Gravity acts on the circulatory system to decrease arterial blood pressure (AP) by causing blood redistribution and reduced venous return. To evaluate roles of the baroreflex and vestibulosympathetic reflex (VSR) in maintaining AP during gravitational stress, we measured AP, heart rate (HR), and renal sympathetic nerve activity (RSNA) in four groups of conscious rats, which were either intact or had vestibular lesions (VL), sinoaortic denervation (SAD), or VL plus SAD (VL + SAD). The rats were exposed to 3 G in dorsoventral axis by centrifugation for 3 min. In rats in which neither reflex was functional (VL + SAD group), RSNA did not change, but the AP showed a significant decrease (-8 ± 1 mmHg vs. baseline). In rats with a functional baroreflex, but no VSR (VL group), the AP did not change and there was a slight increase in RSNA (25 ± 10% vs. baseline). In rats with a functional VSR, but no baroreflex (SAD group), marked increases in both AP and RSNA were observed (AP 31 ± 6 mmHg and RSNA 87 ± 10% vs. baseline), showing that the VSR causes an increase in AP in response to gravitational stress; these marked increases were significantly attenuated by the baroreflex in the intact group (AP 9 ± 2 mmHg and RSNA 38 ± 7% vs. baseline). In conclusion, AP is controlled by the combination of the baroreflex and VSR. The VSR elicits a huge pressor response during gravitational stress, preventing hypotension due to blood redistribution. In intact rats, this AP increase is compensated by the baroreflex, resulting in only a slight increase in AP.
- renal sympathetic nerve
- feedback control
- feedforward control
- predictive control
gravity has a major effect on the circulatory system. In the supine position, a steady state of blood distribution is achieved, but when the position is changed to standing, gravity causes blood redistribution toward the lower part of the body, resulting in hypotension or loss of consciousness when not counteracted by compensatory mechanisms (11, 20, 22). Furthermore, the development of modern technology has imposed another type of gravitational stress. For example, although space is the only place in which we are free of the effects of gravity, astronauts experience hypergravity of up to 3 G during launch. Even though this degree of hypergravity is within the compensation limit for healthy persons, it still poses a potential risk for space travel (15).
Several candidates have been proposed as compensatory mechanisms that maintain arterial blood pressure (AP) during gravitational stress, namely, the baroreflex, vestibulosympathetic reflex (VSR), somatosympathetic reflex, and central command (9). Of these, the baroreflex is a well-known feedback control system for AP, playing a role in maintaining AP during bleeding, postural changes, and exercise (3, 10, 25). Vestibular receptors are also considered to be involved in maintaining AP through the VSR (27). In cats, moving the head up without changing the overall body position (“nose-up” vestibular stimulation) elicits an increase in AP, which is eliminated by vestibular lesions (26). Furthermore, hypotension induced by head-up tilting is augmented by vestibular lesions, showing that vestibular input plays an important role in maintaining AP during postural change (4, 8). Accordingly, the baroreflex and VSR may cooperate to maintain AP during gravitational stress. Ray (17) showed in humans an additive effect of baroreflex and VSR on muscle sympathetic nerve activity during head-down vestibular stimulation and lower body negative pressure.
The goal of the present study was to clarify the role of the vestibular and baroreflex systems in maintaining AP during gravitational stress. For this purpose, we measured AP, heart rate (HR), and renal sympathetic nerve activity (RSNA) in conscious rats during a gravitational stress of 3 G induced by centrifugation. Experiments were performed on rats with intact reflexes (intact), bilateral vestibular lesions (VL), bilateral sinoaortic denervation (SAD), and VL plus SAD (VL + SAD).
Male Sprague-Dawley rats weighing 320-360 g (12-14 wk old) were used (n = 23) and were maintained in accordance with the “Guiding Principles for Care and Use of Animals in the Field of Physiological Science” of the Physiological Society of Japan. The experiments were approved by the Animal Research Committee of Gifu University.
Two to three weeks before the start of the experiment, surgical intervention to induce SAD was performed on the SAD and VL + SAD groups. The rats were anesthetized with pentobarbital sodium (50 mg/kg body wt ip). Under a surgical microscope, the aortic depressor nerve was isolated and dissected by a midcervical incision, and the carotid sinus was isolated from the surrounding connective tissues. After surgery, an antibiotic (cefpimizole sodium, 0.3 g/kg body wt im) was administered for 3 days to prevent infection.
