+ All Categories
Home > Documents > Calorimetric Phe Analysis

Calorimetric Phe Analysis

Date post: 04-Jun-2018
Category:
Upload: ajr266
View: 245 times
Download: 0 times
Share this document with a friend

of 4

Transcript
  • 8/13/2019 Calorimetric Phe Analysis

    1/4

    ANALYIlCAI,BIOC HEMISLRY 180,91-94 (1989)

    Monitoring of Phenylketonuria: A Calorimetric Methodfor the Determination of Plasma PhenylalanineUsing L-Phenylalanine DehydrogenaseUdo Wendel,* Werner Hummel,? and Ulrich Langenbeckz*Kinderkl inik, Universi t i it Dusseldorf, D-4000 Di isseldorf, Federal Republ ic of Germany; t lnsti tut fiir Enzymtechnologie,Unive rsittit Diisseld orf, in der KFA Jiil ich, D-51 70 Jii l ich, Federal R epublic of Germany; and Insti tut fiir Humangenetik,Kl inikum der Universi t i it Frankfurt/M ., D-6000 Frankfurt 70, Federal Republ ic of Germany

    Received November 4,1988

    A simple, rapid, accurate, and precise calor imetr icass ay for the determ ination of L-phenylalanine inplasma samples using t-phenylalanine dehydroge-nase [L-phenylalanine:NAD+-oxidoreductase (deami-nating)] from Rhodo coccus sp. M 4 is descr ibed. Theenzyme catalyzes the NAD-dependent oxidative deami-nation of L-phenylalanine. How ever, the equilibrium ofreaction favors L-phenylalanine formation. By stoi-chiom etric coupling of this reaction with diaphoraseiio-doni tro tetrazol ium chloride ( INT) the formed NADHconverts INT to a formazan whereby the reaction is dis-placed in favor of phenylpyruvate. Using a kinetic ap-proach the increase in absorbance at 492 nm shows l in-ear i ty over more than 30 min. Deproteinized standardsolutions of L-phenylalanine in the range from 30 to1200 Kmol/ l iter show a l inear ity between the dAformaza J30 min and the substrate concentration. In phenylketo-nur ia (PKU) plasma samples no inter ferences caused byL-tyrosine or phen ylpyruvic acid are seen . Applicabil-i ty is demonstrated by comparative determination ofplasma L-phenylalanine of treated PKU patients by thecalor imetr ic method and automated amino acidanalysis. (0 1999 Academic Press, Inc.

    Severa l methods have been reported in the literaturefor the quantitative determination of L-phenylalanine inphysiological fluids, an important parameter in diagno-sis and therapy of disorders of phenylalanine catabo-lism. The Guthrie test (l), which is used worldwide as asemiquantitative test in newborn screening and occa-siona lly also in therapy control in patients with hyper-phenylalaninemias, is of low precision. Blood phenylala-nine leve ls in PKU patients on low-phenylalanine diets

    1 Abbreviations used: PKU, phenylketonuria; INT, iodonitro tetra-zolium chloride; PheDH, L-phenylalanine dehydrogenase.0003.2697/89 3.00Copyright icj 1989 by Academic Press, Inc.All rights of reproduction in any form reserved.

    are most commonly monitored by spectrofluorometricmethods (2,3) or by column chromatography usingamino acid analyzers (4). Spectrophotometric methodsfor quantitative estimation of L-phenylalanine usingsnake-venom L-amino acid oxidase (5) phenylalanineammonia lyase (6), L-phenylalanine oxidase (7), or de-rivative spectrophotometry (8) are not routinely applied.Recently Hummel et al. (9) reported on an optical ki-netic test for the determination of L-phenylalanine byuse of a newly detected NAD(H)-dependent PheDHfrom a strain of Brevibacterium sp.Departing from the latter enzymatic assay we devel-oped a calorimetric method to determine L-phenylala-nine in plasma or serum. Our method couples simulta-neously the reaction of a NAD(H)-dependent PheDH,an enzyme that catalyzes the oxidative deamination ofL-phenylalanine to phenylpyruvate, and a second reac-tion in which initia lly formed NADH and diaphoraseconvert INT to a formazan. This product is measured inthe visible range at 492 nm. The catalyzed reactions are

