Baroreflex sensitivity varies during the rat estrous cycle: role of gonadal steroids

Abstract

Baroreflex sensitivity (BRS) increases in women during the luteal phase of the menstrual cycle, when gonadal hormones are elevated, but whether a similar cycle-dependent variation in BRS occurs in rats is unknown. In addition, whether cyclic BRS changes depend on gonadal steroids has not been previously investigated. To test these hypotheses, BRS was determined in cycling female rats using two approaches: 1) baroreflex control of renal sympathetic nerve activity (RSNA) in anesthetized rats; 2) cardiovagal spontaneous BRS (sBRS) in conscious rats instrumented for continuous telemetric measurements of mean arterial pressure (MAP) and heart rate (HR). MAP, HR, and sBRS were also measured in rats 2–3 and 5–6 wk following ovariectomy (OVX), to eliminate gonadal steroids. In anesthetized rats, RSNA BRS gain was increased (P < 0.01) during proestrus (−4.8±0.5% control/mmHg) compared with diestrus/estrus (−2.8 ± 0.3% control/mmHg). Similarly, a proestrous peak in sBRS was observed in conscious rats (1.66 ± 0.07 ms/mmHg, proestrus; 1.48 ± 0.06 ms/mmHg, diestrus/estrus; P < 0.001). OVX eliminated estrous cycle-induced variation in sBRS. In addition, OVX reduced (P < 0.05) diurnal variations in MAP (5.9 ± 0.3 vs. 3.9 ± 0.5 mmHg) and HR [54 ± 4 vs. 39 ± 3 beats per minute (bpm)], and abolished diurnal variations in sBRS. Finally, while MAP, HR, and sBRS were decreased 2–3 wk following OVX, ∼3 wk later, MAP and sBRS increased, and HR decreased further. No changes in MAP, HR, or sBRS were seen with time in sham OVX controls. In summary, RSNA and cardiovagal sBRS vary during the rat estrous cycle, and this variation is abolished by OVX. We conclude that sex steroid hormones are required for both cyclic and diurnal changes in BRS in rats.

considerable evidence indicates that gonadal steroids can influence mean arterial pressure (MAP), heart rate (HR), and baroreflex sensitivity (BRS). First, sex differences have been documented (1, 2, 5, 16, 20, 27). Second, MAP and HR vary during the estrous cycle of the rat (53), and, in humans, sympathetic BRS increases during the midluteal phase, when gonadal steroid hormone concentrations are higher, compared with the early follicular phase of the menstrual cycle (32). Nevertheless, whether BRS changes during the rat estrous cycle has not been studied. Moreover, whether changes in gonadal steroids underlie the reproductive cycle variability in MAP, HR, and BRS is unclear. Ovariectomy (OVX) and administration of estrogen and/or progesterone have been shown to alter MAP and HR in women and experimental animals; however, results have been conflicting (12, 19, 31, 36, 42, 53, 58). Furthermore, estrogen increases BRS (25, 34, 40, 47); in contrast, a neurosteroid metabolite of progesterone, allopregnanolone, decreases BRS (18). Thus, a clear understanding of the influence of cyclic changes in endogenous gonadal steroids on MAP, HR, and BRS is lacking.

Therefore, the primary goal of this study was to determine if BRS varies during the rat estrous cycle and if this variation depends on gonadal steroids. Because numerous questions remain regarding the cardiovascular actions of exogenous sex steroids in young and postmenopausal women (30, 33), further studies in rats will likely continue to be useful in elucidating underlying physiological and pathophysiological mechanisms. We first tested the hypothesis that BRS in rats is elevated during the proestrous phase (high levels of estrogen) compared with the diestrous phase (low levels of estrogen), in a similar fashion as women (32). To test this hypothesis, we determined whether baroreflex control of renal sympathetic nerve activity (RSNA) is enhanced during proestrus in anesthetized female rats. We also determined whether cardiovagal BRS varies during the estrous cycle of conscious rats, by estimating spontaneous BRS (sBRS) using sequence method analysis of telemetric, continuous arterial blood pressure recordings. Second, to test the hypothesis that increases in gonadal steroids mediate estrous cycle variations, we determined whether OVX abolishes cardiovagal sBRS variations, as well as cycle-related changes in MAP and HR. Finally, to determine whether the variable effects of OVX on MAP and HR previously reported resulted from the differing intervals allowed between OVX and experimentation, we assessed MAP, HR, and sBRS telemetrically at two times following OVX.

