the filtrate (30 /~1) was injected into the carrier stream (water). The time taken to prepare the sample solution was about 3 min. The present method was compared with liquid chromatography (amino acid analyser, Kyowa Seimitsu K-201; column, TSKgel SCX, 150 × 6.0 mm i.d.; mobile phase, citrate buffer; solvent gradient, post-column derivatization with ninhydrin).

3. Results and discussion

3.1. Reactor performance The effect of pH on the activity of the reactor was studied in the pH range 9.0-11.0 using glycine buffer. The optimum pH was about 10.0. The peak height in the glycine buffer was about 1.2 times that in carbonate buffer at the same pH. Borate buffer consisting of 0.2 M borax-0.1 M NaOH (pH 9.5-11.0) inhibited the enzymatic reaction completely. The temperature dependence of the reactor was investigated over the range 15-30°C. The reactor exhibited maximum activity at 20°C. The effect of NAD + concentration on the activity was studied over the range 1-12 mM at the 600 /~M Phe level. The response increased with increasing concentration, first rapidly and then gradually. Above 6 mM, the response was almost constant. A concentration of 8 mM NAD + was chosen to prevent interference of phenylpyruvate and ammonium ions, which are products of the enzymatic reaction; at this concentration, the NAD + concentration in the reactor is 2.7 mM. The peak height was measured by changing the flow-rates of the glycine buffer, the NAD + solu-

133

tion and the carrier solution (water), keeping the flow-rate ratio of these solutions at 1:1:1. The peak height decreased linearly as the total flowrate increased from 0.3 to 0.9 ml min ~. The peak height at 0.3 ml min-~ was about 2.4 times that at 0.9 ml min -1. A total flow-rate of 0.6 ml min was selected as a compromise between sensitivity and sample throughput; at this flow-rate, the sample throughput was 25 h i. Under the conditions of 8 mM NAD + at pH 10.0 and 20°C, the conversion efficiency of the reactor was 90% immediately after the preparation of the reactor. The efficiency was measured by using NADH. Under the same conditions, the response for Phe (10/~M) was not affected by the presence of 30 /~M phenylpyruvate and 100/~M ammonium ion. Also, no interferences was recorded with 1 mM L-ascorbic acid, 2 mM uric acid and 50 mM glucose. Amino acids normally found in proteins, other than Phe, did not give any response. The operational stability of the reactor was evaluated over 4 weeks. The reactor was used for analyses of 50 samples (20 /~M Phe) for 2 h per day and then washed with 10% (v/v) glycerol in 0.1 M phosphate buffer (pH 10.0) and stored 4°C when not in used. The decrease in activity obeyed first-order kinetics. The kinetic constant was 2.9 x 10 -2 day t; the half-life period of the reactor was 23 days. 3.2. Calibration The calibration graph of peak height against Phe concentration was linear over the range 0.9600/~M with a correlation coefficient of 0.999 (12 data points) under the same conditions as described in section 2.3; since the serum sample was diluted five fold, this method can be applied to the assay of samples containing 5 /zM-3 mM Phe. Below a concentration of 0.9/tM a concave graph was obtained and above a concentration of 600 /~M the graph was convex. The relative standard deviation (RSD) for seven replicate injections of 10.0/~M Phe was 0.65% with a reactor having a conversion efficiency of 88%. By the use of a reactor having a conversion efficiency of 50%, the RSD for the same runs was 1.3%. To obtain precise results, the reactor must be used within a

134

N. Kiba et al. / Talanta 44 (1997) 131-134

Table 1 Recovery of phenylalanine added to pooled seruma Added (/t M)

Recovered (/t M)b

Recovery (%)

5.00 10.0 50.0 100 500 1000 2500

4.87 10.3 49.0 101 500 998 2520

97 103 98 101 100 100 101

Values corrected for Phe (63.6 p M) already present in serum. b All values are means (n = 5). coversion efficiency o f 50%. T h e limit o f detection (signal-to-noise ratio = 3) with a r e a c t o r having a c o n v e r s i o n efficiency o f 80% was 0 . 3 / ~ M (2 ng in a 3 0 / t l injection) with R S D 5.6%. 3.3. Application

