The Formation of Sulfur Trioxide Radical Anion ...

THEJOURNAL 01 BIOLOGICALCHEMISTRY Vol. 257, No. 9. Issue of May 10, pp. 5050-5055. 1982 Prrnted in U.S A .

The Formation of Sulfur Trioxide Radical Anion during the Prostaglandin Hydroperoxidase-catalyzedOxidation of Bisulfite (Hydrated Sulfur Dioxide)* (Received for publication, July 21, 1981)

Carolyn MottleyS, Ronald P. Mason, ColinF. Chignell, Kandiah Sivarajah8, and Thomas E. Elings From the Laboratory of Environmental Biophysics and §Laboratoryof Pulmonary Function and Toxicology, National Institute of Environmental Health Sciences, Research Triangle Park, NorthCarolina 27709

.

Sulfur oxides are pervasivein the environment. Sulfur dioxide is a major air pollutant near large cities (l),while the ionized forms, bisulfite and sulfite, are used as preservatives in food, wine, and drugs. These sulfur oxides are metabolized through the oxidation of (bi)sulfite to sulfate, followedby excretion in the urine (2, 3). The nomenclature (bi)sulfite is used when it is not known which species is involved in a reaction. Most of the oxidation probably occurs as a result of the action of the mitochondrial enzyme sulfite oxidase which oxidizes (bi)sulfite to sulfate in a direct two-electron oxidation (4). * The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “aduertisernent” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Permanent address, Department of Chemistry, Luther College, Decorah, IA 52101.

+

In the nasal passages and lung, sulfur dioxide is hydrated rapidly according to the equation Hz0

+ SO, e HS03 + H+

with the equilibrium constant 1.7 X lo2mol/liter (3). Bisulfite, which predominates at physiological pH, is a weak acid which dissociates according to thereaction HSO:

+ H,O e H,O+ + SO;*

with an equilibrium constant 1.02 X mol/liter. Thus, although there is always an equilibrium between sulfite and bisulfite, at pH values greater than 7 the equilibrium lies to the right and sulfite predominates. Extensive studies on the autooxidation of (bi)sulfite to sulfate have shown that thereaction mechanism involvesfree radicals (5,6).(Bi)sulfitefree radicals formed by autooxidation are involved in a number of biologically important reactions: the oxidation of diphosphopyridine nucleotide (7) and methionine (a), the destruction of tryptophan (9) and p-carotene (lo), the addition of SO; across the double bonds of alkenes (11, 12) and of various nucleotides and nucleic acids (13), the peroxidation of corn oil (14) and rat liver homogenate (15), and the cleavage of DNA (16). This paper reports an ESR investigation of the oxidation of (bi)sulfite to theS03 radical anion by prostaglandin synthase. The SO; radical is thought to be one of the species produced byautooxidation. Prostaglandin synthase consists of two enzymatic components: a fattyacid cyclo-oxygenase which converts arachidonic acid to the hydroperoxide, prostaglandin Gz, and a hydroperoxidase which reduces prostaglandin G2 to the alcohol, prostaglandin Hz. The hydroperoxidase often catalyzes many of the reactions catalyzed by horseradish peroxidase (17). Horseradish peroxidase (18)catalyzes the one-electron oxidation of (bi)sulfite tothe sulfur trioxide anion radical (SO,). In this paper, a similar oxidation of (bi)sulfite by prostaglandin synthase is described. This free radical metabolism of (bi)sulfite by prostaglandin synthase is important in light of the widespread occurrence of this enzyme in mammalian systems and the prevalence of (bi)sulfite in the environment. For example, the human lung provides one possible site for this reaction since the lung contains prostaglandin synthase activity (19) and is exposed to at least some of the SOz breathed in from polluted air (1, 20). EXPERIMENTALPROCEDURES

Materials Sodium sulfite and hydrogen peroxide (30%)were obtained from Fisher Scientific. Indomethacin, Type VI horseradish peroxidase, and diethylenetriamine pentaacetic acid were obtained from Sigma. Cat-