METHODS

Animals

Female Sprague-Dawley rats (12 wk old; Charles River Laboratories) were acclimated to the laboratory for at least 7 days prior to surgery or experimentation. Animals were housed with a 12:12-h light-dark cycle with food (LabDiet 5001) and water provided ad libitum. All of the procedures were conducted in accordance with the National Institutes of Health Guide for the Health and Use of Laboratory Animals and were approved by Oregon Health & Science University or University of Texas Health Science Center at San Antonio animal care and use committees.

Survival Surgery

Telemetric catheter implantation.

Anesthesia was induced with 5% isoflurane in 100% oxygen and maintained with 1.5–2% isoflurane in 100% oxygen. A single intramuscular dose of 30,000 U Penicillin G (Hanford's United States Veterinary Products) was administered 10–15 min prior to incision. Using sterile technique, we made an inguinal incision, and the catheter and transmitter (PAC 40, 10-cm length; Data Sciences International, St. Paul, MN) were inserted into the femoral artery. The tip of the catheter was advanced retrograde into the distal abdominal aorta. The catheter was secured, the transmitter was placed into a subcutaneous pocket created in the flank, and the skin was closed. Animals received codeine (1 mg/100 ml) in their drinking water ad libitum for 3 days postprocedure. Rats recovered a minimum of 7 days prior to data collection.

Ovariectomy.

One week before catheter implantation, some rats first underwent either bilateral OVX or sham OVX. Briefly, after induction of anesthesia and administration of prophylactic antibiotic, small posterior flank incisions were made. Each ovary was exteriorized through the wound, separated from the uterus, and removed. A single 3-0 silk tie was used to control the vascular pedicle. The muscle and skin layers were individually closed with interrupted 3-0 silk suture. For sham OVX rats, the same procedure was performed, except the ovaries were left intact. Rats were provided with oral codeine as after the telemetry surgery.

Determination of Baroreflex Function

RSNA baroreflex curves.

Rats were anesthetized and instrumented with femoral arterial and venous catheters, for MAP measurement and drug infusions, respectively. A multifiber renal nerve branch was placed on a bipolar platinum recording electrode and fixed with Wacker Sil-Gel. Nerve activity was amplified (10,000–50,000 times), filtered (band pass between 30 and 1,000–3,000 Hz) with a Grass P511 band-pass amplifier and a Grass HIP511 high-impedance probe, and integrated using a Grass model 7P10G integrator. Computer data acquisition was performed with an analog-to-digital converter using Labtech Notebook software. For baroreflex curve construction, RSNA was recorded while MAP was increased to ∼200 mmHg over a period of 1- 2 min by adjusting the rate of an intravenous infusion of phenylephrine over a dose range of 10–100 μg·kg⁻¹·min⁻¹. After allowing time for MAP and RSNA to return to baseline levels, we recorded RSNA while MAP was lowered to ∼50 mmHg by adjusting the rate of an intravenous infusion of sodium nitroprusside over a dose range of 1–5 μg·kg⁻¹·min⁻¹. The level of nerve activity following a lethal intravenous injection of pentobarbital was taken as background noise and subtracted from the nerve activity recorded during the experimental protocol. Baroreflex curves for RSNA were analyzed with a four parameter sigmoidal equation, RSNA = p4 + p1/{1 + exp[p2(MAP − p3)]}, which provided values for maximal RSNA gain, maximal baroreflex RSNA, minimum baroreflex RSNA, RSNA range, and MAP at midrange (MAP₅₀).

Cardiovagal sBRS.

Arterial pressure was measured continuously via the telemetry catheter at 1,000 Hz, and sBRS was determined post hoc using the sequence method (Hemolab, Iowa City, IA) (4, 52). From the arterial waveform, the software application identified ramps of heart beats in which systolic blood pressure and pulse interval both either increased or decreased. A delay of three beats was selected based on previous studies (38). Data were analyzed in 1-h increments. Artifacts and at most one sequence outlier, which was defined as a value at least 3 standard deviations from the mean, were removed. The software application calculates a regression line between systolic pressure and pulse interval (minimum r threshold was set as 0.8) for each sequence, the slope of which is the sBRS (ms/mmHg) for that individual sequence. Subsequently, the hourly means for all sBRS sequences, as well as MAP and HR (Dataquest ART 4.0, Data Sciences International, St. Paul, MN), were calculated. From these data, 6 or 12-h means were determined for graphical or statistical analysis, respectively.

