Chlorine Dioxide Oxidations of Tyrosine - American Chemical Society

biological warfare agents such as anthrax (4). Health and nutrition ..... (1) Katz, A., and Narkis, N. (2001) Removal of chlorine dioxide disinfection...
0 downloads 0 Views 231KB Size
Download PDF
Chem. Res. Toxicol. 2005, 18, 501-508

Chlorine Dioxide Oxidations of Tyrosine, N-Acetyltyrosine, and Dopa Michael J. Napolitano, Brandon J. Green, Jeffrey S. Nicoson, and Dale W. Margerum* Department of Chemistry, Purdue University, West Lafayette, Indiana 47907-2084 Received November 3, 2004

The reactions of aqueous ClO2 with tyrosine, N-acetyltyrosine, and dopa (3,4-dihydroxyphenylalanine) are investigated from pH 4 to 7. The reaction rates increase greatly with pH to give a series of oxidized products. Tyrosine and N-acetyltyrosine have similar reactivities with second-order rate constants (25.0 °C) for their phenoxide forms equal to 1.8 × 108 and 7.6 × 107 M-1 s-1, respectively. Both species generate phenoxyl radicals that react rapidly with a second ClO2 at the 3 position to give observable but short-lived adducts with proposed C(H)OClO bonding. The decay of these phenoxyl-ClO2 adducts also is rapid and is base-assisted to form dopaquinone (from tyrosine) and N-acetyldopaquinone (from N-acetyltyrosine) as initial products. The consumption of two ClO2 molecules corresponds to a four-electron oxidation that gives ClO2- in the first step and HOCl in the second step. The reaction between ClO2 and the deprotoned-catechol form of dopa is extremely fast (2.8 × 109 M-1 s-1). Dopa consumes two ClO2 to give dopaquinone and 2ClO2- as products. Above pH 4, dopaquinone cyclizes to give cyclodopa, which in turn is rapidly oxidized to dopachrome. A resolved first-order rate constant of 249 s-1 is evaluated for the cyclization of the basic form of dopaquinone that leads to dopachrome as a product with strong absorption bands at 305 and 485 nm.

Introduction Aqueous ClO2 has been used in water treatment plants in the United States and Europe (1, 2). Uses of gaseous ClO2 include chemosterilization (3) and disinfection of biological warfare agents such as anthrax (4). Health and nutrition concerns have contributed to a 27% increase in the consumption of fresh fruits and vegetables from 1970 to 1993 (5). Pathogenic outbreaks associated with fresh produce (5-7) have caused increased concerns about the safety of our foods. The use of ClO2 as an aqueous sanitizing agent in a multitude of food industries is growing with applications in vegetable processing (810), citrus packing (11), fish (12-14), meat (15), and poultry processing (16, 17). Our purpose in studying the reaction of ClO2 with tyrosine (Tyr), N-acetyltyrosine (NAT) and dopa (Scheme 1) is to determine the reaction products and to gain mechanistic insight about the reactions of ClO2 with amino acids and related compounds. Previous work showed that tyrosine, cysteine, methionine, proline, tryptophan, and histidine react much more rapidly than other amino acids with ClO2 (18, 19), but their reaction products have not been well characterized. Other oxidations of tyrosine and dopa have been performed by tyrosinase/O2 (20-22), IO4- (20), HOCl (23-25), and silver oxide (21). Our work shows that from p[H+] 4 to 7 dopaquinone (Scheme 1) is an intermediate product of Tyr and dopa reactions with ClO2. Similarly, the reaction between NAT and ClO2 produces N-acetyldopaquinone. Dopaquinone cyclizes to cyclodopa, which then forms dopachrome at pH > 4 in the presence of an oxidizing agent (26). Our experimental evidence shows that 2 mol of ClO2 react with each mole of Tyr, NAT, or dopa. These results agree

Scheme 1. Structures of Species and Their UV-Vis λmax Values

10.1021/tx049697i CCC: $30.25 © 2005 American Chemical Society Published on Web 02/18/2005

501

502

Chem. Res. Toxicol., Vol. 18, No. 3, 2005

with studies of phenol and hydroquinone reactions with ClO2 by Wajon and co-workers (27). Phenol shows a parallel reactivity with tyrosine and NAT, and hydroquinone parallels the reactivity of dopa. Wajon proposed rapid formation and decay of a phenoxyl-OClO intermediate, but was not able to observe these reactions. In the present studies of ClO2 with Tyr and NAT, we find spectral evidence of a phenoxyl-OClO intermediate that forms as a second ClO2 is consumed and decays to give the initial quinone products. The reaction of dopa and ClO2 is similar to the reaction of hydroquinone and ClO2. Wajon reported a second-order rate constants of 2.4 × 107 M-1 s-1 for the reaction of phenolate ion with ClO2 and 6.5 × 109 M-1 s-1 for the reaction of hydroquinone anion. Subsequently, Neta et al. (28) reported secondorder constants of 2.9 × 107 M-1 s-1 for the phenol/ClO2 reaction and 1.5 × 109 M-1 s-1 for the ClO2 reaction with hydroquinone. Neta also reported rate constants for the reactions of p-cresol, p-methoxyphenol, and resorcinol with ClO2 as 7.4 × 108, 2.6 × 108 and 1.4 × 109 M-1 s-1, respectively. The reported kinetics of phenols and substituted phenols can serve as benchmarks for our rate constants with ClO2. Our ability to monitor these fast reactions through the use of stopped-flow diode -array instruments allow us to observe a complex series of reactions following ClO2 decay.

