Hoppe-Seyler's Z. Physiol. Chem. Bd. 356, S. 727 - 737, Juni 1975

Problems Concerning the Biochemical Action of Superoxide Dismutase (Erythrocuprein) Wulf Paschen and Ulrich Weser

(Received 2 December 1974) Dedicated to Prof. Dr. Dr. G. Weitzelon the occasion of his 60th birthday Summary: The decay of the tetraperoxochromate(V) complex (Cr083e) was examined to study the substrate specificity of erythrocuprein (superoxide dismutase). The decay of CrO 8 e proved rather complex in aqueous solutions. Apart from the two known oxygen species 02 and singlet oxygen ( 1 A g 0 2 ), H 2 0 2 and probably OH* radicals were formed. No unequivocal evidence for the appearance of Superoxide was obtained. The possible electron transfer from Cr s ® to Fe3® (cytochrome c) was also discussed. In Tris buffer, pH 7.8, there were absolutely no signs of superoxide or OH* radical formation. In fact, pulse radiolysis measurements employing a homogeneous OH" source demonstrated that the Tris and OH* radicals react with each other. One mol of H 2 0 2 was generated from 1 mol of Cr083e in Tris buffer. By contrast, only 0.5 mol H 2 O 2 could be determined when the Cr083e decay was carried out in 2-[4-(2-hydroxyethyl)-lpiperazinyl]-ethanesulfonic acid (HEPES) buffer,

pH 7.8. The phenomenon of reducing oxidized cytochrome c could not fully be assigned to a superoxide-mediated reduction, since erythrocuprein was unable to inhibit this cytochrome c reduction efficiently. The energetic oxygen species ( 1 A g 0 2 , OH* etc.) appearing during the Cr083e decay gave rise to a clearly detectable chemiluminescence. In this system, erythrocuprein was very active regardless of which buffer was used. Even in the absence of a chemiluminescent mediating agent, which might have interferred with the enzyme, erythrocuprein proved capable of inhibiting the Cr083einduced chemiluminescence in a rather specific way. No such specificity was seen in the presence of low molecular weight Cu-chelates including Cu(Tyr)2, Cu(Lys)2 and Cu(His)2. The ability to suppress chemiluminescence was approximately 3 orders of magnitude less pronounced than that of the native enzyme. It is presumed that erythrocuprein reacts with oxygen species other than the Superoxide radical.

Address: Prof. Dr. U. Weser, Anorganisch-Biochemische Arbeitsgruppe, Physiologisch-chemisches Institut der Universität, D-74 Tübingen, Hoppe-Seyler-Str. 1. Enzymes Catalase, hydrogen-peroxide:hydrogen-peroxide oxidoreductase (EC 1.11.1.6.) Peroxidase, donor:hydrogen-peroxide oxidoreductase (EC 1.11.1.7.) Superoxide dismutase (erythrocuprein), Superoxide:Superoxide oxidoreductase (EC 1.15.1.1.) Xanthine oxidase, xanthine:oxygen oxidoreductase (EC 1.2.3.2.) A bbreviations: Me2SO = Dimethylsulfoxide; HEPES = 2-[4-(2-hydroxyethyl)-l-piperazinyllethanesulfonic acid; Tris = Tris(hydroxymethyl)aminomethane;Nitro-BT = Nitroblue tetrazolium.

Brought to you by | Purdue University Li Authenticated Download Date | 5/20/15 9:56 PM

728

W. Paschen and U. Weser

Bd. 356(1975)

Parallel conclusions were drawn concerning the portion normally bound with the flavin residue of decay of both the complex of reduced flavin and many flavoenzynW *'. The different mode of 3e molecular oxygen and the decay of CrO8 . It has decay of CrO8 was dictated by the buffer solubeen suggested that the mode of decay of a flavin- tions, i.e. Tris or HEPES. O2 complex is dictated by the particular protein Zur Problematik der biochemischen Reaktivität der Superoxid-Dismutase (Erythrocuprein) Zusammenfassung: Um einen Einblick in die Substratspezifttät von Erythrocuprein (SuperoxidDismutase) zu erhalten, wurde der Zerfall des Tetraperoxochromat(V)-Komplexes (CrO83e) untersucht. Der Zerfall verläuft in wäßriger Lösung sehr komplex. Dabei werden verschiedene energiereiche Sauerstoffspezies gebildet: H 2 0 2 , Singulet-Sauerstoff (1 %£>2) and OH-Radikale.

Cytochrom cox übertragen werden, wird diskutiert.

