0013-7227/91/1292-0901$03.00/0 Endocrinology Copyright (c> 1991 by The Endocrine Society
Vol. 129, No. 2
Printed in U.S.A.
Neuroendocrine Regulation of Plasma Angiotensinogen* TROY KJOS, EIJI GOTOHf, NANCY TKACS, ROY SHACKELFORD, AND WILLIAM F. GANONGJ Department of Physiology, University of California, San Francisco, California 94143-0444
ABSTRACT. In previous studies we found that plasma angiotensinogen levels were reduced by lesions of the hypothalamic paraventricular nuclei. To determine if the decrease was caused by decreased secretion of hormones that normally stimulate angiotensinogen secretion by the liver, we correlated the changes in plasma angiotensinogen produced by paraventricular lesions with changes in plasma LH, ACTH, and thyroid hormones; compared the changes in plasma angiotensinogen and other hormones to those produced by hypophysectomy; and determined the effects of treatment with ACTH and T4 in animals with paraventricular lesions. In male Sprague-Dawley rats, bilateral lesions destroying more than 50% of the paraventricular nuclei decreased plasma angiotensinogen to 787 ± 52 ng angiotensin-I/ml in 7 days compared to 1576 ± 142 ng angiotensin-I/ ml in sham-operated controls. Plasma T 3 and T4 were also
reduced, whereas there were no statistically significant changes in plasma ACTH or LH. Hypophysectomy produced a comparable decline in plasma angiotensinogen and thyroid hormone levels. Daily administration of a single dose of ACTH had no effect on plasma angiotensinogen in rats with paraventricular lesions, but T4 treatment restored plasma angiotensinogen to normal levels. The data indicate that the decline in circulating angiotensinogen produced by lesions of the paraventricular nuclei is caused by the decrease in the secretion of thyroid hormones produced by these lesions. They also demonstrate that in addition to regulating circulating renin via the sympathetic nervous system, the brain has an effect on circulating angiotensinogen via neuroendocrine control of thyroid function. {Endocrinology 129: 901-906, 1991)
I
of the decrease in plasma angiotensinogen in rats with lesions and compared it to the time course of the decrease in plasma angiotensinogen produced by hypophysectomy. Finally, we have determined the effect of replacement therapy with thyroid hormone and ACTH on plasma angiotensinogen in rats with paraventricular lesions. Parts of the results have been published in two abstracts (6, 7).
N PREVIOUS studies of the role of the hypothalamus in the regulation of renin secretion, we found that bilateral destruction of the paraventricular nuclei caused a decline in plasma angiotensinogen (1). This decline was presumably due to decreased secretion of angiotensinogen from the liver, since PRA was unaffected, indicating that consumption of renin substrate was not increased. The effect of the lesion could be initiated neurally, or it could be due to disruption of neuroendocrine function, since thyroid hormones, glucocorticoids, and estrogens all stimulate angiotensinogen secretion from the liver (2), and the paraventricular nuclei regulate thyroid hormones via TRH and TSH (3) and glucocorticoids via CRF and ACTH (4). In addition, areas near the paraventricular nuclei are involved in the regulation of estrogen secretion via GnRH and LH (5). To explore the possibility of neuroendocrine control, we have correlated changes in plasma angiotensinogen after paraventricular lesions with changes in plasma T4, T3, ACTH, and LH. In addition, we have determined the time course
Materials and Methods Male Sprague-Dawley rats, weighing approximately 250 g (Bantin, Kingman, Fremont, CA), were studied. They were provided with water and Purina rat chow ad libitum and maintained in an animal room at 20-22 C on a 14-h light, 10-h dark cycle (lights on at 0600 h). Experiments were performed in the morning unless otherwise noted. The numbers of rats in each group in each experiment are indicated in the tables. Rats were anesthetized with pentobarbital (65 mg/kg, ip, supplemented as necessary). In some of the rats an electrode was implanted stereotaxically in one paraventricular nucleus. A direct current anodal lesion was produced in the nucleus. The electrode was then removed and inserted on the other side, with production of a similar lesion. The details of the technique were previously described (1). The current was 1.0 mamp for 10 sec for each lesion. In sham-lesioned rats electrodes were implanted, but no current was passed. One, 3, 7, or 14 days later, lesioned and sham-lesioned rats were killed by decapitation, and 5 ml trunk blood were collected from each animal
Received January 14,1991. * This work was supported by USPHS Grant HL-29714, NASA Grant NAGW-1611, and the Smokeless Tobacco Research Council. t Present address: Second Department of Internal Medicine, Yokohama City University School of Medicine, 3-46 Viafunecho, Minamiku, Yokohama 232, Japan. t To whom all correspondence and requests for reprints should be addressed. 901
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 19 November 2015. at 22:25 For personal use only. No other uses without permission. . All rights reserved.
902
HYPOTHALAMUS AND PLASMA ANGIOTENSINOGEN
into a chilled tube containing 0.5 ml 0.3 M EDTA. The plasma was separated from this blood by centrifugation and stored in the frozen state until analyzed. In another experiment rats were anesthetized with xylazine (16 mg/kg) and ketamine (20 mg/kg) im, and their pituitary glands were removed via the parapharyngeal approach. Shamhypophysectomized rats were prepared in a similar fashion. The dura was opened, but the pituitary was not removed. Rats from both groups were killed by decapitation 1, 3, 7, or 14 days later. Trunk blood was collected, and plasma was separated and frozen. In a third group of rats, paraventricular lesions or sham lesions were produced, and the rats were permitted to recover for 1 week. Starting on the seventh postoperative day, half of the sham-operated and half of the lesioned groups were treated with synthetic ACTH-(l-24) (Cortrosyn, Organon, West Orange, NJ; 1.2 Mg/100 g, im) at 1400 h each day for 3 days. This dose was chosen after pilot experiments indicated that it would produce a sharp afternoon increase in plasma ACTH with a return to normal by the next morning. The other animals received saline. On the tenth postoperative day, 24 h after the last dose of ACTH or saline, the rats were killed by decapitation. Trunk blood was collected, and plasma was separated and frozen. Finally, paraventricular and sham lesions were produced, as described above, and the rats were permitted to recover for 7 days. On each of the next 5 days, half of the lesioned and half of the sham-operated rats received an injection of T4 (levothyroxine sodium, Schein; 2.0 Mg/100 g, sc). The other rats received saline. All were killed by decapitation 1 day after the last injection of T4 or saline. Trunk blood was collected, and plasma was separated and frozen. The plasma from all of the rats was analyzed for PRA, plasma renin concentration (PRC), and plasma angiotensinogen by the method of Menard and Catt (8). After the first two experiments (paraventricular lesions, hypophysectomy), the old angiotensin-I (AI) standard was replaced with a new one. Therefore, absolute angiotensinogen values were slightly lower in the ACTH and T4 replacement experiments. Actual values are presented in all four experiments, and no correction factors have been applied. However, the intraassay coefficients of variation for PRA, PRC, and angiotensinogen are 13%, 8%, and 7%, respectively, and care was taken to run all specimens from a given experiment in a single assay. Plasma ACTH was measured using commercially available RIA kits (Incstar, Stillwater, MN). For this assay, the intraand interassay coefficients of variation using rat plasma were 1.5% and 5.4%. Plasma LH was measured using a RIA, previously described (9). The reference standard was RP-2. The intra- and interassay coefficients of variation for this assay as currently performed are 8.5% and 10.1%, respectively. Plasma T4 and T 3 were measured in the hypophysectomy and initial paraventricular lesion experiments with kits purchased from Cambridge Nuclear (Billerica, MA). The kits were developed for measurement of human T4 and T3, but gave reasonable and reproducible values. However, the kits were subsequently changed and became unsuitable for measuring T 3 in rats. At that point, we changed to kits produced by Baxter (Cam-
Endo'1991 Voll29«No2
bridge, MA). These also gave low values for T3, but the Cambridge and Baxter kits gave comparable values for T4 in rats. For the Baxter kits, the intra- and interassay coefficients of variation for rat plasma T4 using rat plasma were 11.3% and 17.2%, respectively. The brains of all lesioned or sham-lesioned rats were removed and fixed in formalin. Subsequently, they were serially sectioned, stained with thionin, and examined. The base of the brain was examined in each of the hypophysectomized rats, and in only one rat was a small remnant of pituitary tissue found. However, this rat had low thyroid hormone, LH, and ACTH values, comparable to those in completely hypophysectomized rats, so this rat was included in the series. In the first experiment lesioned rats were divided into groups. Those with destruction of more than 50% of the paraventricular nuclei and those with lesions posterior to the paraventricular nuclei with no damage to these nuclei were compared to shamoperated rats. In the ACTH and T4 replacement experiments, rats with more than 50% of their paraventricular nuclei destroyed and sham-operated animals were included. For statistical analysis, undetectable assay values were called zero. In Exp 1, the data were analyzed by one-way analysis of variance, with Newman-Keuls as a post-hoc test. In the hypophysectomy experiment, statistical analysis was performed by the two-tailed t test. In both of these experiments, P < 0.05 was set as the level of statistical significance. In the ACTH and T4 experiments, statistical analysis was carried out by twotailed t test, with the level of statistical significance adjusted downward to P < 0.0125 (10).
