ARCHIVES
OF BIOCHEMISTRY
AND
BIOPHYSICS
Vol. 288, No. 2, August 1, pp. 427-434, 1991
Effect of pH on the Oxidation Catalyzed by Tyrosinase
Pathway of Dopamine
Manuela Garcia-Moreno, * Jo& Neptuno Rodriguez-L6pez, Ram6n Var&r, * and Francisco Garcia-C%novas$gl
* Francisco
Martinez-Ortiz,?
Jo& Tudela,$
*Ca’tedra de Quimica-Fisica, E. U. Polikknica, Universidad de Castilla-La Mancha, Albacete, Spain; t Departamento de QuimicaFkica, Facultad de Ciencias Quimicas, Universidad de Murcia, Murcia, Spain; and $ Departamento de Bioquimica y Biologia Molecular, Facultad de Biologia, Universidad de Murcia, E-30100 Espinardo, Murcia, Spain
Received October 26, 1990, and in revised form March 12, 1991
The oxidation of 3,4-dihydroxyphenylethylamine (dopamine) by O2 catalyzed by tyrosinase yields 4-(2aminoethyl) - 1,2-benzoquinone (o-dopaminequinone) , which evolves nonenzymatically through two branches or sequences of reactions, whose respective operations are determined by the pH of the medium. The cyclization branch of o-dopaminequinone takes place in the entire range of pH and is the only significant branch at pH > 6. The hydroxylation branch of o-dopaminequinone only operates significantly at pH < 6, and involves the accumulation of 2,4,5-trihydroxyphenylethylamine (6-hydroxydopamine) and 5- (2-aminoethyl) -2-hydroxy1,4benzoquinone (p-topaminequinone) , identified from cyclic voltammetry assays. The kinetic characterization of the hydroxylation branch of o-dopaminequinone has been carried out by spectrophotometric and oxymetric assays. The successful fitting of data to the kinetic behavior predicted by the kinetic analysis at both pH > 6 and pH < 6 confirms the overall oxidation pathway proposed for the dopamine oxidation catalyzed by tyrosinase. The antitumoral power of dopamine is possibly enhanced by the high cytotoxicity of 6-hydroxydopamine and ptopaminequinone, accumulated at the acidic pH characteristic of melanosomes and melanome Cells. 0 1991 Academic Press,
Inc.
The existence of a minor route of dopamine oxidation by tyrosinase has been established ( 1). Tyrosinase is a copper protein widely distributed in the phylogenetic scale ( 2-3 ) . It has a bifunctional activity: the hydroxylation of monophenols to o-diphenols and the oxidation of o-diphenols to o-quinones.
’ To whom correspondence
should be addressed. Fax: 34-68-835418.
0003.9861/91 $3.00 Copyright G 1991 by Academic Press, All rights of reproduction in any form
Dopamine and related compounds have been described as new antitumoral agents, which show significant activity on melanome cells (4-5). 6-Hydroxydopamine (GOHDA) ,’ for example, has shown great cytotoxicity against melanogenic cells ( 6 ) , although its neurotoxicity has ruled out its clinical application. The cytotoxicity of dopamine and GOHDA has been attributed to their selective uptake by melanocytic cells ( 7)) and to the formation of reactive quinones and semiquinones during their metabolic activation catalyzed by tyrosinase ( 7-g), an enzyme widely present in melanome cells (10-12). The corresponding quinones and semiquinones react strongly with nucleophilic groups of cellular molecules and macromolecules (6,8,13-17). Other studies have considered the possible role of neuromelanins in neuron damage, or in the promotion of the selective vulnerability of neurons in Parkinson’s disease ( 18). Tyrosinase catalyzes the oxidation of dopamine (19) by molecular oxygen to yield o-dopaminequinone, which evolves nonenzymatically through two branches consisting of cyclization or hydroxylation reactions (Fig. 1) . The production of semiquinones may be significant because of the presence of spin-stabilizing metal ions (9). This pathway has been characterized at the pH > 6 range, where only the intermediates of the cyclization branch have been identified and the corresponding rate constants determined ( 19 ) . The aim of this paper is to complete the characterization of the oxidation pathway of dopamine catalyzed by tyrosinase. The experimental study attempts to identify the intermediates of the hydroxylation branch and to determine the corresponding rate constants, which would permit the quantitative description of the effect of the pH on this pathway. The intermediates of the hydrox-
’ Abbreviation
used: 60HDA-6,6-hydroxydopamine. 427
Inc. reserved.
428
GARCiA-MORENO
ET AL.
ylation branch, such as those detected in melanosomes (20-22) and in tumoral cells (23-24)) have great cytotoxic power (25)) and could have physiological significance in slightly acidic media. MATERIALS
AND
METHODS
Mushroom tyrosinase (o-diphenol: O2 oxidoreductase, EC 1.14.18.1, 3300 units/mg), dopamine chlorohydrate and GOHDA chlorohydrate were purchased from Sigma (USA). All other chemicals were of analytical grade and supplied by Merck (Germany). Spectra were recorded with an Aminco DW-2 spectrophotometer equipped with a Hewlett-Packard recorder with kinetic response, allowing the dead recording time to be minimized. The scan speed was 20 nm s-i, and the first recording was started at 20 s from the beginning of the reaction. The assay medium contained 50 mM phosphate buffer, pH 5.0. Reference cuvettes contained, in all cases, all the components except substrate. Other conditions are detailed in the legends of the figures. Matrix analysis of the iterative spectra was carried out by application of the test for two or three absorbing species in solution, with stoichiometric restrictions (26). Product accumulation was spectrophotometrically followed at 480 nm using a Perkin-Elmer Lambda-3 spectrophotometer interfaced on-line with a PerkinElmer DS-3600 computer. The reaction medium was 50 mM phosphate buffer with 0.2 M NOsK in order to obtain a constant ionic strength. Other reagents and conditions are detailed in the legends of the figures. Oxygen consumption was followed by a Hansatech DW oxymeter, based on the Clark electrode. Cyclic voltammetry was carried out using a INELECSA PCD1212 microprocessor-controlled system. The working electrode was a hanging mercury drop electrode (METROHM EA 290) with radius equal to 0.06 cm. The reference was a saturated calomel electrode and the counter was a platinum electrode. The enzyme or NaIO, was incubated with the substrate inside the cell with the electrodes. Nitrogen was bubbled through the solution before starting the record. In all cases, scan rate was 60 mV s-l. Temperature was controlled at 20°C using a Hetofrig circulating bath with a heater/cooler and checked using a Cole-Parmer digital thermometer with a precision of fO.l’C. Protein concentration was determinated by a modified Lowry method (27). RESULTS
AND
DISCUSSION
Rapid-Scan Assays Dopamine was oxidized by either O,/tyrosinase or NaI04, in order to contrast the nonenzymatic breakdown of o-dopaminequinone (Fig. 1) at different pH values. Dopamine oxidation by tyrosinase. An iterative spectrum for oxidation of dopamine at pH 5.0 (Fig. 2A) showed the initial formation of an absorbance maximum at 390 nm and the further appearance of another maximum at 485 nm. The high enzyme concentration used caused rapid oxygen exhaustion, characterized by a crossing between tracings with no clearly defined isosbestic point. A graphical analysis of these recordings by the rank matrix method (26) confirms the presence of a minimum of three absorbing species in a solution with stoichiometric restrictions (Fig. 2B). At pH values greater than 6.0 ( 19)) absorption maxima were obtained at 390 and 480 nm, and one isosbestic point
A
FIG. 1. Pathway proposed for the oxidation of dopamine catalyzed by tyrosinase, taking into consideration the effect of pH. E, tyrosinase; D, dopamine; QH, o-dopaminequinone-H+; Q, o-dopaminequinone; OSQ, o-dopaminesemiquinone; L, leukodopaminechrome; T, GOHDA; PQ, p-topaminequinone; PSQ, p-topaminesemiquinone; DC, dopaminechrome. (A ) Cyclization branch of o-dopaminequinone. (B ) Hydroxylation branch of o-dopaminequinone.
