Synthesis of several 2-substituted 3-(p-hydroxyphenyl)-1-propenes and their characterization as mechanism-based inhibitors of dopamine beta-hydroxylase.

Three substrate analogs of dopamine beta-hydroxylase, viz. 2-X-3-(p-hydroxyphenyl)-1- propenes (where X = Br, Cl, H), have been synthesized, and all behave as substrates requiring O2 and ascorbate for the enzyme-catalyzed hydroxylation reaction. The products have been characterized by mass spectrometry as the respective 2-X-3-hydroxy-3-(p-hydroxyphenyl)-1- propenes . The relative kcat values for these compounds at pH 5.5, 0.25 mM O2 are 49 min-1 (2-H), 8.6 min-1 (2-Cl), and 7.0 min-1 (2-Br). All three compounds have the characteristics of mechanism-based inhibitors of dopamine beta-hydroxylase since incubation of enzyme with these compounds under turnover conditions leads to a time-dependent loss of activity. The kinact values at pH 5.5, 0.25 mM O2 are 0.08, 0.20, and 0.51 min-1, respectively, for the 2-Br-, 2-Cl-, and 2-H-substituted analogs. No reactivation was observed after exhaustive dialysis of enzyme inactivated by 2-Br-3-(p-hydroxyphenyl)-1-propene, suggesting irreversible inactivation of dopamine beta-hydroxylase.

Dopamine 0-hydroxylase (EC 1.14.17.1) is a copper-containing monooxygenase involved in the biosynthesis of norepinephrine (Kaufman and Friedman, 1965;Skotland and Ljones, 1979;Rosenberg and Lovenberg, 1980), a neurotransmitter whose physiological level is critical for the normal functioning of the central nervous system (Molinoff and Axelrod, 1971). Thus, dopamine @-hydroxylase is an interesting target for the development of mechanism-based inactivators (Rando, 1977;Abeles and Maycock, 1976;Walsh, 1977Walsh, ,1982. In the presence of molecular oxygen and ascorbate as electron donor, dopamine @-hydroxylase catalyzes in vitro the conversion of several substituted 0-phenethylamines to phenethanolamines. Recent investigations have focused on a number of structural analogs of dopamine which, in addition to being substrates, behave as mechanism-based or suicide inhibitors of this enzyme (Baldoni and Villafranca, 1980;Colombo et al., 1984a;Klinman and Krueger, 1982;May et al., 1983). These analogs can provide information pertinent to the bond making and breaking steps in the enzymatic reaction, can Grant GM-29139. The costs of publication of this article were de-* This work was supported in part by National Institutes of Health frayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. f National Research Service Awardee supported by Grant GM-09195.
3 Fellow of the Jane Coffin Childs Memorial Fund for Medical Research. ll Established Investigator of the American Heart Association. To whom correspondence should be addressed. serve as useful tools to study structure-function relationships in catalysis, and also help to identify amino acid residues at the active site of the enzyme.
Previous work from our laboratory showed that p-hydroxybenzyl cyanide specifically inactivates dopamine @-hydroxylase (Baldoni and Villafranca, 1980). Recent experiments demonstrate that several ring-substituted benzyl cyanides also inactivate dopamine P-hydroxylase in a mechanismbased fashion (Colombo et aL, 1984a). The proposed mechanism involves a tight binding complex and/or covalent adduct formation as a result of activation of an enzyme-bound species produced during catalysis (Colombo et ul., 1984b). Our present report deals with several substrate analogs that contain a potential allene or allylic radical functionality as the latent reactive group which could be unraveled during catalysis. This electrophilic moiety could react with an amino acid residue a t the active site resulting in alkylation. Data in this paper show that 3-~-hydroxyphenyl)-l-propene (VI) and the corresponding 2-bromo and 2-chloro congeners are substrates of dopamine P-hydroxylase (Equation 1) and have the characteristics of mechanism-based inhibitors of this enzyme. We report here the synthesis of these compounds and their kinetic data as substrates.  ylation reaction catalyzed by dopamine P-hydroxylase, we have synthesized propenyl substrate analogs that also have the potential of being mechanism-based inhibitors of this enzyme. The preparation and characterization of compounds I, 111, and V are given in Miniprint.
Product Identification-If dopamine @-hydroxylase catalyzes the benzylic hydroxylation of I, the expected product would be 11. Initially, two separate experiments were performed at pH 5.5 in the presence of 02, ascorbate, and I. One reaction mixture contained dopamine P-hydroxylase and the other did not. These separate reaction mixtures were incubated with 100% O2 for 30 min, extracted with ether, concentrated, and subjected to mass spectrometric analysis (Fig. 1). The mass spectra obtained from both experiments show that I is indeed converted to I1 only in the presence of the enzyme. Similar results were obtained with 2-C1-3-(p-hydroxylphen-y1)-1-propene (111) and 3-(p-hydroxyphenyl)-l-propene (V), i.e. 111 is converted to IV, and V is converted to VI.

