0361-9230/92 $5.00 + .OO 1992 Pergamon Press Lid.

Brain Research Bulletin. Vol. 28, pp. 135-742, 1992 Printed in the USA. All rights reserved.

Cerebral Blood Flow and Metabolism in Soman-Induced Convulsions TSUNG-MING

SHIH*’ AND OSCAR U. SCREMINt2

*U.S. Army Medical Research Institute of Chemical Defense, Aberdeen Proving Ground, MD 21010 7Veterans Aflairs Medical Center and Department of Neurology, Universityof New Mexico, Albuquerque, NM 87108 Received 1 August 199 1 SHIH, T. -M. AND 0. U. SCREMIN. Cerebral bloodflow and metabolism in soman-inducedconvulsions.BRAIN RES BULL 28(5) 735-742, 1992.-Regional cerebral blood flow (CBF) and regional cerebral glucose utilization (CCXJ) were studied by quantitative autoradiographic techniques in rats. Animals were treated either with a toxic dose of soman, an irreversible organophosphorus cholinesterase inhibitor, that produced convulsions or with saline as controls. An increased arterial blood pressure (mean increase = 41% of control) always preceded onset of convulsions. Convulsive activity was associated with an increase of plasma glucose concentration and marked increases over controls of CGU [average of all regions: control = 75 + 5 pmol - 100 -1 g * min-‘, n = regions/animals (304/8); seizures = 45 1 k 20 pmol * 100 g-’ . min-‘, n = 190/5] and CBF [average of all regions: control = 135 + 6 ml- 100 g-‘amin-‘, n = 190/5; seizures = 619 & 29 ml+ 100 g-‘emin-‘, n = 190/5). Regional distribution of these effects revealed a greater proportional increase of CBF over CGU in cingulate, motor, and occipital cortex and caudateputamen. In contrast, a lower proportional increase of CBF over CGU in CA3 region of hippocampus, dentate gyrus, medial thalamus, and substantia nigra was observed, implying the existence of a relative &hernia in these brain areas. These findings may be relevant to the pathogenesis of brain lesions associated with soman-induced convulsions.

Soman

Organophosphorus cholinesterase inhibitor

Cerebral blood flow

Cerebral glucose utilization

Convulsions

Brain lesions

SOMAN is a highly toxic organophosphorus (OP) cholinesterase (ChE) inhibitor whose toxic effects are thought to be related to excess build-up of acetylcholine resulting from the irreversible inhibition of ChE (50). With carbamate pretreatment and atropine/pralidoxime therapy, it is possible to prevent lethality (9,12,19); however, convulsive activity still persists (9,12,18). Soman is known to induce brain lesions when given at doses that generally induced seizure/convulsive activity (20,26,27,36,49). The distribution of this pathology predominates in piriform cortex, amygdala, hippocampus, and thalamus (6,27,28,49). Although the pathogenesis of brain damage associated with epileptic seizures induced by chemical toxicants is far from being understood, failure of energy supply-demand mechanisms (53), decrease in intracellular pH (16), increase in intracellular calcium (30), and excessive activity of excitatory neurotransmitters (5) have been considered. It has been suggested that the brain damage observed afIer status epilepticus in rats may be due to the inability of blood flow to keep pace with metabolic demands (45). For example,

Autoradiography Rat

Plasma glucose

the substantia nigra is known to undergo complete necrosis in this condition, presumably by a hypermetabolic mechanism ( 17). Although regional cerebral glucose utilization (CGU) increases have been reported in soman-induced seizures (22,32,39), no information exists in the literature regarding coupling of blood flow and metabolism in this condition. The possible role of regional cerebral blood flow (CBF) in soman neuropathology then cannot be ascertained. The present investigation was undertaken to evaluate the regional distribution of CBF and CGU by use of autoradiographic techniques with a high degree of spatial resolution. Animals that developed convulsions following soman were subjected to CBF or CGU measuring procedures and the results compared to saline-treated controls. METHOD

Animals Male (Crl:CDBRR VAF/PlusR) Sprague-Dawley rats (Rattus norvegicus) weighing 200-300 g were used. Rats were quaran-

The experiments reported herein were conducted according to the Guide for Care and Use of Laboratory Animals (1985) as prepared by the Committee on Care and Use of Laboratory Animals, National Research Council, NIH Publication No. 85-23. The opinions or assertions contained herein are the private views of the authors and are not to be construed as official or as reflecting views of the Department of the Army or the Department of Defense.. ’ Requests for reprints should be addressed to Tsung-Ming Shih, Commander, U.S. Army Medical Research Institute of Chemical Defense, ATTN: SGRD-UV-PB (Dr. T.-M. Shih), Aberdeen Proving Ground, MD 21010-5425. 2 Current address: VA Medical Center, West Los Angeles, Wilshire and Sawtelle Blvds., Los Angeles, CA 90073.

735

736

SHIH AND SCREMIN TABLE 1 CGU AND CBF IN ANIMALS THAT RECEIVEDSALINE (CONTROL)OR SOMAN AT

A DOSE THAT

CGU (pmol . 100g-’ . min-‘) C0ntr01 Brain Area Cingulate c. Motor c. Somatosens. c. Temporal c. Occip. c. (I 7) Occip. c. (I 8) Occip. c. (18a) Olfactory c. Hip. CA 1 Hip. CA2 Hip. CA3 Dentate gyrus Med. D. thal. Ce. Med. thal.

