ANALYTICALBIOCHEMISTRY

A Radiometric Jeffrey S. Wiseman’ Merrell

184,55-58

Kynurenine and James

Dow Research Institute,

Received

April

(1990)

Monooxygenase

Assay

S. Nichols

2110 East Galbraith

Road, Cincinnati,

Ohio 45215

14, 1989

Kynurenine 3-monooxygenase is a flavin-dependent monooxygenase that catalyzes the oxidation of L-kynurenine to 3-hydroxy-L-kynurenine in the kynurenine pathway of tryptophan metabolism. The enzyme requires NADH or NADPH as a cofactor. A discontinuous assay that utilizes L-t3H]kynurenine as substrate is described. The assay offers high precision and a wide range of accessible substrate and cofactor concentrations. The assay was used to measure kinetic isotope effects and the stereospecificity of oxidation of the cofactor. Hydride is transferred from the A-side (pro-R) of NADH and NADPH since primary deuterium isotope effects were observed for both cofactors when they were deuterated on the A-side but not on the B-side. The large isotope effect on V,.J& for NADH is sensitive to the concentration of kynurenine, which indicates that NADH can bind before kynurenine. o 1990 Academic

surements. There is also a significant background rate of cofactor oxidation in the absence of substrate, which limits the usable substrate or cofactor concentration and the ability to accurately measure enzyme inhibition. Assay procedures have also been described for measuring oxygen uptake (6) or fluorescently labeling the product (7). Neither combines features of sensitivity over a wide range of substrate concentrations, precision, and specificity when substrate or product analogs are examined as possible inhibitors, however. We, therefore, developed an alternative assay that follows the disappearance of [G-3H]kynurenine. A regenerating system for reduced nicotinamide cofactor is included in the assay. We describe the use of this assay to measure isotope effects in the reaction. EXPERIMENTAL

PROCEDURES

L- Kynurenine, L- 3- hydroxykynurenine, NAD(H), NADP(H), glutathione disulfide, glutathione reductase, yeast and liver alcohol dehydrogenases, glucose-6-phosphate dehydrogenase (Leuconostoc mesenteroides), hexokinase, glucose 6-phosphate, and [ 1-‘HIglucose were Kynurenine 3-monooxygenase (EC 1.14.13.9) cata- obtained from Sigma. Ethanol-d6 was obtained from lyzes the oxidation oft-kynurenine to 3-hydroxy-L-kynAldrich. L-[G-3H]Kynurenine was prepared by nonspeurenine. The monooxygenase is known to be present in cific 3Hz exchange (Amersham Corp.) and was a generthe outer membrane of rat liver mitochondria (1); it re- ous gift from Dr. R. Schwartz. The [G-3H]kynurenine quires FAD as cofactor (a), incorporates oxygen from O2 was purified by chromatography using 0.1% HOAc on into the products (3), and requires either NADH or Cis Bondapak (Waters Associates, 0.8 X 10 cm, lo-pm NADPH as coreductant (4). Although this enzyme has particles). The final radiochemical purity was at least been purified, the pure enzyme is unstable and has not 99% by chromatography. In addition, the fact that only a been useful for kinetic studies (5). We have been inter5% tritium isotope effect was observed for this substrate ested in studying the inhibition of this monooxygenase. (see below) indicates that the substrate is at least 95% These studies limited us to using the more stable, unpukinetically homogeneous, i.e., at least 95% of the labeled rified enzyme, and we required a sensitive, precise assay compound is a substrate for the enzyme. The stock subfor the crude enzyme. The enzyme can be assayed spec- strate concentration was adjusted to 2 &i/ml in 0.1% trophotometrically by following the oxidation of nicotinHOAc and unlabeled L-kynurenine was added to a final amide cofactor (6), but the particulate nature of the concentration of 0.25 mM. The substrate was stored at unpurified enzyme interferes with absorbance mea- -20°C and heated for 2 min at 95°C prior to use to remove an unidentified volatile impurity. Mitochondria from livers of male Sprague-Dawley 1 To whom reprint requests should be addressed at Glaxo Research rats were prepared in 0.25 M sucrose according to stanLaboratories, 5 Moore Drive, Research Triangle Park, NC 27709. Press,

Inc.