Napolitano et al.

Figure 1. Photodiode difference spectra for the reaction between ClO2 and Tyr shown on a log time scale (The difference spectrum is obtained by subtracting the spectrum of Tyr under these conditions from the spectral data for the reaction mixture). The intermediate at 245 forms as ClO2 decays. ClO2 decays in 50 ms while dopachrome (305, 485 nm) forms in ∼10 s. Conditions: [Tyr] 2.00 mM, [ClO2] 0.187 mM, p[H+] 6.31, [PO4]T 25.0 mM, µ ) 1.0 M (NaClO4), 25.0 °C, and a path length of 0.962 cm.

Experimental Procedures Reagents. All solutions were prepared with doubly deionized, distilled water. Sodium perchlorate (NaClO4) was recrystallized and standardized gravimetrically prior to use. Tyrosine and NAT (Bachem) and dopa (Acros Organics) were used without further purification. Tyrosine stock solutions (∼0.1 M) were freshly prepared with addition of NaOH (∼0.2 M) to increase solubility. Dopa and NAT stock solutions did not require the addition of NaOH to enhance solubility and were freshly prepared to prevent microbial growth. ClO2 was prepared and standardized as previously reported (29) and stored in a dark refrigerator to hinder decomposition. Commercially available sodium chlorite (NaClO2) was recrystallized and standardized before use (30). All other chemicals were reagent grade quality and were used as received. Methods. The pH measurements were corrected to give p[H+] values ()-log[H+]) based on electrode calibration at 1.0 M ionic strength controlled by addition of NaClO4 (29). UV-vis spectra and kinetic spectra were obtained from a Perkin-Elmer Lamda-9 UV-vis-NIR spectrophotometer (1.00 cm path length) or an Applied Photophysics Stopped-Flow SX.18 MV (APPSF) spectrophotometer (0.962 cm path length) with PD.1 photodiode array or single-wavelength detector. Photodiode array difference spectra as a function of time were obtained by subtraction of the spectrum of Tyr or NAT from that of the reaction mixture of these species with ClO2. Single wavelength stopped-flow reactions on the APPSF instrument with pseudofirst-order rate constants greater than 80 s-1 were corrected for mixing by use of a mixing rate constant of kmix ) 4620 s-1, where kcorr ) (1/kobs - 1/kmix)-1 (31). Chromatographic analysis was performed with a Varian 5020 HPLC with a Hewlett-Packard 1050 diode array detector. A Whatman Partisil 5 ODS-3 C18 column was used for the reversed-phase (RPLC) determination of tyrosine.

Results and Discussion Spectral Observations of the ClO2 + Tyrosine/ NAT/Dopa Reactions. The reactions of ClO2/Tyr, ClO2/ NAT and ClO2/dopa were investigated from p[H+] 4 to 7. The kinetics of ClO2 oxidations were studied by stopped-

Figure 2. Photodiode difference spectra for the reaction between ClO2 and NAT. Intermediate at 240 nm which form as ClO2 decays, spectra were collected every 5.0 ms. The loss of the intermediate is concurrent with the growth of N-acetyldopaquinone, 393 nm. Conditions: [NAT] 1.00 mM, [ClO2] 0.20 mM, p[H+] 5.58, [PO4]T 50.0 mM, µ ) 1.0 M (NaClO4), 25.0 °C, and a path length of 0.962 cm.

flow methods with photodiode array or single-wavelength detection. Kinetic photodiode difference spectra are shown in Figure 1 for the reaction of ClO2 (0.187 mM) with tyrosine (2.00 mM) at p[H+] 6.31. The noise at 275 nm is due the large signal from Tyr that is subtracted. Spectral changes during the first 50 ms are due to the loss of ClO2 at 359 nm and the formation of an intermediate at 245 nm. Subsequent large absorbance increases in 10 s occur at 305 and 485 nm, where dopachrome has absorption bands. Kinetic photodiode difference spectra for the reaction between ClO2 (0.20 mM) and NAT (1.00 mM) at p[H+] 5.58 are shown in Figure 2. The loss of ClO2 at 359 nm (3-100 ms) and the formation of an intermediate at 240 nm is observed initially (3-70 ms). The 240 nm intermediate reaches a maximum absorbance in 70-100 ms and then decays during 0.1 to 2.0 s. This is accompanied

Chlorine Dioxide Oxidations of Tyr, NAT, and Dopa

Chem. Res. Toxicol., Vol. 18, No. 3, 2005 503 Table 1. Products and λmax Values of Catechol Oxidations by O2 and Observed Absorbances of Tyrosine and N-Acetyltyrosine Oxidations by ClO2 substrate

oxidant

pyrochatechola 4-tert-butylcatechola

O2 O2

N-acetyldopaminea

O2

N-acetyldopa ethylesterb N-acetyltyrosinec dopaa dopac tyrosinec

O2 ClO2 O2 ClO2 ClO2

product o-benzoquinone 4-tert-butyl-o-benzoquinone N-acetyldopamine quinone N-acetyldopaquinone ethylester N-acetyldopaquinone dopachrome dopachrome dopachrome

λmax (nm) 390 400 392 390 393 305, 480 305, 485 305, 485

a Enzymatic oxidation (phenol oxidase), ref 32. b Enzymatic oxidation (mushroom tyrosinase), ref 33. c Present study.

