- Email: [email protected]
B Causes Oxidative Damage to Mitochondrial DNA
ARCHIVES OF BIOCHEMISTRY AND BIOPHYSICS
Vol. 335, No. 2, November 15, pp. 295–304, 1996 Article No. 0510
The Metabolism of Tyramine by Monoamine Oxidase A/B Causes Oxidative Damage to Mitochondrial DNA Nils Hauptmann, Joseph Grimsby, Jean C. Shih, and Enrique Cadenas1 Department of Molecular Pharmacology and Toxicology, School of Pharmacy, University of Southern California, Los Angeles, California 90033
Received June 11, 1996, and in revised form August 23, 1996
Monoamine oxidases A/B (EC 1.4.3.4, MAO), flavoenzymes located on the outer mitochondrial membrane, catalyze the oxidative deamination of biogenic amines, such as dopamine, serotonin, and norepinephrine. In this study, we examined whether the H2O2 formed during the two-electron oxidation of tyramine [4-(2-aminoethyl)phenol] (a substrate for monoamine oxidases A/B) may contribute to the intramitochondrial steady-state concentration of H2O2 ([H2O2]ss) and, thus, be involved in the oxidative impairment of mitochondrial matrix components. Supplementation of intact, coupled rat brain mitochondria with benzylamine, b-phenylethylamine, or tyramine showed initial rates of H2O2 production ranging from 0.4- to 1.6 nmol H2O2/min/mg protein. ESR analysis of the oxidative deamination of tyramine by intact rat brain mitochondria revealed the formation of hydroxyl (HO•) and carbon-centered radical adducts — the latter probably originating by the HO•mediated oxidation of mannitol. The signals were substantially enhanced upon addition of FeSO4 and were abolished by catalase. The intramitochondrial [H2O2]ss calculated in terms of glutathione peroxidase activity during the metabolism of tyramine was 48-fold higher (7.71 { 0.25 1 1007 M) than that obtained during the oxidation of succinate via complex II in the presence of antimycin A (1.64 { 0.2 1 1008 M). Oxidative damage to the brain mtDNA was assessed by single strand breakage. The ratio of nicked DNA for the preparations treated with tyramine and those without the amine was 1.5 { 0.29 (n Å 4), 2.12 { 0.28 (n Å 8, P £ 0.05), and 3.12 { 0.69 (n Å 3, P £ 0.05) at 15, 30, and 60 min, respectively. Preincubation of mitochondria with tranylcypromine (trans-2-phenylcyclopropylamine), an inhibitor to MAO A/B, abolished mtDNA oxidative damage. Catalase inhibited mtDNA strand breakage by approximately 60%. Incu-
1
To whom correspondence should be addressed.
bation of intact, coupled rat brain mitochondria with chlorodinitrobenzene (CDNB) depleted mitochondrial GSH by 72%. Tyramine-dependent damage of mtDNA was decreased by 68% in CDNB-treated mitochondria (with 28% remaining GSH). The [H2O2]ss was slightly increased in CDNB-treated mitochondria: 1.38- and 1.28-fold increase during the oxidation of succinate in the presence of antimycin A and during the oxidation of tyramine, respectively. These results suggest that the H2O2 generated during the MAO-catalyzed oxidation of biogenic amines and possibly certain neurotransmitters at the outer mitochondrial membrane contributes to the intramitochondrial [H2O2]ss and may cause oxidative damage to mtDNA. This is effected by the intramitochondrial concentration of GSH and might have potential implications for aging and neurodegenerative processes. q 1996 Academic Press, Inc. Key Words: monoamine oxidase; mitochondrial DNA; glutathione; neurodegenerative diseases; hydrogen peroxide.
Monoamine oxidases (EC 1.4.3.4, MAO-A/B),2 enzymes present in the outer mitochondrial membrane, catalyze the oxidation of biogenic amines accompanied by release of H2O2 (reaction [1]).
[1] 2 Abbreviations used: MAO, monoamine oxidase; GSH, glutathione; GSSG, glutathione disulfide; mtDNA, mitochondrial DNA; ESR, electron spin resonance spectroscopy; DMPO, dimethyl-1-pyrrolineN-oxide; CDNB, 1-chloro-2,4-dinitrobenzene.
295
0003-9861/96 $18.00 Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
AID
ARCH 9698
/
6b23$$$461
10-28-96 07:52:16
arca
AP: Archives
296
HAUPTMANN ET AL.
H2O2 produced during the oxidative deamination of catecholamines (e.g., dopamine) appears to be involved in the progress of neurodegenerative disorders, such as Parkinson disease and, presumably, via oxidative damage to the mitochondrial membrane (1). Several studies were undertaken to assess the extent of impairment of brain mitochondrial functions by this species and its relationship to neurodegenerative disorders. H2O2 produced during the metabolism of benzylamine, tyramine, and phenylethylamine by monoamine oxidases decreases significantly the intramitochondrial GSH content and increases the Ca2/ efflux (2). Likewise, intramitochondrial GSSG formation increased following the benzylamine metabolism by MAO-B; this effect was hardly affected upon supplementation with catalase. This suggests that the H2O2 produced at the outer mitochondrial membrane diffuses partly into the mitochondrial matrix and contributes to the oxidation of GSH (3). However, in the presence of iron, H2O2 formed by the metabolism of benzylamine by MAO-B leads to severe membrane lipid peroxidation, suggesting the homolytic cleavage of H2O2 by Fe2/ within a Fenton reaction (4, 5). It was proposed that lipid peroxidation products can mediate mtDNA damage (6). In addition to these effects, it has been suggested that H2O2 might have a stimulatory effect on cyclooxygenase (7) and MAO-B (8). Over the past years, evidence has accumulated that several human disorders, such as blindness, deafness, dementia, diabetes, and neurodegenerative diseases (e.g., Alzheimer, Parkinson, and Huntington diseases) are accompanied by mtDNA mutations leading mainly to dysfunctional mitochondria with an impaired oxidative phosphorylation activity (9). Furthermore, there is an age-related increase of MAO-B, which is associated with an augmentation of mtDNA damage and a decline in oxidative phosphorylation (10). Mitochondria possess a far less efficient DNA repair system than nuclei and mtDNA is not protected by histones (11, 12). mtDNA is involved in the coding of 13 essential genes and, expectedly, oxidative damage to mtDNA caused by an increased intramitochondrial H2O2 steady-state concentration may result in devastating effects for the mitochondrial functions. In this study, we examined whether the metabolism of tyramine ([4-(2-aminoethyl)phenol]) by MAO-A/B influences the intramitochondrial [H2O2]ss and whether this is associated with an oxidative impairment of mtDNA. We have also assessed the role of intramitochondrial GSH content on the [H2O2]ss and mtDNA single strand breakage.
sulfonic acid (Hepes), bovine serum albumin, 1-chloro-2,4-dinitrobenzene (CDNB), dansyl chloride, g-glu-glu (internal standard for GSH measurements), Nagarse, 4-[2-aminoethyl]phenol (tyramine), trans2-phenylcyclopropylamine (tranylcypromine), benzylamine, and phenylethylamine were purchased from Sigma (St. Louis, MO). GSH, horseradish peroxidase, glucose oxidase, proteinase K, ethidium bromide, and RNase A/T were supplied from Boehringer (Mannheim, Germany). Potassium phosphate, m-cresol purple sodium salt, lithium hydroxide, iodoacetic acid, Tris(hydroxymethyl)aminomethane hydrochloride, acetic acid, and sodium acetate were purchased from Fluka (Buchs, Switzerland). Other reagents were ordered as follows: sucrose and sodium dodecyl sulfate from GibcoBRL (Grand Island, NY); 5,5-dimethyl-1-pyrroline-N-oxide (DMPO), H2O2 , and p-hydroxyphenylacetic acid from Aldrich Chemical Co. (Milwaukee, WI); magnesium chloride and ferrous sulfate from Baker (Phillipsburg, NY); HPC-grade methanol from Baxter (Irvine, CA); agarose from FMC Bioproducts (Rockland, ME); [32P]ATP and [32P]CTP from ICN (Irvine, CA); and BglII from New England Biolabs (Mississauga, Canada).
MATERIALS AND METHODS
Single Strand Break Assay
Chemicals and Biochemicals Mannitol, glucose, ethylenediaminetetraacetic acid (EDTA), succinic acid, glutamic acid, N-2-hydroxyethylpiperazine-N*-2-ethane-
AID
ARCH 9698
/
6b23$$$462
10-28-96 07:52:16
Isolation of Mitochondria and General Assay Conditions Brain mitochondria from male Sprague–Dawley rats (300–350 g) were isolated and the P/O ratio was determined as described in established procedures (13, 14). The assay reaction buffer consisted of 250 mM mannitol, 70 mM sucrose, 1 mM EDTA, 10 mM potassium phosphate, pH 7.4. Protein concentration was determined with the bicinchoninic acid (BCA) protein assay (Pierce Co., Rockford, IL)
Spectrophotometric Measurements Fluorometric. The formation of H2O2 was measured fluorometrically using the horseradish peroxidase/H2O2-coupled dimerization of p-hydroxyphenylacetic acid (lexcitation 315 nm; lemission 410 nm) in an Aminco–Bowman spectrofluorometer (American Instrument Co., Silver Spring, MD) under gentle stirring. The H2O2-generating system (glucose/glucose oxidase) was used as standard for the detection of the continuous extramitochondrial H2O2 release. Electron spin resonance. ESR spectra were recorded on a Bruker ECS 106 equipped with TM 8810 microwave cavity. Measurements were carried out at a microwave frequency of 9.81 GHz and 100 KHz field modulation. DMPO was used as spin trap after extensive purification and produced no ESR signal by itself.
Depletion of Mitochondrial GSH GSH depletion was carried out by the glutathione transferasecatalyzed conjugation of the thiol with CDNB (15). Briefly, mitochondria (0.6 mg protein/ml) were incubated for 6 min with 60 mM CDNB dissolved in ethanol (final concentration 0.6%) and centrifuged for 2 min at 10,500 rpm. GSH content was evaluated by HPLC with fluorometrical detection as previously reported (16), using a 250 1 4.6-mm Spheri-5-amino column (Applied Biosystems Co., Forster City, CA). The latter was connected to a two-pump delivery system (Model LC-600), an autoinjector (Model SIL-9A), and a fluorometric detector (Model RF-535) from Shimadzu Scientific Instruments Inc. (Columbia, MD). For experiments requiring a further use of GSHdepleted mitochondria, the pellet was washed once with reaction buffer and the P/O ratio was checked again using a Clark-type oxygen electrode (Yellow Springs Instrument Co., Yellow Springs, OH).
