Cardiovascular deconditioning occurs in astronauts after spaceflight or in individuals subjected to bed rest. It is characterized by an increased incidence of orthostatic intolerance. The mechanisms responsible for orthostatic intolerance are likely multifactorial and may include hypovolemia, autonomic dysfunction, and vascular and cardiac alterations. The arterial baroreflex is an important compensatory mechanism in the response to an orthostatic stress. In a previous study, we demonstrated that arterial baroreflex mediated sympathoexcitation was blunted in hindlimb-unloaded (HU) rats, a model of cardiovascular deconditioning. The arterial baroreflex also contributes to the regulation of vasoactive hormones including vasopressin and angiotensin II. In the present study, we tested the hypothesis that the neurohumoral response to hypotension is also attenuated in rats after 14 days of hindlimb unloading. To test this hypothesis, the vasodilator diazoxide (15 or 25 mg/kg) or saline (0.9%) was administered to produce hypotension or control conditions, respectively, in conscious HU and control rats. Plasma samples were collected and assayed for vasopressin and plasma renin activity (PRA). Diazoxide (25 mg/kg) produced significant increases in vasopressin and PRA compared with saline controls. HU rats exhibited significantly higher levels of vasopressin at rest and the increase in vasopressin levels during hypotension was enhanced by hindlimb unloading. Neither resting nor hypotension-induced PRA was altered by hindlimb unloading. These data suggest that although baroreflex-mediated sympathoexcitation is blunted by hindlimb unloading, hypotension-induced vasopressin release is enhanced and hypotension-induced PRA is unaffected. Increased circulating vasopressin may serve to compensate for blunted baroreflex regulation of sympathetic nervous activity produced by hindlimb unloading or may actually contribute to it.
- neurohumoral control
- antidiuretic hormone
- renin-angiotensin system
astronauts returning from space, or individuals recovering from prolonged bed rest often experience symptoms of orthostatic intolerance (5, 8, 16, 31, 54, 65). To delineate the possible mechanisms mediating orthostatic intolerance, individuals have been categorized as presyncopal or nonpresyncopal (17, 31, 66), finishers and nonfinishers (5), or simply, orthostatic tolerant and intolerant (54). No matter what nomenclature is used, the common characteristic among these individuals is an inability to maintain arterial pressure during an orthostatic challenge.
Compensations for orthostatic stress include baroreflex-mediated increases in heart rate in an attempt to maintain cardiac output, as well as increases in total peripheral resistance to maintain adequate arterial pressure (47). Initially, vasoconstriction is accomplished through activation of the sympathetic nervous system to maintain arterial pressure (47). In addition to the activation of the sympathetic nervous system, increases in circulating vasoconstrictor agents, such as AVP and ANG II, occur in response to an orthostatic challenge, especially when prolonged, or blood pressure begins to decrease (47, 48). Therefore, the compensatory response to an orthostatic challenge uses both neural and humorally mediated vasoconstriction as a means to maintain arterial pressure.
Although orthostatic intolerance and baroreflex dysfunction do not always occur after spaceflight (25) or bed rest (41), previous work does indicate a relationship between the presence of orthostatic intolerance in certain individuals and the attenuation of baroreflex control of sympathetic activity (12, 17, 31, 54). In addition, hindlimb unloading in rats, which simulates the effect of spaceflight or bed rest (21, 36), also produces blunted baroreflex activation of the sympathetic nervous system (14, 34). These data are consistent with findings that deficits in peripheral vasoconstriction contribute importantly to orthostatic intolerance observed in susceptible individuals after spaceflight or bed rest (5, 31, 66).
It is possible that spaceflight or bed rest blunts baroreflex control of neurohumoral factors, similar to what occurs with baroreflex control of sympathetic outflow. The purpose of the current study was to test the hypothesis that exposure to a period of simulated microgravity or bed rest results in attenuated hypotension-induced increases in circulating concentrations of vasopressin and plasma renin activity (PRA). We used the hindlimb-unloaded (HU) rat in the current experiments because it has been shown to mimic responses observed after spaceflight and bed rest (4, 21, 27, 36). To test our hypothesis, we measured plasma levels of vasopressin and PRA in response to decreases in arterial pressure in conscious rats subjected to 14 days of hindlimb unloading or control conditions.
