0013-7227/92/1313-1417$03.00/0 Endocrinology Copyright 0 1992 by The Endocrine
Vol. 131, No. 3 Society
Printed
Arginine Vasopressin Stimulates Atria1 Peptide Gene Expression and Secretion Diencephalic Neurons* ELLIS R. LEVIN,
REN-MING
Departments of Medicine Beach Veterans Hospital,
HU, MARY
ROSSI, AND MICHAEL
and Pharmacology, University Long Beach, California 90822
of California,
ABSTRACT The central nervous system modulates cardiovascular function and fluid and electrolyte balance in part through the actions of vasoactive peptides/neurotransmitters. The presence of several vasoactive peptides and their receptors in the hypothalamus suggests a possible interaction at this site. One level at which vasoactive peptides such as arginine vasopressin (AVP) and atria1 natriuretic peptide (ANP) might interact is through the mutual regulation of production and secretion in the hypothalamus. To determine whether AVP modulates ANP gene expression and secretion, we cultured fetal rat diencephalic neurons in the presence of AVP. AVP induced a significant increase in ANP secretion in dose-related fashion (mean + SEM basal ANP, 87 f 4 pg/ ml; maximal mean AVP-stimulated ANP, 146 + 6 pg/ml; P < 0.05,by analysis of variance). Neither oxytocin nor the vasoactive neuropeptide angiotensin-II had any effect on ANP secretion. The stimulatory effect of AVP was significantly blocked by coincubation with a V1 receptor anatgonist, but was unaffected by a VZ receptor antagonist. The im-
A
TRIAL natriuretic peptide (ANP) is a vasoactive hormone that is secreted from the atrium of the heart and circulates in the plasma (1). ANP also has been found to be produced by both the in vivo brain and in cultured diencephalic neurons (2,3). Communication between the brain and peripheral ANP systems and coordination of the actions of ANP in these two compartments are probably mediated through ANP receptors in the circumventricular organs of the central nervous system, which lie outside the blood-brain barrier (4, 5). Nevertheless, the brain and peripheral pools of ANP are separateand distinct, sinceANP doesnot appreciably cross the blood-brain barrier (6, 7). This distinction is further highlighted by the different posttranslational processingin the heart compared to that in the brain, since a more truncated form of ANP is found in the central nervous system (3, 8). The actions of ANP in the central nervous system include roles as a central neuromodulator of blood pressure (9-ll), a regulator of cerebrospinal fluid production (12), and a neurotransmitter that modifies pituitary hormone secretion (13, 14). Some of these effects result from the modulation of Received February 27, 1992. Address all correspondence and requests for reprints to: Dr. Ellis R. Levin, Medical Service (Ill-I), Long Beach Veterans Hospital, 5901 East 7th Street, Long Beach, California 90822. *This work was supported by a Merit Review Grant from the V.A. and grants from both the National and California American Heart Associations.
Irvine,
in U.S.A.
Natriuretic from Rat
PICKART California
92717; and the Long
munoreactive ANP secreted in response to AVP was the major brain isoform, ANP-(103-126). Coincubation with a calcium channel antaeonist, nifedipine, had no effect on AVP-induced ANP secretion, whiie ryanodine, an inhibitor of intracellular calcium mobilization, significantly reduced the stimulatory effect of AVP. AVP induced a doserelated, nearly 3-fold maximal increase in ANP mRNA expression at 4 h. Coincubation of the neurons with a V1 receptor antagonist also significantly attenuated the increased ANP gene expression induced by AVP. These results indicate that AVP acts directly through V1 receptors on cultured fetal rat dienceohalic neurons to aument ANP eerie expression and secretion of the peptide. The effects are probably relited to AVP-stimulated mobilization of intracellular calcium and not the result of calcium influx into the cell. These studies provide the first evidence that AVP modulates ANP production from cultured neurons. In the central nervous system, these two vasoactive neuropeptides might interact in part through the regulation of ANP production by AVP. (Endocrinology 131: 1417-1423,1992)
the receptor-mediated actions of other vasoactive neuropeptides (15-17). However, our recent finding that endothelin increasesANP secretion and production in cultured diencephalic neurons indicates that vasoactive neuropeptides also can modulate each other’s production and secretion (18). This potential mutual regulation indicates another mechanism through which these neuropeptides can modulate the complex functions of the brain. Arginine vasopressin (AVP; with neurophysin) is synthesized in the supraoptic, paraventricular, and suprachiasmatic nuclei of the hypothalamus. This hormone is ultimately packaged and transported to the neurohypohysis, where it is secreted in response to changes in osmolality (19). AVP receptors are also present in high concentrations in the hypothalamus (20). We, therefore, postulated that this peptide might modulate the production and secretion of ANP from this part of the brain. We also determined whether oxytocin and angiotensin stimulate ANP secretion. We report that AVP significantly stimulates the expression of ANP mRNA and the secretion of this peptide in cultured fetal rat diencephalic neurons. Materials and Methods Materials AVP, oxytocin, angiotensin, Laboratories (Belmont, CA),
and ANP were obtained and [lZ5]NaI from New
from Penninsula England Nuclear
1417
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 18:11 For personal use only. No other uses without permission. . All rights reserved.
