Peptides,Vol. 11, pp. 557-563. ©PergamonPress plc, 1990. Printedin the U.S.A.
0196-9781/90$3.00 + .00
Focal Metabolic Effects of Angiotensin and Captopril on Subregions of the Rat Subfornical Organ S T E V E N W . SHAVER,1 M A S S A K O K A D E K A R O * A N D P A U L M. G R O S S
Neurosurgical Research Unit, Departments of Surgery and Physiology Queen's University and Kingston General Hospital, Kingston, Ontario, Canada and *Division of Neurosurgery, University of Texas Medical Branch, Galveston, TX Received 15 D e c e m b e r 1989
SHAVER, S. W., M. KADEKARO AND P. M. GROSS. Focalmetaboliceffects of angiotensin and captoprilon subregions of the rat subfornicalorgan. PEPTIDES 11(3) 557-563, 1990.--Angiotensin infusion increased glucose metabolism in 4 of 7 subdivisions of the rat subfomical organ, the effect being stronger in ventromedial compared to dorsolateral zones across the rostrocaudal axis. [SarkLeuS]Angiotensin II attenuated metabolic responses to intravenous angiotensin in all subfomical organ subregions. Brattleboro rats, having high circulating levels of angiotensin, displayed greater rates of glucose metabolism than Long-Evans rats in all subregions, differences that were eliminated by captopril, an inhibitor of angiotensin converting enzyme. The studies reveal focal subfornical organ zones where in vivo metabolic activity corresponds to cytoarchitectonic evidence for topographical processing within this angiotensin-sensitive structure. Deoxyglucose Brattleboro rat
Quantitative autoradiography Glucose metabolism [Sarl-LeuS]Angiotensin II Blood pressure regulation
Image analysis
Circumventricular organs
tions of function within this small structure (28). The concept of distinct subregional compartments in the SFO is also based on studies showing that the density of neurons (4) and binding sites for angiotensin (30), immunocytochemical staining for angiotensin perikarya and afferent fiber terminals (14), and cellular responsiveness to angiotensin (2) are greater in the SFO "ventromedial" zone compared to its supradjacent "dorsolateral" zone. With this evidence for neuronal and capillary densities varying systematically across distinct compartments in the SFO, we hypothesized that blood-borne angiotensin may affect SFO metabolic activity more in some subregions than in others. We thus examined the possibility that metabolic function during angiotensin infusion varies over the rostrocaudal axis of the SFO in relation to its cytoarchitectural subdivisions. Such studies could disclose within this small nucleus specific functional units having the highest level of stimulation and neural processing of an angiotensin stimulus. Further analyses for the densest subregional locations of vascular and neuronal angiotensin receptors in the SFO could arise from such experiments. We tested the above hypothesis further by using captopril to systemically inhibit angiotensin converting enzyme in an animal with normally high circulating levels of renin and angiotensin, the homozygous Brattleboro rat (18). These vasopressin-deficient rats have unusually high rates of SFO glucose metabolism that can be
CENTRALLY mediated responses to systemic angiotensin II include increases in arterial blood pressure, fluid intake, and vasopressin secretion (22). Important for initiating these effects is stimulation of a small endocrine-like nucleus in the roof of the third cerebral ventricle, the subfornical organ (SFO) (16,22). Among brain structures, SFO has the highest activity of angiotensin coverting enzyme (3) and is one of the most avid binding sites for angiotensin (19). It is thus thought to be a primary cerebral target for circulating or central angiotensin. SFO neurons are excited by angiotensin given by local or intravenous injection, into the cerebral ventricles, or into carotid arterial blood (9, 16, 22). Specific electrical stimulation of the SFO and systemic angiotensin acting through its efferent neural pathways elevate blood pressure by activation of peripheral sympathetic nerves and secretion of vasopressin (8, 16, 17). Extensive connections of the SFO throughout the neuraxis assure broadly integrated responses to angiotensin involving behavioral, cardiovascular, and endocrine sytems (15,21). Microscopic histological inspection of the SFO reveals that its cytoarchitecture is not homogeneous (4,23). Along its rostrocaudal extent, four subregions containing differentiated morphology can be identified (28). Evident in light microscopic sections analyzed with stereological methods are focal areas of differential capillary density, findings that implicate topographical specializa-
~Requests for reprints should be addessed to S. W. Shaver, Neurosurgical Research Unit, LaSalle Building, 146 Stuart Street, Kingston, Ontario, Canada, K7L 3N6.
