You are on page 1of 6



153, 536-541


A Sensitive Spectrophotometric Superoxide Dismutase

Method for the Determination Activity in Tissue Extracts


Received October

3. 1985

Superoxide dismutase (EC I. IS. I. I) has been assayed by a spectrophotometric method based
on the inhibition of a superoxide-driven N.ADH oxidation. The assay consists ofa purely chemical reaction sequence which involves EDTA. Mn(II). mercaptoethanol. and molecular oxygen. re-

quiring neither auxiliary enzymes nor sophisticated equipment. The method is very flexible and
rapid and is applicable with high sensitivity to the determination of both pure and crude superoxide dismutase preparations. The decrease of the rate of NADH oxidation is a function of enzyme concentration. and saturation levels are attainable. Fifty percent inhibition. corresponding to one unit ofthe enzyme, is produced by approximately I5 ng of pure superoxide dismutase. Experiments on rat liver cytosol have shown the specificity of the method for superoxide dismutase. Moreover. common cellular components do not interfere with the measurement. except for hemoglobin when present at relatively high concentrations. The assay is performed at physiological pH and is unaffected by catalase. 15 1986 Academtc Press. Inc KEY WORDS: superoxide dismutase; spectrophotometric determination: chemical assay: superoxide: NADH oxidation: metal complex.

Superoxide dismutase (SOD)’ (EC is a family of metalloenzymes which is known to accelerate spontaneous dismutation of the superoxide radical to hydrogen peroxide and molecular oxygen (1). SOD is widely distributed among aerobically living organisms and has been inferred to play an important role in controlling superoxide levels in cellular compartments (2,3). The direct measurement of SOD activity (47) is possible but its application is hampered by the requirement of special apparatus not commonly available in the typical laboratory. The other methods employed for enzyme determination are indirect and rely on the ability of SOD to inhibit a superoxide-driven reaction. The extent to which SOD reduces the rate of that reaction is taken as a measure of enzymic activity. Either enzymatic or nonenzymatic systems are used for the generation
’ Abbreviations used: SOD. superoxide Dea. triethanolamine-diethanolamine; tetrazolium. 0003-2697/86 $3.00 dismutdse: TeaNBT. nitro blue

of superoxide (see Refs. (8-9) for review): detection is then accomplished by luminometric ( 10). polarographic (1 l), or calorimetric ( 1,12,13) techniques, depending on the different approaches and experimental requirements. Notwithstanding the large number of methods available, the need still exists to increase the specificity. accuracy, and simplicity of the assay. In this paper we describe a spectrophotometric method for SOD determination based on a purely chemical reaction sequence which eventually leads to NADH oxidation. This procedure, involving stable and inexpensive reagents. allows a rapid and highly sensitive measurement of SOD activity in pure and crude enzyme preparations, with negligible interference by cellular components.

Chemicals. Reduced adenine dinucleotide

disodium salt) was obtained from

Copyright 2 1986 by Academic Press. Inc All rights of reproductmn in any form reserved.

