Plasma Extracellular Superoxide Dismutase Levels in an Australian Population With Coronary Artery Disease
Abstract—In vitro experiments suggest that free radicals may contribute importantly to atherogenesis. Superoxide dismutase (SOD), particularly extracellular SOD (EC-SOD), which accounts for the majority of SOD biological activity, is a major superoxide scavenger. We explored factors that may affect plasma EC-SOD levels measured by ELISA and assessed the association between plasma EC-SOD and coronary artery disease documented angiographically in 590 white Australian patients ≤65 years old. Mean±SEM plasma EC-SOD in female patients (113.6±13.2 ng/mL) was significantly higher than in male patients (86.6±5.1 ng/mL, P<0.0001), and all 19 patients with levels >400 ng/mL were heterozygous for the Arg213→Gly mutation at the EC-SOD gene; there was also a positive correlation with age (r=0.131, P=0.0016). Plasma EC-SOD in current smokers (75.0±9.3 ng/mL) was much lower than in nonsmokers (111.7±8.2 ng/mL, P<0.01), and ex-smokers had intermediate levels (84.3±7.1 ng/mL). Levels were significantly lower in patients with than in those without a history of acute myocardial infarction (MI) (76.1±7.5 versus 110.1±6.0 ng/mL, P<0.05), and low plasma EC-SOD was independently associated with an increased likelihood of a history of MI (OR, 2.04; 95% CI, 1.10 to 3.82); higher EC-SOD levels also tended to be associated with delayed onset of MI. In conclusion, our study establishes that in patients assessed by coronary angiography, circulating EC-SOD is lower in men than in women and in smokers of each sex and that low levels are independently associated with a history of MI. These findings are consistent with EC-SOD’s being protective and contributing to reduced coronary risk.
- extracellular superoxide dismutase
- extracellular superoxide dismutase Arg213→Gly mutation
- coronary disease
- myocardial infarction
- Received February 25, 1998.
- Accepted May 27, 1998.
Despite the efforts and successes in exploring and implementing pharmacological and nonpharmacological interventions to reduce coronary risk, coronary artery disease (CAD) remains the leading cause of mortality and morbidity in most developed countries. Among the many factors known to contribute to cardiovascular risk, in vitro experiments have suggested a possible link between reactive oxygen species or free radicals and the pathogenesis of atherosclerosis. Reactive oxygen species include superoxide anions, the hydroxyl radical, hydrogen peroxide, hypochlorous acid, and peroxynitrites; these may be produced by many cells, eg, endothelial cells, smooth muscle cells, neutrophils, monocytes, and platelets.1 Superoxide radicals may react with various molecules and result in either direct damage or potentially harmful products. They also react avidly with nitric oxide (NO), which is constantly produced by the endothelium and maintains basal vascular tone, to form peroxynitrite, a potent oxidant.1 2 3 4 An abundance of peroxynitrite, as identified by nitrotyrosine, has been demonstrated in atherosclerotic lesions, and overproduction of peroxynitrite has been implicated in atherogenesis.3 5 6 7
However, protective mechanisms are also available that oppose the deleterious effects of these reactive oxygen species. The balance of these 2 processes is critical in the pathogenesis of many disorders. This protective scavenging function against superoxide is provided mainly by superoxide dismutase (SOD) present in the extracellular space and within cells of the vascular wall.8 9 10 There are 3 isoenzymes of SOD: the secreted extracellular SOD (EC-SOD), cytosolic Cu,Zn-SOD, and mitochondrial Mn-SOD.8 9 10 EC-SOD is a secretory, tetrameric, copper- and zinc-containing glycoprotein. More than 90% of EC-SOD is found in the interstitial spaces of tissues and extracellular fluids, and this accounts for the majority of the SOD activity of plasma, lymph, and synovial fluid.8 11 12 It has a high affinity for heparan sulfate proteoglycan, which appears to be the most important physiological ligand of EC-SOD.13 14 15 The proteoglycans are present in the connective tissue matrix and on cell surfaces, particularly of endothelial cells. The EC-SOD located on the surface of the endothelium bound to proteoglycan accounts for only a small proportion of the total body EC-SOD, but it remains in equilibrium with plasma levels.11 12 16 An Arg213→Gly mutation located at the heparin-binding domain has been identified in a small proportion of the healthy population.17 18 19 20 The mutation is associated with very high plasma EC-SOD levels17 18 19 20 and impairs the affinity of EC-SOD for heparin at the endothelial cell surface.20 21 22
Because a link between reactive oxygen species and atherogenesis is supported mainly by in vitro findings, we sought to explore the association in vivo and investigated the relevance of EC-SOD, an important free-radical scavenger, to atherogenesis. We assessed associations between plasma EC-SOD levels and the occurrence and severity of CAD in an Australian population with documented coronary artery disease status and risk factor profiles. We also explored the relation between the Arg213→Gly point mutation and EC-SOD levels in these patients.