5050

Downloaded from www.jbc.org by guest, on July 12, 2011

The mechanism of prostaglandin synthase-dependent (bi)sulfite (hydrated sulfur dioxide) oxidation was investigated using an enzyme preparation derived from ram seminal vesicles. The horseradish peroxidase-catalyzed oxidation of (bi)sulfite was used as a model system. Incubation of (bi)sulfite with prostaglandin synthase and arachidonic acid, 15-hydroperoxyarachidonic acid, or Hz02 results in the formation of the reactive sulfur trioxide anion radical (SO;). The horseradish peroxidase/HzOzsystem also oxidizes (bi)sulfite to SO;. This free radical reacts with oxygen resulting in oxygen consumption by these incubations. The free radical was detected with the indirect electron spin resonance technique of spin trappin The $0; radical adduct formed by the reaction of 0; with the spin gives a nitroxtrap, 5,5-dimethyl-l-pyrroline-N-oxide, ide free radical with a nearly unique electron spin resonance spectrum (a”= 16.0 G and aN = 14.7 G ) . Using the spin-trapping technique, the $0; could be detected even in incubations of guinea pig lung microsomes. When arachidonic acid-derived prostaglandin Gz was the source of hydroperoxide, formation of SO; could be inhibited by indomethacin. When 15-hydroperoxyarachidonic acid or hydrogen peroxide was used to drive the enzymatic oxidation of (bi)sulfite, indomethacin had no effect. This hydroperoxidase activity was not nearly as heat-labile as the cyclo-oxygenase reaction which forms prostaglandin Gz. Finally, the peroxidatic oxidation of (bi)sulfite may occur in vivo in competition with the mitochondrial sulfite oxidase, which oxidizes(bi)sulfite to sulfate without the formation of free radicals.

Free Metabolism Radical

of

Bisulfite

5051

alase was obtained from Boehringer and arachidonic acid (99% pure) was obtained from Nu Chek Prep., Inc. (Elysian, MN). 5,5-DimethylI-pyrroline-N-oxide was obtained from Aldrich Chemical Co. and distilled before use. 15-Hydroperoxyarachidonicacid was prepared from arachidonic acid using soybean lipoxygenase (Sigma) according to published methods (21). [l-'4C]Arachidonic acid was obtained from New England Nuclear.

cavity. In those experiments in which indomethacin was used as an inhibitor of prostaglandin synthase activity, the microsomeswere preincubated in the buffer with indomethacin (-500pM). Next, sulfite, DMPO, and cofactor were added, in that order. In experiments with horseradish peroxidase, 1 mg/ml of enzyme, 10 PM HzO~,1 mM NazS03, and 270 m~ DMPO were used. Prostaglandin Analysis-Microsomes (1.0 mg of ram seminal vesicle protein) were suspended in 1.0 ml of 100 mM sodium borate/boric acid buffer, pH 7.9, containing 1 r n diethylenetriamine ~ pentaacetic acid. After warming to 37 "C, [l-'4C]arachidonic acid (200 nmol in 5 pl of ethanol) was added to initiate the reaction and the mixture was incubated for 5 min. The reaction was terminated by addition of 1 N HCl to pH 3.5 and themixture was extracted twice with 4 ml of ethyl acetate. The organic layer was removed, dried over anhydrous NazS04, and evaporated. The residue was redissolved in 0.1 ml of acetone and aliquots were analyzed. Prostaglandins were separated by high pressure liquid chromatography as described previously (45) and quantitated by measurement of radioactivity in eluted peaks by liquid scintillation techniques. Oxygen Uptake-Components were added one at a time to a 2-ml chamber equipped with a Clark electrode. Order of addition varied, although final concentrations were as follows: 1m g / d of ram seminal vesicle microsomal protein, 1m~ Na2S03,500 p~ indomethacin, and 410 p~ arachidonic acid or 10 p~ Hz& in sodium borate/boric acid buffer, pH 7.9, containing 1 m~ diethylenetriamine pentaacetic acid. Measurements were made at room temperature with a Yellow Springs Instrument Oxygen Monitor (model 53).