557
558
SHAVER, KADEKARO AND GROSS
neutralized by restoring body fluid balance with exogenous vasopressin (13). We considered Brattleboro rats an effective model for exploring whether glucose metabolism stimulated by angiotensin in specific SFO subregions was altered by inhibition of the converting enzyme. The present report, therefore, is an analysis of subregional metabolic activity in the SFO, unlike our previous studies (5) in which we described angiotensin-related responses in the whole SFO irrespective of its subregions. METHOD
Preparation of Animals Groups consisted of 3-6-month-old (207-507 g) male SpragueDawley (n= 13), Long-Evans (n=7), and homozygous Brattleboro (n= 9) rats which were maintained on a 12-hour light-dark cycle and housed 2-3/cage with free access to food and water. We anesthetized the rats with halothane (3% in oxygen) and implanted polyethylene catheters in both femoral arteries and veins. Following a standard procedure for cerebral metabolic studies of conscious rats (27), we wrapped a plaster cast loosely around the hindquarters of the animals, leaving the legs, tail and catheters protruding. The cast was taped to a lead block to facilitate manipulation of the catheters and constrain mobility of the animals during the two-hour recovery period and experiment. We monitored arterial blood pressure and rectal temperature throughout the procedure and measured arterial glucose concentration, hematocrit, blood gas tensions, and pH periodically.
Administration of Angiotensin H and [Sarl-LeuS]Angiotensin H Sprague-Dawley rats (n = 5) were given angiotensin II acetate salt (Sigma Chemical Co., St. Louis, MO) dissolved in 0.9% saline at an intravenous dose rate of 2.5 txg/min delivered at 34 txl/min (5). The infusion was 45 min in duration, beginning 10 min prior to initiating the deoxyglucose experiment. Thus, a total of 112.5 wg of angiotensin II in 1.53 ml of saline was infused. Control rats were given 1.53 ml of saline over the same period (n = 4). An additional group of Sprague-Dawley rats (n = 4) received [Sarl-LeuS]angiotensin II (Sigma Chemical Co., St. Louis, MO), 1.12 mg dissolved in 0.5 ml of 0.9% saline, by intravenous injection over 10 min immediately preceding infusion of angiotensin as described above. Therefore, the dose ratio of antagonist: agonist was 10:1.
Administration of Captopril Captopril (E. R. Squibb and Sons, Princeton, NJ) was dissolved in 0.5 ml of 0.9% saline, and the pH adjusted to 7.5 with 0.1 N NaOH. One hour prior to the deoxyglucose experiment, Brattleboro (n=5) and Long-Evans (n=4) rats were injected intraperitoneally with the solution at a dose of 50 mg/kg. An additional 1/3dose of captopril (17 mg/kg) was given 10 min before initiating the experiment to ensure effective plasma concentrations of the drug (5). Sham-treated Brattleboro (n = 4) and Long-Evans (n = 4) rats were injected with equivalent volumes of 0.9% saline and studied in parallel with the captopril-treated rats.