One day before the experiment, another surgical intervention was performed on all four groups. The rats were anesthetized using pentobarbital sodium (50 mg/kg body wt ip); then a polyethylene catheter (PE-50, Becton Dickinson, Sparks) for AP measurement was inserted into the abdominal aorta at a level immediately below the diaphragm through the femoral artery, and another for drug administration was inserted into the inferior vena cava through the femoral vein. These catheters were exteriorized at the back of the neck. For RSNA recordings, the postganglionic renal sympathetic nerve was isolated by a right flank incision, and two stainless steel electrodes (AS633, Cooner Wire, Chatworth) were placed around it (13). The nerve and electrodes were covered and fixed with silicone gel (Semicosil 932 A and B, Wacker Chemie, Munich, Germany), and the electrodes were exteriorized at the back of the neck. To avoid accidental damage, the electrodes and catheters were covered with a flexible spring (OD 3.0 mm) that was sutured to the neck skin using a plastic connector. Due to the flexibility of the spring, rats could move almost freely inside the cage. In the VL and VL + SAD groups, vestibular lesions were produced by injection of sodium arsanilate solution into the bilateral middle ear cavities (100 mg/ml, 50 μl/ear) (5, 6). Each rat was placed in a custom-made aluminum cage (35 × 25 × 17 cm), which was placed inside the rotating box of the centrifuge immediately after surgery to allow the rats to become accustomed to the environment of the experimental room. The temperature in the experimental room was maintained at 24°C.
One day later, gravitational stress in the dorsoventral direction was applied to the rats in the prone posture by centrifugation using a custom-made gondola-type rotating box (Shimadzu, Kyoto, Japan; Fig. 1). The AP catheter was connected to a pressure transducer (MP5200, Baxter). It was vital for the transducer probe to be fixed at the same level as the catheter tip, since a difference in level between the two would result in a hydrostatic pressure difference on a change in gravity. The signal from the AP transducer was transmitted to an amplifier (AP-621G, Nihon Kohden, Tokyo). The electrodes for RSNA recording were connected to another amplifier (AP-651J, Nihon Kohden) equipped with a 50- to 2,000-Hz band-pass filter. A gravity sensor (MTS-050, Mitec, Hiroshima, Japan) was placed on the cage floor to measure gravity. All signals were recorded using a DAT data recorder (RD-145T, TEAC, Tokyo). Each experiment consisted of three 3-min periods: a control period at 1 G, a 3-G period, and a recovery period at 1 G. The centrifuge automatically controlled its speed to maintain gravity at the set point.
After completion of the experiment, SAD was confirmed by the absence of an HR response to an increase in AP caused by injection of an α1-receptor-selective agonist (phenylephrine hydrochloride, 3 μg/kg body wt iv) (23). VL was confirmed by a swimming test, in which rats were gently put into the small tub filled with warm water (5, 6). When the lesion was complete, the rat was not able to recognize the direction of the water surface and continued to turn around inside the water. In contrast, when the lesion was incomplete, the rat was able to float on the water. Although we did not confirm VL by the histological means, it is known that the injection of sodium arsanilate solution into middle ear cavity elicits a degeneration of the vestibular nerve (1).
To examine the detailed time course of the change, the data for the first minute of gravitational stress were averaged over 0-10, 10-20, 20-30, and 30-60 s after onset of the gravitational stress, and the remainder was averaged over 1 min. To analyze the time course of changes, repeated-measures two-way ANOVA was used, followed by Fisher's protected least significant difference test. For comparisons between the four groups, the differences between the levels during the first 30 s of gravitational stress and the baseline levels were calculated and compared using ANOVA followed by the Scheffé's test. The significance level was set at P < 0.05.
Table 1 shows the baseline levels for each group before application of gravitational stress. Although the AP was not significantly different between the four groups, the HR was significantly higher in the VL group than in the intact group.
