Chemiluminescent Aldehyde and - American Chemical Society

MAY 2000 VOLUME 13, NUMBER 5 © Copyright 2000 by the American Chemical Society

Articles Chemiluminescent Aldehyde and β-Diketone Reactions Promoted by Peroxynitrite Fernanda S. Knudsen,† Carlos A. A. Penatti,† Leandro O. Royer,† Karine A. Bidart,† Marcelo Christoff,† Denise Ouchi,‡ and Etelvino J. H. Bechara*,† Departamento de Bioquı´mica and Departamento de Quı´mica Fundamental, Instituto de Quı´mica, Universidade de Sa˜ o Paulo, CP 26077, 05599-970 Sa˜ o Paulo, SP, Brazil Received October 28, 1999

Peroxynitrite is shown here to promote the aerobic oxidation of isobutanal (IBAL) and 3-methyl-2,4-pentanedione (MP) in a pH 7.2 phosphate buffer into acetone plus formate and biacetyl plus acetate, respectively. These products are expected from dioxetane intermediates, whose thermolysis is known to be chemiluminescent (CL). Accordingly, the extent of total oxygen uptake by IBAL at different concentrations parallels the corresponding CL maximum intensities. The pH profile based on oxygen uptake data for the MP reaction matches the titration curve of peroxynitrous acid (pKa ∼ 7), indicating that peroxynitrite anion is the oxidizing agent. Energy transfer studies with IBAL and the 9,10-dibromoanthracene-2-sulfonate ion, a triplet carbonyl detector, indicates that triplet acetone (τ ) 19 µs) is the energy donor. It is postulated that IBAL- or MP-generated triplet carbonyls are produced by the thermolysis of dioxetane intermediates, which are formed by the cyclization of R-hydroperoxide intermediates produced by insertion of dioxygen into the IBAL or MP enolyl radicals, followed by their reduction. Accordingly, EPR spin-trapping studies with 3,5-dibromo-4-nitrosobenzenesulfonic acid (DBNBS) and 2-methyl-2-nitrosopropane (MNP) revealed the intermediacy of carboncentered radicals, as expected for one-electron abstraction from the enol forms of IBAL or MP by peroxynitrite. The EPR data obtained with IBAL also reveal formation of the isopropyl radical produced by competitive nucleophilic addition of ONOO- to IBAL, followed by homolytic cleavage of this adduct and β-scission of the resulting Me2CHCH(O-)O•. Superstoichiometric formation of fragmentation products from IBAL or MP attests to the prevalence of an autoxidation chain reaction, here proposed to be initiated by one-electron abstraction by ONOOfrom the substrate. This work reveals the potential role of peroxynitrite as a generator of electronically excited species that may contribute to deleterious and pathological processes associated with excessive nitric oxide and aldehyde production.

Introduction Peroxynitrite, formed by a diffusion-controlled reaction of nitric oxide with the superoxide anion radical [k ) 2 * To whom correspondence should be addressed: Instituto de Quı´mica, Universidade de Saõ Paulo, Caixa Postal 26077, 05599-970 Saõ Paulo, SP, Brazil. Telephone: 55-11-818-3869. E-mail: ebechara@ quim.iq.usp.br.

× 1010 M-1 s-1; pKa(ONOOH) ) 6.8] (1), has been shown to perform one- and two-electron oxidations, nitrations, and nucleophilic addition to carbonyls (2 and references therein). A significant fate of peroxynitrite in biological systems appears to be its reaction with dissolved CO2 † ‡

Departamento de Bioquı´mica. Departamento de Quı´mica Fundamental.

10.1021/tx990176i CCC: $19.00 © 2000 American Chemical Society Published on Web 04/06/2000

318

Chem. Res. Toxicol., Vol. 13, No. 5, 2000

Knudsen et al.

Scheme 1. Generation of Triplet Acetone from Isobutanal (IBAL) Initiated by HRP/H2O2 or Peroxynitritea

a

Route A, from Uppu et al. (21) and Va´squez-Vivar et al. (23). Route B, from Bechara et al. (24). Route C, from this work.

(k ) 1.0 × 104 M-1 s-1) (3), whose concentration may be as high as 1-2 mM (4). However, peroxynitrite has been observed to react in vivo directly with lipids, DNA, proteins, and several other biomolecules such as deoxyribose, glutathione, R-tocopherol, phenols, and amino acids (5-16). Such reactions can trigger important deleterious, protective, or regulatory biological responses (4, 17). In addition, peroxynitrite-mediated reactions may involve reactive intermediates resulting from the decomposition of the peroxynitrite-CO2 adduct (14, 18-20). Recent studies have shown that the nucleophilic addition of ONOO- to linear aliphatic aldehydes produces mainly reformed aldehyde and nitrate, plus minor products such as H2O2 and organic fragmentation products (organic acids) (21, 22). Nucleophilic addition of ONOOto the keto group of pyruvate has also been observed to result in decarboxylation of pyruvate into acetate, with the production of the CO2•- radical (23). These studies are relevant, considering the ubiquity of aldehydes and ketones in biological systems, where they may act as reactive intermediate sources. We examine here whether peroxynitrite, in competition with nucleophilic addition to easily enolizable carbonyl compounds, can act as a one-electron oxidant of the corresponding enols, initiating an oxygen-dependent chemiluminescent reaction similar to those previously reported for the horseradish peroxidase-catalyzed (HRP/ H2O2) oxidation of isobutanal (IBAL) and 3-methyl-2,4pentanedione (MP) (24, 25). Scheme 1 illustrates the two oxidation pathways initiated by peroxynitrite, nucleophilic addition (route A) versus electron abstraction (routes B and C), in the case of IBAL. The HRP-catalyzed reactions of IBAL and MP have been shown to produce electronically excited triplet acetone and triplet biacetyl, respectively. Triplet species have been reputed to be of significant biological relevance because triplet carbonyls can transfer energy to several biomolecules and, being endowed with the chemical behavior of radicals, can trigger typically photochemical reactions in the dark (26). Our hypothesis is that carbonyl compounds that (i) are capable of fast enolization either directly, as is the case of MP [2.8% enol form exists in aqueous medium (27)], or indirectly, by phosphate catalysis as with IBAL (28), and (ii) bear an additional R-alkyl substituent, which is expected to lower the reduction potential (in the case of both MP and IBAL), may prefer to react with peroxynitrite via route C in Scheme 1. For comparison, parallel studies were carried out with 3-methyl-2-butanone and nonsubstituted substrates such as 2-pentanone and 2,4pentanedione. If route C (Scheme 1) were operative, the reactions with IBAL and MP would lead to triplet acetone plus formate and triplet biacetyl plus acetate, respec-

tively, the fragmentation products expected from the thermolysis of dioxetane intermediates. The R-methyl substitution in MP precludes its subsequent nitration, as reported to occur with ethyl acetoacetate treated with peroxynitrite (29). Superstoichiometric formation of substrate fragmentation products (based on added peroxynitrite), as well as formation of tertiary carbon-centered radical intermediates that can be detected by EPR spin trapping, can be predicted by this route.

