The Japanese Journal of Psychiatry and Neurology, Vol. 46, No. 4, 1992
Synaptosomal Membrane Fluidity, Lipid Peroxidation and Superoxide Dismutase Activity in the Brain of Amygdala-Kindled Rats Takashi Itoh, M.D., Kazufumi Akiyama, M.D., Midori Hiramatsu, Ph.D.* and Saburo Otsuki, M.D. Department of Neuropsychiatry, Okayama University Medical School, Okayama * Yamagata Technopolis Foundation, Yamagata
Abstract : Synaptosomal membrane fluidity, lipid peroxide (LPO) and cytosolic superoxide dismutase (SOD)activity were examined in various brain regions (amygdala, hippocampus, striatum and frontal cortex) of amygdala-kindled rats. At 24 h after the last seizure, a significant increase of membrane fluidity was observed in all the regions examined, whereas the LPO level was sigdflcantly decreased in the four regions with enhanced activity of cytosolic SOD.At 7 days after the last seizure, membrane fluidity was decreased only in the hippocampus. At 6 weeks after the last seizure, there were no changes in membrane fluidity between control and kindled rats. These results suggest that membrane fluidity and lipid peroxidation are modulated transiently by a kindled seizure, but not at a steady state of kindling with enduring seizure susceptibility. Key Words: kindling, lipid peroxide, thiobarbituric acid reaction, SOD, synaptosomal membrane fluidity, ESR, epilepsy Jpn J Psychiatr Neurol46: 957466,1992
INTRODUCTION
The biological membrane is not only a constituent of cellular structures, but is also involved in the modulation of signal transmission and membrane functions. According to the fluid mosaic model, the biological membrane consists of a lipid bilayer and membrane proteins. The lipid bilayer provides a physical and biochemical environment in which the membrane proteins can function. Received for publication on July 24, 1992. Mailing address: Takashi Itoh, M.D., Department of Neuropsychiatry, Okayama University Medical School, 2-5- 1, Shikata-cho, Okayama 700, Japan.
Thus, membrane fluidity, one of the properties of the lipid bilayer, is closely associated with membrane enzyme activity, ion conductance and membrane receptor activity. Therefore, it is likely that membrane fluidity plays an important role in the modulation of a variety of membrane functions. Kindling is a phenomenon whereby a progressively increasing epileptiform discharge emerges after repeated electrical stimulation of a limbic brain area, finally culminating in a generalized seizure.' * 26 This phenomenon has been well established not only as a model of human complex partial seizure with secondary generalization, but also as a strategic tool for studying neuronal plasticity.
T. Itoh ef al.
958
Several lines of evidence have suggested that membrane fluidity is altered in epilepsy models, such as El mice." Furthermore, drugs with anticonvulsant action, such as diazepam2', valproic acid2' and proconvulsants, such as pentylenetetrazol (PTZ), may affect membrane fl~idity'~. In the present study, we examined membrane fluidity, lipid peroxidation, superoxide, dismutase in the brain of kindled rats. SUBJECTS AND METHODS
Chemicals 5- and 16-doxylstearic acid ( 5 - and 16DS), superoxide dismutase (SOD) from bovine erythrocytes, hypoxanthine (HPX) and diethylenetriaminepenta acetic acid (DETAPAC) were obtained from Sigma Chemical Co. (St. Louis, Mo.). 5,5-Dimethyl-l-pyrroline N-oxide (DMPO) was obtained from Daiichi Pure Chemical Co. (Tokyo) and xanthine oxidase (XOD) was obtained from Boehringer Mannheim GmbH (Germany). All other chemicals and reagents were of the highest grade available from commercial suppliers.