Caudate-p. Claustrum Amygdala N. basalis Preoptic area Lat. hypoth. Dor. hypoth. D. M. hypoth. Med. em. M. mam. n. Interped. n. s, sup. toll. D. sup. toll. Per. gray M. raphe Pont. ret. n. Pant. n. S. nigra M. genie. Inf. toll. Cerebellum Int. caps.

Hip. commis. C. callosum

Stereotaxic Coordinates 0.7,0.5, 0.7 -0.3, 3, 0.7 -0.3, 5.5, 0.7 -5.8, 7, 0.7 -5.8, 4, 0.7 -5.8, 2, 0.7 -5.8, 6, 0.7 -1.3, 5.5, 0.7 -5.3, 3, 3 -5.3, 5.5, 6 -3.3, 4.3, I -3.3, I, 4.2 - 1.8,0.7,5 -2.3, 0, 6 0.7, 2.5, 6 2.7, 2, 5 -1.8, 3.8, 8 -1.3, 3, 8 -0.3, 1.4, 8.3 -1.3, 2.2, 8.8 -2.3, 0.5, 8.5 -3.3, 0.5, 9 -2.8,0, IO -4.8, 0, 9 -5.8, 0, 9 -6.8, I, 3 -6.8, I, 4.2 -7.3, 0.5, 5.5 -7.3, 0, 8.5 -7.8, 1.3, 8.5 -7.3, 1, IO -5.8, 2, 8 -5.3, 3.5, 6 -8.8, 1.5, 4.4 - 10.3, 0, 5 -1.3, 2.5, 4 -1.3,0,4 0.2,0, 3.5

SE

91.3 98.2 100.3 127.9 96.0 89.2 94.6 64.9 67.8 67.6 54.1 45.1 87.0 92.9 91.4 111.4 48.7 45.6 64.3 44.3 48.2 49.3 34.8 112.2 93.8 83.7 79.8 63.0 94.9 59.3 53.2 72.8 108.5 162.1 61.4 35.4 23.0 34.2

2.2 4.1 3.8 9.0 7.2 7.0 6.8 4.0 5.3 5.4 3.5 3.0 4.5 6.3 3.9 5.4 4.1 2.4 4.3 4.0 3.7 2.9 1.9 10.3 7.9 7.3 7.8 5.2 5.4 3.6 3.2 7.1 8.1 12.7 4.8 4.8 3.4 3.2

Control

Mean 497.5 529.6 534.9 594.8 471.9 455.0 499.8 475.0 453.3 439.6 449.7 457.4 680.0 678.6 575.6 585.3 492.0 405.9 332.0 356.8 343.3 338.1 323.4 575.5 552.7 654.3 394.8 333.4 347.6 227.6 254.0 672.7 534.0 442.3 327.3 316.5 248.9 285.9

CONVULSIONS

CBF(ml. 100g-l. min-I)

Soman

Mean

INDUCED

SE 115.6 110.4 121.1 126.4 101.5 93.4 104.9 78.5 95.5 65.9 69.9 77.1 106.2 113.5 109.4 127.0 99.1 95.7 91.0 68.7 44.3 48.3 64.5 121.0 98.2 113.6 64. I 58.1 71.3 59.0 58.3 114.8 110.09 117.2 62. I 39.0 41.5 63.4

Mean 141.2 141.7 181.2 198.8 131.2 112.7 146.8 126.8 100.8 115.8 95.2 94.4 164.6 166.2 125.0 152.8 115.2 100.4 123.8 116.4 119.3 139.8 106.7 180.7 161.2 158.6 150.0 128.2 189.0 126.2 138.6 122.5 167.6 239.6 124.6 76.5 71.0 13.2

Soman SE 14.9 13.0 22.1 14.1 9.8 8.8 14.2

11.0 11.9 8.9 9.6 6.3 13.4 15.2 13.4 10.2 8.8 7.5 11.5 9.0 9.1 9.3 7.1 25.5 22.3 14.9 15.5 12.2 18.0 9.1 12.2 15.1 24.5 32.9 7.7 7.9 5.3 5.8

Mean 857.0 860.4 803.6 864.0 811.0 791.6 834.4 481.8 510.8 530.6 374.2 506.4 178.4 870.6 767.6 817.4 530.8 585.4 504.4 479.4 512.0 486.2 424.8 650.8 705.2 850.8 678.4 428.0 615.2 369.0 442.0 618.2 770.5 909.6 434.0 385.2 387.4 353.2

SE 46.3 75.4 51.8 69.4 49.2 27.3 81.7 49.8 48.9 66.4 40.0 65.9 47.1 79.4 18.4 95.8 69.2 76.9 61.9 57.4 73.2 83.5 42.8 28.2 29.6 91.5 106.4 33.4 53.0 41.6 77.1 70.0 24.7 155.0 97. I 55.0 50.2 39.5

Values represent means and SE of five to eight rats per group. Stereotaxic coordinates (mm) represent bregma plane, horizontal coordinate, and distance from pial surface for all cortical regions (top eight in the list) and bregma plane, horizontal coordinate, and vertical coordinate for the rest (34). Statistical analysis of the data is described in the text.

