ELSEVIER Regulatory Peptides 57 (1995) 55-63

Generation of cyclic guanosine monophosphate in brain slices incubated with atrial or C-type natriuretic peptides: comparison of

the amplitudes and cellular distribution of the responses

Jason Gon~alves, Kevin L. Grove, Christian F. Deschepper* MR C Multidisciplinary Research Group on Hypertension, Neurobiology and Vasoactive Peptide Laboratory, Institut de Recherches Cliniques

de Montr~al (IRCM) and Universit~ de Montrial, Montrlal, Quebec, H2 W 1R7 Canada

Received 25 January 1995; revised version received and accepted 12 February 1995

Abstract

Natriuretic peptides have been demonstrated to induce a variety of effects when administered into the brain. Most studies to date have tested the effects of 'atrial' natriuretic peptide (ANP), but C-type natriuretie peptide (CNP) has recently been suggested to be the predominant form of natriuretic peptides within the brain. We therefore have compared the ampli- tudes of the cyclic guanosine monophosphate (cGMP) responses induced by either ANP or CNP in slices from differ- ent rat brain regions. Whereas both peptides induced the generation of cGMP, CNP-evoked responses were never greater than those obtained with ANP, regardless of the brain region used or the age of the animal. In diencephalon, ANP even induced a significarttly higher cGMP response than CNP. To test which cells were targets to the actions of the peptides, brain slices were incubated with fluorocitrate (a drug that selectively blocks the metabolism of glial cells). Fluorocitrate totally blocked the ANP-evoked cGMP responses in brain slices. In contrast, fluorocitrate reduced only partially the responses evoked by sodium nitroprusside (a drug that stimulates soluble guanylate cyclase, which is contained predomi- nantly in neurons). ]Likewise, the cGMP response induced by CNP was only partially affected by fluorocitrate. These results indicate that: (1) CNP is not more potent than ANP in terms of its ability to generate cGMP in rat brains; (2) brain cells generating cGMP upon exposure to ANP are predominantly glial; and (3) CNP-responsive cells are partly glial, but be- long at least in part to a different compartment than ANP-responsive cells. All together, these data suggest that there is a functional separation between the effects of ANP and CNP in rat brains.

Keywords: Atrial natriuretic peptide (ANP); C-type natriuretic peptide (CNP); Cyclic guanosine monophosphate (cGMP); Brain; Glial cells; Fluorocitrate

* Corresponding author. Fax: + 1 (514) 9875717.

0167-0115/95/$9.50 @ 1995 Elsevier Science B.V. All rights reserved SSD1 0 1 6 7 - 0 1 1 5 ( 9 5 ) 0 0 0 1 8 - 6

56 J. Gon(alves et al. / Regulatory Peptides 57 (1995) 55-63

1. Introduction

The natriuretic peptides are a family of structur- ally related molecules that affect systemic blood pressure and intravascular volume via a combined effect on a variety of organs (for reviews, see [1,2]). To date, three major subtypes have been identified, i.e., 'atrial', 'brain', and 'C-type' natriuretic peptides (abbreviated as ANP, BNP, and CNP, respectively). These peptides exert most of their effects via guany- late cyclase linked receptors. Two kinds of such re- ceptors have been characterized: ANP and BNP bind preferentially to guanylate cyclase-A (GC-A), whereas CNP recognizes guanylate cyclase-B (GC-B) preferentially [3,4]. CNP is the most re- cently identified member of this family of peptides [5]. By Northern blot analysis, its mRNA was found to be much more abundant in the central nervous system than in any other tissue [5,6]. Furthermore, the concentration of the CNP peptide in human and porcine brains is also much higher than that of ANP or BNP [7-9]. Taken together, these data have led to the suggestion that CNP was the predominant natriuretic peptide in the brains of mammals.

A variety of studies performed in rats have shown that natriuretic peptides control a variety of brain functions, including neuroendocrine secretions [ 10,11 ], salt appetite [ 12], thirst [ 13,14], blood pres- sure [15-17], sympathetic tone [18], dopamine turnover [19] and amiloride-sensitive sodium up- take [20]. Most of these studies have utilized ANP, sometimes at very high doses. If these previously described effects resulted from a cross-recognition of the CNP-specific GC-B receptor by ANP, they may possibly have underestimated the biopotency of natriuretic peptides or even failed to appreciate the full spectrum of their activity [2]. However, the bio- potency of CNP vs. that of ANP in rat brain has not been carefully investigated yet, and it is therefore important to compare the activity of both peptides in order to know which one is physiologically rel- evant.

