0.161~92X),/92 $5.00 c .w Copyright (cl 1991 Petgamon Press Ltd.
Bruin Re\eorch Brr/l6~i/n,Vol. 28. pp. xX9-535, 1992 Printed in the USA. 411rights reserved.
Effect of Drinking on Angiotensin-II-Induced Shifts in Regional Cerebral Blood Flow in the Rat DONALD
A. CZECH*’
AND
ELLIOT
A. STElNt
*Hirjpsychoiogy Laboratory, Department ofP~sychology, MarqWtc C~niversit~~, Miluaukee. WI 53233 and fDc>partmmts of Psychiutry, Pharmacologic, and CeNlrlar Biologic and ilnutom!‘, Me&al College qf U’isconsin. Milwultkw U7 532% Received
19 September
199 1
CZECH, D. A. AND E. A. STEIN. lz@ci t$drinkittg on an,~iorrt?.rin-ll-ittd1ct.c~ .sii$i.sin rqiond trrdml hioctd /h, in i/w ml. BRAIN RES BULL 28(4) 529-535, 1992.-A map of brain regions affected by central administration of the octapeptide angiotensin II (All) and that would further reflect the consequences of All’s well-known dipsugenic action was developed. Regional cerebral blood flow (rCBF) as an indicator ofneuronalactivity was measured in conscious rats shortly after an ICV bolus injection of 100 ng All or saline vehicle (VEH). AII-treated animals were further subdivided into two groups that were either permitted to drink [All (Wt)] or whose water was removed when drinking was attempted [AI1 (W--j]. When compared to VEH condition. blood flow increased significantly within I min after All treatment in 33 of 53 regions sampled in those rats not given an opportunit) to drink. In I I of these 33 regions. ingestion of a small amount of water was associated with a reversal of All-induced elevation in blood llow [i.e., AII (W+) < AI1 (W-)]; these regions included the organum vasculosum lamina terminalis. rostra1 lateral hypothalamus. supraoptic nuclei, rostra1 zona incerta, and median eminence. A group of similarly treated rats exhibited a significant elevation of mean arterial biood pressure following AI1 treatment without significant shifts in artenal blood gases, pH, or bicarbonate. These data are consistent with prominent involvement of the anteroventral third ventricular region of the rat brain. The results further indicate that rCBF may be a sensitive measure for the identi~cation ofcentralsites ofaction of AII as a dipsogenic agent and may reveal distinctions between regions associated primariiy with Initiation of drinking and those reflecting the results of subsequent behavioral events. Cerebral blood flow
Angiotensin II
Circumventricular organs
THE octapeptide angiotensin 11 (All) is involved in a number of central integrative functions linked to body-fluid homeostasis, including stimulation of drinking (16,28,3 1), elevation of arterial blood pressure (9.19,32), increased salt appetite (5,34,43), and release of several pituitary hormones (21,38). Specifically, for example. All is known to be a potent dipsogen when administered either centrally or ~~pheraliy (11,19,28). Althou~ various aspects of All’s role in fluid regulation and drinking remain controversial and unclear, it is widely accepted that All’s dipsogenic and other regulatory effects are mediated, at least in part, through stimulation of receptor sites located within the circumventricular organs (CVOs) of the brain. These regions. with the apparent exception of the subcommissural organ, contain moderate to high concentrations of All receptors [e.g., see ( 12.36)], lie outside the blood-brain barrier, and are potentially accessible to both blood- and cerebrospinal fluid (CSF)-borne AIL Previous studies point to the subfornical organ (SFO) as the primary site of action for the dipsogenic activity of blood-
Drinking
Autoradiography
Iodoantipyrine
borne AI1 whereas All in CSF appears to initiate dipsogenic activity principally at the organum vasculosum lamina tetminaiis (OVLT) and hypothalamic median preoptic (MnPo) area located in the anteroventral third ventricular region (AV3V) (28,29). An extensive literature exists with a focus on mapping neural circuitry between the CVOs and surrounding CNS areas believed to be involved in AII’s dipsogenic effect(s) (17.23.30), as well as on the behavioral, physiological, and pharmacological effects of manipulating these areas. In contrast, attempts to establish temporal linkage between neuronal activity in these and other more distant sites with ongoing behavioral state have received less attention. Our concern was to further study All’s putative role in drinking through developing a broader map that would highlight those brain regions that participate in a functional system(s) underlying particular behavioral states. In the present study, we attempted to identify regions that would be activated in response to an ICV bolus injection of All and to distinguish those linked to All’s dipsogenic activity; the
’ Requests for reprints should be addressed to Donald A. Czech. repayment Milwaukee, WI 53233.
529
of Psychology, Marquette University, Schroeder Complex 454.
530
CZI:C’li AND Sl EIN
latter was accomplished by either offering rats the opportunity to demonstrate an intent to drink or following a brief drinking bout. To this end, we measured regional cerebral blood flow (rCBF) as an indicator of neuronal activity following ICV administration of AII. The [‘4C]iodoantipyrine (IAP) autoradiographic technique used for measuring rCBF (37) has a number of advantages over other metabolic procedures, including good spatial and temporal resolution. In providing a brief time window (30 s) within which to measure metabolic consequences, this rCBF technique allows one to look at discrete time points rather than being limited to measuring average shifts (as, e.g., with the 45-min 2-deoxyglucose method for glucose utilization). Such temporal resolution is clearly critical in being able to focus and lock in on behavioral/biological changes relatively quickly and thereby minimize obscuring specific and transient neuronal events with “background” activity. Rapid changes in rCBF as a consequence of neuronal activity thus provide an opportunity to closely monitor and link behavioral changes to activity in underlying neural systems. Under most conditions, rCBF is believed to be tightly coupled to local metabolic demand ( 13.37). Furthermore, numerous and widespread integrated systems can be assessed simultaneously with autoradio~aphic procedures. In attempting to isolate those patterns of blood flow that reflect AII’s dipsogenic action, we manipulated opportunity to drink. Such manipulation might thereby permit distinguishing between so-called stimulus-driven and behavior-driven circuitry in the brain.
METHOD
Animals and Preparation Male Sprague-Dawley-derived rats (Sasco, Madison, WI) weighing 300-325 g were individually housed in polycarbonate tub-type cages in a tem~rature~ontrolled room with lights on between 0830- 1900 h. Pelleted food and tap water were availabie ad lib. After several days of acclimation to the laboratory environment, a single 26-ga stainless steel guide cannula was stereotaxically implanted under Chloropenf anesthesia (2.5 ml/kg, IP), with tip positioned -0.4 mm above a lateral ventricle; a minimum of 4 days was allowed for recovery. Animals then experienced a restraint procedure of increasing duration for 5 days progressing from 1 to 5 h per day prior to measurement of rCBF or physiological variables. The restraint consisted of wrapping in a terry cloth jacket that immobilized both fore and hind limbs. Rats rapidly acclimated to the restraint, most ceasing to struggle by the third day. During this period and while under restraint, animals were also tested for drinking in response to an ICV injection of 100 ng AI1 (Sigma, St. Louis, MO) delivered in 2.0 ~1 sterile saline over 15 s. On the day of final testing and sacrifice, rats were again anesthetized with Chloropent and femoral arterial and venous catheters (PE 10) were implanted. The incisions were treated with a local anesthetic (2% xylocaine gel) and closed. Animals were then placed in the restraint condition and allowed to recover from anesthesia for a minimum of 5 h. An intravenous injection of 500 IV/kg heparin sodium in 0.5 ml sterile saline was given 1 h prior to start of testing. Tap water was provided during the recovery period via a sipper tube positioned directly in front of the animal. Animals could move their heads freely under restraint and readily drank. Access to water during subsequent testing was dependent on experimental group ~signment, as indicated next.
Animals were assigned to one of three treatment groups (II = 4lgroup) and manipulated postrecovery as follows: Two groups were injected ICV with 100 ng All in 2.0 ~1 normal saline and blood flowdetermined starting 1 min after injection termination. In one of these groups [AI1 (W-)], water was immediatelt removed when the rat attempted to drink (operationally deined as first fick at sipper tube). In the second group [Ali (Wi)]. water continued to be avaiiable ad lib throughout the entire procedure. Animals in a third group (VEH) were injected 10’ with 2.0 ~1 normal saline vehicle prior to initiation of blood flow procedures. Water was again present: however. no animal in the VEH condition attempted to drink. rCBF was measured by the method of Sakurada et al. (37). Briefly, this automdiograph~c method involved the infusion of 0.5 mf normal saline containing fO0 .&i/kg ~‘4C]i~oantipy~ne (45.5 mCi/mmol. Amersham) at a constant rate over 30 s into the femoral vein. Arterial blood samples were collected every 5 s onto preweighed filter paper throughout the infusion period, Immediately after the infusion period, rats were rapidly decapitated, the lower jaw and skin covering the skull were quickly (~30 s) removed, and the skull was immersed in isopentanc (-SO’C) for several minutes, then stored at --80°C. Filter papers were reweighed in counting vials, Aquassure (New England Nuclear) added to the vials, and the radioactivity of samples determined by liquid scintillation spectroscopy on a Beckman LS6000TA counter. Brains were subsequently removed from the skull in a cryostat chamber at -20°C and stored at ~80°C prior to sectioning. Brains were later cut into 30-pm sections at --20°C in a cryostat (Reichert-Jung, Model 850), thaw mounted onto glass slides, and dried on a slide warmer (5O’C). Slides were apposed to X-ray film (Kodak MR-1) with calibrated methacrylate [14C] standards (Amersham) in standard X-ray cassettes for an appropriate period (generally 4-6 weeks). Following film development, slides were stained with thionin for subsequent anatomic localization of structures of interest. Brain regions were analyzed on an MCID Image Analyzer (Imaging Research. St. Catherine. Ontario) and reported as cerebral blood flow in ml/ 100 g/min.