AVP AND
1418
Endo. Voll31.
ANP
1992 No 3
(Wilmington, DE). The Vi receptor antagonist d(CH&-Tyr(Me)AVP was obtained from Bachem (Torrance, CA), and the VZ receptor antagonist d(CH&-[u-Gle’-Gle4-Ala(NH2)9]AVP from M. Manning (Toledo, OH). Nifedipine and ryanodine were obtained from Research Biochemical, Inc. (Natick, MA).
acidified, boiled, then, after lyophilization, reconstituted in RIA buffer. In parallel studies trypan blue exclusion was used to establish cell viability at the end of the incubation period (>90% excluded this dye).
Cell culture
RNA from neurons (or glia) plated on 100~mm plates was extracted by the guanidinium thiocyanate-cesium chloride method (22). As we have previously described (lt?), the extracted RNA was then hybridized with an antisense 32P-labeled cRNA probe for rat ANP made from a KpnI-linearized cDNA template for rat ANP in PBS. The cDNA was originally obtained from Drs. David Gardner (University of CaliforniaSan Francisco) (23) and John Lewicki (Cal BioTechnology, Mountain View, CA) (24) in pUC-9. The 300-base probe spans exon 3 of the rat ANP gene and is predicted to protect 270 bases of the ANP transcript. Insertion into PBS allowed construction of RNA probes complementary to the cDNA for ANP, using Tj or T, RNA polymerase (antisense and sense), according to the methods of Krieg and Melton (25). Hybridization was carried out for 20 h at 58 C in buffer containing 80% formamide (high stringency). Nonhybridized nucleic acid was digested with Sl nuclease (Pharmacia, Piscataway, NJ), then separated by electrophoresis on a 7-~ urea denaturing polyacrylamide gel, and the size of the protected band was determined by comparison to labeled HinfI-digested pBR 322. We previously determined that this probe protected a band of approximately 270 bases, using RNA extracted from the neurons. This band was identical in size to ANP transcripts from the atrium of the heart, detected after hybridization with our probe (18). A HindIIIdivested and 32P-labeled cRNA for H-ras served as an RNA-loadine stindardization probe. The gel was opposed to film with intensify& screens for 48-72 h, and the autoradiographic bands were compared by laser densitometry (LKB, Rockville, MD). Sense probes produced no hybridization.
Cultures of neurons from fetal rat diencephalon were established as previously described (3, 18). Briefly, the diencephalic area of the brain was carefully isolated from 16-day gestational fetal rats and placed into sterile Hanks’ Balanced Salt Solution medium. The tissue consisted mainly of hypothalamus with some thalamus (estimated to be 15% of the total tissue) attached and was mechanically and enzymatically dispersed. The cells were then plated at an optimal density of 5 X lo5 cells/cm’ in Dulbecco’s Modified Eagle’s Medium-Ham’s F-12 medium with 10% fetal bovine serum and kept in a 37 C incubator under 95% air and 5% CO2 conditions. Medium was changed every thud day, and 5-fluorodeoxyuridine (100 P(M) was added on day 3 after plating to decrease glial cell overgrowth. This approach yields a culture that is approximately 85% neurons, as determined by neurofilament antibody and specific enolase staining (3). The neurons were used 7-10 days after plating. We also established glial cultures with a density comparable to the amount of glia in our neuronal cultures, as previously described (21). These gIia1 cultures were then subjected to trypsinization and replated, yielding cultures that were virtually 100% glia, determined by staining with an antibody to glial fibrillary acidic protein (21).
ANP immunoreactivity (ANP-IR) was determined as previously described, using a double antibody, nonequilibrium RIA (3, 18). The primary antisera (dilution, 1:125,000) was generated in rabbits (Bunny B). The sensitivity of the assay is 15 pg/ml. The antiserum detects ANP(102-126) and -(103-126) 85% comparably to ANP (99-126) on a molar basis, but does not recognize porcine brain natriuretic peptide (BNP) and sees porcine C-type natriuretic peptide (C-NP) 10% as well as ANP.
Chromatography Reverse phase HPLC was used to characterize the molecular forms of ANP, as previously described (3, 18). Secretion medium from basal or AVP-stimulated neurons (two loo-mm plates each) was acidified in 0.1 ml 1 M acetic acid, boiled for 10 rain, then sonicated and centrifuged at 35,000 x g for 20 min. The supematant was subsequently freezedried (Savant Instruments, Hicksville, NY) and reconstituted in starting HPLC buffer at the time of analysis. After injection onto a Cl8 ODS column, samples were eluted over 45 min using a nonlinear gradient of 15-60% acetonitrile-0.1% trifluroacetic acid in water. Samples were collected and freeze-dried before reconstitution in RIA buffer. The column was subsequently washed with one gradient over 10 min between sample runs and before marking the column with 1 pg each ANP(102-126), ANP-(103-126), and C-NP.