HRP

+ H202 -+

HRP-compound I

HRP-compound I

+ (bi)sulfite

4

+ Hz0

HRP-compound I1 + SO;

HRP-compound I1 + (bi)sulfite + HRP + SO;

(1)

(2)

(3)

where HRP is horseradish peroxidase. Although the $0; radical anion wasobserved directly in this system, under anaerobic conditions, no direct ESR signal was observed under aerobic conditions (18). The reactions of $0, with oxygen apparently lowered the $0; concentration below the limit of detection for direct ESR. The rapid consumption of oxygen by these incubations supports this hypothesis (Fig. 2). The detection of the DMPO-SO; radical adduct in horseradish peroxidase incubations, under aerobic conditions, implies that

OG

RESULTS

Attempts to detect the$0; radical by direct ESR in incubations at pH 7.9 containing (bi)sulfite, ram seminal vesicle microsomes, and either Hz02,t-butylhydroperoxide, or 15hydroperoxyarachidonicacid under anaerobic conditions were unsuccessful. Presumably, the $0; radical anion concentrations in these incubations were below the detection limit for direct ESR (-lo-' M). In order to detect this labile free radical intermediate, the technique of spin trapping was used. The addition of $0, across the nitroso or nitrone double bond of a spin trap forms

B

A

A .,

A

C

D

RG.1. The ESR spectrum of DMPO-trapped $0; produced in a system containing horseradish peroxidase. A, incubation containing 1 m~ NaZS03, 1 mgof horseradish peroxidase/& of incubation, 10 PM HzOZ,and 270 m~ DMPO. Sodium borate/boric acid buffer, pH 7.88, 1 mM in diethylenetriamine pentaacetic acid was used. B, same as A but with no addition of H~OZ. C, same as A but with no horseradish peroxidase. D, same as A but with no Na2SO3. ' The abbreviation used is: DMPO, 5,5-dimethyl-l-pyrroline-N-o~-Instrumental conditions: modulation amplitude, 0.33 G ; sweep time, 2 min; time constant, 0.128 s; microwave power,20 milliwatts. ide.


a relatively stable nitroxide free radical adduct (24, 25). The choice of the spin trap is of critical importance, because the nitroso spin traps, nitrosobenzene (18,26,27) and 2-methyl-2nitrosopropane (28), form nitroxide adducts with SOT that are indistinguishable from the adductsformed with SO;. The spin trap DMPO was usedin this study for two reasons. First, the DMPO-SO; nitroxide adduct was predominant, i.e. other Enzyme Preparation trapped radicals were in very low concentration and were Ram seminal vesicles were obtained from a local slaughterhouse, essentially unobservable. Second, the $0; radical was not trimmed of excess fat and connective tissue, and stored at -70 "C trapped with DMPO, so there was noproblem in determining until used. Lungs were removed from adult male guinea pigs (350-450 which adduct was present. This conclusion was based onthe g of body weight) obtained from Carsworth Breeders, Portage, MI, and freed of extraneous tissue. Microsomes were prepared from the failure to observe the DMPO-SO; nitroxide adduct in soluvesicles and lungs as previously described (22). Ram seminal vesicle tions of DMPO and sodium dithionite, which is in equilibrium microsomes were stored at -70 "C and wereused within 4 h of with SO; (29). The ESR spectrum of the SOT radical adduct of the spin thawing. Guinea pig lung microsomes were used within 3 h of preparation. Microsomal prostaglandin synthase activity was determined trap DMPO is shown in Fig. 1. The hyperfiie coupling conby measuring arachidonic acid-dependent oxygen uptake using a stants were uH= 16.0 G and a N = 14.7 G, and the g value was Clark electrode. Protein concentration was measured by the method 2.0056 (determined with di-tert-butylnitroxide as an internal of Lowry et al. (23) using bovine serum albumin as a standard. standard with g = 2.0055) (30).These parameters are in good agreement with those reported for the DMPO-SOT nitroxide Incubation Procedures ESR-Microsomes (0.5 mg of ram seminal vesicle protein or 5 mg adduct in photochemical systems (31,32). Free radicals sufiof guinea pig lung protein) were suspended in 0.5 ml of 100 mM ciently stable for observation by direct ESR are not usually sodium borate/boric acid buffer, pH 7.9, containing 1 mM diethyl- reactive enough to add across double bonds, and, therefore, enetriamine pentaacetic acid (to inhibit the autooxidation of are not spin trapped. The sulfur trioxide anion radical (SOT) (bi)sulfite), 1 mM Na2S03,and 270 RIM DMPO.' Arachidonic acid (200 is one of several exceptions to this generalization. nmol in 2.5 plof ethanol), 15-hydroperoxyarachidonicacid (40 nmol The spectrum in Fig. 1 was the result of the oxidation of in 5 p1of ethanol), or Hz02 (5 nmol in 15 p1 of buffer) was added to initiate the enzymatic reaction. The mixture was transferred to a flat (bi)sulfite by the horseradish peroxidase/Hz02 system under cell and the ESR spectrum recorded at room temperature using a aerobic conditions, in the presence of DMPO. The following Varian Century Series E-I04 spectrometer equipped with a TMIM equations account for SO, formation in this system (18),