Materials and Methods Reagents and Preparations. All the reagents, purchased from Sigma-Aldrich, were of purest analytical grade. Peroxynitrite was prepared in 0.6 M HCl from NaNO2 and H2O2 in a 1.5 M NaOH quenched-flow reaction according to Beckman et al. (30) and quantified spectrophotometrically at 302 nm ( ) 1670 M-1 cm-1). Sodium 9,10-dibromoanthracene-2-sulfonate (DBAS) and sodium 9,10-diphenylanthracene-2-sulfonate (DPAS) (405 ) 7750 M-1 cm-1 and 397 ) 8050 M-1 cm-1, respectively) were prepared according to the methods of Catalani et al. (31). Solutions were prepared in Milli-Q-purified water and the buffers pretreated with Chelex. Techniques. Oxygen uptake was followed in a Yellow Spring Instruments model 53 oxygen monitor and chemiluminescence measured with a Hamamatsu TVC 767 photon counter equipped with a red-sensitive Thorn EMI (9658AM) photomultiplier cooled to -12 °C by a Thorn EMI (model Fact MKIII) thermoelectric cooler. Acetone was analyzed by GC with a Shimadzu CH-14A chromatograph with flame ionization detection. A SPB-1 polymethylsilicone capillary column (25 m × 0.25 mm o.d. × 0.25 µm i.d.) was used. Temperatures were as follows: injection, 200 °C; detector, 250 °C; and column oven, 35 °C. Helium was used as a carrier gas; the sample size (headspace) was 1.0 mL. The integrator was a Shimadzu chromatopac C-R4A. Biacetyl was detected by two methods: (a) derivation with 6-hydroxy-2,4,5-triaminopyridine, followed by HPLC/ fluorescence detection analysis, according to the method of Mansilla-Espinosa et al. (32); and (b) GC on a Shimadzu CH 14A chromatograph with flame ionization detection. The experimental conditions employed were the same as those used for the analysis of acetone. EPR spectra were recorded at room temperature using a Bruker EMX spectrometer, 4 min after the addition of peroxynitrite. Capillary electrophoresis was performed on a Perkin-Elmer model 270A-HT apparatus equipped with a UV/vis detector set at 254 nm, a temperature control device maintained at 25 °C, and a data acquisition and analysis system (Turbochrom). A fused-silica capillary (Polymicro Technologies) was employed (75 µm i.d., 375 µm o.d., and 72 cm length). A detection window of approximately 0.2 cm was created at 50 cm from the capillary inlet by removing the polyamide coating. Samples were injected by the hydrodynamic method, at 5 mmHg for 1-5 s. The electrolyte was 0.2 mM CTAB (hexadecyltrimethylammonium bromide) and 10 mM Na2CrO4. The electrophoresis system was operated under reversed polarity and a constant voltage of -30

Carbonyl Oxidation into Triplets by Peroxynitrite

Figure 1. Total oxygen uptake by aldehydes and ketones (all at 10 mM) treated with peroxynitrite (500 µM) in 100 mM phosphate buffer (pH 7.2) at 25 °C. 2-P, pentanone; MB, 3-methyl-2-butanone; IBAL, isobutanal; MP, 3-methyl-2,4-pentanedione; PD, 2,4-pentanedione. Table 1. Enolization of Carbonyls in an Aqueous Medium propertya

isobutanal

2,4-pentanedione

3-methyl2,4-pentanedione

Ke Epad (V) pKa k (M-1 s-1)

1.8 × 10-4 b >2.75 11.6e 8.5 × 10-5 g

0.19c 2.53 8.87f 2.3 × 103 e

0.029c 1.99 10.65f 17e

a K , enolization equilibrium constant; pK , enol dissociation e a constant; Epa, anodic potential; k, second-order rate constant (in b c d the presence of phosphate). Ref 33. Ref 27. Epa was determined by cyclic voltammetry in acetonitrile containing 0.1 M NaClO4 with Pt bead (working), Ag/AgI (reference), and Pt plate (auxiliary) electrodes and expressed vs NHE (this work). e Ref 34. f Ref 28. g Ref 35.

kV. External calibration was carried out with stock solutions of organic solutes: formate (retention time of 2.3 min), acetate (2.8 min), nitrite (1.9 min), nitrate (2.0 min), and carbonate (2.5 min). Aliquots of peroxynitrite and HCl (sufficient to obtain a given pH in the reaction mixture) were deposited separately onto the inner walls of an Eppendorf tube containing 1.5 mL of aqueous substrate, followed by homogenization via vigorous vortexing. The desired final pH was checked in each case. An Applied Photophysics (model SX 18 MV) stopped-flow spectrophotometer was used for kinetic measurements. Peroxynitrite decay was monitored at 340 nm, where interference due to the IBAL and MP substrates is minimal. Stock solutions of substrates in deionized water were mixed with the phosphate buffer less than 1 h before being used.