A separate group of rats was treated in a manner similar to that just noted; however, rCBF was not measured. Samples of arterial blood (150 al) were drawn just before and I min after a 2.0 gl ICV injection of 100 ng AI1 in sterile normal saline and following 0.5 ~1 of vehicle only. The vehicle-only condition preceded the AI1 injection by approximately 10 min in a withinsubjects design (n = 6). The smaller volume was used in the vehicle condition to avoid intr~ucing a relatively large volume of fluid into the ventricle within a rather short period of time. All animals were allowed to drink, as in the rCBF AI1 (W+) condition. Blood samples were assayed for paC02. paOz, HCO; , and pH on a blood gas analyzer (Corning-CIBA, Model 168). Heart rate (HR) and mean arterial pressure (MAP) were continuously monitored via a Statham pressure transducer and appropriate electronics on a Grass Model 79D ~ly~aph. These measures were also evaluated before and 1 min after treatment. As noted, water was available ad lib prior to and during the sampling/test period. Rats were sacrificed with an overdose of anesthetic at the completion of data collection. Statistical Ana&ses rCBF data were evaluated separately in each brain region with one-way analyses of variance (ANOVAs); pairwise com-
ANGIOTENSIN
II DRINKING
AND
CEREBRAL
BLOOD
parisons were made using Student’s f tests (protected with Bonferroni correction). Pre- to postdifferences in HR, MAP, pH. and blood gases were evaluated with dependent measures Student’s I tests. Minimally acceptable significance level was set at p < 0.05.
RESULTS
Table I summarizes the effects oftreatment on rCBF in each of the 53 brain regions sampled. Pair-wise comparisons revealed that, across treatment groups, changes in rCBF fell into five basic patterns. These included the following: 1) significantly higher blood llow in Ail-treated groups compared to VEH controls irrespective of opportunity to drink, but no difference between AI1 groups. This pattern was observed, for example, in the rostra1 lateral preoptic area, basolateral amygdala, and gustatory cortex, and is designated as pattern u in Table 1; 2) significantiv higher blood flow in AII-treated animals that attempted to drink but were prevented from doing so [All (W-t], as compared to both VEH and AI1 (W+) conditions. In two areas, blood flow remained signiiicantly higher in the AI1 (W+) group than in VEH controls. Conscyuently. two subpatterns emerged, which are designated as patterns h and c’. This general pattern was observed in I I of 53 CNS regions sampled. Representative pattern h regions included the rostra1 lateral hypothalamus, supraoptic and mammiliary nuclei. rostra1 zona incerta and OVLT: the two areas that exhihited pattern (‘. in which AI1 (W+) > VEH. included the median eminence and agranular insular cortex; 3) significant difference only between VEH and AI1 (W--) (higher rCBF) conditions. In all these regions, blood flow was lower in the AI1 (W+) than in the AI1 (W-) condition; however. these differences were not statistically signi~cant. At the same time. AI1 (W+) values were not significantly different from the VEH control condition. This pattern, designated as pattern d. was observed in a relatively broad sample (16 of 53) ofCNS regions, including. for example. median preoptic, paraventricular hypothalamus. caudal lateral preoptic area, olfactory tubercie, cingulate cortex (regions 1.3). corticomediai amygdala and hip~campus (CA I -CA3); 4) lack of any signi~cant treatment effect ( 19of 53 areas). represented by choroid plexus, subcommissural organ ventromedial hypothalamus, and white matter. Mean rCBF in animals treated with AI1 but not allowed to drink [AI1 (W--)1 was significantly increased relative to VEH treatment in 33 of 53 areas sampled. In animals permitted to drink following AI1 treatment [AI1 (W+)], mean rCBF in I1 of these 33 regions was signi~cantiy lower than in group AI1 (W-). reflecting a reversal associated with ingestion or act of ingestion of a small amount of water: 9 of these 11 regions dropped to near control levels. The magnitude of increases across all areas sampled ranged from 2% (medial habenula) to 103% (median eminence). the latter considerably higher than the next highest increase of 668 seen in cingulate cortex (Cg 1,3). Finally, with the sole exception of the paravent~cuiar thaiamus, no region exhibited lower rCBF following AI1 relative to VEH treatment. Arterial blood parameters measured in a separate control experiment are shown in Table 2. A dose of 100 ng AII elevated MAP - I5 mmHg above baseline level (144 vs. 159 mmHg): this shift was significantly higher than the pre to postshift (+ I mmHg) following adn~lnlstration of saline vehide. r(5) = 4.32, /I = O.oO7. Ail did not significantly alter HR. pH, paC02* p,O1. or HCO, when compared to vehicle treatment. Physiological parameter values were within the normal range for rat. All animals drank foliowing AI1 administration; latencies to drink from start of injection ranged from -20 to -65 s.
531
FLOW
DISCUSSION
Principal findings in the present study were that rC’BF was significantly altered by an acute ICV injection of AI1 in a broad spectrum of brain regions that have consistently been implicated in body-~uid homeostasis and that ingestion of a small amount of water partially or nearly completely reversed these rCBF shifts in a number of regions. The short time frame within which putative drinking-induced shifts in rCBF occurred strongly suggests neuronai integration. This study appears to be the first to examine the effect(s) ofcentrally administered AI1 on cerebrai blood flow in discrete brain regions along with concurrent measurement of an associated behavioral response. Notewo~hy and striking is the extent to which AII significantly influenced rCBF. Indeed. as already noted, these shifts were seen in 33 of 53 areas analyzed. Furthermore. even in regions that failed to reach significance, the predominant response to AI1 [AI1 (W-)] was an increase in rCBF, implying an AU-induced increase in neuronai activity. In view of the quite widespread neuronai changes seen. it seems appropriate to address several issues that could impact on an interpretation of these findings. The operation of several factors other than neuronal activation could also be consistent with these data. It might, for example, be argued that such widespread effects can be attributed to passive diffusion of peptide out of the ventricular space, permitting it to indiscriminately influence rCBF in widely distributed regions containing AII receptors. This, however. seems unlikely for a variety of reasons. including consideration of the short time lag between start of ICV injection of peptide and end of blood sampling and sacrifice (105 s) and the distance from the ventricles of a number of significantly affected regions. Regions relatively distant from the ventricles that exhibited significant increases in rCBF included cingulate cortex. corticomedial and basolaterai amygdaioid nuclei, olfactory tubercle, and zona incerta. In contrast. rCBF in the anatomically closer medial habenuiar and suprachiasmatic nuclei, regions containing moderate to high densities of AI1 receptors (24.42) was virtually unaffected. Compared to the VEH condition. mean rCBF in the AI1 (W-) condition was 1.8% higher and 2.8% lower, respectively. in medial habenular and suprachiasmatic nuclei. This evidence against passive diffusion is consistent with observations of Landas et al. ( 15). who reported intense fluorescence in OVLT in rats receiving 100 ng fluorescein thiocarbamyl-labeled AI1 into lateral ventricle 3 min prior to sacrifice, whereas adjacent third ventricle showed additional lowlevel fluorescence only along the ependymal wall. Furthermore, we found no evidence of relationship between magnitude of effect (percent change in rC’BF) and either distance from ventricular space or density of AI1 receptors, the latter as determined by receptor binding ( 12.24,36) and immunohistochemicai ( 18) studies. Note also that CSF-borne peptides do not seem to have ready access into CVOs. which are distinguished by specialized ependymai cells (tanycytes) with tight junctions between adjacent cells, thus forming an effective CSF-CVO hnrrier (3627.29). Landas et al. failed to find fluorescence in the SFO. although it is noted that sampling was restricted to mom ventral SF0 tissue. Indirect evidence exists for some penetration ofthis barrier by All (40): however, its extent and signiticance is unclear. C‘SFborne 411 may gain access to AII receptors in brain regions adjacent to the ventricles through gap junctions associated with the ependymal surface of most of the non-CVO regions of the ventricles (4.27) or be transported into 0’0s by tanycytes (26,29). It has been suggested that in the AV3V region, which has been implicated as the area of central interaction of CSFborne AU, AII might gain entry into the MnPo region adjacent
532
C‘ZkC H .4ND S’l’E:IN
I-ABLE REGIONAL CEREBRAL
BLOOD FLOW FOLLOWING
I
ICV SALINE VEHICLE, Ail WITH, OR AII WITHOUT
INGESTION OF WATER’
Statlstlcal Region Circumventricular Choroid plexus Median eminence Organum vasculosum lamina termin ialis
Subcommissural organ Subfomical organ Hypothalamic Anterior area Arcuate nucleus Dorsomedial nucleus Lateral area (rostral) Lateral area (caudal) Lateral preoptic area (rostral) Lateral preoptic area (caudal) Mammillary nucleus Medial preoptic area (rostral) Medial preoptic area (caudal) Median preoptic (nucleus medianusf Pamvent~cul~ nucleus Posterior nucleus Suprachiasmatic nucleus Supraoptic nucleus Ventromedial nucleus Thalamic Lateral habenula Medial habenula Paraventricular nucleus Reuniens nucleus Subthalamus Zona incerta (rostral) Zona incerta (caudal) Mesocorticolimbic Nucleus accumbens Bed nucleus of stria terminalis Basolateral amygdala Central amygdala Corticomedial amygdala Cingulate cortex (Cg 1,2) Cingulate cortex (Cg 1,3) Diagonal band of Broca Lateral septum Medial septum Media1 frontal cortex (Fr2) Olfactory tubercle Ventral tegmental area Cortical Gustatory cortex Piriform cortex Agranuiar preinsular cortex Dorsal endopiriform nucleus Dentate gyrus Hippocampus (CA 1-CA3) Nigrostriatal/extrapyramidal Caudate-putamen (rostral) Caudate-putamen (caudal) Globus pallidus Substantia nigra pars reticulata pars compacta White matter
AII(W--)
Saline
AII(W
t1
256 i 5 962 13 165k 8 144 r I2 163 IL 15
274 rt 1952 205 rt: 168 I 203 t
141 _f- 9 123+ 8 129 t IO 143 zk IO 1444 2 141 rt 6 152+ 5 223 I I4 1242 9 1292 II
143 i. IO 126 Y!YIO
192 1- 14 133t 5 158 + t5 1941 3 169 + 13 207 -c I I 202 It I4 349 -t 29 l63r 8 158 I I? 153r 8 249 + I5 210 rt 13 134-t 5 204 rtr 8 132-t II
I66 t 123t 1392 155i151 t1865 ll4t 236 ir 143 t 14?i131 I 217 +175i li8rt 168+ 12srt
10 6 4 6 3 9 6 15 10 2 6 I2 5 3 2 5
303221 269 + 13 203 t 4 182 t II l89+- 4 199t 3 169k 6
325 274 224 254 211 239 237
+- 31 F 18 F 18 -t 14 1? 14 t II rir I2
264 rf: 2305 183zt 202 + 18211942 198-t
3 ii 4 II 5 4 9
1042 135+ 209 k 231 + 189 t 152 I2042 199 rt 1982 163 i:
299 -+ 8 137-i- 8 I81 + 13 I48 k 16 ISOk 8 301 rt 25 383 rt 31 2.58 i 8 1832 II 263 It I I 316 + 29 258 r I3 207r 5
260 r l22k 184+ l35k 1572 268 + 318& 211 -t 158k 220 t 286 t 210 Itr I72 t
10 7 6 8 7 IO II 5 4 5 8 14 IO
1411 4 219 r 16 198k 3 148 * 13 119rt II Illk 4
221 L 287 + 285 i 204 t 159 rt: 1445
18 6 6 10 18 9
242 t 225 _t 234 ‘1772 l38k 122i-
9 IO 5 7 7 5
l80+ 157-e 882
9 8 2
222 rt 197rt 110+
3 7 5
218+ 3 I81 t 11 832 4
124t 135It 711
4 3 5
1662 4 202 zk I1 88t4
t39* 11 161 +- 14 85~ 4
109+ 5 174 F JO 196k 6 1382 II
8 8 10 31 14 IO 17 I5 9 I2
7 3 8 15 I5
Slgntficance?