    PheDHL-phenylalanine + NAD+ + H,O ti phenylpyruvate+ NADH + NH: 111

    NADH + INT + H+ diap? NAD+ + formazan [2]This simple, rapid , accurate, and precise photometricmethod can be applied for clin ical routine and gives re-sults identical to those of automated amino acid anal-ysis.MAT ER IALS AN D MET H OD SReagents

    All chemica ls were purchased from Merck (Darm-stadt, FRG) or Boehringer (Mannheim, FRG). Standardserum (Prec inorm S) was from Boehringer.91

  • 8/13/2019 Calorimetric Phe Analysis

    2/4

    92 WENDEL, HUMMEL, AND LANGENBECKTABLE 1

    Substrate Speci f ic i ty and Kinetic Parameters of Part iallyPur i fied L-PheDH from Rhodoco ccus sp. M 4 (10)

    Oxidative deamination (pH 9.5)Substrate V max Km [M l

    L-Phenylalanine 100 7.5 x 1om4L-Tyrosine 12 3.1 x 1om3

    The reactions were carried out at 21C and followedby measurement of the absorption of NADH at 340 nm.Enzyme activity is expressed in International Units asthe consumption or production of 1 ymol NADH/minusing a molar extinction coefficient of 6.22 X lo3 litermol- cm- . Activity of oxidative deamination was foundabout 10 per cent that of reductive amination. In the textPheDH activity is given as reductive aminating activity,throughout.

    Note. V,,, is given in relation to the value for L-phenylalanine. K,,,(NAD), 2.7 X 10m4M; K, (phenylpyruvate), 7.0 X 10. M.

    For the enzymatic determination of L-phenylalaninewe used solution 1 (potassium phosphate, 25 mmol/li-ter/triethanolamine 0.2 mol/liter buffer, pH 8.6; TritonX-100, 13.2g/liter), solution 2 (diaphorase from pigheart/NAD; stabil izers), and solution 3 (INT) of the testcombination for the calorim etric determination of L-glu-tamic acid (Cat. No. 139 092, Boehringer).Enzyme Preparation

    PheDH [L-phenylalanine:NAD+-oxidoreductase (de-aminating)] was prepared from a culture of Rhodococcussp. strain M 4 grown aerobically in a bioreactor by theprocedure of Hummel et al . (10). The crude extract ofthe cells containing 25-30 U/mg of PheDH activi ty waspurified further by a two-step procedure (liquid-liquidextraction; DEAE-cel lulose chromatography) to give aspeci fic activ ity of more than 1000 U/mg (as describedby Hummel et al. (10). Less purified enzyme prepara-tions are also suitable, provided the enzyme inducer L-phenylalanine is completely removed. Substrate specifityof PheDH for oxidative deamination reaction is given inthe Table 1.This partially purified enzyme preparation stored at4C at a concentration of 3.0 mg of protein/ml (50 U/ml) in 0.2 M potassium phosphate buffer containing be-taine (pH 9.5) demonstrated about 35 loss of activi tyin 8 months. Dilutions were made with potassium phos-phate/triethanolamine buffer to give approximately 0.05U/1.0 ml of assay mixture for measurements of stan-dards and unknowns.PheDH Assay

    The mixture for the assay of reductive amination con-tained 738 mmol/liter (NH&SO4 buffer, pH 8.5, 0.1mmol/liter NADH, 1.0 mmol/liter phenylpyruvate, andlimiting amounts of the enzyme preparation. Oxidativedeamination was measured in potassium phosphate (8.3mmol/liter)/triethanolamine (67 mmol/liter) buffer, pH8.6, with Triton X-100 (4.4 g/liter) containing 1 mmol/liter NAD+, 2 mmol/liter L-phenylalanine, and limitingamounts of the enzyme preparation.