Results and Discussion Oxygen Uptake and Chemiluminescence. Total oxygen depletion occurs within seconds when IBAL, MP, or 2,4-pentanedione (but not 2-pentanone or 3-methyl2-butanone) (Figure 1) is treated with peroxynitrite under standard experimental conditions, i.e., 10 mM carbonyl compound and 500 µM peroxynitrite in normally aerated 0.10 M phosphate buffer (pH 7.2). Relevant physicochemical constants of these compounds such as enolization equilibrium constants (Ke), enolization rate constants (k1), enol pKa values, and anodic potentials (Epa) are summarized in Table 1 (27, 28, 33-35). The Ke values of β-diketones (such as MP and 2,4-pentanedione) are much higher than those of monocarbonyls because their enol forms are stabilized by intramolecular hydrogen bonds and by conjugation of the carbon-carbon double bond with the carbonyl function. Figure 1 shows that the two β-diketones, MP and 2,4-pentanedione, which exhibit

Chem. Res. Toxicol., Vol. 13, No. 5, 2000 319

relatively high concentrations of the enol form in aqueous medium (2.8 and 15.9%, respectively, from Table 1), and the aldehyde IBAL (Ke ) 1.8 × 10-4 in water) (27), which undergoes fast enolization under bifunctional phosphate catalysis (28), are readily oxidized by the dissolved oxygen. The “bifunctional” phosphate catalysis of IBAL enolization mentioned here results from the concomitant proton donor (to the carbonyl oxygen of IBAL) and proton acceptor (from the methyne hydrogen of IBAL) capacity of the H2PO4- anion. The poorly enolizable monoketones (Ke ∼ 10-8 in water) (36-39), although prone to nucleophilic addition of peroxynitrite, do not appreciably consume oxygen. This is also the case of 2,2-dimethylpropanal, a nonenolizable IBAL homologue, treated with peroxynitrite under identical experimental conditions (not shown). Similarly, acetaldehyde (22), whose Ke is 10-5 in water (36), and pyruvate (23) consume little oxygen when treated with peroxynitrite. The reactivity order toward oxygen (β-diketones > IBAL > acetaldehyde, 2,2-dimethylpropanal, 2-pentanone, 3-methylbutanone, and pyruvate) appears to be dictated by the easy enolization of substrate and of one-electron oxidation (see anodic oxidation peaks in Table 1). Moreover, the lack of oxygen consumption by acetaldehyde and monoketones, which might be expected to undergo peroxynitrite nucleophilic addition followed by adduct homolytic cleavage, β-scission, oxygen addition to resulting carbon-centered radicals, and Russell reactions (21, 22), suggests that the oxygen uptake observed with β-diketones and IBAL cannot be associated with the operation of route A in Scheme 1. Partial oxygen depletion by IBAL may be explained by the competition between nucleophilic attack of peroxynitrite on IBAL and electron abstraction (route A vs routes B and C in Scheme 1). The extent of total oxygen consumption by IBAL as a function of either peroxynitrite or IBAL concentration roughly parallels the accompanying chemiluminescence (maximum intensity) elicited upon addition of a triplet energy detector (31), the 9,10-dibromoanthracene-2-sulfonate ion (DBAS) (panel A or B of Figure 2, respectively). A double-reciprocal plot of the DBAS-sensitized chemiluminescence intensity versus DBAS concentration (1-10 µM) gave a straight line, the slope of which reflects the product of the excited donor lifetime, τ, and the rate constant kET of triplet-singlet energy transfer from acetone to DBAS (31). The slope kETτ was determined to be 7.5 × 104 M-1, from which the lifetime of triplet acetone could be estimated to be roughly 19 µs assuming kET ) 4 × 109 M-1 s-1 (31). This agrees well with triplet acetone lifetime values reported in the literature (31), which range from 2 to 20 µs, depending on the concentration of dissolved oxygen, a triplet quencher. Addition of 9,10-diphenyl-2-anthracenesulfonate ions, capable of detecting excited singlets but not triplets (31), did not enhance chemiluminescence, attesting to the triplet nature of the excited donor. The reaction of MP with peroxynitrite in a phosphate buffer also emits light, but the rapid kinetics (>90% CL is released within 4 s) did not allow for accurate measurements by our photoncounting equipment (not shown). The extent of total oxygen consumption by the IBAL/ peroxynitrite system is 12-fold higher in 100 mM arsenate buffer (pH 7.2) than in phosphate buffer at the same concentration (not shown). Arsenate is known to act more efficiently than phosphate as a bifunctional catalyst of carbonyl enolization (24, 28).

320


Knudsen et al.

Figure 4. pH dependence of the observed pseudo-first-order rate constant (kobs) for isobutanal reaction in 500 (1), 250 (2), and 100 mM (3) phosphate buffer (pH 7.2). Peroxynitrite (500 µM) was allowed to react at 25 °C with 10 mM IBAL. The points were adjusted using the following Lorentzian function: y ) yo + (2A/π)[W/[4(X - Xc)2 + W2]], where yo ) offset, Xc ) center, W ) width, and A ) area.

Figure 2. (A) Total oxygen consumption and maximum chemiluminescence intensity by 20 mM isobutanal (IBAL) at different peroxynitrite concentrations in 100 mM phosphate buffer (pH 7.2) at 25 °C. (B) Oxygen consumption and maximum chemiluminescence by 500 µM peroxynitrite at different IBAL concentrations, in 100 mM phosphate buffer (pH 7.2).

Figure 3. pH profile of isobutanal (IBAL) and 3-methyl-2,4pentanedione (MP) treated with 500 µM peroxynitrite in 500 mM phosphate buffer (pH 7.2) at 25 °C with 10 mM IBAL and 1 mM MP.

pH Profile. The pH profiles of the peroxynitritepromoted oxidations of IBAL and MP were obtained by two methods: total oxygen consumption and stopped-flow spectrophotometry. The pH profile of the MP oxygenconsuming reaction (Figure 3) roughly describes the peroxynitrous acid titration curve (pKa ) 6.8 in 100 mM phosphate) (11), pointing to peroxynitrite anion as the initiator of MP aerobic oxidation (route B in Scheme 1). With IBAL, a subtle increase in the extent of oxygen uptake in the pH region close to the peroxynitrous acid pKa value may be related to the loss of enol substrate by nucleophilic addition of peroxynitrite to the substrate

Figure 5. pH dependence of the observed pseudo-first-order rate constant (kobs) for 3-methyl-2,4-pentanedione (MP) reaction in 500 mM phosphate buffer (pH 7.2). Peroxynitrite (500 µM) was allowed to react with 10 mM MP at 25 °C.