Figure 3. Photodiode difference spectra for the reaction between ClO2 and dopa shown on a log time scale. The decay of ClO2 is very fast and not observed at this pH, but growth of dopaquinone (400 nm) and subsequent loss to form dopachrome (305, 485 nm) in ∼10 s. Conditions: [dopa] 1.00 mM, [ClO2] 0.20 mM, p[H+] 6.37, [PO4]T 0.100 M, µ ) 1.0 M (NaClO4), 25.0 °C, and a path length of 0.962 cm.

by the growth of a species at 393 nm. Absorbance changes at 305 and 485 nm are not observed in this reaction because dopachrome is not formed due to the presence of an acetyl group that prevents cyclization. The assignment for the absorbance at 393 nm is N-acetyldopaquinone. Structures of the reactants and products are given in Scheme 1. The initial reaction between dopa and ClO2 is much faster than for Tyr or NAT. Figure 3 shows results for 0.20 mM ClO2 with 1.00 mM dopa at p[H+] 6.37 where the loss of ClO2 is not shown because it is too fast at this pH. Subsequent spectral changes are similar to those with Tyr except that no intermediate is seen at 240245 nm. A slower loss of absorbance at 395 nm with an accompanying large increase in absorbance at 305 and 485 nm is found due to the formation of dopachrome within 10 s (Figure 3). Our assignment of the ClO2 oxidation products as dopaquinone (395 nm), dopachrome (305 and 485 nm), and N-acetyldopaquinone (393 nm) is possible because of previous studies by Waite and co-workers (32, 33). They have shown that the oxidation of catechols by both IO4- and mushroom tyrosinase/O2 to their corresponding o-quinones give products with absorption maxima near 400 nm. The results of our work and Waite’s work are summarized in Table 1. Waite’s oxidation of N-acetyldopamine to the quinone and the oxidation of NAT with ClO2 have similar λmax values of 392 and 393 nm, respectively. A comparison of the oxidation of dopa by Waite, and the oxidations of tyrosine and dopa with ClO2 show that dopachrome is formed. Similar studies by other groups (20, 34-36) gave comparable results. Kerwin (37) used tandem negative ion electrospray mass spectrometry (ESMS/MS) to show that the (M - H)-1 fragment corresponds to dopachrome when it was generated by oxidation of dopa using tyrosinase, Ag2O, or NaIO4. Thus, the assignments from the UV-Vis data and the mass spectroscopic data agree. Reaction Stoichiometry. Stopped-flow studies of dopachrome formation at 485 nm as various ratios of

Figure 4. Absorbance at 485 nm due to to dopachrome formation upon mixing Tyr with ClO2. The absorbance increases linearly up to a Tyr/ClO2 ratio of 1:2. Excess ClO2 causes a decrease in the observed absorbance. Data collected with the APPSF spectrophotometer with single wavelength detection. Conditions: [Tyr] 1.00 mM, [PO4]T 75.0 mM, p[H+] 6.93(2), µ ) 1.0 M (NaClO4), 25.0 °C, and a path length of 0.962 cm. Table 2. HPLC Measurements of Tyrosine Recovery after Flow Mixing of 5 mL Each of ClO2 and Tyr Solutionsa initial [ClO2], mM

initial [Tyr], mM

% Tyr recovery

p[H+]

0.25 0.50 0.75

0.25 0.25 0.25

40 5.9 0

7.02 7.03 7.03

a Analyses were performed 15 min after mixing. Reaction conditions: [PO4]T ) 20 mM, p[H+] ) 7.03, µ ) 0.1M (NaClO4). Separation conditions: eluent 10% (v/v) methanol/H2O, flow rate 0.5 mL/min, detection 225 nm.

ClO2 are added to Tyr at p[H+] 6.93 are shown in Figure 4. The maximum absorbance increases linearly until a 2.0 ratio of [ClO2]added to [Tyr]initial is reached. The decrease in absorbance as more ClO2 is added is probably due to its reaction with dopachrome. In our other kinetic studies, the limiting reagent is ClO2 so that we do not see its reaction with the products because all the ClO2 is consumed before dopachrome forms. The stoichiometry was also tested by rapid-flow T-mixing of the reactants under 1:1, 2:1, and 3:1 molar ratios of ClO2 to Tyr (Table 2) with HPLC determination of tyrosine recovered after 15 min. All the Tyr is consumed when ClO2 is initially added in a 3:1 ratio, while only 5.9% is found with a 2:1 ratio, and 40% of the Tyr is found with the 1:1 ratio. The data are in general agreement with the above stoppedflow studies, but the results are less satisfactory. The

504

Chem. Res. Toxicol., Vol. 18, No. 3, 2005

Napolitano et al. Table 3. Rate Constants for the Reaction between ClO2 and Tyr, NAT, and Dopa and NAT-OClO Adduct Decay Rate Constants and Dopaquinone Ring Closurea

Figure 5. Plot of kobs for the loss of ClO2 vs p[H+] (a) and vs [Tyr] (b). Conditions: (a) [ClO2] ) 0.187 mM, [PO4]T ) 25 mM, µ ) 1.00 M (NaClO4), λ ) 360 nm, 25.0 °C, and a path length of 0.962 cm; (b) [ClO2] ) 0.075 mM, [PO4]T ) 0.100 M, p[H+] 6.18, µ ) 1.00 M (NaClO4), λ ) 360 nm, 25.0 °C, and a path length of 0.962 cm.