261 + 24 158zk 8 171 tr 5 162 t 24 177 1+_12
ns i‘
h
flS ns
ns ns ns b ns a d b d ns d d ns ns b ns ns RS
c; saline
AII(W+)
d ns b d d nS a
ns d : b IIS
b & d
d d ns
* Values are mean it SEM in ml/100 g/min. $ Patterns ofstatistical significance: (a), AII(W-), AII{W+) > saline;(b), AI&W-) > saline, AII(W+); (c), AI&W-) > saline, AII(Wt) and AII(W+) > saline; fd), AII(W-) > saline; ns, no statistically significant difherences.
ANGIOTENSIN
ii DRINKING
AND CEREBRAL
BLOOD FLOW
TABLE 2 EFFECTS OF ICV SALINE VEHICLE
AND AI1 (100 ng) ON PHYSlOLOCXAL
VARIABLES
A11
Saline Variable (mmHg) HR (hpm) PH P.COZ (mmHg) PA (mmHg) MAP
HCO, (mmoI/l)
143 488 7.43 33.9 19.4
+-6 + 14 f. 0.01 * 0.7 I? 0.8 22.5 F 0.3
Baseline
A
Easelme t1 -7 +0.01 -1.0 +023 PO.2
Values are mean t SEM. (A). magnitude and direction of pre- to postmeasurement * p < 0.0 I, significantly different from pre to post shift following saline
to the OVLT via “leaky” gap junctions and possibly aiso via tanycyte transport directly into the OVLT (30). Consequently. AI1 could reach receptor sites in both these regions. We observed significant elevations in rCBF in group AI1 (W-) in OVLT and median eminence in the present study and suggest that this reflects neuronal integrative activity. It might be worth noting that, although not statistically significant, rCBF also increased in SFO. None of these points, however, is as ~orn~llin~ as the dramatic reversal of rCBF in many of the affected regions when animals were permitted to drink. Because this effect, which occurred within a maximum of - 95 s of drinking a small amount of water [latencies to drink from start of AI1 injection in AI1 (W+) animals were - 10 to -45 s], was seen at a time prior to absorption from the gut, it suggested an action of anticipatory neural “satiety“ signals o~ginating from receptors at the periphery. A separate control experiment measured several physiological parameters in response to ICV injection of both VEH and 100 ng AI1 under conditions essentially identical to those used for rCBF measurement. These data failed to reveal any pre- to posttreatment shifts that would suggest biological activity leading to nonspecific and global shifts in rCBF. As shown in Table 2. blood gases showed only minor shifts. For example, a difference of < 1.O mmHg was found between p,COz pre- to postshifts (reductions) in VEH (- I .O mmHg) and AI1 (- 1.9 mmHg) conditions. As expected, AI1 increased MAP. This increase, however, remained within the range of autoregulation of CBF (10) and is not likely to have led to the observed global blood flow shifts. In a related study, Rettig et al. (33) reported no change in total CBF in unrestrained, mildly water-deprived rats permitted to drink following ICV injection of 100 ng AIL In contrast, significant blood flow shifts were observed in various peripheral organs. The present findings stand in relatively sharp contrast to those recently reported by Tuor et al., also using [‘4C]IAP to measure rCBF in the rat (39). These authors found no significant shifts in any CVO or non-CVO areas following peripheral infusion of AI1 at a gradually increasing rate (-0.02-0.05 &kg-‘/min -I), including many regions also sampled in the present study. It is not clear, however, whether or not these data can be directly compared to those of the present study. For example, the authors report that brains were frozen within 2 min after death by bolus injection ofpentobarbital. This would seem to be near, or perhaps exceed, the upper limit for guarding against postmortem diffusion of tracer. The degree of diffusion within that time could obscure differences in smallerbrain structures( 13,37,46). By comparison, we were able to immerse the skull into -50°C isopentane within
144 471 7.43 33.7 78.5 22.4
+I zk8 + 0.01 to.9 f 1.9 i: 0.3
+7 + 18 f 0.01 i I.2 i 1.7 + 0.7
.I
-
11s -I +o.o I - I.9
If. 3* :?I6
rt 0.02 IL0.9
t4.i
ir1.9
-0.9
i- 0.X
shift
30 s (usually ~20 s) of a sharply defined time point. namely, decapitation. Tuor et al. did not measure drinking or attempt to drink; however, an elevation in mean arterial blood pressure of 24% was reported following AI1 infusion, indicating an All response sufficient to have resulted in drinking if tested. Nonetheless, it remains surprising that no hint of change in SF0 blood flow was observed if, as has been suggested, the SF0 is the principal site of dipsogenic action of blood-~rne Aff. As noted, the widespread shifts in rCBF following All administration [AI1 (W-) condition] correspond to many of the sites having been implicated in aspects of body fluid homeostasis. In view of the aforementioned discussion, we suggest that these effects might likely reflect neuronal activity originating from one or more of the CVOs, along with other elements of the AV3V region, the latter including OVLT and MnPo area. As it is known that alterations in regional metabolic activity reflect principally the activity of synaptic inputs to that region (22), it is therefore conceivable that increases in blood flow seen in a number of regions reflect synaptic activity in projection fields. Indeed, a map of the regions significantly affected coincides to substantial degree with the dist~bution of efferent projections of CVO and AV3V areas ( 17,23,30). It has been suggested that the MnPo nucleus of AV3V might play a key role in integrating and relaying the effects of AI1 from both brain and blood sources ( 14,29). The MnPo is reciprocally connected with both the SF0 and the OVLT and sends efferents to a number of other regions including, in part, supraoptic nucleus, lateral hypothaiamus, and paravent~cular hypothalamus (2,44). With the exception ofthe SFO, all these regions exhibited significant increases in blood flow following AI1 treatment in the present study. Furthermore, elevated blood flow was reversed in the supraoptic nucleus and rostra1 lateral hypothalamic region by ingestion of a small amount ofwater. Projections from MnPo not significantly affected in the present study included bed nucleus of the stria terminalis and arcuate nuclei. Lesions of the MnPo/OVLT region have been shown to severely impair AIIinduced drinking (I ,3,14,20), and thirst-inducing extracellular fluid depletion with polyethylene glycol enhances norepinephrine turnover in the MnPo region (45). Furthermore, MnPo lesions have been shown to disrupt circadian influences on drinking (6). It is interesting to note earlier behavioral studies linking medial amygdala (7,47) and zona incerta (8,35,41) with drinking and salt preference in the rat. Electrophysiological data have implicated the rostra1 zona incerta in neural integration of inputs from osmotic and extracellular thirst (25). Corticomedial amygdala and zona incerta receive projections from SF0 ( 17) and exhibited significant moderate to high increases
534
in blood flow with AI1 treatment; in the rostra1 zona incerta. the AI1 effect was reversed in the group that drank. Data from the rat also indicate that efferents from OVLT terminate in the cingulate gyrus and hippocampal dentate gyrus (30); blood flow in the cingulate cortex was significantly elevated following AI1 treatment. The metabolic consequences of excitatory and inhibitory postsynaptic activity are indistinguishable using blood flow procedures alone; consequently, questions of synaptic polarity in
affected regions and of functional significance cannot be addressed with current protocols. The present study, however, appeared to develop a potentially useful map for further study of the dipsogenic properties of angiotensin II. ACKNOWl.ElXEMEN'IS
This research was supported in part by grants from the Graduate School of Marquette University (D.A.C.) and USPHS-DA 05012 (E.A.S.). Technical supportby Scott Fuller is gratefully acknowledged.
REFERENCES 1. Bellin, S. 1.; Landas, S. K.; Johnson, A. K. Localized injections of
2.
3. 4.
5.
6. 7.
8. 9. 10. il. 12.
13. 14.
15. 16.
17. 18.
19.
