Lippincott Williams & Wilkins, Inc., Philadelphia
Inhibition of Peroxynitrite-Mediated Oxidation of Dopamine by Flavonoid and Phenolic Antioxidants and Their Structural Relationships
Article first published online: 18 JAN 2002
Journal of Neurochemistry
Volume 73, Issue 1, pages 247–253, July 1999
How to Cite
Kerry, N. and Rice-Evans, C. (1999), Inhibition of Peroxynitrite-Mediated Oxidation of Dopamine by Flavonoid and Phenolic Antioxidants and Their Structural Relationships. Journal of Neurochemistry, 73: 247–253. doi: 10.1046/j.1471-4159.1999.0730247.x
Abbreviations used: Rt, retention time; λmax, spectral maximum.
- Issue published online: 18 JAN 2002
- Article first published online: 18 JAN 2002
- Caffeic acid;
- Ferulic acid;
Abstract: The interaction between peroxynitrite and dopamine and the inhibition of this reaction by plant-derived antioxidants have been investigated. Peroxynitrite promoted the oxidation of dopamine to 6-hydroxyindole-5-one as characterised by HPLC and photodiode array spectra, akin to the products of the tyrosinase-dopamine reaction, but no evidence of dopamine nitration was obtained. Although peroxynitrite did not cause nitration of dopamine in vitro, the catecholamine is capable of inhibiting the formation of 3-nitrotyrosine from peroxynitrite-mediated nitration of tyrosine. The plant-derived phenolic compounds, caffeic acid and catechin, inhibited peroxynitrite-mediated oxidation of dopamine. This effect is attributed to the ability of catechol-containing antioxidants to reduce peroxynitrite through electron donation, resulting in their oxidation to the corresponding o-quinones. The antioxidant effect of caffeic acid and catechin was comparable to that of the endogenous antioxidant, glutathione. In contrast, the structurally related monohydroxylated hydroxycinnamates, p-coumaric acid and ferulic acid, which inhibit tyrosine nitration through a mechanism of competitive nitration, did not inhibit peroxynitrite-induced dopamine oxidation. The findings of the present study suggest that certain plant-derived phenolics can inhibit dopamine oxidation.
Oxidative damage to neuronal cells and DNA is implicated in the pathogenesis of neurodegenerative diseases (Götz et al., 1994; Jenner and Olanow, 1996, 1998; Beal, 1997). This observation is supported by numerous biochemical markers of oxidative stress in the parkinsonian brain, including elevation in products of lipid peroxidation and dopamine oxidation, iron deposition in the substantia nigra, as well as decreased levels of glutathione and ascorbate (Dexter et al., 1986; Riederer et al., 1989; Jenner et al., 1992). Oxidising systems that generate superoxide and hydroxyl radicals in vitro mediate the oxidation of dopamine to o-quinone, which in the presence of L-cysteine can undergo covalent addition reactions to form 5-S-cysteinyldopamine (Palumbo et al., 1995; Spencer et al., 1998).
The potential role of reactive nitrogen species, in particular peroxynitrite, in the pathophysiology of neurodegenerative disease is relevant, because its precursors, nitric oxide and superoxide anion, are colocalised in brain tissue. Levels of 3-nitrotyrosine, a biomarker for reactive nitrogen species, are elevated in neurodegenerative diseases, such as Parkinson’s, Alzheimer’s, and amyotrophic lateral sclerosis (Abe et al., 1997; Smith et al., 1997; Su et al., 1997; Good et al., 1998; Hensley et al., 1998).
Several in vitro studies have identified 6-nitro derivatives as the major product following the incubation of catecholamines (dopamine, adrenaline, and noradrenaline) with reactive nitrogen species derived from nitrous acid and nitrogen oxides in acid (Roth and Volkmann, 1969; de la Breteche et al., 1994; d’Ischia and Costantini, 1995; Daveu et al., 1997). The hydroxycinnamic acids, in particular caffeic acid, are structurally very similar to catecholamines (Fig. 1) and are synthesised in plants from the common precursor, tyrosine via the shikimate pathway (Ibrahim and Barron, 1989), whereas in sympathetic neurons, tyrosine is hydroxylated to L-DOPA, which is decarboxylated enzymatically to dopamine. The main structural differences between dopamine and caffeic acid are the absence of an amino group and the presence of an unsaturated carbon side chain and carboxylic acid moiety in the latter compound. A number of flavonoids contain the catechol moiety within the B-ring of their structures. Flavonoids and hydroxycinnamates, major constituents of the human diet, are powerful electron-donating antioxidants, the catechol moiety being one of the main structural features responsible for their antioxidant activities (Bors and Saran, 1987; Rice-Evans et al., 1996). Recent studies have shown that caffeic acid, a catechol-containing antioxidant derived from plants, preferentially inhibits peroxynitrite-mediated tyrosine nitration by a mechanism of electron donation (Kerry and Rice-Evans, 1998), whereas related monophenolic hydroxycinnamates (ferulic acid and p-coumaric acid) inhibit nitration of tyrosine through competitive nitration (Pannala et al., 1998).