volatility of ClO2 is a source of error in the initial mole ratios used and the time delay before HPLC elution of the Tyr is another cause of concern because any HOCl formed could also react with Tyr (23-25). Hence these data are presented as supporting evidence that the ratio of reactants consumed is 2ClO2 per Tyr. It has been shown that reactions between ClO2 and aromatic compounds require two molecules of ClO2 per aromatic molecule (27, 38-40). Two examples are the reactions of ClO2 with phenol (27) and 1-(3,4-dimethoxyphenyl)ethanol (39). Wajon and co-workers have shown that ClO2 oxidizes phenol to benzoquinone. Other products observed were chlorinated phenols and chlorite. Although HOCl was not observed, the formation of the chlorinated phenols was attributed to the reaction of phenol with HOCl. Kinetic Analysis of ClO2 Loss and Product Formation. A first-order rate loss of ClO2 (eq 1) is observed

-

d[ClO2] ) kobs[ClO2] dt

(1)

when excess tyrosine is present. Plots of kobs for pseudofirst-order loss of ClO2 as a function of p[H+] and [Tyr] are shown in Figure 5a and 5b. The kobs values increase as the p[H+] increases, with a 1/[H+] dependence, and as [Tyr] increases. These very large first-order rate constants are corrected for mixing in the stopped-flow instrument (31). The rate expression was shown to be independent of the chlorite concentrations up to 5 mM ClO2- (at p[H+] 6.0, 0.10 M [PO4]T, and µ ) 1.0 M NaClO4) for reactions of 1.00 mM tyrosine with 0.20 mM ClO2. It is concluded that the rate of loss of ClO2 is first-order in

a

52.

Conditions: 25.0 °C, µ ) 1.0 M. b Reference 42. c Reference pKa dopa’s amino group. e Units are s-1.

d

[ClO2], first order in [Tyr] (or [NAT]), and inversely dependent on [H+]. The data are fit to eq 2, based on the

kobs ) 2k1Ka1[Tyr]/[H+]

(2)

2:1 stoichiometry and the acid dissociation constant, Ka, for the phenolic group. The Ka values for these systems and the measured second-order rate constants are given in Table 3. Wajon reported a second-order rate constant for the reaction between ClO2 and phenol to be 0.24 M-1 s-1. This value is 108 smaller than the resolved second-order rate constant for the ClO2/phenoxide reaction. Wajon used a pH range of 0-2 to obtain reliable values for the phenol reaction. It is probable that the phenol form of tyrosine also reacts with ClO2, but based on Wajon’s values we assume that this pathway does not contribute appreciably to the rate of ClO2 loss for the reactions between Tyr, NAT, and ClO2 in the pH range studied. Mechanisms of ClO2 Loss and Formation of Phenoxy-ClO2 Adducts. The proposed mechanism for the Tyr and NAT reactions is shown in Scheme 2, where the phenoxide form of these amino acids is the reactive species with ClO2. This accounts for the inverse [H+] dependence of the reactions (eq 2). The phenoxide species react with one molecule of ClO2 to generate a tyrosyl radical and one chlorite ion. A subsequent fast step between the tyrosyl radical and a second molecule of ClO2 generates a phenoxyl-ClO2 adduct that is observed at 245 nm. A first-order decay of this adduct leads to the formation of dopaquinone (from Tyr) or N-acetyldopaquinone (from NAT). Wajon et al. (27, 40) proposed a

Chlorine Dioxide Oxidations of Tyr, NAT, and Dopa

Chem. Res. Toxicol., Vol. 18, No. 3, 2005 505 Table 4. Observed Rate Constants for the Loss of ClO2 (2k1obs), Growth (k1obs) and Decay (k2obs) of the Intermediate, and Formation of N-Acetyldopaquinone (k2obs)a

p[H+]

2k1obs (359 nm), s-1

k1obs (240 nm), s-1

k2obs (240 nm), s-1

k2obs (359 nm), s-1

5.21b 5.62b 5.67c 5.83d 6.23c

6.8(1) 16.7(2) 20.1(1) 26.5(1) 72.4(5)

3.32(2) 8.40(5) 9.92(5) 13.17(8) 35.5(3)

2.3(1) 2.46(5) 2.47(3) 2.56(2) 3.16(3)

2.24(2) 2.30(1) 2.39(1) 2.41(1) 2.98(1)

a Reaction conditions: [ClO ] 0.20 mM, [NAT] 2.00 mM, µ ) 2 1.0 M (NaClO4). Data were collected using APPFS spectrophotometer using photodiode array detection; all traces were the average of three pushes. b Acetate buffer, [OAc]T ) 25.0 mM. c Phosphate buffer [PO ] ) 25.0 mM. d Phosphate buffer [PO ] 4 T 4 T ) 50.0 mM.

Figure 6. Kinetic traces of intermediate growth and decay at 240 nm at 359 nm (a). ClO2 decay and growth of N-acetyldopaquinone at 393 nm (b). Traces were obtained by APPSF spectrometer using photodiode array detection, each trace is the average of three pushes. Data shows growth of the intermediate and decay of ClO2 are concurrent and the loss of the intermediate corresponds to the formation of N-acetyldopaquinone. Conditions: [ClO2] ) 0.20 mM, [NAT] 2.00 mM [PO4]T ) 50.0 mM, p[H+] 5.83, µ ) 1.00 M (NaClO4), 25.0 °C, and a path length of 0.962 cm.

