INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

IN FEMALE RATS, a number of behaviors have been shown to be influenced by the stage of the estrous cycle. Among these behaviors are sexual receptivity, spontaneous locomotor activity, food ingestion, and water intake. Female rats display sexual receptive behaviors (13) and exhibit increased wheel running activity when estrogen levels are high at proestrus and estrus (26). Conversely, elevated plasma estrogen levels are associated with a reduction in food consumption as well as an attenuation of water intake (9, 27).

With regard to water intake, it has been demonstrated in the female rat that spontaneous drinking as well as thirst that has been elicited via the activation of the renin-angiotensin system is attenuated at estrus (9, 29). More specifically, it has also been shown that acute administration of estrogen to ovariectomized (OVX) female rats results in a reduction in spontaneous (9) and polyethylene glycol-induced (28) water intake. Chronic estrogen treatment has also been shown to attenuate water intake in response to peripheral (10, 12) and intracerebroventricular (11) administration of angiotensin II (ANG II), as well as to isoproterenol, a -adrenergic agonist that activates the renin-angiotensin system (11, 12). In contrast, thirst induced by cellular dehydration is not affected by estrogen (9, 16, 18).

The reduction in response to intracerebroventricular pulse (pICV) injection of ANG II has led to the hypothesis that estrogen regulates ANG II-induced thirst through a central mechanism. The possibility of a central interaction between ANG II and estrogen is supported by findings in several labs. For instance, Jonklaas and Buggy (17) implanted OVX female rats with crystalline estrogen directly into the medial preoptic area and demonstrated that it was sufficient to attenuate thirst elicited by a central injection of ANG II. Their results also revealed that several doses of intracerebroventricular estradiol benzoate (EB) could attenuate intracerebroventricular ANG II-induced thirst in female but not male rats (16). In addition, it has also been shown that injection of ANG II directly into the preoptic area of cycling female rats is less effective at stimulating water intake at estrus compared with other stages of the cycle (18, 19).

The purpose of the present set of experiments was to explore more thoroughly the antidipsogenic effects of estrogen, to provide the first evidence that would elucidate mechanisms underlying estrogen regulation of water intake, and to investigate the role of progesterone in regulating ANG II-induced thirst. OVX female rats were treated with a range of doses of systemic estrogen for a length of time intended to mimic the estrous cycle, and their responsiveness to an intracerebroventricular injection of ANG II was tested. In addition, the antiestrogen CI-628 was employed in a subset of experiments to determine whether the effects of estrogen on water intake were reversible and whether they involved a genomic mechanism, because CI-628 is believed to exert its effects through interactions with the estrogen receptor. Finally, the first studies that examine the possibility of coordinate regulation of water intake by exogenous administration of both estrogen and progesterone are presented.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Animals

Surgery

Before undergoing hormone treatments, animals were tested for correct cannula placement and patency. They were given a pICV injection of 6 ng ANG II (Sigma; 1-µl vol) via a Hamilton syringe connected with PE-10 tubing to an injector that terminated 1 mm beyond the guide cannula. When animals were not being tested, an obturator of the same gauge as the injector was inserted into the cannula. Water intake was measured for 15 min after the pICV ANG II injection using a burette with 0.2-ml divisions that was fitted with a stainless steel spout. Animals that did not drink the criterion amount of 5 ml by the end of the testing period were not included in subsequent experiments. Approximately 10% of the animals failed to drink the criterion amount, and dye injections followed by dissection revealed that the cannulas in these animals were not located in the third ventricle.

Hormone Treatments

Experiment 1: ANG II vs. carbachol-induced thirst. Animals (n = 5 for each group) were given propylene glycol (PG) or 10 µg EB (100-µl vol sc) daily for 48 h. Twenty-four hours after the last injection, water intake was measured for 15 min following a pICV injection of either 6 ng ANG II or 365.2 ng carbachol (carbamylcholine chloride; Sigma; 1-µl vol). A dose of carbachol was chosen whose dipsogenic potency was equivalent to ANG II.

Experiment 2: Dose-response curve for EB. Animals were divided into five groups (n = 5-9 per group) and were given either an injection of EB (1, 2, 10, or 25 µg) or PG (100-µl vol sc) daily for 48 h. Twenty-four hours later, water intake tests were performed with ANG II and cumulative intakes were recorded after 5, 10, and 15 min.

