- Email: [email protected]
Attenuated drinking response induced by angiotensinergic activation of subfornical organ projections to the paraventricular nucleus in estrogen-treated rats
Neuroscience Letters 324 (2002) 242–246 www.elsevier.com/locate/neulet
Attenuated drinking response induced by angiotensinergic activation of subfornical organ projections to the paraventricular nucleus in estrogen-treated rats Junichi Tanaka a,*, Katsuhide Kariya b,c, Hiroko Miyakubo a,d, Kazuhiro Sakamaki c,d, Masahiko Nomura c a
Department of Curriculum, Teaching and Memory, Neuroscience Program, Naruto University of Education, Naruto, Tokushima 772-8502, Japan b Clinical Research Department, Pharmaceutical Division, Japan Tobacco, Inc., Minato-ku, Tokyo 105-8422, Japan c Department of Physiology, Saitama Medical School, Iruma-gun, Saitama 350-0495, Japan d Department of Human Development, Naruto University of Education, Naruto, Tokushima 772-8502, Japan Received 15 January 2002; received in revised form 17 February 2002; accepted 21 February 2002
Abstract The present study was carried out to examine whether estrogen modulates the drinking response caused by activation of neural pathways from the subfornical organ (SFO) to the hypothalamic paraventricular nucleus (PVN) in the female rat. Microinjection of angiotensin II (ANG II) into the SFO elicited drinking in ovariectomized female rats that were treated with either propylene glycol (PG) vehicle or estradiol benzoate (EB). The amount of water intake induced by the ANG II injection was significantly greater in the PG-treated than in the EB-treated animals. In both groups, previous injections of either saralasin, an ANG II antagonist, or phentolamine, an a-adrenoceptor antagonist, bilaterally into the PVN resulted in the significant attenuation of the drinking response to ANG II, whereas similar injections of saline vehicle into the PVN were without effect. These results suggest that the circulating estrogen may act to reduce the drinking response that is mediated through angiotensinergic and a-adrenergic mechanisms in the PVN in response to angiotensinergic activation of SFO efferent projections. q 2002 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Drinking; Subfornical organ; Paraventricular nucleus; Angiotensin II; Saralasin; Phentolamine
There is experimental evidence that neural pathways from the subfornical organ (SFO), a target site for circulating angiotensin II (ANG II) [10], to the hypothalamic paraventricular nucleus (PVN) participate in the regulation of body fluid balance and cardiovascular function. Destruction of the PVN [3] or inactivation of either the angiotensinergic or noradrenergic system in the PVN [15] has been shown to reduce the drinking and pressor responses to electrical or chemical (ANG II) stimulation of the SFO. Activation of SFO efferents has been shown to enhance noradrenaline (NA) release in the PVN [15,16]. It has been demonstrated that SFO neurons contain both ANG II AT1 [6,12] and estrogen [6,12,13] receptors, and that estrogen decreases ANG II binding to AT1 receptors [6] and the number of SFO neurons that express AT1 receptors * Corresponding author. Tel.: 181-88-687-6277; fax: 181-88687-6277. E-mail address: [email protected] (J. Tanaka).
[12]. Previous studies have shown that central interactions between estrogen and angiotensinergic systems are important for modulating ingestive behavior [2,5,7]. The purpose of this study was to clarify whether estrogen modulates the drinking response induced by angiotensinergic activation of the SFO projections to the PVN. Since it has been demonstrated that the drinking response is partially mediated through angiotensinergic and a-adrenoceptor mechanisms in the PVN [14], we compared the amount of water intake caused by ANG II injected into the SFO before and after the treatment with saralasin, an ANG II antagonist, or phentolamine, an a-adrenoceptor antagonist, in the PVN between ovariectomized (OVX) female rats that were treated with propylene glycol (PG) and OVX female rats that were treated with estradiol benzoate (EB). The present study was carried out according to the guiding principles of the Physiological Society of Japan. Female Wistar rats (n ¼ 43) weighing 220–280 g were used for the experiments. They were housed individually in hanging
0304-3940/02/$ - see front matter q 2002 Elsevier Science Ireland Ltd. All rights reserved. PII: S0 30 4- 39 40 ( 02) 0 02 03- 3
J. Tanaka et al. / Neuroscience Letters 324 (2002) 242–246
wire cages for at least 2 weeks before testing. Wayne lab chow and tap water were available ad libitum except where noted. Lights were on in the animal rooms for 12 h per day, and temperature was maintained at 23–25 8C. The animals were anaesthetized with sodium pentobarbital (60 mg/kg, i.p.), and bilaterally OVX. The rats were then placed in a stereotaxic frame. Twenty-six-gauge stainless steel cannulae were stereotaxically lowered into the SFO and PVN (Fig. 1). The intracranial cannulae tips were embedded in dental acrylic anchored by small jeweller’s screws fixed in the skull. The 26-gauge cannula served as a guide for a 33-gauge stainless steel injector cannula, which was inserted just before injections. The injector tip was level with the end of the guide cannula when inserted. Each guide cannula was filled with an obturator of the same gauge as the injector cannula when the animals were not being tested. The animals were given 1 week to regain body weight to pre-surgical levels and reestablish a normal pattern of 24 h food and water intake before treatment was begun. The OVX animals were divided into two groups and received either a subcutaneous vehicle injection of PG (100 ml; Sigma, St. Louis, MO; n ¼ 22) or a subcutaneous injection of EB (10 mg dissolved in PG; Sigma, St. Louis, MO; n ¼ 21) on 5 days. ANG II (Asp 1-Ile 5-ANG II) salt, saralasin (Sar 1-Val 5Ala 8-ANG II) salt and phentolamine mesylate were purchased from the Sigma (St. Louis, MO), Peptide Institute (Osaka, Japan) and the Nippon Chiba Geigy (Tokyo, Japan), respectively. These substances were dissolved in isotonic saline. The second day after the start of PG or EB treatment, each animal was put into a metabolism cage and baseline drinking behavior was observed. All testing was done at least within 3 h after the start of the light part of each rat’s light/dark cycle. On the next day, each rat was removed from its home cage, and the obturator was removed. The injectors, filled with injectate and connected to two 5-ml Hamilton gas chromatography syringes via approximately 1 m of polyethylene tubing, were inserted into the implanted guide cannulae. The injectate within the tip of the injector was separated from the tip of the cannula, and therefore from the rat, by a 0.02-ml air bubble. The rat was then placed in the metabolism cage. Immediately following placement of the rat in the cage, the ANG II solution was injected into the SFO. The injection was administered in a dose of 10 210 M. The latency to the onset of drinking was recorded, and water intake was then monitored for 20 min following the injection. On the next day, a portion of the animals (four PG-treated and four EB-treated rats) were examined for a response to saline vehicle injections into the SFO. Two days after the ANG II injection, the effect of pretreatment with the antagonist or saline in the PVN on the ANG II-induced drinking response was tested. Previous observations have shown that injections of ANG II in a dose of 10 210 M elicit a robust drinking response [15] and that the response is attenuated by the
243
Fig. 1. The location of the cannula tips in OVX female rats that were treated with either PG (B,E) or EB (C,F). (A,D) Photographs from Neutral Red-stained coronal sections illustrate the location of the cannula (arrows) in the SFO (A) and the hypothalamic PVN (D) of the EB-treated rat. (B,C) Circles on schematic transverse sections (8.0 and 7.6 mm anterior to the interaural line) depict the loci of the center of cannula tips used for injections of ANG II. Triangles indicate the loci of cannulae utilized for the ANG II and saline vehicle injections. (E,F) Circles, triangles and squares on two representative transverse sections (7.2 and 6.9 mm anterior to the interaural line) indicate the loci of the center of cannula tips used for injections of saralasin, phentolamine and saline, respectively. AH, anterior hypothalamic area; LH, lateral hypothalamic area; PVN, hypothalamic paraventricular nucleus; SFO, subfornical organ; SM, stria medullaris thalamus; TS, triangular septal nucleus; 3V, third ventricle. Scale bar, 1 mm.
244
J. Tanaka et al. / Neuroscience Letters 324 (2002) 242–246
saralasin or phentolamine treatment with a dose of 10 210 M in the PVN [15]. In this study, injections of ANG II and the antagonists were thus administered in a dose of 10 210 M. The antagonist or saline treatment was achieved 30 s before the ANG II injection. As it is crucial to minimize diffusion of injectate in neuroanatomic localization experiments, all injections of the drug solutions or vehicle were given in a volume of 0.2 ml. Injections were achieved at a rate of 0.02 ml/s. At the completion of the experiment, injections of isotonic saline (0.2 ml) containing 2% Pontamine sky blue dye were made in order to confirm more precisely the location of cannula tips and to measure the spread of the injected solution. Each animal was sacrificed with an overdose of sodium pentobarbital and perfused through the heart with isotonic saline to clear blood, which was followed by 3.7% paraformaldehyde in 0.9% NaCl solution for fixation. The brain was then removed and stored in the formalin saline for 24 h. The location of cannula tips was confirmed histologically in 50 mm sections stained with Neutral red. The stereotaxic coordinates for the locations were determined according to the atlas of Paxinos and Watson [11]. All values reported as means ^ SEM. Drinking response data were analyzed with a one-way and two-way repeated measures analysis of variance and subsequent Tukey’s protected t-test. P , 0:05 was required for significance. Histological analysis of the brains of the rats showed that the cannula tips of four out of 22 PG-treated and three out of 21 EB-treated rats were located in the sites 0.3–0.6 mm away from the main body of the SFO and/or PVN (data not shown). The sites approximately 0.2 mm away from the center of the cannula tip were stained with Pontamine sky blue dye. The data from these seven animals were not included in the further analysis. The animals (18 PG-treated and 18 EB-treated rats) having the cannula tip either within or just above the SFO (Fig. 1A–C) showed a robust drinking in response to the ANG II injection (Fig. 2). The amount of water intake was significantly greater in the PG-treated (water volume, 5.4 ^ 0.4 ml; ranging 2.6–8.4 ml) than in the EB-treated (water volume, 3.7 ^ 0.4 ml; ranging 1.6–6.1 ml) animals (Fð1;34Þ ¼ 30:321, P , 0:001; Fig. 2A). There was no significant difference in the latency of drinking response between the PG-treated (30 ^ 8 s; ranging 4–97 s) and EB-treated (33 ^ 7 s; ranging 5–103 s) animals. No apparent topographical differences were observed between the PG-treated and EB-treated animals in the positions of cannula tips (Fig. 1B,C). Microinjections of saline vehicle into the SFO did not cause water intake in both PG- and EB-treated groups (four animals for each animal group; water volume, less than 0.3 ml; Fig. 2A). These results suggest that estrogen may reduce the drinking response elicited by angiotensinergic activation of the SFO. The effects of previous injections of saralasin, phentolamine or saline into the PVN (Fig. 1D–F) on the drinking response were investigated. In both groups, the water intake