tined on arrival and screened for evidence of disease before they were released for the experiments. They were maintained, under an American Association for Accreditation of Laboratory Animal Care (AAALAC) program, in plastic cages (Lab Products, Inc., Maywood, NJ) on hardwood chip contact bedding (BetaChip, Northeastern Products Corp., Warrensburg, NY) changed two times each week, and were provided commercial certified rodent ration (Zeigler Bros., Inc., Gardners, PA) and tapwater ad lib. Animal holding rooms were maintained at 2 1 + 2°C with 50 +_ 10% relative humidity using at least 10 air changes per hour of 100% conditioned fresh air. Rats were on a 12 L: 12 D full-spectrum lighting cycle, which was provided between 0600 and 1800 h. Four groups of animals were used: saline-treated

controls, CBF measurements (n = 5); saline-treated controls, CGU measurements (n = 8); soman-induced seizures, CBF measurements (n = 5); and soman-induced seizures, CGU measurements (n = 5). Materials Soman, 97% pure as determined by (3’P) nuclear magnetic resonance (NMR), was obtained from the Chemical Research, Development and Engineering Center (Aberdeen Proving Ground, MD). Stock solutions (2 mg/ml) were prepared in icecold 0.154 M sodium chloride (saline), stored at -8O”C, and further diluted in saline to a concentration of 220 CLg/mljust

137

EFFECTS OF SOMAN SEIZURE ON CBF-CGU

prior to administration. All injections were made subcutaneously at a site over the mid dorsum at least 2 h after discontinuation of halothane anesthesia. The volume of injection was 0.5 ml/kg body weight. [‘4C]2-Deoxyglucose (2DG, specific activity 58 mCi/mmol) was obtained from Amersham Corp. (Arlington Heights, IL) and iodo-[‘4C]antipyrine (IAP; specific activity 55 mCi/mmol) was obtained from du Pont Co. (Boston, MA).

heart was arrested with a bolus of T-6 1 euthanasia solution [N(2-(m-methoxy-phenyl)-( 1))~gamma-hydroxybutyramide, 200 mg/ml; 4,4’-methylene-bis(cyclohexyl-trimethyl-ammonium iodide), 50 mg/ml; and tetracaine hydrochloride, 5 mg/ml] (American Hoechst Co.). This produced a precipitous fall of arterial blood pressure within 3 s. The exact timing of this event was determined from a continuous record of iliac artery blood pressure. For determination of CGU, a bolus of saline containing 100 pCi/kg body weight of the tracer 2DG was injected intravenously followed by arterial blood sampling at 0, 0.3, 0.7, 1.O, 1.5, 3.5, 7.5, 15, 25, and 30 min after injection for plasma glucose and radioactivity measurements. Plasma glucose concentrations were determined calorimetrically by the glucose oxidase method (Sigma Diagnostic kit No. 510-A). The animal was sacrificed after the last sample by an intravenous bolus of T-6 1 euthanasia solution. At the end of CBF or CGU procedures, brains were rapidly removed, frozen in methylbutane at -7O”C, and cut in a cryostat in 20-pm thick slices. These sections were subsequently heat dried on glass slides and exposed to Kodak NMC film along with eight radioactive standards. Optical density of standards and tissue images on films was determined with a video-digitizing system consisting of a Chroma-Pro 45 IAIS “Dumas” film illumination system, a Phillips CCD monochrome imaging module coupled to an AT&T Targa M8 digitizing board on a Tandon PCA/I2 microcomputer, and JAVA (Jandel Scientific Corp., Corte Madera, CA) software. Tissue radioactivity was calculated by interpolation from the radioactivity-optical density relationship defined by the standards. Measurements were obtained from 38 regions listed in Table 1 and defined by the stereotaxic coordinates (mm) of the region’s center after the anatomical atlas of Paxinos and Watson (34) listed in the same table.

Procedures

Data Analysis

y = 0.64

+ 0.60

x

r = 0.97

p < 0.001

. .’

l

.. . ., . .*: 3% ’

0 Controls

.

/ I.”

,

1.25

0 Soman I

1.50

Seizures

I

1.75

2.bo

2.;5

2.;0

2.;5

3.bo

Log CGU FIG. 1. A highly significant correlation between CGU and CBF is found both in controls and convulsing animals. A single slope defines the relationship between these variables for both conditions. Every point rep resents paired means of log CGU and log CBF of five to eight animals for all regions listed in Table 1.

Regional CBF was studied with the autoradiographic IAP technique (38) and regional CGU was determined with the quantitative autoradiographic 2DG technique (47). These determinations were performed separately in conscious animals that had been implanted with arterial and venous catheters under halothane anesthesia (2% in air for induction and 1.5% for maintenance) and left to recover in a restmining device (Bollman cage). The device entraps animals between nontraumatic plastic bars that permit free movement of the extremities. Two hours after discontinuation of halothane anesthesia, rats were treated with either saline or soman. Initially, a dose of 99 &kg (equivalent to 0.9 X LDso) of soman was injected. If no convulsions developed in 20 min, additional soman at incremental doses of 11 pg/kg (0.1 X LDSo) was given every 20 min. As soon as convulsions were fully developed, the CGU measuring procedures were started 4.5 + 1.2 min after commencement of convulsive activity and carried out for 30 min. CBF measuring procedures were started later after commencement of seizure activity (14 + 4.5 min) to occur within the first half of the CGU procedure. Those animals that did not exhibit convulsions or died before completion of the CGU or CBF procedure were not included in this report. The IAP technique was implemented by continuous intravenous infusion of 0.6 ml saline containing 100 &i/kg body weight of the tracer IAP over 30 s. Timed blood samples were obtained every 2-3 s from a free-flowing arterial catheter throughout the infusion time and processed for liquid scintillation counting of radioactivity in a Beckman LS scintillation spectrometer (Fullerton, CA). At the end of the 30-s period, the