keto form. The IBAL pH profile shown in Figure 3 also suggests that both the peroxynitrite anion and the peroxynitrous acid act as oxidants in this case. The rate of disappearance of peroxynitrite upon addition of excess MP or IBAL (both of which absorb strongly in the 200-320 nm range), monitored in a stopped-flow instrument at 340 nm, follows first-order kinetics. The slope gave an apparent second-order constant of 166 M-1 s-1 for MP and 344 M-1 s-1 for IBAL, in 100 mM phosphate buffer (pH 7.2) at 25 °C. These values are lower than those obtained for simple aliphatic aldehydes: propionaldehyde, 530 M-1 s-1 (21); and acetaldehyde, 680-830 M-1 s-1 (21, 22). The plot of the first-order constant (kobs) for peroxynitrite disappearance, obtained by subtracting the rate constant values for spontaneous peroxynitrite decomposition from the values measured with the complete system, versus the pH roughly delineates a bell-shaped curve for both IBAL (Figure 4) and MP (Figure 5). As previously reported in the literature, the pH profile for spontaneous peroxynitrite decomposition follows the titration curve of peroxynitrous acid, with observed rate constants of 0.55 s-1 at pH 7.0, 1.0 s-1 at pH 6.5, and 1.1 s-1 at pH 6.0 (0.43, 0.9, and 1.3 s-1, respectively, from ref 14). The catalytic effect of increasing phosphate concentration (100-500 mM) on the rate constants of the IBAL reaction with peroxynitrite is ratified by this experiment (differences in the ionic strength were compensated by addition of NaCl). Unlike the profiles reported for acetaldehyde and pyruvate (21, 22), which match the peroxynitrite titration curve, the bell-shaped pH profiles of the reaction of IBAL and MP with peroxynitrite may result from the contribu-



Table 2. Product Analysis for the Carbonyl (10 mM)/ Peroxynitrite (500 µM) Reactions substrate

no. of experiments

[RCO2-] (mM) formate

IBAL control IBAL (pH 7.5) IBAL (pH 11)

3 3 6

0.06 ( 0.01 0.32 ( 0.03 0.66 ( 0.11a

[R2CdO] (mM) acetone

0.80

substrate

no. of experiments

[RCO2-] (mM) acetate

MP control MP (pH 7.5) MP (pH 11)

3 3 6

0.78 ( 0.04 1.40 ( 0.30 6.36 ( 0.87b

[R2CdO] (mM) biacetyl

9

substrate

no. of experiments


2,4-pentanedione (pH 7.5) 2,4-pentanedione (pH 11)

3 3

0.50 ( 0.02 1.20 ( 0.10

substrate

no. of experiments


2-pentanone (pH 11) 2-pentanone (pH 7.5)

3 3

ND 0.14 ( 0.04

a Acetone detected by gas chromatography within 20% error. Biacetyl detected by gas chromatography and HPLC/fluorescence within 20% error.

b

tion of opposite effects: (i) The ascending part of the curve can be related to the increase in the ONOOconcentration, here assumed to be a one-electron oxidant of the substrate in route B in Scheme 1 or a nucleophilic agent in route A, while (ii) the descending part of the curve may be determined by either a decrease in H2PO4concentration above pH 7 (pKa2 for phosphoric acid ) 7.2, H2PO4- presumably being a more effective bifunctional catalyst of IBAL enolization than its conjugate base, which favors of route B) or a decrease in carbonyl concentration due to base-catalyzed enolization at increased pH values (which would favor route A). The latter hypothesis, i.e., an increased level of enolization upon the pH being raised, is more attractive given that, when enol/ enolate formation is followed spectrophotometrically at 280 nm, a batochromic shift and a strong increase in absorbance with increasing pH can be observed above pH 7. Ascribing the descending part of the pH profile to deprotonation of the substrate (enol IBAL), giving a less reactive form (enolate IBAL), as suggested previously in the case of the reaction of peroxynitrite with ascorbate (pKa ) 4.25) (40), collides with the fact that enol IBAL is a very weak acid (pKa ) 11.6; Table 1). We are, thus, inclined to conclude that the pH profiles for the reaction of peroxynitrite with IBAL and MP, accompanied by following the disappearance of peroxynitrite in the stopped-flow apparatus, reflect the influence of pH on the nucleophilic addition of peroxynitrite to carbonyl. For comparative purposes, we reproduced the pH profile for the reaction of peroxynitrite with acetaldehyde in 500 mM phosphate buffer, previously shown by Nakao et al. (22) to mirror the peroxynitrous titration curve, in 100 mM phosphate buffer under comparable experimental conditions (not shown). One can see a descending trend at pHs above 7, as in the case of IBAL and MP, which might result from the loss of the aldehyde form by phosphate-catalyzed enolization at a high phosphate concentration. On the other hand, the higher kobs values for peroxynitrite disappearance obtained at higher phosphate concentrations may be attributed to phosphateassisted addition (base catalysis) of peroxynitrite to acetaldehyde.

Figure 6. EPR spectra of DBNBS radical adducts obtained after incubation for 3 min at room temperature of 20 mM DBNBS and 500 µM peroxynitrite in phosphate buffer (pH 7.2) containing 0.1 mM diethylenetriaminepentaacetic acid (DTPA) in the presence of 10 mM IBAL (A, C, and D), 10 mM 3-methyl2-butanone (MB), and 20 mM HCO3-. Instrumental conditions were as follows: microwave power, 2 mW; modulation amplitude, 1 G; time constant, 0.32 s; and gain, 1 × 106.

Due to the presence of CO2 in the reaction mixtures, resulting from bicarbonate contamination in both the buffer and peroxynitrite stock solution, one might attribute the initiation of the oxygen consumption to CO2ONOO- or to one of its oxidizing decomposition products (3, 18). The possibility that this is not the case, however, is suggested by the reduction (30 and 50%) of the extent of total oxygen uptake caused by the addition of bicarbonate (1 and 5 mM) to the MP (0.5 mM)- and IBAL (10 mM)-containing standard reaction mixtures. Spin-trapping studies described below confirm that the addition of bicarbonate (20 mM) to the IBAL (10 mM)containing system does lead to higher amounts of CO2ONOO-, which competes with nucleophilic addition of peroxynitrite to IBAL. Product Analysis. Formate and acetate concentrations were determined by capillary electrophoresis in the spent reaction mixtures of air-equilibrated aqueous IBAL and MP (both at 10 mM) treated with 500 µM peroxynitrite (containing 0.5 M NaOH) under vigorous stirring at final pHs of 7-8 (adjusted with HCl) and 11 (Table 2). Capillary electrophoresis also showed the presence of nitrate (0.4-0.7 mM), nitrite (0.5-2.5 mM), and carbonate (1-2 mM) in the spent reaction mixture. The latter values are not reliable in a discussion of the data because peroxynitrite preparations are normally contaminated with these ions at concentrations in the millimolar range and carbonate varies considerably in buffers and peroxynitrite preparations. Acetate was analyzed in the reaction mixtures of 2,4pentanedione and 2-pentanone, for comparison. Acetone was identified in the phosphate (100 mM, pH 7.2)buffered IBAL-containing reaction mixture by mass spectroscopy by comparison with an authentic sample of

322


Knudsen et al.