All procedures were performed according to the guidelines stated in National Institutes of Health's Guide for the Care and Use of Laboratory Animals. All protocols were reviewed and approved by the University of Missouri–Columbia Animal Care and Use Committee. All animals received food (Formulab Diet, 0.28% sodium, Purina, St. Louis, MO,) and water ad libitum.
Male Sprague-Dawley rats (260–350 g, Harlan, Indianapolis, IN) were HU or maintained under control conditions for 14 days, according to methods described previously (34, 36, 38). Briefly, rats were initially tail suspended for short periods (1–3 h/day) over 3 days to acclimate them to hindlimb unloading conditions. The following day, all animals were anesthetized briefly (<10 min.) with 2% halothane, and a thoracic cast (Schering-Plough Animal Health, Union, NJ) was applied. Thoracic casts were applied to both groups and were designed to reduce lordosis in HU animals and prevent them from reaching the tail apparatus. Rats that underwent hindlimb unloading had a tail harness attached that consisted of a curved rigid support affixed to the tail by cloth tape and moleskin. Animals were suspended by attaching the tail harness to a swivel apparatus which allowed the animals access to food and water. Animals were suspended at an angle of ∼30–35° (34, 36) and were able to move about the cage freely without their hindlimbs making contact with the cage floor. Control animals, fitted similarly with thoracic casts, were returned to their home cages where they maintained normal cage activity. Animals that exhibited excessive weight loss (>10%) or overt signs of stress were removed from the study, similar to previous studies (34, 38) and according to the recommendation of others (36). Animals were monitored on a daily basis. Normal food and water intake, grooming, defecation, and urination were used as indications that animals were not under overt stress. On the basis of a previous study using these same criteria, there was no difference in adrenal gland weight between control and HU rats (14), which suggests animals used in the current study were not under overt chronic stress.
After 12 days of hindlimb unloading or control conditions, rats were instrumented to record mean arterial pressure (MAP) and heart rate (HR). Surgery was performed using aseptic techniques under halothane anesthesia (2%). For determination of arterial pressure a polyethylene catheter (PE; PE-50 fused to PE-10) was placed in the abdominal aorta through the left femoral artery. For drug administration, a similar catheter was placed in the femoral vein. After implantation, both catheters were tunneled subcutaneously and exteriorized at the nape of the neck. Catheters were then filled with heparinized saline (10 U/ml) and sealed with plugs. Animals were given subcutaneous fluids (10 ml saline) before being allowed to regain consciousness. Rats were then immediately returned to hindlimb unloading or control conditions to minimize weight bearing by the hindlimbs in HU rats. Animals recovered under control or HU conditions for 2 days before experimentation was begun.
After the 2-day recovery period, conscious HU or control rats were removed from their cages and weighed. Tail harnesses were also removed from HU rats. Rats were then placed in an experimental cage that contained bedding from their home cage. Animals did not have access to food or water during experimentation. The femoral arterial catheter was connected to a pressure transducer positioned at the level of the heart to record pulsatile arterial pressure. MAP was derived electronically using a low-pass filter, and heart rate was determined from the pulsatile arterial pressure signal by a cardiotachometer. Animals were studied in the horizontal position (i.e., all limbs weight bearing) to simulate a return from spaceflight or to normal upright posture after prolonged bed rest. Experiments were conducted between 1 and 3 h after removal from control or HU conditions and were performed in the same isolated, quiet room to minimize external influences on hemodynamic measurements.