was defined at 418 nm. The graphical analysis of these recordings showed that there were two kinetically related species. These data, which were in accordance with other experimental results ( 19)) enabled the two absorbing species (Fig. 1) to be identified as o-dopaminequinone ( Lax = 390 nm) and dopaminechrome (X,,, = 480 nm). The appearance of one isosbestic point revealed that leukodopaminechrome was not accumulated in the assay medium, in accordance with other authors and techniques (9,28). The nonaccumulation of o-semiquinones was due to the absence of spin-stabilizing metal ions (9) _The formation of dopaminechrome (Fig. 1) was simultaneous with the regeneration of dopamine (9,19,28), for which reason the o-diphenol was not considered as one linearly independent substance by the matrix analysis (26). At pH values lower than 6.0 (Figs. 2A and 2B), the intermediates of both the cyclization and the hydroxylation branches of Fig. 1 were present. Thus, the three absorbing species detected by the matrix analysis were identified as o-dopaminequinone (X,,, = 390 nm), dopaminechrome (X,,, = 480 nm) and p -topaminequinone ( Lax = 485 nm (29)). The fact that the regenerated dopamine, leukodopaminechrome, and o- and p -semiqui-
OXIDATION
PATHWAY
OF TYROSINASE-CATALYZED
0 350
450
h(nm)
550
429
DOPAMINE
1.2 0.6 Amj -Amjr/Anj
1.8 -Anj
FIG. 2. Rapid-scan assays for the oxidation of dopamine catalyzed by tyrosinase. Reagents: 1.0 mM dopamine, 0.26 mM O,, and 40 pg/ml tyrosinase. Other conditions as detailed under Materials and Methods. (A) Iterative spectrum. (B) Matrix analysis of the spectrum. A,, = Absorbance value at wavelength i and tracingj. n = 475 nm, n = 392 nm; i (nm) = 371 (O), 402 (A), 423 (), 450 (O), 491 (a), 511 (+), 532 (m);j’ = third recording.
nones were not detected can be explained by the same reasons as described for the cyclization branch. The hydroxylation branch, however, involved the accumulation of another intermediate, as was deduced from the nonappearance of an isosbestic point (Fig. 2A). This compound might have been GOHDA, which is not an absorbing species in the visible range. It is well-known that Dopamine oxidation byperiodate. sodium periodate oxidizes o-diphenols to their corresponding o-quinones (29). When [ dopamine] was greater than [ NaIO,] at slightly acid pH (Fig. 3A) the iterative spectrum was similar to that carried out with high tyrosinase concentrations (Fig. 2A). The presence of at least three absorbing species in solution was detected by matrix analysis (Fig. 3B). At higher pH values ( 19)) a great similarity with the respective enzymatic assays (described in the above section) was also obtained.
0 350
When [NaIO,] was greater than [dopamine] at pH values lower than 6.0, a set of recordings which showed the presence of two absorbance maxima at 390 and 470 nm was obtained, as well as one isosbestic point at 398 nm (Fig. 4A). The existence of a minimum of two absorbing species in solution was detected by matrix analysis (26)) a set of straight lines intersecting at the origin of the coordinates being obtained (Fig. 4B). Similar results were reported at pH values higher than 6.0 ( 19)) although the isosbestic point appeared at 403 nm. The periodate anion might act as oxidant in the following reactions of the pathway depicted in Fig. 1: dopamine + o-dopaminequinone, leukodopaminechrome + dopaminechrome, and GOHDA + o-topaminequinone. The nonaccumulation of o-semiquinones in these experiments (Figs. 3A and 4A) was also due to the absence of spin-stabilizing metal ions ( 9 ) . When [ NaIO, ] was lower
-0.5L -0.8 450
htnm)
550
-0.7
-0.6 Amj -A,jf/A"j
-0.5 -Anjp
FIG. 3. Rapid-scan assays for the oxidation of dopamine by a deficit of NaIO,. Reagents: 2.7 mM dopamine and 0.03 mM NaIO,. Other conditions as detailed under Materials and Methods. (A) Iterative spectrum. (B) Matrix analysis of the spectrum. m = 510 nm, n = 390 nm; i (nm) = 405 (O), 417 (B), 444 (O), 460 (0). 482 (A), 530 (A);j’ = first recording.
430
GARCiA-MORENO
ET AL.
FIG. 4. Rapid-scan assays for the oxidation of dopamine by an excess of NaIO*. Reagents: 0.27 mM dopamine and 2.7 mM NaI04. Other conditions as detailed under Materials and Methods. (A) Iterative spectrum. (B) Matrix analysis of the spectrum. i’ = 475 nm; i (nm) = 370 (O), 415 (A), 435 (Cl), 494 (m), 505 (A), 530 (0);j’ = first recording.
than [dopamine] the periodate anion was exhausted in the stoichiometric formation of o-dopaminequinone (Fig. 3A), in a parallel way to the oxygen depletion catalyzed by tyrosinase (Fig. 2A). The similarity between the tyrosinase and periodate assays confirms the nonenzymatic breakdown of o-dopaminequinone (Fig. 1) at slightly acid pH (Figs. 2 and 3 ) and neutral pH values ( 19 ) . When [ NaIO,] was higher than [dopamine], a stoichiometric quantity of periodate was consumed in the formation of o-dopaminequinone (Fig. 1) , which underwent nonenzymatic cyclization or hydroxylation to yield leukodopaminechrome or GOHDA, respectively. These intermediates were not accumulated in solution since they were also oxidized by the excess of periodate to provide the corresponding dopaminechrome or o-topaminequinone. The nonaccumulation of the intermediates explained the appearance of one isosbestic point, whose hipsochromic displacement at slightly acid pH (Fig. 4A) was due to the simultaneous operation of both branches of the pathway (Fig. 1). Under these conditions, the matrix analysis detected at least two absorbing species, odopaminequinone and dopaminechrome/o-topaminequinone. These two products were generated in a constant ratio during the whole assay time, this ratio being determined by the cyclization and the hydroxylation steps, respectively. Thus, one of the two compounds was considered as linearly dependent on the other by the matrix analysis (26).
The cyclic voltammogram corresponding to the dopamine oxidation by OJtyrosinase at pH 7.0 (Fig. 5) showed one cathodic peak and its corresponding anodic peak. These peaks corresponded to the reduction of the dopaminechrome to leukodopaminechrome (cathodic peak) and to the oxidation of the leukodopaminechrome to dopaminechrome (anodic peak) (28) ; therefore, at these pH values, only the cyclization branch of the pathway was significant. However, cyclic voltammograms obtained for pH values lower than 6.0 after oxidation of dopamine by both O,/ tyrosinase and sodium periodate, consisted of two cathodic and two anodic peaks (Figs. 6A and 6B). Figure 6A shows the cyclic voltammogram corresponding to the dopamine oxidation by tyrosinase at pH 4.2. The cathodic peak obtained at a potential of -0.06 V vs SCE was due to the reduction of dopaminechrome to leukodopaminechrome; its anodic peak appeared at -0.02 V vs SCE and was due to the oxidation of leukodopaminechrome to dopaminechrome, in accordance with the cyclic voltammogram of dopaminechrome at this pH (results not shown). The cathodic peak located at more positive po-
0.5 I (JJN
Cyclic Voltammetry The oxidation pathway of dopamine catalyzed by tyrosinase leads to the production of p-topaminequinone and dopaminechrome (Fig. 1) . Since the X,,, for these compounds are very similar (485 and 480 nm, respectively) , it is difficult to determine spectrophotometrically whether the two products are formed at the same time and pH values. However, cyclic voltammetry provides a solution to this question.
-0.51, I -0.1 -0.2
I I -0.3 -04 E(V)
I
FIG. 5. Cyclic voltammogram for the oxidation of dopamine at neutral pH. Reagents: dopamine (1.0 mM) with tyrosinase (5.0 fig/ml) in 0.1 M phosphate buffer, at pH 7.0.
OXIDATION
03
PATHWAY
0
- 0.1
OF TYROSINASE-CATALYZED
-0.2
01
431
DOPAMINE
0
-0.1
- 0.2
E (VI
E(V)
FIG. 6. Cyclic voltammogram for the oxidation of dopamine at acid pH. (A) Reagents: dopamine (1.0 mM) with tyrosinase M acetate buffer, at pH 4.2 (B) Reagents: dopamine ( 1.0 mM) with NaI04 (0.3 mM) in 0.1 M acetate buffer, at pH 3.5.
tential, +0.05 V vs SCE, could be ascribed to the reduction of p -topaminequinone to GOHDA. Its corresponding anodic peak appeared at the potential value of +0.06 V vs SCE, and represented the oxidation of GOHDA to p-topaminequinone, in accordance with the cyclic voltammogram of GOHDA at this pH (results not shown). Similar results were obtained when dopamine was oxidized by a deficit of NaI04 at pH 3.5 (Fig. 6B ) . These results confirm the oxidation pathway of dopamine here proposed (Fig. 1) . Kinetic
(15 yg/ml)
in 0.1
(19). This kinetic behavior has not yet been analyzed and could be useful for the quantitative characterization of the pathway. The absolute values of the steady-state rate of the enzymatic step were determined by measuring oxygen consumption (Fig. 8) due to the presence of two absorbing products at 480 nm, whose relative proportions are not known. These rates were one-half that of o-dopaminequinone-H+ formation (Fig. 1) , since one molecule of oxygen generated two molecules of quinone in each turnover (30).