Table 3. Hyperfine Splitting Constants (G) for Adduct Radicals with either DBNBS or MNP Obtained by Treatment of Carbonyls with Peroxynitritea

b

a

The spectra were obtained at room temperature in 100 mM phosphate buffer (pH 7.2). b MB, 3-methyl-2-butanone.

acetone, and quantified by gas chromatography; 0.8 mM acetone was found in the spent reaction mixture obtained from 10 mM IBAL and 500 µM peroxynitrite. Acetone was not found in unbuffered reaction mixtures, probably because of the importance of phosphate catalysis in the oxygen-consuming reaction. In addition, biacetyl was analyzed in the MP (10 mM)-containing reaction mixture by both HPLC/fluorescence detection (32) and gas chromatography/flame ionization detection, calibrated against authentic samples of biacetyl; values of 9.0 and 8.8 mM were found with the two methods, respectively. Examination of product yields (listed in Table 2) and the enolization properties of IBAL, MP, and 2,4-pentanedione (given in Table 1) allows for the following conclusions. (1) In every case, the observed yield of carboxylic products (Table 2) is greater than the upper limit predicted from the quantitative stoichiometric reaction of peroxynitrite (500 µM added peroxynitrite) with the substrate, suggesting that a chain reaction occurs in the presence of oxygen. (2) The more readily enolizable and oxidizable the carbonyl substrate (Table 1), the higher the product yield; thus, reactivity decreases in the following order: MP > 2,4-pentanedione > IBAL > 2-pentanone (Figure 1 and Table 2). (3) The observed stoichiometries for acetone/formate (roughly 1:1) and biacetyl/acetate (roughly 1.5:1) mixtures formed in the IBAL and MP reactions, respectively, do not support product formation exclusively in reactions initiated by ONOO- nucleophilic addition (Table 2). EPR Spin-Trapping Studies. EPR spin-trapping studies with 3,5-dibromo-4-nitrosobenzenesulfonic acid (DBNBS) and 2-methyl-2-nitrosopropane (MNP) were conducted to detect and possibly identify carbon-centered radical intermediates. Figure 6 summarizes the spintrapping studies conducted with IBAL/peroxynitrite in the presence of DBNBS. The observed spectrum origi-

nates from two spin adducts: a six-line spectrum (aN ) 14.5 G; aH ) 9.7 G) that can be assigned to a •CHR2 [•CH(CH3)2]-DBNBS adduct (Table 3) and a triplet (aN ) 14.5 G) that can be attributed to an adduct with a tertiary radical (IBAL•). Formation of an isopropyl radical is predicted by route I in Scheme 2 (reaction initiated by addition of ONOO- to IBAL) and the tertiary radical by electron abstraction from IBAL enolate (route II in Scheme 2). Because the aH constant of the six-line spectrum does not agree with the only value reported for an isopropyl adduct (aH ) 5.85 G) (41), we carried out the same experiment with 3-methyl-2-butanone (MB), an IBAL analogue, which is expected to yield the isopropyl radical by route I in Scheme 2, initiated by the addition of ONOO- to the carbonyl group. Indeed, an equivalent six-line spectrum appears to favor our assignment. The triplet signal seen after 4 min does not appear at 4 min in the control experiment without peroxynitrite and tends to increase with time (30 min) in the complete reaction mixture, probably due to the accumulation of the oxidized “ene” adduct of enol IBAL with DBNBS, which is expected to exhibit the same EPR adduct signal. The fact that the six-line spectrum may be due to an adduct with an isopropyl radical formed from route I (Scheme 2) is indicated by quenching of its signal upon addition of 20 mM bicarbonate, a source of CO2, that should compete with IBAL in ONOO- nucleophilic carbonyl addition (21). As already mentioned, the inhibitory effect of bicarbonate as a sink for removing ONOO- can also be evaluated by the observed decline in the extent of oxygen consumption by IBAL plus peroxynitrite (∼50% with 5 mM bicarbonate). Spin-trapping experiments with the MP/peroxynitrite/ DBNBS system revealed only a triplet signal (Table 3), which can be assigned to a tertiary radical adduct with MP• (route II in Scheme 3). Control experiments in the



Scheme 2. Generation of Triplet Acetone by Reaction of Isobutanal (IBAL) with Peroxynitrite

absence of peroxynitrite show no triplet signal at 4 min, although this signal does develop with time due to accumulation of the oxidized “ene” adduct of enol MP with DBNBS and is enhanced upon addition of 20 µM ferricyanide (not shown). Similar experiments carried out with MNP (Figure 7) revealed further significant details that are important to confirm the contribution of route II (Scheme 3) to the chemiluminescent reaction. Besides a triplet signal (aN ) 15.2 G), which can be attributed to an adduct with the tertiary MP• radical, these experiments revealed the presence of a triplet that can be assigned to an adduct with acetyl radical (aN ) 8.3 G), previously characterized (aN ) 8.3 G) during treatment of 2,4-pentenedione with HRP/H2O2 (42). The “ene” adduct is almost absent. Without peroxynitrite, a triplet signal (aN ) 17.0 G) that can be assigned to the MNP oxidation product can be detected; extensive MNP oxidation may be coupled to substrate autoxidation. For comparison, MP oxidation was also conducted with HRP/H2O2 as an oxidant in the presence of MNP, and the same acetyl radical adduct was detected (triplet, aN ) 8.3 G). This triplet can also be obtained in high yields by reaction of biacetyl with peroxynitrite, as

shown in Figure 7 and illustrated in Scheme 3. Biacetyl in the triplet state has previously been shown to be a major product of the MP/HRP/H2O2 system (25). We propose here that EPR-detected acetyl radicals from MP/HRP/H2O2 or MP/peroxynitrite may arise from carbonyl-carbonyl cleavage of electronically excited triplet biacetyl (43), thus reinforcing the occurrence of route II in Scheme 3.