We studied 590 white patients ≤65 years old, both men and women, consecutively referred to the Eastern Heart Clinic at Prince Henry Hospital for clarification of a provisional diagnosis of ischemic heart disease and for whom coronary angiography was performed. Patients receiving warfarin or heparin therapy at the time of study were excluded because this therapy may interfere with quantitative measurements of plasma EC-SOD. Written consent was obtained from every patient after a full explanation of the study, which was approved by the Ethics Committee of the University of New South Wales.
A 4-mL venous blood sample was drawn from a catheter immediately before the angiogram and after a ≥6-hour fast. The blood was collected into an EDTA sample tube and maintained at 4°C for a maximum of 5 hours before centrifugation at 3500 rpm (2000g) for 10 minutes at 4°C. The plasma was stored at –70°C in aliquots for up to 24 months until the EC-SOD analysis.
Measurements of EC-SOD Levels and Biochemical Analyses
Circulating plasma EC-SOD levels were measured as described previously.23 Plasma levels of total cholesterol (TC), HDL cholesterol (HDL-C), triglyceride, glucose, and creatinine were measured by the hospital’s Clinical Chemistry Department by standard enzymatic methods. LDL cholesterol levels were calculated by use of the Friedewald formula. We measured levels of apolipoprotein (apo) A-I, apoB, and Lp(a) using ELISA methods developed in our laboratory.24
Genotyping of Arg213→Gly Mutation at EC-SOD Locus
To assess the relationship between the Arg213→Gly mutation and plasma EC-SOD levels in our population, we genotyped all patients with plasma EC-SOD levels >400 ng/mL (n=19) and a randomly selected a group of 38 patients with levels below the cutoff level (range, 39.6 to 328.2 ng/mL). The method for genotyping was described by Sandstrom and colleagues.20 The DNA fragment of the EC-SOD gene containing the mutant site was amplified by polymerase chain reaction with the primers EC3 and EC5, followed by digestion with the restriction enzyme MwoI. The digestion products were separated on a 12% polyacrylamide gel and stained with silver for viewing.
Documentation of CAD and Other Medical Conditions
Two cardiologists unaware of the EC-SOD findings assessed the angiograms. The severity of CAD was determined by the number of significantly stenosed coronary arteries. Each angiogram was classified as either revealing normal coronary arteries or having no coronary lesion with >50% luminal stenosis or as having 1, 2, or 3 major epicardial coronary arteries with >50% luminal obstructions.
The relevant history was obtained for each patient by use of a questionnaire with standardized choices of answers to be checked during the interview. We recorded the presence or absence (yes/no) of a history of myocardial infarction (MI) and the age of first onset, hypertension requiring treatment, diabetes, and angina pectoris. The presence of each medical condition was confirmed by a review of the hospital medical records for each patient. We recorded the presence and severity of angina and a quantitative assessment of family history of premature CAD, as described previously.24 The lifetime smoking dose was calculated by multiplying the mean number of cigarettes smoked daily and the number of years of smoking. The patient was considered to be a current daily smoker if she or he had regularly smoked ≥5 cigarettes/d for at least the previous 3 months or had stopped smoking for <1 year. Patients who had stopped smoking for ≥1 year were classified as ex-smokers.