Assays
The progress of dopamine oxidation catalyzed by tyrosinase was followed by measuring the appearance of products at 480 nm throughout the assay time. The experimental recordings (Fig. 7A) presented a marked lag period (7). At slightly acid pH, the lag period decreased when the enzyme concentration rose (Fig. 7B ) , whereas no dependence between them was obtained at greater pH values
Kinetic
Analysis
The oxidation pathway of dopamine catalyzed by tyrosinase at values of pH lower than 6.0 is depicted in Fig. 1. This scheme involves the steady-state rate of the enzymatic step (V,) , first (k, and k,) and second (k-, , k,, k3, ke3, k4, kp4, kg, kp5, kg, and ke6) order rate constants. The nonappearance of an isosbestic point in the rapid scan assays at pH < 6 (Figs. 2 and 3) unlike in the same experiments at pH > 6 (19) supported the accumulation
A
FIG. 7. Kinetic assays of product accumulation for the oxidation of dopamine catalyzed by tyrosinase. Reagents: 2.7 mM dopamine and 0.26 mM O2 at, pH 5.22. Other conditions as detailed under Materials and Methods. (A) Spectrophotometric recording. Tyrosinase (pg/ml) : 2.5, 3.3, 5.0, 6.7, 8.3. (B, 0) Corresponding values of the lag period (7) vs enzyme concentration (Es). (B, A ) Plot of products accumulation rate in steady-state conditions against E,,.
FIG. 8. (A) Kinetic assays of oxygen consumption for the oxidation of dopamine catalyzed by tyrosinase. Reagents: 2.7 mM dopamine, 0.26 mM O2 and tyrosinase (@g/ml): 2.5, 3.3, 5.0, 6.7, 8.3 at pH 5.22 (B) Corresponding values of V,, vs E,.
432
GARCiA-MORENO
of one intermediate. This compound should have a diphenolic structure since it was oxidized by an excess of NaI04, as shown by the appearance of one isosbestic point in the corresponding rapid-scan assays at pH < 6 (Fig. 4). This intermediate could be GOHDA, since leukodopaminechrome was not accumulated in the assay medium (9, 19, 28). The cyclic voltammetry assays at pH < 6 showed the simultaneous operation of both the cvclization and the hydroxylation branches of o-dopaminequinone (Fig. 6)) in contrast to a separate pattern for GOHDA. Therefore, the accumulated intermediate could be GOHDA (Fig. 1) , which would imply a significant contribution of the second-order step controlled by k4 and km4, to the kinetic behavior of the pathway. However, there are no analytical solutions to describe the evolution of the products during the transient phase of the kinetic assays shown in Fig. 7A. At the steady-state of the pathway, however, it is possible to derive explicit equations relating the product concentrations with the assay time. Kinetic analysis of this pathway requires that the dopamine consumption as well as the breakdown of the products should be negligible during the reaction time. Both conditions have been verified in the experimental assays. The mass balance between the reagents states that the consumed o-diphenol (Vat) yields intermediates ( Z[ I ] ) , products ( [DC ] and [ PQ ] ) and regenerated odiphenol, through the cyclization and the hydroxylation branches of the pathway: V,t+
k,[D][QH]t
ET AL.
accumulated during the assays (see Rapid-Scan section). Therefore, the net formation of PQ is equivalent to k4[ T] [ QH] t, whereas the net production of DC is determined by k3[ QH ] [ L ] t, and Eq. [l] is transformed to 2([DC]
+ [PQ])
[PQ]t
[ Qfi ]
= Vo+ k-dH+l [&I - MQHI - kc[Ql - %z[QHl - MQHI [Tl = 0 [a]
= k,[QH]
- (k-,[H+l
- k,[QH][L]t
= Z[I]
- k,[ + [DC
2111= [&HI + [&I + IL1 + [OSQI + [PSQ + [Tl. The generation and breakdown steps of semiquinones in rapid equilibrium ( 9 ) :
[21 are
biDI [QHI = k+,[OSQ12,kITI [PQI = kps[PSQ]‘.
[3]
There are reversible steps displaced toward the regeneration of dopamine ( 19) :
MQHI [Ll B LADCI [Dl, JZdQHl[Tl ti kp,[PQ
[Dl.
[41
Furthermore, the contribution of L, OSQ, and PSQ to Eq. [ 2 ] is negligible since these intermediates are not
= 0
= 0,
[61
where k; = k,[ Hz01 . Note that the term k3[ QH] [L] has been replaced by k,[Q ] in Eq. [ 61, since in the steady state both terms are equivalent. From the above system of differential equations, expressions for the concentration of the intermediates at the steady state are obtained:
[QHI,, = 2klkc
Consideration expression
(k-,[H+l + k,)Vo + 2k’,(k-,[H+]
+ k,)
k,Vo
[Qlss= 2klk,
+ 2kL(kp,[H+]
ITl,,
rl [QHlt + [PQI PI
+ k,)[Q]
[?I = k’,[QHl - h[QHI[Tl
+ km,[D][DC]t
- kpg[OSQ]‘t
151
Z[I].
During the steady state of the pathway, the concentration o f the intermediates remains constant:
_
+ k-,[D]
= Vat-
_“I
= k:lk,. _,
+ k,) 171
I
of Eqs. [l] - [ 51 and [ 71 leads to the
[DClss+ [PQlss kl + kp,[H+] + k, 4 2klk, + 2k’,(k-,[H+] + k,) + k4Vo
11
. [81
Thus, the accumulation of the products of the pathway during the steady state follows a straight line (Eq. [ 81) , whose intercept with the time axis defines the lag period (7): kl + k-,[H+] + kc 4 7 = 2klk, + 2k’,(kp,[H+] + kc) + $V, ’
[91
Since k, % kl ( 19)) the previous equation becomes k-,[H+] + k, kh ’ = 2k1k, + 2k;(k-,[H+] + k,) + k4V0.
WI
OXIDATION
PATHWAY
l/v,
OF TYROSINASE-CATALYZED
433
DOPAMINE
[H’l pM
PM-‘s
FIG. 9. Kinetic assays of the oxidation of dopamine catalyzed by tyrosinase, at different pH values for several enzyme concentrations. Values of pH 4.98, 5.16, 5.22, 5.48, 5.68, 6.60, 6.90; tyrosinase (fig/ml) for each pH: 1.7, 2.5, 3.3, 5.0, 6.7, 8.3, 10.0, 11.7, 13.3, 16.7. Other assay conditions are as detailedunder Materials and Methods. (A) Dependence of the lag period (7) on l/V,, kinetic parameters obtained from spectrophotometric and oxymetric recording, respectively. Experimental data and linear regression fittings are shown. (B) Corresponding values of 2cu and of fl vs [H-l. (0, 0) Experimental data of 2u and of 0, respectively. (0 ----- 0) Data calculated using Eq. [14], with the initial estimations of their kinetic constants. (0 0) Data calculated using Eq. (141, with the final estimations of their kinetic constants, obtained from nonlinear regression fitting. (0 ~ 0) Mean value of fi vs [H+] , according to Eq. [15]
This equation concentration experimental ues, however, tration (19). + k, ) < k,k,,
predicts that r decreases when the enzyme ( E0 involved in V,) is raised, according to data at acid pH (Fig. 7). At greater pH valr is not dependent on the enzyme concenThus, when [H+] decreases, 2k’,( kmi[H+] ( k; 4 k, ) and Eq. [lo] is transformed into 7=
kp,[H+]
+ k, +
2k,k,
k’, kvo.