This m e t h o d was a p p l i e d to the d e t e r m i n a t i o n o f Phe in serum. P o o l e d h u m a n serum was r e p e a t e d l y a n a l y s e d for 30 d a y with the reactor. T h e r e a c t o r was used for analyses o f 50 samples in a d a y a n d s t a n d a r d s were m e a s u r e d at 25-sample intervals, in o r d e r to correct the v a r i a t i o n o f the c o n v e r s i o n efficiency. T h e r e a c t o r was renewed every l0 d a y s ( a b o u t 500 injections) because m o r e precise results were o b tained. T h e m e t h o d gave satisfactorily precise a n d r e p r o d u c i b l e results; for serum c o n t a i n i n g 62.7 ~tM Phe, the w i t h i n - d a y R S D was 0.88% a n d the d a y - t o - d a y R S D was 1 . 2 ° . Serum o f k n o w n Phe c o n c e n t r a t i o n was supplem e n t e d with Phe to give final c o n c e n t r a t i o n s o f 0 . 0 7 3 - 2 . 5 0 m M . T h e recoveries were in the range range 9 7 - 1 0 3 % , as shown in T a b l e 1. The results ( n - - 2 1 , from 51 to 849 / t M ) were c o m p a r e d with those o b t a i n e d by liquid c h r o m a t r o g r a p h y . The c a l c u l a t e d linear regression a n d c o r r e l a t i o n coefficient were y = 0.997x + 1.58 a n d r = 0.998, respecively.

4. Conclusion The flow-injection system with i m m o b i l i z e d P h e D H r e a c t o r a n d fluorescence d e t e c t i o n is useful for the sensitive a n d reliable m e a s u r e m e n t o f Phe a n d can easily be used for the analysis o f serum. T h e P h e D H i m m o b i l i z e d on p o l y m e r beads is stable e n o u g h to p e r m i t the m e a s u r e m e n t o f m o r e t h a n 500 samples.

References [1] T. Ohshima, H. Sugimoto and K. Soda, Anal. Lett., 21 (1988) 2205. [2] W. Hummel, H. Schutte and M.-R. Kula, Anal. Biochem., 170 (1988) 397. [3] U. Wendel, W. Hummel and U. Langenbeck, Anal. Biochem., 180 (1989) 91. [4] U. Wendel, M. Koppelkamm, W. Hummel, J. Sander and U. Langenbeck, Clin. Chim. Acta, 192 (1990) 165. [5] H. Naruse, Y.Y. Ohshi, A. Tsuji, M. Maeda, K. Nakamura, T. Fujii, A. Yamaguchi, M. Matsumoto and M. Shibata, Screening, 1 (1992) 63. [6] K.C. Dooley, Clin. Biochem., 25 (1992) 271. [7] M.A. Vilaseca, C. Farre and F. Ramon, Clin. Chem., 39 (1993) 129. [8] R. Taylar, I.C. Smith, S.J. Standing, Clin, Chim. Acta, 218 (1993) 207. [9] S. Girotti, E. Ferri, S. Ghini, R. Budini, G. Carrea, R. Bovara, S. Piazzi, R. Merighi and A. Roda, Talanta, 40 (1993) 425. [10] W. Hummel, H. Schutte, E. Schmidt, C. Wandrey and M.-R. Kula, Appl. Microbiol. Biotechnol., 26 (1987) 409. [11] H. Misono, J. Yonezawa, S. Nagata and S. Nagasaki, J. Bacteriol. 171 (1989)30. [12] W. Hummel, S. Tauschensky, U. Spohn, U. Wendel and U. Langenbeck, in R.D. Schmid and F. Scheller (Eds.), Biosensors: Application in Medicine, Environmental Protection and Process Control, GBF Monographs, Vol. 13. VCH, Weinheim, 1989, p. 313. [13] W. Hummel and U. Wendel, in R.D. Schmid (Ed.), Flow Injection Analysis (FIA) Based on Enzymes or Antibodies, GBF Monographs, Vol. 14, VCH, Weinheim, 1991, p. 135. [14] E.P. Plueddemann, Silane Coupling Agents, Plenum Press, New York, 1982, p. 225. [15] N. Kiba, A. Kato and M. Furusawa, Anal. Chim. Acta, 311 (1995) 71.