0003~2697/90

$3.00

Copyright 0 1990 by Academic Press, Inc. All rights of reproduction in any form reserved.

55

56

WISEMAN

AND

NICHOLS

tivity was quantitated with a Flo-One Beta flow detector with a 2.5 ml flow cell (Radiomatic Instruments and Chemical Co.) using Tru-Count scintillation Auid (TruCount Laboratory Supply Co.) at 3 ml/min. Only two radioactive peaks were observed in the chromatograms, and these were completely resolved. The peaks were the product, L-3-hydroxykynurenine, eluting at 4 min and the substrate eluting at 5.5 min. Ethanol enriched with deuterium is also enriched with tritium. When [2H]ethano1 was used to reduce the nicotinamide cofactors, the peak of contaminating [3H]ethanol overlapped with the product peak and was large enough to interfere with its quantitation. In this case, two columns in series were used (eluant flow rate 4 ml/min) to separate the ethanol .005 ,010 .015 .02 peak from the product peak. L-3-Hydroxykynurenine formation was also monil/[KYNURENINE] (l/uM) tored by a fluorometric procedure. Assays were perFIG. 1. Kinetics of oxidation of kynurenine catalyzed by kynurenformed and quenched as described above. L-3-Hydroxyine 3-monooxygenase at (0) 2000, (A) 100, and (Cl) 50 pM NADPH. kynurenine was measured in the supernatants by reaction with tosyl chloride (7). Fluorescence in the reaction mixtures was compared to that in controls in which enzyme was added after the acid quench. dard procedures with care to prevent contamination with microsomes (8). The mitochondria, 110 mg of proOne unit of kynurenine monooxygenase is defined as tein/ml, were stored at -20°C in small aliquots; the en- the amount of enzyme required to form 1 pmol of product per minute. Units of other enzymes are according to zyme was unstable to repeated freezing and thawing. All enzyme assays were performed at 37°C in 0.1 M the supplier’s specifications. Protein was determined by Tris acetate, pH 8.0, containing 10 mM KCl, and 1 mM the method of Bradford (9) relative to bovine serum alEDTA (3). In addition, 1 mM KCN was added to assays bumin. utilizing NADH as reductant. The normal assay conKinetic parameters for the Michaelis-Menton equatained an NAD(P)H regenerating system consisting of 3 tion were calculated by nonlinear least squares. The mM glucose 6-phosphate and 1 U/ml glucose-6-phos- least-squares fitting routine accepted three independent phate dehydrogenase. L-[G-3H]Kynurenine was present variables so that the data at all substrate and cofactor at 0.04 &i/ml and the total kynurenine concentration concentrations could be combined and fitted simultaneously. was adjusted with unlabeled L-kynurenine. Deuterated nicotinamide cofactors were generated in situ. Deuterium was introduced into the pro-S position RESULTS AND DISCUSSION (B-side) from [l-2H]glucose, 1 mM, in the presence of 5 The chromatographic assay using L-[G-3H]kynurenmM MgC12, 1 mM ATP, 2 U/ml hexokinase, and 2 U/ml ine that was developed for this work allowed assays to glucose-6-phosphate dehydrogenase. Deuterium in the be performed over a wide range of concentrations of subpro-R position (A-side) was introduced from [2H]ethanol, 2%, using 20 U/ml alcohol dehydrogenase (from yeast for NADH and from liver for NADP). For the deTABLE 1 termination of isotope effects, reaction via the labeled Kinetic Parameters for Kynurenine 3-Monooxygenase” cofactor was compared to unlabeled cofactor generated identically except from unlabeled reducing agent. The V enzyme activities in the cofactor reducing systems were vobe = 1 + KJ[Kyn] + &/[NAD(PF] + KabIIKynlIPJADWHl high enough to achieve complete reduction in less than 1 min, compared to assay times of 30 min. NADPH NADH Assays were quenched with 5 ml of acetic acid per mil5.3 f 0.2 3.1+ 0.3 liliter, heated for 2 min at 95”C, and centrifuged at V,, hU/md log*8 180 f 20 15,000g for 1 min to remove protein. An aliquot of the Ka (PM) 53f I 21o-t50 quenched reaction mixture, 0.4 ml, was analyzed by & (PM) 12,500f 1000 50,000f4000 &a (PM)' chromatography on a Cl8 Bondapak column (Waters Associates, lo-pm particles, 0.8 X 10 cm) in 20 mM NaOAc, a Kynurenine was varied from 50 to 250 pM, NADPH was varied pH 5.5, in 2% MeOH at a flow rate of 3 ml/min. Radioac- from 33 to 2000 pM, and NADH was varied from 100 to 1000 PM.