Conclusions Our hypothesis is that, as in the reaction of IBAL and MP with HRP/H2O2, a reaction with ONOO- can be initiated by phosphate-catalyzed enolization of the aldehyde or ketone, followed by one-electron oxidation of the enol by peroxynitrite to give the resonance-stabilized enolyl radical (route II in Schemes 2 and 3). With MP, the phosphate-catalyzed enolization is not as strong a determinant as with IBAL because the MP enol form is already present to the extent of about 2.8% in water (Table 1) (hence, the ease with which its products can be analyzed by capillary electrophoresis and the product yield data correlated using GC and HPLC techniques).

324


Knudsen et al.

Scheme 3. Generation of Triplet Biacetyl by Reaction of 3-Methyl-2,4-pentanedione with Peroxynitrite

One-electron oxidation of either IBAL or MP appears to compete with nucleophilic addition of ONOO- to the carbonyl substrate, followed by fragmentation to radicals (route I in Schemes 2 and 3). The greater the facility of substrate enolization (in the case of β-diketones and phosphate catalysis) and/or oxidation (in the case of R-methyl substitution), the more important the contribution of one-electron abstraction. For both MP and IBAL, bell-shaped pH profile curves suggest two rate-limiting reactions. The ascending curve matches the onset of the titration curve of peroxynitrous acid and, hence, points to the ONOO- concentration as rate-limiting. This dependence is also shown when the reaction is monitored by oxygen uptake, which indicates that ONOO- is the one-electron oxidant of the substrates. The curve obtained spectrophotometrically following the rate of peroxynitrite disappearance reaches a maximum at pH 7.2 (IBAL) or 7.0 (MP) and then decreases, accompanying both the second half of the H2PO4- titration curve and the loss of carbonyl substrate by base-catalyzed enolization. In

contrast, linear aldehydes exhibit a pH profile (20) that is interpreted from the reaction kinetics as being dominated by nucleophilic addition to the carbonyl compound. We propose that the pH profiles of the compounds studied here, obtained by the stopped-flow technique, also mirror nucleophilic addition to the substrate (route I in Scheme 2), whereas those determined by oxygen consumption refer to the competing reaction initiated by electron abstraction from the enol substrates, leading to triplet product formation and hence chemiluminescence (route II in Scheme 2). Insertion of dissolved molecular oxygen into the enolyl radical is expected to produce an R-hydroperoxyl carbonyl radical, which may, in turn, abstract a hydrogen atom from another IBAL or MP molecule, thereby propagating a chain reaction and producing a dioxetane intermediate by cyclization (route II in Scheme 2). Thermolysis of the dioxetane intermediates then yields triplet acetone plus formic acid from IBAL and biacetyl plus acetate from MP. Nucleophilic addition of ONOO- to the substrate should



cytochrome c, myeloperoxidase, lipoxygenase, and prostaglandin-H synthase, have been shown to catalyze such processes. (ii) The second is dismutation of alkoxyl radicals formed by one-electron reduction of alkyl hydroperoxides (Φ3 ) 10%) (45). (iii) The third is disproportionation of alkylperoxyl radicals formed in lipid peroxidation chains (Φ3 < 0.01%) (46, 47). (iv) The final reaction is one-electron oxidation of “ene” hydroperoxides by horseradish peroxidase (HRP) or peroxynitrite, also via dioxetane intermediates (48). Final Remarks. Considering that (i) aldehydes are ubiquitous compounds in biological systems under physiological and pathological conditions, (ii) peroxynitrite has increasingly been implicated in cell toxicity and signaling, and (iii) n,π* triplet carbonyls may be viewed as alkoxyl radicals and, therefore, play a deleterious role in biomolecules, it is tentatively proposed that carbonyl compound reactions with peroxynitrite may constitute another chemical source of long-living reactive triplets operative in “photobiochemical processes without light”.

Figure 7. EPR spectra of MNP radical adducts obtained after incubation for 4 min at room temperature of 10 mM MNP, 10 µM HRP (C), and 500 µM peroxynitrite in 100 mM phosphate buffer (pH 7.2) containing 0.1 mM DTPA in the presence of 10 mM MP (A-C and E) and 10 mM biacetyl. Instrumental conditions were as follows: microwave power, 20 mW; modulation amplitude, 1 G; time constant, 0.32 s (A-C and E) and 0.16 s (D); gain, 5.0 × 104 (A and D) and 5.0 × 105 (B, C, and E).

also generate the same products. However, chemical yields of biacetyl and acetone greatly exceed the added peroxynitrite concentration, suggesting that the oxygenconsuming chain reaction initiated by electron abstraction from the substrate enols predominates over that initiated by nucleophilic addition of ONOO-. With regard to the weak chemiluminescence associated with generation of triplet species in the reaction of peroxynitrite with enolizable carbonyl compounds, it should be kept in mind that, due to their intrinsically long lifetimes (microsecond), triplet species have been considered suitable candidates for participation in bimolecular processes, such as energy transfer and chemical reactions (e.g., isomerization, cycloaddition, Norrish cleavage, and hydrogen abstraction) (26). In addition to the peroxynitrite-driven reaction described in this work, various chemical reactions leading to triplet carbonyl formation may also be of biological relevance, including the following. (i) One is heme protein-catalyzed aerobic oxidation of carbonyl compounds bearing an R-hydrogen, such as IBAL (24) and MP (25), via dioxetane intermediates. Dioxetanes have long been known as clean and efficient sources of triplet carbonyls (Φ3 ∼ 10-60%) (26, 44). Using the IBAL/HRP system as a source of triplet acetone, Cilento (26) demonstrated over the past two decades the feasibility of “photochemistry without light” with, for example, colchicine, riboflavin, chlorophyll, phytochromes, lysozyme, isolated or intracellular DNA, and several other synthetic electronic energy acceptors (e.g., dibromoanthracene and xanthene dyes). Not only HRP but also other proteins and enzymes, including myoglobin,

Acknowledgment. This work was supported by grants from the Fundaç aõ de Amparo a` Pesquisa do Estado de Saõ Paulo (FAPESP), the Programa de Apoio a Nućleos de Exceleˆncia (PRONEX), Conselho Nacional de Desenvolvimento Cientifico e Tecnolo´gico (CNPq), and the von Humboldt Foundation. We are grateful to Dr. Omar El Seoud, Dr. Wilson F. Jardim, and Dr. Lilian R. F. de Carvalho for providing access to the stopped-flow spectrophotometer, the gas chromatograph, and capillary electrophoresis system, respectively. We also thank Dr. Frank Quina, Dr. Brian Bandy, and Ms. Beatrice Allain for reading the manuscript and Dr. The´re`se Wilson and Ohara Augusto for helpful discussions. Samples of MB were kindly provided by Dr. Bruce Freeman and Dr. Enrique Cadenas.