The results are presented as mean±SEM. We used unpaired Student’s t tests for 2-group comparisons and ANOVA when >2 groups were compared. We used logistic regression analysis to assess the independent contributions of various factors to the response variables. Because the distributions of EC-SOD levels were skewed, a nonparametric Kruskal-Wallis 1-way ANOVA was used, and parametric comparisons were also used for the logarithmically transformed EC-SOD levels that were normally distributed. A χ2 test was used for comparisons between categorical variables.
Arg213→Gly Mutation and Plasma EC-SOD Levels
Although every patient with EC-SOD levels >400 ng/mL was heterozygous for the mutation, none of the randomly selected patients with levels <400 ng/mL were found to have the mutant allele. No mutant homozygotes were detected. Given these clear-cut findings, we did not proceed to genotype the remaining patients.
Sex and Plasma EC-SOD Levels
The distribution of EC-SOD levels was highly skewed (Table 1⇓). The majority of the patients had plasma EC-SOD <100 ng/mL, with only a small proportion having high levels and only 2 patients with levels >1000 ng/mL. The EC-SOD levels in the 160 women were ≈31% higher than those in the 430 male patients (Table 1⇓, P<0.0001, power=0.8916) in both parametric and nonparametric comparisons. The proportion of female patients with EC-SOD levels above the 75th percentile of the total population (83.3 ng/mL) was 38.8% and significantly more than the 20.7% for male patients (χ2=20.134, P<0.0001). The same was true when the 90th percentile (113.3 ng/mL) was used as the cutoff point to classify patients into high and low EC-SOD groups (χ2=11.935, P=0.0006).
Cigarette Smoking and EC-SOD Levels
There was a significant effect of smoking on EC-SOD. Nonsmokers had the highest EC-SOD levels, current smokers had the lowest, and the ex-smokers had intermediate levels (Table 2⇓, power=0.9115). This association was also highly significant when EC-SOD levels were categorized according to the 75th or 90th percentile levels in that more nonsmokers had high EC-SOD levels. The percentages of patients above the 75th percentile (83.2 ng/mL) among nonsmokers, ex-smokers, and current smokers were 32.1%, 22.4%, and 21.8%, respectively (χ2=6.727, P=0.034), and for those above the 90th percentile (113.3 ng/mL), they were 15.8%, 8.4%, and 6.8% (χ2=9.043, P=0.0109). The smoking effect on lowering plasma EC-SOD was consistent among groups with high or low plasma EC-SOD levels (according to the 75th percentile level), but there were small numbers in each group for the subgroup analysis (Table 2⇓). However, EC-SOD levels were not related to the lifetime smoking dose in either the whole patient population (r=−0.0687, P=0.1074) or subgroup analysis of only smokers (r=0.0366, P=0.4883), current smokers (r=0.0441, P=0.6035), or ex-smokers (r=0.0729, P=0.2826).
Age and EC-SOD Levels
There was a significant increase in EC-SOD levels with increase in age (r=0.1306, F=10.10, P=0.0016, power=0.8094) in this patient population. This was also true for both men (r=0.096, F=3.974, P=0.046) and women (r=0.188, F=5.804, P=0.017).
Biochemical Variables and EC-SOD Levels
In univariate analysis, EC-SOD levels were also correlated negatively with waist-to-hip ratio (r=−0.1115, P=0.0080), levels of triglyceride (r=−0.160, P=0.0001), the TC/HDL-C ratio (r=−0.1173, P=0.0047), and glucose (r=−0.0621, P=0.3314) and positively with levels of HDL-C (r=0.1459, P=0.0004) and of apoA-I (r=0.159, P=0.0001).