Animals were randomly assigned to three different treatment groups: saline treated (control n = 9, HU n = 9); 15 mg/kg diazoxide (control n = 11, HU n = 11); and 25 mg/kg (control n = 10, HU n = 12). Hemodynamic variables were monitored during an acclimation period (∼1 h), and data were recorded for at least 30–60 min before any experimental intervention to ensure stable MAP and HR. Blood pressure was considered to be stable if there were no obvious changes for a minimum of 10 min. A minimum of 10 min of resting data was obtained before the injection of saline (0.9%) or the long-acting vasodilator diazoxide (15 or 25 mg/kg Hyperstat, Schering Plough, Kenilworth, NJ). We used diazoxide (15 mg/ml) as a hypotensive agent based on previous studies in conscious rats that reported that it produces selective arterial vasodilation (13) without affecting central venous pressure (13) or plasma osmolality (13, 50, 58). In addition, diazoxide has also been shown in conscious rats to produce hypotension-induced increases in vasopressin (50) and PRA (57, 58).
Twenty minutes after injections of diazoxide or saline, animals were anesthetized rapidly with pentobarbital sodium (50 mg/kg iv), were removed from their cage, and blood samples were obtained from trunk blood after decapitation. On the basis of previous reports, decapitation was deemed to be the most rapid and straightforward method to obtain the necessary size blood sample (4.4 ml) that was required for determination of plasma vasopressin and renin activity (19, 23, 60). Blood samples were collected in chilled centrifuge tubes containing ∼10 μl/ml EGTA (15%), sealed, and centrifuged at 5000 g for 10 min in a refrigerated centrifuge (internal temperature <4°C). Plasma samples were aliquoted into individual sample tubes and stored at −70°C until assayed for plasma vasopressin levels or PRA. All assays were performed at the University of Iowa General Clinical Research Center Analytical Laboratory on a fee basis. Plasma vasopressin concentrations were measured by specific RIA after acetone-petroleum ether extraction similar to previous studies (28). The antibody to vasopressin was a generous gift of Dr. Willis Samson, St. Louis University (St. Louis, MO) and shows <0.1% cross reactivity with oxytocin, ANG I, ANG II, CRF, and atrial natriuretic factor (49). Sensitivity of the vasopressin RIA was 0.087 pg, which corresponded to 0.170 pg/ml when factoring in the volume of plasma extracted, extraction efficiency, and volume of extract assayed. The intra- and interassay coefficients of variation averaged 11% and 17%, respectively. PRA was measured using an ANG I (125-I) RIA kit (NEN Life Science Products, PerkinElmer Life Sciences, Boston, MA) designed to measure PRA by the quantitation of generated ANG I over an incubation time of 1 h at 37°. This method has been used in previous publications (19, 60). Sensitivity of the method was 0.1 ng·ml−1·h−1, and the intra- and interassay coefficients of variation averaged 8% and 10%, respectively.
Assessment of hindlimb unloading.
Soleus and plantaris muscles were removed from the noncatheterized leg, blotted dry, and then weighed. Along with the observation of resting tachycardia in HU animals (30, 34, 38) significant decreases in hindlimb postural muscle weights and muscle weights relative to final body weight served as verification of the effectiveness of the hindlimb-unloading procedure, as described previously (34, 36, 38, 61).
Data collection and analysis.
All experimental data were obtained on a chart recorder, written to paper, and analyzed by hand. To determine group differences in baseline MAP, HR, and body and muscle weights, data were analyzed by Student's t-test. To determine the effect of diazoxide on MAP, HR, vasopressin, and PRA between groups, data were analyzed by two- or three-way ANOVA with repeated measures where appropriate. Plasma vasopressin values were square root transformed to correct for nonnormal distribution before being analyzed by ANOVA. When ANOVA indicated a significant main effect or significant interaction, post hoc Tukey tests were performed to test for pairwise comparisons or differences between individual means, according to a commercially available software package (SigmaStat, SPSS, Chicago, IL). A probability of P < 0.05 was considered statistically significant. Data are expressed as means ± SE.
Effects of hindlimb unloading.
Table 1 contains data regarding the effects of 14 days of hindlimb unloading compared with control conditions. Similar to previous reports (34, 36–39, 61), HU rats exhibited signs of hindlimb muscle atrophy, as evidenced by significantly lower soleus and plantaris muscle weights and muscle weights relative to body weight. HU rats had lower body weights than cage controls, and the percent difference in body weight between groups (14.8%) was within the range expected for HU rats of this age (<20%) (36). Hematocrits between groups were not significantly different, as reported previously for rats that had been hindlimb unloaded for 7 days (4).