Brain mtDNA, isolated as previously reported (17), was digested with BglII and a 1.9-kb fragment was cloned into pBSK to obtain a probe for the single strand break assay. The clone was verified by
arca
AP: Archives
297
METABOLISM OF TYRAMINE BY MONOAMINE OXIDASE A/B TABLE I
Rates of H2O2 Formation during the MAO A/B-Catalyzed Oxidative Deamination of Different Amines
Tyramine Phenylethylamine Benzylamine
MAO isoform
v (nmol H2O2 /min/mg protein)
A/B B/ B
1.6 1.0 0.4
Note. Values represent initial velocities and assay conditions were as described under Materials and Methods.
FIG. 1. Formation of H2O2 during the oxidation of various MAO substrates. Intact rat brain mitochondria (P/O 2.8 and 1.7 for glutamate and succinate, respectively) in the reaction buffer described under Materials and Methods were incubated with 0.2 mM substrate for 10 min and H2O2 was measured. (l) tyramine; (j) phenylethylamine; (m) benzylamine. Other assay conditions were as described under Materials and Methods.
restriction analysis. Following the treatment of mitochondria according to the experimental designs described under Results, mtDNA was isolated, electrophoresed in an agarose gel, and transferred (Southern blotting) to a Bio-Rad GT membrane (Bio-Rad, Hercules, CA). The transfer was hybridized to the 1.9-kb BglII radioactive labeled probe overnight, and washed in 21 SSC/0.1% SDS. Analysis was carried out with a Packard Instant Imager (Packard Instruments Co., Meriden, CT) and the occurrence of single strand breaks was expressed by the formula shown below as the number of single strand breaks per 104 base pairs. The statistically significance between tyramine-treated and control mitochondria was assessed by a paired t test. single strand break Å 0ln
MAO-A/B and, hence, its higher rate of metabolism and H2O2 formation may be accounted by the effective oxidation of this substrate by both isoforms of MAO. The second one is a preferred substrate for MAO-B, whereas the third is exclusively metabolized by MAO-B (Table I). A KM of 1.54 mM for tyramine was estimated from a double reciprocal plot (Fig. 2, inset). H2O2 accumulation could not be detected during the metabolism of dopamine due to an interference with the detection system; however, O2 uptake experiments (ascribed to O2 reduction to H2O2) revealed rates 2.7-fold lower than those obtained with tyramine. Tranylcypromine (trans-2-phenylcyclopropylamine) is a nonselective, nonhydrazine, irreversible inhibitor of MAO-A and B (18); preincubation of intact, coupled brain mitochondria caused a 96% inhibition of tyramine metabolism and, hence, of H2O2 formation.
supercoiled DNA . supercoiled DNA / nicked DNA
RESULTS
Monoamine Oxidase-Catalyzed Formation of Hydrogen Peroxide The rate of formation of H2O2 during the MAOcatalyzed metabolism of different biogenic amines was linearly related to the concentration of intact, coupled brain mitochondria (Fig. 1). The substrates examined were tyramine, phenylethylamine, and benzylamine, and the efficiency of H2O2 production followed the order: ktyramine Å 1.6; kphenylethylamine Å 1.0; kbenzylamine Å 0.4 nmol H2O2/min/mg protein, respectively. These rates are higher than those published elsewhere, probably due to the sensitivity of the method and the constant stirring allowing greater diffusibility of O2 . The first amine is metabolized by
AID
ARCH 9698
/
6b23$$$462
10-28-96 07:52:16
FIG. 2. Dependence of the rate of formation of H2O2 on tyramine concentration. Assay conditions: Intact brain mitochondria (0.3 mg protein/ml) in mannitol/sucrose/EDTA/potassium phosphate buffer, pH 7.4, were supplemented with various amounts of tyramine and H2O2 formation measured as described under Materials and Methods. The inset shows a double reciprocal plot, which permits to calculate a KM for tyramine of 0.65 mM.
arca
AP: Archives
298
HAUPTMANN ET AL.
Calculation of Intramitochondrial Steady-State Concentration of H2O2 during the Metabolism of Tyramine by MAO and Comparison to That Built during Mitochondrial Electron Transport The assessment of intramitochondrial H2O2 steadystate concentration has a deep implication for the understanding of oxidative damage to mitochondrial matrix components, such as oxidative injury of DNA, protein modifications, and the intramitochondrial thiol status. The intramitochondrial H2O2 steady-state concentration ([H2O2]ss) was calculated using the steady-state approximation method (19) by which the rate of production of H2O2 equals its rate of consumption or removal: /d[H2O2]/dt Å 0d[H2O2]/dt. The rate of formation of H2O2 equals that of consumption during the metabolism of 2 mM tyramine (Fig. 2) (concentration used in experiments involving mtDNA). This rate was 2.71 nmol/min/mg protein. Assuming that 1 mg mitochondrial protein Å 1 ml (20), the above rate corresponds to 4.52 1 1005 M s01. At least in the brain mitochondrial matrix, the removal of H2O2 is the domain of glutathione peroxidase (GPx); catalase appears to play no role in the removal of H2O2 under these experimental conditions because, on the one hand, its level is far lower than that described for heart mitochondria (21, 22) and, on the other hand, it appears to be localized in particles described as microperoxisomes (23). The concentration of GSH is not considered in this equation because it is present in large amounts (about 5–8 mM) and does not represent a limiting step. Mitochondria contain about 1.17 1 1006 M glutathione peroxidase, and the rate of H2O2 reduction by the enzyme is k Å 5 1 107 M01 s01 (19). Hence, the steady-state concentration of intramitochondrial H2O2 according to the equation below is 7.71 1 1007 M (average of three individual experiments).
05
01
4.51 1 10 M s 5 1 10 M01 s01 [1.17 1 1006 7
Å 7.71 { 0.26 1 1007
M]
.
M
The rate of H2O2 formation by the rat brain respiratory chain during the oxidation of succinate in the presence of antimycin A was 57 pmol/min/mg protein, corresponding to 9.5 1 1007 M s01. This value is in the range of that previously reported (180 pmol/min/mg protein) (24). Higher rates were reported when the assay medium was potassium phosphate in the absence of man-
AID
ARCH 9698
/
/d[H2O2]/dt (M s01 1 1006)
[H2O2]ss (M (1007))
3.33 5.50 10.21 0.95
0.57 0.94 1.75c 0.16
45.2 26.7 16.7 6.7
7.71 — — —
Inner mitochondrial membrane Electron transfer chain Rat liver a Rat lung a Rat heartb Rat brain Outer mitochondrial membrane Monoamine oxidase activity Tyramined Tyraminee Phenylethylaminee Benzylaminee a
Values taken from Ref. (32). Values taken from Ref. (31). c This value considers the high catalase activity in the heart intramitochondrial matrix (concentration of catalase Å 0.72 1 1006 M; k Å 4.6 1 107 M s01 (Ref. 2). d Value obtained with 2 mM tyramine. e Values representing the initial rate and therefore not utilized to calculate the [H2O2]ss . Assay conditions for brain inner mitochondrial membrane activities were 0.17 mg/ml mitochondrial protein in buffer, pH 7.4, supplemented with succinate (7 mM) and antimycin A (4.7 nmol/mg protein). Assay for outer mitochondrial membrane activity was as described in the legend to Fig. 1. b
nitol (223 pmol/min/mg protein) (25). Therefore, the [H2O2]ss built in the brain mitochondrial matrix under the above conditions was: [H2O2]ss Å Å
/d[H2O2]/dt k[GPx] 9.5 1 1007 M s01 5 1 107 M01 s01 [1.17 1 1006
Å 1.62 { 0.26 1 10
/d[H2O2]/dt [H2O2]ss Å k[GPx] Å
TABLE II
Production of H2O2 by Rat Brain Mitochondria and the Calculated Matrical [H2O2]ss during the Electron Transfer in the Presence of Antimycin A and during the MAO-Catalyzed Oxidation of Amines
6b23$$$462
10-28-96 07:52:16
08
M]
M.
This value is close to the ‘cellular’ [H2O2]ss reported by Chance et al. (26) and in agreement with the intramitochondrial value recently reported by Boveris and Cadenas (19) (Table II). In conclusion, the [H2O2]ss obtained during the metabolism of biogenic amines by the outer membrane MAO activity in brain mitochondria is about 48-fold higher than that originating during the oxidation of substrates via complex II of the electron transfer chain in the presence of antimycin A. Oxidative Impairment of mtDNA The [H2O2]ss built by the metabolism of tyramine by MAO-A/B is expected to contribute to oxidative damage
arca
AP: Archives
METABOLISM OF TYRAMINE BY MONOAMINE OXIDASE A/B
299
FIG. 3. Induction of single strand breaks during the metabolism of tyramine by rat brain mitochondria. General assay conditions: intact rat brain mitochondria (0.3 mg protein/ml) in the buffer described in the legend to Fig. 2 were incubated with 2 mM tyramine at 377C under gentle stirring for 60 min in the absence and presence of different compounds. Electrophoretic distribution of form II (nicked) and form I (supercoiled) mtDNA: (A) Ethidium bromide staining. Left lane, molecular size markers (lDNA/Hind III). Lane 1, mtDNA plus tyramine; lane 2, mtDNA. (B) Southern blotting hybridization. Lane 1, mtDNA plus tyramine; lane 2, mtDNA; lane 3, mtDNA plus tyramine plus catalase (400 Urml01); lane 4, mtDNA plus tyramine plus tranylcypromine. (C) Time-dependent formation of mtDNA strand breaks. Under the general assay conditions described above, mitochondria were incubated with tyramine for 15 (n Å 4), 30 (n Å 8), and 60 (n Å 3) min. Inset shows the linear dependence on strand break formation on time.
of intramitochondrial components and, likewise, to contribute substantially to the cytosolic [H2O2]ss . Intact, coupled rat brain mitochondria incubated with tyramine produced H2O2 at a rate of 2.71 nmol/min/mg protein. This elicited a significant induction of intramitochondrial DNA single strand breaks. The intramitochondrial H2O2 may undergo a site-specific homolytic scission by copper pools other than cytochrome oxidase (27) or by mtDNA-bound transition metals. This, along with the restricted diffusibility (about 6–10 molecular diameters) and the high reactivity of HO• provides the conditions for an oxidative attack on bases (preferentially guanosine) and sugar residues in mtDNA. The former implies an addition reaction usually leading to the formation of 8-hydroxydeoxyguanosine; the latter mainly entails a C4 mechanism involving H abstraction
AID
ARCH 9698
/
6b23$$$462
10-28-96 07:52:16
from the sugar moiety eliciting single strand break formation (28–30). Single strand breakage of form I (supercoiled or close circular) mtDNA leads to form II (nicked or open circular) mtDNA, represented in the lower and upper bands, respectively, of the ethidium bromide staining (Fig. 3A) and Southern blotting hybridization (Fig. 3B). Control or untreated mitochondria show a single strand break background value of 0.75/104 bp (Fig. 3B, lane 2), which increases to 3.12 { 0.69/104 bp upon incubation with tyramine for 60 min (Fig. 3B, lane 1). The mitochondrial membrane is impermeable to catalase, but the hemoprotein affects the intramitochondrial [H2O2]ss upon displacement of the equilibrium; hence, a substantial decrease (Ç60%) of single strand break formation was obtained in the presence of catalase (1.27/
arca
AP: Archives
300
HAUPTMANN ET AL. TABLE III
[H2O2]ss in Control and CDNB-Treated Rat Brain Mitochondria [H2O2]ss (M 1 1007)
[GSH] (nmol/mg protein)
Inner mitochondrial membrane Electron transfer chain Outer mitochondrial membrane MAO-A/B activity
Control
/CDNB
Control
/CDNB
7.95 { 0.42
2.19 { 0.37
0.16 { 0.02
0.22 { 0.03
7.95 { 0.42
2.19 { 0.37
7.71 { 0.25
9.91 { 0.97
Note. Assay conditions as described under Materials and Methods.