6-hydroxydopamine into lamina terminalis-associated structures: Effects on experimentally induced drinking and pressor responses. Brain Res. 416:75-83; 1987. Brown, E. R.; Standaert, D. G.; Saper, C. B. Efferent connections of the medial preoptic region in the rat. Sot. Neurosci. Abstr. 1I: 827; 1985. Buggy, J.; Johnson, A. K. Angiotensin-induced thirst: Effects of third ventricle obstruction and periventricular ablation. Brain Res. 149: 117-128; 1978. Fenstermacher, J. D.; Blasberg, R. G.; Path& C. S. Methods for quantifying the transport of drugs across brain barrier systems. Pharmacol. Ther. 142 17-248; 198 1. Fhiharty, S. J.; Epstein, A. N. Sodium appetite elicited by intracerebrovent~cular infusion of an~otensin II in the tat: II. Synetgistic interaction with systemic rnine~~~o~icoi~. Behav. Neurosci. 97: 746-758; 1983. Gardiner, T. W.; Snicker, E. M. Impaired drinking responses of rats with lesions of nucleus medianus: Circadian dependence. Am. J. Physiol. 248:R224-R230; 1985. Gentil, C. G.; Antunes-Rodrigues, J.; Negro-Vilar, A.; Covian, M. R. Role of amygdaloid complex in sodium chloride and water intake in the rat. Physiol. Behav. 3:981-985: 1968. Grossman, S. P. A massessment ofthe brain mechanisms that control thirst. Neurosci. Biobehav. Rev. 8:95-104; 1984. Hartle, D. K.; Brody, M. J. The angiotensin II pressor system of the rat forebrain. Circ. Res. 54:355-366: 1984. Herpin, D. The effects of antihypertensive drugs on the cerebral blood flow and its regulation. Prog. Neurobiol. 35:7.5-83; 1990. Hsiao. S.: Eostein. A. N.: Camardo. J. S. The dipsoaenic notencv of peripheral angiotensin II. Horm. Behav. 8: 129-i&; 1977. . Israel, A.; Plunkett, L. M.; Saavedra, J. M. Quantitative autoradiographic characterization of receptors for angiotensin II and other neuropeptides in individual brain nuclei and peripheral tissues from single-rats. Cell. Mol. Neurobiol. 5:21 l-222;. 1985. Jav. ,. T. M.: Lucinnani. _ ,,G.: Crane. A. M.: Jehle. J.: Sokolaff. L. Measurement of local cerebral blood flow with [‘“Cl Iodoantipyrine in the mouse. J. Cereb. Blood Flow Metab. 8: 12 I-l 29; 1988. Johnson, A. K. The periventricular anteroventral third ventricle (AV3V): Its relationship with the subfomical organ and neural systems involved in maintaining body fluid homeostasis. Brain Res. Bull. 15:595-601; 1985. Landas, S.; Phi&s, M. I.; Stamler, J. F.; Raizada, M. K. Visualization of specific angietensin II binding sites in the brain by fluorescent microscopy. Science 210:791-793; 1980. Lang, R. E.; Unger, T.; Rascher, W.; Ganten, D. Brain angiotensin. In: Iversen. L. L.: Iversen. S. D.: Snvder. S. H.. eds. Handbook of psychoph&macoiogy, vol: 16: Neur&eptides. New York: Plenum Press; 1983:307-361. Lind, R. W. The subfomical organ: Neural connections. In: Gross, P. M., ed. Circumventricular organs and body fluids, vol. I. Boca Raton, FL: CRC Press; 1987127-42. Lind, R. W.; Swanson, L. W.; Ganten, D. Organimtion of angiotensin II immunoreactive cells and fibers in the rat central nervous system: An immunohistochemical study. Neuroendoctinology 40:2-24; 1985. Mangiapane, M. L.; Simpson, J. B. Subfomical organ: Forebrain site of pressor and dipsogenic action of angiotensin II. Am. J. Physiol. 239:R382-R389: 1980.
20. Man~a~ne, M. L.; Thrasher, T. N.; Keil, L. C.; Simpson, J. B.: Ganong, W. F. Deficits in drinking and vasopressin secretion after lesions of the nucleus medianus. Neuroendocrinology 37:73-77; 1983. 21. Mangiapane, M. L.; Thrasher, T. N.; Keil, L. C.; Simpson, J. B.: Ganong, W. F. Role for the subfomical organ in vasopressin secretion. Brain Res. Bull. 13:43-47; 1984. 22. Mata, M.; Fink, D. J.; Gainer, H.; Smith, C. B.; Davidsen, L.; Savaki, H.; Schwartz, W. J.; Sokoloff, L. Activitydependent energy metabolism in rat posterior pituitary primarily reflects sodium pump activity. J. Neurochem. 342 13-2 15; 1980. 23. McKinley, M. J.; MeAllen, R. M.; Mendelsohn, F. A. 0.; Allen, A. M.; Chai, S. Y.; Oldfield, B. J. Circumventricular organs; neuroendocrine interfaces between the brain and the hemal milieu. Front. Neur~nd~nol. 11:91-127; 1990. 24, Mendel~hn, F. A. 0.; Quirion, R.; Saavedra, J. M.; Aguilera, G.; Catt, K. J. Autoradiomdio~phic localization of angiotensin II receptors in rat brain. Proc. Nat]. Acad. Sci. USA 81:15X-1579; 1984. 25. Mok, D.; Mogenson, G. J. Convergence of signals in the zona incerta for angiotensin-mediated and osmotic thirst. Brain Res. 407:332340; 1987. 26. Palkovits, M. Summary of structural and functional aspects of the circumventricular organs. In: Gross, P. M., ed. Circumventricular organs and body fluids, vol. II. Boca Raton, FL: CRC Press; 1987: 209-218. 27. Peters, A.; Palay, S. L.; Webster, H. deF. The fine structure of the nervous system: Neurons and their supporting cells. 3rd ed. Oxford: Oxford University Press; I99 1. 28. Phillips, M. I. Functions of an~otensin in the central nervous system. Annu. Rev. Phvsiol. 49413-435: 1987. 29. Phihips, M. I. Brain an~otensin.‘In: Gross, P. M., ed. Circumventricular organs and body fluids, vol. III. Boca Raton. FL: CRC Press; 1987:163-182. 30. Phillips, M. 1.; Camacho, A. Neural connections of the organum vasculosum of the lamina terminalis. in: Gross, P. M., ed. Circumventricular organs and body fluids, vol. I. Boca Raton, FL: CRC Press 1987:157-170. 31. Ramsay, D. J.; Thrasher, T. N. Thirst and water balance. In: Stricker, E. M., ed. Handbook of behavioral neurobiology, vol. 10: Neurobiology of food and fluid intake. New York: Plenum Press; 1990: 353-386. 32. Rettig, R.; Ganten, D.; Lang, R. E.; Unger, Th. The renin-angiotensin system in the central cot&o1 of blood pressure. Eur. Heart J. 8(suppl. B):129-132; 1987. 33. Rettig, R.; White, F. C.; Printz, M. P. Effects of~n~ng and central an~otensin II stjmulation on organ blood flow in conscious rats. Brain Res. 50.5:25l-256; f989. 34. Rowland, N. E.; Fregly, M. J. Sodium appetite: Species and strain differences and role of renin-angiotensin-aldosterone system. Appetite i 1:143-178; 1988. 35. Rowland, N.; Grossman, S. P.; Grossman, L. Zona incerta lesions: Regulatory drinking deficits to intravenous NaCI, angiotensin, but not to salt in the food. Physiol. Behav. 23:745-750; 1979. 36. Saavedra, J. M.; Israel, A.; Plunkett, L. M.; Kurihara, M.; Shigematsu, K.: Correa, F. M. A. Quantitative distribution of anniotensin 11 binding sites in rat brain by autoradiography. Peptides?:679-687; 1986. 37. Sakurada, 0.; Kennedy, C.; Jehle, J.: Brown, J. D.; Carbin. G. L.; Sokoloff, L. Measurement of local cerebral blood
ANGIOTENSIN
38.
39.
40.
4 I.
42.
I1 DRINKING
AND CEREBRAL
BLOOD FLOW
flow with iodo[r4C]antipyrine. Am. J. Physiol. 234:H59-H66: 1978. Steele, M. K.: Negro-Vilar, A.; McCann, S. M. Effect ofangiotensin II on in viva and in vitro release of anterior pituitary hormones in the female rat. Endocrinology 109:893-899; 198 1. Tuor. U. I.; Kondysar, M. H.: Harding, R. K. Effect of angiotensin II and peptide YY on cerebral and circumventricular blood flow. Peptides 9:141-149; 1988. Van Houten. M.: Mangiapane, M. L.; Reid, I. A.; Ganong, W. F. [Sar’.Ala”] angiotensin II in cerebrospinal fluid blocks the binding of blood-borne [“5]angiotensin II to the circumventricular organs. Neuroscience 10:1421-1426: 1983. Walsh L. L..;Grossman, S. P. Electrolytic lesions and knife cuts in the region of the zona incerta impair sodium appetite. Physiol. Behav. l&587-596: 1977. Wamsley. J. K.: Herblin. W. F.: Alburges. M. E.; Hunt, M. Evidence for the presence of angiotensin B-type I receptors in brain. Brain Res. Bull. 25:397-400: 1990.
535
43. Weiss, M. L.; Moe, K. E.; Epstein. A. N. Interference with central actions ofangiotensin II suppresses sodium appetite. Am. J. Physiol. 250:R250-R259: 1986. 44. Wilkin. L. D.: Mitchell, L. D.; Ganten, D.; Johnson, A. K. The supraoptic nucleus: Afferents from areas involved in control of body fluid homeostasis. Neuroscience 28:573-584: 1989. 45. Wilkin, L. D.; Patel, K. P.; Schmid, P. G.; Johnson. A. K. Increased norepinephrine turnover in median preoptic nucleus following reduced extracellular fluid volume. Brain Res. 423:369372; 1987. 46. Williams J. L.; Shea, M.: Furlan, A. J.: Little. J. R.; Jones, S. C. Importance of freezing time when iodoantipyrine is used for measurement of cerebral blood flow. Am. J. Physiol. 261:H252-H256: 1991. 47. Zolovick. A. J.; Avrith, D.; Jalowiec, J. E. Reversible colchicineinduced disruption of amygdaloid function in sodium appetite. Brain Res. Bull. 5:35-39: 1980.
