INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

STATIC (ISOMETRIC) HANDGRIP EXERCISE evokes an increase in systemic arterial blood pressure and heart rate (HR) (14, 18). Arresting the forearm blood flow just before cessation of the exercise results in the blood pressure remaining above the resting level (1, 23, 26). These results suggest that accumulation of metabolites within the muscle triggers chemosensitive afferents (group III and IV afferents) and reflexively raises arterial blood pressure (18, 25). This reflex is commonly called the muscle metaboreflex, and it is thought to have an important role in cardiovascular regulation during exercise.

In humans, the muscle metaboreflex has been shown to induce an increase in mean arterial blood pressure (MAP) and to enhance sympathetic activity to the resting muscles (18, 22, 30). It has been known for some time that the changes in sympathetic efferent nerve activity are different in different organs during a given physiological stress (6, 29); thus it would not be surprising if the muscle metaboreflex were to evoke different autonomic responses in different organs. Yamashita et al. (35) showed that activation of group III and IV afferents elicits an increase in the activity of neurons in the supraoptic nucleus in cats, suggesting that the muscle metaboreflex may stimulate the posterior pituitary gland to secrete arginine vasopressin (AVP) into the plasma. However, O'Leary et al. (24) found that activation of the muscle metaboreflex during dynamic exercise (by decreasing aortic flow to the active muscle) did not increase AVP or plasma renin activity (PRA) in dogs. In fact, only when they attenuated the pressor response during the activation of the muscle metaboreflex did AVP release occur. Vissing et al. (32) showed that an increase in plasma adrenocorticotropic hormone (ACTH) occurred on stimulation of group III and IV muscle afferents in cats, suggesting that the muscle metaboreflex may increase activity in the anterior pituitary. These data support the hypothesis that the muscle metaboreflex not only enhances sympathetic activity to the various organs but also elicits neurohumoral responses. In humans, most is known about the changes in sympathetic nervous activity to muscle during the muscle metaboreflex, with the sympathetic effects on other areas being less well studied. Furthermore, the neurohumoral effects of the muscle metaboreflex are not well understood in humans.

The purpose of this study was to determine whether activation of the muscle metaboreflex in humans does elicit neurohumoral responses and, if so, which glands and hormones are involved. To this end, we measured PRA and the plasma levels of AVP, ACTH, epinephrine (Epi), and norepinephrine (NE) during sustained activation of the muscle metaboreflex in human volunteers.

	METHODS

Top Abstract Introduction Methods Results Discussion References

We studied eight normal male volunteers whose mean age was 22 ± 1 yr. Their weight was 60 ± 2 kg, and their height 170 ± 2 cm. None of the subjects was receiving medication, and none smoked. The study was approved by the Human Subjects Committee in our institute, and each subject gave informed written consent.

Procedures. The study involved two protocols (control and occlusion protocols) (Fig. 1), with the two being performed at least 1 wk apart by a given subject. All experiments were conducted in the morning after an overnight fast. On arrival in the laboratory, the subjects underwent a 1-h seated rest period before any measurements were made. After entering the test room, in which the ambient temperature was kept at 26 ± 1°C, the subject lay down in a semisupine position. Thereafter, the subject performed handgrips at maximum voluntary contraction (MVC) by using a handgrip dynamometer, enabling 50% MVC to be determined. After this, a 22-gauge catheter was placed in a left forearm vein. While the subject rested for at least 30 min, the electrodes used to measure HR (i.e., lead II) were applied and the cuffs were fitted. Rapidly inflatable cuffs for the purpose of occlusion were fitted to both upper arms, with another cuff being placed on the left thigh for the measurement of arterial blood pressure.

View larger version (14K): [in this window] [in a new window]   Fig. 1.   Experimental protocols. HG, handgrip; RE, RO: right-hand exercise and occlusion, respectively; LE, LO: left-hand exercise and occlusion, respectively; 1, 2, 3: 1st, 2nd, and 3rd, respectively.

Measurements. During the rest, occlusion, and recovery periods, measurements of systolic (SAP) and diastolic (DAP) arterial blood pressure were taken every minute from the left thigh (which was positioned at heart level) by an oscillometric method (Dynamap-8100, Critikon). MAP was calculated according to the formula MAP = DAP + (SAP  DAP)/3. HR was averaged at 1-min intervals at rest and during occlusion and recovery periods, and for the last 10 s at 30-s intervals during exercise periods. Immediately after each exercise bout and once during each occlusion, the subject informed us of his rating of perceived exertion (RPE).