To confirm that sBRS represents largely the gain of parasympathetic control of the heart, in a subset of rats, hourly averages of sBRS, the number of identified baroreflex sequences, MAP, and HR were determined 4 h before and 4 h after the administration of methscopolamine (0.5 mg/kg iv, n = 4, 2.0 mg/kg iv, n = 3) during the dark phase, 4 h before lights on.

Experiment Protocols

Does gain of baroreflex control of RSNA increase during proestrus in anesthetized rats?

The estrous cycle of each rat was determined by examining the cytology of vaginal smears for two consecutive estrous cycles prior to baroreflex testing. Rats were anesthetized with an intraperitoneal injection of Inactin (100 μg/kg) and prepared with femoral artery and vein catheters and with an electrode implanted around a branch of the renal sympathetic nerve. After a 1-h stabilization period, baroreflex relationships between MAP and RSNA were then determined.

Does cardiovagal sBRS increase during proestrus in conscious rats?

The estrous cycle of each rat was first established by examining the cytology of vaginal smears for at least two consecutive 4- or 5-day cycles. To avoid the potential effects of swabbing on the continuous measurements of sBRS, we measured arterial pressure telemetrically and determined sBRS during the subsequent two cycles; data were analyzed for the second of these cycles.

Do cycle- and circadian-dependent variations in MAP, HR, or sBRS depend on gonadal hormones?

MAP, HR, and cardiovagal sBRS were determined on four consecutive days (12 ± 1 to 15 ± 1 days) following OVX (Figs. 2–4). In a second cohort (Figs. 5–7), these variables were determined on four consecutive days at two time points following the initial OVX or sham OVX surgery: 18 ± 1 to 21 ± 1 days and 37 ± 4 to 40 ± 4 days following OVX, and 19 ± 1 to 22 ± 1 days and 33 ± 2 to 36 ± 2 days following sham OVX. The first time period is termed “early” and examines the initial effects of the loss of gonadal hormones, whereas the second period is termed “late” and determines the delayed effects of gonadal hormone loss.

Documentation of OVX

At the completion of the study, sham OVX and OVX rats were deeply anesthetized with 5% isoflurane in 100% oxygen. Blood (∼1 ml) was removed by cardiac puncture, the abdomen was opened, and the uterus was excised. Excess fat was removed, and the uterus was weighed. The blood was allowed to clot, and serum was collected following centrifugation. Sera samples were stored at −80°C, until estradiol-17β levels were measured in one radioimmunoassay by the Endocrine Technology and Support Core at the Oregon National Primate Research Center using a previously published procedure (43). Briefly, each serum sample (150 –220 μl), together with control samples containing water blank and tritiated estradiol-17β tracer for determination of recovery, were extracted with 5 ml of redistilled diethyl ether. After centrifugation, the supernatant was dried under an air stream, dissolved in 200 μl of solvent (hexane:benzene:methanol, 62:20:13) and subjected to chromatography on 1.0 gram Sephadex LH-20 in glass columns to isolate estradiol-17β (eluted between 9.5–15.5 ml). Concentrations of estradiol-17β in samples were assayed with an ultrasensitive estradiol-17β double antibody radioimmunoassay with a sensitivity of 0.5 pg/tube and were corrected for recovery; intra-assay variation is routinely <10%.

Statistical Analysis

In the experiments in anesthetized rats, results from animals studied during diestrus and estrus were grouped and compared with results obtained from proestrous rats using a Student's t-test. For experiments in which daily telemetric determinations of MAP, HR, and sBRS were obtained, between-group differences in 12-h averages were first evaluated using three-way ANOVA for repeated measures [factors were group (cycling and early OVX), day (1–4) and time of day (dark and light)], and the Newman-Keuls post hoc test was applied to ascertain specific within- and between-group differences. When this analysis revealed a significant three-way interaction, separate two-way ANOVA were used to assess within-group differences between days and time of day. To determine whether differences exist between the early and late periods following OVX or sham-OVX, two separate three-way ANOVA for repeated measures were performed for each group [factors were period (early and late), day (1–4), and time of day (dark and light)]. The significant results of ANOVA analyses are provided in the figure legends. Finally, uterine weights and estrogen concentrations were compared between OVX and sham OVX rats with a Student's t-test. In all cases, P < 0.05 was considered statistically significant.