The interaction of catecholamines with peroxynitrite has not been widely investigated. The aim of the present study was to investigate the peroxynitrite-mediated oxidation of dopamine and the competitive inhibition of tyrosine nitration by dopamine. The ability of phenolic antioxidants, both monophenolic and catecholate structures, to inhibit the reaction of dopamine with peroxynitrite was also investigated in relation to their structural dependence, as well as comparative reactivities with ascorbate and glutathione.
MATERIALS AND METHODS
The following chemicals were obtained from Sigma Chemical Company (Poole, Dorset, U.K.): dopamine (3,4-dihydroxy-phenethylamine), mushroom tyrosinase (4,800 U/mg of solid; EC 184.108.40.206), sodium nitrite, EDTA, L-cysteine, glutathione, ascorbic acid, caffeic acid, ferulic acid, p-coumaric acid, and (+)-catechin. Hydrogen peroxide solution (30%, wt/vol), citric acid, and potassium dihydrogen orthophosphate were purchased from BDH Laboratory Supplies (Poole, Dorset, U.K.). The sodium n-nonyl sulphate was obtained from Lancaster Synthesis Ltd. (Morecambe, Lancashire, U.K.). HPLC grade acetonitrile was purchased from Rathburn (Walkerburn, Scotland, U.K.). All the compounds were prepared using ultrapure water (18.2 MΩ Maxima Ultrapure water).
Peroxynitrite reaction with dopamine
The reaction between dopamine and peroxynitrite was investigated. Peroxynitrite was prepared by a modification of the method described by Beckman et al. (1994). Two syringes, analogous to a stopped flow apparatus, were used to rapidly mix acidified hydrogen peroxide (20 ml of 1 M H2O2 in 0.7 M HCl, final concentration) with sodium nitrite (20 ml of 0.2 M, final concentration), resulting in the formation of peroxynitrous acid, which was quenched immediately with ice-cold potassium hydroxide (1.5 M in 40 ml) to form peroxynitrite anion. Manganese dioxide was used to remove unreacted hydrogen peroxide. The solution was filtered, and the peroxynitrite concentration was determined from its UV absorbance at 302 nm using ε302 = 1,670 M-1 cm-1 (Hughes and Nicklin, 1968). The typical yields of freshly prepared peroxynitrite were 40 mM. Peroxynitrite was made fresh each day and was stored on ice for no longer than 1 h before experimentation, to minimise any degradation of peroxynitrite in alkaline solutions.
Peroxynitrite (final concentration, 4 mM) was reacted with dopamine (final concentration, 4 mM) in 50 mM phosphate buffer at pH 7.0 at room temperature for 15 min. In some experiments, L-cysteine (final concentration, 8 mM) was added to samples after the addition of peroxynitrite. The samples were diluted 1:1 (vol/vol) with 0.05 M HCl before HPLC injection.
The inhibition of peroxynitrite-mediated tyrosine nitration by dopamine
The ability of dopamine to inhibit peroxynitrite-mediated 3-nitrotyrosine formation was investigated. A single concentration of tyrosine (100 μM) was reacted with peroxynitrite (500 μM) in the presence of varying concentrations of dopamine (0-250 μM), and the formation of 3-nitrotyrosine was quantified by HPLC. The ability of dopamine to inhibit peroxynitrite-mediated reactions was expressed as the percentage inhibition of 3-nitrotyrosine formation.