Superoxide dismutase121, originally designated as erythrocuprein^, is known to react with the Superoxide radical at a diffusion-controlled rate. 9 A second-order rate constant of 2.4 ± 2 *!"1 s~ * for this enzyme-catalyzed reaction was measured by several groups employing pulse

radiolysisf4'7'9!. There is no doubt that erythrocuprein reacts with Superoxide, but the specificity of this reaction remains obscure. It has been successfully demonstrated that Cu(Tyr)2^ was active in exactly the same way as the native Cu, Zn-protein. Surprisingly, the Superoxide dis-

Bei dem CrOs39 Zerfall kommt es zu einer starken Chemilumineszenz. In Gegenwart von HEPESPuffer kann diese Chemilumineszenz sogar ohne zusätzlichen Scintillator (Luminol) gemessen werden. Unkontrollierte Phosphoreszenzreaktionen des Luminols wurden dadurch ausgeschlossen. 3e Der Zerfallsmodus von Cr08 in wäßriger Lösung Erythrocuprein hemmt in beiden Fällen diese hängt von der Wahl des Puffers ab. Es entstehen Chemilumineszenz spezifisch. Verglichen mit unterschiedliche Zerfallsprodukte in 2-[4-(2-Hydem nativen Enzym ist die Aktivität der niederdroxyäthyl)piperazinyl]äthansulfonsäure-Puffer molekularen Cu-Komplexe Cu(Tyr)2, Cu(Lys)2 (HEPES) oder in Tris-Puffer. Beim Zerfall in Tris- und Cu(His)2 um etwa 3 Größenordnungen gePuffer gibt es keine Hinweise dafür, daß Of- und ringer. Im Vergleich dazu ist die Superoxid-DisOH-Radikale entstehen. Es wird durch Pulsradio- mutase-Aktivität von Erythrocuprein fast idenlysemessungen mit einer reinen OH-Radikalquelle tisch mit'der Superoxid-Dismutase-Aktivität diegezeigt, daß Tris mit OH-Radikalen reagiert. ser Cu-Komplexe. Es wird daher geschlossen, daß beim CrOg39 Zerfall Erythrocuprein nicht unbeDie quantitativen Untersuchungen der verschiedingt mit O2ö-Radikalen sondern mit anderen denen Zerfallsprodukte ergaben, daß in HEPES3e Sauerstoff-Spezies reagiert. Puffer aus l mol Cr08 0.5 mol H 2 0 2 , in TrisDer Zerfallsmodus von CrO83e wurde mit dem Puffer dagegen l mol H 2 0 2 entstehen. Ob eventuell auch O2e-Radikale auftreten, wurde mit Zerfall der verschiedenen Komplexe von redudem Nitrotetrazoliumblau-Test und mit oxidierziertem Flavin und molekularem Sauerstoff vertem Cytochrom c untersucht. Während der Zerglichen. Der Zerfallsmodus des Flavin-Sauerstofffallsreaktion von CrO83e in Tris-Puffer konnten Komplexes scheint durch den Proteinrest bekeine 0^-Radikale nachgewiesen werden. Auch stimmt zu werden, der Modus des Cr083e Zerin HEPES-Puffer verlief der Nitrotetrazoliumfalls durch die Wahl des Puffers. In beiden Fällen blau-Test negativ, oxidiertes Cytochrom c wurde entstehen verschiedene energiereiche Sauerstoffjedoch teilweise reduziert. Erythrocuprein konnte Spezies. Cr08e erwies sich daher als brauchbares die Cytochrom-c-Reduktion nur teilweise hemund doch problematisches Modell zur Untersumen. Inwieweit diese Reduktion durch 2 oder chung der Reaktivität von Erythrocuprein mit andere Sauerstoffradikale zustande kommt, oder energiereichen Sauerstoff-Spezies, die bei vielen 50 3e Enzymreaktionen entstehen können. ob Elektronen direkt vom Cr auf das Fe des

Brought to you by | Purdue University Li Authenticated Download Date | 5/20/15 9:56 PM

Bd. 356 (1975)

Problems Concerning Superoxide Dismutase

mutase activity of hydrated Cu2® at neutral pH was even twice as higlJ9'. No direct evidence for the presence of Superoxide in intact biological tissues is known, although cytochromec reduction was inhibited by erythiocuprein^8'. A specific inhibition of Superoxide dismutase in biological systems, with the consequence of a possible rise of detectable Superoxide, was also not found. In several studies^10"14!, the main physiological function of erythrocuprein was proposed to be the scavenging of highly reactive species of singlet oxygen. The removal of these energetic singlet oxygen species must be guaranteed to avoid the breakdown of vital metabolic processes. In general, it is known that singlet oxygen can be generated from superoxidei15! during the dismutation of this radical. Peters et al.l16! presented evidence that 1Ag02 was exclusively formed from Cr083e at neutral pH. We used this transition metal ion peroxo complex as a convenient substrate for erythrocuprein'14'; and indeed, erythrocuprein proved a most powerful metalloprotein, able to suppress the luminol-mediated chemiluminescence of ι^02· Although the Superoxide dismutase activity of low molecular weight copper chelates is virtually identical to that of erythrocuprein, no such reactivity was seen when these complexes were examined in the CrO83e assay^14'. The ability of the low molecular weight Cu-chelates to quench the luminol-mediated chemiluminescence was remarkably low; several orders of magnitude less than that of native erythrocuprein.