RESULTS

Estrous cycle induced variability in baroreflex control of RSNA in anesthetized female rats.

Compared with rats in the diestrous and estrous phases, rats in proestrus exhibited elevations in maximal RSNA gain (−4.8 ± 0.5%/mmHg, proestrus; −2.8 ± 0.3%/mmHg, diestrus/estrus; P < 0.01), in maximal baroreflex RSNA (118 ± 19%, proestrus; 57 ± 11%, diestrus/estrus; P < 0.05) and in RSNA range (173 ± 20%, proestrus; 107 ± 9%, diestrus/estrus; P < 0.05) (Fig. 1). In contrast, neither minimum baroreflex RSNA (−55 ± 5%, proestrus; −50 ± 4%, diestrus/estrus; P > 0.50) nor BP₅₀ (113 ± 5 mmHg, proestrus; 114 ± 5 mmHg, diestrus/estrus; P > 0.50) was significantly different. Thus, baroreflex regulation of RSNA was enhanced during proestrus.

Fig. 1.

Baroreflex gain and maximum RSNA increase in anesthetized rats during proestrus (n = 6) compared with diestrus/estrus (n = 6) (see text for details). *P < 0.05 between groups, baroreflex maximum.

Effects of Methscopolamine on sBRS

Treatment of rats with methscopolamine almost completely eliminated the detection of sBRS sequences (187 ± 22 to 13 ± 4 sequences/h; P < 0.05), reduced sBRS of the remaining sequences (1.93 ± 0.01 to 0.92 ± 0.14 ms/mmHg; P < 0.05), increased MAP (105 ± 10 to 124 ± 12 mmHg) and increased HR [395 ± 12 to 429 ± 11 bpm; P < 0.05]. (See Supplemental Fig. 1 in the online version of the article for the complete data set.) Therefore, sBRS primarily reflects changes in parasympathetic activity to the heart.

Diurnal Variations of MAP, HR, and Cardiovagal sBRS in Conscious Cycling and OVX Female Rats

In intact rats, as expected, MAP and HR displayed diurnal variation (P < 0.0001), with a rise of MAP and HR during the dark phase (Figs. 2 and 3). In addition, sBRS was reduced during the dark phase (Fig. 4; P < 0.05).

Fig. 2.

Mean arterial pressure varies during the estrous cycle of the rat (n = 5), but not in OVX animals (n = 6). Three-way ANOVA revealed a significant (P < 0.05) diurnal (light vs. dark) effect, as well as group [cycling vs. ovariectomy (OVX)] by diurnal and group by day by diurnal interactions. Subsequent two-way ANOVA of the cycling group indicated a significant diurnal effect, as well as a day by diurnal interaction. Two-way ANOVA of the OVX group indicated a significant diurnal effect only. *P < 0.05 compared with all other days within group (comparisons made among either light periods or dark periods only); †P < 0.05 OVX compared with cycling at the same time. The dark phase of the day-night cycle is indicated by gray shading.

Fig. 3.

Heart rate does not vary significantly between days in either intact (n = 5) or OVX (n = 6) rats. Three-way ANOVA revealed a significant (P < 0.05) diurnal effect, as well as significant group by day and group by diurnal interactions. Nevertheless, no specific within-group differences were detected following post hoc testing. †P < 0.05 OVX compared with cycling at the same time. The dark phase of the day-night cycle is indicated by gray shading.

Fig. 4.

Spontaneous baroreflex sensitivity increases during proestrus in conscious rats (n = 5), and this cyclic variation is abolished following OVX (n = 6). Three-way ANOVA revealed significant (P < 0.05) group, diurnal, and between-day effects, as well as a group-by-day interaction. *P < 0.05 compared with at least one other day within group (comparisons made among either light or dark periods only); †P < 0.05 OVX compared with cycling at the same time. The dark phase of the day-night cycle is indicated by gray shading.