Most of the cellular actions of ANP are mediated

via the intracellular generation of cyclic guanosine monophosphate (cGMP) [1 ]. The identity of brain cells generating cGMP in response to exposure to ANP has recently been investigated by using immu- nocytochemistry [22,23]. By then comparing the distribution of cGMP-containing cells in brain slices to that of glial acidic fibrillary protein (GFAP, an astrocyte-specific marker), it was determined that almost all ANP-responsive cells were in fact astro- cytes. This observation is important, because if the brain actions of ANP do indeed require the genera- tion of cGMP, it means that ANP might affect neu- ronal functions via an effect on non-neuronal cells. The validity of these conclusions depends on how well the immunocytochemical techniques previously used were able to detect all cGMP that was gener- ated in the tissue. Nonetheless, these data are in good agreement with other observations showing that in primary cultures, guanylate-cyclase coupled receptors were present mostly on glial, but not on neuronal cells [24]. Again, little is known about how CNP compares to ANP in this regard.

The aim of our studies was three-fold. We wanted to: (1) compare the amplitudes of the cGMP re- sponses elicited by ANP and CNP in rat brain slices obtained from different regions; (2) use a different method than immunocytochemistry to investigate whether brain cells generating cGMP in response to ANP were predominantly glial; and (3) test whether cells generating cGMP in response to CNP belonged to the same compartment as ANP-responsive cells. To address the latter two questions, we investigated the effects of fluorocitrate, a drug that selectively blocks the metabolism of glial cells in brain slices [25,26], on both ANP- and CNP-evoked cGMP responses.

2. Materials and methods

2.1. Animals

All rats were maintained on a 12-12 h light-dark cycle (lights on at 7:00 a.m.) with rat chow and tap

J. Gonqalves et al. /Regulatory Peptides 57 (1995) 55-63 57

water available ad libitum. Most studies used 200- 250 g male Sprague-Dawley rats (Charles River, Montreal, Quebec, Canada). For ontogeny studies, pregnant Sprague-Dawley rats were obtained from the same vendor, and pups of both sex were used at 3, 6 or 9 days of age.

2.2. Brain slice pi,ocedures

Rats were decapitated and brains were removed and placed in cold Krebs-Ringer solution (118 mM NaC1, 25 mM NaHCO 3, 4.7 mM KC1, 1.2 mM KH2PO4, 2.5 mM CaC12, 1.2 mM MgSO4, 10 mM glucose, at pH 7.4). The diencephalon was dissected out starting from the optic chiasm (anteriorly) up to the mamillary bodies (posteriorly). In certain cases, the frontal cortical lobe was also collected and dis- sected free from the hippocampus and the striatum. All dissections were carried out with cooled dissec- tion instruments, and the regions of interest were then sliced into 200-250/~m transverse slices using a Sorvall tissue chopper (DuPont, Wilmington, DE). The procedure used to incubate and maintain the slices in viable condition was adapted from the con- ditions that had l~)een used previously by others for kidney slices [27]. Briefly, 12 slices were collected from each piece of either diencephalon or frontal cortex and transfi,~rred to borosilicate glass vials con- raining 4 ml of cold Krebs-Ringer supplemented with 1 mM isobutylmethylxanthine (IBMX; Aldrich, Mil- waukee, WI) to :inhibit phosphodiesterase activity. Each vial received its own supply of a 5 ~o COz- 95 ~o 02 gas mixture via a 20-gauge needle inserted through a plastic: snap-cap and fitted to a plastic tubing that was connected to the gas tank. The slices were preincubate,:l under such conditions for 90 min in a shaking water bath at 37°C, either in the pres- ence or absence of 1 mM fluorocitrate (Sigma, St. Louis, MO), a selective glial toxin (a stock solution was freshly dissolved in hot 0.25 M HC1 before each experiment). To :remove fluorocitrate at the end of the preincubation periods, each slice was washed three times in Krebs-Ringer buffer and then trans-