729

established singlet oxygen source^15!. The central question, whether or not erythrocuprein reacts with other oxygen species capable of producing singlet oxygen, is still open. In this context a comprehensive study using Cr083e as a singletoxygen-producing system in the absence of a chemiluminescent mediator such as luminol was performed. Cr08 θ proved best suited, since its structure has been elucidated by X-ray diffraction measurements^21'. Some interesting results on the CrO8 θ decay in aqueous solutions have beenpublishedl14'22', indicating substantial discrepancies regarding the presence or absence of Superoxide. Therefore, a comprehensive investigation was carried out by examining the decay of this transition metal peroxo complex in more detail, to focus attention on the problems of the biochemical action of Superoxide dismutase with different oxygen species. Experimental Section Reagents Crystalline potassium tetraperoxochromate (K CrO ) was prepared from alkaline potassium chromate and icecold hydrogen peroxide( 23 J. Cu(His>2, Cu(Lys)2 and Cu(Tyr)2 were made according tol9!. Erythrocuprein was isolated from freshly prepared bovine erythrocytesi24'. Only reagents of analytical quality were used. Deionized water was distilled over quartz. The purchased chemicals were from the companies listed below: Boehringer Mannheim GmbH: equine cytochrome c; horse radish peroxidase (EC 1.11.1.7), grade I; crystalline catalase (EC 1.11.1.6), analytical grade; xanthine oxidase (EC 1.2.3.2) analytical grade; xanthine. FerakChemie, Berlin: leuco crystal violet. K & K laboratories, Hollywood: potassium Superoxide. Merck, Darmstadt: diphenyl carbazide, f-butyl alcohol and dimethylsulfoxide. Roth, Karlsruhe: luminol, reagent grade, nitroblue tetrazolium (nitro-BT). Serva, Heidelberg: 2-[4(2-hydroxyethyl)-l-piperazinyl]-ethane sulfonic acid (HEPES), reagent grade.

In the light of both the exclusive formation of 1 A g 0 2 using Cr08e[161 and the indirect assay of singlet oxygen by the luminol-mediated chemiluminescencel14], the preliminary conclusion was drawn that erythrocuprein reacts directly with iA g 0 2 . A final decision regarding the direct reaction of 1ΔΚ02witn erythrocuprein cannot be made at the moment, since all methods employed Hydrogen peroxide assay are based on secondary reactions of singlet Any H2O2 generated during the CrO83e decay was deteroxygen. Thus, all statements^14'17"20' rejecting mined using leuco crystal violet and horse radish peror considering the possibility of a direct interoxidasel25!. A millimolar absorption coefficient of the action of 1 A g 0 2 and erythrocuprein should be oxidized crystal violet €595 = 75 was used' 25 '. H2O2 postponed until improved experimental techniwas measured after the CrOg30 decay was completed. ques for studying this question are available. The assay mixture was composed of 155μΜ leuco cryNevertheless, it is commonly accepted that erystal violet and 0.14juM horse radish peroxidase in IM throcuprein prevents the deleterious action of acetate buffer. The total volume was 1.2 ml. Chromate did not disturb this assay. The temperature was 12 °C. singlet oxygen by scavenging Superoxide, an

Brought to you by | Purdue University Lib Authenticated Download Date | 5/20/15 9:56 PM

730

W. Paschen and U. Weser

Bd. 356 (1975)