Table 2 contains resting hemodynamic data for each treatment group. Similar to previous reports (34, 37, 38), HU rats from all groups exhibited resting tachycardia relative to control animals. Similarly HU rats had higher resting arterial blood pressures, as reported previously (14, 37), although this did not reach significance in the group of animals treated with 25 mg/kg diazoxide.
Figure 1 demonstrates MAP and HR responses to diazoxide (15 or 25 mg/kg) in control and HU rats. Diazoxide produced a rapid, dose-related decrease in MAP in both groups. MAP was similar across groups within 1 min of injection of either 15 mg/kg or 25 mg/kg doses of diazoxide, and this similarity persisted through the 20-min time period. During hypotension in response to 15 mg/kg diazoxide, HR significantly increased compared with baseline in both groups but remained higher in HU compared control rats throughout the protocol. Hypotension produced by 25 mg/kg diazoxide also increased HR in both groups. HR was similar between HU and control rats within 1 min of injection. Saline injections had no significant effect on MAP or HR in either group (control: ΔMAP = −1 ± 3 mmHg; ΔHR = 14 ± 3 bpm; HU: ΔMAP = 0 ± 3 mmHg; ΔHR = 8 ± 6 bpm at the 20-min time point).
Vasopressin levels in saline-treated animals were very similar to baseline vasopressin levels reported previously (46, 50, 56). Figure 2 illustrates the effects of two levels of hypotension due to diazoxide (15 or 25 mg/kg) or control conditions (saline injection) on plasma vasopressin levels in HU or control rats. ANOVA revealed a significant interaction between treatment (saline or diazoxide) and condition (control or HU). Post hoc analysis of basal vasopressin (i.e., saline-treated animals) indicated that HU rats had higher resting vasopressin levels. Vasopressin levels were not significantly altered by the lower dose of diazoxide (15 mg/kg) compared with saline for either control or HU rats. These data suggest that the hypotensive stimulus produced by the low dose was not sufficient to elicit an increase in circulating vasopressin. In contrast, vasopressin levels were significantly elevated in both groups of rats treated with the higher dose of diazoxide (25 mg/kg) compared with animals treated with saline or the lower dose of diazoxide. These data suggest that the higher dose of diazoxide produced a hypotensive stimulus large enough to elicit an increase in circulating vasopressin in both groups. Furthermore, because resting arterial pressures were not significantly different in HU and control rats given 25 mg/kg diazoxide (Table 2), both absolute levels and changes in MAP over time were similar between control and HU rats given 25 mg/kg diazoxide. The interaction between condition and treatment also suggests that differences in vasopressin levels between control and HU rats were even greater in response to hypotension (25 mg/kg diazoxide, Fig. 2) than under resting conditions. These data suggest that hindlimb unloading enhances both resting and hypotension-induced vasopressin release.
PRA in saline-treated animals was very similar to baseline PRA reported in previous studies (56–58, 60). Figure 3 demonstrates the effects of two levels of hypotension due to diazoxide (15 or 25 mg/kg) or control conditions due to saline injection on PRA in HU or control rats. Unlike vasopressin responses, PRA responses were dose dependent. For example, PRA was significantly higher in both groups of rats treated with either dose of diazoxide (15 or 25 mg/kg) compared with saline. In addition, animals from both groups treated with the higher dose of diazoxide (25 mg/kg) had higher PRAs compared with animals given the lower dose of diazoxide (15 mg/kg). Importantly, PRAs were not significantly affected by hindlimb unloading compared with controls under any condition.
The purpose of the current study was to test the hypothesis that exposure to a period of hindlimb unloading results in attenuated hypotension-induced increases in circulating concentration of vasopressin and PRA. This hypothesis was based on a previous study from our laboratory that demonstrated blunted sympathoexcitation in response to hypotension in HU rats (34). However, the results of our experiments are inconsistent with this hypothesis and suggest that hindlimb unloading enhances both resting and hypotension-induced levels of vasopressin. In contrast, neither resting nor hypotension-induced increases in PRA were affected by hindlimb unloading. Collectively, these data suggest that baroreflex control of sympathetic nervous system activity, vasopressin release, and PRA are affected differentially by hindlimb unloading. Finally, we speculate that differential sympathetic nerve and vasopressin responses may contribute importantly to the response to an orthostatic challenge in astronauts after spaceflight or individuals recovering from bed rest.
