EFFECTS OF ANGIOTENSIN II ON ARGININE-VASOPRESSIN IN PHYSIOLOGICAL AND PATHOLOGICAL SITUATIONS IN MAN P. L. PADFIELD AND J. J. MORTON Medical Research Council, Blood Pressure Unit, Western Glasgow, Gil 6NT, Scotland
(Received 29
Infirmary,
October 1976)
SUMMARY
designed to determine whether angiotensin II has a direct stimulatory effect arginine-vasopressin in man and to determine the role, if any, played by angiotensin II in the control of vasopressin release in physiological and pathological conditions. Acute infusion of angiotensin II in normal volunteers produced small but definite increases in plasma levels of arginine-vasopressin (5\m=.\4\m=+-\0\m=.\3(s.e.m.) to 6\m=.\4\m=+-\0\m=.\2pg/ml) only when plasma angiotensin II levels were supraphysiological. Concurrent measurements of plasma arginine-vasopressin and angiotensin II were made during acute changes in fluid balance and posture in normal volunteers and in clinical conditions characterized by high plasma levels of angiotensin II (Addison's disease and Bartter's syndrome). The results of these studies allow us to conclude that there is little to suggest a direct effect of angiotensin II which is likely to be relevant to the normal physiological control of arginine-vasopressin in man. Studies
were
on
INTRODUCTION
As changes in renal sodium and water excretion can be produced by both argininevasopressin (Light «fe Du Vigneaud, 1958; Berliner «fe Bennett, 1967) and angiotensin II (Brown & Peart, 1962; Del Greco, 1962; Brown, Chinn, Lever & Robertson, 1969), it would be reasonable to suspect that control of the two hormones might be linked in some way. Angiotensin II has been shown by some workers to stimulate vasopressin release in laboratory animals (Bonjour & Malvin, 1970; Mouw, Bonjour, Malvin & Vander, 1971;
man (Uhlich, Weber, Eigler & Groschel-Stewart, 1975), although this view is not supported by other studies (Claybaugh, Share «fe Shimizu, 1972; Coussineau, Gagnon «fe Sirois, 1973; Shade «fe Share, 1975). A suppressant effect of vasopressin on angiotensin II seems more clearly established both in laboratory animals (Bunag, Page «fe McCubbin, 1967; Vandogen, 1975) and in man (Khokhar, Slater, Forsling ^200 =
33
COË Ö I S 120
Posture
[-
Supine
Tilt
-*+««-
-Supine-
3 Time (h)
4
5
-
l_
12
6
Fig. 2. Effect of posture on levels of angiotensin II and vasopressin. Six normal subjects were tilted to 85° for 2 h after spending a 2 h period supine. Results are represented as the mean ± s.e.m. DISCUSSION
Bonjour
& Malvin
(1970) were the first to demonstrate a small but significant increase in antidiuretic activity in the circulation after i.v. infusion of angiotensin II (10 ng/kg/min) in conscious dogs. In attempting to repeat this work in anaesthetized dogs (Mouw et al. 1971), the same workers were unable to confirm the rise in plasma vasopressin during i.v. infusion of angiotensin II, but produced variable rises after intracarotid infusions. Other workers
have since been unable to substantiate these original findings with anaesthetized dogs (Claybaugh et al. 1972; Coussineau et al. 1973; Shade & Share, 1975), although Keil et al. (1975) were able to show a striking increase in plasma vasopressin (measured by radio¬ immunoassay) after intraventricular injections of angiotensin II in conscious rats. In man, Uhlich et al. (1975) have shown a two- to fivefold increase in plasma vasopressin (radio¬ immunoassay) 30 min after the start of an intravenous infusion of angiotensin II designed to raise systolic arterial pressure by 20 mmHg. In our experiments, i.v. infusion of angiotensin II produced a small but systematic rise in plasma vasopressin. This increase achieved statistical significance only at the highest rate of infusion, although this does not rule out the possibility that the smaller earlier increases were in fact real. A prompt fall of vasopressin to basal values was seen 1 h after stopping the infusion. The overall changes were small but the accompanying changes in urine osmolality and flow suggest biological significance. It is of course difficult to separate the antidiuretic effects of angiotensin II and vasopressin. In water-loaded normal volunteers or in patients with pituitary diabetes insipidus, the effect of angiotensin II on urine osmolality is, however, modest (Del Greco, 1962; Brown et al. 1969); values higher than plasma osmolality are rarely achieved. Our results therefore support the original view of Bonjour «fe Malvin (1970), and it is likely that failure to observe an