104 bp; Fig. 3B, lane 3). Tranylcypromine, an effective irreversible inhibitor of MAO, abolished strand break formation (i.e., the values in the presence of the inhibitor are those obtained under control (background level) conditions (0.75/104 bp; Fig. 3B, lane 4). Figure 3C shows a time-dependent increase of single strand breakage of mtDNA during incubation with tyramine (linearly related with time; Fig. 3C, inset) obtained from data from Southern blot hybridization. Each value represents the ratio of tyramine treated sample/control of the relative increase of form II (nicked) mtDNA, according to the formula described under Materials and Methods. The time-dependent increase of single strand breaks was as follows: 1.5 { 0.29, 2.12 { 0.28 (P £ 0.05), and 3.12 { 0.69 (P £ 0.05)-fold at 15-, 30-, and 60 min incubation, respectively (Fig. 3C). Effect of the Intramitochondrial GSH on the [H2O2]ss and the Oxidative Damage to mtDNA Incubation of intact, coupled rat brain mitochondria with 1-chloro-2,4-dinitrobenzene for 8 min elicited 72% depletion of the intramitochondrial GSH content (7.95 nmol/mg protein in control mitochondria and 2.38 nmol/mg protein in CDNB-treated mitochondria). This appears to be the maximal GSH depletion level to be reached in rat brain mitochondria: exposure of brain mitochondria to longer incubation periods or higher CDNB concentrations did not change significantly this percentage, but altered the mitochondrial functions. Depletion of GSH requires a GSH transferase-catalyzed conjugation between the thiol and CDNB. This treatment as described above did not affect the P/O and respiratory control ratios of mitochondria when tested with glutamate or succinate as electron donors. Table III lists the GSH content in different mitochondrial preparations and the corresponding [H2O2]ss . A slight increase (1.38-fold, P õ 0.05) of [H2O2]ss in the CDNB-treated mitochondria was observed during the metabolism of succinate in the presence of antimycin A. Likewise, a 1.28-fold increase in the [H2O2]ss associated
AID
ARCH 9698
/
6b23$$$463
10-28-96 07:52:16
with the metabolism of tyramine in CDNB-treated mitochondria was calculated. A 68% lower than controls single strand break formation was observed in GSH-depleted rat brain mitochondria during the oxidation of tyramine with MAO A/B for 60 min (not shown). Regardless of the GSH content of brain mitochondria, it is to be noted that the [H2O2]ss built up during the MAO-catalyzed metabolism of tyramine is about 48-fold higher (average from the data shown in Table III) than that built during the oxidation of succinate in the presence of antimycin A. It is also expected that the H2O2 formed during the outer membrane MAO reaction would contribute to the cytosolic [H2O2]ss . Electron Spin Resonance Studies ESR in conjunction with the spin trap 5,5-dimethyl1-pyrroline-N-oxide revealed after supplementation of intact brain mitochondria with tyramine a spectrum consisting of a composite of DMPO-OH (aN Å 14.9 G; aH Å 14.9 G) and DMPO-CH3 (aN Å 15.8 G; aH Å 22.5 G) adducts in a ratio 2:1 (Fig. 4). The latter, a carboncentered radical, is expected to originate from the reaction of HO• with mannitol (k Å 2.7 1 109 M01 s01) (31), present in a concentration of 250 mM in the buffer. These signals were substantially amplified in a timedependent manner by the addition of FeSO4 suggesting the occurrence of a Fenton-driven mechanism. The formation of HO• in this experimental model appears irrelevant to the mechanism of mtDNA oxidative damage on the following basis: (a) HO• is a strong electrophile which diffuses 10 molecular diameters from its site of formation before reacting with the target; consequently, a contribution of HO• to mtDNA oxidative damage is highly unlikely. (b) The outer mitochondrial membrane itself is expected to be the primary target for HO• (5, 6). (c) Conversely, the HO• formed upon decomposition of MAO-generated H2O2 in the bulk solution may elicit oxidative damage to extramitochondrial targets within
arca
AP: Archives
METABOLISM OF TYRAMINE BY MONOAMINE OXIDASE A/B
FIG. 4. HO• and carbon-centered radical formation during the oxidative deamination of tyramine by mitochondrial monoamine oxidase. Assay conditions: intact rat brain mitochondria (0.27 mg/ml) in the buffer described in the legend to Fig. 2 were supplemened with 0.2 M DMPO. (A) No substrate added, (B) plus 0.2 mM tyramine, (C) plus 0.2 mM tyramine and 60 mM FeSO4 . (D) As in C after 10 min incubation. (E) As in C after 20 min incubation. (F) As in C in the presence of catalase (400 Urml01).
the diffusion distance of this species. Evidently, the metabolism of tyramine by outer membrane-located MAO contributes to the cytosolic [H2O2]ss because the signal was abolished by catalase (Fig. 4). Likewise, a freely diffusion of H2O2 into the intramitochondrial matrix appears to occur and may contribute to the tyramine-dependent, tranylcypromine-sensitive oxidative damage of mtDNA described above (see Fig. 3B and Table II). DISCUSSION
At least two sources may be considered to contribute to the intramitochondrial [H2O2]ss : on the one hand, the mitochondrial respiratory chain is a significant source of H2O2 and O2r0, whose formation accounts for about 4% of the total O2 consumed in mitochondrial preparations (26). Mammalian tissues produce H2O2 at rates of 0.08–0.1 nmol/min/mg protein in the presence of succinate and this values increase to about 0.2–1.0
AID
ARCH 9698
/
6b23$$$463
10-28-96 07:52:16
301
nmol/min/mg protein upon supplementation with antimycin A (32, 33). As shown above, intact rat brain mitochondria produce 0.056 nmol/min/mg protein during the oxidation of succinate (state 4) and in the presence of antimycin A; this value is 3.6-fold lower than that obtained with rat liver mitochondria. The intramitochondrial [H2O2]ss calculated for brain mitochondria under these conditions was 1.62 { 0.26 1 1008 M. On the other hand, H2O2 stemming from the oxidation of biogenic amines by the outer membrane MAO-A/B appears also to contribute partly to an intramitochondrial [H2O2]ss . This view is substantiated by (a) a previous report associating MAO-mediated production of H2O2 with the formation of intramitochondrial GSSG, hardly affected by catalase, but abolished by MAO inhibitors (4); (b) the tyramine-dependent, tranylcypromine-inhibitable mtDNA single strand break formation (Fig. 3B); (c) the inhibition of ESR signals (Fig. 4) by catalase, suggesting that H2O2 is formed outside the mitochondria and it diffuses across the inner mitochondrial membrane into the matrix. The intramitochondrial [H2O2]ss calculated during the oxidation of tyramine by the outer membrane MAO was 7.71 1 1007 M. In conclusion, the intramitochondrial [H2O2]ss obtained during the metabolism of tyramine by the outer membrane MAO-A/B is about 48-fold higher than that originating by the complex II-mediated oxidation of succinate in the presence of antimycin A. The rates of H2O2 formation originating from the mitochondrial electron-transfer chain and MAO activity are increased after partial depletion of GSH by CDNB. This is in disagreement with previous reports, in which CDNB treatment hardly affected the H2O2 formation by complex II of the respiratory chain (34). Furthermore, glutathione deficiency obtained by treatment of newborn rats with L-buthionine (S, R)-sulfoximine leads to mitochondrial damage in the brain visualized as a destroyed internal structure of mitochondria (35). The time-dependent formation of single strand breaks during the metabolism of tyramine by MAO-A/ B in vitro can be considered as a direct evidence of oxidative damage to mtDNA, due to the diffusion of H2O2 across the inner mitochondrial matrix. HO• presumably formed in a Fenton-like reaction catalyzed by DNA-bound transitions metals (e.g., Cu) attacks mainly the thymine and guanosine residues (k Å 109 – 1010 M01 s01), forming a primary reducing or oxidizing radical. The radical character is transferred to the sugar moiety resulting in single strand break formation (30). Alternatively, a direct H-atom abstraction of the sugar moiety by HO• leads to (36) single strand breaks. It may be surmised that the effect of HO• could be restricted by the presence of mannitol, an effective scavenger of HO•, in the incubation buffer. Replacement of mannitol by Tris buffer showed only a slight elevation of single strand breaks because also this com-
arca
AP: Archives
302
HAUPTMANN ET AL.