ferred to new vials containing freshly aerated Krebs- Ringer buffer prewarmed at 37 ° C and supplemented with 1 mM IBMX. The slices were then further in- cubated for a total of 40 min, with some of the fol- lowing agents being added for the last 10 min of the incubation: isoproterenol (Sigma), sodium nitro- prusside (Sigma), rat ANP 1-28 (Peninsula, Bel- mont, CA) or porcine CNP-22 (Peninsula). At the end of the incubation, each slice was rapidly trans- ferred to a microcentrifuge tube containing 250 #1 of boiling 0.1 M HC1 and boiled for 5 min. The tubes were then placed on ice for 20 min and the tissues were subsequently homogenized by a 3 s burst of sonication. The proteins in the samples were pre- cipitated by 3 min of centrifugation at 10,000 rpm. The clear supernatants were transferred to a new tube and stored at 4 °C for subsequent measurement of cGMP by radioimmunoassay. The protein pellets were dissolved in 250 #1 of 1.0 M NaOH and main- tained at 4 °C for 48 h before determination of total protein content.

2.3. Biochemical assays

Total protein in the redissolved pellets was deter- mined according to the Bradford method, using bo- vine thyroglobulin as a standard [28]. To determine the concentrations of cyclic nucleotides, the super- natants were first acetylated by adding 12.5/~1 of a 1:2 mix of acetic anhydride and triethylamine to 50 #1 aliquots. After 3 min of incubation, the acetylation reaction was stopped by diluting the samples with sodium acetate buffer (pH 6.2) to a final volume of 500/A. Cyclic G MP or AMP were then measured in 50 #1 aliquots of acetylated sample. The antibodies used for the assays were raised at the Clinical Re- search Institute of Montreal, using conditions simi- lar to what had been described previously [29]. The conditions used for the radioimmunoassay have been described in detail previously [30].

58 J. Gonqalves et al. / Regulatory Peptides 57 (1995) 55-63

2.4. Data analysis

All results are expressed as means + S.D. Differ- ences between groups were tested by one-way analy- sis of variance (ANOVA) followed by Fisher's least- significant difference (LSD) test, or by two-way ANOVA followed by Scheffe's post-hoc test.

3. R e s u l t s

In diencephalon, both ANP and CNP (10 - 7 M) induced a clear elevation of cGMP (Fig. 1). The effect was found to be highly significant (P< 0.001) by one-way ANOVA analysis, and post-hoc com- parisons by Fisher's LSD test revealed that both peptides induced a response that was significantly different from control, but that the response obtained with ANP was greater than the one obtained with CNP (P< 0.001). In cortex, both peptides induced a significant elevation of cGMP (P< 0.005), although the responses were not significantly different from each other. Additional experiments were conducted using diencephalon slices from 3-day-, 6-day- and

9-day-old animals (data not shown). Although the basal levels of cGMP were higher in brains from younger animals, the pattern and amplitudes of the responses induced by ANP and CNP were similar to what was obtained with brain slices from adult rats.

All further experiments were performed with brain slices from diencephalons. To verify that fluoroci- trate effectively blocked the metabolism of glial cells in brain slices, we tested the effect of the drug on the isoproterenol-induced generation of cyclic adenos- ine monophosphate (cAMP) (Fig. 2). By two-way ANOVA analysis, we found a significant effect of preincubation with fluorocitrate (P<0.001), a significant effect of treatment with isoproterenol (P < 0.01), and the interaction between fluorocitrate and isoproterenol was significant as well (P < 0.025). In the slices incubated in the absence of fluoroci- trate, isoproterenol (10-4 M) induced a 2.5-fold in- crease in cAMP, and the difference was significant by Scheffe's post-hoc test (P<0.025). In contrast, there was no significant increase in cGMP in the groups that had been preincubated in the presence of fluorocitrate. These data indicated that fluoroci-

I CON ~ CNP ~ ANP

8

c

6 o.

-6 4 E o.

o. 2

# - - 1

.k~r ~r

rt=18

n=23

D I E N

n=8 /r/r

n=6

CTX

Fig. 1. Effects of ANP and CNP (10 -7 M)on cGMP accumu- lation in incubated rat brain slices from either diencephalon (DIEN) or cortex (CTX). All results are expressed as means _+ S.D. The number of experimental samples in each group (n) is as indicated. *P<0.05 vs. control (CON); **P<0.005 vs. control; ***P<0.001 vs. control; #P<0.001 vs. the CNP group.

lOO

© 80

~, so

~ 4o o .

13.

~ 20

o

I n = 9 I ~ con

I S O

n = 9 n = 9

CTL FC

Fig. 2. Effect 10 -4 M isoproterenol (ISO) on cAMP accumula- tion in rat brain diencephalon slices preincubated either in the absence (control = CTL) or in the presence of fluorocitrate (FC). All results are expressed as means + S.D. **P<0.025 vs. control (CTL).