in response to ANG II injected into the SFO was significantly attenuated by previous injections of either saralasin (Fð1;12Þ ¼ 18:325, P , 0:01 for the PG-treated animals, n ¼ 7; Fð1;12Þ ¼ 21:650, P , 0:01 for the EB-treated animals, n ¼ 7; Fig. 2B) or phentolamine (Fð1;12Þ ¼ 12:993, P , 0:05 for the PG-treated animals, n ¼ 7; Fð1;12Þ ¼ 11:285, P , 0:05 for the EB-treated animals, n ¼ 7; Fig. 2C). No complete blockage of the
Fig. 2. Total water intake (ml/20 min) in response to microinjection of ANG II (18 PG-treated and 18 EB-treated animals) or saline vehicle (Saline, four PG-treated and four EB-treated animals) into the SFO (A) and the effects of pretreatment with saralasin (Sar; seven PG-treated and seven EB-treated animals; B), phentolamine (Phe; seven PG-treated and seven EB-treated animals; C) or saline vehicle (Saline; four PG-treated and four EB-treated animals; D) in the PVN on the ANG II-induced water intake in the PG-treated (PG) and EB-treated (EB) animals. Results are expressed as means ^ SEM. Total water intake after ANG II injection into the SFO was significantly lower in the EB-treated than in PG-treated animals (A–D). †P , 0:05, ††P , 0:01, †††P , 0:001. In both groups, water intake to ANG II injected directly into the SFO was significantly diminished by pretreatment with Sar (B) or Phe (C), but not by Saline (D), in the PVN. *P , 0:05, **P , 0:01 compared with those of ANG II. ††P , 0:01.
J. Tanaka et al. / Neuroscience Letters 324 (2002) 242–246
drinking response by the pretreatment was found in any animals. The amount of water intake after the treatment with either saralasin (Fð1;12Þ ¼ 18:962, P , 0:01) or phentolamine (Fð1;12Þ ¼ 17:001, P , 0:01) in the PVN was significantly greater in the PG-treated than in the EB-treated animals. There was no significant difference between the PG-treated (the saralasin treatment, 55 ^ 7% decrease compared with the pretreatment; the phentolamine treatment, 32 ^ 4% decrease compared with the pretreatment) and EB-treated (the saralasin treatment, 53 ^ 6% decrease compared with the pretreatment; the phentolamine treatment, 34 ^ 4% decrease compared with the pretreatment) animals in the amount of the attenuation of total water intake by the drug treatment. Previous injections of saline vehicle into the PVN, on the other hand, had no significant influence on the drinking response in both groups (four PGtreated and four EB-treated animals; Fig. 2D). Pretreatment with either saralasin or phentolamine in the PVN did not cause significant differences in the latencies to drinking in both groups (the PG-treated animals, 32 ^ 10 s in the pretreatment vs. 34 ^ 11 s the saralasin treatment, 30 ^ 10 s in the pretreatment vs. 35 ^ 8 s in the phentolamine treatment; the EB-treated animals, 28 ^ 10 s in the pretreatment vs. 35 ^ 11 s the saralasin treatment, 34 ^ 10 s in the pretreatment vs. 35 ^ 9 s in the phentolamine). These results show that an increase in the circulating level of estrogen may attenuate the drinking behavior induced by angiotensinergic activation of the pathways from the SFO to the PVN. On the other hand, since the antagonist injection into the PVN is not able to abolish completely the ANG II-induced drinking response and to produce any significant changes in the response latency, it is possible to speculate that other neural circuits that are activated by ANG II may be involved in the drinking response. Immunohistochemical observations have revealed that ANG-immunoreactive neurons in the SFO project directly to the PVN [9]. Electrophysiological data have demonstrated that SFO neurons projecting to the PVN are activated by ANG II and their responsiveness to ANG II is suppressed by estrogen [14]. Recent studies have shown that an increase in the circulating level of estrogen decreases ANG II binding to AT1 receptors in the SFO [6] and the expression of both the AT1 and estrogen receptors [12]. Previous [15] and our findings in which saralasin injected into the PVN reduced the drinking response induced by ANG II injections into the SFO indicate that the angiotensinergic system in the PVN is implicated in mediating the response. Although there is no direct evidence showing the co-existence of ANG II AT1 and estrogen receptors in SFO neurons with angiotensinergic projections to the PVN, it might be expected that estrogen may act to attenuate the drinking response by decreasing the sensitivity of direct angiotensinergic SFO projections to ANG II. It has been postulated that the NA system in the PVN is implicated in the regulation of drinking behavior. Adminis-
245
tration of noradrenergic agonists into the PVN has been shown to increase drinking behavior [8]. The NA blockade of the hypothalamic sites has been shown to reduce the ANG II-induced drinking [4]. Microdialysis investigations have indicated that electrical [16] or chemical (ANG II) [15] stimulation of the SFO facilitates NA release in the PVN. Although the precise mechanisms underlying an activation of the noradrenergic system in the PVN in response to ANG II acting at the SFO cannot be determined from our data, these findings and our data suggest that estrogen may serve to attenuate the ANG II-induced drinking response that is mediated through an a-adrenoceptor mechanism in the PVN. It has been known that estrogen diminishes the drinking induced by systemic or intraventricular injections of ANG II. Intracerebroventricular administration of estrogen has been shown to reduce the increased drinking elicited by ANG II [5]. In addition, the ANG II-induced drinking has been shown to be attenuated by estrogen treatment [2,7]. The present data strongly suggest that the neural circuits from the SFO to the PVN may be involved in the attenuation in the drinking behavior observed in the previous studies and in female rats during estrous cycle [1].
