| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J. Biol. Chem., Vol. 279, Issue 11, 9693-9697, March 12, 2004
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
¶
From the
Department of Pharmacology and Toxicology, Faculty of Medicine, Universiteit Maastricht; P.O. Box 616, NL 6200 MD Maastricht, The Netherlands and the Department of Medical Oncology, Vrije Universiteit Medical Center; De Boelelaan 1117, NL 1081 HV Amsterdam, The Netherlands
Received for publication, November 10, 2003 , and in revised form, December 16, 2003.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
1-antiproteinase (1-AP) inactivation, and glutathione S-transferase P1-1 (GST-P1-1) inactivation. In addition, nitration of tyrosine was measured to assess protection of the sulfur-containing compounds against ONOOH. For protection against HOCl, the efficacy of the antioxidant was controlled by the ratio of the reaction rates of the antioxidant and the detector molecule with the oxidant. The rank order of the activity of the antioxidants (GSH > DHLA 
LA GSMe > GSSG) appeared to be independent of the detector used. However, the rank order of the antioxidants against ONOOH-induced damage is strongly dependent on the detector. LA was 40 times less active than GSH in the inhibition of ONOOH-induced DHR-123 oxidation, whereas LA was 20 times more active than GSH in preventing the inhibition of GST-P1-1 by ONOOH. This points to different molecular mechanisms of ONOOH damage to DHR-123 compared with ONOOH damage to GST-P1-1. LA is a poor antioxidant in protecting against the form of ONOOH damage involved in DHR-123 oxidation. In the form of ONOOH toxicity involved in GST-P1-1 inhibition, LA is the most potent sulfur-containing antioxidant in our series. It is proposed that an intermediate product in which both sulfur atoms of LA have reacted is involved in the reaction of ONOOH with LA. The high potency of LA to protect GST-P1-1 against ONOOH might be of therapeutic interest. |
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
It has been found that lipoic acid (LA) is a very potent protector against ONOOH-mediated damage (1–4). In contrast, preliminary studies in our laboratory indicated that LA is a very poor protector against ONOOH-mediated oxidation of dihydrorhodamine-123 (DHR-123). This controversy might originate from differences in the procedures used to assess the antioxidant activity of LA. The activity of an antioxidant is usually determined by quantifying the ability of the antioxidant to prevent damage induced by a reactive species to a detector. Den Hartog et al. (7) emphasized that the efficacy of an antioxidant depends on the reaction rate of the oxidant with the detector molecule relative to the reaction rate of the oxidant with the antioxidant. This was illustrated by the difference in protective activity against the HOCl of several sulfur-containing antioxidants using different detectors (7). The slower the reaction of the detector with the oxidant, the higher was the apparent activity of a protector. The objective of the present study is to further examine the antioxidant activity of sulfur-containing compounds in the presence of different detectors (protein and non-protein) and different oxidants (ONOOH and HOCl) in order to resolve the apparent discrepancy in the efficacy of LA in several antioxidant assays. The results show that, with HOCl, the rate of the reaction of HOCl with the oxidant relative to that of the detector is essential. With ONOOH, the molecular mechanism of damage and protection is decisive for the efficacy of the antioxidant.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
1-antiproteinase (1-AP), glutathione S-transferase P-1–1 (GST-P1-1), GSH, GSSG, GSMe, 1-chloro-2,4-dinitrobenzene (CDNB) L-tyrosine, 3-nitro-L-tyrosine, 5,5-dithio-bis(2-nitrobenzoic acid) (DTNB), and sodium hypochlorite (NaOCl) were obtained from Sigma. DHR-123 and KO2 were obtained from Fluka Chemie GmbH (Buchs, Switzerland). LA and 
-LA were obtained from Asta Medica AG (Frankfurt, Germany). Nitrogen monoxide was obtained from AGA (Hamburg, Germany). All other chemicals were of the highest grade of purity. Synthesis of Potassium Oxoperonitrate—Potassium oxoperonitrate was produced from the reaction of solid KO2 with NO gas as described by Koppenol et al. (16). Briefly, the NO gas was slowly led over the mixture of KO2 and quartz sand, which was constantly stirred and kept on ice. The mixture was poured into a cold potassium hydroxide solution. Manganese dioxide was used to remove the hydrogen peroxide that results from decomposition of the excess of potassium superoxide. A solution with ONOOH was obtained by filtering off the sand and manganese dioxide. The concentration of ONOOH was determined spectrophotometrically at 302 nm.