effect of angiotensin II on vasopressin may be related to the use of anaesthetized animals as experimental models. Some forms of anaesthesia are known to increase vasopressin secretion (see Ginsburg, 1968). From our experiments large amounts of angiotensin II, resulting in supraphysiological plasma levels, are required to produce even a small rise in plasma vasopressin. It thus seems unlikely that such a regulating mechanism would operate at the physiological level in man. Another approach to the problem is to study pathological conditions in man where marked increases of plasma angiotensin II occur. Acute blood loss is a known stimulus to vasopressin release (Share, 1961) and renin secretion (Brown, Davies, Lever, Robertson & Verniory, 19666). It is possible that the marked vasopressin response seen in our patient was potentiated by angiotensin II in the blood. However there is good evidence in the dog, that antidiuretic activity increases markedly after haemorrhage when renin secretion (Claybaugh «fe Share, 1972) or the conversion of angiotensin I to angiotensin II (Morton, Lever, Ledingham, Tehrani, Stuart, Reyes «fe McGarrity, 1976) is prevented, that the increase in vasopressin can precede the rise in plasma renin activity (Claybaugh & Share, 1972) and that the rise of vasopressin after haemorrhage in nephrectomized dogs is not increased by angiotensin infusion (Shade & Share, 1975). It seems more likely, therefore, that a fall in blood volume (Share, 1961) and/or blood pressure is the main stimulus to vaso¬ pressin release after haemorrhage. There is some debate as to whether patients or animals with adrenal insufficiency have increased levels of vasopressin (see Share «fe Travis, 1970) and we know of no data on vaso¬ pressin levels in Bartter's syndrome. Despite the high levels of angiotensin II in the three patients with this syndrome, vasopressin levels were not high. Conversely, at a time when plasma osmolality and serum sodium were low, both angiotensin II and vasopressin were high in the patient with Addison's disease. The fall of vasopressin without a fall of angio¬ tensin II after water loading in this patient, suggests that mechanisms other than a stimu¬ latory effect of angiotensin II were responsible for the increased level of vasopressin. In normal man, the assumption of the erect posture is accompanied by a rise in plasma renin concentration (Brown, Davies, Lever, McPherson & Robertson, 1966 a). Evidence for a rise in vasopressin in these circumstances is conflicting (see Kimura, Minai, Matsui, Mouri, Sato, Yoshinaga «fe Hoshi, 1976). In our subjects studied here and previously (Morton et al. 1975), a tilt to 85° resulted in definite increases in angiotensin II in all subjects
but the small increases seen in plasma vasopressin were statistically insignificant. It is possible that our inability to see a more definite rise was related to the mild fluid-load conditions of the experiment (Robertson, 1974). Even if vasopressin secretion is increased on standing, it is unlikely that the relatively low levels of angiotensin II could be stimulatory. The known fall in plasma volume which occurs on standing (Fawcett & Wynn, 1960) could stimulate the release of both peptides. Finally, the increase of vasopressin after fluid deprivation in normal volunteers is not paralleled by a consistent increase in angiotensin II, again suggesting that the rise in vasopressin is not contributed to by a stimulatory effect of angiotensin II. In conclusion, therefore, while we have shown that extremely high levels of plasma angiotensin II produced by acute infusion of this peptide do have a stimulatory effect on arginine-vasopressin, we have no evidence to suggest that such a mechanism of control of arginine-vasopressin is operative at the physiological level in man. REFERENCES
Bartter, F. C. (1962). Hyperplasia of the juxtaglomerular complex with hyperaldosteronism and hypokalaemia. A new syndrome. American Journal of Medicine 33, 811-817. Berliner, R. W. & Bennett, C. M. (1967). Concentration of urine in the mammalian kidney. American Journal of Medicine 42, 777-789. Bonjour, J. P. & Malvin, R. L. (1970). Stimulation of ADH release by the renin-angiotensin system. American Journal of Physiology 218, 1555-1559. Brown, J. J., Chinn, R. H., Lever, A. F. & Robertson, J. I. S. (1969). Renin and angiotensin as a mechanism of diuretic-induced antidiuresis in diabetes insipidus. Lancet ii, 237-239. Brown, J. J., Davies, D. L., Lever, A. F., McPherson, D.
