Mechanisms of Ageing and Development, 52 (1990) 11--26
Elsevier Scientific Publishers Ireland Ltd.
E F F E C T OF A G I N G ON P U L M O N A R Y S U P E R O X I D E D I S M U T A S E
HARRY ISCHIROPOULOS, KIKKAWA*
C H R I S T I N E E. N A D Z 1 E J K O
and Y U T A K A
Department of Pathology, University of California, Irvine, California College of Medicine, Irvine, CA, 92717 (U.S.A.)
(Received May 22nd, 1989)
SUMMARY P u l m o n a r y C u , Z n superoxide dismutase was examined in young (1-month-old), adult (4--5-months-old) and aged (24-months-old) rats to determine if partially inactive forms o f the enzyme accumulate in the lung with age. Measurement of Cu,Zn superoxide dismutase activity in lung homogenates showed that total Cu,Zn superoxide dismutase activity/mg D N A was essentially the same in adult and aged rats. The average value of Cu,Zn superoxide dismutase/mg D N A for young rats was less than half that of adult and aged rats. Cu,Zn superoxide dismutase was purified from the lung homogenates and fractionated into isoelectric variants by either isoelectric focusing or chromatofocusing. Three main isoelectric variants of Cu,Zn superoxide dismutase were recovered with pI values of 5.15, 4.88 and 4.75. In all age groups studied, the pI 4.88 variant had a markedly higher specific activity than the other two variants, as well as the highest metal content and greatest resistance to inactivation of all three variants. The pI 4.88 variant declined from 88% of the total Cu,Zn superoxide dismutase activity in the young animals to only 70% in the aged animals. The results of this study indicate that the proportion of the relatively inactive forms of p u l m o n a r y Cu,Zn superoxide dismutase increased with age.
K e y words: Isoelectric variants; Chemiluminescence; Cu,Zn superoxide dismutase;
Mn superoxide dismutase. INTRODUCTION The free radical hypothesis of aging postulates that senescence results from the damaging effects o f superoxide radicals and other reactive oxygen species formed as *To whom all correspondence should be addressed. 0047-6374/90/$03.50 Printed and Published in Ireland
© 1990 Elsevier Scientific Publishers Ireland Ltd.
12 a consequence of cellular metabolism . This theory has received support from studies showing that the lifespan of a species correlates with the ratio of metabolic rate to intracellular levels of superoxide dismutase [2,3], an enzyme that catalyzes the conversion of superoxide radical to hydrogen peroxide . Studies on houseflies showed that longevity was correlated with levels of superoxide dismutase within the same cohort group . The aging rate of different tissues may also be related to the amount of superoxide dismutase. Mizuno and Ohta  have shown that thiobarbituric acid reactive products increased significantly in areas of the brain known to show changes with aging. These areas also showed a decrease in superoxide dismutase activity with aging whereas the other areas of the brain showed no change or an increase in superoxide dismutase activity. The importance of superoxide dismutase as a free radical scavenger during normal metabolism is indicated by a report  which showed that spontaneous chemiluminescence, an indicator of intracellular peroxidation reactions, increased six to ten times in isolated hepatocytes when intracellular Cu,Zn superoxide dismutase was inhibited by diethyldithiocarbamate. Although there is evidence that the superoxide dismutases, particularly Cu,Zn superoxide dismutase, play a critical role in the defense against free radicals, it is not known if defenses against free radicals decline in old age . Studies on age-related changes in superoxide dismutase have yielded diverse results. Reiss and Gershon reported that C u , Z n superoxide dismutase activity was decreased in the liver of aged rats [9--11] and mice , while little change was seen in heart and brain tissues of aged animals . Cu,Zn superoxide dismutase purified from the liver of aged rats was reported to have decreased specific activity and less thermostability [9,11]. In contrast, other studies have shown no decrease in Cu,Zn superoxide dismutase activity in the liver [12--14], brain [12,14] and heart [12,13] of aged rats and mice but did find a significant decrease in Mn superoxide dismutase in the livers of aged