Experiment 3: Effects of the antiestrogen CI-628. Twenty-four hours before hormone administration, animals (n = 6 for each group) were each injected with 2 mg nitromifene citrate ip (CI-628; Parke-Davis; dissolved in distilled water) or with the distilled water vehicle. Two micrograms of CI-628 was then administered just before injections of PG or EB. PG or 2 µg EB was subcutaneously injected daily for 48 h. Twenty-four hours after the last treatment, animals were given a 15-min ANG II water intake test. The doses of CI-628 and EB and the schedule of administration were chosen on the basis of previous studies that showed that 2 mg CI-628 can effectively block the induction of hypothalamic-preoptic progesterone receptors by 2 µg EB (25). CI-628 is also known to produce mild anorexic effects (5, 6); thus body weights were measured daily throughout the CI-628 studies as independent confirmation that the CI-628 injections were physiologically effective. Each animal was used as its own control, in that the body weight data were expressed as a percentage of pretreatment body weight, and these percentages were compared between experimental groups.

Experiment 4: Effects of progesterone on water intake. In experiment 4A animals were divided into four treatment groups (n = 5 for each group). Group 1 received daily injections of PG for 48 h. Rats were given a pICV ANG II water intake test 24 h after the last dose. Group 2 received 2 µg EB daily for 48 h. A pICV ANG II water intake test was given 24 h after the last injection. Group 3 received 2 µg EB daily for 48 h, and 48 h after the second EB injection they were given 500 µg progesterone subcutaneously (Sigma; dissolved in PG). Animals were tested 4 h after progesterone injection with the standard pICV ANG II drinking test. Group 4 received 500 µg progesterone and were tested 4 h later with the pICV ANG II. All hormonal treatments were timed so that all animals were given the pICV ANG II water intake test at the same time.

Statistical analysis. The results are presented as means ± SE. ANOVA was performed on all of the data, and the Student-Newman-Keuls test was used as a post hoc test with the significance level set at P < 0.05.

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

Experiment 1: ANG II vs. Carbachol-Induced Thirst

View larger version (12K): [in this window] [in a new window]   Fig. 1.   Comparison of 15-min water intakes induced by either a central injection of angiotensin II (ANG II, 6 ng) or carbachol (365.2 ng) in ovariectomized (OVX) female rats given propylene glycol vehicle (PG) or 10 µg of estradiol benzoate (EB) for 48 h. **Significantly different from vehicle-treated animals (P < 0.01 by ANOVA and Student-Newman-Keuls post hoc test; n = 5 for each group).

Experiment 2: Dose-Response Curve for EB

View larger version (21K): [in this window] [in a new window]   Fig. 2.   Central ANG II-induced drinking in OVX female rats after 48-h administration of PG vehicle or increasing doses of EB (1-25 µg). In 1 set of animals, isotonic saline (vehicle; ) was injected instead of ANG II. *Significantly different from PG treatment group (P < 0.05); **significantly different from PG treatment group (P < 0.01; n = 5-9 for each group).

Experiment 3: Effects of the Antiestrogen CI-628

View larger version (12K): [in this window] [in a new window]   Fig. 3.   A: systemic administration of 2 mg anti-estrogen CI-628 blocks effects of 2 µg EB on ANG II-induced thirst. NS, not significant; **significantly different from PG-treated controls (P < 0.01; n = 6 for each group). B: systemic administration of 2 µg CI-628 and 2 µg EB, separately and together, significantly affects weight gain in OVX female rats. *Significantly different from vehicle (PG)-treated controls (P < 0.05; n = 6 for each group).

Experiment 4: Effects of Progesterone on Water Intake

View larger version (15K): [in this window] [in a new window]   Fig. 4.   Treatment with 2 µg EB and 500 µg progesterone (P) for 4 h results in attenuation of water intake similar to attenuation seen with 2 µg EB alone. Progesterone administered without estrogen priming does not significantly attenuate ANG II-induced thirst. *Significantly different from vehicle (PG)- and P-treated groups (P < 0.05; n = 5 for each group).

Although in experiment 4A it was shown that progesterone did not facilitate the effects of EB, it is possible that progesterone regulates water intake by reversing the effects of estrogen. In reproductive behavior studies, it has been shown that progesterone initially enhances and then inhibits the effects of estrogen on lordosis behavior in the female rat. The synergistic effects of progesterone are seen 4-6 h after injection, but these effects are gone within 24 h, and additional progesterone does not facilitate lordosis (21, 22). Experiment 4B was conducted to examine whether progesterone had a biphasic effect on water intake similar to its effect on reproductive behavior when given to estrogen treated rats. In experiment 4B, we were not able to establish that progesterone reversed the effects of EB. The effects of 2 µg EB persisted 48 h following the last injection of estrogen. However, it is clear from Fig. 5 that 4-h progesterone treatment did not enhance and 24-h progesterone treatment did not reverse the effects of 48-h estrogen administration. Animals tested 24 h after the progesterone injection still had significantly attenuated water intakes compared with control animals, but the antidipsogenic effects were not as pronounced as in other treatment groups.