Experiments Secretion studies were carried out in Krebs-Ringer bicarbonate buffer. On the day of experimentation, Dulbecco’s Modified Eagle’s MediumHam’s F-12 medium with fetal bovine serum was removed and replaced with Krebs-Ringer buffer containing 2 mu calcium without serum. Studies for RNA were conducted in lOO-mm culture dishes, originally plated with 25 X lo6 cells/dish, while secretion studies were carried out in 6-we11 plates (7 million cells/well). Medium alone (control) or medium containing AVP at various concentrations, AVP receptor antagonists (Vi or V,), 50 $M nifedipine, or 1 PM ryanodine, alone or in combination, was added to the cells. In separate experiments the effects of oxytocin (10e8 and 10e6 M) or angiotensin (same concentrations) were examined. Three dishes each per condition were then incubated with AVP (10m6 M) for 4 or 24 h (time course) to determine ANP gene expression. Subsequent studies of ANP mRNA expression or secretion were carried out after 4 h of incubation. The medium from the secretion experiments was
Solution
hybridization
and 5’1 nuckase protection
Statistics Data from at least two secretion studies were pooled and analyzed by calculating a mean and SE for each condition, which were then compared by analysis of variance (ANOVA) and a multiple range test; Scheffe’s test was used for significant F values. Significance was defined as a P < 0.05 level. All studies were carried out two or three times. RNA comparisons were made by laser densitometry of autoradiographs, and data were normalized for RNA loading by comparing the density of the experimental condition RNA hybridized with the ANP probe divided by the same RNA hybridized with a probe for H-rus. A ratio was then established by comparing normalized experimental RNA to normalized control RNA. This resulted in values expressing the relative densities of the experimental conditions.
Results Secretion
studies
In dose-related fashion, AVP induced a significant increase in ANP secretion from cultured diencephalic neurons (mean + SEM basalANP, 87 f 4 pg/ml; mean 10m6 M AVP-stimulated ANP, 146 + 6 pg/ml; P < 0.05, by ANOVA; Fig. 1). In contrast, neither oxytocin nor angiotensin-II stimulated ANP secretion (mean basal ANP, 85 + 8; 1O-6M oxytocin, 87 f 6; 10e6M angiotensin, 91 + 5 pg/ml). AVP (lo-’ M) alsoinduced a significant, greater than 50% increase in ANP secretion. These findings occurred after 4 h of incubation with AVP, indicating that AVP may also affect ANP synthesis in these cells. The effects of AVP appeared to be mediated through V1 receptors, since coincubation with the Vi receptor antagonist d(CH&-Tyr(Me)AVP significantly impaired the ability of AVP to stimulate ANP secretion (Fig. 2). The Vi receptor antagonist competitiveIy inhibited greater than 50% of the
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 18:11 For personal use only. No other uses without permission. . All rights reserved.
AVP
AND
ANP
1419
150 a T :
125 h z . CI) VP a T !i a
150
4 6
a
100
z 0
L
0
100
: f
75
50 C
AVP 1 o-6
AVP
AVP
VIA
25
PEPTlDElSUSSTANCE
AVP 1 o-10
PEPTIDE
AVP 1 o-8
AVP 1 o-6
(MI
ANP geneexpression AVP also increased the mRNA for ANP. Due to the lack of sufficient neurons to conduct a more comprehensive time course, a limited study was carried out to determine the time
(Ml
FIG. 2. The effect of AVP and AVP receptor antagonists VI or Vz on ANP secretion from diencephalic neurons after 4 h of incubation. Each point represents the mean f SEM difference compared to the control level of ANP secretion, using pooled data from two experiments (n = 10 wells/condition). C, Secretion in the absence of AVP. *, P < 0.05 comnared to control. bv ANOVA ~1~s Scheffe’s test. +. P < 0.05 for 10e6*~ AVP plus ~I+-M Vl receptor antagonist compared to AVP alone.
FIG. 1. Effect of AVP on the secretion of ANP from cultured rat diencephalic neurons incubated for 4 h. Each value represents the mean f SEM of data from two experiments combined (n = lO/experimental condition). *, P < 0.05 vs. control cells (C) incubated in the absence of AVP, by ANOVA plus Scheffe’s test.
augmented secretion of ANP caused by 10m6M AVP. In contrast, the Vz receptor antagonist d(CH&-[u-Gle’-Gle4Ala(NH&]AVP had little effect on AVP-stimulated ANP secretion. The effects of AVP were probably mediated through a postreceptor-binding signal involving cytoplasmic calcium, sinceryanodine, an inhibitor of intracellular calcium mobilization, significantly and almost completely impaired the stimulatory action of AVP (Fig. 3). Related to this issue, L-type calcium channel-mediated calcium influx into the cell was probably not involved, since nifedipine had no effect on AVP’s actions (Fig. 3). HPLC followed by RIA profiling of the secretedmedium revealed that ANP-(103-126) was the predominant isoform secreted (Fig. 4). In addition, a higher mol wt form of ANP was secreted in fraction 15, possibly the 102-126 isoform, as well as a smaller unidentifiable fragment of ANP-IR detected in fraction 28. The basal neurons contained both ANP-(102-126) and ANP-(103-126). No ANP-IR was detected at the elution position of the standard C-NP using our ANP antibody. Previously, we had shown that ANP-(l-28), the predominant form in plasma, is not present in cultured rat hypothalamic neurons (3, 18).