Determination of chromate buffer, pH 7.8. In this case luminol was required. The reaction was started using 50 μ/ of a concentrated Two different methods were employed to quantitate 0 3e K 3CrO8 solution, 0.82mM, dissolved in 0.1N NaOH. CrO^ formation during the CrOg decay. DiphenylChemiluminescence was detected in a Packard scintillacarbazide reacts with chromatei 2 ^] to form a red-violet tion counter model 2002. The adjustment was exactly carbazonel 27 ) which has an absorption maximum at the same as for 3H counting. The coincidence circuit was 546 nm (e 546 = 28 400). The assay volume was 1.2m/. The reaction was performed in 0.1N HC1 and in the pres- turned on. The first readings were taken 20 sec after the addition of K3CrO8. All measurements were repeated at ence of 0.39mM diphenylcarbazide. The alternative least four times. The maximal deviation between parallel method was the direct recording of CrO42e at pH 7.8 experiments was less than ±5%. using the absorption maximum at 375 nm. 6375 =4 800 was determined from a calibration curve of different a) Counting of CrO83e decay in Tris buffer (partially amounts of weighed CrO42e probes. integrated counting): The measurement was performed until the CrO83e decay was completed, usually after Reduction of cytochrome c 6 min. Every 15 sec, counting was carried out for a period of 12 sec. At the end of the experiment, all countThe rate of reduction of oxidized cytochrome c was ing rates were summarized. Polypropylene counting vials monitored at 550 nm. Readings were taken in the preswere employed. The assay volume was 2.65 ml and was ence of both HEPES and Tris buffers, each 83mM, pH composed of 95mM Tris buffer, pH 7.8; 0.2mM luminol, 7.8. Usually 32μΜ cytochrome c and 280μΜ K3Cr08 0.42μΜ catalase and 15μΜ K3CrO8. were present. The reaction was started with Cr083e (dissolved in 0.1 m/ 0.1N NaOH). b) Counting of CrO83e decay in HEPES buffer (fully integrated counting): Without luminol, the counting Nitro blue tetrazolium test yield was markedly diminished. Thus, counting was The most sensitive assay for Superoxide proved to be the continued for at least 5 min. The assay mixture was com28 e posed of 220μΜ K3CrO8; 95mM HEPES buffer, pH 7.8. nitro-BT assayl ). In the presence of O2' , nitro-BT is The total volume was 3.15 ml. reduced to the formazan color, which can be measured at 560 nm. A calibration curve using potassium superoxide was linear from 0 -125μΜ Ο2Θ. Estimation of KO2 Results was carried out in an indirect way by measuring the potasAccording to Riesenfeld[23], the transition persium concentration of analytical grade KO2 by atomic 3e absorption spectroscopyi29!. The results were confirmed oxo complex Cr08 decomposes into chromate, molecular oxygen and hydroxyl ions. It has been by the peroxide assay following the spontaneous dismutation of KO2 in water according to demonstrated that all or portions of this mole2 Ό2Θ -* Ο22θ + O2. cular oxygen is initially generated as singlet oxygen ( 1 A g 0 2 ) ll6 l Cr083e has proved a most conPulse radiolysis venient substrate for erythrocuprein^ 14 ' 22 '. HowPulse radiolysis measurements were performed using the ever, the molecular mechanisms are still not fully same equipment described in the preceeding study! 91. understood. The discrepancies regarding the oc•OH radicals were produced by irradiating aqueous solucurrence of oxygen species other than the singlet tions saturated with N 2 O with 1.81-MeV electrons. The types became apparent. In this context we exaconcentration of OH radicals was determined to be mined the Cr083e decay in more detail. 24μΜ. Recording of the spectra was carried out by measuring the absorbance 2 ms after the pulse. Data First of all it was shown that CrO83e forms exwere collected stepwise every 5 -10 nm from 210 to clusively CrO42e in aqueous solutions at pH 7.8, 325 nm. regardless of which buffer was employed. The Measurement of chemiluminescence decay was completed after approximately 8 min 3e (Fig. 1). The search for a greater variety of oxygen During the decay of CrOg , singlet oxygen, and probubly, as we shall see later, · OH are being gradually species yielded H 2 0 2 . There was, however, a dis16 formed^ l They can be detected by measuring the tinct difference in the amount of peroxide forchemiluminescence caused either directly or via secmation (Table 1), depending on the buffer. ondary reactions of these oxygen species. If CrOg3e was In HEPES buffer, only one half mole of H 2 0 2 allowed to decompose in 95mM HEPES buffer, pH 7.8, was generated per mole Cr083e. Provided that no chemiluminescent probe had to be added. ChemiOH-radicals were being formed during the CrO83e luminescence was not strong enough to be detected decay, they would be in a position to react with when the CrOg3e decay was performed in 95mM Tris

Brought to you by | Purdue University Li Authenticated Download Date | 5/20/15 9:56 PM

Bd. 356(1975)

731

Problems Concerning Superoxide Dismutase

peroxide according to the Haber-Weiss cycle1301. The consequence would be a lower yield of H 2 0 2 . However, in the presence of a suitable OH* scavenger, much more H 2 0 2 should be detectable. Indeed, the addition of r-butanol, a powerful OH* scavenger14'61, doubled the H 2 0 2 yield. This result gave rise to the assumption that Tris itself is an efficient OH" scavenger. Pulse radiolytic measurements using exclusively OH-radicals and Tris

buffer confirmed the assumption, as a reaction of Tris and OH* could be clearly detected at 290 nm. Neither Tris nor the OH-radical absorbs light at this wavelength (Fig. 2).