(Received 29
Infirmary,
October 1976)
SUMMARY
designed to determine whether angiotensin II has a direct stimulatory effect arginine-vasopressin in man and to determine the role, if any, played by angiotensin II in the control of vasopressin release in physiological and pathological conditions. Acute infusion of angiotensin II in normal volunteers produced small but definite increases in plasma levels of arginine-vasopressin (5\m=.\4\m=+-\0\m=.\3(s.e.m.) to 6\m=.\4\m=+-\0\m=.\2pg/ml) only when plasma angiotensin II levels were supraphysiological. Concurrent measurements of plasma arginine-vasopressin and angiotensin II were made during acute changes in fluid balance and posture in normal volunteers and in clinical conditions characterized by high plasma levels of angiotensin II (Addison's disease and Bartter's syndrome). The results of these studies allow us to conclude that there is little to suggest a direct effect of angiotensin II which is likely to be relevant to the normal physiological control of arginine-vasopressin in man. Studies
were
on
INTRODUCTION
As changes in renal sodium and water excretion can be produced by both argininevasopressin (Light «fe Du Vigneaud, 1958; Berliner «fe Bennett, 1967) and angiotensin II (Brown & Peart, 1962; Del Greco, 1962; Brown, Chinn, Lever & Robertson, 1969), it would be reasonable to suspect that control of the two hormones might be linked in some way. Angiotensin II has been shown by some workers to stimulate vasopressin release in laboratory animals (Bonjour & Malvin, 1970; Mouw, Bonjour, Malvin & Vander, 1971;
man (Uhlich, Weber, Eigler & Groschel-Stewart, 1975), although this view is not supported by other studies (Claybaugh, Share «fe Shimizu, 1972; Coussineau, Gagnon «fe Sirois, 1973; Shade «fe Share, 1975). A suppressant effect of vasopressin on angiotensin II seems more clearly established both in laboratory animals (Bunag, Page «fe McCubbin, 1967; Vandogen, 1975) and in man (Khokhar, Slater, Forsling ^200 =
33
COË Ö I S 120
Posture
[-
Supine
Tilt
-*+««-
-Supine-
3 Time (h)
4
5
-
l_
12
6
Fig. 2. Effect of posture on levels of angiotensin II and vasopressin. Six normal subjects were tilted to 85° for 2 h after spending a 2 h period supine. Results are represented as the mean ± s.e.m. DISCUSSION
Bonjour
& Malvin
(1970) were the first to demonstrate a small but significant increase in antidiuretic activity in the circulation after i.v. infusion of angiotensin II (10 ng/kg/min) in conscious dogs. In attempting to repeat this work in anaesthetized dogs (Mouw et al. 1971), the same workers were unable to confirm the rise in plasma vasopressin during i.v. infusion of angiotensin II, but produced variable rises after intracarotid infusions. Other workers
have since been unable to substantiate these original findings with anaesthetized dogs (Claybaugh et al. 1972; Coussineau et al. 1973; Shade & Share, 1975), although Keil et al. (1975) were able to show a striking increase in plasma vasopressin (measured by radio¬ immunoassay) after intraventricular injections of angiotensin II in conscious rats. In man, Uhlich et al. (1975) have shown a two- to fivefold increase in plasma vasopressin (radio¬ immunoassay) 30 min after the start of an intravenous infusion of angiotensin II designed to raise systolic arterial pressure by 20 mmHg. In our experiments, i.v. infusion of angiotensin II produced a small but systematic rise in plasma vasopressin. This increase achieved statistical significance only at the highest rate of infusion, although this does not rule out the possibility that the smaller earlier increases were in fact real. A prompt fall of vasopressin to basal values was seen 1 h after stopping the infusion. The overall changes were small but the accompanying changes in urine osmolality and flow suggest biological significance. It is of course difficult to separate the antidiuretic effects of angiotensin II and vasopressin. In water-loaded normal volunteers or in patients with pituitary diabetes insipidus, the effect of angiotensin II on urine osmolality is, however, modest (Del Greco, 1962; Brown et al. 1969); values higher than plasma osmolality are rarely achieved. Our results therefore support the original view of Bonjour «fe Malvin (1970), and it is likely that failure to observe an effect of angiotensin II on vasopressin may be related to the use of anaesthetized animals as experimental