Superoxide Dismutase Activity in Serum of Patients With Acute Cerebral Ischemic Injury

Subjects and Methods

The antioxidative activity in serum was measured sequentially in 41 patients with acute ischemic stroke who were admitted to the Department of Neurology within 24 hours after onset of symptoms (13 men and 28 women, aged 28 to 91 years [mean, 62.8 years]). Twenty-four age-matched patients who were being treated at the same time in the neurological department for nonvascular diseases (eg, intervertebral disk protrusion, polyneuropathies, or muscular diseases) served as controls (10 men and 14 women, aged 35 to 81 years [mean, 60.3 years]).

The study was approved by the local ethics committee. Patients with (1) concomitant cardiac, renal, hepatic, or cancerous disease; (2) recent head trauma; (3) recent history of transient ischemic attacks; or (4) CT and/or MRI results inconclusive for the location of the ischemic lesion were excluded from this study. All patients were evaluated by CT and/or MRI on day 1 and on day 4 after stroke.

Clinical examination was performed on admission and daily thereafter and was scored according to the 58-point Scandinavian Stroke Scale (SSS).^{12 13} On admission and 10 days after stroke, the neurological deficit was assessed by the SSS.

Blood samples were drawn between 8 am and noon from indwelling venous catheters within the first 24 hours after stroke and again 72 hours, 5 days, and 10 days thereafter. Samples were immediately centrifuged (1500g, 10 minutes), and serum was stored at −80°C until assayed. Serum SOD activity was measured with a chemiluminometric method described in detail before.¹⁴ Briefly, xanthine oxidase (grade III from buttermilk, 420 mU/mL, Sigma), 0.1 mmol/L lucigenin (Sigma), and serum samples were diluted in 50 mmol/L potassium phosphate buffer, pH 10.0. Serum samples were diluted 1:10.⁴ At this concentration, the influence of natural reducing agents such as ascorbate or epinephrine is excluded in the assay,¹⁴ and the inhibition of chemiluminescence is caused by SOD. Mn-SOD (Sigma) dilutions or buffer was used as control. The superoxide-producing reaction was initiated by the automated dispensing of 60 μL of 1.45 mmol/L xanthine. All reagents were freshly prepared in 50 mmol/L potassium phosphate buffer, pH 10.0. The buffer and xanthine solutions were kept at room temperature, the other solutions on ice. Chemiluminescence was measured for 20 minutes in 1-minute cycles in BioLumat LB 9501. Mean blank values given by cuvettes without xanthine oxidase were subtracted from the peak counts per minute. Results are expressed as micrograms per milliliter SOD according to a standard curve with bovine Mn-SOD.

Results

Of the 41 patients with stroke, 9 had large hemispheric infarcts, in 8 patients due to cardiac embolism occluding the internal carotid artery and in 1 due to dissection of the internal carotid artery with supraocclusional middle cerebral artery embolism. Three of these patients died from transtentorial herniation during the 10 days of investigation. Nine patients showed a medium-sized infarct on cranial CT. Five of these patients had an occlusion of the internal carotid artery; in 4 of them the middle cerebral artery was occluded. Twenty-three patients had small cerebral infarcts due to either MCA branch occlusion or lacunar infarcts. Initial SSS ranged from 2 to 56 (mean, 25.98±14.3). Ten days after stroke, the mean SSS was 29.88±18.0 (range, 6 to 58).

The chemiluminescence triggered by xanthine/xanthine oxidase was significantly less inhibited by serum of patients after acute stroke compared with age-matched control patients (1489.12±28.7 versus 1335.9±32.6, mean±SEM; P<.034, Wilcoxon rank test), indicating less SOD activity in their serum. SOD activity was decreased most profoundly within 24 hours after onset of neurological deficits and recovered to values not different from control subjects within 5 days after stroke (Fig 1⇓). We found a significant correlation between the infarct volume on the cranial CT scan and the SOD activity in the serum obtained within 24 hours after stroke (P=.01, Spearman correlation). Patients with large infarcts (>150 cm³) had a higher chemiluminescence and thus a lower SOD activity than those with small infarcts (P<.001, Wilcoxon rank test) (Fig 2⇓). Accordingly, the degree of neurological deficit of stroke patients scored by SSS on admission correlated well with the SOD activity (r=.599, P<.001, Spearman correlation). In patients with low SSS scores on admission a higher chemiluminescence was observed (Fig 3⇓).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1.