3.6-

3M-

r

3.2-

3.0-

210

Fig. 1. K 3 Cr0 8 decay in HEPES buffer. The decay was performed at 12 °C in 1.05 ml 95 mM HEPES buffer, pH 7.8. The reaction was started with 50 μΐ of a 6.5mM K3CrO8 solution in 0.1N NaOH. Chromate was recorded at 375 nm using a quartz cell with a 1-cm light path. Table 1. Hydrogen peroxide formation. The K3CrO8 decay was performed in HEPES and Tris buffers at 12 °C in a volume of 1.05 ml. The concentration of K3CrO8 in 95mM HEPES or Tris buffer was 0.12mM. After the complete decay of K3CrO8 in the presence and absence of f-butyl alcohol, an OH-radical scavenger, samples were taken and the concentration of the hydrogen peroxide measured using leuco crystal violet as a cosubstrate and horse radish peroxidase. The resulting crystal violet formation was recorded at 596 nm.

250

λ I ml —*-

300

Fig. 2. Reaction of Tris and OH-radicals: ( ) spectrum of OH-radicals, ( ) spectrum of the reaction product Tris-OH, ( ) absorption of Tris. OH-radicals were produced by irradiating an aqueous solution saturated with N 2 O with a 40-ns pulse of 1.81-MeV electrons. The OH concentration was 24μΜ, the concentration of Tris was ImM. The pH was adjusted with NaOH to 10 throughout. The recording was performed in quartz cells with 2-cm light paths 2 ms after each pulse.

f-Butyl alcohol [mol//]

0.8

Tris

HEPES 10 4 x[H 2 0 2 ] [mol//]

1.04 1.00

0.51 1.01

Brought to you by | Purdue University Lib Authenticated Download Date | 5/20/15 9:56 PM

732

W. Paschen and U. Weser

Bd. 356 (1975)

Superoxide or no Superoxide During the decay of Cr083e in phosphate buffer, oxidized cytochrome c was progressively reducedl22!. Due to the complete inhibition of cytochrome c reduction by erythrocuprein (superoxide dismutase)^22', the reducing factor was assumed to be 02Θ. However, we were unable to see any cytochrome c reduction at all when the CrO83e decay was carried out in Tris buffer. This observation supports the above conclusion, that Tris imposes a mode of Cr083e decay substantially different from those observed in phosphate or HEPES buffer. The possibility that 02Θ may react with Tris was excluded by control experiments using established Ο2Θ sources in Tris buffer. Both K02 and enzymically produced 02Θ in the xanthine, xanthine-oxidase system produced a strong reduction of cytochrome c which was fully inhibited by erythrocuprein. The CrO83e decay in HEPES buffer was for some reason quite surprising. At 25 °C no initial cytochrome c reduction was measured, and the conclusion could be drawn that 02Θ and other reducing factors were absent. Upon lowering the temperature to 12 °C, some but not all cytochrome c was reduced in the first few seconds. This was followed by a quick reoxidation (Fig.3) of the cytochrome c. In the presence of erythrocuprein, no such reduction was seen. Surprisingly, the complete reduction of all oxidized cytochrome c was observed after a lag phase of 2 min. In contrast to the cytochrome c reduction, this reduction could not be inhibited by erythrocuprein. Even the presence of catalase had no influence on the above reactions. This initial reduction of cytochrome c, which was inhibited by erythrocuprein, could suggest that 02Θ is the reducing agent. However, no positive reaction regarding the presence of Superoxide was measured employing the nitro-BT assayi28). Cr083e in concentrations up to 1 ΟΟΟμΜ gave no measurable positive reaction. Cr083e and Cr042e did not disturb this assay, since the addition of 1 - ΙΟμΜ Κ02 resulted in a dramatic and clearly detectable formazan color. Unfortunately, the direct characterization of the Superoxide radical by EPR spectroscopy failed due to the strong interference of Cr 5 ®.

Fig. 3. Reduction of cytochrome c in HEPES buffer: The reduction of cytochrome c was monitored at 550 nm. The assay mixture was composed of 32μΜ cytochrome c and 280μΜ K 3 CrO 8 in 83mM HEPES buffer, pH 7.8 in a total volume of 1.2 ml. 14μΜ erythrocuprein added ( ), erythrocuprein omitted ( ).

At the moment, there is no unequivocal evidence that Ο 2 Θ is being generated during the Cr083e decay. Formation of Ο2Θ via OH" would be both attractive and plausible, but a final proof is lacking. The reduction of cytochrome c could have occurred by a direct electron transfer from Cr5e to the Fe3® of cytochrome c. The inability of erythrocuprein to inhibit the second cytochrome c reduction of Fig.3 could be taken as evidence for this. CrO/e decay monitored by chemiluminescence High-energy singlet oxygen species and/or OH* were expected to produce a clearly detectable chemiluminescence. When Cr083e was allowed to decompose in HEPES buffer, no chemiluminescent mediating reagent had to be added (Fig. 4). In the presence of erythrocuprein, the chemiluminescence was dramatically diminished (Table 2). However, the low molecular weight Cu-chelates displayed a markedly lower reactivity. The difference was almost three orders of

Brought to you by | Purdue University Li Authenticated Download Date | 5/20/15 9:56 PM