rodents [12,14]. Further studies on lens , heart and brain tissue  showed that although the amount of enzyme activity in homogenates of these tissues was unchanged in aged animals, larger quantities of antiserum to Cu,Zn superoxide dismutase were required to precipitate equal amounts of enzyme activity in tissue homogenates from aged subjects. This finding, that enzyme activity per antigenic unit declined with age, suggests that altered superoxide dismutase with reduced catalytic activity accumulates in aged animals. During the course of our studies on the isoelectric variants of Cu,Zn superoxide dismutase from rat lung, we found that some of the variants of SOD had relatively little enzyme activity. It has been known for over a decade that purified Cu,Zn superoxide dismutase from a wide variety of species separates into several bands during electrophoresis [1 l, 16--18]. These bands each have the same molecular weight and are antigenically cross-reactive  but they have different isoelectric points, hence the term isoelectric variants. We have examined age-related changes in the isoelectric variants of Cu,Zn superoxide dismutase of the lung to see if relatively inactive
13 isoelectric variants of Cu,Zn superoxide dismutase accumulate with age. The effect of in vitro exposure of Cu,Zn superoxide dismutase to activated oxygen species was examined to determine if age-related changes in the pattern of isoelectric variants could be the result of oxidative damage. MATERIALSAND METHODS
Materials Fisher F344 male rats were obtained from Harlan Sprague Dawley Inc. and Sprague--Dawley CD strain male rats were obtained from Charles River Breeding Laboratories. Young rats were 25--30 days old and weighed 35--40 g, adult rats were 4--5 months old and 300--350 g, and aged rats were 24 months old and weighed 400--500 g. Chemicals for high performance liquid chromatography, cytochrome c (type VI from horse heart) and xanthine oxidase (grade III from buttermilk) were from Sigma Chemical Company. Polybuffer 74 and the Mono P FPLC column were purchased from Pharmacia, Piscataway, NJ. The DEAE 500 HPLC column was obtained from Perkin Elmer, Norwalk CT.
Purification of Cu, Zn Superoxide dismutase Rats were anesthetized with sodium pentobarbitol 70 mg/kg i.p. Lungs were perfused with 0.9°70 saline via the portal vein with the thoracic cavity closed after transection of the abdominal aorta. The lungs were removed and the lung parenchyma was dissected free of the trachea and major airways. The lung lobes were blotted gently and weighed. Subsequent steps were performed at 4 °C unless otherwise stated. The tissue was added to 2.5 mM potassium phosphate buffer (pH 7.8), (1 g tissue/7 ml) with 10 mM mannitol and 440 U/ml catalase and homogenized in a microblender (Eberbach Corporation) for 30 s followed by homogenization in a Potter Elvehjem Teflon pestle glass homogenizer for 15 strokes. Cu,Zn superoxide dismutase was extracted from the homogenate by the method of Crapo, McCord and Fridovich . Cu,Zn superoxide dismutase was purified by HPLC ion exchange (Perkin Elmer Isopure) using a DEAE-500 anion exchange column eluted with a gradient of 20 mM Tris--HCL (pH 8.5)/20 mM Tris--HCL and 0.5 M NaCI. We found previously that this purification method results in the recovery of approximately 60070 of the extractable Cu,Zn superoxide dismutase. The fractions containing Cu,Zn superoxide dismutase activity were pooled, dialyzed and separated into isoelectric variants by either preparative isoelectic focusing or FPLC chromatofocusing. Preparative isoelectric focusing was done on a 110-ml LKB 8100 preparative column using a sucrose gradient and 2070 Pharmalyte (pH range 4--6.5). The purified Cu,Zn superoxide dismutase was electrofocused for 6 h after constant current was obtained and 1-ml fractions were collected and analyzed for pH and superoxide dismutase activity. Chromatofocusing was performed by FPLC using a Mono P column equilibrated with 25 mM piperazine buffer (pH 6.8) and eluted with
14 Polybuffer (pH 4.8) diluted 1:14 with H20. Thin layer gel isoelectric focusing was performed using 5% polyacrylamide gels containing 2.20/0 ampholine, (pH 4--6.5) (LKB) using a Bio-Rad 111 Mini IEF cell. Gels were stained with a silver stain kit from Bio-Rad using the manufacturer's protocol.