FIG. 5. Intramitochondrial [H2O2]ss , mtDNA single strand break, and a possible role for GSH during oxidative deamination of tyramine by MAO brain mitochondria.
pound is known to act as radical scavenger (results not shown). Moreover, the mechanism proposed entails a site-specific formation of HO• (not formation in the bulk solution) and, hence, the interference of mannitol or other HO• quenchers is debatable. In vivo single strand breakage may lead to mutagenicity resulting from single nucleotide substitutions during base repair. However, also certain repair endonucleases, which specifically recognize free radical-induced DNA base modifications, incise the DNA at the site of the modification, leading to single strand breaks (37). The presence of these enzymes in brain mitochondria remains to be proven. The accumulation of muta-
AID
ARCH 9698
/
6b23$$$463
10-28-96 07:52:16
tions of mtDNA in postmitotic cells is favored by the continuous turnover of mtDNA independent of cell replication and a yet unknown mechanism accelerating the division of those mitochondria with decreased or defective ATP synthesis (11). It has been reported that incubation of intact rat liver mitochondria with succinate/antimycin A leads to an increase of the formation of 8-hydroxydeoxyguanosine, another widely used approach to assess DNA oxidative damage (38). It is also established that the rate of O2r0 and H2O2 production of the respiratory chain increases with age and this correlates with higher levels of 8-hydroxydeoxyguanosine in mtDNA (39, 40).
arca
AP: Archives
METABOLISM OF TYRAMINE BY MONOAMINE OXIDASE A/B
Formation of 8-hydroxydeoxyguanosine causes exclusive misreading during DNA synthesis in vitro (G r T transversion) (41). Preliminary data obtained in our laboratory suggest also an increase of 8-hydroxydeoxyguanosine formation during the oxidative deamination of tyramine by brain mitochondria MAO-A/B. However, no unambiguous results could be obtained with this approach, probably due to the sensitivity of the assay, which, however, proved successful with rat liver mitochondria. Recently, a direct correlation was established between accumulation of 8-hydroxydeoxyguanosine and single strand break formation when plasmid DNA was exposed to oxidative conditions (42). Despite the extensively described antioxidant effect of cellular thiols, the depletion of 72% of the intramitochondrial GSH led to a substantially diminished single strand break formation (68% less compared to the control) during tyramine-mediated formation of H2O2 . These data are supported by previous reports, which assessed single strand breaks formation by 1O2 (43), presumably because thiols keep the DNA-bound transition metals in a reduced state, facilitating the cleavage of H2O2 in a Fenton site-specific manner. This is expected to occur even at lower [H2O2]ss ; furthermore, thiols may be autoxidized in the presence of DNAbound CuII, forming oxygen free radicals as well as thiyl radicals (43–45). In summary, intramitochondrial GSH appears to have prooxidant functions at least in terms of protection against mtDNA oxidative damage. The exact mechanism by which a decrease of intramitochondrial GSH is associated with a decrease (and not an increase) of single strand break formation remains to be elucidated, but it is tempting to suggest that the thiol is involved in the reduction of transition metals essential for mtDNA strand breakage. These views are summarized in Fig. 5. The results presented here offer the possibility of an oxidative damage of mtDNA in vivo during the MAOA/B-catalyzed metabolism of biogenic amines with the consequential impairment of oxidative phosphorylation and, hence, the energy level of the cell. Neurodegenerative diseases, such as Parkinson’s disease, are associated with an increase of MAO-B and a selective less active complex I of the respiratory chain. An impaired complex I activity seems not accompanied by deletions of mtDNA (46); hence, it might be likely related to a direct HO• attack stemming from site specific cleavage of H2O2 by, for example, redox active copper pools other than those in cytochrome oxidase. Alternately, it has been proved that dopamine by itself may inhibit mitochondrial NADH-dehydrogenase activity (47), a fact which might raise importance if the MAO-A/B activity is impaired. It has been shown that monoamines and their metabolites (e.g., dihydroxyphenylacetic acid, dopamine, homovanillic acid) may act as radical scavengers (48). Following this, the significance of our results
AID
ARCH 9698
/
6b23$$$463
10-28-96 07:52:16
303
in vivo must be considered under the possible impact of an equilibrium encompassing various factors, such as the age-related increase of MAO activity, the H2O2 production, the intramitochondrial GSH status, the radical quenching activity of MAO products and substrates, and the possible age-related decrease of the synthesis of certain neurotransmitters (e.g., dopamine) (46). An impairment of this sensitive equilibrium results in an accumulation of somatic mtDNA mutations, which along with the progress of germinal (inherited) mutations may lead above the 90% threshold of mutated mtDNA (11), a value required to make apparent dysfunctional oxidative phosphorylation and to observe clinical symptoms. Furthermore, H2O2 generated during the outer membrane monoamine oxidase activity-catalyzed oxidation of catecholamines may have important implications for the ischemia–reperfusion syndrome. This may bear relevance in connection with the increased levels of catecholamines related to myocardial ischemia/reperfusion injury (49, 50) and the proposed role of oxygen radicals arising from enzyme-catalyzed catecholamine oxidations (51), rather than autooxidation. ACKNOWLEDGMENT Work was supported by Grant HL53467 from NIH. Note added in proof. Detection of H2O2 during the metabolism of tyramine by the method based on the HRP/H2O2-coupled dimerization of p-hydroxyphenylacetic acid may lead to an overestimation due to the simultaneous contribution to fluorescence by the HRP/ H2O2-mediated oxidation and subsequent dimerization of tyramine. However, measurement of H2O2 formation by the HRP/scopoletin method – free of the above interference – yielded the same values reported here.
REFERENCES 1. Cohen, G. (1983) J. Neural Transmembr. Suppl. 19, 89–103. 2. Sandri, G., Panfili, E., and Ernster, L. (1990). Biochim. Biophys. Acta 1035, 300–305. 3. Werner, P., and Cohen, G. (1991). FEBS Lett. 280, 44–46. 4. Savov, V. M., Prilipko, L. L., and Gorkin, V. Z. (1983). Acta Physiol. Pharmacol. Bulg. 9, 3–13. 5. Kagan, V. E., Smirnov, A. V., Savov, V. M., and Gorkin, V. Z. (1984). Vopr. Med. Khim. 30, 112–118. 6. Hruszkewycz, A. M. (1988). Biochem. Biophys. Res. Commun. 153, 191–1977. 7. Seregi, A., Serfozeo, P., and Mergl, Z. (1983). J. Neurochem. 40, 407–413. 8. Konradi, C., Riederer, P., and Youdim, M. B. (1986). J. Neural. Transmembr. Suppl. 22, 61–73. 9. Wallace, D. C. (1992). Annu. Rev. Biochem. 61, 1175–1212. 10. Saura, J., Richards, J. G., and Mahy, N. (1994). Neurobiol. Aging 15, 399–408. 11. Kadenbach, B., Mu¨nscher, C., Frank, V., Mu¨ller-Ho¨cker, J., and Napiwotzki, J. (1995). Mutat. Res. 338, 161–172. 12. Richter, C. (1995). Int. J. Biochem. Cell Biol. 27, 647–653. 13. Basford, R. E. (1967). Methods Enzymol. 10, 97–101.
arca
AP: Archives
304
HAUPTMANN ET AL.
14. Sciamanna, M. A., and Lee, C. P. (1993). Arch. Biochem. Biophys. 305, 215–224. 15. Jocelyn, P. C., and Cronshaw, A. (1985). Biochem. Pharmacol. 34, 1588–1590. 16. Martin, J., and White, I. N. H. (1991). J. Chromatogr. 568, 219– 224. 17. Shigenaga, M. K., Aboujaoude, E. N., Chen, Q., and Ames, B. N. (1994). Methods Enzymol. 234, 16–33. 18. Houlihan, P., and Gershon, P. (1994). in Monoamine Oxidase Inhibitors in Neurological Diseases (Lieberman, A., Olanow, C. W., Youdim, M. B. H., and Tipton, K., Eds.), pp. 290–309, Dekker, New York. 19. Boveris, A., and Cadenas, E. (1996). in Oxygen, Gene Expression, and Cellular Function (Clerch, L. B., and Massaro, D., Eds.), Dekker, New York, in press. 20. Tyler, D. D. (1975). Biochem. J. 147, 493–504. 21. Radi, R., Turrens, J. F., Chang, L. Y., Bush, K. M., Crapo, J. D., and Freeman, B. A. (1991). J. Biol. Chem. 266, 22028–22034. 22. Nohl, H., and Hegener, D. (1978). FEBS Lett. 89, 126–130. 23. Sinet, P. M., Heikkila, R. E., and Cohen, G. (1980). J. Neurochem. 34, 1421–1428. 24. Adamo, A. M., Llesuy, S. F., Pasquini, J. M., and Boveris, A. (1989). Biochem. J. 263, 273–277. 25. Patole, M. S., Swaroop, A., and Ramasarma, T. (1986). J. Neurochem. 47, 1–8. 26. Chance, B., Sies, H., and Boveris, A. (1979). Physiol. Rev. 59, 527–605. 27. Massa, E. M., and Giulivi, C. (1993). Free Radicals Biol. Med. 14, 559–565. 28. Steenken, S. (1989). Chem. Rev. 89, 503–520. 29. von Sonntag, C. (1987). The Chemical Basis of Radiation Biology, pp. 116–166, Taylor & Francis, London. 30. Symons, M. C. R. (1987). J. Chem. Soc. Faraday Trans. I 83, 1– 11. 31. Halliwell, B. (1978). FEBS Lett. 92, 321–326.
AID
ARCH 9698
/
6b23$$$463
10-28-96 07:52:16
32. Cadenas, E., and Boveris, A. (1980). Biochem. J. 188, 31–37. 33. Forman, H. J., and Boveris, A. (1982). in Free Radicals in Biology (Pryor, W. A., Ed.), Vol. 5, pp. 65–90, Academic Press, New York. 34. Zoccarato, F., Cavallini, L., Deana, R., and Alexandre, A. (1988). Biochem. Biophys. Res. Commun. 154, 727–734. 35. Jain, A., Ma˚rtensson, J., Stole, E., Auld, P. A. M., and Meister, A. (1991). Proc. Natl. Acad. Sci. USA 88, 1913–1917. 36. Meunier, B., Pratviel, G., and Bernadou, J. (1994). Bull. Soc. Chim. Fr. 131, 933–943. 37. Epe, B., and Hegler, J. (1994). Methods Enzymol. 234, 122–131. 38. Giulivi, C., Boveris, A., and Cadenas, E. (1995). Arch. Biochem. Biophys. 316, 909–916. 39. Sohal, R. S., and Brunk, U. T. (1992). Mutat. Res. 275, 295–304. 40. Ames, B. N., Shigenaga, M. K., and Hagen, T. M. (1993). Proc. Natl. Acad. Sci. USA 90, 7915–7922. 41. Kasai, H., and Nishimura, S. (1991). in Oxidative Stress: Oxidants and Antioxidants (Sies, H., Ed.), pp. 99–116, Academic Press, San Diego. 42. Toyokuni, S., and Sagripanti, J.-L. (1996). Free Radicals Biol. Med. 20, 859–864. 43. Devasagayam, T. P. A., DiMascio, P., Kaiser, S., and Sies, H. (1991). Biochim. Biophys. Acta 1088, 409–412. 44. Reed, C. J., and Douglas, K. T. (1989). Biochem. Biophys. Res. Commun. 162, 1111–1117. 45. Misra, H. P. (1974). J. Biol. Chem. 249, 2151–2155. 46. Fahn, S., and Cohen, G. (1992). Ann. Neurol. 32, 804–812. 47. Ben-Shachar, D., Zuk, R., and Glinka, Y. (1995). J. Neurochem. 64, 718–723. 48. Liu, J., and Mori, A. (1993) Arch. Biochem. Biophys. 302, 118– 127. 49. Singal, P. K., Kapus, N., Dhillon, K. S., Beamish, R. E., and Dhalla, N. S. (1982). Can. J. Physiol. Pharmacol. 60, 1390–1397. 50. Flaherty, J. T., and Weisfeldt, M. L. (1988). Free Radicals Biol. Med. 5, 409–419. 51. Jewet, S. L., Eddy, L. J., and Hochstein, P. (1989). Free Radicals Biol. Med. 6, 185–188.