RADIOMETRIC

KYNURENINE

TABLE 2 Isotope Effects for Reaction of Kynurenine 3-Monooxygenase and Isotopically Labeled Nicotinamide Cofactors Kyn (PM)

NADPH (Of)

50 250 50 250 50 250 50 250 50 250

100 100 100 100 -

-

NADH (PM)

Stereochem. of labeling B B

500 500

100 100 1000 1000

B B A A A A A A

k&o 1.20 1.17 0.97 1.02 2.5 4.0 39 8.8 2.2 2.1

+ k It f + f f + + f

0.08 0.06 0.03 0.05 0.2 0.3 8 1.7 0.1 0.2

&rate and nicotinamide cofactors. Representative data (Fig. 1) demonstrates the high precision of the assay. As discussed above, we chose to use mitochondria as the source of enzyme for these assays, and reduced nicotinamide cofactors are unstable in the presence of mitochondria. EDTA was added to the assay buffers to block hydrolysis of NADPH to NADH by alkaline phosphatase. In addition, CN- was added to assays utilizing NADH to block rapid nonspecific oxidation of this cofactor, and a regeneration system was used to compensate for further cyanide-insensitive oxidation of both NADH and NADPH. With these precautions NADPH and NADH concentrations as low as 30 and 50 PM, respectively, could be achieved, with no limit on the highest concentration of cofactor, up to at least 5 mM. There was no observed lower limit on the usable L-kynurenine concentration, but marked nonlinearity with time was observed at concentrations of L-kynurenine above 250 PM, ascribable primarily to product inhibition. This limits the utility of a discontinuous assay such as this at high substrate concentrations. The assay was linear with time up to at least 120 min and linear with protein concentration up to approximately 1 mg/ml. Conveniently, protein concentrations of 0.25-0.5 mg/ml and assay times of 30 min were used. The tritiated substrate was prepared by nonspecific tritium exchange, and the positions of labeling have not been determined. It was thus possible that tritium was introduced into a kinetically sensitive position. When the chromatographic assay was compared to a fluorometric assay for formation of L-3-hydroxykynurenine (7), however, rates of conversion of unlabeled substrate differed from rates for [3H]kynurenine by only a factor of 1.05, which was not statistically significant. The tritiated substrate is, therefore, kinetically equivalent to unlabeled substrate. Kinetic parameters for the monooxygenase with both NADPH and NADH as cofactors are presented in Table