Measurement of Glucose Metabolism in Subregions of the SFO The rate of glucose utilization in the SFO was measured using the method of Sokoloff et al. (27); rate and lumped constants for gray matter structures were applied to derive values for glucose metabolism in the SFO as discussed previously (6). Intravenous
injection of 2-deoxy-D-[1-~4C]glucose (specific activity, 50-55 mCi/mmol; New England Nuclear Boston, MA), 125 IxCi/kg in 0.5 ml saline, initiated the procedure. Fourteen timed blood samples of approximately 70 txl each were drawn into microcentrifuge tubes over the course of the experiment. The plasma samples were analyzed to determine the arterial time course for [~4C]deoxyglucose and glucose concentrations by scintillation counting and a glucose oxidase assay, respectively. The experiment was terminated after 45 min by intravenous injection of an overdose of sodium pentobarbital. The brain was quickly extracted and frozen in isopentane cooled to - 45°C in dry ice. Twenty lxm thick coronal sections of brain in the region of the SFO were cut in a cryostat at - 20°C, picked up on glass coverslips, dried instantly at 70°C, and exposed together with calibrated [~4C]methylmethacrylate standards (Amersham, Arlington Heights, IL) to X-ray film (Kodak SB-5) in cassettes for five days. Quantitative densitometric analysis of the autoradiographs was performed with an MCID system (Micro-Computer Imaging Device, Imaging Research, Inc., St. Catharines, ON, Canada), which allowed determinations of tissue 14C concentrations by referencing optical densities of the SFO to those of the standards. Using the tissue 14C concentrations, the rate constants and the gray matter "lumped constant" together with the time courses for plasma [~4C]deoxyglucose and glucose concentration, we computed values for cerebral tissue glucose metabolism by the operational equation of the method (27). Magnified serial images of the SFO were analyzed for glucose metabolism within subregions according to anatomical boundaries defined by the appearance of ependymal cells, capillary density, and the extent of pericapillary spaces, as detailed previously [see Figs. 1-2 in (28)]. To optimize precision of the intraorgan analysis, we used an overlay program in the MCID system software allowing superimposition of histological sections with the autoradiographs (Imaging Research Inc.). Because this overlay method depends on accurate matching of edges in the histological and autoradiographic sections, there was no perceptible distortion in the alignments. We also applied a morphometric analysis of the SFO subregions using the MCID system. For both the autoradiographic and morphometric analyses along the rostrocaudal axis, the SFO was divided into four subregions: "rostral," "transitional," "central," and "caudal" (28). Each subregion spanned approximately 100 txm. With the exception of the rostral subregion, all subregions were divided in the coronal plane into a "ventromedial" and an overlying "dorsolateral" zone (28). Thus, rates of glucose metabolism were obtained for each animal in 7 subdivisions of the SFO.
Statistical Analysis Glucose utilization was measured in each animal as the mean from three or more serial brain sections for each SFO subregion and zone. Data were analyzed by standard procedures using a one-factor analysis of variance (ANOVA) for the Sprague-Dawley rats and a two-factor ANOVA for the Long-Evans and Brattleboro rats. Tukey "critical difference" values were calculated to determine minimum differences between rat groups when significant F-ratios (p
0196-9781/90$3.00 + .00
Focal Metabolic Effects of Angiotensin and Captopril on Subregions of the Rat Subfornical Organ S T E V E N W . SHAVER,1 M A S S A K O K A D E K A R O * A N D P A U L M. G R O S S
Neurosurgical Research Unit, Departments of Surgery and Physiology Queen's University and Kingston General Hospital, Kingston, Ontario, Canada and *Division of Neurosurgery, University of Texas Medical Branch, Galveston, TX Received 15 D e c e m b e r 1989
SHAVER, S. W., M. KADEKARO AND P. M. GROSS. Focalmetaboliceffects of angiotensin and captoprilon subregions of the rat subfornicalorgan. PEPTIDES 11(3) 557-563, 1990.--Angiotensin infusion increased glucose metabolism in 4 of 7 subdivisions of the rat subfomical organ, the effect being stronger in ventromedial compared to dorsolateral zones across the rostrocaudal axis. [SarkLeuS]Angiotensin II attenuated metabolic responses to intravenous angiotensin in all subfomical organ subregions. Brattleboro rats, having high circulating levels of angiotensin, displayed greater rates of glucose metabolism than Long-Evans rats in all subregions, differences that were eliminated by captopril, an inhibitor of angiotensin converting enzyme. The studies reveal focal subfornical organ zones where in vivo metabolic activity corresponds to cytoarchitectonic evidence for topographical processing within this angiotensin-sensitive structure. Deoxyglucose Brattleboro rat
Quantitative autoradiography Glucose metabolism [Sarl-LeuS]Angiotensin II Blood pressure regulation
Image analysis
Circumventricular organs