[Ill
Furthermore, since V, increases when pH is raised, a negligible contribution of the second quotient as regards the first leads to T = (km,[H+]
+ k,)/2klk,,
1 21
an expression not dependent on the enzyme concentration. On the other hand, Equation [lo] can be rearranged as 7 = 2a + 0(1/V,,),
P31
where
2a =
$ [H+] 2 [H+]
+ $$ 1 2 +e
Kinetic
Data Analysis
Two series of spectrophotometric (Fig. 7) and electrometric (Fig. 8) assays were carried out by using the same set of enzyme concentrations. From AdBOvs t recordings, data of the final portion were fitted by linear regression and the corresponding r vs E0 values were calculated. The initial zone of the [ 0,] vs t recording was fitted by linear regression, yielding the respective values of V,,/2 vs EO. The kinetic assays, therefore, provide T vs ( 1 lb’,,) data at the same E, values. This procedure was applied to several pH values (Fig. 9A) and the data fitted by linear regression according to Eq. [ 13 1. The great similarity between the slopes of the straight lines (Fig. 9A) confirmed the nondependence of p on pH (Eq. 15). Thus, the mean values of the slopes enabled k’,/ k4 to be calculated (Fig. 9B). The intercepts on the ordinate axis, however, showed a hyperbolic behavior on [H+] (Fig. 9B ) . The nonlinear regression fitting of 2cuvs [H’ ] data to Eq. [14] permitted the determination of all the rate constants of the pathway, by considering K, = 2.51 10 -I1 M (31) . Initial estimations of kl , kb , and k, were obtained, from the lower limit, higher limit and half-saturation value of the hyperbola, respectively (Fig. 9B). The values of the rate constants of the pathway (Fig. 1) are described in Table I.
[I41 Effect of pH on the Dopamine
2
and
P = k;/k,
K, = k, / km,. These expressions are useful to determine the rate constants of the pathway.
[I51
Oxidation
Pathway
The spectral and the kinetic assays described above support and characterize the proposed oxidation pathway of dopamine (Fig. 1) . The pH of the medium determines the differential contribution of the cyclization and the
434
GARCIA-MORENO TABLE
I
Values of the Rate Constants for the Oxidation Pathway of Dopamine Catalyzed by Tyrosinase (Fig. l), at 20°C Constant k, k, k:j k4
Units s ~’ M-’ 5-l
8 -’
M-' s-1
ET AL. 3. 4. 5. 6.
Value 0.12 (8.11 849.7 640.1
f 0.02 f 0.65) X 10m5 t- 178.4 + 96.2
7. 8. 9. 10.
hydroxylation branches, through the protonation/deprotonation of the key intermediate, o-dopaminequinoneH+. Thus, the above experimental data show that both branches are only simultaneously significant at slightly acid pH. Indeed, there is evidence concerning the acidic character of melanosomes ( 20-22) and tumoral cells (23-24). Antitumoral activity has been found much greater for catechols and catecholamines forming noncyclizable quinones than for those forming cyclizable quinones (5), perhaps due to the stability of noncyclizable quinones (32) that react with nucleophilic groups of cellular molecules and macromolecules (6, 8, 13-17). Dopamine at neutral pH formed a cyclizable quinone, but at slightly acid pH the addition of water to the quinone ring generates GOHDA (Fig. 1) , which forms a quinone that cyclizes slowly (32). This increases the cytotoxic and antitumoral power of dopamine (8). The production of semiquinones might be increased under physiological conditions, due to the presence of spin-stabilizing ions. These ions have been found in a variety of melanized structures (33)) including the choroid of the eye (34-35)) black hair ( 36)) and isolated melanosomes from Harding-Passey, horse, and human melanomes (37). The reaction between semiquinones and sulf hydryl aminoacids such as cysteine probably triggers a cascade of free-radical reactions. Therefore, the production of even small amounts of semiquinones in the pathway of oxidation of catechols and catecholamines may have toxicological significance ( 9 ) . Furthermore, reagents causing changes in the pH of the physiological microenvironment might determine the actual cytotoxicity of dopamine and other catecholamines.
11. 12. 13. 14. 15. 16. 17. 18. 19.
20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30.
31. 32.
ACKNOWLEDGMENTS This paper has been partially supported by a grant from the CICYT (ALI89-674)) DGICYT (PB 87-0700)) and CARM (PB 90/37). J. N. Rodriguez-Lopez has received a fellowship from the Comunidad Autonoma de Castilla-La Mancha (Spain).
REFERENCES 1. Harrison, W. H. (1963) Arch. Biochem. Biophys. 101, 116-130. 2. Robb, D. A. ( 1984) in Copper Proteins and Copper Enzymes (Lontie, R., ed.), Vol. II, pp. 2077240, CRC Press, Boca Raton.
33. 34. 35. 36. 37.
Lerch, K. (1983) Mol. Cell. Biochem. 52, 125-138. Wick, M. M. (1978) J. Invest Dermatol. 71, 163-164. Wick, M. M. (1979) Cancer Treat. Rep. 63, 991-997. Wick, M. M., Byers, L., and Ratliff, J. (1979) J. Incest. Dermatol. 72,67-69. Wick, M. M., Byers, L., and Frei, F. (1977) Science 197,468-469. Graham, D. G., Tiffany, S. M., and Vogel, S. F. (1978) J. Invest. Dermatol. 70, 113-116. Korytowski, W., Sarna, T., Kalyanaraman, B., and Sealy, R. C. (1987) Biochim. Biophys. Acta 924.383-392. Pomerantz, S. H., and Li, J. P. C. (1973) Yale J. Biol. Med. 46, 541-552. Nishioka, K. (1978) Eur. J. Biochem. 85, 137-146. Burnett, J. B. (1971) J. Biol. Chem. 246,3079-3091. Saner, A., and Thoener, H. (1971) Mol. Pharmacol. 7, 147-154. Rothman, A., Daly, J. W., and Creveling, C. R. (1976) Mol. Pharmacol. 12,887-899. Jimenez, M., Garcia-Canovas, F., Garcia-Carmona, F., Iborra, J. L., and Lozano, J. A. (1986) Znt. J. Biochem. 18, 161-166. Cabanes, J., Garcia-Canovas, F., and Garcia-Carmona, F. (1987) Biochim. Biophys. Acta 914, 190-197. Garcia-Carmona, F., Cabanes, J., and Garcia-Canovas, F. (1987) Biochim. Biophys. Acta 914, 198-204. Langston, J. W. (1988) Trends. Pharmacol. Sci. 9,347-348. Jimenez, M., Garcia-Carmona, F., Garcia-Canovas, F., Iborra, J. L., Lozano, J. A., and Martinez, F. (1984) Arch. Biochem. Biophys. 235,438-448. Sacki, H., and Oikawa, A. (1983) J. Cell. Physiol. 116, 93-97. Sacki, H., and Oikawa, A. (1985) J. Znuest. Dermatol. 85,423-425. Moelhman, G., Slominski, A., Kuklinska, E., and Lerner, A. B. ( 1988) Pigment. Cell. Res. Suppl. 1, 79-87. Wike-Hooley, J. L., Haveman, J., and Reinhold, H. S. (1984) Radiother. Oncol. 2, 343-366. Hedley, D. W., and Jorgensen, H. B. (1989) Exp. Cell. Res. 180, 106e116. Liang, Y. O., Plotsky, P. M., and Adams, R. N. (1977) J. Med. Chem. 20, 581-583. Coleman, J. S., Varga, L. P., and Mastin, S. H. ( 1970) Znorg. Chem. 9,1015-1020. Hartree, E. (1972) Anal. Biochem. 46422-427. Adams, R. N., Murril, E., McCreery, R., Blank, L., and Karolczak, M. (1972) Eur. J. Pharmacol. 17, 287-292. Graham, D. G., and Jeffs, P. W. (1977) J. Biol. Chem. 252,57295734. Galindo, J. D., Pedreiio, E., Garcia-Carmona, F., Garcia-Canovas, F., Solano, F., and Lozano, J. A. (1983) Znt. J. Biochem. 15,14551461. Gray, D. O., and Weitzman, P. D. J. (1968) Data for Biochemical Research, p. 22, Clerendon Press, Oxford. Blank, C. L., McCreery, R. L., Wightman, R. M., Chey, W., Adams, R. M., Reid, J. R., and Smissman, E. E. (1976) J. Med. Chem. 19, 178-180. Palumbo, A., D’Ischia, M., Misuraca, G., and Prota, G. ( 1987) Biochim. Biophys. Acta 925, 203-209. Bowness, J. M., and Morton, R. A. (1953) Biochem. J. 53, 620626. Underwood, E. J. ( 1971) in Trace Elements in Human and Animal Nutrition, 3rd Ed., Academic Press, New York. Dorea, J. G., and Pereira, S. E. (1983) J. Nutr. 113,2375-2381. Horcicko, J., Borovansky, J., Duchon, J., and Prochazkova, B. (1983) Hoppe-Seyler’s Z. Physiol. Chem. 354, 203-204.