Bruin Re\eorch Brr/l6~i/n,Vol. 28. pp. xX9-535, 1992 Printed in the USA. 411rights reserved.
Effect of Drinking on Angiotensin-II-Induced Shifts in Regional Cerebral Blood Flow in the Rat DONALD
A. CZECH*’
AND
ELLIOT
A. STElNt
*Hirjpsychoiogy Laboratory, Department ofP~sychology, MarqWtc C~niversit~~, Miluaukee. WI 53233 and fDc>partmmts of Psychiutry, Pharmacologic, and CeNlrlar Biologic and ilnutom!‘, Me&al College qf U’isconsin. Milwultkw U7 532% Received
19 September
199 1
CZECH, D. A. AND E. A. STEIN. lz@ci t$drinkittg on an,~iorrt?.rin-ll-ittd1ct.c~ .sii$i.sin rqiond trrdml hioctd /h, in i/w ml. BRAIN RES BULL 28(4) 529-535, 1992.-A map of brain regions affected by central administration of the octapeptide angiotensin II (All) and that would further reflect the consequences of All’s well-known dipsugenic action was developed. Regional cerebral blood flow (rCBF) as an indicator ofneuronalactivity was measured in conscious rats shortly after an ICV bolus injection of 100 ng All or saline vehicle (VEH). AII-treated animals were further subdivided into two groups that were either permitted to drink [All (Wt)] or whose water was removed when drinking was attempted [AI1 (W--j]. When compared to VEH condition. blood flow increased significantly within I min after All treatment in 33 of 53 regions sampled in those rats not given an opportunit) to drink. In I I of these 33 regions. ingestion of a small amount of water was associated with a reversal of All-induced elevation in blood llow [i.e., AII (W+) < AI1 (W-)]; these regions included the organum vasculosum lamina terminalis. rostra1 lateral hypothalamus. supraoptic nuclei, rostra1 zona incerta, and median eminence. A group of similarly treated rats exhibited a significant elevation of mean arterial biood pressure following AI1 treatment without significant shifts in artenal blood gases, pH, or bicarbonate. These data are consistent with prominent involvement of the anteroventral third ventricular region of the rat brain. The results further indicate that rCBF may be a sensitive measure for the identi~cation ofcentralsites ofaction of AII as a dipsogenic agent and may reveal distinctions between regions associated primariiy with Initiation of drinking and those reflecting the results of subsequent behavioral events. Cerebral blood flow
Angiotensin II
Circumventricular organs
THE octapeptide angiotensin 11 (All) is involved in a number of central integrative functions linked to body-fluid homeostasis, including stimulation of drinking (16,28,3 1), elevation of arterial blood pressure (9.19,32), increased salt appetite (5,34,43), and release of several pituitary hormones (21,38). Specifically, for example. All is known to be a potent dipsogen when administered either centrally or ~~pheraliy (11,19,28). Althou~ various aspects of All’s role in fluid regulation and drinking remain controversial and unclear, it is widely accepted that All’s dipsogenic and other regulatory effects are mediated, at least in part, through stimulation of receptor sites located within the circumventricular organs (CVOs) of the brain. These regions. with the apparent exception of the subcommissural organ, contain moderate to high concentrations of All receptors [e.g., see ( 12.36)], lie outside the blood-brain barrier, and are potentially accessible to both blood- and cerebrospinal fluid (CSF)-borne AIL Previous studies point to the subfornical organ (SFO) as the primary site of action for the dipsogenic activity of blood-
Drinking
Autoradiography
Iodoantipyrine
borne AI1 whereas All in CSF appears to initiate dipsogenic activity principally at the organum vasculosum lamina tetminaiis (OVLT) and hypothalamic median preoptic (MnPo) area located in the anteroventral third ventricular region (AV3V) (28,29). An extensive literature exists with a focus on mapping neural circuitry between the CVOs and surrounding CNS areas believed to be involved in AII’s dipsogenic effect(s) (17.23.30), as well as on the behavioral, physiological, and pharmacological effects of manipulating these areas. In contrast, attempts to establish temporal linkage between neuronal activity in these and other more distant sites with ongoing behavioral state have received less attention. Our concern was to further study All’s putative role in drinking through developing a broader map that would highlight those brain regions that participate in a functional system(s) underlying particular behavioral states. In the present study, we attempted to identify regions that would be activated in response to an ICV bolus injection of All and to distinguish those linked to All’s dipsogenic activity; the
’ Requests for reprints should be addressed to Donald A. Czech. repayment Milwaukee, WI 53233.
529
of Psychology, Marquette University, Schroeder Complex 454.
530
CZI:C’li AND Sl EIN
latter was accomplished by either offering rats the opportunity to demonstrate an intent to drink or following a brief drinking bout. To this end, we measured regional cerebral blood flow (rCBF) as an indicator of neuronal activity following ICV administration of AII. The [‘4C]iodoantipyrine (IAP) autoradiographic technique used for measuring rCBF (37) has a number of advantages over other metabolic procedures, including good spatial and temporal resolution. In providing a brief time window (30 s) within which to measure metabolic consequences, this rCBF technique allows one to look at discrete time points rather than being limited to measuring average shifts (as, e.g., with the 45-min 2-deoxyglucose method for glucose utilization). Such temporal resolution is clearly critical in being able to focus and lock in on behavioral/biological changes relatively quickly and thereby minimize obscuring specific and transient neuronal events with “background” activity. Rapid changes in rCBF as a consequence of neuronal activity thus provide an opportunity to closely monitor and link behavioral changes to activity in underlying neural systems. Under most conditions, rCBF is believed to be tightly coupled to local metabolic demand ( 13.37). Furthermore, numerous and widespread integrated systems can be assessed simultaneously with autoradio~aphic procedures. In attempting to isolate those patterns of blood flow that reflect AII’s dipsogenic action, we manipulated opportunity to drink. Such manipulation might thereby permit distinguishing between so-called stimulus-driven and behavior-driven circuitry in the brain.
METHOD
Animals and Preparation Male Sprague-Dawley-derived rats (Sasco, Madison, WI) weighing 300-325 g were individually housed in polycarbonate tub-type cages in a tem~rature~ontrolled room with lights on between 0830- 1900 h. Pelleted food and tap water were availabie ad lib. After several days of acclimation to the laboratory environment, a single 26-ga stainless steel guide cannula was stereotaxically implanted under Chloropenf anesthesia (2.5 ml/kg, IP), with tip positioned -0.4 mm above a lateral ventricle; a minimum of 4 days was allowed for recovery. Animals then experienced a restraint procedure of increasing duration for 5 days progressing from 1 to 5 h per day prior to measurement of rCBF or physiological variables. The restraint consisted of wrapping in a terry cloth jacket that immobilized both fore and hind limbs. Rats rapidly acclimated to the restraint, most ceasing to struggle by the third day. During this period and while under restraint, animals were also tested for drinking in response to an ICV injection of 100 ng AI1 (Sigma, St. Louis, MO) delivered in 2.0 ~1 sterile saline over 15 s. On the day of final testing and sacrifice, rats were again anesthetized with Chloropent and femoral arterial and venous catheters (PE 10) were implanted. The incisions were treated with a local anesthetic (2% xylocaine gel) and closed. Animals were then placed in the restraint condition and allowed to recover from anesthesia for a minimum of 5 h. An intravenous injection of 500 IV/kg heparin sodium in 0.5 ml sterile saline was given 1 h prior to start of testing. Tap water was provided during the recovery period via a sipper tube positioned directly in front of the animal. Animals could move their heads freely under restraint and readily drank. Access to water during subsequent testing was dependent on experimental group ~signment, as indicated next.
Animals were assigned to one of three treatment groups (II = 4lgroup) and manipulated postrecovery as follows: Two groups were injected ICV with 100 ng All in 2.0 ~1 normal saline and blood flowdetermined starting 1 min after injection termination. In one of these groups [AI1 (W-)], water was immediatelt removed when the rat attempted to drink (operationally deined as first fick at sipper tube). In the second group [Ali (Wi)]. water continued to be avaiiable ad lib throughout the entire procedure. Animals in a third group (VEH) were injected 10’ with 2.0 ~1 normal saline vehicle prior to initiation of blood flow procedures. Water was again present: however. no animal in the VEH condition attempted to drink. rCBF was measured by the method of Sakurada et al. (37). Briefly, this automdiograph~c method involved the infusion of 0.5 mf normal saline containing fO0 .&i/kg ~‘4C]i~oantipy~ne (45.5 mCi/mmol. Amersham) at a constant rate over 30 s into the femoral vein. Arterial blood samples were collected every 5 s onto preweighed filter paper throughout the infusion period, Immediately after the infusion period, rats were rapidly decapitated, the lower jaw and skin covering the skull were quickly (~30 s) removed, and the skull was immersed in isopentanc (-SO’C) for several minutes, then stored at --80°C. Filter papers were reweighed in counting vials, Aquassure (New England Nuclear) added to the vials, and the radioactivity of samples determined by liquid scintillation spectroscopy on a Beckman LS6000TA counter. Brains were subsequently removed from the skull in a cryostat chamber at -20°C and stored at ~80°C prior to sectioning. Brains were later cut into 30-pm sections at --20°C in a cryostat (Reichert-Jung, Model 850), thaw mounted onto glass slides, and dried on a slide warmer (5O’C). Slides were apposed to X-ray film (Kodak MR-1) with calibrated methacrylate [14C] standards (Amersham) in standard X-ray cassettes for an appropriate period (generally 4-6 weeks). Following film development, slides were stained with thionin for subsequent anatomic localization of structures of interest. Brain regions were analyzed on an MCID Image Analyzer (Imaging Research. St. Catherine. Ontario) and reported as cerebral blood flow in ml/ 100 g/min.