[1] Findlay, A.L.R., Fitzsimons, J.T. and Kucharczyk, J., Dependence of spontaneous and angiotensin-induced drinking in the rat upon the oestrous cycle and ovarian hormones, J. Endocrinol., 82 (1979) 215–225. [2] Fregly, M.J. and Thrasher, T.N., Attenuation of angiotensininduced water intake in estrogen-treated rats, Pharmacol. Biochem. Behav., 9 (1978) 509–514. [3] Gutmann, M.B., Jones, D.L. and Ciriello, J., Effect of paraventricular nucleus lesions on drinking and pressor responses to ANG II, Am. J. Physiol., 255 (1988) R882–R887. [4] Jones, D.L., Injections of phentolamine into the anterior hypothalamic–preoptic area of rats blocks both pressor and drinking responses produced by central administration of angiotensin II, Brain Res. Bull., 13 (1984) 127–133. [5] Jonklass, J. and Buggy, J., Angiotensin–estrogen interaction in female brain reduces drinking and pressor responses, Am. J. Physiol., 247 (1984) R167–R172. [6] Kisley, L.R., Sakai, R.R. and Fluharty, S.J., Estrogen decreases hypothalamic angiotensin II AT1 receptor binding and mRNA in the female rat, Brain Res., 44 (1999) 34–42. [7] Kisley, L.R., Sakai, R.R. and Fluharty, S.J., Ovarian steroid regulation of angiotensin II induced water intake in the rat, Am. J. Physiol., 276 (1999) R90–R96. [8] Leibowitz, S.F., Paraventricular nucleus: a primary site mediating adrenergic stimulation of feeding and drinking, Pharmacol. Biochem. Behav., 8 (1978) 163–175. [9] Lind, R.W., Swanson, L.W. and Ganten, D., Angiotensin II immunoreactivity in neural afferents and efferents of the subfornical organ of the rat, Brain Res., 321 (1984) 209–215. [10] McKinley, M.J., McAllen, R.M., Mendelsohn, F.A.O., Allen, A.M., Chai, S.Y. and Oldfield, B.J., Circumventricular organ: neuroendocrine interfaces between the brain and the hemal milieu, Front. Neuroendocrinol., 11 (1990) 91–127. [11] Paxinos, G. and Watson, C., The Rat Brain in Stereotaxic Coordinates, Academic Press, Sydney, 1986. [12] Rosas-Arellano, M.P., Solano-Flores, L.P. and Ciriello, J., Co-localization of estrogen and angiotensin receptors
246
J. Tanaka et al. / Neuroscience Letters 324 (2002) 242–246
within subfornical organ neurons, Brain Res., 837 (1999) 254–262. [13] Simerly, R.B., Chang, C., Muramatsu, M. and Swanson, L.W., Distribution of androgen and estrogen receptor mRNA-containing cell in the rat brain; an in situ hybridization study, J. Comp. Neurol., 294 (1990) 76–95. [14] Tanaka, J., Hayashi, Y., Miyakubo, H., Sakamaki, K., Okumura, T. and Nomura, S., Estrogen decreases the responsiveness of subfornical organ neurons projecting to the hypothalamic paraventricular nucleus to angiotensin II in female rats, Neurosci. Lett., 307 (2001) 155–158.
[15] Tanaka, J., Hayashi, Y., Nomura, S., Miyakubo, H., Okumura, T. and Sakamaki, K., Angiotensinergic and noradrenergic mechanisms in the hypothalamic paraventricular nucleus participate in the drinking response induced by activation of the subfornical organ in rats, Behav. Brain Res., 118 (2001) 117–122. [16] Tanaka, J., Hayashi, Y., Sakamaki, K., Okumura, T. and Nomura, M., Activation of the subfornical organ enhances extracellular noradrenaline concentrations in the hypothalamic paraventricular nucleus in the rat, Brain Res. Bull., 54 (2001) 421–425.