Oxidation of DHR-123—The use of DHR-123 as a detector to monitor protection against ONOOH- and HOCl-induced oxidation was based on the procedure described by Kooy et al. (17). In short, 100 µl of the desired concentration of the test compound was added to 0.9 ml of 100 mM phosphate buffer at 37 °C. DHR-123 was added to give a final concentration of 5 µM. During vortexing, 10 µl of ONOOH or HOCl was added to give a final concentration of 0.6 or 5 µM, respectively. The fluorescent product, rhodamine-123, was measured by fluorescence detection with excitation and emission wavelengths of 500 and 536 nm, respectively. The effects are expressed as the concentration giving 50% inhibition of the oxidation of DHR-123 (IC50). Control experiments showed that the test compounds in a concentration of 500 µM did not affect the fluorescence.
Elastase Assay—Elastase activity was used as a detector according to the procedure described by Haenen and Bast (6) with minor modifications. The reagents were dissolved in potassium phosphate buffer (19 mM), pH 7.4, containing 140 mM NaCl. Twenty micrograms of 1-AP was preincubated at 37 °C for 5 min with the test compounds in the desired concentration. During vortexing, 10 µl of ONOOH was added. The final concentration of ONOOH was 150 µM. After 5 min incubation, 5 µg of elastase was added. After 10 min of incubation, 50 µl of a 10 mM solution of N-tert-butoxycarbonyl-L-alanine p-nitrophenol ester in methanol was added. The increase in absorption at 410 nm was determined. Control experiments showed that the test compounds in a concentration of 100 µM did not affect the activity of control or ONOOH-inactivated 1-AP.
Gluthathione-S-transferase P1-1 Assay—GST-P1-1 (5 µg/ml) was mixed with the test compound in 100 mM potassium phosphate buffer (pH 6.5), and the mixture was incubated at 37 °C for 1 min. During vortexing, ONOOH (10 µM, final concentration) was added. After 2 min of incubation, 1 mM 1-chloro-2,4-dinitrobenzene and 1 mM GSH were added. GST-P1-1 activity was monitored spectrophotometrically by recording the increase in absorbance at 340 nm (18).
Nitrotyrosine Assay—Under vortexing, 10 µl of ONOOH solution (final concentration 150 µM) was added to a solution containing 100 µM L-tyrosine in 100 mM potassium phosphate buffer (pH 7.4) and the test compound in the desired concentration. This mixture was incubated for 5 min. Measurement of 3-nitrotyrosine was carried out by using HPLC. The eluent was 50 mM KH2PO4/K2HPO4, pH 7, with 5% acetonitrile (v/v). The column was a Hypersil BDS-C18 column (150 x 4.6 mm) (Supelco Inc., Bellefonte, PA). The UV detector was set at 278 nm.
LA and GSH Consumption by ONOOH and HOCl—One hundred microliters of the antioxidant (dissolved in 100 mM phosphate buffer) was added to 700 µl of 100 mM phosphate buffer at 37 °C, giving a final concentration of 100 µM. During vortexing, 100 µl of the desired concentration of ONOOH or HOCl was added. GSH was determined spectrophotometrically by the addition of 100 µl of 6 mM 5,5-dithio-bis(2-nitrobenzoic acid) at 412 nm (19). LA was determined using HPLC and UV detection at 333 nm. A Hypersil BDS-C18 column (150 mm x 4.6 mm) (Supelco Inc.) was used. The mobile phase was 10 mM phosphate solution in water (pH 2.7) plus acetonitrile (65:35) (v/v). The samples were acidified before HPLC analysis (pH 2.6–2.9). Determination of -LA was performed on the same HPLC system at 230 nm. The mobile phase was 0.15% acetic acid in water/acetonitrile (75:25) (v/v).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
1-AP, and GST-P1-1 (Table I). Each of the sulfur-containing compounds tested had a prominent protective effect against HOCl except GSSG, which only had a moderate protective effect against HOCl with DHR-123 as a detector molecule.