5052

Free Radical Metabolismof Bisulfite IO0

Fig. 4, D and B ) . In the absenceof arachidonic acid, catalase inhibited oxygen uptake by ram seminal vesicles in the presence of (bi)sulfite (cf: Fig. 4, C and B ) , and H 2 0 2 supported (bi)sulfite-dependentoxygen uptake by indomethacin-treated microsomes (Fig. 4 E ) . The peroxidatic activity of the ram seminal vesicles could be initiated either by hydrogen peroxide or 15-hydroperoxy-

s so -

60 -

40 -

20

0-

under these conditions SO; adds across the nitrone double with oxygen. bond of DMPO at a greater rate than it reacts $0; + DMPO + DMPO-SOB nitroxide

FIG. 3. The ESR spectrum of DMPO-trapped $0, produced by ram seminal vesicle microsomes. A, incubation containing I

m Na'SO,, 1 mg of ram seminal vesicle microsomal protein/ml of

(4)

incubation, 410 ~ L Marachidonic acid, and 270 mM DMPO. B, same as A but microsomes were preincubated with 500 PM indomethacin for This conclusion is supported by oxygen uptake studies,which 10 min at 4 " C . C, same as A but with no arachidonic acid. 0,same as show that theprior addition of 270 m DMPO totally inhibits C with 8OOO units of catalase/ml of incubation. E, same as A but microsomes were heat-denatured by heating for 2 min in boiling oxygen uptake in these incubations(Fig. 2). Fig. 3 shows the ESR spectrum of the $0: adduct of water. Instrumental conditions: modulation amplitude, 0.33 G ; sweep time, 4 min; time constant, 0.25 s; microwave power,20 milliwatts. DMPOinanincubationmixture containing ramseminal

vesicle microsomes and arachidonic acid. The spectrum is identical with that in Fig. 1. Indomethacin, a prostaglandin cyclo-oxygenase inhibitor which inhibits prostaglandin GPformation, significantly decreases the ESR signal intensity (Fig. 3B). There was some peroxidase activity in the microsomes, which was independent of arachidonic acid (Fig. 3 C ) . This peroxidase activity was largely driven by H202 as shown by inhibition with catalase (Fig. 3 0 ) . The H 2 0 2 was not added but may be formed by autooxidation of the (bi)suKite stock solution. The entire ESR signal was a result of enzymatic activity since heat-denatured microsomes yielded no signal (Fig. 3 E ) , but peroxidases other than prostaglandin synthase cannot be excluded, because no speciiic inhibitor of the peroxidase activity of prostaglandin synthaseis known. It should be noted that a similar, if not identical, spectrum to that of Fig. 3A has been reported from incubations containing ram seminal vesicles, DMPO, and arachidonic acid (33). Although (bi)sulfite was not intentionally added to the system it may acid. have been present as an antioxidant in the arachidonic The additionof arachidonic acid results in prostaglandin GZ formation, which consumes oxygen (Fig. 4A). Addition of sulfite to ram seminal vesicle microsomes and arachidonic acid stimulated oxygen uptake (cf Fig. 4, B and A ) . Neither DMPO nor diethylenetriamine pentaacetic acid affected the arachidonic acid-dependent rate of oxygen uptake. Preincubating the microsomes with indomethacin inhibited both arachidinic acid- and (bi)sulfite-dependent oxygen uptake (cf.

0

FIG. 4. Oxygen uptake curves for the sulfite/ram seminal vesicle microsomal system. Sodium borate/boric acid buffer (pH 7.88 and 1 mM diethylenetriamine pentaacetic acid) was placed in the chamber and other components added at points indicated. Final concentrations were 1 rn NazS03, 1 mg of microsomal protein/& of incubation, 400 p~ indomethacin, 820 ~ L Marachidonic acid, 10 p~ H,O,. 30.000 units of catalase/ml of incubation. RSV. ram seminal v≤ arachidonic acid; CAT, catalase.