View larger version (14K): [in this window] [in a new window]   Fig. 5.   Forty-eight-hour treatment with 2 µg EB followed by 24-h treatment with 500 µg P does not reverse effects of estrogen on water intake. Effects of estrogen treatment on ANG II-induced drinking are apparent 48 h after last hormone injection. *Significantly different from vehicle (PG)-treated animals (P < 0.05; n = 6 for each group).

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

Previous studies that examined thirst over the course of the estrous cycle of the rat have revealed that a reduction in water intake occurs when plasma estradiol levels are elevated. At estrus, it has been repeatedly demonstrated that spontaneous water intake is decreased compared with other stages of the cycle (9, 27). In addition, the efficacy of agents, such as isoprenaline, that induce thirst via activation of the renin-angiotensin system has been shown to be reduced at proestrus and estrus (9). Thirst elicited by the direct injection of ANG II into the preoptic area is also attenuated at the proestrus and estrus phases of the cycle (9). These behavioral changes have been attributed to the actions of the elevated circulating levels of estrogen associated with the proestrus and estrus stages of the cycle. However, because these experiments relied on the use of hormonally intact animals, they were unable to directly examine the relative contributions of each of the ovarian steroid hormones to the regulation of water intake in the normally cycling female rat.

An alternative approach to examining the contributions of ovarian steroid hormones to behavior has been to remove endogenous sources of hormones by ovariectomy and replace estrogen, progesterone, or both in a clearly defined manner. A number of experiments have examined peripheral and central ANG II-induced water intakes in ovariectomized rats receiving estrogen in Silastic capsules for several weeks (10-12). Although these studies clearly showed that estrogen had the potential to regulate water intake, they failed to mimic a natural hormonal state for the animal, which raises the question of the biological significance of these earlier results. A series of experiments was performed on ovariectomized female rats in which estrogen replacement was given in crystalline form directly into the brain 24 h before testing (17). These studies identified the medial preoptic area as a potential site for estrogen regulation of ANG II-induced thirst, as estrogen administration to other areas of the hypothalamus was not antidipsogenic. However, this treatment paradigm did not represent a natural state for the rat and thus did not offer a physiologically relevant model of how estrogen may regulate water intake in the intact, cycling rat.

By ovariectomizing female rats and replacing hormones systematically over a shorter time frame, we have provided evidence to support the idea that estrogen regulation of the central renin-angiotensin system may represent a physiological mechanism by which thirst is regulated over the course of the estrous cycle. In the present studies, EB given in a time course that mimics the estrous cycle attenuated ANG II-induced water intake in a dose-dependent manner. As such, this is the first study that examines the contribution of estrogen to the regulation of water intake using a physiologically relevant hormone treatment regimen.

In other studies, it has been shown that the daily injection of 10 µg EB for 48 h results in proestrus levels of estrogen 24 h after the last injection (2, 30). In our experiments, this dose regimen was sufficient to significantly attenuate water intake that had been elicited with a central injection of ANG II. The magnitude of the attenuation with 10 µg EB was comparable to the reduction in water intake that has been observed at proestrus and estrus after central ANG II infusions (9). Supraphysiological doses were not shown to be more effective at attenuating water intake.

The antidipsogenic effects of estrogen were specific for ANG II-induced thirst, inasmuch as water intake induced by the cholinomimetic carbachol was unaffected by hormone treatment. This specificity has been demonstrated after estrogen administration directly into the medial preoptic area (17) and at estrus in the course of the female rat reproductive cycle (9). The lack of effect on carbachol-induced thirst is particularly interesting, as it has been demonstrated that ANG II and carbachol infusions activate many of the same brain nuclei thought to be important in the regulation of body fluid homeostasis but utilize pharmacologically distinct neural pathways (1, 20, 24).

Our studies have also provided evidence to suggest that the antidipsogenic effects of estrogen may be mediated by a genomic mechanism, as pretreatment with the antiestrogen CI-628 reverses the effects of physiological doses of EB. The systemic dose of CI-628 (2 mg) has been shown to effectively penetrate the blood-brain barrier and modulate other estrogen-regulated behaviors. When given in the same time frame as described for our water intake experiments, CI-628 prevents the estrogen stimulation of lordosis and the luteinizing hormone surge in female rats (4, 8, 15) and also blocks estrogen induction of progesterone receptors in both the hypothalamus and the pituitary (8, 25). In addition to its effects on water intake, CI-628, separately and in combination with EB, significantly altered body weight. Animals treated with EB, CI-628, or both failed to gain weight during the treatment period. This effect has been demonstrated by other investigators and can serve as an independent indication of the efficacy of the CI-628 injections (5, 6).