The inhibition of peroxynitrite reaction with dopamine
The ability of compounds to inhibit the reaction between dopamine and peroxynitrite was studied. In control experiments, dopamine (100 μM) was reacted with peroxynitrite (500 μM). For inhibition experiments, dopamine was coincubated with the relevant antioxidant (50-1,000 μM) before the addition of peroxynitrite. The compounds investigated for inhibitory activity were ascorbic acid, glutathione, catechin, caffeic acid, ferulic acid, and p-coumaric acid. Inhibition of peroxynitrite-dependent depletion of dopamine was expressed as the percentage of dopamine reacted, such that a low percentage of dopamine reacted was equated with an increased peroxynitrite scavenging by the antioxidant. The concentration of substrate, either dopamine or antioxidant (caffeic, ferulic, p-coumaric acids, and catechin), remaining in the samples after the reaction with peroxynitrite was determined by HPLC with photodiode array detection. Calibration curves for dopamine (10-100 μM) and antioxidants (10-1,000 μM) versus peak areas were constructed. Glutathione, L-cysteine, and ascorbic acid were undetected in the HPLC conditions used for these experiments.
Tyrosinase-mediated oxidation of dopamine
Preparative experiments were performed to identify and characterise the products formed following the reaction of dopamine with tyrosinase, based on methods by Rosengren et al. (1985). Dopamine (final concentration, 10 mM) was incubated with 1,000 U of tyrosinase in 50 mM phosphate buffer, pH 6.0. The samples were incubated at 25°C for 15 min, after which time 50 μl of sample was diluted in 950 μl of 0.05 M HCl ready for injection onto HPLC. To synthesize cysteinyl-dopamine, a twofold higher concentration of L-cysteine (final concentration, 20 mM) was coincubated with dopamine and tyrosinase at 25°C for 30 min.
Acidified nitrite-mediated nitration of dopamine
The synthesis of 6-nitrodopamine was performed by incubating dopamine with acidic nitrite. The reaction conditions were similar to those described by Oldrieve et al. (1998), whereby dopamine (final concentration, 1 mM) was reacted with an equimolar concentration of sodium nitrite in 0.5 M HCl. The formation of reaction products was analysed by HPLC.
The detection of 3-nitrotyrosine, antioxidants, dopamine, and reaction products was performed by HPLC with photodiode array detection (Waters 996 Photodiode Array Detector). For quantitative experiments, peak areas were obtained by detection at 280 nm with the spectra of the resolved peaks obtained between 200 and 600 nm. The HPLC conditions used to separate compounds was based on the method described by Fornstedt et al. (1990) using a Novapak C18 4 μm column (250 × 4.6 mm; Waters, Watford, U.K.) and a mobile phase consisting of 0.32 mmol/L sodium n-nonyl sulphate, 0.054 mmol/L EDTA, 3.5 mmol/L KH2PO4, and 46 mmol/L citric acid, pH 2.2. An isocratic system comprising 90% buffer and 10% acetonitrile was pumped at a flow rate of 0.8 ml/min. The injection volume of samples was 50 μl (Waters 717 plus Autosampler), and data acquisition was performed with Chromatography Manager software (Waters, Milford, MA, U.S.A.).
All results are expressed as means ± SD of three separate experiments unless stated otherwise. The UV-visible photodiode array spectra (see Fig. 2) are representative spectra of products separated by reverse-phase HPLC. Statistical evaluation of the dopamine-dependent inhibition of 3-nitrotyrosine formation and of the effect of antioxidants on dopamine oxidation was by parametric one-way analysis of variance (ANOVA) with Tukey multiple comparisons using GraphPad Prism (GraphPad Software, San Diego, CA, U.S.A.). Significance was defined as p < 0.05. The relationship between the concentration of phenolics (caffeic acid and catechin) and dopamine after the reaction with dopamine was determined by linear regression analysis.
The reactions of dopamine with tyrosinase and with acidified nitrite were undertaken as model reactions that yield oxidation and nitration products, respectively, and the results of these experiments were used to confirm the identity of products derived from the reaction of peroxynitrite with dopamine. The spectral characteristics of the products were confirmed by HPLC with photodiode array detection by comparison with previously published data that have characterised dopamine oxidation products by electrochemical methods using cyclic voltammetry and UV-visible and mass spectroscopy (Graham, 1978; Zhang and Dryhurst, 1993).