Results Plasma angiotensinogen and other hormone values in rats at various times after lesions destroying more than 50% of the paraventricular nuclei, lesions behind the paraventricular nuclei, and sham lesions are summarized in Table 1. On the first postoperative day, the mean plasma angiotensinogen level in rats with paraventricular lesions was comparable to the level in unoperated controls. Plasma angiotensinogen was elevated at this time in rats with posterior lesions and rats with sham lesions. On the third postoperative day, plasma angiotensinogen had declined in the rats with paraventricular lesions, but since there were only two rats in this group, no statistical comparison was made to the comparable posterior lesion and sham lesion groups. By days 7 and 14, plasma angiotensinogen was approximately 50% of the values in sham-operated and posterior lesion animals. A representative paraventricular lesions is shown in Fig. 1. Damage to the nuclei was usually extensive in this group, and there were too few partial lesions to permit any conclusions about which subdivision of the nucleus was involved. PRA and PRC were variable in the lesioned animals, but no statistically significant differences were observed in the various groups. Plasma T4 and T 3 were significantly reduced in the rats with paraventricular lesions.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 19 November 2015. at 22:25 For personal use only. No other uses without permission. . All rights reserved.
HYPOTHALAMUS AND PLASMA ANGIOTENSINOGEN
903
TABLE 1. Effect of paraventricular lesions on plasma angiotensinogen and hormone levels Group
Status
Controls
n
Angiotensinogen (ng Al/ml)
PRA (ngAI/ml-2h)
PRC (ngAI/ml-2h)
T (Mg/dl)
T3 (ng/ml)
ACTH (pg/ml)
LH (ng/ml)
2
1423
24.1
139
1.8
45.5
216.6
0.30
Day 1 postop
PVL PostL Sham L
4 4 5
1517 ± 345 2242 ± 145 2161 ± 206
26.4 ± 10.1 17.5 ± 1.5 21.9 ± 4.6
132 ± 32 92 ± 5 99 ± 7
0.5 ± 0.3° 2.3 ± 0.4 2.2 ± 0.5
20.8 ± 8.8° 56.1 ± 4.6 50.7 ± 6.0
76.8 ± 14.4 0.51 ± 0.05 110.1 ± 18.6 0.32 ± 0.06 145.5 ± 33.0 0.31 ± 0.06
Day 3 postop
PVL PostL Sham L
2 8 5
1198 1964 ± 199 1705 ± 218
8.5 16.9 ± 3.4 14.6 ± 3.1
70 94 ±16 126 ± 21
1.5 2.5 ± 0.2 2.2 ± 0.4
39.7 53.0 ± 4.3 49.9 ± 5.2
136.8 177.6 ± 47.6 167.2 ± 66.8
Day 7 postop
PVL PostL Sham L
6 9 8
787 ± 52° 1533 ± 145 1576 ± 142
16.4 ± 3.3 16.0 ± 2.4 15.2 ± 3.9
208 ± 36 118 ± 20 151 ± 24
1.0 ± 0.3a 34.3 ± 1.8* 3.0 ± 0.4 50.0 ± 4.9 2.6 ± 0.1 64.2 ± 8.2
101.3 ± 21.2 0.35 ± 0.05 102.8 ± 14.4 0.19 ± 0.07* 201.5 ± 61.4 0.55 ± 0.07
Day 14 postop
PVL PostL Sham L
3 5 6
802 ± 112° 1413 ± 173 1507 ± 145
15.0 ± 5.1 15.2 ± 4.8 11.0 ± 1.9
125 ± 32 113 ± 27 84 ± 4
1.1 ± 0.2° 3.1 ± 0.3 2.9 ± 0.3
55.4 ± 2.2 0.38 ± 0.30 148.6 ± 69.1 0.43 ± 0.11 133.8 ± 38.1 0.56 ± 0.12
38.9 ± 1.7° 66.0 ± 3.3 58.3 ± 3.9
0.25 0.42 ± 0.06 0.41 ± 0.08
PVL, More than 50% of paraventricular nuclei destroyed; Post L, lesions posterior to paraventricular nuclei; sham L, sham lesion. P < 0.05 us. sham L and post L. 6 P < 0.05 vs. sham L. 0
gen was elevated in the sham-operated animals 1 day after surgery, and compared to unoperated controls, the increase was statistically significant. The increase did not occur if the pituitary was removed, and values had returned to normal in the sham-operated rats by the third postoperative day. PRA and PRC were elevated 1 and 3 days after hypophysectomy compared to those in
FIG. 1. Coronal section of the hypothalamus showing a representative bilateral paraventricular lesion.