arca
AP: Archives
Vol. 335, No. 2, November 15, pp. 295–304, 1996 Article No. 0510
The Metabolism of Tyramine by Monoamine Oxidase A/B Causes Oxidative Damage to Mitochondrial DNA Nils Hauptmann, Joseph Grimsby, Jean C. Shih, and Enrique Cadenas1 Department of Molecular Pharmacology and Toxicology, School of Pharmacy, University of Southern California, Los Angeles, California 90033
Received June 11, 1996, and in revised form August 23, 1996
Monoamine oxidases A/B (EC 1.4.3.4, MAO), flavoenzymes located on the outer mitochondrial membrane, catalyze the oxidative deamination of biogenic amines, such as dopamine, serotonin, and norepinephrine. In this study, we examined whether the H2O2 formed during the two-electron oxidation of tyramine [4-(2-aminoethyl)phenol] (a substrate for monoamine oxidases A/B) may contribute to the intramitochondrial steady-state concentration of H2O2 ([H2O2]ss) and, thus, be involved in the oxidative impairment of mitochondrial matrix components. Supplementation of intact, coupled rat brain mitochondria with benzylamine, b-phenylethylamine, or tyramine showed initial rates of H2O2 production ranging from 0.4- to 1.6 nmol H2O2/min/mg protein. ESR analysis of the oxidative deamination of tyramine by intact rat brain mitochondria revealed the formation of hydroxyl (HO•) and carbon-centered radical adducts — the latter probably originating by the HO•mediated oxidation of mannitol. The signals were substantially enhanced upon addition of FeSO4 and were abolished by catalase. The intramitochondrial [H2O2]ss calculated in terms of glutathione peroxidase activity during the metabolism of tyramine was 48-fold higher (7.71 { 0.25 1 1007 M) than that obtained during the oxidation of succinate via complex II in the presence of antimycin A (1.64 { 0.2 1 1008 M). Oxidative damage to the brain mtDNA was assessed by single strand breakage. The ratio of nicked DNA for the preparations treated with tyramine and those without the amine was 1.5 { 0.29 (n Å 4), 2.12 { 0.28 (n Å 8, P £ 0.05), and 3.12 { 0.69 (n Å 3, P £ 0.05) at 15, 30, and 60 min, respectively. Preincubation of mitochondria with tranylcypromine (trans-2-phenylcyclopropylamine), an inhibitor to MAO A/B, abolished mtDNA oxidative damage. Catalase inhibited mtDNA strand breakage by approximately 60%. Incu-
1
To whom correspondence should be addressed.
bation of intact, coupled rat brain mitochondria with chlorodinitrobenzene (CDNB) depleted mitochondrial GSH by 72%. Tyramine-dependent damage of mtDNA was decreased by 68% in CDNB-treated mitochondria (with 28% remaining GSH). The [H2O2]ss was slightly increased in CDNB-treated mitochondria: 1.38- and 1.28-fold increase during the oxidation of succinate in the presence of antimycin A and during the oxidation of tyramine, respectively. These results suggest that the H2O2 generated during the MAO-catalyzed oxidation of biogenic amines and possibly certain neurotransmitters at the outer mitochondrial membrane contributes to the intramitochondrial [H2O2]ss and may cause oxidative damage to mtDNA. This is effected by the intramitochondrial concentration of GSH and might have potential implications for aging and neurodegenerative processes. q 1996 Academic Press, Inc. Key Words: monoamine oxidase; mitochondrial DNA; glutathione; neurodegenerative diseases; hydrogen peroxide.
Monoamine oxidases (EC 1.4.3.4, MAO-A/B),2 enzymes present in the outer mitochondrial membrane, catalyze the oxidation of biogenic amines accompanied by release of H2O2 (reaction [1]).
[1] 2 Abbreviations used: MAO, monoamine oxidase; GSH, glutathione; GSSG, glutathione disulfide; mtDNA, mitochondrial DNA; ESR, electron spin resonance spectroscopy; DMPO, dimethyl-1-pyrrolineN-oxide; CDNB, 1-chloro-2,4-dinitrobenzene.
295
0003-9861/96 $18.00 Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
AID
ARCH 9698
/
6b23$$$461
10-28-96 07:52:16
arca
AP: Archives
296
HAUPTMANN ET AL.
H2O2 produced during the oxidative deamination of catecholamines (e.g., dopamine) appears to be involved in the progress of neurodegenerative disorders, such as Parkinson disease and, presumably, via oxidative damage to the mitochondrial membrane (1). Several studies were undertaken to assess the extent of impairment of brain mitochondrial functions by this species and its relationship to neurodegenerative disorders. H2O2 produced during the metabolism of benzylamine, tyramine, and phenylethylamine by monoamine oxidases decreases significantly the intramitochondrial GSH content and increases the Ca2/ efflux (2). Likewise, intramitochondrial GSSG formation increased following the benzylamine metabolism by MAO-B; this effect was hardly affected upon supplementation with catalase. This suggests that the H2O2 produced at the outer mitochondrial membrane diffuses partly into the mitochondrial matrix and contributes to the oxidation of GSH (3). However, in the presence of iron, H2O2 formed by the metabolism of benzylamine by MAO-B leads to severe membrane lipid peroxidation, suggesting the homolytic cleavage of H2O2 by Fe2/ within a Fenton reaction (4, 5). It was proposed that lipid peroxidation products can mediate mtDNA damage (6). In addition to these effects, it has been suggested that H2O2 might have a stimulatory effect on cyclooxygenase (7) and MAO-B (8). Over the past years, evidence has accumulated that several human disorders, such as blindness, deafness, dementia, diabetes, and neurodegenerative diseases (e.g., Alzheimer, Parkinson, and Huntington diseases) are accompanied by mtDNA mutations leading mainly to dysfunctional mitochondria with an impaired oxidative phosphorylation activity (9). Furthermore, there is an age-related increase of MAO-B, which is associated with an augmentation of mtDNA damage and a decline in oxidative phosphorylation (10). Mitochondria possess a far less efficient DNA repair system than nuclei and mtDNA is not protected by histones (11, 12). mtDNA is involved in the coding of 13 essential genes and, expectedly, oxidative damage to mtDNA caused by an increased intramitochondrial H2O2 steady-state concentration may result in devastating effects for the mitochondrial functions. In this study, we examined whether the metabolism of tyramine ([4-(2-aminoethyl)phenol]) by MAO-A/B influences the intramitochondrial [H2O2]ss and whether this is associated with an oxidative impairment of mtDNA. We have also assessed the role of intramitochondrial GSH content on the [H2O2]ss and mtDNA single strand breakage.
sulfonic acid (Hepes), bovine serum albumin, 1-chloro-2,4-dinitrobenzene (CDNB), dansyl chloride, g-glu-glu (internal standard for GSH measurements), Nagarse, 4-[2-aminoethyl]phenol (tyramine), trans2-phenylcyclopropylamine (tranylcypromine), benzylamine, and phenylethylamine were purchased from Sigma (St. Louis, MO). GSH, horseradish peroxidase, glucose oxidase, proteinase K, ethidium bromide, and RNase A/T were supplied from Boehringer (Mannheim, Germany). Potassium phosphate, m-cresol purple sodium salt, lithium hydroxide, iodoacetic acid, Tris(hydroxymethyl)aminomethane hydrochloride, acetic acid, and sodium acetate were purchased from Fluka (Buchs, Switzerland). Other reagents were ordered as follows: sucrose and sodium dodecyl sulfate from GibcoBRL (Grand Island, NY); 5,5-dimethyl-1-pyrroline-N-oxide (DMPO), H2O2 , and p-hydroxyphenylacetic acid from Aldrich Chemical Co. (Milwaukee, WI); magnesium chloride and ferrous sulfate from Baker (Phillipsburg, NY); HPC-grade methanol from Baxter (Irvine, CA); agarose from FMC Bioproducts (Rockland, ME); [32P]ATP and [32P]CTP from ICN (Irvine, CA); and BglII from New England Biolabs (Mississauga, Canada).
MATERIALS AND METHODS
Single Strand Break Assay
Chemicals and Biochemicals Mannitol, glucose, ethylenediaminetetraacetic acid (EDTA), succinic acid, glutamic acid, N-2-hydroxyethylpiperazine-N*-2-ethane-
AID
ARCH 9698
/
6b23$$$462
10-28-96 07:52:16
Isolation of Mitochondria and General Assay Conditions Brain mitochondria from male Sprague–Dawley rats (300–350 g) were isolated and the P/O ratio was determined as described in established procedures (13, 14). The assay reaction buffer consisted of 250 mM mannitol, 70 mM sucrose, 1 mM EDTA, 10 mM potassium phosphate, pH 7.4. Protein concentration was determined with the bicinchoninic acid (BCA) protein assay (Pierce Co., Rockford, IL)
Spectrophotometric Measurements Fluorometric. The formation of H2O2 was measured fluorometrically using the horseradish peroxidase/H2O2-coupled dimerization of p-hydroxyphenylacetic acid (lexcitation 315 nm; lemission 410 nm) in an Aminco–Bowman spectrofluorometer (American Instrument Co., Silver Spring, MD) under gentle stirring. The H2O2-generating system (glucose/glucose oxidase) was used as standard for the detection of the continuous extramitochondrial H2O2 release. Electron spin resonance. ESR spectra were recorded on a Bruker ECS 106 equipped with TM 8810 microwave cavity. Measurements were carried out at a microwave frequency of 9.81 GHz and 100 KHz field modulation. DMPO was used as spin trap after extensive purification and produced no ESR signal by itself.
Depletion of Mitochondrial GSH GSH depletion was carried out by the glutathione transferasecatalyzed conjugation of the thiol with CDNB (15). Briefly, mitochondria (0.6 mg protein/ml) were incubated for 6 min with 60 mM CDNB dissolved in ethanol (final concentration 0.6%) and centrifuged for 2 min at 10,500 rpm. GSH content was evaluated by HPLC with fluorometrical detection as previously reported (16), using a 250 1 4.6-mm Spheri-5-amino column (Applied Biosystems Co., Forster City, CA). The latter was connected to a two-pump delivery system (Model LC-600), an autoinjector (Model SIL-9A), and a fluorometric detector (Model RF-535) from Shimadzu Scientific Instruments Inc. (Columbia, MD). For experiments requiring a further use of GSHdepleted mitochondria, the pellet was washed once with reaction buffer and the P/O ratio was checked again using a Clark-type oxygen electrode (Yellow Springs Instrument Co., Yellow Springs, OH).