up to 7 I ml with water. 0. RESULTS Description oJ‘t/re method.4-7. pH 7. 350 mg/ml ammonium sulfate suspension) was provided by Boehringer-Mannheim. (4) Mercaptoethanol. 7.100 ml: mix thoroughly and read against air at 340 nm for a stable baseline: then add mercaptoethanol solution. Make a stock solution of EDTA.2 M. Coenzyme oxidation occurs in the presence of suitable concentrations of EDTA. Dilute 14. Mix and monitor the decrease in absorbance using 0. The supernatant is dialyzed against cold homogenization buffer and then employed for enzymatic assays. To perform the assay sequentially add the following (see Materials and Methods) to a cuvette: Tea-Dea buffer. even at room temperature. and mercaptoethanol are quite stable.85 g of EDTA-acid in a final volume of 100 ml.040 ml.e. 0. 3300 U/mg protein) and from Sigma Chemical Company (human erythrocytes. once made up. Protein determination was carried out by the L. All other chemicals were analytical grade and used without further purification. in a purely chemical system recently developed in our laboratory. sample (or water or buffer) 0.3 M (i. Eqrr. Darmstad (West Germany). dissolve 20 mg of reduced nucleotide (disodium salt) in 4 ml of water. MnCl?. Assays were performed with a Gilford spectrophotometer (Model 350) connected to a recorder.pmenr.he Results section). ( 1) Triethanolamine-diethanolamine (Tea-Dea) buffer. 1 where the kinetics of NADH oxidation in the absence (control) and in the pres- . adjusting the pH to around 7 with molar NaOH solution) and of MnCL2.1 M (by dissolving 1. Catalase (beef liver.100 ml.035 ml. The final volume in the cuvette is 1. A typical analysis of SOD activity is shown in Fig. pH 7. 10 mrvr. A detailed explanation of the reaction mechanism is beyond the scope of the present paper and will be reported elsewhere (manuscript submitted). ca.62 g of MnC12. taking care to maintain the pH around 7. mediated by superoxide radical. For 100 assays. 0. For longer storage. 100 mM/50 mM. The NADH solution is stable for at least 3 days in the refrigerator.. and then cleared by centrifugation at 30.000 rpm for 60 min at 4°C. can be used over a 2week period (see further comments in l. Dilute 0. HCI up to 1 liter with water.800 ml: NADH solution. ‘The addition of SOD to the reaction mixture causes a proportionate inhibition.05 ml of concentrated thiol. 7Hz0 in 100 ml water).5.9 g Tea. keep it at ~20°C. 100 mM each. 13.5-8 with molar NaHCOI solution. 0. 10. Pure SOD preparations were obtained from Diagnostic Data Inc. thus confirming the involvement of superoxide in the process and providing the basis for SOD activity determination. 0. the tissue extract is first prepared by homogenizing the liver in 3 vol of Tea-Dea buffer 25 mM. and ca. For the measurement of SOD in rat liver. (3) NADH. 2500 U/ mg protein). (beef erythrocytes. The mixture of equal volumes of these two stock solutions yields our third reagent (EDTA/MnCI:). Reagent 3.8 ml of 37%. Rcqynts l~inrlsolutions.4. dissolve 5. for months. of the rate of NADH oxidation. Preparation qf’ rat liver c~msoi. 2H20..065 ml and the light path is IO mm.5 g Dea. The principle of this method is based on the oxidation of NADH. while MnClz . The solutions of EDTA.SUPEROXIDE DISMUTASE DETERMINATION 537 Boehringer-Mannheim (West Germany).9% NaCl buffered at pH 7.4.owry method ( 14) on samples dialyzed against 0. according to the following scheme. 0. (3) EDTA/MnCI. ethylenedinitrilotetraacetic acid (EDTA). and 2-mercaptoethanol were supplied by Merck. Mr? and mercaptoethanol (see below) through a free-radical chain of reactions involving thiol oxidation and univalent 02 reduction.5 mM.5-I full scale deflection. 14. All solutions were made up with deionized or well-aerated distilled water. EDTA/MnC12 solution.