tions of function within this small structure (28). The concept of distinct subregional compartments in the SFO is also based on studies showing that the density of neurons (4) and binding sites for angiotensin (30), immunocytochemical staining for angiotensin perikarya and afferent fiber terminals (14), and cellular responsiveness to angiotensin (2) are greater in the SFO "ventromedial" zone compared to its supradjacent "dorsolateral" zone. With this evidence for neuronal and capillary densities varying systematically across distinct compartments in the SFO, we hypothesized that blood-borne angiotensin may affect SFO metabolic activity more in some subregions than in others. We thus examined the possibility that metabolic function during angiotensin infusion varies over the rostrocaudal axis of the SFO in relation to its cytoarchitectural subdivisions. Such studies could disclose within this small nucleus specific functional units having the highest level of stimulation and neural processing of an angiotensin stimulus. Further analyses for the densest subregional locations of vascular and neuronal angiotensin receptors in the SFO could arise from such experiments. We tested the above hypothesis further by using captopril to systemically inhibit angiotensin converting enzyme in an animal with normally high circulating levels of renin and angiotensin, the homozygous Brattleboro rat (18). These vasopressin-deficient rats have unusually high rates of SFO glucose metabolism that can be
CENTRALLY mediated responses to systemic angiotensin II include increases in arterial blood pressure, fluid intake, and vasopressin secretion (22). Important for initiating these effects is stimulation of a small endocrine-like nucleus in the roof of the third cerebral ventricle, the subfornical organ (SFO) (16,22). Among brain structures, SFO has the highest activity of angiotensin coverting enzyme (3) and is one of the most avid binding sites for angiotensin (19). It is thus thought to be a primary cerebral target for circulating or central angiotensin. SFO neurons are excited by angiotensin given by local or intravenous injection, into the cerebral ventricles, or into carotid arterial blood (9, 16, 22). Specific electrical stimulation of the SFO and systemic angiotensin acting through its efferent neural pathways elevate blood pressure by activation of peripheral sympathetic nerves and secretion of vasopressin (8, 16, 17). Extensive connections of the SFO throughout the neuraxis assure broadly integrated responses to angiotensin involving behavioral, cardiovascular, and endocrine sytems (15,21). Microscopic histological inspection of the SFO reveals that its cytoarchitecture is not homogeneous (4,23). Along its rostrocaudal extent, four subregions containing differentiated morphology can be identified (28). Evident in light microscopic sections analyzed with stereological methods are focal areas of differential capillary density, findings that implicate topographical specializa-
~Requests for reprints should be addessed to S. W. Shaver, Neurosurgical Research Unit, LaSalle Building, 146 Stuart Street, Kingston, Ontario, Canada, K7L 3N6.
557
558
SHAVER, KADEKARO AND GROSS
neutralized by restoring body fluid balance with exogenous vasopressin (13). We considered Brattleboro rats an effective model for exploring whether glucose metabolism stimulated by angiotensin in specific SFO subregions was altered by inhibition of the converting enzyme. The present report, therefore, is an analysis of subregional metabolic activity in the SFO, unlike our previous studies (5) in which we described angiotensin-related responses in the whole SFO irrespective of its subregions. METHOD
Preparation of Animals Groups consisted of 3-6-month-old (207-507 g) male SpragueDawley (n= 13), Long-Evans (n=7), and homozygous Brattleboro (n= 9) rats which were maintained on a 12-hour light-dark cycle and housed 2-3/cage with free access to food and water. We anesthetized the rats with halothane (3% in oxygen) and implanted polyethylene catheters in both femoral arteries and veins. Following a standard procedure for cerebral metabolic studies of conscious rats (27), we wrapped a plaster cast loosely around the hindquarters of the animals, leaving the legs, tail and catheters protruding. The cast was taped to a lead block to facilitate manipulation of the catheters and constrain mobility of the animals during the two-hour recovery period and experiment. We monitored arterial blood pressure and rectal temperature throughout the procedure and measured arterial glucose concentration, hematocrit, blood gas tensions, and pH periodically.
Administration of Angiotensin H and [Sarl-LeuS]Angiotensin H Sprague-Dawley rats (n = 5) were given angiotensin II acetate salt (Sigma Chemical Co., St. Louis, MO) dissolved in 0.9% saline at an intravenous dose rate of 2.5 txg/min delivered at 34 txl/min (5). The infusion was 45 min in duration, beginning 10 min prior to initiating the deoxyglucose experiment. Thus, a total of 112.5 wg of angiotensin II in 1.53 ml of saline was infused. Control rats were given 1.53 ml of saline over the same period (n = 4). An additional group of Sprague-Dawley rats (n = 4) received [Sarl-LeuS]angiotensin II (Sigma Chemical Co., St. Louis, MO), 1.12 mg dissolved in 0.5 ml of 0.9% saline, by intravenous injection over 10 min immediately preceding infusion of angiotensin as described above. Therefore, the dose ratio of antagonist: agonist was 10:1.