Preparation of Amygdala-Kindled Rats Male Sprague-Dawley rats (Charles River Japan) weighing 280-350 g at the time of surgery were used. The animals were housed with free access to food and water under a 12-h light/l2-h dark cycle. A twisted tripolar stimulating/recording electrode was implanted stereotaxically into the left amygdala (AM) under pentobarbital anesthesia (50mg/ kg, i.p.) using the stereotaxic coordinates (AP: 0, L: 5.0 and D: 8.0 mm), in the atlas of Pellegrino et aLZ4A screw electrode was placed on the right frontal skull as a reference. At least one week aftkr the implantation, electrical stimulation was started. The stimulus current consisted of a 1-s train of sine wave pulses of 1 ms at 60 Hz and a current intensity of 200 pA.Kindling seizure development was assessed according to
Racine's classification.z6 Electrical stimulation was administered once daily until every animal manifested at least 5 consecutive generalized seizures at stage 5 in every experiment. Control animals underwent an electrode implantation in the left AM, but were not stimulated. Measurement of Synaptosomal Membrane Fluidity The AM-kindled rats were decapitated either 24 h, 7 days or 6 weeks after the last kindled seizure, with the matched sham controls. The brains were removed immediately and divided into four regions (amygdala, hippocampus, striatum and frontal cortex) on a plate over ice. All samples were weighed and added to 20 volumes of a 0.32 M sucrose buffer solution containing 1 mM EDTA and 10 mM Tris-HCI (pH 7.4). Synaptosomal fractions, which were prepared according to the method of Zaleska et al?, were washed by centrifugation in the 0.32 M sucrose buffer at 17,000X g for 10 min. The final pellets were suspended in the 0.32 M sucrose buffer and stored for a maximum of 4 weeks at - 80 "C until analysis of membrane fluidity. 5-DS and 16-DS, N-oxyl-4, 4'-dimethyloxazolidine derivatives of stearic acid with the doxy1 groups in the C-5 and (2-16 positions, respectively, were employed as the spin label probes (Fig. 1-c). They were stored in a freezer at - 20 "C as stock solutions containing 10 mg/125 ml ethanol. Before spin-labeling, 20 pl of the stock solution was put into a glass test tube, and dried with a stream of nitrogen. The 100 pl of the synaptosomal fraction were added and mixed well with a vortex mixer for 1 min. Thirty microliters of the sample were put into a capillary tube, and each spin label parameter was assayed using an electron spin resonance (ESR) spectrometer (JES-FElXG, JEOL Ltd., Tokyo) according to the methods of Nagy et ~ 1Conditions . ~ ~ of ESR spectrometry were as follows; for 16DS, magnetic field, 327.7*5 mT; modulation, 0.2 mT; power, 4 mW; response, 0.3 s; amplitude, 7.9 X 100, sweep time, 4 min; tem-
Kindling and Membrane Fluidity perature, 37°C; for 5-DS, magnetic field, 327.7f 10 mT; modulation, 0.63 mT; power, 4 mW; response, 1 s; amplitude, 1.6X 1,OOO; sweep time, 4 min; temperature, 37 "C. Calculation of Parameters for Evaluation of Membrane Fluidity The order parameter, S,which was used to evaluate the membrane fluidity of 5-DS, showed anisotropic motion and was calculated using the following eq~ation:'~
S=
TI/ - T I Tzz - ( (Txx Tyy)/2)
+
.- an a"
where Txx (6.3 G), Tyy (5.8 G) and Tzz (33.6 G) are the hyperfine principle values of the nitroxide radical, and
+ +
aN = (Txx Tyy Tzz)/3 aN'=(T// +2T1)/3 T /I and T I were measured as shown in Fig. l-a. The order parameter, S,was found in the range 0.0 to 1.0. A reduction of S indicates an increase in the mobility of the spin label, that is, an increase of membrane fluidity. In the deep fluid regions of the membrane labeled with 16-DS, S could not be measured, since this probe showed a rather isotropic motion. Therefore, the apparent rotational correlation time, to,was obtained according tothe following equation':