Oxidation of dopamine by tyrosinase
Tyrosinase-mediated oxidation of dopamine (10 mM) yielded a bright orange chromophore. HPLC analysis of the reaction products revealed a peak with a retention time (Rt) of 5.2 min and spectral maxima (λmax) at 289 and 479 nm (Fig. 2A). The Rt and λmax for dopamine and products are summarised in Table 1. In the presence of L-cysteine (20 mM), a pale yellow solution was obtained. No peak corresponding to Rt of 5.2 min and λmax of 479 nm was detected by HPLC. The major product of the tyrosinase-mediated dopamine oxidation in the presence of L-cysteine eluted at Rt of 31.9 min with λmax at 232, 252, and 292 nm (Fig. 2B, Table 1). These features were consistent with 5-S-cysteinyldopamine previously described (Rosengren et al., 1985; Zhang and Dryhurst, 1993; Spencer et al., 1998). A minor product eluting at Rt of 21.0 min had a similar absorption profile (λmax at 234, 256, and 293 nm), which may correspond to 2-S-cysteinyldopamine (Fig. 2B, Table 1). The addition of cysteine 15 min after incubating dopamine and tyrosinase resulted in the formation of detectable levels of 6-hydroxyindole-5-one, 5-S-cysteinyldopamine, and 2-S-cysteinyldopamine. Incubating cysteine alone with tyrosinase did not yield detectable products by HPLC in the present experimental conditions.
|Incubation||Rt (min)||λmax (nm)||Product name|
|Dopamine + tyrosinase||5.2||289, 479||6-Hydroxyindole-5-one|
|Dopamine + cysteine + tyrosinase||31.9||232, 252, 292||5-S-Cysteinyldopamine|
|21.0||234, 256, 293||2-S-Cysteinyldopamine|
|Dopamine + peroxynitrite||5.0||285, 469||6-Hydroxyindole-5-one|
|Dopamine + acidic nitrite||10.8||233, 279, 388||6-Nitrodopamine|
Nitration of dopamine by acidified nitrite
The reaction of dopamine (1 mM) with acidified nitrite (1 mM in 0.5 M HCl) resulted in the formation of a bright yellow chromophore. HPLC analysis of this sample detected a single product, Rt of 10.8 min, which exhibited a similar spectrum to that of dopamine but with an additional absorption peak at 388 nm in the acidic conditions, indicative of a nitrated aromatic compound (Fig. 2C, Table 1) compared with dopamine itself (Fig. 2D, Table 1).
Reaction of dopamine with peroxynitrite
The formation of dopamine-derived oxidation and/or nitration products following the reaction of dopamine with peroxynitrite was analysed by HPLC with photodiode array detection. The direct interaction of dopamine with peroxynitrite (molar ratio 1:1) led to the formation of a product with Rt of 5.0 min and λmax of 285 and 469 nm (Fig. 2A). Nitrodopamine was not detected by HPLC analysis following the reaction of dopamine with peroxynitrite. The product obtained from the reaction of peroxynitrite with dopamine was confirmed as an oxidation product by the identification of the orange chromophore, 6-hydroxyindole-5-one (dopaminochrome), which exhibits a characteristic λmax at 470 nm. This finding indicates that peroxynitrite and tyrosinase mediate the same reactions, leading to the formation of 6-hydroxyindole-5-one. This chemical oxidation and cyclisation of dopamine to form indolic compounds are supported by the observation that samples stored over a short period of time formed a black insoluble material.
In contrast to the reaction of dopamine and cysteine with tyrosinase, we could not demonstrate the formation of 5-S-cysteinyldopamine by reacting dopamine with peroxynitrite in the presence of excess cysteine (molar ratio 1:2:1, dopamine/cysteine/peroxynitrite). The scavenging activity of thiol compounds toward peroxynitrite precluded the formation of detectable levels of 5-S-cysteinyldopamine. The addition of cysteine, either before or after reacting dopamine with peroxynitrite (molar ratio 1:2:1, dopamine/cysteine/peroxynitrite), resulted in the formation of low levels of 6-hydroxyindole-5-one exhibiting a Rt of 5.0 min and λmax of 469 nm. However, this product was not detected on interaction of cysteine with peroxynitrite before addition of dopamine. Several other products were detected by HPLC following the reaction of dopamine with peroxynitrite in the presence of cysteine. The spectral characteristics and Rt of these products were comparable to those of the products identified from the reaction of cysteine with peroxynitrite and postreaction addition of dopamine, indicating that these products may be derived from cysteinyl radical-mediated reactions (data not shown). Products of the reaction of cysteine alone with peroxynitrite were not detected using the above HPLC procedure.