References (1) Kissner, R., Nauser, T., Bugnon, P., Lye, P. G., and Koppenol, W. H. (1997) Formation and properties by peroxynitrite as studied by laser flash photolysis. Chem. Res. Toxicol. 10, 1285-1292. (2) Squadrito, G. L., and Pryor, W. A. (1998) Oxidative chemistry of nitric oxide: the roles of superoxide peroxynitrite, and carbon dioxide. Free Radical Biol. Med. 25, 392-403. (3) Lymar, S. V., and Hurst, J. K. (1995) Rapid reaction between peroxynitrite ion and carbon dioxide: implications for biological activity. J. Am. Chem. Soc. 117, 8867-8868. (4) Radi, R. (1998) Peroxynitrite reactions and diffusion in biology. Chem. Res. Toxicol. 11, 720-721. (5) Shi, H., Noguchi, N., Xu, Y., and Niki, E. (1999) Formation of phospholipid hydroperoxides and its inhibition by R-tocopherol in rat brain synaptosomes induced by peroxynitrite. Biochem. Biophys. Res. Commun. 257, 651-656. (6) Burney, S., Caufield, J. L., Niles, J. C., Wishnok, J. S., and Tanninbaum, S. R. (1999) The chemistry of DNA damage from nitric oxide and peroxynitrite. Mutat. Res. 424, 37-49. (7) Alvarez, B., Ferrer-Sueta, G., Freeman, B. A., and Radi, R. (1999) Kinetics of peroxynitrite reaction with amino acids and human serum albumin. J. Biol. Chem. 274, 842-848. (8) Kato, Y., Kawakishi, S., Aoki, T., Itakura, K., and Osawa T. (1997) Oxidative modification of tryptophan residues exposed to peroxynitrite. Biochem. Biophys. Res. Commun. 234, 82-84. (9) Radi, R., Beckman, J. S., Bush, K. M., and Freeman, B. A. (1991) Peroxynitrite oxidation of sulfhydryls. The cytotoxic potential of superoxide and nitric oxide. J. Biol. Chem. 266, 4244-4250. (10) Radi, R., Beckman, J. S., Bush, K. M., and Freeman, B. A. (1991) Peroxynitrite-induced membrane lipid peroxidation: The cytotoxic potencial of superoxide and nitric oxide. Arch. Biochem. Biophys. 288, 481-487.

326


(11) Koppenol, W. H., Moreno, J. J., Pryor, W. A., Ischiropoulos, H., and Beckman, J. S. (1992) Peroxynitrite, a cloaked oxidant formed by nitric oxide and superoxide. Chem. Res. Toxicol. 5, 834-842. (12) Padmaja, S., Squadrito, G. L., and Pryor, W. A. (1998) Inactivation of glutathione by peroxynitrite. Arch. Biochem. Biophys. 349, 1-6. (13) Zhang, H., Squadrito, G. L., Uppu, R. M., Lemercier, J.-N., Cueto, R., and Pryor, W. A. (1997) Inhibition of peroxynitrite-mediated oxidation of glutathione by carbon dioxide. Arch. Biochem. Biophys. 339, 183-189. (14) Lemercier, J. N., Padmaja, S., Cueto, R., Squadrito, G. L., Uppu, R. M., and Pryor, W. A. (1997) Carbon dioxide modulation of hydroxylation and nitration of phenol by peroxynitrite. Arch. Biochem. Biophys. 345, 160-170. (15) Ischiropoulos, H., Zhu, L., Chen, J., Tsai, M., Martin, J. C., Smith, C. D., and Beckman, J. S. (1992) Peroxynitrite-mediated tyrosine nitration catalyzed by superoxide dismutase. Arch. Biochem. Biophys. 298, 431-437. (16) Grune, T., Blasig, I. E., Sitte, N., Roloff, B., Haseloff, R., and Davies, K. J. A. (1998) Peroxynitrite increases the degradation of aconitase and other cellular proteins by proteasome. J. Biol. Chem. 273, 10857-10862. (17) Winck, D. A., and Mitchell, J. B. (1998) Chemical biology of nitric oxide: insights into regulatory, cytotoxic and cytoprotective mechanism of nitric oxide. Free Radical Biol. Med. 25, 434-456. (18) Lymar, S. V., Jiang, Q., and Hurst, J. K. (1996) Mechanism of carbon dioxide-catalyzed oxidation of tyrosine by peroxynitrite. Biochemistry 35, 7855-7861. (19) Jourd’heuil, D., Miranda, K. M., Kim, S. M., Espey, M. G., Voddovotz, Y., Laroux, S., Mai, C. T., Miles, A. M., Grisham, G., and Wink, D. A. (1999) The oxidative and nitrosative chemistry of nitric oxide/superoxide reaction in the presence of bicarbonate. Arch. Biochem. Biophys. 365, 92-100. (20) Bonini, M. G., Radi, R., Ferrer-Sueta, G., Ferreira, A. M., and Augusto, O. (1999) Direct EPR detection of the carbonate radical anion produced from peroxynitrite and carbon dioxide. J. Biol. Chem. 274, 10802-10806. (21) Uppu, R. M., Winston, G. W., and Pryor, W. A. (1997) Reactions of peroxynitrite with aldehydes as probes for reactive intermediates responsible for biological nitration. Chem. Res. Toxicol. 10, 1331-1337. (22) Nakao, L. S., Ouchi, D., and Augusto, O. (1999) Oxidation of acetaldehyde by peroxynitrite and hydrogen peroxide-iron(II). Production of acetate and methyl radicals. Chem. Res. Toxicol. 12, 1010-1018. (23) Va´squez-Vivar, J., Denicola, A., Radi, R., and Augusto, O. (1997) Peroxynitrite-mediated decarboxylation of pyruvate to both carbon dioxide and carbon dioxide radical anion. Chem. Res. Toxicol. 10, 786-794. (24) Bechara, E. J. H., Oliveira, O. M. M. F., Duran, N., Baptista, R. C., and Cilento, G. (1979) Peroxidase-catalyzed generation of triplet acetone. Photochem. Photobiol. 30, 101-110. (25) Bechara, E. J. H., and Soares, C. H. (1982) Enzymatic generation of triplet biacetyl. Photochem. Photobiol. 36, 117-119. (26) Cilento, G. (1995) Enzymatic generation of electronically excited triplet species. Photochemistry without light. Cienc. Cult. (Sao Paulo) 47, 312-319. (27) Alcais, P., and Brouillard, R. (1972) Influence of structure on β-diketo-enol equilibria: a kinetic study by the relaxation technique. J. Chem. Soc., Perkin Trans. 2, 1214-1219. (28) Bruice, P. Y. (1990) Role of the acidity of the ketone in determining the mechanism of enolization via proton abstraction from ketone carbinolamine or imine: Catalyis of the enolization of 2,4pentanedione and 3-methyl-2,4-pentanedione by oxyanions and by primary, secondary and tertiary amines. J. Am. Chem. Soc. 112, 7361-7368. (29) Uppu, R. M., and Pryor, W. A. (1996) Carbon dioxide catalysis of the reaction of peroxynitrite with ethyl acetoacetate: an example of aliphatic nitration by peroxynitrite. Biochem. Biophys. Res. Commun. 229, 764-769.