Effects of Sex, Smoking, Age, and Biochemical Variables on EC-SOD Levels in Multivariate Analysis
To assess the confounding effects of the measured factors, we used a multivariate regression analysis. In the model, EC-SOD was the response variable, and smoking status, age, past history of MI, and levels of plasma lipids and apolipoproteins were entered as effect variables. The relationships between plasma EC-SOD levels and age (P=0.0106), smoking status (P=0.0439), and sex (P=0.0181) remained statistically independent. However, none of the relationships between biochemical variables, waist-to-hip ratio, and the EC-SOD levels were statistically significant after control for age, sex, smoking, and history of MI. As shown in Table 2⇑, the sex difference in EC-SOD levels was constant in subgroups of nonsmokers, current smokers, and ex-smokers. This was also true for the effect of smoking status on EC-SOD levels after control for sex, even though the degrees of the differences and statistical significance tests varied according to the size of the subgroup population and the variance in EC-SOD levels. The adjusted EC-SOD levels (log EC-SOD levels) derived from the multivariate regression model for men and women were 87.4±6.3 ng/mL (1.85±0.01 ng/mL) and 108.8±9.2 ng/mL (1.89±0.01 ng/mL), respectively. The adjusted levels for nonsmokers, current smokers, and ex-smokers were 118.4±9.2 ng/mL (1.91±0.01 ng/mL), 80.8±10.9 ng/mL (1.83±0.02 ng/mL), and 95.1±8.8 ng/mL (1.87±0.01 ng/mL), respectively.
The Arg213→Gly Mutation and Relationships Between EC-SOD Levels and Demographic Variables
Because higher EC-SOD levels (>400 ng/mL) are associated with the Arg213→Gly mutation, patients with levels above or below the 400 ng/mL level may belong to 2 different genetic populations. When we confined the statistical analysis to those with plasma levels <400 ng/mL, the independent association between the EC-SOD levels and age (P<0.0001) and sex (P=0.0006) remained. Current smokers still had the lowest EC-SOD (68.8±2.6 ng/mL; log EC-SOD, 1.80±0.01 ng/mL; n=145) compared with nonsmokers (75.3±2.4 ng/mL; log EC-SOD, 1.85±0.01 ng/mL; n=179) and ex-smokers (73.0±2.0 ng/mL; log EC-SOD, 1.83±0.01 ng/mL; n=244, P=0.018). When the comparisons were conducted among patients with EC-SOD >400 ng/mL, levels were still the lowest among current smokers (520±145.9 ng/mL, n=2) compared with nonsmokers (790.5±62.2 ng/mL, n=11) and ex-smokers (709.2±84.3 ng/mL, n=6). Furthermore, in the subgroup analysis confined to patients with levels <400 ng/mL, plasma concentrations of creatinine (r=0.094, P=0.028), of triglyceride (r=−0.2014, P=0.0067), of apoA-I (r=0.1846, P=0.00001), and of glucose (r=−0.140, P=0.0238) were significantly associated with the plasma EC-SOD levels. These relationships remained significant after age, sex, smoking, and history of MI were controlled for. In the multivariate analysis, none of the other variables were significantly correlated with the EC-SOD levels in this subgroup, even though some of the variables, including TC/HDL ratio, apoB, and waist-to-hip ratio, were significantly correlated with EC-SOD in univariate analyses.