There were no statistically significant differences in plasma LH in rats with paraventricular lesions and sham-operated animals. However, posterior lesions produced a significant reduction in plasma LH at 7 days without a significant reduction in plasma angiotensinogen. Plasma ACTH was lower in the rats with paraventricular lesions, but the SEs were large, and none of the differences was statistically significant. Thus, plasma angiotensinogen fell slowly after paraventricular lesions to about 50% of the control value on days 7 and 14 after the lesions, and this correlated with a marked reduction in plasma T4 and T3. Hypophysectomy produced a decrease in plasma angiotensinogen that paralleled the decrease produced by paraventricular lesions (Table 2). Plasma angiotensino-
sham-operated controls, possibly because of hypovolemia secondary to the temporary diabetes insipidus produced by the operation. On the seventh day, PRC was still elevated, while PRA had declined as the plasma angiotensinogen level fell, presumably decreasing All production and, consequently, the negative feedback effect of All on renin secretion. Hypophysectomy produced the expected marked declines in plasma T4, T3, and LH. The plasma ACTH level fell, but not to zero; the values after hypophysectomy probably represent the nonspecific blank of the method. The results of the experiment in which ACTH was injected into rats with paraventricular lesions are shown in Table 3. The dose of ACTH administered (1.2 /ug/100 g-day for 3 days) did not increase the low levels of plasma angiotensinogen in the rats with paraventricular lesions, and it had no effect on PRA and PRC. The adrenals of the lesioned rats were somewhat smaller than those of the sham-lesioned rats, and there was a small increase in adrenal weight in rats treated with ACTH, but none of these differences was statistically significant. Plasma ACTH was not elevated at the time of death, 24 h after the last injection of the hormone. As expected, plasma T4 values were lower in the animals with lesions, and there was no difference between ACTH- and saline-
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 19 November 2015. at 22:25 For personal use only. No other uses without permission. . All rights reserved.
HYPOTHALAMUS AND PLASMA ANGIOTENSINOGEN
904
Kndo • 1991 Vol 129-No 2
TABLE 2. Effect of hypophysectomy on plasma angiotensinogen and hormone levels Status
Group Controls
n
Angiotensinogen (ng Al/ml)
PRA (ngAI/ml-2h)
PRC (ngAI/ml-2h)
T4 (Mg/dl)
T3 (ng/ml)
ACTH (pg/ml)
LH (ng/ml)
168.2 ± 30.7
0.62 ± 0.15
10
1560 ± 91
14.8 ± 1.5
80 ± 4
3.5 ± 0.2
89.7 ± 3.4
Day 1 postop
Hyp Sham
9 10
1363 ± 94° 3646 ± 656*
61.5 ± 4.8° 25.4 ± 4.4
301 ± 25° 95 ± 6
2.0 ± 0.2° 3.0 ± 0.2
54.2 ± 3.5 54.6 ± 3.5
36.8 ± 2.4° 0.16 ± 0.02° 81.1 ± 8.3 0.28 ± 0.02
Day 3 postop
Hyp Sham
20 18
1193 ± 34° 1366 ± 47
33.5 ± 3.7° 21.8 ± 2.0
132 ± 13° 88 ± 7
0.2 ± 0.04° 2.3 ± 0.1
22.2 ± 2.5° 51.2 ± 2.7
57.5 ± 5.0° 0.01 ± 0.01° 89.3 ± 6.9 0.17 ± 0.04
Day 7 postop
Hyp Sham
8
771 ± 37° 1386 ± 78
22.2 ± 4.6 27.6 ± 5.0
252 ± 27° 140 ± 23
0.1 ± 0.02°
8
17.6 ± 3.1 1010 ± 47° 10 41.0 ± 14.0 1272 ± 90 5 Hyp, Hypophysectomy; Sham, sham hypophysectomy. ° P < 0.05 vs. sham. ° P < 0.05 vs. controls.
Day 14 postop
112 ± 14 226 ± 88
Hyp Sham
2.7 ± 0.2
8.7 ± 1.0°
47.4 ± 5.5° 122.2 ± 24.9
57.0 ± 7.2
0±0°
0±0°
3.0 ± 0.2
67.4 ± 2.6
0.07 ± 0.03° 0.24 ± 0.02
40.0 ± 2.9" 0.04 ± 0.02" 132.5 ± 16.7 0.48 ± 0.07
TABLE 3. Effect of ACTH on plasma angiotensinogen and hormone levels
Status
Treatment
n
PVL Sham L
ACTH ACTH
4
PVL Sham L
Saline Saline
5
4
4
Angiotensinogen (ng Al/ml)
PRA (ng Al/ml • 2 h)
PRC (ng Al/ml • 2 h)
486 ± 103° 906 ± 23
4.6 ± 0.5 6.1 ± 1.2
34 ±7 30 ±3
589 ± 33° 848 ± 61
4.5 ± 0.8 5.8 ± 1.7
32 ±2 35 ±4
T
Adrenal wt (mg/100 g)
ACTH (Pg/ml)
(Mg/dl)
Left
Right
1.8 ± 0.3 2.6 ± 0.3
79.3 ± 11.7 97.8 ± 9.2
6.6 ± 0.8 8.6 ± 0.4
7.1 ± 0.9 7.7 ± 0.8
1.2 ± 0.2° 2.9 ± 0.2
86.4 ± 13.7 255.6 ± 93.2
6.3 ± 0.4 8.1 ± 0.8
5.6 ± 0.7 7.8 ± 0.5
See Table 1. ACTH, Cortrosyn, 1.2 Mg/100 g, im, at 1400 h for each of 3 days, death at 1000 h on fourth day. " P < 0.0125 us. corresponding sham L.
treated animals with lesions. The results of the thyroid hormone replacement experiment in rats with paraventricular lesions are summarized in Table 4. The dose of T4 used produced plasma T4 values that were moderately but significantly greater than the values in saline-treated controls. As in the previous experiments, plasma angiotensinogen was markedly reduced in the animals with lesions. T4 replacement treatment restored plasma angiotensinogen to values that were in the normal range and almost as high as those in the T4-treated sham-lesioned animals. There were no significant changes in PRA, PRC, and ACTH.
Discussion What is the explanation for the decline in plasma angiotensinogen produced by paraventricular lesions? There is little if any evidence for direct neural control of angiotensinogen secretion. Furthermore, in experiments carried out with J. P. Porter (unpublished data), we found that electrical stimulation of the paraventricular nuclei for 15 min failed to increase plasma angiotensinogen, even though it produced a prompt increase in PRA and PRC. Therefore, a more likely explanation is that the decline in plasma angiotensinogen is neuroen-
TABLE 4. Effect of T4 on plasma angiotensinogen and hormone levels Status
Treatment
PVL Sham L
T4 T4
PVL Sham L
Saline Saline
n 11
8 7
8
Angiotensinogen (ng Al/ml)
PRA (ngAI/ml-2h)
PRC (ngAI/ml-2h)
T4 (Mg/dl)
1007 ± 63" 1231 ± 58
5.3 ± 0.8 5.3 ± 1.2
29 ± 5 28 ± 4
4.0 ± 0.4° 4.1 ± 0.3°
ACTH (pg/mO 82.8 ± 17.9 89.2 ± 14.4
552 ± 65C 927 ± 52
4.7 ± 0.9 6.0 ± 1.1
40 ± 9 31 ± 5
2.0 ± 0.3 2.6 ± 0.1
69.4 ± 7.0 63.4 ± 6.5
See Table 1. T4, 2.0 Mg/100 g, sc, for each of 5 days, death on sixth day. " P < 0.0125 vs. PVL saline. b P < 0.0125 vs. sham L saline. c P < 0.0125 vs. corresponding sham L.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 19 November 2015. at 22:25 For personal use only. No other uses without permission. . All rights reserved.