V2A
V:A 10-6
;t A 1 o-8
200
7
*
150 h ii . : a T
100
5 a
50
C
AVP 10-6
AVP + N 5x10-5
PEPTlDElSUBSTANCE
N
AVP
R
lz 10-B (M)
FIG. 3. The effect of a calcium channel blocker, nifedipine (50 pM), or the intracellular calcium mobilization inhibitor ryanodine (1 pM) on AVP-induced ANP secretion. Data (mean f SEM) are combined from two experiments (n = 10 wells/experimental variable). *, P < 0.05 compared to control (C), by ANOVA plus Scheffe’s test. +, P < 0.05 for 10e6 M AVP plus lo-’ M ryanodine compared to AVP alone.
when AVP augmented ANP gene expression. This study indicated that AVP increasedANP gene expressionby 2.9 + 0.3 &EM)-fold at 4 h, with a return to baselineat 24 h (Fig. 5; n = 2 experiments). The size of the protected transcript
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 18:11 For personal use only. No other uses without permission. . All rights reserved.
Endo. 1’392 VoI 131 l No 3
AVP AND ANP
B c ir 1
After determining that AVP increased ANP gene expression at 4 h of incubation, we carried out a dose-response study at this time (Table 1). AVP (10e6 M) induced a 2.5 + 0.2-fold increase in ANP gene expression (Fig. 6, lane 2), while AVP (lo-’ M; lane 3) resulted in a 50% increase compared to control after adjustment for RNA loading by comparison to H-YUS. The stimulation of ANT’ gene expression by AVP was significantly reversed by coincubation with a VI receptor antagonist (Fig. 6, lane 2 vs. 4). Cultured diencephalic glia, at the density present in our neuronal cultures, did not secrete ANP or express ANP mRNA basally or in response to AVP (data not shown). These latter results are nearly identical to previously reported findings (18).
iI II ‘I ‘I I’ I’ I’ I’
Discussion We have demonstrated that AVP directly causesan increase in ANP mRNA levels and peptide secretion from cultured fetal rat diencephalic neurons. These are the first results indicating that AVP can modulate ANT’ production and secretion in the brain. Since these findings were derived
0
50
TABLE 1. Increase in mRNA for ANP induced by AVP in cultured fetal rat diencephalic neurons mRNA 10
20 FRACTION
30
40
NUMBER
4. Reverse phase HPLC profile of ANP contained within cultured diencephalic neurons (0) or secreted into the medium from neurons incubated with lo-’ M AVP (0). Fractions represent pooled media from two lOO-mm plates of neurons per each condition subjected to HPLC, followed by RIA. Arrows A, B, and C represent the elution positions of 1 pg each of rat ANP-(102-126), ANP-(103-126), and porcine C-NP standards, respectively. FIG.
298-
2
Control AVP (lo* M) AVP (10” M) AVP (10” M) + Vi antagonist (10e6 M) V1 antagonist (IO-’ M)
for ANP
ratio
1 2.5 + 0.2” 1.5 + 0.4 1.2 + 0.3” 0.78 + 0.2
Each value was determined from pooled extracted RNA from three IOO-mm plates of neurons per condition during a representative experiment, the study was repeated, and the data were combined as the average of the two experiments. Each protected band was quantitated by laser densitometry and normalized for RNA loading by comparison to protected bands after hybridization of a 32P-labeled cRNA probe for H-ras with 30 pg neuronal RNA from each experimental condition. These normalized values were then compared to the level in control neurons, designated 1, and this ratio was used for the table values. a P < 0.05 for AVP (10e6 M) us. control. * P < 0.05 for AVP (10e6 M) vs. AVP plus Vi antagonist.
3
5. Time course of the effects of AVP on ANP gene expression in cultured rat diencephalic neurons, determined by Sl nuciease protection. Lanes l-3 are RNA (40 rg) extracted from neurons cultured with 10” M AVP for 0,4, and 24 h, respectively. The protected band is 270 bases, determined by comparison with 32P-labeled Hi&I-digested pBR 322 (digested band of 298 bases, indicated by arrow). A second study was similar in appearance. FIG.
was approximately 270 bases, which we have previously shown is identical in size to the ANP transcript in rat atrium (18). The return to baseline of ANP mRNA may be related to degradation of the AVP over 24 h. However, we determined that only about 20-25% of the AVP was degraded during this period by HPLC profile of the incubation medium, and that the reduced AVP concentration should have stimulated ANP gene expression, as did lo-’ M AVP. Possibly, AVPinduced receptor down-regulation blunted the stimulatory effects over 24 h.
