Temporal changes in superoxide dismutase, glutathione peroxidase ...

5 downloads 40 Views 1MB Size Report
Authors; Authors and affiliations. Sahebarao P. Mahadik; Tapas K. Makar; Jayasimha N. Murthy; Astrid Ortiz; Chandramohan G. Wakade; Steven E. Karpiak.

©Copyright 1993 by Humana Press Inc. All rights of any nature whatsoever reserved. 1044-7393/93/1801-02-0001 $02.80

Temporal Changes in Superoxide Dismutase, Glutathione Peroxidase, and Catalase Levels in Primary and Peri-Ischemic Tissue Monosialoganglioside (GM 1) Treatment Effects SAHEBARAO P. MAHADIK,*• 2.3 TAPAS K. MAKAR,' JAYASIMHA N. MURTHY,' ASTRID ORTIZ,' CHANDRAMOHAN G. WAKADE,' AND STEVEN E. KARPIAK''

2

'Division of Neuroscience, New York State Psychiatric Institute; Departments of Psychiatry, Biochemistry, and Molecular Biophysics, College of Physicians and Surgeons, Columbia University, New York; 3 Present address: Department of Psychiatry and Health Behavior, Medical College of Georgia, Augusta, GA 30912

2

Received December 9, 1991; Accepted March 27, 1992

ABSTRACT Time-dependent changes in levels of the antioxidant enzymes, superoxide dismutasae (SOD), glutathione peroxidase (GSHPOD), and catalase (CAT) after cortical focal ischemia in rat indicate that: (1) primary and peri-ischemic tissues differ in both rate and the magnitude of oxyradical-induced ischemic injury, and (2) ischemic tissue remains vulnerable to oxyradical damage as long as 72 h after ischemia since the antioxidant enzyme levels remain at or below basal levels. After 72 h, the increased levels of these enzymes are sufficient to protect tissue against oxyradical damage. GM1 ganglioside (10 mg/kg, im) further increased the already elevated levels of the enzymes after ischemia, thereby indicating the GM1 treatment increases the capacity of ischemic tissue to protect against oxyradical injury. *Author to whom all correspondence and reprint requests should be addressed.

Molecular and Chemical Neuropathology

I

Vol. 18, 1993

2

Mahadik et al. Index Entries: CNS ischemia; GMI ganglioside; oxyradical enzymes; superoxide dismutase; glutathione peroxidase; catalase; antioxidant defense; lipid peroxidation.

INTRODUCTION Central nervous system (CNS) injury as a result of either hemorrhagic or ischemic stroke is initiated by a series of biochemical alterations in the cellular microenvironment, which include ionic imbalances, cellular acidosis, altered phospholipid metabolism, increases in free fatty acids, oxygen free radical production, and energy failure (for review, see Bazan, 1976; Nemoto, 1978; Siesjo, 1981; Raichle, 1983; Shui et al., 1983; Mahadik and Karpiak, 1988; Braughler and Hall, 1989; Karpiak et al., 1990a). These processes, if not prevented or ameliorated during their initial stages, can lead to cell death. It is known that aerobic metabolism following ischemia, particularly after reperfusion, generates high levels of toxic oxygen derivatives (oxyradicals): the superoxide (O Z the hydroxyl radical (OH ), and hydrogen peroxide (H 2 0 2 Freeman and Crapo, 1982; Cao et al., 1988; Floyd, 1990). These oxyradicals induce membrane lipid peroxidation, resulting in reduced membrane fluidity, increased membrane permeability to ions, the inactivation of several membrane enzymes, DNA strand breaks, and finally cell death (Demopoulos et al., 1980; Leibovitz and Siegal, 1980; Braughler and Hall, 1989). It is hypothesized that the CNS is highly vulnerable to oxyradical injury since the brain is enriched in substrates for lipid peroxidation, particularly unsaturated fatty acids (Cohen, 1982). In normal physiological conditions, cells are protected from oxyradical toxicity by the synergistic action of oxyradical metabolizing (antioxidant) enzymes. An increase in these enzymes above basal levels occurs when oxyradical levels increase. Therefore, the levels of these antioxydant enzymes reflect the magnitude of the cellular defense response (oxyradical enzyme metabolism) against injury. We undertook a study of the time-dependent changes in the levels of the three oxyradical metabolizing enzymes in ischemic injury to assess the contribution of oxyradical injury to the pathophysiology of CNS stroke. Such data could assist in identifying those biochemical process that might be amenable to therapeutic intervention. Evidence that describes oxyradical-associated injury in stroke is based on studies of increased levels of lipid peroxidation, which lead to losses of membrane structure and function (Braughler and Hall, 1989). Those studies have prompted treatment strategies (lazaroids) aimed at achieving tissue protection by inhibition of lipid peroxidation (for reviews, see Braughler and Hall, 1989; Hall and Braughler, 1989). However, measures of lipid peroxidation are problematic because of methodological limitations, and assay - ),