OF BIOCHEMISTRY
AND
BIOPHYSICS
Vol. 288, No. 2, August 1, pp. 427-434, 1991
Effect of pH on the Oxidation Catalyzed by Tyrosinase
Pathway of Dopamine
Manuela Garcia-Moreno, * Jo& Neptuno Rodriguez-L6pez, Ram6n Var&r, * and Francisco Garcia-C%novas$gl
* Francisco
Martinez-Ortiz,?
Jo& Tudela,$
*Ca’tedra de Quimica-Fisica, E. U. Polikknica, Universidad de Castilla-La Mancha, Albacete, Spain; t Departamento de QuimicaFkica, Facultad de Ciencias Quimicas, Universidad de Murcia, Murcia, Spain; and $ Departamento de Bioquimica y Biologia Molecular, Facultad de Biologia, Universidad de Murcia, E-30100 Espinardo, Murcia, Spain
Received October 26, 1990, and in revised form March 12, 1991
The oxidation of 3,4-dihydroxyphenylethylamine (dopamine) by O2 catalyzed by tyrosinase yields 4-(2aminoethyl) - 1,2-benzoquinone (o-dopaminequinone) , which evolves nonenzymatically through two branches or sequences of reactions, whose respective operations are determined by the pH of the medium. The cyclization branch of o-dopaminequinone takes place in the entire range of pH and is the only significant branch at pH > 6. The hydroxylation branch of o-dopaminequinone only operates significantly at pH < 6, and involves the accumulation of 2,4,5-trihydroxyphenylethylamine (6-hydroxydopamine) and 5- (2-aminoethyl) -2-hydroxy1,4benzoquinone (p-topaminequinone) , identified from cyclic voltammetry assays. The kinetic characterization of the hydroxylation branch of o-dopaminequinone has been carried out by spectrophotometric and oxymetric assays. The successful fitting of data to the kinetic behavior predicted by the kinetic analysis at both pH > 6 and pH < 6 confirms the overall oxidation pathway proposed for the dopamine oxidation catalyzed by tyrosinase. The antitumoral power of dopamine is possibly enhanced by the high cytotoxicity of 6-hydroxydopamine and ptopaminequinone, accumulated at the acidic pH characteristic of melanosomes and melanome Cells. 0 1991 Academic Press,
Inc.
The existence of a minor route of dopamine oxidation by tyrosinase has been established ( 1). Tyrosinase is a copper protein widely distributed in the phylogenetic scale ( 2-3 ) . It has a bifunctional activity: the hydroxylation of monophenols to o-diphenols and the oxidation of o-diphenols to o-quinones.
’ To whom correspondence
should be addressed. Fax: 34-68-835418.
0003.9861/91 $3.00 Copyright G 1991 by Academic Press, All rights of reproduction in any form
Dopamine and related compounds have been described as new antitumoral agents, which show significant activity on melanome cells (4-5). 6-Hydroxydopamine (GOHDA) ,’ for example, has shown great cytotoxicity against melanogenic cells ( 6 ) , although its neurotoxicity has ruled out its clinical application. The cytotoxicity of dopamine and GOHDA has been attributed to their selective uptake by melanocytic cells ( 7)) and to the formation of reactive quinones and semiquinones during their metabolic activation catalyzed by tyrosinase ( 7-g), an enzyme widely present in melanome cells (10-12). The corresponding quinones and semiquinones react strongly with nucleophilic groups of cellular molecules and macromolecules (6,8,13-17). Other studies have considered the possible role of neuromelanins in neuron damage, or in the promotion of the selective vulnerability of neurons in Parkinson’s disease ( 18). Tyrosinase catalyzes the oxidation of dopamine (19) by molecular oxygen to yield o-dopaminequinone, which evolves nonenzymatically through two branches consisting of cyclization or hydroxylation reactions (Fig. 1) . The production of semiquinones may be significant because of the presence of spin-stabilizing metal ions (9). This pathway has been characterized at the pH > 6 range, where only the intermediates of the cyclization branch have been identified and the corresponding rate constants determined ( 19 ) . The aim of this paper is to complete the characterization of the oxidation pathway of dopamine catalyzed by tyrosinase. The experimental study attempts to identify the intermediates of the hydroxylation branch and to determine the corresponding rate constants, which would permit the quantitative description of the effect of the pH on this pathway. The intermediates of the hydrox-
’ Abbreviation
used: 60HDA-6,6-hydroxydopamine. 427
Inc. reserved.
428
GARCiA-MORENO
ET AL.
ylation branch, such as those detected in melanosomes (20-22) and in tumoral cells (23-24)) have great cytotoxic power (25)) and could have physiological significance in slightly acidic media. MATERIALS
AND
METHODS
Mushroom tyrosinase (o-diphenol: O2 oxidoreductase, EC 1.14.18.1, 3300 units/mg), dopamine chlorohydrate and GOHDA chlorohydrate were purchased from Sigma (USA). All other chemicals were of analytical grade and supplied by Merck (Germany). Spectra were recorded with an Aminco DW-2 spectrophotometer equipped with a Hewlett-Packard recorder with kinetic response, allowing the dead recording time to be minimized. The scan speed was 20 nm s-i, and the first recording was started at 20 s from the beginning of the reaction. The assay medium contained 50 mM phosphate buffer, pH 5.0. Reference cuvettes contained, in all cases, all the components except substrate. Other conditions are detailed in the legends of the figures. Matrix analysis of the iterative spectra was carried out by application of the test for two or three absorbing species in solution, with stoichiometric restrictions (26). Product accumulation was spectrophotometrically followed at 480 nm using a Perkin-Elmer Lambda-3 spectrophotometer interfaced on-line with a PerkinElmer DS-3600 computer. The reaction medium was 50 mM phosphate buffer with 0.2 M NOsK in order to obtain a constant ionic strength. Other reagents and conditions are detailed in the legends of the figures. Oxygen consumption was followed by a Hansatech DW oxymeter, based on the Clark electrode. Cyclic voltammetry was carried out using a INELECSA PCD1212 microprocessor-controlled system. The working electrode was a hanging mercury drop electrode (METROHM EA 290) with radius equal to 0.06 cm. The reference was a saturated calomel electrode and the counter was a platinum electrode. The enzyme or NaIO, was incubated with the substrate inside the cell with the electrodes. Nitrogen was bubbled through the solution before starting the record. In all cases, scan rate was 60 mV s-l. Temperature was controlled at 20°C using a Hetofrig circulating bath with a heater/cooler and checked using a Cole-Parmer digital thermometer with a precision of fO.l’C. Protein concentration was determinated by a modified Lowry method (27). RESULTS
AND
DISCUSSION
Rapid-Scan Assays Dopamine was oxidized by either O,/tyrosinase or NaI04, in order to contrast the nonenzymatic breakdown of o-dopaminequinone (Fig. 1) at different pH values. Dopamine oxidation by tyrosinase. An iterative spectrum for oxidation of dopamine at pH 5.0 (Fig. 2A) showed the initial formation of an absorbance maximum at 390 nm and the further appearance of another maximum at 485 nm. The high enzyme concentration used caused rapid oxygen exhaustion, characterized by a crossing between tracings with no clearly defined isosbestic point. A graphical analysis of these recordings by the rank matrix method (26) confirms the presence of a minimum of three absorbing species in a solution with stoichiometric restrictions (Fig. 2B). At pH values greater than 6.0 ( 19)) absorption maxima were obtained at 390 and 480 nm, and one isosbestic point
A
FIG. 1. Pathway proposed for the oxidation of dopamine catalyzed by tyrosinase, taking into consideration the effect of pH. E, tyrosinase; D, dopamine; QH, o-dopaminequinone-H+; Q, o-dopaminequinone; OSQ, o-dopaminesemiquinone; L, leukodopaminechrome; T, GOHDA; PQ, p-topaminequinone; PSQ, p-topaminesemiquinone; DC, dopaminechrome. (A ) Cyclization branch of o-dopaminequinone. (B ) Hydroxylation branch of o-dopaminequinone.