HYPOTHALAMUS AND PLASMA ANGIOTENSINOGEN docrine in origin. Evidence in favor of this conclusion is the fact that the decline developed slowly over 7 days, and that a parallel decrease was seen when the pituitary was removed. Since estrogens, adrenal corticosteroids, and thyroid hormones all increase the secretion of angiotensinogen from the liver (2), a decrease in plasma angiotensinogen might be secondary to a decrease in plasma estrogens, glucocorticoids, or thyroid hormones. A decrease in plasma estrogen is an unlikely explanation, since the relevant tropic hormone, LH, was not reduced by the paraventricular lesions. Furthermore, LH was reduced by lesions behind the paraventricular nuclei, but in these rats plasma angiotensinogen remained in the normal range. Plasma ACTH values were lower in the animals with paraventricular lesions than in the animals with posterior or sham lesions, but the SEs were large, and the differences were not statistically significant. In addition, there was a small nonsignificant decrease in adrenal weight. Previous investigators have reported that paraventricular lesions prevent the increase in ACTH secretion produced by stress, since they destroy CRF-secreting neurons, but do not lower resting ACTH values or produce adrenal atrophy (4, 11). We made no effort to avoid stress when killing the rats, and the lower ACTH values in rats with paraventricular lesions are probably explained by the blunted stress response. The absence of a statistically significant decline in plasma ACTH makes it unlikely that the decline in plasma angiotensinogen can be explained by glucocorticoid deficiency. However, plasma ACTH levels were presumably relatively constant during the day in the rats with paraventricular lesions, and it could be that incidental stress-induced increases in plasma ACTH and the diurnal increase that occurs in normal animals are necessary for normal angiotensinogen secretion. Therefore, we administered ACTH to rats with paraventricular lesions each afternoon for 3 days in a dose that produced a prompt, but transient, increase in ACTH and glucocorticoid output, then killed the rats on the fourth day. However, ACTH treatment had no effect on plasma angiotensinogen. On the other hand, there was a clear correlation between depressed thyroid function and depressed plasma angiotensinogen. Bouhnik and associates (12) found that hypothyroid rats had low plasma angiotensinogen levels that were restored to normal by administration of thyroid hormones. In our rats with paraventricular lesions and low plasma angiotensinogen levels, plasma T4 and T 3 were reduced, and replacement therapy with T4 restored plasma angiotensinogen to normal values. Thus, our data indicate that the decline in circulating angiotensinogen produced by paraventricular lesions is due to the destruction of TRH-secreting neurons, with a consequent re-
905
duction in the secretion of thyroid hormones. Thyroid hormones may also affect the renin-angiotensinogen system in other ways. Circulating angiotensinconverting enzyme activity has been reported to be increased in hyperthyroidism and decreased in hypothyroidism (13). In addition, PRC as well as plasma angiotensinogen were increased by T3 in the studies by Bouhnik et al. (12), and there are clinical reports suggesting that renin secretion is increased in hyperthyroidism (14). However, our replacement dose of T4 did not change the PRC, and it is difficult to visualize a relation between changes in angiotensin-converting enzyme activity and angiotensinogen without a change in PRC. The increase in plasma angiotensinogen 24 h after sham hypophysectomy deserves comment. This increase does not occur if the pituitary is removed. There was a similar but smaller increase 24 h after the production of posterior hypothalamic lesions or sham lesions, and the rise did not occur when the paraventricular nuclei were damaged. This suggests that some aspect of the production of the sham or posterior lesions or sham hypophysectomy produces a transient increase in angiotensinogen that is mediated by one or more of the hormones regulated by the paraventricular nuclei and secreted by the anterior pituitary. The occurrence of the increase has been confirmed on several occasions, and its cause is now under active investigation in our laboratory. It is important to remember that circulating All levels depend not only on the rate of secretion of renin, but also on the amount of angiotensinogen in the plasma. The concentration of angiotensinogen is a limiting factor for the generation of All (2), and therefore, a decline in plasma angiotensinogen results in lower plasma All values even though renin secretion is unchanged. It is well established that an important component in the regulation of renin secretion is the sympathetic nervous system, and numerous psychological and other stimuli produce sympathetically mediated increases in renin secretion (15). The present data demonstrate that plasma angiotensinogen is also under neural control via neuroendocrine pathways. Therefore, the plasma All level is affected by not one but two different neural mechanisms. References 1. Gotoh E, Murakami K, Bahnson TD, Ganong WF 1987 Role of brain serotonergic pathways and hypothalamus in regulation of renin secretion. Am J Physiol 253:R179-R185 2. Menard J, Bouhnik J, Clauser E, Richoux JP, Corvol P 1983 Biochemistry and regulation of angiotensinogen. Clin Exp Hypertension [A] 5:1005-1019 3. Lechan RM, Jackson IMD 1982 Immunohistochemical localization of thyrotropin-releasing hormone in the rat hypothalamus and pituitary. Endocrinology 111:55-65 4. Makara GB, Stark E, Kapocs G, Papocs Antoni FA 1986 Longterm, effects of hypothalamic paraventricular lesion on CRF content and stimulated ACTH secretion. Am J Physiol 250:E319E324 5. Witkin JW, Paden CM, Silverman A-J 1982 The luteinizing hor-
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 19 November 2015. at 22:25 For personal use only. No other uses without permission. . All rights reserved.
906
6. 7. 8. 9. 10. 11.
HYPOTHALAMUS AND PLASMA ANGIOTENSINOGEN mone-releasing hormone (LHRH) systems in the rat brain. Neuroendocrinology 35:429-438 Ganong WF, Gotoh E, Shackelford R, Kjos T 1989 Neuroendocrine control of renin substrate. ASGSB Bull 2:28 (Abstract) Ganong WF, Kjos T, Shackelford R, Tkacs N, Gotoh E 1990 Evidence that neural control of circulating angiotensinogen is mediated by thyroid hormones. ASGSB Bull 4:84 (Abstract) Menard J, Catt KJ 1972 Measurement of renin activity, concentration and substrate in rat plasma by radioimmunoassay of angiotensin I. Endocrinology 90:422-430 Gallo RV 1981 Pulsatile LH release during periods of low level LH secretion in the rat estrous cycle. Biol Reprod 24:771-777 Dawson-Saunders B, Trapp RG 1990 Basic and Clinical Biostatistics. Appleton and Lange, Norwalk Ganong WF 1963 The central nervous system and the synthesis
12. 13.
14. 15.
Endo-1991 Vol 129 • No 2
and release of ACTH. In: Nalbandov AV (ed) Advances in Neuroendocrinology. University of Illinois Press, Urbana, pp 92-149 Bouhnik J, Galen F-X, Clauser E, Menard J, Corvol P 1981 The renin-angiotensin system in thyroidectomized rats. Endocrinology 108:647-650 Ehlers MRW, Riordan JF 1990 Angiotensin-converting enzyme: biochemistry and molecular biology. In: Laragh JH, Brenner BM (eds) Hypertension: Pathophysiology, Diagnosis, and Management. Raven Press, New York, pp 1217-1231 Ganong WF 1982 Thyroid hormones and renin secretion. Life Sci 30:561-569 Ganong WF, Barbieri C 1982 Neuroendocrine components in the regulation of renin secretion. In: Ganong WF, Martini L (eds) Frontiers in Neuroendocrinology. Raven Press, New York, pp 231262
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 19 November 2015. at 22:25 For personal use only. No other uses without permission. . All rights reserved.
