| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Other Original Articles |
Third Department of Internal Medicine, Kyorin University School of Medicine, Mitaka, Tokyo 181-8611, Japan
Address all correspondence and requests for reprints to: Eiji Itagaki, M.D., Ph.D., Third Department of Internal Medicine, Kyorin University School of Medicine, 6-20-2 Shinkawa, Mitaka, Tokyo 181-8611, Japan. E-mail: ita{at}kyorin-u.ac.jp
Abstract
To clarify the mechanism for the potentiation of CRH-induced ACTH response by the infusion of hypertonic saline, we investigated changes in plasma ACTH concentration after infusion of 5% hypertonic saline in five patients with untreated central diabetes insipidus (DI). Basal levels of plasma ACTH and cortisol in the DI group were not significantly different from those in normal control subjects. The infusion of hypertonic saline produced an increase in plasma arginine vasopressin (AVP) in controls, but did not elevate ACTH. However, in patients with DI, the plasma AVP concentration did not change, but circulating ACTH increased 3.6-fold (7.7 ± 1.5 to 23.0 ± 2.7 pmol/liter; P < 0.01), and plasma cortisol also increased significantly (298 ± 99 to 538 ± 124 nmol/liter; P < 0.05). Moreover, a positive correlation was observed between plasma ACTH and osmolality (r = 0.72; P < 0.005). These results indicate that ACTH secretion in DI patients is regulated by a mechanism distinct from that in healthy subjects. It seems possible that the increase in plasma osmolality promotes ACTH secretion in DI patients through AVP and/or urocortin via the hypophyseal portal system, independent of the AVP secretion from magnocellular neurons.
THE SECRETION OF ACTH is primarily regulated by the release of CRH from the hypothalamus. However, it has been suggested that arginine vasopressin (AVP) acts in concert with CRH to promote ACTH secretion (1, 2). It has recently been reported that water deprivation (3) or infusion of hypertonic saline (4, 5) in healthy subjects enhances the ACTH response to CRH. As an increase in plasma AVP occurs in parallel with the ACTH response in such cases, it can be speculated that endogenous AVP, the secretion of which has been promoted by elevated osmolality, may augment CRH-induced ACTH secretion. Alternatively, it is also possible that the ACTH response-enhancing mechanism through hypertonic saline infusion may not involve AVP, because this phenomenon is similarly seen in patients with central diabetes insipidus (DI), in whom AVP secretion is attenuated (6). Furthermore, the ACTH response in DI patients to CRH alone or combined hypertonic saline and CRH was recently reported to be greater than that in healthy subjects (6, 7). Hence, it can be hypothesized that ACTH secretion in DI patients may be regulated via a different mechanism than that in healthy controls, and that an elevation of plasma osmolality may promote ACTH secretion without the involvement of AVP. However, this mechanism has yet to be elucidated in detail, and this osmotic effect itself on ACTH secretion remains to be investigated in DI patients.
Therefore, we investigated the effect of increased plasma osmolality on ACTH secretion in DI patients and normal subjects after infusion of 5% hypertonic saline. The present study showed that ACTH secretion was significantly increased by the elevation of plasma osmolality in DI patients. As this ACTH response was not associated with an increase in plasma AVP, it can be concluded that it is at the very least unrelated to the secretion of AVP from magnocellular neurons.
Materials and Methods
Subjects and protocol
The study was performed with five untreated central DI patients
(four with idiopathic DI and one with infundibulo-neurohypophysitis;
Table 1
). The patients were two men and
three women (mean age, 53.6 yr), and duration of DI ranged from 4–11
months. All patients were admitted to our hospital because of
polydipsia and polyuria. Urinary osmolality in the morning for these
patients ranged from 83–185 mmol/kg, and it was lower than plasma
osmolality in every patient. After 6.5 h of fluid deprivation,
urinary osmolality for these patients ranged from 117–464 mmol/kg, and
the ratio of urinary osmolality to plasma osmolality ranged from
0.38–1.56. Hence, with the administration of 5 U aqueous vasopressin,
urinary osmolality increased from 24–120% in these patients. Based on
these findings, two patients were diagnosed as complete DI (patients 3
and 4) and three as partial DI (patients 1, 2 and 5) (8).