J. Gon¢alves et al. / Regulatory Peptides 57 (1995) 55-63 59

c

o

o E. a .

o..

8 !

6 -

4 -

2 -

- - ' 1 n-8

n-7 n - 9

I ICON

I A N P

CTL FC

c

o .

o E Q.

13-

c

o o .

D o E o .

13.

~o

n-7

8 "

6 "

4 "

2 - n - .

CTL

3 r

~t~t

FC

0 CTL FC

I ICON

I SNP

CON

I CNP

Fig. 3. Effects of three different agents on the cGMP accumula- tion in rat brain diencephalon slices preincubated either in the absence (control = CTL) or in the presence of fluorocitrate (FC). All results are expressed as means + S.D. (a, top) The cGMP- inducing agent was 10- 7 M ANP; *P< 0.05 between the groups indentified by the brackets. (b, middle) The cGMP-inducing agent was 10 -4 M sodium nitroprusside (SNP); ***P< 0.001, **P<0.025 between Lhe groups indentified by the brackets. (c, bottom) The cGMP-inducing agent w a s 1 0 - 7 M CNP.

trate did indeed effectively block the generation of cAMP within glial cells.

We then tested the effect of fluorocitrate preincu- bation on the generation of c G M P induced by 3 different agents (Fig. 3). When ANP was used as the treating agent (Fig. 3, top), we observed a significant fluorocitrate effect (P < 0.05), a significant treatment effect (P<0.05), and a significant interaction be- tween fluorocitrate preincubation and isoproterenol treatment (P < 0.05). Post-hoc analysis revealed that ANP induced a significant c G M P response in con- trol slices ( P < 0.05), but not in slices preincubated with fluorocitrate. The effect of fluorocitrate on so- dium nitroprusside (SNP)-induced c G M P genera- tion was quite different (Fig. 3, middle). Similar to what we had observed with ANP, there was a sig- nificant fluorocitrate effect (P<0.01), a significant treatment effect (P<0.001), and a significant inter- action between fluorocitrate preincubation and SNP treatment ( P < 0.05). However, SNP (10 -4 M ) in- duced a significant c G M P elevation both in slices preincubated without ( P < 0.001) and with fluoroci- trate (P<0.001). Nonetheless, fluorocitrate did at- tenuate the effect of SNP, since the amount o f c G M P in slices preincubated with fluorocitrate and treated with SNP was significantly lower ( P < 0.025) than in SNP-treated slices preincubated in the absence of fluorocitrate.

The general pattern of the interaction between fluorocitrate and CNP was similar to what we have observed with SNP (Fig. 3, lower). Thus, we ob- served a significant fluorocitrate effect ( P < 0.001), a significant treatment effect ( P < 0.001), and a signif- icant interaction between fluorocitrate preincubation and CNP treatment ( P < 0.025). CNP (10-7 M) in- duced a significant c G M P elevation both in slices preincubated without ( P < 0.001) and with fluoroci- trate ( P < 0.025). However, the amount of c G M P in slices preincubated with fluorocitrate and treated

***P<0.001, **P<0.025 between the groups indentified by the brackets.

60 J. Gon~alves et al. / Regulatory Peptides 57 (1995) 55-63

with CNP was significantly lower (P< 0.001) than in CNP-treated slices preincubated in the absence of fluorocitrate.

4. Discussion

The data presented in the present paper allow us to make three important conclusions: (1) CNP is not more potent than ANP in terms of its ability to generate cGMP in rat brains; (2) brain cells gener- ating cGMP upon exposure to ANP are predomi- nantly glial; and (3) brain cells generating cGMP upon exposure to CNP are partly glial, but belong at least in part to a different compartment than ANP- responsive cells.