15 ng of pure superoxide dismutase from beef erythrocytes. \ -11 ’ E c \ ’ ‘1 ‘\ PAOLETTI ET AL -----__ -. at 340 nm. The assay conditions are the same as described for pure SOD and the percentage of NADH oxidation is reported as a function of the total proteins in the extract.-SOD 3 --__ -SOD 2 z 0 z 1.-*-i rnemuwnmt. Almost complete saturation levels (99% inhibition) are obtained with 400 ng of the enzyme.9869. To test the applicability of this method to SOD determination in tissue extracts.9925.366 . Four assays are carried out simultaneously in the absence (control) and in the presence of SOD (sample 1. Deterrninution ofSOD in liver euxtructs. Fifty percent inhibition is produced by approximately 10 pg of liver extract which means an average of 100 units of SOD per mg protein of cell cytosol.150.619 s. 3).038.. Assay mixtures are prepared as described in the text and decreases in absorbance.4 -. 180-200 yg proteir. --_ --__ ---_ .). 2 for the purified enzyme and in this case too. 80. \ \ L 0 0 8 16 INCUBATION (mtnl ‘\CONIROI 24 FIG. the correlation coefficient Y = -0. Culihration cm’e \t’ii/l pure SOD. The curve thus obtained shows that inhibition is not directly proportionate to SOD concentration. 2. sample 2. The linear regression equation.638 . while the same amount of heated SOD fails to inhibit the reaction. The curve obtained is essentially identical to that shown in Fig. EDTA or other chelators for Mn2+ (not EGTA) may alter the optimum . Under our conditions Ai1340 over an S-min interval is 0. the boiled sample fails to inhibit NADH oxidation at any rate. yields AA values of 0.10 min) which are used for calculations. Relative rates of NADH oxidation are reported as a function of the amount of enzyme in the assay mixture. and 380 ng of SOD in the assay mixture. n = 22. Least-square linear regression analysis was used to obtain a best-fitting curve over the range 4-40 ng. while the saturation level is reached with ca. at room temperature. One unit of the enzyme corresponds to ca.538 . 5 nmol per min. n = 28. SgAq 1. 80 ng: sample 3. Effect of superoxide dismutase on the rate of NADH oxidation.0.67. One unit of the enzyme is the amount of SOD capable of inhibiting by = -0. is c = 115. experiments were carried out using rat liver cytosol as the sample (Fig.2 I ' \ \ \ \ \ \ \ \ \ \ \ \ \ \ 'SOD. by transforming SOD values into logarithms. then increase progressively (usually 2-4 min after mercaptoethanol addition) to yield good linear kinetics (5. 2. are recorded for about 15 min after mercaptoethanol addition. over a range of 4-40 pg protein.250 for the control.008. and 0. Curve fitting to experimental data is comparable to that described in the comments to Fig. The reaction is started by mercaptoethanol and changes in absorbances are recorded for about 15 min. respectively. IO ng. I. 50%’ the rate of NADH oxidation observed in the control. The equation is ?. but rather follows an exponential-like function. while the presence of 10. 380 ng of pure enzyme from bovine erythrocytes. Rates of NADH oxidation are initially low.55. The determination of SOD activity in pure enzyme preparations from bovine erythrocytes is shown in Fig. the correlation coefficient t ence of superoxide dismutase (SODrm3) are compared by simultaneous assay using a multicell holder. = 1 16.687 s. Diagnostic Data Inc. Precautions and optimal conditiom . The rate of NADH oxidation in the control is ca. For calculations the maximal rates obtained are expressed as a percentage of the control (ordinate) and plotted against a suitable reference (abscissa).

) are added to incubation mixtures and assayed for activity as previously reported. owing to handle a large number of samples. The rate of NADH oxidation (g-min linear kinetic) is expressed as a percentage of the control.+ ..-.SUPEROXIDE DISMUTASE DETERMINATION 539 -‘-t-l--c-i-+20 SUPEROXIDE DISMlJiASE lngl 40 400 FIG. to concentrations suitable for the assay.. Samples containing inactive SOD (0). Closed dots (0) represent the average of four separate determinations and bars (=) indicate the range of experimental values. faster reaction rates. . are observed. Values (0: n = 27) refer to separate determinations carned out individually by three of us using different stock solutions of the enzyme.4. samples are usually diluted by such a large factor that NADH-oxidase or any other interfering activity is practically undetectable. do not start the reaction. excess of Mn” ions in the sample could slow down the rate of NADH oxidation. Protein content of the sample is within a range of O-400 pg as determined by the Lowry method.. 2. are shown for comparison. To avoid all interference by compounds mentioned above. Titration of SOD in rat liver cytosol. For precise calculations. Titration curve with pure superoxide dismutase. but may compete for the chelator. should be evaluated before mercaptoethanol addition and subtracted from the final rate. i. Open squares (0) refer to assays with boiled samples. In 0 L’+0 ++8 LIVER I6 t~-+--+-~. EDTA/Mn’+ ratio and affect the rate of reaction. because of the high sensitivity of the method.g pratml --+ 32 184 FIG. dialysis could be conveniently replaced by a rapid desalting through a small Sephadex G-25 (coarse) column. 3. However. Measurements are carried out as described in the text. mercaptoethanol. Increasing amounts of SOD (O-400 ng) from bovine erythrocytes (Diagnostic Data Inc. which is always run in each set of assays. NADH-oxidase activity. In addition. if present in the sample.e. Other divalent cations of the second transition series.++ 24 CYTOSOL . samples must be dialyzed against suitable me- dia before the analysis. pH 7. Alternatively. at concentrations comparable to that of Mn”. as compared to the control. cysteine and reduced glutathione. Moreover.. The liver extract 1’see Materials and Methods) is diluted with 100 mM Tea-Dea buffer. when the sample contains free thiols. heated at 100°C for 2 min.