FIG. 6. Sl nuclease protection after solution hybridization of total RNA (30 pg) from neurons incubated with AVP by itself or with a V, receptor antagonist for 4 h. Top, Lanes l-5 are control, 10’ M AVP, 10” M AVP, lo4 M AVP plus lObe M Vi receptor antagonist, and the Vl receptor antagonist alone, respectively. This autoradiograph is representative of a second study. Bottom, Control hybridization with a cRNA for H-ros. Densitometry of protected hybridized bands are normalized for the density of H-ras, as described in Materials and Methods.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 18:11 For personal use only. No other uses without permission. . All rights reserved.
AVP
AND
from an in vitro preparation of fetal neurons, extrapolation of these results to the in vivo adult brain are tentative. If our findings are representative of the normal physiology in the adult brain, they support the idea that AVP produced in hypothalamic nuclei might regulate ANP in a paracrine fashion. In contrast, oxytocin and the vasoactive neuropeptide angiotensin did not stimulate ANP secretion. Several investigators have shown that iv administered AVP (which does not cross the blood-brain barrier) results in increased plasma levels of ANP which is secretedfrom the heart. This probably occurs asa responseto altered intravascular volume and atria1 stretch (26, 27). Therefore, AVP potentially increasesANP secretion in both compartments of the body, through direct and indirect means. The stimulation of ANP secretion and gene expression by AVP appearsto be mediated through Vi and not VI receptors on cultured diencephalic neurons. Cultured glia, at the density present in our neuronal cultures, did not produce measurable ANP secretion or expressANP mRNA and, therefore, probably did not contribute to the results in this study; these findings are corroborated by our previous demonstration that endothelin stimulates ANP secretion and mRNA in neurons only (18). Additionally, vasopressin receptors have been found only on neurons and not glia in the hypothalamic area (28). Vasopressinreceptors have traditionally been cIassified into V1 and VZ subtypes basedupon their signal transduction mechanisms(29). Vi receptors are found in liver and vascular smooth muscle, where they mediate the glycogenolytic and vasoconstrictive actions of vasopressin (29, 30). V2 receptors are found primarily in the kidney, where AVP acts as an antidiuretic hormone (31). Previous characterization of AVP receptors in the brain indicates that mainly classicVi receptors are present (31), which is consistent with our results. V1 receptors are reported to signal through phosphatidylinositol hydrolysis, leading to cytosolic calcium mobilization (30), while Vz receptors in the periphery signal through CAMP generation (31). Our findings that nifedipine does not influence the effects of AVP, yet ryanodine inhibits AVP-stimulated ANP secretion support the idea that intracellular calcium mobilization is necessary for AVP to stimulate ANP secretion from cultured diencephalic neurons. Recently, it hasbeen shown that histamine forms inositol 1,4,5-trisphosphate and activates a ryanodine-sensitive mechanism of increased intracellular calcium in adrenal chromaffin cells, comparable to the postulated effects of AVP in these studies (32). Interestingly, angiotensin-II, a neuropeptide that acts through postreceptor signaling mechanismsinvolving intracellular calcium mobilization (33), did not increase ANP secretion, while endothelin, a peptide that in part signals through similar mechanisms, does stimulate ANP secretion and transcription (18, 34). Therefore, mobilization of intracellular calcium per se does not appear to be a sufficient stimulus for increased ANP production. The possibility that AVP acted through oxytocin receptors on the diencephalic neurons is unlikely, in view of the fact that oxytocin itself had little effect on ANP secretion. In this study we showed that concentrations of AVP as low as lo-’ M significantly stimulated ANP secretion. Shewey
ANP
and Dorsa (35) showed that AVP 1O-8-1O-6M induced as much asa 65% increasein inositol phospholipid metabolism in rat septal membranesby binding to Vi receptors. Although the fold stimulation of ANP gene expression by 10m6 M AVP exceeded the increase in secretion, the two processesoften do not increase in a stoichiometric fashion. This is because ANP transcripts are not necessarily translated, and some degradation of secreted ANP protein can occur over 4-h exposure to cellsin culture. It is possiblethat AVP contributes to increased ANP gene expression in part by stabilizing the primary transcript. Previously, we have shown that the vasoconstrictor peptide endothelin augments ANP gene expression and peptide secretion from these sameneurons (18), while ANP hasbeen proposed to directly inhibit AVP secretion from supraoptic and paraventricular neurons (36). In cultured aortic endothelial cells, AVP and angiotensin increase endothelin gene expression and protein secretion (37). These