Scheme 2. Mechanism for the Formation of a Phenoxyl-ClO2 Adduct that Decays to Give Dopaquinone and N-Acetyldopaquinone

show the formation of an adduct with an absorption band at 245 nm, followed by its decay to give dopaquinone and HOCl. Similarly, one ClO2 reacts with NAT to form ClO2- and a NAT radical that reacts with a second ClO2 to form an adduct with an absorption band at 240 nm. The decay of this adduct also is observed (Figure 6a and b). We propose for the Tyr and NAT systems that the decay of the intermediate generates quinones and HOCl. Kinetic traces of the absorbance changes for the reaction between 2.00 mM NAT and 0.20 mM ClO2 at p[H+] 5.83 are shown in Figure 6a at 240 nm and in 6b at 359 nm. Figure 6a shows the growth and decay of the phenoxylClO2 adduct (i.e. NAT-OClO) at 240 nm. Figure 6b shows the loss of ClO2 and the formation of N-acetyldopaquinone at 359 nm. Although 359 nm is the λmax for ClO2, N-acetyldopaquinone also absorbs at this wavelength (see Figure 2). The kinetic traces are fit on the basis of the reaction steps given in eqs 3 and 4 (also see Scheme 2)

2ClO2 + NAT (10-fold excess) f

similar mechanism for the reaction between phenol and ClO2, where one ClO2 reacts with a phenoxide ion to give ClO2- and a phenoxy radical. This radical in turn reacted very rapidly with a second ClO2 to yield p-benzoquinone and released HOCl. In this mechanism Wajon suggested that an intermediate from the phenoxy radical and ClO2 might form, but no direct evidence was given. However, our results with tyrosine at a higher pH clearly

NAT-OClO + ClO2-

k1obs (3)

NAT-OClO f N-acetyldopaquinone + HOCl

k2obs (4)

for a sequence of two first-order reactions where NATOClO is an intermediate that builds up and decays. The functions used to fit each trace (41) are modified (see appendix) to take into account the 2:1 stoichiometry in eq 3. The observed rate constants obtained from the traces are summarized in Table 4 for p[H+] values from 5.21 to 6.23. At 240 nm k1obs corresponds to the growth of the NAT-OClO adduct as well as to the loss of ClO2 at 359 nm (2k1obs). Furthermore, the decay of the intermediate at 240 nm (k2obs) is concurrent with the growth of the product, N-acetyldopaquinone, observed at 359 nm. The decay of the NAT-OClO adduct (Scheme 2) is proposed to occur by proton abstraction by OH- or H2O. In Figure 7, k2obs is plotted against [OH-] (data from Table 4) to show that the decay of the intermediate is [OH-] dependent with a nonzero intercept (eq 5). The resolved k2obs ) k2 + kOH 2 [OH ]

(5)

constants are 2.16(2) s-1 for k2 and 2.1(1) × 107 M-1 s-1 for kOH 2 . As shown in Scheme 2, the first ClO2 molecule

506

Chem. Res. Toxicol., Vol. 18, No. 3, 2005

Napolitano et al. Scheme 3. Proposed Mechanism for the pH Dependent Formation of Dopachrome from Dopaquinone via Cyclodopa

Figure 7. Resolved first-order rate constants for the decay of the NAT-OClO adduct. Each data point is the average k2obs values for data measured at 240 and 359 nm (Table 4). Reaction conditions: [ClO2] 0.20 mM, [NAT] 2.00 mM, µ ) 1.0 M (NaClO4), [OAc]T ) 25.0 mM (O), [PO4]T ) 50.0 mM (4), and [PO4]T ) 25.0 mM (0).

undergoes a one electron reduction while the second ClO2 combines with the phenoxyl radical and then reacts to release HOCl (eq 4). Thus, the second ClO2 molecule undergoes a three-electron reduction. Overall, the fourelectron oxidation by two ClO2 molecules gives one ClO2and one HOCl as the reduced products and dopaquinone or N-acetyldopaquinone as the oxidized products. Dopachrome Formation. Additional reactions occur readily with dopaquinone to give dopachrome, a bright red product with absorption maximums at 305 and 485 nm as shown in Figure 1. Studies of tyrosine and dopa oxidations by mushroom tyrosinase/O2 at varying pH give similar results (20-22, 26, 42). On the other hand, NAT does not generate dopachrome because its amide group is much less nucleophilic than the amine group in dopaquinone and is less subject to cyclization. The proposed mechanism for the formation of dopachrome is shown in Scheme 3. A very reactive intermediate, cyclodopa, is formed via an internal Michael cyclization by nucleophilic attack of the amino group of dopaquinone (k3). The pH dependence of this reaction is due to the equilibrium distribution of the amine given (Ka2), where the ammonium form is unreactive toward cyclization. Once cyclodopa forms, it is rapidly oxidized by oxygen or by dopaquinone (33) to form dopachrome. The rate of formation of dopachrome is first order in the amine form of dopaquinone (DQ-) as shown in Scheme 3. A plot of kobs vs p[H+] is shown in Figure S1. The kinetic expression for dopachrome (DC) formation is given in eq 6 and the observed rate constant for [DQ]T )

d[DC] ) k3[DQ-] dt

Figure 7 to kobs ) k3Ka210p[H+] gives a value of 249 s-1 for the cyclization rate constant (k3). Kinetic Analysis and Proposed Mechanism for the Reaction of Dopa and ClO2. The two-electron oxidation of dopa by ClO2 (Figure 3) shows that the product of this reaction is dopaquinone that reacts further to generate dopachrome. Dopaquinone is also formed by the four-electron oxidation of Tyr by ClO2. However, the dopa reaction is approximately 16 times faster. In Figure 8, the observed loss ClO2 at λ ) 359 nm is first order when dopa is in excess. The rate constant for the of loss of ClO2 increases with pH and a nonzero intercept is found for the plot of kobs vs 1/[H+] (Figure 8). As the dopa concentration is increased from 1 to 5 mM, the observed pseudo first-order rate constant for the loss ClO2 increases linearly with a zero intercept (Figure S2). From these data the kinetic expression and observed rate constant of ClO2 decay are given in eqs 8

(6)

[DQH] + [DQ-] is given in eq 7. The value of k3 is

kobs )

k3Ka2 [H+]

(7)

estimated by assuming Ka2 is equal to the Ka of the amino group of dopa, pKa ) 8.71 (43, 44). Fitting the data in

Figure 8. First-order rate constants vs 1/[H+] for the reaction between ClO2 and excess dopa. Conditions: [ClO2] ) 0.20 mM, [dopa] ) 1.00 mM, [OAc]T ) 0.100 M, µ ) 1.00 M (NaClO4), λ ) 359 nm, 25.0 °C, and a path length of 0.962 cm.