A separate group of rats was treated in a manner similar to that just noted; however, rCBF was not measured. Samples of arterial blood (150 al) were drawn just before and I min after a 2.0 gl ICV injection of 100 ng AI1 in sterile normal saline and following 0.5 ~1 of vehicle only. The vehicle-only condition preceded the AI1 injection by approximately 10 min in a withinsubjects design (n = 6). The smaller volume was used in the vehicle condition to avoid intr~ucing a relatively large volume of fluid into the ventricle within a rather short period of time. All animals were allowed to drink, as in the rCBF AI1 (W+) condition. Blood samples were assayed for paC02. paOz, HCO; , and pH on a blood gas analyzer (Corning-CIBA, Model 168). Heart rate (HR) and mean arterial pressure (MAP) were continuously monitored via a Statham pressure transducer and appropriate electronics on a Grass Model 79D ~ly~aph. These measures were also evaluated before and 1 min after treatment. As noted, water was available ad lib prior to and during the sampling/test period. Rats were sacrificed with an overdose of anesthetic at the completion of data collection. Statistical Ana&ses rCBF data were evaluated separately in each brain region with one-way analyses of variance (ANOVAs); pairwise com-
ANGIOTENSIN
II DRINKING
AND
CEREBRAL
BLOOD
parisons were made using Student’s f tests (protected with Bonferroni correction). Pre- to postdifferences in HR, MAP, pH. and blood gases were evaluated with dependent measures Student’s I tests. Minimally acceptable significance level was set at p < 0.05.
RESULTS
Table I summarizes the effects oftreatment on rCBF in each of the 53 brain regions sampled. Pair-wise comparisons revealed that, across treatment groups, changes in rCBF fell into five basic patterns. These included the following: 1) significantly higher blood llow in Ail-treated groups compared to VEH controls irrespective of opportunity to drink, but no difference between AI1 groups. This pattern was observed, for example, in the rostra1 lateral preoptic area, basolateral amygdala, and gustatory cortex, and is designated as pattern u in Table 1; 2) significantiv higher blood flow in AII-treated animals that attempted to drink but were prevented from doing so [All (W-t], as compared to both VEH and AI1 (W+) conditions. In two areas, blood flow remained signiiicantly higher in the AI1 (W+) group than in VEH controls. Conscyuently. two subpatterns emerged, which are designated as patterns h and c’. This general pattern was observed in I I of 53 CNS regions sampled. Representative pattern h regions included the rostra1 lateral hypothalamus, supraoptic and mammiliary nuclei. rostra1 zona incerta and OVLT: the two areas that exhihited pattern (‘. in which AI1 (W+) > VEH. included the median eminence and agranular insular cortex; 3) significant difference only between VEH and AI1 (W--) (higher rCBF) conditions. In all these regions, blood flow was lower in the AI1 (W+) than in the AI1 (W-) condition; however. these differences were not statistically signi~cant. At the same time. AI1 (W+) values were not significantly different from the VEH control condition. This pattern, designated as pattern d. was observed in a relatively broad sample (16 of 53) ofCNS regions, including. for example. median preoptic, paraventricular hypothalamus. caudal lateral preoptic area, olfactory tubercie, cingulate cortex (regions 1.3). corticomediai amygdala and hip~campus (CA I -CA3); 4) lack of any signi~cant treatment effect ( 19of 53 areas). represented by choroid plexus, subcommissural organ ventromedial hypothalamus, and white matter. Mean rCBF in animals treated with AI1 but not allowed to drink [AI1 (W--)1 was significantly increased relative to VEH treatment in 33 of 53 areas sampled. In animals permitted to drink following AI1 treatment [AI1 (W+)], mean rCBF in I1 of these 33 regions was signi~cantiy lower than in group AI1 (W-). reflecting a reversal associated with ingestion or act of ingestion of a small amount of water: 9 of these 11 regions dropped to near control levels. The magnitude of increases across all areas sampled ranged from 2% (medial habenula) to 103% (median eminence). the latter considerably higher than the next highest increase of 668 seen in cingulate cortex (Cg 1,3). Finally, with the sole exception of the paravent~cuiar thaiamus, no region exhibited lower rCBF following AI1 relative to VEH treatment. Arterial blood parameters measured in a separate control experiment are shown in Table 2. A dose of 100 ng AII elevated MAP - I5 mmHg above baseline level (144 vs. 159 mmHg): this shift was significantly higher than the pre to postshift (+ I mmHg) following adn~lnlstration of saline vehide. r(5) = 4.32, /I = O.oO7. Ail did not significantly alter HR. pH, paC02* p,O1. or HCO, when compared to vehicle treatment. Physiological parameter values were within the normal range for rat. All animals drank foliowing AI1 administration; latencies to drink from start of injection ranged from -20 to -65 s.
531
FLOW
DISCUSSION
Principal findings in the present study were that rC’BF was significantly altered by an acute ICV injection of AI1 in a broad spectrum of brain regions that have consistently been implicated in body-~uid homeostasis and that ingestion of a small amount of water partially or nearly completely reversed these rCBF shifts in a number of regions. The short time frame within which putative drinking-induced shifts in rCBF occurred strongly suggests neuronai integration. This study appears to be the first to examine the effect(s) ofcentrally administered AI1 on cerebrai blood flow in discrete brain regions along with concurrent measurement of an associated behavioral response. Notewo~hy and striking is the extent to which AII significantly influenced rCBF. Indeed. as already noted, these shifts were seen in 33 of 53 areas analyzed. Furthermore. even in regions that failed to reach significance, the predominant response to AI1 [AI1 (W-)] was an increase in rCBF, implying an AU-induced increase in neuronai activity. In view of the quite widespread neuronai changes seen. it seems appropriate to address several issues that could impact on an interpretation of these findings. The operation of several factors other than neuronal activation could also be consistent with these data. It might, for example, be argued that such widespread effects can be attributed to passive diffusion of peptide out of the ventricular space, permitting it to indiscriminately influence rCBF in widely distributed regions containing AII receptors. This, however. seems unlikely for a variety of reasons. including consideration of the short time lag between start of ICV injection of peptide and end of blood sampling and sacrifice (105 s) and the distance from the ventricles of a number of significantly affected regions. Regions relatively distant from the ventricles that exhibited significant increases in rCBF included cingulate cortex. corticomedial and basolaterai amygdaioid nuclei, olfactory tubercle, and zona incerta. In contrast. rCBF in the anatomically closer medial habenuiar and suprachiasmatic nuclei, regions containing moderate to high densities of AI1 receptors (24.42) was virtually unaffected. Compared to the VEH condition. mean rCBF in the AI1 (W-) condition was 1.8% higher and 2.8% lower, respectively. in medial habenular and suprachiasmatic nuclei. This evidence against passive diffusion is consistent with observations of Landas et al. ( 15). who reported intense fluorescence in OVLT in rats receiving 100 ng fluorescein thiocarbamyl-labeled AI1 into lateral ventricle 3 min prior to sacrifice, whereas adjacent third ventricle showed additional lowlevel fluorescence only along the ependymal wall. Furthermore, we found no evidence of relationship between magnitude of effect (percent change in rC’BF) and either distance from ventricular space or density of AI1 receptors, the latter as determined by receptor binding ( 12.24,36) and immunohistochemicai ( 18) studies. Note also that CSF-borne peptides do not seem to have ready access into CVOs. which are distinguished by specialized ependymai cells (tanycytes) with tight junctions between adjacent cells, thus forming an effective CSF-CVO hnrrier (3627.29). Landas et al. failed to find fluorescence in the SFO. although it is noted that sampling was restricted to mom ventral SF0 tissue. Indirect evidence exists for some penetration ofthis barrier by All (40): however, its extent and signiticance is unclear. C‘SFborne 411 may gain access to AII receptors in brain regions adjacent to the ventricles through gap junctions associated with the ependymal surface of most of the non-CVO regions of the ventricles (4.27) or be transported into 0’0s by tanycytes (26,29). It has been suggested that in the AV3V region, which has been implicated as the area of central interaction of CSFborne AU, AII might gain entry into the MnPo region adjacent
532
C‘ZkC H .4ND S’l’E:IN
I-ABLE REGIONAL CEREBRAL
BLOOD FLOW FOLLOWING
I
ICV SALINE VEHICLE, Ail WITH, OR AII WITHOUT
INGESTION OF WATER’
Statlstlcal Region Circumventricular Choroid plexus Median eminence Organum vasculosum lamina termin ialis
Subcommissural organ Subfomical organ Hypothalamic Anterior area Arcuate nucleus Dorsomedial nucleus Lateral area (rostral) Lateral area (caudal) Lateral preoptic area (rostral) Lateral preoptic area (caudal) Mammillary nucleus Medial preoptic area (rostral) Medial preoptic area (caudal) Median preoptic (nucleus medianusf Pamvent~cul~ nucleus Posterior nucleus Suprachiasmatic nucleus Supraoptic nucleus Ventromedial nucleus Thalamic Lateral habenula Medial habenula Paraventricular nucleus Reuniens nucleus Subthalamus Zona incerta (rostral) Zona incerta (caudal) Mesocorticolimbic Nucleus accumbens Bed nucleus of stria terminalis Basolateral amygdala Central amygdala Corticomedial amygdala Cingulate cortex (Cg 1,2) Cingulate cortex (Cg 1,3) Diagonal band of Broca Lateral septum Medial septum Media1 frontal cortex (Fr2) Olfactory tubercle Ventral tegmental area Cortical Gustatory cortex Piriform cortex Agranuiar preinsular cortex Dorsal endopiriform nucleus Dentate gyrus Hippocampus (CA 1-CA3) Nigrostriatal/extrapyramidal Caudate-putamen (rostral) Caudate-putamen (caudal) Globus pallidus Substantia nigra pars reticulata pars compacta White matter
AII(W--)
Saline
AII(W
t1
256 i 5 962 13 165k 8 144 r I2 163 IL 15
274 rt 1952 205 rt: 168 I 203 t
141 _f- 9 123+ 8 129 t IO 143 zk IO 1444 2 141 rt 6 152+ 5 223 I I4 1242 9 1292 II
143 i. IO 126 Y!YIO
192 1- 14 133t 5 158 + t5 1941 3 169 + 13 207 -c I I 202 It I4 349 -t 29 l63r 8 158 I I? 153r 8 249 + I5 210 rt 13 134-t 5 204 rtr 8 132-t II
I66 t 123t 1392 155i151 t1865 ll4t 236 ir 143 t 14?i131 I 217 +175i li8rt 168+ 12srt
10 6 4 6 3 9 6 15 10 2 6 I2 5 3 2 5
303221 269 + 13 203 t 4 182 t II l89+- 4 199t 3 169k 6
325 274 224 254 211 239 237
+- 31 F 18 F 18 -t 14 1? 14 t II rir I2
264 rf: 2305 183zt 202 + 18211942 198-t
3 ii 4 II 5 4 9
1042 135+ 209 k 231 + 189 t 152 I2042 199 rt 1982 163 i:
299 -+ 8 137-i- 8 I81 + 13 I48 k 16 ISOk 8 301 rt 25 383 rt 31 2.58 i 8 1832 II 263 It I I 316 + 29 258 r I3 207r 5
260 r l22k 184+ l35k 1572 268 + 318& 211 -t 158k 220 t 286 t 210 Itr I72 t
10 7 6 8 7 IO II 5 4 5 8 14 IO
1411 4 219 r 16 198k 3 148 * 13 119rt II Illk 4
221 L 287 + 285 i 204 t 159 rt: 1445
18 6 6 10 18 9
242 t 225 _t 234 ‘1772 l38k 122i-
9 IO 5 7 7 5
l80+ 157-e 882
9 8 2
222 rt 197rt 110+
3 7 5
218+ 3 I81 t 11 832 4
124t 135It 711
4 3 5
1662 4 202 zk I1 88t4
t39* 11 161 +- 14 85~ 4
109+ 5 174 F JO 196k 6 1382 II
8 8 10 31 14 IO 17 I5 9 I2
7 3 8 15 I5
Slgntficance?