Steady State Kinetics with 2-Br-3-(p-hydroxyphenyl)-l-pro-
pene-In initial velocity experiments at pH 5.0 and 37 "C, I is a linear competitive inhibitor versus tyramine with a K, value of 4.9 mM (Fig. 2).
When 1 was incubated with 0 2 , ascorbate, and enzyme, a small rate of oxygen uptake was detected employing the oxygraph assay, indicating that I is a very poor substrate for dopamine P-hydroxylase. The low sensitivity of this assay did not permit reliable evaluation of kinetic constants. Thus, to quantitate the amount of product formed and to determine the steady state kinetic parameters for the dopamine P-hydroxylase-catalyzed hydroxylation of I, an HPLC method was developed. A typical HPLC analysis of a reaction mixture, as described in Miniprint, showed three elution peaks. The peaks with retention times of 6.6 and 2.6 min corresponded, respectively, to the substrate I and product 11. If the enzyme was omitted from the assay mixture, no peak at 2.6 min was observed. The third peak with a retention time of 2 min contained the remaining components of the assay mixture.
Kinetic data obtained under identical conditions (25 "C, pH 5.5, 100% 0,) for the dopamine P-hydroxylase-catalyzed oxidation of compound I and of tyramine showed that both I and tyramine have similar K,,, values (5.9 and 4.3 mM, respec- tively). However, I is processed at a rate which is only 1.1% that of tyramine (V, = 0.19 and 17.3 Fmol of Oz/min/mg, respectively). It should be kept in mind that the actual values of V , depend on the magnitude of the K, for oxygen. The data above were obtained at a single oxygen concentration (1.21 mM). The K,,, and kcat values for compounds I, 111, and V were obtained at 0.25 mM 0, under identical conditions for a direct comparison and are presented in Table I.
Inhibition of Dopamine P-Hydroxylase by Compounds I, III, and V-Experiments were conducted to evaluate whether all three 3-(p-hydroxyphenyl)-l-propenes were mechanismbased inhibitors of dopamine &hydroxylase. This was the case as established by several criteria. 1) Inactivation was first order and displayed saturation kinetics (Fig. 3 for compound I). The value for kinact, the rate of inactivation at infinite inhibitor concentration, was evaluated from a plot of l/koh versus l/inhibitor (Fig. 3, inset) to be 0.09 rnin". The value for the dissociation constant of the enzyme inhibitor complex calculated from this plot is 4.9 mM. The inactivation of dopamine &hydroxylase by compounds I11 and V also followed first order kinetics. The kin,& and K d values for all  Table 1. Thus, the three substrate analogs inactivate dopamine @-hydroxylase in a manner consistent with the following model:

TABLE I Kinetic parameters for 2-X-3-(p-hydroxyphenyl)-l-propenes as substrates and mechanism-based inhibitors of dopamine j3-
2) Inactivation required the presence of the other two substrates, i.e. oxygen and ascorbate (Table 11, Table I1 clearly show that inactivation is strictly dependent on catalysis. Inspection of the data in Table I1 (incubation conditions 4 and 5 ) also shows that the observed rate of inactivation by I increased about 4-fold when the concentration of oxygen was increased from 21 to 100%. This indicates that, as was the case with the benzyl cyanide analogs (Colombo et al., 1984a), the value for the K,,, for O2 is higher when compound I is the substrate than when tyramine is the substrate. Because inactivation follows saturation kinetics, i.e. a reversible complex is formed prior to the first order inactivation step, the true value of the K,,, for O2 with I was determined from incubation experiments. Fig. 4 shows that at saturating ascorbate, a plot of Ilkob. versus 1411 at three different levels of O2 gives a set of straight lines intersecting to the left of the vertical axis and above the horizontal axis. This pattern is similar to one expected for a sequential mechanism in initial velocity experiments. The K,,, for O2 and the maximal rate of inactivation were 2.7 mM and 0.77 min", respectively, as calculated from the intercept replot (Fig. 4, inset).
Partition between Inactivation and Turnover-The kinetic constants for inactivation by all three compounds are given in Table I. In addition, the steady state kinetic parameters for these compounds were determined under the same conditions as were used for inactivation, allowing partition coefficients (k,,JkinaCt) to be calculated. The previous kinetic model accounts for this partitioning. The partition coefficients were also determined directly by measuring the amount of product formed after complete inactivation of the enzyme using HPLC to analyze for the product (Table I, Method B). These numbers are given in Table I   107 "Dopamine P-hydroxylase (250 pg/ml) was incubated at room temperature in acetate/MES (0.1 M each) buffer, pH 5.5, containing 20 mM fumarate, 14% DMF, and 156 wg/ml of catalase. As indicated in the left column, the incubation solutions were equilibrated with 21% 02, 100% 02, 100% N,, or 100% CO. When present, the concentration of ascorbate was 38 mM and the concentrations of the analogs were 4.5 mM (compound I) and 0.33 mM (compound V). The value for &b was determined from semilog plots of activity uersccs time. -1.0 -0.8 -0.6 -0.4 -0.2  0.2 Q4 0.6 0.8 1.0 1.