Attenuated drinking response induced by angiotensinergic activation of subfornical organ projections to the paraventricular nucleus in estrogen-treated rats Junichi Tanaka a,*, Katsuhide Kariya b,c, Hiroko Miyakubo a,d, Kazuhiro Sakamaki c,d, Masahiko Nomura c a
Department of Curriculum, Teaching and Memory, Neuroscience Program, Naruto University of Education, Naruto, Tokushima 772-8502, Japan b Clinical Research Department, Pharmaceutical Division, Japan Tobacco, Inc., Minato-ku, Tokyo 105-8422, Japan c Department of Physiology, Saitama Medical School, Iruma-gun, Saitama 350-0495, Japan d Department of Human Development, Naruto University of Education, Naruto, Tokushima 772-8502, Japan Received 15 January 2002; received in revised form 17 February 2002; accepted 21 February 2002
Abstract The present study was carried out to examine whether estrogen modulates the drinking response caused by activation of neural pathways from the subfornical organ (SFO) to the hypothalamic paraventricular nucleus (PVN) in the female rat. Microinjection of angiotensin II (ANG II) into the SFO elicited drinking in ovariectomized female rats that were treated with either propylene glycol (PG) vehicle or estradiol benzoate (EB). The amount of water intake induced by the ANG II injection was significantly greater in the PG-treated than in the EB-treated animals. In both groups, previous injections of either saralasin, an ANG II antagonist, or phentolamine, an a-adrenoceptor antagonist, bilaterally into the PVN resulted in the significant attenuation of the drinking response to ANG II, whereas similar injections of saline vehicle into the PVN were without effect. These results suggest that the circulating estrogen may act to reduce the drinking response that is mediated through angiotensinergic and a-adrenergic mechanisms in the PVN in response to angiotensinergic activation of SFO efferent projections. q 2002 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Drinking; Subfornical organ; Paraventricular nucleus; Angiotensin II; Saralasin; Phentolamine
There is experimental evidence that neural pathways from the subfornical organ (SFO), a target site for circulating angiotensin II (ANG II) [10], to the hypothalamic paraventricular nucleus (PVN) participate in the regulation of body fluid balance and cardiovascular function. Destruction of the PVN [3] or inactivation of either the angiotensinergic or noradrenergic system in the PVN [15] has been shown to reduce the drinking and pressor responses to electrical or chemical (ANG II) stimulation of the SFO. Activation of SFO efferents has been shown to enhance noradrenaline (NA) release in the PVN [15,16]. It has been demonstrated that SFO neurons contain both ANG II AT1 [6,12] and estrogen [6,12,13] receptors, and that estrogen decreases ANG II binding to AT1 receptors [6] and the number of SFO neurons that express AT1 receptors * Corresponding author. Tel.: 181-88-687-6277; fax: 181-88687-6277. E-mail address: [email protected] (J. Tanaka).
[12]. Previous studies have shown that central interactions between estrogen and angiotensinergic systems are important for modulating ingestive behavior [2,5,7]. The purpose of this study was to clarify whether estrogen modulates the drinking response induced by angiotensinergic activation of the SFO projections to the PVN. Since it has been demonstrated that the drinking response is partially mediated through angiotensinergic and a-adrenoceptor mechanisms in the PVN [14], we compared the amount of water intake caused by ANG II injected into the SFO before and after the treatment with saralasin, an ANG II antagonist, or phentolamine, an a-adrenoceptor antagonist, in the PVN between ovariectomized (OVX) female rats that were treated with propylene glycol (PG) and OVX female rats that were treated with estradiol benzoate (EB). The present study was carried out according to the guiding principles of the Physiological Society of Japan. Female Wistar rats (n ¼ 43) weighing 220–280 g were used for the experiments. They were housed individually in hanging
0304-3940/02/$ - see front matter q 2002 Elsevier Science Ireland Ltd. All rights reserved. PII: S0 30 4- 39 40 ( 02) 0 02 03- 3
J. Tanaka et al. / Neuroscience Letters 324 (2002) 242–246
wire cages for at least 2 weeks before testing. Wayne lab chow and tap water were available ad libitum except where noted. Lights were on in the animal rooms for 12 h per day, and temperature was maintained at 23–25 8C. The animals were anaesthetized with sodium pentobarbital (60 mg/kg, i.p.), and bilaterally OVX. The rats were then placed in a stereotaxic frame. Twenty-six-gauge stainless steel cannulae were stereotaxically lowered into the SFO and PVN (Fig. 1). The intracranial cannulae tips were embedded in dental acrylic anchored by small jeweller’s screws fixed in the skull. The 26-gauge cannula served as a guide for a 33-gauge stainless steel injector cannula, which was inserted just before injections. The injector tip was level with the end of the guide cannula when inserted. Each guide cannula was filled with an obturator of the same gauge as the injector cannula when the animals were not being tested. The animals were given 1 week to regain body weight to pre-surgical levels and reestablish a normal pattern of 24 h food and water intake before treatment was begun. The OVX animals were divided into two groups and received either a subcutaneous vehicle injection of PG (100 ml; Sigma, St. Louis, MO; n ¼ 22) or a subcutaneous injection of EB (10 mg dissolved in PG; Sigma, St. Louis, MO; n ¼ 21) on 5 days. ANG II (Asp 1-Ile 5-ANG II) salt, saralasin (Sar 1-Val 5Ala 8-ANG II) salt and phentolamine mesylate were purchased from the Sigma (St. Louis, MO), Peptide Institute (Osaka, Japan) and the Nippon Chiba Geigy (Tokyo, Japan), respectively. These substances were dissolved in isotonic saline. The second day after the start of PG or EB treatment, each animal was put into a metabolism cage and baseline drinking behavior was observed. All testing was done at least within 3 h after the start of the light part of each rat’s light/dark cycle. On the next day, each rat was removed from its home cage, and the obturator was removed. The injectors, filled with injectate and connected to two 5-ml Hamilton gas chromatography syringes via approximately 1 m of polyethylene tubing, were inserted into the implanted guide cannulae. The injectate within the tip of the injector was separated from the tip of the cannula, and therefore from the rat, by a 0.02-ml air bubble. The rat was then placed in the metabolism cage. Immediately following placement of the rat in the cage, the ANG II solution was injected into the SFO. The injection was administered in a dose of 10 210 M. The latency to the onset of drinking was recorded, and water intake was then monitored for 20 min following the injection. On the next day, a portion of the animals (four PG-treated and four EB-treated rats) were examined for a response to saline vehicle injections into the SFO. Two days after the ANG II injection, the effect of pretreatment with the antagonist or saline in the PVN on the ANG II-induced drinking response was tested. Previous observations have shown that injections of ANG II in a dose of 10 210 M elicit a robust drinking response [15] and that the response is attenuated by the
243
Fig. 1. The location of the cannula tips in OVX female rats that were treated with either PG (B,E) or EB (C,F). (A,D) Photographs from Neutral Red-stained coronal sections illustrate the location of the cannula (arrows) in the SFO (A) and the hypothalamic PVN (D) of the EB-treated rat. (B,C) Circles on schematic transverse sections (8.0 and 7.6 mm anterior to the interaural line) depict the loci of the center of cannula tips used for injections of ANG II. Triangles indicate the loci of cannulae utilized for the ANG II and saline vehicle injections. (E,F) Circles, triangles and squares on two representative transverse sections (7.2 and 6.9 mm anterior to the interaural line) indicate the loci of the center of cannula tips used for injections of saralasin, phentolamine and saline, respectively. AH, anterior hypothalamic area; LH, lateral hypothalamic area; PVN, hypothalamic paraventricular nucleus; SFO, subfornical organ; SM, stria medullaris thalamus; TS, triangular septal nucleus; 3V, third ventricle. Scale bar, 1 mm.
244
J. Tanaka et al. / Neuroscience Letters 324 (2002) 242–246
saralasin or phentolamine treatment with a dose of 10 210 M in the PVN [15]. In this study, injections of ANG II and the antagonists were thus administered in a dose of 10 210 M. The antagonist or saline treatment was achieved 30 s before the ANG II injection. As it is crucial to minimize diffusion of injectate in neuroanatomic localization experiments, all injections of the drug solutions or vehicle were given in a volume of 0.2 ml. Injections were achieved at a rate of 0.02 ml/s. At the completion of the experiment, injections of isotonic saline (0.2 ml) containing 2% Pontamine sky blue dye were made in order to confirm more precisely the location of cannula tips and to measure the spread of the injected solution. Each animal was sacrificed with an overdose of sodium pentobarbital and perfused through the heart with isotonic saline to clear blood, which was followed by 3.7% paraformaldehyde in 0.9% NaCl solution for fixation. The brain was then removed and stored in the formalin saline for 24 h. The location of cannula tips was confirmed histologically in 50 mm sections stained with Neutral red. The stereotaxic coordinates for the locations were determined according to the atlas of Paxinos and Watson [11]. All values reported as means ^ SEM. Drinking response data were analyzed with a one-way and two-way repeated measures analysis of variance and subsequent Tukey’s protected t-test. P , 0:05 was required for significance. Histological analysis of the brains of the rats showed that the cannula tips of four out of 22 PG-treated and three out of 21 EB-treated rats were located in the sites 0.3–0.6 mm away from the main body of the SFO and/or PVN (data not shown). The sites approximately 0.2 mm away from the center of the cannula tip were stained with Pontamine sky blue dye. The data from these seven animals were not included in the further analysis. The animals (18 PG-treated and 18 EB-treated rats) having the cannula tip either within or just above the SFO (Fig. 1A–C) showed a robust drinking in response to the ANG II injection (Fig. 2). The amount of water intake was significantly greater in the PG-treated (water volume, 5.4 ^ 0.4 ml; ranging 2.6–8.4 ml) than in the EB-treated (water volume, 3.7 ^ 0.4 ml; ranging 1.6–6.1 ml) animals (Fð1;34Þ ¼ 30:321, P , 0:001; Fig. 2A). There was no significant difference in the latency of drinking response between the PG-treated (30 ^ 8 s; ranging 4–97 s) and EB-treated (33 ^ 7 s; ranging 5–103 s) animals. No apparent topographical differences were observed between the PG-treated and EB-treated animals in the positions of cannula tips (Fig. 1B,C). Microinjections of saline vehicle into the SFO did not cause water intake in both PG- and EB-treated groups (four animals for each animal group; water volume, less than 0.3 ml; Fig. 2A). These results suggest that estrogen may reduce the drinking response elicited by angiotensinergic activation of the SFO. The effects of previous injections of saralasin, phentolamine or saline into the PVN (Fig. 1D–F) on the drinking response were investigated. In both groups, the water intake