Vol. 129, No. 2
Printed in U.S.A.
Neuroendocrine Regulation of Plasma Angiotensinogen* TROY KJOS, EIJI GOTOHf, NANCY TKACS, ROY SHACKELFORD, AND WILLIAM F. GANONGJ Department of Physiology, University of California, San Francisco, California 94143-0444
ABSTRACT. In previous studies we found that plasma angiotensinogen levels were reduced by lesions of the hypothalamic paraventricular nuclei. To determine if the decrease was caused by decreased secretion of hormones that normally stimulate angiotensinogen secretion by the liver, we correlated the changes in plasma angiotensinogen produced by paraventricular lesions with changes in plasma LH, ACTH, and thyroid hormones; compared the changes in plasma angiotensinogen and other hormones to those produced by hypophysectomy; and determined the effects of treatment with ACTH and T4 in animals with paraventricular lesions. In male Sprague-Dawley rats, bilateral lesions destroying more than 50% of the paraventricular nuclei decreased plasma angiotensinogen to 787 ± 52 ng angiotensin-I/ml in 7 days compared to 1576 ± 142 ng angiotensin-I/ ml in sham-operated controls. Plasma T 3 and T4 were also
reduced, whereas there were no statistically significant changes in plasma ACTH or LH. Hypophysectomy produced a comparable decline in plasma angiotensinogen and thyroid hormone levels. Daily administration of a single dose of ACTH had no effect on plasma angiotensinogen in rats with paraventricular lesions, but T4 treatment restored plasma angiotensinogen to normal levels. The data indicate that the decline in circulating angiotensinogen produced by lesions of the paraventricular nuclei is caused by the decrease in the secretion of thyroid hormones produced by these lesions. They also demonstrate that in addition to regulating circulating renin via the sympathetic nervous system, the brain has an effect on circulating angiotensinogen via neuroendocrine control of thyroid function. {Endocrinology 129: 901-906, 1991)
I
of the decrease in plasma angiotensinogen in rats with lesions and compared it to the time course of the decrease in plasma angiotensinogen produced by hypophysectomy. Finally, we have determined the effect of replacement therapy with thyroid hormone and ACTH on plasma angiotensinogen in rats with paraventricular lesions. Parts of the results have been published in two abstracts (6, 7).
N PREVIOUS studies of the role of the hypothalamus in the regulation of renin secretion, we found that bilateral destruction of the paraventricular nuclei caused a decline in plasma angiotensinogen (1). This decline was presumably due to decreased secretion of angiotensinogen from the liver, since PRA was unaffected, indicating that consumption of renin substrate was not increased. The effect of the lesion could be initiated neurally, or it could be due to disruption of neuroendocrine function, since thyroid hormones, glucocorticoids, and estrogens all stimulate angiotensinogen secretion from the liver (2), and the paraventricular nuclei regulate thyroid hormones via TRH and TSH (3) and glucocorticoids via CRF and ACTH (4). In addition, areas near the paraventricular nuclei are involved in the regulation of estrogen secretion via GnRH and LH (5). To explore the possibility of neuroendocrine control, we have correlated changes in plasma angiotensinogen after paraventricular lesions with changes in plasma T4, T3, ACTH, and LH. In addition, we have determined the time course
Materials and Methods Male Sprague-Dawley rats, weighing approximately 250 g (Bantin, Kingman, Fremont, CA), were studied. They were provided with water and Purina rat chow ad libitum and maintained in an animal room at 20-22 C on a 14-h light, 10-h dark cycle (lights on at 0600 h). Experiments were performed in the morning unless otherwise noted. The numbers of rats in each group in each experiment are indicated in the tables. Rats were anesthetized with pentobarbital (65 mg/kg, ip, supplemented as necessary). In some of the rats an electrode was implanted stereotaxically in one paraventricular nucleus. A direct current anodal lesion was produced in the nucleus. The electrode was then removed and inserted on the other side, with production of a similar lesion. The details of the technique were previously described (1). The current was 1.0 mamp for 10 sec for each lesion. In sham-lesioned rats electrodes were implanted, but no current was passed. One, 3, 7, or 14 days later, lesioned and sham-lesioned rats were killed by decapitation, and 5 ml trunk blood were collected from each animal
Received January 14,1991. * This work was supported by USPHS Grant HL-29714, NASA Grant NAGW-1611, and the Smokeless Tobacco Research Council. t Present address: Second Department of Internal Medicine, Yokohama City University School of Medicine, 3-46 Viafunecho, Minamiku, Yokohama 232, Japan. t To whom all correspondence and requests for reprints should be addressed. 901
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 19 November 2015. at 22:25 For personal use only. No other uses without permission. . All rights reserved.
902
HYPOTHALAMUS AND PLASMA ANGIOTENSINOGEN
into a chilled tube containing 0.5 ml 0.3 M EDTA. The plasma was separated from this blood by centrifugation and stored in the frozen state until analyzed. In another experiment rats were anesthetized with xylazine (16 mg/kg) and ketamine (20 mg/kg) im, and their pituitary glands were removed via the parapharyngeal approach. Shamhypophysectomized rats were prepared in a similar fashion. The dura was opened, but the pituitary was not removed. Rats from both groups were killed by decapitation 1, 3, 7, or 14 days later. Trunk blood was collected, and plasma was separated and frozen. In a third group of rats, paraventricular lesions or sham lesions were produced, and the rats were permitted to recover for 1 week. Starting on the seventh postoperative day, half of the sham-operated and half of the lesioned groups were treated with synthetic ACTH-(l-24) (Cortrosyn, Organon, West Orange, NJ; 1.2 Mg/100 g, im) at 1400 h each day for 3 days. This dose was chosen after pilot experiments indicated that it would produce a sharp afternoon increase in plasma ACTH with a return to normal by the next morning. The other animals received saline. On the tenth postoperative day, 24 h after the last dose of ACTH or saline, the rats were killed by decapitation. Trunk blood was collected, and plasma was separated and frozen. Finally, paraventricular and sham lesions were produced, as described above, and the rats were permitted to recover for 7 days. On each of the next 5 days, half of the lesioned and half of the sham-operated rats received an injection of T4 (levothyroxine sodium, Schein; 2.0 Mg/100 g, sc). The other rats received saline. All were killed by decapitation 1 day after the last injection of T4 or saline. Trunk blood was collected, and plasma was separated and frozen. The plasma from all of the rats was analyzed for PRA, plasma renin concentration (PRC), and plasma angiotensinogen by the method of Menard and Catt (8). After the first two experiments (paraventricular lesions, hypophysectomy), the old angiotensin-I (AI) standard was replaced with a new one. Therefore, absolute angiotensinogen values were slightly lower in the ACTH and T4 replacement experiments. Actual values are presented in all four experiments, and no correction factors have been applied. However, the intraassay coefficients of variation for PRA, PRC, and angiotensinogen are 13%, 8%, and 7%, respectively, and care was taken to run all specimens from a given experiment in a single assay. Plasma ACTH was measured using commercially available RIA kits (Incstar, Stillwater, MN). For this assay, the intraand interassay coefficients of variation using rat plasma were 1.5% and 5.4%. Plasma LH was measured using a RIA, previously described (9). The reference standard was RP-2. The intra- and interassay coefficients of variation for this assay as currently performed are 8.5% and 10.1%, respectively. Plasma T4 and T 3 were measured in the hypophysectomy and initial paraventricular lesion experiments with kits purchased from Cambridge Nuclear (Billerica, MA). The kits were developed for measurement of human T4 and T3, but gave reasonable and reproducible values. However, the kits were subsequently changed and became unsuitable for measuring T 3 in rats. At that point, we changed to kits produced by Baxter (Cam-
Endo'1991 Voll29«No2
bridge, MA). These also gave low values for T3, but the Cambridge and Baxter kits gave comparable values for T4 in rats. For the Baxter kits, the intra- and interassay coefficients of variation for rat plasma T4 using rat plasma were 11.3% and 17.2%, respectively. The brains of all lesioned or sham-lesioned rats were removed and fixed in formalin. Subsequently, they were serially sectioned, stained with thionin, and examined. The base of the brain was examined in each of the hypophysectomized rats, and in only one rat was a small remnant of pituitary tissue found. However, this rat had low thyroid hormone, LH, and ACTH values, comparable to those in completely hypophysectomized rats, so this rat was included in the series. In the first experiment lesioned rats were divided into groups. Those with destruction of more than 50% of the paraventricular nuclei and those with lesions posterior to the paraventricular nuclei with no damage to these nuclei were compared to shamoperated rats. In the ACTH and T4 replacement experiments, rats with more than 50% of their paraventricular nuclei destroyed and sham-operated animals were included. For statistical analysis, undetectable assay values were called zero. In Exp 1, the data were analyzed by one-way analysis of variance, with Newman-Keuls as a post-hoc test. In the hypophysectomy experiment, statistical analysis was performed by the two-tailed t test. In both of these experiments, P < 0.05 was set as the level of statistical significance. In the ACTH and T4 experiments, statistical analysis was carried out by twotailed t test, with the level of statistical significance adjusted downward to P < 0.0125 (10).