Superoxide dismutase (SOD) activity in serum was measured as described in “Subjects and Methods.” Serum from control patients contained significantly more SOD activity than that of stroke patients measured within 24 hours after onset of symptoms. Five days after stroke, SOD activity returned to control values. Data are given as mean±SEM.*P=.034, Wilcoxon rank test.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2.

Patients with larger strokes had less free radical scavenging activity than those with smaller infarcts. Data are given as mean±SEM (*P<.005, **P<.001 vs control, Wilcoxon rank test).

• Download figure
• Open in new tab
• Download powerpoint

Figure 3.

Superoxide dismutase (SOD) activity in serum obtained within 24 hours after stroke correlated significantly (P=.001, Spearman correlation) with the neurological deficit scored by the Scandinavian Stroke Scale on admission (SSS-d1). Patients with severe neurological deficits had a lower free radical scavenging SOD activity.

Discussion

We investigated the SOD activity in the serum of patients with acute cerebral ischemia and observed a significant decrease within 24 hours after stroke compared with age-matched control subjects. This seems to be in contrast to previous studies, which, however, used a different methodological approach. In previous investigations of the concentration of cytoplasmic CuZn-SOD in the serum of patients with stroke, either no significant difference was found¹⁵ or there was an increase in plasma and cerebrospinal fluid.¹⁶ The method we used is specific for SOD, excluding other antioxidants as a cause for reduced chemiluminescence created by the xanthine/xanthine oxidase system.¹⁴ However, it measures the total SOD activity and does not differentiate between the different SOD isoforms. In extracellular fluids, including serum, extracellular SOD, which is secreted by endothelial cells, is the major SOD isozyme, and the activity of SOD in serum closely parallels the concentration of extracellular SOD.¹⁷ Therefore, mitochondrial Mn-SOD and cytoplasmic CuZn-SOD might not be as relevant in serum, and a strong reduction in extracellular SOD activity could account for the reduction in the antioxidative capacity we found in stroke patients. Strand and Marklund¹⁸ reported an increased CuZn-SOD activity in the cerebrospinal fluid and found a good correlation with the size of the infarct on cranial CT scan and functional impairment. They suggested that CuZn-SOD activity in the cerebrospinal fluid is a good marker for the extent of brain damage, since this small molecule leaks easily from the injured brain tissue into the cerebrospinal fluid. However, no serum analysis was performed. In our study decreased SOD activity was already apparent within 24 hours after onset of symptoms and recovered subsequently over the next few days. Therefore, it seems likely that antioxidants are depleted as a consequence of an excessive production of oxygen free radicals very early after the ischemic insult. The generation of superoxide anion usually occurs at the time of reperfusion. Since it is unlikely that the occluded arteries recanalized at the same time in all patients investigated in this study, reactive oxygen metabolites from other sources including activated granulocytes in the penumbra of the ischemic infarct may add to the observed depletion of radical scavenging enzymes. The low SOD activity in patients with large infarcts and poor outcome might reflect the increased amount of free oxygen radicals released from a severe ischemic injury. Thus, increasing the antioxidative capacity in serum within the first day after the onset of symptoms might be a therapeutic option to minimize the oxidative injury caused by oxygen free radicals until the endogenous free radical scavenging systems recover. In animal models of cerebral ischemia, polyethylene glycol–conjugated or liposome-entrapped CuZn-SOD was used successfully to reduce brain injury by ischemia and reperfusion. Infarct volume^{4 19} and postischemic blood-brain barrier permeability^{4 20} were reduced after administration of SOD in piglets that had been subjected to focal cerebral ischemia. Further evidence for the important role of SOD in the defense of free oxygen radical damage in reperfusion injury came from transgenic animal experiments. Mice overexpressing SOD were highly resistant to reperfusion injury and damage of the blood-brain barrier.^{5 21 22} The decrease in the infarct volume in transgenic mice correlated well with an improved neurological outcome, whereas mice with a knockout mutation for SOD had larger infarcts after focal ischemia.⁶ Supplementation of depleted antioxidative capacity thus may also be beneficial in stroke patients.