261 + 24 158zk 8 171 tr 5 162 t 24 177 1+_12
ns i‘
h
flS ns
ns ns ns b ns a d b d ns d d ns ns b ns ns RS
c; saline
AII(W+)
d ns b d d nS a
ns d : b IIS
b & d
d d ns
* Values are mean it SEM in ml/100 g/min. $ Patterns ofstatistical significance: (a), AII(W-), AII{W+) > saline;(b), AI&W-) > saline, AII(W+); (c), AI&W-) > saline, AII(Wt) and AII(W+) > saline; fd), AII(W-) > saline; ns, no statistically significant difherences.
ANGIOTENSIN
ii DRINKING
AND CEREBRAL
BLOOD FLOW
TABLE 2 EFFECTS OF ICV SALINE VEHICLE
AND AI1 (100 ng) ON PHYSlOLOCXAL
VARIABLES
A11
Saline Variable (mmHg) HR (hpm) PH P.COZ (mmHg) PA (mmHg) MAP
HCO, (mmoI/l)
143 488 7.43 33.9 19.4
+-6 + 14 f. 0.01 * 0.7 I? 0.8 22.5 F 0.3
Baseline
A
Easelme t1 -7 +0.01 -1.0 +023 PO.2
Values are mean t SEM. (A). magnitude and direction of pre- to postmeasurement * p < 0.0 I, significantly different from pre to post shift following saline
to the OVLT via “leaky” gap junctions and possibly aiso via tanycyte transport directly into the OVLT (30). Consequently. AI1 could reach receptor sites in both these regions. We observed significant elevations in rCBF in group AI1 (W-) in OVLT and median eminence in the present study and suggest that this reflects neuronal integrative activity. It might be worth noting that, although not statistically significant, rCBF also increased in SFO. None of these points, however, is as ~orn~llin~ as the dramatic reversal of rCBF in many of the affected regions when animals were permitted to drink. Because this effect, which occurred within a maximum of - 95 s of drinking a small amount of water [latencies to drink from start of AI1 injection in AI1 (W+) animals were - 10 to -45 s], was seen at a time prior to absorption from the gut, it suggested an action of anticipatory neural “satiety“ signals o~ginating from receptors at the periphery. A separate control experiment measured several physiological parameters in response to ICV injection of both VEH and 100 ng AI1 under conditions essentially identical to those used for rCBF measurement. These data failed to reveal any pre- to posttreatment shifts that would suggest biological activity leading to nonspecific and global shifts in rCBF. As shown in Table 2. blood gases showed only minor shifts. For example, a difference of < 1.O mmHg was found between p,COz pre- to postshifts (reductions) in VEH (- I .O mmHg) and AI1 (- 1.9 mmHg) conditions. As expected, AI1 increased MAP. This increase, however, remained within the range of autoregulation of CBF (10) and is not likely to have led to the observed global blood flow shifts. In a related study, Rettig et al. (33) reported no change in total CBF in unrestrained, mildly water-deprived rats permitted to drink following ICV injection of 100 ng AIL In contrast, significant blood flow shifts were observed in various peripheral organs. The present findings stand in relatively sharp contrast to those recently reported by Tuor et al., also using [‘4C]IAP to measure rCBF in the rat (39). These authors found no significant shifts in any CVO or non-CVO areas following peripheral infusion of AI1 at a gradually increasing rate (-0.02-0.05 &kg-‘/min -I), including many regions also sampled in the present study. It is not clear, however, whether or not these data can be directly compared to those of the present study. For example, the authors report that brains were frozen within 2 min after death by bolus injection ofpentobarbital. This would seem to be near, or perhaps exceed, the upper limit for guarding against postmortem diffusion of tracer. The degree of diffusion within that time could obscure differences in smallerbrain structures( 13,37,46). By comparison, we were able to immerse the skull into -50°C isopentane within
144 471 7.43 33.7 78.5 22.4
+I zk8 + 0.01 to.9 f 1.9 i: 0.3
+7 + 18 f 0.01 i I.2 i 1.7 + 0.7
.I
-
11s -I +o.o I - I.9
If. 3* :?I6
rt 0.02 IL0.9
t4.i
ir1.9
-0.9
i- 0.X
shift
30 s (usually ~20 s) of a sharply defined time point. namely, decapitation. Tuor et al. did not measure drinking or attempt to drink; however, an elevation in mean arterial blood pressure of 24% was reported following AI1 infusion, indicating an All response sufficient to have resulted in drinking if tested. Nonetheless, it remains surprising that no hint of change in SF0 blood flow was observed if, as has been suggested, the SF0 is the principal site of dipsogenic action of blood-~rne Aff. As noted, the widespread shifts in rCBF following All administration [AI1 (W-) condition] correspond to many of the sites having been implicated in aspects of body fluid homeostasis. In view of the aforementioned discussion, we suggest that these effects might likely reflect neuronal activity originating from one or more of the CVOs, along with other elements of the AV3V region, the latter including OVLT and MnPo area. As it is known that alterations in regional metabolic activity reflect principally the activity of synaptic inputs to that region (22), it is therefore conceivable that increases in blood flow seen in a number of regions reflect synaptic activity in projection fields. Indeed, a map of the regions significantly affected coincides to substantial degree with the dist~bution of efferent projections of CVO and AV3V areas ( 17,23,30). It has been suggested that the MnPo nucleus of AV3V might play a key role in integrating and relaying the effects of AI1 from both brain and blood sources ( 14,29). The MnPo is reciprocally connected with both the SF0 and the OVLT and sends efferents to a number of other regions including, in part, supraoptic nucleus, lateral hypothaiamus, and paravent~cular hypothalamus (2,44). With the exception ofthe SFO, all these regions exhibited significant increases in blood flow following AI1 treatment in the present study. Furthermore, elevated blood flow was reversed in the supraoptic nucleus and rostra1 lateral hypothalamic region by ingestion of a small amount ofwater. Projections from MnPo not significantly affected in the present study included bed nucleus of the stria terminalis and arcuate nuclei. Lesions of the MnPo/OVLT region have been shown to severely impair AIIinduced drinking (I ,3,14,20), and thirst-inducing extracellular fluid depletion with polyethylene glycol enhances norepinephrine turnover in the MnPo region (45). Furthermore, MnPo lesions have been shown to disrupt circadian influences on drinking (6). It is interesting to note earlier behavioral studies linking medial amygdala (7,47) and zona incerta (8,35,41) with drinking and salt preference in the rat. Electrophysiological data have implicated the rostra1 zona incerta in neural integration of inputs from osmotic and extracellular thirst (25). Corticomedial amygdala and zona incerta receive projections from SF0 ( 17) and exhibited significant moderate to high increases
534
in blood flow with AI1 treatment; in the rostra1 zona incerta. the AI1 effect was reversed in the group that drank. Data from the rat also indicate that efferents from OVLT terminate in the cingulate gyrus and hippocampal dentate gyrus (30); blood flow in the cingulate cortex was significantly elevated following AI1 treatment. The metabolic consequences of excitatory and inhibitory postsynaptic activity are indistinguishable using blood flow procedures alone; consequently, questions of synaptic polarity in
affected regions and of functional significance cannot be addressed with current protocols. The present study, however, appeared to develop a potentially useful map for further study of the dipsogenic properties of angiotensin II. ACKNOWl.ElXEMEN'IS
This research was supported in part by grants from the Graduate School of Marquette University (D.A.C.) and USPHS-DA 05012 (E.A.S.). Technical supportby Scott Fuller is gratefully acknowledged.
REFERENCES 1. Bellin, S. 1.; Landas, S. K.; Johnson, A. K. Localized injections of
2.
3. 4.
5.
6. 7.
8. 9. 10. il. 12.
13. 14.
15. 16.
17. 18.
19.