The inhibition of peroxynitrite-mediated tyrosine nitration by dopamine
Approximately 32% of tyrosine was nitrated (32.4 ± 8.9 μM 3-nitrotyrosine) following the reaction of tyrosine (100 μM) with peroxynitrite (500 μM). The nitration of tyrosine by peroxynitrite was inhibited by dopamine in a concentration-dependent manner (p < 0.001; Fig. 3). Approximately 50% inhibition of 3-nitrotyrosine formation was achieved with 37.5 μM dopamine, whereas complete inhibition was observed with a dopamine concentration of 250 μM. The ability of dopamine to react with peroxynitrite was also calculated as the percentage of dopamine reacted with peroxynitrite. Dopamine (initial concentration, 100 μM) was 71.8 ± 1.6% (n = 6 experiments) depleted by the reaction with 500 μM peroxynitrite. In the presence of tyrosine (100 μM), the concentration of dopamine reacted with peroxynitrite remained unchanged at 70.5 ± 0.5% (n = 3).
The inhibition of peroxynitrite reaction with dopamine
The inhibition of peroxynitrite-mediated oxidation of dopamine by a range of antioxidants is illustrated in Fig. 4. Ascorbic acid was a very potent inhibitor of peroxynitrite-mediated dopamine oxidation. At equimolar concentrations of dopamine and ascorbic acid (100 μM), the percentage concentration of dopamine reacted with peroxynitrite was decreased from ∼72% in control incubations to 35.2 ± 3.8% (p < 0.001). Complete inhibition of dopamine oxidation by peroxynitrite was achieved in the presence of a 2.5 molar excess of ascorbic acid (p < 0.001). At 100 μM, glutathione, catechin, and caffeic acid significantly inhibited dopamine oxidation (p < 0.001), but were ∼30% as effective as ascorbate. At higher concentrations up to 1,000 μM, caffeic acid exhibited greater peroxynitrite scavenging activity than catechin (p < 0.01) and glutathione (p < 0.001), whereas catechin was also more effective than glutathione (p < 0.05). In contrast, the monohydroxylated hydroxycinnamic acids, ferulic and p-coumaric acids, had no inhibitory effect on the reaction between dopamine and peroxynitrite.
The relationship between the concentration of substrates after reaction with peroxynitrite is shown in Fig. 5. A significant inverse correlation exists between the concentration of dopamine and caffeic acid reacted (linear regression, r2 = 0.979, p = 0.0002) and between dopamine and catechin reacted (r2 = 0.916, p = 0.0027), indicating that competitive inhibition of peroxynitrite-mediated dopamine oxidation is achieved by the phenolic compounds. No such relationship was apparent for the monohydroxylated compounds, p-coumaric and ferulic acids, neither of which inhibited peroxynitrite-mediated oxidation of dopamine.
The studies described here support the peroxynitrite-mediated oxidation of dopamine and formation of the o-quinone, the main oxidation product being 6-hydroxy-indole-5-one. Nitration of dopamine by peroxynitrite was not observed in the present study, consistent with the findings of Daveu et al. (1997) who, on the basis of spectroscopic studies, suggested oxidation of dopamine. Zhang and Dryhurst (1993) have characterised the electrochemical oxidation of dopamine to an o-quinone intermediate and the subsequent deprotonation and intramolecular cyclisation reactions that lead to the formation of 6-hydroxyindole-5-one. Although the formation of 5-S-cysteinyldopamine from the nucleophilic addition of L-cysteine with dopamine o-quinone has been reported previously (Palumbo et al., 1995; Spencer et al., 1998), the direct scavenging activity of thiols toward peroxynitrite means that cysteinyl-adduct formation is unlikely to occur following the reaction of catecholamines with peroxynitrite.
Phenolic compounds can inhibit peroxynitrite-mediated nitration of tyrosine either by acting as alternative substrates for nitration, in the case of monohydroxylated structures such as p-coumaric acid and ferulic acid, or by reducing reactive nitrogen species as has been demonstrated for catechol structures such as caffeic acid (Pannala et al., 1997, 1998; Kerry and Rice-Evans, 1998). In the present study, the peroxynitrite-mediated oxidation of dopamine was not inhibited by p-coumaric acid and ferulic acid. In contrast, caffeic acid was an effective inhibitor of peroxynitrite-mediated dopamine oxidation, and this is attributed to its ability to reduce reactive nitrogen species and to form an o-quinone (Kerry and Rice-Evans, 1998). Similarly, as shown here, dopamine inhibits the nitration of tyrosine induced by peroxynitrite not by competitive nitration, but by a coupled oxidation-reduction reaction.