Measurement of tissue radioactivity for calculation of CBF and CGU was obtained from multiple readings in at least two different tissue sections and averaged for every region. Regional group means and SE were computed. Statistical significance of CBF and CGU means between animals treated with saline or soman were assessed by a two-factor (treatments and regions) with interaction (treatments - regions) analysis of variance (ANOVA) after a log transformation (54) to homogenize variances. In a second type of statistical analysis, and to eliminate variability between animals while preserving the pattern of regional variation in CGU and CBF values, data of CGU and CBF were transformed to ratios between every region value and a global value calculated by averaging all 38 studied regions in each animal. Protected t comparisons for every region between control and seizure groups were then performed. Due to the large number of contrasts, the probability level required to declare a difference as significant was set at 0.0 1. Comparisons of plasma glucose levels between control and seizure groups (average of measurements during CGU procedures) was done by Student’s t-test. Comparisons of arterial blood pressure levels before and at onset of convulsions and at the end of the experiment were performed by the Bonferroni procedure (52). RESULTS The toxicity of soman was characterized by chewing movements, salivation, head and trunk tremors, Straub tail, and generalized tonic-clonic convulsions. The convulsive activity consisted primarily of continuous repetitive jerking of head, body, and the extremities. Total dose of soman given per animal at

738

SHIH AND SCREMIN

0

CONTROL m

SEIZURE3

ecu

CBF

FIG. 2. nCGU (top) and nCBF (bottom) in brain regions of control and convulsing animals. nCBF exceeds nCGU in cingulate, motor, and two regions of occipital cortex of convulsing animals, while the opposite is true for CA3 field of Ammon’s horn and dentate gyms. *Statistically significant difference

from control, p < 0.0 I. the time when convulsions developed ranged from 99-121 pg/ kg SC (equivalent to 0.9- 1.1 X LDSo). Once started, convulsive activity continued unabated throughout the duration of CBF or CGU measurement periods. Animals also showed gasping and copious secretion of saliva and mucus. A significant increase in arterial blood pressure (mean + SE; n = 10) preceded the onset of convulsions: before soman injection = 104.1 -t- 3.8 mm Hg, after soman injection but before onset of convulsions = 147.0 + 4.0 mm Hg (p < O-01), and at the end of experiment

= 128.8 ? 8.4 mm Hg (NS.) (n = 5). Plasma glucose was 142.2 & 6.24 mg/dl (n = 8) in control rats and 277.2 + 29.2 mg/dl (n = 5) (p < 0.005) in soman-treated animals. A large increase in CGU and CBF levels was observed in all brain regions of convulsing animals (Table 1). Global mean CGU, obtained by averaging the values of all regions in all animals of every group, was 75 + 5 pm01 . 100 g-’ - min-’ [n (regions/animals) = 304/8] in controls and 45 1 + 20.1 pmol - 100 g-’ - min-’ (n = 190/5) during soman seizures. A similar behavior

739

EFFECTS OF SOMAN SEIZURE ON CBF-CGU 0

CONTROL

_

SEIZURES

2.0 ,

CGU

1

ratios of region to global CGU and CBF (nCGU, nCBF) (see the Method section). This transformation eliminated the variance between animals and allowed better study of the pattern of regional distribution of CGU and CBF. Means of nCGU and nCBF in animals that developed soman seizures were contrasted with those of controls at the p < 0.01 level (Figs. 2-4). This analysis revealed five different patterns: 1) increased (with regard to controls) nCGU and no change in nCBF in CA3 region of hippocampus, dentate gyms, medial thalamus, and substantia nigra; 2) no change in nCGU and decreased nCBF in central (periacqueductal) gray matter; 3) no change in nCGU and increased nCBF in cingulate, motor, and occipital (areas 17 and 18) cortex and caudate-putamen; 4) decreased nCGU and no change in nCBF in temporal cortex, inferior colliculus, and medial geniculate; and 5) a parallel decrease in nCGU and nCBF in medial raphe and pontine reticular nucleus. For the remaining 23 regions depicted in Figs. 2-4, no differences were observed between nCGU or nCBF of control and seizing animals at the significance level selected. The lack of matching between CGU and CBF in some of the regions was apparent from the observation of the primary data, as exemplified in Fig. 5. DISCUSSION

FIG. 3. nCGU (top) and nCBF (bottom) in brain regions of control and convulsing animals. The increase in nCGU over controls observed in medial thalamus of convulsing animals is not matched by nCBF. The opposite pattem is observed in caudate-putamen. *Statistically significant difference from control, p < 0.0 1. was observed for global CBF: 135 + 6 ml - 100 g-’ - min-’ (n = 190/5) in controls and 619 f 29 ml. 100 g-‘-min-’ (n = 190/ 5) during soman seizures. Considerable variations of both CBF and CGU were observed among regions. Since the variance of the seizure groups was greater than that of the control groups, a log transformation was applied (54) to the data to achieve homoscedasticity. When the paired mean values of CGU and CBF of every region were plotted, it became evident that the correlation between these variables under control and seizure conditions could be defined by a single slope (Fig. 1). A clear separation of the two conditions was also apparent. ANOVA (two-factor model with interaction) of log-transformed data showed significance for the treatment and region factors, as well as for their interaction for both CGU and CBF [CGU: treatments, F(1) = 2293, p < 0.0001, regions, F(37) = 10.45, p < 0.0001, interaction; fl37) = 2.19, p < 0.000 1; CBF: treatments, F(l) = 2733, p < 0.0001, regions, fl37) = 9.11,~ < 0.0001, interaction, F(37) = 1.59, p < 0.021. This type of analysis, although useful to characterize the global behavior of the variables under study, does not provide detailed information on the regional variations of CGU and CBF and the way in which these are individually affected by the seizure activity. The pattern of distribution of CGU and CBF among the regions was analyzed by calculating