Bd. 356 (1975)

Table 2. Inhibition of K3CrOg-mediated chemiluminescence in HEPES buffer. The formation of highly energetic Required equivalent of Cu2® chelate oxygen species during the K3CrOg chelated copper to yield decay was detected by measuring 50% inhibition of 1 A g O 2 the chemiluminescence in a liquid induced chemiluminescence scintillation counter. The assay mixCu2e [nrnol//) ture was composed of 220μΜ K 3 Cr0 8) 95mM HEPES buffer, erythrocuprein 4.1 pH 7.8. The total volume was 890 Cu(His)2 3.15 ml. The reaction was started 1100 Cu(Lys)2 with 100 μ/ of a 6.7mM solution Cu(Tyr)2 1800 of K3Cr08 in 0.1N NaOH. The total counts were compared with those obtained after the addition of erythrocuprein or the Cu-chelates

0.3-

733

Problems Concerning Superoxide Dismutase

ratio Cu20 (erythrocuprein) : Cu 2 ® chelate

_ _ 1 : 217 1 :268 1 :428

magnitude. This result is very significant, especially in the light of the identical Superoxide dismutase activities of Cu(Tyr)2 and erythrocuprein£u2e[9] jt may ke concluded that during this CrO83e decay, erythrocuprein reacts with some oxygen species other than the Superoxide radical. The same picture was obtained when the Cr083e decay was performed in Tris buffer and in the presence of luminol. The reactivity of Cu2® using low molecular weight Cu2e chelates was only 0.5% ofthat of erythrocuprein Cu2e (Table 3). As we have shown above, there was absolutely no sign of the presence of Superoxide during the decay of Cr083e in Tris buffer. Furthermore, the yield of OH*, the possible powerful source for chemiluminescence, should be substantially diminished by the reaction of Tris with OH*. Thus, luminol had to be added to improve the net counting rate. Discussion

The reader may ask why we used a transition metal peroxo complex rather than a biological system for studying the substrate specificity of erythrocuprein. Our knowledge on the structure Fig.4. Time courses of K^CrOg decay and chemilumines- of most biochemically relevant compounds is very limited, while the molecular architecture of cence. Cr083e is known precisely1211- From X-ray difThe curves include both the chromate formation and fraction measurements, it was deduced that the the K3CrOg mediated chemiluminescence. The decomchromium ion is present in the oxidation state position of K3CrO8 was carried out in 95mM HEPES + 5 and tetrahedrally surrounded by four identibuffer, pH 7.8, containing 0.3mM K3CrO8. cal 022Θ groups. The 022θ distance is 1.846 A ( ) K3Cr08 decay monitored at 375 nm; from the Cr5®. No indication whatsoever was ( ) K3CrOg mediated chemiluminescence.

Brought to you by | Purdue University Lib Authenticated Download Date | 5/20/15 9:56 PM

734

W. Paschen and U. Weser

Bd. 356 (1975)

Table 3. Inhibition of K^CrOg-mediated chemiluminescence in Tris buffer. The rather weak chemiluminesCu 2 ® chelate Required equivalent of ratio Cu 2 ® cence observed during the chelated copper to yield (erythrocuprein) : Cu2® decay in Tris buffer was markedly 1 chelate 50% inhibition of Δ§Ο2 improved by adding luminol. The induced chemiluminescence assay volume was 2.65 ml and was 2 Cu ®[nmol//] composed of 95mM Tris buffer, pH _ 7.8, 0.2mM luminol, 0.42μΜ catalerythrocuprein 16.4 ase, and 15μΜ K3CrO8. The reac3000 1 : 183 Cu(His)2 tion was started with 50 μΐ of 4500 1 : 274 Cu(Lys)2 0.82mM K3CrO8 solution disCu(Tyr)2 7800 1 :487 solved in 0.1 N NaOH. Counting was carried out every 15 sec for a period of 12 sec. At the end of the experiment all counting rates were summarized. These total counts were compared with those obtained after the addition of erythrocuprein or the Cu-chelates.

seen for the presence of Superoxide in crystalline K3CrO8. There is not even unequivocal evidence that Superoxide is generated during the decay of Cr083e in aqueous solutions. The possibility of a Superoxide formation via a secondary reaction induced by OH" might be discussed (Fig. 5). Unfortunately, EPR measurements during the aqueous CrO83e decay gave no conclusive results, due to the overlapping chromium signal. Nevertheless, the fruitless attempts to observe a formazan color formation using the nitro-BT assay as well as the inability to suppress the cytochrome c reduction after 2 min in HEPES buffer are not convincing arguments for the presence of Ο 2 Θ . It must be emphasized that in Tris buffer, both assays for Superoxide proved absolutely negative, and the yield of free OH" was minimized by its reaction with Tris. As it is known that elevated H 2 O 2 concentrations inactivate Superoxide dismutase'10^, it was ascertained that none of the