Plasma EC-SOD Levels and CAD
As shown in Table 3⇓, plasma EC-SOD levels were 44.9% lower in patients with than in those without a past history of MI (power=0.6585). When we used the 75th or the 90th percentile levels to classify patients into either high– or low–EC-SOD groups, there were many more patients with high EC-SOD and no history of MI than patients with MI (χ2=6.577, P=0.0103 for the 75th percentile level cut point and χ2=6.410, P=0.0113 for the 90th percentile level). Those with lower EC-SOD levels (ie, below the 90th percentile level of 112 ng/mL) had an increased likelihood of having had an MI (OR, 2.04; 95% CI, 1.10 to 3.82), as assessed in a multivariate logistic regression analysis after other independent-effect variables were controlled for. Furthermore, although the association between EC-SOD levels and age at onset of MI falls short of statistical significance (r=0.1204, β=1.699, SEM=0.080, F=3.34, P=0.0689), the higher EC-SOD levels tended to be associated with late onset of MI. The ages of onset of MI for male and female patients were 48.8±0.6 and 52.3±1.1 years, respectively (F=8.04, P=0.0048). Because the effects of sex and smoking on EC-SOD levels were highly significant, we have also presented the subgroup data after these 2 factors were controlled for (Table 4⇓). Although with the smaller number of patients in each subgroup, and consequently reduced power, none of the univariate comparisons between the subgroups were statistically significant, EC-SOD levels were higher in the subgroups without MI history, the female patients, and the nonsmokers (Table 4⇓). Moreover, the association between the EC-SOD levels and past history of MI remained significant in a logistic regression analysis among all 590 patients in whom sex, age, and smoking status were controlled for (χ2=5.390, P=0.0202). The adjusted EC-SOD levels were 99.8±6.5 ng/mL (1.88±0.01 ng/mL) for those without and 80.3±8.0 ng/mL (1.84±0.01 ng/mL) for those with a positive history of MI.
With regard to CAD severity, although more patients without significantly diseased vessels tended to have higher plasma EC-SOD levels, levels were not related to the number of significantly diseased vessels (Table 5⇓). The proportion of patients without significantly diseased vessels with high EC-SOD (>75th percentile level) (30.8%) was greater than that for patients with 1, 2, or 3 significantly diseased vessels (23.6%), but this difference was not statistically significant. Because there were fewer women among patients with more severe CAD who were also older (Table 5⇓), we adjusted for these variables in a stepwise logistic regression model. The association between the number of significantly diseased vessels and EC-SOD levels was not significant (F=0.2357, P=0.8715). We also reclassified our patient groups by comparing those with triple-vessel disease with the rest of the patients. There was no statistically significant difference in EC-SOD levels between the 2 groups. The adjusted EC-SOD levels after age, sex, and smoking status were controlled for were not different (F=2.34, P=0.126), although those with triple-vessel disease (88.9±5.8 ng/mL) tended to have lower levels than the rest of the patients (106.1±10.5 ng/mL).
Other Medical Conditions and EC-SOD Levels
We found no associations between EC-SOD levels and the presence of diabetes, hypertension, or a positive family history of premature CAD in this patient population. The EC-SOD levels in patients without angina (105±8.2 ng/mL, n=242) tended to be higher than in those with stable angina (88.8±9.0 ng/mL, n=202), and patients with unstable angina had the lowest EC-SOD levels (82.7±10.6 ng/mL, n=242). However, these differences were not statistically significant (F=1.72, P=0.180).
Although SOD functions primarily within cells and in extracellular matrixes and counteracts damaging oxidative effects of superoxide, circulating EC-SOD is also likely to have an antioxidation role; it may also be a useful indicator of EC-SOD bioavailability in the artery wall. This hypothesis is supported by a recent finding that arterial wall EC-SOD expression was elevated in human and rabbit atherosclerotic lesions and that this is accompanied by increased inducible NO synthase expression.25 In our study, we found that EC-SOD levels were lower among patients with a history of MI. These depressed levels could result from reduced production of EC-SOD, thereby rendering an individual more susceptible to oxidative damage. It could also occur with an increased production of superoxide, which interacts either directly with circulating EC-SOD or with EC-SOD in the vascular wall, which is in equilibrium with plasma levels. Furthermore, among patients who had MI, higher EC-SOD levels tended to be associated with a later onset of MI. Although it was not statistically significant, patients with unstable angina also tended to have lower EC-SOD levels. Irrespective of the mechanism, our results suggest that EC-SOD could be protective in relation to the development of CAD and that plasma EC-SOD may serve as an indicator of the balance between the damaging effects and the bioscavenging capability of superoxide. This is further supported by our findings of the associations between plasma EC-SOD levels and other established CAD risk factors, particularly sex and smoking habits.