    Automated Amino Acid AnalysisPlasma samples and standards (dilution sequences ofphenylalanine dissolved in Precinorm S) were depro-

    teinized with 4 vol of 3.0 sulfosalicylic acid in waterand the mixture was centrifuged. Determinations of L-phenylalanine were performed on a Biotron ic 6001amino acid analyzer (Biotron ic, Munich) using a shortprogram with the following details: column 0.4 X 30 cmfilled with BTC 2710 (Biotronic) resin; column tempera-ture, 46C; buffer flow, 20 ml/h. Buffers, 0.47 N lithiumcitrate buffer (pH 3.3) and 0.50 N lithium citrate buffer(pH 4.1).Phenylalanine retention time was 36 min. The con-centrations were quantified with a Shimadzu CR 3A dig-ital integrator (Kyoto, Japan) by comparison to a cal-ibrated commercia l standard (Benson H-standard,Biotron ic). The sample volume applied to the columnwas 100 ~1. The protein-free supernatant was applied di-rectly to the sample collector or stored frozen (-20C)for a short time until analys is.Enzymatic Determination of L-Phenylalanine

    Plasma samples and standards (phenylalanine dis-solved in Precinorm S) were deproteinized with an equalvolume of HC104 (1.6 mol/liter) and centrifuged and theacid extract was neutralized by the addition of an appro-priate volume of KHC03 solution (6 mol/liter). Aftercentrifugation, 0.1 ml of the supernatant was insertedinto the assay. The reaction mixture (1.0 ml) consistedof 217 ~1 of solution 1 (potassium phosphate/triethanol-amine buffer, pH 8.6), 433 ~1 redistilled water, 100 ~1 ofsolution 2 (NAD/diaphorase), 100 ~1 of solution 3 (INT)of the test combination Cat. No. 139 092 (Boehringer),and 100 ~1 of the deproteinized sample with a final L-phenylalanine concentration of 1.5-60 pmol/liter. Finalconcentrations in the assay were for triethanolamine 43mmol/liter, potassium phosphate 5.4 mmol/liter, TritonX-100 2.87 g/liter, NAD 0.75 mmol/liter, INT 0.08mmol/liter, and diaphorase 0.15 U. The reaction wasstarted by addition of 50 ~1 PheDH solution. In general,enzyme activity amounted to 0.05 U/assay. Incubationwas at 21C. Increase of absorbance at 492 nm was mea-sured after exactly 30 min using a Beckman DU-8 spec-trophotometer. Appropriate blanks and calibrationstandards were likew ise prepared each day using stan-

  • 8/13/2019 Calorimetric Phe Analysis

    3/4

    COLORIMETRIC ASSA Y FOR PLASM A L-PHENYLALANINE 93

    I I I30 45 60t ime ot incubat ion Im inI

    FIG. 1. Time depende nce of absorbance (A 492 nm) for assays con-taining different amounts of L-phenylalanine. L-Phenylalanine con-centrations in standard serum were adjusted as indicate d. Sample swere processed and enzymatic analysis with 0.1 ml serum extract wasperformed as given under Materials and Methods. After addition ofenzyme (0.05 U/assay) absorbance at 492 nm was recorded againsttime. The amounts of L-phenylalanine in the assay solution (1.0 ml)are given in nanomo les (7.5 nmol/l .O ml assay solution correspond to0.15 mmol/liter or 2.5 mg/dl L-phenylalanine in the serum sample).Absorbance (dA/30 min) for blanks (assays without L-pheny lalanine)was constantly below 0.01.

    dard serum (Precinorm S) containing defined amountsof L-phenylalanine.Calculations

    Plasma concentrations of L-phenylalanine were esti-mated by the use of appropriate calibration curves. Forthe determination of recovery , standard serum wasspiked with L-phenylalanine to give concentrations of0.06 and 0.48 mmol/liter, respectively. Results are pre-sented as means +- SD of n determinations. Linear re-gression (least-square method) was used for statisticalanalysis.RESULTSEnzymatic Determination of L-Phenylalanine