Soman is a potent convulsive agent ( 11,2 1,22,43,44). Its action is presumed to be mediated through the excess acetylcholine stimulation at the central choline+ receptors (50). The convulsions consist of continuous repetitive jerking of the head, shoulders, and fore and hind limbs. Soman-induced seizures can persist for several hours although the intensity and frequency of the movements may diminish over time and become somewhat episodic. If the convulsions are effectively blocked by pharmacologic intervention early in the process, the neuropathology may be prevented (3,13,23,24,28,37,48). On the other hand, if the convulsions are allowed to continue neuronal degeneration occurs (20,26,27,36,49). Associated with soman seizures, a marked increase in CGU has been reported with rates doubling in most areas (22,32) and a four- to sixfold increase of CGU in substantia nigra, dentate gyrus, hippocampal body, and septum (39). Maxwell et al. (29) reported a 388% increase in global CBF following soman, but did not study regional distribution for this variable. Prolonged and excessive increases in neuronal activity and relative &hernia generated by uncoupling of blood flow and metabolism has been postulated as a possible mechanism in seizure-induced brain lesions (45,53). The present study was initiated to evaluate the role of this mechanism in soman-induced neuropathology. In this study, convulsions following soman administration, although with varied time of onset, were maintained throughout the duration of the experiments. This, added to the facts that a shortened protocol was used for CGU measurements and that CBF was timed with regard to onset of convulsive activity to fall within the first half of the period during which CGU was measured, assures that the two variables were assessed under comparable conditions. In addition, the short interval between commencement of convulsive activity and CGU or CBF measurements assures that the maximal levels of activation of both variables have been detected. Previous studies with other models of status epilepticus show that maximal metabolic and vascular changes are achieved soon after onset of convulsions to decrease slowly in magnitude afterward (2,3 1). The CGU technique utilized in this study is relatively insensitive to variations in plasma glucose concentration. Although marked changes in the lumped constant used in the calculation of CGU are associated with hypoglycemia, little effect is observed in hyperglycemia (40).

740

SHIH AND SCREMIN 0

CONTROL

m

SEIZURES

CGU

1

CBF

been termed “cerebral blood flow-metabolism coupling” and is a well-established manifestation of cerebral physiology. This phenomenon was confirmed by the present experiments, which showed that paired values of CGU and CBF of control and convulsing animals were correlated with a common slope. Close analysis of this relationship, however, disclosed some interesting characteristics of the changes in the patterns of regional distribution of these variables between both conditions. Twenty-five of 38 sampled regions showed either no change or a parallel change in nCGU and nCBF between controls and soman seizures, which confirms the notion put forward above. In 13 regions, however, a discrepancy was found between changes in nCGU and nCBF that in the case of the CA3 region of hip-

IAP ALJTORADIOGFIAPHY(CBF)

2DG AUTORADIOGRAPHY (CGU)

FIG. 4. nCGU (top) and nCBF (bottom) in brain regions of control and convulsing animals. The increase in nCGU observed in substantia nigra of convulsing animals is not matched by nCBF. No evidence of relative ischemia is observed in the rest of the regions. *Statistically significant difference from control, p < 0.01.

Our own evaluation in rats with hyperglycemia induced by a continuous infusion of physostigmine showed no significant differences of the lumped constant value between controls and hyperglycemic animals (4 1). An increase in arterial blood pressure consistently preceded motor convulsive activity and lasted for the duration of the experiment. This phenomenon was useful in directing the attention to animal behavior to accurately determine the time of onset of convulsions. The hypertensive response to ChE inhibitors, including soman, has been reported in rats (4,7,8,25,29,5 1). The pressor response to soman has been shown to be mediated by central muscarinic receptors acting through increased sympathetic activity (4). The increase in arterial blood pressure observed, however, was within the limits of cerebral autoregulation ( 14) and therefore it is unlikely that it may have been responsible for the observed increase. in CBF. The incidence of seizures in soman-treated animals was accompanied by large increases in plasma glucose levels and in both CGU and CBF in all brain regions studied. This observation is in agreement with the well-known interdependence between blood flow and metabolism in the brains of experimental animals and humans (46). There is normally a tight correlation between the levels of regional CBF and CGU. This phenomenon has

FIG. 5. Autoradiographs generated with 1AP(CBF) (top) and 2DG (CGU) (middle) in animals convulsing after soman administration. The bottom panel provides a key for the regions shown. It is apparent that the substantia nigra (SN) hypermetabolism makes this region stand out clearly over the rest of brain stem in the CGU image (middle). No such phenomenon is observed in the CBF image (top), indicating a failure of blood flow to the substantia nigra to raise above that of other brainstem structures. A lack of correspondence between CBF and CGU is also found in occipital cortex (17,18,18A) and temporal cortex (TE), where the CBF image is darker than the CGU one, and in the entorhinal cortex (ENT), where the opposite is observed. For a quantitative description of this phenomenon, see text and preceding figures.