OVX significantly altered the diurnal rhythm of all cardiovascular variables (Fig. 2–4). For both MAP and HR, the magnitude of the light-dark difference was reduced in OVX rats (MAP: 5.9 ± 0.3 mmHg, intact; 3.9 ± 0.5 mmHg, OVX; P < 0.01; HR: 54 ± 4 bpm, intact; 39 ± 3 bpm, OVX; P < 0.01). In particular, OVX reduced MAP and HR during the 12-h dark phase (Figs. 2 and 3; P < 0.05). On the other hand, while HR values during the light phase were unaffected by OVX, light-phase MAP during estrus was elevated compared with OVX rats (Fig. 2; P < 0.05). Thus, a reduced increase in MAP and HR during the dark phase largely accounts for the decreased diurnal variation seen in the OVX rats. In addition, OVX abolished the diurnal variation of sBRS, which was due primarily to a failure for sBRS to increase during the light phase (Fig. 4).

Daily Variations of MAP, HR, and Cardiovagal sBRS in Conscious Cycling and OVX Female Rats

MAP.

Three-way ANOVA of MAP revealed a significant group by day by time interaction (P < 0.05). To determine whether this significance reflected differences between days, two-way ANOVAs were performed on the cycling and on the OVX rats. In the intact rats, dark-phase MAP during proestrus was elevated (P < 0.05) compared with all other cycle day dark-phase values, whereas light-phase MAP during proestrus was reduced compared with all other cycle day light-phase values (Fig. 2; P < 0.05). Consequently, the light-dark MAP variation during proestrus (8.4 ± 0.5 mmHg) was larger (P < 0.05) than on the other days of the cycle (estrus, 5.0 ± 0.8 mmHg; metestrus, 5.0 ± 0.5 mmHg; diestrus, 5.0 ± 0.3 mmHg). Thus, in the cycling rat, MAP achieved higher levels in the dark phase during proestrus. In contrast, MAP was not different between days in the OVX rats (Fig. 2).

HR.

HR did not vary significantly between days in either intact cycling rats or OVX rats (Fig. 3).

Cardiovagal sBRS.

Cardiovagal sBRS also varied during the estrous cycle (Fig. 4). During the light phase, proestrous sBRS was significantly increased compared with all other days (P < 0.05). During the dark phase, proestrous sBRS was increased only compared with metestrus (P < 0.05). In contrast, in the OVX group, no daily variation in cardiovagal sBRS was observed (Fig. 4).

Time-dependent effects of OVX on MAP, HR, and cardiovagal sBRS

While OVX rats initially exhibited reductions in MAP, HR, and sBRS compared with intact rats (Figs. 2–4), with time, MAP increased (Fig. 5; P < 0.01), HR decreased further (Fig. 6; P < 0.05) and sBRS increased (Fig. 7; P < 0.0001). In contrast, MAP, HR, and sBRS did not change during the study in sham OVX rats (Figs. 5–7). Importantly, in the OVX group, the diurnal variations in MAP and HR did not change, and the lack of diurnal and daily variation in sBRS persisted over the entire study period (Figs. 5–7). Indeed, the large BRS variability observed in the sham OVX rats (Fig. 7) is likely due in part to estrous cycle variability, as this was not controlled in this cohort.

Fig. 5.

Mean arterial pressure increases late post-OVX compared with early post-OVX (n = 4). In contrast, mean arterial pressure does not change following sham OVX (n = 3). Three-way ANOVA of the OVX group revealed significant (P < 0.05) time after OVX and diurnal effects. In the Sham OVX group, only a significant (P < 0.05) diurnal effect was detected. *P < 0.05 late OVX compared with early OVX (ANOVA, time after OVX effect). Shaded symbols, dark phase of the day-night cycle; open symbols, light phase of the day-night cycle.

Fig. 6.

Heart rate decreases late post-OVX compared with early post-OVX (n = 4). In contrast, heart rate does not change following sham OVX (n = 3). Three-way ANOVA of the OVX group revealed significant (P < 0.05) time after OVX and diurnal effects, as well as a significant day by diurnal interaction. Three-way ANOVA of the sham group revealed a significant (P < 0.05) diurnal effect, as well as day by diurnal and time after sham by day by diurnal interactions. *P < 0.05 late OVX compared with early OVX (ANOVA, time after OVX effect). Shaded symbols, dark phase of the day-night cycle; open symbols, light phase of the day-night cycle.

Fig. 7.