It had been suggested previously that CNP was the predominant natriuretic peptide in brain [6-9], but its potency has not been compared to that of ANP in rat brains. We therefore compared the abil- ity of both peptides to generate cGMP in rat brain slices. The effects of BNP were not investigated be- cause this peptide (in contrast to the previous two) is either not present or undetectable in rat brains [ 5 ]. We observed that the cGMP responses elicited by CNP were never higher than those evoked by ANP. This was somewhat in contrast with results we have obtained previously with cultured cells. Indeed, we have observed that CNP induced much greater cGMP responses in cultured rat astrocytes than ANP, and that these cells expressed preferentially CNP-specific GC-B receptors rather than GC-A re- ceptors [31 ]. Since these cells were derived from the brains of neonatal animals, we investigated whether the brain might be more responsive to CNP during early stages of development. However, there was no difference in the relative potency of CNP vs. ANP in brains taken from rats as young as 3-day-old. The predominance of GC-B vs. GC-A receptors in cul- tured astrocytes might therefore represent a culture- related artifact, and not be truly representative of the distribution of receptors on cells in their native brain environment. Similar differences between in vivo or cultured cells have been previously reported for vas-

cular smooth muscle cells, where a shift from GC-A to GC-B receptors occurs after repeated passages in culture [32].

We also observed regional differences in the re- sponses of brain slices. Indeed, ANP induced sta- tistically higher cGMP responses than CNP in slices taken from diencephalons. The higher sensitivity of the diencephalon to ANP is consistent with previous reports indicating that ANP is much more abundant in hypothalamic nuclei than in other brain nuclei [33,34], and that it may modulate the activity of a variety ofhypothalamic neurons [35,36]. Since ANP was used at doses (10-7 M) where it does not bind to GC-B receptors [ 31,37], these data indicate that the effects of ANP do not result from a mere cross- recognition of CNP-preferring GC-B receptors, and that this peptide has specific effects on its own in rat brains. This is in agreement with other data show- ing that: (1) bona fide ANP gene mRNA transcripts are present within the rat central nervous system [38]; (2) in rat brains, the concentration of ANP is comparable in magnitude to that of CNP [5]; and (3) some brain functions are affected by central ad- ministration of anti-ANP monoclonal antibodies which do not cross-react with CNP [17].

We were particularly interested in verifying which brain cells generated cGMP in response to ANP and CNP, since it has been reported that ANP- responsive cells were mainly astrocytes [22-24]. We therefore took advantage of the properties of fluo- rocitrate, a drug that is taken up selectively by glial cells and reversibly inhibits their metabolic activity [26]. Using this drug, it has been possible for in- stance to demonstrate that adenylate-cyclase- coupled fl-adrenergic receptors were mainly located in glial cells, since fluorocitrate inhibits the genera- tion of cAMP in brain slices incubated with isopro- terenol [25,39]. After verifying that fluorocitrate could indeed inhibit the isoproterenol-induced gen- eration of cAMP, we tested the effect of this drug on brain slices treated with ANP. Fluorocitrate totally inhibited the ANP-induced generation of cGMP, in- dicating that the majority of guanylate-cyclase

J. Gonfalves et aL / Regulatory Peptides 57 (1995) 55-63 61

coupled ANP-receptors are indeed located in glial cells. Specificity of the effect of flurocitrate was fur- ther investigated by testing its effect on sodium ni- troprusside, a drug that specifically stimulates soluble guanylate cyclase but does not activate par- ticulate guanylate cyclase [40]. Whereas glial cells (at least in culture) do contain soluble guanylate cy- clase [41], the btdk of this enzyme within the brain is contained with:in the neuronal compartment [42]. Accordingly, the nitroprusside-induced generation of cGMP was only partly affected by fluorocitrate. These data indicate that fluorocitrate: (1) did not affect the generation of cGMP by a general non- specific effect, and (2) did not totally inhibit the ac- tivity of a guanylate cyclase that was partly con- tained in a non-glial compartment.

Similar to what we had observed with sodium nitroprusside, we found that fluorocitrate affected only partially the generation of cGMP induced by CNP. This indicated that guanylate-cyclase coupled CNP receptors belonged at least in part to a different compartment than ANP-responsive cells (either neurons, or other fluorocitrate-insensitive cells). This functional separation is consistent with data show- ing that centrally administered CNP has effects that are different, and sometimes opposite, from those obtained from ANP [43-45]. The importance of CNP in the brain may also vary greatly between species. For instance, it has recently be shown that only ANP or BNP could stimulate the generation of cGMP in guinea pig cerebellum, and there was no evidence for responses mediated through the CNP- specific GC-B receptor [46]. In contrast, CNP is much more abundant than ANP or BNP [7-9] in human and porcine brains, although the relative po- tencies of these peptides on biological responses is not known. All together, the picture that emerges is not that of one peptide being more important than the other in brain, but that of two peptides mediat- ing specific, different (and maybe species-specific) effects. Further work is needed to understand which precise structures and brain functions are preferen- tially affected by either ANP or CNP.