in response to ANG II injected into the SFO was significantly attenuated by previous injections of either saralasin (Fð1;12Þ ¼ 18:325, P , 0:01 for the PG-treated animals, n ¼ 7; Fð1;12Þ ¼ 21:650, P , 0:01 for the EB-treated animals, n ¼ 7; Fig. 2B) or phentolamine (Fð1;12Þ ¼ 12:993, P , 0:05 for the PG-treated animals, n ¼ 7; Fð1;12Þ ¼ 11:285, P , 0:05 for the EB-treated animals, n ¼ 7; Fig. 2C). No complete blockage of the
Fig. 2. Total water intake (ml/20 min) in response to microinjection of ANG II (18 PG-treated and 18 EB-treated animals) or saline vehicle (Saline, four PG-treated and four EB-treated animals) into the SFO (A) and the effects of pretreatment with saralasin (Sar; seven PG-treated and seven EB-treated animals; B), phentolamine (Phe; seven PG-treated and seven EB-treated animals; C) or saline vehicle (Saline; four PG-treated and four EB-treated animals; D) in the PVN on the ANG II-induced water intake in the PG-treated (PG) and EB-treated (EB) animals. Results are expressed as means ^ SEM. Total water intake after ANG II injection into the SFO was significantly lower in the EB-treated than in PG-treated animals (A–D). †P , 0:05, ††P , 0:01, †††P , 0:001. In both groups, water intake to ANG II injected directly into the SFO was significantly diminished by pretreatment with Sar (B) or Phe (C), but not by Saline (D), in the PVN. *P , 0:05, **P , 0:01 compared with those of ANG II. ††P , 0:01.
J. Tanaka et al. / Neuroscience Letters 324 (2002) 242–246
drinking response by the pretreatment was found in any animals. The amount of water intake after the treatment with either saralasin (Fð1;12Þ ¼ 18:962, P , 0:01) or phentolamine (Fð1;12Þ ¼ 17:001, P , 0:01) in the PVN was significantly greater in the PG-treated than in the EB-treated animals. There was no significant difference between the PG-treated (the saralasin treatment, 55 ^ 7% decrease compared with the pretreatment; the phentolamine treatment, 32 ^ 4% decrease compared with the pretreatment) and EB-treated (the saralasin treatment, 53 ^ 6% decrease compared with the pretreatment; the phentolamine treatment, 34 ^ 4% decrease compared with the pretreatment) animals in the amount of the attenuation of total water intake by the drug treatment. Previous injections of saline vehicle into the PVN, on the other hand, had no significant influence on the drinking response in both groups (four PGtreated and four EB-treated animals; Fig. 2D). Pretreatment with either saralasin or phentolamine in the PVN did not cause significant differences in the latencies to drinking in both groups (the PG-treated animals, 32 ^ 10 s in the pretreatment vs. 34 ^ 11 s the saralasin treatment, 30 ^ 10 s in the pretreatment vs. 35 ^ 8 s in the phentolamine treatment; the EB-treated animals, 28 ^ 10 s in the pretreatment vs. 35 ^ 11 s the saralasin treatment, 34 ^ 10 s in the pretreatment vs. 35 ^ 9 s in the phentolamine). These results show that an increase in the circulating level of estrogen may attenuate the drinking behavior induced by angiotensinergic activation of the pathways from the SFO to the PVN. On the other hand, since the antagonist injection into the PVN is not able to abolish completely the ANG II-induced drinking response and to produce any significant changes in the response latency, it is possible to speculate that other neural circuits that are activated by ANG II may be involved in the drinking response. Immunohistochemical observations have revealed that ANG-immunoreactive neurons in the SFO project directly to the PVN [9]. Electrophysiological data have demonstrated that SFO neurons projecting to the PVN are activated by ANG II and their responsiveness to ANG II is suppressed by estrogen [14]. Recent studies have shown that an increase in the circulating level of estrogen decreases ANG II binding to AT1 receptors in the SFO [6] and the expression of both the AT1 and estrogen receptors [12]. Previous [15] and our findings in which saralasin injected into the PVN reduced the drinking response induced by ANG II injections into the SFO indicate that the angiotensinergic system in the PVN is implicated in mediating the response. Although there is no direct evidence showing the co-existence of ANG II AT1 and estrogen receptors in SFO neurons with angiotensinergic projections to the PVN, it might be expected that estrogen may act to attenuate the drinking response by decreasing the sensitivity of direct angiotensinergic SFO projections to ANG II. It has been postulated that the NA system in the PVN is implicated in the regulation of drinking behavior. Adminis-
245
tration of noradrenergic agonists into the PVN has been shown to increase drinking behavior [8]. The NA blockade of the hypothalamic sites has been shown to reduce the ANG II-induced drinking [4]. Microdialysis investigations have indicated that electrical [16] or chemical (ANG II) [15] stimulation of the SFO facilitates NA release in the PVN. Although the precise mechanisms underlying an activation of the noradrenergic system in the PVN in response to ANG II acting at the SFO cannot be determined from our data, these findings and our data suggest that estrogen may serve to attenuate the ANG II-induced drinking response that is mediated through an a-adrenoceptor mechanism in the PVN. It has been known that estrogen diminishes the drinking induced by systemic or intraventricular injections of ANG II. Intracerebroventricular administration of estrogen has been shown to reduce the increased drinking elicited by ANG II [5]. In addition, the ANG II-induced drinking has been shown to be attenuated by estrogen treatment [2,7]. The present data strongly suggest that the neural circuits from the SFO to the PVN may be involved in the attenuation in the drinking behavior observed in the previous studies and in female rats during estrous cycle [1].