Baroreflex sensitivity increases late post-OVX compared with early post-OVX (n = 4). In contrast, baroreflex sensitivity does not change with time following sham OVX (n = 3). Three-way ANOVA of the OVX group revealed a significant (P < 0.05) effect of time after OVX. *P < 0.05 late OVX compared with early OVX (ANOVA, time after OVX effect). Shaded symbols, dark phase of the day-night cycle; open symbols, light phase of the day-night cycle.

Effects of OVX on Plasma Estradiol and on Uterine Weight

Both plasma estradiol levels (12.8 ± 3.6 pg/ml, sham OVX, n = 6; 3.5 ± 1.2 pg/ml, OVX, n = 9; P < 0.05) and uterine weights (0.81 ± 0.14 g, sham OVX, n = 6; 0.10 ± 0.01 g, OVX, n = 6; P < 0.001) were reduced in OVX rats compared with sham OVX animals.

DISCUSSION

The purpose of the present study was to test the hypothesis that BRS varies during the estrous cycle of the rat and that this variation depends on gonadal steroids. The major new findings are 1) OVX reduced diurnal changes in MAP and HR and eliminated diurnal variations in sBRS; 2) gain of baroreflex control of RSNA and cardiovagal sBRS were both elevated during proestrus; 3) OVX eliminated cycle-induced variations in sBRS; and 4) OVX initially decreased MAP, HR, and sBRS, but with time MAP and sBRS returned toward normal, and HR decreased further. We conclude that gonadal steroids are essential for cycle-dependent and circadian changes in BRS in rats.

The present results confirm the diurnal variations in MAP, HR, and BRS previously observed in male rats; during the dark phase, MAP and HR increase, whereas BRS declines (7, 38, 57). A new finding is that OVX reduced the light-dark variation in MAP and HR and abolished the circadian rhythm in BRS. Interestingly, menopausal women also exhibit reduced circadian variation in MAP, due to a smaller fall in pressure at night (dipping) (49, 51). In addition, hormone replacement therapy (HRT) or estrogen treatment has been reported to increase nocturnal dipping in postmenopausal women (9, 10, 29). Collectively, the current and previous studies suggest that gonadal steroids enhance the diurnal variation in cardiovascular regulation. This conclusion is consistent with the well-documented requirement of estrogen for the precisely timed surge of luteinizing hormone during the afternoon of proestrus, as well as for diurnal changes in locomotor activity (11, 26). For these latter effects, current evidence implicates an interaction of estrogen with the biological clock emanating from the superchiasmatic nucleus (11, 26); therefore, we speculate that the effects of OVX observed in the present study may use similar neuronal circuitry.

The primary goal of the present experiments was to determine whether MAP, HR, or BRS varies during the rat reproductive cycle. In agreement with Takezawa et al. (53), who also used continuous telemetric recordings of arterial pressure, we found that MAP decreases during the light phase of proestrus but increases during the dark proestrous phase and that these daily differences are abolished by OVX. In addition, the present results document for the first time that the gain and maximum of baroreflex control of RSNA, as well as sBRS, are increased during rat proestrus. Because methscopolamine pretreatment nearly abolished the detection of sBRS sequences, indicating that sBRS reflects largely vagal HR baroreflex control, these results also indicate that proestrus enhances baroreflex regulation of both sympathetic and parasympathetic nerves. In agreement, gains of both baroreflex control of muscle sympathetic nerve activity and cardiovagal BRS are elevated in women during the high estradiol phases of the menstrual cycle (21, 32, 46, 54).

A further finding in the present study was that the cycle variation in BRS is abolished by OVX, indicating that gonadal steroids are required. It is unlikely that progesterone is the main gonadal steroid involved, since a neurosteroid metabolite, allopregnanolone, decreases gain of baroreflex control of RSNA (18). In addition, a prominent effect of allopregnanolone is to suppress the RSNA baroreflex maximum (18), but we observed that the RSNA maximum was elevated during proestrus. Interestingly, however, progesterone demonstrates a secondary rise during metestrus (8), and sBRS reached its lowest point during the evening of this day in the cycle; thus, progesterone via its metabolite allopregnanolone may contribute to this suppression. Alternatively, since progesterone can reduce hypothalamic estrogen receptors (6), the estrogen-to-progesterone ratio may critical. However, while both estrogen and progesterone receptors are present in brain stem regions that convey baroreflex information, in these regions, the expression of estrogen receptors is not influenced by progesterone (17). In addition, significant evidence implicates estrogen. First, the estrogen profile closely mirrors the pattern of changes in sBRS during the rat estrous cycle (3, 8); a similar direct correlation between the changes in plasma estrogen and baroreflex control of HR has been observed in women during the menstrual cycle (54). Second, numerous studies have demonstrated that estrogen, via an action in the hindbrain, increases BRS (14, 23, 34, 39, 47, 48). Therefore, we suggest that during proestrus the late afternoon peak in estrogen mediates the increased BRS in the present study.