Although we observed no association between the plasma EC-SOD levels and CAD severity as assessed by the number of significantly diseased vessels, factors associated with CAD severity may not be the same as those associated with CAD occurrence. Free radicals could be involved more in the initiation than the progression of atherosclerotic lesions. Our data indicate that EC-SOD is relevant to the occurrence of MI, and our findings are consistent with the notion that increased levels are protective.26 27 28
A missense mutation (Arg213→Gly) has been found to be responsible for very high EC-SOD levels in a small proportion of healthy subjects.17 18 19 20 Our genotype analysis has also confirmed that every individual with levels >400 ng/mL (by our ELISA method) was heterozygous for the mutation. However, the genetic contribution to the population distribution of plasma EC-SOD is not known and is currently under investigation in our laboratory. Furthermore, it is also not known whether elevated plasma EC-SOD in individuals with the Arg213→Gly variant is associated with an increased or decreased superoxide scavenging capability and therefore its relevance to CAD risk.
Our study has demonstrated a wide-ranging effect of smoking on plasma EC-SOD, in that current smokers had low levels, 67% of nonsmokers’ levels, and there were intermediate concentrations in ex-smokers. This effect was also demonstrable in subgroups of high or low EC-SOD. Thus, there was an important general effect of smoking on plasma EC-SOD. Marklund and colleagues29 recently reported a somewhat smaller but similar smoking effect in the general population, in that smokers had 91% to 95% of the nonsmokers’ levels. Their survey did not take into account the graded effects of current smoking and ex-smoking on EC-SOD as observed in our study. Although the mechanism(s) for the smoking-induced low plasma EC-SOD levels is unknown, inhaled NO or O2− produced by cigarette smoking30 may decrease circulating EC-SOD or, alternatively, other components of smoking may downregulate EC-SOD production.
The high plasma EC-SOD levels we found in female patients (30% higher than men) in this population are consistent with the fact that women have a lower risk of CAD than men. The observation was independent of other demographic and measured risk factors in our patient population. This was also found recently by Marklund et al,29 who reported a significant difference (10% higher in women) in a large, healthy population. They also observed increased plasma EC-SOD levels with increasing age, as we did in our patient population. Furthermore, the negative associations between EC-SOD and triglycerides and the positive association with HDL-C we found in the patients are consistent with the findings of Marklund et al in the healthy population. However, the association they observed with waist-to-hip ratio was not significant in our patient population after control for sex and age in a multivariate analysis. Although we also found significant correlations with levels of apoA-I, creatinine, and glucose, we did not find in our patients the significant association between EC-SOD and body mass index identified in the healthy Swedish population. This could be because our patients were largely overweight. However, all of these observed positive and negative correlations are consistent with the notion that EC-SOD may be protective in relation to CAD.
However, it should also be noted that the averages and ranges of plasma EC-SOD levels in our patient population are ≈30% to 40% lower than those reported for a healthy Swedish population,20 29 most likely because of differences in the analytical methods used. The method we used was developed by Yamada et al19 and Adachi et al,21 23 and the EC-SOD levels in a Japanese population by this method were also lower than those reported for the Swedish group20 29 and not different from the levels in our patients. This difference is also reflected in the different cutoff levels (600 ng/mL for Swedish and 400 ng/mL for Japanese subjects) to adjust for high and low phenotypes.19 29 Of our patients, 19 (3.22%) were above the 400-ng/mL and 12 (2.03%) above the 600-ng/mL cutoff levels. Our percentage of patients in the high group is close to that reported by Marklund et al (3.8%) if the 400-ng/mL cutoff is used.
In summary, our study shows that in patients requiring investigation by coronary angiography, plasma EC-SOD is lower in men than in women; that smoking is associated with depressed levels in both sexes; and that low levels are independently associated with a history of MI. Although further studies are needed, eg, a follow-up of these patients, which is currently ongoing in our laboratory, our present findings are consistent with EC-SOD’s being associated with a reduced coronary risk.