741

EFFECTS OF SOMAN SEIZURE ON CBF-CGU

pocampus, dentate gyms, medial thalamus, and substantia nigra (increased nCGU with no change in nCBF) and central gray (no change in nCGU with decreased nCBF) reflected a relative &hernia. This phenomenon is reminiscent of a similar dissociation between CBF and CGU previously shown by Ackermann et al. (1) in the hippocampus and piriform cortex of rats under amygdala-kindled seizures. In contrast, cingulate, motor, and occipital cortex (areas 17 and 18), as well as caudate-putamen, of convulsing animals showed greater nCBF values than controls while no changes in nCGU were present in these same regions. This behavior was also observed in previous experiments with subtoxic doses of soman (42) in which we found a significant increase in CBF in these regions without a concomitant increase in CGU. Temporal cortex, inferior colliculus, and medial geniculate exhibited relatively lower nCGU values in convulsing rats than controls while no changes in nCBF were present. Although it is true that the relative &hernia found in the CA3 field of Ammon’s horn, dentate gyms, and medial thalamic nucleus is consistent with reported soman-induced neuronal lesions, a complementary one-to-one regional correlation between brain lesions reported by others (20,26,27,49) and CBF-CGU uncoupling found in this study cannot be established. For example, amygdala, piriform cortex, and claustrum were reported to have severe damage after soman, yet our results were not able to identify a statistically reliable dissociation between nCBF and nCGU. Contrary to the relative ischemia hypothesis, we observed a greater nCBF over nCGU in cingulate, motor, and occipital cortex, where severe neuronal necroses have been reported. Thus, in general the data presented suggest that a lack of correlation between blood flow and metabolism in brain regions may not be the primary mechanism for the pathogenesis of soman-induced brain damage. Therefore, mechanisms other than blood

flow-metabolism uncoupling, such as decrease in intracellular pH, increase in intracellular calcium, and excessive activity of excitatory neurotransmittets, should be considered as an explanation for regional brain damage following intense seizure activity (5,16,24,26,30,35). We found significant relative ischemia in substantia nigra during soman seizures. Although only limited soman neuropathology data is available for this region in the literature (33,35), substantia nigra is known to undergo necrosis, presumably induced by relative ischemia, in other models of seizures (17,45,53) and has emerged as an important structure in the propagation of seizure activity (10,15). The dissociation of CBF and metabolism found in the present study cannot fully explain the regional brain lesions reported in the literature. We have, nevertheless, identified hypermetabolism that is not matched by CBF in several key brain regions during soman-induced convulsions. Thus, we suggest that this phenomenon deserves further exploration in terms of the correlation of lesions with CBF-CGU uncoupling in a more thorough and systematic manner, particularly since no report available in the literature describes regional distribution of pathologic changes with the same detail used here for the description of CBF-CGU matching. Furthermore, the possible mechanisms that mediate the limitation of the CBF increase in the affected regions presented in this study require further investigation. ACKNOWLEDGEMENTS

We thank Deborah Heuser, Thomas Koviak, and Elsa Romero for excellent technical assistance and Andris Kaminskis for plasma glucose measurements. We are also indebted to Dr. Clifford Quails, Department of Mathematics, University of New Mexico, for help with statistical analysis. This work was supported by a VA-DOD Merit Review from Medical Research Service, Department of Veterans Affairs.

REFERENCES I. Ackermann, R. F.; Chugani, H. T.; Finch, D. M.; Babb, T. L.; Lear,

2.

3.

4.

5.

6.

7.

8. 9.

J. L.; Engel, J. Jr. Autoradiographic studies of stimulation-induced changes in rat hippocampal metabolism and blood flow. J. Cereb. Blood Flow Metab. 3(Suppl l):S238-S239; 1983. Borgstrom, L.; Chapman, A. G.; Siesjo, B. K. Glucose consumption in the cerebral cortex of rat during bicuculline-induced status epilenticus. J. Neurochem. 27:971-973: 1976. Branman, D. J.; Jaax, N. K.; Sparenbbtg, S. MK-80 I protects against seizures and brain damage induced by the cholinesterase inhibitor soman. Sot. Neurosci. Abstr. 14:240; 1988. Brezenoff, H. E.; McGee, J.; Knight, V. The hypertensive response to soman and its relation to brain acetylcholinestemse inhibition. Acta Pharmacol. Toxicol. 55:270-277; 1984. Chapman, A. G.; Westerbrook, E.; Premachandera, M.; Meldrum, B. S. Changes in regional neurotransmitter aminoacid levels in rat brain during seizures induced by L-allylglycine, bicuculline, and kainic acid. J. Neurochem. 43:62-70; 1984. Churchill. L.: Pazdernik. T. L.: Jackson. J. L.: Nelson. S. R.: Samson. F. E.; M&mot&, J. H. Jr.: McLeod, C.‘G. Jr. Soman-induced brain lesions demonstrated by muscarinic receptor autoradiogmphy. Neurotoxicology 6:81-90; 1985. De Neef, J. H.; Jo&an, K. M.; Pot&s, A. J. The effects of paraoxon on blood pressure in the anesthetized and in the conscious rat. Br. J. Pharmacol. 77:223-232; 1982. Dimhuber, P.; Cullumbine, H. The effect of anticholinesterase agents on the rat’s blood pressure. Br. J. Pharmacol. Chemother. 10: 1215; 1955. Dimhuber, P.; French, M. C.; Green, D. M.; Leadbeater, L.; Stratton, J. A. The protection of primates against soman poisoning by pretreatment with pyridostigmine. J. Pharm. Pharmacol. 31:295-299; 1979.