Magnetic resonance imaging scans showed an absence of hyperintense
signals in the neurohypophysis in all cases, and enlargement of the
pituitary stalk was seen in one subject (patient 2). In another subject
(patient 5), a 5-mm contrast-poor mass was seen in the pituitary stalk,
but the relationship with DI was unknown. Anterior pituitary function,
as evaluated by the four-hormone loading test (CRH, TRH, GHRH, and
LH-releasing hormone), was maintained in all subjects.
|
Hypertonic saline (HS) was administered according to the following protocol. After 20 min of bed rest, 5% saline was given iv for 10 min at a rate of 0.24 ml/kg·min. Blood samples were collected before infusion and at 10 and 20 min after starting infusion. Tests were commenced at 0900 h, and food and drink were freely allowed until immediately before the tests. In both groups some subjects complained of mild thirst during the test, but no other adverse reactions associated with the HS infusion were observed, and every subject completed the test without any problem.
Hormone and other assays
All hormone assays were performed by RIA techniques using commercially available kits [ACTH, Allegro ACTH Kit (Nichols Institute Diagnostics, San Juan Capistrano, CA); AVP, AVP RIA Mitsubishi (Mitsubishi Chemical, Tokyo, Japan); cortisol, Cortisol Eiken (Eiken, Tokyo, Japan); aldosterone, Aldosterone RIA Kit II (Dainabot, Tokyo, Japan); and PRA, Renin RIA Kit (Dainabot, Tokyo, Japan)]. The RIA kit for the AVP assay required extraction of plasma using a Sep-Pak C18 cartridge (9). The inter- and intratest coefficients of variation for this kit range from 7.8–17.3% (mean, 12%) and from 4.4–22.5% (mean, 8.2%), respectively. Plasma osmolality was determined by freezing point depression (10).
Statistical methods
The results are expressed as the mean ± SEM. Statistical analysis was carried out using paired and unpaired t tests. Correlation was evaluated with Pearson’s correlation coefficient. P < 0.05 was considered significant.
Results
Effects of HS infusion on plasma osmolality and AVP
The basal levels of plasma osmolality in each group were not
significantly different (291.2 ± 1.2 mmol/kg in the control group
and 287.0 ± 1.2 mmol/kg in the DI group). Plasma osmolality was
significantly elevated in both groups 10 min after starting HS infusion
(298.0 ± 2.5 mmol/kg in the control group and 301. ± 1.8 mmol/kg
in the DI group; P < 0.05 and P <
0.01, respectively; Fig. 1), but the
between-group difference was not significant. However, the degree of
increase in osmolality 10 min after the start of HS infusion was
significantly greater for the DI group than for the control group
(14.8 ± 1.9 and 6.8 ± 2.1 mmol/kg, respectively;
P < 0.05).
|
).
|
The basal plasma ACTH levels were not significantly different
between the groups (5.2 ± 0.6 pmol/liter in the control group and
7.7 ± 1.5 pmol/liter in the DI group). At 10 and 20 min after the
start of HS infusion, plasma ACTH showed a nonsignificant tendency to
decrease in the control group to 5.0 ± 0.3 and 4.4 ± 0.2
pmol/liter, respectively. In the DI group, plasma ACTH levels markedly
increased at 10 and 20 min after the start of HS infusion to 23.0
± 2.7 pmol/liter (P < 0.01) and 17.3 ± 1.4
pmol/liter (P < 0.05), respectively (Fig. 3). This pattern of behavior of plasma
ACTH, peaking at 10 min after starting HS infusion, was consistently
observed in all patients, and the individual increases in plasma ACTH
ranged from about 2- to 6-fold, with a mean increase of 3.6-fold.
Interestingly, even in the patient whose plasma AVP level was lower
than 0.3 pmol/liter both before and after HS infusion (patient 4), the
increase in plasma ACTH was 2.3-fold. Although there was no correlation
between plasma osmolality and plasma ACTH in the control group (data
not shown), a positive correlation (r = 0.72; P <
0.005) was observed in the DI group (Fig. 4). No correlations between plasma AVP
and ACTH were seen in either the control group or the DI group.
|
|
).