The exclusive glial localization of ANP-responsive GC-A receptors suggests that ANP may act in rat brains in a very particular way, and that it may modulate neuronal function via its effects on astro- cytes. However, it is not known which actions of ANP in the brain truly depend on the generation of cGMP. Conceivably, the possibility still remains that ANP exerts its effects on neurons independently of the generation of intracellular cGMP, since ANP has been shown to affect calcium conductances in adrenal glomerulosa [47] and in renal cortical thick ascending limb cells [48] in a cGMP-independent manner. It will be important to determine whether such effects may occur in central neurons as well. Since the two effects are not mutually exclusive, we will also need to assess how the generation of cGMP in astrocytes may affect the functions of neighbour- ing neurons.

Acknowledgements

This work was supported by a grant of the Medi- cal Research Council of Canada to the Multidisci- plinary Research Group on Hypertension and a grant from the 'fondation des maladies du coeur du Qu6bec' to C.F.D. We thank Dr. G. Thibault for providing the anti-cyclic nucleotides antisera, Mich- eline Vachon for performing some radioimmunoas- says, and Sylvie Picard for her outstanding techni- cal help. C.F.D. was supported by the F.R.S.Q. as 'chercheur-boursier senior'.

References

[ 1] Brenner, B.M., Ballerman, B.J., Gunning, M.E. and Zeidel, M.L., Diverse biological actions of atrial natriuretic peptide, Physiol. Rev., 70 (1990) 665-699.

[ 2] Samson, W.K., Natriuretic hormones. A family of hor- mones, Trends Endocrinol. Metab., 3 (1992) 86-90.

[ 3] Wong, K.-F. and Garbers, D.L., Receptor guanylyl cycla- ses, J. Clin. Invest., 90 (1992) 299-305.

[ 4] Garbers, D.L., Guanylyl cyclase-linked receptors, Pharma- col. Ther., 50 (1991) 337-345.

62 J. Gonqalves et al. / Regulatory Peptides 57 (1995) 55-63

[ 5] Minamino, N., Aburaya, M., Kojima, M., Miyamoto, K., Kangawa, K. and Matsuo, H., Distribution of C-type natri- uretic peptide and its mRNA in rat central nervous system and peripheral tissue, Biochem. Biophys. Res. Commun., 197 (1993) 326-335.

[ 6] Kojima, M., Minamino, N., Kangawa, K. and Matsuo, H., Cloning and sequence analysis of a cDNA encoding a pre- cursor for rat C-type natriuretic peptide (CNP), FEB S Lett., 276 (1990) 209-213.

[ 7] Togashi, K., Kameya, T., Kurosawa, T., Hasegawa, N. and Kawakami, M., Concentrations and molecular forms of C-type natriuretic peptide in brain and cerebrospinal fluid, Clin. Chem., 38 (1992) 2136-2139.

[ 8] Ueda, S., Minamino, N., Aburaya, M., Kangawa, K., Mat- sukura, S. and Matsuo, H., Distribution and characteriza- tion of immunoreactive porcine C-type natriuretic peptide, Biochem. Biophys. Res. Commun., 175 (1991) 759-767.

[ 9] Minamino, N., Makino, Y., Tateyama, H., Kangawa, K. and Matsuo, H., Characterization of immunoreactive human C-type natriuretic peptide in brain and heart, Bio- chem. Biophys. Res. Commun., 179 (1991) 535-542.

[ 10] Samson, W.K., Atrial natriuretic factor inhibits dehydration and hemorrhage-induced vasopressin release, Neuroendo- crinology, 40 (1985) 277-279.

[11] Januszewicz, P., Thibault, G., Garcia, R., Gutkowska, J., Genest, J. and Cantin, M., Effect of synthetic atrial natri- uretic factor on arginine vasopressin release by the rat hypothalamo-neurohypophyseal complex in organ culture, Biochem. Biophys. Res. Commun., 134 (1986) 6552-6558.

[ 12] Antunes-Rodrigues, J., McCann, S.M. and Samson, W.K., Central administration of atrial natriuretic factor inhibits saline preference in the rat, Endocrinology, 118 (1986) 1726-1728.

[13] Antunes-Rodrigues, J., McCann, S.M., Rogers, L.C. and Samson, W.K., Atrial natriuretic factor inhibits dehydra- tion- and angiotensin II-induced water intake in the con- scious, unrestrained rat, Proc. Natl. Acad. Sci. USA, 82 (1985) 8720-8723.