where AHo is the peak width of the central signal, and h(o, and h(-l, are the peak heights of the signals at central and higher magnetic fields, respectively, as shown in Fig. l-b. As to is proportional tothe microviscosity of the membrane, there is a negative correlation between t, and membrane fluidity. Preparation for Thiobarbituric Acid Reaction and Measurement of SOD Activity The AM-kindled rats were decapitated either 24 h or 7 days after the last kindled
959
seizure, with the matched sham controls. The brains were removed immediately and divided into four regions (amygdala, hippocampus, striatum and frontal cortex) on a plate over dry ice. Each sample was placed in a micro test tube, quickly fro en with liquid nitrogen and stored at -80 for a maximum of 2 weeks until assay. Samples were weighed and added to 10 volumes of an icecold 0.1 M phosphate buffer (pH 7.8). The samples were then homogenized using a Teflon-glass homogenizer. One hundred microliters of the homogenate for the thiobarbituric acid reaction were removed, and the remainder was centrifuged at 105,000X g for 60 min. The supernatant containing the cytosol fraction was used for measurement of
\
SOD. Thiobarbituric Acid Reaction (Ohkawa, et a 1 . p The lipid peroxide (LPO) values were determined by measuring the accumulation of thiobarbituric acid (TBA)-reactive substances or malondialdehyde, since TBA-reactive substances have been reported to arise from lipid peroxides of polyunsaturated fatty acids with three or more double bonds by heating under acidic conditions. The reaction mixture contained 100 ,ul of homogenate sample, 200 ,u 1 of 8.1% sodium dodecyl sulfate (SDS), 1.5 ml of 20% acetic acid buffer (pH 3.5) and 1.5 ml of 1% TBA. The mixture was finally made up to 4.0 ml with phosphate buffer, and heated at 95 "C for 60 min with a hot-plate bath. After cooling quickly with ice-cold water, 1.0 ml of phosphate buffer and 5.0 ml of a mixture of nbutanol and pyridine (151, v/v) were added, and the mixture was mixed vigorously with a vortex mixer. After centrifugation at 4,000 rpm for 10 min, the fluorescence of the organic upper layer was measured using a Florence Spectrophotometer Model 650-10s (Hitachi Ltd., Tokyo). The assay was performed at wavelengths of 515 nm and 553 nm for excitation and emission, respectively. 1,1,3,3Tetraethoxypropane was used as an internal
T. Itoh
960
standard,
et
al. a
A
Measurement of SOD Activity Wiramatsu, et ai.)14 To each fraction was added 4 volumes of phosphate buffer. Then, 50 pl of 2 mM HPX, 35 pl of 5.5 mM DETAPAC, 50 pl of the diluted sample, 15 pl of DMPO and 50 pl of XOD were added to a test tube and mixed with a vortex mixer. The solution was then placed in a special flat cell (volume 160 pl, Labotec Co., Tokyo) and incubated for 1 min. The DMPO-0; spin adduct was analyzed by ESR spectrometry. A standard curve was made using 3.125 to 50 unit/ml SOD, and manganese oxide was used as an internal standard. Conditions of ESR spectrometry for measurement of SOD activity were as follows: magnetic field, 335.7f 5 mT; power, 4 mW; response, 0.3 s; modulation, 0.2 mT; amplitude, 3.2X 1,OOO; sweep time, 2 min; temperature, room temperature. The protein concentration was measured using Bio-Rad Protein Assay kits (Richmond, CA). Statistics The results were usually expressed as the mean fstandard deviation, and statistical significance was determined by the Student's t test (two tailed). RESULTS
Synaptosomal Membrane Fluidity Fig. 1-a, b shows the ESR spectra of spin labels embedded in the synaptosomal membrane. The surface and deep layers of the membranes were probed with 5-DSand 16DS,respectively (Fig. 1-c) . Figs. 2-a and 3-a show synaptosomal membrane fluidity in the brain of control and kindled rats at 24 h after the last seizure. The rotational correlation time, to, at the hydrophobic core was significantly decreased in the amygdala, hippocampus and frontal cortex (p