    An endpoint method was not feasible because of thelimited amount of the PheDH enzyme. We thereforechose a kinetic approach and measured in the linearrange of the absorbance change at 492 nm.The changes in absorbance were linear with reactiontime over more than 30 min, at which time about 30 ofphenylalanine was converted. A linear relationship wasalso obtained between the increase in absorbance andthe phenylalanine concentration over a range from 1.5to 60 nmol/ml assay corresponding to 30 to 1200 pmol/liter (0.5 to 20 mg/dl) in the serum sample (Fig. 1). This

    is the range of phenylalanine concentrations expectedin the plasma of healthy individuals and treated PKUpatients.The within-run coefficients of variation (four samplesanalyzed seven times in succession) for 3,9, 30, and 60nmol/ml assay, respectively, were 4.0,4.3,2.7, and 2.3 .The between-run coefficients (seven determinationsover 2 weeks) for 5, 22, and 42 nmol/ml assay were 5.5,4.5, and 3.4 , respectively. Recovery of L-phenylalaninefrom sp iked serum was 103 ? 10 (serum concentration,0.06 mmol/liter) and 98 5 5 (serum concentration,0.48 mmol/liter), respec tively.Measurements of a dilution sequence of L-tyrosine instandard serum up to a concentration of 0.55 mmol/literdid not yield any change in absorbance compared tothose in the blank. Addition of different amounts of L-tyrosine up to 30 nmol or L-methionine up to 25 nmol inthe assay (containing 1,30, and 60 nmol phenylalanine/ml) did not influence at all the absorbance brought aboutby L-phenylalanine.Automated Amino Acid Analysis

    Using a short program for the determination of L-phe-nylalanine the coefficients of variation within and be-tween runs (four standard concentrations from 0.12 to1.2 mmol/liter, each sample analyzed five times) werebelow 3 and 5 , respec tively. Recovery of L-phenylala-nine from spiked serum was 103 ?I 4 (n = 12; concen-tration, 0.48 mmol/liter).

    .

    .i

    .rI I I I0 03 0.6 0.9 12

    L- phe cone I by colonmetric assay, mmo l/ I IFIG. 2. Correlation between the phenylalanine values obtained byautoma ted amin o acid analysis (y-axis) and the calorimetric determi-nation (x-axis) in plasma samples of PKU patie nts under therapy.Dotted line, linear least-square regression (y = 0.020 + 1.013 x; r= 0.998).

  • 8/13/2019 Calorimetric Phe Analysis

    4/4

    94 WENDEL, HUMMEL,Comparison of Methods

    Using automated amino acid analysis and our enzy-matic method, we estimated in parallel the L-phenylala-nine content in 41 plasma samples obtained from 36treated PKU patients. The data are presented in Fig. 2.In the range from 0.03 to 1.28 mmol/liter plasma the cor-relation coefficient was r = 0.998. With automatedamino acid analysis as the dependent variable the re-gression was y = 0.020 + 1.013 X. The intercept is notsignificantly different from zero.DISCUSSION

    We describe an enzymatic calorim etric method for thedetermination of plasma phenylalanine which is basedon the oxidative deamination reaction of PheDH. It wasour aim to develop this assay as a simple, accurate, andprecise method for monitoring PKU patients during d i-etary therapy. Present-day column chromatographicand spectrofluorometric methods are time-consumingand require high-grade instrumentation and sophis ti-cated technical knowledge. By using one of the just re-cently isolated L-phenylalanine dehydrogenases (10-12)and by designing the phenylalanine assay according tothe princip les of commercia l Test Combinations, thefoundation has been laid for quick monitoring and effi-cient counseling of children with PKU in an outpatientclinic environment.The formerly described enzymatic uv method for thedetermination of phenylalanine (9) is based on the in-crease in absorbance of NADH during the first minuteof reaction. In the present assay the NADH is recycledto NAD with diaphorase and INT. Therefore, the equi-librium which normally lies far toward phenylalanine isdisplaced to phenylpyruvic acid and the lineari ty in timeof formation of product is much extended. In addition,the sensitivity of the assay is increased by a factor ofthree, because the product, INT-formazan, has a molarextinction coefficient t of about 20,000 (13).Compared to the enzyme from Breuibacterium sp. (9)the Rhodococcus sp. M 4 enzyme has a much higher K,,,for phenylalanine. This assures first-order kinetics overa broader range of substrate concentrations and is an-other advantage of our calorimetric assay. Furthermore,this method compares favorably with automated aminoacid analys is. There is no indication of any systematicerror in either system and accuracy and precision do notdiffer significantly for both methods.