Less is known about the products formed from the reaction between catechin and peroxynitrite. The ability of catechin to inhibit competitively the reaction between dopamine and peroxynitrite in the present study suggests that this compound is also acting as a reducing agent rather than a substrate for nitration. The comparable levels of dopamine “sparing” achieved by catechin and caffeic acid support a similar mechanism of action for these two catechol-containing phenolics. Further work is required to confirm the formation of catechin oxidation products in peroxynitrite-mediated reactions.
The most potent compound to inhibit peroxynitrite-mediated oxidation of dopamine was ascorbic acid. It is already known that ascorbic acid is an important antioxidant in the CNS, because animal studies have shown that a diet depleted of vitamin C reduces brain levels of this antioxidant and increases the formation of 5-S-cysteinyldopamine (Fornstedt and Carlsson, 1991). The effectiveness of ascorbic acid as an antioxidant may be also due to its ability to scavenge peroxynitrite in addition to reducing dopamine o-quinone to its original catechol structure (Tse et al., 1976).
A modest level of inhibition of dopamine oxidation was achieved by glutathione that is comparable to the competitive inhibition exhibited by caffeic acid and catechin. The inhibitory or scavenging action of glutathione toward peroxynitrite is well documented. Peroxynitrite is known to form a glutathionyl radical via the one-electron oxidation of the thiol, which can dimerise forming glutathione disulphide (Radi et al., 1990; Karoui et al., 1997). The formation of S-nitrosothiol is also a consequence of reactions between thiols and nitrogen oxides (Gow et al., 1997). To date, there is no evidence to suggest that glutathione is capable of reducing o-quinone to catechol structure (Nappi and Vass, 1994). The ability of glutathione to react with peroxynitrite is distinct from glutathione’s ability to undergo a nucleophilic addition reaction with dopamine o-quinone to form 5-S-glutathionyldopamine.
The results of this study indicate that a selective mechanism seems to exist whereby compounds susceptible to oxidation will react with peroxynitrite in preference to compounds that are substrates for nitration. This is consistent with our observation that catecholamines, like dopamine, are more susceptible to peroxynitrite-mediated damage than monohydroxylated compounds, such as tyrosine. At equimolar concentrations, dopamine reacts more rapidly with peroxynitrite than does tyrosine. This observation may have implications for the oxidation of dopamine by peroxynitrite that may contribute to the depletion of dopamine observed in Parkinson’s disease.
The ability of catecholamines, dopamine analogues and the larger tetrahydroisoquinoline derivatives, to bind to the dopamine transporter in striatal cells has been investigated previously, with particular emphasis on the role of the catechol moiety and the terminal amine functionality as mediators of binding and translocation, respectively (Meiergerd and Schenk, 1994; Kawai et al., 1998). From the current understanding of the structure-activity relationships for striatal dopamine transport, one could predict that caffeic acid and catechin, but not monohydroxylated p-coumaric acid and ferulic acid, are potential substrates for the dopamine transporter, because the catechol moiety mediates recognition. However, the absence of an amine group would slow the rate of transport of the hydroxycinnamate and catechin in striatal suspensions as previously observed for the deaminated catechol derivative, 4-ethylcatechol (Meiergerd and Schenk, 1994). These structural considerations may also have implications for the ability of caffeic acid to act as a substrate for aromatic amino acid transporters that facilitate the passage of compounds such as L-DOPA across the blood-brain barrier into the CNS (Wade and Katzman, 1975). The role of antioxidants, such as caffeic acid and catechin, in neuroprotection requires further investigation, with particular emphasis on the bioavailability of catechol derivatives in the CNS and uptake into neurons.
Financial support from Ministry of Agriculture, Fisheries and Food is gratefully acknowledged.
- 21997) Oxidative damage in neurodegenerative disease. Neuroscience 3 21–27.(
- 161968) The chemistry of peroxonitrites. Part I. The kinetics of decomposition of peroxynitrous acid. J. Chem. Soc. (A) 450–453.& (
- 171989) Phenylpropanoids, in Methods in Plant Biochemistry (Dey P. M. and Harborne J. B., eds), pp. 75–111. Academic Press, London.& (