-

) (

Molecular and Chemical Neuropathology

Vol. 18, 1993

GMI Ganglioside Increases Antioxidant Enzymes

3

imprecision because of the high turnover of lipid peroxides. Consequently, lipid peroxidation levels do not adequately reflect the complex physiological response (defense) of injured CNS issue. Levels of the three antioxidant enzymes associated with oxyradical metabolism have not been detailed in CNS ischemic tissue. Since these enzymes must be made de novo or protected from degradation, their levels can be useful in assessing the metabolic potential of a cell to respond to injury. The temporal pattern of their increases should delineate those periods of maximal vulnerability to, and optimal defense against, oxyradical injury after the ischemic episode. We have been studying ischemic injury (cortical focal ischemic) in the rat (Tamura et al., 1981; Chen et al., 1986; Karpiak et al., 1990b, 1991) and have described the time-dependent changes that occur in edema formation and tissue ions changes (Na+, K+, and Ca++) (Karpiak et al., 1991), membrane fatty acids (Mahadik et al., 1989, 1991), and plasma membrane ATPases (Mahadik et al., 1992) in primary and peri-ischemic tissue. The ATPase activity losses are indicators of altered plasma membrane structure and failure of function (Tanaka and Teruya, 1973; Schwartz et al., 1976; Kimelberg, 1977). This membrane failure is partially a consequence of lipid peroxidation. We have found that these structural and functional membrane losses in ischemic pathology parallel behavioral deficits (Ortiz et al., 1990; Bharucha et al., 1991). These biochemical and behavioral pathologies can be significantly ameliorated by acute treatment with monosialoganglioside (GM1 ganglioside). Since GM1's neuroprotective mechanism has not been identified, its effects on in vivo antioxidant enzymes and lipid peroxidation should prove valuable. We report here the time-dependent (3, 24, 48, and 72 h and 2 wk) changes in levels of SOD, CAT, and GSHPOD levels in primary and periischemic cerebral cortical tissue and the effects of GM1 ganglioside treatment on those enzyme levels.

METHODS Cortical Focal Ischemia The procedure has been described in detail (Chen et al., 1986; Karpiak et al., 1990b, 1991). Male Sprague-Dawley rats (265±25 g) were allowed ad libitum food and water up to 3 h before surgery. Rats are anesthetized (im) with Ketamine HCl (87.5 mg/kg) and Rompun (7.5 mg/kg). Surgery is done using a Nikon type 102 stereo-microscope with fiber optic bifurcated illumination and ring lens illumination. The entire procedure is done on a heated (37°C) operating table (Harvard Apparatus, South Natick, MA) to maintain normal body temperature. A 2-cm midline cervical incision is made and the skin is retracted. Submaxillary salivary glands

Molecular and Chemical Neuropathology

Vol. 18, 1993

4

Mahadik et al.

are separated. Both common carotid arteries (CCAs) are separated from their accompanying vagus and sympathetic nerves. Silk sutures (3-0) are placed over the ipsilateral (left) CCA. Using a saline cooled dentral drill, a 2 x 2 mm craniectomy is performed 1 mm superior and rostral to the junction of the zygoma and squamosal bones. The middle cerebral artery (MCA) is distinguished from its accompanying veins by having fewer branches and being more straight. The dura is dissected using a fine needle. The ipsilateral (left) CCA is ligated next. The MCA is coagulated (Butcher Hyfrecator Model 733: setting =40) with biopolar radiofrequency forceps (Tieman No. 160-1841) 1 mm below the rhinal fissure, and, at the bifurcation of the MCA (4 mm above the rhinal fissure). The MCA is cut with microscissors at the two points of occlusion. Gelfoam is placed into the craniectomy, the temporalis muscle is allowed to fall back into position and the skin is sutured. A nontraumatic micro-aneurysm clip (BRI34-3550) is applied for 1 h to the contralateral CCA and then removed. If CCA blood flow is not restored, the animal is discarded. Controls (sham operated) are rats in which the surgery is terminated immediately after dissection of the dura.

GM1 Ganglioside Treatment Animals were injected (im) within 10 min of the MCAo + CCAo with either saline or ganglioside GM1 (Fidia Research Laboratories). Subsequent injections were given after 24, 48, and 72 h. The dose of GM1 (10 mg/kg) was chosen since in earlier studies using this model we found it to be optimally effective in reducing biochemical and behavioral dysfunctions (Ortiz et al., 1990; Bharucha et al., 1991; Karpiak et al., 1991; Mahadik et al., 1992). Six animals were used in each experimental group. The effects of GM1 treatment were assayed at 24 h, 72 h, and 2 wk after the MCAo + CCAo.