Synaptosomal Membrane Fluidity, Lipid Peroxidation and Superoxide Dismutase Activity in the Brain of Amygdala-Kindled Rats Takashi Itoh, M.D., Kazufumi Akiyama, M.D., Midori Hiramatsu, Ph.D.* and Saburo Otsuki, M.D. Department of Neuropsychiatry, Okayama University Medical School, Okayama * Yamagata Technopolis Foundation, Yamagata
Abstract : Synaptosomal membrane fluidity, lipid peroxide (LPO) and cytosolic superoxide dismutase (SOD)activity were examined in various brain regions (amygdala, hippocampus, striatum and frontal cortex) of amygdala-kindled rats. At 24 h after the last seizure, a significant increase of membrane fluidity was observed in all the regions examined, whereas the LPO level was sigdflcantly decreased in the four regions with enhanced activity of cytosolic SOD.At 7 days after the last seizure, membrane fluidity was decreased only in the hippocampus. At 6 weeks after the last seizure, there were no changes in membrane fluidity between control and kindled rats. These results suggest that membrane fluidity and lipid peroxidation are modulated transiently by a kindled seizure, but not at a steady state of kindling with enduring seizure susceptibility. Key Words: kindling, lipid peroxide, thiobarbituric acid reaction, SOD, synaptosomal membrane fluidity, ESR, epilepsy Jpn J Psychiatr Neurol46: 957466,1992
INTRODUCTION
The biological membrane is not only a constituent of cellular structures, but is also involved in the modulation of signal transmission and membrane functions. According to the fluid mosaic model, the biological membrane consists of a lipid bilayer and membrane proteins. The lipid bilayer provides a physical and biochemical environment in which the membrane proteins can function. Received for publication on July 24, 1992. Mailing address: Takashi Itoh, M.D., Department of Neuropsychiatry, Okayama University Medical School, 2-5- 1, Shikata-cho, Okayama 700, Japan.
Thus, membrane fluidity, one of the properties of the lipid bilayer, is closely associated with membrane enzyme activity, ion conductance and membrane receptor activity. Therefore, it is likely that membrane fluidity plays an important role in the modulation of a variety of membrane functions. Kindling is a phenomenon whereby a progressively increasing epileptiform discharge emerges after repeated electrical stimulation of a limbic brain area, finally culminating in a generalized seizure.' * 26 This phenomenon has been well established not only as a model of human complex partial seizure with secondary generalization, but also as a strategic tool for studying neuronal plasticity.
T. Itoh ef al.
958
Several lines of evidence have suggested that membrane fluidity is altered in epilepsy models, such as El mice." Furthermore, drugs with anticonvulsant action, such as diazepam2', valproic acid2' and proconvulsants, such as pentylenetetrazol (PTZ), may affect membrane fl~idity'~. In the present study, we examined membrane fluidity, lipid peroxidation, superoxide, dismutase in the brain of kindled rats. SUBJECTS AND METHODS
Chemicals 5- and 16-doxylstearic acid ( 5 - and 16DS), superoxide dismutase (SOD) from bovine erythrocytes, hypoxanthine (HPX) and diethylenetriaminepenta acetic acid (DETAPAC) were obtained from Sigma Chemical Co. (St. Louis, Mo.). 5,5-Dimethyl-l-pyrroline N-oxide (DMPO) was obtained from Daiichi Pure Chemical Co. (Tokyo) and xanthine oxidase (XOD) was obtained from Boehringer Mannheim GmbH (Germany). All other chemicals and reagents were of the highest grade available from commercial suppliers.