Measurement of superoxide dismutase activity Superoxide dismutase activity was measured using the xanthine-xanthine oxidase, cytochrome c reduction assay . One unit of superoxide dismutase activity was defined as a 50% decrease in the rate of reduction of cytochrome c at pH 7.8. Except where indicated otherwise, all measurements of superoxide dismutase activity were made at pH 7.8. Mn superoxide dismutase has the same catalytic activity and is assayed by the same methods as Cu,Zn superoxide dismutase. It was necessary to determine how much of the superoxide dismutase activity in a tissue homogenate was due to Cu,Zn superoxide dismutase and how much was due to Mn superoxide dismutase. Two approaches were used to distinguish between the two forms of the enzyme. First, Cu,Zn superoxide dismutase activity was measured directly after it was extracted exhaustively from tissue homogenates by organic solvents , a procedure which inactivated Mn superoxide dismutase while having no effect on Cu,Zn superoxide dismutase activity . Second, selective inactivation of Cu,Zn superoxide dismutase but not Mn superoxide dismutase by 1 mM CN- was used to derive the activity of Cu,Zn and Mn superoxide dismutase from the total dismutase activity in tissue homogenates [19,21]. Calculation of Cu,Zn superoxide dismutase activity/mg DNA in dialyzed tissue homogenates using 1 mM CN- was in close agreement with results obtained by direct measurement of Cu,Zn superoxide dismutase after exhaustive extraction.
Analytical procedures DNA was measured using diphenylamine  and protein was quantitated by the method of Lowry et al.  Cu (II) and Zn(II) content of Cu,Zn superoxide dismutase was measured using a Perkin Elmer 603 atomic absorption spectrophotometer. The capacity of mitochondrial and microsomal fractions of lung homogenates to generate reactive oxygen species was detected by chemiluminescence. The mitochondria and microsomes were isolated from lung homogenates as previously reported . Mitochondrial or microsomal protein (1 mg) was suspended in 900/~1 of 50 mM Tris--HC1 (pH 7.4), 150 mM KCI, and 1 mM MgCI2, supplemented with 0.I mM N A D H for mitochondria or 0.1 mM N A D P H for microsomes. After a 30-min incubation at 37 °C, lucigenin amplified chemiluminescence was measured by adding 100 /A of 2 mM lucigenin to the subcellular fractions. Chemiluminescence was measured for 3 min with 30-s integration intervals using a Turner Model 20e luminometer. Oxidative DNA damage was quantitated by extracting DNA from the lung homogenates  and measuring thiobarbituric acid-reactive products of deoxyribose degradation per mg of DNA as described by Gutteridge and Bannister
Statistical significance was determined by a parametric one-way analysis of variance and the non-parametric Kruskal-Wallis one-way analysis o f variance. The Scheffe F-test was used to compare group means when both tests of A N O V A indicated statistical significance, . A P value of less than 0.05 was considered statistically significant. Statistical analysis was performed on a Macintosh II using StatView 5~2÷software.
SOD U/g Wet Weight
SOD U/mg DNA
60 B Y
Age in Months
Fig. 1. Superoxide dismutase (SOD) content o f the lungs of young, adult and aged rats. (A) Cu,Zn and Mn SOD activity expressed as U / g wet lung weight. (B) Cu,Zn and Mn SOD activity expressed as U / m g D N A . Y-statistically significant compared to the young group. A-statistically significant compared to the adult group. Values for N are shown in parentheses in Fig. lA. Error bars indicate l standard deviation.