was defined at 418 nm. The graphical analysis of these recordings showed that there were two kinetically related species. These data, which were in accordance with other experimental results ( 19)) enabled the two absorbing species (Fig. 1) to be identified as o-dopaminequinone ( Lax = 390 nm) and dopaminechrome (X,,, = 480 nm). The appearance of one isosbestic point revealed that leukodopaminechrome was not accumulated in the assay medium, in accordance with other authors and techniques (9,28). The nonaccumulation of o-semiquinones was due to the absence of spin-stabilizing metal ions (9) _The formation of dopaminechrome (Fig. 1) was simultaneous with the regeneration of dopamine (9,19,28), for which reason the o-diphenol was not considered as one linearly independent substance by the matrix analysis (26). At pH values lower than 6.0 (Figs. 2A and 2B), the intermediates of both the cyclization and the hydroxylation branches of Fig. 1 were present. Thus, the three absorbing species detected by the matrix analysis were identified as o-dopaminequinone (X,,, = 390 nm), dopaminechrome (X,,, = 480 nm) and p -topaminequinone ( Lax = 485 nm (29)). The fact that the regenerated dopamine, leukodopaminechrome, and o- and p -semiqui-
OXIDATION
PATHWAY
OF TYROSINASE-CATALYZED
0 350
450
h(nm)
550
429
DOPAMINE
1.2 0.6 Amj -Amjr/Anj
1.8 -Anj
FIG. 2. Rapid-scan assays for the oxidation of dopamine catalyzed by tyrosinase. Reagents: 1.0 mM dopamine, 0.26 mM O,, and 40 pg/ml tyrosinase. Other conditions as detailed under Materials and Methods. (A) Iterative spectrum. (B) Matrix analysis of the spectrum. A,, = Absorbance value at wavelength i and tracingj. n = 475 nm, n = 392 nm; i (nm) = 371 (O), 402 (A), 423 (), 450 (O), 491 (a), 511 (+), 532 (m);j’ = third recording.
nones were not detected can be explained by the same reasons as described for the cyclization branch. The hydroxylation branch, however, involved the accumulation of another intermediate, as was deduced from the nonappearance of an isosbestic point (Fig. 2A). This compound might have been GOHDA, which is not an absorbing species in the visible range. It is well-known that Dopamine oxidation byperiodate. sodium periodate oxidizes o-diphenols to their corresponding o-quinones (29). When [ dopamine] was greater than [ NaIO,] at slightly acid pH (Fig. 3A) the iterative spectrum was similar to that carried out with high tyrosinase concentrations (Fig. 2A). The presence of at least three absorbing species in solution was detected by matrix analysis (Fig. 3B). At higher pH values ( 19)) a great similarity with the respective enzymatic assays (described in the above section) was also obtained.
0 350
When [NaIO,] was greater than [dopamine] at pH values lower than 6.0, a set of recordings which showed the presence of two absorbance maxima at 390 and 470 nm was obtained, as well as one isosbestic point at 398 nm (Fig. 4A). The existence of a minimum of two absorbing species in solution was detected by matrix analysis (26)) a set of straight lines intersecting at the origin of the coordinates being obtained (Fig. 4B). Similar results were reported at pH values higher than 6.0 ( 19)) although the isosbestic point appeared at 403 nm. The periodate anion might act as oxidant in the following reactions of the pathway depicted in Fig. 1: dopamine + o-dopaminequinone, leukodopaminechrome + dopaminechrome, and GOHDA + o-topaminequinone. The nonaccumulation of o-semiquinones in these experiments (Figs. 3A and 4A) was also due to the absence of spin-stabilizing metal ions ( 9 ) . When [ NaIO, ] was lower
-0.5L -0.8 450
htnm)
550
-0.7
-0.6 Amj -A,jf/A"j
-0.5 -Anjp
FIG. 3. Rapid-scan assays for the oxidation of dopamine by a deficit of NaIO,. Reagents: 2.7 mM dopamine and 0.03 mM NaIO,. Other conditions as detailed under Materials and Methods. (A) Iterative spectrum. (B) Matrix analysis of the spectrum. m = 510 nm, n = 390 nm; i (nm) = 405 (O), 417 (B), 444 (O), 460 (0). 482 (A), 530 (A);j’ = first recording.
430
GARCiA-MORENO
ET AL.
FIG. 4. Rapid-scan assays for the oxidation of dopamine by an excess of NaIO*. Reagents: 0.27 mM dopamine and 2.7 mM NaI04. Other conditions as detailed under Materials and Methods. (A) Iterative spectrum. (B) Matrix analysis of the spectrum. i’ = 475 nm; i (nm) = 370 (O), 415 (A), 435 (Cl), 494 (m), 505 (A), 530 (0);j’ = first recording.
than [dopamine] the periodate anion was exhausted in the stoichiometric formation of o-dopaminequinone (Fig. 3A), in a parallel way to the oxygen depletion catalyzed by tyrosinase (Fig. 2A). The similarity between the tyrosinase and periodate assays confirms the nonenzymatic breakdown of o-dopaminequinone (Fig. 1) at slightly acid pH (Figs. 2 and 3 ) and neutral pH values ( 19 ) . When [ NaIO,] was higher than [dopamine], a stoichiometric quantity of periodate was consumed in the formation of o-dopaminequinone (Fig. 1) , which underwent nonenzymatic cyclization or hydroxylation to yield leukodopaminechrome or GOHDA, respectively. These intermediates were not accumulated in solution since they were also oxidized by the excess of periodate to provide the corresponding dopaminechrome or o-topaminequinone. The nonaccumulation of the intermediates explained the appearance of one isosbestic point, whose hipsochromic displacement at slightly acid pH (Fig. 4A) was due to the simultaneous operation of both branches of the pathway (Fig. 1). Under these conditions, the matrix analysis detected at least two absorbing species, odopaminequinone and dopaminechrome/o-topaminequinone. These two products were generated in a constant ratio during the whole assay time, this ratio being determined by the cyclization and the hydroxylation steps, respectively. Thus, one of the two compounds was considered as linearly dependent on the other by the matrix analysis (26).
The cyclic voltammogram corresponding to the dopamine oxidation by OJtyrosinase at pH 7.0 (Fig. 5) showed one cathodic peak and its corresponding anodic peak. These peaks corresponded to the reduction of the dopaminechrome to leukodopaminechrome (cathodic peak) and to the oxidation of the leukodopaminechrome to dopaminechrome (anodic peak) (28) ; therefore, at these pH values, only the cyclization branch of the pathway was significant. However, cyclic voltammograms obtained for pH values lower than 6.0 after oxidation of dopamine by both O,/ tyrosinase and sodium periodate, consisted of two cathodic and two anodic peaks (Figs. 6A and 6B). Figure 6A shows the cyclic voltammogram corresponding to the dopamine oxidation by tyrosinase at pH 4.2. The cathodic peak obtained at a potential of -0.06 V vs SCE was due to the reduction of dopaminechrome to leukodopaminechrome; its anodic peak appeared at -0.02 V vs SCE and was due to the oxidation of leukodopaminechrome to dopaminechrome, in accordance with the cyclic voltammogram of dopaminechrome at this pH (results not shown). The cathodic peak located at more positive po-
0.5 I (JJN
Cyclic Voltammetry The oxidation pathway of dopamine catalyzed by tyrosinase leads to the production of p-topaminequinone and dopaminechrome (Fig. 1) . Since the X,,, for these compounds are very similar (485 and 480 nm, respectively) , it is difficult to determine spectrophotometrically whether the two products are formed at the same time and pH values. However, cyclic voltammetry provides a solution to this question.
-0.51, I -0.1 -0.2
I I -0.3 -04 E(V)
I
FIG. 5. Cyclic voltammogram for the oxidation of dopamine at neutral pH. Reagents: dopamine (1.0 mM) with tyrosinase (5.0 fig/ml) in 0.1 M phosphate buffer, at pH 7.0.
OXIDATION
03
PATHWAY
0
- 0.1
OF TYROSINASE-CATALYZED
-0.2
01
431
DOPAMINE
0
-0.1
- 0.2
E (VI
E(V)
FIG. 6. Cyclic voltammogram for the oxidation of dopamine at acid pH. (A) Reagents: dopamine (1.0 mM) with tyrosinase M acetate buffer, at pH 4.2 (B) Reagents: dopamine ( 1.0 mM) with NaI04 (0.3 mM) in 0.1 M acetate buffer, at pH 3.5.
tential, +0.05 V vs SCE, could be ascribed to the reduction of p -topaminequinone to GOHDA. Its corresponding anodic peak appeared at the potential value of +0.06 V vs SCE, and represented the oxidation of GOHDA to p-topaminequinone, in accordance with the cyclic voltammogram of GOHDA at this pH (results not shown). Similar results were obtained when dopamine was oxidized by a deficit of NaI04 at pH 3.5 (Fig. 6B ) . These results confirm the oxidation pathway of dopamine here proposed (Fig. 1) . Kinetic
(15 yg/ml)
in 0.1
(19). This kinetic behavior has not yet been analyzed and could be useful for the quantitative characterization of the pathway. The absolute values of the steady-state rate of the enzymatic step were determined by measuring oxygen consumption (Fig. 8) due to the presence of two absorbing products at 480 nm, whose relative proportions are not known. These rates were one-half that of o-dopaminequinone-H+ formation (Fig. 1) , since one molecule of oxygen generated two molecules of quinone in each turnover (30).