10. Garant, D. S.; Gale, K. Lesions of substantia nigra protect against experimentally induced seizures. Brain Res. 273:156-161; 1983. 11. Glenn, J. F.; Hinman, D. J.; McMaster, S. B. Electmencephalographic correlates of nerve agent poisoning In: Dun, N. J.; Perhnan, R. L., eds. Neurobiology of acetylcholine. New York Plenum; 1987:503-534. 12. Gordon, J. J.; Leadbeater, L.; Maidment, M. P. The protection of animals against organophosphate poisoning by pretreatment with a carbamate. Toxicol. Appl. Pharmacol. 43:207-2 16; 1978. 13. Hayward, I. J.; Wall, H. G.; Jaak, N. K.; Wade, J. V.; Marlow, D. D.; Nold, J. B. Decreased brain pathology in organophosphateexposed rhesus monkeys following benzodiazepine therapy. J. Neurol. Sci. 98:99-106; 1990. 14. Heistad, D. D.; Kontos, H. A. Cerebral circulation. In: Shepherd, J. T.; Abboud, F. M., eds. The cardiovascular system, vol. 3, peripheral circulation and organ blood flow. Bethesda, MD: American Physiological Society; 1983: 137- 182. 15. Iadarola, M. J.; Gale, K. Substantia nigra: Site of anticonvulsant activity mediated by gamma-aminobutyric acid. Science 2 18: 12371240; 1982. 16. Inamura, K.; Smith, M.-L.; Hansen, A. J.; Siesjo, B. K. Seizureinduced damage to substantia nigra and globus pallidus is accompanied by pronounced intra- and extracellular acidosis. J. Cereb. Blood Flow Metab. 9:821-829; 1989. 17. Ingvar, M.; Folbergrovl, J.; Siesjii, B. K. Metabolic alterations underlying the development of hypermetabolic necrosis in the substantia nigra in status epilepticus. J. Cereb. Blood Flow Metab. 7: 103-108; 1987. 18. Kluwe, W. M.; Chinn, J. C.; Feder, P.; Olson, C.; Joiner, R. Efficacy of pyridostigmine pretreatment against acute soman intoxication in a primate model. Proc. Sixth Med. Chem. Defense Biosci. Rev. (AD Bl21516):227-234; 1987.

742

19. Leadbeater, L.; Inns, R. H.; Rylands, J. M. Treatment of poisoning by soman. Fund. Appl. Toxicol. 5:S225-S23 1; 1985. 20. Lemercier, G.; Carpentier, P.; Sentenac-Roumanou, H.; Morelis, P. Histological and histochemical changes in the central nervous system of the rat poisoned by an irreversible anticholinesterase organophosphorus compound. Acta Neuropath. Berl. 6 I : 123- 129; 1983. 2 1. Lipp, __ J. A. Cerebral electrical activitv followine soman administration. Arch. Int. Pharmacodyn. 175:i61-169; l!968. 22. McDonough, J. H. Jr.; Hackley, B. E. Jr.; Cross, R. S.; Samson, F. E.; Nelson, S. R. Brain regional glucose use during soman-induced seizures. Neurotoxicology 4203-210; 1983. 23. McDonough, J. H. Jr.; Jaax, N. K.; Crowley, R. A.; Mays, M. Z.; Modrow, H. E. Atropine and/or diazepam therapy protects against soman-induced neural and cardiac pathology. Fund: Appl. Toxicol. 13:256-276; 1989. 24. McDonough, J. H. Jr.; McLeod, C. G.; Nipwoda, M. T. Direct microinjection of soman and VX into the amygdala produces repetitive limbic convulsions and neuropathology. Brain Res. 435: 123-l 37; 1987. 25. McGee, J.; Brezenoff, H. E. Protection by physostigmine against the pressor effect of soman in the rat. Life Sci. 41:65-69; 1987. 26. McLeod, C. G. Jr. Pathology of nerve agents: Perspectives on medical management. Fund. Appl. Toxicol. 5:SlO-Sl6; 1985. 27. McLeod, C. G. Jr.; Singer, A. W.; Harrington, D. G. Acute neuropathology in soman poisoned rats. Neurotoxicology 5:53-58; 1984. 28. Martin, L. J.; Doebler, J. A.; Shih, T.-M.; Anthony, A. Protective effect of diazepam pretreatment on soman-induced brain lesion formation. Brain Res. 325:287-289; 1985. 29. Maxwell, D. M.; Lenz, D. E.; &off, W. A.; Kaminskis, A.; Froehlich, H. L. The effects of blood flow and detoxification on in vivo cholinesterase inhibition by soman in rats. Toxicol. Appl. Pharmacol. 88:66-76; 1987. 30. Meldrum, B. S. Cell damage in epilepsy and the role of calcium in cytotoxicity. In: Delgado-Escueta, A. V.; Ward, A. A.; Woodbury, D. M.; Porter, R. J., eds. Basic mechanisms of the epilepsies. New York: Raven Press; 1986:849-855. 31. Meldrum, B. S.; Nilsson, B. Cerebral blood flow and metabolic rate early and late in prolonged seizures induced in rats by bicuculline. Brain 99~523-542; 1976. 32. Miller, A. L.; Medina, M. A. Cerebral metabolic effects of organophosphorus anticholinesterase compounds. Metabol. Brain Dis. 1: 147-156; 1986. 33. Olney, J. W.; Price, M. T.; Zorumski, C. F.; Clifford, D. B. Cholinotoxic syndromes: Mechanisms and prevention. In: Proceedings of the Workshop on Convulsions and Related Brain Damage Induced by Organophosphorus Agents. Aberdeen Proving Ground, MD: U.S. Army Medical Research Institute of Chemical Defense; 1990:147162. 34. Paxinos, G.; Watson, C. The rat brain in stereotaxic coordinates. Sidney: Academic Press; 1982. 35 Pazdernik, T. L.; Cross, R.; Giesler, M. P.; Nelson, S. R.; Samson, F. E.; McDonough, J. H. Jr. Delayed effects of soman: Brain glucose use and pathology. Neurotoxicology 6:6 l-70; 1985. 36. Petms, J. M. Soman neurotoxicity. Fund. Appl. Toxicol. 1:242; 1981. 37. Price, M. T.; Stewart, G. R.; Olney, J. W. Procyclindine protects against soman neurotoxicity, even when administered alter onset of convulsions. Sot. Neurosci. Abstr. 15:1349; 1989.