In reproductive behavior studies in the female rat, administration of progesterone after estrogen pretreatment has been shown to initially enhance, and then reverse, the effects of estrogen (21, 22). Using the same hormone replacement paradigms that have typically been used to study female rat sexual behavior, we replaced estrogen and progesterone and administered water intake tests with ANG II. Although progesterone acts synergistically with estrogen to enhance sexual receptivity in female rats (21, 22), we have not been able to establish a similar relationship with regard to ovarian steroid regulation of water intake. Progesterone did not appear to increase the antidipsogenic effects of estrogen on ANG II-induced water intake. When ineffective doses of estrogen (1 µg) were given in conjunction with progesterone, water intakes were not reduced significantly below control levels.

By contrast, it has been shown that long-term treatment of ovariectomized female rats with progesterone results in a decrease in water and salt intake (7). However, it is difficult to compare these data with our own because the progesterone was administered without estrogen and testing took place after 21 days of daily progesterone injections. In our experiments, progesterone was given in a single dose after estrogen priming in an effort to mimic the estrous cycle of the rat. It is likely that the difference between our results and those after long-term progesterone treatment is due to the difference in experimental design.

Female rats initially display increased sexual receptivity after EB and progesterone treatment, but 24 h later, the effect is absent (21, 22). Animals are no longer receptive, and additional injections of progesterone are ineffective at eliciting lordosis responses. In our water intake experiments, we were unable to demonstrate that progesterone caused either a facilitation or a reversal of the antidipsogenic effects of estrogen. These results suggest that although progesterone is critical in the precise regulation of female sexual behavior, it does not seem to be importantly involved in the regulation of thirst, at least in the context of these studies. It is possible that the fluctuations in estrogen levels over the course of the estrous cycle are sufficient to attenuate spontaneous and ANG II-induced water intake. It is also possible that, in conjunction with estrogen, other unidentified factors associated with the estrous cycle may influence the cyclicity of water consumption.

Although the mechanism by which estrogen regulates ANG II-induced thirst was not addressed in these experiments, there are a number of studies that suggest that estrogen reduces the dipsogenic response to ANG II by reducing the number of hypothalamic receptors available to bind ANG II (3, 17). Radioligand binding assays on homogenates of microdissected hypothalamic tissue localized the angiotensin receptor population that is sensitive to estrogen to the medial preoptic area.

Although changes in receptor binding have been detected, the mechanisms of these changes may be genomic or reflect more rapid posttranslational downregulation. Our CI-628 data provide support for the notion of a genomic mechanism involved in the estrogen regulation of the central renin-angiotensin system. However, it is not clear whether the CI-628 is interfering with estrogen regulation at the level of the ANG II receptors or at another level, perhaps by altering the expression of a synthetic component of ANG II. For instance, it has been shown that the precursor of ANG II, angiotensinogen, is located in the hypothalamus and is sensitive to estrogen treatment (14). Any change in central ANG II levels would be an indirect mechanism by which receptors could potentially be regulated by estrogen. Alternatively, if ANG II is indeed a neurotransmitter utilized by neurons that regulate thirst, then estrogen may change the availability of ANG II, which would, in turn, alter the efficacy of the pathway. It is also possible that estrogen attenuates ANG II-induced thirst indirectly through the activation of another neurochemical system, such as the tachykinins, which have been shown to inhibit central ANG II-induced water intake (23).

The CI-628 treatment has opposing effects on food and water intake and appears to provide a means to dissociate these ingestive behaviors. This dissociation can be attributed to CI-628 having an antagonist effect on ANG II-induced thirst but a partial agonist effect on the anorexic effects of estrogen. Our data are in agreement with previously published studies that demonstrate that CI-628 is as effective as estrogen at reducing body weight, which, they have shown, results from decreased food intake (5, 6). This observation strongly supports the notion that the estrogen effect on water intake seen over the estrous cycle is not merely a by-product of reduced food intake but is an independently regulated system. Future studies, perhaps utilizing quantitative autoradiography, that localize the brain nuclei in which ANG II receptor binding changes occur should help define a specific role for estrogen in the modulation of thirst. In situ hybridization or other methods that quantitatively examine mRNA levels of ANG II receptors and the synthetic components of the central renin-angiotensin system should also provide further insight into the mechanisms by which estrogen regulates water intake.

In summary, EB, but not progesterone, given in physiologically significant doses and lengths of time has been shown to attenuate water intake induced by an intraventricular injection of ANG II in a dose-dependent and reversible manner, possibly through a genomic mechanism. Confirming prior observations, carbachol-induced water intake, which involves overlapping neural structures but a different neurochemical pathway, was not affected by estrogen treatments. This work demonstrates a central interaction between the peptide hormone ANG II and estrogen that involves changes in gene transcription and may elucidate the hormonal mechanism for the cyclicity of ingestive behavior in female rats.