[14] Katsuura, G., Nakamura, M., Inouye, K., Kono, M., Nakao, K. and Imura, H., Regulatory role of atrial natri- uretic polypeptide in water drinking in rats, Eur. J. Phar- macol., 121 (1986) 285-287.

[15] Shimizu, T., Katsuura, G., Nakamura, M., Nakao, K., Morii, N., Itoh, Y., Shiono, S. and Imura, H., Effect of intracerebroventricular atrial natriuretic polypeptide on blood pressure and urine production in rats, Life. Sci., 39 (1986) 1263-1270.

[16] Levin, E.R., Weber, M.A. and Mills, S., Atrial natriuretic factor-induced vasodepression occurs through central ner- vous system, Am. J. Physiol., 255 (1988) H616-H622.

[ 17] Yang, R.H., Jin, H., Chen, Y.F., Wyss, J.M. and Oparil, S.,

Blockade of endogenous anterior hypothalamic atrial natri- uretic peptide with monoclonal antibody lowers blood pres- sure in spontaneously hypertensive rats, J. Clin. Invest., 86 (1990) 1985-1990.

[18] Schultz, H.D., Steele, M.K. and Gardner, D.G., Central administration of atrial peptide decreases sympathetic out- flow in rats, Am. J. Physiol., 258 (1990) R1250-R1256.

[19] Kadowaki, K., Hirota, K., Koike, K., Ohmichi, M., Miy- ake, A. and Tanizawa, O., Atrial and brain natriuretic pep- tides enhance dopamine accumulation in cultured rat hy- pothalamic cells including dopaminergic neurons, Neuroendocrinology, 56 (1992) 11-17.

[20] Kanda, F., Sarnacki, P. and Arieff, A.I., Atrial natriuretic peptide inhibits amiloride-sensitive sodium uptake in rat brain, Am. J. Physiol., 263 (1992) R279-R283.

[21] Mc Carthy, R.T., Isales, C.M., Bollag, W.B., Rasmussen, H. and Barrett, P.Q., Atrial natriuretic peptide differentially modulates T- and L-type calcium channels, Am. J. Physiol., 258 (1990) F473-F478.

[22] de Vente, J., Manshanden, C.G., Sikking, R.A., Ramaek- ers, F.C.S. and Steinbusch, H.W.M., A functional param- eter to study heterogeneity of glial cells in rat brain slices: cyclic guanosine monophosphate production in atrial natri- uretic factor (ANF)-responsive cells, Glia, 3 (1990) 43-54.

[23] De Vente, J., Bol, J.G.J.M. and Steinbusch, H.W.M., cGMP-producing, atrial natriuretic factor-responding cells in the rat brain, Eur. J. Neurosc., 1 (1989) 436-460.

[24] Levin, E.R., Frank, H.J.L., Gelfand, R., Loughlin, S.E. and Kaplan, G., Natriuretic peptide receptors in cultured rat diencephalon, J. Biol. Chem., 265 (1990) 10019-10024.

[25] Stone, E.A., Sessler, F.M. and Weimin, L., Glial localiza- tion of adenylate-cyclase-coupled fl-adrenoreceptors in rat forebrain slices, Brain. Res., 530 (1990) 295-300.

[26] Paulsen, R.E., Contestabile, A., Villani, L. and Fonnum, F., The effect of fluorocitrate on transmitter amino acid release from striatal slices, Neurochem. Res., 13 (1988) 637-641.

[27] Urban, J.H. and Van de Kar, L.D., An improved method for the measurement of renin release from coronal, vibratome-cut kidney slices, J. Pharmacol. Methods, 15 (1986) 245-253.

[28] Bradford, M.M., A rapid and sensitive method for the quan- titation of microgram quantities of protein utilizing the prin- ciple of protein-dye binding, Anal. Biochem., 72 (1976) 248-254.

[29] Steiner, A.L., Parker, C.W. and Kipnis, D.M., Radio- immunoassay for cyclic nucleotides, J. Biol. Chem., 247 (1972) 1106-1113.

[30] Qing, G. and Garcia, R., Characterization of plasma and tissue atrial natiruretic factor during development of mod- erate high output heart failure in the rat, Cardiovasc. Res., 27 (1993) 464-470.