Preparation of Amygdala-Kindled Rats Male Sprague-Dawley rats (Charles River Japan) weighing 280-350 g at the time of surgery were used. The animals were housed with free access to food and water under a 12-h light/l2-h dark cycle. A twisted tripolar stimulating/recording electrode was implanted stereotaxically into the left amygdala (AM) under pentobarbital anesthesia (50mg/ kg, i.p.) using the stereotaxic coordinates (AP: 0, L: 5.0 and D: 8.0 mm), in the atlas of Pellegrino et aLZ4A screw electrode was placed on the right frontal skull as a reference. At least one week aftkr the implantation, electrical stimulation was started. The stimulus current consisted of a 1-s train of sine wave pulses of 1 ms at 60 Hz and a current intensity of 200 pA.Kindling seizure development was assessed according to
Racine's classification.z6 Electrical stimulation was administered once daily until every animal manifested at least 5 consecutive generalized seizures at stage 5 in every experiment. Control animals underwent an electrode implantation in the left AM, but were not stimulated. Measurement of Synaptosomal Membrane Fluidity The AM-kindled rats were decapitated either 24 h, 7 days or 6 weeks after the last kindled seizure, with the matched sham controls. The brains were removed immediately and divided into four regions (amygdala, hippocampus, striatum and frontal cortex) on a plate over ice. All samples were weighed and added to 20 volumes of a 0.32 M sucrose buffer solution containing 1 mM EDTA and 10 mM Tris-HCI (pH 7.4). Synaptosomal fractions, which were prepared according to the method of Zaleska et al?, were washed by centrifugation in the 0.32 M sucrose buffer at 17,000X g for 10 min. The final pellets were suspended in the 0.32 M sucrose buffer and stored for a maximum of 4 weeks at - 80 "C until analysis of membrane fluidity. 5-DS and 16-DS, N-oxyl-4, 4'-dimethyloxazolidine derivatives of stearic acid with the doxy1 groups in the C-5 and (2-16 positions, respectively, were employed as the spin label probes (Fig. 1-c). They were stored in a freezer at - 20 "C as stock solutions containing 10 mg/125 ml ethanol. Before spin-labeling, 20 pl of the stock solution was put into a glass test tube, and dried with a stream of nitrogen. The 100 pl of the synaptosomal fraction were added and mixed well with a vortex mixer for 1 min. Thirty microliters of the sample were put into a capillary tube, and each spin label parameter was assayed using an electron spin resonance (ESR) spectrometer (JES-FElXG, JEOL Ltd., Tokyo) according to the methods of Nagy et ~ 1Conditions . ~ ~ of ESR spectrometry were as follows; for 16DS, magnetic field, 327.7*5 mT; modulation, 0.2 mT; power, 4 mW; response, 0.3 s; amplitude, 7.9 X 100, sweep time, 4 min; tem-
Kindling and Membrane Fluidity perature, 37°C; for 5-DS, magnetic field, 327.7f 10 mT; modulation, 0.63 mT; power, 4 mW; response, 1 s; amplitude, 1.6X 1,OOO; sweep time, 4 min; temperature, 37 "C. Calculation of Parameters for Evaluation of Membrane Fluidity The order parameter, S,which was used to evaluate the membrane fluidity of 5-DS, showed anisotropic motion and was calculated using the following eq~ation:'~
S=
TI/ - T I Tzz - ( (Txx Tyy)/2)
+
.- an a"
where Txx (6.3 G), Tyy (5.8 G) and Tzz (33.6 G) are the hyperfine principle values of the nitroxide radical, and
+ +
aN = (Txx Tyy Tzz)/3 aN'=(T// +2T1)/3 T /I and T I were measured as shown in Fig. l-a. The order parameter, S,was found in the range 0.0 to 1.0. A reduction of S indicates an increase in the mobility of the spin label, that is, an increase of membrane fluidity. In the deep fluid regions of the membrane labeled with 16-DS, S could not be measured, since this probe showed a rather isotropic motion. Therefore, the apparent rotational correlation time, to,was obtained according tothe following equation':