Brain mtDNA, isolated as previously reported (17), was digested with BglII and a 1.9-kb fragment was cloned into pBSK to obtain a probe for the single strand break assay. The clone was verified by
arca
AP: Archives
297
METABOLISM OF TYRAMINE BY MONOAMINE OXIDASE A/B TABLE I
Rates of H2O2 Formation during the MAO A/B-Catalyzed Oxidative Deamination of Different Amines
Tyramine Phenylethylamine Benzylamine
MAO isoform
v (nmol H2O2 /min/mg protein)
A/B B/ B
1.6 1.0 0.4
Note. Values represent initial velocities and assay conditions were as described under Materials and Methods.
FIG. 1. Formation of H2O2 during the oxidation of various MAO substrates. Intact rat brain mitochondria (P/O 2.8 and 1.7 for glutamate and succinate, respectively) in the reaction buffer described under Materials and Methods were incubated with 0.2 mM substrate for 10 min and H2O2 was measured. (l) tyramine; (j) phenylethylamine; (m) benzylamine. Other assay conditions were as described under Materials and Methods.
restriction analysis. Following the treatment of mitochondria according to the experimental designs described under Results, mtDNA was isolated, electrophoresed in an agarose gel, and transferred (Southern blotting) to a Bio-Rad GT membrane (Bio-Rad, Hercules, CA). The transfer was hybridized to the 1.9-kb BglII radioactive labeled probe overnight, and washed in 21 SSC/0.1% SDS. Analysis was carried out with a Packard Instant Imager (Packard Instruments Co., Meriden, CT) and the occurrence of single strand breaks was expressed by the formula shown below as the number of single strand breaks per 104 base pairs. The statistically significance between tyramine-treated and control mitochondria was assessed by a paired t test. single strand break Å 0ln
MAO-A/B and, hence, its higher rate of metabolism and H2O2 formation may be accounted by the effective oxidation of this substrate by both isoforms of MAO. The second one is a preferred substrate for MAO-B, whereas the third is exclusively metabolized by MAO-B (Table I). A KM of 1.54 mM for tyramine was estimated from a double reciprocal plot (Fig. 2, inset). H2O2 accumulation could not be detected during the metabolism of dopamine due to an interference with the detection system; however, O2 uptake experiments (ascribed to O2 reduction to H2O2) revealed rates 2.7-fold lower than those obtained with tyramine. Tranylcypromine (trans-2-phenylcyclopropylamine) is a nonselective, nonhydrazine, irreversible inhibitor of MAO-A and B (18); preincubation of intact, coupled brain mitochondria caused a 96% inhibition of tyramine metabolism and, hence, of H2O2 formation.
supercoiled DNA . supercoiled DNA / nicked DNA
RESULTS
Monoamine Oxidase-Catalyzed Formation of Hydrogen Peroxide The rate of formation of H2O2 during the MAOcatalyzed metabolism of different biogenic amines was linearly related to the concentration of intact, coupled brain mitochondria (Fig. 1). The substrates examined were tyramine, phenylethylamine, and benzylamine, and the efficiency of H2O2 production followed the order: ktyramine Å 1.6; kphenylethylamine Å 1.0; kbenzylamine Å 0.4 nmol H2O2/min/mg protein, respectively. These rates are higher than those published elsewhere, probably due to the sensitivity of the method and the constant stirring allowing greater diffusibility of O2 . The first amine is metabolized by
AID
ARCH 9698
/
6b23$$$462
10-28-96 07:52:16
FIG. 2. Dependence of the rate of formation of H2O2 on tyramine concentration. Assay conditions: Intact brain mitochondria (0.3 mg protein/ml) in mannitol/sucrose/EDTA/potassium phosphate buffer, pH 7.4, were supplemented with various amounts of tyramine and H2O2 formation measured as described under Materials and Methods. The inset shows a double reciprocal plot, which permits to calculate a KM for tyramine of 0.65 mM.
arca
AP: Archives
298
HAUPTMANN ET AL.
Calculation of Intramitochondrial Steady-State Concentration of H2O2 during the Metabolism of Tyramine by MAO and Comparison to That Built during Mitochondrial Electron Transport The assessment of intramitochondrial H2O2 steadystate concentration has a deep implication for the understanding of oxidative damage to mitochondrial matrix components, such as oxidative injury of DNA, protein modifications, and the intramitochondrial thiol status. The intramitochondrial H2O2 steady-state concentration ([H2O2]ss) was calculated using the steady-state approximation method (19) by which the rate of production of H2O2 equals its rate of consumption or removal: /d[H2O2]/dt Å 0d[H2O2]/dt. The rate of formation of H2O2 equals that of consumption during the metabolism of 2 mM tyramine (Fig. 2) (concentration used in experiments involving mtDNA). This rate was 2.71 nmol/min/mg protein. Assuming that 1 mg mitochondrial protein Å 1 ml (20), the above rate corresponds to 4.52 1 1005 M s01. At least in the brain mitochondrial matrix, the removal of H2O2 is the domain of glutathione peroxidase (GPx); catalase appears to play no role in the removal of H2O2 under these experimental conditions because, on the one hand, its level is far lower than that described for heart mitochondria (21, 22) and, on the other hand, it appears to be localized in particles described as microperoxisomes (23). The concentration of GSH is not considered in this equation because it is present in large amounts (about 5–8 mM) and does not represent a limiting step. Mitochondria contain about 1.17 1 1006 M glutathione peroxidase, and the rate of H2O2 reduction by the enzyme is k Å 5 1 107 M01 s01 (19). Hence, the steady-state concentration of intramitochondrial H2O2 according to the equation below is 7.71 1 1007 M (average of three individual experiments).
05
01
4.51 1 10 M s 5 1 10 M01 s01 [1.17 1 1006 7
Å 7.71 { 0.26 1 1007
M]
.
M
The rate of H2O2 formation by the rat brain respiratory chain during the oxidation of succinate in the presence of antimycin A was 57 pmol/min/mg protein, corresponding to 9.5 1 1007 M s01. This value is in the range of that previously reported (180 pmol/min/mg protein) (24). Higher rates were reported when the assay medium was potassium phosphate in the absence of man-
AID
ARCH 9698
/
/d[H2O2]/dt (M s01 1 1006)
[H2O2]ss (M (1007))
3.33 5.50 10.21 0.95
0.57 0.94 1.75c 0.16
45.2 26.7 16.7 6.7
7.71 — — —
Inner mitochondrial membrane Electron transfer chain Rat liver a Rat lung a Rat heartb Rat brain Outer mitochondrial membrane Monoamine oxidase activity Tyramined Tyraminee Phenylethylaminee Benzylaminee a
Values taken from Ref. (32). Values taken from Ref. (31). c This value considers the high catalase activity in the heart intramitochondrial matrix (concentration of catalase Å 0.72 1 1006 M; k Å 4.6 1 107 M s01 (Ref. 2). d Value obtained with 2 mM tyramine. e Values representing the initial rate and therefore not utilized to calculate the [H2O2]ss . Assay conditions for brain inner mitochondrial membrane activities were 0.17 mg/ml mitochondrial protein in buffer, pH 7.4, supplemented with succinate (7 mM) and antimycin A (4.7 nmol/mg protein). Assay for outer mitochondrial membrane activity was as described in the legend to Fig. 1. b
nitol (223 pmol/min/mg protein) (25). Therefore, the [H2O2]ss built in the brain mitochondrial matrix under the above conditions was: [H2O2]ss Å Å
/d[H2O2]/dt k[GPx] 9.5 1 1007 M s01 5 1 107 M01 s01 [1.17 1 1006
Å 1.62 { 0.26 1 10
/d[H2O2]/dt [H2O2]ss Å k[GPx] Å
TABLE II
Production of H2O2 by Rat Brain Mitochondria and the Calculated Matrical [H2O2]ss during the Electron Transfer in the Presence of Antimycin A and during the MAO-Catalyzed Oxidation of Amines
6b23$$$462
10-28-96 07:52:16
08
M]
M.
This value is close to the ‘cellular’ [H2O2]ss reported by Chance et al. (26) and in agreement with the intramitochondrial value recently reported by Boveris and Cadenas (19) (Table II). In conclusion, the [H2O2]ss obtained during the metabolism of biogenic amines by the outer membrane MAO activity in brain mitochondria is about 48-fold higher than that originating during the oxidation of substrates via complex II of the electron transfer chain in the presence of antimycin A. Oxidative Impairment of mtDNA The [H2O2]ss built by the metabolism of tyramine by MAO-A/B is expected to contribute to oxidative damage
arca
AP: Archives
METABOLISM OF TYRAMINE BY MONOAMINE OXIDASE A/B
299
FIG. 3. Induction of single strand breaks during the metabolism of tyramine by rat brain mitochondria. General assay conditions: intact rat brain mitochondria (0.3 mg protein/ml) in the buffer described in the legend to Fig. 2 were incubated with 2 mM tyramine at 377C under gentle stirring for 60 min in the absence and presence of different compounds. Electrophoretic distribution of form II (nicked) and form I (supercoiled) mtDNA: (A) Ethidium bromide staining. Left lane, molecular size markers (lDNA/Hind III). Lane 1, mtDNA plus tyramine; lane 2, mtDNA. (B) Southern blotting hybridization. Lane 1, mtDNA plus tyramine; lane 2, mtDNA; lane 3, mtDNA plus tyramine plus catalase (400 Urml01); lane 4, mtDNA plus tyramine plus tranylcypromine. (C) Time-dependent formation of mtDNA strand breaks. Under the general assay conditions described above, mitochondria were incubated with tyramine for 15 (n Å 4), 30 (n Å 8), and 60 (n Å 3) min. Inset shows the linear dependence on strand break formation on time.
of intramitochondrial components and, likewise, to contribute substantially to the cytosolic [H2O2]ss . Intact, coupled rat brain mitochondria incubated with tyramine produced H2O2 at a rate of 2.71 nmol/min/mg protein. This elicited a significant induction of intramitochondrial DNA single strand breaks. The intramitochondrial H2O2 may undergo a site-specific homolytic scission by copper pools other than cytochrome oxidase (27) or by mtDNA-bound transition metals. This, along with the restricted diffusibility (about 6–10 molecular diameters) and the high reactivity of HO• provides the conditions for an oxidative attack on bases (preferentially guanosine) and sugar residues in mtDNA. The former implies an addition reaction usually leading to the formation of 8-hydroxydeoxyguanosine; the latter mainly entails a C4 mechanism involving H abstraction
AID
ARCH 9698
/
6b23$$$462
10-28-96 07:52:16
from the sugar moiety eliciting single strand break formation (28–30). Single strand breakage of form I (supercoiled or close circular) mtDNA leads to form II (nicked or open circular) mtDNA, represented in the lower and upper bands, respectively, of the ethidium bromide staining (Fig. 3A) and Southern blotting hybridization (Fig. 3B). Control or untreated mitochondria show a single strand break background value of 0.75/104 bp (Fig. 3B, lane 2), which increases to 3.12 { 0.69/104 bp upon incubation with tyramine for 60 min (Fig. 3B, lane 1). The mitochondrial membrane is impermeable to catalase, but the hemoprotein affects the intramitochondrial [H2O2]ss upon displacement of the equilibrium; hence, a substantial decrease (Ç60%) of single strand break formation was obtained in the presence of catalase (1.27/
arca
AP: Archives
300
HAUPTMANN ET AL. TABLE III
[H2O2]ss in Control and CDNB-Treated Rat Brain Mitochondria [H2O2]ss (M 1 1007)
[GSH] (nmol/mg protein)
Inner mitochondrial membrane Electron transfer chain Outer mitochondrial membrane MAO-A/B activity
Control
/CDNB
Control
/CDNB
7.95 { 0.42
2.19 { 0.37
0.16 { 0.02
0.22 { 0.03
7.95 { 0.42
2.19 { 0.37
7.71 { 0.25
9.91 { 0.97
Note. Assay conditions as described under Materials and Methods.