Results Plasma angiotensinogen and other hormone values in rats at various times after lesions destroying more than 50% of the paraventricular nuclei, lesions behind the paraventricular nuclei, and sham lesions are summarized in Table 1. On the first postoperative day, the mean plasma angiotensinogen level in rats with paraventricular lesions was comparable to the level in unoperated controls. Plasma angiotensinogen was elevated at this time in rats with posterior lesions and rats with sham lesions. On the third postoperative day, plasma angiotensinogen had declined in the rats with paraventricular lesions, but since there were only two rats in this group, no statistical comparison was made to the comparable posterior lesion and sham lesion groups. By days 7 and 14, plasma angiotensinogen was approximately 50% of the values in sham-operated and posterior lesion animals. A representative paraventricular lesions is shown in Fig. 1. Damage to the nuclei was usually extensive in this group, and there were too few partial lesions to permit any conclusions about which subdivision of the nucleus was involved. PRA and PRC were variable in the lesioned animals, but no statistically significant differences were observed in the various groups. Plasma T4 and T 3 were significantly reduced in the rats with paraventricular lesions.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 19 November 2015. at 22:25 For personal use only. No other uses without permission. . All rights reserved.
HYPOTHALAMUS AND PLASMA ANGIOTENSINOGEN
903
TABLE 1. Effect of paraventricular lesions on plasma angiotensinogen and hormone levels Group
Status
Controls
n
Angiotensinogen (ng Al/ml)
PRA (ngAI/ml-2h)
PRC (ngAI/ml-2h)
T (Mg/dl)
T3 (ng/ml)
ACTH (pg/ml)
LH (ng/ml)
2
1423
24.1
139
1.8
45.5
216.6
0.30
Day 1 postop
PVL PostL Sham L
4 4 5
1517 ± 345 2242 ± 145 2161 ± 206
26.4 ± 10.1 17.5 ± 1.5 21.9 ± 4.6
132 ± 32 92 ± 5 99 ± 7
0.5 ± 0.3° 2.3 ± 0.4 2.2 ± 0.5
20.8 ± 8.8° 56.1 ± 4.6 50.7 ± 6.0
76.8 ± 14.4 0.51 ± 0.05 110.1 ± 18.6 0.32 ± 0.06 145.5 ± 33.0 0.31 ± 0.06
Day 3 postop
PVL PostL Sham L
2 8 5
1198 1964 ± 199 1705 ± 218
8.5 16.9 ± 3.4 14.6 ± 3.1
70 94 ±16 126 ± 21
1.5 2.5 ± 0.2 2.2 ± 0.4
39.7 53.0 ± 4.3 49.9 ± 5.2
136.8 177.6 ± 47.6 167.2 ± 66.8
Day 7 postop
PVL PostL Sham L
6 9 8
787 ± 52° 1533 ± 145 1576 ± 142
16.4 ± 3.3 16.0 ± 2.4 15.2 ± 3.9
208 ± 36 118 ± 20 151 ± 24
1.0 ± 0.3a 34.3 ± 1.8* 3.0 ± 0.4 50.0 ± 4.9 2.6 ± 0.1 64.2 ± 8.2
101.3 ± 21.2 0.35 ± 0.05 102.8 ± 14.4 0.19 ± 0.07* 201.5 ± 61.4 0.55 ± 0.07
Day 14 postop
PVL PostL Sham L
3 5 6
802 ± 112° 1413 ± 173 1507 ± 145
15.0 ± 5.1 15.2 ± 4.8 11.0 ± 1.9
125 ± 32 113 ± 27 84 ± 4
1.1 ± 0.2° 3.1 ± 0.3 2.9 ± 0.3
55.4 ± 2.2 0.38 ± 0.30 148.6 ± 69.1 0.43 ± 0.11 133.8 ± 38.1 0.56 ± 0.12
38.9 ± 1.7° 66.0 ± 3.3 58.3 ± 3.9
0.25 0.42 ± 0.06 0.41 ± 0.08
PVL, More than 50% of paraventricular nuclei destroyed; Post L, lesions posterior to paraventricular nuclei; sham L, sham lesion. P < 0.05 us. sham L and post L. 6 P < 0.05 vs. sham L. 0
gen was elevated in the sham-operated animals 1 day after surgery, and compared to unoperated controls, the increase was statistically significant. The increase did not occur if the pituitary was removed, and values had returned to normal in the sham-operated rats by the third postoperative day. PRA and PRC were elevated 1 and 3 days after hypophysectomy compared to those in
FIG. 1. Coronal section of the hypothalamus showing a representative bilateral paraventricular lesion.