|
1-AP, GST-P1-1, and tyrosine (Table II). GSH efficiently protected these detector molecules against ONOOH; the IC50 of GSH was practically independent of the detector. GSSG could not protect any of the detector molecules used against ONOOH. Methylation of GSH (giving GSMe) abolished the ability to protect against oxidation of DHR-123 and nitration of tyrosine with ONOOH. GSMe still had a considerable protective activity against ONOOH when 1-AP was used as a detector and, to a lesser extent, when GST P1-1 was used as a detector. Dihydrolipoic acid (DHLA), a dithiol, had a protective effect against ONOOH that was slightly lower than that of GSH when DHR-123 was used as detector. DHLA efficiently protected 1-AP or tyrosine against ONOOH-mediated inhibition. LA, a compound that contains an intramolecular disulfide, was hardly effective in the prevention of DHR-123 oxidation by ONOOH, but it potently protected against ONOOH-induced 1-AP damage and tyrosine nitration. Remarkably, both LA and DHLA had a very potent protective activity against ONOOH when GST-P1-1 was used as a detector. Their protective effect against ONOOH with this detector was 
20 times more potent than that of GSH.
|
1, i.e. one molecule of GSH reacts with one molecule of ONOOH. The apparent stoichiometry of LA was >1. With a 5-fold excess of ONOOH over LA, less than half of the LA was consumed. The product formed was -LA, the thiosulfinate of LA that has four possible isomers (4, 6).
|
-LA (Fig. 2).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
1-AP and GST-P1-1, against HOCl damage. Similarly, Den Hartog et al. (7) reported that acetylcholinesterase is efficiently protected by GSSG against HOCl-induced damage. This can also be explained by the relatively slow reaction of the target, acetylcholinesterase, with HOCl. Also, other proclaimed antioxidant activities of GSSG, e.g. the ability of GSSG to protect against NO2Cl-induced DNA damage (20), are probably the result of a relatively slow reacting detector (DNA). In vivo NO2Cl will probably react with other cellular targets besides DNA. In such a case, the reported antioxidant effect of GSSG is, of course, physiologically irrelevant.
GSH protected all the tested detectors efficiently against ONOOH, whereas GSSG is not a protector against ONOOH (Table II). GSMe did not protect DHR-123 and tyrosine, whereas it did protect the targets GST-P1-1 and 1-AP against ONOOH-induced inhibition. One of the vulnerable groups in 1-AP is methionine. Oxidation of methionine in 1-AP results in a loss of activity (21). Because in methionine and GSMe a chemically identical group, i.e. the S-methyl group (21), is attacked by ONOOH, it was anticipated that 1-AP is protected by GSMe. The relatively high IC50 of GSMe with DHR-123 and tyrosine can be explained by arguing that the reaction rate of these detectors with ONOOH is higher than that of GSMe, whereas the lower IC50 of GSMe with 1-AP or GST-P1-1 indicates that the reaction rate of ONOOH with these detectors is comparable with the reaction rate of ONOOH with GSMe. This explanation is in line with the explanation described above for the difference in activity of an antioxidant toward HOCl in different assays. The efficacy of an antioxidant depends on the detector used.