Chlorine Dioxide Oxidations of Tyr, NAT, and Dopa

Chem. Res. Toxicol., Vol. 18, No. 3, 2005 507

Scheme 4. Proposed Mechanism for the Reaction between ClO2 and Dopa

ne decays to form a pink or yellow product(s) dependent on the ClO2/NAT ratio. At a 1:10 ClO2/NAT ratio, a pink product(s) is observed (Figure S3) and at a ratio of 2:1 a yellow product(s) is observed (Figure S4). Andersen reports similar results when N-acetyldopamine and pyrocatechol are oxidized (49) but showed this as pH dependent. Furthermore, Andersen had isolated dimers of the oxidized catechols. Dopa-containing peptides have been shown to produce dimers and trimers under oxidative conditions having similar UV-vis spectra to our results of N-acetyldopaquinone decay (50, 51).

Conclusions

and 9. The values for the resolved rate constants are in Table 3.

-

( (

) )

d[ClO2] k5Ka3 [dopa][ClO2] ) 2 k4 + dt [H+] kobs ) 2 k4 +

k5Ka3 [H+]

[dopa]

(8)

(9)

The proposed mechanism for the ClO2/dopa reaction, Scheme 4, accounts for the trends observed for the rate of loss of ClO2. The mechanism is analogous to that proposed by Wajon for the ClO2 and hydroquinone reaction (27). We propose that the two moles of ClO2 react per mole of dopa. As Scheme 4 shows, dopa reacts as its catechol form (k4), supported by the intercept of the pH data (Figure 8), to generate a radical cation which rapidly deprotonates forming the neutral radical. In addition, the deprotonated catechol, which is in a pre-equilibrium (Ka3), also reacts with ClO2 (k5) as shown by the increase of kobs vs pH, to generate the same neutral radical. The neutral radical is then rapidly oxidized by another ClO2, to generate dopaquinone, ClO2-, and H+. Contrary to the Tyr and NAT systems a ClO2 adduct intermediate is not observed because the second one-electron oxidation generates stable products. Furthermore, the removal of one electron from the radical is more energetically favorable then forming a ClO2-radical adduct. In the Tyr and NAT systems the removal of a second electron would lead to a carbocation intermediate, therefore the phenoxyl-ClO2 adduct is the favored pathway. Slow Reactions of Dopachrome and N-Acetyldopaquinone. We have shown that dopachrome results from reactions of ClO2 with Tyr and with dopa. Upon standing (1 day), the reaction solutions form either a black precipitate (pH 4) or a brown solution (pH 6). These products are attributed to the oxidative polymerization of dopachrome to melanin. Similar results have been reported by other groups (45-48). The reaction between ClO2 and NAT produces Nacetyldopaquinone. Similar to the dopachrome polymerization, we believe an analogous process occurs with N-acetyldopaquinone. Upon standing, N-acetyldopaquino-

Our results show that 2 mol of ClO2 react with 1 mol of Tyr or NAT with evidence for the formation of phenoxyl-OClO adducts that decay to give dopaquinone or N-acetyldopaquinone. Thus ClO2 can react by electron transfer and by radical-radical bond formation, but it does not react directly as a chlorinating agent. The combined reactions with two ClO2 molecules are fourelectron oxidations to give ClO2- and HOCl as the reduction products. HOCl is a well-known chlorinating agent that can react further (23-25). Hence we cannot exclude the possibility that halogenated aromatic products or chloramine derivatives are formed after ClO2 is consumed. The quinone products are identified by their UV-vis spectral characteristics and by the parallel reactivity of Tyr and phenol oxidations by ClO2. Our results show that the rate of reaction between ClO2 and dopa is nearly diffusion controlled. This study shows that ClO2 oxidation of Tyr is very fast at biological pH. The oxidation of NAT by ClO2 forms an o-quinone product and provides a possible model for the reaction of Tyr residue within a protein. Furthermore, it has been shown that dopa residues in peptides are oxidized to their corresponding quinones (50, 51). Therefore ClO2 reactions with Tyr residues in a protein could play a role in the disinfection of pathogens.

Appendix Intermediate formation and decay for consecutive firstorder reactions with 2:1 stoichiometries for ClO2/NAT are calculated from the following equation, which is adapted from ref 40 based on a 1:1 stoichiometry.

At )

a(2k1obs) + bk2obs exp(-2k1obst) + k2obs - 2k1obs c(2k1obs) exp(-k2obst) + d k2obs - 2k1obs

where a, b, c, and d are constants.

Acknowledgment. This work was supported by National Science Foundation Grants CHE-0332072 and CHE-0139876. Supporting Information Available: Observed first-order rate constants of dopachrome formation vs p[H+] (λ ) 490 nm), first-order rate constants vs [dopa] for the ClO2/dopa system (λ ) 359 nm), and UV-vis spectra of 10:1 and 2:1 ClO2/NAT reaction mixtures. This material is available free of charge via the Internet at http://pubs.acs.org.