1

1

Statistics. A two-way analysis of variance for repeated measurements was used for comparison of data. Post hoc tests to determine the significance of differences between means were performed by using Tukey's test. All values are presented as means ± SE, and the null hypothesis was rejected at P < 0.05.

	RESULTS

Top Abstract Introduction Methods Results Discussion References

Figure 2A shows the level of MAP at rest, during the interval (with or without occlusion) between the 1-min bouts of isometric handgrip exercise, and during the recovery after each exercise protocol. During occlusion, MAP was significantly higher, by ~20-30 mmHg, than the resting level or the corresponding value in the control protocol. Figure 2B shows the level of HR at rest, at 30 s and 60 s into each bout of isometric handgrip exercise, during the interval (with or without occlusion) between the bouts of isometric handgrip exercise, and 3 and 5 min into the recovery after the control and occlusion protocols. HR increased by ~20-25 beats/min during each exercise bout and returned to the resting level during each interval between the bouts both with and without occlusion (Fig. 2B). There were no significant differences between the control and occlusion protocols with respect to the level of HR.

View larger version (23K): [in this window] [in a new window]   Fig. 2.   Levels of mean arterial pressure (A) and heart rate (B) at rest, during isometric handgrip exercise at 50% maximum voluntary contraction in control protocol (control) and occlusion protocol (occlusion), and in recovery period after each exercise protocol. Handgrip Ex, handgrip exercise. * Significant difference from corresponding value at rest, P < 0.05.  Significant difference from control protocol, P < 0.05.

The remaining variables were measured at rest, at 14 and 25 min into the two protocols, and after 5 min of recovery. Figure 3A shows that, during and after the control protocol, PRA was not significantly different from its resting level, indicating that several bouts of isometric handgrip exercise did not in themselves increase PRA. In the occlusion protocol, however, PRA was significantly increased, from 1.78 ± 0.3 at rest to 2.34 ± 0.3 at 14 min, 2.79 ± 0.32 at 25 min, and 2.78 ± 0.34 ng · ml¹ · h¹ during the recovery. Significant differences between the PRA values in the control and occlusion protocols were seen at 14 and 25 min and during the recovery. These differences show that the imposition of an occlusion after the isometric handgrip exercise led to an increase in PRA. Although ACTH did not increase as a result of the control protocol, it increased significantly during the occlusion protocol (Fig. 3B). Similarly, AVP did not increase during the control protocol; however, in the occlusion protocol, it was significantly raised at 25 min (Fig. 3C).

View larger version (13K): [in this window] [in a new window]   Fig. 3.   Levels of plasma renin activity (PRA; A), ACTH (B), and arginine vasopressin (AVP; C) at rest, at 14 and 25 min into each exercise protocol, and during recovery. * Significant difference from corresponding value at rest, P < 0.05.  Significant difference from control protocol, P < 0.05.

As shown in Fig. 4, during the control protocol both NE and Epi were slightly increased at 14 min, but they returned to the resting level thereafter. In the occlusion protocol, NE was increased from 266.9 ± 22.4 at rest, to 520.0 ± 40.6 at 14 min, and to 536.0 ± 39.3 pg/ml at 25 min. It then declined somewhat, to 389.3 ± 28.3 pg/ml, during the recovery. In the occlusion protocol, Epi was increased from 44.8 ± 9.0 pg/ml at rest to 134.0 ± 21.8 pg/ml at 14 min. It then declined, to 80.5 ± 12.3 pg/ml at 25 min and to 41.5 ± 8.8 pg/ml during the recovery.

View larger version (15K): [in this window] [in a new window]   Fig. 4.   Plasma levels of norepinephrine (NE; A) and epinephrine (Epi; B) at rest, at 14 and 25 min into each exercise protocol, and during recovery. * Significant difference from corresponding value at rest, P < 0.05.  Significant difference from control protocol, P < 0.05.

Table 1 shows the data for the plasma La and glucose, together with Hct, Hb, TP, and osmolality. La increased significantly in both exercise protocols. Glucose did not change during the control protocol, whereas during and after the occlusion protocol it was significantly higher than it had been at rest. Hct, Hb, and TP were all significantly increased during the occlusion protocol. Plasma osmolality was unchanged by either protocol. Plasma volume at 25 min was decreased 2.8 ± 1.0% during the control protocol and 7.4 ± 0.6% during the occlusion protocol. The subjects' RPE values were higher during the occlusion protocol (15 ± 1) than during the corresponding control protocol (11 ± 1) in the period after the second left-hand exercise (i.e., from ~9 min into the protocol until 24 min).