A,


FIG. 2. Oxygen uptake with and without spin trap DMPO. Sodium borate/boric acid buffer, pH 7.88,l mM in diethylenetriamine pentaacetic acid (-) or buffer containing 270 mM DMPO (---) was placed in chamber and all other components added at times indicated. The oxygen consumption that occurs prior to the addition of Hz02 can be inhibited by catalase (18).This result implies the (bihlfitestock solution contains HzOz,which presumably forms as the result of (bi)sulfite autooxidation. Final concentrations were: 3.3 mMSOB', 0.5mg of horseradish peroxidase (HRP)/ml,10 PM HZ02 in a total volume of 3 ml.

5053

Free Radical Metabolism of Bisulfite arachidonic acid as shown in Fig. 5. As a peroxidatic cofactor, 15-hydroperoxyarachidonicacid is exceeded only by prostaglandin Gz (34). In both of these series of spectra, the microsomes were preincubated with indomethacin to inhibit any cyclo-oxygenase activity. At higher concentrations, H202 reacts directly with (bi)sulfite to produce $0; nonenzymatically (11, 18) Hz02 + (bi)sulfite + SOi

+ OH + OH-

(or H20)

(5)

F

I G

FIG. 6. The ESR spectrum of DMPO-trapped $0; produced in system containing guineapig lung microsomes. A, incubation containing 1 mM Na2S03, 10 mg of guinea pig lung microsomal p r o t e i n / d of incubation, 410 p~ arachidonic acid, and 270 mM DMPO. B, same as A but microsomes were preincubated with 529 PM indomethacin for 30 min a t 4 "C. C, same asA but with 7000 units of catalase/& of incubation. 0,same as B but with 7000 units of catalase/& of incubation. E, same as A but microsomes were heatdenatured by heating in boiling water for 2min. F, incubation consists of 1 mM Na~S03,410 PM arachidonic acid, 7000 units of catalase/ml of incubation, and 270 mM DMPO. G, incubation consists of 1 mM Na2S03, 10 mg of guinea pig lung microsomal protein/& of incubation, 7000 units of catalase/& of incubation, and 270 m~ DMPO. H, same as A but no Na2S03. Instrumental conditions: modulation amplitude, 0.33 G ; sweep time, 4 min; time constant, 0.5 s; microwave power, 20 milliwatts.

hydroxyl adduct of DMPO, which is often present as an impurity in solutions of DMPO (35). The othercontrols show no detectable nitroxide signals (Fig. 6, F-H). DISCUSSION

B

C

I

20 G

I

FIG. 5. The ESR spectrum of DMPO-trapped SO; produced by a 15-hydroperoxyarachidonicacid-initiated peroxidaseactivity in ram seminal vesicle microsomes. A, incubation containing 1 mM Na2S03, 1 mg of microsomal protein/& of incubation preincubated with 400 p~ indomethacin, 270 mM DMPO, and 15hydroperoxyarachidonic acid. E , same as A but with no microsomes or indomethacin. C, same asA but with heat-denatured microsomes. Instrumental conditions: modulation amplitude, 0.33 G; sweep time, 8 min; time constant, 0.5 s; microwave power, 20 milliwatts.

The recent direct observation of the one-electron oxidation of (bi)sulfite to $0; by horseradish peroxidase (18) led to an extension of that work to theprostaglandin synthase enzyme system, which often oxidizes aromatic chemicals in a manner identical with the oxidation by horseradish peroxidase (12). Prostaglandin synthase is widespread in biological systems including the lungs, which are exposed to SOZfrom air pollution. The peroxidase activity of prostaglandin synthase is responsible for the oxidation of (bi)sulfite to $0,. This is apparent since indomethacin, which inhibits the cyclo-oxygenase activity, decreases the ESR signal observed when arachidonic acid is used to initiate enzyme activity. The DMPOSO; nitroxide adduct was observed when either 15-hydroperoxyarachidonic acid or Hz02 was used as the initiator. Either of these two hydroperoxides support peroxidase activity without need of the cyclo-oxygenaseactivity.