508

Chem. Res. Toxicol., Vol. 18, No. 3, 2005

References (1) Katz, A., and Narkis, N. (2001) Removal of chlorine dioxide disinfection by-products by ferrous salts. Water Res. 35, 101-108. (2) Elphick, A. (1998) The growing use of chlorine dioxide. Processing March, 24. (3) Rosenblatt, D. H. (1983) Use of chlorine dioxide as a chemosterolizing agent. U.S. Patent 4,504,442. (4) Ritter, S. K. (2001) Ousting anthrax. Chem. Eng. News 79, 2426. (5) National Advisory Committee on Microbiological Criteria for Foods. (1999) Microbiological safety evaluation and recommendations on fresh produce. Food Control. 10, 117-143. (6) Nguyenthe C., and Carlin F. (1994) The microbiology of minimally processed fresh fruits and vegetables. Crit. Rev. Food Sci. Nutr. 34, 371-401. (7) Beuchat, L. R. (1992) Surface disinfection of raw produce. Dairy Food Environ. Sanitat. 12, 6-9. (8) Costilow, R. N., Uebersax, M. A., and Ward, P. J. (1984) Use of chlorine dioxide for controlling microorganisms during handling and storage of fresh cucumbers. J. Food Sci. 49, 396-401. (9) Reina, L. D., Fleming, H. P., and Humphries, E. G. (1995) Microbiological control of cucumber hydrocooling water with chlorine dioxide. J. Food Prot. 58, 541-546. (10) Zhang, S., and Farber, J. M. (1996) The effects of various disinfectants against Listeria monocytogenes on fresh-cut vegetables. Food Microbiol. 13, 311-321. (11) Brown, G. E., and Wardowski, W. F. (1986) Use of chlorine dioxide in Florida citrus packing houses to reduce inoculum of decay pathogens. Citrus Ind. 67, 48-56. (12) Key, A. M., Grodner, R. M., Park, D. L., Luizo, J., Kilgen, M. L., and Andrews, L. S. (1996) The use of a combination high-pressure spray wash/chlorine dioxide wash tank system for the processing of seafood products. IFT annual meeting: book of abstracts, p 133, Institute of Food Technologists, Chicago, IL. (13) Lin, W., Huang, T., Cornell, J. A., Lin, C., and Wei, C. (1996) Chlorine dioxide solution in a fish model system. J. Food Sci. 61, 1030-1034. (14) Kim, J. M., Huang, T. S., Marshall, M. R., and Wei, C. I. (1999) Chlorine dioxide treatment of seafoods to reduce bacterial loads. J. Food Sci. 64, 1089-1093. (15) Cutter, C. N., and Dorsa, W. J. (1995) Chlorine dioxide spray washes for reducing fecal contamination on beef. J. Food Prot. 58, 1294-1296. (16) Villarreal, M. E., Baker, R. C., and Regenstein, J. M. (1990) The incidence of Salmonella on poultry carcasses following the use of slow release chlorine dioxide (Alcide). J. Food Prot. 53, 464-467, 510. (17) Tsai, L., Higby, R., and Schade, J. (1995) Disinfection of poultry chiller water with chlorine dioxide: consumption and byproduct formation. J. Agric. Food Chem. 43, 2768-2773. (18) Noss, C. I., Hauchman, F. S., and Olivieri, V. P. (1986) Chlorine dioxide reactivity with proteins. Water Res. 20, 351-356. (19) Tan, H., Sen, A. C., Wheeler, W. B., Cornell, J. A., and Wei, C. I. (1987) A kinetic study of the reaction of aqueous chlorine and chlorine dioxide with amino acids, peptides and proteins. J. Food Sci. 52, 1706-1711. (20) Graham, D. G., and Jeffs, P. W. (1977) The Role of 2, 4, 5-trihydroxyphenylalanine in melanin Biosynthesis. J. Biol. Chem. 252, 5729-5734. (21) Mason, H. S. (1948) The chemistry of melanin. III. Mechanism of the oxidation of dihydroxyphenylalanine by tyrosinase. J. Biol. Chem. 172, 83-99. (22) Raper, H. S. (1928) The aerobic oxidases. Physiol. Rev. 8, 245282. (23) Domigan, N. M., Charlton, T. S., Duncan, M. W., Winterbourn, C. C., and Kettle, A. J. (1995) Chlorination of tyrosyl residues in peptides by myeloperoxidase and human neutrophils. J. Biol. Chem. 270, 16542-16548. (24) Kettle, A. J. (1996) Neutrophils convert tyrosyl residues in albumin to chlorotyrosine. FEBS Lett. 379, 106-106. (25) Winterbourn, C. C. (2002) Biological reactivity and biomarkers of the neutrophil oxidant, hypochlorous acid. Toxicology 181-182, 223-227. (26) Canovas, F. G., Garcia-Carmona, F., Sanchez, J. V., Iborra Pastor, J. L., and Teruel, A. L. (1982) The role of pH in the melanin biosynthesis pathway. J. Biol. Chem. 257, 8738-8744. (27) Wajon, J. E., Rosenblatt, D. H., and Burrows, E. P. (1982) Oxidation of phenol and hydroquinone by chlorine dioxide. Environ. Sci. Technol. 16, 396-402. (28) Alfassi, Z. B., Huie, R. E., and Neta, P. (1986) Substituent effects on rates of one-electron oxidation of phenols by the radicals ClO2, NO2, and SO3-. J. Phys. Chem. 90, 4156-4158.