There were no statistically significant differences in plasma LH in rats with paraventricular lesions and sham-operated animals. However, posterior lesions produced a significant reduction in plasma LH at 7 days without a significant reduction in plasma angiotensinogen. Plasma ACTH was lower in the rats with paraventricular lesions, but the SEs were large, and none of the differences was statistically significant. Thus, plasma angiotensinogen fell slowly after paraventricular lesions to about 50% of the control value on days 7 and 14 after the lesions, and this correlated with a marked reduction in plasma T4 and T3. Hypophysectomy produced a decrease in plasma angiotensinogen that paralleled the decrease produced by paraventricular lesions (Table 2). Plasma angiotensino-
sham-operated controls, possibly because of hypovolemia secondary to the temporary diabetes insipidus produced by the operation. On the seventh day, PRC was still elevated, while PRA had declined as the plasma angiotensinogen level fell, presumably decreasing All production and, consequently, the negative feedback effect of All on renin secretion. Hypophysectomy produced the expected marked declines in plasma T4, T3, and LH. The plasma ACTH level fell, but not to zero; the values after hypophysectomy probably represent the nonspecific blank of the method. The results of the experiment in which ACTH was injected into rats with paraventricular lesions are shown in Table 3. The dose of ACTH administered (1.2 /ug/100 g-day for 3 days) did not increase the low levels of plasma angiotensinogen in the rats with paraventricular lesions, and it had no effect on PRA and PRC. The adrenals of the lesioned rats were somewhat smaller than those of the sham-lesioned rats, and there was a small increase in adrenal weight in rats treated with ACTH, but none of these differences was statistically significant. Plasma ACTH was not elevated at the time of death, 24 h after the last injection of the hormone. As expected, plasma T4 values were lower in the animals with lesions, and there was no difference between ACTH- and saline-
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 19 November 2015. at 22:25 For personal use only. No other uses without permission. . All rights reserved.
HYPOTHALAMUS AND PLASMA ANGIOTENSINOGEN
904
Kndo • 1991 Vol 129-No 2
TABLE 2. Effect of hypophysectomy on plasma angiotensinogen and hormone levels Status
Group Controls
n
Angiotensinogen (ng Al/ml)
PRA (ngAI/ml-2h)
PRC (ngAI/ml-2h)
T4 (Mg/dl)
T3 (ng/ml)
ACTH (pg/ml)
LH (ng/ml)
168.2 ± 30.7
0.62 ± 0.15
10
1560 ± 91
14.8 ± 1.5
80 ± 4
3.5 ± 0.2
89.7 ± 3.4
Day 1 postop
Hyp Sham
9 10
1363 ± 94° 3646 ± 656*
61.5 ± 4.8° 25.4 ± 4.4
301 ± 25° 95 ± 6
2.0 ± 0.2° 3.0 ± 0.2
54.2 ± 3.5 54.6 ± 3.5
36.8 ± 2.4° 0.16 ± 0.02° 81.1 ± 8.3 0.28 ± 0.02
Day 3 postop
Hyp Sham
20 18
1193 ± 34° 1366 ± 47
33.5 ± 3.7° 21.8 ± 2.0
132 ± 13° 88 ± 7
0.2 ± 0.04° 2.3 ± 0.1
22.2 ± 2.5° 51.2 ± 2.7
57.5 ± 5.0° 0.01 ± 0.01° 89.3 ± 6.9 0.17 ± 0.04
Day 7 postop
Hyp Sham
8
771 ± 37° 1386 ± 78
22.2 ± 4.6 27.6 ± 5.0
252 ± 27° 140 ± 23
0.1 ± 0.02°
8
17.6 ± 3.1 1010 ± 47° 10 41.0 ± 14.0 1272 ± 90 5 Hyp, Hypophysectomy; Sham, sham hypophysectomy. ° P < 0.05 vs. sham. ° P < 0.05 vs. controls.
Day 14 postop
112 ± 14 226 ± 88
Hyp Sham
2.7 ± 0.2
8.7 ± 1.0°
47.4 ± 5.5° 122.2 ± 24.9
57.0 ± 7.2
0±0°
0±0°
3.0 ± 0.2
67.4 ± 2.6
0.07 ± 0.03° 0.24 ± 0.02
40.0 ± 2.9" 0.04 ± 0.02" 132.5 ± 16.7 0.48 ± 0.07
TABLE 3. Effect of ACTH on plasma angiotensinogen and hormone levels
Status
Treatment
n
PVL Sham L
ACTH ACTH
4
PVL Sham L
Saline Saline
5
4
4
Angiotensinogen (ng Al/ml)
PRA (ng Al/ml • 2 h)
PRC (ng Al/ml • 2 h)
486 ± 103° 906 ± 23
4.6 ± 0.5 6.1 ± 1.2
34 ±7 30 ±3
589 ± 33° 848 ± 61
4.5 ± 0.8 5.8 ± 1.7
32 ±2 35 ±4
T
Adrenal wt (mg/100 g)
ACTH (Pg/ml)
(Mg/dl)
Left
Right
1.8 ± 0.3 2.6 ± 0.3
79.3 ± 11.7 97.8 ± 9.2
6.6 ± 0.8 8.6 ± 0.4
7.1 ± 0.9 7.7 ± 0.8
1.2 ± 0.2° 2.9 ± 0.2
86.4 ± 13.7 255.6 ± 93.2
6.3 ± 0.4 8.1 ± 0.8
5.6 ± 0.7 7.8 ± 0.5
See Table 1. ACTH, Cortrosyn, 1.2 Mg/100 g, im, at 1400 h for each of 3 days, death at 1000 h on fourth day. " P < 0.0125 us. corresponding sham L.
treated animals with lesions. The results of the thyroid hormone replacement experiment in rats with paraventricular lesions are summarized in Table 4. The dose of T4 used produced plasma T4 values that were moderately but significantly greater than the values in saline-treated controls. As in the previous experiments, plasma angiotensinogen was markedly reduced in the animals with lesions. T4 replacement treatment restored plasma angiotensinogen to values that were in the normal range and almost as high as those in the T4-treated sham-lesioned animals. There were no significant changes in PRA, PRC, and ACTH.
Discussion What is the explanation for the decline in plasma angiotensinogen produced by paraventricular lesions? There is little if any evidence for direct neural control of angiotensinogen secretion. Furthermore, in experiments carried out with J. P. Porter (unpublished data), we found that electrical stimulation of the paraventricular nuclei for 15 min failed to increase plasma angiotensinogen, even though it produced a prompt increase in PRA and PRC. Therefore, a more likely explanation is that the decline in plasma angiotensinogen is neuroen-
TABLE 4. Effect of T4 on plasma angiotensinogen and hormone levels Status
Treatment
PVL Sham L
T4 T4
PVL Sham L
Saline Saline
n 11
8 7
8
Angiotensinogen (ng Al/ml)
PRA (ngAI/ml-2h)
PRC (ngAI/ml-2h)
T4 (Mg/dl)
1007 ± 63" 1231 ± 58
5.3 ± 0.8 5.3 ± 1.2
29 ± 5 28 ± 4
4.0 ± 0.4° 4.1 ± 0.3°
ACTH (pg/mO 82.8 ± 17.9 89.2 ± 14.4
552 ± 65C 927 ± 52
4.7 ± 0.9 6.0 ± 1.1
40 ± 9 31 ± 5
2.0 ± 0.3 2.6 ± 0.1
69.4 ± 7.0 63.4 ± 6.5
See Table 1. T4, 2.0 Mg/100 g, sc, for each of 5 days, death on sixth day. " P < 0.0125 vs. PVL saline. b P < 0.0125 vs. sham L saline. c P < 0.0125 vs. corresponding sham L.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 19 November 2015. at 22:25 For personal use only. No other uses without permission. . All rights reserved.