but at 20 PM HzOz the $0; radical cannot be detected in the absence of ram seminal vesicles. The peroxidatic activity of prostaglandin synthase was much less heat-labile than the cyclo-oxygenase activity. In experiments initiated by arachidonic acid no signal was observed if the microsomes were kept at room temperature for approximately 1h. On the other hand,hydroperoxidase activity was initiated with 15-hydroperoxyarachidonic acid in microsomes which had been at room temperature for 24 h. The enzyme was totally denatured by heating the microsomes in a boiling water bath for 3 min. Guinea pig lung microsomes were also used as a source of prostaglandin synthase, because this tissue is more representative of activity of prostaglandin synthase in mammalian tissue. Furthermore the human lung provides one possible site for this reaction since the lung contains prostaglandin synthase (19) and is exposed to at least some of the SO2 in polluted air (20). Despite the fact that the guinea pig lung microsomes contain approximately 200 times less prostaglandin synthase activity than ram seminal vesicles, the ESR spectra obtained from guinea pig lung microsomes indicate the formation of SO: (Fig. 6 A ) . The signal was inhibited by indomethacin (Fig. 6s)and therewas some peroxidase activity, which could be inhibited by catalase (Fig. 6C). Together indomethacin and catalasetotally suppressed the DMPOsulfur trioxide adduct signal (Fig. 6 0 ) . The four-line spectrum obtained from the heat-denatured control (Fig. 6 E ) is the

Free Radical Metabolism of Bisulfite

5054

The mechanism for oxidation of (bi)sulfite must involve one-electron oxidationsince the intermediate $0: is produced. This is contrary to O'Brien and Rahimtula, who reported the oxidation of (bi)sulfite by prostaglandin synthase to be a one-step, two-electron oxidation that forms sulfate directly (36).The initial step in the mechanism is the reaction of the hydroperoxide, prostaglandin G2 (formed from arachidonic acidand oxygen by cyclo-oxygenase),or another hydroperoxide, 15-hydroperoxyarachidonicacid or H202,with the peroxidase to produce the corresponding dcohol reduction product and a compound I-type form of the enzyme or its equivalent. The compound I-type intermediate is presumably two oxidizing equivalents above the resting state of the enzyme. The protein is then reduced back to theresting state by oxidizing (bi)sulfite. Once the $0, radical is formed, it reacts rapidly with oxygen via two reactions (5,6, 18).

so; + 0 2 -+ so, + 0; so,-+ 02 + so;

(6) (7)

6; + SOT

+ 2H'

+ H20, so, + so; + so,-+ so;; -+

$0:

(8)

(9)

In addition to reacting with oxygen, the $0: radical can undergo disproportionation (Equation 10) and/or dimerization (Equation 12) (38).

so3 + so; SO3 + Hz0 + SO; + 2H+ 2s0, + s,o; 230;

+

(10) (11)

(12)

These reactions terminate the free radical chain reaction and do not reduce molecular oxygen. In addition to these known chemical reactions (Equations 6-12), a variety of biochemical factors must be considered in the interpretation of the oxygen consumption results of Fig. 4. The HnOz-initiated oxygen consumption of Fig. 4E is certainly due to the enzymatic formation of SO;, but the importance of propagation reactions (Equations 8 and 9) in the formation of $0; is unknown. The arachidonic acid-initiated (bi)sulfate oxidation is even more complicated (Fig. 4B), because many electron donors are known to increase prostaglandin formation by ram seminal vesicle microsomes, whichalso increase oxygen consumption (39). In this case (bi)sulfite actually inhibits prostaglandin formation (Table I). Since DMPO alone has no effect on prostaglandin biosynthesis, but reverses the inhibition by (bi)sulfite, reactions of the $0, radical are implicated in the (bifsulfite inhibition of prostaglandin formation. Therefore, the increase in oxygenconsumption seen in the presence of bisulfite (compare Fig. 4, B and A ) must be due to the reactions of SO, with oxygen (Equations 6 and 7), because the bioxygenation of arachidonic acid is inhibited by (bi)sulfite (Table I). Whether competition for oxygen bythe SO, radical, an inactivating reaction of SOY with prostaglandin synthase, or some other reaction of SO; is responsible for the inhibition of prostaglandin biosynthesis is unknown. There is apparently enough sulfite oxidase in the lung to oxidize reasonable concentrations of (bi)sulfite directly to sulfate with no free radical intermediate (40). However, it should be noted that in studies of dogs which have inhaled SOz, SOzdoes reach the blood plasma (41). Once in the blood