Preparation of Tissue Extracts The cerebral hemispheres were dissected and two 7-mm diameter core punches (cortex only, excluding white matter) were sampled from the primary ischemic area (Area 1, parietal cortex) and peri-ischemic area (Area 2, occipito-temporal cortex). The corresponding contralateral Area 1 was dissected (punch) as a control (Karpiak et al., 1991). After weighing the tissue (60±3 mg wet wt) the enzymes were extracted (>95%) by homogenizing tissue in 30 vol of buffer (0.1M phosphate buffer, EDTA, 0.1 mM, pH 7.8) followed by quick freezing and thawing. The enzyme extract was collected by centrifugation at 100,000g. SOD activity was assayed after removal of interfering substances by passage through Sephadex G-10 columns. GSHPOD and CAT activities were assayed using the re-

Molecular and Chemical Neuropathology

Vol. 18, 1993

GM1 Ganglioside Increases Antioxidant Enzymes

5

maining extract. All enzyme activities were expressed as units per milligrams of total protein. Protein was determined in the total homogenate by the method of Lowry et al. (1951). Tissues were also obtained and assayed from animals perfused (transcardially) with phosphate buffered saline. This was done in order to remove blood from the tissue since the oxyradical enzyme levels are high in erythrocytes (six shams and six ischemic rats were used at 72 h post MCAo + CCAo).

Antioxidant Enzyme Assays SOD was determined using a slightly modified spectrophotometric procedure (Nishikimi et al., 1972; Fried, 1975). SOD activity is determined from the degree of inhibition of NBT reduction by superoxides that are produced by nonenzymatic reduction (NADH-phenozine methosulfate, PMS-system) under aerobic conditions. The cytochrome C oxidase does not interfere in this assay procedure. Extract (300 µL) was added to a 1.5-mL reaction mixture (200 mM phosphate buffer, pH 7.8; 1 mM diethyenetriamine pentaacetic acid [DETPAC]; 0.41 mM nitroblue tetrazolium [NBT]; 1.1 mM phenazine methosulfate; 1 mg bovine serum albumin; and 50 mM NADH) and read at 540 nm after 25 min (formazan formed as a result of NBT reduction). SOD activity was determined from the percent NBT reduction in presence of sample, using the plot of percent NBT reduction with increasing concentrations of standard reference SOD (3300 U/mg protein, Sigma Chemical Co., St. Louis, MO). Increasing amounts of SOD (0.25-2.5 U) inhibited linearly NBT reduction. Activity was expressed as standard SOD U/mg protein. GSHPOD was determined using a 50-µL extract in a 1-mL reaction mixture (50 mM phosphate buffer, pH 7.0; 1 mM DETPAC; 0.24 U of glutathione reductase; 10 mM glutathione; 1.5 mM NADPH; and 12 mM t-butyl hydroperoxide) (Flohe and Gumzler, 1984). Activity was expressed as units/mg protein (1 U =1 nmol of NADPH oxidized/min, monitored by absorption at 340 nm at 37°C. CAT was also determined with 50 µL of extract of 1 mL of reaction mixture (50 mM phosphate buffer, pH 7.0 and 59 mM hydrogen peroxide) (Aebi, 1984). CAT activity was expressed as units/mg protein (1 U =1 umol of hydrogen peroxide used/min, monitored by absorption at 240 nm at 25°C). All data were analyzed by an ANOVA (TIME) with all statistical posthoc comparisons (Scheffe') made by assessing whether each time-point was different (*p< 0.05; p< 0.01) from sham-controls (basal enzyme activity levels). GM1 effects were assessed by ANOVA (Treatment x Time) with post-hoc (Scheffe') tests comparing GM1 effects to enzyme levels at each time-point.

Molecular and Chemical Neuropathology

Vol. 18, 1993

Mahadik et al.

6

Peri - Ischemic

Ischemic 4

4

Y3

3

rn E2

2

C 0

a C

1 n

- Cont 3hrs 24hr 48hr 72hr 2wks

Cont 3hrs 24hr 48hr 72hr 2wks

Days After Ischemia

Days After Ischemia

Fig. 1. Time-dependent changes (mean ± SEM) in levels of SOD in ischemic and peri-ischemic tissue. SOD activity is expressed as standard SOD units. All statistical differences are based on comparisons of each enzyme level to basal levels. GM1 levels are compared to the enzyme level for the time-point studied (*p