findings collectively imply that in vim, several vasoactive peptides modulate each other’s actions at least partly through mutually regulating gene expression and protein production and secretion throughout the body. What processesin the brain might be mediated through the interaction of AVP and ANP? In the adult rat brain, AVP has been shown to modulate the central nervous system regulation of blood pressure, acting as a pressor agent (38). This effect probably occurs through AVP stimulating sympathetic outflow through the hypothalamus and brainstem (39, 40). In contrast, ANP is a vasodepressor and acts to decreaseperipheral sympathetic tone through actions again in both hypothalamic and brainstem nuclei (11, 41). Thus, the releaseof ANP engenderedby AVP in the hypothalamus might serve to limit the extent of AVP-induced sympathetic stimulation and dampen the resulting increased blood pressure. Support for this idea comesfrom studies in the nucleus of the solitary tract, where AVP has been shown to decrease blood pressurethrough the releaseof an unidentified vasodepressorsubstance(ANP?) (42). Another functional overlap for these two peptides is in the releaseof ACTH from the pituitary. AVP is a known secretagogue for ACTH (43), while ANP has been shown to inhibit both CRF-induced and AVP-stimulated ACTH secretion(44, 45). This latter effect in vim might occur in part as the result of ANP inhibiting the secretion of AVP (19, 46, 47). The interactions of AVP and ANP may also occur through postreceptor binding signaling interactions. AVP has been shown to inhibit ANP-induced cGMP generation (ANP’s second messenger)in smooth muscle cells, possibly through stimulating phosphatidylinositol hydrolysis and subsequent protein kinase-C activation (48). The stimulation of protein kinase-C has been shown in several cell systems to inhibit ANP-generated cGMP (49, 50). ANP, in turn, can inhibit AVP-stimulated increases in intracellular calcium (51, 52), potentially attenuating the action of AUP. If these interactions also occur in the in vivo brain, the two peptides could modulate each other’s actions through several mechanisms. Our studies indicate that AVP mainly causesthe release of ANP-(103-126), the predominant form of ANP secreted
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 18:11 For personal use only. No other uses without permission. . All rights reserved.
1422
AVP AND ANP
from neurons in response to several secretagogues (18, 53). The ANP-IR found in neuronal culture medium could not be brain natriuretic peptide (BNP), since our antibody does not recognize BNP, and BNP is not present in rat brain (54). Neither could we be substantially detecting C-NP, the recently described member of the natriuretic peptide family that is found in abundance in porcine brain (55), since HPLCfractionated ANP-IR from our neuronal cultures does not elute at the position of standard C-NP. In summary, our findings provide the first evidence that AVP can directly stimulate ANP production in a brain preparation. A dynamic interaction of these two hypothalamic peptides at several levels probably contributes to important brain functions mediated through the hypothalamus. References AJ 1985 Atria1 natriuretic factor: a hormone produced by 1. DeBold the heart. Science 230:767-768 2. Gardner DG, Vlasuk GP, Baxter JD, Fiddes JC, Lewicki J 1987 Identification of atria1 natriuretic factor gene transcripts in the central nervous system of the rat. Proc Nat1 Acad Sci USA 84:21752179 3. Levin ER, Loughlin SE, Kaplan G 1990 Atria1 natriuretic peptide secretion from fetal rat diencephalon in culture. J Neuroendocrinol 2:318-321 4. Saper CB, Standaert DG, Currie MG, Schwartz D, Needleman P 1985 Atriopeptin-immunoreactive neurons in the brain: presence in cardiovascular regulatory areas, Science 227:1047-1049 5. Kuihara M, Saavedra JM, Shigematsu K 1987 Localization and characterization of atria1 natriuretic peptide binding sites in discrete areas of rat brain and pituitary by quantitative autoradiography. - _ Brain Res 408:31-39 * . . _ 6. Levin ER, Frank HJL, Weber MA, Ismail M, Mills SD 1987 Studies of the penetration of the blood brain barrier by atria1 natriuretic factor. Biochem Biophys Res Commun 147:1226-1231 7. Levin ER, Weber MA, Mills S 1988 Atria1 natriuretic factor induced vasodepression occurs through the central nervous system. Am J Physiol255:H616-H622 8. Shiono S, Nakao K, Morii N, Yamada T, Itoh H, Sakamoto M, Sugawara A, Saito Y, Katsuura G, Imura H 1986 Nature of atria1 Inatriuretic polvpeptide in rat brain, Biochem Biophys _ . Res Commun 135:728-734 ‘& _ 9. Ermirio E, Ruggeri I’, Cogo CE, Molinari C, Calaresu FR 1989 Neuronal and cardiovascular responses to ANF microiniected into the solitary nucleus. Am J Physioi 256:R577-R582 ’ 10. Levin ER, Mills S, Weber MA 1989 Central nervous system mediated vasodepressor action of atria1 natriuretic factor. Life Sci 44:1617-1624 S 1990 Blockade of 11. Yang R-H, Jin H, Chen Y-F, Wyss JM, Oparil endogenous anterior hypothalamic atria1 natriuretic