Control + 1mM Nad303 + 270 mM DMPO + 1m~ Na2S03and 270 mM DMPO Boiled control

nmol/min/mgprotein 8.0 f 1.3" 3.2 f 1.0" 7.4 f 1.2'

9.5 k 1.4' 0.2 k 0.01'

"N=3. bN=2.

plasma, bisulfite is converted to S-sulfocysteine compounds including protein S-sulfonates (3, 42), which clearly indicates that (bi)sulfite is not oxidizedby sulfite oxidase under all circumstances. An equilibrium exists leading to a mechanism for bisulfite storage and distribution (3). RSSR

+ HSOB

RSSOB + R'SH

The S-sulfocysteine compounds can also be reduced back to bisulfite in the presence of NADPH or cleaved to form thiosulfate, which can be reduced to (bi)sulfite by the enzyme rhodanase or by thiosulfate reductase and reduced glutathione in the liver (2). It seems that the prostaglandin synthase-catalyzed oxidation may bea possible pathway for the conversion of (bi)sulfite to sulfate for elimination, and that some potentially toxic reactions could be caused by(bi)sulfite oxidation through this pathway as a result of reactions between the SO; radical and many biologically important molecules. In particular, prostaglandin synthase- and other peroxidase-catalyzed oxidations of (bi)sulfite to the $07 radical may be responsible for some of the injuries which result to humans born without sulfite oxidase (43), the enzyme which normally detoxifies(bi)sulfite via a two-election oxidation of (bi)sulfite to SOT (44). Acknowledgment-We wish to thank Dr. Charles E. Carter for suggesting the experiments with the guinea pig lung microsomes. REFERENCES 1. Rall, D. P. (1974)Enuiron. Health Perspect. 8,97-121 2. Petering, D. H., and Shih, N. T. (1975)Enuiron. Res. 9,55-65 3. Petering, D. H. (1977) in Biochemical Effects of Environment Pollutants (Lee, S. D., ed) pp. 293-306,Ann Arbor Science Publishers, Ann Arbor, MI 4. Rajagopalan, K. V., and Johnson, J. L.(1977) in Biochemical Effects of Environmental Pollutants (Lee, S. D., ed) pp. 307314, Ann Arbor Science Publishers, Ann Arbor, MI 5. Fridovich, I., and Handler, P. (1960) J . Biol. Chem. 235, 18351838 6. Hayon, E., Treinin, A., and W a f ,J. (1972) J . Am. Chem. SOC.94, 47-57 7. Klebanoff, S. J. (1961)Biochim. Biophys. Acta 48,93-103 8. Yang, S. F. (1970) Biochemistry 9,5008-5014 9. Yang, S. F. (1973) Enuiron. Res. 6,395-402 10. Peiser, G. D., and Yang, S. F. (1979) J. Agric. Food Chem. 27, 446-449 11. Flockhart, B. D., Ivin, K. J., Pink, R. C., and Sharma,B. D. (1971) Chem. Commun. 339-340 12. Norman, R. 0. C., and Storey, P. M.(1971) J. Chem. Soc. fB) 1009-1013 13. Hayatsu, H. (1976)Prog. Nucleic Acid Res. Mol. Biol.16,75-124 14. Kaplan, D., McJilton, C., and Luchtel, D. (1975)Arch. Enuiron. Health. 30,507-509 15. Inouye, B., Ikeda, M., Ishida, T., Ogata, M., Akiyama, J., and


These reactions with oxygen account for the rapid oxygen uptake by the full system. Indomethacin inhibits the oxygen uptake since it blocks the cyclo-oxygenasesystem. Once initiated, oxygen consumption can be sustained by either of two propagation reactions (6, 37).

TABLEI Effectof NazS03 and DMPO on prostaglandin biosynthesis Ram seminal vesiclemicrosomes (1 mg/ml) were incubated at 37 "C for 5 min with [l-'4C]arachidoNc acid (200 p ~ in) 1.0 ml of 100 m~ sodium borate/boric acid buffer, pH 7.9, containing 1m~ diethylenetriamine pentaacetic acid. Prostaglandins were estimated by high pressure liquid chromatographic analysis of its acidic organic extract as described under "Experimental Procedures." Values are mean f S.D.