Figure 2 shows typical AP, HR, RSNA, and MRSNA responses during the first 30 s of gravitational stress. A rapid increase in gravity from 1 to 3 G was induced by centrifugation. In the intact rat, the AP started to increase at the onset of gravitational stress and then remained above the baseline level, while the HR gradually decreased and RSNA increased. These responses were modified by VL and/or SAD. In the VL rat, the RSNA increase was attenuated, and the pressor and bradycardial responses were completely abolished. In contrast, in the SAD rat, the AP and RSNA increases were markedly augmented, while the HR decrease was reduced compared with in intact rats. In the VL + SAD rat, the AP decreased, while the HR and RSNA did not change.
The averaged data for the whole experimental period (1-G, 3-G, and 1-G recovery period) are presented in Fig. 3. In the intact group, the AP increased during the initial 30 s of gravitational stress and then returned to the baseline level. In the SAD group, the AP increased markedly and peaked at 20 s of gravitational stress and then gradually returned toward the baseline level. In the VL group, the AP was stable throughout gravitational stress, whereas in the VL + SAD group, it showed a significant decrease. It is interesting to note that the increase in AP in the intact and SAD groups occurred immediately after the onset of gravitational stress, whereas the AP decrease in the VL + SAD group was delayed. A marked bradycardia was seen in the intact and SAD groups, but not in the VL and VL + SAD groups, showing that it was caused by vestibular input. Bradycardia continued even after cessation of gravitational stress. RSNA increased in the intact, VL, and SAD groups but was unaffected in the VL + SAD group.
Because an increase in AP was observed only during the initial 30 s of gravitational stress in the intact group, we compared changes during this period between the four groups (Fig. 4). The increase in AP was significantly larger in the SAD group than in the intact group, showing that the baroreflex suppressed the AP increase in the intact group. Moreover, the decrease in HR seen in the intact group was augmented by SAD. The RSNA increase observed in the intact group was augmented by SAD and attenuated by VL, suggesting that the VSR augmented, and the baroreflex attenuated, RSNA in the intact group.
The present study demonstrates the importance of the vestibular and baroreflex systems in controlling AP during gravitational stress. The major findings of the present study are that 1) the AP increase observed in the intact group was completely abolished by VL and augmented by SAD; 2) although the AP did not change in the VL group, it was decreased in the VL + SAD group; 3) vestibular input was essential for eliciting bradycardia; and 4) the RSNA increase observed in the intact group was augmented by SAD, attenuated by VL, and completely abolished by VL + SAD.
There have been few reports that evaluated the vestibular effect on HR change during gravitational stress (2, 17, 28). In the present study, a marked bradycardia was observed in the intact and SAD groups but not in the VL and VL + SAD groups, indicating that this response was elicited by vestibular inputs. This finding is consistent with the report of Yates et al. (28) that gravitational stress induced by linear acceleration in humans elicits an AP increase and an HR decrease and that these responses are attenuated in the patients with vestibular dysfunction. In contrast, HR change was not reported in the experiment using caloric stimulation, suggesting that signal from otolith organs may be the afferent input eliciting this response (2).
In the present study, the roles of the baroreflex and VSR on RSNA during gravitational stress were assessed by lesion experiments. Afferent signals that can modify sympathetic nerve activity are from the eyes, baroreceptors, vestibular receptors, nonspecific receptors in the head, muscle proprioceptors, mechanoreceptors in the skin, and central command (7, 18-20, 24, 27). Because RSNA did not change in response to gravitational stress in rats lacking functional baroreceptor and vestibular receptors (i.e., the VL + SAD group), these receptors can be considered the two major components affecting RSNA under the conditions of our present study. The RSNA response observed in rats with functional vestibular receptors, but no baroreceptor activity (i.e., the SAD group), indicates that vestibular input elicited a marked RSNA increase in response to gravitational stress. In contrast, the role of the baroreflex in controlling RSNA is complex. In the intact group, gravitational stress elicited an increase in RSNA, which was significantly augmented by SAD, suggesting that the baroreflex suppressed the RSNA increase in the intact group. In the VL + SAD group, RSNA was not affected by gravitational stress, but in the VL group, i.e., rats with a functional baroreflex, but no functional vestibular receptors, RSNA increased in response to gravitational stress, suggesting that the baroreflex augmented RSNA. These opposing effects of the baroreflex on RSNA may be due to the bidirectional input to baroreceptors. The baroreceptor is loaded in rats with a functioning VSR (i.e., the intact and SAD groups) due to the pressor effect of the VSR but is unloaded in rats lacking a functioning VSR (i.e., the VL and VL + SAD groups)
As mentioned above, in the intact group, the VSR and baroreflex had opposing effects on RSNA, i.e., the VSR augmented, and the baroreflex suppressed, RSNA. This result contrasts with the findings of Ray (17), who demonstrated that the baroreflex and VSR in humans have an additive effect on muscle sympathetic nerve activity. The discrepancy can be explained if the direction of the input to the baroreceptor is considered. In the present study, we demonstrated that the baroreceptor was loaded due to the pressor response induced by the VSR, whereas in Ray's study, the baroreceptor was probably unloaded because of the lower body negative pressure. Thus the baroreflex and VSR might function cooperatively or competitively, depending on the direction of the input to the baroreceptor.