Napolitano et al. (29) Furman, C. S., and Margerum, D. W. (1998) Mechanism of chlorine dioxide and chlorate ion formation from the reaction of hypobromous acid and chlorite ion. Inorg. Chem. 37, 4321-4327. (30) Jia, Z., Margerum, D. W., and Francisco, J. S. (2000) Generalacid-catalyzed reactions of hypochlorous acid with acetyl hypochlorite with chlorite ion. Inorg. Chem. 39, 2614-2620. (31) Nicoson, J. S., and Margerum, D. W. (2002) Kinetics and mechanisms of aqueous chlorine reactions with chlorite ion in the presence of chloride ion and acetic acid/acetate buffer. Inorg. Chem. 41, 342-347. (32) Waite, J. H. (1976) Calculating extinction coefficients for enzymatically produced o-quinones. Anal. Biochem. 75, 211-218. (33) Rzepecki. L. M., and Waite, J. H. (1989) A chromogenic assay for catecholoxidases Based on the addition of L-proline to quinones. Anal. Biochem. 179, 375-381. (34) Arnow, L. E. (1937) Colorimetric determination of the components of 3,4-dihydroxyphenylalanine-turosine mixtures. J. Biol. Chem. 118, 531-537. (35) Sugumaran, M., Semensi, V., Dali, H., and Mitchell, W. (1989) Novel transformations of enzymatically generated carboxymethylo-benzoquinone to 2,5,6-trihydroxybenzofuran and 3,4-dihydroxymandelic acid. Bioorg. Chem. 17, 86-95. (36) Taylor, S. W., Molinski, T. F., Rzepecki L. M., and Waite, J. H. (1991) Oxidation of peptidyl 3,4-dihydroxyphenylalanine analogues: Implications for the biosynthesis of tunichromes and related oligopeptides. J. Nat. Prod. 54, 918-922. (37) Kerwin, J. L. (1996) Negative ion electrospray mass spectrometry of polyphenols, catecholamines and their oxidation products. J. Mass Spectrom. 31, 1429-1439. (38) Grimley, E., and Gordon, G. (1973) The kinetics and mechanism of the reaction between chlorine dioxide and phenol in acidic aqueous solution. J. Inorg. Nucl. Chem. 35, 2383-2392. (39) Svenson, D., Kadla, J. F., Chang, H., and Jameel, H., (2002) Effect of pH on the mechanism of OClO• oxidation of aromatic compounds. Can. J. Chem. 80, 761-766. (40) Ganiev, I. M., Suvorkina, E. S., and Kabal’nova, N. N. (2003) Reaction of chlorine dioxide with phenol. Russ. Chem. Bull. 52, 1123-1128. (41) Espenson, J. H. (1995) Chapter 4. Consecutive reactions; the steady-state and other approximations. In Chemical kinetics and reaction mechanisms, 2nd ed. (Speer, J. B., and Morriss, J. M., Eds.) pp 70-75, McGraw-Hill, Inc., New York. (42) Garcia-Carmona, F., Gracia, F., Iborra, J. L., and Lozano, J. A. (1982) Kinetic study of the pathway of melanization between L-dopa and dopachrome. Biochim. Biophys. Acta 717, 124-131. (43) Young, T. E., Griswold, J. R., and Hulbert, M. H. (1974) Melanin. I. Kinetics of the oxidative cyclization of dopa to dopachrome. J. Org. Chem. 39, 1980-1982. (44) Martell, A. E., and Smith, R. M. (1974) Critical Stability Constants Vol. 1, Amino Acids, pp 31 and 33, Plenum Press, New York. (45) Heacock, R. A., and Mattok, G. L. (1963) The chemistry of the aminochromes Part V. Rearrangements in the presence of sodium hydroxide and zinc acetate. Can. J. Chem. 41, 139-147. (46) Land, E. J., Ransden, C. A., and Riley, P. A. (2003) Tyrosinase autoactivation and the chemistry of ortho-quinone amines. Acc. Chem. Res. 36, 300-308. (47) Pezzella A., Napolitano, A., d'Ischia, M., and Prota, G. (1996) Oxidative polymerisation of 5,6-dihydroxy-2-carboxylic acid to melanin: A new insight. Tetrahedron 52, 7913-7920. (48) Peter, M. G. (1989) Chemical modifications of biopolymers by quinones and quinone methides. Angew. Chem., Int. Ed. Engl. 28, 555-570. (49) Andersen, S. O., Jacobsen, J. P., Bojesen, G., and Roepstorff, P. (1992) Phenoloxidase catalyzed coupling of catechols. Identification of novel coupling products. Biochim. Biophys. Acta 1118, 134-138. (50) Burzio, L. A., and Waite, J. H. (2001) Reactivity of peptidyltyrosine to hydroxylation and cross linking. Protein Sci. 10, 735740. (51) Michon, T., Chenu, M., Kellershon, N., Desmeril, M., and Gueguen, J. (1997) Horseradish peroxidase oxidation of tyrosine containing peptides and their subsequent polymerization: A kinetic study. Biochemistry 36, 8504-8513. (52) Wang, W., Noel, S., Desmadril, M., Gueguen, J., and Michon, T. (1999) Kinetic evidence for the formation of a Michaelis-Mentenlike complex between horseradish peroxidase compound II and di-(N-acetyl-L-tyrosine). Biochem. J. 340, 329-336.

TX049697I