While OVX initially reduced MAP, HR, and sBRS, after a few weeks, MAP and sBRS returned toward basal values, and HR decreased further. These findings resemble the changing effects of short-term vs. longer-term gonadal steroid treatment on MAP and HR (42, 53). The changes are unlikely due to aging alone (12), since the 2–3 wk duration between samples was relatively short, and, in sham OVX animals, MAP, HR, and sBRS remained unchanged. In addition, the results are not due to recovery in estrogen levels, since measurements at the end of the study confirmed low steroid levels and uterine weights. Nevertheless, these data may explain, in part, the previously observed variability of the effects of OVX on cardiovascular regulation, since studies have been performed at differing times following OVX.

The mechanism for this time-dependent effect is unknown, though, multiple factors may be involved (for a review, see Ref. 42). The “early” changes observed around 3 wk following OVX may largely stem from estrogen loss, since El-Mas and Abdel-Rahman (15) demonstrated that the suppression of BRS observed 2–3 wk following OVX is reversed with estrogen treatment. The mechanisms producing the “late” changes about 5–6 wk following OVX may result from an interaction of steroid loss with nitric oxide (NO). Multiple studies have established that estrogen supports NO (for a review, see Refs. 22 and 55). In addition, in humans and rats, MAP and vascular resistance are elevated several weeks after bilateral oophorectomy, and these changes are reversed with estrogen replacement (19, 29). Importantly, the ability of OVX to decrease vascular conductance depends on a loss of estrogen-induced NO bioavailability (19), and this effect appears to require several weeks for maximal development (35). A slow decline in NO could also explain the delayed increases in sBRS and decreases in heart rate that we observed, since blockade of NO production induces bradycardia and increases baroreflex gain (13, 28, 37, 41).

In summary, both RSNA and cardiovagal BRS vary during the estrous cycle of the rat, with the highest levels achieved during proestrus. These cycle-induced variations, as well as diurnal variations, are dependent on gonadal steroids, since each is abolished by OVX.

Significance and Perspectives

The menopause heralds a significant rise of hypertension and its attendant cardiac and vascular disease (50). Given the documented effects of estrogen to improve vascular function and baroreflex sensitivity, it has been hypothesized that HRT would diminish cardiovascular morbidity and mortality in these women. Unexpectedly, however, in two large prospective randomized trials, female gonadosteroid hormone replacement to postmenopausal women did not reduce cardiovascular events (24, 45). A key question is, why? Data from this study suggest two clinical implications that may be significant factors in the postmenopausal development of cardiovascular disease. First, many women in these trials were started on HRT years following menopause. Our data demonstrate time-dependent cardiovascular effects following surgical menopause in the rat. Therefore, we speculate that HRT at earlier time points following menopause would produce superior effects on the blood pressure and on the cardiovascular system. Recent studies support this idea (30, 44). Second, impairment of BRS and the loss of normal autonomic-regulated circadian changes in blood pressure (dipping), are associated with adverse cardiovascular events, including sudden death (for a review, see Ref. 56). Because OVX reduced diurnal changes in MAP and HR, we speculate that menopause-induced blunting of the circadian regulation is another mechanism contributing to postmenopausal cardiovascular disease. Thus, studies of the impact of sex steroid hormones on the regulation of blood pressure and BRS are providing a deeper understanding of the underlying mechanisms affecting cardiovascular disease; the results of these studies may then form the foundation for future clinical trials.

GRANTS

This work was supported, in part, by funding from the Medical Research Foundation (to R. K. Goldman and V. L. Brooks), the Collins Foundation (to R. K. Goldman), the American Heart Association (to V. L. Brooks), and National Institutes of Health Grants HL-088552 (to V. L. Brooks) and HL-03153 (to C. Hinajosa-Laborde).