3-

oxygen species arising during the CrO83e decay affected the enzyme. When the decay was completed, erythrocuprein was separated by gel filtration and no detectable loss of Superoxide dismutase activity was seen. As has been shown, the decay of CrO83e in aqueous solutions proved far from simple. Apart from the complex nature of this CrO83e decay, erythrocuprein displayed a distinct and specific reactivity. From the present study, it may be concluded that erythrocuprein does not react exclusively with 02Θ. This suggestion was made in the light of the virtually identical Superoxide dismutase activity of low molecular weight Cu2® complexes!91 and the above-mentioned specific reactivity of erythrocuprein during the decomposition of CrO83e. In Fig.6, the problems of the reaction of an electron donor of biological significance with molecular oxygen and the relationship with a strictly inorganic system are sum-

2-

0-, Cr 0,

Cr Ο Ο Η

OH

0, Cr 0

3-

3-

-OH 02 Fig.5. Suggested proteolytic attack of CrO83e in HEPES buffer.

Brought to you by | Purdue University Li Authenticated Download Date | 5/20/15 9:56 PM

Bd. 356(1975)

735

Problems Concerning Superoxide Dismutase

marized. During the course of the reaction, all sorts of oxygen species may be generated. Some of these oxygen species may be considered beneficial for many metabolic processes of the aerobic cell. By contrast, the singlet oxygen type and OHradicals would be extremely deleterious in many pathways of the cellular metabolism. Reducing

agents such as membrane lipids, unsaturated substrates and coenzymes including the flavins could well be irreversibly damaged, Some striking similarities between the decay of both the oxidized flavins and the chromium V peroxochromate complex can be seen.

X0 0

•OH

Ό2~

022~

Ο2"

X:Cr5*

Fig. 6. Possible pathways of XC>2 breakdown. The scheme flavinre£i. · C>2 is taken from ref J11.

Brought to you by | Purdue University Lib Authenticated Download Date | 5/20/15 9:56 PM

736

W. Paschen and U. Weser

In either system, electrons are transferred to molecular oxygen. The specificity of the decay of the flavin-oxygen complex was attributed to specific proteins (denoted as R) which are bound with the flavin moiety and are in a position to dictate the particular breakdown of either flavinoxygen complex. A great number of intermediates may be formed (02Θ, OH*, H 2 0 2 etc.). On the other hand, it was striking to see that the breakdown of the Cr5e peroxo complex can be manipulated by a simple buffer change. In Tris buffer, no Ο2Θ or OH* species were formed. The situation was different when HEPES or phosphate buffers were employed instead.

Bd. 356(1975)