Our chemical assay seems particularly suitable to that purpose since it allows the measurement of minute amounts of SOD. Each set of assays must refer to its own control and best measurements are obtained when the values of . this molecule must be removed from the sample before assaying.e.. are ca.540 PAOLETTI tT AL. These inconveniences are frequently observed in assays involving the reduction of NBT.150-0. a specific activity of ca.000 unit/mg protein and ca. addition it is worth mentioning that NADPH reacts as well as NADH in the system (data not shown) without serving as a substrate for NADH-oxidase. The latter result implies that in our system the same value for catalytic activity is obtained by using either 50% or half-maximal inhibition for calculation.626. does not interfere with our assay and manganese-SOD (Mn-SOD) can be easily determined by differential measurement. such as 2 ng. 66. In addition to the experiments with rat liver cytosol. as determined by the xanthine-oxidase/ cytochrome c ( 1). we have found that fifty percent inhibition. However. So far. This lack of interference is clearly shown by the fact that calibration curves for pure SOD and rat liver cytosol are almost identical and saturation levels (99%) are attainable in both cases. relies on the oxidation of the detector (NADH in this case) and therefore will not be affected by the presence of reductants which are known to occur in tissue extracts ( 12). A substantial improvement of the NBT assay has been obtained by Buettner (16). In addition.and manganese-form of the enzyme. Changes of temperature. NADH-diaphorase/hydroxylamine ( 13) and xanthine-oxidase/nitro blue tetrazolium (NBT) ( 12) systems. i. as also pointed out by Oberley (15). but are without effect on the relative degree of inhibition observed. respectively. over an S-min interval. or other suitable detector and ascribed to the action of aspecific electron donors on chromogenic substrates. at concentrations used for inhibiting the cuprozinc enzyme. DISCUSSION Any reaction inhibitable by superoxide dismutase could potentially provide the basis for an indirect assay of the enzyme and. the present method has been applied successfully to the measurement of SOD in a variety of other cell extracts and body fluids (data not shown). Reactivity of reagent 3 is increased by storage: therefore aged solutions immediately yield maximal rates of NADH oxidation without the initial delay shown by fresh-made preparation of the complex. Cyanide. with his method. An important problem in SOD analysis is the discrimination between the cuprozinc. the only major inconvenience encountered comes from hemoglobin. is produced by 15 ng of pure protein.. thus. 200. This fact confirms the flexibility of our method and may turn out very useful when assaying for SOD in fractions containing high levels of NADH-oxidase. several methods have been developed over the years..&4+. pH and oxygen tension in the system may influence the absolute rate of NADH oxidation. In addition. Our method. and 630 ng. according to that principle. especially when dealing with samples having low SOD levels. which are far below the detection limit of most published methods. are within the range of 0. respectively. while values of catalytic activity. one unit of the enzyme. this method has proved ofgreat value in determining traces of Mn-SOD separated . on the contrary. while hydrogen peroxide inhibits it at a millimolar concentration level. saturation levels are not attainable and different values for half-maximal inhibition are obtained for pure and crude SOD preparations. However.400. 100 unit/mg protein can be calculated for SOD in pure beef erythrocytes preparations (Diagnostic Data Inc. Catalase does not interfere with the assay. only a few procedures permit the sensitive and reliable determination of enzymatic activity in tissue extracts with low SOD levels. cytochrome (‘. of the control. From our data. This avoids the necessity of running a full calibration curve each time and is of valuable practical importance. for reliable determination of SOD in hemolysates.: see Materials and Methods) and rat liver cytosol. whose procedure yields exactly the same sensitivity reported here.