104 bp; Fig. 3B, lane 3). Tranylcypromine, an effective irreversible inhibitor of MAO, abolished strand break formation (i.e., the values in the presence of the inhibitor are those obtained under control (background level) conditions (0.75/104 bp; Fig. 3B, lane 4). Figure 3C shows a time-dependent increase of single strand breakage of mtDNA during incubation with tyramine (linearly related with time; Fig. 3C, inset) obtained from data from Southern blot hybridization. Each value represents the ratio of tyramine treated sample/control of the relative increase of form II (nicked) mtDNA, according to the formula described under Materials and Methods. The time-dependent increase of single strand breaks was as follows: 1.5 { 0.29, 2.12 { 0.28 (P £ 0.05), and 3.12 { 0.69 (P £ 0.05)-fold at 15-, 30-, and 60 min incubation, respectively (Fig. 3C). Effect of the Intramitochondrial GSH on the [H2O2]ss and the Oxidative Damage to mtDNA Incubation of intact, coupled rat brain mitochondria with 1-chloro-2,4-dinitrobenzene for 8 min elicited 72% depletion of the intramitochondrial GSH content (7.95 nmol/mg protein in control mitochondria and 2.38 nmol/mg protein in CDNB-treated mitochondria). This appears to be the maximal GSH depletion level to be reached in rat brain mitochondria: exposure of brain mitochondria to longer incubation periods or higher CDNB concentrations did not change significantly this percentage, but altered the mitochondrial functions. Depletion of GSH requires a GSH transferase-catalyzed conjugation between the thiol and CDNB. This treatment as described above did not affect the P/O and respiratory control ratios of mitochondria when tested with glutamate or succinate as electron donors. Table III lists the GSH content in different mitochondrial preparations and the corresponding [H2O2]ss . A slight increase (1.38-fold, P õ 0.05) of [H2O2]ss in the CDNB-treated mitochondria was observed during the metabolism of succinate in the presence of antimycin A. Likewise, a 1.28-fold increase in the [H2O2]ss associated
AID
ARCH 9698
/
6b23$$$463
10-28-96 07:52:16
with the metabolism of tyramine in CDNB-treated mitochondria was calculated. A 68% lower than controls single strand break formation was observed in GSH-depleted rat brain mitochondria during the oxidation of tyramine with MAO A/B for 60 min (not shown). Regardless of the GSH content of brain mitochondria, it is to be noted that the [H2O2]ss built up during the MAO-catalyzed metabolism of tyramine is about 48-fold higher (average from the data shown in Table III) than that built during the oxidation of succinate in the presence of antimycin A. It is also expected that the H2O2 formed during the outer membrane MAO reaction would contribute to the cytosolic [H2O2]ss . Electron Spin Resonance Studies ESR in conjunction with the spin trap 5,5-dimethyl1-pyrroline-N-oxide revealed after supplementation of intact brain mitochondria with tyramine a spectrum consisting of a composite of DMPO-OH (aN Å 14.9 G; aH Å 14.9 G) and DMPO-CH3 (aN Å 15.8 G; aH Å 22.5 G) adducts in a ratio 2:1 (Fig. 4). The latter, a carboncentered radical, is expected to originate from the reaction of HO• with mannitol (k Å 2.7 1 109 M01 s01) (31), present in a concentration of 250 mM in the buffer. These signals were substantially amplified in a timedependent manner by the addition of FeSO4 suggesting the occurrence of a Fenton-driven mechanism. The formation of HO• in this experimental model appears irrelevant to the mechanism of mtDNA oxidative damage on the following basis: (a) HO• is a strong electrophile which diffuses 10 molecular diameters from its site of formation before reacting with the target; consequently, a contribution of HO• to mtDNA oxidative damage is highly unlikely. (b) The outer mitochondrial membrane itself is expected to be the primary target for HO• (5, 6). (c) Conversely, the HO• formed upon decomposition of MAO-generated H2O2 in the bulk solution may elicit oxidative damage to extramitochondrial targets within
arca
AP: Archives
METABOLISM OF TYRAMINE BY MONOAMINE OXIDASE A/B
FIG. 4. HO• and carbon-centered radical formation during the oxidative deamination of tyramine by mitochondrial monoamine oxidase. Assay conditions: intact rat brain mitochondria (0.27 mg/ml) in the buffer described in the legend to Fig. 2 were supplemened with 0.2 M DMPO. (A) No substrate added, (B) plus 0.2 mM tyramine, (C) plus 0.2 mM tyramine and 60 mM FeSO4 . (D) As in C after 10 min incubation. (E) As in C after 20 min incubation. (F) As in C in the presence of catalase (400 Urml01).
the diffusion distance of this species. Evidently, the metabolism of tyramine by outer membrane-located MAO contributes to the cytosolic [H2O2]ss because the signal was abolished by catalase (Fig. 4). Likewise, a freely diffusion of H2O2 into the intramitochondrial matrix appears to occur and may contribute to the tyramine-dependent, tranylcypromine-sensitive oxidative damage of mtDNA described above (see Fig. 3B and Table II). DISCUSSION
At least two sources may be considered to contribute to the intramitochondrial [H2O2]ss : on the one hand, the mitochondrial respiratory chain is a significant source of H2O2 and O2r0, whose formation accounts for about 4% of the total O2 consumed in mitochondrial preparations (26). Mammalian tissues produce H2O2 at rates of 0.08–0.1 nmol/min/mg protein in the presence of succinate and this values increase to about 0.2–1.0
AID
ARCH 9698
/
6b23$$$463
10-28-96 07:52:16
301
nmol/min/mg protein upon supplementation with antimycin A (32, 33). As shown above, intact rat brain mitochondria produce 0.056 nmol/min/mg protein during the oxidation of succinate (state 4) and in the presence of antimycin A; this value is 3.6-fold lower than that obtained with rat liver mitochondria. The intramitochondrial [H2O2]ss calculated for brain mitochondria under these conditions was 1.62 { 0.26 1 1008 M. On the other hand, H2O2 stemming from the oxidation of biogenic amines by the outer membrane MAO-A/B appears also to contribute partly to an intramitochondrial [H2O2]ss . This view is substantiated by (a) a previous report associating MAO-mediated production of H2O2 with the formation of intramitochondrial GSSG, hardly affected by catalase, but abolished by MAO inhibitors (4); (b) the tyramine-dependent, tranylcypromine-inhibitable mtDNA single strand break formation (Fig. 3B); (c) the inhibition of ESR signals (Fig. 4) by catalase, suggesting that H2O2 is formed outside the mitochondria and it diffuses across the inner mitochondrial membrane into the matrix. The intramitochondrial [H2O2]ss calculated during the oxidation of tyramine by the outer membrane MAO was 7.71 1 1007 M. In conclusion, the intramitochondrial [H2O2]ss obtained during the metabolism of tyramine by the outer membrane MAO-A/B is about 48-fold higher than that originating by the complex II-mediated oxidation of succinate in the presence of antimycin A. The rates of H2O2 formation originating from the mitochondrial electron-transfer chain and MAO activity are increased after partial depletion of GSH by CDNB. This is in disagreement with previous reports, in which CDNB treatment hardly affected the H2O2 formation by complex II of the respiratory chain (34). Furthermore, glutathione deficiency obtained by treatment of newborn rats with L-buthionine (S, R)-sulfoximine leads to mitochondrial damage in the brain visualized as a destroyed internal structure of mitochondria (35). The time-dependent formation of single strand breaks during the metabolism of tyramine by MAO-A/ B in vitro can be considered as a direct evidence of oxidative damage to mtDNA, due to the diffusion of H2O2 across the inner mitochondrial matrix. HO• presumably formed in a Fenton-like reaction catalyzed by DNA-bound transitions metals (e.g., Cu) attacks mainly the thymine and guanosine residues (k Å 109 – 1010 M01 s01), forming a primary reducing or oxidizing radical. The radical character is transferred to the sugar moiety resulting in single strand break formation (30). Alternatively, a direct H-atom abstraction of the sugar moiety by HO• leads to (36) single strand breaks. It may be surmised that the effect of HO• could be restricted by the presence of mannitol, an effective scavenger of HO•, in the incubation buffer. Replacement of mannitol by Tris buffer showed only a slight elevation of single strand breaks because also this com-
arca
AP: Archives
302
HAUPTMANN ET AL.