The most striking result is the finding that the antioxidants LA and DHLA protect GST-P1-1 far more efficiently against ONOOH than GSH (Table II). In contrast, LA provides a very poor protection against ONOOH-induced DHR-123 oxidation (Table II). GSSG was not able to protect the ONOOH-mediated oxidation of DHR-123, 1-AP, GST-P1-1, or tyrosine (Table II). Moreover, the rank order in potency of the antioxidants is not similar for the four detectors used (Table II). This implies that a difference in the protective activity of LA is not solely due to the difference in reactivity of the detectors. The low consumption of LA by ONOOH compared with GSH is peculiar (Fig. 1). At equimolar concentrations of LA and ONOOH (100 µM), the consumption of LA was only 13%. The relatively low consumption of LA by ONOOH was also noted previously. Trujillo and Radi (4) reported that only 25% of LA was consumed when 5 mM of LA was mixed with 5 mM ONOOH. This denotes that the protection of GST-P1-1 against ONOOH by LA, which is superior to the protection offered by GSH, is not due to direct scavenging of ONOOH by LA. Apparently, the mechanism of protection of GST-P1-1 against ONOOH by LA differs from the protection by GSH against ONOOH-mediated DHR-123 oxidation. Also, the difference in rank order of the potencies of LA and GSH obtained with the different detectors points to different mechanisms of damage and protection.
Tyrosine residues play a critical role in GST inactivation by ONOOH (22). Previous studies (2, 3) and the results presented in Table II revealed that LA offered an efficient protection against ONOOH-induced tyrosine nitration. The IC50 value of LA is six times lower than the initial concentration of ONOOH with tyrosine as a detector (IC50 = 24 µM; initial concentration of ONOOH = 150 µM) and >10 times lower than that of GST-P1-1 (IC50 = 0.9 µM; initial concentration of ONOOH = 10 µM). A similar difference was reported in the study of Whiteman et al. (2); the IC50 value of LA for the inhibition of tyrosine nitration was 10 times as low as the initial concentration of ONOOH. Theoretically, in a standard competition assay the minimal IC50 of the antioxidant is equal to half of the initial concentration of oxidant, i.e. when the antioxidant reacts much faster than the detector.
The explanations of Whiteman et al. (2) for the very low IC50 of LA compared with that of the initial concentration of ONOOH were, first, that one molecule of LA can scavenge several ONOOH molecules and, second, that LA can combine with reactive intermediates. The first explanation, viz. that LA scavenges several ONOOH molecules, does not seem probable. The oxidation products of LA that is formed by ONOOH is identical to that formed by HOCl (Figs. 1 and 2), i.e. -LA. This two-electron oxidation product of LA is formed after a reaction of one LA with probably one ONOOH (4) or one HOCl (23). This makes it unlikely that LA reacts with several ONOOH molecules. The second explanation of Whiteman et al. (2), i.e. that LA can combine with the reactive intermediates of ONOOH, seems more plausible. Whiteman et al. (2) stated that these intermediate products of ONOOH could be 
and 
.
Interestingly, Nakagawa et al. (3) reported that LA is a selective inhibitor of tyrosine nitration by ONOOH, but LA does not inhibit ONOOH-induced dityrosine formation from tyrosine. They argued that ONOOH has two reactive forms, i.e. a caged biradical [ONO·... ·OH] and a caged bipolar [ONO+... –OH] form. The caged biradical form of ONOOH yields dityrosine, and the caged bipolar form of ONOOH yields nitrotyrosine. LA was suggested as protecting against the bipolar form only, whereas thiol-containing compounds such as GSH and DHLA protect against both the bipolar and the biradical form. Although this explanation is highly speculative, it fits nicely within our experimental data. The oxidation of DHR-123 could, just as the oxidation of tyrosine into dityrosine, be mediated by the biradical form of ONOOH and, therefore, be inhibited by sulfhydryl-containing compounds and not by LA. Based on the prominent protective activity of LA, GST-P1-1 inactivation by ONOOH would be mediated by the bipolar form of ONOOH.