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

In the control protocol, subjects performed a 1-min bout of isometric handgrip exercise at 50% MVC six times with 3-min intervals of rest. PRA, AVP, and ACTH did not increase, and NE and Epi were slightly increased only at 14 min into the protocol. Nazar et al. (20) demonstrated that AVP increased, but ACTH did not increase, after 9 min of static handgrip exercise at 30% MVC, with the two hands being used alternately without a rest interval. Few et al. (7) reported that holding a 20-kg weight in the hand for 5 min evoked an elevation in ACTH. Kjær et al. (12) found that ACTH did not increase during 10-min static exercise (a 2-leg knee isometric extension exercise at 15% MVC), whereas both NE and Epi did increase. These findings support the notion that the endocrine responses to static exercise are dependent both on the active muscle mass and on the exercise duration (11, 26). It seems likely that because of the smaller muscle groups used (handgrip) and the shorter duration (1 min) of the exercise bouts (together with the 3-min rest intervals), the present exercise protocol did not result in a stimulus strong enough to elicit a significant secretion of PRA, AVP, or ACTH.

In the occlusion protocol, the MAP (measured during each occlusion period) was ~25 mmHg higher than the corresponding value in the control protocol. After such isometric exercise, occlusion is known exclusively to stimulate the so-called muscle metaboreflex. The metaboreflex is thought to be stimulated by several metabolites, with lactic acid, H⁺, and adenosine (3, 22, 30) being likely candidates. In both the control and occlusion protocols, the increase in plasma La over and above the value at rest demonstrated that the repeated 1-min bouts of isometric handgrip exercise at 50% MVC were sufficient to produce such metabolites. Previous data show that the same form of exercise decreases cellular pH in the exercising muscle from 7.2 to 6.7 units (23). When trapped by the forearm occlusion, the metabolites in the forearm could keep on activating the muscle metaboreflex, and the differences seen between the control and occlusion protocols in terms of hormone levels would represent the selective effects of the activation of the muscle metaboreflex. Because levels of PRA, AVP, ACTH, NE, and Epi were significantly higher during the occlusion protocol than during the control protocol, our data suggest that sustained muscle metaboreflex activation may lead to a significant secretion of these substances in humans.

During muscle metaboreflex activation in humans, peripheral vascular resistance and muscle sympathetic nerve activity are known to increase (22, 27, 30). In cats, renal sympathetic activity is increased by direct stimulation of muscle afferents, including group III and IV afferents, which are known to be important in the initiation of the reflex regulation of the cardiovascular system during exercise (31). Because it is known that an increase in renal sympathetic nervous activity promotes renin release (4, 13), the increase in PRA during our occlusion protocol would seem to suggest that muscle metaboreflex stimulation increases renal sympathetic activity in humans as well as in cats. However, O'Leary et al. (24) reported that, during dynamic exercise in dogs, the muscle metaboreflex response provoked by decreasing aortic flow to the active muscle did not include an increase in PRA. They suggested that the renal baroreflex and arterial baroreflex may have counteracted the effects of the muscle metaboreflex on renin secretion in their dogs, because the arterial pressure increased by 63 mmHg during the activation of the muscle metaboreflex. In the present experiment, the MAP rose by ~25 mmHg more in the occlusion protocol than in the control protocol. Thus, although it may be true that when the arterial blood pressure increases substantially, the arterial baroreflex and/or renal baroreflex may act to modulate the magnitude of the muscle metaboreflex responses (11, 13, 24), the present data clearly show that, at least in humans, the action of the metaboreflex on renin secretion can be strong enough to overcome the effects of the arterial and renal baroreflexes. Wallin et al. (33) recently showed a positive correlation between muscle sympathetic activity and renal norepinephrine spillover in humans. They also suggested that in healthy human subjects the resting sympathetic activity in the kidney is similar to, or in proportion to, that in skeletal muscle. All this is consistent with the idea that, during the activation of the muscle metaboreflex, sympathetic activity to the kidney would be enhanced in humans.

Yamashita et al. (35) showed that stimulation of group III and IV afferents from skeletal muscles caused an increase in the activity of AVP neurosecretory cells within the supraoptic nucleus in anesthetized animals. In dogs, O'Leary et al. (24) reported that the muscle metaboreflex provoked during dynamic exercise by decreasing aortic flow to the active muscle did not increase AVP. However, when the pressor response was attenuated (63 mmHg under normal conditions vs. 23 mmHg under attenuated-pressure conditions) during the activation of the muscle metaboreflex, AVP release did occur. Because in our experiment the increase in MAP was ~25 mmHg, which is similar to the change under their attenuated-pressure conditions (23 mmHg), our results are consistent with the idea that an AVP release may be seen only in association with a moderate increase in MAP during activation of the muscle metaboreflex (24).