FIG. 5. Intramitochondrial [H2O2]ss , mtDNA single strand break, and a possible role for GSH during oxidative deamination of tyramine by MAO brain mitochondria.
pound is known to act as radical scavenger (results not shown). Moreover, the mechanism proposed entails a site-specific formation of HO• (not formation in the bulk solution) and, hence, the interference of mannitol or other HO• quenchers is debatable. In vivo single strand breakage may lead to mutagenicity resulting from single nucleotide substitutions during base repair. However, also certain repair endonucleases, which specifically recognize free radical-induced DNA base modifications, incise the DNA at the site of the modification, leading to single strand breaks (37). The presence of these enzymes in brain mitochondria remains to be proven. The accumulation of muta-
AID
ARCH 9698
/
6b23$$$463
10-28-96 07:52:16
tions of mtDNA in postmitotic cells is favored by the continuous turnover of mtDNA independent of cell replication and a yet unknown mechanism accelerating the division of those mitochondria with decreased or defective ATP synthesis (11). It has been reported that incubation of intact rat liver mitochondria with succinate/antimycin A leads to an increase of the formation of 8-hydroxydeoxyguanosine, another widely used approach to assess DNA oxidative damage (38). It is also established that the rate of O2r0 and H2O2 production of the respiratory chain increases with age and this correlates with higher levels of 8-hydroxydeoxyguanosine in mtDNA (39, 40).
arca
AP: Archives
METABOLISM OF TYRAMINE BY MONOAMINE OXIDASE A/B
Formation of 8-hydroxydeoxyguanosine causes exclusive misreading during DNA synthesis in vitro (G r T transversion) (41). Preliminary data obtained in our laboratory suggest also an increase of 8-hydroxydeoxyguanosine formation during the oxidative deamination of tyramine by brain mitochondria MAO-A/B. However, no unambiguous results could be obtained with this approach, probably due to the sensitivity of the assay, which, however, proved successful with rat liver mitochondria. Recently, a direct correlation was established between accumulation of 8-hydroxydeoxyguanosine and single strand break formation when plasmid DNA was exposed to oxidative conditions (42). Despite the extensively described antioxidant effect of cellular thiols, the depletion of 72% of the intramitochondrial GSH led to a substantially diminished single strand break formation (68% less compared to the control) during tyramine-mediated formation of H2O2 . These data are supported by previous reports, which assessed single strand breaks formation by 1O2 (43), presumably because thiols keep the DNA-bound transition metals in a reduced state, facilitating the cleavage of H2O2 in a Fenton site-specific manner. This is expected to occur even at lower [H2O2]ss ; furthermore, thiols may be autoxidized in the presence of DNAbound CuII, forming oxygen free radicals as well as thiyl radicals (43–45). In summary, intramitochondrial GSH appears to have prooxidant functions at least in terms of protection against mtDNA oxidative damage. The exact mechanism by which a decrease of intramitochondrial GSH is associated with a decrease (and not an increase) of single strand break formation remains to be elucidated, but it is tempting to suggest that the thiol is involved in the reduction of transition metals essential for mtDNA strand breakage. These views are summarized in Fig. 5. The results presented here offer the possibility of an oxidative damage of mtDNA in vivo during the MAOA/B-catalyzed metabolism of biogenic amines with the consequential impairment of oxidative phosphorylation and, hence, the energy level of the cell. Neurodegenerative diseases, such as Parkinson’s disease, are associated with an increase of MAO-B and a selective less active complex I of the respiratory chain. An impaired complex I activity seems not accompanied by deletions of mtDNA (46); hence, it might be likely related to a direct HO• attack stemming from site specific cleavage of H2O2 by, for example, redox active copper pools other than those in cytochrome oxidase. Alternately, it has been proved that dopamine by itself may inhibit mitochondrial NADH-dehydrogenase activity (47), a fact which might raise importance if the MAO-A/B activity is impaired. It has been shown that monoamines and their metabolites (e.g., dihydroxyphenylacetic acid, dopamine, homovanillic acid) may act as radical scavengers (48). Following this, the significance of our results
AID
ARCH 9698
/
6b23$$$463
10-28-96 07:52:16
303
in vivo must be considered under the possible impact of an equilibrium encompassing various factors, such as the age-related increase of MAO activity, the H2O2 production, the intramitochondrial GSH status, the radical quenching activity of MAO products and substrates, and the possible age-related decrease of the synthesis of certain neurotransmitters (e.g., dopamine) (46). An impairment of this sensitive equilibrium results in an accumulation of somatic mtDNA mutations, which along with the progress of germinal (inherited) mutations may lead above the 90% threshold of mutated mtDNA (11), a value required to make apparent dysfunctional oxidative phosphorylation and to observe clinical symptoms. Furthermore, H2O2 generated during the outer membrane monoamine oxidase activity-catalyzed oxidation of catecholamines may have important implications for the ischemia–reperfusion syndrome. This may bear relevance in connection with the increased levels of catecholamines related to myocardial ischemia/reperfusion injury (49, 50) and the proposed role of oxygen radicals arising from enzyme-catalyzed catecholamine oxidations (51), rather than autooxidation. ACKNOWLEDGMENT Work was supported by Grant HL53467 from NIH. Note added in proof. Detection of H2O2 during the metabolism of tyramine by the method based on the HRP/H2O2-coupled dimerization of p-hydroxyphenylacetic acid may lead to an overestimation due to the simultaneous contribution to fluorescence by the HRP/ H2O2-mediated oxidation and subsequent dimerization of tyramine. However, measurement of H2O2 formation by the HRP/scopoletin method – free of the above interference – yielded the same values reported here.
REFERENCES 1. Cohen, G. (1983) J. Neural Transmembr. Suppl. 19, 89–103. 2. Sandri, G., Panfili, E., and Ernster, L. (1990). Biochim. Biophys. Acta 1035, 300–305. 3. Werner, P., and Cohen, G. (1991). FEBS Lett. 280, 44–46. 4. Savov, V. M., Prilipko, L. L., and Gorkin, V. Z. (1983). Acta Physiol. Pharmacol. Bulg. 9, 3–13. 5. Kagan, V. E., Smirnov, A. V., Savov, V. M., and Gorkin, V. Z. (1984). Vopr. Med. Khim. 30, 112–118. 6. Hruszkewycz, A. M. (1988). Biochem. Biophys. Res. Commun. 153, 191–1977. 7. Seregi, A., Serfozeo, P., and Mergl, Z. (1983). J. Neurochem. 40, 407–413. 8. Konradi, C., Riederer, P., and Youdim, M. B. (1986). J. Neural. Transmembr. Suppl. 22, 61–73. 9. Wallace, D. C. (1992). Annu. Rev. Biochem. 61, 1175–1212. 10. Saura, J., Richards, J. G., and Mahy, N. (1994). Neurobiol. Aging 15, 399–408. 11. Kadenbach, B., Mu¨nscher, C., Frank, V., Mu¨ller-Ho¨cker, J., and Napiwotzki, J. (1995). Mutat. Res. 338, 161–172. 12. Richter, C. (1995). Int. J. Biochem. Cell Biol. 27, 647–653. 13. Basford, R. E. (1967). Methods Enzymol. 10, 97–101.
arca
AP: Archives
304
HAUPTMANN ET AL.
14. Sciamanna, M. A., and Lee, C. P. (1993). Arch. Biochem. Biophys. 305, 215–224. 15. Jocelyn, P. C., and Cronshaw, A. (1985). Biochem. Pharmacol. 34, 1588–1590. 16. Martin, J., and White, I. N. H. (1991). J. Chromatogr. 568, 219– 224. 17. Shigenaga, M. K., Aboujaoude, E. N., Chen, Q., and Ames, B. N. (1994). Methods Enzymol. 234, 16–33. 18. Houlihan, P., and Gershon, P. (1994). in Monoamine Oxidase Inhibitors in Neurological Diseases (Lieberman, A., Olanow, C. W., Youdim, M. B. H., and Tipton, K., Eds.), pp. 290–309, Dekker, New York. 19. Boveris, A., and Cadenas, E. (1996). in Oxygen, Gene Expression, and Cellular Function (Clerch, L. B., and Massaro, D., Eds.), Dekker, New York, in press. 20. Tyler, D. D. (1975). Biochem. J. 147, 493–504. 21. Radi, R., Turrens, J. F., Chang, L. Y., Bush, K. M., Crapo, J. D., and Freeman, B. A. (1991). J. Biol. Chem. 266, 22028–22034. 22. Nohl, H., and Hegener, D. (1978). FEBS Lett. 89, 126–130. 23. Sinet, P. M., Heikkila, R. E., and Cohen, G. (1980). J. Neurochem. 34, 1421–1428. 24. Adamo, A. M., Llesuy, S. F., Pasquini, J. M., and Boveris, A. (1989). Biochem. J. 263, 273–277. 25. Patole, M. S., Swaroop, A., and Ramasarma, T. (1986). J. Neurochem. 47, 1–8. 26. Chance, B., Sies, H., and Boveris, A. (1979). Physiol. Rev. 59, 527–605. 27. Massa, E. M., and Giulivi, C. (1993). Free Radicals Biol. Med. 14, 559–565. 28. Steenken, S. (1989). Chem. Rev. 89, 503–520. 29. von Sonntag, C. (1987). The Chemical Basis of Radiation Biology, pp. 116–166, Taylor & Francis, London. 30. Symons, M. C. R. (1987). J. Chem. Soc. Faraday Trans. I 83, 1– 11. 31. Halliwell, B. (1978). FEBS Lett. 92, 321–326.
AID
ARCH 9698
/
6b23$$$463
10-28-96 07:52:16
32. Cadenas, E., and Boveris, A. (1980). Biochem. J. 188, 31–37. 33. Forman, H. J., and Boveris, A. (1982). in Free Radicals in Biology (Pryor, W. A., Ed.), Vol. 5, pp. 65–90, Academic Press, New York. 34. Zoccarato, F., Cavallini, L., Deana, R., and Alexandre, A. (1988). Biochem. Biophys. Res. Commun. 154, 727–734. 35. Jain, A., Ma˚rtensson, J., Stole, E., Auld, P. A. M., and Meister, A. (1991). Proc. Natl. Acad. Sci. USA 88, 1913–1917. 36. Meunier, B., Pratviel, G., and Bernadou, J. (1994). Bull. Soc. Chim. Fr. 131, 933–943. 37. Epe, B., and Hegler, J. (1994). Methods Enzymol. 234, 122–131. 38. Giulivi, C., Boveris, A., and Cadenas, E. (1995). Arch. Biochem. Biophys. 316, 909–916. 39. Sohal, R. S., and Brunk, U. T. (1992). Mutat. Res. 275, 295–304. 40. Ames, B. N., Shigenaga, M. K., and Hagen, T. M. (1993). Proc. Natl. Acad. Sci. USA 90, 7915–7922. 41. Kasai, H., and Nishimura, S. (1991). in Oxidative Stress: Oxidants and Antioxidants (Sies, H., Ed.), pp. 99–116, Academic Press, San Diego. 42. Toyokuni, S., and Sagripanti, J.-L. (1996). Free Radicals Biol. Med. 20, 859–864. 43. Devasagayam, T. P. A., DiMascio, P., Kaiser, S., and Sies, H. (1991). Biochim. Biophys. Acta 1088, 409–412. 44. Reed, C. J., and Douglas, K. T. (1989). Biochem. Biophys. Res. Commun. 162, 1111–1117. 45. Misra, H. P. (1974). J. Biol. Chem. 249, 2151–2155. 46. Fahn, S., and Cohen, G. (1992). Ann. Neurol. 32, 804–812. 47. Ben-Shachar, D., Zuk, R., and Glinka, Y. (1995). J. Neurochem. 64, 718–723. 48. Liu, J., and Mori, A. (1993) Arch. Biochem. Biophys. 302, 118– 127. 49. Singal, P. K., Kapus, N., Dhillon, K. S., Beamish, R. E., and Dhalla, N. S. (1982). Can. J. Physiol. Pharmacol. 60, 1390–1397. 50. Flaherty, J. T., and Weisfeldt, M. L. (1988). Free Radicals Biol. Med. 5, 409–419. 51. Jewet, S. L., Eddy, L. J., and Hochstein, P. (1989). Free Radicals Biol. Med. 6, 185–188.
arca
AP: Archives