6-hydroxydopamine into lamina terminalis-associated structures: Effects on experimentally induced drinking and pressor responses. Brain Res. 416:75-83; 1987. Brown, E. R.; Standaert, D. G.; Saper, C. B. Efferent connections of the medial preoptic region in the rat. Sot. Neurosci. Abstr. 1I: 827; 1985. Buggy, J.; Johnson, A. K. Angiotensin-induced thirst: Effects of third ventricle obstruction and periventricular ablation. Brain Res. 149: 117-128; 1978. Fenstermacher, J. D.; Blasberg, R. G.; Path& C. S. Methods for quantifying the transport of drugs across brain barrier systems. Pharmacol. Ther. 142 17-248; 198 1. Fhiharty, S. J.; Epstein, A. N. Sodium appetite elicited by intracerebrovent~cular infusion of an~otensin II in the tat: II. Synetgistic interaction with systemic rnine~~~o~icoi~. Behav. Neurosci. 97: 746-758; 1983. Gardiner, T. W.; Snicker, E. M. Impaired drinking responses of rats with lesions of nucleus medianus: Circadian dependence. Am. J. Physiol. 248:R224-R230; 1985. Gentil, C. G.; Antunes-Rodrigues, J.; Negro-Vilar, A.; Covian, M. R. Role of amygdaloid complex in sodium chloride and water intake in the rat. Physiol. Behav. 3:981-985: 1968. Grossman, S. P. A massessment ofthe brain mechanisms that control thirst. Neurosci. Biobehav. Rev. 8:95-104; 1984. Hartle, D. K.; Brody, M. J. The angiotensin II pressor system of the rat forebrain. Circ. Res. 54:355-366: 1984. Herpin, D. The effects of antihypertensive drugs on the cerebral blood flow and its regulation. Prog. Neurobiol. 35:7.5-83; 1990. Hsiao. S.: Eostein. A. N.: Camardo. J. S. The dipsoaenic notencv of peripheral angiotensin II. Horm. Behav. 8: 129-i&; 1977. . Israel, A.; Plunkett, L. M.; Saavedra, J. M. Quantitative autoradiographic characterization of receptors for angiotensin II and other neuropeptides in individual brain nuclei and peripheral tissues from single-rats. Cell. Mol. Neurobiol. 5:21 l-222;. 1985. Jav. ,. T. M.: Lucinnani. _ ,,G.: Crane. A. M.: Jehle. J.: Sokolaff. L. Measurement of local cerebral blood flow with [‘“Cl Iodoantipyrine in the mouse. J. Cereb. Blood Flow Metab. 8: 12 I-l 29; 1988. Johnson, A. K. The periventricular anteroventral third ventricle (AV3V): Its relationship with the subfomical organ and neural systems involved in maintaining body fluid homeostasis. Brain Res. Bull. 15:595-601; 1985. Landas, S.; Phi&s, M. I.; Stamler, J. F.; Raizada, M. K. Visualization of specific angietensin II binding sites in the brain by fluorescent microscopy. Science 210:791-793; 1980. Lang, R. E.; Unger, T.; Rascher, W.; Ganten, D. Brain angiotensin. In: Iversen. L. L.: Iversen. S. D.: Snvder. S. H.. eds. Handbook of psychoph&macoiogy, vol: 16: Neur&eptides. New York: Plenum Press; 1983:307-361. Lind, R. W. The subfomical organ: Neural connections. In: Gross, P. M., ed. Circumventricular organs and body fluids, vol. I. Boca Raton, FL: CRC Press; 1987127-42. Lind, R. W.; Swanson, L. W.; Ganten, D. Organimtion of angiotensin II immunoreactive cells and fibers in the rat central nervous system: An immunohistochemical study. Neuroendoctinology 40:2-24; 1985. Mangiapane, M. L.; Simpson, J. B. Subfomical organ: Forebrain site of pressor and dipsogenic action of angiotensin II. Am. J. Physiol. 239:R382-R389: 1980.
20. Man~a~ne, M. L.; Thrasher, T. N.; Keil, L. C.; Simpson, J. B.: Ganong, W. F. Deficits in drinking and vasopressin secretion after lesions of the nucleus medianus. Neuroendocrinology 37:73-77; 1983. 21. Mangiapane, M. L.; Thrasher, T. N.; Keil, L. C.; Simpson, J. B.: Ganong, W. F. Role for the subfomical organ in vasopressin secretion. Brain Res. Bull. 13:43-47; 1984. 22. Mata, M.; Fink, D. J.; Gainer, H.; Smith, C. B.; Davidsen, L.; Savaki, H.; Schwartz, W. J.; Sokoloff, L. Activitydependent energy metabolism in rat posterior pituitary primarily reflects sodium pump activity. J. Neurochem. 342 13-2 15; 1980. 23. McKinley, M. J.; MeAllen, R. M.; Mendelsohn, F. A. 0.; Allen, A. M.; Chai, S. Y.; Oldfield, B. J. Circumventricular organs; neuroendocrine interfaces between the brain and the hemal milieu. Front. Neur~nd~nol. 11:91-127; 1990. 24, Mendel~hn, F. A. 0.; Quirion, R.; Saavedra, J. M.; Aguilera, G.; Catt, K. J. Autoradiomdio~phic localization of angiotensin II receptors in rat brain. Proc. Nat]. Acad. Sci. USA 81:15X-1579; 1984. 25. Mok, D.; Mogenson, G. J. Convergence of signals in the zona incerta for angiotensin-mediated and osmotic thirst. Brain Res. 407:332340; 1987. 26. Palkovits, M. Summary of structural and functional aspects of the circumventricular organs. In: Gross, P. M., ed. Circumventricular organs and body fluids, vol. II. Boca Raton, FL: CRC Press; 1987: 209-218. 27. Peters, A.; Palay, S. L.; Webster, H. deF. The fine structure of the nervous system: Neurons and their supporting cells. 3rd ed. Oxford: Oxford University Press; I99 1. 28. Phillips, M. I. Functions of an~otensin in the central nervous system. Annu. Rev. Phvsiol. 49413-435: 1987. 29. Phihips, M. I. Brain an~otensin.‘In: Gross, P. M., ed. Circumventricular organs and body fluids, vol. III. Boca Raton. FL: CRC Press; 1987:163-182. 30. Phillips, M. 1.; Camacho, A. Neural connections of the organum vasculosum of the lamina terminalis. in: Gross, P. M., ed. Circumventricular organs and body fluids, vol. I. Boca Raton, FL: CRC Press 1987:157-170. 31. Ramsay, D. J.; Thrasher, T. N. Thirst and water balance. In: Stricker, E. M., ed. Handbook of behavioral neurobiology, vol. 10: Neurobiology of food and fluid intake. New York: Plenum Press; 1990: 353-386. 32. Rettig, R.; Ganten, D.; Lang, R. E.; Unger, Th. The renin-angiotensin system in the central cot&o1 of blood pressure. Eur. Heart J. 8(suppl. B):129-132; 1987. 33. Rettig, R.; White, F. C.; Printz, M. P. Effects of~n~ng and central an~otensin II stjmulation on organ blood flow in conscious rats. Brain Res. 50.5:25l-256; f989. 34. Rowland, N. E.; Fregly, M. J. Sodium appetite: Species and strain differences and role of renin-angiotensin-aldosterone system. Appetite i 1:143-178; 1988. 35. Rowland, N.; Grossman, S. P.; Grossman, L. Zona incerta lesions: Regulatory drinking deficits to intravenous NaCI, angiotensin, but not to salt in the food. Physiol. Behav. 23:745-750; 1979. 36. Saavedra, J. M.; Israel, A.; Plunkett, L. M.; Kurihara, M.; Shigematsu, K.: Correa, F. M. A. Quantitative distribution of anniotensin 11 binding sites in rat brain by autoradiography. Peptides?:679-687; 1986. 37. Sakurada, 0.; Kennedy, C.; Jehle, J.: Brown, J. D.; Carbin. G. L.; Sokoloff, L. Measurement of local cerebral blood
ANGIOTENSIN
38.
39.
40.
4 I.
42.
I1 DRINKING
AND CEREBRAL
BLOOD FLOW
flow with iodo[r4C]antipyrine. Am. J. Physiol. 234:H59-H66: 1978. Steele, M. K.: Negro-Vilar, A.; McCann, S. M. Effect ofangiotensin II on in viva and in vitro release of anterior pituitary hormones in the female rat. Endocrinology 109:893-899; 198 1. Tuor. U. I.; Kondysar, M. H.: Harding, R. K. Effect of angiotensin II and peptide YY on cerebral and circumventricular blood flow. Peptides 9:141-149; 1988. Van Houten. M.: Mangiapane, M. L.; Reid, I. A.; Ganong, W. F. [Sar’.Ala”] angiotensin II in cerebrospinal fluid blocks the binding of blood-borne [“5]angiotensin II to the circumventricular organs. Neuroscience 10:1421-1426: 1983. Walsh L. L..;Grossman, S. P. Electrolytic lesions and knife cuts in the region of the zona incerta impair sodium appetite. Physiol. Behav. l&587-596: 1977. Wamsley. J. K.: Herblin. W. F.: Alburges. M. E.; Hunt, M. Evidence for the presence of angiotensin B-type I receptors in brain. Brain Res. Bull. 25:397-400: 1990.
535
43. Weiss, M. L.; Moe, K. E.; Epstein. A. N. Interference with central actions ofangiotensin II suppresses sodium appetite. Am. J. Physiol. 250:R250-R259: 1986. 44. Wilkin. L. D.: Mitchell, L. D.; Ganten, D.; Johnson, A. K. The supraoptic nucleus: Afferents from areas involved in control of body fluid homeostasis. Neuroscience 28:573-584: 1989. 45. Wilkin, L. D.; Patel, K. P.; Schmid, P. G.; Johnson. A. K. Increased norepinephrine turnover in median preoptic nucleus following reduced extracellular fluid volume. Brain Res. 423:369372; 1987. 46. Williams J. L.; Shea, M.: Furlan, A. J.: Little. J. R.; Jones, S. C. Importance of freezing time when iodoantipyrine is used for measurement of cerebral blood flow. Am. J. Physiol. 261:H252-H256: 1991. 47. Zolovick. A. J.; Avrith, D.; Jalowiec, J. E. Reversible colchicineinduced disruption of amygdaloid function in sodium appetite. Brain Res. Bull. 5:35-39: 1980.