The major finding that is in conflict with this theory is the reported reaction of ONOOH by LA with a stoichiometry of 1:1 and a second order rate constant of 1400 M–1 s–1, which is almost identical to that of GSH (1350 M–+ s–1) (4). Trujillo and Radi (4) determined this rate constant by monitoring ONOOH consumption in a stopped-flow spectrophotometer. In the same study, the relatively low consumption of LA by ONOOH, identical to the observation depicted in Fig. 1, was also reported. This finding, i.e. the low consumption of LA in this study and study of Trujillo and Radi (4), is in conflict with the very high second order rate constant reported. The consumption of LA by increasing amounts of ONOOH should have been identical to that of GSH, as shown in Fig. 2 for HOCl. As shown in Fig. 1, this is clearly not the case for LA consumption by ONOOH. The low consumption of LA by ONOOH is more in line with the hypothesis that only a part of the ONOOH, i.e. the bipolar form, reacts with LA. It should be noted that Trujillo and Radi (4) calculated the apparent second order rate constant by dividing the observed rate of ONOOH consumption by the concentration of LA and the concentration of ONOOH. If only part of the ONOOH is able to react with LA, i.e. the bipolar form, the concentration of the bipolar form should have been used in the calculation instead of the total amount of ONOOH. Consequently, the apparent second order rate constant of the reaction of the bipolar form of ONOOH with LA would have been higher than the reported value of the rate of the second order reaction of ONOOH with LA. GSH reacts with both forms of ONOOH, and, therefore, the second order rate constant of GSH cannot be adjusted in the same way as for LA. This would make the true second order rate constant of the reaction of the bipolar form of ONOOH with LA higher than that with GSH. This could explain the higher efficacy of LA compared with GSH in the protection of GST-P1-1.
The question arises of from where the high protection of LA originates. It is not the reactivity of the separate sulfur atoms; they are far less reactive than the sulfur in GSH. The most prominent feature of LA is that the sulfur atoms are neighboring in a five-member ring. The orientation of the sulfur atoms in the 1,2-dithiolane ring causes a partial overlap of the lone pz electron pairs of the sulfur atoms. As a consequence, the pz orbitals combine into binding and anti-binding orbitals. The filled anti-binding orbital facilitates the ability of LA to donate electrons; in other words, this facilitates the ability of LA to reduce oxidants such as ONOOH and act as antioxidant. This is the explanation for the reactivity of LA as compared with a linear disulphide, such as GSSG, where an overlap of the pz electron pair is partially absent (6).
In the reaction scheme proposed (Fig. 3), the anti-binding electron pair binds one of the sulfur atoms to the positively charged nitrogen of the bipolar form of ONOOH. The binding electron that already has a partial double bond character (because of interchange with the free d-levels of the neighboring sulfur atoms) forms a double bond between the sulfur atoms. As a result, the sulfur atom that is not bound to the nitrogen becomes positively charged, and the negatively charged oxygen of the bipolar form of ONOOH binds to this sulfur atom. The intermediate that is formed in this way decomposes to give -LA and 
as shown in Fig. 3. In this reaction scheme it is crucial that the pz orbitals of the two sulfur atoms have been converted into a filled binding orbital and an anti-binding orbital. The net reaction of the proposed reaction scheme is the two-electron oxidation of LA by ONOOH yielding -LA and , which is consistent with Trujillo and Radi (4).
|
|
FOOTNOTES |
|---|
¶ To whom correspondence should be addressed.: Tel.: 31-43-3881418; Fax: 31-43-3884149; E-mail: a.bast{at}farmaco.unimaas.nl.
1 The abbreviations used are: ONOOH, peroxynitrite; HOCl, hypochlorous acid; LA, lipoic acid; DHR-123, dihydrorhodamine-123; 1-AP, anti-proteinase; GST-P1-1, glutathione S-transferase P1-1; GSMe, S-methylglutathione; HPLC, high pressure liquid chromatography; DHLA, dihydrolipoic acid.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  H. Ha, J.-H. Lee, H.-N. Kim, H.-M. Kim, H. B. Kwak, S. Lee, H.-H. Kim, and Z. H. Lee {alpha}-Lipoic Acid Inhibits Inflammatory Bone Resorption by Suppressing Prostaglandin E2 Synthesis J. Immunol., January 1, 2006; 176(1): 111 - 117. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| All ASBMB Journals | Molecular and Cellular Proteomics |
| Journal of Lipid Research | ASBMB Today |