Assays
The progress of dopamine oxidation catalyzed by tyrosinase was followed by measuring the appearance of products at 480 nm throughout the assay time. The experimental recordings (Fig. 7A) presented a marked lag period (7). At slightly acid pH, the lag period decreased when the enzyme concentration rose (Fig. 7B ) , whereas no dependence between them was obtained at greater pH values
Kinetic
Analysis
The oxidation pathway of dopamine catalyzed by tyrosinase at values of pH lower than 6.0 is depicted in Fig. 1. This scheme involves the steady-state rate of the enzymatic step (V,) , first (k, and k,) and second (k-, , k,, k3, ke3, k4, kp4, kg, kp5, kg, and ke6) order rate constants. The nonappearance of an isosbestic point in the rapid scan assays at pH < 6 (Figs. 2 and 3) unlike in the same experiments at pH > 6 (19) supported the accumulation
A
FIG. 7. Kinetic assays of product accumulation for the oxidation of dopamine catalyzed by tyrosinase. Reagents: 2.7 mM dopamine and 0.26 mM O2 at, pH 5.22. Other conditions as detailed under Materials and Methods. (A) Spectrophotometric recording. Tyrosinase (pg/ml) : 2.5, 3.3, 5.0, 6.7, 8.3. (B, 0) Corresponding values of the lag period (7) vs enzyme concentration (Es). (B, A ) Plot of products accumulation rate in steady-state conditions against E,,.
FIG. 8. (A) Kinetic assays of oxygen consumption for the oxidation of dopamine catalyzed by tyrosinase. Reagents: 2.7 mM dopamine, 0.26 mM O2 and tyrosinase (@g/ml): 2.5, 3.3, 5.0, 6.7, 8.3 at pH 5.22 (B) Corresponding values of V,, vs E,.
432
GARCiA-MORENO
of one intermediate. This compound should have a diphenolic structure since it was oxidized by an excess of NaI04, as shown by the appearance of one isosbestic point in the corresponding rapid-scan assays at pH < 6 (Fig. 4). This intermediate could be GOHDA, since leukodopaminechrome was not accumulated in the assay medium (9, 19, 28). The cyclic voltammetry assays at pH < 6 showed the simultaneous operation of both the cvclization and the hydroxylation branches of o-dopaminequinone (Fig. 6)) in contrast to a separate pattern for GOHDA. Therefore, the accumulated intermediate could be GOHDA (Fig. 1) , which would imply a significant contribution of the second-order step controlled by k4 and km4, to the kinetic behavior of the pathway. However, there are no analytical solutions to describe the evolution of the products during the transient phase of the kinetic assays shown in Fig. 7A. At the steady-state of the pathway, however, it is possible to derive explicit equations relating the product concentrations with the assay time. Kinetic analysis of this pathway requires that the dopamine consumption as well as the breakdown of the products should be negligible during the reaction time. Both conditions have been verified in the experimental assays. The mass balance between the reagents states that the consumed o-diphenol (Vat) yields intermediates ( Z[ I ] ) , products ( [DC ] and [ PQ ] ) and regenerated odiphenol, through the cyclization and the hydroxylation branches of the pathway: V,t+
k,[D][QH]t
ET AL.
accumulated during the assays (see Rapid-Scan section). Therefore, the net formation of PQ is equivalent to k4[ T] [ QH] t, whereas the net production of DC is determined by k3[ QH ] [ L ] t, and Eq. [l] is transformed to 2([DC]
+ [PQ])
[PQ]t
[ Qfi ]
= Vo+ k-dH+l [&I - MQHI - kc[Ql - %z[QHl - MQHI [Tl = 0 [a]
= k,[QH]
- (k-,[H+l
- k,[QH][L]t
= Z[I]
- k,[ + [DC
2111= [&HI + [&I + IL1 + [OSQI + [PSQ + [Tl. The generation and breakdown steps of semiquinones in rapid equilibrium ( 9 ) :
[21 are
biDI [QHI = k+,[OSQ12,kITI [PQI = kps[PSQ]‘.
[3]
There are reversible steps displaced toward the regeneration of dopamine ( 19) :
MQHI [Ll B LADCI [Dl, JZdQHl[Tl ti kp,[PQ
[Dl.
[41
Furthermore, the contribution of L, OSQ, and PSQ to Eq. [ 2 ] is negligible since these intermediates are not
= 0
= 0,
[61
where k; = k,[ Hz01 . Note that the term k3[ QH] [L] has been replaced by k,[Q ] in Eq. [ 61, since in the steady state both terms are equivalent. From the above system of differential equations, expressions for the concentration of the intermediates at the steady state are obtained:
[QHI,, = 2klkc
Consideration expression
(k-,[H+l + k,)Vo + 2k’,(k-,[H+]
+ k,)
k,Vo
[Qlss= 2klk,
+ 2kL(kp,[H+]
ITl,,
rl [QHlt + [PQI PI
+ k,)[Q]
[?I = k’,[QHl - h[QHI[Tl
+ km,[D][DC]t
- kpg[OSQ]‘t
151
Z[I].
During the steady state of the pathway, the concentration o f the intermediates remains constant:
_
+ k-,[D]
= Vat-
_“I
= k:lk,. _,
+ k,) 171
I
of Eqs. [l] - [ 51 and [ 71 leads to the
[DClss+ [PQlss kl + kp,[H+] + k, 4 2klk, + 2k’,(k-,[H+] + k,) + k4Vo
11
. [81
Thus, the accumulation of the products of the pathway during the steady state follows a straight line (Eq. [ 81) , whose intercept with the time axis defines the lag period (7): kl + k-,[H+] + kc 4 7 = 2klk, + 2k’,(kp,[H+] + kc) + $V, ’
[91
Since k, % kl ( 19)) the previous equation becomes k-,[H+] + k, kh ’ = 2k1k, + 2k;(k-,[H+] + k,) + k4V0.
WI
OXIDATION
PATHWAY
l/v,
OF TYROSINASE-CATALYZED
433
DOPAMINE
[H’l pM
PM-‘s
FIG. 9. Kinetic assays of the oxidation of dopamine catalyzed by tyrosinase, at different pH values for several enzyme concentrations. Values of pH 4.98, 5.16, 5.22, 5.48, 5.68, 6.60, 6.90; tyrosinase (fig/ml) for each pH: 1.7, 2.5, 3.3, 5.0, 6.7, 8.3, 10.0, 11.7, 13.3, 16.7. Other assay conditions are as detailedunder Materials and Methods. (A) Dependence of the lag period (7) on l/V,, kinetic parameters obtained from spectrophotometric and oxymetric recording, respectively. Experimental data and linear regression fittings are shown. (B) Corresponding values of 2cu and of fl vs [H-l. (0, 0) Experimental data of 2u and of 0, respectively. (0 ----- 0) Data calculated using Eq. [14], with the initial estimations of their kinetic constants. (0 0) Data calculated using Eq. (141, with the final estimations of their kinetic constants, obtained from nonlinear regression fitting. (0 ~ 0) Mean value of fi vs [H+] , according to Eq. [15]
This equation concentration experimental ues, however, tration (19). + k, ) < k,k,,
predicts that r decreases when the enzyme ( E0 involved in V,) is raised, according to data at acid pH (Fig. 7). At greater pH valr is not dependent on the enzyme concenThus, when [H+] decreases, 2k’,( kmi[H+] ( k; 4 k, ) and Eq. [lo] is transformed into 7=
kp,[H+]
+ k, +
2k,k,
k’, kvo.