where AHo is the peak width of the central signal, and h(o, and h(-l, are the peak heights of the signals at central and higher magnetic fields, respectively, as shown in Fig. l-b. As to is proportional tothe microviscosity of the membrane, there is a negative correlation between t, and membrane fluidity. Preparation for Thiobarbituric Acid Reaction and Measurement of SOD Activity The AM-kindled rats were decapitated either 24 h or 7 days after the last kindled
959
seizure, with the matched sham controls. The brains were removed immediately and divided into four regions (amygdala, hippocampus, striatum and frontal cortex) on a plate over dry ice. Each sample was placed in a micro test tube, quickly fro en with liquid nitrogen and stored at -80 for a maximum of 2 weeks until assay. Samples were weighed and added to 10 volumes of an icecold 0.1 M phosphate buffer (pH 7.8). The samples were then homogenized using a Teflon-glass homogenizer. One hundred microliters of the homogenate for the thiobarbituric acid reaction were removed, and the remainder was centrifuged at 105,000X g for 60 min. The supernatant containing the cytosol fraction was used for measurement of
\
SOD. Thiobarbituric Acid Reaction (Ohkawa, et a 1 . p The lipid peroxide (LPO) values were determined by measuring the accumulation of thiobarbituric acid (TBA)-reactive substances or malondialdehyde, since TBA-reactive substances have been reported to arise from lipid peroxides of polyunsaturated fatty acids with three or more double bonds by heating under acidic conditions. The reaction mixture contained 100 ,ul of homogenate sample, 200 ,u 1 of 8.1% sodium dodecyl sulfate (SDS), 1.5 ml of 20% acetic acid buffer (pH 3.5) and 1.5 ml of 1% TBA. The mixture was finally made up to 4.0 ml with phosphate buffer, and heated at 95 "C for 60 min with a hot-plate bath. After cooling quickly with ice-cold water, 1.0 ml of phosphate buffer and 5.0 ml of a mixture of nbutanol and pyridine (151, v/v) were added, and the mixture was mixed vigorously with a vortex mixer. After centrifugation at 4,000 rpm for 10 min, the fluorescence of the organic upper layer was measured using a Florence Spectrophotometer Model 650-10s (Hitachi Ltd., Tokyo). The assay was performed at wavelengths of 515 nm and 553 nm for excitation and emission, respectively. 1,1,3,3Tetraethoxypropane was used as an internal
T. Itoh
960
standard,
et
al. a
A
Measurement of SOD Activity Wiramatsu, et ai.)14 To each fraction was added 4 volumes of phosphate buffer. Then, 50 pl of 2 mM HPX, 35 pl of 5.5 mM DETAPAC, 50 pl of the diluted sample, 15 pl of DMPO and 50 pl of XOD were added to a test tube and mixed with a vortex mixer. The solution was then placed in a special flat cell (volume 160 pl, Labotec Co., Tokyo) and incubated for 1 min. The DMPO-0; spin adduct was analyzed by ESR spectrometry. A standard curve was made using 3.125 to 50 unit/ml SOD, and manganese oxide was used as an internal standard. Conditions of ESR spectrometry for measurement of SOD activity were as follows: magnetic field, 335.7f 5 mT; power, 4 mW; response, 0.3 s; modulation, 0.2 mT; amplitude, 3.2X 1,OOO; sweep time, 2 min; temperature, room temperature. The protein concentration was measured using Bio-Rad Protein Assay kits (Richmond, CA). Statistics The results were usually expressed as the mean fstandard deviation, and statistical significance was determined by the Student's t test (two tailed). RESULTS
Synaptosomal Membrane Fluidity Fig. 1-a, b shows the ESR spectra of spin labels embedded in the synaptosomal membrane. The surface and deep layers of the membranes were probed with 5-DSand 16DS,respectively (Fig. 1-c) . Figs. 2-a and 3-a show synaptosomal membrane fluidity in the brain of control and kindled rats at 24 h after the last seizure. The rotational correlation time, to, at the hydrophobic core was significantly decreased in the amygdala, hippocampus and frontal cortex (p