|
Basal PRA was 0.28 ± 0.17 ng/ml·h in the control group and
1.08 ± 0.39 ng/ml·h in the DI group, showing a tendency to
increase in the latter group (P = 0.076). After HS
infusion, there were no significant changes in PRA in the control
group. In the DI group, PRA showed a tendency toward lower values 10
and 20 min after starting HS infusion (0.75 ± 0.19 and 0.61
± 0.19 ng/ml·h, respectively), but the differences were not
significant (Fig. 6). At 10 min after
starting HS infusion, PRA was significantly higher in the DI group
(P < 0.05).
|
).
|
Although ACTH secretion is primarily regulated by CRH, AVP has been found together with CRH in the parvocellular neurons of the paraventricular nucleus (PVN) and is known to act in concert with CRH on ACTH secretion (1, 2). It has been reported that the ACTH response to CRH is enhanced in patients with central DI, and that infusion of HS further potentiates this response (6, 7). However, the details of this mechanism remain unclear.
Our data showed that HS infusion itself increases both plasma osmolality and plasma ACTH concentration in DI patients. The finding that this increase in ACTH was absent in normal controls suggests that the regulation of ACTH in DI patients differs from that in healthy subjects. A further finding of interest was that this ACTH response was not associated with an increase in plasma AVP; instead, there was a positive correlation between plasma ACTH and osmolality. Thus, it is possible that ACTH secretion in DI patients is promoted directly by an osmotic stimulus or via some factor other than AVP.
However, Rittmaster and colleagues (5) analyzed the ACTH and cortisol responses to HS infusion and reported that the levels of ACTH and cortisol increased in nine healthy individuals. Furthermore, Bähr et al. (11) administered HS to six healthy subjects and reported that although the plasma AVP level increased significantly, levels of ACTH and cortisol did not increase significantly. In both of these reports, however, HS (5% NaCl) was infused for 120 min at a rate of 0.06 ml/kg·min. As this administration protocol is apparently different from that in the present study (0.24 ml/kg·min for 10 min), further investigation is needed to ascertain whether an osmotic stimulus alone can facilitate ACTH secretion in healthy individuals.
It has been recently demonstrated that there are at least two secretion pathways for AVP (12, 13). In one pathway, AVP is secreted into the posterior pituitary from magnocellular neurons in the supraoptic nucleus (SON) and PVN, and AVP released into peripheral blood via this pathway plays a key role in regulating water balance. The other pathway involves the secretion of AVP from parvocellular neurons in the PVN into the hypophyseal portal system, suggesting involvement with the regulation of anterior pituitary function. Hence, although it is apparent that the release of AVP from magnocellular neurons into peripheral blood is depressed in DI patients, it seems probable that AVP release from parvocellular neurons into the anterior pituitary is preserved. Mazza and colleagues (7) reported that the ACTH response to CRH is enhanced in DI patients and suggested, as a possible mechanism for this phenomenon, that the secretion of AVP from parvocellular neurons may be maintained in DI patients. In other words, they proposed that the impaired AVP secretion in DI patients might be restricted to the magnocellular neurons, and that AVP secretion from parvocellular neurons is enhanced. As a result, an exaggerated ACTH response occurs because AVP levels in the hypophyseal portal system would be higher. From autopsies of patients with hereditary DI, Bergeron et al. (14) reported that the elimination of vasopressin neurons mainly occurs in the magnocellular component, but that parvocellular vasopressin neurons are relatively well preserved. In a study using mice, although the plasma level of ACTH increased in response to salt loading, the level of CRH mRNA in the hypothalamus remained unchanged, whereas the level of AVP mRNA was shown to increase in the PVN and SON. The authors of that study drew the conclusion that the increase in ACTH was mostly attributable to an increase in AVP release from parvocellular neurons in the PVN (15). In addition, in another experiment using rats, ip administration of HS was found to increase the expression of vasopressin heteronuclear RNA in the parvocellular and magnocellular neurons in PVN (16). These findings indicate that an increase in osmolality might facilitate AVP production not only in magnocellular neurons, but also in parvocellular neurons (15, 16). Accordingly, the present results suggest that elevated plasma osmolality may be responsible for the increased secretion of AVP from parvocellular neurons. However, further studies are needed to clarify the details of this mechanism.