[1] Findlay, A.L.R., Fitzsimons, J.T. and Kucharczyk, J., Dependence of spontaneous and angiotensin-induced drinking in the rat upon the oestrous cycle and ovarian hormones, J. Endocrinol., 82 (1979) 215–225. [2] Fregly, M.J. and Thrasher, T.N., Attenuation of angiotensininduced water intake in estrogen-treated rats, Pharmacol. Biochem. Behav., 9 (1978) 509–514. [3] Gutmann, M.B., Jones, D.L. and Ciriello, J., Effect of paraventricular nucleus lesions on drinking and pressor responses to ANG II, Am. J. Physiol., 255 (1988) R882–R887. [4] Jones, D.L., Injections of phentolamine into the anterior hypothalamic–preoptic area of rats blocks both pressor and drinking responses produced by central administration of angiotensin II, Brain Res. Bull., 13 (1984) 127–133. [5] Jonklass, J. and Buggy, J., Angiotensin–estrogen interaction in female brain reduces drinking and pressor responses, Am. J. Physiol., 247 (1984) R167–R172. [6] Kisley, L.R., Sakai, R.R. and Fluharty, S.J., Estrogen decreases hypothalamic angiotensin II AT1 receptor binding and mRNA in the female rat, Brain Res., 44 (1999) 34–42. [7] Kisley, L.R., Sakai, R.R. and Fluharty, S.J., Ovarian steroid regulation of angiotensin II induced water intake in the rat, Am. J. Physiol., 276 (1999) R90–R96. [8] Leibowitz, S.F., Paraventricular nucleus: a primary site mediating adrenergic stimulation of feeding and drinking, Pharmacol. Biochem. Behav., 8 (1978) 163–175. [9] Lind, R.W., Swanson, L.W. and Ganten, D., Angiotensin II immunoreactivity in neural afferents and efferents of the subfornical organ of the rat, Brain Res., 321 (1984) 209–215. [10] McKinley, M.J., McAllen, R.M., Mendelsohn, F.A.O., Allen, A.M., Chai, S.Y. and Oldfield, B.J., Circumventricular organ: neuroendocrine interfaces between the brain and the hemal milieu, Front. Neuroendocrinol., 11 (1990) 91–127. [11] Paxinos, G. and Watson, C., The Rat Brain in Stereotaxic Coordinates, Academic Press, Sydney, 1986. [12] Rosas-Arellano, M.P., Solano-Flores, L.P. and Ciriello, J., Co-localization of estrogen and angiotensin receptors
246
J. Tanaka et al. / Neuroscience Letters 324 (2002) 242–246
within subfornical organ neurons, Brain Res., 837 (1999) 254–262. [13] Simerly, R.B., Chang, C., Muramatsu, M. and Swanson, L.W., Distribution of androgen and estrogen receptor mRNA-containing cell in the rat brain; an in situ hybridization study, J. Comp. Neurol., 294 (1990) 76–95. [14] Tanaka, J., Hayashi, Y., Miyakubo, H., Sakamaki, K., Okumura, T. and Nomura, S., Estrogen decreases the responsiveness of subfornical organ neurons projecting to the hypothalamic paraventricular nucleus to angiotensin II in female rats, Neurosci. Lett., 307 (2001) 155–158.
[15] Tanaka, J., Hayashi, Y., Nomura, S., Miyakubo, H., Okumura, T. and Sakamaki, K., Angiotensinergic and noradrenergic mechanisms in the hypothalamic paraventricular nucleus participate in the drinking response induced by activation of the subfornical organ in rats, Behav. Brain Res., 118 (2001) 117–122. [16] Tanaka, J., Hayashi, Y., Sakamaki, K., Okumura, T. and Nomura, M., Activation of the subfornical organ enhances extracellular noradrenaline concentrations in the hypothalamic paraventricular nucleus in the rat, Brain Res. Bull., 54 (2001) 421–425.