Kjær et al. (12) carried out epidural blockade to evaluate the importance in humans of afferent nerve feedback via unmyelinated nerves, which include group III and IV afferents. Such blockade attenuated the increase in ACTH secretion otherwise seen during voluntary dynamic exercise. They therefore suggested that ACTH secretion is enhanced by the afferent feedback from working muscles. Vissing et al. (32) confirmed that stimulation of group III and IV muscle afferents in cats increases plasma ACTH. Our results are consistent with these findings and, moreover, demonstrate that a sustained muscle metaboreflex can promote the secretion of AVP and ACTH in humans. Furthermore, Vissing et al. found that stimulation of group III and IV muscle afferents in cats increased plasma glucose. Because it is known that ACTH is secreted by the pituitary gland (10) and has a role in promoting the mobilization of glucose, the increase in glucose that occurred during our occlusion protocol could also reflect an increase in ACTH secretion.

In the present experiment, AVP increased from 4.29 pg/ml at rest to 18.2 pg/ml during the occlusion protocol. Although nonhuman species (e.g., rats and dogs) show large pressor responses to an increase in AVP, human subjects are known to exhibit the smallest pressor response (4). Because it was reported that blood pressure regulation was affected by the increase in PRA (21) and that the blood pressure was increased when ACTH was infused intravenously in humans (34), it is thought that the increase in PRA and ACTH may have contributed to the increase in MAP during the occlusion in the present study. However, MAP returned to the resting level during recovery, whereas the levels of PRA and ACTH were still elevated. Thus, in the present experiment, the pressor effect secondary to the increase in AVP, PRA, and ACTH would be expected to be small.

Our subjects experienced some discomfort during occlusion, as shown by the fact that the RPE during the occlusion protocol was higher than the corresponding control value. It is well known that ACTH and AVP are secreted in association with severe mental stress or pain (9). However, sustained long-term vascular occlusion alone produced pain equivalent to that produced by exercise in the presence of occlusion, yet it did not increase MAP (15) and presumably did not increase AVP or ACTH. Furthermore, although mental stress usually induces an increase in HR (8), the HR responses were no different whether subjects were under occlusion or control conditions in the present experiment. Thus the mental stress experienced by our subjects may have been too slight to cause any significant systemic effects. However, we cannot exclude the possibility that the increase in ACTH and AVP could have been due to the discomfort or mental stress caused by the occlusion per se.

It is known that the skeletal muscles and kidneys are the main sources of plasma NE at rest (5) and that the active muscles are the main source during dynamic exercise (2). Because the elevated level of MAP, which is mainly due to the increase in total peripheral resistance evoked by the muscle metaboreflex (22), was sustained during the occlusion protocol, we can assume that sympathetic activity was enhanced to resistance vessels in the nonactive muscles and other organs (including the kidneys, as suggested by the increase in PRA). A sustained enhancement of sympathetic activity to those regions may also have caused greater NE release from these tissues and organs during the occlusion protocol and account for the elevated plasma NE seen here. Epi, on the other hand, is known to be mainly secreted by the adrenal glands in association with increased sympathetic efferent activity. During the occlusion protocol, Epi was elevated at 14 min, but it had declined toward the resting level by 25 min, a time at which NE still remained elevated. These data suggest that, during a sustained activation of the muscle metaboreflex, the pattern of change in sympathetic activity to the adrenal glands could be different from that to other regions (e.g., blood vessels and kidneys).

During the occlusion protocol, Hct, Hb, and TP were all increased. The increases could have been secondary to the sustained increase in MAP (25 mmHg) that occurred during the occlusion protocol because such an increase can shift water from intravascular to extravascular regions by Starling forces. To judge from the changes in Hct and Hb (28), there was a reduction in plasma volume of ~3% in the control protocol and of ~7% in the occlusion protocol, and the estimated reduction in total blood volume was by 2 and 4%, respectively. Although a reduction in blood volume could evoke a cardiopulmonary baroreflex, it is not certain whether the cardiopulmonary baroreflex can induce changes in AVP and PRA (16, 17, 19) and, therefore, whether this reflex could underlie the changes we observed.

In conclusion, when muscle metaboreflex activation was produced in humans by repeated forearm occlusion after isometric handgrip exercise, significant increases in PRA, AVP, ACTH, NE, and Epi occurred. These data demonstrate that such a sustained activation of the muscle metaboreflex leads to the secretion of several hormones originating from different regions.