Free Metabolism Radical

5055

31. Chignell, C. F., Kalyanaraman, B., Si, R. H., and Mason, R. P. (1981) Photochem. Photobiol. 34, 147-156 32. Kirino, Y., Ohkuma, T., and Kwan, T. (1981) Chem. Pharm. Bull. 29.29-34 33. Carpenter, M. P. (1981) Fed. Proc. 40, 189-194 34. Marnett, L. J., and Reed, G . A. (1979) Biochemistry 18, 29232929 35. Floyd, R. A., and Wiseman, B. B. (1979) Biochim. Biophys. Acta 586, 196-207 36. O’Brien, P. J., andRahimtula, A. D. (1980) in Biochemistry, Biophysics and Regulation of Cytochrome P-450 (Gustafsson, J. A., Carlstedt-Duke, J . , Mode, A., and Rafter, J., eds) pp. 273281, Elsevier/North-Holland, Biomedical Press, New York 37. Hegg, D. A., and Hobbs, P. V. (1978) Atmos. Enuiron. 12, 241253 38. Eriksen, T. E. (1974) J. Chem. Soc. Faraday Trans. Z 70, 208215 39. Egan, R.W., Paxton, J., and Kuehl, F. A., Jr. (1976) J. Biol. Chem. 251,7329-7335 40. Cohen, H. J., Drew, R. T., Johnson, J. L., and Rajagopalan, K. V. (1973) Proc. Natl. Acad.Sei. U. S. A. 70,3655-3659 41. Yokoyama, E., Yoder, R. E., and Frank, N. R. (1971) Arch. Enuiron. Health 22,389-395 42. Shapiro, R. (1977) Mutat. Res. 39, 149-176 43. Johnson, J. L., Waud, W. R., Rajagopalan, K.V., Duran, M., Beemer, F. A., and Wadman, S. K. (1980) Proc. Natl. Acad. Sei. U. S. A. 77,3715-3719 44. Cohen, H. J., and Fridovich, I. (1971) J. Biol. Chem. 246,359-366 45. Eling, T., Warnock, R., Dick, D., and Tainer,B. (1980) Prost. and Med. 5, 345-355


Utsumi, K. (1978) Toxicol. Appl. Pharmacol. 46.29-38 16. Hayatsu, H., and Miller, R. C., Jr. (1972) Biochem. Biophys. Res. Commun. 46,120-124 17. Lasker, J. M., Sivarajah, K., Mason, R.P., Kalyanaraman, B., Abou-Donia, M. B., and Eling, T. E. (1981) J. Biol. Chem. 256, 7764-7767 18. Mottley, C., and Mason, R. P. (1982) Mol. Pharmacol., in press 19. Sivarajah, K., Lasker, J . M., and Eling, T. E. (1981) Cancer Res. 41, 1834-1839 20. Andersen, I., Lundqvist, G. R., Jensen, P. L., and Proctor, D. F. (1974) Arch. Environ. Health 28, 31-39 21. Funk, M. O., Isaac, R., and Porter, N. A. (1976) Lipids 11, 113117 22. Sivarajah, K., Anderson, M. W., and Eling, T. E. (1978) Life Sei. 23,2571-2578 23. Lowry, 0. H., Rosebrough, N. J., F m , A. L., and Randall, R. J. (1951) J. Biol. Chem. 193,265-275 24. Janzen, E. G. (1980) in Free Radicals in Biology (Pryor, W. A., ed) Vol. IV, pp. 115-154, Academic Press, New York 25. Kalyanaraman, B. K. (1982) Rev. Biochem. Toxicol. 4, in press 26. Lakatos, B., Turcsanyi, B., and Tudos, F. (1971) Acta Chim. Acad. Sei.Hung. 70, 225-234 27. Mulvey, D., and Water, W. A. (1974) J. Chem. SOC.Perkin Trans. II, 772-773 28. Chignell, C. F., Kalyanaraman, B., Mason, R. P., and Sik, R. H. (1980) Photochem. Photobwl. 32,563-572 29. Lambeth, D. O., and Palmer, G. (1973) J. Biol. Chem. 248.60956103 30. Kominami, S., Rokushima, S., and Fiatano, H. (1976) Int. J. Radiat. Biol.30, 525-534

of Bisulfite