Specific Activity of Isoelectric Variants of Cu,Zn SOD 4000
• • •
1 Month 4-5 Months 24 Months
• • •
1 Month 4-5 Months 24 Months
iS pl 5.15
Cu Content of Isoelectric Variants of CuZn SOD 2
Fig. 2. Analysis of isoelectric variants of C u , Z n SOD from the lungs of young, adult and aged rats. (A) Specific activity, N = 3,4,2 for young, adult, and aged rats. (B) Copper content, N = 2 for each age group. The values shown for N refer to the number of separate experiments performed. For each experiment, lung homogenates from several animals were pooled, the C u , Z n SOD was purified and separated into isoelectric variants. T h e average value for each group of experiments is indicated by the closed symbol. Error bars enclose all measurements.
Age-related changes in the activity o f CuZn and Mn superoxide dismutase in the lung The content of Cu,Zn and Mn superoxide dismutase was measured in the lungs of 25-day-old (young), 4--5-month-old (adult) and 24-month-old (aged) rats. These three age groups were chosen because they corresponded to the growth phase, mature phase and post-mature phase of lung development. The results in Fig. 1A show the amount of catalytically active Mn and Cu,Zn superoxide dismutase/g of wet lung weight in the different age groups. The adult and aged groups had significantly higher levels of Cu,Zn superoxide dismutase activity than the young group. Mn superoxide dismutase activity was slightly but significantly higher in the adult rats than the other two groups. Figure 1B shows Mn and Cu,Zn superoxide dismutase activity expressed as units/mg DNA. Here, the level of Cu,Zn superoxide dismutase/mg DNA was essentially the same in adult and aged animals and significantly lower in young animals. Mn superoxide dismutase activity/rag DNA was the same as in Fig. 1A. The two different denominators for superoxide dismutase activity were used to rule out the possibility that age-related changes in superoxide dismutase levels were merely the result of changes in the denominator. Lung weight and DNA content of the lung increased throughout the life span as reported previously . Isolation and characterization o f isoelectric variants of Cu, Zn superoxide dismutase: relationship to aging The purified Cu,Zn superoxide dismutase from the lungs of each age group was separated into isoelectric variants as described in the methods. Three main variants were obtained with pI values of 5.15, 4.88 and 4.75. The specific activity and copper content of the three variants is shown in Fig. 2. The catalytic activity of the three variants differed widely within each age group. The specific activity of the pI 4.88 variant was much higher than the other two variants of Cu,Zn superoxide dismutase in all age groups. In all age groups, only the pI 4.88 variant contained the full complement of copper (2 atoms/molecule) that is needed for the enzyme to be fully active . The zinc content of each variant was essentially the same as the copper content. A number of experiments were done to rule out the possibility that these partiallyinactive variants were generated during purification. The same pattern of variants was obtained when either cytosolic fractions or acetone precipitates of tissue homogenates were electrofocused and assayed for enzyme activity. Cu,Zn superoxide dismutase extracted from tissues using salt precipitation  without organic solvents resulted in the same variants with the same differences in specific activity and metal contents. Metal content and specific activity was essentially the same whether the variants were isolated by chromatofocusing or isoelectric
100 - A
50 25 Apl 4.75 "~ i=1
15 80 °
50 25 05
Fig. 3. Inactivation of isoelectric variants of Cu,Zn SOD from adult rat lung. (A) Inactivation by H,O in 50 mM sodium carbonate at 25°C. (B) Inactivation by D D C after 2 h incubation at 25°C. (C) Inactivation at 80°C. (D) Inactivation by CN-. SOD was measured by the xanthine-xanthine oxidase cytochrome c reduction assay at pH 10. The closed symbols enclose duplicate measurements.