Although contrary to our hypothesis, the results from our vasopressin experiments are consistent with findings from studies in astronauts postflight (31) and individuals subjected to head-down bed rest (64). In both of these studies, vasopressin levels in response to orthostatic challenges (i.e., head-up tilt or stand test) were significantly enhanced in presyncopal astronauts or bed rest subjects. In the present study, 25 mg/kg diazoxide produced similar decreases in MAP and reduced MAP to similar levels in control and HU rats, suggesting that the hypotensive stimulus for vasopressin release was similar between groups. In addition, in a previous study, we reported that baroreceptor afferent sensitivity to hypotension was similar in HU and control animals (35). These data support the hypothesis that enhanced vasopressin release is due to centrally mediated alterations in reflex control of vasopressin release, as has been suggested previously (31). Furthermore, given the blunted sympathoexcitation reported previously in HU rats (34, 34), astronauts (17), and bed rest subjects (54), these data suggest that hindlimb unloading, spaceflight, and bed rest produce an uncoupling of the response to a hypotensive signal that initiates reflex changes in sympathetic outflow and vasopressin.
The mechanisms responsible for greater vasopressin levels in the HU rat are likely to be multifactorial as vasopressin release is regulated by numerous factors, including decreases in blood pressure or blood volume, increases in plasma osmolality, and increased ANG II (52, 53, 59). Hindlimb unloading (14, 34, 37, 38), like spaceflight and bed rest (15, 16, 31, 42, 44) does not appear to be associated with decreased resting arterial pressure. In the current study, resting arterial pressure was not lower in HU rats and PRA were similar in saline-treated HU and control rats. Thus changes in pressure or ANG II are not likely to contribute to increased vasopressin. Reduced plasma volume or an increase in plasma osmolality could contribute to enhanced hypotension-induced vasopressin release (45). There appears to be less evidence for changes in plasma osmolality after spaceflight (24) or bed rest (26), including studies in which AVP was enhanced (64). Cardiovascular deconditioning in rats (4, 10, 11, 27) and humans (6, 7, 65) is associated with reductions in plasma volume. Although hematocrits were similar between groups, we did not measure plasma volume and cannot eliminate decreased volume as a contributing factor. Future studies will be necessary to determine whether decreases in plasma volume or possible changes in plasma osmolality contribute to enhanced vasopressin release in HU rats. In addition, it is possible that increased vasopressin levels in response to hypotension are due to an enhanced central response to a similar stimulus. Whatever the initiating mechanism, higher vasopressin levels in HU rats could be mediated by greater activity of hypothalamic magnocellular neurons involved in vasopressin release, differences in vasopressin metabolism, or both.
On a functional basis, elevated vasopressin levels might be expected to help maintain arterial pressure during an orthostatic challenge via increased vasoconstriction in peripheral resistance vessels (22). It is also possible that the enhanced vasopressin release occurs as a compensatory mechanism in an attempt to counteract less activation of the sympathetic nervous system. However, several studies have now demonstrated that the direct vasoconstrictor actions of vasopressin are offset by centrally mediated reductions in sympathetic outflow (1, 3, 20). In fact, reflex sympathoexcitation in response to hypotension in conscious rabbits is reduced in a concentration-dependent manner by intravenous infusion of vasopressin (2, 40). Although the effect of vasopressin in the rat may differ from the rabbit (43, 51, 63), it is intriguing to speculate, as others have (31), that higher vasopressin levels during hypotension contribute to blunted baroreflex-mediated sympathoexcitation, which has been observed previously in HU rats (34), postflight astronauts (17, 31), and bed rest subjects (54).