SHIH AND SCREMIN 38. Sakurada, 0.; Kennedy, C.; Jehle, J.; Brown, J. D.; Carbin, G. L.; Sokoloff, L. Measurement of local cerebral blood flow with iodo[‘4C]antipyrine. Am. J. Physiol. 234:H59-H66; 1978. 39. Samson, F. E.; Pazdemik, T. L.; Cross, R. S.; Giesler, M. P.; Mewes, K.; Nelson, S. R.; McDonough, J. H. Jr. Soman induced changes in brain regional glucose use. Fund. ADDS. . . Toxicol. 4:S173-S183; 1984. 40. Schuier, F.; Orzi, F.; Suds, S.; Lucignani, G.; Kennedy, C.; Sokoloff, L. Inthience of plasma glucose concentration on lumped constant of the deoxyglucose method: Effects of hyperglycemia in the rat. J. Cereb. Blood Flow Metab. 10:765-773; 1990. 41. Scremin, 0. U.; Allen, K.; Torres, C.; Scremin, A. M. E. Physostigmine enhances blood flow-metabolism ratio in neocortex. Neuropsychopharmacology 1:297-303; 1988. 42 Scremin, 0. U.; Shih, T. -M.; Corcoran, K. D. Cerebral blood flowmetabolism coupling after administration of soman at non-toxic levels. Brain Res. Bull. 26:353-356; 1991. 43. Shih, T.-M. Anticonvulsant effects of diazepam and MK-801 in soman poisoning. Epilepsy Res. 7:105-l 16; 1990. 44. Shih, T.-M. Choline& actions of diazeoam and atrooine sulfate in soman poisoning. Brain Res. Bull. 26:565-573; 199i. 45. Siesjo, B. K.; Ingvar, M.; Folbergrova, J.; Chapman, A. G. Local cerebral circulation and metabolism in bicuculline-induced status epilepticus: Relevance for develoument of cell damaae. Adv. Neurol. 34:2i7-230; 1983. 46. Sokoloff, L. Circulation and energy metabolism of the brain. In: Sieael. G. J.: Aaranoff. B. W.: Albers R. W.: Molinoff. P.. eds. Basic neurochemist&. New’York: ‘Raven Press; l989:565-596. 47. Sokoloff, L.; Reivich, M.; Kennedy, C.; et al. The [C’4]deoxyglucose method for the measurement of local cerebral glucose utilization: Theory, procedure and normal values in the conscious and anesthetized albino rat. J. Neurochem. 28:897-916; 1977. 48. Spare&or& S.; Brennecke, L. H.; Braitman, D. J. Forebrain damage induced bv ACh-esterase inhibitor soman is blocked bv MK-801 even when given after convulsion-onset. Sot. Neurosci’Abstr. 16: 281; 1990. 49. Switzer, R. C. III; Campbell, S. K.; Murphy, M. R.; Kerenyi, S. Z.; Miller, S. A.; Hartgraves, S. K. Soman-induced convulsions and brain damage as a function of: Chronic and acute exposure in rats and diazepam therapy in rhesus monkeys. In: Proceedings of the Workshop on Convulsions and Related Brain Damage Induced by Organophosphorus Agents. Aberdeen Proving Ground, MD: U.S. Army Medical Research Institute of Chemical Defense; 1990:3370. 50. Taylor, P. Anticholinesterase agents. In: Gilman, A. G.; Rall, T. W.; Nies, A. S.; Taylor, P., eds. Goodman and Gilman’s the phannacoloaical basis of theraneutics. New York: Peraamon Press; 1990: 130-149. 51 Varagic, V. The action of eserine on the blood pressure of the rat. Br. J. Pharmacol. 10:349-353; 1955. 52. Wallenstein, S.; Zucker, C. L.; Fleiss, J. L. Some statistical methods useful in circulation research. Circ. Res. 47: 1-9; 1980. 53. Wasterlain, C. G. Epileptic seizures. In: Siegel, G. J.; Agranoff, B. W.; Albers, R. W.; Molinoff, P., eds. Basic neurochemistry. New York Raven Press; 1989:797-810. 54. Winer, B. J. Statistical principles in experimental design, seconded., New York: McGraw-Hill: 1971:397-402.