focusing. The decreased catalytic activity of the pI 5.15 and 4.75 variants could not be accounted for by contaminating proteins: SDS gel electrophoresis of each variant (adult rat lung) resulted in a single band (under reduced conditions) corresponding to an apparent molecular weight of 16 000 Da . Amino acid analysis showed that each variant had a similar amino acid composition to that reported previously  for Cu,Zn superoxide dismutase (data not shown). The isoelectric variants of Cu,Zn superoxide dismutase were further characterized by examining the stability of the variants under several different conditions. Cu, Zn superoxide dismutase is a very stable enzyme but it is irreversibly inactivated by hydrogen peroxide. It is also inactivated by substances that affect copper, namely the copper chelator diethyldithiocarbamate (DDC), and cyanide. Figure 3 shows the effect of heat, H202, DDC, and CN-, on the activity of the three main variants of Cu,Zn superoxide dismutase from adult rat lung. The pI 4.88 variant was the most thermostable of the variants and showed the greatest resistance to inactivation by H202, DDC, and CN-. The data on metal content combined with the relative activities and stabilities of the three variants suggest that the pI 4.88 variant contained the highest proportion of fully active enzyme and that the pl 5.15 and 4.75 variants contained relatively little active enzyme. Figure 4 shows an isoelectric focusing gel of pulmonary Cu,Zn superoxide dismutase from young, adult and aged rats. Although equal amounts of protein were applied for each age group, the pI 5.15 and 4.75 bands are progressively more prominent in superoxide dismutase from adult and aged rats. The relative
5.15 438 4.75
Fig. 4. Isoelectric focusing gel of pulmonary Cu,Zn SOD from young, adult and aged rats. (A) aged, (B) adult, (C) young.
% Total CuZn SOD Activity 100 "
60 • • •
1 Month 4-5 Months 24 Months
14 pl 5 1 5
pl 4 7 5
% CuZn SOD Protein 80
• • •
1 Month 4-5 Months 24 Months
Variant Fig. 5. Relative contribution of each of the isoelectric variants to Cu,Zn SOD activity and Cu,Zn SOD protein. (A) °7o total Cu,Zn SOD activity. (B) 070 Cu,Zn SOD protein. The relative amount of protein contained by each variant was calculated from the specific activity of each variant and 07o total SOD activity. N = 3,4,2 independent experiments done on young, adult, and aged rats respectively. The closed symbols indicate the average value for each group of experiments. The error bars enclose all measurements.
contribution of each variant to both the total activity and protein content of pulmonary Cu,Zn superoxide dismutase was quantified to determine if the relatively inactive variants, pI 5.15 and pI 4.75, accumulated with age. The data shown in Fig. 5 indicates that the enzymatically active variant, pI 4.88, declined relative to the other two less active variants in terms of percentage of total activity as the animal aged. In young animals, 88°70 of the total Cu,Zn superoxide dismutase activity was due to the pI 4.88 variant; but in aged animals, only 70% of the total activity could be attributed to the pI 4.88 variant. This trend was even more striking when each variant was expressed in terms of percentage of Cu,Zn superoxide dismutase protein. In young animals, 60% of the Cu,Zn superoxide dismutase protein was in the pI 4.88 form as compared to only 37% in the adult and 30% in the aged rats. The variant with a pI of 4.75 was markedly increased in aged rats compared to both young and adult rats. As shown in Fig. 6, the heat stability of each individual variant of pulmonary Cu,Zn superoxide dismutase was not decreased in aged animals. The variant with a pI of 4.88 was the most stable in all age groups and the pI 4.75 variant was the least stable of the variants in all age groups with the exception of the pl 5.15 variant isolated from adult rat lung. This one variant was least thermostable of all variants in all age groups.
Formation of partially inactivated b~oelectric variants of Cu, Zn superoxide dismutase by activated oxygen species in vitro To determine if the partially inactive variants of Cu,Zn superoxide dismutase were the result of exposure of superoxide dismutase to reactive oxygen species, we used lung microsomes as a physiologic source of activated oxygen species. Microsomes were isolated from lung homogenates and washed free of superoxide dismutase activity. Cu,Zn superoxide dismutase (7125 U) from adult rat tung was 100
pl 5.15 L