.. D. P. (1978) Sc~rc~nc~c 201. 93-104. and Fridovich. Buettner. and Massey. R. ( 1’383) irz Methods of Enzymatic Analysis (Bergmmeyer. M. Ch~7 121. 1. IO. M. plant physiology. 105. and Kaplan. Misra. 244. 457-464. 0.I& Biodwnz.. Rotilio. 247. Rosebrough. J. ( 1969) J. 105. 68. J. ( 195 I ) J. V.. 1. Boddi for his continuous suggestions and statistical analyses. 8. 5. I4?. 12.. H. L. 7. J. J. R. Vol.... 293-302. Rigo.. (1975) B~odw~~~. and Fridovich. McCord. ( 197 I ) . (1975) A& Biodxw.. I. ( 1984) in Methods in Enzymology (Colowick. C51 68. pp. (1976) J Bird 18.SUPEROXIDE DISMUTASE DETERMINATION 541 from rat liver cytosol by means of gel filtration (data not shown). ( 1984) in Methods in Enzymology (Colowick.. and clinical chemistry.?. In this regard it is worth recalling that our assay is carried out at pH 7. (1972) Birxhirn. H. . 15. 3170-3175. J. and Dr.). 116. Youngman.. 1-X. 251. New York. D.. ( 1978) Phr~toc~hcr. and Gtting.. D.. REFERENCES I. 875-880. (1971) 3. McCord. I. and Fridovich.. R. Oberley. Photohid 28. and Randall.1ctu 268. S. Rabani. Proi.-I(& SC. and Fridovich.. Jr. P.18). (1977) in Superoxide and Superoxide Dismutases (Michelson A. H. 1024-1027. Flohe. S. easily performed on a time scale of minutes. BiuphJ. Bray. 13.. 404-420.145. E. M. 4. I. Tyler. I I. and Kaplan. J. Marklund. Weinheim. Crapo. F. pp. 898-904. J.6049-6055.. and OBwald.. pp. pp. W. C.. 2. M. M. Klug. ACKNOWLEDGMENTS We would like to thank Professor A. Farr. eds. III. H. and Johnson. J. Blol. eds. 9. W. (1969) Biochrm. .). Viglino. 6.. 493-504.. 16. 276-187.. Fridovich. Lowry. It appears particularly suitable for application in the field of biochemistry. P. 17. Academic Press. and Fielden. and Fridovich. M..nchrtn. B. McCord. C. E. and Rotilio. C‘hn. 44. L. (1972) J Bid C7rrm. Academic Press. I. Ballou. 247. N. Nar. W.. G. ( I98 I ) . P.. Bioqlzm R~. E.. (1972) d Bid Chc177 .). U. Fonnesu. G. N. Vol. D. G. and Spitz.17. McCord.4 which favors the detection of Mn-SOD. This research was supported by a grant from the Minister0 della Pubblica Istruzione (6Og’).. &airman of the Institute. Beuchamp. 605-609. Elstner... Oberley.. 14.r.. B. 7504-7507. L. D. Viol C’11crn. Bensinger. R. I. G. New York. Palmer.. I. J.. 0. L. eds. J. S. 4839-4442. In fact.. On the whole.. physiological pH is the most suitable for optimal activity of Mn-SOD. D. Keele. R. A.. which is not reliably assayed at elevated pH values as required by other sensitive methods ( 17. S. and Fridovich. M. H. Academic Press. Verlag Chemie. C. B. New York. W.). R. F. ed. A. and Leuthauser. 147. I l. Vol.L Cornrnm 36. V. 693695. C. and Fridovich. the present procedure for SOD determination involves stable and inexpensive reagents and consists of a single spectrophotometric step.