HYPOTHALAMUS AND PLASMA ANGIOTENSINOGEN docrine in origin. Evidence in favor of this conclusion is the fact that the decline developed slowly over 7 days, and that a parallel decrease was seen when the pituitary was removed. Since estrogens, adrenal corticosteroids, and thyroid hormones all increase the secretion of angiotensinogen from the liver (2), a decrease in plasma angiotensinogen might be secondary to a decrease in plasma estrogens, glucocorticoids, or thyroid hormones. A decrease in plasma estrogen is an unlikely explanation, since the relevant tropic hormone, LH, was not reduced by the paraventricular lesions. Furthermore, LH was reduced by lesions behind the paraventricular nuclei, but in these rats plasma angiotensinogen remained in the normal range. Plasma ACTH values were lower in the animals with paraventricular lesions than in the animals with posterior or sham lesions, but the SEs were large, and the differences were not statistically significant. In addition, there was a small nonsignificant decrease in adrenal weight. Previous investigators have reported that paraventricular lesions prevent the increase in ACTH secretion produced by stress, since they destroy CRF-secreting neurons, but do not lower resting ACTH values or produce adrenal atrophy (4, 11). We made no effort to avoid stress when killing the rats, and the lower ACTH values in rats with paraventricular lesions are probably explained by the blunted stress response. The absence of a statistically significant decline in plasma ACTH makes it unlikely that the decline in plasma angiotensinogen can be explained by glucocorticoid deficiency. However, plasma ACTH levels were presumably relatively constant during the day in the rats with paraventricular lesions, and it could be that incidental stress-induced increases in plasma ACTH and the diurnal increase that occurs in normal animals are necessary for normal angiotensinogen secretion. Therefore, we administered ACTH to rats with paraventricular lesions each afternoon for 3 days in a dose that produced a prompt, but transient, increase in ACTH and glucocorticoid output, then killed the rats on the fourth day. However, ACTH treatment had no effect on plasma angiotensinogen. On the other hand, there was a clear correlation between depressed thyroid function and depressed plasma angiotensinogen. Bouhnik and associates (12) found that hypothyroid rats had low plasma angiotensinogen levels that were restored to normal by administration of thyroid hormones. In our rats with paraventricular lesions and low plasma angiotensinogen levels, plasma T4 and T 3 were reduced, and replacement therapy with T4 restored plasma angiotensinogen to normal values. Thus, our data indicate that the decline in circulating angiotensinogen produced by paraventricular lesions is due to the destruction of TRH-secreting neurons, with a consequent re-
905
duction in the secretion of thyroid hormones. Thyroid hormones may also affect the renin-angiotensinogen system in other ways. Circulating angiotensinconverting enzyme activity has been reported to be increased in hyperthyroidism and decreased in hypothyroidism (13). In addition, PRC as well as plasma angiotensinogen were increased by T3 in the studies by Bouhnik et al. (12), and there are clinical reports suggesting that renin secretion is increased in hyperthyroidism (14). However, our replacement dose of T4 did not change the PRC, and it is difficult to visualize a relation between changes in angiotensin-converting enzyme activity and angiotensinogen without a change in PRC. The increase in plasma angiotensinogen 24 h after sham hypophysectomy deserves comment. This increase does not occur if the pituitary is removed. There was a similar but smaller increase 24 h after the production of posterior hypothalamic lesions or sham lesions, and the rise did not occur when the paraventricular nuclei were damaged. This suggests that some aspect of the production of the sham or posterior lesions or sham hypophysectomy produces a transient increase in angiotensinogen that is mediated by one or more of the hormones regulated by the paraventricular nuclei and secreted by the anterior pituitary. The occurrence of the increase has been confirmed on several occasions, and its cause is now under active investigation in our laboratory. It is important to remember that circulating All levels depend not only on the rate of secretion of renin, but also on the amount of angiotensinogen in the plasma. The concentration of angiotensinogen is a limiting factor for the generation of All (2), and therefore, a decline in plasma angiotensinogen results in lower plasma All values even though renin secretion is unchanged. It is well established that an important component in the regulation of renin secretion is the sympathetic nervous system, and numerous psychological and other stimuli produce sympathetically mediated increases in renin secretion (15). The present data demonstrate that plasma angiotensinogen is also under neural control via neuroendocrine pathways. Therefore, the plasma All level is affected by not one but two different neural mechanisms. References 1. Gotoh E, Murakami K, Bahnson TD, Ganong WF 1987 Role of brain serotonergic pathways and hypothalamus in regulation of renin secretion. Am J Physiol 253:R179-R185 2. Menard J, Bouhnik J, Clauser E, Richoux JP, Corvol P 1983 Biochemistry and regulation of angiotensinogen. Clin Exp Hypertension [A] 5:1005-1019 3. Lechan RM, Jackson IMD 1982 Immunohistochemical localization of thyrotropin-releasing hormone in the rat hypothalamus and pituitary. Endocrinology 111:55-65 4. Makara GB, Stark E, Kapocs G, Papocs Antoni FA 1986 Longterm, effects of hypothalamic paraventricular lesion on CRF content and stimulated ACTH secretion. Am J Physiol 250:E319E324 5. Witkin JW, Paden CM, Silverman A-J 1982 The luteinizing hor-
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 19 November 2015. at 22:25 For personal use only. No other uses without permission. . All rights reserved.
906
6. 7. 8. 9. 10. 11.
HYPOTHALAMUS AND PLASMA ANGIOTENSINOGEN mone-releasing hormone (LHRH) systems in the rat brain. Neuroendocrinology 35:429-438 Ganong WF, Gotoh E, Shackelford R, Kjos T 1989 Neuroendocrine control of renin substrate. ASGSB Bull 2:28 (Abstract) Ganong WF, Kjos T, Shackelford R, Tkacs N, Gotoh E 1990 Evidence that neural control of circulating angiotensinogen is mediated by thyroid hormones. ASGSB Bull 4:84 (Abstract) Menard J, Catt KJ 1972 Measurement of renin activity, concentration and substrate in rat plasma by radioimmunoassay of angiotensin I. Endocrinology 90:422-430 Gallo RV 1981 Pulsatile LH release during periods of low level LH secretion in the rat estrous cycle. Biol Reprod 24:771-777 Dawson-Saunders B, Trapp RG 1990 Basic and Clinical Biostatistics. Appleton and Lange, Norwalk Ganong WF 1963 The central nervous system and the synthesis
12. 13.
14. 15.
Endo-1991 Vol 129 • No 2
and release of ACTH. In: Nalbandov AV (ed) Advances in Neuroendocrinology. University of Illinois Press, Urbana, pp 92-149 Bouhnik J, Galen F-X, Clauser E, Menard J, Corvol P 1981 The renin-angiotensin system in thyroidectomized rats. Endocrinology 108:647-650 Ehlers MRW, Riordan JF 1990 Angiotensin-converting enzyme: biochemistry and molecular biology. In: Laragh JH, Brenner BM (eds) Hypertension: Pathophysiology, Diagnosis, and Management. Raven Press, New York, pp 1217-1231 Ganong WF 1982 Thyroid hormones and renin secretion. Life Sci 30:561-569 Ganong WF, Barbieri C 1982 Neuroendocrine components in the regulation of renin secretion. In: Ganong WF, Martini L (eds) Frontiers in Neuroendocrinology. Raven Press, New York, pp 231262
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 19 November 2015. at 22:25 For personal use only. No other uses without permission. . All rights reserved.