    Partially purified PheDH from Rhodococcus sp. exhib-its some catalytic activi ty on tyrosine (10). However, be-cause of low V,,,,, and very high K,, the integrated rateequation predicts that l ess than 1 of tyrosine is con-

    AND LANGENBECK

    verted within 30 min under the conditions of our assay.Accord ingly, no interference was observed.Another potential source of interference is endoge-nous phenylpyruvic acid. Its concentration in PKU sera,

    however, is very low: 7 + 6 and 52 + 28 pmol/liter at 500and 1100 Fmol/liter phenylalanine, respec tively (14,15).The 20-fold dilution of the serum samples thereforebrings the concentrations in the test solution well belowthe Ki of 70 pmol/liter. It is for this reason that no sys-tematic deviation is observed in our assay at high phe-nylalanine levels in PKU sera. Possible interferences byascorb ic acid and pyruvate were shown not to be relevantclinically.The detection limit of this method for phenylalanineis of the order of 1.5 to 3 pmol/liter in the assay mixture.This is sufficient also for the quantification of the nor-mal plasma level of phenylalanine. The test can thus beused for the quantitative determination of plasma phe-nylalanine at concentrations ranging from 0.03 to 1.2mmol/liter without any further modification. It shouldbe applicable to the determination of phenylalanine inwhole venous and capil lary blood and in paper-driedblood and may also be amenable to automation.AC KN OWLED GMEN T S

    The authors acknowledge the skillful help of Mrs. E. Diekmann,Mrs. H. Goddertz, Mrs. S. Loslein , and Mr. J. Vernau.

    REFERENCES1.2.3.4.5.

    6.7.8.9.

    10.11.12.13.14.15.

    Guthrie, R., and Susi, A. (1963) Pedia trics 32,338-34 3.McCaman, M. W., and Robins, E. (1962) J. Lab. Clin. Med. 59,885-890.Wong, P. W. K., OFlynn, M. E., andInouye, T. (1964) Clin. C/rem.10,1098-1104.Piez, K. A., and Morris, L. (1960) Anal. Biochem. 1,187-201.La Du, B. M., and Michael, P. J. (1960) J. Lab. Clin. Med. 55,491-496.Shen, R., and Abell, C. W. (1977) Science 198,665-667.Koyama, H. (1984) Clin. Chim. Acta 136,131-136.Kullmann, K.-H., Endres, W., Kirzinger, S., and Schmidt, H.-L.(1982) J. Clin. Chem. Clin. Biochem. 20,181-183.Hummel, W., Schiitte, H., and Kula, M.-R. (1988) Anal. Biochem.170,397-401.Hummel, W., Schiitte, H., Schmidt, E., Wandrey, C., and Kula,M.-R. (1987) Appl. Microbial. Biotechnol. 26,409-416.Asano, Y., and Nakazawa, A. (1985) Agric. Biol. Chem. 49,3631-3632.Humm el, W., Weiss, N., and Kula , M.-R. (1984) Arch. Microbial.137,47-52.Babson, A., and Babson, G. (1973) Clin. Chem. 19,766-769.Langenbeck, U., Behbe hani, A. W., and Luthe, H. (1981) J. Znher-ited Metab. Dis. 4,69-70.Langenbeck, U., and Behbehani, A. W. (1988) in Proceedings,26th SSIEM An nual Symposium, Glasgow.


Recommended