[Ill
Furthermore, since V, increases when pH is raised, a negligible contribution of the second quotient as regards the first leads to T = (km,[H+]
+ k,)/2klk,,
1 21
an expression not dependent on the enzyme concentration. On the other hand, Equation [lo] can be rearranged as 7 = 2a + 0(1/V,,),
P31
where
2a =
$ [H+] 2 [H+]
+ $$ 1 2 +e
Kinetic
Data Analysis
Two series of spectrophotometric (Fig. 7) and electrometric (Fig. 8) assays were carried out by using the same set of enzyme concentrations. From AdBOvs t recordings, data of the final portion were fitted by linear regression and the corresponding r vs E0 values were calculated. The initial zone of the [ 0,] vs t recording was fitted by linear regression, yielding the respective values of V,,/2 vs EO. The kinetic assays, therefore, provide T vs ( 1 lb’,,) data at the same E, values. This procedure was applied to several pH values (Fig. 9A) and the data fitted by linear regression according to Eq. [ 13 1. The great similarity between the slopes of the straight lines (Fig. 9A) confirmed the nondependence of p on pH (Eq. 15). Thus, the mean values of the slopes enabled k’,/ k4 to be calculated (Fig. 9B). The intercepts on the ordinate axis, however, showed a hyperbolic behavior on [H+] (Fig. 9B ) . The nonlinear regression fitting of 2cuvs [H’ ] data to Eq. [14] permitted the determination of all the rate constants of the pathway, by considering K, = 2.51 10 -I1 M (31) . Initial estimations of kl , kb , and k, were obtained, from the lower limit, higher limit and half-saturation value of the hyperbola, respectively (Fig. 9B). The values of the rate constants of the pathway (Fig. 1) are described in Table I.
[I41 Effect of pH on the Dopamine
2
and
P = k;/k,
K, = k, / km,. These expressions are useful to determine the rate constants of the pathway.
[I51
Oxidation
Pathway
The spectral and the kinetic assays described above support and characterize the proposed oxidation pathway of dopamine (Fig. 1) . The pH of the medium determines the differential contribution of the cyclization and the
434
GARCIA-MORENO TABLE
I
Values of the Rate Constants for the Oxidation Pathway of Dopamine Catalyzed by Tyrosinase (Fig. l), at 20°C Constant k, k, k:j k4
Units s ~’ M-’ 5-l
8 -’
M-' s-1
ET AL. 3. 4. 5. 6.
Value 0.12 (8.11 849.7 640.1
f 0.02 f 0.65) X 10m5 t- 178.4 + 96.2
7. 8. 9. 10.
hydroxylation branches, through the protonation/deprotonation of the key intermediate, o-dopaminequinoneH+. Thus, the above experimental data show that both branches are only simultaneously significant at slightly acid pH. Indeed, there is evidence concerning the acidic character of melanosomes ( 20-22) and tumoral cells (23-24). Antitumoral activity has been found much greater for catechols and catecholamines forming noncyclizable quinones than for those forming cyclizable quinones (5), perhaps due to the stability of noncyclizable quinones (32) that react with nucleophilic groups of cellular molecules and macromolecules (6, 8, 13-17). Dopamine at neutral pH formed a cyclizable quinone, but at slightly acid pH the addition of water to the quinone ring generates GOHDA (Fig. 1) , which forms a quinone that cyclizes slowly (32). This increases the cytotoxic and antitumoral power of dopamine (8). The production of semiquinones might be increased under physiological conditions, due to the presence of spin-stabilizing ions. These ions have been found in a variety of melanized structures (33)) including the choroid of the eye (34-35)) black hair ( 36)) and isolated melanosomes from Harding-Passey, horse, and human melanomes (37). The reaction between semiquinones and sulf hydryl aminoacids such as cysteine probably triggers a cascade of free-radical reactions. Therefore, the production of even small amounts of semiquinones in the pathway of oxidation of catechols and catecholamines may have toxicological significance ( 9 ) . Furthermore, reagents causing changes in the pH of the physiological microenvironment might determine the actual cytotoxicity of dopamine and other catecholamines.
11. 12. 13. 14. 15. 16. 17. 18. 19.
20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30.
31. 32.
ACKNOWLEDGMENTS This paper has been partially supported by a grant from the CICYT (ALI89-674)) DGICYT (PB 87-0700)) and CARM (PB 90/37). J. N. Rodriguez-Lopez has received a fellowship from the Comunidad Autonoma de Castilla-La Mancha (Spain).
REFERENCES 1. Harrison, W. H. (1963) Arch. Biochem. Biophys. 101, 116-130. 2. Robb, D. A. ( 1984) in Copper Proteins and Copper Enzymes (Lontie, R., ed.), Vol. II, pp. 2077240, CRC Press, Boca Raton.
33. 34. 35. 36. 37.
Lerch, K. (1983) Mol. Cell. Biochem. 52, 125-138. Wick, M. M. (1978) J. Invest Dermatol. 71, 163-164. Wick, M. M. (1979) Cancer Treat. Rep. 63, 991-997. Wick, M. M., Byers, L., and Ratliff, J. (1979) J. Incest. Dermatol. 72,67-69. Wick, M. M., Byers, L., and Frei, F. (1977) Science 197,468-469. Graham, D. G., Tiffany, S. M., and Vogel, S. F. (1978) J. Invest. Dermatol. 70, 113-116. Korytowski, W., Sarna, T., Kalyanaraman, B., and Sealy, R. C. (1987) Biochim. Biophys. Acta 924.383-392. Pomerantz, S. H., and Li, J. P. C. (1973) Yale J. Biol. Med. 46, 541-552. Nishioka, K. (1978) Eur. J. Biochem. 85, 137-146. Burnett, J. B. (1971) J. Biol. Chem. 246,3079-3091. Saner, A., and Thoener, H. (1971) Mol. Pharmacol. 7, 147-154. Rothman, A., Daly, J. W., and Creveling, C. R. (1976) Mol. Pharmacol. 12,887-899. Jimenez, M., Garcia-Canovas, F., Garcia-Carmona, F., Iborra, J. L., and Lozano, J. A. (1986) Znt. J. Biochem. 18, 161-166. Cabanes, J., Garcia-Canovas, F., and Garcia-Carmona, F. (1987) Biochim. Biophys. Acta 914, 190-197. Garcia-Carmona, F., Cabanes, J., and Garcia-Canovas, F. (1987) Biochim. Biophys. Acta 914, 198-204. Langston, J. W. (1988) Trends. Pharmacol. Sci. 9,347-348. Jimenez, M., Garcia-Carmona, F., Garcia-Canovas, F., Iborra, J. L., Lozano, J. A., and Martinez, F. (1984) Arch. Biochem. Biophys. 235,438-448. Sacki, H., and Oikawa, A. (1983) J. Cell. Physiol. 116, 93-97. Sacki, H., and Oikawa, A. (1985) J. Znuest. Dermatol. 85,423-425. Moelhman, G., Slominski, A., Kuklinska, E., and Lerner, A. B. ( 1988) Pigment. Cell. Res. Suppl. 1, 79-87. Wike-Hooley, J. L., Haveman, J., and Reinhold, H. S. (1984) Radiother. Oncol. 2, 343-366. Hedley, D. W., and Jorgensen, H. B. (1989) Exp. Cell. Res. 180, 106e116. Liang, Y. O., Plotsky, P. M., and Adams, R. N. (1977) J. Med. Chem. 20, 581-583. Coleman, J. S., Varga, L. P., and Mastin, S. H. ( 1970) Znorg. Chem. 9,1015-1020. Hartree, E. (1972) Anal. Biochem. 46422-427. Adams, R. N., Murril, E., McCreery, R., Blank, L., and Karolczak, M. (1972) Eur. J. Pharmacol. 17, 287-292. Graham, D. G., and Jeffs, P. W. (1977) J. Biol. Chem. 252,57295734. Galindo, J. D., Pedreiio, E., Garcia-Carmona, F., Garcia-Canovas, F., Solano, F., and Lozano, J. A. (1983) Znt. J. Biochem. 15,14551461. Gray, D. O., and Weitzman, P. D. J. (1968) Data for Biochemical Research, p. 22, Clerendon Press, Oxford. Blank, C. L., McCreery, R. L., Wightman, R. M., Chey, W., Adams, R. M., Reid, J. R., and Smissman, E. E. (1976) J. Med. Chem. 19, 178-180. Palumbo, A., D’Ischia, M., Misuraca, G., and Prota, G. ( 1987) Biochim. Biophys. Acta 925, 203-209. Bowness, J. M., and Morton, R. A. (1953) Biochem. J. 53, 620626. Underwood, E. J. ( 1971) in Trace Elements in Human and Animal Nutrition, 3rd Ed., Academic Press, New York. Dorea, J. G., and Pereira, S. E. (1983) J. Nutr. 113,2375-2381. Horcicko, J., Borovansky, J., Duchon, J., and Prochazkova, B. (1983) Hoppe-Seyler’s Z. Physiol. Chem. 354, 203-204.