peptide with monoclonal antibody lowers blood pressure in spontaneously hypertensive rats. J Clin Invest 86:1985-1990 JA 1987 Brain barrier tissues: end organs for 12. Steardo L, Nathanson atriopeptins. Science 235:470-473 R, Mogg R 1988 Evidence for a dopaminergic 13. Samson WK, Bianchi mechanism for the prolactin inhibitory effect of atria1 natriuretic factor. Neuroendocrinology 47:269-271 WK, Aguila MC, Bianchi R 1988 Atria1 natriuretic factor 14. Samson inhibits luteinizmg hormone secretion in the rat: evidence for a hvuothalamic site of action. Endocrinolonv 122:1573-1582 K, Morii N, Yamada TyShiono S, Sakamoto M, 15. Itbh H, Nakao Sugawara A, Saito Y, Katsuura G, Shiomi T, Eigyo M, Matsushita A, Imura H 1986 Central actions of atria1 natriuretic polypeptide on blood pressure in conscious rats. Brain Res Bull 16:745-749 16. Antunes-Rodriguez J, McCann SM, Rodgers LC, Samson WK 1985 Atria1 natriuretic factor inhibits dehydration and angiotensin II-induced water intake in the conscious unrestrained rat. Proc Nat1
Endo. 1992 Vol131. No 3
Acad Sci USA 82:8720-8723 17. Levin ER 1991 Atria1 natriuretic peptide and endothelin: interactions in the central nervous system and periphery. Mol Cell Neurosci 2:189-201 18. Levin ER, Isackson PJ, Hu R-M 1991 Endothelin increases atria1 natriuretic peptide production in cultured rat diencephalic neurons. Endocrinology 128:2925-2930 LW, Sawchencko PE 1983 Hypothalamic integration: 19. Swanson organization of the paraventricular and supraoptic nuclei. Annu Rev Neurosci 6~269-324 HI, Gee KW, Brinton RE, Davis TP, Hadley M, Wam20. Yamamura sley JK 1983 Light microscopic autoradiographic visualization of [3H]-arginine vasopressin binding sites in rat brain. Life Sci 32:19191924 ER, Frank HJL, Gelfand R, Loughlin S, Kaplan G 1990 21. Levin Natriuretic peptide receptors in cultured rat diencephalon. J Biol Chem 265:10019-10024 22. Chomczynski P, Sacchi N 1987 Single step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem 162:156-159 23. Gardner DG, Deschepper CF, Ganong WF, Satoshi H, Fiddes J, Baxter JD, Lewicki J 1986 Extra atria1 expression of the gene for atria1 natriuretic factor. Proc Nat1 Acad Sci USA 83:6697-6701 M, Greenberg B, Johnson L, Seilhamer J, Brewer M, 24. Yamanaka Friedemann T, Miller J, Atlas S, Laragh J, Lewicki J, Fiddes J 1984 Cloning and sequence analysis of the cDNA for the rat atria1 natriuretic factor precursor. Nature 309:719-722 25. Krieg A, Melton DA 1984 Functional messenger RNAs are produced by SP6 in vitro transcription of cloned cDNAs. Nucleic Acids Res 12:7057-7070 PT, Schwartz D, Katsube D, Holmberg SW, Needleman 26. Manning P 1985 Vasopressin-stimulated release of atriopeptin: endocrine antagonists in fluid homeostasis. Science 229:395-400 27. Itoh H, Nakao K, Yamada T, Morii N, Shiono S, Sugawara A, Saito Y, Mukoyama M, Arai H, Katsuura G, Eigyo M, Matsushita A, Imura H 1987 Modulatory role of vasopressin in secretion of atria1 natriuretic polypeptide in conscious rats. Endocrinology 120:2186-2188 28 Jeulin AC, Nicolaides S 1988 Evidence for vasopressin Vl receptors of rostrodiencephalic neurons: iontophoretic studies in the in vim rat. Responses to oxytocin and angiotensin. Brain Res Bull 20:817823 29 Michell RH, Kirk CJ, Billah MM 1979 Hormonal stimulation of phosphotidylinositol breakdown with particular reference to the hepatic effects of vasopressin. Biochem Sot Trans 7:861-865 30 de Wulf H, Keppens S, Vandeheede JR, Haustraete F, Proust C, Carton H 1980 Cyclic AMP-independent regulation of liver glycogenolysis. In: Nunez J, Dumont J (eds) Hormones and Cell Regulation. Elsevier/North Holland, Amsterdam, vol 4:47-71 mechanisms of receptor activation. Prog. 31 Jard S 1983 Vasopressin: Brain Res 60:383-394 KA, Murawsky MM 1991 The inositol 1,4,5trisphos32. Stauderman phatate-forming agonist histamine activates a ryanodine-sensitive Ca+’ release mechanism in bovine adrenal chromaffin cells. 1 Biol Chem 266:19150-19153 AM, Lew PD, Jornot L, Vallotin MB 1984 Correlation 33. Capponi between cytosolic free Ca*+ and aldosterone production in bovine adrenal glomerulosa cells. J Biol Chem 259:8863-8869 HE, Harris HN, McDonough PM, Knowlton KU, Glem34. Shubeita botski C. Brown TH, induction of , , Chien KR 1990 Endothelin sarcomere assembly and cardiac gene expression in ventricular myocytes: a paracrine mechanism for myocardial cell hypertrophy. J Biol Chem 265:20555-20562 LM, Dorsa DM 1988 W-type vasopressin receptors in rat 35. Shewey brain septum: Binding characteristics and effects on inositol phospholipid metabolism. J Neurosci 8:1671-1677 DG, Cechetto DF, Needleman P, Saper CB 1987 Inhi36. Standaert bition of the firing of vasopressin neurons by atriopeptin. Nature 329:151-153 37. Resink TJ, Hahn AWA, Scott-Burden T, Powell J, Weber E, Buhler FR 1990 Inducible endothelin mRNA expression and peptide secretion in cultured human vascular smooth muscle cells.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 18:11 For personal use only. No other uses without permission. . All rights reserved.