models. Some forms of anaesthesia are known to increase vasopressin secretion (see Ginsburg, 1968). From our experiments large amounts of angiotensin II, resulting in supraphysiological plasma levels, are required to produce even a small rise in plasma vasopressin. It thus seems unlikely that such a regulating mechanism would operate at the physiological level in man. Another approach to the problem is to study pathological conditions in man where marked increases of plasma angiotensin II occur. Acute blood loss is a known stimulus to vasopressin release (Share, 1961) and renin secretion (Brown, Davies, Lever, Robertson & Verniory, 19666). It is possible that the marked vasopressin response seen in our patient was potentiated by angiotensin II in the blood. However there is good evidence in the dog, that antidiuretic activity increases markedly after haemorrhage when renin secretion (Claybaugh «fe Share, 1972) or the conversion of angiotensin I to angiotensin II (Morton, Lever, Ledingham, Tehrani, Stuart, Reyes «fe McGarrity, 1976) is prevented, that the increase in vasopressin can precede the rise in plasma renin activity (Claybaugh & Share, 1972) and that the rise of vasopressin after haemorrhage in nephrectomized dogs is not increased by angiotensin infusion (Shade & Share, 1975). It seems more likely, therefore, that a fall in blood volume (Share, 1961) and/or blood pressure is the main stimulus to vaso¬ pressin release after haemorrhage. There is some debate as to whether patients or animals with adrenal insufficiency have increased levels of vasopressin (see Share «fe Travis, 1970) and we know of no data on vaso¬ pressin levels in Bartter's syndrome. Despite the high levels of angiotensin II in the three patients with this syndrome, vasopressin levels were not high. Conversely, at a time when plasma osmolality and serum sodium were low, both angiotensin II and vasopressin were high in the patient with Addison's disease. The fall of vasopressin without a fall of angio¬ tensin II after water loading in this patient, suggests that mechanisms other than a stimu¬ latory effect of angiotensin II were responsible for the increased level of vasopressin. In normal man, the assumption of the erect posture is accompanied by a rise in plasma renin concentration (Brown, Davies, Lever, McPherson & Robertson, 1966 a). Evidence for a rise in vasopressin in these circumstances is conflicting (see Kimura, Minai, Matsui, Mouri, Sato, Yoshinaga «fe Hoshi, 1976). In our subjects studied here and previously (Morton et al. 1975), a tilt to 85° resulted in definite increases in angiotensin II in all subjects
but the small increases seen in plasma vasopressin were statistically insignificant. It is possible that our inability to see a more definite rise was related to the mild fluid-load conditions of the experiment (Robertson, 1974). Even if vasopressin secretion is increased on standing, it is unlikely that the relatively low levels of angiotensin II could be stimulatory. The known fall in plasma volume which occurs on standing (Fawcett & Wynn, 1960) could stimulate the release of both peptides. Finally, the increase of vasopressin after fluid deprivation in normal volunteers is not paralleled by a consistent increase in angiotensin II, again suggesting that the rise in vasopressin is not contributed to by a stimulatory effect of angiotensin II. In conclusion, therefore, while we have shown that extremely high levels of plasma angiotensin II produced by acute infusion of this peptide do have a stimulatory effect on arginine-vasopressin, we have no evidence to suggest that such a mechanism of control of arginine-vasopressin is operative at the physiological level in man. REFERENCES
Bartter, F. C. (1962). Hyperplasia of the juxtaglomerular complex with hyperaldosteronism and hypokalaemia. A new syndrome. American Journal of Medicine 33, 811-817. Berliner, R. W. & Bennett, C. M. (1967). Concentration of urine in the mammalian kidney. American Journal of Medicine 42, 777-789. Bonjour, J. P. & Malvin, R. L. (1970). Stimulation of ADH release by the renin-angiotensin system. American Journal of Physiology 218, 1555-1559. Brown, J. J., Chinn, R. H., Lever, A. F. & Robertson, J. I. S. (1969). Renin and angiotensin as a mechanism of diuretic-induced antidiuresis in diabetes insipidus. Lancet ii, 237-239. Brown, J. J., Davies, D. L., Lever, A. F., McPherson, D.