Literature

1 Massey, V., Palmer, G. & Ballou, D. (1971) in Flavins and Flavoproteins (Kamin, H., ed.) pp. 349 - 364, University Park Press, Baltimore. 2 McCord, J. M. & Fridovich, I. (1969) J. Biol. Chem. 244, 6049 - 6055. 3 Mann, T. & Keilin, D. (1939) Proc. Roy. Soc. Ser. B 126,303-315. 4 Rotilio, G., Bray, R. C. & Fielden, E. M. (1972) Biochim. Biophys. Acta 268, 605 - 609. 5 Fielden, E. M., Roberts, P. B., Bray, R. C., Lowe, D. J., Mautner, G. N., Rotilio, G. & Calabrese, L. (1974) Biochem.J. 139,49-60. 6 Klug-Roth, D., Fridovich, I. & Rabani, J. (1973) /. Amer. Chem. Soc. 95, 2786 - 2790. 7 Bannister, J. V., Bannister, W. H., Bray, R. C, The assays for the reactivity of erythrocuprein in Fielden, E. M., Roberts, P. B. & Rotilio, G. (1973) the presence of about 12 different flavoproFEBS Lett. 32, 303 - 306. 8 teinsl3 *' were rather limited. They were restricted Babior, B. M., Kipnes, R. S. & Curnutte, J. T. (1973) J. Clin. Invest. 52, 741 - 744. to the detection of possible Superoxide by em9 Brigelius, R., Hartmann, H.-J., Bors, W., Saran, M., ploying the inhibition of the cytochrome c reE. & Weser, U. (1975) this J. 356,739-745. ductase assay1321. The reactivity of erythrocuprein Lengfelder, 10 Finazzi-Agro, A., Giovagnoli, C, de Sole, P., with other oxygen species was never examined. Calabrese, L., Rotilio, G. & Mondovi, B. (1972) FEBS We are far from understanding in full detail either Lett. 21, 183-185. the decay of the different flavin-oxygen com1 * Weser, U. & Paschen, W. (1972) FEBS Lett. 27, plexes or the reactivity of erythrocuprein. The 248 - 250. 12 challenging task of revealing the physiological Joester, K. E., Jung, G., Weber, U. & Weser, U. function of erythrocuprein prompted us to use (1972) FEBS Lett. 25, 25 - 28. 3e 13 the well defined Cr08 system. Indeed, there Zimmermann, R., Flohe, L., Weser, U. & Hartmann, H. J. (197'3) FEBS Lett. 29, 117- 120. are some indications that this Cu, Zn-protein re14 Paschen, W. & Weser, U. (1973) Biochim. Biophys. acts with one or more high energy oxygen species Acta 327, 217 -222. other than the Superoxide radical. Last but not 15 Khan, A. U. (1970) Science 168, 476 - 477. least, the possibility that erythrocuprein reacts 16 Peters, J. W., Pitts, J.N., Jr., Rosenthal, I. &Fuhr, H. with the CrO83e to form a complex cannot be (1972) /. Amer. Chem. Soc. 94, 4348 - 4350. completely excluded. 17 Michelson, A. M. (1974) FEBS Lett. 44, 97 - 100. 18 Meyeda, E. A. & Bard, A. J. (1974) /. Amer. Chem. A promising approach to the study of the reac96, 4023 - 4024. tivity of Superoxide dismutase with singlet oxygen Soc. 19 Schaap, A. P., Thayer, A. L., Faber, G. R., Goda, K. might be the use of photochemically induced sing& Kimura, T. (1974) /. Amer. Chem. Soc. 94, let oxygen in 2 H 2 0, which is known to prolong 4025-4026. the half-life time of 1 A g 0 2 . This interesting ques- 20 Goda, K., Kimura, T., Thayer, A. L., Kees, K. & tion is currently under investigation, and again Schaap, A. P. (1974) Biochem. Biophys. Res. Commun. there are strong indications that the specificity 58,660-666. of Superoxide dismutase is pronounced compared 21 Stomberg, R. & Brosset, C. (1960) Acta Chem. Scand. 14,441-452. to the low molecular weight copper chelates. 22 Hodgson, E. K. & Fridovich, I. (1974) Biochemistry 13,3811-3815. W.P. is a recipient of a predoctoral fellowship (Graduier23 Riesenfeld, E. H., Wohlers, H. E. & Kutsch, W. A. ten F rderungsprogramm des Landes Baden- W rttem(1905) Chem. Ber. 38, 1885 - 1898. berg). This study was aided by Deutsche Forschungs24 gemeinschaft grants No We 401/9 and 11 awarded to Weser, U. (1973) in Structure and Bonding U.W.. Special thanks go to Drs. Bors, Lengfelder and (Hemmerich, P., Jtfrgensen, C. K., Neilands, J. B., Saran (Ges. f. Strahlen- und Umweltforschung, NeuNyholm, R. S., Reinen, D. & Williams, R. J. P., eds.) herberg) for their generous help and valuable discussions Vol. 17, pp. l - 65, Springer-Verlag, Berlin, Heidelberg, during the course of the pulse radiolysis measurements. New York.

Brought to you by | Purdue University Li Authenticated Download Date | 5/20/15 9:56 PM

Bd. 356(1975)

Problems Concerning Superoxide Dismutase

25 Mottola, H. A., Simpson, B.E. &Gorin,G. (1970) Anal. Chem. 42,410-411. 26

Lang-Mainz, K. (1955) in Handbuch der Physiologisch- und Pathologisch-Chemischen Analyse (HoppeSeyler and Thierfelder, eds.) Vol. HI/1, pp. l - 190, Spring er-Verlag, Berlin, Heidelberg. 27

Cazeneuve, M. P. (1900) Bull. Soc. Chim. 23, 701 - 706. 28 Beauchamp, C. & Fridovich, I. (1971) Anal. Biochem. 44, 276 - 287.

737

29

Reynolds, R. J. & Aldous, K.,(1970) Atomic Absorption Spectroscopy, Griffin, London. 30 Haber, F. & Weiss, J. (1934) Proc. Roy. Soc. Ser. A 147,332-351. 31 Massey, V., Strickland, S., Mayhew, S. G., Howell, L. G., Engel, P. C., Matthews, R. G., Schumann, M. & Sullivan, P. A. (1969) Biochem. Biophys. Res. Commun. 36,891-897. 32 Weser, U., Bohnenkamp, W., Cammack, R., Hartmann, H. J. & Voelcker, G. (1972) thisJ. 353, 1059 -1068.

Brought to you by | Purdue University Li Authenticated Download Date | 5/20/15 9:56 PM

Brought to you by | Purdue University Li Authenticated Download Date | 5/20/15 9:56 PM

Problems concerning the biochemical action of superoxide dismutase (erythrocuprein).

The decay of the tetraperoxochromate- (V) complex (CrO83theta) was examined to study the substrate specificity of erythrocuprein (super-oxide dismutas...
1MB Sizes 0 Downloads 0 Views