While it is well known that AP, as well as many other variables in a living body, is maintained at a constant level by the feedback control system, some variables are controlled by a feedforward control system (12, 14, 21). In the present study, we demonstrated that the vestibular system is involved in maintaining AP. If gravitational stress occurs, it is detected by the vestibular receptors, which reflexively (i.e., VSR) control the AP before it decreases due to blood redistribution. Thus the vestibular system acts as a feedforward AP controller against gravitational stress. However, the AP increase observed in the intact group indicates that the AP was overcompensated rather than compensated, and this effect was more readily seen in the SAD group lacking the baroreflex. While feedforward control has the advantage of a short response delay, the major disadvantages are the instability of the response and the overcorrect error (16). Comparison of the AP response in the intact and SAD groups showed that the overcompensated AP was compensated by the baroreflex. It is interesting to note that such overcompensation is also observed in other feedforward control systems in the living body, i.e., K+ excretion by the kidney and the body temperature control (12, 21). Thus the AP control system during gravitational stress is a combination of the vestibular feedforward system and the baroreflex feedback system (Fig. 5).
This overcompensation by the VSR feedforward control system can be quantified. Because a step input of gravity was applied to the rats, the gain (input-to-output ratio) can be calculated. The important issue is that, because the baroreflex loop was open and the VSR eliminated in the VL + SAD group (Fig. 6A), the AP response in this group is caused only by the effect of gravitational stress on AP via blood redistribution. The estimated gain between the AP change and gravitational stress due to blood redistribution (Gd) was -3.2 mmHg/G. Similarly, the gain between the AP change and gravitational stress due to the VSR (Gv) was 6.8 mmHg/G (Fig. 6B). In the case of Gd + Gv = 0, the AP decrease due to blood redistribution matches the AP increase due to VSR, resulting in no change in AP. However, because Gd + Gv > 0 in the present study, the AP increased in the SAD group.
Because only one input, i.e., the 3-G step input, was used in the present study, we could not determine whether the AP response to gravity was linear. Furthermore, in the present study, the gravitational stress was applied in the dorsoventral axis, whereas in the previous animal experiments, VSR was examined during the gravitational stress in the head-to-foot axis by head-up tilting (4, 8). Because blood pooling by gravity depends on the height difference between the hydrostatic indifferent point and the pooling site, it can be supposed that blood pooling is more prominent when the gravity acts in head-to-foot axis than in dorsoventral axis. Thus the Gd would be greater if the rats were set in head-up tilting position in the present study. Further study is necessary to elucidate the relationship between the Gd and direction of the gravitational stress.
In conclusion, the control system for protecting the AP from gravitational stress is composed of two subsystems, i.e., the baroreflex feedback control system and the vestibular feedforward control system. Vestibular inputs elicit rapid and marked sympathetic outflow before the change in AP, thus preventing the development of hypotension. Thus the AP is controlled by the VSR in a predictive manner. The overcorrect error caused by the VSR is compensated by the baroreflex.
This study was supported by a grant-in-aid for Scientific Research from the Ministry of Education, Culture, Sports, Science, and Technology of Japan, and a Ground Research for Space Utilization research grant from the National Space Development Agency of Japan and the Japan Space Forum.
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 © 2004 the American Physiological Society