Administration of Captopril Captopril (E. R. Squibb and Sons, Princeton, NJ) was dissolved in 0.5 ml of 0.9% saline, and the pH adjusted to 7.5 with 0.1 N NaOH. One hour prior to the deoxyglucose experiment, Brattleboro (n=5) and Long-Evans (n=4) rats were injected intraperitoneally with the solution at a dose of 50 mg/kg. An additional 1/3dose of captopril (17 mg/kg) was given 10 min before initiating the experiment to ensure effective plasma concentrations of the drug (5). Sham-treated Brattleboro (n = 4) and Long-Evans (n = 4) rats were injected with equivalent volumes of 0.9% saline and studied in parallel with the captopril-treated rats.
Measurement of Glucose Metabolism in Subregions of the SFO The rate of glucose utilization in the SFO was measured using the method of Sokoloff et al. (27); rate and lumped constants for gray matter structures were applied to derive values for glucose metabolism in the SFO as discussed previously (6). Intravenous
injection of 2-deoxy-D-[1-~4C]glucose (specific activity, 50-55 mCi/mmol; New England Nuclear Boston, MA), 125 IxCi/kg in 0.5 ml saline, initiated the procedure. Fourteen timed blood samples of approximately 70 txl each were drawn into microcentrifuge tubes over the course of the experiment. The plasma samples were analyzed to determine the arterial time course for [~4C]deoxyglucose and glucose concentrations by scintillation counting and a glucose oxidase assay, respectively. The experiment was terminated after 45 min by intravenous injection of an overdose of sodium pentobarbital. The brain was quickly extracted and frozen in isopentane cooled to - 45°C in dry ice. Twenty lxm thick coronal sections of brain in the region of the SFO were cut in a cryostat at - 20°C, picked up on glass coverslips, dried instantly at 70°C, and exposed together with calibrated [~4C]methylmethacrylate standards (Amersham, Arlington Heights, IL) to X-ray film (Kodak SB-5) in cassettes for five days. Quantitative densitometric analysis of the autoradiographs was performed with an MCID system (Micro-Computer Imaging Device, Imaging Research, Inc., St. Catharines, ON, Canada), which allowed determinations of tissue 14C concentrations by referencing optical densities of the SFO to those of the standards. Using the tissue 14C concentrations, the rate constants and the gray matter "lumped constant" together with the time courses for plasma [~4C]deoxyglucose and glucose concentration, we computed values for cerebral tissue glucose metabolism by the operational equation of the method (27). Magnified serial images of the SFO were analyzed for glucose metabolism within subregions according to anatomical boundaries defined by the appearance of ependymal cells, capillary density, and the extent of pericapillary spaces, as detailed previously [see Figs. 1-2 in (28)]. To optimize precision of the intraorgan analysis, we used an overlay program in the MCID system software allowing superimposition of histological sections with the autoradiographs (Imaging Research Inc.). Because this overlay method depends on accurate matching of edges in the histological and autoradiographic sections, there was no perceptible distortion in the alignments. We also applied a morphometric analysis of the SFO subregions using the MCID system. For both the autoradiographic and morphometric analyses along the rostrocaudal axis, the SFO was divided into four subregions: "rostral," "transitional," "central," and "caudal" (28). Each subregion spanned approximately 100 txm. With the exception of the rostral subregion, all subregions were divided in the coronal plane into a "ventromedial" and an overlying "dorsolateral" zone (28). Thus, rates of glucose metabolism were obtained for each animal in 7 subdivisions of the SFO.
Statistical Analysis Glucose utilization was measured in each animal as the mean from three or more serial brain sections for each SFO subregion and zone. Data were analyzed by standard procedures using a one-factor analysis of variance (ANOVA) for the Sprague-Dawley rats and a two-factor ANOVA for the Long-Evans and Brattleboro rats. Tukey "critical difference" values were calculated to determine minimum differences between rat groups when significant F-ratios (p