J. Gon~alves et aL / Regulatory Peptides 57 (1995) 55-63 63

[31] Deschepper, C.F. and Picard, S., Effects of C-type natri- uretic peptide oa rat astrocytes: regional differences and characterization of receptors, J. Neurochem., 62 (1994) 1974-1982.

[32] Suga, S.-I., Nakao, K., Kishimoto, I., Hosoda, K., Mukoyama, M., Arai, H., Shirakami, G., Ogawa, Y., Ko- matsu, Y., Nakagawa, O., Hama, N. and Imura, H., Phenotype-related alteration in expression of natriuretic peptide receptor:; in aortic smooth muscle cells, Circ. Res., 71 (1992) 34-39.

[33] Jin, H., Chen, Y.F., Yang, R.-H., Meng, Q.C. and Oparil, S., Impaired release of atrial natriuretic factor in NaCI- loaded spontaneously hypertensive rats, Hypertension, 11 (1994) 739-744.

[34] Bahner, U., Geiger, H., Palkovits, M. and Heidland, A., Atrial natriuretic factor in specific brain areas of spontane- ously hypertensive rats, Hypertension, 12 (1988) 519-524.

[ 35 ] Yamashita, H. and Kannan, H., Inhibition of hypothalamic neurons by the atrial natriuretic peptide family, News Physiol. Sci., 7 (1992) 75-79.

[36] Oparil, S. and Wyss, J.M., Atrial natriuretic factor in cen- tral cardiovascular control, News Physiol. Sci., 8 (1993) 223-228.

[37] Koller, K.J., Lowe, D.G., Bennett, G.L., Minamino, N., Kangawa, K., Matsuo, H. and Goeddel, D.V., Selective activation of the B natriuretic peptide receptor by C-type natriuretic pepti6e, Science, 252 (1991) 120-123.

[38] Gardner, D.G., '~qasuk, G.P., Baxter, J.D., Fiddes, J.C. and Lewicki, J.A., Identification of atrial natriuretic factor gene transcripts in the central nervous system of the rat, Proc. Natl. Acad. Sci. USA, 84 (1987) 2175-2179.

[39] Stone, E.A. and John, S.M., Further evidence for a glial localization of rat cortical fl-adrenoreceptors: studies of in

vivo cyclic AMP responses to catecholamines, Brain Res., 549 (1991) 78-82.

[40] Waldmann, S.A. and Murad, F., Cyclic GMP synthesis and function, Pharmacol. Rev., 39 (1987) 163-196.

[41] Simmons, M.L. and Murphy, S., Induction of nitric oxide synthase in glial cells, J. Neurochem., 59 (1992) 897-905.

[42] De Vente, J. and Steinbusch, H.W.M., On the stimulation of soluble and particulate guanylate cyclase in the rat brain and the involvement of nitric oxide as studied by cGMP immunocytochemistry, Acta Histochem., 92 (1992) 13-38.

[43] Samson, W.K., Skala, K.D. and Huang, F.-L.S., CNP-22 stimulates, rather than inhibits, water drinking in the rat: evidence for a unique biological action of the C-type natri- uretic peptides, Brain Res., 568 (1991) 285-288.

[44] Charles, C.J., Richards, A.M. and Espiner, E.A., Central C-type natriuretic peptide but not atrial natriuretic factor lowers blood pressure and adrenocortical secretion in nor- mal conscious sheep, Endocrinology, 131 (1992) 1721-1726.

[45] Samson, W.K., Huang, F.-L. and Fulton, R.J., C-type natriuretic peptide mediates the hypothalamic actions of the natriuretic peptides to inhibit luteinizing hormone secretion, Endocrinology, 132 (1993) 504-509.

[46] Hernandez, F., Alexander, S.P.H. and Kendall, D.A., Natriuretic peptide-induced cyclic GMP accumulation in adult guinea-pig cerebellar slices, Br. J. Pharmacol., 113 (1994) 216-220.

[47] Isales, C.M., Lewicki, J.m., Nee, J.J. and Barrett, P.Q., ANP-(7-23) stimulates a DHP-sensitive Ca z÷ conductance and reduces cellular cAMP via a cGMP-independent mechanism, Am. J. Physiol., 263 (1992) C334-C342.

[48] Dai, L.-J. and Quamme, G.A., Atrial natriuretic peptide initiates Ca 2 ÷ transients in isolated renal cortical thick as- cending limb cells, Am. J. Physiol., 265 (1993) F592-F597.