I/[lnhibitor]
FIG. 4. Oxygen dependence of the inactivation of dopamine @-hydroxylase with I. Dopamine @-hydroxylase (0.12 mg/ml, 11.4 Cu/tetramer) was incubated at room temperature at the indicated oxygen saturation in 110 mM MES, pH 5.5, containing 14% DMF, 156 Fg/ml of catalase, and 38 mM ascorbate. The concentration of I was varied from 0.9 to 5.1 mM. At intervals, aliquots were removed and assayed for dopamine 8-hydroxylase activity in the 37 "C standard assay at pH 5.0 as indicated under "Experimental Procedures." At each concentration of I, the value of &was obtained from semilog plots of residual enzyme activity uers'sus time.

TABLE rrr
Irreversibility of dopamine &hydroxylase inactivation by 2-Br-3-(phydroxyphenyll-1 -propene Before dialysis After dialysis that there is one active site/dopamine @-hydroxylase monomer. Irreversibility of the Inhibition of Dopamine @-Hydroxyluse by Compound I-Dopamine @-hydroxylase inactivated in the presence of 0 2 , ascorbate, and 4.5 mM compound I did not regain significant activity when dialyzed against a 500-fold excess of buffer (four changes) for 48 h (Table 111). These results indicate that compound I causes a time-dependent, first, order irreversible inactivation of the enzyme.

DISCUSSION
The data presented in this paper establish that the newly synthesized compounds I, 111, and V are alternate substrates for dopamine 0-hydroxylase and are also inhibitors of the enzyme during catalysis (suicide substrates).
Steady state kinetic constants obtained a t a single oxygen concentration showed that the kcat for 3-(p-hydroxyphenyl)-1-propene is -6 times higher than that of the 243-and 2-Brsubstituted analogs (Table I). The V/K values are, respectively, 50.5, 2.2, and 3.9 min-' m"' for compounds V, 111, and I. In initial velocity studies at pH 5.0 (Fig. 2), I was found to be a competitive inhibitor versus tyramine. When assayed under identical conditions (100% 02), the 2-Br-substituted analog was hydroxylated at a rate much slower than tyramine (relative V , < 1.1%). The true V, for I is probably higher inasmuch as the true K , for O2 estimated from inactivation experiments is 2.7 mM (Fig. 4). In this respect, compound I behaves similarly to p-hydroxybenzyl cyanide (Colombo et al., 1984a) in that the K , value for O2 is 2.8 mM with this compound.
In incubation experiments, the interaction of 3-(p-hydrox-ypheny1)-1-propene and of the 2431-and 2-Br-substituted compounds with dopamine @-hydroxylase had the characteristics of a mechanism-based inhibitor (Rando, 1974;Abeles and Maycock, 1976;Walsh, 1982). The rate of inactivation by these compounds was strictly dependent on catalysis (Table  11) and followed saturation kinetics (Fig. 3). For each analog, the time-dependent loss of activity was first order for several half-lives, indicating that the processing of compounds I, 111, and V is accompanied by irreversible enzyme inactivation. Also, the partition ratio for these compounds ranged from -40 to -100 (Table I). Last, no recovery of enzyme activity occurred after prolonged dialysis of dopamine P-hydroxylase inactivated by incubation with I under turnover conditions (Table 111).
The requirement for all the substrates (ascorbate and 02) indicates that dopamine 0-hydroxylase catalyzes the first step or steps in the oxidation of compounds I, 111, or V prior to inactivation. An inactivation mechanism involving an activated intermediate should account for the fact that compound V, the unsubstituted analog, is also a suicide substrate of the enzyme. This observation rules out a mechanism in which an electrophilic allene is generated prior to product release from the enzyme-product complex by the abstraction of the benzylic proton followed by halogen (Br-or C1-) elimination. Several other mechanisms for the inactivation reaction can be postulated. Kinetic data for the dopamine P-hydroxylasecatalyzed hydroxylation of ring-substituted phenethylamines and benzyl cyanides analyzed as a function of the a substituent constant using the Hammett equation suggest a radical-cation intermediate in the enzyme-catalyzed hydroxylation rea~tion.~ This mechanism is also reasonable for the B. Rajashekhar and J. J. Villafranca, unpublished results. benzylic hydroxylation of compounds I, 111, and V. Thus, following initial enzyme reduction, binding of substrate and 0 2 occurs as demonstrated by the kinetic results of Miller and Klinman (1983). Recent data from our laboratory demonstrate that maximum activity occurs with a stoichiometry of two copper ions at the active site (Ash et al., 1984). With 0;bound between two Cu2+, removal of one electron from bound substrate would yield a cation radical which could ultimately rearrange to an enzyme-bound intermediate that could partition between formation of product and inactivation. The inactivation may result from attack by an active site base, resulting in formation of a covalent enzyme adduct. The mechanism of inactivation is still unknown, and experiments designed to further explore the inactivation reaction and the enzyme residues modified during the inactivation reaction are underway in this laboratory.