Neither resting nor hypotension-induced levels of PRA were altered by hindlimb unloading. These data are consistent with previous studies examining the effect of acute (24-h) (29, 60) and chronic (7-day) hindlimb unloading in rats (62). The most straightforward explanation of these data is that hindlimb unloading has no effect on the regulation of plasma renin. However, on the basis of previous reports, there does appear to be an uncoupling of control of renin and other volume-regulating hormones, such as aldosterone in HU rats (60) and in bed rest subjects (64). We were also surprised by the lack of effect of hindlimb unloading on hypotension-induced increases in PRA, especially given our previous study, in which hypotension-induced sympathoexcitation was blunted in HU rats (34). Alterations in the regulation of the renin-angiotensin system have been reported previously after spaceflight and bed rest. In fact, most bed rest studies have noted an enhancement of PRA (18, 33, 55), especially in response to an orthostatic challenge (32, 33, 64). The reasons for the lack of difference in the present study are unknown but may be due to offsetting effects of hindlimb unloading on different aspects of the regulation of plasma renin, including renin release regulated by renal sympathetic nerves (9). For example, blunted renal sympathoexcitation produced by hindlimb unloading (34) may have offset enhanced renin release mediated by other mechanisms. Alternatively, regulation of renin activity may not be affected in HU rats but is enhanced in humans after spaceflight or bed rest.
Only the higher dose of diazoxide produced vasopressin levels that were significantly elevated in both groups compared with saline-treated animals. This was true despite the fact that the lower dose of diazoxide (15 mg/kg) produced significant and sustained hypotension (∼25 mmHg) for 20 min in both groups. It is possible that the brief period of anesthesia before obtaining blood samples attenuated hypotension-induced vasopressin release. It is also possible that obtaining our samples at a single time point reduced our chances of seeing a significant increase in vasopressin. We believe these to be unlikely, given the half-life of vasopressin (∼7 min. in rat) and the extended period of hypotension (20 min) that preceded sample collection. In addition, 15 mg/kg diazoxide increased PRA significantly, suggesting that the hypotension was sufficient to elicit a reflex increase in ANG II. We examined the 10-min time point based on preliminary studies and previous studies (50, 58), suggesting that vasopressin levels and PRA peak within 10–30 min of hypotension. Finally, it may be that there is a threshold for hypotension-induced vasopressin release similar to what has been suggested for increases in osmolality and hypovolemia (52, 53).
In summary, hindlimb unloading in rats appears to alter neurohumoral regulation of the circulation by enhancing levels of vasopressin during control or hypotensive conditions. In contrast, hindlimb unloading does not appear to alter resting or hypotension-induced increases in PRA. Interestingly, these data suggest that baroreceptor reflex control of sympathetic outflow, vasopressin, and ANG II may be uncoupled after hindlimb unloading. Finally, the increased levels of vasopressin may actually contribute to reduced baroreflex-mediated sympathoexcitation via central nervous system-mediated effects of vasopressin to decrease sympathetic outflow.
This research was supported by grants from the American Heart Association (AHA 265264Z to P. J. Mueller; AHA-97–30246 to M. J. Sullivan), the National Institutes of Health (RO1 HL-55306 to E. M. Hasser; RO1 DK-57822 to M. J. Sullivan; KO2 HL03620 to J. T. Cunningham) and the National Aeronautics and Space Administration (NAGW-4991 to E. M. Hasser). This investigation was conducted in a facility constructed with support from Research Facilities Improvement Program Grant Number C06 RR-16498 from the National Center for Research Resources, National Institutes of Health.
Present addresses: J. T. Cunningham; University of Texas Health Science Center-San Antonio, Department of Pharmacology, 7703 Floyd Curl Dr., San Antonio, TX 78229; R.R. Grindstaff, College of Health Sciences, Tennessee State University, 3500 John A. Merritt Blvd., Nashville, TN 37209.
The authors wish to thank Jodie Smith and Sarah Friskey for technical assistance and Donna Farley and the University of Iowa General Clinical Research Center Analytical Laboratory for the plasma vasopressin and renin activity analyses. We also thank members of the Neurohumoral Control of the Circulation group at the University of Missouri–Columbia for their input on this project.
↵† Deceased 21 June, 2001.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2006 the American Physiological Society