As there is a tendency toward dehydration in DI patients, the renin-angiotensin II (A II) system is enhanced (17). It is known that A II acts centrally or directly on the anterior pituitary, stimulating the secretion of ACTH (18). In fact, our study showed that PRA tended to be higher in DI patients than in controls, and at 10 min after starting HS infusion, PRA was significantly higher in the DI group. However, the reduced tendency of PRA in our DI patients after infusion of HS argues against the possibility that enhancement of the renin-A II system is involved in the increased ACTH response. In addition, Milsom and co-workers (4) found no changes in the ACTH response to CRH in salt-restricted normal subjects even when endogenous A II secretion was significantly heightened.
Elias et al. (6) reported that the increased ACTH response to CRH in DI patients was further augmented by infusion of HS. Plasma atrial natriuretic peptide (ANP) levels were lower than those in controls, and they hypothesized that the decrease in ANP might be associated with the enhanced ACTH response to CRH, as ANP has been suggested to inhibit ACTH secretion. We did not measure plasma ANP levels, but circulating ANP levels were increased after HS infusion in their DI patients, even though the levels were still lower than those in normal controls. Hence, it seems unlikely that the HS-induced increase in ACTH in this study is associated with ANP. As reported by Elias et al. (6), the results of the their study also showed that the infusion of HS facilitated ACTH secretion in DI patients. However, they compared the results after CRH administration and those after combined administration of CRH with HS, whereas in the present study we investigated the effects of HS infusion alone. In other words, the finding of the present study seems to be unique, in that an osmotic stimulus clearly increased ACTH secretion in DI patients.
Recently, urocortin (UCN), a novel neuropeptide in the CRH family, has been isolated from the rat midbrain (19, 20). UCN has a 45% sequence identity with CRH and can stimulate ACTH secretion more potently than that of CRH both in vitro and in vivo (19). UCN stimulates ACTH secretion via CRH receptors in the anterior pituitary, and its action is more potent than that of CRH (20). Additionally, UCN has been reported to act synergistically with AVP in terms of ACTH secretion (21). UCN-like immunoreactivity has been observed in the rat SON, PVN, and median eminence. In rats, UCN-like immunoreactivity in these regions of hypothalamus is markedly increased by hypophysectomy, as it is by salt loading (22) or dehydration (23). These findings suggest that like AVP, UCN plays a key role in controlling ACTH secretion from the anterior pituitary and regulating the body fluid balance. Furthermore, it has been reported that intracerebroventricular administration of UCN in the rat decreases the HS-induced release of AVP into the peripheral bloodstream (24). This seems to suggest that some feedback mechanism may exist between AVP and UCN in the regulation of the water balance. Hence, given the experimental results described above in rats, we can hypothesize that the ACTH secretion-promoting effect of HS infusion in our DI patients may have been due to the release of UCN to the anterior pituitary as a result of the elevation of plasma osmolality. However, although plasma osmolality levels after HS infusion in normal healthy subjects were comparable to those in DI patients, an increase in ACTH release was evident only in DI patients. One possible reason is that the production of UCN in the hypothalamus in DI patients is increased to compensate for a decrease in the supply of AVP to the anterior pituitary. It may also be that UCN production induced by the osmotic elevation is further augmented to the extent that it reaches the anterior pituitary, where its potent ACTH secretion-promoting effect is manifested. To explore this hypothesis further, it necessary to investigate UCN production in DI patients and in the Brattleboro rat, an animal model of DI.
In summary, we found that blood levels of ACTH in DI patients were increased by the elevation of plasma osmolality and that the regulatory mechanism of ACTH secretion in DI patients was distinct from that in healthy subjects. Given that this increase in ACTH is not associated with a rise in AVP levels in peripheral blood, it may occur independently of AVP secretion from magnocellular neurons.
Acknowledgments
Footnotes
Abbreviations: A II, Angiotensin II; ANP, atrial natriuretic peptide; AVP, arginine vasopressin; DI, diabetes insipidus; HS, hypertonic saline; PVN, paraventricular nucleus; SON, supraoptic nucleus; UCN, urocortin.
Received December 18, 2000.
Accepted September 2, 2001.
References
-receptor agonist U50488H on the secretion
of arginine vasopressin. Neuroendocrinology 48:658–662[Medline]
infusion increases vasopressin, ACTH and
11-hydroxycorticosteroid secretion. Am J Physiol 234:66–71
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
) and DI patients (•). Plasma osmolality was
significantly elevated in both groups. Each bar
represents the mean ± SEM. *, P
< 0.05; **, P < 0.01 (vs. 0
min).