Knudsen et al. (30) Beckman, J. S., Beckman, T. W., Chen, J., Marshall, P. A., and Freeman, B. A. (1990) Apparent hydroxyl radical production of peroxynitrite. Implications for endothelial injury from nitric oxide and superoxide. Proc. Natl. Acad. Sci. U.S.A. 87, 1620-1624. (31) Catalani, L. H., Wilson, T., and Bechara, E. J. H. (1987) Watersoluble fluorescence probes for chemiexcitation studies: sodium 9,10-dibromoanthracene-2-sulfonate and 9,10-diphenylanthracene2-sulfonate. Synthesis, properties and application to triplet acetone and tetramethyldioxetane. Photochem. Photobiol. 45, 273-281. (32) Mansilla-Espinosa, A., Duran-Me´ras, I., and Salinas, F. (1998) High perfomance liquid chromatographic-fluorometric determination of glyoxal, methylglyoxal and diacetyl in urine by prederivatization to pteridinic rings. Anal. Biochem. 255, 263-273. (33) Baader, W. J., Bohne, C., Cilento, G., and Dunford, B. H. (1985) Peroxidase-catalyzed formation by triplet acetone and chemiluminescence of isobutyraldehyde and molecular oxygen. J. Biol. Chem. 260, 10217-10225. (34) Chiang, Y., Kresge, A. J., and Walsh, P. A. (1986) Generation of simple enols in aqueous-solution from alkali-metal enolates. Some chemistry of isobutyraldehyde enol. J. Am. Chem. Soc. 108, 63146320. (35) Bohne, C., Macdonald, I. D., and Dunford, H. B. (1987) Transient state kinetics of the reaction of isobutyraldehyde with compound I and compound II of horseradish peroxidase. J. Biol. Chem. 262, 3572-3578. (36) Guthrie, J. P., and Cullimore, P. A. (1979) Enol content of simple carbonyl compounds: a thermochemical approach. Can. J. Chem. 57, 240-248. (37) Dubois, J.-E., El-Alaoui, M., and Toullec, J. (1981) Kinetics and thermodynamics of keto-enol tautomerism of simple carbonyl compounds: An approach based on a kinetic study of halogenation at low concentration. J. Am. Chem. Soc. 103, 5393-5401. (38) Toullec, J. (1984) Constants of ketone-enol equilibrium of aliphatic, alicyclic and aromatic ketones in water. Tetrahedron Lett. 25, 4401-4402. (39) Mills, S. G., and Beak, P. (1985) Solvent effects on keto-enol equilibria. Tests of quantitative models. J. Org. Chem. 50, 12161224. (40) Squadrito, G. L., Jin, X., and Pryor, W. A. (1995) Stopped-flow kinetics study of the reaction of ascorbic acid with peroxynitrite. Arch. Biochem. Biophys. 322, 53-59. (41) Ettinger, K. V., Forrester, A. R., and Hunter, C. H. (1982) Lyoluminescence and spin trapping. Can. J. Chem. 60, 15491559. (42) Mottley, C., Robinson, R. E., and Mason, R. P. (1991) Free radical formation in the oxidation of malondialdehyde and acetylacetone by peroxidase enzymes. Arch. Biochem. Biophys. 289, 153-160. (43) Bentrude, W. G., and Darnall, K. R. (1968) Biacetyl photochemistry: products in solution. J. Chem. Soc., Chem. Commun., 810812. (44) Cilento, G., and Adam, W. (1995) From free radicals to electronically excited species. Free Radical Biol. Med. 19, 103-114. (45) De la Fuente, U., and Lissi, E. A. (1992) Excited carbonyl formation in the combination and disproportionation of free radicals. J. Biolumin. Chemilumin. 7, 27-35. (46) Cadenas, E. (1989) Biochemistry of oxygen toxicity. Annu. Rev. Biochem. 58, 79-110. (47) Mendenhall, G. D., Sheng, X. C., and Wilson, T. (1991) Yields of excited carbonyl species from alkoxyl and from alkylperoxyl radical dismutation. J. Am. Chem. Soc. 113, 8976-8977. (48) Timmins, G. S., Santos, R. E., Whitwood, A. C., Catalani, L. H., Di Mascio, P., Gilbert, B., and Bechara, E. J. H. (1997) Lipid peroxidation-dependent chemiluminescence from cyclization of alkylperoxyl radicals to dioxetane radical intermediates. Chem. Res. Toxicol. 10, 1090-1096.

TX990176I

Chemiluminescent Aldehyde and - American Chemical Society

Recommend Documents