This work was supported by a grant from the National Health and Medical Research Council of Australia. We wish to thank Lily Fenech, Shirley Brown, Steven Brouwer, Dr Greg Cranney, Dr Ahmed Farshid, and all the nurses in the Eastern Heart Clinic for their assistance in clinical data collection. We are also particularly grateful to Dr Roger Allan and the cardiologists in the Department for allowing us to study their patients.
Beckman JS, Koppenol WH. Nitric oxide, superoxide, and peroxynitrite: the good, the bad, and the ugly. Am J Physiol. 1996;271:C1424–C1437.
White CR, Brock TA, Chang L-Y, Crapo J, Briscoe P, Ku D, Bradley WA, Gianturco SH, Gore J, Freeman BA, Tarpey MM. Superoxide and peroxynitrite in atherosclerosis. Proc Natl Acad Sci U S A. 1994;91:1044–1048.
Beckman JS, Beckman TW, Chen J, Marshall PA, Freeman BA. Apparent hydroxyl radical formation by peroxynitrite: implications for endothelial injury from nitric oxide and superoxide. Proc Natl Acad Sci U S A. 1990;86:1620–1624.
Stralin P, Karlsson K, Johansson BO, Marklund SL. The interstitium of the human arterial wall contains very large amounts of extracellular superoxide dismutase. Arterioscler Thromb Vasc Biol. 1995;15:2032–2036.
Marklund SL. Human copper-containing superoxide dismutase of high molecular weight. Proc Natl Acad Sci U S A. 1982;79:7634–7638.
Weisiger RA, Fridovich I. Mitochondrial superoxide dismutase: site of synthesis and intramitochondrial localization. J Biol Chem. 1973;248:4793–4796.
Abrahamsson T, Brandt U, Marklund SL, Sjoqvist PO. Vascular bound recombinant extracellular superoxide dismutase type C protects against the detrimental effects of superoxide radicals on endothelium-dependent arterial relaxation. Circ Res. 1992;70:264–271.
Sandstrom J, Karlsson K, Edlund T, Marklund SL. Heparin-affinity patterns and composition of extracellular superoxide dismutase in human plasma and tissues. Biochem J. 1993;294:853–857.
Karlsson K, Marklund SL. Heparin-induced release of extracellular-superoxide dismutase to human blood plasma. Biochem J. 1987;242:55–59.
Folz RJ, Peno-Green L, Crapo JD. Identification of a homozygous missense mutation (Arg to Gly) in the critical binding region of the human EC-SOD gene (SOD3) and its association with dramatically increased serum enzyme levels. Hum Mol Genet. 1994;3:2251–2254.
Sandstrom J, Nilsson P, Karlsson K, Marklund SL. 10-fold increase in human plasma extracellular superoxide dismutase content caused by a mutation in heparin-binding domain. J Biol Chem. 1994;269:19163–19166.
Adachi T, Yamada H, Yamada Y, Morihara N, Yamazaki N, Murakami T, Futenma A, Kato K, Hirano K. Substitution of glycine for arginine-213 in extracellular-superoxide dismutase impairs affinity for heparin and endothelial cell surface. Biochem J. 1996;313:235–239.
Wang XL, Tam C, McCredie RM, Wilcken DEL. Determinants of severity of coronary artery disease in Australian men and women. Circulation. 1994;89:1974–1981.
Luoma JS, Stralin P, Marklund SL, Hiltunen TP, Sarkioja T, Yla-Herttuala S. Expression of extracellular SOD and iNOS in macrophages and smooth muscle cells in human and rabbit atherosclerotic lesions: colocalization with epitopes characteristic of oxidized LDL and peroxynitrite-modified proteins. Arterioscler Thromb Vasc Biol. 1998;18:157–167.
Homans DC, Asinger R, Pavek T, Crampton M, Lindstrom P, Peterson D, Bache RJ. Effect of superoxide dismutase and catalase on regional dysfunction after exercise-induced ischemia. Am J Physiol. 1992;263:H392–H398.