MONOOXYGENASE

ASSAY

57

1. Both the V,,,,, and the K,,, for L-kynurenine (K, in Table 1) varied with the type of cofactor, but the change was small. The kinetics were clearly sequential as opposed to ping-pong with respect to binding of L-kynurenine and nicotinamide cofactor since plots of l/V vs l/NADPH are not parallel at different kynurenine concentrations (Fig. 1). The dependence on oxygen concentration was not examined. Isotope effects for the oxidation of L-kynurenine by kynurenine 3-monooxygenase are shown in Table 2. In order to best mimic the conditions of the normal enzyme assay, isotope effects were measured with deuterated reduced cofactors generated in situ with an NAD(P)H regenerating system. When a regenerating system is used, there is the danger of introducing label on both sides of the cofactor due to multiple turnovers of the cofactor and thus invalidating measurements of stereospecificity. Multiple turnovers could arise either from specific oxidation of kynurenine or from nonspecific background oxidation of the cofactor. Care was taken, therefore, to measure the effects under conditions such that turnover of the cofactor due to all causes was less than 10%. When deuterium was introduced into the B-side of the cofactor there was only a small, presumably secondary, effect for NADPH and no effect for NADH at the substrate concentrations used. In contrast, the large isotope effects for transfer of 2H from the A-side of both of the cofactors clearly indicates that hydride is transferred from this side. This stereochemistry is common to all of the flavoprotein monooxygenases so far characterized that act on aromatic substrates (10-12). With NADH it was possible to use low enough concentrations of cofactor to measure a V,,,,,/K,,, isotope effect. The isotope effect measured at 100 PM NADH is sensitive to the concentration of L-kynurenine. This sensitivity indicates that L-kynurenine can bind after NADH but before reduction of the flavin (13). The result does not distinguish between a compulsory or a random order of binding for L-kynurenine (the distinction could be made only at kynurenine concentrations higher than we can achieve in the present assay). The V,,,/Km isotope effect measured at 50 PM kynurenine and 100 PM NADH is unusually large, 39 f 8. The cause of such a large effect has not been determined in this case, but similar results for other enzymes and their possible sources have been summarized (14,15). REFERENCES 1.

Okamoto, H., Yamamoto, S., Nozaki, M., and Hayaishi, 0. (1967) Biochem.

Biophys.

Res. Commun.

26,309-314.

2. Okamoto, H., and Hayaishi, 0. (1967) Commun.

Biochem.

Biophys.

3. Saito, Y., Hayaishi, O., and Rothberg, S. (1957) J. 229,921-934.

Res.

29,394-399. Biol.

Chem.

58

WISEMAN

4. Okamoto,

H.,

and Hayaishi,

0. (1969)

Arch.

B&hem.

Biophys.

131,603-608. 5. Nishimoto, Y., Takeuchi, F., and Shibata, Y. (1979) J. Chromatogr. 169,357-364. 6. Okamoto, H. (1970) in Methods in Enzymology (Tabor, H., and Tabor, C. W., Eds.), Vol. 17, Part A, pp. 460-463, Academic Press, New York. 7. Watanabe, M., Watanabe, Y., and Okada, M. (1970) C&z. Chim. Acta 27,461-466. 8. Hogaboom, G. H. (1955) in Methods in Enzymology (Colowick, S. P., and Kaplan, N. O., Eds.), Vol. 1, pp. 16-19, Academic Press, New York.

AND

NICHOLS 9. Bradford,

M. M. (1976)

10. Yu, Y., Wang, 1110.

L.-H.,

Anal.

andTu,

11. Reyerson, C. C., Ballou, 21,1144-1151.

B&hem. S.-C.

72,248-254.

(1987)

Biochemistry26,1105-

D. P., and Walsh,

C. (1982)

Biochemistry

12. You, K. (1982) in Methods in Enzymology (Purich, Vol. 87, pp. 101-126, Academic Press, New York. 13. Klinman, J. P., Humphries, Chem. 266,11,648-11,651. 14. Northrup, 15. Klinman,

D. B. (1981) J. P. (1978)

H., and

Annu. Ah.

Voet,

J. G. (1980)

Reu. Biochem.

Enzymol.

D. L., Ed.),

46,415494.

50,103-131.

J. Bid.