AVP
AND
Biochem Biophys Res Commun 168:1303-1310 38. Bereck KH. Swords BH 1990 Central role for vasouressin in cardiovascular regulation and the pathogenesis of hype;tension. Hoper__ tension 16:2’i3-224 39. Matsueuchi H. Sharabi FM. Gordon FL lohnson AK, Schmid PG 1982 Blood pressure and heart rate responses to microinjection of vasopressin into the nucleus tractus solitarius region of the rat. Neuropharmacology 21:687-693 40. Feurstein G, Zerbe RL, Faden AI 1983 Central cardiovascular effects of vasotocin, oxytocin and vasopressin in conscious rats. J Pharmacol Exp Ther 228348-353 41. Levin ER 1989 Atria1 natriuretic factor: neuromodulator of the central nervous system regulation of blood pressure. Am J Hypertension 2:594-598 42. Brattstrom A, DeJong W, DeWeid D 1988 Vasopressin microinjections into the nucleus tractus solitarii decreases heart rate and blood pressure in anesthetized rats. J Hypertension [Suppl4] 6:S521-S524 43. Raff H, Merrill DC, Skelton MM, Brownfield MS, Cowley Jr AW 1990 Control of adrenocorticotropin secretion and adrenocortical sensitivity in neurohypohysectomized conscious dogs: effects of acute and chronic vasopressin replacement. Endocrinology 119:1126-1133 44. King MS, Baertschi AJ 1990 Physiologic concentrations of atria1 natriuretic factors with intact N-terminal sequences inhibit corticotropin-releasing factor-stimulated adrenocorticotropin secretion from cultured interior pituitary cells. Endocrinology i24:286-291 45. Kovacs KI. Antoni FA 1990 Atrioueptin inhibits stimulated secretion of adrenocorticotropin in rats: ekdence for a pituitary site of action. Endocrinology 127:3003-3008 46. Iitake K, Share L, Crofton JT, Brooks DP, Ouchi Y, Blaine EH 1986 Central atria1 natriuretic factor reduces vasopressin secretion in the rat. Endocrinology 119:438-440 47. Crandall ME, Greeg CM 1986 In vitro evidence for an inhibitory -
1
__
ANP
48.
49.
50.
51.
1423
effect of atria1 natriuretic peptide on vasopressin release. Neuroendocrinology 44:439-445 Nambi P, Whitman M, Gessner G, Aiyar N, Crooke ST 1986 Vasopressin-mediated inhibition of atria1 natriuretic factor-stimulated cGMP accumulation in an established smooth muscle cell line. Proc Nat1 Acad Sci USA 83:8492-8495 Jaiswal RK, Jaiswal N, Sharma RK 1988 Negative regulation of atria1 natriuretic factor receptor coupled membrane guanylate cyclase by phorbol ester. Potential protein kinase C regulation of cyclic GMP signal in isolated adrenocortical carcinoma cells of rat. FEBS Lett 22747-50 Mukhopadhyay AK, Leidenberger FA 1988 Effect of a tumorpromoting phorbol ester on atrial-peptide-induced testosterone production and cGMP accumulation bv isolated mouse Levdia . Y cells. Mol Cell Endocrinol56:171-176 * Cornwell TL, Lincoln TM 1989 Regulation of intracellular Ca*+ levels in cultured vascular smooth muscle cells. J Biol Chem 264:1146-1155
Meyer-Lehnert H, Caramel0 C, Tsai P, Schrier RW 1988 Interaction of atriopeptin III and vasopressin on calcium kinetics and contraction of aortic smooth muscle. J Clin Invest 82:1407-1414 53. Lim AT, Dean B, Copolov DL 1990 Evidence for post-translational processing of auriculin B to atriopeptin III immediately prior to secretion by hypothalamic neurons in culture. Endocrinology 52.
127:2598-2600
Aburaya M, Minamino N, Hino J, Kangawa K, Matsuo H 1989 Distribution and molecular forms of brain natriuretic peptide in the central nervous system, heart and peripheral tissue of rat. Biochem Biophys Res Commun 165:880-887 55. Sudoh T, Minamino N, Kantgawa K, Matsuo H 1990 C-Tvpe natriuretic peptide (CNP): a new member of natriuretic peptide family identified in porcine brain. Biochem Biophys Res Commun 54.
168:863-870
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 18:11 For personal use only. No other uses without permission. . All rights reserved.